2011 33 RD ANNUAL INFORMS MARKETING SCIENCE CONFERENCE

2011 33RD ANNUAL INFORMS MARKETING SCIENCE CONFERENCE June 9-11, 2011 Hotel InterContinental, in Houston, Texas business.rice.edu BOSTON2012 2012 INFORMS Marketing Science Conference June 7 – June 9, 2012 Boston, MA Join us in historic Boston, MA, USA www.bu.edu/marketingscience2012 Conference Co-Chairs: Shuba Srinivasan and Patrick Kaufmann Boston University School of Management 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas OUR SINCERE THANKS TO OUR SILVER LEVEL UNDERWRITER. GUROBI OPTIMIZATION, INC. GET GUROBI OPTIMIZER 4.5 NOW WWW.GUROBI.COM ii JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas INFORMS SOcIeTy OF MaRkeTINg ScIeNce BOaRD aND aDVISORy cOUNcIL The Institute of Operations Research and Management Sciences (INFORMS) is an international, not-for-profit scientific society with 10,000 members, including Nobel laureates, dedicated to applying scientific methods to help improve decision-making, management, and operations. The Marketing Science Conference falls under the auspices of the INFORMS Society for Marketing Science (ISMS) sub-branch whose major purpose is to foster the development, dissemination, and implementation of knowledge, basic and applied research, and science and technologies that improve the understanding and practice of marketing. INFORMS SOcIeTy FOR MaRkeTINg ScIeNce BOaRD aDVISORy cOUNcIL President Donald R. Lehmann, Columbia University Scott A. Neslin, Dartmouth College Jean-Pierre Dubé, University of Chicago Treasurer Gerard (Gerry) J. Tellis, University of Southern California Baohong Sun, Carnegie Mellon University Secretary Brian T. Ratchford, University of Texas at Dallas Newsletter Editor Luc Wathieu, Georgetown University Vice President of Meetings Fred Feinberg, University of Michigan Vice President of Electronic Communications Peter T. L. Popkowski Leszczyc, University of Alberta Vice President of Practice V. Kumar, Georgia State University Vice President of Education Bart Bronnenberg, Tilburg University Vice President of Membership Min Ding, Pennsylvania State University Vice President of External Relations Gary Lilien, Pennsylvania State University President Elect Kannan Srinivasan, Carnegie Mellon University Past President Rick Staelin, Duke University INFORMS Liaison Officer Paul Messinger, University of Alberta business.rice.edu iii 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas INFORMS MaRkeTINg ScIeNce cONFeReNce 1979 Stanford University David B. Montgomery, Dick Wittink 1997 University of California, Berkley Tulin Erdem, Miguel Villas-Boas, Russell Winer 1980 University of Texas, Austin Robert Leone 1998 INSEAD, France Erin Anderson, Hubert Gatignon 1981 New York University John Keon 1999 Syracuse University Amiya Basu, T. Mazumdar, S. P. Raj 1982 University of Pennsyilvania Vijay Mahajan, Yoram Wind 2000 University of California, Los Angeles Randolph Bucklin, Donald Morrison 1983 University of Southern California Fred Zufreyden 2001 University of Mainz Oliver Heil 1984 University of Chicago Steven Shugan 2002 University of Alberta Peter T. L. Popkowski Leszczyc 1985 Vanderbilt University Russell Winer, Allan Shocker 2003 University of Maryland Brian Ratchford, Roland Rust, Venky Shankar 1986 University of Texas, Dallas Ram Rao 2004 Erasmus University, Rotterdam Stefan Stremersch 1987 HEC, France Dominique Hanssens, Gilles Laurent 2005 Emory University Sundar Bharadwaj, Douglas Bowman, Sandy Jap 1988 University of Washington Allan Shocker, Robert Jacobson 2006 University of Pittsburgh Rabi Chatterjee, Jeff Inman, R. Venkatesh 1989 Duke University John McCann, Richard Staelin 2007 Singapore Management University Sundar Bharadwaj, Jin K. Han, David B. Montgomery, Chin Tiong Tan 1990 University of Illinois S. Sudharshan 1991 University of Delaware/Dupont Meryl Gardner, John Frey 1992 London Business School Mark Uncles, Gerald Goodhardt 1993 Washington University Chakravarthi Narasimhan 1994 University of Arizona Dipankar Chakravarti, Ambar Rao 1995 University of New South Wales John Roberts, Pamela Morrison 1996 University of Florida Steven Shugan, Barton Weitz iv 2008 University of British Columbia Charles B. Weinberg, Darren Dahl, Daniel Putler 2009 University of Michigan Eugene Anderson, Fred Feinberg 2010 University of Cologne Werner Reinartz, Karen Gedenk, Franziska Völckner 2011 Rice University Richard R. Batsell, Sharad Borle, Ajay Kalra, Amit Pazgal 2012 Boston University Shuba Srinivasan, Patrick Kaufmann 2013 Ozyegin University Tulin Erdem (NYU), Koen Pauwels JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas ReSTaURaNTS IN HOUSTON On the pages which follow we include some representative Houston restaurants as well as other attractions. RANDY’S PICKS LOCATION American/Continental Brennan’s Houston Old world charm with Southern style. 3300 Smith St Price Range: $$$ www.brennanshouston.com Da Marco The braised short ribs are excellent. 1520 Westheimer Rd Price Range: $$$ www.damarcohouston.com Bar-B-Q 5109 Kirby Dr Price Range: $$ www.goodecompany.com Goode Co. Bar B Q Excellent Bar-B-Q in a very authentic Texas setting. Italian Carrabba’s Italian Grill The ribeye steak, osso buco on Wednesdays, the calamari and the quail are all excellent. 3115 Kirby Dr Price Range: $$ www.carrabbas.com French Café Rabelais Authentic French restaurant with the menu on a blackboard and a fine wine selection. 2442 Times Blvd Price Range: $$ www.caferabelais.com Brasserie Max and Julie Excellent and reasonable for brunch on Saturday or Sunday. The Skate fish is also an unusual item. 4315 Montrose Blvd Price Range: $$ www.maxandjulie.net Latin 1800 Post Oak Blvd Price Range: $$$ www.cordua.com Américas Américas is a progressive Pan-Latin dining experience highlighting the gifts of the new world. Mexican (Mex-Mex & Tex-Mex) Pico’s Best margaritas in town – get Top Shelf Frozen. Tasty appetizer: Chilorio. Entres: Red Snapper and Shrimp Adobado are very good. Sometimes there are fresh soft-shell crabs. business.rice.edu 5941 Bellaire Blvd Price Range: $ www.picos.net v 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas ReSTaURaNTS IN HOUSTON (continued) RANDY’S PICKS LOCATION Seafood Pesce High quality fresh seafood. Chef Mark Holley is a master. 3029 Kirby Dr Price Range: $$$ www.pescehouston.com Goode Co. Seafood Very fresh and perfectly seasoned seafood. Also, you will see the only set of four different restaurants owned by one person all located within one block of each other. 2621 Westpark Dr Price Range: $$ www.goodecompany.com Vietnamese Miss Saigon Café Authentic. The Vietnamese coffee is an unusual treat. 5503 Kelvin Dr Price Range: $$ www.miss-saigoncafe-houston.com MANY OTHER GOOD RESTAURANTS IN HOUSTON LOCATION American Mo’s…A Place for Steaks Culinary delights don’t end with the steaks – you’ll find a wide selection of delicious entrées, appetizers, sides and desserts. 1801 Post Oak Blvd Price Range: $$$ www.mosaplaceforsteaks.com Grand Lux Café Eclectic menu offers extraordinary variety and selection. Also offered is an impressive selection of Specialty Cocktails and Martinis, as well as an array of Beer, Wine and After Dinner Drinks. 5000 Westheimer Rd #690 Price Range: $$ www.grandluxcafe.com The Cheesecake Factory The Cheesecake Factory menu boasts more than 200 menu selections made fresh from scratch each day. 5015 Westheimer Rd #3406 Price Range: $$ www.thecheesecakefactory.com Kenny & Ziggy’s Delicatessan Combines traditional New York deli food with contemporary cuisine. 2327 Post Oak Blvd Price Range: $ www.kennyandziggys.com Sullivan’s Steakhouse Vibrant neighborhood American Steakhouse featuring the finest steaks, seafood, hand-shaken martinis and live music. 4608 Westheimer Rd Price Range: $$$ www.sullivansteakhouse.com vi JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas ReSTaURaNTS IN HOUSTON (continued) MANY OTHER GOOD RESTAURANTS IN HOUSTON LOCATION Bar-B-Q 9797 Westheimer Rd; 8777 South Main St Price Range: $ www.pappas.com Pappas Bar-B-Q Old world charm with Southern style. Mexican Lupe Tortilla Authentic Tex-Mex fresh food in a casual, family friendly environment. 2414 Southwest Freeway Price Range: $$ www.lupetortilla.com Seafood Pappas Seafood 3001 S. Shepherd Price Range: $$ www.pappasseafood.com Pappadeaux Seafood Kitchen Several locations Price Range: $$ www.pappadeaux.com Chinese Café Chino Known for its authentic Hunan and Pan Asian cuisine. 3285 Southwest Frwy Price Range:$$ www.cafechinohouston.com P.F. Chang’s China Bistro The menu gives Chinese classics a contemporary tweak. 4094 Westheimer Rd Price Range: $$ www.pfchangs.com Red Pepper Chinese Restaurant Casual Chinese eatery. 5626 Westheimer Rd Price Range: $ www.redpepperchineserestaurant.com Greek Alexander The Great Greek classics and entrees which make use of traditional ingredients in innovative ways. business.rice.edu 3055 Sage Rd Price Range: $$ www.alexanderthegreat.cc vii 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas ReSTaURaNTS IN HOUSTON (continued) MANY OTHER GOOD RESTAURANTS IN HOUSTON LOCATION Greek (continued) 4747 San Felipe St Price Range: $$ www.yiayiamarys.com Yia Yia Mary’s Pappas Greek Kitchen Greek, Mediterranean Indian Bombay Brassierie North Indian cuisine. 3005 West Loop South Price Range: $$ www.thebombaybrasserie.com Kiran’s Restaurant & Bar 4100 Westheimer Rd Price Range: $$$ www.kiranshouston.com Mayuri Indian Restaurant South-North Indian specialties, Indian Chinese. 5857 Westheimer Rd Price Range: $$ www.mayuri.com Madras Pavilon 3910 Kirby Dr #130 Price Range: $$ Indika Progressive Indian cuisine using local ingredients. 516 Westheimer Rd Price Range: $$ www.indikausa.com Italian Ciao Bello A delightful Italian restaurant that delivers serious Italian food in a casual, fun setting that’s the perfect place for family and friends. 5161 San Felipe St #100 Price Range: $$$ www.ciaobellohouston.com Maggiano’s Little Italy Authentic, Italian-American restaurant featuring Chef-prepared dishes and premium wines. 2019 Post Oak Blvd Price Range: $$ www.maggianos.com D’Amico’s Italian Market Café Authentic Northern and Southern Sicilian cuisine; an Italian Deli and Imported Food Market. Casual and relaxed Italian dining setting in the heart of Rice Village. 5626 Westheimer Rd Price Range: $ www.redpepperchineserestaurant.com Valentino Vin Bar Located inside the Hotel Derek in the Galleria area. Italian artfully done, an excellent wine list. 2525 West Loop South Price Range: $$$ www.valentinorestaurantgroup.com/valentinohouston viii JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas ReSTaURaNTS IN HOUSTON (continued) MANY OTHER GOOD RESTAURANTS IN HOUSTON LOCATION Latin 3055 Sage Rd Price Range: $ Phone: 713-622-8877 Argentina Cafe Old world charm with Southern style. Pizza Star Pizza Authentic Tex-Mex fresh food in a casual, family friendly environment. 2111 Norfolk St Price Range: $$ www.starpizza.net $ = Under $10 $$ = $11 - $30 $$$ = $31 - $60 $$$$ = Over $61 business.rice.edu ix 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas SIgHTSeeINg IN aND aROUND HOUSTON LOCATION ATTRACTION Houston Museum of Fine Arts Houston Oldest comprehensive museum in the southwest US. Features exhibits from antiquity to present. Has outstanding collections of European art from antiquity to 1920. 1001 Bissonnet Tue-Wed 10am-5pm Thu 10am-9pm Fri-Sat 10am-7pm Sun 12:15pm-7pm www.mfah.org Houston Museum of Natural Science Activities include the Wortham IMAX Theatre, Burke Baker Planetarium, Cockrell Butterfly Center & Insect Zoo, and the museum’s exhibit halls. 1 Hermann Circle Drive Mon, Wed-Sun 9am-5 pm Tue 9am-8 pm www.hmns.org Bayou Bend Collection and Gardens Bayou Bend is a collection of American fine and decorative arts in the restored mansion of philanthropist and Houstonian Ima Hogg surrounded by large formal gardens. 1 Westcott Street Open Tue-Sun. Closed Mon. www.mfah.org/visit/bayou-bend-collection-andgardens Hermann Park Picnic area, jogging trails, McGovern Lake, pedal boats, miniature train, Japanese Garden, children’s playground. 6000 Blk of Fannin www.hermannpark.org Houston Zoo 55 acres in Hermann Park. More than 3,100 animals and more than 500 species. Visit the recently opened African Forest. 1513 N. MacGregor Dr. Daily 9am-7pm www.houstonzoo.org Houston Arboretum & Nature Center 155 acres of wildlife sanctuary. Self guided tours. 4501 Woodway Daily Grounds & Trails 7am-7pm www.houstonarboretum.org Space Center Houston Space history exhibits, giant screen theater, hands-on activities tell the story of NASA’s manned space flight program. SCH is the only place in the world where visitors can see astronauts train for missions, touch a real moon rock, land a shuttle, and take a behind-the-scenes tour of NASA. Approximately 25 miles south of downtown Houston in the NASA/Clear Lake area. 1601 NASA Parkway Daily 10am-7pm www.spacecenter.org Houston Ballet One of the nation’s best ballet companies. Has toured nationally and internationally to overwhelming critical acclaim. “Taming of the Shrew” June 9, 11 & 12 Wortham Theater 501 Texas Avenue www.houstonballet.org x JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas SIgHTSeeINg IN aND aROUND HOUSTON (continued) LOCATION ATTRACTION Houston (continued) 3055 Sage Rd Price Range: $ Phone: 713-622-8877 Argentina Cafe Old world charm with Southern style. Galveston Moody Gardens Looking for island fun? Explore the Aquarium Pyramid®, Discovery Museum or 3D, 4D and Ridefilm theaters. Enjoy the Colonel Paddlewheel Boat or a little summer fun at beautiful Palm Beach. One Hope Blvd. Daily 10am-6pm www.moodygardens.com Schlitterbahn Galveston Island Waterpark The park offers over 33 amazing rides and attractions, including uphill water coasters, thrilling speed slides, kid’s water playgrounds, whitewater rapids, relaxing hot tubs, family raft rides and the Boogie Bahn surf ride. Next to Moody Gardens Daily 10am-8pm www.schlitterbahn.com/gal Railroad Museum and Terminal Includes exhibits of a model train layout of the port of Galveston, Pullman sleepers, cabooses, a diner car, a meal care, steam locomotives and more. The Strand at 25th Daily 10am-5pm www.galvestonrrmuseum.com Texas Seaport Museum/1877 Tall Ship Elissa Share the adventure of the high seas at the Texas Seaport Museum, home of the 1877 tall ship Elissa, a floating National Historic Landmark. Pier 21 #8 Daily 10am-5pm www.tsm-elissa.org Kemah Kemah Boardwalk Known for its exceptional restaurants, Kemah Boardwalk is a 40acre family-oriented entertainment complex. Amusements include a carousel, Ferris wheel, games and rides. business.rice.edu 2nd St. & Bradford Ave. & Waterfront St. Daily www.kemahboardwalk.com xi 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas SUMMaRy ScHeDULe Wednesday, June 8, 2011 8:30am-8:00pm Doctoral Consortium Rice University Conference Registration Continental Breakfast Session 1 (TA) AM Break Session 2 (TB) Lunch Session 3 (TC) PM Break Session 4 (TD) Opening Reception Legends Ballroom Foyer Discovery Center A & B All 15 Meeting Rooms Legends Ballroom Foyer All 15 Meeting Rooms Discovery Center A & B All 15 Meeting Rooms Legends Ballroom Foyer All 15 Meeting Rooms Champions & Legends Foyer Conference Registration Continental Breakfast Session 1 (FA) AM Break Session 2 (FB) Lunch Session 3 (FC) PM Break Session 4 (FD) Cocktails Dinner Served Legends Ballroom Foyer Discovery Center A & B All 15 Meeting Rooms Legends Ballroom Foyer All 15 Meeting Rooms Discovery Center A & B All 15 Meeting Rooms Legends Ballroom Foyer All 15 Meeting Rooms Legends Ballroom Foyer Legends Ballroom Conference Registration Continental Breakfast Session 1 (SA) AM Break Session 2 (SB) Lunch Session 3 (SC) Legends Ballroom Foyer Discovery Center A & B All 15 Meeting Rooms Legends Ballroom Foyer All 15 Meeting Rooms Discovery Center A & B All 15 Meeting Rooms Thursday, June 9, 2011 7:00am-5:00pm 7:30-8:30am 8:30-10:00am 10:00-10:30am 10:30am-12:00pm 12:00-1:30pm 1:30-3:00pm 3:00-3:30pm 3:30-5:00pm 5:30-7:30pm Friday, June 10, 2011 7:00am-5:00pm 7:30-8:30am 8:30-10:00am 10:00-10:30am 10:30am-12:00pm 12:00-1:30pm 1:30-3:00pm 3:00-3:30pm 3:30-5:00pm 6:00-7:00pm 7:00-9:30pm Saturday, June 11, 2011 7:00am-3:00pm 7:30-8:30am 8:30-10:00am 10:00-10:30am 10:30am-12:00pm 12:00-1:30pm 1:30-3:00pm xii JONES GRADUATE SCHOOL OF BUSINESS 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas INFORMaTION FOR SeSSION cHaIRS, PReSeNTeRS, aND PaRTIcIPaNTS Guidelines for session chairs: 1. 2. 3. 4. 5. 6. 7. Please arrive at the room where you are chairing a session at least 10 minutes before its official start time (during the break prior to the session). Plan to be present for the entire session. Choose a seat in the front of the room where a presenter can easily observe you. Introduce the session title at the beginning of the session and announce the allotted time per presenter (3 presentations per session: 30 min; 4 presentations per session: 22 min; 5 presentations per session: 18 min). The allotted time already includes Q&A. If presenters do not show up before the start of the session, you are free to allocate their time to the other presenters. Introduce each speaker at the beginning of his or her talk. Use the provided cards to let speakers know when there are 5 minutes and 2 minutes left in their allotted time. Guidelines for presenters: 1. 2. 3. 4. 5. 6. 7. 8. Please arrive at the room where you will be presenting at least 10 minutes before its official start time (during the break prior to the session). Bring your presentation on a USB key in appropriate format (generally Power Point). Insert your USB key into one of the slots of the laptop and copy your presentation to the desktop. Open your file to ensure it works, then close it. A student helper will be there to assist you. All conference rooms are equipped with a laptop with a USB-port and a beamer. If you are not going to be present for the entire session, please let the session chair know. Depending on the number of presentations in your session, you will have either 30 minutes (3 presentations), 22 minutes (4 presentations), or 18 minutes (5 presentations) to present. The allotted time already includes Q&A. If you do not show up before the start of the session, the chair is free to allocate your time to the other presenters. You will be prompted when there are 5 and 2 minutes left of your allotted time. Guidelines for participants: 1. 2. If you want to attend presentations of two different sessions that run at the same time, we kindly ask you to leave the sessions only at the beginning or end of a talk. Please leave and enter the rooms quietly. business.rice.edu xiii 33RD ANNUAL MARKETING SCIENCE CONFERENCE June 9-11, 2011, Houston, Texas INTeRcONTINeNTaL HOTeL & cONVeNTION ceNTeR MaP www.ichouston.com Tracks 1-10 G RO U N D L E V E L - A D J A C E N T TO L O B B Y Speakers Chairmans Room Boardroom Service Area Business Center IV III Restrooms II Trk 3 Restrooms Tr k8 Tr k9 Tr k1 0 III I Trk 2 Tr k7 F Ba oun llr de oo rs m Gift Shop Discovery Center VI Trk 5 Awards Banquet I Counselors Trk 6 Legends Ballroom IV II Directors Room B VII Breakfast Lunch A V Trk 4 Trk 1 Restrooms Registration Bar Elevators Valet Lobby Lounge Front Desk InterContinental Houston (Near the Galleria) 2222 West Loop South Houston, Texas 77027 (713) 627-7600 Lobby Restaurant Entrance Summit Valet 2 N D F L O O R - A B OV E L E G E N D S B A L L RO O M Tracks 11-15 Service Area III Trk 13 II Restrooms Champions Boardroom IV V Trk 15 VI Trk 12 Trk 14 I VII Trk 11 Champions Conference Center Elevators xiv JONES GRADUATE SCHOOL OF BUSINESS Discovery Center (Below) 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 8.30-10.00 (TA) TA01 – Legends Ballroom I TA02 – Legends Ballroom II TA03 – Legends Ballroom III TA04 – Legends Ballroom V Marketing Science Institute I Google WPP Award Papers Internet Bayesian Econometrics I: Methods & Application Chair: Don Lehmann Chair: Shuba Srinivasan Chair: Anita Elberse The History of Marketing Science: The Early Years Russ Winer A Structural Model of Employee Behavioral Dynamics in Enterprise Social Media Yan Huang, Anindya Ghose Information Available versus Information Acquired? Implications for Consumer Choice Models S. Siddarth, Imran Currim, Ofer Mintz Media Aggregators and the Link Economy: Strategic Hyperlink Formation in Content Networks Chrysanthos Dellarocas, Zsolt Katona, William Rand Investigating the Dynamic Impact of Advertising on Online Search and Offline Sales Jeffrey Dotson, Sandeep Chandukala, Qing Liu, Stefan Conrady The Broadcast Window Effect: Information Discovery and Cross-channel Substitution Patterns for Media Content Rahul Telang, Anuj Kumar, Michael Smith Not to Click Through: The Benefits of Search Engine Advertising - Combination of Old and New Media German Zenetti, Tammo Bijmolt, Daniel Klapper, Peter Leeflang Are Audience Based Online Metrics Leading Indicators of Brand Performance? Shuba Srinivasan, Randolph Bucklin, Koen Pauwels, Oliver Rutz Viral Videos: The Dynamics of Online Video Advertising Campaigns Anita Elberse, Clarence Lee, Lingling Zhang TA05 – Legends Ballroom VI TA06 – Legends Ballroom VII TA07 – Founders I TA08 – Founders II New Product I: Introduction Competition I: Measuring the Impact of Competition in Retail Markets I ASA Special Session on the Marketing-Statistics Interface – I Innovation I: Open Innovation Chair: A. Yesim Orhun Chair: Peter J. Lenk The Impact of Competition on Endogenous Product Provision: The Case of Motion Picture Exhibition Market A. Yesim Orhun, Pradeep Chintagunta, Sriram Venkataraman Dynamic Market Segmentation Models and Methods Timothy J. Gilbride, Peter J. Lenk Chair: Keyvan Dehmamy New Product Design under Channel Acceptance: Brick-and-Mortar, Online Exclusive, or Brick-and-Click Lan Luo, Jiong Sun The Evolution of Research on New Products, Innovation, and Growth Don Lehmann Optimizing the Three Dimensions of New FMCG’s Market Success by Means of the Marketing Mix Tilo Halaszovich, Christoph Burmann An Empirical Analysis of New Product Launch Michael Cohen, Rui Huang Dyadic Choice and Compromise Effects: Implications for Decision Optimality Lin Bao, Neeraj Arora, Qing Liu Choice from Simulated Store Shelves – How Similar are Two Identical SKUs? Keyvan Dehmamy, Thomas Otter Chair: Sanjay Sisodiya Chair: Michael Cohen Observational Learning and Networking Externality in Decision-making Dongling Huang, Andrei Strijnev, Yuanping Ying Dyadic Patent Citation and Firm Performance Yantao Wang, Yi Qian, Sha Yang Empirical Investigation of Retail Expansion and Cannibalization in a Dynamic Environment .S Sriram, V. Kumar, Joseph Pancras How Wal-Mart’s Entry Affects Incumbent Retailers Huihui Wang, Carl Mela, Andrés Musalem Retail Market Expansion and Substitution Effects: Evidence from a Field Experiment Frederico Rossi, Eric Anderson, Ralf Elsner, Duncan Simester Viral Marketing: Understanding the Diffusion of User Generated Content Within and Across Networks Yuchi Zhang, Wendy W. Moe Bayesian Model Selection and Simulation Bias of the Harmonic Mean Estimator of Integrated Likelihoods Peter J. Lenk Open Innovation Practices and Market Outcomes: The Moderating Role of Product Capabilities Deepa Chandrasekaran, Gaia Rubera, Andrea Ordanini Network and Knowledge Asset Alignment in Open Innovation Tanya Tang, Eric Fang, William Qualls Innovative Capability: Investigating Open Innovation, Marketing Capability, and Firm Performance Sanjay Sisodiya, Yany Grégoire, Jean Johnson 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 8.30-10.00 (TA) TA09 – Founders III TA10 – Founders IV TA11 – Champions Center I TA12 – Champions Center II Promotions I Consumer Behavior: Perceptions Direct Marketing Branding Chair: Dinesh Gauri Chair: Jana Diels Chair: Eric Schwartz Chair: Xiaoying Zheng Timing of Retailer Price-promotions Huseyin Karaca, Anne T. Coughlan, Lakshman Krishnamurthi, Vincent Nijs The Effects of New Product Introduction to Different Category Context on Price Evaluations Akihiro Nishimoto, Sayaka Ishimaru, Sotaro Katsumata, Eiji Motohashi Calibration? Definition, Motivation and Insights Learned from a Direct Marketing Setting Kristof Coussement, Wouter Buckinx Customer Based Multidimensional Brand Equity and Asymmetric Risk Kyoung Nam Ha, Gary Erickson, Robert Jacobson Optimizing Target Selection of Direct Mailing by Charities Remco Prins, Bas Donkers Brand Equity and Product Recalls Sheila Goins, Cathy Cole, Qiang Fei, Lopo Rego Test and Learn: A Reinforcement Learning Perspective Eric Schwartz Factors Enhancing a Brand’s Competitive Clout: A Two-step Empirical Analysis Juan Carlos Gázquez-Abad, Agustí Casas-Romeo, Rubén Huertas-García, Francisco J. Martínez-López Stars, Leaders, Free-riders and Losers: Roles in the Category Expansion during Sales Promotions Sergio Meza The Impact of Retailer Promotional Activities on Store Traffic Shyda Valizade-Funder, Oliver Heil, Kamel Jedidi An Empirical Investigation of Retailer Pass-throughs Across Categories Dinesh Gauri, Joseph Pancras, Debabrata Talukdar How Surprisingly Little Thoughts Count on Receiver’s Motivated Appreciation for Giver’s Thoughts Yan Zhang When Looks Can Be Deceptive: Consumer Response to Unfamiliar Product Packaging Descriptors Rishtee Batra The Impact of Marketing Capability on Customer Responses: A Customer-based Brand Equity Perspective Xiaoying Zheng, Siqing Peng, Yi Xie Understanding Customers` Substitution Patterns when Branded Items Become Unavailable Jana Diels, Lutz Hildebrandt, Nicole Wiebach TA13 – Champions Center III TA14 – Champions Center VI TA15 – Champions Center V Quantifying the Profit Impact of Marketing I Customers' Willingness-to-pay Research B2B: Relationships Chair: Xueming Luo Chair: Neil Biehn The Impact of Strategic Alliance on the Innovator’s Financial Value in Markets with Network Effects and Standard Competition Qi Wang, Ashwin Malshe, Jinhong Xie Exploring Consumer Heterogeneity with Respect to Seasonal Shifts of Demand Ali Umut Guler Chair: Alfred Zerres The Dynamic Effects of Service Recovery Strategies on Customer Satisfaction Xueming Luo, Fang Zheng The Information Content of Marketing Investments: The Case of Sales Force Resizing Announcements Anne T. Coughlan, Joseph Kissan, M. Babajide Wintoki Panel Discussion: Criticisms and Future Research Opportunities for Marketing’s Profit Impact Moderator: Dominique Hanssens, Panelists: David Reibstein, Gerard J. Tellis A Structural Model for a "Name Your Own Price" Mechanism with a Fixed Price Option Kerem Yener Toklu Framing Effects and Consumers' Reactions to Corporate Social Responsibility Carmelo J. Leon, Jorge Araña, Christine Eckert Pricing and Willingness-to-pay Estimation in B2B Markets Neil Biehn A Theory of Bargaining Costs and Price Terms in the Absence of RelationshipSpecific Investments Desmond (Ho-Fu) Lo, Giorgio Zanarone Assessment of Purchasing Maturity in Small Business Jeffery Adams, Ralph Kauffman Integrative Negotiation Training: Enduring Effects of Asymmetrical and Symmetrical Training Alfred Zerres, Klaus Backhaus, Alexander Freund, Joachim Hüffmeier 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 10.30-12.00 (TB) TB01 – Legends Ballroom I TB02 – Legends Ballroom II TB03 – Legends Ballroom III TB04 – Legends Ballroom V Marketing Science Institute II Empirical Modeling Social Networks and Profitability Bayesian Econometrics II: Methods & Application Chair: Don Lehmann Chair: Eric Schwartz Chair: Michael Haenlein Research on Branding: Issues and Outlook Jan-Benedict Steenkamp Modeling Online Visitation and Conversion Dynamics Chang Hee Park, Young-Hoon Park Co-chair: Barak Libai Research on Brands: Latest Findings and Future Opportunities Marc Fischer Speed of Product Updates in Online Games Paulo Albuquerque The Evolution of Research on Marketing Metrics Dominique Hanssens Estimating Preferences from Configured Choice Sanjog Misra Chair: Sudhir Voleti Test and Learn: A Reinforcement Learning Perspective Eric Schwartz How Customer Word of Mouth Affects the Benefits of New Product Exclusivity to Distributors Christophe Van den Bulte, Renana Peres Evolving Viral Marketing Strategies William Rand, Forrest Stonedahl Determinants of Social Influence on Adoption in Customer Ego Networks Hans Risselada, P.C. (Peter) Verhoef, Tammo Bijmolt Customer Acquisition in a Connected World: Revenue vs. Opinion Leaders Michael Haenlein, Barak Libai Simultaneous Scaling of Multiple Domains: Application to Country-of-origin Effects in Asia Luming Wang, Giana Eckhardt, Terry Elrod A Dynamic Spatial Hierarchical Model of Theater Level Box-office Performance Shyam Gopinath, Pradeep Chintagunta, Hedibert Lopes, Sriram Venkataraman Attribute-level Heterogeneity Peter Ebbes, John Liechty, Rajdeep Grewal, Matthew Tibbits A Nonparametric Model of Attribute Based Inter-product Competition Sudhir Voleti, Pulak Ghosh, Praveen Kopalle TB05 – Legends Ballroom VI TB06 – Legends Ballroom VII TB07 – Founders I TB08 – Founders II New Product II: Diffusion Competition II: More on Measuring the Impact in Retail Markets ASA Special Session on the Marketing-Statistics Interface – II Innovation II Chair: A. Yesim Orhun Chair: Anindya Ghose Modeling Seasonality in New Product Diffusion Yuri Peers, Philip Hans Franses, Dennis Fok Entry with Social Planning Stephan Seiler, Pasquale Schiraldi, Howard Smith Assessing the Validity of Market Structure Analysis Derived from Text Mining Data Oded Netzer, Ronen Feldman, Moshe Fresko, Jacob Goldenberg Empirical Test of the Bass Diffusion Model using Exogenous Shocks on Word of Mouth Effect Sungjoon Nam Sleeping with the “Frenemy": The Agglomeration-differentiation Tradeoff in Spatial Location Choice Sumon Datta, K. Sudhir Spatiotemporal Analysis of New Product Diffusion Li Zheng Does Reducing Spatial Differentiation Increase Product Differentiation? Effects of Zoning on Retail Entry and Format Variety K. Sudhir, Sumon Datta Chair: Chander Velu Chair: Li Zheng What Drives Me? A Novel Application of the Conjoint Adaptive Ranking Database System to Vehicle Consideration Set Formation using Population Statistics Ely Dahan How is the Mobile Internet Different? Search Costs and Local Activity Sangpil Han, Avi Goldfarb, Anindya Ghose Evaluating Financial Risk from Cross Border M&A Activities on Brand Identity Sustainability Sixing Chen, Xiaoqi Yang, Ronald W. Cotterill An Analysis of the Financial Performance of Radical, Complex and Financially Risky Innovations Lisa Schöler, Bernd Skiera, Gerard J. Tellis Promoting Growth and Innovation through Acquisition: A Choice Modeling Approach Yu Yu, Vithala Rao Product Portfolio Effects of Innovation: A Diversification Perspective on Innovation Value Creation Fredrika Spencer, Richard Staelin Entrepreneurs as Owner-managers, Ownership Concentration and Business Model Innovation Chander Velu, Arun Jacob 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 10.30-12.00 (TB) TB09 – Founders III TB10 – Founders IV TB11 – Champions Center I TB12 – Champions Center II Promotions II Decision Making Response to Advertising Brand Identity Chair: Francesca Sotgiu Chair: Robert Rooderkerk Chair: Ho Kim Chair: Stefan Worm On the Timing and Depth of a Manufacturer’s Sales Promotion Decisions with Forward-looking Consumers Yan Liu, Subramanian Balachander, Sumon Datta Some Empirical Evidence on Predicted versus Reported Behavior: The Role of Attitudes and Situational William Putsis, Preethika Sainam, Gal Zauberman Selling the Drama: Death-related Publicity and its Impact on Music Sales Leif Brandes, Stephan Nüesch, Egon Franck Brand Extensions Frequency and Brand Performance Helena Allman Empirical Investigation of Consumer Impulse Purchases from Television Home Shopping Channels Sang Hee Bae, Sang-Hoon Kim, Sungjoon Nam The Impact of Free-trial Promotions on Adoption of a High-tech Consumer Service Bram Foubert, Els Gijsbrechts, Charlotte Rolef Units Versus Numbers Ashwani Monga, Rajesh Bagchi Resource Abundance and Conservation in Consumption Meng Zhu, Ajay Kalra Optimizing the Assortment Layout: The Effect of Categorization Congruency on Purchase Incidence Robert Rooderkerk Is Beauty in the Eye of Beholders? Linking Facial Features to Source Credibility in Advertising Li Xiao, Min Ding Creativity in Advertising and Implications for Product Sales Performance Peter Saffert, Werner Reinartz Priming vs. Wearout: Early Prelaunch Advertising, Online Buzz and Newproduct Sales Ho Kim, Dominique Hanssen Promotion Effectiveness in Economic Turbulence: From Price Wars to Economic Downturns Francesca Sotgiu, Katrijn Gielens TB13 – Champions Center III TB14 – Champions Center VI TB15 – Champions Center V Quantifying the Profit Impact of Marketing II Dynamic Pricing Issues No Session Chair: Xueming Luo Chair: Jonathan Zhang An Econometric Model of Firms’ Participation Decisions Across CSR Activities Nitin Mehta, Vikas Mittal, Christopher Groening Online Content Pricing Anita Rao The Case Stock Market Rewards for Customer and Competitor Orientations: of Initial Public Offerings Alok R. Saboo, Rajdeep Grewal The Impact of Marketing Strategy on Corporate Bankruptcy Niket Jindal, Leigh McAlister Conspicuous Consumption and Dynamic Pricing Richard Schaefer, Raghunath Rao Estimating Dynamic Pricing Decisions in Markets with State Dependent Demand Koray Cosguner, Tat Y. Chan, P. B. Seetharaman Dynamic Targeted Pricing in B2B Settings Jonathan Zhang, Oded Netzer, Asim Ansari Material Values and Consumer Personality Effects on Brand Personality Perceptions Tiffany Ting-Yu Wang Cross-cultural Differences in Brand Engagement Antonieta Reyes, Felipe Korzenny What Makes a Strong B2B Brand? The Role of Tangible versus Intangible Brand Attributes Stefan Worm 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 1.30-3.00 (TC) TC01 – Legends Ballroom I TC02 – Legends Ballroom II TC03 – Legends Ballroom III TC04 – Legends Ballroom V Choice I: New Models of … Online Advertising - I Internet: Social Influence Econometric Methods I: General Chair: Peter Stuettgen Chair: Laura Kornish Chair: Jun Yang Chair: Sridhar Narayanan Assessing Two Alternative Methods for Modelling Heterogeneity in Stated Preference Data Paul Wang, Jordan Louviere, Kyuseop Kwak How Do Advertising Standards Affect Online Advertising? Avi Goldfarb Successful Social Networkers: Impact of Activities and Network Positions Lucas Bremer, Florian Stahl, Asim Ansari, Mark Heitmann A Cigarette, a Six Pack or Porn? The Complementarity of Vices Rachel Shacham, Peter Golder, Tulin Erdem The Evolution of Switch Customers in E-Commerce: Understanding When and How Customers Switch Fan Zhang, Tat Chan, Qin Zhang Handling Endogenous Regressors by Joint Estimation Using Copulas Sungho Park, Sachin Gupta A Direct Utility Model for Asymmetric Complements Sanghak Lee, Greg Allenby, Jaehwan Kim Utility-based Model of Asymmetric Competitive Structure using Store-level and Forced Switching Data Paul Messinger, Fang Wu A Satisficing Choice Model Peter Stuettgen, Peter Boatwright, Robert Monroe Internet Display Advertising and Consumer Purchase Behavior: Do Ad Platforms Matter? Paul Hoban, Randolph Bucklin The Effect of Banner Exposures on Memory for Established Brands Titah Yudhistira, Eelko Huizingh, Tammo Bijmolt Is it a Fad or Necessity? Measuring the Effectiveness of Social Media on E-tailers Jun Yang, Jungkun Park Website Ad Quantities: An Empirical Analysis of Traffic, Competition, and Business Model Laura Kornish, Jameson Watts Improving Predictive Validation Steve Shugan Regression Discontinuity with Unobserved Score Sridhar Narayanan, Kirthi Kalyanam TC05 – Legends Ballroom VI TC06 – Legends Ballroom VII TC07 – Founders I TC08 – Founders II New Product III: Adoption Competition III: General ASA Special Session on the Marketing-Statistics Interface – III Innovation III Chair: Mark Ratchford Chair: Vincent Mak Chair: Merle Campbell Chair: Michael Braun An Investigation of Scales for Consumer Innovativeness Masataka Yamada, Toshihiko Nagaoka What if Marketers Put Customers Ahead of Profits? Scott Shriver, V. "Seenu" Srinivasan A Multivariate Analysis of Pre-acquisition Drivers of Technology Adoption Mark Ratchford, Jeffrey Dotson Gaining from Imitative Entry: Dynamic Durable Pricing with Rational Consumer Expectations Lu Qiang, Wei-yu Kevin Chiang KFC and McDonald’s Entry in China: Competitors or Companions? Qiaowei Shen, Ping Xiao Dominance and Innovation in a Dynamic Macro Environment Vincent Mak, Jaideep Prabhu, Rajesh Chandy, Chander Velu Customer Waiting Time and Purchasing Behavior: An Empirical Study of Supermarket Queues Andrés Musalem, Yina Lu, Marcelo Olivares, Ariel Schilkrut Forecasting Customer Purchase Rates Incorporating Temporal Variation Luo Lu, Zainab Jamal Optimal Mailing in a Beta-geometric Beta-binomial (BG/BB) Model George Knox, Rutger van Oest Modeling Customer Lifetimes with Multiple Causes of Churn Michael Braun, David Schweidel The Chinese Knockoff Effect: How Do Consumers Perceive “Shanzhai” Cellphones? Shu-Chun Ho, Huang Shang-Jui When do Firms Benefit from Alliance Specialization in Either Innovation or Marketing? Jongkuk Lee, Young Bong Chang The Financial Determinants of Adopting Radically Innovative Information Technology: An Empirical Anarchy Merle Campbell, Tim Bohling, Maureen Schumacher, V Kumar 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 1.30-3.00 (TC) TC09 – Founders III TC10 – Founders IV TC11 – Champions Center I TC12 – Champions Center II Promotions III Consumer Behavior Advertising Strategy Brand Equity Chair: Ty Henderson Chair: Yi-Yun Shang Chair: Sabita Mahapatra Chair: U.N. Umesh How Effective Are Conditional Promotions Tobias Langer, Kusum Ailawadi, Karen Gedenk, Scott Neslin My Brain is Tired. Can I Make Inference Spontaneously? Xiaoning Guo, Inigo Arroniz The Impact of Advertising on Brand Trial in Experience Good Markets Raimund Bau Improving the Image of Countries, Cities and Tourist Destinations, Using Media and Branding Strategies Eduardo Oliveira Coupon Expiration and Redemption Joseph Pancras, Rajkumar Venkatesan Theories of Emotion in Consumer Behavior Khalil Rohani, Laila Rohani, Joe Barth Advertising during Recession: Role of Industry Characteristics, Strategy Type, and Market Orientation Peren Ozturan, Aysegul Ozsomer Gifts with a Gab: A Multivariate Poisson Analysis of the Effects of Gifts on Customer Acquisitions Sudipt Roy, Purushottam Papatla Carrot or Stick? - Asymmetric Evaluation on Counterfeit Products Under Different Self Construal Xi Chen Promoting a Brand Portfolio with a Social Cause: Findings from an In-market Natural Experiment Ty Henderson, Neeraj Arora Remedying Reverse Self-control Effect of Hyperopic Consumers on Vicious Avoidance Yi-Yun Shang, Kuen-Hung Tsai TC13 – Champions Center III TC14 – Champions Center VI TC15 – Champions Center V Quantifying the Profit Impact of Marketing III Consumer Responses to Pricing CRM I: Customer Lifetime Value Chair: Xueming Luo Chair: Anja Lambrecht Chair: Zainab Jamal The Market Valuation of Company Initiated Customer Engagement Sander F. M. Beckers, Jenny van Doorn, P.C. (Peter) Verhoef Starting Prices as Catalysts for Consumer Response to Customization Marco Bertini, Luc Wathieu Payments as a Virtual Lock-in: Customers’ Profitability over Time in the Presence of Payments Irit Nitzan, Barak Libai, Danit Ein-Gar Impact of Price Change on Profitability: Theory and Empirical Evidence Vinay Kanetkar The Value Relevance of Marketing Expenditures Min Chung Kim, Leigh McAlister Assortment Diversification in the Retail Industry: The Impact on Market-based and Accounting-based Performance Timo Sohl, Thomas Rudolph Free vs. Fee: Pricing of Online Content Services Kanishka Misra, Anja Lambrecht Private Label Response to National Brand Promotions: A Field Experiment Eric Anderson, Karsten Hansen, Duncan Simester Paying with Money or with Effort: Pricing When Customers Anticipate Hassle Anja Lambrecht, Catherine Tucker A Study on the Effectiveness of Emotional Versus Rational Appeals on Consumer of Eastern India Sabita Mahapatra Competitive Advantage through Internal Branding Constituents: Developing Critical Component Framework Anurag Kansal, Prem Dewani Conceptualization and Development of Scale for Power of Brand in a Brand-consumer Relationship Roopika Raj, Abraham Koshy Patent Data and Marketing Science U.N. Umesh, Monte Shaffer A New Model Proposal to Churn Management Omer Faruk Seymen, Abdulkadir Hiziroglu Hazards of Ignoring Involuntary Customer Churn Zainab Jamal, Randolph Bucklin 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 3.30-5.00 (TD) TD01 – Legends Ballroom I TD02 – Legends Ballroom II TD03 – Legends Ballroom III TD04 – Legends Ballroom V Choice II: Effects on ... Online Advertising – II Internet: Car Buying Econometric Methods II: General Chair: Linda Court Salisburyua Chair: Harald van Heerde Chair: Chen Lin Chair: Martin Spann Incorporating State Dependence in Aggregate Market Share Models Polykarpos Pavlidis, Dan Horsky, Minjae Song Connecting Social Media with Television Advertising and Online Search Yanwen Wang Modeling the Volume of Positive Online Word of Mouth for Automobiles Jie Feng, Purushottam Papatla Empirical Regularity in Academic Marketing Research Productivity Patterns Vijay Ganesh Hariharan, Debabrata Talukdar, Chanil Boo Investigating Advertisers’ View of Online and Print Media: Complements or Substitutes? Shrihari Sridhar, S. Sriram External Search in Secondary Markets and Impact of Internet Search on Seller Choice Sonika Singh Does Television Advertising Influence Online Search? Mingyu Joo, Kenneth Wilbur, Yi Zhu Media, Finance and Automotive: A Latent Trait Model of Consumption in Seemingly Disparate Categories Chen Lin, Douglas Bowman Complexity Effects on Choice Experiment-based Model Performance Benedict Dellaert, Bas Donkers, Arthur van Soest The Interplay of Reference Dependence and Choice Set Formation in Replacement Decisions Paul Messinger, Joffre Swait, Lianhua Li Does Choice Set Formation Drive the Diversification Effect? A Model and Experimental Evidence Linda Court Salisbury, Fred M. Feinberg Does Online Advertising Help or Hurt Offline Sales? A Nation-wide Field Experiment Harald van Heerde, Isaac Dinner, Scott Neslin Investigating the Performance of a Dynamic Budget Allocation Heuristic: A Simulation based Analysis Nils Wagner Social Network Based Judgmental Forecasting Martin Spann, Christian Pescher, Gary Lilien, Gerrit Van Bruggen TD05 – Legends Ballroom VI TD06 – Legends Ballroom VII TD07 – Founders I TD08 – Founders II New Product IV: Strategy Competition IV: Quality Services Innovation IV Chair: Dinah Vernik Chair: S. Chan Choi Chair: Kimmy Wa Chan Chair: Anna S. Cui Strategic Product Line Design with Product Concept Demonstration Taewan Kim, Eunkyu Lee The Impact of Competition and the Cost of Overstating Quality on the Optimal Quality, Quality Claims Praveen Kopalle, Don Lehmann Service Worker Role in Encouraging Customer Equity: Dyadic Analysis Yu-Li Lin, Hsiu-Wen Liu Patent Rank and Firm Performance Monte Shaffer, U.N. Umesh The Strategic Role of Exchange Programs Bo Zhou, Debu Purohit, Preyas Desai Merging in Spatial Competition Tieshan Li Price and Inventory Competition between New and Old Technologies Dinah Vernik, Preyas Desai, Fernando Bernstein A Structural Analysis on Service Quality and Pricing Tradeoff in Airlines Chen Zhou, Rajdeep Grewal Hedonic Quality Differentiation and Channel Choice S. Chan Choi Perceptions of Service Failures: A Test and Extension of Affective Forecasting Theory Muyu Wei, Geng Cui Can I Do It? Can You Do It? Roles of Self-efficacy and Other-efficacy of Customers and Employees Kimmy Wa Chan, Bennett C. K. Yim, Simon Lam What You Don’t Know Can’t Hurt You: Effects of Knowledge Limitations on Technological Innovativeness Stav Rosenzweig, David Mazursky Alliance Portfolio Resource Diversity and Firm Innovation Anna S. Cui, Gina O’Connor 2011 INFORMS Marketing Science Conference Thursday, June 9th, 2011 3.30-5.00 (TD) TD09 – Founders III TD10 – Founders IV TD11 – Champions Center I TD12 – Champions Center II Retailing I: General Consumer Behavior: Decision Making Advertising Content Bidding Chair: Umut Konus Chair: Berna Basar Chair: Larry Garber Chair: Ming Cheng The Effect of Brand Assortment Shares on National Brand Performance Across U.S. Supermarkets Minha Hwang, Raphael Thomadsen Consumer Gratitude and Customer Loyalty: Moderating Effect of Stage of Relationship, Gender and Age Prem Dewani, Anurag Kansal The Role of Brand Construal and Affect Valence in Comparative Advertising Ying Ho, Candy K. Y. Ho Coordinating Traditional and Search Advertising Alex Kim, Subramanian Balachander Validating Suppliers of Retailer’s Resources in Augmenting Product Safety Performance Wei-Che Hsu, Ming-Chih Tsai Impact of Visual and Tactile Input on Variety Seeking Behavior Subhash Jha, S. (Sivkumaran) Bhardawaj The Influence of Product-placement Clutter and Other Context Variables on Brand Attitude and Memory Pola Gupta Modeling Price Dynamics in Simultaneous Auctions: A Bayesian Factor Analytic Approach Norris Bruce Assortment Selection in Retailing: Strict Return Policies Call for Eccentric Products Aydin Alptekinoglu, Elif Akcali, Alex Grasas Turkish Gift Buying Attitudes in Today's Marketing Environment Berna Basar, A. Banu Elmadag Bas The Effects of Shape Complexity and Presentation Larry Garber, Eva Hyatt, Unal Boya An Investigation of Market Learning and its Implications for an IP Auction House Joseph Derby, Mayukh Dass An Empirical Investigation of Sponsored Search Engine Advertising Pricing Ming Cheng, Lei Wang, S. Chan Choi Tracking Holistic Customer Experience in Realtime Umut Konus, Emma MacDonald, Hugh Wilson TD13 – Champions Center III TD14 – Champions Center VI TD15 – Champions Center V Quantifying the Profit Impact of Marketing IV Pricing and Competition CRM II: Customer Loyalty Chair: Maxim Sinitsyn Chair: Harmeen Soch Inferring Competitor Pricing with Incomplete Information Marcel Goic, Alan Montgomery Allocating Optimal Multi-period Budget to Loyalty and Sales Promotion Programs Hsiu-Yuan Tsao, Li-Wei Chen, Hsiu-Feng Yan Chair: Xueming Luo The More Efficient the Better: Advertising Efficiency and its Impact on Firm’s Financial Performance Jin-Woo Kim, Traci Freling Marketing Spending, Analyst Coverage, and Firm Performance in the IPO Market Monica Fine, Kimberly Gleason Can Stock Markets Really Predict the Future? Case of Product Innovations M. Berk Talay, M. Billur Akdeniz Total Recall: Investor and Consumer Response Following Toyota's Automotive Recall Robert Evans Jr. Resale Price Maintenance when Retailers are Heterogeneous Charles Ingene, Mark Parry, Zibin Xu MAP and RPM: Determinants of Violations Ayelet Israeli, Eric Anderson, Anne T. Coughlan Coordination of Price Promotions in Complementary Categories Maxim Sinitsyn The Impact of Loyalty Program on Loyalty Transfer within the Partnership Network So Young Lee, Hyang Mi Kim, Jae Wook Kim Influence of Perceived Relationship Investment and Cross-buying on Share-of-Wallet Harmeen Soch, Navneet Multani 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 8.30-10.00 (FA) FA01 – Legends Ballroom I FA02 – Legends Ballroom II FA03 – Legends Ballroom III FA04 – Legends Ballroom V Choice III: More Effects on … UGC-I (The Evolution and Impact of Online Opinions) Internet: Customer Response Dynamic Models I Chair: Zheyin (Jane) Gu Chair: Ashish Sood Chair: Aditya Billore Chair: Rene Algesheimer Capturing the Unobserved Comparison Effects in Consumer Choices: A Hierarchical ME Model Ping Wang, Jaihak Chung, Meng Su, Luping Sun Online Product Opinions: Incidence, Evaluation and Evolution Wendy W. Moe, David Schweidel Sales Tax and Online Consumer Behavior Nicholas Lurie, Sriram Venkataraman, Peng Huang Applying Conditional Three-level Nonlinear Growth Curve Modeling to Innovation Diffusion Margot Loewenberg, Markus Meierer, Rene Algesheimer Learning Dynamics in Product Relaunch Sue Ryung Chang, Tulin Erdem Consumer Attribute-based Learning and Retailer Category Management Strategies Zheyin (Jane) Gu, Sha Yang A Firm's Optimal Response to Negative Rumors Dina Mayzlin, Yaniv Dover, Jiwoog Shin Empirically Investigating the Relationship between What Brands Do and What Consumers Say (Social Media), Sense (Mindset), and Do (Purchase) Douglas Bowman, Manidh Tripatthi Power of Customer Voice: Shap Analysis of Online Product Reviews to Predict Diffusion in Sequential Channels Ashish Sood, Mayukh Dass, Wolfgang Jank, Yue Tian Trajectory-based Consumer Segmentation and Product Recommender System in the Online Market Youngsoo Kim, Ramayya Krishnan The Impact of Personalization and Interactivity on Choice Goal Attainment and Decision Satisfaction Sally McKechnie, Prithwiraj Nath Consumer Demographics & Changing Perception to Online Advertising: Applying Learning Curve Mechanism Aditya Billore, Anurag Kansal An Asymmetric Threshold Error Correction Model of Pass-through in the U.S. Supermarket Industry Miguel Gomez, Christopher Lanoue, Timothy Richards A Bayesian DYMIMIC Model for Forecasting Movie Viewers Dong Soo Kim, Jaehwan Kim, Duk Bin Jun Measuring Individual’s Growth in Achievement Over Time under Changing Group Affiliations Rene Algesheimer, Markus Meierer, Egon Franck, Leif Brandes FA05 – Legends Ballroom VI FA06 – Legends Ballroom VII FA07 – Founders I FA08 – Founders II New Product V: Design & Development Channels I: General Panel Session: Cases? Projects? Simulations? Problem Sets? What's the Best Way to Teach Marketing Science? The Long Run Consequences of Short Run Decisions I Chair: Gary Lilien Co-chair: Ahmed Khwaj Co-chair: Arvind Rangaswamy Taste and Health: Balancing and Highlighting in Choices Across Complementary Categories Hai Che, Botao Yang, K. Sudhir Chair: Sudheer Gupta Chair: Wooseong Kang Marketing Instrument Innovations and Their Impact on New Product Performance Wenzel Drechsler, Martin Natter Investigating the Relationship between R&D and Marketing in the New Product Development Process Suj Chandrasekhar, Srinath Gopalakrishna Embedding Product Development Accelerations in Environmental Uncertainty Tao Wu Consumer Opinion of Product Design Dimensions Wooseong Kang, Janell Townsend, Mitzi Montoya Information Sharing and New Product Development in a Non-integrated Distribution Channel Shan-Yu Chou Distributor Support in New Product Launch Wei Guan, Jakob Rehme Long-term Asymmetric Buyer-seller Relationship: An Empirical Study Yuying Shi, Qiong Wang, Bart Weitz Inventories, Incentives, and Channel Structure Sudheer Gupta Cases? Projects? Simulations? Problem Sets? What's the Best Way to Teach Marketing Science? Moderators: Gary Lilien, Arvind Rangaswamy, Panelists: Arnaud De Bruyn, Dominique Hanssens, Ujwal Kayande, Charlotte Mason Chair: K. Sudhir Information Acquisition and Ex-ante Moral Hazard Jian Ni, Nitin Mehta Changing the Tone: The Dynamics of Political Advertising over the Election Cycle Ron Shachar, Paul Ellickson, Mitch Lovett A Dynamic Model of Thirst and Beverage Consumption Ahmed Khwaja, K. Sudhi, Guofang Huang 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 8.30-10.00 (FA) FA09 – Founders III FA10 – Founders IV FA11 – Champions Center I FA12 – Champions Center II Retailing II: General Measurement Issues Using Endorsers in Advertising Aesthetics Chair: Manish Gangwar Chair: Julie Lee Chair: Debasis Pradhan Chair: Elea McDonnell Feit The Impact of Retailers’ Corporate Social Responsibility on Price Fairness Perceptions and Loyalty Kusum Ailawadi, Jackie Luan, Scott Neslin, Gail Taylor Customer Innovation: A Combined Lead User and Conjoint Analysis Approach Alexander Sänn; Daniel Baier Alienating the Mainstream: Does the Inclusion of Gay and Lesbian Imagery Diminish Brand Perception? Anthony Perez, Helene Caudill Inferring Color Preferences: A Utility Model Approach Seth Orsborn, Peter Boatwright, Jonathan Cagan Consumer Perceptions of Corporate Gayfriendly Activities: The Role of Gender and Gay Identity Gillian Oakenfull Product Aesthetics is Must or Plus? Trade-offs Between Product Aesthetic and Functional Attributes Jesheng Huang, Chia Ming Hu Attitude Towards Celebrity Endorsement and Brand Loyalty: Mediating Effect of Celebrity Credibility Debasis Pradhan, Duraipandian Israel Shaping Product Perceptions Tanuka Ghoshal, Peter Boatwright To Kill Two Birds with One Long Queue Wenqing Zhang, Chun (Martin) Qiu Shopper Loyalty to Whom? Chain and Outlet Loyalty in a Dynamic Retail Environment Arjen van Lin, Els Gijsbrechts Accountability of Biological-response Measures for Advertising Effects Akihiro Inoue Using Augmented Best-worst Scaling to Test Schwartz’ Theory of Values Julie Lee, Jordan Louviere, Geoff Soutar Examining Store Attractiveness as a Category-specific Trait Manish Gangwar, Qin Zhang, P. B. Seetharaman FA13 – Champions Center III FA14 – Champions Center VI FA15 – Champions Center V Marketing Finance Interface I Pricing and Consumer Behavior CRM IV: Customer Loyalty Chair: Michal Herzenstein Chair: Marcus Kunter Chair: Janghyuk Lee Going Public: How Stock Market Participation Changes Firm Product Innovation Behavior Christine Moorman, Simone Wies Produce Line Obfuscation Lin Liu, Anthony Dukes Understanding Whether and How Marketing Efforts Drive Loyalty in the Car Industry of Emerging Markets Guillermo Armelini, Hernán Román Media Expenditure Effectiveness and Firm Performance Lopo Rego, Lisa Schöler, Bernd Skiera The Impact of Capital Structure on Customer Satisfaction Reo Song, Gautham Vadakkepatt The Use of Advertising for Capital Market Benefits Michal Herzenstein, Tzachi Zach, Dan Horsky A Simple Metric that Really Matters: Including the Share of Customer Business in Financial Reports Christian Schulze, Manuel Bermes, Bernd Skiera Choosing the Right Plan? Asymmetric Biases in 3-part Tariff Plan Choices Vardit Landsman, Itai Ater A Model of the Consumer Pricing Decision Process under Pay-what-you-want Marcus Kunter Is Rewarding VIPs Profitable? Steven Sangwoo Shin, Jia Li The Effects of Effort Level on Reward Redemption Behavior Jiyoon Kim, Janghyuk Lee, Sang Yong Kim Modeling the Impact of Visual Design in Consumer Choice Model Elea McDonnell Feit, Jeffrey Dotson, Mark Beltramo, Randall Smith 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 10.30-12.00 (FB) FB01 – Legends Ballroom I FB02 – Legends Ballroom II FB03 – Legends Ballroom III FB04 – Legends Ballroom V Choice IV: Market Structure & Substitution UGC-II (Quest for Comprehension and Integration) Internet Relationship Dynamic Models II Chair: Kamer Toker-Yildiz Chair: Manish Tripathi Chair: Ingrid Poncin Chair: Roopa Choodamani Measuring How Different Marketing Instruments Affect Competition: The Role of Choice Model Specifications Qiang Liu, Thomas Steenburgh, Sachin Gupta Listening in on Online Conversations: Measuring Consumer Sentiment with Social Media David Schweidel, Wendy W. Moe How Websites Can Create Trust: The Mechanisms that Build Initial Trust in ECommerce Environments Paul Driessen, Marcel van Birgelen, Eric Rongen Advertising Strategies by Multinational Firms Wiebke Schlabohm, Barbara Deleersnyder Optimal Dynamic Pricing Strategies: Consumer Cross-category Incidence/Purchase Quantity Decisions Sri Devi Duvvuri, Praveen Kopalle Market Delineation Strategies in Consumer Goods Market Sebastian Gabel, Raimund Bau Social Tag Maps: A New Approach For Understanding Brand Association Networks Hyoryung Nam The Quest for Content: The Role of User Generated Links in Online Content Shachar Reichman, Jacob Goldenberg, Gal Oestreicher The Quality of Electronic Customer-tocustomer Interaction: Classification and Consequences Moritz Mink, Dominik Georgi Video Ads Virality Thales Teixeira Avatar Identification on 3d Commercial Website: Gender Issues Ingrid Poncin, Marion Garnier Modeling Dynamics of Consumer Preference and Promotion Effect in Brand Choices Eiji Motohashi, Tomoyuki Higuchi Be Careful When Using the Mover-stayer Conceptual Framework in Brand Choice Mode Kanghyun Yoon Morphing Marketing Response Optimization – Advocating a Next Practice Roopa Choodamani, Pradeep Kumar The Influence of Willingness-to-pay on Consumer’s Cross Category Purchase Behavior Kamer Toker-Yildiz, Sri Devi Duvvuri, Minakshi Trivedi A Framework for Unifying Differentiated User-generated Content: What I Say, Where I Go, and What I Think Manish Tripathi, Ashish Sood FB05 – Legends Ballroom VI FB06 – Legends Ballroom VII FB07 – Founders I FB08 – Founders II Game Theory I: Decisions Under Limited Information Channels II: Relationship Management Panel Session: Collaborative Research: Reasons Why, Difficulties and Potential Models The Long Run Consequences of Short Run Decisions II Chair: Jeffrey D. Shulman Chair: Sara Valentini Chair: K. Sudhir Chair: Glen Urban How Hidden Add-on Pricing Can Reduce Profit Jeffrey D. Shulman, Xianjun Geng Salesforce Compensation under Inventory Considerations Kinshuk Jerath, Tinglong Dai Memories and Rules Juanjuan Zhang, Jeanine Miklós-Thai The Model of Buzz Jiwoong Shin, Arthur Campbell, Dina Mayzlin Investigating Impact of Multiple Communication & Marketing Mix Elements in Multichannel Environment Ashish Kumar, Ram Bezawada, Minakshi Trivedi Return on Channel Investments for Customer Acquisition – A Cross-channel Analysis Maik Eisenbeiss, Monika Käuferle, Peter Saffert, Werner Reinartz Does Multichannel Usage Produce More Profitable Customers Sara Valentini, Elisa Montaguti, Scott Neslin Insights Into the Role of the Internet in a Multichannel Customer Management Strategy Tanya Mark, Katherine N. Lemon, Jan Bulla, Antonello Maruott, Mark Vandenbosch Co-chair: Ahmed Khwaja Collaborative Research: Reasons Why, Difficulties and Potential Models (Data Base sharing and Prospective Meta Analysis) Moderator: Glen Urban, Panelists: Eric Bradlow, Gary Lilien, Don Lehmann, Catherine Tucker, Stefan Stremersch, Jan-Benedict Steenkamp, Jerry Wind, Gui Liberali Dynamic Competition between New and Used Durable Goods Without Physical Depreciation Masakazu Ishihara, Andrew Ching A Dynamic Model of Competition with Bundling Vineet Kumar, Timothy Derdenger A Dynamic General Equilibrium Model of User Generated Content Carl Mela, Dae-Yong Ahn A Dynamic Structural Analysis of Enterprise Knowledge Sharing Baohong Sun, Yingda Lu, Param Vir Singh 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 10.30-12.00 (FB) FB09 – Founders III FB10 – Founders IV FB11 – Champions Center I FB12 – Champions Center II Retailing III: Competition Bayesian Applications Salesforce I Internet: Unique Topics Chair: Bruce McWilliams Chair: Ralf van der Lans Chair: James Hess Chair: Agustí Casas-Romeo If You Build It, Will They Come?: Anchor Store Quality and Competition in Shopping Malls Ravi Shanmugam Variety Seeking in Movie Choice: The Role of Ratings Joon Ro, Romana Khan DEA with Econometrically Estimated Iindividual Coefficients: A Pharmaceutical Sales Force Application Soenke Albers, Andre Bielecki Quantifying Transaction Costs in Online / Offline Grocery Channel Choice Junhong Chu, Pradeep Chintagunta, Javier Cebollada Assessing Salesforce Performance: An Empirical Approach Wei Zhang, Ajay Kalra The Effect of Banner Exposures on Memory for Established Brands Titah Yudhistira, Eelko Huizingh, Tammo Bijmolt Dynamic Competitive Intensity in Retail Markets: Drivers and Implications on Retailer Performance Geunhye Yang, Katrijn Gielens, Jan-Benedict Steenkamp Money-back Guarantees: The Great Brand Equalizer Bruce McWilliams Inferring Competition in Search Engine Advertising with Limited Information Sha Yang The Multiple Effects of Social Comparisons on Consumer Expenditure Rafael Becerril-Arreola Partner Selection in Brand Alliances Ralf van der Lans, Bram Van den Bergh, Evelien Dieleman Sales Contests and Quotas with Imbalanced Territories - A Model and Experiments James Hess, Niladri Syam, Ying Yang A Study of Consumer Interest in Innovative Products Across Developed and Emerging Markets Gauri Kulkarni Application of Case Study on the Quality of Public Transport in European Cities with a Tool for Digital Ethnography Agustí Casas-Romeo, Rubén Huertas-García, Juan Carlos Gázquez-Abad FB13 – Champions Center III FB14 – Champions Center VI FB15 – Champions Center V Marketing Finance Interface II Pricing Research CRM III: Customer Loyalty Chair: Michael Sorell Chair: R. Mohan Pisharodi Chair: Radu Dimitriu Preventing Raised Voices from Echoing: Advertising as Response to Shareholder Activism Simone Wies, Arvid O. I. Hoffmann, Jaakko Aspara, Joost M. E. Pennings Service Refund as a Price Discrimination Mechanism Zelin Zhang, Weishi Lim Do Reward Programs Affect Consumer Behavior? Ricardo Montoya, Oded Netzer, Ran Kivetz The Impact of Consulting on Buying Behavior – The Case of Attention Behavior Nicolas Bourbonus, Dominik Georgi, Olaf Stotz Wedded Bliss or Tainted Love?: Stock Market Reactions to the Introduction of Co-branded Products Zixia(Summer) Cao, Alina Sorescu The Value of a Global Brand: Is Perception Reality? Michael Sorell, Arturo Bris, Willem Smit Determinants of Gain and Loss Parameters in Store-level Data: A Cross-category Analysis Sebastian Oetzel, Daniel Klapper The Timing and Speed of New Product Price Landings Carlos Hernandez Mireles, Dennis Fok, Philip Hans Franses Price Pressure and Supplier Relations: Industry-Specific Findings R. Mohan Pisharodi, Ravi Parameswaran, John Henke, Jr. Shortcuts to Glory? Exploring When and Why Attribute Performance Can Directly Drive Loyalty Johannes Boegershausen, Christophe Haon, Daniel Ray How do E-Commerce Interfaces Affect Customer Satisfaction and Loyalty? Hsiu-Wen Liu, Yu-Li Lin Investigating Multipurpose Customers Radu Dimitriu, Fred Selnes 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 1.30-3.00 (FC) FC01 – Legends Ballroom I FC02 – Legends Ballroom II FC03 – Legends Ballroom III FC04 – Legends Ballroom V Choice V: Empirical Results UGC-III (Content and Impact) Analytic Models of Online Behavior Structural Models I Chair: Christian Schlereth Chair: Raji Srinivasan Chair: J. Miguel Villas-Boas Chair: Wenbo Wang Measuring Scale Attraction Effects in Charitable Donations: An Application to Optimal “Laddering” Kee Yeun Lee, Fred M. Feinberg Bimodal Distribution of Emotional Content in Customer Reviews: Emotional Biases in Online Customer Reviews Wonjoon Kim The Interplay between Sponsored Search and Display Advertising Kannan Srinivasan, Kinshuk Jerath, Amin Sayedi A Dynamic Model of Consumers’ Optimal Default on Financial Products: A Case of Subprime Mortgages Minjung Park, Patrick Bajari, Sean Chu, Denis Nekipelov Data or Structure? Using a Field Experiment to Assess the Determinants of Counterfactual Demand Predictive Performance Manuel Hermosilla, Yi Qian, Eric Anderson Ad Revenue and Content Commercialization: Evidence from Blogs Monic Sun, Feng Zhu Optimal “Last-minute” Selling by a Monopolist Facing Forward-looking and Risk Averse Consumers Ori Marom, Abraham Seidmann Modeling Consumer Learning of Attribute-specific Preferences Jihong Min, Subramanian Balachander Does Advertising Affect Chatter? Assessing the Dynamics of Advertising on Online Word-of-mouth Seshadri Tirunillai , Gerard J. Tellis Sampling Paid Content Florian Stahl, Don Lehmann, Oded Koenigsberg, Daniel Halbheer Social Influence in the Evolution of Online Ratings of Service Firms Raji Srinivasan Optimal Search for Product Information J. Miguel Villas-Boas, Monic Sun, Fernando Branco FC05 – Legends Ballroom VI FC06 – Legends Ballroom VII FC07 – Founders I FC08 – Founders II Game Theory II: Market Entry Channels III: Competition New Directions in Word of Mouth Dynamic Models in Marketing Chair: Matthew Selove Chair: Jaime Romero Chair: Jonah Berger Chair: Paul Ellickson Cross-market Experience and Market Entry Dai Yao, Yakov Bart When and How Do Coordinating Contracts Improve Channel Efficiency? Ernan Haruvy Co-chair: Andrew Stephen The Benefit of Increased Competition David Soberman, Amit Pazgal Channel Structure and Performance under Co-marketing Alliance Xiao Zuhui, Liu Lming, Zhang Xubing How the Frequency and Pattern of Social Influence Over Time Shape Product Adoption Raghu Iyengar, Jeffrey Cai, Jonah Berger Learning About Entertainment Products: A Dynamic Consumer Decision Model with Learning About Changing Match-Values Mitch Lovett, William Boulding, Richard Staelin The Complementary Roles of Traditional and Social Media Publicity in Driving Marketing Performance Andrew Stephen, Jeff Galak Determining Consumers' Discount Rates with Field Studies Song Yao, Jeongwen Chiang, Yuxin Chen, Carl Mela Promotional Reviews Yaniv Dover, Dina Mayzlin Does AMD Spur Intel to Innovate More? Ronald Goettler, Brett Gordon Multichannel Word of Mouth: The Effect of Brand Characteristics Renana Peres, Ron Shachar Dynamics of Pricing Strategy and Repositioning Costs Paul Ellickson, Sanjog Misra, Harikesh Nair Estimation of Willingness to Pay Intervals by Discrete Choice Experiments Christian Schlereth, Christine Eckert, Bernd Skiera A Dynamic Model of Competitive Entry Response Matthew Selove Should Be Close to or Away from Your Competitors? Store Location Choice by Gravity Model Wei-Jhih Yang, Jesheng Huang, Lichung Jen Price Competition in the Spanish Nondurable Retail Industry Jaime Romero, Daniel Klapper, Martin Natter Studying the Switching Behavior of Electricians: Assessing the Impact of a Loyalty Program Madhu Viswanathan, Ranjan Banerjee, Om Narasimhan Green Lifestyle Adoption: Shopping Without Plastic Bags Wenbo Wang, Yuxin Chen 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 1.30-3.00 (FC) FC09 – Founders III FC10 – Founders IV FC11 – Champions Center I FC12 – Champions Center II Retailing IV: Competition Segmentation Salesforce II Word of Mouth and Marketing Strategy Chair: Aharon Hibshoosh Chair: David Norton Chair: Steven Lu Chair: Yogesh Joshi Product Variety Decision: When Specialty Stores Meet with Big-box Retailers Jiong Sun, Tao Chen Loyalty to Service Providers in the Very Short Run and in the Very Long Run: The Impact of Ageing and Cohort Gilles Laurent , Raphaëlle LambertPandraud Integrated Versus Specialized Salesforce: When Hunting-farming is Harming Ying Yang, Niladri Syam, James Hess Impact of Company Announcements on the Evolution of Online Word-of-mouth Omer Topaloglu, Piyush Kumar, Dennis Arnett, Mayukh Dass Individual-based or Group-based Tournaments? An Experimental Study Hua Chen, Noah Lim, Michael Ahearne Antecedents and Consequences of Prerelease C2C Buzz Evolution: A Functional Analysis Guiyang Xiong, Sundar Bharadwaj Variety and Cost Pass-through Among Supermarket Retailers Timothy Richards, Stephen Hamilton, William Allender Pricing, Package Size, Advertising and Trade Areas in Spatial Competition of Retail Warehouse Clubs Aharon Hibshoosh The Dynamics of Brand Preferences Along Consumers’ Life Paths Tingting Fan, Peter Golder Limited Editions: When Snobs Behave Like Conformists and Conformists Behave Like Snobs Sergio Moccia, Oliver Heil Investigating Salespeople Turnover in a Dynamic Structural Framework Steven Lu, Ranjit Voola One Size Fits Others: The Role of Label Ambiguity in Targeting Diverse Consumer Segments David Norton, Randy Rose, Caglar Irmak FC13 –Champions Center III FC14 – Champions Center VI FC15 – Champions Center V Financial Decision Making Price Discounting CRM V: Customer Satisfaction Chair: Carlos Lourenco Chair: Kamel Jedidi Chair: Nima Jalali What You Know, What You Do or Who You Know? A Model of Individual Investor Returns Thomas Gruca, Sheila Goins An Empirical Investigation of the Longterm Effects of Price Discrimination in Business Markets Hernan Bruno, Shantanu Dutta The Utility of DLF Binary Ratings in Customer Satisfaction Measurement and Modeling Keith Chrzan, Jeremy Loscheider Investing for Retirement: The Moderating Effect of Fund Assortment Size on the 1/N Heuristic Jeff Inman, Susan Broniarczyk, Mimi Morrin Volume Based Discounts and Sequential Choice: Structural Estimation and Determination of Optimal Pricing James Reeder, Sanjog Misra One-stop Shopping: A Double Edged Sword? Xiaojing Dong, Pradeep Chintagunta Individual Investors Risk Behavior in Times of Crisis: A Cross-cultural Study Nikos Kalogeras, Joost M.E. Pennings, Joost Kuikman, Koert van Ittersum Improving Investment Advice Using Preferred Outcome Distributions Carlos Lourenco, Bas Donkers, Benedict Dellaert, Dan Goldstein A Conjoint Model of Quantity Discounts Kamel Jedidin, Raghu Iyengar Dynamics of Satisfaction: A Regime Switching Ordinal Model for Affective and Cognitive Factors Nima Jalali, Purushottam Papatla Underpromising and Overdelivering Competitive Implications of Word of Mouth Yogesh Joshi, Andrés Musalem 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 3.30-5.00 (FD) FD01 – Legends Ballroom I FD02 – Legends Ballroom II FD03 – Legends Ballroom III FD04 – Legends Ballroom V Choice VI: Applications UGC-IV (Content and Impact) Online Search Structural Models II Chair: Paola Mallucci Chair: Janghyuk Lee Chair: Alan Montgomery Chair: Yi Zhao Modeling Consumer Demand for Type, Form, and Package Size in the Seafood and Fish Industry Benaissa Chidmi Understanding the Dynamic Process of Online WOM: A HB Choice Model for Online Response Behavior Luping Sun, Ping Wang, Meng Su Consumer Search and Propensity to Buy Ofer Mintz The Impact of the Marketing Mix on Durable Product Replacement Decisions Dinakar Jayarajan, S. Siddarth, Jorge Silva-Risso Determinants of Complement Exclusivity in Platform Markets: A Study of the U.S. Videogame Market Srabana Dasgupta, Souvik Datta, Nilesh Saraf User-generated Content in News Media T. Pinar Yildirim, Esther Gal-Or, Tansev Geylani Manufacturers’ e-B2B Platform Choices – Relational Risk Threshold Chen-Han Yang, Ming-Chih Tsai, Chieh-Hua Wen Contractual Choices and their Consequences in a Time Inconsistent World Paola Mallucci, George John, Om Narasimhan Online Reviews and Consumers’ Willingness-to-pay: The Role of Uncertainty Yinglu Wu, Jianan Wu Failed Diffusion on Weak Tie Bridges Janghyuk Lee, Seok-Chul Baek, Jonghoon Bae, Sukwon Kang, Hyung Noh Return on Quality Improvements in Search Engine Marketing Nadia Abou Nabout, Bernd Skiera Which Link to Click—Sponsored or Organic? An Empirical Investigation on Consumer’s ‘Clickability” Amalesh Sharma, Sourav Borah Predicting Purchase Conversion Rates for Online Search Advertisements Using Text Mining Alan Montgomery, Kinshuk Jerath, Qihang Lin Economic Value of Celebrity Endorsement:Tiger Woods’ Impact on Sales of Nike Golf Balls Kevin Chung, Timothy Derdenger, Kannan Srinivasan Determination of Brand Assortment: An Empirical Entry Game with Post-choice Outcome Li Wang, Tat Y. Chan, Alvin Murphy An Empirical Model of Dynamic Re-entry, Advertising and Pricing Strategies in the Wake of Product Yi Zhao, Ying Zhao, Yuxin Chen FD05 – Legends Ballroom VI FD06 – Legends Ballroom VII FD07 – Founders I FD08 – Founders II Game Theory III: General Channels V: Strategy Meet the Editors Marketing Science / Management Science Chair: Niladri Syam Chair: Volker Trauzettel Managerial Myopia and Real Activity Mis-Management: Consequences for Marketing and Firm Performance A Model of the "It" Products in Fashion Kangkang Wang, Dmitri Kuksov Name-your-own-price as a Competitive Distribution Channel in the Presence of Posted Prices Xiao Huang, Greys Sosic Chair: Richard Batsell Would “False” Promotions be Profitable? Evidence from Experimental Data Yiting Deng, William Boulding, Richard Staelin Facts and Slant in News Production Yi Zhu, Anthony Dukes, Kenneth Wilbur Production Networks in Co-creation Niladri Syam, Amit Pazgal The Optimal Online Common Agency Strategy in the Presence of In-store Display Advertising Hao-An Hung, I-Huei Wu Slotting Allowance and Marketing Channel Strategy: An Empirical Analysis Using Quantile Regression Joo Hwan Seo, Ravi Achrol Price-matching and Retailing Strategies Volker Trauzettel Chair: Natalie Mizik Meet the Editors Co-chair: Anindita Chakravarty Performance Benchmarks as Drivers of Marketing: The Role of Analyst Forecasts Anindita Chakravarty, Rajdeep Grewal Dynamics of Marketing Effort Valuation: High-Frequency Stock Market Data Analysis Isaac Dinner, Natalie Mizik, Don Lehmann Changing the Rules of the Game: The Impact on Firm Value of Adopting an Aggressive Marketing Strategy Following Equity Offerings Didem Kurt, John Hulland Customer Satisfaction and the CEO’s Long-term Equity Incentives Don O’Sullivan, Vincent O’Connell Managing for the Moment: Role of Real Activity Manipulation Versus Accruals in SEO Over-valuation Natalie Mizik, Sugata Roychowdhury, S.P Kothari 2011 INFORMS Marketing Science Conference Friday, June 10th, 2011 3.30-5.00 (FD) FD09 – Founders III FD10 – Founders IV FD11 – Champions Center I FD12 – Champions Center II Retailing V: Location Decisions Survey Research Sports and Fashion Models of Word of Mouth Processes Chair: Jungki Kim Chair: Songting Dong Chair: Hema Yoganarasimhan Chair: Ingmar Nolte The Effect of in-Store Travel Distance on Unplanned Purchase with Applications to Shopper Marketing Sam Hui, Yanliu Huang, Jeff Inman, Jacob Suher The Impact of Different Scaling Techniques on Dropout Rates in Online Surveys Petra Wilczynski, Marko Sarstedt An Empirical Investigation of Sports Sponsorship Yupin Yang, Avi Goldfarb Modeling Promotional Word-of-mouth Backhun Lee, Minhi Hahn Accounting for Unobserved Heterogeneity in Models with Strategic Interactions Zheng Li, Maria Ana Vitorino A Machine Learning Approach to Analyzing Multi-attribute Data: The OrdEval Algorithm Sandra Streukens, Koen Vanhoof, Marko Robnik-Sikonja The Consumption of Live Sporting Events: Satisfaction of Very Important Fans Dennis Ahrholdt, Claudia Höck, Christian Ringle Testing Firms’ Conditional Differentiation Behaviour: Quantitative Evidence in Fashion Advertising Kitty Wang Demand Growth Patterns of Individual Consumers in a Geographically Expanded Retail Market Jungki Kim, Duk Bin Jun, Myoung Hwan Park Voice Analysis for Measuring Consumer Preferences Hye-jin Kim, Min Ding Estimating Nonresponse Bias in Survey Data Songting Dong, Ujwal Kayande Identifying the Presence and Cause of Fashion Cycles in the Choice of Given Names Hema Yoganarasimhan FD13 – Champions Center III FD14 – Champions Center VI FD15 – Champions Center V Network Effects Marketing Strategy I: General CRM VI: Customer Satisfaction Chair: Harikesh Nair Chair: Neil Bendle Chair: Jiana-Fu Wang Online Consumer-to-consumer Communication and Marketing Strategy Ganesh Iyer, Zsolt Katona Does Market Potential Always Attract New Market Entry? A Contingency View Namwoon Kim, Ge Zhan, Sungwook Min Identifying High Value Customers in a Network: Individual Characteristics Versus Social Influence Sang-Uk Jung, Qin Zhang, Gary J. Russell Repositioning via Abstraction Using Categorical Data Jonathan Lee, Heungsun Hwang Modeling Determinants of the Satisfaction-loyalty Relationship: Theoretical and Empirical Evidence Young Han Bae, Gary J. Russell, Lopo Rego Brand Value and Indirect Network Effects in a Two-sided Platform Yutec Sun Social Ties and User Generated Content: Evidence from an Online Social Network Harikesh Nair, Reto Hofstetter, Scott Shriver Business is in My Blood: Do Family Firms Outperform Non-family Firms During Economic Recessions? Saim Kashmiri, Vijay Mahajan Are Your Customers Crazy? Neil Bendle Does the Variance in Customer Satisfaction Matter for Firm Performance? Eun Young Lee, Shijin Yoo, Dong Wook Lee, Sundar Bharadwa The Impact of Online Railway Ticket Cancellation Policy on Revenue and Customer Satisfaction Jiana-Fu Wang Where Do the Joneses Go on Vacation? Social Comparison and the Weighting of Information Ingmar Nolte, Sandra Nolte, Leif Brandes 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 8.30-10.00 (SA) SA01 – Legends Ballroom I SA02 – Legends Ballroom II SA03 – Legends Ballroom III SA04 – Legends Ballroom V Conjoint Analysis: Improving the Process Twitter and Social Media Effects of Online Medium on Consumer Behavior Structural Models III Chair: Dan Horsky Chair: Abishek Borah Chair: Jie Zhang Chair: Andre Bonfrer Best-worst Conjoint Analysis as a Remedy for Lexicographic Choosers Joseph White, Keith Chrzan Methodology for Codifying Qualitative Twitter Content into Categorical Data Stephen Dann Disentangling the Effects of Online Shopping Decision Time on Website Conversion Dimitrios Tsekouras, Benedict Dellaert An Equilibrium Analysis of Online Social Content-sharing Websites Tony Bao, David Crandall Using Additional Data Collection and Analysis Steps to Improve the Validity of Online-based Conjoint Sebastian Selka, Daniel Baier Gossip: Can It Kill a Giant? Liwu Hsu, Shuba Srinivasan, Susan Fournier Estimation of Individual Level Multiattribute Utility from Ordinal Paired Preference Comparisons Dan Horsky, Paul Nelson, Sangwoo Shin Structural Dynamic Factor Analysis for Quantitative Trendspotting Rex Du, Wagner Kamakura Is All That Twitters Gold? Market Value of Digital Conversations in Social Media Abishek Borah, Gerard Tellis Clicks to Conversion: The Impact of Product and Price Information Vandana Ramachandran, Siva Viswanathan, Hank Lucas Retargeting – Investigating the Influence of Personalized Advertising on Online Purchase Behavior Alexander Bleier, Maik Eisenbeiss Usage Experience with Decision Aids and Evolution of Online Purchase Behavior Jie Zhang, Savannah Wei Shi Uncovering the Dynamics of Product and Process Innovation: An Analysis of Dynamic Discrete Games Xi Chen, John Dong Market Size, Quality, and Competition in Portuguese Driving Schools David Muir, Maria Ana Vitorino, Katja Seim Investigating Income Dynamics using the BLP Market Share Model Andre Bonfrer, Anirban Mukherjee SA05 – Legends Ballroom VI SA06 – Legends Ballroom VII SA07 – Founders I SA08 – Founders II Game Theory IV: Signaling Channels VI: General Entertainment Marketing I: Movies Continous-Time Marketing Chair: Yuanfang Lin Chair: Joseph Lajos Chair: Tom Fangyun Tan Chair: Olivier Rubel The Green Monoploist Kyung Jin Lim, Sridhar Balasubramanian, Pradeep Bhardwaj Impact of Consumer Returns on the Manufacturer's Optimal Returns Policy Thanh Tran, Ramarao Desiraju Demand Lifting through Pre-launch Marketing Activities Shijin Yoo, Tae Ho Song, Janghyuk Lee Life-cycle Channel Coordination Issues in Launching and Innovative Durable Product Xiuli He, Gutierrez J. Gutierrez The Effects of Asymmetric Interdependence on Asymmetric Conflict - Using Response Surface Analysis Hyang Mi Kim, Jae Wook Kim Awareness and Preference-based Consumer Segmentation in Forecasting Movie Box-office Performance Sangkil Moon, Barry Bayus, Youjae Yi, Junhee Kim Mass Behavior in a World of Connected Strangers Jurui Zhang, Yong Liu, Yubo Chen Informational Effect of Soldout Products on Consumer Search Behavior and Product Evaluation Yuanfang Lin, Paul Messinger, Xin Ge Do Channel Roles and the Salesdistribution Relationship Differ Between Countries? Joseph Lajos, Hubert Gatignon, Erin Anderson Can Star Actors and Directors Reduce the Risk of Box Office Failure? An Analysis of Risk Effects Alexa Burmester, Michel Clement, Steven Wu An Empirical Study of the Effects of Production Timing Decisions on Movie Financial Performance Tom Fangyun Tan, Kartik Hosanagar, Jehoshua (Josh) Eliashberg An Exact Method for Estimating Structural Continuous-time Models with Discrete-time Data Prasad A. Naik Advertising Investments under Competitive Clutter Olivier Rubel 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 8.30-10.00 (SA) SA09 – Founders III SA10 – Founders IV SA11 – Champions Center I SA12 – Champions Center II Retailing VI: Auto Industry Health Care Marketing I Social Influence I Online Word of Mouth Research Chair: Tae-kyun Kim Chair: Yansong Hu Chair: Jose-Domingo Mora Chair: Mounir Kehal Financial Incentives and Adoption of Hybrid Cars Sriram Venkataraman, Anindya Ghose Future Challenges for eHealth Concept Based on Market Analysis Lenka Jakubuv, Juraj Borovsky, Karel Hana The Silent Signals: Implicit User Generated Content and Implications for Consumer Decision Making Sunil Wattal, Anindya Ghose, Gordon Burtch Get Something for Nothing: Designing Optimal Free Sampling Strategy for Online Communities Shuojia Guo, Lei Wang, Yao Zhao Auto Industry Crisis and Firm Outcomes O. Cem Ozturk, Sriram Venkataraman, Pradeep Chintagunta Variation in Retailer Competition in Durable Goods Markets: An Empirical Study Tae-kyun Kim, S. Siddarth, Jorge Silva-Risso Exploring Relationships Among Marketing Effort, Customer’s Personality and Hospital Brand Experience Ravi Kumar, Shailendra Singh, Prem Purwar, Satyabhushan Dash Offering Pharmaceutical Samples: The Role of Physician Learning and Patient Payment Ability Ram Bala, Pradeep Bhardwaj, Yuxin Chen From Invention to Innovation: Technology Licensing by New Ventures in the Biopharmaceutical Industry Yansong Hu A Model of Social Dependence and Intra-group Interaction Youngju Kim, Jaehwan Kim, Neeraj Arora You May Have Influenced My Next Purchase: Social Influence in Food Purchase Behavior Jayati Sinha, Gary J. Russell, Dhananjay Nayakankuppam Intra and Cross-household Influences as Predictors of Individual Consumption Jose-Domingo Mora SA13 – Champions Center III SA14 – Champions Center VI SA15 – Champions Center V Private Labels I: General Marketing Strategy II: Firm Performance CRM VII: Customer Equity Chair: Anita Basalingappa Chair: Murali Mantrala Chair: Soumya Sarkar Implementing Online Store for National Brand When Competing Against Private Label Naoual Amrouche, Ruiliang Yan Measuring Cross-category Spillover Effects of Private Label Branding in U.S. Supermarket Retailing Sophie Theron, Timothy Richards, Geoffrey Pofahl What Drives Private-label Margins? Anne ter Braak, Inge Geyskens, Marnik G. Dekimpe The Dynamic Impact of Increasing Pricegap And Assortment-imitation on Private Label Performance Murali Mantrala, Elina Tang, Srinath Beldona, Shrihari Sridhar, Suman Basuroy Various Strategic Orientations: Theoretical Comparison, Construct Refinement and Empirical Analyses Christian Hoops, Michael Bücker Effect of Advertising Capital and R&D Capital on Sales Growth, Profit Growth and Market Value Growth Gautham Vadakkepatt, Venkatesh Shankar, Rajan Varadarajan Influence of Market Orientation on Corporate Brand Performance: Evidences from Indian B2B Firms Soumya Sarkar, Prashant Mishra The Formation of Impulse Buying: A Perspective on Self-control Failure of Consumer Behavior in CRM Kok Wei Khong, Hui-I Yao Monetizing UGC: A Hybrid Content Approach Theodoros Evgeniou, Kaifu Zhang, Paddy Padmanabhan, Inyoung Chae Competitiveness of Customer Relationship Management: Does Profitability Really Matter? Tae Ho Song, Sang Yong Kim Differential Influences of Market Structures on Cognition and Affect Anita Basalingappa, M. S. Subhas The Impact of Online Referrals on Consumer Choice in the Context of Charity Donations Kyuseop Kwak, Luke Greenacre, Valeria Noguti, Alicia Tan eWom Conducing Text-based Knowledge Diffusion through the Social Web: An Empirical Study Mounir Kehal 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 10.30-12.00 (SB) SB01 – Legends Ballroom I SB02 – Legends Ballroom II SB03 – Legends Ballroom III SB04 – Legends Ballroom V Advertising: Strategy Social Networks Online Consumer Behavior Product Management: General Chair: Gangshu Cai Chair: Christian Barrot Chair: Donna L. Hoffman Chair: Yeong Seon Kang Competing In Hollywood Claudio Panico, Sebastiano Delre Stimulus and Mutual Interaction Stochastic Bass Model Tolga Akcura, Kemal Altinkemer Post-consumption Satisfaction with Movies: A Multivariate Poisson Analysis of Online Ratings Ruijiao Guo, Purushottam Papatla Voice Banking: An Exploratory Study of the Access to Banking Services Using Natural Speech Mauro Arancibia, Claudio Villar, Jorge Marshall, Natalia Arancibia, Sergio Meza Advertising and Pricing Strategies for Luxury Brands with Social Influence and Brand Maintainance Jin-Hui Zheng, Chun-Hung Chiu, Tsan-Ming Choi Assessing Value in Product Networks Barak Libai, Eyal Carmi, Ohad Yassin, Gal Oestreicher Persuading Consumers With Social Attitudes Daniel Halbheer, Stefan Buehler Consumers as Active Participants in Viral Marketing Campaigns – Analyzing Forwarding Behavior Christian Pescher, Martin Spann, Philipp Reichhart The Role of Trust in the Firm-hosted Virtual Community in Purchase Intentions Formationt Illaria Dalla Pozz Modeling Unobserved Drop-out Rate to Optimize e-Panelist Lifetime Value Arnaud De Bruyn R&D Spillover and Product Differentiation in Fully Covered Market Xin Wang, Yuying Xie Downsizing or Price Competition Responding to Increasing Input Cost Yeong Seon Kang Efficacy of Advertising Structures and Cost Sharing Formats in a Competing Channel Gangshu Cai, Bin Liu, Zhijian (Zj) Pei An Empirical Comparison of Seeding Strategies for Viral Marketing Christian Barrot Why People Use Social Media: How Motivations Influence Goal Pursuit Donna L. Hoffmann SB05 – Legends Ballroom VI SB06 – Legends Ballroom VII SB07 – Founders I SB08 – Founders II Game Theory V: General Improving Efficiency in Marketing Negotiations Entertainment Marketing II Privacy and Marketing Chair: Ralf Wagner Chair: Erik Bushey Chair: Catherine Tucker Co-chair: Katrin Bloch The Impact of Product Placement on Ad Avoidance Natasha Foutz, David Schweidel, Robin Tanner The Impact of Relative Standards on the Propensity to Disclose Alessandro Acquisti, Leslie John, George Loewenstein The Advertising Role of Professional Critics in the Book Industry Michel Clement, Marco Caliendo Privacy as Resistance to Segmentation Luc Wathieu Chair: Ruhai Wu Within-firm and Across-firm Search: The Impact on Firms’ Product Lines and Prices Anthony Dukes, Lin Liu Bounded Rationality in Dynamic Games: Insights into Strategy Optimization Amid Player Uncertainty Jennifer Cutler Firm Strategies in the “Mid Tail” of Platform-based Retailing Baojun Jiang, Kinshuk Jerath, Kannan Srinivasan Repeated Consumption Pattern under Different Pricing Schemes Ruhai Wu, Suman Basuroy Measuring Efficiency of Negotiated Exhanges: An Evaluation, Refinement and Extension P.V. (Sundar) Balakrishnan, Charles Patton, Robert Wilken Benefits of Mediating Lawyers in Negotiations Olivier Mesly The Role of Intuition and Deliveration in Negotiations Katrin Bloch, Ralf Wagner The Role of Team Composition in Cross-cultural Business Negotiations Robert Wilken, Frank Jacob, Nathalie Prime Facing Bargaining Power Ralf Wagner, Katrin Bloch US Holidays in Non-US Markets: Moderating Role of Movie Nationality in Demand Fluctuation Joonhyuk Yang, Wonjoon Kim Modeling Head-to-head Competition and Quality Decisions in Television Program Scheduling Erik Bushey, Udatta Palekar Misplaced Confidences: Privacy and the Control Paradox Laura Brandimarte, Alessandro Acquisti, George Loewenstein Social Networks, Personalized Advertising, and Privacy Controls Catherine Tucker 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 10.30-12.00 (SB) SB09 – Founders III SB10 – Founders IV SB11 – Champions Center I SB12 – Champions Center II International Marketing I: General/Emerging Markets Health Care Marketing II Social Influence II No Session Chair: Jaap Wieringa Chair: Duraipandian Israel Product Bundling in Patent-protected Markets Eelco Kappe, Stefan Stremersch Co-creation of Social Value in an Online Brand Community Kwok Ho Poon, Leslie S.C. Yip How Generic Drugs Affect Brands Before and After Entry Jaap Wieringa, Peter Leeflang, Ernst Osinga What is There to ‘Like’ About Facebook? K N Rajendran, Steven B Corbin, Ciara Pearce, Matthew Bunker Chair: Sameer Mathur Global Expansion to vs. from Emerging Markets: An Empirical Study of Crossborder M & A's Completion Chenxi Zhou, Qi Wang, Jinhong Xie Going Global: Why Some Firms from Emerging Markets Internationalize More than Others Sourindra Banerjee, Rajesh Chandy, Jaideep Prabhu Multinational Strategic Alliance Models between Taiwan and China Shih-Wei Huang, Wun-Hwa Chen, Ai-Hsuan Chiang Quantity Discounts in Emerging Markets Sameer Mathur, Kannan Srinivasan, Preyas Desai How, When and to Whom Should Pharmaceutical Innovations be Promoted? Katrin Reber, Peter Leeflang, Philip Stern, Jaap Wieringa Optimal Allocation of Marketing Resources: Employing Spatially Determined Social Multiplier Effects between Physicians Sina Henningsen, Soenke Albers, Tammo Bijmolt Too Much or Not Enough - How the Degree of Interpersonal Similarity Forces Compliance with Requests Johannes Hattula, Sven Reinecke, Stefan Hattula User Personality, Perceived Benefits and Usage Intensity of Social Networking Sites: An Indian Study Duraipandian Israel, Debasis Pradhan SB13 – Champions Center III SB14 – Champions Center VI SB15 – Champions Center V Private Labels II: Effect on the Distribution Channel Marketing Strategy III: General CRM VIII: Customer Lifetime Value Chair: Nipun Agarwal Chair: Peter Pal Zubcsek To Research or to Execute? Analysis of the Drivers of Marketing Performances Chiara Saibene, Fabio Ancarani Churn Prediction Using Bayesian Ensemble in Telecommunications Market Jaewook Lee, Namhyong Kim Recall Now or Recall Later: Investigating Drivers of a Firm’s Decision to Delay a Recall Meike Eilert, Kartik Kalaignanam, Satish Jayachandran Improved Churn Prediction with More Effective Use of Customer Data Özden Gür Ali, Umut Ariturk, Hamdi Ozcelik Chair: Alexei Alexandrov Retailer Brand: To Keep it Private or Not? Yunchuan Liu, Liwen Chen, Steve Gilbert Retailer Brand Introduction with Consumer Evaluation Ying Xiao, Yunchuan Liu Market Expansion Effort in a Common Retailer Channel with Asymmetric Manufacturers Serdar Sayman, Gangshu Cai Effects of Manufacturers' Advertising on Volumes, Retail Margins, and Retail Profits Alexei Alexandrov Virtual Events: An Emerging Tactic that Complements the World of Experience Marketing Nipun Agarwal Information Communities: The Network Structure of Communication Peter Pal Zubcsek, Imran Chowdhury, Zsolt Katona 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 1.30-3.00 (SC) SC01 – Legends Ballroom I SC02 – Legends Ballroom II SC03 – Legends Ballroom III SC04 – Legends Ballroom V Advertising and Two Sided Markets Auctions and Pricing Meet the Editor: Journal of Service Research Unique Topics 2 Chair: Kaifu Zhang Chair: Woochoel Shin Chair: Nithya Rajamani Chair: Richard Batsell The Impact of Advertising on Media Bias Tansev Geylani, T. Pinar Yildirim, Esther Gal-Or Lemony Prices: An Online Field Experiment on Price Dispersion Zemin Zhong, David Ong Matching Markets for Contextual Advertising: The Tao of Taobao and the Sense of AdSense Chunhua Wu, Kaifu Zhang, Tat Y. Chan Two-dimensional Auctions for Sponsored Search Amin Sayedi, Kinshuk Jerath Is Online Content Worth Paying For?: A Two-sided Market Approach Jinsuh Lee, Manohar Kalwani Meet the Editors Role of Government in Marketing Sustainable Development: An Exploratory Investigation V. Mukunda Das, Saji K B Linkages between Infrastructure and Consumption Demand in Emerging Markets Puja Agarwal Does Higher Transparency Lead to More Search in Online Auctions? Peter T. L. Popkowski Leszczyc, Ernan Haruvy Poverty (Tenure) Track Daniel Shapira, Eran Manes The Nature of Informal Garments Markets: An Empirical Examination in Emerging Economy Prashant Mishra, Gopal Das Contextual Advertising Kaifu Zhang, Zsolt Katona First-page Bid Estimates and Keyword Search Advertising: A Strategic Analysis Woochoel Shin, Preyas Desai, Wilfred Amaldoss SC05 – Legends Ballroom VI SC06 – Legends Ballroom VII SC07 – Founders I SC08 – Founders II No Session Consumer Preferences Entertainment Marketing III No Session Chair: Doug Walker Chair: Dominik Papies Awareness and Ability to Express Preferences and its Impact on the Establishment of Causal Relations Rubén Huertas-García, Paloma Miravitlles-Matamor, Esther Hormiga, Jorge Lengler Co-chair: Sohyoun Shin A Bayesian Approach to Estimating Demand for Product Characteristics: An Application to Coffee Purchase in Boston Margil Funtanilla, Benaissa Chidmi An Anti-ideal Approximation of the Mixed Logit Model Robert Bordley Can CRM Create Goal Incongruence Among Salespeople and their Firms? An Agency Theory Perspective Doug Walker, Eli Jones, Keith Richards Influence of Film Adaptation on Motion Picture Performance: Experiences on SF Films in Hollywood Sunghan Ryu, Young-Gul Kim, Jae Kyu Lee Buy-now Prices at Entertainment Shopping Auctions Jochen Reiner, Martin Natter, Bernd Skiera Testing Strategies in Hollywood: A Duopolistic Game vs. an Agent Based Model Sebastiano Delre, Claudio Panico An Experimental Analysis of Price Elasticities for Music Downloads Dominik Papies, Martin Spann, Michel Clement 2011 INFORMS Marketing Science Conference Saturday, June 11th, 2011 1.30-3.00 (SC) SC09 – Founders III SC10 – Founders IV SC11 – Champions Center I SC12 – Champions Center II International Marketing II No Session No Session No Session SC13 – Champions Center III SC14 – Champions Center VI SC15 – Champions Center V Private Labels III: Effect on Market Shares Marketing Strategy IV: Firm Performance No Session Chair: Hyeong-Tak Lee Chair: Sohyoun Shin The Long Term Impact of a Recession on Brand Shares Satheeshkumar Seenivasan, Debabrata Talukdar, K. Sudhir Drivers of International Growth: Analysis of U.S. Franchisors’ International Growth Strategies Bart Devoldere, Venkatesh Shankar The Introduction of a Store Brand in a High-quality Market Segment: Analysis of a Natural Experiment Elena Castellari, Rui Huang Analyzing the Dynamics of Satisfaction, Recommendation and Customer Acquisition Henning Kreis, Till Dannewald Investigation of Determinants of Private Label Success in an Integrated Framework Hyeong-Tak Lee, Thomas Gruca Exploring the Components of Marketing Process Capability & Confirming its Relationship w/Performance Sohyoun Shin Chair: Fareena Sultan Beyond Globalization: Effectiveness of Technology Strategies of Foreign Firms in China Bennett C. K. Yim, Caleb Tse, Eden Yin National Influencers on Adoption and Usage of Online Auction Websites: New Zealand, Germany & Korea Tony Garrett, Jong-Ho Lee, Stefan Bodenberg Unraveling the Internationalizationprofitability Paradox Joseph Johnson, Debanjan Mitra, Eden Yin Consumers Un-tethered: A Three-market Study of Consumer Acceptance of Mobile Marketing Fareena Sultan, Andrew J. Rohm, Tao (Tony) Gao, Margherita Pagani Conference Sessions Thursday, 8:30am - 10:00am How to Navigate the Contributed Sessions ■ TA01 There are four primary resources to help you understand and navigate the Conference Sessions: Legends Ballroom I • This contributed session listing provides the most detailed information. The listing is presented chronologically by day/time, showing each session and the papers/abstracts/authors within each session. Cluster: Special Sessions Invited Session Marketing Science Institute I Chair: Don Lehmann, Columbia University, Columbia Business School, New York, NY, United States of America, [email protected] 1 - The History of Marketing Science: The Early Years Russ Winer, New York University, New York, NY, United States of America, [email protected] • The Author, Session Chair and Session indices provide cross-reference assistance (pages 93-99). In the late 1950s, the Ford Foundation sponsored a number of programs to increase the rigor of the research being conducted in business schools. Up to that point, business school research was largely focused on case studies and professional, highlypractical topics. An exception was the science-based approach at the Graduate School of Industrial Administration at Carnegie Mellon University. A result of those programs focused on bringing a more scientific approach to analyzing marketing programs was the publication of a number of books including one in 1961 by Frank Bass and others, Mathematical Models and Methods in Marketing and another by Ronald Frank, Alfred Kuehn, and William Massy, Quantitative Techniques in Marketing Analysis. An additional result of the push towards making marketing decision-making more scientific was the establishment of the Marketing Science Institute in 1961. In this talk, I will trace the beginnings of the field of marketing science focusing on the decades of the 1950s, 1960s, and 1970s just up to the founding of the journal Marketing Science. I will highlight the development of research in several areas including brand switching models, advertising response, buyer behavior, and comprehensive models such as those developed for new product forecasting and sales management. Key people and publications will be noted. • The map and floor plans included in the Front Matter show you where technical session tracks are located. • The “Master Track Schedule” is on the back cover. This is an overview of the tracks (general topic areas) and when/where they are scheduled. Quickest Way to Find Your Own Session Use the Author Index (pages 94-97) — the session code for your presentation(s) will be shown. Then refer to the full session listing for the room location of your session(s). 2 - New Product Design under Channel Acceptance: Brick-and-Mortar, Online Exclusive, or Brick-and-Click Lan Luo, University of Southern California, Los Angeles, CA, United States of America, [email protected], Jiong Sun The Session Codes SB01 The day of the week Track number. Coordinates with the room locations shown in the Master Track Schedule. Room locations are also indicated in the listing for each session. In today’s marketplace, many product markets are characterized by the existence of powerful retailers (e.g., Home Depot and Toys R Us) that serve as gatekeepers of new product introductions. In recent years, virtually all such “power retailers” started to establish online stores to further expand their shelf-spaces as well as customer bases. The rising popularity of such online stores provides manufacturers with a new opportunity as well as a new challenge in determining how to design their new products for powerful retailers that operate in multiple channels. We develop a gametheoretical model to show that, in the presence of the online store, the manufacturer may introduce a product of higher quality to be carried offline (or in the retailer’s brick-and-click stores) than in the absence of the online store. When the offline exclusive (or brick-and-click) option is just slightly more desirable than the online exclusive option for the manufacturer, the retailer enjoys the most leverage in his relationship with the manufacturer and reaps the highest possible profit. The manufacturer’s profit and the channel efficiency may also improve with the introduction of the online store. This occurs when the retailer’s offline reservation profit or the online shopping benefit is high. In such cases, the manufacturer will design a product of a lower quality for the retailer’s online store. Our analysis also yields managerial insights for retailers that operate both offline and online stores. Time Block. Matches the time blocks shown in the Master Track Schedule. Time Blocks THURSDAY Session Session Session Session A B C D 8:30-10:00am 10:30am-12:00pm 1:30-3:00pm 3:30-5:00pm 3 - The Evolution of Research on New Products, Innovation, and Growth Don Lehmann, Columbia University, Columbia Business School, New York, NY, United States of America, [email protected] FRIDAY Session Session Session Session A B C D Work in the area of new products, innovation, and growth has evolved over the last fifty plus years. This brief talk uses two sources, MSI’s research priorities and award winning papers, along with judgment to trace the evolution of this research area. The topic is not a new one as classics by Katz and Lazarsfeld (1955), Fourt and Woodlock (1960), and Bass (1969) demonstrate. Interestingly, however, tracing MSI’s research priorities shows a remarkable pattern. From 1974 to 1992, new products per se were not a principle priority; rather, the focus was on improving marketing mix efficiency. Beginning in 1992, however, new products and innovation have consistently been a top priority. Three interesting trends in the nature of this priority are evident. First, the focus has moved from the lone inventor to teams and cross functional efforts to external collaborations with partner firms and special customers (i.e. lead users) to solution spotting and customer engagement. Second, while forecasting (predicting success) continues to be a popular topic, more emphasis has been placed on understanding the adoption process, creation (e.g., design), managing the process, and valuing innovation efforts. Finally, the focus has evolved from new products to really new products to innovation in general to (organic) growth. These trends suggest several future research areas including structured creativity, design production efficiency, repair, upgrades, and disposal/re-purposing, internal firm implementation, and organic creation of new products via information technology. 8:30-10:00am 10:30am-12:00pm 1:30-3:00pm 3:30-5:00pm SATURDAY Session A Session B Session C 8:30-10:00am 10:30am-12:00pm 1:30-3:00pm Room Locations All session rooms are located in the InterContinental Houston on the Ground Level and 2nd Floor. 1 TA02 MARKETING SCIENCE CONFERENCE – 2011 ■ TA02 4 - Are Audience Based Online Metrics Leading Indicators of Brand Performance? Shuba Srinivasan, Associate Professor of Marketing, Boston University, School of Management, 595 Commonwealth Avenue, Boston, MA, 02215, United States of America, [email protected], Randolph Bucklin, Koen Pauwels, Oliver Rutz Legends Ballroom II Google WPP Award Papers Cluster: Internet and Interactive Marketing Invited Session This study analyzes the added explanatory value of including audience-based online metrics in a sales response model that already accounts for short and long-term effects of the traditional marketing mix. We also investigate the relationships among ‘behavioral’ intermediate measures (such as click-through) and ‘attitudinal’ intermediate metrics, such as brand liking. Finally, we assess whether online metrics are leading indicators of brand performance. Dynamic systems models connect marketing mix actions, online and offline mindset metrics and sales over time for a leading consumer packaged good. We find that including online marketing metrics adds significant predictive ability to models of sales performance. However, online success may also generate online backlash and both positive as negative online affect drives performance. In turn, online metrics are driven by marketing actions, including advertising and price. Our findings suggest that customer engagement in online specific campaigns indeed helps build brands. Chair: Shuba Srinivasan, Associate Professor of Marketing, Boston University, School of Management, 595 Commonwealth Avenue, Boston, MA, 02215, United States of America, [email protected] 1 - A Structural Model of Employee Behavioral Dynamics in Enterprise Social Media Yan Huang, PhD Student, Carnegie Mellon University, Pittsburgh, PA, 15213, United States of America, [email protected], Anindya Ghose We develop and estimate a dynamic structural framework to analyze social media content creation and consumption behavior by employees within an enterprise. We focus in particular on employees’ blogging behavior. The model is flexible enough to handle trade-offs between blog posting and blog reading as well as between workrelated content and leisure-related content, all of which are ubiquitous in actual blogging forums. We apply the model to a unique dataset that comprises of the complete details of blog posting and reading behavior of 2396 employees over a 15month period at a Fortune 1000 IT services and consulting firm. We find that blogging has a significant long-term effect in that it is only in the long term that the benefits of blogging outweigh the costs. There is also evidence of strong competition among employees with regard to attracting readership for their posts. While readership of leisure posts provides little direct utility, employees still post a significant amount of these posts because there is a significant spillover effect on the readership of work posts from leisure posts. In the counterfactual experiment, we find that a policy of prohibiting leisure-related activities can hurt the knowledge sharing in enterprise setting. By demonstrating that there are positive spillovers from work-related blogging to leisure-related blogging, our results suggest that a policy of abolishing leisure-related content creation can inadvertently have adverse consequences on work-related content creation in an enterprise social media setting. ■ TA03 Legends Ballroom III Internet Contributed Session Chair: Anita Elberse, Associate Professor, Harvard Business School, Soldiers Field, Boston, MA, 02163, United States of America, [email protected] 1 - Information Available versus Information Acquired? Implications for Consumer Choice Models S. Siddarth, University of Southern California, Marshall School of Business, 3660 Trousdale Parkway, Los Angeles, CA, 90089, United States of America, [email protected], Imran Currim, Ofer Mintz 2 - Media Aggregators and the Link Economy: Strategic Hyperlink Formation in Content Networks Chrysanthos Dellarocas, Associate Professor, Boston University, Boston, MA, United States of America, [email protected], William Rand, Zsolt Katona Consumers face daily decisions about the products they want to buy. The digital revolution has significantly enhanced consumers’ accessibility to information and hence the value they derive from it. Commercial websites have the ability to data offer the consumer a choice to acquire product feature information and analyze this data to provide insights that could improve their ability to convert visitors into buyers. In this research we seek to understand how consumers use product information and how best to model their choice process. We analyze the choices of a sample of shoppers who visited a website and had an opportunity to choose one out of three products, for which the firm provided information on a large set of product features. A unique aspect of the data is that the attribute values in the corresponding cells were hidden and shoppers had to explicitly click on cells to acquire this information. We estimate a set of choice models to infer the impact of the features on consumer choice including a standard multinomial logit model that incorporates all of the available alternative and attribute information, a choice set model that accounts only for those alternatives that consumers explicitly consider, a restricted attribute model that only incorporates the attribute information that was explicitly acquired, and, finally, a model that accounts for both choice sets and attribute restriction. We compare the predictive performance of the models on a holdout sample and also compare the parameter estimates in order to gain insights into how inferences about the impact of different attributes on consumer choice depend upon the information acquired. A key property of the World Wide Web is the possibility for firms to place virtually costless links to third-party content as a substitute or complement to their own content. This ability to hyperlink has enabled new types of players, such as search engines and content aggregators, to successfully enter content ecosystems, attracting traffic and revenue by hosting links to the content of others. This, in turn, has sparked a heated controversy between content producers and aggregators regarding the legitimacy and social costs/benefits of uninhibited free linking. This work is the first to model the implications of interrelated and strategic hyper-linking and content investments. Our results provide a nuanced view of the, so called, “link economy”. We show that it is possible for content sites of similar ability to reduce competition and improve profits by forming links to each other; in these networks one site would invest heavily in content and other sites would link to it. Interestingly, competitive dynamics often preclude the formation of these link networks, even in settings where they would improve joint profits. Within these networks, aggregators can have both positive and negative effects. On the positive side aggregators make it easier for consumers to find good quality content, and thus increase the appeal of the entire content ecosystem, relative to alternative media. At the same time, their market entry takes away some of the attention and revenue that would otherwise go to content sites. Finally, by placing links to only a subset of available content, aggregators increase competitive pressure on content sites, inducing them to create better content but generate lower profits. 2 - Investigating the Dynamic Impact of Advertising on Online Search and Offline Sales Jeffrey Dotson, Assistant Professor of Marketing, Vanderbilt University, 401 21st Avenue South, Nashville, TN, 37203, United States of America, [email protected], Qing Liu, Sandeep Chandukala, Stefan Conrady 3 - The Broadcast Window Effect: Information Discovery and Cross-channel Substitution Patterns for Media Content Rahul Telang, Carnegie Mellon University, Pittsburgh, PA, 15213, United States of America, [email protected], Michael Smith, Anuj Kumar Although the conditions that motivate individuals to buy, sell, search, and post on the internet are diverse, the information generated as a byproduct of these activities has the potential to help marketers develop a better understanding of consumer and firm behavior. In this paper we collect and utilize online product consideration data in order to build a dynamic model of sales and advertising. The objectives of this research are twofold: first, to incorporate secondary data collected from online sources into a model of demand, thus improving our ability to forecast sales; and second, to develop a better understanding of the role of advertising in the sales generation process. We illustrate the benefits of our approach using data for a luxury automobile brand where we show that the role of advertising is one of demand creation rather than purchase acceleration. Several recent papers in the literature have shown that for movies and music, a small proportion of titles account for the majority of sales. In this paper we collect DVD sales data for movies broadcast on premium pay cable channels to analyze the degree to which information asymmetry is driving this skewness in the sales distribution for movies. Our data show that while broadcasting movies on pay cable channels increases demand for those movies, the increase in demand is disproportionately large for less popular movies. This suggests an information spillover effect of the movie broadcast: movie broadcast increases awareness about lesser-known movies and thus helps uninformed consumer discover the movies. Since popular movies are highly advertised during the theatrical and DVD release, there is less scope of discovery during its broadcast. In contrast, less successful movies are more likely to be discovered during its broadcast and thus experience higher increase in DVD sales during this period. We further estimate learning based model of DVD demand to precisely quantify the proportion of uninformed customers and thus lost DVD sales due to incomplete information. Our study contributes to the growing literature on impact of information provision on market outcomes and on the dynamics of long tail markets. 3 - Not to Click Through: The Benefits of Search Engine Advertising Combination of Old and New Media German Zenetti, Goethe-University Frankfurt, Grueneburgplatz 1, Frankfurt, Germany, [email protected], Tammo Bijmolt, Peter Leeflang, Daniel Klapper The technological process together with the widespread integration of the internet makes new communication possibilities available, especially for e-commerce products online adverting channels become more attractive. Thus, an important question for firms is to find out which combination of communication channels should be used to 2 MARKETING SCIENCE CONFERENCE – 2011 TA05 address a target group. i.e., should an advertising channel, such as search engine advertising, be integrated in the existing advertising planning? Which synergy effects will take place in that case with traditional media, such as TV? In this approach pre and post campaign survey data are used to measure the impact and synergy of TV, online and search engine adverting on four marketing metrics simultaneously. In doing so, it is accounted for the connectivity of the cognitive, affective and conative dimensions of advertising effectiveness, which might occur due to unmeasured variables that jointly influence advertising effectiveness. The results show that next to TV advertising also search engine advertising has significant effects on marketing metrics. Additionally, online advertising enhances the impact of TV advertising and search engine advertising enlarges the effectiveness of TV and online advertising. Thus, depending on the context managers should consider the inclusion of online and search engine advertising as marketing tool. methods. A series of experiments are conducted to empirically test the model. Our initial results demonstrate that dyads indeed exhibit enhanced compromise effects than individuals. An adaptive study design customized for individual member preferences will be used to demonstrate that dyadic compromise effects enhance the likelihood of suboptimal decisions made by dyads. We link our findings to product design, dyadic welfare and decision quality of groups. We suggest tactics groups can adopt to minimize suboptimal choice decisions. 4 - Viral Videos: The Dynamics of Online Video Advertising Campaigns Anita Elberse, Associate Professor, Harvard Business School, Soldiers Field, Boston, MA, 02163, United States of America, [email protected], Clarence Lee, Lingling Zhang Multiple facings of identical SKUs are the rule in commercial retailing and the available shelf space that determines the number of facings of a particular offering is scarce and expensive. Recently, various commercial market researchers have attempted to experimentally study the effect of the number of facings on the demand for individual SKUs. The basic idea is to conduct choice experiments where the choice sets vary the number of facings of individual offerings as an additional factor. We take this data as an opportunity to investigate primitive aspects of choice behavior. For example, we measure the degree of similarity competition b/w multiple facings of the same SKU and across different SKUs and the joint effect of satiation and similarity on the chosen quantities. At this point we are not concerned with the external validity of these experiments but interested in learning about fundamental aspects of choice behavior. 3 - Choice from Simulated Store Shelves - How Similar are Two Identical SKUs? Keyvan Dehmamy, Goethe University Frankfurt, Grüneburgplatz 1, Frankfurt a.M., 60323, Germany, [email protected], Thomas Otter Firms increasingly promote their products using advertisements posted on online video-sharing sites such as YouTube. There, web users often redistribute these “advertiser-seeded” advertisements, either in their original form or as altered, derivative works. Such user-generated “viral” placements can significantly enhance the true number of an advertising campaign’s impressions – in fact, across our data for movie and video game trailers, the number of views generated by viral placements is eight times greater than the number of views for the original “advertiser-seeded” placements. In this study, in order to help advertisers understand how their online video advertisements spread, we investigate the dynamics of viral video campaigns, modeling the accumulation of advertiser-seeded and user-generated views as two interrelated processes. We find that several instruments under the control of advertisers – notably the intensity, timing, and coverage of original video placements – influence the extent to which campaigns benefit from user-generated content. Our results suggest that, with the right strategy, advertisers can substantially and inexpensively increase the number of impressions that their online video campaigns yield. ■ TA05 Legends Ballroom VI New Product I: Introduction Contributed Session Chair: Michael Cohen, Assistant Professor in Residence, University of Connecticut, 1376 Storrs Road Unit 4021, Storrs, CT, 06269-4021, United States of America, [email protected] 1 - Observational Learning and Networking Externality in Decision-making Dongling Huang, Rensselaer Polytechnic Institute, 110 8th Street, Pitts., Troy, NY, 12180, United States of America, [email protected], Yuanping Ying, Andrei Strijnev ■ TA04 Legends Ballroom V Bayesian Econometrics I: Methods & Application Contributed Session Chair: Keyvan Dehmamy, Goethe University Frankfurt, Grüneburgplatz 1, Frankfurt a.M., 60323, Germany, [email protected] 1 - Dyadic Patent Citation and Firm Performance Yantao Wang, Marketing Department, Kellogg School of Management, 2001 Sheridan Rd, Evanston, IL, 60208, United States of America, [email protected], Sha Yang, Yi Qian We investigate the early momentum of new product release. When making a purchase decision, consumers often turn to two different sources of information: consumer word-of-mouth communications and other consumers’ purchase behavior. A third important non-infomational factor consumers may also consider is the fact that the number of existing users of a product may bring additional social or other functional benefits. In reality, the following three factors may all be in play: information sharing, observational learning, and network externalities. We attempt to account for all three factors in the consumer’s decision making process using movie sales data. Specifically, we propose a simple yet effective identification approach to separating the effects of observational learning and network externalities, while controlling for word-of-month. We condition our analysis on movie quality and opening momentum. Using our identification scheme, we find that both observational learning and network externalities have a considerable impact on consumers’ movie going decisions. In particular, network externalities have a stronger effect than observational learning. Firms in an industry are likely to interact with each other in different ways. Because of these interactions, one firm’s performance is not independent from the influence of other firms. One important form of firm interaction is patent citation. Dyadic patent citation reflects firm interaction in the form of innovative interdependence between two firms and the diffusion of knowledge among firms. Therefore, the first goal of this paper is to use annual information on the number of patents firm i (sender) cites from firm j (receiver) to measure knowledge interdependence, and to explain the interdependence by the dyad –specific characteristics of the two firms, the characteristics of the sender and the receiver, and latent dyad-, sender-, and receiverspecific factors. Knowledge diffusion can have important implications on the invention of new products, which in turn impacts the firm’s performance. Therefore, our second goal in this paper is to predict firm performance using firm’s sensitivity to its role as a sender as well as a receiver in the citation process while controlling for other time-varying characteristics of the firm. We build a two-step econometric model to simultaneously estimate the citation and performance equations using Bayesian MCMC method that also accounts for firm heterogeneity. 2 - Optimizing the Three Dimensions of New FMCG’s Market Success by Means of the Marketing Mix Tilo Halaszovich, Assistant Professor, University of Bremen, Hochschulring 4, Bremen, 28359, Germany, [email protected], Christoph Burmann “Innovate or die”, as stated by Robert G. Cooper since the 1980s, seems to become “innovate and die” in light of new products failing at an alarming rate of up to 80% in most FMCG markets. Therefore, in today’s marketplaces the firm’s ability to commercialize new products successfully is a central success factor. The disproportion between their importance and missing success can partly be attributed to a lack of knowledge concerning the influence of the marketing mix instruments on the outcome of new FMCG launches. Market performance of a new product is thereby a three dimensional construct including the product’s value share and its ability to generate customer satisfaction as well as customer acceptance. Given the fact that these three dimensions are closely connected with each other, understanding the marketing mix influence on the new FMCG’s market success requires simultaneous control for each success dimension and their interactions. Therefore, the authors develop a dynamic structural equation model which allows measuring the marketing mix impact in a holistic structure. The model is fitted to a sample of new FMCG launches in laundry and home-care markets in Germany and based on scanner-panel data. The results of the model reveal the interactions between all three measures and how they are influenced by the marketing mix in different ways. Combining these findings, an optimization of the whole marketing mix is possible throughout the introduction phase of new FMCG products. 2 - Dyadic Choice and Compromise Effects: Implications for Decision Optimality Lin Bao, PhD Student, University of Wisconsin-Madison, 4181 Grainger Hall, 975 University Ave, Madison, WI, 53706, United States of America, [email protected], Neeraj Arora, Qing Liu The compromise effect has been extensively studied in the marketing literature (e.g. Simonson 1989, Kivetz et al 2004). Its basic premise is that individuals are more likely to choose an intermediate option versus an extreme option. In this research we study how the compromise effect manifests itself in dyadic choices. In particular we investigate two questions: 1) Do dyads exhibit enhanced compromise effects than individuals? 2) Do such dyadic compromise effects enhance the likelihood that a suboptimal decision will be made by a dyad? At a theoretical level, an investigation of dyadic compromise effect is of interest because extant research on group polarization (Myers and Lamm 1976) suggests that groups are more likely to choose the extreme options. In contrast, evidence also exists that dyads with divergent preferences tend to negotiate and settle on an intermediate option (Menasco and Curry 1989). Also, it is known that the perception of being evaluated by group members enhances individual susceptibility to the compromise effect (Simonson 1989). In this research we develop a dyadic choice model that incorporates individual and dyad level compromise effects. Heterogeneity in compromise effects is captured using Bayesian 3 TA06 MARKETING SCIENCE CONFERENCE – 2011 3 - An Empirical Analysis of New Product Launch Michael Cohen, Assistant Professor in Residence, University of Connecticut, 1376 Storrs Road Unit 4021, Storrs, CT, 06269-4021, United States of America, [email protected], Rui Huang with the incremental sales percentage ranging from a low of 59% to a high of 85%. As regards the individual stores that get cannibalized, we find significant decay in cannibalization with distance. For example, the average cannibalization rate at a distance of one mile or less is approximately 4%. This figure drops to less than 1% at a distance of 7-10 miles. We discuss how managers can use the model presented in this paper to make two key decisions: (a) isolating locations that can be closed by identifying stores that yield the lowest marginal benefit to the chain and (b) choosing the best location for a new store from a set of viable alternatives. Validating the decision to launch a new product and its accompanying introduction strategy is a keystone component of the marketing planning process. Launch includes product development as well as the roll-out and marketing campaign. A common product introduction strategy is to develop products from the attributes of existing products. For example, Reese’s brand Whipps candy bar fuses the soft nougat of Three Musketeers with the peanut butter flavor of Reese’s Cups. This low cost product development strategy is typical in many experience-product markets, most prominently in packaged food products. This research introduces an empirical method to analyze market data for testing new product introduction strategies in an established marketplace. Our paper takes the approach of estimating a limited consumer awareness discrete choice demand model that specifies television advertising, display, coupons, and feature advertisements in a way that augments a product’s salience in the marketplace. We use Nielsen household purchase histories and gross rating point data for popular chocolate candy bars to estimate the model. The data are compiled from the New York, Chicago, and Los Angeles television markets and span 2006-2008. We use demand estimates to investigate the relative effectiveness of in-store promotion versus television advertising and demonstrate how to predict the performance of a new product before committing to launch. Thanks to Reese’s introduction of the Whipps candy bar during our data period, we are able to validate our model’s predictions. Results from these analyses provide guidance for both product development and promotion planning. 3 - How Wal-Mart’s Entry Affects Incumbent Retailers Huihui Wang, Duke University, Fuqua School of Business, 100 Fuqua Drive, Durham, NC, United States of America, [email protected], Carl Mela, Andrés Musalem Walmart is America’s largest company (Fortune 2010). In light its magnitude, the competition induced by Walmart’s expansion constitutes a sizable economic phenomenon. Surprisingly, the majority of the academic literature, in contrast to the media reports, concludes that price responses of incumbent retailers are mixed or minimal. However, this need not imply that there is no price reaction. On one hand, Walmart’s low price strategy may lead incumbent stores to lower their prices to compete with the new entrant. On the other hand, Wal-Mart may attract the more price sensitive customers from the incumbent firms’ customer mix. This would leave the incumbents with customers that are less price sensitive allowing them to raise their prices. Given that price response is a complex interplay of horizontal and vertical differentiation and costs, our objective is to disentangle these factors to provide insights into the nature of competitive response to Walmart’s entry. We start by formulating a model of consumer store choice which extends Kim et al. (2002). Given this model we consider how Walmart’s entry affects consumer behavior and prices in the market. Combining the results from different policy experiments we decompose how these different drivers (marginal costs, horizontal and vertical differentiation) affect incumbent retailers’ reactions to Walmart’s entry. ■ TA06 Legends Ballroom VII 4 - Retail Market Expansion and Substitution Effects: Evidence from a Field Experiment Frederico Rossi, University of North Carolina, Kenan-Flagler Business School, Chapel Hill, NC, United States of America, [email protected], Eric Anderson, Ralf Elsner, Duncan Simester Competition I: Measuring the Impact of Competition in Retail Markets I Cluster: Special Sessions Invited Session Chair: A. Yesim Orhun, University of Chicago, Booth School of Business, 5807 Woodlawn Avenue, Chicago, IL, 60637, United States of America, [email protected] 1 - The Impact of Competition on Endogenous Product Provision: The Case of Motion Picture Exhibition Market A. Yesim Orhun, University of Chicago, Booth School of Business, 5807 Woodlawn Avenue, Chicago, IL, 60637, United States of America, [email protected], Pradeep Chintagunta, Sriram Venkataraman Price discounts and price cues typically increase demand but is this due to: substitution between products, substitution between retailers, or market expansion? We conduct a large-scale field experiment with a European direct mail firm to answer these questions. We experimentally vary price discounts and price cues (“Sale” claims) for twenty books and CDs sold at one retailer. A unique aspect of the data is that we measure sales not just at the retailer that conducted the test, but also at two competing retailers who offer the same twenty items at the control price. This yields direct measures of retail store substitution and reveals findings that differ from outcomes elsewhere in the literature. We find evidence of retail store substitution and this is largest (i) when a discount is combined with a price cue (ii) for established items that have been previously sold by the retailer (iii) among store switching customers. Moreover, we find evidence of a shopping basket effect that leads to positive spillovers and increased demand of non-promoted items. We find evidence of market expansion for discounted items, but there is little evidence of overall market expansion. Our results offer empirical support for theories of retail price competition, including loss leader pricing and loyal/switcher models of promotion. In industries where products have short life cycles, and where products offered are a primary driver of demand to the firm, it is important to study the impact of a change in market structure on product choices. A canonical example of such an industry is the motion picture exhibition market. Using weekly U.S and Canada movie revenues at the theater level for the September 2003 - March 2005 period, we provide evidence on how a new theater entry affects an incumbent firms product choices and its revenues. As expected, we find significant negative impact of entry on incumbents revenues. Interestingly, the decrease in revenues is larger for incumbents that face entry of a theater of the same chain compared to incumbents that face entry by a rival chain. We show that as a result of entry of a co-owned firm, not only does the number of titles decrease more so than when the entrant is a rival, but also that the average tickets sold per title is lower. Importantly, we find that the decline in the ticket sales per title of the co-owned incumbent is fully explained by the decrease in the inherent box office potential of movies it screens. While we present movie choices of the incumbent as an explanation for the difference in business stealing and cannibalization, we note that our findings can shed empirical light on the larger issue of the impact of competition on product quality provision. We discuss the drivers of the observed quality response to competition. ■ TA07 Founders I ASA Special Session on the MarketingStatistics Interface - I Cluster: Special Sessions Invited Session Chair: Peter J. Lenk, University of Michigan, 701 Tappan Street, Ann Arbor, MI, United States of America, [email protected] 1 - Dynamic Market Segmentation Models and Methods Timothy J. Gilbride, University of Notre Dame, 399 Mendoza College of Business, Notre Dame, IN, United States of America, [email protected], Peter J. Lenk 2 - Empirical Investigation of Retail Expansion and Cannibalization in a Dynamic Environment S. Sriram, University of Michigan, Ross School of Business, Ann Arbor, MI, 48109, United States of America, [email protected], Joseph Pancras, V. Kumar Managers seeking to modify their portfolio of retail locations either by adding new stores or closing existing ones need to know the net impact of a store’s opening/closure on the overall chain performance. The answer to this question lies in inferring the extent to which each store generates incremental sales as opposed to competing with other stores belonging to the chain for the same set of customers. However, when the chain is experiencing a growth or a decline in sales, not accounting for these dynamics in brand preference is likely to yield misleading estimates regarding the magnitude of incremental sales versus cannibalization. We develop a demand model that accounts for dynamics in brand preferences and spatial competition between geographically proximate retail outlets. We calibrate the model parameters on both attitudinal and behavioral data for a fast food chain in a large U.S. city. The data reveal that the chain experienced increases in (a) overall chain sales, (b) number of stores, and (c) average sales per store during the period of our analysis. The results suggest that, on average, 69.7% of a store’s sales constitute incremental purchases with the rest derived from cannibalized sales from nearby stores belonging to the chain. However, there is significant heterogeneity across stores Commercial market segmentation studies can be expensive, time consuming projects that are infrequently updated. Because of the time between subsequent studies, managers may miss important changes in the market. This research suggests that smaller, less expensive studies can be used to track changes in market segments. The proposed Bayesian methodology adapts a general state-space model where the space equations relate the cross-sectional data to population-level parameters at a given point in time, and the state model captures the time dynamics in the population-level parameters. Improved algorithms for two generalized dynamic models are detailed. Sequential analysis is appropriate when the goal is to estimate model parameters as new data arrives without re-estimating past results. For the sequential analysis, we propose a novel particle filter samplings scheme that is more efficient and better suited for the structure of the data than methods proposed in the literature. When several data sets are available and the goal is to quantify trends and simultaneously estimate parameters for all time periods, a retrospective or full-information approach is preferred. We find that substantially smaller samples can be used in the tracking studies because information is shared across studies. The paper identifies the important 4 MARKETING SCIENCE CONFERENCE – 2011 TA09 been embracing open innovation through which they successfully bring knowledge outside the boundary of the firm into internal knowledge systems and integrate both external and internal knowledge to create high impact innovations. Even though there is growing interest in this new innovation paradigm, empirical academic research on this area is still very scant (Chesbrough 2006).We intend to integrate two most prominent perspectives in marketing strategy, social network theory and knowledge-based view, by exploring the alignments of a firm’s internal knowledge asset and external network asset in order to improve open innovation performance. We disaggregate knowledge asset and network asset into depth and breadth (Luca and Atuahene-Gima 2007). And argue the depth and breadth of knowledge and network assets form the “building blocks” of open innovation, which may depend on the alignment or “fit” of these assets. We refer to this as network-knowledge alignment strategy. Specifically, we explore network-knowledge alignment strategy in the context of Open Source Software (OSS) development, which has emerged as an important open innovation phenomenon (von Hippel 2005). By collecting the project embeddedness network data as well as developer’s knowledge across different knowledge domains from SourceForge.Net, the largest OSS development website, we investigate how OSS development teams can align its network and knowledge asset depth and breadth in order to achieve high project performance in terms of project internal efficiency and project external effectiveness. role that prior distributions play in the Bayesian analysis and suggests the predictive log likelihood be used for model selection when the managerial goal is to smooth parameter estimates and make short term predictions. The methods are illustrated with a unique 20 year data set on U.S. consumers’ attitudes towards marketing. 2 - Viral Marketing: Understanding the Diffusion of User Generated Content Within and Across Networks Yuchi Zhang, University of Maryland, 3330J Van Munching Hall, College Park, MD, 20742, United States of America, [email protected], Wendy W. Moe This research proposes a methodology for modeling the diffusion of user-generated content (UGC). The diffusion of UGC can be characterized as a process that is highly dependent on word-of-mouth referrals, more so than other contexts. Specifically, we examine the daily views for a sample of videos posted on YouTube. We assume that the audience for each video is drawn from multiple diffusion networks in the population. The UGC spreads within each network according to network-specific Weibull processes and across networks according to a stochastic process that allows each network to begin its diffusion process at different points in time. We examine covariate effects and correlations between networks in this process. Our results indicate that UGC tends to diffuse simultaneously across different subsets of the population. We identify three drivers of these diffusion processes: content, video poster, and the poster’s initial network and show the impact of each on the overall diffusion process. We find that when videos diffuse rapidly at the onset, the overall diffusion of the content is less likely to include secondary diffusion networks. Furthermore, if these secondary diffusion networks exist, they tend to view the video much later in time. We also find significant effects (that vary across genres) of the size of the poster’s subscriber base on secondary diffusion networks. Our analysis provides insight into how managers should tailor both their content and diffusion strategy to more effectively employ viral marketing tactics. 3 - Innovative Capability: Investigating Open Innovation, Marketing Capability, and Firm Performance Sanjay Sisodiya, Assistant Professor of Marketing, University of Idaho, 875 Campus Drive / P.O. Box 443161, Moscow, ID, 83844, United States of America, [email protected], Jean Johnson, Yany Grégoire The recently emerging trend of open innovation is gaining considerable interest. Here firms seek out external inputs for innovation and identify external paths for internally developed innovations (Chesbrough 2003, 2006). While there is evidence that firms following open innovation a successful (e.g., Huston and Sakkab, 2006), firms must also maintain a closed innovation perspective in order to develop unique innovations. Firms that are capable of performing both open and closed innovation may benefit by combining these two capabilities into a higher order innovative capability that could lead to even greater levels of firm performance. Firms with high innovative capability should outperform other firms that excel at either but not both open or closed innovation. While an innovative capability is intriguing, maintaining an innovative capability might not be enough to provide a firm with a competitive advantage. Thus it is important to also consider other combinative firm capabilities that could lead to heightened levels of firm success. Critical to success with an innovative capability, is the firms marketing capability (e.g., Vorhies and Morgan 2005). Those firms that maintain both an innovative capability, as manifested through open and closed innovation, combined with a marketing capability should achieve superior levels of firm performance as compared to those firms with a weak marketing capability. These hypotheses are tested on a sample of over 210 publicly traded technology firms. Results support the notion that while following open innovation firms may achieve a competitive advantage, it is the combination of maintaining an innovative capability and a marketing capability that contributes to greater levels of performance. 3 - Bayesian Model Selection and Simulation Bias of the Harmonic Mean Estimator of Integrated Likelihoods Peter J. Lenk, University of Michigan, 701 Tappan Street, Ann Arbor, MI, United States of America, [email protected] Bayesian model selection depends on the integrated likelihood of the data given the model. Newton and Raftery’s harmonic mean estimator (HME) is simple to implement by computing the likelihood of the data at MCMC draws from the posterior distribution. Alternative methods in the literature require additional simulations or more extensive computations. In theory HME is consistent but can have an infnite variance. In practice, the computed HME is simulation biased. This talk identifies the source of the bias and recommends several algorithms for adjusting the HME to remove it. The bias can be substantial and can negatively affect HME’s ability to select the correct model in Bayesian model selection. The bias often causes the computed HME to overestimate the integrated likelihood, and the amount of bias tends to be larger for more complex models. When the computed HME errs, it tends to select models that are too complex. Simulation studies of linear and logistic regression models demonstrate that the adjusted HME effectively removes the simulation bias, is more accurate, and indicates more reliably the best model. ■ TA08 ■ TA09 Founders II Founders III Innovation I: Open Innovation Promotions I Contributed Session Contributed Session Chair: Sanjay Sisodiya, Assistant Professor of Marketing, University of Idaho, 875 Campus Drive / P.O. Box 443161, Moscow, ID, 83844, United States of America, [email protected] 1 - Open Innovation Practices and Market Outcomes: The Moderating Role of Product Capabilities Deepa Chandrasekaran, Assistant Professor of Marketing, Lehigh University, 621 Taylor St., Bethlehem, PA, 18015, United States of America, [email protected], Gaia Rubera, Andrea Ordanini Chair: Dinesh Gauri, Syracuse University, 721 University Ave, Syracuse, NY, 13214, United States of America, [email protected] 1 - Timing of Retailer Price-promotions Huseyin Karaca, Northwestern University, 2001 Sheridan Road, Evanston, IL, United States of America, [email protected], Lakshman Krishnamurthi, Vincent Nijs, Anne T. Coughlan Sales promotions have been the subject of numerous studies in the literature. A quick overview of this literature, however, reveals surprisingly little research on the subject of timing of promotions. Our paper addresses retailers’ promotion timing problem by studying the strategic behavior of a retailer in scheduling its price promotions for frequently purchased packaged goods around peak demand periods, facing consumer segments that differ in their propensity to purchase in and out of peak demand periods. We present and analyze an analytical framework from the perspective of a profit-maximizing retailer facing demand for two brands within a product category, i.e. a national brand and a private label, from two different segments of utilitymaximizing consumers over a two-period time frame. Our paper makes both theoretical and substantive contributions to the literature. The consideration of dynamically varying segment participation in the market is novel from a theoretical perspective. As opposed to the static segments considered in the previous literature, the dynamic segmenting approach in this research considers inter-temporal changes in the structure of aggregate demand. On the substantive front, the research brings to light a particular segmented consumption behavior in the empirical data and uses this managerial observation to explain and predict retail promotion activity across markets and products. Our research has immediate managerial implications. Considering the time pressure due to scheduling of promotions of hundreds of categories in any given week, the retailers might resort to rule-of-thumb promotional decisions rather than optimal ones. Our analytic framework, however, generates sensible rules retailers can use when scheduling their price-promotions. While the use of open innovation (OI) has gained importance in the last decade, empirical evidence about its effects is largely anecdotal. Particularly how OI interacts with a firm’s new product (NPD) capabilities is not clear. While an absorptive capacity perspective leads us to believe that only firms with good NPD capabilities can benefit from OI; the Non-Invented-Here Syndrome literature suggests that OI is more useful for firms with poor NPD capabilities. The authors contend that the effect of OI in fact depends on the interaction of NPD capabilities with the nature of input acquired via OI. They identify three types of OI practices: ideas-centric; technology-centric and product-centric. Their empirical analysis combines primary data from a survey of 239 Italian firms with secondary data on innovation and financial outcomes from proprietary databases. They demonstrate that the effect of OI on innovation rate is indeed contingent upon the level of a firm’s NPD capabilities and differs according to the type of OI. Theoretical and managerial implications are provided. 2 - Network and Knowledge Asset Alignment in Open Innovation Tanya Tang, University of Illinois at Urbana-Champaign, 350 Wohlers Hall, 1206 South Sixth Street, Champaign, IL, 61820, United States of America, [email protected], Eric Fang, William Qualls Open innovation paradigm has received a lot of attention in managerial practices during the last several years. Companies like IBM, Cisco Systems, DuPont have all 5 TA10 MARKETING SCIENCE CONFERENCE – 2011 2 - Stars, Leaders, Free-riders and Losers: Roles in the Category Expansion during Sales Promotions Sergio Meza, Assistant Professor, University of Toronto, Mississauga, Toronto, ON, Canada, [email protected] ■ TA10 In the marketing literature there are opposite views of whether sales promotions expand the category sales or not. Increase in sales may be due to within-category brand switching (Gupta, 1988; Sun et al, 2003) or category expansion (Chintagunta, 1993; Van Heerde, 1999; Nijs et al, 2001). Category expansion can be the result of anticipated consumption (van Heerde et al, 2004; Mace and Neslin, 2004; Bell et al, 2002) or additional consumption (Ailawadi and Neslin, 1998; van Heerde, 1999; van Heerde et al, 2004, Blattberg et al, 1995). The issue is complicated by the fact that sales promotions may be scheduled to coincide with peaks of demand (see Chevalier et al. 2003, Meza 2011)). Are expansions of the category during sales promotions due to the effect of the sales promotions? Or are they due to the fact that sales promotions are scheduled when expansions of the category are expected? In this paper we investigate which brands have a larger effect in expanding the category when promoted, and which brands sell more when others are promoted. Brands that expand the most (when promoted) and sell the most (when others are promoted) are called “stars”. Brands that expand the most (when promoted), but sell the least (when others are promoted) are called “Leaders”. Brands that expand the least (when promoted), but sell the most (when others are promoted) are called “Free-riders”. Brands that do not expand nor sell are called “losers”. Contributed Session Founders IV Consumer Behavior: Perceptions Chair: Jana Diels, Humboldt-Universitaat zu Berlin, Wirtschaftswissenschaftliche Fakultaat, Spandauer Str. 1, Berlin, Germany, [email protected] 1 - The Effects of New Product Introduction to Different Category Context on Price Evaluations Akihiro Nishimoto, Otaru University of Commerce, 3-5-21 Midori, Otaru, Japan, [email protected], Sayaka Ishimaru, Sotaro Katsumata, Eiji Motohashi The purpose in this research is to demonstrate the effects of new product introduction to different category context on price evaluations. In particular, we focus on the phenomenon that the heterogeneity of category knowledge among consumers, which is activated in the evaluations of the new product, affects the willingness to pay of it. That is, we investigate on the effects of category knowledge as the priming effects in recognizing the new product. From the past studies, the priming has three effects. The first is assimilation effect. This is the priming effect based on the interpretation of the new product. The second is contrast effect. This priming effect is based on the comparison between the prior knowledge and the new product. And the third is correction contrast effect. This is the priming effect for the experts based on correction for context effect. In this research, we run three experiments. In the first experiment, we have verified the effects of new product introduction to different category context on price evaluations, taking into account the strategic relationship with the priming category and the target category. In the second experiment, we have investigated the priming effects diluting the evaluation of the new product. And the third experiment, we have confirmed the effects of different category context as the priming effect to generalize our results. From the three experiments, we have demonstrated the relationship between the category context as the priming effect and the level of category knowledge. We will show some of the results and implications in the presentation. 3 - The Impact of Retailer Promotional Activities on Store Traffic Shyda Valizade-Funder, University of Mainz, Jakob- Welder-Weg 9, Mainz, 55099, Germany, [email protected], Oliver Heil, Kamel Jedidi Building store traffic through marketing activities is one of the major goals for retailers seeking sales growth. By understanding how to build traffic, retailers can assess the productivity of their advertising and promotional activities. This enables retailers to assess which communication media are effective in targeting their customers, determine which promotion activities to implement, and optimize resource allocation across these marketing efforts. The objective of our study is to assess the impact of marketing activities on store traffic for a service provider. Most studies examined the impact of promotions on traffic in a supermarket setting using transaction data. In contrast, we investigate these questions using actual store traffic, which we collected using an advanced video technology that accurately measures head count, overall and by gender. In addition, our unique data encompass all the marketing activities the sponsoring firm uses to drive store traffic (e.g., flyers, radio, outdoor, print, TV, billboard, and window advertising). We propose a set of hypotheses based on the economics of information and promotional literature. Our results show that except for radio advertising, all promotional activities have a positive impact on sales. Moreover, the depth and breadth of promotional activities have a positive and significant effect on female store traffic but did not significantly affect the traffic of male shoppers. This is in contrast to print ads, which have a positive and significant effect on male store traffic but no effect on female traffic. We conclude the paper by discussing managerial implications for retailers as well as suggesting directions for future research. 2 - How Surprisingly Little Thoughts Count—On Receiver’s Motivated Appreciation for Giver’s Thoughts Yan Zhang, National University of Singapore Business School, 15 Kent Ridge Drive, Singapore, Singapore, [email protected] Gift-giving is a social exchange that includes both the objective value of a gift as well as the symbolic meaning of the exchange itself. The objective value of a gift is sometimes considered to be of secondary importance in people’s evaluations of a gift, as when people claim, “it’s the thought that counts.” Because it often takes motivation and attentional resources to consider another person’s thoughts, we predicted that thoughts would count for very little in evaluating gift exchanges unless gift receivers were motivated or otherwise triggered to consider a gift giver’s thoughts. Three experiments demonstrate that others’ thoughts are likely to be triggered when a friend’s gift has relatively little objective value, or is considered to be objectively undesirable. Gift givers, however, are directly aware of the amount of thought they put into their gift, and therefore predict that their thoughts will “count” more than they actually do. The fourth experiment found that although thought counts very little in most cases, investing thoughts into a gift made givers feel more socially connected with receivers, which may help maintain and develop the relationship between givers and receivers. 4 - An Empirical Investigation of Retailer Pass-throughs Across Categories Dinesh Gauri, Syracuse University, 721 University Ave., Syracuse, NY, 13214, United States of America, [email protected], Debabrata Talukdar, Joseph Pancras Retail pass-through has been extensively analyzed analytically and empirically, and recent empirical work has stressed the importance of appropriate methodology and data for inferring correct retail pass-through. However the literature on retail passthrough has interpreted ‘pass-through’ as being confined to a specific product category, and to brands belonging to this category. This category restriction has been derived from a tradition of modeling the retailer as a ‘category profit maximizer’. Yet it is widely accepted that retailers strive to maximize profits across categories, with several categories specifically functioning as ‘loss leaders’. In this paper we argue that this view of the retailer makes it necessary to reevaluate retailer pass-through from being a ‘within category’ phenomenon to a ‘cross category phenomenon’. Using a unique dataset we empirically evaluate cross pass-throughs with a variety of categories – selected on the basis of profitability. We find that by and large loss leader categories have negative cross category pass-throughs (average 21.5%) while profitable categories have positive cross category pass-throughs (average 21.98%). We also find empirical evidence of the existence of significant cross-category brand level pass-throughs. Category characteristics such as price elasticity and proportion of loss leaders affect the cross-category pass-throughs. We conclude that future work on retailer pass-throughs needs to incorporate cross category analysis in order to capture the ‘true’ strategic behavior of the retailer. 3 - When Looks Can Be Deceptive: Consumer Response to Unfamiliar Product Packaging Descriptors Rishtee Batra, Assistant Professor, Indian School of Business, AC2 Level 1 Office 2116, Indian School of Business- Gachibowli, Hyderabad, AP, 500032, India, [email protected] Marketing scholars are increasingly paying attention to a product’s packaging, often described as a silent salesman that influences consumer decisions. A product’s package informs consumers about many different aspects of product, including the quantity of the goods contained inside. The current paper investigates consumers’ responses to packaging containing unfamiliar measurement units and argues that consumers will anchor on the face value of the measurement unit in arriving at their perceptions of product quality and willingness to pay. Study 1 hypothesizes that when a consumer encounters packaging information that is presented in a foreign unit whose face value is higher than the equivalent in their home units (e.g. an American consumer evaluating a package containing 454 grams of a product versus 1 pound), that consumer will have an overall higher perception of the product quality and will have a higher willingness to pay due to inaccuracies in perceived quantity. Study 2 hypothesizes that the over- or under-estimation of overall product quality and willingness to pay will be moderated by the extent to which consumers can access a similar packaging from their memory. In the case that the product’s packaging is typical of the category, the results of Study 1 are expected to be mitigated. Although consumers might have difficulty interpreting foreign units, they still have a visual anchor to which they can compare the given product and make more accurate judgments. If, however, the packaging is novel, the effect is expected to endure because the consumer has no point of comparison. 6 MARKETING SCIENCE CONFERENCE – 2011 4 - Understanding Customers` Substitution Patterns when Branded Items Become Unavailable Jana Diels, Humboldt-Universitaat zu Berlin, Wirtschaftswissenschaftliche Fakultaat, Spandauer Str. 1, Berlin, Germany, [email protected], Lutz Hildebrandt, Nicole Wiebach TA12 targeted direct marketing. For instance, in customer relationship management, marketers may predict future customer value, but which actions they should select for which customers and in which contexts or channels? To provide new insight on this problem, we use a model-based algorithm to determine an optimal policy of how to run a sequence of tests to learn about the effectiveness of those actions in the most cost-efficient way. To do this, we introduce the reinforcement learning framework to the marketing literature. This approach generalizes methods commonly used in marketing to solve dynamic optimization problems. Unlike prior methods, our approach accommodates a selection among many different actions. By considering the similarity of those actions, the firm may learn about the effectiveness of action without before ever testing it. As one example, we investigate firms with a customerbase engaging in activity for free with the goal of converting some of them into paying customers. Finally, we discuss the design of field experiments for implementation of the optimal sequential marketing tests. At the point-of-sale customers are often faced with unexpected situations of reduced choice sets, e.g. caused by delistings or temporary out-of-stocks. Accordingly, it is of major importance for retailers and manufacturers to gain insight into individual substitution patterns if choice sets are reduced. Previous experimental research, predominantly directed to product introduction, has revealed that changes in the set of alternatives significantly affect customers` preferences and hence, their product choice. The objective of this research is to analyze whether the unavailability of an item induces comparable systematic shifts in choice probabilities as in the product entry case. Furthermore, a theoretical framework to understand, predict and use substitution patterns is derived by using context and phantom theory. In a series of experiments, we find strong support for the existence of certain context effects when items are permanently removed or temporarily out-of-stock. Thereby, the importance of context theory as major theoretical approach to explain switching patterns for product unavailability is approved. Moreover, implications for retailers when making assortment and inventory decisions can be derived. The results of a real-life quasiexperiment further suggest that manufacturers may encounter substantially larger losses than retailers when items are delisted from the assortment. With regard to temporal unavailability, the outcome of an online experiment extends existing outof-stock research by (1) demonstrating that a negative similarity effect arises in outof-stock situations, (2) highlighting the relevance of promotion as an essential driver for out-of-stock reactions, and (3) including a so far neglected option – branch switching. ■ TA12 Champions Center II Branding Contributed Session Chair: Xiaoying Zheng, PhD Student, Guanghua School of Management, Peking University, Zhong Guan Xin Yuan No.4 Building, Beijing, 100871, China, [email protected] 1 - Customer Based Multidimensional Brand Equity and Asymmetric Risk Kyoung Nam Ha, Doctoral Candidate, University of Washington, 24003 SE 12th Pl, Sammamish, WA, 98075, United States of America, [email protected], Robert Jacobson, Gary Erickson ■ TA11 Based on market asset theory, we investigate brand asset in influencing a firm’s performance, in particular, a firm’s risk. Building on a literature in marketing that suggests several components in constituting brand value and in finance that emphasizes the role of asymmetries in the systematic risk, we assess the extent to which dimensions of brand equity (Differentiation, Relevance, Esteem, Knowledge, and Energy) influence the downside risk, upside risk, and the differential between upside and downside risk. We find that (i) Esteem has a negative effect on the risk differential, which comes from a more pronounced positive effect on downside risk, and (ii) Energy has a positive effect on the risk differential, which comes from a more pronounced positive effect on upside risk. We also find that (iii) Knowledge has a negative effect on downside risk as well as upside risk, and, thus, it does not have a statistically significant effect on the risk differential. The empirical analysis highlights the limitations of working with an aggregate measure of beta (which aggregates over both upturns and downturns) as well as an aggregate measure of brand equity (which aggregates distinct brand equity dimensions). By showing which brand characteristics indeed affect a firm’s risk beyond simply proving the theory in marketing that brand equity will have effect on a firm’s performance, this study provides managers more pragmatic information and sophisticated insights in making strategic decisions. Champions Center I Direct Marketing Contributed Session Chair: Eric Schwartz, The Wharton School, 3730 Walnut Street, JMHH 700, Philadelphia, PA, 19103, United States of America, [email protected] 1 - Calibration? Definition, Motivation and Insights Learned from a Direct Marketing Setting Kristof Coussement, IESEG School of Management, 3 Rue de la Digue, Lille, 59000, France, [email protected], Wouter Buckinx Calibration refers to the adjustment of the posterior probabilities output by a classification algorithm towards the true prior probability distribution of the target classes. This adjustment is necessary to account for the difference in prior distributions between the training set and the test set. This article proposes a new calibration method, called the probability-mapping approach. Two types of mapping are proposed: linear and non-linear probability mapping. These new calibration techniques are applied to 9 real-life direct marketing datasets. The newly-proposed techniques are compared with the original, non-calibrated posterior probabilities and the adjusted posterior probabilities obtained using the rescaling algorithm of Saerens, Latinne, & Decaestecker (2002). The results recommend that marketing researchers must calibrate the posterior probabilities obtained from the classifier. Moreover, it is shown that using a ‘simple’ rescaling algorithm is not a first and workable solution, because the results suggest applying the newly-proposed non-linear probabilitymapping approach for best calibration performance. 2 - Brand Equity and Product Recalls Sheila Goins, Assistant Professor, University of Iowa, S320 Pappajohn Business Building, Iowa City, IA, 52242-0944, United States of America, [email protected], Qiang Fei, Lopo Rego, Cathy Cole Marketing managers and scholars have long recognized brands as fundamental market based firm assets. The marketing literature provides compelling theoretical rationale and empirical evidence linking strong brands with competitive advantages, customer commitment and decreased price sensitivity, thus contributing to firm performance. However, little is known about how customer brand perceptions influence customer response and firm performance in light of negative events such as product recalls. The literature offers competing views with both “customer brand commitment insulate the company,” and “the bigger they are, the harder they fall” arguments being offered. Using a multi-method approach and customer-based brand equity metrics, we explore how customer brand perceptions influence the firm’s product and financial market performance following a recall announcement. We empirically test the product market using experimental data, while the financial market is tested via an event study method. We also examine competitive dynamics in product recalls. Our experimental findings reveal brand quality as critical to recall effects: lower quality recalled brands exhibit smaller decreases in brand perceptions than higher quality recalled brands. We also identify asymmetric competitive dynamics: recalling a high quality brand causes positive increases in lower quality brand evaluations, while high quality brand evaluations do not change when a low quality brand is recalled. Findings from the event study confirm the relevance of customer perceptions: firms with strong brands experience less negative abnormal returns in response to brand recalls. Our findings also indicate competitive asymmetries and distinct time horizons for shareholder responses for focal and competitor firms. 2 - Optimizing Target Selection of Direct Mailing by Charities Remco Prins, VU University Amsterdam, FEWEB Marketing - Office 2E-19, De Boelelaan 1105, Amsterdam, 1081 HV, Netherlands, [email protected], Bas Donkers When determining the effectiveness of existing target selection procedures in direct mailing campaigns, correcting for the resulting endogenous sample composition can be a complicated procedure. In the present study, we propose a method to determine the effectiveness of target selection procedures using an additional experimental mailing to a randomly selected group of customers. Through this field experiment, we can provide clear cut directions for improving target selection rules, without the need for explicit corrections for the impact of previous target selection activities. The proposed approach also easily allows an estimation of the net effect of sending an additional mailing, after correcting for possible cannibalization effects. We apply this method to the direct mail campaigns of four large Dutch charitable organizations. The results provide directions for improvement for each of the charities, in terms of selection rules and mail frequency. 3 - Test and Learn: A Reinforcement Learning Perspective Eric Schwartz, The Wharton School, 3730 Walnut Street, JMHH 700, Philadelphia, PA, 19103, United States of America, [email protected] Firms run experiments to test and learn about marketing actions. However, running many so-called A/B or multivariate tests can be slow and costly, and the results from one test do not give necessarily inform what the next test should be. We frame this as an optimization problem, known as the multi-armed bandit problem, involving the tradeoff between exploring uncertain actions and exploiting what has been learned so far. The problem occurs in a range of domains from personalized website design to 7 TA13 MARKETING SCIENCE CONFERENCE – 2011 3 - Factors Enhancing a Brand’s Competitive Clout: A Two-step Empirical Analysis Juan Carlos Gázquez-Abad, Asistant Professor of Marketing, University of Almería, Department of Management & Business Adm., Ctra. Sacramento s/n. Cañada San Urbano, Almería, 04120, Spain, [email protected], Rubén Huertas-Garcìa, Francisco J. Martínez-Lopez, Agustí Casas-Romeo These and other findings offer important managerial implications for innovators on how to efficiently form strategic alliances and increase new product success in markets with network effects and standards competition. 2 - The Dynamic Effects of Service Recovery Strategies on Customer Satisfaction Xueming Luo, Eunice & James L. West Distinguished Professor of Marketing, The University of Texas at Arlington, Arlington, TX, United States of America, [email protected], Fang Zheng The market power (or ‘competitive clout’) of a brand is an increasingly important component of modern marketing strategies. However, the factors that enhance a brand’s competitive clout (BCC) are poorly understood. This study therefore suggests an integrated model of BCC and three factors that are proposed to play a role in its formation: (i) consumer price sensitivity; (ii) brand market share; and (iii) consumer brand preferences. These variables are examined both individually and simultaneously to demonstrate the direct effect of each on BCC and how the interrelationships among them contribute to BCC. In doing so, a two-step empirical analysis is conducted. First, a multi-nomial logit model provides an own- and crossprice response matrix for a chosen set of competitive brands. Secondly, BCC is regressed against the variables of market share, intrinsic preferences, and price sensitivity using an interaction effects regression model. The results of the analysis show that market share is not the only way to increase BCC; in particular, consumer preferences are shown to play a key role in developing a strong brand. Even though the existing literature has ample studies on the types service recovery efforts and their impact on customer evaluations, little research attention has been paid to how dynamic the effects are when recouping customer satisfaction. We evaluate dynamic effects in terms of the long tail or short tail, high or low buildup, as well as quick or slow peak point timing. Different from previous studies using perceptual soft data with survey or experiments, this paper utilizes company archival hard data in a field study. The setting of our dataset is interesting because it is related to a major Chinese telecom company and how this company tries to pull back its customer satisfaction after a major service failure caused by the deadly earthquake in 2008. The developed time-series model account for the endogenous problems due to reverse causality, full interactions among endogenous variables, autoregression effects, market competition, and many alternative explanations. This research is crucial for managers to pulse when and what rescue initiatives can be considered a success in combating customer satisfaction loss. 4 - The Impact of Marketing Capability on Customer Responses: A Customer-based Brand Equity Perspective Xiaoying Zheng, PhD Student, Guanghua School of Management, Peking University, Zhong Guan Xin Yuan No.4 Building, Beijing, 100871, China, [email protected], Yi Xie, Siqing Peng 3 - The Information Content of Marketing Investments: The Case of Sales Force Resizing Announcements Anne T. Coughlan, John L. & Helen Kellogg Professor of Marketing, Northwestern University, 2001 Sheridan Road, Evanston, IL, United States of America, [email protected], Joseph Kissan, M. Babajide Wintoki A large number of studies have shown the significant role of marketing capability in attaining corporate competitive advantage. However, we know few about how marketing capability influences customer behaviors. Drawing upon customer-based brand equity theory, this study proposes a model regarding brand equity, which is conceptualized as a multidimensional construct of brand familiarity, brand reputation and brand trust, as the key factor connecting marketing capability and customer behavioral intentions. Two distinct types of marketing capability, product development capability and marketing communication capability, are investigated due to their importance for brand performance and accessibility to consumers. Data was collected through a face-to-face survey at three major railway stations in Beijing city in the context of electronic product industry. Structural equation modeling (SEM) was employed to conduct data analysis. Results indicate that two categories of marketing capability have varying impact on brand equity and influence customer responses through different brand equity components. Specifically, product development capability has positive influence on all the three brand equity components with its greatest impact on brand trust, whereas marketing communication capability has positive relationship with brand familiarity and brand reputation but not with brand trust. Brand reputation is directly related with brand trust, while brand familiarity does not necessarily lead to brand trust. Finally, all the three brand equity components can improve customer behavioral intentions. In sum, results suggest that a firm’s efforts on marketing capability can enhance brand equity through distinct routes, and subsequently facilitate favorable customer responses. This paper is concerned with the ability of marketing investments to convey information from the “black box” of the firm to participants in the financial markets. We focus on the context of sales force resizing announcements in the pharmaceutical industry because the sales force comprises the main marketing instrument for firms in this industry. We hypothesize that sales force resizing announcements will generate a concomitant market reaction because they provide information about future demand. Moreover, we predict stronger absolute effects for “increase” announcements than for “decrease” announcements. This is because the mean-shifting and uncertainty reducing aspects of new information work in concert for increase announcements, but in opposing ways for decrease announcements. Our theory also predicts a more pronounced market reaction when investors are more uncertain about the firm’s future performance and the announcement contradicts the direction of investors’ prior beliefs. Employing an event-study analysis, we find an unusually strong twoday market reaction of 3.2% for increase announcements and no significant market reaction for decrease announcements. The hypothesized effects with respect to uncertainty and direction of investors’ prior beliefs are also supported. We conclude by discussing implications for managing the investor relations function. 4 - Panel Discussion: Criticisms and Future Research Opportunities for Marketing’s Profit Impact Moderator: Dominique Hanssens, University of California-Los Angeles, 110 Westwood Plaza, Los Angeles, CA, 90095, United States of America, [email protected], Panelists: Gerard J. Tellis, David Reibstein ■ TA13 Champions Center III Both scholars and practitioners have long been concerned about the accountability of marketing expenditures. To rebuild confidence in marketing, researchers should analyze marketing’s impact on stock prices in a new paradigm of assessing the value of marketing activities. However, the burgeoning field is not without criticism. First, there is concern that prior studies are not rigorous enough in terms of econometric modeling. Thus, this session will organize some modelers to share their sights on how to deal with endogeneity, reserve causality, heterogeneity, and other issues. Second, some criticize that there is little substance on the underlying processes and boundary conditions for the marketing-finance interface. As such, the panel members will present some insights into future research topics and build a more comprehensive research agenda on the importance of marketing in the financially oriented reality world. Quantifying the Profit Impact of Marketing I Cluster: Special Sessions Invited Session Chair: Xueming Luo, Eunice & James L. West Distinguished Professor of Marketing, The University of Texas at Arlington, Arlington, TX, United States of America, [email protected] 1 - The Impact of Strategic Alliance on the Innovator’s Financial Value in Markets with Network Effects and Standard Competition Qi Wang, Binghamton University, Binghamton, Binghamton, NY, United States of America, [email protected], Jinhong Xie, Ashwin Malshe Innovating firms in markets with network effects and standards competition face significantly higher risk due to the challenge of attracting buyers at start up and the possibility of a “winner take all” market outcome going to a competitor. In contrast to previous research that has addressed these challenges from the perspective of marketing performance (e.g., increasing a product’s user base), this study examines how an innovator in such markets can increase its financial value by forming strategic alliances, both within the industry (i.e., with other manufactures) and across industries (e.g., with its suppliers, complementary product producers, and distributors). Based on data collected in the high-definition DVD player market, our empirical analyses find general support for a positive impact of strategic alliance announcements on the innovator’s financial value. More interestingly, our results reveal that not all strategic alliances are equally valuable to the innovator. Rather, (i) an alliance with producers of complementary products creates a higher value than an alliance with manufactures producing the same product and (ii) an alliance with downstream channel members such as distributors and retailers creates more value than an alliance with upstream channel members such as R&D and engineering. Furthermore, we illustrate the different dynamic pattern of the impact by a horizontal strategic alliance (with partners in the same or complementary industry) and by a vertical strategic alliance (with upstream or downstream channel partners). ■ TA14 Champions Center VI Customers’ Willingness-to-pay Research Contributed Session Chair: Neil Biehn, Senior Director, PROS Pricing, 3100 Main Street, Houston, TX, 77002, United States of America, [email protected] 1 - Exploring Consumer Heterogeneity with Respect to Seasonal Shifts of Demand Ali Umut Guler, PhD Student, London Business School, Regent’s Park, London, NW14SA, United Kingdom, [email protected] In this research, I provide an alternative explanation for why the prices of seasonal products might be falling during their high demand periods. This phenomenon of countercyclical pricing is well documented in the economics literature. It goes against intuition, as basic economic theory predicts a price increase when there is an outward 8 MARKETING SCIENCE CONFERENCE – 2011 TA15 ■ TA15 shift in the demand curve. Most of the explanations proposed for this phenomenon are based on competitive interactions among firms. I focus on consumer-level insights that could provide a rationale for a countercyclical pricing scheme. I show theoretically that the optimal pricing scheme for a seasonal product could be countercyclical due to consumer heterogeneity with respect to seasonal shifts in demand. The monopolist then makes use of this heterogeneity to sell only to the higher valuation segment during low season. I also provide empirical support for this theory using two different datasets of canned soup sales. Using a store-level aggregate sales dataset, I provide evidence against alternative explanations and I document that the demand appears to become more price inelastic during high season. I show that this finding can be explained by the existence of a break in the demand curve due to presence of different consumer segments. On a different dataset of consumer-level purchases, I show that the fit of a purchase incidence model improves when it allows for consumer segments that differ in their valuations across seasons. The estimates obtained for segment specific product valuations are in line with the predictions of the proposed model. Champions Center V B2B: Relationships Contributed Session Chair: Alfred Zerres, Institute of Business-to-Business Marketing, University of Muenster, Am Stadtgraben 13-15, Münster, 48143, Germany, [email protected] 1 - A Theory of Bargaining Costs and Price Terms in the Absence of Relationship-specific Investments Desmond (Ho-Fu) Lo, Assistant Professor, Santa Clara University, 500 El Camino Real, Santa Clara, CA, 95050, United States of America, [email protected], Giorgio Zanarone Firms often bring pre-existing assets such as brand and market strength into joint business ventures. We model a bargaining process and price terms in these ventures as a tension between saving ex ante contract design costs and facing ex post opportunism, in the shadow of judicial behavior. On the one hand, to save on design costs, parties can leave price open for future negotiations and face ex post dilution in their pre-existing assets through wasteful bargaining actions. On the other hand, parties can go through the effort of fixing the contract price ex ante and thus prevent opportunistic renegotiations through a credible threat of having the contract reinstated by courts. We show that firms prefer to specify the contract terms ex ante when the value of their pre-existing assets is high and the environment is not complex. Our theory adds to Transaction Cost Economics (TCE) by (1) formalizing a bargaining process that is observed in the real world and its effect, together with the judicial principle of contract reinstatement, on price terms and (2) extending the TCE rationale to settings where the assets brought into production are not relationshipspecific. 2 - A Structural Model for a “Name Your Own Price” Mechanism with a Fixed Price Option Kerem Yener Toklu, Rice University, 6100 Main St, Houston, TX, United States of America, [email protected] This paper provides an empirical analysis for a version of the “Name Your Own Price” mechanism used in an online marketplace where buyers also have the option to purchase the item at a fixed price determined by the seller. This particular sale type with the additional fixed price feature has not been paid much attention in the literature. Structural econometric techniques are employed to estimate the distribution of buyers’ valuations which is then used to find the optimal fixed price for the seller and quantify the value of information. First, a decision theoretic model is used and the optimal buying strategy is characterized. A nonparametric method is then proposed to estimate the distribution of values. To generalize the model for different buying strategies a behavioral model is also considered under weaker assumptions and parametric estimation is used to capture a number of sale and item covariates. Calibrated models are then used to run counterfactual simulations to find revenue maximizing fixed prices for sellers. Aside from characterizing buyer behavior and optimal selling strategies, the paper also studies the information asymmetry between sellers and buyers, and provides a way to quantify the value of information. It is prevalent in online market places that the quality of the item being sold may not be reflected perfectly on the sale webpage which raises an information asymmetry between sellers and buyers. The uncertainty in the quality of the item directly influences the value of buyers, and knowing this, sellers reveal information using different channels on the webpage. Amount of this information is used as a variable and simulations are run to measure the effect of information asymmetry on sale prices and quantify the value of information for buyers. 2 - Assessment of Purchasing Maturity in Small Business Jeffery Adams, Assistant Professor, University of Houston-Downtown, College of Business, 320 N Main St, Houston, TX, 77002, United States of America, [email protected], Ralph Kauffman Relatively little research has been done on B-to-B buyer-supplier relationships in small and medium size business, and perhaps even less has been done on how to assess the level of purchasing development in such businesses. Successful marketing relationships with these smaller firms can hinge on the ability of marketers to assess the degree of purchasing maturity in such customers. “Purchasing maturity” can be defined as the readiness and ability of customer firms to engage in more highly developed complex and longer-term buyer-seller relationships. The primary objective of this study is to develop a means that marketers can use to measure purchasing maturity in particular small and medium size business customers or potential customers. Using data collected from a cross-sectional sample of small and medium size businesses in four NAICS sub-sectors, a measurement model is developed. Factor analysis, Pearson correlation, and stepwise regression are used in development of the model to identify factors that, to the degree that they are present, give indication of the level or degree of purchasing maturity in a business. A scorecard is developed for application of the model to individual firms. The scorecard can be used by both marketers and customer management to assess the level of purchasing maturity that exists in the buying firm. The primary contribution of this research is the development of an additional tool for B-to-B marketers to assess the readiness of existing or potential small and medium size customers to engage in higher-level buyer-seller relationships and thereby increase the seller’s market potential. 3 - Framing Effects and Consumers’ Reactions to Corporate Social Responsibility Carmelo J. Leon, Department of Applied Economic Analysis, University of Las Palmas de Gran Canaria, C/Juan de Quesada, n∫ 30, Las Palmas de Gran Canaria, Spain, [email protected], Jorge Araña, Christine Eckert Corporate social responsibility (CSR) might raise different reactions across consumers depending on the way they are framed in their marketing and selling process. In this paper we focus on the effects of alternative frames for CSR measures. Whereas previous research only focused on single evaluations of each CSR policy separately and thus ignored potential trade-offs consumers face when deciding between different CSR-policies companies offer, we utilize discrete choice experiments (DCE) that more realistically picture consumers’ every-day choices. Subjects were presented with alternative profiles of CSR measures and were divided into two treatments that varied according to the definition of the status quo alternative and the valuation question. The design of the experiments allowed us to compare the willingness to pay (WTP) a higher price for a product undertaking CSR measures and the willingness to accept (WTA) a lower price for a product with a lower CSR profile. By disentangling the effects of differences in choice uncertainty (i.e. choice scale) and preferences between the two scenarios, we are able to show that the WTA/WTP gap exists only for some CSR policies but doesn’t show up for others. Our results further demonstrate that measures for WTA and WTP are significantly more dispersed in choice experiments with joint evaluations of different CSR policies than in single evaluation – which is likely to be caused by the availability of reference alternatives consumers use to form preferences in the former situation. 3 - Integrative Negotiation Training: Enduring Effects of Asymmetrical and Symmetrical Training Alfred Zerres, Institute of Business-to-Business Marketing, University of Muenster, Am Stadtgraben 13-15, Münster, 48143, Germany, [email protected], Joachim Hüffmeier, Alexander Freund, Klaus Backhaus Almost every business deal in practice is shaped by negotiations as a dominant transaction mechanism (Anderson/Narus/Narayandas, 2009; Thompson, 2009). Thus, negotiation skills are crucial to ensure successful transactions for every business marketing manager and companies consequentially try to improve their marketing managers’ negotiation skills through respective training programs. As a consequence, negotiation training has become a multi-billion dollar industry alone in the US. But are these trainings worth their money? To address this question, we focus on two gaps in the negotiation training literature. While research has shown that negotiation training can successfully increase negotiation outcomes immediately after the training, little is known about its effectiveness across time. Furthermore, in all previous studies both conflicting parties were trained. From an applied perspective, this may well be a problematic feature of the literature, as firms can hardly train both parties in business negotiations. To address the related open questions, we conducted an integrative negotiation training experiment with a total of 360 participants (180 dyads). We used a one-factorial design with training (no training, only seller trained, only buyer trained, and both parties trained) as between- and three measurement times as within-subjects factor (assessments before, immediately after, and 30 days after the training). Results reveal that integrative negotiation training is not only effective, but even increases in periods of non-practice. However, one-party negotiation training was only successful if the trained party was the seller and failed if the buyer received the training. This results pattern is discussed with respect to the broader negotiation literature. 4 - Pricing and Willingness-to-pay Estimation in B2B Markets Neil Biehn, Senior Director, PROS Pricing, 3100 Main Street, Houston, TX, 77002, United States of America, [email protected] There has been considerable research, as well as the application of various marketing techniques, to capture consumers’ Willingness-to-Pay. In this talk, we discuss why many of the common techniques are difficult to implement in a B2B environment. We will also explore a more data driven approach to capturing the Willingness-to-Pay of a B2B customer and present results from its implementation. 9 TB01 MARKETING SCIENCE CONFERENCE – 2011 Thursday, 10:30am - 12:00pm dimension), marketing metrics need to bridge the attitudinal domain and the financial domain. As an illustration, consumer preference for a brand is an attitudinal metric that ultimately needs to be translated in monetary terms. Second, I propose that metrics should be forward looking. For example, market share may be a useful metric of past achievement, but what does it mean for a brand’s expected future performance? Third, metrics need to be standardized, so that comparisons can be made across industries and across firms in any sector. I will review several recent research contributions that have made substantial progress in these directions. ■ TB01 Legends Ballroom I Marketing Science Institute II Cluster: Special Sessions Invited Session ■ TB02 Chair: Don Lehmann, Columbia University, Columbia Business School, New York, NY, United States of America, [email protected] 1 - Research on Branding: Issues and Outlook Jan-Benedict Steenkamp, C. Knox Massey Distinguished Professor of Marketing, University of North Carolina at Chapel Hill, Kenan-Flagler Business School, Chapel Hill, NC, 27599, United States of America, [email protected] Legends Ballroom II Empirical Modeling Contributed Session Chair: Eric Schwartz, The Wharton School, 3730 Walnut Street, JMHH 700, Philadelphia, PA, 19103, United States of America, [email protected] 1 - Modeling Online Visitation and Conversion Dynamics Chang Hee Park, Cornell University, Ithaca, NY 14853, United States of America, [email protected], Young-Hoon Park Branding is clearly at the core of marketing. In fact, it is difficult to envisage marketing activities that do not directly or indirectly affect the firm’s brands. Therefore, it is not surprising that for the last quarter century, branding has been a research priority of the Marketing Science Institute, where the specific aspects of branding that were singled out for meriting special research attention reflected the dynamics in the marketplace. A milestone in branding research was the 1988 MSI working paper “Defining, Measuring, and Managing Brand Equity,” written by Lance Leuthesser. MSI-sponsored research on brand equity changed the way we think about brands, and was recognized with multiple awards. In the current list of research priorities, MSI calls for research on managing brands in a marketplace that is transformed by competitive threats emanating from (1) brands that originate in emerging markets, (2) store brands, and (3) economic downturns. In this presentation, I will do three things. First, I will provide a concise overview of past branding research, with special emphasis on MSI research priorities and publications. Second, I will provide a reflection on past research. What are some key learnings? What are strengths and limitations of previous research efforts? Going back in history is done too rarely in our field. To quote philosopher George Santayana: “Those who cannot remember the past are condemned to repeat it.” Finally, I will outline some issues in branding research that I believe are in urgent need of future research, and relate them to MSI’s current branding priorities. This paper develops a multi-event timing model that captures the lumpy shopping patterns of online customers and infers the formation of latent visit clusters constituting the arrival processes. Because the start and the end of each visit cluster are unobserved, we employ a changepoint modeling framework and statistically infer the cluster formation on the basis of customer visit patterns through data augmentation in Bayesian approach. Our model provides a set of novel inferences about the patterns underlying online shopping behavior, including (1) the number of visits constituting a visit cluster, (2) the arrival rate within a visit cluster, (3) the time length of a visit cluster, (4) the number of visit clusters in a given time period, and (5) the arrival rate between visit clusters, at the individual customer level. We apply our modeling framework to the data of customer visits and purchases at VictoriasSecret.com. The main results suggest strong empirical evidence of lumpy shopping patterns by online customers with significant heterogeneity in the extent of the lumpiness. As a result, the proposed model offers excellent fit and predictive performance. By linking customer visits to purchases, we find that on average a visit cluster with purchase conversions contains more visits than one without purchase, the likelihood of purchase conversions is greatest for a customer’s second arrival in a visit cluster, and a considerable portion of purchase conversions occur in later visits in clusters. We also demonstrate how our cluster-based inferences can be utilized for attribution management in online marketing. 2 - Research on Brands: Latest Findings and Future Opportunities Marc Fischer, Universitaat Passau, Passau, D-94030, Germany, [email protected] A brand is an important asset for many firms that contributes to sustainable superior performance. Unsurprisingly, researchers from different schools of thought have been dealing with brand related questions for many years. Although the field has produced many papers and research reports, some more and some less influential, there is still vital interest in brand research. A number of questions with high practical relevance have not been answered, yet. This presentation summarizes the latest findings on brands and brand management. Because even the latest research findings are so numerous, the presentation focuses on those results with special relevance for practitioners. Particularly, the author covers findings that help explain why brands are important to customers, which is a prerequisite for gaining management relevance. In addition, he summarizes latest findings on the relevance of brands to firms and their stakeholders. The review includes important questions of how to integrate brands into the financial management and controlling system of firms. The presentation finishes with an outlook on future priorities in brand research. The author identifies several underresearched areas and questions that have not been anwered, yet. Again, the suggested agenda for future research focuses on topics that are not only interesting from an academic standpoint but also have practical relevance. 2 - Speed of Product Updates in Online Games Paulo Albuquerque, Assistant Professor, University of Rochester, Simon Graduate Business School, Carol Simon Hall 3-110 R, Rochester, NY, 14627, [email protected] This paper proposes an empirical model studying the managerial decision regarding the speed of product innovation in the online computer gaming industry. The demand for game content is modeled as a function of consumer forward looking expectations and response to new content, as well as their previous participation in the game. The launch of new content involves a trade-off. Consumers value content innovation, which leads to more participation, but faster product updates increase firm costs and make old content obsolete, which may lead to lower product usage as well. We use individual-level data collected online from the popular computer game “World of Warcraft” to empirically test our model, and provide recommendations about the scheduling of product updates. 3 - Estimating Preferences from Configured Choice Sanjog Misra, William E. Simon School of Business, University of Rochester, Rochester, NY, 14627, United States of America, [email protected] 3 - The Evolution of Research on Marketing Metrics Dominique Hanssens, University of California-Los Angeles, 110 Westwood Plaza, Los Angeles, CA, 90095, United States of America, [email protected] The traditional choice framework imagines consumers picking from a small set of well defined alternatives. The current online choice environment, on the other hand, allows consumers to be active participants in both the supply and demand aspects of choice by giving them configuration tools to help them design their own optimal products. In this paper I describe, model and estimate preferences from data that is generated by product configuration tools. The paper generalizes exiting methods and approaches by explicitly modeling interactions across design elements and allowing for consumer heterogeneity in them. A novel Bayesian econometric algorithm is presented that allows estimation under configured choice regimes that may often be characterized by intractable likelihoods. I demonstrate the use of this methodology first using simple (toy) examples and then with real data. The paper concludes with a discussion of extensions and other potential applications. “What you can measure, you can manage”, is a popular saying that applies particularly well to the marketing discipline and that has motivated the Marketing Science Institute to invest substantial research and dissemination resources in marketing metrics over the past few decades. By the 1950s, marketing had established an effective vocabulary, with root concepts such as segmentation, the four P’s and the product life cycle. It then naturally turned to the definition of relevant metrics to quantify these and other constructs. Among the early contributors was the PIMS initiative (e.g. Buzzell, Gale & Sultan 1975) which, among other things, highlighted the importance of market share, advertising/sales ratios, return to marketing investment and the like. Since then, a substantial number of new metrics have been advanced, covering marketing assets (such as brand equity), marketing activities (such as innovation activity) and marketing outcomes (such as sales revenue and firm value), see Lehmann and Reibstein (2005) for a review. That trend has arguably led to “metricflation”, with some publications listing as many as 103 key marketing metrics (Davis 2007). Perhaps as a defensive reaction to this proliferation, senior management has embraced “metric simplification”, all the way down to “the one number you need to know”, Net Promoter Score (Reichheld 2003). Where do we go from here ? In my view, we need to end up somewhere between “1” and “103”, and probably closer to the former. To do so, we will want to formulate key conditions for metrics to be declared strategically relevant. I propose three such critera. First, unlike finance, whose metrics are largely unidimensional (in the monetary 4 - Test and Learn: A Reinforcement Learning Perspective Eric Schwartz, The Wharton School, 3730 Walnut Street, JMHH 700, Philadelphia, PA, 19103, United States of America, [email protected] Firms run experiments to test and learn about marketing actions. However, running many so-called A/B or multivariate tests can be slow and costly, and the results from one test do not give necessarily inform what the next test should be. We frame this as an optimization problem, known as the multi-armed bandit problem, involving the tradeoff between exploring uncertain actions and exploiting what has been learned so far. The problem occurs in a range of domains from personalized website design to targeted direct marketing. For instance, in customer relationship management, 10 MARKETING SCIENCE CONFERENCE – 2011 TB04 determinants of actual social influence that is exerted over others. In this study we analyze determinants of social influence on adoption within the ego network of a customer while we control for traditional variables that are known to affect adoption behavior. We use three sources of data: 1) communication-based network data to determine the ego network of adopters of a mobile phone service, 2) personality traits and customer-firm relationship data of those adopters, obtained by an online survey, and 3) customer and customer-firm relationship data of those in the ego networks. We estimate a multilevel hazard model for the adoption behavior of the individuals in the ego networks of the initial adopters. This allows us to analyze the social influence that an initial adopter exerts over the others in his/her ego network and by which factors this influence is determined. We contribute to the customer management literature in two ways. First, we are among the first that empirically investigate determinants of actual social influence on adoption within a customer’s ego network; characteristics of the initial adopter, characteristics of the customer-firm relationship, and the attitude towards the service. Second, by showing that customers influence others in their ego network we illustrate that customer value management is truly enriched by a network oriented concept like customer engagement. marketers may predict future customer value, but which actions they should select for which customers and in which contexts or channels? To provide new insight on this problem, we use a model-based algorithm to determine an optimal policy of how to run a sequence of tests to learn about the effectiveness of those actions in the most cost-efficient way. To do this, we introduce the reinforcement learning framework to the marketing literature. This approach generalizes methods commonly used in marketing to solve dynamic optimization problems. Unlike prior methods, our approach accommodates a selection among many different actions. By considering the similarity of those actions, the firm may learn about the effectiveness of action without before ever testing it. As one example, we investigate firms with a customerbase engaging in activity for free with the goal of converting some of them into paying customers. Finally, we discuss the design of field experiments for implementation of the optimal sequential marketing tests. ■ TB03 Legends Ballroom III 4 - Customer Acquisition in a Connected World: Revenue vs. Opinion Leaders Michael Haenlein, Associate Professor, ESCP Europe, Paris, France, [email protected], Barak Libai Social Networks and Profitability Cluster: Special Sessions Invited Session A fundamental principle of informed customer acquisition is that firms should give priority to attracting customers that will supply the most value. In doing so, companies face a fundamental dilemma: On the one hand, firms today have an increasing ability to assess the lifetime value of their customers and to understand how it is distributed. This information can be used to assess which are the best potential customers to acquire and to focus on potential clients with high expected customer lifetime value (revenue leaders). On the other hand, customers provide the firm value not only through what they buy, but also in the way they affect others via social influence such as word of mouth. Firms might therefore be well advised to attract clients with a high number of social connections (opinion leaders), which have been shown to exert a disproportional effect on others. While the acquisition of revenue leaders results in higher direct value, the acquisition of opinion leaders leads to higher social value. Our study analyzes the tradeoff between focusing on the acquisition of higher lifetime value customers (revenue leaders) and higher social influence customers (opinion leaders). Using an agent-based model, we highlight the complexity of this trade off, esp. in situations where both sources of value are not independent, and show under which conditions focusing on revenue leaders can lead to higher value than focusing on opinion leaders. Chair: Michael Haenlein, Associate Professor, ESCP Europe, Paris, France, [email protected] Co-Chair: Barak Libai, Leon Recanati Graduate School of Business Administration, Tel Aviv, Israel, [email protected] 1 - How Customer Word of Mouth Affects the Benefits of New Product Exclusivity to Distributors Christophe Van den Bulte, Associate Professor, University of Pennsylvania, Philadelphia, PA, United States of America, [email protected], Renana Peres Marketing executives often face the decision whether and for how long to grant exclusivity to distributors of their new products. Using an agent based model, we assess how word of mouth among customers and the competition among structurally equivalent distributors with partially-overlapping customer bases influence the profitability of granting exclusivity for new products. Our results show that the presence of communication spillovers among customers of different distributors can make exclusivity undesirable for the distributors and the industry overall. This reversal of the conventional wisdom occurs because, though exclusivity protects the favored distributor from market share losses to competitors, it also precludes him from benefiting from the positive word of mouth generated by customers of other distributors. The effect is magnified by both the level of cross-distributor communication among customers and the level of structural equivalence among distributors. The forces at work and the result we obtain apply not only to exclusivity in distribution but also to selling to original equipment manufacturers (OEMs). ■ TB04 Legends Ballroom V Bayesian Econometrics II: Methods & Application Contributed Session 2 - Evolving Viral Marketing Strategies William Rand, University of Maryland, College Park, MD, United States of America, [email protected], Forrest Stonedahl Chair: Sudhir Voleti, Assistant Professor, Indian School of Business (ISB), 2118, AC2 L1, ISB Campus, Gachibowli, Hyderabad, AP, 50032, India, [email protected] 1 - Simultaneous Scaling of Multiple Domains: Application to Country-of-origin Effects in Asia Luming Wang, University of Manitoba, Asper School of Business, Winnipeg, Canada, [email protected], Giana Eckhardt, Terry Elrod Viral marketing is based on the idea that consumer discussions about a product are more powerful than traditional advertising. However, who to seed in a viral marketing campaign in order to maximize profit based on the amount and rate of product adoption is not obvious. Given an arbitrary network and a limited seeding budget choosing the optimal seeding strategy has been shown to be computationally intractable (NP-Hard). Furthermore, it is not clear what the proper seeding budget should be for a particular network since additional seeds cost more but also encourage more rapid adoption, and thus can directly affect the profitability of the campaign. In order to address these problems, we define a strategy space for consumer seeding based upon network characteristics. We measure strategy effectiveness by simulating adoption using a Bass-like agent-based model. We examine six different social network structures: four classic theoretical models (random, lattice, small-world, and preferential attachment) and two empirical datasets (extracted from Twitter friendship data and an alumni social networking website). We use an evolutionary algorithm to simultaneously optimize the seeding strategy and budget. Our results show that a simple strategy (ranking by node degree) is near-optimal for the four theoretical networks, but that a more nuanced strategy performs significantly better on the empirical networks. Moreover, we find that the exact same strategy appears to work well on different empirical networks even when collected from very different sources. Finally, when we look at all of the networks together, we find a correlation between the optimal seeding budget for a network (the number of individuals to seed), and the inequality of the degree distribution. Recent years have seen a proliferation of applications of market structure analysis, especially studies for inferring market structure from consumer preference and choice data. The analyst infers brand positions in an (intangible) attribute space from the data, given a market in which consumers have heterogeneous tastes for these attributes. Two critical issues have drawn the authors’ attention. First, most market structure analyses are two-mode (i.e., brands and consumers) on existing brands within a single product category. However, brands often are extended beyond their original categories to reduce the cost and risk of entering a new product category. Therefore, a macro-view market structure analysis across product categories on both served and unserved markets may provide more insight into the cross-category competition situation and suggest further strategic moves. Second, the interpretation of map dimensions is a two-step approach (i.e., generating dimensions first and then labeling them) with some limitations in term of data usage efficiency and measurement error management. An integrated method is preferable. The current research presents a probabilistic spatial model on partially overlapped domains (a) to provide a flexible approach to dealing with more complex market structures, (b) to both examine the served and explore the unserved marketplaces, and (c) to make the choice map self-explainable (by simultaneously scaling spaces with possibly different natures). The authors demonstrate the proposed method using country of origin as an application area due to its complexity in term of modes involved (e.g., consumer, country, product) and interrelated domains (e.g., the preference map and the perception map). 3 - Determinants of Social Influence on Adoption in Customer Ego Networks Hans Risselada, University of Groningen, Groningen, Netherlands, [email protected], P.C. (Peter) Verhoef, Tammo Bijmolt Traditionally, customer value management is concerned with individual customers analyzed in isolation. The recently introduced concept of customer engagement behavior broadens the scope of customer value management by capturing post transaction behavior of a customer that potentially influences the behavior of others in his/her ego network. However, in the literature little is known about the 11 TB05 MARKETING SCIENCE CONFERENCE – 2011 ■ TB05 2 - A Dynamic Spatial Hierarchical Model of Theater Level Box-office Performance Shyam Gopinath, Doctoral Student, Kellogg School of Management, 2001 Sheridan Road, Evanston, IL, United States of America, [email protected], Pradeep Chintagunta, Hedibert Lopes, Sriram Venkataraman Legends Ballroom VI New Product II: Diffusion Contributed Session Chair: Li Zheng, ESSEC, 7, Rue Du General Henrys, Esc. 6; Apt. 67, Paris, 75017, France, [email protected] 1 - Modeling Seasonality in New Product Diffusion Yuri Peers, Erasmus University Rotterdam, Erasmus School of Economics, P.O. Box 1738, Room H11-26, Rotterdam, 3000DR, Netherlands, [email protected], Philip Hans Franses, Dennis Fok Most of the prior research on the movie industry has focused on the box office performance at the movie level. The goal of this research is to investigate the drivers of theater level box-office performance. One key challenge for theaters is the competition faced by other theaters in the same region. We provide a novel modeling framework that takes into account this spatial relationship. Moreover, we allow this relationship to change over time. The first key objective of this research is to compare the relative importance of the different box-office drivers. In addition, we investigate whether there is seasonality in the impacts i.e., are some factors more important than others during certain times of the year like holidays (e.g., summer and Christmas). The second key objective is to develop a forecasting model and compare it with several benchmark models. The authors propose a method to include seasonality in any diffusion model that has a closed-form solution. The resulting diffusion model captures seasonality in a way that naturally matches the overall S-shaped pattern. The method assumes that additional sales at seasonal peaks are drawn from previous or future periods. This implies that the seasonal pattern does not influence the underlying diffusion pattern. The model is compared with alternative approaches through simulations and empirical examples. As alternatives we consider the standard Generalized Bass Model [GBM] and the basic Bass model, which ignores seasonality. One of our main findings is that modeling seasonality in a GBM generates good predictions, but gives biased estimates. In particular, the market potential parameter is underestimated. Ignoring seasonality, in cases where data of the entire diffusion period is available, gives unbiased parameter estimates in most relevant scenarios. However, when only part of the diffusion period is available, estimates and predictions become biased. We demonstrate that our model gives correct estimates and predictions even if the full diffusion process is not yet available. 3 - Attribute-level Heterogeneity Peter Ebbes, The Ohio State University, Fisher College of Business, 2100 Neil Ave, Columbus, OH, 43210, United States of America, [email protected], John Liechty, Rajdeep Grewal, Matthew Tibbits This research studies finite mixture specifications for modeling consumer heterogeneity where each regression coefficient has its own finite mixture, that is, an attribute finite mixture model. This attribute specification has two advantages over traditional vector finite mixture specifications, in which the finite mixture applies to the whole vector of regression coefficients. First, as our results demonstrate, the attribute finite mixture model can identify heterogeneity structures using less data. Second, the attribute model enables managers to get a better understanding of highly complex market structures. By allowing for different numbers of support points per covariate, the proposed method easily uncovers complex market structures in simulated and real data. We use recent advances in reversible jump Markov Chain Monte Carlo (MCMC) to estimate the parameters for the attribute-based finite mixture model. The model allows for a different number of components per attribute, so the possible total number of vector components across all attributes may become quite large. Furthermore, in practice, we do not know a priori how many support points each attribute level has. We solve this computational problem by placing a prior on the space of the attribute-based models, assuming that the number of components per attribute level is a discrete random variable. The Reversible Jump MCMC algorithm moves around this variable dimension’s model space and simultaneously estimates the posterior distributions for the number of components and the mixture components, conditional on the number of components. It thus provides an established framework for model choice. 2 - Empirical Test of the Bass Diffusion Model using Exogenous Shocks on Word of Mouth Effect Sungjoon Nam, Assistant Professor, Rutgers University, 1 Washington St., #992, Newark, NJ, 07102, United States of America, [email protected] This paper empirically tests whether the Bass Diffusion Model really captures the communication structure by using a video-on-demand service adoption data. Previous attempts to identify the communication structure suffer from aggregation errors and endogeneity biases caused by the “correlated” errors. In contrast, the new data features enable us identify the different communication structure of a diffusion process using micro level diffusion data with exogenous shocks on word of mouth effects. The data has objective measures on exogenously determined signal quality that influences the number of new movies updated. Since the signal quality is not observed for new adopters but only could be inferred from their neighbors, it only influences the rate of imitation. In contrary to BDM’s assumptions, the estimated results show that exogenous shocks on word of mouth effect affect the coefficient of innovation, not the coefficient of imitation. The Bass model’ ability to capture the communication structure is not supported at the micro level adoption of video on demand service. Consequently we should be very careful to interpret these parameters in BDM to represent the communication structure in new product diffusion. 4 - A Nonparametric Model of Attribute Based Inter-product Competition Sudhir Voleti, Assistant Professor, Indian School of Business (ISB), 2118, AC2 L1, ISB Campus, Gachibowli, Hyderabad, AP, 50032, India, [email protected], Pulak Ghosh, Praveen Kopalle We propose an approach to assess attribute based inter-product competition that employs a novel nonparametric statistical method, the nested Dirichlet Process (nDP), to effect a hierarchical clustering of our units of analysis – brands and SKUs. Our method exploits the information content in both the observed and the latent attributes in the data to induce a pattern of dependence that groups ‘similar’ (and hence, more substitutable) brands into similar-brand clusters and, simultaneously, groups similar SKUs within the similar-brand clusters into similar-SKU clusters. Subsequently, linear programming methods yield attribute weights that best discriminate between each pair of similar-unit clusters. These attribute weights feed into a comprehensive competition measure that easily finds place in most generalized linear models of sales or market response. We then demonstrate the utility of the proposed competition measure by applying it in a category-profit maximization problem. We find that the competition measure proposed improves fit, explained variance, prediction and managerial insight in the application context. Our approach bears several advantages over extant methods - it is parsimonious, flexible, avoids distributional assumptions on model terms, avoids the model selection problem by endogenously determining the appropriate number of similarity clusters, allows inference on the clusters obtained, is able to accommodate restrictions defined a priori based on category structure, accommodates asymmetric competitive effects between pairs of products, and yields realistic cross-product substitution effects, intuitive marginal effects as well as own- and cross-attribute elasticities. 3 - Spatiotemporal Analysis of New Product Diffusion Li Zheng, ESSEC, 7, Rue Du General Henrys, Esc. 6; Apt. 67, Paris, 75017, France, [email protected] The classical diffusion model implies two key assumptions: 1) all members of a given population equally affect and are affected by one another; 2) the potential influence of prior adoption remains constant through time from the beginning of their occurrence. This article will challenge these two key assumptions, and attempt to develop a spatiotemporal hierarchical model of diffusion that allows heterogeneity both within the population and over time. In the space dimension, spatial heterogeneity and spatial dependence play a role in innovation diffusion. Captured by space-specific covariates, the spatial heterogeneity refers to the aspects that influence the marginal diffusion speed across spaces. Several kinds of regional characteristics such as economy, demography and telecommunication infrastructure are modeled as drivers of diffusion growth. The spatial dependence refers to the spillover influence of region j on region i, which connects the current adopters in each region to the past adopters in related regions (i.e., from the ‘influence set’). Several kinds of proximity such as geographic, transportation, technological and demographic similarity help in defining such an ‘influence set’. In the time dimension, the power of earlier adopters’ influence on the potential adopters decays over time. The intuition behind the model structure is that not all the adopters have influence on the potential adopters’ decision-making, and such impact may vary across space and over time. 12 MARKETING SCIENCE CONFERENCE – 2011 TB07 ■ TB06 ■ TB07 Legends Ballroom VII Founders I Competition II: More on Measuring the Impact in Retail Markets ASA Special Session on the Marketing-Statistics Interface - II Cluster: Special Sessions Invited Session Cluster: Special Sessions Invited Session Chair: A. Yesim Orhun, University of Chicago, Booth School of Business, 5807 Woodlawn Avenue, Chicago, IL, 60637, United States of America, [email protected] 1 - Entry with Social Planning Stephan Seiler, London School of Economics, Department of Economics, Houghton Street, London, WC2A 2AE, United Kingdom, [email protected], Pasquale Schiraldi, Howard Smith Chair: Anindya Ghose, New York University, 44 W 4th Street, Suite 8-94, New York, NY, 10012, United States of America, [email protected] 1 - Assessing the Validity of Market Structure Analysis Derived from Text Mining Data Oded Netzer, Philip H. Geier Jr. Associate Professor, Columbia Business School, 3022 Broadway, New York, NY, 10027, United States of America, [email protected], Ronen Feldman, Moshe Fresko, Jacob Goldenberg In 1996 a regulatory reform was introduced in the UK that made it more difficult to open large out-of-town supermarkets. The idea behind this planning regulation was to protect town centre vitality. In this paper we analyze the consequences for consumers and firms of alternative planning policies. We start by analyzing demand for the UK supermarket industry. The industry is characterized by stores that are differentiated along various dimensions: they offer different ranges of product and products of different quality, they vary in size etc. As most consumers regularly visit several supermarkets in the same week it is important to carefully model the interactions between different stores. To this end we propose a demand system in which we allow consumers to visit up to two stores in each week and do not impose a priori restrictions as to whether two different types of supermarkets are substitutes or complements. We use individual level data on store choice as well as expenditure to estimate the model. In the case of two-stop shopping we also use information on how weekly expenditure is split up between the two stores. In the estimation we carefully disentangle correlation in preferences from true complementarity. Finally we use the demand estimates in order to compute supermarket profits which are used in a model of store entry. We explicitly model the decision of the planning authority as we have data on planning applications (both accepted and rejected ones). We simulate a counterfactual in which we remove the asymmetric treatment of large and small stores in order to analyze the effect on store profits and consumer welfare. Can one analyze the information posted by consumers on the Internet to allow managers to assess market structure? Web 2.0 provides gathering places for internet users in blogs, forums, and chat-rooms. These gathering places leave footprints in the form of colossal amounts of data. This type of information offers the firm an opportunity to “listen” to consumers in the market in general and to its own customers in particular. Our objective is to utilize the large-scale consumer generated data posted on the Web, in order to allow firms to understand consumers’ brand associative network and the implied market structure insights. We first text-mine the Web exploratory data and convert them into quantifiable perceptual association and similarity between brands and products. We use network analysis techniques to convert the text-mined data into a semantic network, which can in turn inform the firm, or the researcher, about the market structure and some meaningful relationships therein. We demonstrate this approach using two cases - sedan cars and diabetes drugs - generating market structure perceptual maps, without interviewing a single consumer. The proposed approach demonstrates high degree of internal validity. We examine the external validity of the proposed approach by comparing the market structure mined from the user-generated content to those obtain from traditional market structure approaches based on sales and survey data. The comparison to traditional market structure approaches provides strong support for the external validity of utilizing online conversations and the text mining approach to derive market structure. 2 - Sleeping with the “Frenemy”: The Agglomeration-differentiation Tradeoff in Spatial Location Choice Sumon Datta, Purdue University, Krannert School of Management, 403 W. State Street, West Lafayette, IN, 47907, United States of America, [email protected], K. Sudhir 2 - What Drives Me? A Novel Application of the Conjoint Adaptive Ranking Database System to Vehicle Consideration Set Formation using Population Statistics Ely Dahan, Princeton University, (Visiting), Princeton, NJ, United States of America, [email protected] A central tradeoff in location choice is the balance between agglomeration and differentiation. Should a firm co-locate (sleep) with a competitor to increase volume (competitor is a “friend” who can draw more customers to the location with agglomeration) or locate far away from a competitor in order to reduce competition (competitor is an “enemy” from who one should spatially differentiate)? Since observed co-location may be consistent with pure differentiation rationales such as (a) high demand at the location; (b) low cost at the location and (c) restrictive zoning regulations which allow entry in only small areas, it is challenging to disentangle the agglomeration- differentiation tradeoff from firms’ location choices. The paper develops a comprehensive structural model of entry and location choice that helps disentangle the agglomeration-differentiation tradeoff by decomposing profits into revenue and cost, and then further decomposing the revenue into its components of consumer choice based volume and competition based price. To capture zoning effects, we introduce a new approach to obtain zoning data, an approach that should be of general interest for a large stream of spatial location applications. Our results show that the agglomeration effect explains a significant fraction of observed colocation. Surprisingly, zoning has little direct effect on co-location. But tighter zoning restrictions interact with the agglomeration effect to explain a surprisingly large fraction of observed co-location. We find that strategic firms respond in complex and nonlinear ways to a small change in zoning which could cause a discontinuous impact on the observed location pattern. A new method of individual, internet-based adaptive choice-based conjoint analysis for vehicles points to a future of highly efficient questioning with a new purpose: helping customers understand themselves. Several novel approaches underpin this method: (1) the development of adaptive choice-based conjoint based on a predefined database of utilities, (2) development of a random utility generator utilizing population statistics that acts as a simulator of the actual market, including the ability to reproduce real market shares, (3) the use of actual products as conjoint stimuli, and (4) fine-tuning the tradeoff between allowing for respondent error versus enforcing consistent answers using computer assistance. 3 - How is the Mobile Internet Different? Search Costs and Local Activity Sangpil Han, New York University, New York, NY, United States of America, [email protected], Avi Goldfarb, Anindya Ghose We explore how internet browsing behavior varies between mobile devices and personal computers. Smaller screen sizes on mobile devices increase the cost to the user of reading information. In addition, a wider range of locations for mobile internet usage suggests that the offline context can be particularly important. Using data on user behavior at a microblogging service (similar to Twitter), we exploit randomization in the ranking mechanism for user-generated microblog posts to identify a random experiment in the cost of reading information. Using a hierarchical Bayesian framework to better control for heterogeneity, our estimates show that search costs are higher on mobile devices compared to PCs. While links that appear at the top of a page are always more likely to be clicked, this effect is much stronger on mobile devices. We also find that the benefit of searching for geographically close matches is higher on mobile devices. Stores located in close proximity to a user are much more likely to be clicked on mobile devices than PCs. We discuss how these changes may affect market outcomes in local commerce and the implications for monetization of user-generated content in social media platforms. 3 - Does Reducing Spatial Differentiation Increase Product Differentiation? Effects of Zoning on Retail Entry and Format Variety K. Sudhir, Yale School of Management, Yale School of Management, New Haven, CT, United States of America, [email protected], Sumon Datta Zoning regulations limit the extent to which a firm can spatially differentiate. Even though the inability to spatially differentiate can lead to lower prices, a common conjecture is that zoning can be anti-competitive because fewer retailers will choose to enter tightly zoned markets. However, retailers also have the choice to profitably compete by choosing a higher level of format differentiation. Using estimates from a structural model of entry and location choice in the presence of zoning restrictions, we are able to perform counterfactuals to evaluate the relative impact of zoning on both the number of firms and format variety. We find that zoning impacts entry significantly more strongly when firms cannot differentiate on formats. Also, given the ability to differentiate on formats, for large ranges of zoning restrictions, retailers respond only with changes in the format mix rather than by reducing the number of firms. This implies that empirically, one may find weak linkages between zoning restrictions and entry, when retailers can differentiate on format. 13 TB08 MARKETING SCIENCE CONFERENCE – 2011 4 - Evaluating Financial Risk from Cross Border M&A Activities on Brand Identity Sustainability Sixing Chen, University of Connecticut, Storrs, CT, United States of America, [email protected], Xiaoqi Yang, Ronald W. Cotterill which it does so. We propose a general theory for how innovation creates new value for a firm and apply this theory to understanding how new value, measured as abnormal stock returns, from innovation is reflected in the changes it manifests in the diversity of a firm’s product portfolio. The empirical analysis of this framework seems to indicate that both innovations achieve value through individual mechanisms (e.g., demand) and portfolio mechanisms (e.g., leverage and cannibalization). The importance of this topic lies in both theory and practice. Theoretically, this work sheds light on the degree to which innovation value is a function of the value accrued to the innovation itself and its interdependency with the firm’s overall product portfolio. Practically, understanding how the new value from innovation incorporates the effect of the innovation on the firm’s portfolio enables firms to grasp how decisions they make regarding innovation influences their expected success. The past several years have witnessed a strong trend of cross border merger and acquisition activities in emerging markets. M&A activity has a deep impact on reconstructing firms’ organizational structure and market structure, facilitating company organizational identity and corporate branding. Most financially mature companies in emerging markets seek technology advances as well as corporate recognition over the cross border M&A activity. However, cross border M&A includes huge financial risks that are related to the sustainability of the brand identity of the corporation both before and after the M&A process. This research examines the relationships between the firm’s financial risk and the sustainability of corporate branding identity during a cross border M&A activity. It chooses companies from Brazil, Russia, India and China, that initiated cross border M&A in the past 10 years. It includes multiple industries and M&A activities that happen between emerging markets and countries from different continents, such as North America, Europe, South East Asia, and Africa. We also focus on controlling some market related and technological determinants of R&D intensity and spending for different industries. We use both the branding rankings in business publications as well as corporate financial data for this empirical study. We expect the findings of the research will provide sufficient support that advertising spending and new leadership structure is among the factors that are highly associated with sustainability of brand identity after a cross border M&A. 4 - Entrepreneurs as Owner-managers, Ownership Concentration and Business Model Innovation Chander Velu, Judge Business School, University of Cambridge, Judge Business School, Trumpington Street, Cambridge, CB2 1AG, United Kingdom, [email protected], Arun Jacob We examine the relationship between ownership and innovation by using data from the US and European bond trading industry. We show that less concentrated ownership and presence of entrepreneurs as owner-mangers positively influences the degree of innovation. In addition, we show that the positive relationship between less concentrated ownership and the degree of innovation is more pronounced in highly competitive environments and becomes negative in less competitive environments. On the other hand, we show that the positive relationship between entrepreneurs as owner-managers and the degree of innovation is stronger in less competitive environments and becomes negative in highly competitive environments. ■ TB08 Founders II ■ TB09 Innovation II Founders III Contributed Session Promotions II Chair: Chander Velu, Judge Business School, University of Cambridge, Judge Business School, Trumpington Street, Cambridge, CB2 1AG, United Kingdom, [email protected] 1 - An Analysis of the Financial Performance of Radical, Complex and Financially Risky Innovations Lisa Schöler, Goethe-University Frankfurt, Grüneburgplatz 1, Frankfurt, Germany, [email protected], Bernd Skiera, Gerard J. Tellis Contributed Session Chair: Francesca Sotgiu, HEC Paris, 1 Rue de la Liberation, Jouy en Josas, France, [email protected] 1 - On the Timing and Depth of a Manufacturer’s Sales Promotion Decisions with Forward-looking Consumers Yan Liu, Assistant Professor, Texas A&M University, 4112 TAMU, 220G Wehner, College Station, TX, 77843, United States of America, [email protected], Subramanian Balachander, Sumon Datta Innovation introduction is an important driver of the financial performance of a firm, but little knowledge exists on the influence of radical, complex and financially risky innovations on stock market returns and how these three important characteristics are moderated by economic conditions and location. This study analyzes 198 product announcements with an event study and shows how these three important characteristics of innovations influence stock market returns. The authors find that radicalness and financial risk have a positive and complexity has a negative impact on stock market returns. These impacts are also influenced by economic conditions and location. This paper investigates a manufacturer’s optimal timing and depth of price promotions over a planning horizon in a frequently purchased packaged goods context. Our empirical analysis comprises of two steps. In the first step, we obtain heterogeneous demand side parameters with a dynamic structural model. In this model, consumers decide whether to buy, which brand to buy and how much to buy conditional on their rational expectations of future promotions. In the second step, we specify a dynamic game between consumers and the manufacturer and solve for the optimal promotion policy, taking the structural demand-side parameters from the first step as given. We obtain the optimal promotion policy as the Markov-perfect equilibrium outcome of the dynamic game. In our empirical application, we develop the optimal promotion schedule for the StarKist brand in the canned tuna category using household-level panel data. We find that it is optimal for the manufacturer to promote when the mean inventory for brand switchers is sufficiently low and that the optimal discount depth decreases in the mean inventory for brand switchers. We also find that Starkist could increase profit by offering more frequent but shallower price promotions. Interestingly, we find that the manufacturer’s profit increases as consumers become more forward-looking (discount the future less). 2 - Promoting Growth and Innovation through Acquisition: A Choice Modeling Approach Yu Yu, Assistant Professor, Georgia State University, 1336 College of Business, Georgia State U, 35 Broad Street, Atlanta, GA, 30303, United States of America, [email protected], Vithala Rao While innovation and growth can be promoted internally through focus on research and development (R&D), many firms find acquisition from external sources to be a fast and attractive alternative. Despite the numerous theories of merger and acquisition in the literature, no empirical study has tackled the problem of target selection in an acquisition. This paper is the first to study the target selection criteria in an empirical setting. It quantifies the elusive concept of synergy by developing novel measures of similarity and complementarity between the acquirer and the target that are more comprehensive than the existing measures in the literature. Using an innovative application of the discrete choice model, the authors find that firms use acquisition to promote growth and innovation in areas of strategic interest. Specifically, acquirers choose targets whose product markets match their own R&D projects, and targets whose R&D projects match their own product markets. These findings provide support for the knowledge based view of the firm and lay the foundation for future research in this area. In spite of having a relative small sample, the authors were able to illustrate the robustness of the estimation results as well as the predictive ability of the model through a set of tests. These tests might be valuable for many empirical researchers who also face small sample constraints on a regular bases. 2 - Empirical Investigation of Consumer Impulse Purchases from Television Home Shopping Channels Sang Hee Bae, PhD Candidate, New York University, 40 West 4th St. Tisch Hall #920, New York, NY, 10012, United States of America, [email protected], Sang-Hoon Kim, Sungjoon Nam Due to retail channel proliferation and mega size retail stores, marketers are putting more efforts in in-store marketing tactics to induce more impulse purchases while customers are at the stores. In particular, customers are more likely to purchase on impulse when they discover a product by chance with a price promotion or a bundling/multi-unit packaging offer. Although impulse buying is a widespread phenomenon, previous literature heavily relies on interrupt survey data or lab experiments on unplanned/impulse purchases. Also, the empirical data have been limited to a small number of product categories such as fashion and grocery. This paper empirically investigates the consumers’ retrospective canceling behavior on previous impulsive purchases for pricing, product bundling, and packaging offer. We use order canceling data from a television home shopping channel to proxy impulse purchase behavior. Order cancellation occurs when a customer places a product order, and subsequently cancels the order even before the product is shipped. The unique aspect of our data can distinguish between impulse purchases and unplanned purchases. We find that a small price discount significantly lowers consumer order cancellation by 30%. Also impulse buying is more likely to occur when products are sold with bundling offers for hedonic goods, but not for utilitarian goods. Furthermore, order cancellation varies by multi-unit packaging of products. These findings give managerial implications for in-store price promotion and bundling practices. 3 - Product Portfolio Effects of Innovation: A Diversification Perspective on Innovation Value Creation Fredrika Spencer, University of North Carolina-Wilmington, 1009 Country Glen Ct, Apex, NC, 27502, United States of America, [email protected], Richard Staelin Organizational researchers have long considered innovation a critical activity. While insightful regarding the nature of the innovation process and the rewards and risk associated with innovation, prior work has neglected the perspective that innovations function within a firm’s wider product portfolio. This perspective enables assessment of when innovations truly generate new value for firms and the mechanisms through 14 MARKETING SCIENCE CONFERENCE – 2011 3 - The Impact of Free-trial Promotions on Adoption of a High-tech Consumer Service Bram Foubert, Assistant Professor, Maastricht University, School of Business and Economics, Department of Marketing and SCM, P.O. Box 616, Maastricht, 6200 MD, Netherlands, [email protected], Els Gijsbrechts, Charlotte Rolef TB10 we examine research from psychology and behavioral decision making to develop our theory and then test our hypotheses using successive waves of the PATS survey. We find that non-users do a good job predicting future non-use. However, those who have previously tried drugs before generally do a poor job of predicting future use, systematically under-predicting the amount that they will use in the future. Further, we find that the greater the amount of prior use, the greater the under-prediction. More generally, we find evidence that situation plays a larger role than attitude in accounting for this systematic under-prediction. With the boost of electronic consumer services like online movie rentals or digital TV, free-trial promotions have gained widespread acceptance. Although a few studies have explored the impact of samples for consumer packaged goods, free trials for electronic consumer services deserve separate attention. For one, a service trial does not involve a fixed consumption amount but a fixed consumption period, such that learning depends on the consumer’s usage intensity. Moreover, because of its hi-tech nature, the quality of the service may evolve over time. In this study, we investigate the effects of free-trial promotions on consumers’ adoption decisions for hi-tech services. We identify the different components of a consumer’s “adoption utility” that are influenced by a free trial, and use a mixed logit structure with Bayesian learning to model the resulting effects on adoption behavior. A key feature of our model is that it incorporates usage-based learning about service quality in a context where the service quality itself evolves over time. We estimate the model on data from a large European telecom operator offering free trials to promote its new interactive digital TV service. Our empirical results yield several key insights for managers. First, our analysis enables us to document how free trials impact both the number of adopters and the total subscription fees – thereby shedding light on the degree of subsidization. Second, we compare the effectiveness of free trials in conveying information about the service, with that of other tools, namely advertising and direct marketing. Last but not least, we generate insights into the appropriate timing of free-trial promotions, and offer guidelines for targeting. 2 - Units versus Numbers Ashwani Monga, University of South Carolina, 1705 College Street, Columbia, SC, 29212, United States of America, [email protected], Rajesh Bagchi Marketers frequently make changes to their product offerings (shipping time, package size, etc.). We study how consumers react to such changes. The physical quantities we consider are duration, height, and weight. Numerosity research suggests that people react more strongly to equivalent changes that are expressed in units that are small (e.g., change in shipping time from 7 to 21 days) rather than large (e.g., change from 1 to 3 weeks), because of the size of the numbers (7 to 21 > 1 to 3). We propose an opposite effect – what we call unitosity – such that people react more strongly to changes in large rather than small units because of the size of the units themselves (few more weeks > few more days). These opposing effects, we argue, could occur due to perceptual salience such that physically prominent numbers elicit numerosity, but prominent units elicit unitosity. Then, we discuss cognitive salience due to mindsets. We argue that people construe numbers at a low level and units at a high level, because of which a concrete mindset yields a numerosity effect whereas an abstract mindset yields a unitosity effect. We observe such numerosity-unitosity reversals in four laboratory studies. 4 - Promotion Effectiveness in Economic Turbulence: From Price Wars to Economic Downturns Francesca Sotgiu, HEC Paris, 1 Rue de la Liberation, Jouy en Josas, France, [email protected], Katrijn Gielens 3 - Resource Abundance and Conservation in Consumption Meng Zhu, Carnegie Mellon University, 5000 Forbes Ave, Posner 385a, Tepper School of Business, CMU, Pittsburgh, PA, 15213, United States of America, [email protected], Ajay Kalra Reports on supermarket price wars are a wide spread theme in the business press, throughout the world and over time. When retailers intensify price competition, brand manufacturers are often at a loss, as they tend to lose control over their price and promotion tactics. Whereas temporary price discounts are manufacturers favorite tool to (temporary) boost their brand’ sales and market shares, it remains unclear how individual promotion actions perform amid a barrage of store-wide pricerollbacks. To address this issue, we examine price promotion effectiveness when brand manufacturers are tied up in supermarket price wars. We look at whether brand managers can benefit from promoting more (or less) during price wars and whether they would be better off refraining from entering price wars, and/or changing their tactics as price wars go on. To do so, we first estimate the effectiveness of individual promotion events using a multiple break analysis (Leone 1987). Next, we relate the individual price promotion sensitivities to different price war scenarios. We analyze 687 individual price promotion events of a multinational CPG manufacturer at four competing supermarket chains between 2001 and 2005 in the Dutch market. Our results reveal that during price wars, promotions are more effective than during a business-as-usual environment. This effect, however, decreases over time during price wars. Still, when price wars coincide with economic contractions promotion effectiveness increases over time. In most consumption contexts, consumers tend to seek convenience, which typically leads to greater acquisition of resources than is necessary. Such over-acquisition of resources is only possible when resources are abundant. When resources are not abundant, consumers need to carefully monitor their acquisition and consumption to not deplete the supply. Therefore, cues indicating non-abundance of a resource can prompt conservation behaviors. Importantly, we posit that the tendency to conserve triggered by non-abundance cues in a prior context can persist into subsequent consumption of unrelated resources. In four experiments, we demonstrate the proposed phenomenon, showing that non-abundance cues regarding one particular resource decrease cognitive accessibility of the general construct of abundance, and subsequently increase participants’ tendency to conserve a different type of resource (e.g., water, energy). Our results suggest that the underlying mechanism for the effect of non-abundance cues on conservation is motivational rather than priming of conservation-related concepts or traits. 4 - Optimizing the Assortment Layout: The Effect of Categorization Congruency on Purchase Incidence Robert Rooderkerk, Tilburg University, Warandelaan 2, Tilburg, NB, 5000 LE, Netherlands, [email protected] Consumer perceptions of the variety and complexity of an assortment affect the decision to buy or not. The study investigates the influence of assortment configuration on purchase incidence through its effect on both assortment perceptions. The layout of an assortment implies a certain externally induced product categorization. Consumers, on the other hand, have their own internal product categorization. This study studies the effect of the congruency between the external and internal categorization on purchase incidence. The data result from a betweensubjects experiment for a biscuit category, conducted with actual shoppers in a grocery store. Results from Bayesian mediation analyses show that higher categorization congruency positively affects purchase incidence via two routes. The first route corresponds to a positive effect of categorization congruency on perceived variety. This is consistent with the notion that higher categorization congruency facilitates consumer recognition of variety. The second route involves a negative effect of categorization congruency on perceived complexity. An assortment lay-out that is more consistent with a consumer’s internal categorization will make it easier for the consumer to understand the assortment. The effect of congruency on the two assortment perceptions is asymmetrically moderated by product knowledge. Whereas higher product knowledge strengthens the effect of categorization congruency on perceived variety, it has no effect on the relationship with perceived complexity. The analyses account for the correlation between the mediators and the repeated measures nature of the data. As such, they provide a generalizable framework to test for multiple mediation in the presence of a binary dependent variable. ■ TB10 Founders IV Decision Making Contributed Session Chair: Robert Rooderkerk, Tilburg University, Warandelaan 2, Tilburg, NB, 5000 LE, Netherlands, [email protected] 1 - Some Empirical Evidence on Predicted versus Reported Behavior: The Role of Attitudes and Situational William Putsis, University of North Carolina at Chapel Hill, KenanFlagler Business School, CB 3490, McColl Bldg #4518, Chapel Hill, NC, 27599, United States of America, [email protected], Preethika Sainam, Gal Zauberman Drug abuse among teenagers is commonplace: for example, in 2002, an estimated 2.6 million Americans used marijuana. So, what factors lead to drug use among teenagers and do situational or attitudinal factors drive anticipated behavior? For example, while there are many reasons for using drugs, peer pressure often plays a major role in initial trial. In our research, we use data from the Partnership for Drug Free America’s (PDFA) Partnership Attitude Tracking Survey (PATS). Using this survey, the main question that we examine is whether individuals, teenagers in particular, are able to correctly predict their future consumption of illegal drugs. We use prior research to provide hypotheses about the role that situation versus attitude play in the accuracy of their prediction, accounting for factors such as peer influence, access to drugs, attitudes about drugs as well as current and prior drug use. The two key aspects of our framework are prediction accuracy and the power of the situation over beliefs and attitudes in accounting for behavior. To provide insight into these issues, 15 TB11 MARKETING SCIENCE CONFERENCE – 2011 ■ TB11 4 - Priming vs. Wearout: Early Prelaunch Advertising, Online Buzz and New-product Sales Ho Kim, PhD Student, University of California-Los Angeles, Anderson School of Management, 110 Westwood Plaza B401, Los Angeles, 90095, United States of America, [email protected], Dominique Hanssens Champions Center I Response to Advertising Contributed Session Chair: Ho Kim, PhD Student, University of California-Los Angeles, Anderson School of Management, 110 Westwood Plaza B401, Los Angeles, 90095, United States of America, [email protected] 1 - Selling the Drama: Death-related Publicity and its Impact on Music Sales Leif Brandes, Senior Research Assistant, University of Zurich, Department of Business Administration, Plattenstrasse 14, Zurich, 8032, Switzerland, [email protected], Stephan Nüesch, Egon Franck Managers sometimes advertise a new product long before its planned launch, a phenomenon we refer to as “early prelaunch advertising.” Such early marketing investment is often justified by a presumed priming effect, i.e. the launch advertising will be more impactful due to the early advertising. On the other hand, the early advertising may also be subject to wearout and lose all or most of its impact by the time the product is launched. Depending on the relative magnitude of these effects, the early advertising investments may or may not be economically defensible. This study examines the effect of early prelaunch advertising on new-product sales, taking into account consumers’ time-discounting behavior. We develop several hypotheses regarding the effect of prelaunch advertising on buzz generation, as well as the effect of prelaunch buzz on new-product sales. The hypotheses are tested by applying an extended geometric distributed lag model to a dataset consisting of weekly advertising spending, weekly number of blog postings and keyword search volumes of 159 widely released movies. Our findings suggest that (1) consumers’ time-discounting behavior plays a critical role in determining the time-varying effects of prelaunch advertising, (2) the wearout effect dominates the priming effect, and (3) the volume of online buzz in the prelaunch period provides incremental predictive power for forecasting movie box-office revenue. These findings provide insights on the allocation of prelaunch advertising budgets and on the value of intermediate online buzz data for revenue forecasting of new products. This paper analyses the sales impact of artist publicity in the music industry. Our identification strategy employs exogenous variation in the timing of information release due to natural deaths of artists. We have observations on weekly sales figures for 441 albums of 77 artists who died in the period 1992 – 2009. Our empirical analysis proceeds in three steps. First, we address the question if product characteristics, such as album quality, moderate the sales impact of death-related information. Our findings show substantial sales increase after artist deaths, and that quality and information are indeed complements. This result obtains independent of whether we measure quality via expert evaluations, or pre-death sales levels. Second, we find after-death sales to be significantly more responsive to album publicity if an artist’s last album was released more than four years before death. This finding lends itself to the interpretation that customers are partly uninformed about products, but contradicts customer-mood related explanations. Third, we analyse in greater detail the relationship between after-death sales and environmental cues, in particular, album media publicity. Duration analysis reveals that spell length of abnormal afterdeath sales is more persistent than death’s impact on media publicity. We hypothesize that death-related information facilitates observational learning, which leads to the documented persistence in sales. Controlling for album publicity, we find that firstorder autocorrelation in sales is indeed substantially higher after-death than before. We relate our findings to the advertising literature and discuss practical implications. ■ TB12 Champions Center II Brand Identity Contributed Session Chair: Stefan Worm, Assistant Professor, HEC Paris, 1, Rue de la Libération, Jouy-en-Josas, 78351, France, [email protected] 1 - Brand Extensions Frequency and Brand Performance Helena Allman, University of South Carolina, Moore School of Business, Marketing Department., 1705 College Street, Columbia, SC, 29208, United States of America, [email protected] 2 - Is Beauty in the Eye of Beholders? Linking Facial Features to Source Credibility in Advertising Li Xiao, The Pennsylvania State University, 418B Business Building, University Park, State College, PA, 16802, United States of America, [email protected], Min Ding Source credibility states that the effectiveness of a marketing communication depends greatly upon who delivers it. Specifically, source credibility refers to the credibility of a source on expertise, trustworthiness and attractiveness. Although not studied in marketing contexts, it has been well documented in psychology that different faces influence how people make inferences of personality traits, including expertise, trustworthiness and attractiveness. The key assumption of this literature is that people are generally homogeneous in making trait inferences from faces and there exists one consensus trait face (CTF) for a specific trait. In other words, if a face appears trustworthy to one person, then it should appear trustworthy to everyone else. However, this assumption is not without controversy and two alternative views exist. Some scholars argue that heterogeneities in gender, ethnicities and social status make people differ in their way of making inferences from faces, and therefore multiple CTFs exist. Some scholars argue that people’s preferences depend highly on the contexts how preferences are elicited, so no consistent CTF exist. It’s important to solve this controversy because different number of CTFs will result in different types of effective advertising campaign. To address this controversy, we conducted an empirical study, where participants were exposed to multiple sets of synthetic faces and asked to evaluate these faces along seven trait dimensions. They are babyfacedness, masculinity, attractiveness, trustworthiness, competence, aggressiveness and warmth, which are shown to contribute to source credibility in the literature. Multiple models were used to analyze the data. Results and implications for advertising campaign were discussed. The topic of this research is to determine whether the strategy of frequent brand extensions results in improved brand performance, and if so, under what conditions. While it has been shown that firms that maintain higher-than-average revenue growth typically continuously introduce new products, the effects of the above-thanaverage number of new product introductions on brand performance, specifically when brand performance is being assessed from the market-based brand equity point of view, have not been adequately explored in the literature. Whether the frequency or intensity of brand extension introductions matters for brand’s market performance remains an unanswered question. This research aims to answer that question. It is hypothesized that frequent brand extensions results in higher market share and price premium for the focal brands. Parent brand strength positively moderates the relationship between the brand extensions frequency and the brand’s market performance. Brands that introduce more innovative products with higher fit between the extensions and the parent brand perform better than brands with less innovative new products and lower fit between the new products and the parent brand. 3 - Creativity in Advertising and Implications for Product Sales Performance Peter Saffert, University of Cologne, Albertus-Magnus-Platz 1, Cologne, 50923, Germany, [email protected], Werner Reinartz Drawing from Materialism Values Theory, “The Big Five Consumer Personality” Theory, and Brand Personality literature, this paper investigates the effects of material values and consumer personality on perceptions of global versus local fashion brand personality. Previous research focused on the effects of consumer materialism on well-being and impulsive shopping behavior. Our study pioneers exploring the link between consumer materialism and luxury brand versus own-brand personality perceptions. We conduct an online survey targeting visitors to fashion brand websites. According to factor analysis and ANCOVA, highly-materialistic consumers are found to be more likely to view luxury global brand, instead of the two mid-priced domestic own-brands, as manifesting competent/trusted and sophisticated personality traits than less materialistic consumers. For both global luxury brand and local own-brands alike, our results confirm that consumer personality dimensions such as “extrovert” and “conscientious” positively predict perception of exciting/sociable and trusted brand personality respectively as shown to be consistent with prior research and selfcongruity theory. Unlike past findings, “neurotic” consumer personality enhances perception of exciting/sociable brand personality regardless of brand type. This study extends previous research on consumer personality and brand personality by providing empirical evidence concerning the impact as well as interactions of material values and consumer personality on brand personality perception. Finally, we discuss the implications for how to align branding strategies with consumer material values and personality in the case of global luxury brand versus local own-brand. 2 - Material Values and Consumer Personality Effects on Brand Personality Perceptions Tiffany Ting-Yu Wang, Associate Professor, KNU, 70-7,1F,Ln16, Xianyan Rd.,Wenshan Dist, Taipei, 11688, Taiwan - ROC, [email protected] Many studies exist investigating when, how, and under which conditions TV ads are successful in influencing the consumer. It is widely believed that creativity plays a key role in the effectiveness of advertising. Yet, the role of creativity in advertising is not fully understood – in particular with respect to behavioral consumer outcomes. The objective of this study is to take a closer look at creativity and its composing elements, i.e., originality, flexibility, elaboration, synthesis, and artistic value, and analyze the importance of each of these creativity dimensions on the commercial success of the product that is being advertised. We try to answer the question, which aspects of creativity are associated with behavioral response and sales performance and thus, to determine on which of these dimensions the advertising agencies should focus on when designing effective TV advertising spots. In particular, we rate advertising award-winning TV spots across a set of different awards and compare them to non-winning commercials of competing brands in several FMCG product categories in Germany. Moreover, we link the rating on the creative dimensions to the sales outcome of the advertised products. 16 MARKETING SCIENCE CONFERENCE – 2011 TB14 3 - Cross-cultural Differences in Brand Engagement Antonieta Reyes, The Florida State University, 4120K University Center, Building C, Tallahassee, FL, 32306-2651, United States of America, [email protected], Felipe Korzenny 2 - The Case Stock Market Rewards for Customer and Competitor Orientations: of Initial Public Offerings Alok R. Saboo, Pennsylvania State University, PA, United States of America, [email protected], Rajdeep Grewal The purpose of this study was to examine brand engagement in self concept (BESC) among people of different cultural groups in the US. We used Sprott, Czellar, & Spangenberg’s BESC scale to examine if and how consumers from different US cultural groups incorporate brands into their self-concepts. Prior research efforts looking at the relationships between culture and brands has been mostly dedicated to brand loyalty. Brand loyalty, however, has to do with specific brand relationships while BESC examines the a more generalized and abstract relationship with brands. This study is particularly salient because it used substantive national online samples of non-Hispanic Whites, Asians, Hispanics, and African Americans. Results revealed that the relationship between ethnicity and BESC was statistically significant. NonHispanic Whites and Hispanics who prefer Spanish showed significantly lower BESC than all the other groups while African Americans and Asians scored highest on the scale. These results provide evidence that cultural background has a relationship with the degree to which consumers engage with brands. Specifically, the results indicate important differences between Hispanics who chose to answer in Spanish and those who chose English. These relationships suggest that brand engagement varies with levels of acculturation as consumers become more sophisticated in their brand appreciations. The authors derive implications for marketing and brand management based on the trends found. Recognizing that initial public offerings (IPOs) represent the debut of private firms on the public stage, we investigate the role of pre-IPO customer and competitor orientations (CCOs) for the IPO performance of the firm. Building on signaling theory, we propose that these orientations influence investors’ sentiments towards an IPO. We test our framework using data collected from Computer Aided Text Analysis, expert coders, and secondary sources for a sample of 543 firms across 43 industries going public between 2000 and 2004. Results from a Bayesian shrinkage model, which accounts for industry-specific effects and uses latent instrumental variables (LIV) to account for endogeneity of CCOs in the IPO context, show that these orientations positively influence IPO performance. Further, these influences are moderated by IPO specific variables and the facets of the organizational task and institutional environments, such that (1) underwriter reputation and venture funding positively moderate the effects of CCOs; (2) technological and market turbulence positively moderate and institutional complexity negatively moderates the effect of customer orientation; and (3) technological turbulence, competitive intensity, and institutional complexity positively moderate the effect of competitor orientation. Finally, we demonstrate that accounting for endogeneity using latent instrumental variables substantially improves the predictive validity of our model relative to alternate model specifications. 4 - What Makes a Strong B2B Brand? The Role of Tangible versus Intangible Brand Attributes Stefan Worm, Assistant Professor, HEC Paris, 1, Rue de la Libération, Jouy-en-Josas, 78351, France, [email protected] 3 - The Impact of Marketing Strategy on Corporate Bankruptcy Niket Jindal, University of Texas at Austin, McCombs School of Business, Austin, TX, United States of America, [email protected], Leigh McAlister Many B2B firms have started to invest systematically in building their brands to differentiate their products more effectively from competition. However, very little is known about successful strategies to build strong B2B brands. For example, B2B marketers struggle to determine which type of attributes they should establish for their brands. The existing literature advises B2B firms to emphasize intangible (non product-related) brand attributes such as trust or reliability over tangible (i.e. product-related) attributes. Our study questions whether it is an effective strategy to rely on intangible attributes as points-of-difference for strong brands. Based on multiindustry survey data, we show that B2B brand strength is more strongly driven by tangible as opposed to intangible brand attributes. We also find a positive interaction between both types of attributes, indicating that intangible attributes become more effective when complemented by salient tangible attributes. B2B Marketers are today often concerned about the diminishing technical differentiation of the products in their industry and thus turn to brand building. Our study also indicates that it is more important to build intangible brand attributes when product differentiation decreases. Tangible brand attributes however, remain similarly important regardless of the level of product differentiation. After controlling for predictors found in the bankruptcy literature, we ask whether a firm’s marketing strategy has an impact on its bankruptcy risk. We represent marketing strategy conventionally, by advertising and R&D intensity, and also propose a new metric indicating whether the firm discloses advertising and R&D expenditures. The disclosure metric allows us to expand our sample from just those firms that disclose advertising and R&D expenditures to consider all firms, removing a source of sample selection bias. We hypothesize that creditors are more likely to renegotiate credit terms with firms that have a strong emphasis on advertising or R&D (due to the associated cash flow benefits), thereby reducing the risk of these firms filing for bankruptcy. For those firms that do go into bankruptcy, the ones emphasizing advertising build assets that may lose a significant amount of value in liquidation due to a lack of a strong secondary market for brands. We hypothesize that these firms are more likely to reorganize and emerge from bankruptcy (versus liquidate). Analysis of all large publicly traded firms in the U.S. from 1980 to 2006 confirms our hypotheses. ■ TB14 Champions Center VI ■ TB13 Champions Center III Dynamic Pricing Issues Quantifying the Profit Impact of Marketing II Contributed Session Chair: Jonathan Zhang, Assistant Professor of Marketing, University of Washington, 547 Paccar Hall Box 353226, Seattle, WA, 98195, United States of America, [email protected] 1 - Online Content Pricing Anita Rao, Stanford University, Graduate School of Business, 518 Memorial Way, Stanford, CA, 94305, United States of America, [email protected] Cluster: Special Sessions Invited Session Chair: Xueming Luo, Eunice & James L. West Distinguished Professor of Marketing, The University of Texas at Arlington, Arlington, TX, United States of America, [email protected] 1 - An Econometric Model of Firms’ Participation Decisions Across CSR Activities Nitin Mehta, University of Toronto, Toronto, ON, Canada, [email protected], Vikas Mittal, Christopher Groening The internet has changed the way we consume and access content – movies, books, videos and music. High bandwidth, high-speed data streaming and Digital Rights Management (DRM) together have made it possible for owners to sell their content through the internet. The pricing of digital content is a challenging problem which may vary widely based on the type of content and customer heterogeneity arising from 1) consumers who want to consume the content once versus repeatedly, 2) consumers who value consuming the content sooner rather than later and 3) varying degrees of price sensitivity. The goal of this paper is to provide a research framework to guide optimal purchase and rental pricing in the digital world. To illustrate factors affecting the relative purchase and rental prices, this paper explicitly considers movies because of the historical prevalence of both options. Currently digitally downloadable movies are priced rigidly resulting in insufficient price variation making it hard to recover underlying consumer preferences. We resort to an experimental design where consumers are asked to trade-off between Buying Now, Renting Now and Postponing their decision in choice tasks where the current and future purchase and rental prices, as well as the time the future prices come into effect, are varied. This enables us to identify the demand parameters governing consumer’s preferences which in conjunction with a dynamic equilibrium framework are used to compute the optimal prices to be charged over time for both the purchase and rental options. We outline a model of firms’ decisions to participate in different sets of activities related to corporate social responsibility (CSR). Specifically, we investigate how firms allocate their resources amongst different sets of activities across three broad areas of CSR: environment, community and employees. To do so, we first propose a two stage econometric model of firm’s decisions. In the first stage, the firm decides on the allocation of its resources across the different broad areas of CSR. In the second stage, the firm decides on how it should allocate its area specific resources amongst the different activities within that CSR area. We estimate our model on the KLD data set that covers the yearly participation decisions of around 1000 firms over 10 years amongst different sets of activities across three different areas of CSR. The questions that we address are: (a) To what extent does a firm’s decision to participate in each CSR activity depend on the firm’s characteristics (such as its financial performance indicators, size, R&D and advertising intensities, impact of external shareholders) and the characteristics of the industry that the firm belongs to (such as the extent of unionization in the industry, the extent of competition in the industry)? (b) To what extent does a firm’s decision to participate simultaneously in any two activities stems from heterogeneity and from complementarity (i.e., the synergies that the firm would enjoy by participating in the two activities simultaneously)? (c) To what extent does a firm’s decision to participate in a CSR activity result from compensatory behavior, whereby the firm tries to compensate its prior poor track record in a CSR area by participating in CSR activities in that area? 17 TC01 MARKETING SCIENCE CONFERENCE – 2011 2 - Conspicuous Consumption and Dynamic Pricing Richard Schaefer, PhD Student, University of Texas at Austin, 1 University Station, B6700, McCombs School of Business, Austin, TX, 78703, United States of America, [email protected], Raghunath Rao Thursday, 1:30pm - 3:00pm We study a producer’s dynamic pricing policy when marketing a durable good, particularly an item that provides consumer utility via two mechanisms; specifically, consumers experience an intrinsic consumption utility and an externality (denoted ‘fashion utility’) that depends upon the conspicuousness of the product and the identity of other consumers of using the same product.In our analytical model, we consider the joint impact of consumption utility and fashion utility, and our results reverse the direction of causality long emphasized in prior studies. We show that products with high intrinsic quality command higher prices due to greater input costs; with a higher retail price, such products become exclusive and, hence, more fashionable when consumption is visible. Two key results emerge from the analysis of our dynamic model: a) whether the producer sells its product via skimming over time depends upon the discount factor, and b) under skimming, more visible products depreciate faster. We extend our model to study the investments that producer might make to separate the cohorts over time and thereby not dilute the intertemporal fashion utility for the early adopters of the product. We find that a producer will be willing to incur higher costs for cohort separation if the product visibility is higher but would decrease this investment if the product is higher in intrinsic quality, all else being equal. The results of our paper are of potential interest to manufacturers of fashion goods for their pricing strategies as well as to the policy makers for studying the welfare impact of identity-related goods that have often been derided for providing very few intrinsic benefits. Legends Ballroom I ■ TC01 Choice I: New Models of … Contributed Session Chair: Peter Stuettgen, Carnegie Mellon University, 5000 Forbes Ave., Pittsburgh, PA, 15213, United States of America, [email protected] 1 - Assessing Two Alternative Methods for Modelling Heterogeneity in Stated Preference Data Paul Wang, Senior Lecturer, University of Technology, Sydney, Marketing Discipline Group, P.O. Box 123, Sydney, Australia, [email protected], Jordan Louviere, Kyuseop Kwak Market segmentation has long been recognized as one of the key concepts in the marketing discipline (e.g., Frank et al. 1972; Wedel and Kamakura 2000). It refers to the process of classifying customers into homogeneous groups known as segments. Stated choice experiment (aka choice-based conjoint analysis) offers a more powerful method to obtain customer preference data than traditional rating method (Louviere et al. 2000; Street et al. 2005). Stated choice models based on the random utility framework are becoming increasingly popular in marketing and applied economics literature (Louviere et al. 2000; Train 2009). The need to account for preference heterogeneity in such models has motivated researchers to develop various ways to solve the problem. Most commonly, researchers make a certain distributional assumption for unknown heterogeneity across respondents and across choice tasks. If a continuous distribution is assumed, mixed logit (Train 2009) or hierarchical Bayesian approaches (Rossi et al. 2005) are used to approximate such unobserved heterogeneity. If a discrete distribution is assumed, latent class model is mostly used (Kamakura and Russell 1989) to identify latent market segments. The primary purpose of this paper is to introduce two alternative methods for modeling preference heterogeneity in stated choice data: (1) Cutler and Breiman’s (1994) archetypal analysis method and (2) Kaufman and Rousseeuw’s (1990) fuzzy analysis clustering method. We compare these less well-known ways of incorporating preference heterogeneity with the traditional latent class modeling approach using health carerelated choice data. We find the two alternative methods have considerable promise for strategic market segmentation applications. 3 - Estimating Dynamic Pricing Decisions in Markets with State Dependent Demand Koray Cosguner, PhD Student in Marketing, Washington University in St. Louis, Olin Business School, Campus Box 1133, Saint Louis, MO, 63130, United States of America, [email protected], Tat Y. Chan, P. B. Seetharaman We propose an empirical approach to examine the pricing behavior of manufacturers in the presence of state dependent demand. To the extent that state dependence characterizes the evolution of brands’ market shares, profit maximizing firms should consider these inter-temporal linkages in demand and be forward-looking in their pricing decisions. In this study, we estimate such a dynamic pricing model of firms. First, we propose a demand model with state dependence and estimate it by using household scanner panel data. This demand model is inputted into our fully structural dynamic pricing model that we use to estimate the marginal costs of each firm in the industry studied. We use approaches proposed by Berry and Pakes (2000) and Bajari, Benkard and Levin (2007) to estimate this dynamic pricing model. Before estimating the dynamic pricing model by using store level data, we perform a simulation study by using Pakes and McGuire (1994) algorithm. This study allows us to check the performance of Berry and Pakes (2000) and Bajari, Benkard and Levin (2007) approaches in terms of recovering the assumed structural parameters of interest. Then, we estimate the marginal costs of each firm in the cola product category by using our proposed dynamic pricing model and compare our estimates with the estimates from two benchmark cases: myopic (maximizing single period profit) and static pricing models (demand without state dependence). Our framework not only helps firms to understand the demand dynamics in the industry, but also helps them to make their pricing decisions optimally. Furthermore, the proposed framework can be used by new potential entrants, incumbent firms or regulators to understand the cost structure in any oligopolistic market. 2 - A Direct Utility Model for Asymmetric Complements Sanghak Lee, The Ohio State University, 2100 Neil Ave, Columbus, 43210, United States of America, [email protected], Greg Allenby, Jaehwan Kim Asymmetric complements refer to goods where one is more dependent on the other, yet consumers receive enhanced utility from consuming both. Examples include garden hoses and sprinklers, chips and dip, and routine versus personalized services where the former has a broader base for utility generation and the latter is more dependent on the other’s presence. Measuring the presence of asymmetries is difficult because it requires longitudinal variation of utility where the degree of interdependency changes. We introduce a utility structure capable of identifying the origin of demand variation, and investigate the presence of asymmetric complementarity using scanner panel data of milk, cereal, ketchup, and yogurt purchases. Implications are explored through counterfactual analyses involving price elasticities, spillover effects and the influence of merchandising variables. 4 - Dynamic Targeted Pricing in B2B Settings Jonathan Zhang, Assistant Professor of Marketing, University of Washington, 547 Paccar Hall Box 353226, Seattle, WA, 98195, United States of America, [email protected], Oded Netzer, Asim Ansari 3 - Utility-based Model of Asymmetric Competitive Structure using Store-level and Forced Switching Data Paul Messinger, University of Alberta, 3-23 Business Building, Edmonton, Canada, [email protected], Fang Wu This research models the impact of firm pricing decisions on different facets of the customer purchasing process in business-to-business (B2B) contexts and develops an integrated framework for inter-temporal targeted pricing to maximize long-term profitability for the firm. B2B pricing often allows considerable flexibility in implementing first degree and inter-temporal price discrimination, and often involve a request for a price quote providing the firm with an opportunity to better assess price sensitivity and unfulfilled demand. The proposed model considers the buyer’s quantity, timing and bid request and acceptance decisions in an integrated fashion while accounting for customer heterogeneity, asymmetric reference price effects, and short- and long-term purchase dynamics. We weave together hierarchical Bayesian copulas with a non-homogeneous hidden Markov model to account for the interrelated decisions and for long-term purchase dynamics. The results reveal that the seller’s pricing decisions can have long-term impact on its buyers by shifting their preferences between a “vigilant” and “relaxed” buying state, Furthermore, price losses relative to a reference price not only have larger negative effects relative to gains on buyers’ buying behavior, but buyers also take longer to adapt to losses. Additionally, the proposed model exhibits superior predictive performance relative to benchmark models. In a price policy simulation the proposed model results in a 52% improvement in profitability, demonstrating the potential to employ value-based pricing policies even in a traditional B2B industry characterized by cost-plus pricing practices. Other policy simulations shed light on how B2B firm should price in the recent economic environment with volatile commodity prices. We propose a utility-based model of competitive structure that accounts for and spatially represents both vertical (quality) and horizontal characteristics, where we show how the vertical characteristics govern observed asymmetric substitution patterns in the data. We find it desirable to estimate our model jointly with two kinds of data, aggregate store-level sales data and forced switching survey (stated preference) data. To facilitate estimation with such two kinds of data, we develop a behavioral justification that establishes a conceptual linkage between these two kinds of data. The intuition idea behind our modeling approach is that consumers’ actual purchase behavior for brands gives us important market structure information regarding brands’ proximities in some latent attribute space, and the same proximity relationships across brands influence the consumers’ forced switching behavior. By imposing a particular distribution on latent ideal points and willingness to pay parameters, we are able to estimate heterogeneity in our proposed model. 4 - A Satisficing Choice Model Peter Stuettgen, Carnegie Mellon University, 5000 Forbes Ave., Pittsburgh, PA, 15213, United States of America, [email protected], Peter Boatwright, Robert Monroe In addition to the standard compensatory choice models like logit and probit models, a new stream of research has recently proposed several non-compensatory choice models, most of them variants of the conjunctive choice rule (e.g., Gilbride and Allenby 2004, Jedidi and Kohli 2005). We contribute to this new stream by formulating a consumer choice model based on Simon’s (1955) idea of “satisficing” decision making, in which alternatives are evaluated sequentially and the first satisfactory alternative is chosen. This requires knowledge (or probabilistic modeling) 18 MARKETING SCIENCE CONFERENCE – 2011 TC03 4 - Website Ad Quantities: An Empirical Analysis of Traffic, Competition, and Business Model Laura Kornish, Associate Professor, University of Colorado-Boulder, Boulder, CO, United States of America, [email protected], Jameson Watts of the sequence of evaluation, a part of decision making that does not appear in previous models. We estimate our model using data from a conjoint-like experiment, in which the stimuli are pictorials of supermarket shelves and the participants’ gaze is recorded via eye-tracking equipment. Eye-tracking not only allows us to know the order of evaluation, but also to only include information into the decision making process that the decision maker was actually aware of (e.g., if the participant never looked at price, we know that price did not affect the decision). Finally, modeling the observed eye path and the decision making jointly lets us (1) differentiate between variable that affect attention vs. variables that affect choice and (2) explore interactions of how choice variables may affect search. In running a website, a firm balances two potential streams of revenue: sales of products, services, or information content to visitors; and sales of advertising space to other organizations offering their own promotional messages. Web-based businesses thus operate in “two-sided markets,” selling something of value to visitors and selling visitors’ attention. The question of how much advertising to sell affects both sides of the market – visitors may find the website less appealing with more advertising – and is therefore both complex and important. In this paper, we empirically test the competing theories about the determinants of ad quantities on websites. Recent theoretical literature, namely Katona and Sarvary (2008) [KS]; Godes, Ofek, and Sarvary (2009) [GOS]; and Kind, Nilssen, and Sorgard (2009) [KNS], have some contradictory predictions, and our goal is to resolve the debates through a data-driven analysis. We focus on three issues. Do sites that have more competition devote more space to advertising? GOS say no and KNS say perhaps yes. Do sites that have more traffic devote more space to advertising? KS say no, KNS say yes, and GOS say it depends. And how do the answers to those questions change depending on the business model of the site, i.e., whether it is an advertising-only business model vs. a hybrid of the two streams? We study these questions with a large sample of websites taken from the top million sites as ranked by quantcast.com. ■ TC02 Legends Ballroom II Online Advertising - I Cluster: Internet and Interactive Marketing Invited Session Chair: Laura Kornish, Associate Professor, University of Colorado-Boulder, Boulder, CO, United States of America, [email protected] 1 - How Do Advertising Standards Affect Online Advertising? Avi Goldfarb, University of Toronto, Toronto, ON, Canada, [email protected] The technological transformation and automation of delivery methods has revolutionized the advertising industry. However, increasing reliance on technology has also led to requirements for standardization of advertising formats. This paper examines how the memorability of banner advertising changed with the advent of new standards regularizing the size of display advertising. Using data from randomized field tests, we find evidence that for most ads, recall of banner advertising declines as a result of standardization. Because we find that the decline is much weaker when a standardized ad is the only ad on the page, and when the ads appear to be more original, a likely explanation is that standardization makes it harder for basic ads to distinguish themselves from their competition. ■ TC03 Legends Ballroom III Internet: Social Influence Contributed Session Chair: Jun Yang, Assistant Professor, University of Houston Victoria, School of Business, 14000 University Blvd, Sugar Land, TX, 77479, United States of America, [email protected] 1 - Successful Social Networkers: Impact of Activities and Network Positions Lucas Bremer, University of Kiel, Department of Marketing, Kiel, 24118, Germany, [email protected], Florian Stahl, Asim Ansari, Mark Heitmann 2 - Internet Display Advertising and Consumer Purchase Behavior: Do Ad Platforms Matter? Paul Hoban, PhD Student, Marketing Area, UCLA Anderson School, 110 Westwood Plaza, Los Angeles, CA, 90059, [email protected], Randolph Bucklin The growth of online social networks and the decreasing effectiveness of traditional marketing have lead to a surge of interest in using such networks for marketing purposes. Actors embedded in such social networks are interested in understanding how to leverage the interactions and relationships among network members to increase their popularity and garner online success. In this paper we analyze how the network activities that an actor engages in influences the actor’s egocentric positions within the network and how such activities in combination with these network positions eventually impact marketing success. In particular, we study how a network of music artists can drive music downloads by engaging in active communication and building of friendship activities. Our sample consists of a set of 440 music artists who actively operate profiles on two independent online social network platforms at the same time. Personal network information on both platforms is tracked monthly over a period of one year. As network positions are endogeneous, and instrumental variables are not available readily, we use a latent instrumental variable approach to model the endogeneity of multiple network influence variables. Our results indicate that online success is determined by both the social network structure and networking activities of the music artists rather than by their outside popularity. Most importantly, the drivers of online success are not limited to the size of artist’s personal network. The findings of our study provide several insights into the use of personal online social networks for promoting products and services online. Total spending on Internet advertising is projected to surpass $100 billion by 2014, making it the second largest advertising medium behind TV. Display advertising (a.k.a. banner ads) is now an integral component of many online marketing campaigns. However, considerable uncertainty exists regarding the effects of online display media on consumer purchase behavior. This has made many managers reluctant to invest greater sums to date. Academic research on the effects of display advertising has also been limited, constrained by access to data and modeling challenges. Existing studies on display advertising have focused on brand affect, stated purchase intent, click through rates, site traffic, and purchase acceleration by existing customers. A limitation of most prior work is that all display ads are treated as equivalent. Among other things, this ignores the potential effect of ad platform. An ad platform determines where and how a display ad is served, how it may be targeted, and the cost to the advertiser. Using a unique dataset from a financial services firm and a continuous time hazard model, we examine the relationship between display ad impressions and individual purchase behavior while accounting for differences across advertising platforms. We do so for both potentially new and existing customers using a very large sample of long-lived tracking cookies for a twomonth period in 2010. Preliminary results show significant differences in ad effectiveness across platforms, despite identical ad copy. For advertisers, our results suggest that the costs associated with each platform can be weighed against effectiveness, leading to potentially more productive spending allocations. For ad platforms, there are implications for pricing and differentiation strategies. 2 - The Evolution of Switch Customers in E-Commerce: Understanding When and How Customers Switch Fan Zhang, Washington University in St.Louis, One Brookings Drive, Campus Box 1133, St.Loius, MO, 63130, United States of America, [email protected], Tat Chan, Qin Zhang 3 - The Effect of Banner Exposures on Memory for Established Brands Titah Yudhistira, University of Groningen, Department of Marketing, Nettelbosje 2, Groningen, 9747AE, Netherlands, [email protected], Eelko Huizingh, Tammo Bijmolt With growing customer bases, many manufacturers start selling products directly to their customers, after initially only selling products through retail stores. This is particularly common in the context of online business. For manufacturers, selling through retailers has the trade-off between reaching more customers and receiving lower profit margins, particularly from the repeated customers. It is beneficial for manufacturers to know how they can utilize the existing customer base of the retailers to build its own customer base who shop directly from them. We propose a model to study this diffusion process. Particularly, we focus on the roles that the network effect and product assortments play on this process. We apply the proposed model to individual level purchase data from a Chinese online company which sells premium fresh fruit in its own online store after initially only selling the products through online retailers. We find that product assortment is a main factor for customers to switch from shopping at the online retailers to directly at the store of the manufacturer. We also find significant network effect in this diffusion process. Although the focus of banner advertising has moved from click-through to brand building, there is little evidence of the effectiveness of banner advertising on brand memory for established brands. The purpose of our study is to provide new evidence on how banner exposures affect ad and brand memory as well as to propose and test a mechanism on how online exposures benefit established brands. Our study is unique since we (i) focus on established brands, (ii) emphasize the difference between ad memory and brand memory, and (iii) study the differential effects of three exposure variables, i.e frequency of exposures, time difference between the last exposure and measurement, and average time between exposures (spacing). We base our analyses on massive empirical data, i.e surveys and exposure data from 59.370 individuals in ten countries that were exposed to actual internet campaigns of 26 well-known brands. We find that exposure frequency and spacing have a beneficial effect and that time difference between the last exposure and measurement has a detrimental effect on ad memory. On the contrary, no effect of the three exposure variables on brand memory is found. However, the same effects of exposure frequency and time difference on brand memory are found by including mediating effects of ad recall and recognition. As a practical implication, this study shows that established brands can benefit from banner advertising by advertising more continuously, while assuring enough space between subsequent exposures. 19 TC04 MARKETING SCIENCE CONFERENCE – 2011 3 - Is it a Fad or Necessity? Measuring the Effectiveness of Social Media on E-tailers Jun Yang, Assistant Professor, University of Houston Victoria, School of Business, 14000 University Blvd, Sugar Land, TX, 77479, United States of America, [email protected], Jungkun Park inferential statistics, a Bayesian decision perspective, the classical scientific method, and the modern machine learning theory perspective. The paper is available from the author. 4 - Regression Discontinuity with Unobserved Score Sridhar Narayanan, Stanford University, 518 Memorial Way, Stanford, CA, 95014, United States of America, [email protected], Kirthi Kalyanam The rapid development of online communities and social networks has dramatically changed the way how marketers usually work. Consumers actively write reviews and share their experiences on social networking sites such as Facebook, Youtube and Twitter. Instead of treating consumers as passive information receivers, marketers begin to consider consumers as active co-producers of content and information. Marketing strategies such as ‘seeding’ campaigns in online communities and firms’ participation in social network sites are commonly adopted in practice. Though social media has recently become the spotlight of both practitioners and academies, there is limited research to study its impact on firms, especially on retailers. Furthermore, since the beginning of ecommerce, many studies have been conducted to identify the key determinants of successful e-tailing. However, there seems to be lack of consensus on some of the factors. This study tries to fill these research gaps. We have collected a dataset on the top 500 e-tailers. The dataset contains detailed information on the features provided by each e-tailer, and its financial performance. We seek to answer the following research questions. First, we empirically identify the determinants for a successful e-tailer, since there is still lack consensus in the literature. Second, after controlling those determinants, we would like to measure the influence from social media on those e-tailers’ performances. Furthermore, we want to test whether those influences will differ across various channel structure types and across different product categories. Empirical results and managerial implications are included. Regression Discontinuity (RD) designs estimate treatment effects in situations where the treatment is based on an underlying continuous ‘score’ variable, with treatment taking place if the score crosses a discete threshold and not otherwise. The local average treatment effect is measured by comparing outcomes for subjects whose scores are just above and just below the threshold, with the latter serving as a control group for the former. We extend the scope of RD to contexts where the score or the threshold are not fully observed, but only components of the score, or covariates that explain treatment are observed. The method involves estimating scores using a first stage empirical approximation, which involves fitting a binary choice model for treatment as a function of observed score components or other covariates. Next, the outcomes for individuals with estimated score just above the threshold are compared with those just below the threshold to obtain the treatment effect, as in a standard RD approach. Since a binary choice model has an inbuilt threshold of zero, this approach works even when the actual threshold on the true score function is unobserved. We show analytically the conditions under which the method uncovers a valid treatment effect. To demonstrate the methodology, we use a casino marketing setting where the exact scores used by the casino to decide on the treatment (offers mailed to consumers) are observed. We use a Monte Carlo simulation to establish the empirical conditions required to estimate the treatment effect. Finally we show an application to pharmaceutical detailing, where the scores are unobserved. The estimates using our our proposed approach generate new insights that add to the literature. ■ TC04 Legends Ballroom V Econometric Methods I: General ■ TC05 Contributed Session Legends Ballroom VI Chair: Sridhar Narayanan, Stanford University, 518 Memorial Way, Stanford, CA, 95014, United States of America, [email protected] 1 - A Cigarette, a Six Pack or Porn? The Complementarity of Vices Rachel Shacham, Carlson School of Management, University of Minnesota, 321 19th Avenue South, Suite 3-150, Minneapolis, MN, 55401, United States of America, [email protected], Peter Golder, Tulin Erdem New Product III: Adoption Contributed Session Chair: Mark Ratchford, Assistant Professor of Marketing, Vanderbilt University, Owen Graduate School of Management, 401 21st Avenue South, Nashville, TN, 37203, United States of America, [email protected] 1 - An Investigation of Scales for Consumer Innovativeness Masataka Yamada, Professor, Nagoya University of Commerce and Business, 4-4 Sagamine, Komenoki-cho, Nissin-shi, 470-0193, Japan, [email protected], Toshihiko Nagaoka Using statistical copulas, we develop an empirical model that allows us to study the levels of complementarity between different vices while accounting for self-selection. The method developed in this study is particularly well suited to the issues that occur while studying vice (in contrast to other types of goods). In particular, we allow the level of complementarity to differ between addicts and other users. We estimate the model using a unique dataset that contains detailed individual-level data over time. Last year at Cologne, we reconsidered not only the consumer innovativeness but also the theory of innovation diffusion itself focusing on the construct of consumer innovativeness. We proposed a new way of viewing the diffusion of innovation using Carnap’s theoretical construct and disposition concept. We introduced an intermediate level of abstraction construct between theoretical construct and disposition concept. We name it T-D mixture. This study reviews the measurement scales of each level of innovativeness. Then we reposition some of the scales as the scales for the T-D mixture. We can expect better predictions of actual adoption behavior by these scales. 2 - Handling Endogenous Regressors by Joint Estimation using Copulas Sungho Park, Assistant Professor, Arizona State University, P.O. Box 874106, Tempe, AZ, 85287, United States of America, [email protected], Sachin Gupta We propose a method to tackle the endogeneity problem without using instrumental variables. The proposed method models the joint distribution of the endogenous regressor and the error term in the structural equation of interest (the structural error) using a copula method, and makes inferences on the model parameters by maximizing the likelihood derived from the joint distribution. Using a series of simulation studies and an empirical example, we show that the proposed method captures the dependence structure between the endogeneous regressor and the structural error well enough to overcome the endogeneity problem. Other properties of the proposed method are discussed. 2 - A Multivariate Analysis of Pre-acquisition Drivers of Technology Adoption Mark Ratchford, Assistant Professor of Marketing, Vanderbilt University, Owen Graduate School of Management, 401 21st Avenue South, Nashville, TN, 37203, United States of America, [email protected], Jeffrey Dotson This study develops and empirically tests a new parsimonious multiple-item scale to measure consumers’ propensities to adopt new technologies. We show that a consumer’s likelihood to embrace new technologies can reliably be measured by a 14item index that combines assessments of consumers’ positive and negative attitudes towards technology. Consistent with prior work on technology readiness, we show four distinct dimensions of consumers’ technology adoption propensity: two inhibiting factors, dependence and vulnerability, and two contributing factors, optimism and proficiency. We develop the index on a cross-sectional dataset of U.S. consumers then establish the validity of each of the four component scales on two dissimilar validation sets. Next, using a multivariate probit model, respondents’ scores on the resulting index sub-scales are combined with demographic factors to examine the antecedents of adoption of a range of technologies. Distinct segments of adopters emerge. The results are of interest to practitioners interested in learning about the drivers of technology adoption for new products. 3 - Improving Predictive Validation Steve Shugan, University of Florida, 2030 NW 24th Avenue, Gainesville, FL, 32605, United States of America, [email protected] Predictive validation is very popular in management science and marketing science for many reasons including the scientific principle that valid theories should make valid predictions. However, when predictive validation and testing use the same data, it has been shown that invalid theories can imply incorrectly specified models that defeat predictive validation. Moreover, wrong models can out-predict the true model (that represent the actual data generating process), both in-sample and out-ofsample. Consequently, focusing only on fit or predictive validation can result in wrong implications. For example, as shown in the paper, better predicting response functions often underestimate optimal expenditures and imply wrong strategies. This paper shows (analytically and through simulation) that combining several predictive metrics allows detection of wrong response functions better than all popular extant metrics (absolute-mean-error, square-error, likelihood, and so on). The analytical proofs show how to improve predictive validation when those wrong response functions employ dampening. Dampening is a special type of bias that (as shown in the paper) allows wrong response functions to out-predict the true response function that represents that actual response to the decision variables (e.g., price, advertising, etc.). The simulations show how to improve predictive validation in the general case. The paper also discusses the interpretation of predictive validation in terms of 20 MARKETING SCIENCE CONFERENCE – 2011 ■ TC06 TC07 4 - Dominance and Innovation in a Dynamic Macro Environment Vincent Mak, Judge Business School, University of Cambridge, Judge Business School, Trumpington Street, Cambridge, CB2 1AG, United Kingdom, [email protected], Jaideep Prabhu, Rajesh Chandy, Chander Velu Legends Ballroom VII Competition III: General Contributed Session One of the most widely contested issues in innovation research is whether dominant firms tend to be lethargic or pioneering with innovations. Both theoretical and empirical evidence is divided on this topic. In this paper, we propose one way to reconcile these differences. We build a game-theoretic model that highlights a key driver of innovation by dominant as well as less dominant firms: the macro environment faced by these firms. Specifically, the model focuses on how changes in market profitability due to exogenous changes in competing firms’ macro environment affect the innovation strategies of these firms. We show that in environments with rapidly declining profitability, dominant firms are likely to be lethargic and innovate incrementally, while in environments with rapidly growing profitability, dominant firms innovate radically and ahead of competitors. Surprisingly, we also find that a firm in an environment with rapidly growing profitability might innovate at a later time (if at all) and earn a lower expected profit when its probability of succeeding with a radical innovation increases. Finally, we examine a case in which the macro environment is profitable throughout, and find that in such situations a hybrid innovation pattern can emerge whereby the dominant firm innovates radically but late. Chair: Vincent Mak, Judge Business School, University of Cambridge, Judge Business School, Trumpington Street, Cambridge, CB2 1AG, United Kingdom, [email protected] 1 - What if Marketers Put Customers Ahead of Profits? Scott Shriver, Stanford GSB, 518 Memorial Way, Stanford, CA, United States of America, [email protected], V. “Seenu” Srinivasan We examine a duopoly where one of the firms does not maximize profit, but instead maximizes customer surplus subject to a profit constraint. The model assumes customer willingness to pay for quality is uniformly distributed and that customers follow a simple decision rule: when presented with two products of known quality and price, purchase one unit of the product which maximizes surplus, or if surplus is negative for both products, elect not to purchase any product. We further assume that firms’ marginal cost of production is convex (quadratic) in quality. Competition between firms is modeled as a two-stage game, which is solvable by backward induction. In the first stage, firms choose product quality levels sequentially, fully anticipating subsequent price competition. In the second stage, firms take qualities as given and choose prices simultaneously in accordance with a Nash equilibrium. Two possibilities are considered: (a) the first mover is the profit maximizing firm, and (b) the first mover is the surplus maximizing firm. We compare the results to the corresponding base case of Moorthy (1988), where both firms are profit maximizing. We find that firms can deliver significant additional value to their customers by forgoing small amounts of profit. The effectiveness of this strategy depends upon which firm is the first mover. In the case that the surplus-maximizing firm moves first, 5%-10% increases in customer surplus “cost” 10%-20% of potential profits. By contrast, when the profit-maximizing firm chooses quality first, we find that sacrificing 20% of profits is sufficient for the surplus-maximizing firm to more than triple the customer surplus it would have provided under a profit-maximizing objective. ■ TC07 Founders I ASA Special Session on the Marketing-Statistics Interface - III Cluster: Special Sessions Invited Session Chair: Michael Braun, Massacusetts Institute of Technology, 100 Main Street, Cambridge, MA, United States of America, [email protected] 1 - Customer Waiting Time and Purchasing Behavior: An Empirical Study of Supermarket Queues Andrés Musalem, Duke University, Fuqua School of Business, 100 Fuqua Drive, Durham, NC, 27708, United States of America, [email protected], Yina Lu, Marcelo Olivares, Ariel Schilkrut 2 - Gaining from Imitative Entry: Dynamic Durable Pricing with Rational Consumer Expectations Lu Qiang, City University of Hong Kong, ACAD P7701, 83 Tat Chee Avenue, Kowloon Tong,Hong Kong - PRC, [email protected], Wei-yu Kevin Chiang This paper investigates the impact of imitative entry on intertemporal pricing strategy of an innovator (brand-name company) selling a new durable in a two-period dynamic framework. While acting as a monopolist in the first period, the innovator faces competitive entry of an imitator in the second one. The market consists of loyal versus non-loyal segments of consumers who are heterogeneous in their valuation of product. The loyal customers buy only from the innovator. Before deciding when to buy, they rationally anticipate the pricing policy to be followed by the innovator who is unable to credibly commit in advance to the second-period price. On the other hand, the non-loyal customers are price-sensitive and only act in the second period. In making a purchase decision, they compare the benefit of buying from the innovator against that from the imitator. Contrary to the conventional wisdom, the equilibrium of a two-stage pricing game between the innovator and the imitator indicates that imitative entry may potentially benefit the innovator. We provide specific explanations and implications to this counter-intuitive finding by looking further into the dynamic pricing strategy of the innovator and the corresponding purchasing choice of the rational consumers. The design of services requires making decisions about service features, employee selection and training, staffing levels and standards of service quality. The traditional approach has been to design and manage services to attain a quantifiable service standard – for example, call centers are designed to guarantee that a fraction of their customers (say 95%) are served within a given waiting time (e.g., 30 seconds). These design decisions usually involve trade-offs between the costs of sustaining a certain service standard versus the value that customers attach to this level of service. In contrast with extant literature that focuses on subjective measures of service quality, in this paper we measure the effect of objective measures of service quality – in particular, waiting time of customers in a queue – on actual customer purchasing behavior. Accordingly we develop a methodology which can be used to attach a financial tag on the implications of having customers waiting for service. Our paper uses a novel data set which was collected through digital cameras and image recognition software. Conducting a pilot study in the fresh deli section of a large supermarket, we collected data on queue lengths and number of staff members providing service every 30 minutes during a 6 month period. In addition, we obtained point-of-sales (POS) data from all relevant transactions during this time period. The time-stamp of the POS transaction is used to link the snapshot data from the queue with the customer purchase data and data augmentation methods are used to account for the uncertainty about the exact queue length that customers experience. A key feature of the supermarket we study is that a large fraction of their purchases are from loyalty card holders. This is used to construct a panel data set of customer purchases, which enables us to control for customer heterogeneity in purchasing behavior using Bayesian methods. 3 - KFC and McDonald’s Entry in China: Competitors or Companions? Qiaowei Shen, Assistant Professor of Marketing, The Wharton School, University of Pennsylvania, 3730 Walnut Street, JMHH 700, Philadelphia, PA, 19104, United States of America, [email protected], Ping Xiao In this paper, we study the entry of McDonald’s and KFC, the two largest fast food chain restaurants in the world, in China, which is the world’s largest emerging market. Our data covers the entire history of entry by the two chains from 1987 when the first KFC outlet opened in Beijing up to year 2007, when more than 200 cities have KFC or McDonald’s or both. We are particularly interested in how the duopoly western fast food chains may influence each other in their decision to enter a new city or open additional outlets in an existing market (city). We find that the scale of rival presence as indicated by the cumulative number of outlets has a positive effect on a chain’s entry decision after controlling for local economic conditions, regional time-invariant unobservables and time trend. We find evidence that such positive effect is due to firm learning about uncertain market demand. Furthermore, the effect varies by time period (e.g. before 1999 vs. after 1999) and by city type (big city vs. small city). 2 - Forecasting Customer Purchase Rates Incorporating Temporal Variation Luo Lu, Stanford University, Sequoia Hall, Stanford, CA, United States of America, [email protected], Zainab Jamal When predicting the future purchasing patterns of customers, the expected number of transactions in a future period is an important factor in the “lifetime value” calculations for each individual customer. Among the extant models that provide such capabilities, the BG/NBD model stands out. Under this framework, a major assumption is that the number of transactions made by a customer follows a Poisson process with a fixed transaction rate. However, in the real world the transaction rates may vary based on time. For example, in our empirical data from an online digital sharing service, we found a “seasonal pattern” which indicates that the customers intend to make more purchases in summer and winter. We also found that the customers make less purchases as time goes by, indicating an “annual pattern”. We replaced the homogeneous Poisson process in the BG/NBD model by a nonhomogeneous one to include the seasonal and yearly factors in the transaction rates and used a Bayesian framework to estimate the model at an individual level. The results demonstrate the effectiveness of the proposed methods. 21 TC08 MARKETING SCIENCE CONFERENCE – 2011 2 - When do Firms Benefit from Alliance Specialization in Either Innovation or Marketing? Jongkuk Lee, Ewha Womans University, 11-1 Daehyun-dong, Seodaemun-gu, Seoul, Korea, Republic of, [email protected], Young Bong Chang 3 - Optimal Mailing in a Beta-geometric Beta-binomial (BG/BB) Model George Knox, Tilburg University, Tilburg, Netherlands, [email protected], Rutger Van Oest How should firms design an optimal dynamic mailing policy when customer response is governed by a “buy til you die” framework? Our starting point is the betageometric beta-binomial model (e.g., Fader, Hardie and Berger 2004), which assumes that customer response to mailings while “alive” is distributed with beta heterogeneity across the population. After each mailing, the customer may irreversibly transition into an absorbing “inactive” state (i.e., Bernoulli death process); this likelihood is also distributed across the population according to a Beta distribution. Whether a given customer is “alive” and how likely he or she is to respond to mailings are unknown quantities to the firm. We embed this model of consumer behavior in a Bayesian dynamic programming problem for the firm that incorporates firm learning, as well as unobserved customer heterogeneity and attrition. Sending a mailing not only results in an immediate gain or loss to the firm but also allows it to learn the “true” parameters governing the customer’s response and drop out behavior, which increase long-term profits. We derive the optimal policy that balances these two forces and provide evidence through simulation about how it changes with respect to response and drop-out rates. Technological innovation and marketing are at the core of firm strategies to create value for customers and appropriate the created value in the marketplace. Forming alliances is increasingly common for technological innovation and marketing activities. A firm has to determine its strategic focus in building is alliance portfolio, i.e., to what extent it forms alliances for innovation and to what extent for marketing. Previous studies have emphasized the superior benefits of balancing the two complementary tasks (i.e., innovation and marketing), compared to specializing in either one. However, such consensus may be somewhat premature and not necessarily logical in all contexts. In particular, the benefits of specializing alliances in R&D or marketing compared to balancing the two have yet to be examined. In response, this study investigates when firms would benefit from specializing alliances in innovation and when from specializing alliances in marketing. Particularly, we focus on how alliance specialization in either innovation or marketing interacts with a firm’s internal investment specialization in either innovation or marketing for the firm profitability. Contrary to the previous emphasis on the superior benefits of balancing innovation and marketing, this study reveals boundary conditions under which specializing both internal investments and alliance formation in either innovation or marketing can generate a positive interaction for the firm profitability. The results of this study highlight the value of specializing in either innovation or marketing across two levels, internal investment and alliance formation. 4 - Modeling Customer Lifetimes with Multiple Causes of Churn Michael Braun, Massacusetts Institute of Technology, 100 Main Street, Cambridge, MA, United States of America, [email protected], David Schweidel Customer retention and customer churn are key metrics of interest to marketers, but little attention has been placed on linking the different reasons for which customers churn to their value to a contractual service provider. In this article, we put forth a hierarchical competing risk model to jointly model when customers choose to terminate their service and why. Some of these reasons for churn can be influenced by the firm (e.g., service problems or price-value tradeoffs), but others are uncontrollable (e.g., customer relocation and death). Using data from a provider of land-based telecommunication services, we examine how the relative likelihood to end service due to different reasons shifts during the course of the customer- firm relationship. We then show how the effect of a firm’s efforts to reduce customer churn for controllable reasons are mitigated by presence of uncontrollable ones. The result is a measure of the incremental customer value that a firm can expect to accrue by delaying churn for different reasons. This “upper bound” on the return of retention marketing is always less than what one would estimate from a model with a single cause of churn and depends on a customer’s tenure to date with the firm. We discuss how our framework can be employed to tailor the firm’s retention strategy to individual customers, both in terms of which customers to target and when to focus efforts on reducing which causes of churn. 3 - The Financial Determinants of Adopting Radically Innovative Information Technology: An Empirical Anarchy Merle Campbell, Georgia State University, 5128 Vinings Estates Way SE, Mableton, GA, 30126, United States of America, [email protected], Tim Bohling, Maureen Schumacher, V Kumar The objective of this research study is to investigate the client-side factors associated with organizations that adopt radical innovation. Specifically, what financial and economic commonalities do firms posses that may influence their proclivity to adopt radical innovation? Previous studies examining an organization’s adoption of radical innovation identify isolated variables of interest and fail to deliver a comprehensive framework. Armed with an understanding of the financial and economic factors associated with radical innovation adopters, marketing managers can build and execute more targeted marketing strategies. The study was developed by building upon Katona’s (1951, 1979a,b) theory of psychological economics, as well as the existing literature regarding economic determinants of innovation adoption (Rosner, 1968). A survey was used to collect a comprehensive sample of both adopter and non-adopters of radically innovative IT – Cloud Computing. Economic and financial measures to operationalize the independent variables were then collected. A hierarchical logit model is used to identify the drivers that are significant in predicting the adopters and non-adopters relative to the latent constructs. The study results support a majority of the theoretical propositions and hypotheses that a difference in the two groups does exist. The results of this study are a necessary first step to further investigate the role that firm specific financial and economic factors have on the adoption of radical innovation. ■ TC08 Founders II Innovation III Contributed Session Chair: Merle Campbell, Georgia State University, 5128 Vinings Estates Way SE, Mableton, GA, 30126, United States of America, [email protected] 1 - The Chinese Knockoff Effect: How Do Consumers Perceive “Shanzhai” Cellphones? Shu-Chun Ho, Assistant Professor, National Kaohsiung Normal University, No.116 Heping 1st Rd. Lingya Dis., Kaohsiung, Taiwan ROC, [email protected], Huang Shang-Jui ■ TC09 Founders III Promotions III Contributed Session China’s electronics manufacturers have been trying to beat the global leading mobile phone manufacturers at its own game by rushing out all kinds of knockoffs. The Chinese knockoff cellphone, also called Shanzhai cellphone, gradually penetrates the global mobile phone market share and has created Chinese knockoff effect. Knockoff cellphone is considered as a disruptive innovation that offers relatively simple, convenient, and low-cost choice to consumers. The objective of this research is to explore the Chinese knockoff effect and how the consumers perceive Shanzhai cellphone. We conducted narrative interview with 12 consumers who have purchased and been using knockoff cellphones. In addition, we further interviewed 3 knockoff mobile phone distributors to unveil the Chinese knockoff effect. Our major findings are as following. First, price and brand preference are the major determinants when consumers plan to buy mobile phones. Second, consumers decide to purchase knockoff cellphones because they are curious about the novelty of Shanzhai cellphones. They want to try new and different technologies and consider Shanzhai cellphones as a real bargain. Third, the knockoff cellphone distributors suggest that functionality and innovation contribute to consumers’ preference of knockoff phones. However, the lack of well-known brand, reputation, and reliability also impede the acceptance of knockoff phones. Our findings provide practical insights to mobile phone manufacturers and distributors in design, research and development, and marketing strategies. Our findings also provide theoretical basis to investigate consumers’ perception of knockoff phones when they purchase mobile phones. Acknowledgement: Shu-Chun Ho thanks the National Science Council in Taiwan for partially supporting this study under Grant #NSC99-2410-H-017-017. Chair: Ty Henderson, University of Texas - Austin, 1 University Station B6700, Austin, TX, 78712, United States of America, [email protected] 1 - How Effective Are Conditional Promotions Tobias Langer, University of Hamburg, 5428 S Woodlawn Ave, Unit 3A, Chicago, IL, 60615, United States of America, [email protected], Kusum Ailawadi, Karen Gedenk, Scott Neslin The term “conditional promotions” refers to promotions that offer a discount to consumers if a condition is met. The condition may be within the control of the consumer or it may be an external event. This research focuses on the latter type, which is becoming increasingly common. For example, a German electronics retailer recently offered the following conditional promotion: buy a TV for at least 500 € before the European soccer championship and get 100 € cash back for every goal scored by the German team in the championship final. Like rebates, conditional promotions offer a delayed reward. But unlike rebates, the reward is uncertain. In order to compare the effectiveness of conditional promotions and traditional rebates, we conduct a choice-based conjoint experiment with two durables product categories. The attributes in the experiment are brand, quality, and promotion (no promotion, rebates and conditional promotions with different discounts), where the conditional promotions are based on a major sports event. We use the choice data to estimate consumers’ utility functions, where utility from a conditional promotion depends on consumers’ subjective probability that the event will occur, on their risk proneness 22 MARKETING SCIENCE CONFERENCE – 2011 and on the transaction utility they derive from the conditional promotion, for example from rooting for their favorite sports team. We use our model estimates to simulate market share effects for conditional promotions and rebates. Our results yield insights on when conditional promotions are successful and why. ■ TC10 2 - Coupon Expiration and Redemption Joseph Pancras, University of Connecticut, 2100 Hillside Road, Unit 1041, Storrs, CT, 06269, United States of America, [email protected], Rajkumar Venkatesan Contributed Session TC10 Founders IV Consumer Behavior Chair: Yi-Yun Shang, National Taipei University, 67, Sec.3, Ming-shen E. Rd., Taipei, 104, Taiwan - ROC, [email protected] 1 - My Brain is Tired. Can I Make Inference Spontaneously? Xiaoning Guo, PhD Candidate, University of Cincinnati, Department of Marketing, Cincinnati, OH, 45220, United States of America, [email protected], Inigo Arroniz Coupon redemption has been shown to be bimodal over the coupon validity period, with modes close to the start and expiration dates. This evidence has utilized aggregate coupon redemption and validity period dates. Using a new dataset of individual level coupon redemption, we demonstrate the ‘coupon expiration rate’ at the individual level. We build an individual level coupon redemption model to estimate the probability of coupon redemption at both the start and end of the validity period, and derive posterior probabilities of individual households of redeeming close to the start date as well as close to the expiration date. We conduct post-hoc analysis of the households which are more likely to exhibit coupon expiration redemption behavior by contrasting their categories purchased, demographics and campaigns from those who are more likely to redeem close to the start date. We discuss implications of these household level insights for retail store managers. Most purchases involve choices among options with incomplete attribute information. In such situations, consumers often have the option not to choose any of the alternatives to avoid uncertainty. Alternatively, consumers can make inferences about the missing attributes. These inferences may occur spontaneously. In this paper, the authors investigate the situational factor (ego depletion) and the personality factor (need for closure) to affect consumers to make inferences about the missing attributes. The findings show that consumers with high need for closure and highly depleted (or consumers with low need for closure and in the low ego depletion condition) are more likely to defer the choice. 3 - Gifts with a Gab: A Multivariate Poisson Analysis of the Effects of Gifts on Customer Acquisitions Sudipt Roy, Assistant Professor of Marketing, Indian School of Business, Gachibowli, Hyderabad, AP, 500032, India, [email protected], Purushottam Papatla 2 - Theories of Emotion in Consumer Behavior Khalil Rohani, PhD Candidate, University of Guelph, Marketing and Consumer Studies, Guelph, ON, N1G 2W1, Canada, [email protected], Laila Rohani, Joe Barth The subject of emotion has been gaining popularity among researchers in the area of consumer behaviour. Past research shows that researchers paid more attention to consumer’s rational behaviour; however, there are a growing number of studies dedicated to the emotion aspect of consumers. In this paper, two theories are introduced regarding emotion in consumer behaviour; Appraisal-Tendency theory and the Hierarchical consumer emotions model. The purpose of this paper is to find relations between consumers’ emotions and their consumption behaviour by exploring two different theories concerning emotions and consumers’ decision making behaviour. This paper first discusses definitions of major terms that are often used to describe consumers feeling such as mood, emotion, and affect. Second, Appraisal Tendency theory, related studies, and their findings are introduced. Third, the Hierarchical consumer emotions model and its empirical research are presented. Last, based on analysis of the two theories, propositions are proposed in order to predict specific consumption behaviour according to consumers’ certain emotional condition. We use a Multivariate Poisson count model to investigate whether receiving products of a firm as gifts can turn recipients into buyers of those products as well as other products from that firm. We use a count model since we examine the number of renewals of magazines by subscribers who first received those magazines via gift subscriptions. Additionally, we investigate the number of times that the recipients of gift subscriptions for some magazines initiate and continuation subscriptions to their sister magazines from the same publisher. Our analysis is based on a Multivariate Poisson analysis of counts of renewals of different magazines using link functions that include the receipt of gift subscriptions, household demographics and recipient lifestyles since the magazines that we analyze are lifestyle-related. Findings from this research are presented along with implications for additional research on gifts as alternative mechanisms of consumer word of mouth which has been has traditionally been defined as “informal, evaluative communication (positive or negative) between at least two conversational participants about characteristics of an organization and/or a brand, product, or service that could take place online or offline” (Carl 2006 p.608). We propose that gifts can be viewed as consumer brand advocacy (Keller 2007; Park and MacInnis 2006) where a consumer actively promotes the brand to other potential consumers. We also suggest that companies can encourage gifts of their products by current customers as an alternative to using sales promotions to reach prospective customers. 3 - Carrot or Stick? - Asymmetric Evaluation on Counterfeit Products under Different Self Construal Xi Chen, Assistant Professor, Business School, China University of Political Science and Law, 1903, #4,Wankexingyuan,Yangshan Rd, Chaoyang District, Beijing, 100107, China, [email protected] 4 - Promoting a Brand Portfolio with a Social Cause: Findings from an In-market Natural Experiment Ty Henderson, University of Texas - Austin, 1 University Station B6700, Austin, TX, 78712, United States of America, [email protected], Neeraj Arora This paper proposes that counterfeit products can be seen as representative of both return and risk. Consumers with different self construal exhibit asymmetric evaluation toward counterfeit product when exposed to uncertainty of return and risk – asymmetric sensitivity to financial risk and social risk under claim that emphasize on risk , and asymmetric sensitivity to financial return and social return under claim that emphasize on return. This paper shows that (a) when risk aspect of counterfeit product is accessible, consumers exhibit different sensitivity to different types of risks, namely the social risk and financial risk. Specifically, interdependent people are more risk averse toward social loss compared with financial loss, the reverse for independent people. (b) When return aspect is activated and accessible, consumers are asymmetrically sensitive to different gains, namely the social gain and financial gain. In particular, independent people consider financial return more than social return, the reverse for interdependent people.This phenomenon occurs because of different weight put on social and financial aspect by different self construal people, different self view becomes underlying antecedent of this asymmetric effect. And this is exhibited within a single counterfeit product because counterfeit product is representative of combination of both risk and return. What happens when a company links its brand to a social cause? Existing research provides process-level and experimental insight into this question. Given the popularity of promotion campaigns where purchases trigger an accompanying public good (e.g. Yoplait Pink Lids for Breast Cancer Research), there is a lack of research investigating the market outcomes of these embedded premium (EP) promotion campaigns. Our research examines the effects on primary and secondary demand of a nationwide in-market implementation of an EP promotion program across product categories. We leverage a specialized self-reported A.C. Nielsen panel dataset marker that classifies participating and non-participating households to quantify the effects of the EP intervention on purchase incidence, quantity and brand choice. The household-level hierarchical Bayes model accounts for participating/non-participating cohort differences and temporal shifts in brand preference to uncover the differential effect of the EP promotion launch in the market. We calibrate the model on four years of data from a national household purchase panel across different product categories. After accounting for temporal shifts in brand preference and buying behavior while also incorporating other marketing mix drivers, we find that the EPbrand association increases brand preference among the EP-participating households. We observe this cohort-conditional EP effect on brand choice amid significant aggregate changes in brand preference and buying behavior that impact both primary and secondary demand. We quantify the impact of the EP promotion relative to other elements of the marketing mix (e.g. coupons) and evaluate the financial implications of the EP strategy vis-á-vis other promotion options. 23 TC11 MARKETING SCIENCE CONFERENCE – 2011 4 - Remedying Reverse Self-control Effect of Hyperopic Consumers on Vicious Avoidance Yi-Yun Shang, National Taipei University, 67, Sec.3, Ming-shen E. Rd., Taipei, 104, Taiwan - ROC, [email protected], Kuen-Hung Tsai refers to businesses’ externally oriented intelligence-related activities and responsiveness simultaneously determine firm conduct and performance (Matsuno and Mentzer 2000). The hypotheses are tested over cross-sectional data collected from Turkey’s largest manufacturing companies, controlling for a number of factors such as firm performance prior to recession, and changes in real prices and wages. The purpose of this study was to remedy reverse self-control of hyperopic consumption and encourage hyperopic consumers to indulge themselves. We examine the behavioral effect of hyperopic consumers on vicious avoidance through the perspective of emotional accounting and the reinforcement theory to remedy hyperopic consumption and achieve behavioral reshaping. Findings – The study results show that positive emotional accounting induces hyperopic consumers to start mental operations before decision-making, which gives them a “Righteous Reason” to choose vice and generates the “Laundering Effect”, which helps hyperopic consumers alleviate the guilt of choosing vice and obtain the rationality and legitimacy of such choices. This makes hyperopic consumers brave enough to choose vice (indulgence) and realizes the aim of remedying the reverse self-control of hyperopic consumers. Implications – The study found that implementing remediation after consumers already have a basic plan is more effective than implementing one before consumers have developed a shopping concept. Give a righteous reason to hyperopic consumers to choose vice: to provide righteous reasons help hyperopic consumers alleviate their feelings of guilt over choosing vice, thereby changing their purchase decisions to more luxurious and hedonistic ones and improving customer satisfaction while bringing more benefits to firms. Originality/value-This study was to expanding the application of the operational conditioning theory into the study of consumer behaviors with the intention of shaping behaviors by remedying the reverse selfcontrol problem of hyperopic consumers, helping hyperopic consumers bravely unfetter themselves and expanding the application of the operational conditioning theory. 3 - A Study on the Effectiveness of Emotional Versus Rational Appeals on Consumer of Eastern India Sabita Mahapatra, Assitant Professor, Indian Institute of Management, Pitampur, Rau, Indore, MP, 453331, India, [email protected] ■ TC11 ■ TC12 The prerequisite of a successful advertisement message is exposure and interpret of ad message in the way the advertiser wants. The message’s impact depends not only upon what is said but also on how it is said. There is a slow down in advertising, and worse still, the relevance of mass media advertising is coming under scrutiny like never before due to other non conventional medias. Besides memorability consumer response to ad is considered as the most important parameter to measure effectiveness. The present study tries to find out the effectiveness of emotional versus rational appeals used in TV ads on the consumers of eastern parts of India. The finding of the study emphatically proved that the emotional appeals used in TV ads had positive effect on consumers for established product even if they are expensive, also for new and less expensive product category. Though for new expensive product category emotional appeals does not play that dominating role for respondents from various demographic and psychographics segment. The investment in TV ads is of gigantic proportion. Such mammoth investments must be justified. The present study hopes to provide direction to the advertising agencies in the matter of allocation of and development of ad message. The finding from the present study hopes to provide a cue to agency people as well as creative departments in the agencies for a proper mix of appeals taking in to consideration various stages and category of products along with demographic and psychographics of the audience targeted. Champions Center I Champions Center II Advertising Strategy Brand Equity Contributed Session Contributed Session Chair: Sabita Mahapatra, Assitant Professor, Indian Institute of Management, Pitampur, Rau, Indore, MP, 453331, India, [email protected] 1 - The Impact of Advertising on Brand Trial in Experience Good Markets Raimund Bau, RWTH, Moritz-Sommer-Str 4, Duesseldorf, Germany, [email protected] Chair: U.N. Umesh, Professor, Washington State University, Classroom (VCLS) 308S, 14204 NE Salmon Creek Avenue, Vancouver, WA, 986869600, United States of America, [email protected] 1 - Improving the Image of Countries, Cities and Tourist Destinations, Using Media and Branding Strategies Eduardo Oliveira, Research Assistant, University of Minho, Rua Comandante Luis Pinto Silva 47 6∫ Es, Povoa de Lanhoso, 4830-535, Portugal, [email protected] “Draw new consumers to the brand” is a popular line in brand manager’s target setting dialogues in FMCG markets. Generally, advertising is considered a tool to accomplish this. Consequently, leading consumer good firms employ ad-hoc market research tools to ensure that advertising has a “shift” effect, which is essentially the transfer of non-buyers into the pool of buyers. Especially in experience goods markets a distinction between two types of consumers has to made: those who have tried the brand before and those who have no experience with the brand. Consumers of the latter group face a distinctively different situation than the former group when considering to buy the brand in question. They have higher levels of insecurity regarding brand attributes which are unobservable prior to the purchase. Hence, many practitioners believe that advertising can overcome brand trial aversion. We use more than 7 years of household panel data on two categories of durable fast moving consumer goods to estimate a random parameter logit model. Upon estimation we test hypotheses on if and how advertising shifts the demand curve of consumers by impacting the disutility associated with brand trial. We find evidence that advertising is effective in lowering the disulitity of the risk associated with a trial purchase. In order to draw managerial implications from this insight we simulate the financial impact of advertising investments compared to a couponing strategy using a realistic cost and revenue scenario for firms in both markets. Countries, states, regions, cities and tourist destinations face increasing competition when they try to attract tourists, inhabitants, investments. Marketing is a universal and dynamic process that can be applied to developing, promoting and improve the image of entities, including products, services, places, personalities, properties, causes, and information. In the case of marketing or branding places, as countries, cities or destinations, informal marketing tasks and more evaluated communications channels has gone on for centuries. The place managers have to be able to develop their selfpromotion strategies to reach the marketing level of the capital and human resources. A clear definition of place branding, developed in a consistent and cooperative way with stakeholders and as a product of a network between local, regional and national stakeholders, can promote the production base, increasing the touristic and human resource flows. The current literature offers an extensive discourse in the field of crisis in general and the role of the marketing, branding and media strategies during crisis in particular. The challenge to the place marketers, place marketing researchers and decision makers are pro-active management approaches, in an ethically and environmentally sustainable perspective with creative media strategies. All these “place stakeholders” needs to find the best strategies, with inspiration and innovation to minimize the consequence caused by an unstable financial system, the economic instability and the natural events in the place’s image and in the decrease in the flow of people and capital. 2 - Advertising during Recession: Role of Industry Characteristics, Strategy Type, and Market Orientation Peren Ozturan, PhD Candidate, Koç University, Rumelifeneri Yolu CASE Sariyer, Istanbul, Turkey, [email protected], Aysegul Ozsomer 2 - Competitive Advantage through Internal Branding Constituents: Developing Critical Component Framework Anurag Kansal, Doctoral Student, Indian Institute of Management, Indore, IIM I, Indore, Indore, MP, India, [email protected], Prem Dewani Past research has shown that countercyclical (vs. procyclical) advertising spending leads to increase in organizational performance (Tellis and Tellis 2009). Although the consequences of cyclical advertising have been investigated to some extent, we lack knowledge on its antecedents. That is, why and under what circumstances are some firms more likely to respond to economic recessions by changing their advertising spending and what are the performance consequences of doing so? Specifically, the authors examine role of industry characteristics (level of competition and turbulence), business strategy type and market orientation as potential antecedents of the advertising spending-performance relationship. The predictions are built on two alternative schools of thought i.e. the Structure-Conduct-Performance paradigm (Bain 1951), positing that firm conduct is determined by the environment of the firm and Contingency Theory (Chandler 1962), asserting that managerial perceptions of an environment vary; therefore firms operating in the same environment end up implementing different strategies which lead to different performance goals. According to this latter perspective, business strategy types (Miles and Snow 1978) that treat strategic orientation as a planned pattern of organizational adaptation to the perceived environment and market orientation (Kohli and Jaworski 1990) which Increasing Globalization and advent of main stream consumerism has led to a saturated market place resulting in paucity of available external resources. This has forced marketers to look inwards for attaining a competitive edge in an increasing cut-throat marketplace. While a lot of research has been conducted in both external and internal branding processes, this paper aims to provide the marketer with a comprehensive framework that not just enhances the use of internal marketing and helps him gain significant competitive advantage, but also helps in identifying various critical components and cogs that are responsible for creation and implementation of an effective Internal Branding process. It uses Focus Group Discussions to compile a list of factors, and follows up with a survey of 238 respondents and culminates with a series of Expert Interviews. Major findings included the significant role of employee perception in the successful implementation of Internal Branding process, and a list of seven significant factors that are integral to the critical component framework. A strong link between the Internal Brand and the external corporate identity was also found in the analysis. The findings and the critical component framework that has 24 MARKETING SCIENCE CONFERENCE – 2011 TC13 engagement is identified by the Marketing Science Institute (2010) as one of the priority research topics for the coming years. For companies, customer engagement can have both a bright and a dark side: For example, engaging customers in the cocreation of a new market offering can be the beginning of a success story, while a negative customer experience can easily reach tons of people through online wordof-mouth. Therefore, a growing number of firms pursue strategies to steer customer engagement and initiate customer engagement themselves (Verhoef, Reinartz and Krafft 2010). In this paper, we investigate the market valuation of firm’s initiatives to steer customer engagement. We apply the event study methodology, which has as benefits that it allows for isolation of the effect we are interested in and is a forward looking performance metric (Geyskens, Gielens, and Dekimpe 2002). We aim to assess the relative effectiveness of various types of customer engagement, such as customer co-creation and word-of-mouth. We also assess the impact of different contexts, such as different industries (e.g. business-to-consumer vs. business-tobusiness), and the type of offering (e.g. products vs. services). Initial empirical results show heterogeneity in the investors’ response to different types of customer engagement initiatives. been developed would help marketers to maintain a checklist during the planning, coordinating as well as the implementation process to achieve an effective internal brand. The paper attempts to provide both inside-out and outside-in perspective which allows for a more robust framework, as well as a strong base for the development and evolution of an effective communications strategy. 3 - Conceptualization and Development of Scale for Power of Brand in a Brand-consumer Relationship Roopika Raj, Doctoral Student, Indian Institute of Management, Ahmedabad, Dorm 2, Room 18, IIMA Old Campus, Vastrapur, Ahmedabad, 380015, India, [email protected], Abraham Koshy Marketers are developing understanding of measures of brand-consumer relationships like love, trust, commitment and attachment to develop strategic and tactical initiatives to ensure that consumers are satisfied with their brand. These measures have been borrowed from the interpersonal relationship literature. And the motivation for this lies in the recent understanding from research in the area of brand-consumer relationships that consumers form relationships with brands the same way they form relationships with each other in a social context with other humans. But research on interpersonal variable of power of brand in the brand consumer dyad has been neglected. The vast social psychology and organizational behavior literature indicates that power is at the heart of understanding the dynamics of social and personal relationships. The paper is an attempt to understand and conceptualize the nature of power of brand and its bases in a brand-consumer relationship context. A total of thirty in-depth interviews were conducted with customers about their relationships with brands to arrive at a typology of power bases of a brand: reinforcement power, customary power, expert power, positional power and associative power. Further it also helps to construct and develop a scale for measuring power of brand on the various bases identified through interviews. A consumer might attribute power to brand on more than one bases hence the scale would be an important tool for brand managers to measure the overall power attributed by consumers to their brand. 2 - Impact of Price Change on Profitability: Theory and Empirical Evidence Vinay Kanetkar, University of Guelph, Guelph, ON, Canada, [email protected] How does one determine the effect of price changes on a firm’s profitability? In this paper, literature on this issue is reviewed. The econometric literature suggests that the effect of price on profitability is small but statistically significant. Alternative analytical and statistical models lead to novel approach to predict effect of price, variable cost and quantity changes on firm’s profitability. As a by-product of this work, it is possible to estimate price elasticity for a firm using publicly available income statements. 3 - The Value Relevance of Marketing Expenditures Min Chung Kim, Hong Kong Polytechnic University, Kowloon, Hong Kong - PRC, [email protected], Leigh McAlister 4 - Patent Data and Marketing Science U.N. Umesh, Professor, Washington State University, Classroom (VCLS) 308S, 14204 NE Salmon Creek Avenue, Vancouver, WA, 98686-9600, United States of America, [email protected], Monte Shaffer Because firms do not publicly report marketing expenditure, most studies considering the link between firm value and marketing use advertising (which is publicly reported for many firms) as a proxy for marketing. We extend those studies in two ways. First, we broaden the proxy for marketing by considering both advertising and salesforce. Second, we offer an explanation for the fact that some studies linking advertising to firm value have found a positive relationship while others have found a negative relationship. The accounting literature suggests that the link to firm value for both unexpected salesforce expenditure and for unexpected advertising expenditure should be negative. We confirm the accounting-hypothesizedrelationship for salesforce expenditure, but find a contingent relationship for advertising expenditure. Firm value and unexpected advertising expenditure are negatively related for firms that advertise below the advertising response threshold, but they are positively related for firms that advertise above that threshold. Perhaps because this contingent relationship is hard for analysts to learn by observation of the stock market, analysts ignore value-relevant advertising expenditure information when they forecast firm value. There is a large amount of interest in conducting research on patents by academicians and industry analysts each year. The late Zvi Griliches (1990) had a vision to make patent data publicly available for academic researchers to utilize. He and his colleagues have made attempts to provide patent data in a useful form (Hall 2001). Although well-intentioned, academic practitioners are looking for much more from patent data than what these current ‘one-off’ datasets can offer, i.e., they are neither comprehensive nor contiguous. To meet the need for a data source that assists in conducting research efficiently, we propose to first identify what variables knowledgeable researchers in the field want in terms of form and structure. This step will be done by using a survey and in-depth interviews of the top researchers in the fields of marketing, economics, technology and related topics. We then propose to organize the data on millions of individual patents into relational tables based on all possible variables that we can parse from the USPTO’s machine-readable data already harvested from the public-domain source (USPTO). In addition, we propose to generate additional tables that will enable researchers to merge this dataset with external databases that deal with finance, economic-statistics, firm-level performance and valuation of patents. We finally propose to engineer a solution to make all of these new variables available to the academic community as well as the business community at the end of each future calendar year (ongoing: 2012, 2013, and so on). We discuss the importance of the project, solicit feedback and suggestions on what kind of data to include, and propose to write a Marketing Science data paper delineating the resulting data. 4 - Assortment Diversification in the Retail Industry: The Impact on Market-based and Accounting-based Performance Timo Sohl, University of St. Gallen, St. Gallen, Switzerland, [email protected], Thomas Rudolph Over the last decades, many grocery retailers have increased their sales share accounted for by non-food products. Despite its high importance for retail firms, the literature has yet been scarce on describing performance implications of assortment diversification into food and non-food retailing. Based on the literature on market power advantages, we state that retail firms who diversify their assortments can benefit from an increasing sales-based market share. Furthermore, drawing from the resource-based view of diversification, we propose that merchandise management skills are capabilities that may be used successfully across different product lines within food or non-food assortments, but may be difficult to leverage across food and non-food assortments. Thus, while we predict a positive relationship between assortment diversification and market share, we hypothesize that the relationship between assortment diversification and profits has its optimum where an increasing marginal cost curve exceeds the marginal sales-based benefits curve. To test our hypotheses, we use a unique dataset of the world’s leading retailers’ assortment diversification behavior over thirteen years (from 1997 to 2009). Results show that retailers’ market share first decreases as they start to diversify their assortments and than increases at an increasingly higher level as they focus more heavily on assortment diversification. Conversely, we find the predicted inverted U-shaped relationship between assortment diversification and profits. Consequently, our findings indicate a trade-off between the maximization of accounting-based and market-based performance measures. Since increasing market share is an important objective of many marketing departments, our findings have important implications for academics and practitioners. ■ TC13 Champions Center III Quantifying the Profit Impact of Marketing III Cluster: Special Sessions Invited Session Chair: Xueming Luo, Eunice & James L. West Distinguished Professor of Marketing, The University of Texas at Arlington, Arlington, TX, United States of America, [email protected] 1 - The Market Valuation of Company Initiated Customer Engagement Sander F. M. Beckers, University of Groningen, Groningen, Netherlands, [email protected], Jenny van Doorn, P.C. (Peter) Verhoef In the light of recent societal developments, for example with respect to information technology and associated popular new social media (e.g. Facebook and Twitter), customer engagement (i.e. non-transactional customer behavior, Van Doorn et al. 2010) has become a very important topic for companies (Verhoef, Reinartz and Krafft 2010). Noteworthy, customer engagement is seen as a ‘new perspective in customer management’ (Verhoef, Reinartz and Krafft 2010) and research on customer 25 TC14 MARKETING SCIENCE CONFERENCE – 2011 ■ TC14 ■ TC15 Champions Center VI Champions Center V Consumer Responses to Pricing CRM I: Customer Lifetime Value Cluster: Special Sessions Invited Session Contributed Session Chair: Zainab Jamal, Hewlett Packard Labs, 1501 Page Mill Road, Palo Alto, CA, United States of America, [email protected] 1 - Payments as a Virtual Lock-in: Customers’ Profitability over Time in the Presence of Payments Irit Nitzan, Tel Aviv University, The Recanati Graduate School, of Business Administration, Tel Aviv, 69978, Israel, [email protected], Barak Libai, Danit Ein-Gar Chair: Anja Lambrecht, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, [email protected] 1 - Starting Prices as Catalysts for Consumer Response to Customization Marco Bertini, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, [email protected], Luc Wathieu In this project, we demonstrate how customer defection is affected by the mere use of a payment mechanism. We show that offering customers an opportunity to pay over time leads them to perceive payments as switching costs. As a result, customers change their defection behavior, and consequently their lifetime value to the firm. We use a combination of aggregate data from a cellular company and experiments to exhibit how customers may perceive payments as switching costs, and how this changes their behavior and profitability over time. Our results are very relevant for the current public discussion on customers’ switching costs in industries such as the cellular industry, and are of key importance to firms given the immediate effect of retention on the bottom line. Consumers often make decisions about products that require a certain degree of customization. From the perspective of the firm, providing customization is costly, but it can lead to greater value creation and surplus extraction. However, success of a customization strategy is contingent on the presence of customers who are responsive and prepared to pay for this benefit. The goal of this research is to propose and test a theoretical link between the presence of a starting (base) price and consumer engagement in customization: starting prices make consumers realize that the good has a strong component of customization. Specifically, we propose that starting prices act to split the total expense into a common component and a component that purchasers can attribute to their idiosyncratic tastes. This perception catalyzes consumer response to customization, as it causes positive process evaluations and a greater willingness to pay for the difference between the starting price and the final price. 2 - A New Model Proposal to Churn Management Omer Faruk Seymen, Sakarya University, Sakarya University Esentepe Kampusu UZEM, Sakarya, Turkey, [email protected], Abdulkadir Hiziroglu 2 - Free vs. Fee: Pricing of Online Content Services Kanishka Misra, London School of Business, Regent’s Park, London, NW1 4SA, United Kingdom, [email protected], Anja Lambrecht Predicting customer churn in the scope of relationship management has been receiving great attention by companies. Acquiring a new customer costs five times more than retaining an existing customer. To reduce the retention cost, prediction of churners has to be as accurate as possible so that retention campaigns and allocating the resources should be appropriate. In churn management, the majority of these models are utilized to predict churn customers on the next period. In this study, a new churn model is proposed within this context that not only next period churners can be detected but possible churners in further periods who could be seen “loyal” for next period can also be identified. Detecting these types of customers might allow companies to allocate their resources efficiently and to reduce the cost associated with the retention programs. In the proposed model, historical data from a supermarket retailer which contains millions of transaction data belong to one million customers are used. We proposed that customers visit the company within a time route and this could bring out a churn route pattern which could be derived from churner customers’ behaviors. This findings will be examined by rule based algorithm, neural network, decision trees and regression. The model is tested on 8340 customers and its results were assessed using the historical data. The results of the proposed model were compared to Bayesian network that is a well-known churn predictive model. Comparison results showed that route pattern of churners obtained from their RFM datas have a great importance to realize customer churn behavior, so the proposed model could be useful for marketing practitioners. Online content providers face the challenge whether to offer content for free, against a fee, or some hybrid of the two. We hypothesize that this varies with the demand for content and the distribution of customer valuation of the service. We develop a model and provide empirical evidence for when firms should price online content. We use empirical data from a website on the amount of content that is free or offered against a fee as well as demand (click stream) of the site across time. Our results suggest that the amount of content offered against a fee varies and increases when content is more attractive to a larger number of customers due to specific demand shocks. 3 - Private Label Response to National Brand Promotions: A Field Experiment Eric Anderson, Hartmarx Professor of Marketing, Northwestern University, Kellogg School of Management, 2001 Sheridan Road, Evanston, IL, 60208, United States of America, [email protected], Karsten Hansen, Duncan Simester Recently there has been an active debate about whether retailers react to trade promotions by lowering the prices of competing products. A simple category pricing model suggests that if a promotion on one item affects demand for other items it will generally be profitable to change the prices of all affected items. We report on the results of a large-scale field experiment that investigated how private label brands should be priced when a national brand is promoted. Our experiment focuses on 28 “copycat” private label brands and the corresponding national brand sold in over 6,000 stores of a national retailer. We experimentally vary the price level of the copycat private label brand when the national brand is promoted. Our econometric model integrates historical and experimental data in a unified framework. The findings confirm that shielding is an effective strategy for preserving the sales volume and profit contribution of the private label items. It can lead to higher profits not just on the private label items, but also in aggregate. The size of the shielding discount is important. If the shielding discounts are too large they erode too much margin, while small shielding discounts may not preserve enough private label demand. 3 - Hazards of Ignoring Involuntary Customer Churn Zainab Jamal, Hewlett Packard Labs, 1501 Page Mill Road, Palo Alto, CA, United States of America, [email protected], Randolph Bucklin We study the impact of ignoring involuntary churn (when a firm terminates the subscription of a customer) on the estimation and diagnosis of voluntary churn (when the customer terminates the relationship). The presence of competing events has been shown to bias the estimates of the hazard rates and survival times of the main event. In our case, the main event is voluntary churn and the competing event is involuntary churn. We estimate a bivariate Weibull survival model that captures the dependency between the two event times and has been proposed in the literature as an approach to the problem of dependent competing risk. We compare this model with a benchmark model – a univariate Weibull model with only voluntary churn. We estimate the models using maximum likelihood techniques. We find significant differences in the prediction of voluntary churn rates. Also the impact of covariates on the voluntary churn rates is different across the two models. Further, the bivariate Weibull survival model does better in predicting the customers who are more likely to churn. An added dimension of the study is that the key covariates that influence voluntary churn rate impact involuntary churn rate differently. Our study highlights the need to incorporate involuntary churn when modeling the voluntary churn process. 4 - Paying with Money or with Effort: Pricing when Customers Anticipate Hassle Anja Lambrecht, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, [email protected], Catherine Tucker For many services, customers subscribe to long-term contracts. We suggest that rather than evaluating multi-period service contracts at the contract-level, customers use period-level bracketing. This means they evaluate the contract as the sum of distinct per-period utilities. This has important consequences when utility varies over the course of the contract, for example due to “hassle costs.” If customers use period-level bracketing, they will value a lower price more in periods where they have hassle than in other periods. We explore this using data from a field experiment for web hosting services. The field experiment had 2 hassle cost priming conditions (present, absent) x 2 discount conditions (offered, not offered). We find that a lower price in the initial period is more attractive to customers when they expect their hassle costs to be high at set-up. In six lab experiments, we support and extend the field experiment’s findings. Importantly, we find evidence for period-level bracketing when customers have hassle costs independently of whether hassle costs occur in the first, an intermediate or the last period of a contract. We rule out alternative explanations, such as hyperbolic discounting. Our findings suggest that in setting prices, firms should consider the timing of hassle costs faced by customers. 26 MARKETING SCIENCE CONFERENCE – 2011 Thursday, 3:30pm - 5:00pm TD02 surpass a (latent) threshold to even be compared with others available. We examine diversification via an experimental choice sequence task and dual-component stochastic model (choice set formation; choice-from-set). The experiment manipulates two key diversification drivers in prior literature: whether choices are made all at once (“simultaneously”) vs. separately (“sequentially”), and the relative attractiveness of available options. We find that diversification may be primarily a phenomenon of choice set formation: people do penalize their previously chosen option, but only for simultaneous choices, and only in the set formation portion of the model. Thus, prior model-based evidence of variety-seeking in diversification may have been misinterpreted as a discounting of an item’s features at the stage of choice, as opposed to a failure to ‘consider’ the option at all. That is, failing to methodologically incorporate choice set formation may erroneously suggest that consumers are using a diversification choice heuristic when they may instead be diversifying the alternatives in their ‘consideration set.’ Furthermore, results suggest that the expected number of items thus ‘considered’ is substantially greater in multiple-, versus single-, item choice. ■ TD01 Legends Ballroom I Choice II: Effects on ... Contributed Session Chair: Linda Court Salisbury, Assistant Professor, Boston College, CSOM, Fulton 441, 140 Commonwealth Ave., Chestnut Hill, MA, 02467, United States of America, [email protected] 1 - Incorporating State Dependence in Aggregate Market Share Models Polykarpos Pavlidis, PhD Candidate, Simon Graduate School of Business, University of Rochester, Rochester, NY, 14627, United States of America, [email protected], Dan Horsky, Minjae Song ■ TD02 Legends Ballroom II An empirical investigation of individual household level data in twenty CPG categories (IRI Marketing Data set, 2008) uncovers that state dependence is a significant determinant of consumers’ choices in all twenty product categories. In light of this finding, we examine whether and how one can incorporate state dependence in a demand model of frequently purchased goods estimated with market level data. Aggregated data is more often available and is frequently used in applied research in marketing and economics. The incorporation of consumer choice dynamics in these market share models can potentially increase their applicability and improve their predictive behavior. Online Advertising - II Cluster: Internet and Interactive Marketing Invited Session Chair: Harald van Heerde, University of Waikato, Private Bag 3105, Hamilton, 3240, New Zealand, [email protected] 1 - Connecting Social Media with Television Advertising and Online Search Yanwen Wang, PhD, Emory University, 1300 Clifton Rd, Atlanta, GA, 30033, United States of America, [email protected] 2 - Complexity Effects on Choice Experiment-based Model Performance Benedict Dellaert, Erasmus University Rotterdam, P.O. Box 1738, Rotterdam, 3000 DR, Netherlands, [email protected], Bas Donkers, Arthur van Soest In an attempt to muscle through turbulent economic times, the U.S. automobile industry has embraced social media. Ford pioneered the practice of hiring syndicated bloggers to serve as brand advocates, while much of BMW’s recent success is attributed to its engagement with consumers across Facebook, Twitter, and YouTube. While there is research looking at the impact of social media on firm outcomes – be they sales or other metrics like stock-returns – little, if any, research is undertaken on understanding how social media impact consumers’ online search for automobiles. Understanding the impact of social media on consumer search is very important as online search informs consumer’s consideration-sets, which in turn affect their final choice of an automobile. By combining three unique datasets, namely: (i) consumerlevel click-stream search data for automobiles, (ii) consumer-directed television advertising and (iii) social media (volume and sentiments of buzz spanning blogs, newspaper articles, newsgroups, online forums, etc.), this study sets out to answer the following questions. One, how does the volume and sentiment of online buzz impact the volume of a consumer’s online branded search? Two, how do these effects compare with the impact of television advertising on online search? Three, are there synergies between online buzz and television advertising on online search? How do these effects systematically vary across consumers and geographic markets? After calibrating our model with these aforementioned data, we offer insights for advertising and targeted advertising interventions to undo negative buzz generated through product recalls. Understanding what drives choice experiment-based model performance helps firms to better evaluate future marketing actions when using such models. This research investigates choice experiment complexity (i.e., the number of alternatives, number of attributes, and utility similarity between the most attractive alternatives) as a key factor to affect choice model performance both within and between complexity conditions. The results show that complexity has a negative direct effect on choice model performance. However, this effect is contingent on consumers’ decision time used relative to the number of normative elementary information processes (EIPs) that the decision task requires, which we take as an approximation of the degree of simplification in the consumer’s decision process. When consumers spend less time per normative EIP, this increases choice model performance within a given choice complexity condition. However, it decreases choice model performance between complexity conditions. We also introduce a practical modeling approach that captures these direct and indirect effects of choice complexity and hence enables predictions between different choice complexity conditions. 3 - The Interplay of Reference Dependence and Choice Set Formation in Replacement Decisions Lianhua Li, PhD Student, University of Alberta, Faculty of Business, Edmonton, AB, Canada, [email protected], Paul Messinger, Joffre Swait 2 - Investigating Advertisers’ View of Online and Print Media: Complements or Substitutes? Shrihari Sridhar, Assistant Professor of Marketing, Michigan State University, Eli Broad College of Business, N300 Marketing Department, East Lansing, MI, 48824, United States of America, [email protected], S. Sriram The literature presents evidence supporting the expectation of a strong preference for the status quo good in replacement decisions. Prospect theory predicts this tendency; however, prospect theory ignores possible choice set formation effect, assuming that all goods (status quo and new goods) will be evaluated, using the status quo as reference point, and then compared to reach a final choice. Nonetheless, several explanations for the exaggerated preference for status quo goods in prior literature predict the possibility of choice set formation revolving around status quo products. We therefore hypothesize that replacement decisions are subject to both referencedependence and choice set formation effect. To test this hypothesis, we develop and compare three choice models: a pure reference-dependence model, a pure choice set formation model, and a hybrid model that accounts for both reference-dependence and choice set formation. Using experimental choice data, we find that the hybrid model greatly outperforms the first two models, which strongly suggests the coexistence of reference-dependence and choice set formation in replacement decisions. We also show that omitting either process leads to estimation bias in the parameters of the included process. Just as the reference-dependence model based on prospect theory has led to improved understanding and prediction of choice, we believe “upgrading” it to include choice set formation effect will further enhance these benefits. Faced with declining profits, print newspapers have turned hybrid, catering to both print and online audiences. As a multi-channel platform, the hybrid newspaper needs readers and advertisers across both print and online channels to be on board. This requires knowledge not only about the interrelatedness in demand between readers and advertisers, but also about whether advertisers perceive the print and online media-channels to be complements or substitutes. Extant research has investigated whether readers find online and print newspapers to be complements or substitutes. But less is known about how advertisers, who provide 80% of newspaper revenue, choose to allocate their media dollars between the online and print version of a multi-channel platform. From a newspaper’s standpoint, understanding advertisers’ preferences for allocating advertising expenses between print and online media would be useful to develop targeted pricing and salesforce strategies. We use advertiser-level data from a large hybrid newspaper in the US from 2005-2010, pertaining to the amount of media dollars purchased across three outlets (print-weekday, print-Sunday and online) by quarter. We estimate a multiple discrete-continuous extreme value model (MDCEV) of advertisers’ simultaneous choice (i.e., which combination of media to advertise in) and the corresponding expenditure in each medium. While performing this analysis, we control for temporal variation in print and online readership as well as salesforce efforts directed at each advertiser. We subsequently investigate how this decision is affected by advertisers’ baseline utility for each medium, growth in the readership base across channels and also by their perceived substitutability/complementarity across channels. 4 - Does Choice Set Formation Drive the Diversification Effect? A Model and Experimental Evidence Linda Court Salisbury, Assistant Professor, Boston College, CSOM, Fulton 441, 140 Commonwealth Ave., Chestnut Hill, MA, 02467, United States of America, [email protected], Fred M. Feinberg Diversification – exhibiting greater variety as multiple choices are made together, in advance of consumption – is a robust and important phenomenon (e.g., purchasing groceries, queuing Netflix films). Studies of diversification tend to focus on choice alone, conceptualizing it as inherently comparative. This precludes the possibility that diversification is driven largely by choice set formation, wherein each item needs to 27 TD03 MARKETING SCIENCE CONFERENCE – 2011 3 - Does Television Advertising Influence Online Search? Mingyu Joo, Syracuse University, 721 University Ave., Suite 311, Syracuse, NY, 13244, United States of America, [email protected], Kenneth Wilbur, Yi Zhu that are not used to search for new cars. The presence of unique Internet information sources and multiple seller types makes search for used cars different from that for new cars. Search for used cars is modeled as a three stage process. The first stage is consumer search for information on makes and models. The second stage is search for used cars available in the secondary market and the third stage involves final choice of a used car. Results show that education, age, experience of buying used cars, price range and trust in seller are important drivers of search and seller choice in secondary car market. Search on dealer specific sources increase the possibility of buying from dealers whereas use of local websites reduces the odds of buying from a dealer. Analysis of interrelationship of use of information sources indicates that use of Internet sources reduce dealer visits. Local websites have a negative impact on use of retail websites and consumers complement dealer and print sources to search for used cars. Traditional advertising influences consumer information search, and consumers increasingly use television and internet simultaneously. This paper finds a significant association between television advertising for financial services brands and consumers’ tendency to choose branded keywords (e.g. “Fidelity”) rather than generic category-related keywords (e.g. “stocks”). This effect is largest for young brands during standard business hours with an elasticity (.07) comparable to extant measurements of advertising’s impact on sales. However, television advertising is not correlated with category search incidence. These findings show that practitioners should account for cross-media synergies when planning, executing, and evaluating both television and search advertising campaigns. The results also show why and how the search advertising literature should enrich its modeling of competition among advertisers. 3 - Media, Finance and Automotive: A Latent Trait Model of Consumption in Seemingly Disparate Categories Chen Lin, Emory University, 1300 Clifton Road NE, Atlanta, GA, 30322, United States of America, [email protected], Douglas Bowman 4 - Does Online Advertising Help or Hurt Offline Sales? A Nation-wide Field Experiment Harald van Heerde, University of Waikato, Private Bag 3105, Hamilton, 3240, New Zealand, [email protected], Isaac Dinner, Scott Neslin The rise of new media and media fragmentation not only provides new experiences for consumption in many product categories, but also imposes challenges for media buyers to target across multiple media platforms. This paper examines whether media consumption vary with non-media product preferences. For example, do customers who prefer fixed income investments consume media differently from customers who prefer equity investments? Do Internet fans drive different types of cars compared to print media audiences? We propose a theory-driven, psychologically realistic model that examines consumer preferences across seemingly disparate product categories using a latent-trait approach for parameter heterogeneity. We calibrate the model with a proprietary dataset on individual behavior and choices in media consumption, financial investments, and automotive purchases. We hypothesize and test two mechanisms to explain drivers of correlated preferences using a set of attitudinal of behavioral measures: System 1, in which hard-to-observe psychological processes governs decision making across multiple categories, and System 2, in which media characteristics and preferences contribute to information processing in other product categories. We demonstrate the power of this model by comparing performance with traditional multi-category choices models and latent-class models. Implications are drawn on media targeting and psychological process modeling. Increasingly many firms sell both via physical stores and the online channel. To drive sales through both channels, many firms advertise both in the traditional media and online. One of the risks of such a dual channel advertising strategy is that while advertising in one channel may enhance sales through that channel, it may cannibalize sales from the other channel. In contrast, there is also the potential for a positive cross-channel effect. Given the strong growth in worldwide online advertising expenditures, one particularly pressing question is whether online advertising helps or hurts offline sales. We study this research question using a unique field experiment. A major upscale US retail chain with a national presence experimentally increased online (banner) advertising in a random subset of its markets, while leaving online advertising unchanged in a control group of markets. We observe sales and marketing data at the weekly level for each of the markets over a two year time period. We estimate a dynamic sales response model for the number of both online and offline customers and their average spend as a function of online and traditional advertising and a number of control variables. The model allows for a detailed understanding of the within- and across-channel dynamic effects of both types of advertising. It also helps managers to understand whether these effects manifest themselves primarily via the number of customers or the average customer spend. ■ TD04 Legends Ballroom V Econometric Methods II: General ■ TD03 Contributed Session Chair: Martin Spann, Ludwig-Maximilians-University Munich, Geschwister-Scholl-Platz 1, Munich, 80539, Germany, [email protected] 1 - Empirical Regularity in Academic Marketing Research Productivity Patterns Vijay Ganesh Hariharan, Assistant Professor of Marketing, Erasmus University, Burgemeester Oudlaan 50, Rotterdam, 3000DR, Netherlands, [email protected], Debabrata Talukdar, Chanil Boo Legends Ballroom III Internet: Car Buying Contributed Session Chair: Chen Lin, Emory University, 1300 Clifton Road NE, Atlanta, GA, 30322, United States of America, [email protected] 1 - Modeling the Volume of Positive Online Word of Mouth for Automobiles Jie Feng, Assistant Professor of Marketing, SUNY Oneonta, 224 Netzer Administration Bldg., Economics & Business Division, Oneonta, NY, 13820, United States of America, [email protected], Purushottam Papatla In any academic discipline, published articles in respective journals represent “production units” of scientific knowledge, and bibliometric distributions reflect the patterns in such productivity across authors or “producers”. We use a comprehensive data set from 11 leading marketing journals to examine if there exists any empirical regularity in the patterns of research productivity in the marketing literature. Our results present strong evidence that there indeed exists a distinct empirical regularity. It is the so called Generalized Lotka’s Law of scientific productivity pattern: the number of authors publishing n papers is about 1/(n power c) of those publishing one paper. We find the empirically estimated value of the exponent c to be 2.05 for the overall bibliometric data across the leading marketing journals. For the individual journals, the estimated values of c range from 2.15 to 2.83, with lower values indicating higher authorship concentration levels. We also find that variations in authorship concentration levels across journals and over time are driven by a journal’s maturity, topical focus, relative attractiveness as a publication outlet and its review process characteristics. Online word of mouth is gaining in importance as a means of promoting automobiles to consumers. Several auto brands, such as Scion xB, Ford Fiesta, and Chevy Tahoe, were reported using online word of mouth campaigns to generate product awareness as well as sales. The literature suggests that product attributes affect positive word of mouth. Automakers, therefore, need to identify the attributes that are likely to affect positive word of mouth for automobiles and ensure that their brands excel on those attributes. In this study, we classify the generic attributes of automobiles into four broad categories: quality, design and performance, newness of the model and body style, and investigate how each category affects the volume of positive online word of mouth. In order to ensure that our findings are not an artifact of the modeling approach, we repeat our investigation using different modeling assumptions. Our results across all the analyses provide a consistent finding that design/performance plays the dominant role in stimulating positive online word of mouth. In addition, designing new models or redesigning existing models also stimulates positive online word of mouth. However, this strategy may not be as effective for luxury and large cars. 2 - Investigating the Performance of a Dynamic Budget Allocation Heuristic: A Simulation Based Analysis Nils Wagner, University of Passau, Aberlestrasse 18, Munich, 81371, Germany, [email protected] The marketing budget allocation process is one of the most important tasks a manager is being charged with. As firms in general sell a portfolio of products and can choose among various marketing activities their profit maximization problem is characterized by high complexity. However, managers prefer to use simple rules to determine the marketing budget as they find it difficult to fully understand sophisticated allocation tools provided by academics. The paper ‘Dynamic Marketing Budget Allocation across Countries, Products, and Marketing Activities’ by Fischer, Albers, Wagner, and Frie (2011, forthcoming in Marketing Science) address this problem and presents a dynamic allocation rule which is suggested to be close to optimum while being easy to understand and to implement. We test the nearoptimality as well as the convergence properties of this allocation rule by conducting a comprehensive simulation study using a large number of data conditions including all factors contained in the profit maximization function. In particular, we analyze changes in performance by imposing estimation error affecting the parameters of 2 - External Search in Secondary Markets and Impact of Internet Search on Seller Choice Sonika Singh, PhD Student, University of Texas-Dallas, 800 W. Campbell Road, SM32, Richardson, TX, 75080-3021, United States of America, [email protected] This research looks at the determinants of use of Internet information sources and the impact of Internet information sources on seller choice in secondary market for cars. Unlike the new car market, used car market is characterized by multiple sellers such as dealers and individuals. The quality of used cars is different even for cars of same make and model. This market is characterized by use of some unique information sources like local websites (craigslist), auction and newspaper websites 28 MARKETING SCIENCE CONFERENCE – 2011 TD06 3 - Merging in Spatial Competition Tieshan Li, Concordia University, 1450 Guy Street, Montreal, QC, Canada, [email protected] interest. To evaluate the performance of the allocation rule we compare the profit measures gained by the allocation rule with the optimal solution as well as simple practitioner rules. Further, the impact of the various factors on the performance of the allocation rule in terms of robustness and convergence properties is investigated by following the concept of a surface regression. The authors find that the allocation rule is quite robust to all types of data manipulation and even performs very well in case of estimation error. Merging is an important business action in the market and it has attracted attention from researchers. Given the existing research on merging focusing more on the supply side, this work investigates the merging effect from the demand side. In the spatial competition with the circle market set-up, we study the effects of merging with direct competitor and merging with indirect competitor. In the case of merging with indirect competitor, the firm does not have motivation to do so unless the merging action can bring efficiency gain on the merging firms. However, in the case of merging with direct competitors, the firm always has motivation to do so no matter the existence of efficiency gain. With efficiency gain from the merging action, the merging firms are always better off while the profits of the other firms depend on the degree of efficiency gain of the merging firms and are not monotonic. Without efficiency gain, all the firms in the market are better off. However, the merging firms are not the ones which benefit most from the merging action. 3 - Social Network Based Judgmental Forecasting Martin Spann, Ludwig-Maximilians-University Munich, GeschwisterScholl-Platz 1, Munich, 80539, Germany, [email protected], Christian Pescher, Gary Lilien, Gerrit Van Bruggen Forecasting is especially challenging in high-uncertainty environments, e.g., due to unstable market conditions. A popular method to generate reliable forecasts in conditions of high uncertainty is judgmental forecasting, in which a number of informants is asked for their estimates. Improvements in forecasting accuracy can be gained by assigning weights to those forecasts that are likely to be more accurate. The structural position of each informant in a social network can be a good indicator for the informant’s access to information, which has not been analyzed in literature yet. The authors conduct two experiments, one laboratory experiment using the markstrat game and one field experiment, in which they show that in conditions of high uncertainty social network based weights can improve forecasting accuracy compared to weighting approaches discussed in recent literature. 4 - Price and Inventory Competition between New and Old Technologies Dinah Vernik, Assistant Professor of Marketing, Rice University, 6400 Main Street, Houston, TX, United States of America, [email protected], Preyas Desai, Fernando Bernstein In this paper we investigate production/ordering decisions across multiple generations of a product. In many consumer durables product categories manufacturer typically needs to innovate and introduce new versions of the product. However, considerable uncertainty exists about the consumers’ reaction to the new product. The consumers might or might not like the new version of the product more than the old one, while the manufacturing costs associated with the new technology are usually higher. The problem is made more complex by the fact that the quality of the new product as perceived by consumers (through reviews, consumer reports, etc.) cannot be realized until the product enters the market while the manufacturing decisions must be made in advance. This uncertainty about new product quality at the time when ordering/manufacturing decisions are made creates some interesting issues of channel conflict. We consider a setup where manufacturer sells both products via a retailer using a price-only contract. Two systems are compared: a centralized system, where the manufacturer makes both production and retail pricing decisions for both products. And a decentralized one, where the manufacturer moves first by offering the wholesale price and the retailer follows by placing an order for both products and determining the retail price. Several interesting results about product inventories and pricing of the two generations of the product are presented and discussed. ■ TD05 Legends Ballroom VI New Product IV: Strategy Contributed Session Chair: Dinah Vernik, Assistant Professor of Marketing, Rice University, 6400 Main Street, Houston, TX, United States of America, [email protected] 1 - Strategic Product Line Design with Product Concept Demonstration Taewan Kim, PhD Candidate, Syracuse University, 721 University Ave., Syracuse, NY, 13244, United States of America, [email protected], Eunkyu Lee Trade shows are a popular venue for firms to showcase their innovative activities to their industry cohorts and to the general public. Product concept demonstrations at trade shows can contribute to the long-term success of the new product by influencing awareness and perception of the new product as well as by encouraging buyers to wait for the new model launch instead of purchasing a currently available model without a delay. However, if the perception created by the product concept demonstration turns out to be not consistent with the quality delivered by the launched model, the company might suffer negative effects on its reputation and the demand for the new product. This paper develops and analyzes a game-theoretic model to examine how a firm’s product positioning strategy interacts with its product concept demonstration strategy in duopoly. Each firm develops and launches a line of two products with the option of demonstrating each product concept at a trade show before the completion of its development. The market characterized by consumer heterogeneity in design taste and willingness-to-pay for technical quality provides an opportunity for each firm to strategically choose the horizontal and vertical product positions for profit maximization. Our analysis of the Nash equilibrium solution provides new insights into the strategic inter-dependence of product positioning and product concept demonstration. We also show that the equilibrium product demonstration strategies may follow the prisoner’s dilemma under certain conditions. ■ TD06 Legends Ballroom VII Competition IV: Quality Contributed Session Chair: S. Chan Choi, Professor, Rutgers Business School, 1 Washington Park, Newark, NJ, 07102, United States of America, [email protected] 1 - The Impact of Competition and the Cost of Overstating Quality on the Optimal Quality, Quality Claims Praveen Kopalle, Professor, Dartmouth College, Tuck School, Dartmouth, NH, United States of America, [email protected], Don Lehmann This paper examines the impact of strategic competition and cost of overstating quality on equilibrium quality and quality claims. We model the situation where two firms simultaneously introduce a new product or add a major new feature to an existing product. Potential customers have an initial quality expectation for the new good based on a combination of public and company provided information. These expectations get updated based on both their actual experience when they buy a product and word of mouth from those who have bought it. The companies make one-time decisions about average quality (which is costly to produce), price, and advertised quality. We formally develop a two period model which allows for a large weight on results in the second period to capture the impact of future period sales. To provide support for our model assumptions and results we (i) conducted a survey of consumers and (ii) analyzed data from the U.S. Federal Trade Commission (FTC) on deceptive advertising cases decided between August 1996 and December 2002. In addition, to evaluate the extent to which managers would behave in line with our model prescriptions in different competitive and legal cost contexts, we conducted a conjoint style study with future managers at a top business school. In the main, managerial intuition is directionally consistent with the model. Competition leads to higher quality, lower price, and higher advertised quality. Heightened costs of overstating quality led to lower advertised quality. Most interesting, it appears that when competitors are constrained to be truthful in their advertising due to legal costs, optimal product quality, and hence potentially consumer welfare, is lower. 2 - The Strategic Role of Exchange Programs Bo Zhou, Fuqua School of Business, 100 Fuqua Drive, Box 90120, Durham, NC, 27708, United States of America, [email protected], Debu Purohit, Preyas Desai In an exchange program, consumers turn in an old item and get store credit or a gift card toward another purchase in the store. To analyze the impact of exchange programs in a competitive setting, we assume that two firms first decide its policy on whether or not to offer an exchange program. If a firm does not offer an exchange program, it simply chooses a retail price for the new good. If it offers an exchange program, it chooses not only the retail price for the new good but also the price compensation for the exchange. In terms of consumers, based on their valuations of the old and new goods, there are two types of consumers: low valuation and high valuation. Each consumer has an old good that can potentially be used in an exchange program. Based on the prices, consumers decide whether to purchase and, if it is offered, whether to participate in the exchange program. In deciding whether to participate in the exchange program, consumers evaluate the costs and benefits of turning in the old product and getting a new one. For the new product, consumers consider the quality and the price. On the other hand, consistent with behavioral decision theory, we assume that consumers view their old product as a part of their endowment. As a result, consumers’ incur a loss that is equivalent to their mental book value when they turn in their old product. We find that whether the retailers offer an exchange program depends on the proportion of low valuation consumers in the market and their valuations of the old and new goods. The equilibrium outcome can be both offering or not offering exchange, neither offering exchange as a Prisoners’ Dilemma, and coordination equilibria. 29 TD07 MARKETING SCIENCE CONFERENCE – 2011 2 - A Structural Analysis on Service Quality and Pricing Tradeoff in Airlines Chen Zhou, Doctoral Candidate, Pennsylvania State University, Smeal College of Business, University Park, PA, 16802, United States of America, [email protected], Rajdeep Grewal self-enhancement, both consumers and service providers tend to over-estimate the valence of their own reactions to service failures (anger vs. apology) given the significant differences in their feelings before and after the event. The results also indicate a significant gap in the perceived severity of the problem and emotional reactions between the two groups. Thus the perceptual gap between consumers and service providers is exacerbated and affect consumer expectations of the recovery efforts and their re-purchase intention. However, past experiences and the use of surrogate information, i.e., insight from others with similar experiences, serves as an effective way to reduce the perceptual gap and can help service providers to minimize consumer discontent. These findings have meaningful implications for understanding consumer reactions to service failures and for improving the management of recovery activities. Aside from price, service quality plays a prominent role in customer acquisition and retention for services firms such as those in airlines and hotel industries. Recognizing the importance of service quality, we reason that a formal model of competitive pricing behaviors of service firms should consider service quality as a managerial decision variable. For the purpose, we develop and estimate a structural model that treats both pricing and service quality as endogenous such that (1) both the variables are important in determining demand and (2) service quality also influence the production function. Thus, in our structural model we estimate systematically estimate demand, production, and profit functions. We test our model in the context of the US Airlines industry, where we conceptualize a route as a market. The data comes from multiple secondary sources, typically records maintained by the US Department of Transportation. Specifically, we obtain data on the 21 airlines on which Airline Quality Rating Reports provided by Purdue University maintains service quality measures. For these airlines, as is typical with structural models in the airlines industry, we obtain pricing and other data (e.g., capacity offered and utilized) on routes where origin and destination cities have population greater than 850,000. The results from the Generalized Method of Moment estimation of the structural model suggest heterogeneity in demand sensitivity of service quality and price across airlines, thereby having important implications for these two strategic resource allocation decisions. 3 - Can I Do It? Can You Do It? Roles of Self-efficacy and Other-efficacy of Customers and Employees Kimmy Wa Chan, Assistant Professor in Marketing, Hong Kong Polytechnic University, Department of Marketing and Management, Li Ka Shing Tower, Hung Hom, Kowloon, Hong Kong - PRC, [email protected], Bennett C. K. Yim, Simon Lam Engaging customer participation (CP) in service production and delivery to cocreate value is gaining credence in both academic writing as well as marketplace practices. However, recent research shows that CP could be a double-edged sword; it could improve service quality to customers and strengthen the relational bond between customers and employees, yet, it could also pose problems for service employees (e.g., increase job stress). CP also could be a challenging endeavor for customers who not only need to have the knowledge, but also the ability, to perform their roles in specific service contexts. CP as a taxing situation for both customers and employees would therefore suggest that appraisals of both parties’ capabilities may affect their participation behaviors and emotional experience. This study examines the roles of both self-efficacy (SE) and other-efficacy (OE) (perceived capabilities of the partner) in the relationship between CP and enjoyment of participation for both customers and employees in the context of professional financial services. Social cognitive and role theories provide the foundations to support how (1) SE moderates the effect of CP on the enjoyment of participation and (2) the synergic effect of SE and OE affects the enjoyment of participation differentially for customers and employees. Empirical results from 223 pairs of customers and service employees of financial services suggest that efficacy perceptions determine the magnitude of CP effects. The match and mismatch of SE and OE also moderates the effect of CP on participation enjoyment, albeit differently for customers versus employees. Significant implications on managing customer-employee collaboration in service participation derived from the results are discussed. 3 - Hedonic Quality Differentiation and Channel Choice S. Chan Choi, Professor, Rutgers Business School, 1 Washington Park, Newark, NJ, 07102, United States of America, [email protected] Quality is a multidimensional construct. In the literature, the quality concept has been used to describe either a whole product or an individual attribute. We employ a more holistic view of quality, in which quality dimensions are classified into either hedonic or utilitarian. Utilitarian qualities included attributes that are useful, practical, and necessary. Hedonic qualities are associated with fantasy, fun, and pleasure. Previous studies show that hedonic attributes have a dominant influence over utilitarian attributes when price information is present. This paper models price competition between two channels when companies are differentiated in hedonic quality. We employ a demand function that is derived from representative consumer utility that captures both horizontal and quality differentiations. We examine the way differentiation in hedonic quality alters the optimal channel choice decisions when companies compete in price when the products are also horizontally differentiated. ■ TD08 ■ TD07 Founders II Founders I Innovation IV Services Contributed Session Contributed Session Chair: Anna S. Cui, Assistant Professor of Marketing, University of Illinois at Chicago, 601 S Morgan Street, Chicago, IL, 60607, United States of America, [email protected] 1 - Patent Rank and Firm Performance Monte Shaffer, PhD Student, Washington State University, 821 Old Moscow, Pullman, WA, 99163, United States of America, [email protected], U.N. Umesh Chair: Kimmy Wa Chan, Assistant Professor in Marketing, Hong Kong Polytechnic University, Department of Marketing and Management, Li Ka Shing Tower, Hung Hom, Kowloon, Hong Kong - PRC, [email protected] 1 - Service Worker Role in Encouraging Customer Equity: Dyadic Analysis Yu-Li Lin, Assistant Professor, Southern Taiwan University, 1, Nan-Tai Street, Yung-Kang City, Tainan, Taiwan - ROC, [email protected], Hsiu-Wen Liu Recently, top scholars in innovation rightfully assessed that traditional patent data is old and tired (Tellis et al. 2009, p. 12): “Some researchers and policy makers consider the registration of patents so important that they equate patents to innovation and often measure the latter with the former. This line of thinking suggests that patents would be an important driver of radical innovation. If markets value patents as highly as many researchers do, patents should have a major influence on financial returns of a firm.” In this manuscript, we intend to show how patents do indeed have a major influence on financial returns of firms. To do so, however, requires new data and a new perspective on patents and innovation. In a previous manuscript, we introduced Patent Rank as an objective measure of radical innovation. Patent Rank is a logical extension of WPC (weighted patent counts) introduced by Trajtenberg (1990). WPC weights each patent by forward citations because each patent’s importance and relevance is determined by how many other patents cite it as prior work. In other words, if more important patents cite a patent then the one that is cited increases, in turn, in its importance. Thus we specify a general model to define the network formation and structure. A ‘local effects’ model is most appropriate when studying financial returns using Fama/French or Carhart specifications. Another measure of importance is how well a patent, and its citations, diffuse over time. Utilizing nonlinear diffusion models, we identify which type of diffusion pattern best fit the innovation, and utilized that model to estimate a patent’s lifetime value (PLV). In this manuscript, we combine our findings to show how patents have a major influence on financial returns of firms. The primary focus of this paper is assessment of the role of the service worker behaviors in encouraging customer equity. The researchers investigate this topic utilizes a dyadic sampling design. The sample includes 398 customer and service provider dyads. The findings of this research support our hypotheses. Customer orientation affects trust toward service provider. Trust toward service provider affect customer equity. Further, trust to the service worker serves as a mediator of the effects of customer orientation and customer equity. Finally, theoretical, managerial and future research implications are included. 2 - Perceptions of Service Failures: A Test and Extension of Affective Forecasting Theory Muyu Wei, Lingnan University, Dept of Marketing and Int’l Business, Tuen Mun, Hong Kong - PRC, [email protected], Geng Cui Service failure occurs when a service provider fails to meet customer expectations. Overall, service providers are believed to have more accurate predictions of the severity of service failures. However, researchers have seldom examined the accuracy of consumer perceptions or the perceptual gaps between consumers and service providers when analyzing service failure. In a 2 (group) x 2 (time) x 2 (surrogate information) experimental study, we test and extend a theory from psychology – affective forecasting – to examine the reactions of consumers and service providers to service failures. Both groups were asked about their perceived severity of the problem and their emotional reactions before and after a service failure. Due to the effect of 30 MARKETING SCIENCE CONFERENCE – 2011 TD09 2 - Validating Suppliers of Retailer’s Resources in Augmenting Product Safety Performance Wei-Che Hsu, Postgraduate Research Assistance, Chung-Hsing University, epartment of Marketing National, 250 Kuo-Kuang Rd., Rm.749, Taichung, 402, Taiwan - ROC, [email protected], Ming-Chih Tsai 2 - What You Don’t Know Can’t Hurt You: Effects of Knowledge Limitations on Technological Innovativeness Stav Rosenzweig, Assistant Professor, Ben Gurion University of the Negev, The Guilford Glazer Faculty of Business, POB 653, Beer Sheva, 84105, Israel, [email protected], David Mazursky Technological innovativeness underlies the development of new technologies. It generates new markets and transforms existing ones. As such, it drives the survival, growth, and success of firms, industries, and countries. A major driver of innovativeness is knowledge, which can be derived either from within the country’s technology or from outside it. What are the innovativeness consequences of limited sources of knowledge? Do knowledge sources – internal or external to a country’s technology – affect the country’s technological innovativeness? We use patent and trade data to answer these questions. We employ more than 280,000 patents issued in the US across 12 technological subcategories and over 16 years. In contrast to a prevalent thinking that bountiful knowledge sources enhance innovation we find that exposure to externally-derived knowledge is negatively associated with technological innovativeness in most technological subcategories. We also find that this negative relationship is reversed for computation and communications related subcategories. Moreover, while one may expect a negative relationship between internally-derived knowledge and innovativeness, we find that this relationship is curvilinear whereby highest levels of innovativeness are observed when internallyderived knowledge is used at moderate levels. We attribute our findings to the consequences of knowledge constraints and limitations, and suggest that they may have positive implications for technological innovativeness. This study examines the effect of relationship-specific resources in affecting product safety. Retailers are increasingly concerned in the monitoring of product safety. When dealing with large numbers of individual suppliers, an effective method validating suppliers’ resources in regulating product safety is needed. Past researches examined relationship-specific invested resources in affecting financial performance, but few examines the effect of retailer and its suppliers’ investment on developing relationship-specific resources in achieving superior product safety. The study hypothesize under different extent of integrated transaction relations, parties investment in developing relationship-specific resource differently affect product safety performance. Leveraging from intellectual capital and resource-based view, we first identify intangible resources, capabilities, and validate their influences on retailer/suppliers’ invested relationship-specific resources, we then examine the effect of these invested resources in affecting food safety performance. A total of 61 valid data collected from suppliers of principle retailers in Taiwan enabled empirical examining through regression analyses. We find invested relationship-specific resources from either retailer or suppliers not to directly affect performance of product safety, but are significant when moderated by integrated relations. Findings from our research assist retailers optimize supplier selection through validating supplier’s intangible resources, and provide retailers an effective allocation of resource in developing mutually beneficial strategic resources in augmenting product safety. 3 - Alliance Portfolio Resource Diversity and Firm Innovation Anna S. Cui, Assistant Professor of Marketing, University of Illinois at Chicago, 601 S Morgan Street, Chicago, IL, 60607, United States of America, [email protected], Gina O’Connor 3 - Assortment Selection in Retailing: Strict Return Policies Call for Eccentric Products Aydin Alptekinoglu, SMU Cox School of Business, 6212 Bishop Blvd, Dallas, TX, 75275, United States of America, [email protected], Elif Akcali, Alex Grasas Despite of firms’ increasingly common simultaneous engagement in multiple partnerships, research in marketing has predominantly focused on individual alliances without considering the important interdependencies among different alliances. This study takes a portfolio approach to examine the resource diversity of multiple alliance partners and its contribution to firm innovation. While diversity is generally viewed as beneficial for innovation, this study argues that resource diversity in an alliance portfolio can only contribute to innovation when diverse resources and information are shared across alliances or with other activities in the firm. Thus the benefit of resource diversity is dependent upon effective coordination across different alliances. This study examines factors that may facilitate or inhibit coordination across alliances and thus influence the realization of the benefit of resource diversity in an alliance portfolio. It identifies a number of moderating factors along three dimensions including the composition of an alliance portfolio, alliance governance, and the market environment. By doing so, this study not only demonstrates the boundary conditions for a firm to benefit from diverse partners, but also highlights the importance of coordination among different alliances suggesting a portfolio approach for partnership research in marketing. Should retailers consider product returns when merchandising? We study how the optimal assortment decision of a price-taking retailer is influenced by its return policy in make-to-order (MTO) and make-to-stock (MTS) environments. We model individual consumer behavior in nested multinomial logit fashion, with purchase decisions in the first stage and keep/return decisions in the second stage. The retailer selects its assortment from an exogenous set of horizontally differentiated products. We call products with high (low) attractiveness popular (eccentric), because they are more (less) likely to be purchased by a typical consumer. Our main finding is that the optimal assortment has a counterintuitive structure for relatively strict return policies: It is optimal to offer a mix of the most popular and the most eccentric products when the refund amount upon return is sufficiently low. In contrast, if the refund is sufficiently high, or when returns are disallowed, optimal assortment is composed of only the most popular products. The structure of the optimal assortment is invariant to operational environment. Moreover, we show that the structure of optimal assortment differs between MTO and MTS environments when offering variety has a negligible fixed cost; argue that a more lenient return policy may not necessarily imply less variety; and take steps to verify the robustness of our findings to the retailer optimizing the return policy, to returned items cannibalizing the sales of new items, to quantity-dependent salvage values, and to the consumers reselling rather than returning. In summary, we conclude that retailers should carefully consider their return policy when merchandising, especially if it is sufficiently more strict than a full-refund policy. ■ TD09 Founders III Retailing I: General Contributed Session Chair: Umut Konus, Assistant Professor, Eindhoven University of Technology (TU/e), School of Industrial Engineering, TU/e ITEM Group IE&IS Pav.M.08, Eindhoven, 5600MB, Netherlands, [email protected] 1 - The Effect of Brand Assortment Shares on National Brand Performance Across U.S. Supermarkets Minha Hwang, Assistant Professor, McGill University, Samuel Bronfman Building, 1001 Sherbrooke Street West, Montreal, QC, H3A1G5, Canada, [email protected], Raphael Thomadsen 4 - Tracking Holistic Customer Experience in Realtime Umut Konus, Assistant Professor, Eindhoven University of Technology (TU/e), School of Industrial Engineering, TU/e ITEM Group IE&IS Pav.M.08, Eindhoven, 5600MB, Netherlands, [email protected], Emma MacDonald, Hugh Wilson Recent research conceptualizes customer experience as the customer’s subjective response to the holistic direct and indirect brand encounter. Studies to date have, however, focused on parts of this holistic encounter such as communications or the retail environment. We propose a new method for tracking real-time experience using SMS (text) messages per encounter. 2506 consumers reported, via a structured SMS message, whenever they encountered either of two focal brands in a tracking period of a week, providing data on both encounter occurrence and encounter positivity (valenced affective response). As compared with survey methods, real-time insight offers the advantage of not relying on memory, which is particularly important for capturing affective response. We apply this method to examine the impact of customer’s encounters with various customer touch-points on brand preference for two soft drink brands. In our model we consider six encounter types: television and online advertisements, in-store and bar/restaurant communications, seeing others drinking, and word-of-mouth. Our results reveal that relative impacts of customer brand encounters through different touch-points vary by brand. Realtime encounter positivity adds explanatory power to our model. Positive encounters with TV ads, in-store communications and bar/restaurant communications have a positive impact on brand preference for both brands. The method may help managers to allocate resources across the marketing plan. This paper extends the literature that discussed the local nature of national brand market shares by empirically investigating the performance of leading national brand market shares across U.S. supermarkets. Variance decomposition analyses of storelevel brand market shares of the top two national brands in six consumer packaged goods categories indicate the presence of large account-level components in national brand market shares. Specifically, we find that chain-level effects account for 24% of variation in brand shares, after controlling for variation across markets. We also note that there is substantial cross-chain variation in the assortments offered by different chain, and that chain-level effects account for 35% of the variation in a brand’s assortment share in a store, even controlling for market-level effects, which explain 51% of a brand’s assortment share. To gain more insight into the origin of the chainlevel effects, we investigate whether the association between market shares and assortment shares is causal. We find that chain’s brand assortment explains, on average, 57% of variance in market shares that can be attributed to chain-level components. We provide evidence that the estimated causal effects of brand assortment shares are robust to potential simultaneity biases. Taken together, these results suggest that the depth of distribution is among the major drivers of national brand market shares across stores. 31 TD10 MARKETING SCIENCE CONFERENCE – 2011 ■ TD10 ■ TD11 Founders IV Champions Center I Consumer Behavior: Decision Making Advertising Content Contributed Session Contributed Session Chair: Berna Basar, Research Assistant, Okan University, Akfirat Kampusu Formula1 Pisti Yani, Istanbul, Turkey, [email protected] 1 - Consumer Gratitude and Customer Loyalty: Moderating Effect of Stage of Relationship, Gender and Age Prem Dewani, Doctoral Student, Indian Institute of Management, Ahmedabad, Vastrapur, Ahmedabad, Gujarat, Ahmedabad, 380015, India, [email protected], Anurag Kansal Chair: Larry Garber, Associate Professor, Elon University, 2075 Campus Box, Elon, NC, 27244, United States of America, [email protected] 1 - The Role of Brand Construal and Affect Valence in Comparative Advertising Ying Ho, Assistant Professor, University of Macau, Faculty of Business Administration, Taipa, Macau, [email protected], Candy K. Y. Ho Comparative advertising is often used to persuade consumers that the advertised brand is relatively superior to its competitors. Focusing on how the advertised brand is compared with its competitors, this research argues that the comparison process may affect the way the advertised brand is construed and hence the persuasiveness of comparative advertisements with different affective appeals. We propose that when consumers are prompted to construe the advertised brand at a low (high) level, negative (positive) affect advertisements induce higher brand evaluation. Our argument is based on the construal fit explanation (Kim et al. 2009; Wong 2009; Zhao and Xie 2010), where a match of the construal level of an object and that of processing style increases the perceived relevance and processing fluency of the information, thereby increasing information persuasiveness. On one hand, with reference to the construal level theory (Trope and Liberman 2000), we argue that the advertised brand can be construed at different levels depending on how the comparison is framed in the comparative advertisement (e.g., comparison within a general product category versus comparison with a specific brand; comparison of feasibility versus desirability features, etc.) On the other hand, research on affect and information processing (e.g., Gasper and Clore 2002) suggests that positive versus negative affect prompts global versus local information processing. We propose that the low versus high construal level of advertised brand will cause a match/mismatch with the processing style prompted by negative versus positive affect advertisements, thereby leading to differential ad persuasiveness. Most theories of relationship marketing (RM) emphasize the mediating role of trust and commitment in business relationships. In spite of the acknowledgement of reciprocity and gratitude as a basis of relationships, applications of gratitude have been limited to explain theories of psychology and sociology. Role of consumer gratitude in developing and maintaining long term business relations are at fledging stage in marketing literature. Paper explains and empirically validates the mediating role of consumer gratitude (along with trust and commitment) in RM investment, customer purchase intention & customer loyalty. Paper further attempts to find the role of stage of relationship, gender and age of customer as moderators of RM investments and consumer gratitude. Confirmatory Factor Analysis (CFA) is done on a sample of 282 participants (140 female and 142 male). The proposed model shows a good model fit (> 0.92) for both combined data set and for male & female data set separately. The mediating role of consumer gratitude in RM investment and Customer Purchase Intension & Customer loyalty is validated empirically. Further, series of Analysis of Variance (ANOVA) is done to validate the moderating effect of stage of relationship, gender and age of customer on customer gratitude. It is found that age of relationship, gender of the customer and age of customers significantly affect consumer gratitude for a given RM investment. It is concluded that gratitude plays key role in development and maintenance of relations and stage of relationship, gender and age of customers are found important segmentation variables. Managerial implications and theoretical contribution of the study are also discussed. 2 - The Influence of Product-placement Clutter and Other Context Variables on Brand Attitude and Memory Pola Gupta, Professor of Marketing, Wright State University, Department of Marketing, 3640 Colonel Glenn Hwy., Dayton, OH, 45435, United States of America, [email protected] 2 - Impact of Visual and Tactile Input on Variety Seeking Behavior Subhash Jha, Visiting Research Scholar, The University of Memphis, 208 Windover Road, Apartment 1, Memphis, TN, 38111, United States of America, [email protected], S. (Sivkumaran) Bhardawaj In response to viewers’ resentment of ad cluster, the decline in free time due to other competing options such as cell phones, text messaging, satellite dishes and the Internet, television advertisers are coming up with more creative and deceptive antizapping strategies (e.g., product placements). With product placements, the ads are camouflaged because of their seamless integration with the programming content. The growth in the body of literature addressing product placements has revealed multiple variables that affect consumer response to placements. Although the clutter in product placements is widely reported in popular press, its effect on brand attitude and memory has not yet been investigated. This research attempts to fill that void by examining the different effects of product placement clutter on memory and attitude. The proposed research will also assess the impact of demographic variables (age, gender, education, income, culture) on perceived product placement clutter in movies, T.V. shows and game shows. Several hypotheses relating to clutter in product placement are proposed in this study. Sensory aspects have proved to be the underpinnings of many feeling based hedonic behaviors in the consumer behavior literature; however there has been little empirical research on the impact of sensory stimuli on variety seeking behavior. Variety seeking behavior is a kind of low-effort feelings-based decision-making, which largely get manifested in affective laden consumption. Prior research on variety seeking is limited to the impact of taste sense only and largely igroned the potential impact of other senses such as touch, vision, smell and sound. Hence, the present work is developed on a Stimulus Organism Response theory to addresses this gap with a conceptual framework incorporating three set of variables related to individuals, products and stores capturing two important sensory dimensions i.e. touch and vision. It provides a more comprehensive depiction of how variety seeking behavior is theoretically related to tactile and visual dimensions of a product and store. Two studies are proposed to be conducted among retail shoppers- a lab experiment and a mall intercept survey- to test the hypotheses. The findings of this work will make a strong theoretical contribution by paving the way for greater understanding of these two common but distinct and complex behaviors from a sensory marketing perspective. It will help marketing managers carefully design the sensory stimuli of a retail store to encourage variety seeking behavior. At the end, it outlines the limitations of research findings and possible avenues for future research. 3 - The Effects of Shape Complexity and Presentation Larry Garber, Associate Professor, Elon University, 2075 Campus Box, Elon, NC, 27244, United States of America, [email protected], Eva Hyatt, Unal Boya In a recent study (Garber, Hyatt and Boya 2009), we found evidence that packages of simple form, such as cylinders, appear larger than packages of equal volume and complex form; that is, those exhibiting distinct contiguous parts such as caps and necks, shoulders, bodies and feet. This result is directly opposite of results of two prior studies (Folkes and Matta 2004; Raghubir and Krishna 1999), who find that packages of complex form appear larger than packages of equal volume but simple form. In both experiments, packages were presented serially in pairs. The purpose of the present series of studies is to provide an explanation by specifying the conditions under which our result or the opposite will occur. Subjects estimated the relative volumes of packages whose shapes ranged from the simple to the complex, presented in a succession of contexts that also ranged in level of visual complexity. Results indicate that shape complexity significantly affects volume appearance independently of height, though the direction of the effect (i.e., which shape type appears larger than the other) depends upon the complexity of the context in which the packages are presented. This reversal suggests that the volume estimation strategies employed by consumers change, becoming more heuristic due to the cognitive effort required when the context in which volume estimation takes place becomes sufficiently visually complex. Theoretical and managerial implications are discussed, as well as future research directions. 3 - Turkish Gift Buying Attitudes in Today’s Marketing Environment Berna Basar, Research Assistant, Okan University, Akfirat Kampusu Formula1 Pisti Yani, Istanbul, Turkey, [email protected], A. Banu Elmadag Bas This study examines gift selection behavior of Y Generation in Turkey’s marketing environment. The motives and factors influencing the gift giving decisions of adults have been explored using qualitative data. Previous studies have not focused on the relation between the gift buying habits and buyers’ reaction to marketing elements in Turkey. In order to get insight about gifting habits of Y Generation, thirty two indepth interviews were carried out. Besides its availability, this group has been chosen since the technology, mass marketing, and popular culture in which today’s youth grew up, differentiated Y Generation from previous youth cultures. Giving gifts to parents and/or romantic partners are considered to be the most important gift giving occasions. While the gifts bought for family members are more utilitarian, bought without any price limit and without reciprocal obligation; gifts bought for romantic others are more expressive, customized and requires some amount of reciprocity. In addition to that, male participants spend the highest amount of money on gifts when they are buying a wedding gift to their romantic partner. Although there is no consensus on the importance of brand name in gift giving, quality is considered as a gift selection criteria by most of the participants. These participants prefer to buy gifts according to the quality standards set within their environment. It is also obvious that some brand stores deliver high quality gifts in the eyes of the customers. 32 MARKETING SCIENCE CONFERENCE – 2011 ■ TD12 TD13 4 - An Empirical Investigation of Sponsored Search Engine Advertising Pricing Ming Cheng, PhD Student, Rutgers Business School, 1 Washington Park, Newark, NJ, 07102, United States of America, [email protected], Lei Wang, S. Chan Choi Champions Center II Bidding Contributed Session Search engine advertising has become a popular way of advertising. Advertisers need to choose a selection of keywords and decide the bidding price for each keyword. Optimizing these decisions can significantly improve the effectiveness of search engine advertising. In this paper, we build an empirical model to describe how keyword selection and bidding price affect different levels of consumer interest. Using data from a leading search engine company in Asia, we study the heterogeneous effects across industries and identify factors that influence advertisers’ optimal keyword selection and bidding strategy. Chair: Ming Cheng, PhD Student, Rutgers Business School, 1 Washington Park, Newark, NJ, 07102, United States of America, [email protected] 1 - Coordinating Traditional and Search Advertising Alex Kim, Purdue University, 403 W. State St, West Lafayette, IN, United States of America, [email protected], Subramanian Balachander Search advertising is a rapidly growing form of online advertising. Unlike traditional media advertising, which is typically sold on the basis of audience demographics, search advertising is sold on the basis of the keyword through an auction. The new world of search advertising raises questions about how media planners should incorporate this in their decision-making. In choosing traditional advertising media, firms typically use measures like CPM (cost-per-thousand consumers reached) to compare different media. However, the question arises as to whether such measures can translate seamlessly to search advertising. i.e., can firms use CPM from traditional media to decide whether to participate and how much to bid in search advertising? In this paper, we analyze this critical issue using a vertically differentiated duopoly model in which firms make decisions on coordinating traditional and search advertising. We find interestingly that firms should not use CPM from traditional media in search advertising. We find that firms may undertake search advertising even if the minimum cost of search advertising exceeds CPM in traditional media. The reason is because of the strategic benefits firms expect to derive from undertaking search advertising. The firm can derive strategic benefits by pre-empting competitors in expanding the awareness through search advertising. On the other hand, the firm may also raise competitors’ cost for a keyword even if it loses the auction. However, we also find that the pursuit of these strategic benefits can result in a prisoner’s dilemma in which the firms’ profits are decreased with search advertising. ■ TD13 Champions Center III Quantifying the Profit Impact of Marketing IV Cluster: Special Sessions Invited Session Chair: Xueming Luo, Eunice & James L. West Distinguished Professor of Marketing, The University of Texas at Arlington, Arlington, TX, United States of America, [email protected] 1 - The More Efficient the Better: Advertising Efficiency and its Impact on Firm’s Financial Performance Jin-Woo Kim, University of Texas at Arlington, Arlingron, TX, United States of America, [email protected], Traci Freling Since there has been a call for more financial accountability in marketing, emphasis has been placed on determining the impacts of marketing activities including customer satisfaction, new product development, corporate social responsibility, and brand equity. Prior research suggests that advertising is positively related to firms’ financial performance. However, several investigations empirically demonstrated that efficiency in marketing can improve a firm’s financial rewards. The current study examines how the stock market reacts to advertising efficiency and advertising effectiveness in the context of Super Bowl advertising. Data Envelopment Analysis (DEA) was used to assess advertising efficiency (advertising executional efficiency). Three advertising executional factors were considered as DEA inputs: (1) advertising length (minutes); (2) frequency (count); and (3) number of brands promoted. Two types of advertising effectiveness were included as DEA outputs: (1) Ad Meter ratings; and (2) Nielsen rating. An event study was conducted to estimate the abnormal stock returns of Super Bowl advertisers. Event study results show that Super Bowl advertising from 2004 to 2008 is positively related to abnormal stock returns for advertisers. Finally, cross-sectional regression analysis was applied to test the impact of advertising efficiency on Super Bowl advertisers’ cumulative abnormal returns. Preliminary results show that ad efficiency is positively associated with abnormal return while Ad Meter and Nielsen ratings are not related to abnormal returns. The findings suggest that being efficient is more important in generating positive abnormal returns. 2 - Modeling Price Dynamics in Simultaneous Auctions: A Bayesian Factor Analytic Approach Norris Bruce, Associate Professor, The University of Texas at Dallas, 800 West Campbell Road, Richardson, TX, 75080-3021, United States of America, [email protected] Auctions have become ubiquitous, partly because of the popularity of internet auctions and that of auctioning public assets, such as oil and gas exploration rights, bus routes, and spectrum licenses. This ubiquity has given us rich sources of data, and the chance but also the challenge to create statistical methods to study features, such as an auction’s price dynamics and the factors that may influence the evolution of prices in various types of auctions. This work thus presents a Bayesian dynamic factor model to study price dynamics in a simultaneous auction; a format in which the seller offers multiple items for sale at the same time, and participants bid over multiple rounds for the items in which they are interested. The factor approach decomposes each bid into two components: 1) temporal component that varies over the rounds, driven by interactions among bidders; and 2) a cross-sectional component that measures how each item and relatedness among items influence willingness to pay. The sequential Bayesian approach helps mitigate the missing or partially observed data problem: here the fact that participants may bid on a subset of the available items in each round, and bid only in a subset of the rounds. Finally, I apply the factor model to a particular example, FCC spectrum auctions, and identify it by exploiting the format of the auction itself: i.e. the fact that participants can bid on multiple items in the each round. The main results reveal the bidding behaviors of the participants, showing how their bids evolve over time as function of not only design variables and temporal competition, but also their desire to acquire related items. 2 - Marketing Spending, Analyst Coverage, and Firm Performance in the IPO Market Monica Fine, Florida Atlantic University, Boca Raton, FL, United States of America, [email protected], Kimberly Gleason Research linking marketing to financial outputs has been gaining significance in the marketing discipline. Marketing is being forced to speak the language of finance and relate expenditures to measures of firm performance (McAlister et al. 2007). Research in marketing often looks at soft measures of sales or customer satisfaction, but tends to ignore the actual impact on the bottom line (hard measures) (Joshi et al. 2010). This research investigates the effects of marketing spending and analyst coverage during IPOs. This paper contributes by investigating advertising’s impact on analyst coverage instead of focusing on soft measures such as customer satisfaction (Ngobo et al. 2009) or media coverage (Rinallo et al. 2009). Theories from marketing and finance, market-based assets theory and signaling theory respectively, serve as the conceptual basis of this paper. Our research questions include: Does marketing spending send a signal to analyst and investors about the financial health of the firm? Does marketing spending lead to more analyst coverage? Lower forecast error? And More favorable analyst coverage? We will also investigate the possibility of marketing spending and analyst coverage joint determination in the model. Therefore, we will test to see if the two variables are endogenous. Our test and previous research will produce skewed results if endogeneity is an issue. Thus, after controlling for the endogeneity of marketing spending and analyst coverage we will uncover more accurate results. Analyst coverage is often found to be endogenous, but to our knowledge marketing spending and analyst coverage have not been tested for endogeneity. 3 - An Investigation of Market Learning and its Implications for an IP Auction House Joseph Derby, Doctoral Student, Texas Tech University, Rawls College of Business, Lubbock, TX, 79409, United States of America, [email protected], Mayukh Dass In this paper, we study how firms modify their operations based on their market learning and how such changes affect their future performance. The focus here is on Intellectual Property (IP) auction house called Ocean Tomo, which recently started selling IP through organized auctions. With IP representing 70% of the asset values, it has become a major influence in the success of US corporations. We collected data from eight IP auctions held between 2006-2008 and longitudinally examine the evolution of IP exchanges. In particular, we examine how the effects of various drivers including market characteristics, product characteristics, and seller characteristics on IP valuation changed over time. Next, we investigate the apparent changes made by the auction house overtime, and how such changes affected the auction outcomes. We conclude by drawing implications for service organizations such as the auction house, which operate businesses in two-sided markets. 33 TD14 MARKETING SCIENCE CONFERENCE – 2011 3 - Can Stock Markets Really Predict the Future? Case of Product Innovations M. Berk Talay, University of Massachusetts-Lowell, 1 University Avenue, Falmouth Hall 207C, Lowell, MA, 01854, United States of America, [email protected], M. Billur Akdeniz some costumers satisfy most of their requirement from one of their distributors, other consistently split their demands among them. Our proposed methodology could also help to guide strategic decision making. In our empirical application we found that price sensitivity of customers making most of their purchases with the focal supplier are less affected by the volume of purchases in previous periods. We expect these results to provide valuable information for vendors to negotiate prices with the customers. Over the last decade, there has been a spurt on the discussion about the value of, and return on, marketing investments. Vast majority of the studies on return on marketing investments are based on the efficient market hypothesis (EMH), which asserts that favorable prospects of a company will drive its stock price up, and vice versa. Nevertheless, both the validity of the EMH and – more importantly - its relevancy has been cast aspersions. Specifically, Hanssens et al. (2009) call for future research to challenge the efficient market hypothesis. We address this call by a multimethod analysis of the link between the stock price of a company and the future success of its marketing actions. On contrary to the previous studies, we analyze the extent to which changes in the stock price can accurately augur the future success (or failure) of a marketing action. Specifically, we focus on the new product introductions and identify important new product introductions and calculate their forecasted ROIs using three different methods: event study, buy-and-hold abnormal returns, and calendar-time portfolio returns. Then, we assess the performance of these new product introductions in both absolute (e.g., total sales) and relative terms (e.g., market share). Finally, we analyze the extent to which a positive (negative) abnormal return on the market value of a company is related with the success (failure) of a new product it launches to the market. 2 - Resale Price Maintenance when Retailers are Heterogeneous Charles Ingene, Chair Professor, The Hong Kong Polytechnic University, M801, Li Ka Shing Tower, Hung Hom, Kowloon, Hong Kong - PRC, [email protected], Mark Parry, Zibin Xu Manufacturers can now legally use Resale Price Maintenance (RPM) to affect the prices charged by their retailers. We derive the optimal RPM strategy for a manufacturer that sells to competing retailers who (i) may be homogeneous or heterogeneous (have equal or different demands); (ii) may or may not be in competition; and (iii) may or may not free ride on a rival’s service. We derive seven major conclusions that hold with or without free riding. First, for most combinations of service effectiveness and retailer heterogeneity, a manufacturer prefers an optimal RPM rule to not using RPM. Second, optimality may entail a maximum (ceiling) or minimum (floor) price, depending on service effectiveness. Third, when ceiling and floor yield about the same profit, the manufacturer does not use RPM. Fourth, an optimal RPM policy may not bind all retailers. Fifth, consumers generally benefit from a manufacturer-optimal RPM strategy relative to non-use of RPM: higher prices from a floor are more than counterbalanced by greater service; lower prices due to a ceiling more than compensate for a lack of service. Sixth, a representative consumer prefers exclusive distribution with RPM over broad distribution without RPM. Seventh, retailers are generally disadvantaged by RPM. 4 - Total Recall: Investor and Consumer Response Following Toyota’s Automotive Recall Robert Evans Jr., Assistant Professor, Texas A&M International University, 5201 University Boulevard, 217C Western Hemespheric Trade Center, Laredo, TX, 78045, United States of America, [email protected] 3 - MAP and RPM: Determinants of Violations Ayelet Israeli, Doctoral Student, Kellogg School of Management, Northwestern University, 2001 Sheridan Road, Evanston, IL, 60208, United States of America, [email protected], Eric Anderson, Anne T. Coughlan On January 21, 2010, Toyota announced the recall of 2.3 million vehicles due to poorly designed brake pedals. As a result, Toyota lost in excess of $25 billion in shareholder value within one month while their stock price plunged from $91.17 on the opening day of the announcement to $73.18 one month later. Investors clearly disliked the news of this costly announcement which was followed by suspended sales of eight models from January 26th to February 5th. Despite the severe and long-lasting reaction by investors, it appears as though consumer reaction to these events was short-lived, as sales quickly returned to pre-recall levels almost immediately following the recall month and the month which included suspended sales. Despite the recall announcement being the single recognized major event during the time period examined, investors and consumers seem to react with different magnitudes and duration, perhaps showing that there are different theoretical bases for explaining behavior of the two distinct groups to a singular event. Traditionally, the efficient market hypothesis (Fama 1970) has been used to explain shareholder reaction to firm news, meaning that investors take information as it is released to the market, evaluate the strength and value of that information, which, in turn, is immediately incorporated into the firm stock price, while customerbased brand equity (Keller 1993) may explain consumer response to the same news. In responding to the call to treat the investment community as consumers (http://www.msi.org/research), delineation of the appropriate theoretical context in which to examine these groups is imperative. Implications for managers, researchers and theory are discussed. Minimum Advertised Price (MAP) and Resale Price Maintenance (RPM) are two types of vertical restraints that manufacturers may use to influence retail prices. While these pricing restraints are common practice in many industries, they have received little empirical attention. In this paper, we study a unique database and provide empirical evidence of compliance with pricing programs. We show that compliance varies significantly among both products and retailers. Moreover, we show that higher prices and higher commitment to the brand are associated with higher compliance rates. Yet, high distribution intensity and shipping rates are associated with lower compliance rates. Compliance rates also vary by product category and by the online venue where the product is sold. Finally, we study how product and retail characteristics relate to the depth of violations. These findings suggest that commitment to brand and distribution intensity determine manufacturers’ ability to effectively monitor and enforce the pricing program, which in turn affects compliance rates. 4 - Coordination of Price Promotions in Complementary Categories Maxim Sinitsyn, McGill University, 855 Sherbrooke St. W, Dept of Economics, Room 443, Montreal, QC, H3A2T7, Canada, [email protected] In this paper, I investigate the outcome of a price competition between two firms, each producing two complementary products. Specifically, I study each firm’s decision to coordinate price promotions of its products. Consumers are divided into loyals, who purchase both products from their preferred firm, and heterogeneous switchers, who can mix and match between the four possible bundles. The switchers are willing to pay some price premium in order to purchase two complementary products produced by the same firm, as they believe that these products are a better match than two products produced by different firms. I find that each firm predominantly promotes its complementary products together. This finding is supported by data in the shampoo/conditioner and cake mix/cake frosting categories. ■ TD14 Champions Center VI Pricing and Competition Contributed Session Chair: Maxim Sinitsyn, McGill University, 855 Sherbrooke St. W, Dept of Economics, Room 443, Montreal, QC, H3A2T7, Canada, [email protected] 1 - Inferring Competitor Pricing with Incomplete Information Marcel Goic, Assistant Professor of Marketing, University of Chile, Republica 701, Santiago, Chile, [email protected], Alan Montgomery ■ TD15 Champions Center V CRM II: Customer Loyalty We study how business customers make multi-product purchase decisions and how the distributors who sell those products can make inferences about their demand functions with incomplete information. The problem is that distributors rarely observe a competitor’s price directly, and must infer competitor response indirectly from their own observations about customer purchases. In this research we propose that customers make their product orders by minimizing procurement costs and we impose first order conditions to characterize regions in the parameter space where consumer will buy from each distributor. We use those conditions to estimate an empirical model of purchase behavior that enables us to identify the likelihood of each consumer buying from the competitor or simply changing his consumption patterns. We apply our proposed model to a wholesale food distributor and we find widespread heterogeneity in purchase patterns. As expected some customers are loyal, while others are not, and the remainder fall in between. The empirical results shed light on the competitive elements of customer demand that cannot be study with traditional reduced form response models. For example we found that while Contributed Session Chair: Harmeen Soch, Assistant Professor, Guru Nanak Dev University, Department of Commerce and Business Mgt., G. T. Road, Amritsar, Pb, 143005, India, [email protected] 1 - Allocating Optimal Multi-period Budget to Loyalty and Sales Promotion Programs Hsiu-Yuan Tsao, Associate Professor, Tamkang University, No.151, Yingzhuan Road, Danshui District, New Taipei, 25137, Taiwan - ROC, [email protected], Li-Wei Chen, Hsiu-Feng Yan This study applies a first order Markov type market share model to examine the relative impact of loyalty and promotion effects on market share, and then determines the allocation of optimal multi-period budget between loyalty and sales promotion programs by nonlinear dynamic programming to minimize budget the 34 MARKETING SCIENCE CONFERENCE – 2011 FA01 Friday, 8:30am - 10:00am while maintaining market growth rate during the multi-period stretch. Applying this approach to data from a consumer panel for 2009 provided by Taylor Nelson Sofres (TNS) Global for three product categories comprising adult milk powder, shampoo, and detergent, sensitivity analysis reveals that the impact of the promotion effect on market share exceeds that of the loyalty effect. However, based on the given initial size of market share and of the loyalty effect, and on an estimation of the promotion effect, with empirical data concerning the outlay budgeted for loyalty and promotion programs, nonlinear dynamic programming is used to demonstrates that the optimal budget decision for the coming year is to allocate more to loyalty than to promotion, thus minimizing budget while maintaining market growth over the multi-period stretch. With this combination of a Markov type market share model and nonlinear dynamic programming, the study provides a platform for exploring the dynamic impacts of the long-term loyalty effect and the short-term promotion effect on market share, and helps determine the allocation of optimal multi-period budget between loyalty and promotion programs. ■ FA01 Legends Ballroom I Choice III: More Effects on … Contributed Session Chair: Zheyin (Jane) Gu, The State University of New York, Albany, 1400 Washington Ave. BA 336, Albany, NY, 12222, United States of America, [email protected] 1 - Capturing the Unobserved Comparison Effects in Consumer Choices: A Hierarchical ME Model Ping Wang, Doctoral Student of Marketing, Peking University, Guanghua School of Management, Beijing, 100871, China, [email protected], Jaihak Chung, Meng Su, Luping Sun 2 - The Impact of Loyalty Program on Loyalty Transfer within the Partnership Network So Young Lee, Korea University, 507, LG-POSCO Bld., Anam-Dong, Sungbuk, Seoul, Korea, Republic of, [email protected], Hyang Mi Kim, Jae Wook Kim Traditional choice models assume that product utility is determined by the attributes of the product and the corresponding preference weights only. However, much research finds that reference points can systematically shift consumers’ preferences. This reference product (RP) can be either external (other products in the current choice set) or internal (ideal products in mind or products that consumers currently own). If a consumer compares products under consideration, the utility of a product evaluated by the consumer is determined not only by the attribute levels of the product itself but also by those of another product in the same choice set or in their minds. This comparison will exert similar effects on utility as product attributes. This study proposes a consumer choice model taking into account comparison effect by capturing the impact of both types of RPs and investigating the source of the RPs. In order to test its validity, we collect the consumer preference data for smart phones with a choice-based conjoint analysis. By applying a hierarchical mixture-of-experts model, we incorporate both consumer heterogeneity and product heterogeneity in the model and identify factors that contribute most to different sources of RPs. We find that the RP can be external. It can be either the most preferred product, the least preferred product or holistic impression of all the products in the choice set. The choice of external RP is determined by consumer characteristics, expertise and product attributes. However, under certain circumstances, the RP can also be internal. The choice of internal RP is determined by usage time, product familiarity and satisfaction level. The findings provide more insights into the unobserved comparison effects with a hierarchical ME model. In recent years, increasing number of businesses are offering partnership loyalty program (PLP). The PLP network generally includes diverse sectors of businesses such as credit card companies and airlines, hotels and telecomm companies as well as family restaurants and beauty parlors. Despite the prevalence of PLP, most previous studies related to the loyalty program have largely focused on the dyadic relationship between a single company and customer group, and little attention has been given to the PLP and issues dealing with relationships among customers, businesses and PLPs. This study examines the impact of PLP on customer loyalty and analyzes the manner in which the loyalty becomes diffused within the loyalty program network. In particular, we explores how (1) the loyalty towards one specific store affects the loyalty towards the PLP, (2) the loyalty towards the PLP transfers to other partner stores, and (3) the loyalty towards the PLP feeds back to the first store. The data obtained from an online survey together with the actual purchase behaviors information extracted from corporate database of 2,021 PLP members shows the following: (1) For customers who have been loyal to a specific firm/store, not only does their loyalty to the PLP increase but also their loyalty to that business is reinforced. (2) The customers’ loyalty formed in such a manner does transfer to other partner firms/stores within the network. Understanding the loyalty diffusion mechanism within the PLP network can help marketers to create effective strategies for relationship-marketing management and support decision on choosing the right partner with which they can share the loyalty program. It can also assist a potential partner firm in deciding whether to join the PLP network or not. 2 - Learning Dynamics in Product Relaunch Sue Ryung Chang, New York University, 40 West 4th Street, New York, NY, United States of America, [email protected], Tulin Erdem 3 - Influence of Perceived Relationship Investment and Cross-buying on Share-of-Wallet Harmeen Soch, Assistant Professor, Guru Nanak Dev University, Department of Commerce and Business Mgt., G. T. Road, Amritsar, Pb, 143005, India, [email protected], Navneet Multani Relaunch is the reintroduction of a product or marketing campaign after it has been discontinued for a period of time and undergone some improvement or change (Barron’s Marketing Term Dictionary 1994). Even though it is a very common and popular strategy in many industries such as consumer packaged goods and automobile industry, there has been little empirical research on product relaunch in either marketing or economics. Relaunch has higher risk than line extension since it involves changes to the existing product and hence there is no guarantee that the customers who used to buy the pre-relaunched product will like and keep buying a relaunched product as well. In this research, we account for the learning dynamics in the context of product relaunch. Specifically, we investigate how the overall preference of an original (pre-relaunch) product is related to that of relaunched product. We do so by allowing consumers to be uncertain about product quality and transfer perceptions of the old product to the relaunched product. Also, we examine how the changes in preference influence optimal advertising decisions of firms. This study develops a conceptual framework for understanding the influence of perceived relationship investment (firm’s effort to strengthen relationship with customers) and cross-buying (purchasing additional products from an existing firm) on the relationship quality (assessment of the strength of relationship) between a firm and customer which in turn affects share-of-wallet (amount and frequency with which the customers make purchases with a particular service provider). This research advances the existing literature on perceived relationship investment which was till now being determined by tangible rewards (firm’s tangible offers), interpersonal communication (firm’s interaction with regular customers), preferential treatment (firm’s special treatment for its regular customers) and direct mail (customer’s perception of extent to which retailer keeps its regular customers informed). The study incorporates other antecedents of perceived relationship investment like store size, product assortment (collection of different products for sale) and promotion (an activity designed to increase visibility or sales) as well. Direct mail moderates cross-buying which is determined by perceived fairness of price, reputation of a firm and the locational convenience for customers. Cross-buying also has a direct effect on share-of-wallet. Keeping in mind these relationships, this study is the first of its kind to develop a unique model for understanding the joint effect of relationship quality and cross-buying on share-of-wallet. This study is proposed to be conducted in shopping malls and will help retailers understand the resources, efforts and attention needed to maintain and enhance relationships with loyal customers which increases share-of-wallet and profitability for firms. 3 - Consumer Attribute-based Learning and Retailer Category Management Strategies Zheyin (Jane) Gu, The State University of New York, Albany, 1400 Washington Ave. BA 336, Albany, NY, 12222, United States of America, [email protected], Sha Yang We develop a joint demand and supply framework to empirically study consumers’ learning behaviors when they face a large number of uncertain choice alternatives in a product category. On the demand side, we apply the Bayesian learning framework and examine consumer learning based on multiple product attributes. This attributebased approach allows us to model both within-product and cross-product learning with a product category.On the supply side, we model retailers as strategic decision makers who choose the optimal price for each product item they carry to maximize total category profits. We apply the model to a household-level data set of ready-toeat cereal purchases obtained from Information Resources, Inc., and we empirically demonstrate strong attribute-based consumer learning activities. First, we find significant within-product learning for all three attributes that we model: brand, flavor, and grain type. In addition the experience signals are much nosier for brand than for flavor or grain type. Second, we find significant cross-product learning for brand and grain type. Our policy experiments suggest that retailers can obtain greater category profits by encouraging cross-product learning. In addition, encouraging cross-product learning is more profitable on the attribute for which consumers perceive greater experience variability and along which the retailer assortment is more diversified. Finally, our empirical analysis suggests that retailers do not consider learning in their current practice of setting prices and that incorporating learning would allow retailers to charge higher prices and obtain greater profits. 35 FA02 MARKETING SCIENCE CONFERENCE – 2011 ■ FA02 hypotheses, we use 507,194 online movie ratings collected from Yahoo and IMDB on 564 movies released from Feb. ‘99 – Jul. ‘09. We then propose a fully functional shape model which links the diffusion curves across sequential channels. Moreover, our model also proposes a functional interaction term which allows us to investigate the moderating effect of online consumer ratings over time. The resulting model allows us to investigate the optimal shape of diffusion curves and to infer the optimal release timing between sequential channels. Legends Ballroom II UGC-I (The Evolution and Impact of Online Opinions) Cluster: Internet and Interactive Marketing Invited Session Chair: Ashish Sood, Emory University, Atlanta, GA, United States of America, [email protected] 1 - Online Product Opinions: Incidence, Evaluation and Evolution Wendy W. Moe, University of Maryland, College Park, MD, United States of America, [email protected], David Schweidel ■ FA03 Legends Ballroom III Internet: Customer Response While recent research has demonstrated the impact of online product ratings and reviews on product sales, we still have a limited understanding of the individual’s decision to contribute these opinions. In this research, we empirically model the individual’s decision to provide a product rating and investigate factors that influence this decision. Specifically, we consider how previously posted opinions in a ratings environment may affect a subsequent individual’s posting behavior, both in terms of whether to contribute (incidence) and what to contribute (evaluation), and identify selection effects that influence the incidence decision and adjustment effects that influence the evaluation decision. Our results indicate that individuals vary in their underlying behavior and their reactions to the product ratings previously posted. Systematic patterns in these behaviors have important implications for the evolution of product opinions at a site, which we illustrate through the use of simulations. We show that posted product opinions can be affected substantially by the composition of the underlying customer base and find that products with polarized customer bases may receive product ratings that evolve in a similar fashion to those with primarily negative customers as a result of the dynamics exhibited by a core group of active customers. Contributed Session Chair: Aditya Billore, Indian Institute of Management, IIM Indore Rau Pithampur Road, Rau, Indore, 453331, India, [email protected] 1 - Sales Tax and Online Consumer Behavior Nicholas Lurie, Assistant Professor, Georgia Institute of Technology, College of Management, 800 W Peachtree St NW, Atlanta, GA, 30306, United States of America, [email protected], Sriram Venkataraman, Peng Huang We examine how local sales tax affects consumers’ online purchase behavior and what this implies for the practice of electronic commerce. Although early surveybased research found that consumers who live in high-tax localities are more likely to shop online, an analysis of online transactions of 88,814 U.S. households in 2006 shows the opposite. Among other effects, we find that higher local tax rates are associated with lower online expenditures, reduced transaction frequency, and a lower probability of making an online purchase. A disaggregate analysis shows that higher sales tax does not significantly boost demand from tax avoiding retailers but significantly lowers demand from online retailers that collect tax. This demandreducing effect is directionally stronger for pure-play Internet retailers, with warehouse or administrative operations in the state where a consumer is located, than for brick-and-click retailers that operate physical as well as on-line stores. 2 - A Firm’s Optimal Response to Negative Rumors Dina Mayzlin, Associate Professor, Yale University, New Haven, CT, United States of America, [email protected], Yaniv Dover, Jiwoong Shin Negative commercial rumors can dramatically affect a company’s sales. While some of the rumors may die out on their own inflicting only minimal damage, others can be severely harmful and may require a response on the firm’s part. The problem is especially important due to the growth of online word of mouth and the capacity of firms to routinely monitor online conversations. What is not clear is when and how a company should act on the information that it receives. Is it better for the firm to respond to a rumor while relatively few people are talking about it or is the firm better off waiting until the rumor is more widespread? We model the diffusion of rumors among consumers who are heterogeneous in their preferences. We find that under certain conditions, a lack of response on the firm’s part can serve as an informative signal that the rumor is false. We also find that in the case of partially true and completely true negative rumors, the firm will respond to the rumor after a finite time interval. Consequently, we find that there is an optimal time interval after which the firm chooses to respond to rumors, and this interval depends on the impact the rumor potentially has on the market as well as the likelihood and the magnitude of contagion in that market. 2 - Trajectory-based Consumer Segmentation and Product Recommender System in the Online Market Youngsoo Kim, Singapore Management University, 80 Stamford Road, Singapore, 178902, Singapore, [email protected], Ramayya Krishnan We aim to understand a distinctive longitudinal online shopping pattern and further suggest how to utilize this better understanding. We first identify the clusters of individuals following similar progressions of online shopping behavior on three dimensions: (1) product intangibility level, (2) product price and (3) the number of transactions. We collected individual-level transaction data of the random sample of consumers at one of the premier online shopping malls in Korea. The data cover four and half years from January 2002 to June 2006. We found the distinctive patterns: (a) consumers purchasing more intangible items tend to expand their online purchased items toward more intangible products over time while consumers purchasing mainly tangible items limit themselves to highly tangible products, (b) average price-based trajectory analysis shows that the slope coefficient of the online small spender is negative but insignificant but online big spenders purchasing relatively expensive products follow a steep downward trajectory and (c) consumers showing high purchasing frequency are likely to increase their purchasing frequency, whereas consumers purchasing with relatively low purchasing frequency show constant purchasing frequency over time. We checked robustness of our analyses with time-varying covariates and dual trajectory analysis. Second, we develop a recommender system to optimize product positioning strategy using a Bayesian approach and statistical information acquired from trajectory analysis. The experimental study indicates that the proposed trajectory-based recommendation algorithm is effective despite requiring less and easily obtainable data in comparison to previous techniques. 3 - Empirically Investigating the Relationship between What Brands Do and What Consumers Say (Social Media), Sense (Mindset), and Do (Purchase) Douglas Bowman, Professor of Marketing, Emory University, 1300 Clifton Road NE, Atlanta, GA, 30322, United States of America, [email protected], Manish Tripathi The paper applies appropriate state-of-the-art methods to empirically investigate the effects on brand sales of the linkages between (1) what brands are doing in the marketplace (advertising effort and messaging; product additions, deletions, enhancements; pricing and promotion; distribution); (2) what consumers are saying (consumer-generated social media content); and, (3) what consumers are sensing (consumer attitudes and mindset towards brands). A Vector Autoregressive modeling framework is used in an initial analysis to explore lead/lag relationships among the variables. A structural model of brand sales is then presented to assess elasticities. 3 - The Impact of Personalization and Interactivity on Choice Goal Attainment and Decision Satisfaction Sally McKechnie, Associate Professor in Marketing, Nottingham University Business School, Jubilee Campus, Wollaton Road, Nottingham, NG8 1BB, United Kingdom, [email protected], Prithwiraj Nath 4 - Power of Customer Voice: Shape Analysis of Online Product Reviews to Predict Diffusion in Sequential Channels Ashish Sood, Emory University, Atlanta, GA, United States of America, [email protected], Mayukh Dass, Wolfgang Jank, Yue Tian Although the effects of interactivity and personalization tools on website browsing experience have been the subject of previous research, the impact of variable levels of such web design features on buyers’ feelings of decision satisfaction has received relatively little attention. Buyers’ satisfaction with decision-making depends on their attainment of choice goals and the choice set which the seller provides. However, there is little research on how a new-to-market e-retailer offering a limited choice set can vary its web design features to influence browsing decision satisfaction of potential buyers. Such inter-relationships are also likely to be influenced by product category knowledge and predispositon of buyers towards maximizing choice options for optimal decision-making satisfaction. Using an experimental methodology this study examines these relationships in the context of complex, high risk purchase situations, where the seller is new to the market and potential buyers are asked to make a purchase selection under a time constraint. Our findings suggest that buyers with low product knowledge perceive better goal choice attainment and are more satisified with decision-making when the e-retailer uses lower levels of interactivity and personalization features in their web design than buyers with high product Firms increasingly launch new products across multiple channels in a sequential manner. For example, following the main theatrical debut, a movie is launched in other distribution channels such as DVDs, cable channels such as PPV and streaming channels such as Netflix. While recent research highlights the impact of online product ratings on product sales, little is known about how such reviews affect the subsequent sales in sequential channels. Online consumer ratings vary significantly in terms of the total number of ratings, valence of opinions, and the dynamics over time. Traditional forecasting models, which do not account for this dynamic and changing voice of the customer fail to incorporate this potentially vital source of information into the model. The underlying thesis of the current paper is that the shape of online consumer ratings over time provides valuable information that can be used to model and predict sales and diffusion of a product when launched in a new channel. We hypothesize that there is a relationship between the diffusion curves across sequential channels, and that the shape of these curves is moderated by the shape of the online consumer ratings pattern. In order to investigate these 36 MARKETING SCIENCE CONFERENCE – 2011 FA05 3 - A Bayesian DYMIMIC Model for Forecasting Movie Viewers Dong Soo Kim, PhD Candidate, KAIST Business School, 87 Hoegiro Dongdaemoon-Gu Seoul 130-722, Seoul, Korea, Republic of, [email protected], Jaehwan Kim, Duk Bin Jun knowledge. However, buyers with a high level of predisposition towards maximizing perceive better choice goal attainment and are more satisfied with decision-making than buyers with a low level of predisposition towards maximizing when the eretailer provides better interactivity and design features on its website. This study offers a significant contribution to research and practice. In this study, we explore the issue of how to enhance forecast of the movie performance, all-time question for managers in movie industry. The conceptual core of our approach is at expected sales. The expected sales of agents in the movie market (i.e., screen managers at supply side or potential moviegoers at demand side) play important role in predicting the actual sales. We pay attention to the uniqueness of the expected sales; it is anyway latent and it is evolving over time. This leads to a quick sense that incorporating these components into the model is a natural choice and thus critical for proper forecast of the movie for future period. Based on this notion, we propose a simple DYMIMIC model for forecasting movie. The model based on a simplified intuitive story for movie consumption - spontaneous demand and socially-driven demand - was calibrated and tested on the actual movie data in United States and other countries. The model allows for evolution of the latent expected sales in the simplest way, and it captures the cross-sectional unobserved heterogeneity across countries. In the empirical analysis, the model outperforms the other benchmark models used in the literature. The model successfully demonstrates the importance of incorporating latency and time-varying construct into the model in addition to having the heterogeneity for better forecasting and it offers a simple yet informative platform of the model that one can add variables (movie attributes and marketing activities). 4 - Consumer Demographics & Changing Perception to Online Advertising: Applying Learning Curve Mechanism Aditya Billore, Indian Institute of Management, IIM Indore Rau Pithampur Road, Rau, Indore, 453331, India, [email protected], Anurag Kansal Through Epsilon technology and methodology we present an exploratory ethnographic case study of the quality of service of public transport (Metro, Bus, Commuter Car, Tram, etc...). In various European cities (London, Rome, Paris, Berlin, Madrid, Lisbon). The degree of reliability and speed with which it has conducted an exploratory analysis revealed the severity and cost reduction makes this technique is effective and efficient consolidation. ■ FA04 Legends Ballroom V Dynamic Models I 4 - Measuring Individual’s Growth in Achievement Over Time under Changing Group Affiliations Rene Algesheimer, Chair of Marketing and Market Research, University of Zurich, IFB, Plattenstrasse 14, Zurich, 8032, Switzerland, [email protected], Markus Meierer, Egon Franck, Leif Brandes Contributed Session Chair: Rene Algesheimer, Chair of Marketing and Market Research, University of Zurich, IFB, Plattenstrasse 14, Zurich, 8032, Switzerland, [email protected] 1 - Applying Conditional Three-level Nonlinear Growth Curve Modeling to Innovation Diffusion Margot Loewenberg, University of Zurich, IFB, Plattenstrasse 14, Zurich, 8032, Switzerland, [email protected], Markus Meierer, Rene Algesheimer Studying change is one of the most challenging, but also one of the most demanded topics in science. In marketing and management there is an increasing interest in examining individual change processes under the influence of changing environmental settings. In particular, scholars are interested in the evolution of customer preferences under changing peer groups or about employee’s growth in achievement while working for changing employers. In general, the influence of changing group affiliations on individual’s performance over time gains more and more interest. When studying individual growth in the presence of individual mobility scholars are faced with important methodological challenges. First, with individual mobility the data structure is not consistently clustered. Hierarchical models, which are traditionally used to correct standard errors in case of nested data structure, are not suitable. In our case individuals belong to more than one higherlevel unit. The data structure is non-hierarchical. Second, endogeneity may be an issue when individuals have the choice to which group they will move. Third, the questions must be raised how to model the growth intercept in the presence of individual mobility. While the slope is modeled as growth across the entire set of group memberships, the intercept is considered as a function of only the first group with which the individual was affiliated. In this paper we present an empirical model that captures individual’s growth in achievement over time in the presence of changing team affiliations. The model is applied with large-scale longitudinal data from the German premier soccer league and tested for robustness with data from the NBA. Implications for research and practice are discussed. Practically all scientific disciplines in the social sciences are concerned about the representation and measurement of change. Recently, using nonlinear growth curve modeling to analyze complex developmental patterns and to capture differences in trajectories over time more appropriately is gaining more and more attention. However, it has seldom been applied to hierarchical data with more than two levels. To our knowledge, this modeling technique has not been used within the marketing discipline, even though it is especially important when analyzing, e.g., quarterly sales figures nested in brands nested in companies. With this paper, we contribute to existing research on growth curve modeling by analyzing a series of conditional three-level nonlinear growth curve models. First, it is evaluated how well different types of nonlinear growth functions on different hierarchical levels are capable of describing the observed pattern of change of innovation diffusion. Second, we estimate the extent to which covariates on company- and country-level influence growth trajectories on each level. The third aim of this paper is to describe and highlight other areas of the marketing discipline where three-level growth curve models can be applied. We illustrate our model using longitudinal data on the number of broadband internet subscribers from 2001 to 2010 on a quarterly basis (level 1) from the 300 largest network operators (level 2) in 50 countries (level 3). We analyze how well nonlinear growth models of logistic, Gompertz, and Richards functions fit the data. Among other covariates, the innovative strength of a country, the corporate parent as well as the first mover effect are included in the model. We conclude with implications for practice and research. ■ FA05 Legends Ballroom VI 2 - An Asymmetric Threshold Error Correction Model of Pass-through in the U.S. Supermarket Industry Miguel Gomez, Cornell University, 246 Warren Hall, Ithaca, BY, 14850, United States of America, [email protected], Christopher Lanoue, Timothy Richards New Product V: Design & Development Contributed Session Chair: Wooseong Kang, Assistant Professor, North Carolina State University, 2332 Nelson Hall, Raleigh, NC, United States of America, [email protected] 1 - Marketing Instrument Innovations and Their Impact on New Product Performance Wenzel Drechsler, Goethe University Frankfurt, Grueneburgplatz 1, R&W Building, Frankfurt, 60323, Germany, [email protected], Martin Natter Manufacturers want retailers to pass trade deals through to consumers but, somewhat paradoxically, they would rather retailers not pass through wholesale price increases. While there is a large literature on incomplete trade-promotion pass-through, in this study we seek to explain retail pass-through of both upward and downward variation in wholesale prices. Previous explanations for incomplete pass-through include inventory costs, market power and demand curvature, among others. Recent theoretical research, however, focuses on the role of consumer search (Tappata 2009; Yuan and Han 2011). We offer an empirical test of the hypothesis that incomplete, asymmetric pass-through behavior is driven largely by consumer search behavior and that suppliers can mitigate incomplete pass-through with properly designed pull strategies. Consumer search implies that retail prices will adjust more rapidly, and completely, upward rather than downward so we estimate an asymmetric threshold error correction model (ATECM). We use data describing four product categories (coffee, yogurt, orange juice and frozen entrees) in four markets (Minneapolis, St. Louis, Pittsburgh and Hartford) on a weekly basis for 2007-2008 to test for both the existence of thresholds in the wholesale-retail price relationship and for the presence of asymmetries in pass-through. Our tests of the existence of threshold effects follow Tsay (1997), while we calculate brand-level threshold values using the grid search method of Chan (1993). Our preliminary results provide evidence of thresholds and asymmetries in pass-through and generally support the consumer search hypothesis. Firms increasingly use innovative marketing instruments such as new pricing or advertising methods to promote their new products. This development increases the importance of assessing the overall effectiveness of marketing instrument innovations with regard to new product performance. Using data from 2,003 firms this study shows that marketing instrument innovations in general positively contribute to new product performance. In particular, innovations with regard to design, sales channel and pricing strongly support the performance of new products. New advertising methods, on the contrary, show no significant effect on new product performance. 37 FA06 MARKETING SCIENCE CONFERENCE – 2011 2 - Investigating the Relationship between R&D and Marketing in the New Product Development Process Suj Chandrasekhar, Principal, Strategic Insights Inc., 162 Collins Road NE, Cedar Rapids, IA, 52402, United States of America, [email protected], Srinath Gopalakrishna to the manufacturer about final demands of a new product. In this paper, we attempt to analyze how information sharing influences new product development in a nonintegrated channel. We build a game-theoretic model where the retailer facing two segments of consumers, has private information about the final demand of the new product, while it is the manufacturer that makes the new product investment before demand realization. We obtain the following results. (i)The value of information sharing depends on the positioning of the new product item relative to the manufacturer’s current product item. Under some regularity conditions, when the new product that the manufacturer is considering to develop is a high-end item, the manufacturer cannot benefit from information sharing from the retailer. (ii) On the other hand, when the new product that the manufacturer is considering to develop is a low-end item and when it is sufficiently costly to increase the success rate of the new product, information sharing helps both the manufacturer and the retailer. (iii) However, when it is not very costly to increase the success rate for the new product, information sharing from the retailer to the manufacturer reduces the retailer’s profit more than it increases the manufacturer’s profit. In the absence of information sharing, the manufacturer in a non-integrated channel will still develop the new product and make an efficient investment that maximizes channel profits. In this research, the authors focus on the R&D/Marketing interface within organizations and investigate the different dimensions and levels of this relationship. Typically, the Marketing team is very concerned with new product launch deadlines and in many instances the expressed view is that meeting deadlines receive priority and resources at the expense of potential breakthrough concepts that require extended deliberation. Building on Kotler, Walcott and Chandrasekhar (2008), we explore this subject empirically by examining the quality of the working relationship between R&D and Marketing in several companies spanning several industries and countries. The data come from surveys of marketers, R&D personnel, engineers and product managers at senior and middle management levels via an online assessment questionnaire. Select executives were also interviewed. In this paper, we examine the perceived levels of contribution of Marketing to NPD as well as the type of relationship between R&D and Marketing. We account for company size, geographic region and industry variability while addressing the following questions: • What are the reported levels of engagement between Marketing and R&D and how do they vary by company size and industry? • What are the perceived contributions of Marketing to the stages of the new product development process? • What steps can a company take to improve the working relationship between R&D and Marketing? We report on Marketing’s contribution to NPD in four areas (ideation, early stage concept refinement, development, and launch/post-launch) and the types of relationship between the two functions. 2 - Distributor Support in New Product Launch Wei Guan, Linköping University, IEI 581 83, Linköping, Sweden, [email protected], Jakob Rehme Many companies usually commit to develop and produce new products and delegate the selling aspects to distributors. It is recognized that distributors play an important role in bringing products to customers and that distributors have their own business objectives. Despite its importance, research regarding distributor support in the new product launch is few in number. This paper seeks to fill the gaps by empirically analyzing how distributors support is obtained for new products. The first purpose of this paper is to investigate how distributors’ selling effort, advertising and technical support are organized for new products and how these differ from more established items. The second objective is to examine the consequent demands and requirements posed to suppliers. The third objective is to explore insights into potential areas for future product or service innovation. This study takes an exploratory, case study research approach. Multiple cases regarding innovative products or solutions recently came out to the British building material market were studied. Data is collected through in depth interviews using an open-ended format. Findings Distributors are starting to understand that they can not make successful new product launch at the store level without a well functioning supplier operations. New product sales of distributors is more and more relying on suppliers in aspects of logistics, commercial and technology. Marketing support regarding pricing, promotion and merchandising from suppliers is extremely important for distributors in selling new products. Previous new product launch studies focus primarily on strategic and tactical decisions made by suppliers, this paper emphasize the importance of distributor support. 3 - Embedding Product Development Accelerations in Environmental Uncertainty Tao Wu, Professor of Marketing, California State University, Long Beach, 1250 Bellflower Boulevard, Long Beach, CA, 90840, United States of America, [email protected] Past new product development (NPD) research suggested that in uncertain environments, contingency planning, concurrent development, early customer feedback, and iterations improve firms’ cycle time and even product performance in general. However, a review of organizational learning literature reveals contradictory demands that these practices place on organizations’ information processing: Contingency planning and concurrent development require firms to have a relative clear understanding of, and be able to accurately predict, the future of the market, while early customer feedbacks and iterations are more effective when firms are vague in their perceptions of the market behaviors. In addition, these NPD practices have been examined mostly with the concept of environmental uncertainty as onedimensional, which precludes contextual understandings of the findings. This research is set out to further explore the impact of each aforementioned practice on NPD performance under different types of environmental uncertainty. It differentiates environmental uncertainty based on the degree to which companies understand and/or can accurately predict the future of the market. Through empirical examination of 164 product development firms, it proposes a typology of combining these practices in uncertain environments to maximize NPD performance. 3 - Long-term Asymmetric Buyer-seller Relationship: An Empirical Study Yuying Shi, PhD Student, University of Florida, Warrington College of Business, 211 Bry, Gainesville, FL, 32606, United States of America, [email protected], Qiong Wang, Bart Weitz Long-term relationships between buyers and suppliers have obtained substantial attention in the past decades. Although much as been written about the benefits of such relationships to firms, little is known about the nature of such relationships. In this study, we empirically assess one type of long-term relationships composed of suppliers and their major buyers, in particular those retailers who purchase at least 10 percent of the total sales of these suppliers. Using cross-sectional and longitudinal data over a 20-year period, we find that suppliers may benefit from having major retailers in terms of their financial performance despite of the power asymmetry. Specifically, the retailers’ relative sizes, R&D and advertising investments, and the lengths of relationships with suppliers may moderate the influence of major retailers onto suppliers. Our study sheds lights on why and how a firm develops long-term relationships with major buyers. We conclude this study with a discussion of implications and future research, specifically the demonstrated importance of investigating asymmetric buyer-seller relationships. 4 - Consumer Opinion of Product Design Dimensions Wooseong Kang, Assistant Professor, North Carolina State University, 2332 Nelson Hall, Raleigh, NC, United States of America, [email protected], Janell Townsend, Mitzi Montoya Product design is a critical component of brand strategy. We develop a conceptual framework illustrating how two critical design factors – form and function – impact consumer opinion, and delineate brand specific effects. We further identify nonmonotonic effects, as well as the interaction effects of the individual factors among the dimensions. A longitudinal model based on objective measures of form and function is tested with a data set we developed for models available in the U.S. automotive market from 1999 – 2007. The data include information on 16 firms, 32 brands and 137 products. The results suggest form and function play a significant role in forming consumer opinions, but have diminishing returns. However, the results suggest brands may benefit from an extension of certain design factors. Trade-offs between form factors generally moderate each other, but that does not appear to be so for function factors. The relationships between factors of form and function are multifarious and complex, but generally do not support the notion of trade-offs between form and function as being a cause for changes in opinion. 4 - Inventories, Incentives, and Channel Structure Sudheer Gupta, Associate Professor, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada, [email protected] A well-known result in channels literature states that competing manufacturers may have an incentive to decentralize downstream retailing to buffer from intense price competition when products are substitutable. We explore this effect in a dynamic setting where retailers can carry inventories forward. We show that retailers will always carry inventories as a credible source of competition for the manufacturers, even in the absence of traditional reasons for inventories. The presence of strategic inventories in dynamic intra-channel relations counters the benefits of strategic decentralization as a buffer against competition. We establish equilibrium outcomes for competing manufacturers selling differentiated products over two periods, and show how degree of product differentiation and the ease of carrying inventories forward affect prices, profits and equilibrium channel structure. ■ FA06 Legends Ballroom VII Channels I: General Contributed Session Chair: Sudheer Gupta, Associate Professor, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada, [email protected] 1 - Information Sharing and New Product Development in a Non-integrated Distribution Channel Shan-Yu Chou, National Taiwan University, 1 Sec. 4 Roosevelt Road, Taipei, Taiwan - ROC, [email protected] A manufacturer, when developing new products, often faces high demand uncertainty. A retailer, closer to consumers, may possess superior information relative 38 MARKETING SCIENCE CONFERENCE – 2011 ■ FA07 FA09 3 - Changing the Tone: The Dynamics of Political Advertising over the Election Cycle Ron Shachar, Arison School of Business, The Interdisciplinary Center, Herzliya, Israel, [email protected], Paul Ellickson, Mitch Lovett Founders I Panel Session: Cases? Projects? Simulations? Problem Sets? What’s the Best Way to Teach Marketing Science? This paper empirically investigates the dynamic incentives determining when and how much to advertise, along with the optimal choice of tone. We focus on political advertising in closely contested U.S. congressional races. In particular, we investigate the tendency for close races to become more negative as election day gets closer and establish that candidates generally match on the tone of their advertising with a tendency to match more on the negative than positive. We offer an explanation for the tendency to go negative that suggests negativity in close races is not a foregone conclusion. We examine empirically the dynamic implications associated with changing the volume and tone of advertising over an election cycle. To do so, we propose and estimate a strategic model of sequential decision-making in which candidates react to the arrival of new information as well as the strategic actions of their rivals in forming an optimal advertising policy. We estimate the structural parameters of the model using a full solution approach, characterized by a system of sequential decision problems with mixed controls. We use the structural estimates from the model to investigate the degree to which government policy can impact the equilibrium tone of the race, thereby altering the tone of political discourse. Cluster: Special Sessions Invited Session Chair: Gary Lilien, Pennsylvania State University, 484 Business Building, University Park, PA, United States of America, [email protected] Co-Chair: Arvind Rangaswamy, Pennsylvania State University, University Park, PA, United States of America, [email protected] 1 - Cases? Projects? Simulations? Problem Sets? What’s the Best Way to Teach Marketing Science? Moderator: Arnaud De Bruyn, ESSEC Business School, Cergy, 95000, France, [email protected], Panelists: Dominique Hanssens, Ujwal Kayande, Charlotte Mason Many of us have struggled to find the best way to teach our specialty, marketing science. Should we use traditional lectures and problem sets? What about cases? What about simulations? What about projects? And what role should generalized or specialized software play? In this special session, each panelist will provide an brief overview of what he or she has been doing in the MBA/EMBA classroom, sharing what has worked, what has not and what developments he or she sees that the marketing science community should be aware of. We will then open the discussion up to the audience to share their experiences and ask questions of the panelists. 4 - A Dynamic Model of Thirst and Beverage Consumption Ahmed Khwaja, Assistant Professor, Yale University, New Haven, CT, United States of America, [email protected], K. Sudhir, Guofang Huang The physical need to consume beverages due to thirst occurs several times a day. Apart from satisfying the physical thirst need, beverages also satisfy a variety of other short term needs such as “quick pickup,” “refreshing fun” etc. Given the frequency with which beverages are consumed, they also have significant long-term health consequences. The goal of the paper is to build and estimate an “as-if” model of thirst and beverage consumption that helps understand the consumer tradeoff between the short-run needs and their long-term consequences. Researchers rarely have consumption data, hence they make inferences about consumer’s utility from consumption thorough purchase data. Here we exploit a rare dataset with information not only what a consumer consumed in every two hour period, but also the context, moods and stated objectives captured in real time over a period of two weeks. This allows us to understand how consumers trade-off long-term and shortterm needs in routine consumption. Our modeling framework can be useful in studying a variety of health-related issues such as obesity, cancer etc., which are significantly affected by routine consumer short-run choices of food, beverages and cigarettes etc. ■ FA08 Founders II The Long Run Consequences of Short Run Decisions I Cluster: Special Sessions Invited Session Chair: K. Sudhir, Yale School of Management, Yale School of Management, New Haven, CT, United States of America, [email protected] Chair: Ahmed Khwaja, Assistant Professor, Yale University, New Haven, CT, United States of America, [email protected] 1 - Taste and Health: Balancing and Highlighting in Choices Across Complementary Categories Hai Che, University of Southern California, Los Angeles, CA, United States of America, [email protected], Botao Yang, K. Sudhir ■ FA09 Behavioral research using laboratory experiments suggests that some consumers “balance” healthy versus tasty choices across complementary categories when having a meal, while others “highlight” one attribute such as taste or health across both categories. To-date, there has been little empirical research using field data. We therefore examine consumers’ choices of tasty versus healthy foods across complementary food categories in grocery stores to identify evidence of such behavior in a field setting. Given only purchase data, we impute household inventory based on past purchases and study how the inventory of “tasty” or “healthy” foods in one category affects the choice of tasty or healthy foods in a complementary category. We find evidence of both highlighting and balancing consumer segments. Thus far the literature on cross-category choices has found that customers tend to have positive correlation in preferences for attributes across categories (i.e., a price or feature sensitive customer tends to be price or feature sensitive across categories). This is because existing models do not allow for heterogeneity in correlations for preferences across categories. Our modeling approach allows for heterogeneity in correlations for attribute preferences; thus allowing us to uncover evidence of balancing between attributes across categories using field data. Founders III Retailing II: General Contributed Session Chair: Manish Gangwar, Assistant Professor, Indian School of Business, ISB Campus, AC2-L1-2113, Gachibowli, Hyderabad, AP, 500032, India, [email protected] 1 - The Impact of Retailers’ Corporate Social Responsibility on Price Fairness Perceptions and Loyalty Kusum Ailawadi, Charles Jordan 1911 TU’12 Professor of Marketing, Tuck School at Dartmouth, Dartmouth College, 100 Tuck Hall, Hanover, NH, 03755, United States of America, [email protected], Jackie Luan, Scott Neslin, Gail Taylor We study the effect of four key dimensions of Corporate Social Responsibility in the grocery retail industry on consumers’ perceptions of the fairness of retailers’ prices and their attitudinal and behavioral loyalty towards those retailers. Our model controls for other key drivers of retail store patronage and for heterogeneity in CSR response, and we estimate it using data on consumers’ shopping behavior and their perceptions of the retailers they patronize. We model price fairness as a (partial) mediator of the CSR-loyalty relationship. CSR may affect price fairness directly whereby a high price is perceived as more acceptable if part of the value from the consumer-firm exchange is seen as benefiting society. It may also affect price fairness indirectly through cost and profit judgments whereby consumers associate CSR with higher costs or/and lower profit for the retailer. We estimate these direct and indirect effects and quantify the net impact of retailers’ CSR on consumer loyalty. Among the four CSR dimensions in our study, we find that environmental friendliness has the weakest effect on both price fairness and behavioral loyalty. Local product sourcing, from which consumers personally benefit in their exchange with the retailer, has the strongest direct effect on behavioral loyalty, though it delivers no added boost through price fairness. The remaining two CSR dimensions, fair treatment of employees and community support, have intermediate direct effects on behavioral loyalty and the indirect effects through price fairness account for approximately 10% of the total effect. 2 - Information Acquisition and Ex-ante Moral Hazard Jian Ni, Johns Hopkins University, Baltimore, MD, United States of America, [email protected], Nitin Mehta In contractual markets like insurance or credit, it is the firm’s interest that the consumer engages frequently in information acquisition, but the consumer may not have the incentive to do so. We use the health insurance market as the example, more particularly chronic illnesses (e.g., prostate cancer). For such illnesses, the cost of treatments increases significantly with the severity of the illness. Thus it is in the insurer’s interest that consumers get regular diagnostic checkups before and after they are diagnosed with the illness (regular checkups from early detection of the illness result in low treatment costs in the future; and regular checkups after the diagnosis of the illness result in the consumer to consume the ‘right’ treatment that is specific to the stage of his illness, which consequently leads to lower cost of treatments in the future). However, consumers may not have the incentive to go for regular checkups if their monetary (out of pocket expenses for getting checkups) or non monetary costs (time and effort) are high. The questions that we address are: (i) what is the impact of consumer’ information acquisition behavior (how regularly they go for checkups) on the firm’s (insurance company’s) long term profits? (ii) What are the magnitudes of monetary and non-monetary costs of information acquisition, and how do they vary over types of contracts (insurance plans) and consumer demographics? (iii) What strategies can firms employ in terms of communication tactics and contract design to reduce consumers’ monetary and non-monetary costs, so that they engage in information acquisition on a more frequent basis? 39 FA10 MARKETING SCIENCE CONFERENCE – 2011 2 - To Kill Two Birds with One Long Queue Wenqing Zhang, PhD Candidate, McGill University, 1001 Sherbrooke West, Montreal, QC, H3A1G5, Canada, [email protected], Chun (Martin) Qiu lead to breakthrough innovations within conjoint analysis. For the empirical setting 140 customers from target, analogue and foreign markets were personally interviewed using conjoint measurement and lead user identification techniques. Based on a former approach by Lüthje, Herstatt and von Hippel in 2005 with the lead user method, the new approach completes the conjoint measurement by a two parted survey covering customers use experience along with their technical skills and providing background information of the own developed innovations (Lüthje et al. 2005). The empirical study showed thirty general new improvements, among them sixteen lead user innovations were found. In fact non-lead users significantly prefer more innovations as opposed to lead users. So the extension leads indeed to different preferred attribute-levels and increases the performance for breakthrough innovations. We investigate two strategic roles of customer queue management for retailers during holiday season shopping. We focus on the exit queue that customers must take before finishing shopping tasks. When retailers can influence the length of the queue through either operational or promotional decisions, it is not necessarily beneficial for retailers to invest in reducing the length of the queue. Rather, we show that, first, a long queue helps retailers effectively segment customers, and target those who are price-sensitive but time-insensitive. Second, a long queue prompts consumers who decide to join the queue to purchase more, which increases the overall profitability. 2 - Accountability of Biological-response Measures for Advertising Effects Akihiro Inoue, Professor, Keio Business School, 4-1-1 Hiyoshi, Kohoku-ku, Yokohama, 223-8526, Japan, [email protected] 3 - Shopper Loyalty to Whom? Chain and Outlet Loyalty in a Dynamic Retail Environment Arjen van Lin, Doctoral Student, Tilburg University, P.O. Box 90153, Tilburg, 5000 LE, Netherlands, [email protected], Els Gijsbrechts The effects of ads have been investigated in several ways. Most of them are the measures based upon social and cognitive psychology, such as recognition, memory, purchase-intention, change in attitude, and so on. However, these measures account for the effects of ads in a limited degree. Recently, some researchers complementarily have adopted the biological-responses in evaluating the ads effects. There exist several ways to measure biological-responses. We can measure brain-blood-flow via fMRI (Functional Magnetic Resonance Imaging), a most expensive technique. Also, we can quantify brain wave by using EEG (Electroencephalography). A well-known measure is GSR (Galvanic Skin Response) where we can use several machines to calculate the GSRs. Another well-known measure is eye-movement with the usage of, usually, eye-cameras. We have run the experiment with a new device that allows us to measure GSR and eye-movement quite easily (Fukushima, Inoue, and Niwa, 2010; Oyama, Miao, Hirohashi, and Mizuno, 2010). The data are collected via a marketing research company located in Tokyo (Japan) with the sample size of 200 in 2009. We have confirmed some significant correlations between biologicalresponses (Lyapunov index and fractal dimension) based upon the logit models where the dependent variables are the ad effects and independent variables are both the traditional and biological-response measures. However, the traditional psychosocial measures account for the ad effects more than these biological-responses at this moment. Besides, we found that the surrounding environments for ad-effect experiments are crucial for the verification. We will show the detail in the presentation. Previous literature has convincingly shown that consumers have a tendency to revisit the same store. While such store loyalty has typically been interpreted as adherence to a retail chain, we suggest that some portion may actually be outlet (or location) loyalty, independent of its affiliation with the chain. This distinction between chain and outlet loyalty is highly relevant in the context of takeovers or market entries (e.g. the competitive acquisition of Safeway (UK) by Morrisons early 2004), where retailers have to anticipate the performance implications of acquiring competing chain outlets. While identification in past research has been limited by the scanner panel data typically available, we propose an approach to disentangle the two forms of loyalty, in a dynamic retail setting involving store closures and competitive acquisitions. Moreover, we allow the strength of outlet loyalty following an acquisition to be moderated by the length of the transition period (i.e. how long the outlet is closed for refurbishment), and by the format of the acquiring chain (sameformat traditional supermarket, or hard-discounter). We estimate our model on a unique scanner panel data set covering up to ± 250 Dutch local markets, characterized by multiple retail takeovers. Preliminary results confirm that, after an acquisition, consumers show a sense of outlet loyalty, irrespective of its (change in) banner and marketing mix. Furthermore, we find that transition length and chainformat dissimilarity is of significant impact, and may lower the consumer’s probability of return. Managerial implications are discussed. 4 - Examining Store Attractiveness as a Category-specific Trait Manish Gangwar, Assistant Professor, Indian School of Business, ISB Campus, AC2-L1-2113, Gachibowli, Hyderabad, AP, 500032, India, [email protected], Qin Zhang, P. B. Seetharaman 3 - Using Augmented Best-worst Scaling to Test Schwartz’ Theory of Values Julie Lee, Winthrop Professor, University of Western Australia, 35 Stirling Highway, Crawley, 6014, Australia, [email protected], Jordan Louviere, Geoff Soutar The literature on grocery store loyalty views a consumer as, possessing store loyalty toward a particular store for her or his overall shopping needs. In this study, we examine store attractiveness as a category-specific trait, i.e., a consumer could be loyal to Store A in category 1, but loyal to Store B in category 2. We call this storecategory loyalty (SCL). We enumerate several key influencing variables, which are related to product assortments and prices of categories at the stores and their expected effects. We use an in-home scanning panel dataset of more than1000 households in more than 150 grocery categories across 16 retail chains over a 53week period to test our hypotheses. We use Hierarchical Bayes estimation technique to control for household and category level unobserved heterogeneity. Our, results show that a variety of category, store and consumer characteristics affect SCL in line with our predictions. Further, we detail how the findings can help retailers make strategic decisions. We conduct managerial exercises to further illustrate the managerial implications of our study to retailers, for example, by deriving revenue consequences from changing some of the marketing levers, i.e., variables related to product assortments and prices, which are in their control. Best-worst scaling (BWS) is now widely used by academics and practitioners to measure subjective/latent quantities. On the plus side, BSW is based on a theory of how humans make best and worst choices, it uses choice-based balanced incomplete block experimental designs, assumes only ordinality of responses, is consistent with random utility theory and does not need various “adjustments/standardizations” to yield comparable measures for different people. However, a potential disadvantage is that BWS yields more information about extreme options and less about interior items that are less frequently chosen. The purpose of this paper is to compare twochoice BWS with a new variant that relies on additional best and worst choices in each choice set; obtaining more information matters if some items have similar subjective values on the underlying scale. We compare traditional two choice BWS (BW) to a four choice (BWaug) variant in the context of testing which best retrieves Schwartz’ hypothesized Values circumplex structure. The comparison involves a sample of respondents evaluating 11 values (items) that are assigned to 11 subsets of 6 items based on a balanced incomplete block design. The BW response task yields three bits of order information (e.g., A>B|C|D|E >F), whereas BWaug yields five bits (e.g., A>B|C>D>E >F) regarding the ranking of the six items in each subset. We find that BWaug requires little extra effort from respondents, while producing significant gains in the variability and correlations between items. It also better recovers Schwartz’ hypothesized circumplex structure. ■ FA10 Founders IV Measurement Issues Contributed Session ■ FA11 Chair: Julie Lee, Winthrop Professor, University of Western Australia, 35 Stirling Highway, Crawley, 6014, Australia, [email protected] 1 - Customer Innovation: A Combined Lead User and Conjoint Analysis Approach Alexander Sänn, Brandenburg University of Technology Cottbus, Erich-Weinert Strasse 1, Cottbus, 03046, Germany, [email protected], Daniel Baier Champions Center I Using Endorsers in Advertising Contributed Session Chair: Debasis Pradhan, Assistant Professor (Marketing), School of Business & Human Resources, XLRI Jamshedpur, Circuit House Area (East), Jamshedpur, Jharkhand State, 831035, India, [email protected] 1 - Alienating the Mainstream: Does the Inclusion of Gay and Lesbian Imagery Diminish Brand Perception? Anthony Perez, St. Edward’s University, 3001 South Congress Ave, Austin, TX, 78704, United States of America, [email protected], Helene Caudill Today lead user method (von Hippel 1986) and conjoint analysis (Green and Srinivasan 1990) are widely applied in new product development using customers’ preferences to achieve optimal product design. The main goal of this paper is to measure the difference between lead user and non-lead user preferences and analyze whether it can be used in new product design for generating breakthrough innovations via conjoint measurement. As a consequence conjoint analysis and the lead user method are combined into a new approach increasing the performance of both methods. The new approach extends traditional conjoint analysis by a lead user identification technique, which identifies leading customers by their use experience, technical skills and own idea development (Spann et al. 2009). The well-known field of mountain biking is used as an application area. We assume that the separation can The buying power of the lesbian, gay, bisexual, and transgender (LGBT) market is approximately $850 billion per year, the largest per capita of any minority group. Thus, this market is very appealing to companies and has become a more open focus of their marketing strategies. For example, Progressive Insurance recently included a gay couple in their television advertising. Although this type of marketing is 40 MARKETING SCIENCE CONFERENCE – 2011 FA12 consumer product choices. Real estate agents routinely suggest that sellers make sure wall and floor colors are neutral in order to appeal to potential buyers, re-painting or re-carpeting if necessary. The availability of specific product colors has empirically been found to affect category sales as well, where the absence of certain previously stocked colors disproportionately reduced category sales volume [1]. The goal of this work is to develop an approach to include product color in a utility function in a way that allows managers to infer color preferences outside the discrete set of colors which were measured. Building on the continuity of color spectrum and on prior models in product forms and colors [2,3], we develop a model that allows retailers and manufacturers to use responses color swatches to infer a superior set of colors to produce and to stock. The results of this research are applicable to both manufacturers (new product decisions) and to retailers (decisions about which products to stock). [1] Boatwright, Peter, Sharad Borle, and Kirthi Kalyanam (2007), “Deconstructing Each Item’s Category Contribution,” Marketing Science, 26, 3 (MayJune), 1-15. [2] Orsborn, S., Cagan, J. & Boatwright, P., 2009, “Quantifying Aesthetic Form Preference in a Utility Function”, ASME Journal of Mechanical Design, 131(6), 061001. [3] Turner, H., Orsborn, S. and Grantham, K., 2009, “Quantifying Product Color Preference in a Utility Function,” Proceedings of 2009 ASEM, October 14-17, Springfield, MO, USA. becoming more acceptable, the dilemma companies face is how to target this unique market without alienating the mainstream and possibly diminishing the perception of their brand. For the academic community, this is a new and emerging area of interest. While studies focusing on gay and lesbian imagery in advertisements first appeared 15 years ago, these early studies used “fake print advertisements” and only surveyed a small heterosexual population. Our study adds to the existing research by using “real print advertisements” in addition to adding responses from the LGBT community. Over 125 respondents to a survey instrument indicate the inclusion of gay and lesbian imagery would likely alienate straight men more so than any other demographic group. Additionally, the results suggest that the inclusion of gay and lesbian imagery may be more appropriate to use for services rather than products. Our results also indicate that marketers should consider using subtle gay and lesbian imagery and symbols (such as a rainbow), as opposed to explicit gay and lesbian imagery (such as obvious gay and lesbian interaction), in order to appeal to both the mainstream and LGBT markets. 2 - Consumer Perceptions of Corporate Gay-friendly Activities: The Role of Gender and Gay Identity Gillian Oakenfull, Associate Professor of Marketing, Miami University, 3038 Farmer School of Business, Oxford, OH, 45056, United States of America, [email protected] 2 - Product Aesthetics is Must or Plus? Trade-offs Between Product Aesthetic and Functional Attributes Jesheng Huang, Assistant Professor, Tamkang University, 151, Yingzhuan Rd., Danshui Dist., New Taipei Cit, Taiwan - ROC, [email protected], Chia Ming Hu Recent corporate recognition of the attractiveness of the $641 billion gay consumer market has lead to a dramatic increase in both gay-oriented marketing activities and corporate policies. Corporate spending within the gay community totaled over $231 million in 2006 (Wilke, 2007). The most commonly employed gay-friendly activities include the provision of domestic partner benefits, corporate support of gay causes, company identification as gay-friendly in marketing communications, and advertising in both gay and mainstream media (Peñaloza, 1996; Oakenfull, 2000; Tuten, 2004, 2006.) While academic research has offered some theoretical discussions of the potential benefits of gay-friendly activities, little is known about factors that may influence gay consumers’ perceptions of the gay-friendliness of different types of gayoriented activities. This research explores the influence of gay identity and gender on gay consumers’ perceptions of the gay-friendliness of gay-oriented corporate activities, both internal and external to the firm. The findings of this research suggest that it is important that companies avoid a treatment of gay consumers as a group with monolithic preferences and perceptions. An individual’s level of gay identity and gender have a significant impact on perceptions of the gay-friendliness of various corporate activities. An individual with a strong gay identity will tend to perceive all gay-oriented activities that are external to the company as more gay-friendly than will an individual with a weak gay identity. Lesbians’ more socio-political identity will lead them to consider activities that are perceived as playing a strong role in legitimizing the gay and lesbian movement in mainstream society as more gayfriendly than will gay males. Previous research has shown that consumer preferences have both hedonic and utilitarian dimensions. The aesthetic aspect of a product is normally taken as a source of hedonic consumption. What we are mostly interested in is whether the product aesthetic attribute will dominate consumer choice rather than product functional attributes? And how a preference of product aesthetic attribute will evoke the critical justification effect on hedonic consumption? Based on the research on loss aversion that demonstrates an asymmetry in evaluations depending on the direction of the proposed trade, the authors predict that differential loss aversion for product attributes may be a function of attribute importance for consumer. Therefore, this article conducted some trade-off decision tasks under the condition of given same price for each pair of comparable aesthetic-functional attributes, in order to examine which of both opposite attributes consumer will forfeiture under the price and other attributes equals. The authors chose mp3 players as the experiment target, and design four pairs of hypothetical products, focusing on salient aesthetic versus salient functional attribute respectively, namely push button with planer vs. convex style, USB connector is hided vs. exposed, with speaker vs. non, and with touch screen vs. non. In addition, due to different aesthetic acumens among consumers, the authors clustered respondents into several groups based on the measurement of the centrality of visual product aesthetic (CVPA). The findings indicate that this aesthetic-functional trade-offs design can derive the asymmetry in preferences of relative loss aversion between product aesthetic and functional attributes, which has significant heterogeneities among consumers with different CVPA. 3 - Attitude Towards Celebrity Endorsement and Brand Loyalty: Mediating Effect of Celebrity Credibility Debasis Pradhan, Assistant Professor (Marketing), School of Business & Human Resources, XLRI Jamshedpur, Circuit House Area (East), Jamshedpur, Jharkhand State, 831035, India, [email protected], Duraipandian Israel 3 - Shaping Product Perceptions Tanuka Ghoshal, Assistant Professor, Indian School of Business, Indian School of Business, AC-2 #2114, Gachibowli, Hyderabad, AP, 500032, India, [email protected], Peter Boatwright Globally, the dependence of marketers on celebrity endorsement is on the rise. Indian advertising industry has seen a huge spurt of celebrity endorsement in the last decade as 60 % of Indian advertisements use celebrity endorsement. A celebrity can play multiple roles such as spokesperson, endorser, testimonial, and actor which may not be mutually exclusive. Traditionally, in extant literature three dimensions of source credibility has been taken such as attractiveness, expertise, and trustworthiness. The present study has taken an additional dimension called popularity to make it complete. While some studies have drawn inferences about the relationships between brand attitude and brand loyalty and some others about source credibility and brand attitude, the mediation effect of credibility of celebrity in the relationship between consumer’s attitude towards celebrity endorsement and brand loyalty has not received any attention. This paper makes a modest attempt to study this mediating role. The Barron & Kenny (1986) process has been used to measure the possible partial or full (if any) mediation effect of celebrity credibility. The relative impact of dimensions of celebrity endorsements, along with the attitude of consumers towards celebrity endorsements on the brand loyalty has been studied. Role of gender has also been one of the areas of our investigation which throws insights both for academician and practitioners in advertising field. In this research we investigate the role of (aesthetically appealing) product shape in influencing product preference and perceptions. We firstly investigate whether individuals have preferences for certain kinds of shapes (curvilinear versus angular) across a range of products. We then study whether superior functional and/or hedonic benefits are attributed to aesthetically preferred shapes, for products where shape should not be associated with functionality. We finally investigate how product shape influences perceptions, and explore whether the impact occurs at a conscious or non-conscious level. We find that for common hedonic and utilitarian products (car, teapot, mp3 player, external hard drive, sports bottle etc.) the curvilinear form is preferred over the angular form by a vast majority. “Visual pleasure” is the most common spontaneous reason provided for the preference. We have preliminary evidence that visually appealing shapes may be associated with perceived superior functional benefits (example, a curvilinear external hard drive is perceived to have more capacity than an angular shaped one), and/or perceived superior hedonic benefits (example, a curvilinear teapot perceived to be “better suited for party”). We additionally find that shape can subtly impact at a non-conscious level, wherein certain functional product attributes are accorded higher (lower) importance when they are associated with a more (less) aesthetically shaped product. The importance of product design as a tool for differentiation has been oft- reiterated, however, there is little to no work isolating and examining the specific role of product shape in impacting perceived functionality, underlining the contribution of this research. ■ FA12 Champions Center II Aesthetics Contributed Session Chair: Elea McDonnell Feit, Research Director, Wharton Customer Analytics Initiative, 256 S. 37th Street, Philadelphia, PA, 19104, United States of America, [email protected] 1 - Inferring Color Preferences: A Utility Model Approach Seth Orsborn, Assistant Professor, Bucknell University, 121 Taylor Hall, Lewisburg, PA, 17837, United States of America, [email protected], Peter Boatwright, Jonathan Cagan Although many quantitative methods have been developed for observable and discretely measurable product characteristics, there is much room to develop quantitative approaches for aesthetic preferences. Color, for example, affects 41 FA13 MARKETING SCIENCE CONFERENCE – 2011 4 - Modeling the Impact of Visual Design in Consumer Choice Model Elea McDonnell Feit, Research Director, Wharton Customer Analytics Initiative, 256 S. 37th Street, Philadelphia, PA, 19104, United States of America, [email protected], Jeffrey Dotson, Mark Beltramo, Randall Smith customer. A firm’s capital structure indirectly influences customer satisfaction through the mediation of marketing assets. Specifically, the authors argue that the potential destruction of value of marketing assets in case of financial distress can lead firms with high financial leverage to invest less in marketing assets, which in turn can negatively influence customer satisfaction. Using a uniquely compiled dataset of 73 firms tracked for a period of eight years (2000-2007), the authors test and find support for these hypotheses. Product designers often want to understand the affect of visual design on product choice, but it hasn’t been clear how to incorporate such complex, multi-dimensional attributes into consumer choice models. Existing approaches either ignore important information about visual design and its effect on vehicle choice, particularly the fact that individuals differ in what they find appealing, or subject respondents to long and difficult tasks where they rate many alternative visual designs. But when the affect of design at the individual-level is ignored, the observed individual behavior is likely to violate the IIA property, a feature of most choice models—including hierarchical choice models. We propose a new model that controls for the fact that customers who prefer one shape are likely to prefer similar-shapes. The model incorporates a an error covariance between alternatives that is parameterized as a function of the visual similarity of the alternatives. Using choice-based conjoint data for crossover utility vehicles, we demonstrate that the proposed model makes better predictions about which products will gain or lose share when a new vehicle enters the market. We will discuss how such methods can be applied to other complex attributes such as speed of a computer, sound quality of a stereo or taste of a snack food. 4 - The Use of Advertising for Capital Market Benefits Michal Herzenstein, Assistant Professor of Marketing, University of Delaware, 319 Lerner Hall, Newark, DE, 19716, United States of America, [email protected], Tzachi Zach, Dan Horsky Do firms engage in marketing activities for reasons other than to increase sales and build up their brand? A natural context to examine this question is the raising of funds through initial public offerings (IPOs) and seasoned equity offerings (SEOs). Are the marketing activities, in particular advertising expenditures, of firms engaged in these offerings different before and after such offerings? Past literature in both marketing and finance domains is equivocal, with some papers showing that companies engage in earning management and reduce advertising expenses to appear more profitable (Mizik and Jackobson 2007); and others showing that advertising budgets are increased in order to reduce the asymmetric information between firms and investors (Chemmanur and Yan 2009). Yearly advertising spending (from COMPUSTAT) was employed in the above papers. Using TNS data which tracks weekly advertising expenses, Joshi and Hanssens (2010) show that ad spending has a positive impact on firms’ market capitalization. But their data is limited to two industries and nine firms. Similar results were shown by Osinga et al. (2011) in the pharmaceutical industry. In the present research we utilize a much larger TNS dataset (all firms engaged in IPO or SEO in 1995-2009, over 2000 firms). By employing weekly data (rather than yearly data as prior research) we are able to shed light on firms’ ad spending patterns and purposes before securing capital. ■ FA13 Champions Center III Marketing Finance Interface I Contributed Session Chair: Michal Herzenstein, Assistant Professor of Marketing, University of Delaware, 319 Lerner Hall, Newark, DE, 19716, United States of America, [email protected] 1 - Going Public: How Stock Market Participation Changes Firm Product Innovation Behavior Christine Moorman, T. Austin Finch, Sr. Professor of Business Administration, Duke University, Fuqua School of Business, 100 Fuqua Way, Durham, NC, 27708, United States of America, [email protected], Simone Wies 5 - A Simple Metric that Really Matters: Including the Share of Customer Business in Financial Reports Christian Schulze, Goethe-University Frankfurt, Grueneburgplatz 1, Frankfurt, 60323, Germany, [email protected], Manuel Bermes, Bernd Skiera Firm’s profit either comes from customer business or non-customer business. Customer business refers to products being sold to customers or services mandated by customers, whereas non-customer business comprises financial activities on the firm’s own authority and for its own ac-count. Knowledge about their share in firm’s profit is important because they rely on different intangible assets. For example, brands are important for customer business, while smart traders are necessary to generate high returns for excess liquidity. Yet, little is known about the size of these two kinds of profits. In our first empirical study, we therefore analyze the degree of transparency about non-customer business for the world’s leading firms across various industries. Surprisingly, our evaluation of their public financial reports reveals that a lack of stringent reporting requirements does not allow shareholders to determine the importance of customer vs. non-customer business for most firms. In our second empirical study, we then focus on the banking industry to analyze the share of profit for customer business and non-customer business of more than 200 of the world’s largest banks. Our findings are highly unintuitive: The share of customer business at bank’s profit on average exceeds 100% - thus, non-customer business on average destroys value. Both, the surprising findings for banks and the lack of transparency in other industries lead us to propose more transparency in financial reporting regarding customer business and the inclusion of customer business’s share of profit as an additional metric. Innovation is a marketing action of high strategic relevance and a source of competitive advantage. Abundant literature documents the positive effects of product introductions on performance and capital market measures. These findings reveal that investors do interpret innovation actions, and therefore suggest that managers incorporate investor behavior in their decision-making. Apart from a few notable exceptions, acknowledging the stock market as antecedent of innovative behavior has so far been very limited in research in marketing. We address this possible reverse causality and consider how firms change their innovative behavior, i.e. level and timing of innovation, following an initial public offering (IPO). We examine our predictions in a longitudinal data set of 263 firms that change their status from private to public during the sample period. This quasi-experimental set-up is complemented by a control panel of 285 firms that remain private throughout the sample period from 1995 until 2007. We obtain a difference-in-differences estimate of the average capital market effect that reveals an increase in the number of introductions per quarter after the firm’s IPO. We also observe a change in the timing of product introductions, with firms introducing new products closer to the end of the quarter after they go public. We contribute to literature by showing a role of the stock market in affecting firm behavior. Likewise, we employ a unique developmental perspective on what the process of “going public” means for the firm’s strategic choices. Finally, by using a quasi-experimental approach with a difference-indifferences estimator, we are able to rule out methodological concerns with regard to unobserved omitted factors. ■ FA14 Champions Center VI Pricing and Consumer Behavior 2 - Media Expenditure Effectiveness and Firm Performance Lopo Rego, Associate Professor, University of Iowa, Tipple College of Business, 108 PBB W274, Iowa City, IA, 52242-1994, United States of America, [email protected], Lisa Schöler, Bernd Skiera Contributed Session Chair: Marcus Kunter, RWTH Aachen, Templergraben 64, Aachen, 52064, Germany, [email protected] 1 - Produce Line Obfuscation Lin Liu, University of Southern California, Marshall School of Business, Los Angeles, CA, 90089, United States of America, [email protected], Anthony Dukes Using a unique dataset detailing advertising expenditures by media type, we examine the association between advertising investments, and short- (cash flows) and longterm financial performance (Tobin’s Q). We find that reliance on aggregate advertising data leads to biased assessments on the influence these investments have on marketplace and financial performance. Using detailed advertising data, we identify media type efficacy in creating firm performance, and detail specific firm and industry conditions moderating these associations. This study contributes to a better understanding of the effectiveness of, and how firms can leverage advertising investments, and how such investments create marketplace and financial performance. There is evidence that consumers sometimes feel confused when shopping because of the large number of products available. Conventional wisdom suggests that a firm takes caution when extending its product line in order to mitigate confusion. However, this wisdom ignores the impact of consumer confusion on competitive interactions among firms. We develop a competitive model of firm competition when a portion of the consumers become confused due to the number of products available. We consider two levels of confusion and find that if some consumers are moderately confused, then even the presence of a small portion of these confused consumers provides benefits to firms in the form of reduced price competition. This effect encourages firms to further obfuscate consumers’ choice by expanding their product lines beyond the level of when there is no confusion. Alternatively, firms shrink their products lines but do not lower prices as consumers become highly confused. Finally, consumer welfare among confused and non-confused consumers, as well as total social welfare, unambiguously suffers as a consequence of either moderate or high levels of confusion. 3 - The Impact of Capital Structure on Customer Satisfaction Reo Song, Assistant Professor, Kansas State University, Marketing Department, Calvin 104, Manhattan, KS, 66506, United States of America, [email protected], Gautham Vadakkepatt This study develops and tests a comprehensive conceptual framework that examines the direct and indirect link between capital structure, i.e. a firm’s mix of financial claims, and firm-level customer satisfaction. The authors hypothesize that a firm’s capital structure can directly and indirectly impact customer satisfaction. Capital structure can directly affect customer satisfaction by imposing liquidation cost on the 42 MARKETING SCIENCE CONFERENCE – 2011 FB01 2 - Is Rewarding VIPs Profitable? Sangwoo Shin, Assistant Professor, Krannert School of Management, Purdue University, West Lafayette, IN, 47907, United States of America, [email protected], Jia Li 2 - Choosing the Right Plan? Asymmetric Biases in 3-part Tariff Plan Choices Vardit Landsman, Tel Aviv University and Erasmus University, P.O. Box 39040, Tel Aviv, 69978, Israel, [email protected], Itai Ater Underlying most VIP/loyalty programs in practice is the belief that a company can maximize profitability by rewarding its best customers. Academia, however, has often cast doubt on this practitioner’s belief, and there has been long-standing controversy over the effectiveness of VIP programs. The primary objective of this paper is to provide empirical evidence that sheds light on the question of whether and how the VIP programs work. More specifically, we attempt to quantify and further decompose the impacts of a VIP program on different aspects of customers’ purchase behavior. Using customer-level panel data sampled from the database of a large single-site department store, we build a structural model to measure the impacts of a VIP program on consumer’s purchase behavior in the two aspects: the frequency of purchase and the amount of spending. Rich customer-level transaction and reward information contained in our data enables us to accomplish our task. Upon estimation, we conduct a number of counterfactual experiments to evaluate how the VIP program affects the firm’s profitability. Our finding suggests that the department store in question benefits mostly from encouraging customers to spend more. This paper investigates consumer plan and usage choices when facing a menu of three-part tariff plans. We find that most consumers do not choose three-part tariff plans that minimize their costs, but rather choose plans with too high allowances. We further show that post choice plan switching behavior is driven by consumers’ attempt to avoid overage payments. We try to broaden our understanding of the sources of this behavior, and identify alternative choice mechanisms that best fit the observed choices. Finally, we examine the implications of consumer choice biases on the profits of firms. We use a panel of 70,000 customers from a large commercial bank that introduced a menu of 6 three-part tariff plans. The new pricing scheme was offered to existing bank customers along with the option to keep the old per-usage pricing scheme. Our customer level data include detailed usage, financial, demographic and marketing information over 31 months, including 6 months before the new three-part tariff menu was introduced. 3 - A Model of the Consumer Pricing Decision Process under Pay-what-you-want Marcus Kunter, RWTH Aachen, Templergraben 64, Aachen, 52064, Germany, [email protected] 3 - The Effects of Effort Level on Reward Redemption Behavior Jiyoon Kim, PhD Student, Korea University, Business School, Anam-dong, Seongbuk-gu, Seoul, 136701, Korea, Republic of, [email protected], Janghyuk Lee, Sang Yong Kim Pay-What-You-Want (PWYW) is a participative pricing mechanism where the customer is allowed to pay what he wants for the product, even nothing, and the supplier must accept the customer’s offer (a dictator game). It is applied in various fields such as restaurants, hotels, cinemas, the music industry, computer games, museums and internet sites (e.g. Wikipedia). Research contributions on PWYW investigate the effect of customer and market characteristics on the customer-chosen PWYW price and the sellers’ unit sales and revenues. We contribute to this literature by analyzing the pricing decision process in the mind of the customer at the point and time of purchase. We adress it in a conceptual model which is tested in a preliminary study with qualitative in-depth interviews. The model investigates the following questions: Firstly, what determinants (e.g. uncertainty) exhibit the strongest impact on the customer’s selection of a reference point for PWYW pricing? Secondly, does the customer also incorporate the supplier’s cost situation in his decision? If so, what is his sharing rule (e.g. equal split)? Finally, what determinants (e.g. fairness) affect the customer’s utility or how does he evaluate alternative PWYW price options? We conduct a laboratory and a field experiment to answer these questions. This study investigates the effects of consumer’s effort level on the reward redemption behavior. Despite the wide spread of loyal program in B2C industry, research on loyalty program usage behavior such as reward accumulation and redemption is limited. Previous research (Smith and Sparks, 2009; Reibstein and Traver, 1982) enlightened influencing factors of coupon redemption rate. This study explores loyalty program behavior in depth by testing empirically the endowment effect (Kahneman et al. 1990) in the context of reward accumulation (effort level) and reward redemption behavior. For testing a large scale transaction data set of a general loyalty program such as Air miles. We operationalize two different variables of effort level: accumulation frequency and type as well point redemption speed is measured with left and right censoring methods. Our findings show that there exists a significant negative effect of effort level on redemption speed. The more consumers exert on effort to cumulate reward points, the slower they redeem points. As well we test the generalizability of censoring methods in loyalty program research with customer cohorts of different tenure duration. It is recommended that loyalty program should provide easy ways to cumulated reward points in order to make its customers redeem reward points quickly. ■ FA15 Friday, 10:30am - 12:00pm Champions Center V CRM IV: Customer Loyalty Contributed Session ■ FB01 Chair: Janghyuk Lee, Associate Professor, Korea University, Business School, Anam-dong, Seongbuk-gu, Seoul, 136701, Korea, Republic of, [email protected] 1 - Understanding Whether and How Marketing Efforts Drive Loyalty in the Car Industry of Emerging Markets Guillermo Armelini, Assistant Professor of Marketing, ESE Business School, Av. Plaza 1905, Santiago, Chile, [email protected], Hernán Román Legends Ballroom I Choice IV: Market Structure & Substitution Contributed Session Chair: Kamer Toker-Yildiz, PhD Student, University at Buffalo, Marketing, School of Management, 215 Jacobs Management Center, Buffalo, NY, 14260, United States of America, [email protected] 1 - Measuring How Different Marketing Instruments Affect Competition: The Role of Choice Model Specifications Qiang Liu, Purdue University, 403 W. State Street, West Lafayette, IN, 47907, United States of America, [email protected], Thomas Steenburgh, Sachin Gupta The purpose of this study is to understand whether changes in advertising and in the size of the brand’s dealer network affect customers’ willingness to buy the same brand again in an emerging market (Chile). Prior research showed that customer loyalty is a key issue in driving customer profitability in the car industry. Most of this literature though defines loyalty in terms of consumers’ purchasing intentions, and therefore little is known about the determinants of loyalty based on actual repurchasing behavior, which is the definition we use. We check our hypotheses using a proprietary data base which contains records of all Chilean cars buyers’ transactions from 2000 to 2010, spending in advertising by media and the size of the car brand’s dealer network during the same period of time. We model loyalty using transition matrices and develop an econometric model to assess how car brand’s marketing efforts affect customer loyalty controlled for other issues that might affect customer election as well, such as consumers’ demographics, economic factors (i.e. exchanges rates), the vertical structure of product lines for each car’s brand, etc. Our first results indicate a moderate effect of both, advertising spending and the size of the brand’s dealers network in customer loyalty. We also found that the vertical structure of product line is a key antecedent of customer loyalty. Limitations and managerial implications for managing loyalty in the car industry of emerging markets are also discussed. A large sub-class of discrete choice models that includes the logit, Generalized Extreme Value, and covariance probit models, possess the Invariant Proportion of Substitution (IPS) property. The IPS property implies that the proportion of demand generated by substitution away from a given competing alternative is the same, no matter which marketing instrument is employed. Indeed our empirical application to prescription writing choices of physicians in the hyperlipidemia category shows this to be the case. We find that three commonly used models that all suffer from the IPS restriction – the homogeneous logit model, the nested logit model, and the mixed logit model – lead to unintuitive estimates of the sources of demand gains due to increased marketing investments in DTCA, detailing, and meetings and events. We then employ an alternative choice model specification that relaxes the IPS property – the so-called “universal” logit model. The mixed universal logit model predicts that increases in DTCA result in sales gains that come primarily from non-drug treatments rather than from other cholesterol lowering drugs. By contrast, the mixed logit model predicts for both DTCA and physician meetings and detailing that gains would come largely from brand switching. 43 FB02 MARKETING SCIENCE CONFERENCE – 2011 ■ FB02 2 - Optimal Dynamic Pricing Strategies: Consumer Cross-category Incidence/Purchase Quantity Decisions Sri Devi Duvvuri, Assistant Professor, University at Buffalo, Marketing, School of Management, 215F Jacobs Management Center, Buffalo, NY, 14260, United States of America, [email protected], Praveen Kopalle Legends Ballroom II UGC-II (Quest for Comprehension and Integration) Cluster: Internet and Interactive Marketing Invited Session The objective of this research is to determine optimal pricing policies from a retailer’s perspective by taking into consideration the dynamics of consumers’ purchase behavior across categories. While research on category management focuses on developing pricing strategies within a category, our approach involves both withinand cross-category interactions. In the consumer demand model, we investigate the effects of including the quantity purchased on estimation and specification of multivariate choice models. Specifically, we estimate a Hierarchical Bayes econometric specification of a multivariate tobit model. A multivariate tobit model simultaneously accounts for purchase incidence and quantity decisions of consumers across multiple categories. The model will be estimated using Markov Chain Monte Carlo (MCMC) methods such as Gibbs Sampling and Metropolis-Hastings Algorithm. Scanner data for different categories are used for estimation. Using the estimated parameters, we develop profit-maximizing dynamic pricing policies over time for the various brands across the two categories. Chair: Manish Tripathi, Emory University, Atlanta, GA, United States of America, [email protected] 1 - Listening in on Online Conversations: Measuring Consumer Sentiment with Social Media David Schweidel, University of Wisconsin, Wisconsin School of Business, Madison, WI, United States of America, [email protected], Wendy W. Moe Businesses are increasingly turning to social media as a listening tool in an effort to gain insights into how consumers perceive their brand. Some recent research has examined the predictive ability of both the volume of social media contributions and the sentiment expressed therein. Much of this work, however, focuses either on comments contributed to a single online venue or on a simple aggregation of comments across venues and domains (e.g., total number of contributions across venues or average sentiment expressed across all venues). In this research, we probe the differences in sentiment that manifest across multiple types of venues, including blogs, forums and social networks. We also investigate differences in sentiment that may exist between contributors with and without direct experience. To this end, we model the posted online sentiment pertaining to a single brand across all social media venues. Our analysis reveals notable differences across venues and time in the sentiment that contributors express, suggesting that measures of social media will depend on the source from which they are gathered. By controlling for the effects of venue, domain, experience and other factors, we proposed an adjusted online sentiment metric that represents underlying consumer sentiment better than metrics based on contributions to a single online venue or a simple aggregation of all online contributions. 3 - Market Delineation Strategies in Consumer Goods Market Sebastian Gabel, Doctoral Student, RWTH Aachen, Virchowstr. 2, Aachen, 52066, Germany, [email protected], Raimund Bau The term market delineation is closely associated with anti-trust research and adjacent literature. Yet, knowledge of the relevant products in a market is of great importance for marketing practitioners and researchers as well. In clearly differentiated markets with small number of alternatives, marketing research and practice can easily incorporate all alternatives into models and competitor strategies. Problem arise when potential markets carry several thousand SKUs that (a) cannot be consolidated on a small set of properties and (b) exhibit parallel product usage as well as full substitution patterns. In these markets practitioners often use heuristics, such as product market shares, to reduce their field of attention to a feasible set of products. However, heuristics may lead to misguided management decisions. For researchers using discrete choice models market delineation is of great importance if the model estimation becomes computationally intensive due to a large number of alternatives. Popular sampling strategies might not fully reflect consumer substitution patterns of a particular firm’s products. Using a market simulation we evaluate how different delineation strategies make it possible to reach both the practitioner’s and researcher’s goals. We evaluate 1) Management heuristics based on aggregate data, 2) Advanced descriptive individual-level data, 3) Causal time series analysis and 4) Random sampling. Based on choice sets derived in each approach we analyze potential bias in parameter estimates and elasticities in an aggregate logit model. A comparison demonstrates which strategy yields best results under given circumstances. Finally, the benefits of using the results of this simulation study will be illustrated in an analysis of a German food market. 2 - Social Tag Maps: A New Approach for Understanding Brand Association Networks Hyoryung Nam, University of Maryland, College Park, MD, United States of America, [email protected] We present a novel technique for inferring consumers’ brand associative networks using data from social tags. Social tags are a manifestation of a consumer’s mental representation of a particular product, brand or firm; and hence provide a rich source of information for assessing how consumers think and conceive different products and brands individually as well as related to each other. Given the massive rise in user generated content in the recent years, many websites provide such information in a readily accessible manner. Marketers have much to gain from processing this information to understand the revealed rather than stated preferences of consumers for their products and brands. We demonstrate that the social tag maps have significant benefits for understanding brand associations as compared to conventional techniques such as metaphor elicitation techniques or brand concept maps. In particular, social tags provide a unique ability to obtain information about dynamic evolution of brand associations and competitive associative maps. 4 - The Influence of Willingness-to-pay on Consumer’s Cross Category Purchase Behavior Kamer Toker-Yildiz, PhD Student, University at Buffalo, Marketing, School of Management, 215 Jacobs Management Center, Buffalo, NY, 14260, United States of America, [email protected], Sri Devi Duvvuri, Minakshi Trivedi 3 - The Quest for Content: The Role of User Generated Links in Online Content Shachar Reichman, Tel-Aviv University, Israel, [email protected], Jacob Goldenberg, Gal Oestreicher Consumer’s cross-category purchase behavior has been investigated by marketing researchers due to its important managerial and theoretical implications. This stream of studies have shown that incorporating cross category effects allows a better understanding of consumer’s purchase behavior. For a typical shopping trip, consumers are likely to make multiple purchases and their purchase decisions depend on various observed and unobserved factors. In addition to studying the effects of observable variables (e.g., marketing mix), recent research has also documented modeling efforts that also estimate the effects of unobservable factors. In this study, we propose a cross-category model that allows us to measure how consumer’s willingness-to-pay influences his actual purchase decision in a multi-category decision making framework. It is important to note that willingness-to-pay is a multidimensional variable and unobservable to the researcher. We build a multivariate tobit model to accommodate both cross-category purchase behavior and the willingness-to-pay construct. We use hierarchical Bayesian inference to estimate a heterogeneous specification of our model. We use scanner panel data for organic and conventional product types across multiple categories. Interestingly, we use survey data from the households in the panel to further investigate and validate our findings. Online content is often presented as a product network, where nodes are product pages linked by hyperlinks (e.g., books on Amazon.com). These links are often generated by collaborative filtering algorithms and are based on aggregated data. Recently, websites have begun to offer social networks and user generated links alongside the product network, creating a dual network structure. We focus on the role of this dual network structure and specifically of the user generated links in facilitating content exploration. We first analyze the YouTube.com dual network and show that user pages have unique structural properties (e.g. high betweenness centrality) and act as better content brokers in the dual network. Next, we create a synthetic dual network and show that random rewiring of the product network cannot replicate the brokering effect of the user generated links of the real world dual network. We conduct five internet experiments to study the effect of different structures on the efficiency of the exploration process (time to desirable outcome), and effectiveness (average product rating and overall satisfaction). Our first two experiments show that the dual network structure leads to faster access to “good” content and to overall higher satisfaction. We subsequently extend the experiments to include dynamic structures, in which the underlying site structure changes as a function of time and in response to participants’ satisfaction. Our findings suggest that using a dynamic mechanism leads to an increase in consumers’ ratings of the products as well as to higher overall satisfaction. 44 MARKETING SCIENCE CONFERENCE – 2011 FB04 4 - A Framework for Unifying Differentiated User-generated Content: What I Say, Where I Go, and What I Think Manish Tripathi, Emory University, Atlanta, GA, United States of America, [email protected], Ashish Sood 3 - Video Ads Virality Thales Teixeira, Assistant Professor, Harvard Business School, Morgan 165, Soldiers Filed Rd, Boston, MA, 02163, United States of America, [email protected] The authors present a comprehensive quantitative model that links multiple types of User-Generated Content (UGC). Typical uses for UGC include announcements, reviews, and geographical-based tracking. The authors employ a novel data set that combines regional data from Twitter (announcements), Yelp (reviews), and Gowalla (geographic) to provide insights on the interaction between these forms of UGC, as well as the path of information flow across these types of UGC. In addition, the authors examine how category and consumer characteristics moderate differentiated UGC interactions. Finally, the research provides insights for the growing industry trend of UGC consolidation. Viral ads are videos placed and shared online. To be successful substitutes of paid media (e.g. TV), viral ads need to generate millions of views, to justify this ‘earned media’ approach to advertising. Consumers opt-in to viewing because the ads provide some utility as opposed to the traditional model in which media content provides the majority of utility and viewers choose which ads to view in an opt-out manner. This research looks at properties of highly successful viral video ads inducing opt-in from two sources of utility: content and social utility. It disentangles the role of the creative, measured by utility of content, from that of the sharer’s social inclinations, in the popular domain of humorous ads. Utility from humor is assessed via feelings of joy evoked during viewing while the viewer’s inclination towards sharing the ad is assessed by psychological traits. In a lab we expose subjects to viral ads, allowing them to freely view and share ads with their acquaintances. In the process, we measure their momentary feelings of joy via facial expressions analysis and use this to calibrate a sequential dynamic model of (1) viewership, (2) sharing incidence and (3) frequency. We find that ads more likely to be fully viewed evoke increasing levels of joy over time, but those that are shared (given full viewing) tend to evoke this emotion in higher amounts towards the beginning and end. Regarding psychological traits, introvert viewers are more likely to watch ads fully, but the extroverts and socially self-centered (non others-directed) are more likely to share viral ads with their acquaintances. This is novel evidence that people who share ads online are not doing so altruistically, i.e. to please the receiver, but in order to benefit themselves. ■ FB03 Legends Ballroom III Internet Relationship Contributed Session Chair: Ingrid Poncin, Professor, SKEMA - Université Lille Nord de France, Avenue Willy Brandt, Lille, 59777, France, [email protected] 1 - How Websites Can Create Trust: The Mechanisms that Build Initial Trust in E-Commerce Environments Paul Driessen, Radboud University Nijmegen, Institute for Management Research, P.O. Box 9108, Nijmegen, 6500 HK, Netherlands, [email protected], Marcel van Birgelen, Eric Rongen 4 - Avatar Identification on 3d Commercial Website: Gender Issues Ingrid Poncin, Professor, SKEMA - Université Lille Nord de France, Avenue Willy Brandt, Lille, 59777, France, [email protected], Marion Garnier Previous studies about online trust have identified website features that increase trust that potential customers have in an e-commerce vendor, but have failed to produce many insights into the mechanisms behind these effects. The objective of this study is to explain why some website features help to build initial online trust. The authors apply findings from relationship management literature in an online context and propose four dimensions of perceived trustworthiness which determine initial online trust: competence, benevolence, credibility and problem-solving ability. They conduct an online experiment using a 2x2x2 design to manipulate three website features: graphic appeal, a security element and a community element. In the experiment, 204 subjects visited one of eight versions of a fictitious website offering holiday villas. The results show that graphic appeal has a significant positive effect on competence, benevolence and problem solving ability. The presence of a security element has a significant positive effect on competence, credibility and problem solving ability. The presence of a community element has a significant positive effect on benevolence, credibility and problem solving ability. Significant interaction effects between graphic appeal and the presence of a security element are found that show that an improvement in graphic appeal will have a positive effect only when a security element is present. Further analysis shows that trustworthiness dimensions fully mediate the relationships between website features and initial online trust. This demonstrates that trustworthiness dimensions are crucial in understanding the online trust building process. New forms of e-commerce and notably three-dimensional (3D) commercial websites considerably change the way customers can shop online. Within this context, the crucial role of the avatar representing the consumer online has to be dealt with. Building on the literature on online gaming and metaverses, identification to the avatar was identified as a central basement of the immersion and online buying process. Three studies were conducted on a 3D shopping mall online. The first study led to identifying identification to the avatar as antecedent of immersion on the website and buying intentions. A qualitative study deepened those results by highlighting gender differences in the identity building process and the relationship between the consumer and his/her avatar. More specifically, three different ways of creating one’s avatar are observed: (1) representation and high similarity between the individual and the avatar, (2) representation of an improved or ideal self and (3) fanciful representation. Women seem to mainly represent themselves as they really are, or in some cases, an improved or ideal self, while men are more prone to an improved representation of themselves or to create a totally fanciful and imaginary character. This has a crucial implication on the shopping process of trying on and buying real products on the 3D shopping, mall as well as an implication on the relationship between the self and consumption. Finally, the third study focused through a longitudinal quantitative approach - on the dynamic process of identification to the avatar and statistically confirmed this influence of gender. 2 - The Quality of Electronic Customer-to-customer Interaction: Classification and Consequences Moritz Mink, Frankfurt-School of Finance & Management, Sonnemannstrasse 9-11, Frankfurt, 60314, Germany, [email protected], Dominik Georgi ■ FB04 Legends Ballroom V Dynamic Models II The quality of electronic customer-to-customer interaction (eCCIq) becomes crucial, either in innovative business models that are intentionally built on such eCCI occurring completely on purpose (e.g., Ebay, Facebook) or at least for providers that are embedding web 2.0 and social media techniques in their service offering (e.g., Fidor Bank AG). This drives us to focus on the conceptualization and operationalization of eCCIq as well as on its impact on some of service marketing’s key concepts. To make it more tangible in a first step we review related concepts and constructs such as C2C relationships, CCI, service quality and especially e-service quality. Further related topics are the concept of customer experience and motivation theory. From these reference points we deduct twelve factors that constitute the domain of eCCIq: tangibles, reliability, responsiveness, empathy, (quality of) content, quantity, positivity, trust, security, privacy, social benefit, and entertainment. Subsequently we give an integrated classification of eCCIq aftermath along a very generic specification of the service profit chain. Our study describes the first step to eliminate one most likely reason for eCCI having received so little scholarly attention: The feeling that their management lies outside the direct control of the business. The next logical step is an empirical conceptualization and operationalization of the eCCIq construct. Results will be presented at the conference. Thereby firms will be enabled to more selectively influence eCCIq. Contributed Session Chair: Roopa Choodamani, Modeling Director, DraftFCB, 633 N.St Clair, Chicago, IL, 60611, United States of America, [email protected] 1 - Advertising Strategies by Multinational Firms Wiebke Schlabohm, University of Hamburg, Lehrstuhl Professor Gedenk, Welckerstr. 8, Hamburg, 20354, Germany, [email protected], Barbara Deleersnyder As firms globalize their businesses, they need to manage their advertising activities in an international market, and allocate funds across the home and foreign markets. Yet, little is known regarding how multinational advertising is managed, and what the implications are on global firm performance. This study empirically examines the patterns of advertising adjustments by multinational firms over time, both in the home and foreign markets, and determines the impact of alternative advertising strategies on global firm performance. For this purpose, data are gathered for a wide sample of 135 global advertising firms over several decades. For each firm, U.S. as well as worldwide advertising spending is examined, and subsequently linked to various firm performance measures (including stock prices and annual revenues). Our findings offer evidence that multinational firms behave myopically, and tend to favor their home market over more distant ones. In particular, firms increase their advertising significantly faster in the domestic market compared to advertising spending in foreign areas. Such behavior is more likely when firms are financially constrained. Moreover, when reducing international advertising budgets, companies tend to restrict budgets more in the foreign as opposed to their home markets. These advertising strategies are found to affect global firm performance. Our results should caution managers of multinational firms not to be trapped in such myopic advertising decisions, but develop strategies to reduce incentives to favor markets nearby. Key words: international, advertising, multinational, myopic marketing management,globalization. 45 FB05 MARKETING SCIENCE CONFERENCE – 2011 ■ FB05 2 - Modeling Dynamics of Consumer Preference and Promotion Effect in Brand Choices Eiji Motohashi, The Graduate University for Advanced Studies, 10-3 Midori-cho, Tachikawa, 1908562, Japan, [email protected], Tomoyuki Higuchi Legends Ballroom VI Game Theory I: Decisions Under Limited Information Cluster: Special Sessions Invited Session This paper develops a choice model that includes time-varying parameters, which are called state variables, based on the state-space approach. We show how to incorporate both invariant individual heterogeneity and variant marketing effects in brand choices into the multinomial logit framework. Marketers are often interested in two kinds of variations on consumer preferences and sensitivities to marketing actions. One is the variation within consumers and the other is one within different time. For a variety of reasons such as change of market conditions, parameters in a choice model may vary over time; therefore, it is important to appropriately model both variations to obtain the correct inferences. The state space approach offers an optimal modeling framework for this problem, so we formulate dynamic change of parameters in the style of it. We consider some of household specific variables for explaining consumer heterogeneity. In empirical analysis, we use the particle filter method to estimate state variables and unknown parameters, and apply our methodology to scanner panel data in instant coffee category, which have been gathered at a super market over three years in Japan. We find that the effects of marketing efforts are not stable during the data period. The results also show time variation in parameters is more important than household specific variables to predict unknown brand choices. In the presentation, we will show the model specification, results of analysis, and computational algorithm for the estimation method. Chair: Jeffrey D. Shulman, University of Washington, Michael G. Foster School of Business, Seattle, WA, United States of America, [email protected] 1 - How Hidden Add-on Pricing Can Reduce Profit Jeffrey D. Shulman, University of Washington, Michael G. Foster School of Business, Seattle, WA, United States of America, [email protected], Xianjun Geng After committing to purchasing a product or service, consumers are often required to pay add-on prices for additional features or services. For example, upon checking into a hotel, consumers may then choose whether to pay to access the internet, park a car, or even to use the pool. This paper uses an analytical model to examine the consequences of these fees when there is a segment of boundedly rational consumers who were unaware of these fees at the time of initial purchase and inaccurately believed the listed price was all-inclusive. The literature on add-on pricing finds that charging separately for add-ons will, at worst, have no effect on profit and in some cases will actually improve profit. However, we demonstrate how and when profit will be lower when firms are able to employ add-on pricing, even in the presence of consumer bounded rationality. Moreover, we identify situations in which firm profit is decreasing in the number of consumers who underestimate the true price for the combined base and add-on products/services. Thus bounded rationality regarding hidden add-on prices may actually hurt firm profitability. 3 - Be Careful When Using the Mover-stayer Conceptual Framework in Brand Choice Mode Kanghyun Yoon, Assistant Professor of Marketing, Long Island University, C.W. Post Campus, 720 Northern Blvd, Roth Hall #114, College of Management, Brookville, NY, 11548, United States of America, [email protected] 2 - Salesforce Compensation under Inventory Considerations Kinshuk Jerath, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected], Tinglong Dai The mover-stayer (MS) framework (Blumen, Kogan and McCarthy 1955) has been popularly used in marketing to account for household heterogeneity, to study consumers’ brand switching patterns, and to investigate consumers’ inertia/varietyseeking tendencies when making brand choice decisions. This framework assumes that the probability of hard-core loyals (or inertia consumers) in the stayer segment making repeat purchases is equal to one. Due to this assumption, MS-based brand choice models provide very limited insights into the purchasing behaviors of consumers in the stayer segment, and most importantly, may result in biased parameter estimates. In this paper, we first demonstrate that the parameter estimates of the MS-based brand choice models – e.g., the variety-seeking and Lightning Bolt models – are indeed biased. Next, we introduce a generalized mover-stayer (GMS) framework which relaxes the aforementioned “probability-one” assumption. Using the GMS framework, we develop various GMS-based brand choice models which capture the differences in the state-dependent purchasing patterns of consumers given the two hidden states (i.e., inertia and variety-seeking tendencies). The calibration results of these models show that the parameter estimates are not subject to the bias problem upon relaxing the probability-one assumption. Managerial implications are discussed along with directions for future research. We study a scenario in which a firm has to decide the compensation contract for a sales manager whose job is to exert effort to increase the level of demand. Demand is uncertain, and actual sales depend on the realized demand and the availability of inventory (with the inventory level decided directly by the firm). In this context, we find that the variable compensation rate of the sales manager can increase with increasing demand uncertainty, which is a result contrary to classical salesforce compensation theory but is observed empirically. We also find that the optimal level of inventory can decrease with demand uncertainty even if backorders are costlier than unsold units, which is a result contrary to classical inventory theory. We also endogenously derive a quota-bonus contract, which is a widely-prevalent compensation contract, as an optimal contract for the sales manager. In summary, we find that studying the salesforce compensation problem along with operational considerations can provide novel insights that explain results that existing theory cannot. 3 - Memories and Rules Juanjuan Zhang, Massachusetts Institute ot Technology, Sloan School of Management, Cambridge, MA, United States of America, [email protected], Jeanine Miklós-Thai 4 - Morphing Marketing Response Optimization - Advocating a Next Practice Roopa Choodamani, Modeling Director, DraftFCB, 633 N.St Clair, Chicago, IL, 60611, United States of America, [email protected], Pradeep Kumar People may remember their decisions but not the information they relied on when arriving at these decisions. However, one can engage in observational learning from her past self by inferring what she knew from what she did. If past information is relevant for future decisions, a forward-looking decision-maker may strategically bias her current choice to make memories more informative. This finding can be applied to corporate settings where management turnover causes organizational memory attrition. To inform future decisions and maximize long-run profits, a firm may prevent managers from responding to short-run market shocks. Traditional sales response functions to marketing drivers are concave (downward). These response functions are identified based on the data estimation range and the best fitting relationship between sales and marketing. The choice of these response functions are critical to answer key managerial questions and perform “What-If” simulations to simulate the impact of marketing drivers on sales at different levels of marketing. Previous work in academia and the industry has focused largely on identifying the single best functional form like S-Curves, Log curves and the like that fits the data. From an industry application perspective, while the ‘What-If’ paradigm helps to evaluate scenarios to a large extent, given the dynamic marketing and media environment, a shift in paradigm to ‘If-Then’ simulations is likely to provide versatility to optimizations, while assessing risk and uncertainty to empower better marketing decision making. This work using the widely published Lydia Pinkham data (Source: Kristian S. Palda) addresses the perils of relying on a single response curve to make managerial decisions. Mathematical Optimizations indicate that combining the response curves using model averaging techniques produces superior results and improves managerial decision making by eliminating idiosyncrasies introduced by choice of response curves. The authors examine some of the best practices in the industry to build market response models and optimizations. The intent is to suggest techniques in response modeling and ptimizations to evolve a next practice in this increasingly applied area of management science.What are some of the best practices in the industry in resource allocation?What would “Next Generation “approaches to marketing mix modeling look like? 4 - The Model of Buzz Jiwoong Shin, Associate Professor, Yale School of Management, New Haven, CT, United States of America, [email protected], Arthur Campbell, Dina Mayzlin In this paper we explore how the firm can use word of mouth marketing and advertising to optimally target information to different groups of consumers in order to maximize the diffusion of information about its product. That is, suppose that there are two types of consumers –the high (or the socially desirable) type and the low (or the socially undesirable) type. A consumer who obtains information about the product is more likely to talk about it if the conversation can serve as a signal that the consumer is from a socially desirable group. Hence, by making it cheaper for the desirable type of consumer to obtain information about the product and by making it more expensive for the socially undesirable type to do so, the firm maximizes the length of time that consumers will pass on the information about the product. This effect is further intensified if the mixing patterns between the types differ. We also show that the firm may benefit from the commitment to not reveal information to the low-type group throughout the information diffusion process. On the other hand, if such a commitment is not possible, the firm will choose to reveal some information to the low-type consumers. Finally, we explore the question of when a …firm may choose to invest resources by communicating product information to the consumers versus having the consumers engage in costly search on their own. 46 MARKETING SCIENCE CONFERENCE – 2011 ■ FB06 FB07 4 - Insights Into the Role of the Internet in a Multichannel Customer Management Strategy Tanya Mark, Assistant Professor, University of Guelph, 50 Stone Road East, Department of Marketing and Consumer Studies, Guelph, ON, N1G 2W1, Canada, [email protected], Katherine N. Lemon, Jan Bulla, Antonello Maruotti, Mark Vandenbosch Legends Ballroom VII Channels II: Relationship Management Contributed Session Chair: Sara Valentini, Assistant Professor, University of Bologna, via Capo di Lucca, 34, Bologna, 40126, Italy, [email protected] 1 - Investigating Impact of Multiple Communication & Marketing Mix Elements in Multichannel Environment Ashish Kumar, PhD Student, State University of New York Buffalo, 232 Jacobs, Buffalo, NY, 14260, United States of America, [email protected], Ram Bezawada, Minakshi Trivedi As the Internet begins to account for an increasingly larger portion of a company’s total sales, it is important to understand the impact of the Internet on customer segments and its role in a multichannel customer management strategy. There is mounting evidence in favour of encouraging multichannel behaviour since it is linked to an increase in the number of purchases (Ansari et al. 2008) and higher customer profitability (Venkatesan et al. 2007). Not all is positive though since research also finds that multichannel enthusiasts are less loyal than retail consumers (Konus et al. 2008) and long-term purchase incidence is negatively affected with increase in Internet usage (Ansari et al. 2008). These mixed findings suggest the Internet has an important role in a multichannel customer management strategy but more research is needed to disentangle the long-term effects of multiple channels on buying behaviour. We begin to address this need by investigating the role of the Internet and its impact on purchase incidence and expenditure decisions on a cohort of customers of a multichannel retailer over a 9 year period. We develop and empirically validate a dynamic hurdle model and find three segments: active buyers, occasional buyers, and inactive buyers. Channel usage has differing effects on purchase incidence and expenditure decisions for each of the segments. A consistent finding across all three segments is that multichannel buying behaviour increases the number of purchases. However, relative to the retail store and the call centre, the Internet is the least important channel for loyal customers. On the other hand, the inactive segment is more likely to buy from the call centre and the Internet rather than the retail store. Fueled by technological advances, existing communication methods such as newspaper and television are now being supplemented with emerging media like emails and mobile communications. Furthermore, such advances also enable firms to expand their modes of distribution beyond their traditional outlets to include multiple channels (e.g., brick and mortar stores and online stores). In such a multicommunication, multi-channel environment where consumers may well exhibit differential responses with respect to the above, several issues emerge. Firstly, how do retailers’ online and offline marketing and communication mix elements impact consumer purchase behavior/channel choice? Secondly, how effective are these marketing/communication elements in a multi-channel environment? Finally, what role do consumer characteristics play in such an environment? To address these issues, we utilize a data set comprising information on consumer exposure to traditional communication media (e.g., newspaper, print catalogs) and emerging media (e.g., emails, e-catalogs) in addition to marketing mix variables (e.g., prices, promotions) and purchase information across both offline and online channels. Specifically, we propose a disaggregate, joint category incidence/channel choice and purchase quantity/order size model to understand the above phenomenon. The insights from this research will be helpful for both manufacturers and retailers to more effectively manage multi-channel consumers in a multi-media environment, critical for both profitability and superior customer interactions. ■ FB07 Founders I 2 - Return on Channel Investments for Customer Acquisition – A Cross-channel Analysis Maik Eisenbeiss, Assistant Professor, University of Cologne, Albertus Magnus Platz 1, Cologne 50923, Germany, [email protected], Monika Käuferle, Werner Reinartz, Peter Saffert Panel Session: Collaborative Research: Reasons Why, Difficulties and Potential Models (Data Base sharing and Prospective Meta Analysis Acquiring customers through different distribution channels has been becoming the norm rather than the exception in many industries. In that context, managers are constantly facing the challenge of effectively allocating resources across a set of multiple channels. To the extent that different channels have different search and purchase attributes, customers may favor one channel for product search and another one for purchase. Hence, a channel’s direct contribution to the firm, e.g. the total number of customer acquisitions through that particular channel, is only telling half the story when evaluating its potential for resource allocation. Instead, managers also need to account for cross-channel effects, i.e., the effects of investments in one channel on purchases made through the other channels. In this paper, we try to quantify direct and cross-channel effects of investments on the number of new customer acquisitions of mobile telecommunication providers operating across various channels – exclusive stores, independent dealers, Internet, and call center. Our data set comprises channel specific investment and acquisition information for a global panel of mobile service providers. Besides confirming the general existence of cross-channel effects, our analysis reveals that their direction depends on the type of investment: While investments directed to the information function of a channel (e.g., general advertising) may induce positive cross-channel effects between various channels, investments into a channel’s transaction function (e.g., distribution system) only lead to negative crosschannel effects on the number of new customer acquisitions of the other channels. We additionally show that positive cross-channel effects are moderated by certain market factors. Chair: Glen Urban, Professor, Massachusetts Institute of Technology, Cambridge, MA, 02139, United States of America, [email protected] 1 - Collaborative Research: Reasons Why, Difficulties and Potential Models (Data Base sharing and Prospective Meta Analysis) Moderator: Glen Urban, Professor, Massachusetts Institute of Technology, Cambridge, MA, 02139, United States of America, [email protected], Panelists: Eric Bradlow, Gary Lilien, Don Lehmann, Catherine Tucker, Stefan Stremersch, Jan-Benedict Steenkamp, Jerry Wind, Gui Liberali Cluster: Special Sessions Invited Session This session will explore the potential benefits of cross university collaborative research in marketing science as well as the impediments to such large scale multiperson research programs. We will examine existing data base sharing (e.g. Individual Analytics program at Wharton and Marketing Science data base sharing) to identify success factors. Then we will apply our structure to a collaborative model popular in medicine called “Prospective Meta Analysis”. Applications to new media allocation and product innovation will serve as a discussion point. We will have a panel, but encourage wide audience participation and sharing of collaborative research experiences. 3 - Does Multichannel Usage Produce More Profitable Customers? Sara Valentini, Assistant Professor, University of Bologna, via Capo di Lucca, 34, Bologna, 40126, Italy, [email protected], Elisa Montaguti, Scott Neslin One of the most intriguing and managerially relevant findings in the multichannel customer management literature is the positive association between the multichannel customer and profits. The question is whether this is an actionable, causal relationship, specifically, whether marketing campaigns can be designed to turn single-channel customers into multichannel customers, and in turn whether these multichannel customers will become more profitable to the firm. The purpose of this research is to conduct a field experiment to investigate this question. We design a field test where we vary the incentive (financial vs. non-financial) for inducing the customer to become multichannel, and the targeting of the incentive (random assignment vs. assignment based on a model developed to identify the customers who are expected to generate the most revenues if they become multichannel). The field test started at the beginning of January 2011. We discuss the design of the experiment, the model-based treatment, and the preliminary results. 47 FB08 MARKETING SCIENCE CONFERENCE – 2011 ■ FB08 4 - A Dynamic Structural Analysis of Enterprise Knowledge Sharing Baohong Sun, Carnegie Mellon University, Pittsburgh, PA, United States of America, [email protected], Yingda Lu, Param Vir Singh Founders II The Long Run Consequences of Short Run Decisions II Enterprise 2.0 systems (such as employee expertise sharing forums, employee blogs, and internal wikis) have been adopted in a number of organizations to overcome the barriers to intra-organizational knowledge sharing that have emerged as a result of knowledge silos formed over the year. However, organizations are still struggling to understand the knowledge sharing behavior of employees on these systems, the design of these systems and devising policies to enhance knowledge sharing on these systems. We propose a theoretically grounded dynamic structural model with endogenized network formation that takes into account the “learning by sharing” and “knowledge spillover,” two salient features enabled by the public social platform. More specifically, our model recognizes the dynamic and inter-dependent nature of knowledge seeking and sharing decisions and allows them to be driven by knowledge increment and social status building in anticipation of future reciprocal reward. Applying the model to a unique panel data on expertise sharing forum, we find that the seemly altruism behavior of knowledge sharing with peers can be better explained by a dynamic and interactive decision making in anticipation of future reward reciprocated by the community. During the competition for social reputation, there forms “Core/Periphery” where employees with high reputation are more likely to answer each other’s questions, discouraging users with low social status from participating. Interestingly, active learning by asking questions is more effective in improving knowledge than reactive learning by reading answers. A sensitivity analysis show that hiding the identity of the knowledge seeker breaks the Core/Periphery structure and improve the knowledge sharing by 35.7%. Cluster: Special Sessions Invited Session Chair: K. Sudhir, Yale School of Management, Yale School of Management, New Haven, CT, United States of America, [email protected] Chair: Ahmed Khwaja, Assistant Professor, Yale University, New Haven, CT, United States of America, [email protected] 1 - Dynamic Competition between New and Used Durable Goods Without Physical Depreciation Masakazu Ishihara, University of Toronto, Toronto, ON, Canada, [email protected], Andrew Ching Both theoretical and empirical studies on used goods markets have consistently assumed that used goods markets are perfectly competitive in a static sense, i.e., the quantity demanded for used goods is equal to the quantity supplied of used goods in every time period. This assumption significantly simplifies the analysis on the interaction between new and used goods markets, as used goods retailers are assumed to be price-takers and do not make a dynamic decision. However, none of the previous studies have examined the validity of this assumption. In this paper, we examine the data from new and used video game markets in Japan and find evidence that this assumption is strongly violated in this market. The data set includes weekly sales and prices of new and used video games as well as weekly quantities of used games bought by used video game retailers from consumers and associated resale values. Our data show that on average, the inventory of used video games carried by used video game retailers keeps increasing even after the quantity demanded for used video games starts to fall. In order to rationalize this observed pattern of the used video game inventory, we propose a dynamic structural equilibrium model of new and used goods competition in which consumers, the manufacturer and used goods retailers are all forward-looking. The manufacturer sets prices of new games, and used good retailers set both prices and resale values of the used game. We then estimate the dynamic model using the Japanese video game data. We plan to use the model to quantify the welfare impacts of killing-off the used video game market. ■ FB09 Founders III Retailing III: Competition Contributed Session Chair: Bruce McWilliams, Professor of Marketing, ITAM (Instituto Tecnologico Autonomo de Mexico), Av. Camino a Santa Teresa No. 930, Mexico, DF, 10700, Mexico, [email protected] 1 - If You Build It, Will They Come?: Anchor Store Quality and Competition in Shopping Malls Ravi Shanmugam, Assistant Professor of Marketing, Santa Clara University, 500 El Camino Real, Santa Clara, CA, 95053, United States of America, [email protected] 2 - A Dynamic Model of Competition with Bundling Vineet Kumar, Harvard Business School, Boston, MA, United States of America, [email protected], Timothy Derdenger We develop a dynamic model of consumer purchases of hardware and software bundles, where forward-looking consumers face a choice of purchasing a console bundled with specific games, or a standalone console that is expected to result in future sales of software (games) to the owner of the hardware (console). We use the Dynamic BLP approach of Gowrisankaran and Rysman (2010) applied to aggregate data that tracks sales of videogame consoles and corresponding software titles over time. The dynamic BLP algorithm simplifies the complexities of estimating a full dynamic model of forward-looking consumers on aggregate data. A short-term hardware console or bundle purchase decision made by the consumer thus has implications for future revenue from software sales. We examine firm policies on bundling, including the choice of software to offer in the bundle, as well as evaluation of whether consumers should be able to self-select and create their own bundle. We also focus on understanding the dynamic implications of first-party bundles, where the console producer includes a game it owns within the bundle, in contrast with third-party bundles, where the hardware and software producers are different firms. The ability of shopping centers to attract customers and increase sales depends in part on their anchor stores, the small number of large-sized, high-profile tenants located in every mall. In this paper, a theoretical model of competition between anchor and non-anchor stores in a shopping mall is developed, with the goal of explaining an observed pattern of choices of anchor-store quality levels made by mall developers. In particular, the relationship between a mall’s anchor-store quality levels, size, and measures of mall performance (visitor traffic and store profits) is examined. It is found that mall size, because of its relationship to the probability that consumers will find a “fit” between their preferences and the non-anchor store’s goods, has varying effects on price competition between the stores, visitor traffic, mall profits, and anchor quality levels chosen by mall developers. The primary analytical result is that mall size has a positive and concave, i.e. inverse U-shaped, relationship with the probability that the developer chooses a high-quality anchor over a low-quality one. The predictions of this model are then validated using a data set containing information about key strategic variables for major North American malls, showing that the proposed relationships are robust to the inclusion of inter-mall competitive effects and additional relevant controls. 3 - A Dynamic General Equilibrium Model of User Generated Content Carl Mela, Duke University, The Fuqua School of Business, 100 Fuqua Drive, Durahm, NC, United States of America, [email protected], Dae-Yong Ahn 2 - Dynamic Competitive Intensity in Retail Markets: Drivers and Implications on Retailer Performance Geunhye Yang, PhD Candidate, University of North Carolina at Chapel Hill, Kenan-Flagler Business School, Chapel Hill, NC, 27599, United States of America, [email protected], Katrijn Gielens, Jan-Benedict Steenkamp User content websites involve two behaviors; consuming content (e.g., reading reviews or viewing videos) and generating content (e.g., writing reviews or uploading videos). Users generate free information content for the reputational effect of being influential or popular. The consumption of content can generate utility via the pleasure of reading or the utility of information. Hence, user engagement involves the joint creation and consumption of content where each respective action can generate utility to a participant. We develop a dynamic general equilibrium model of joint consumption and generation of information content based on rational expectations. We estimate this model and conduct policy simulations using a proprietary data from a web site where users generate and consume content in the form of reviews and forum postings. Our approach is general and applies to contexts ranging from chat rooms to journal publications to video sharing sites. Our model has managerial implications relevant to web sites that seek to maximize site traffic and participation; these outcomes being relevant to the advertisers and users alike. Grocery retail markets have become highly concentrated. This trend has substantial implications for how retailers compete and perform. Still, competitive interactions between retailers in these markets have largely been neglected, the most important exception being the change in competitive interaction due to the entry of a “giant” retailer. Therefore, little is known about how powerful players in highly concentrated, ‘on-going’ markets compete and how it affects their performance. In this study, we measure (1) to what extent retailers react to the competitive actions of their rivals, and (2) how competitive reaction sensitivities affect retailers’ category and overall performance. More specifically, we emphasize how retailers react to each other’s pricing and assortment decisions for their private labels (these are the brands for which retailers control the marketing mix entirely). We use reaction functions proposed by Leeflang and Wittink (1992, 1996) to capture competitive sensitivities and extend their approach to allow for time-varying reaction elasticities using Kalman filter inferences. These dynamic reaction elasticities allow us to assess when and how often retailers respond to competitive moves by (1) (strong) accommodation (e.g., if a rival retailer decreases its price, the focal retailer reacts by increasing its price), (2) (strong) retaliation (e.g., the focal retailer decreases its price when a rival does so), or (3) by doing nothing. Next, we relate these different types of reactions to category performance and store traffic. As such, we can evaluate when and to what extent competitive strategies are beneficial or detrimental. 48 MARKETING SCIENCE CONFERENCE – 2011 FB11 3 - The Multiple Effects of Social Comparisons on Consumer Expenditure Rafael Becerril-Arreola, UCLA Anderson School of Management, 110 Westwood Plaza Suite B401, Los Angeles, CA, 90024, United States of America, [email protected] 3 - Money-back Guarantees: The Great Brand Equalizer Bruce McWilliams, Professor of Marketing, ITAM (Instituto Tecnologico Autonomo de Mexico), Av. Camino a Santa Teresa No. 930, Mexico, DF, 10700, Mexico, [email protected] Existing literature on Money-Back Guarantees (MBGs) based on signaling literature suggests that they will be adopted by high quality firms. However, MBGs are ubiquitous in many retailing environments, thus requiring a new analysis to explain their prevalence. We use game theory to examine the impact of adopting MBGs for high and low quality retailers in a competitive environment where consumers are fully informed and returned products have a positive salvage value to the retailer. If salvage values are low enough, neither retailer will offer an MBG. However, if salvage values are high enough, the low quality retailer unconditionally gains from offering MBGs while the high quality retailer loses relative to the No MBG environment. When the low quality retailer offers an MBG, it is Nash equilibrium for the high quality retailer to also offer an MBG. When retailers are allowed to adjust their quality levels, the optimal retailers’ qualities will be more dispersed with MBGs than without them. This work breaks down the effects of social comparisons on positional expenditures into the effects of the frequency of the comparisons and the effects of the comparison discrepancy. We estimate an expenditure system model with expenditure data from 16 geographies and 36 product categories. The results indicate that the frequency of comparisons, as approximated by the frequency of involvement in social activities, is positively associated with levels of spending on expensive visible goods. The comparison discrepancy, as approximated by income inequality, correlates negatively with levels of spending on goods that are both expensive and visible. In addition, we find that social class, as measured by occupational prestige, does correlate with the expenditure shares of goods that are both expensive and visible. We thus show that all income, income distribution, class, and social involvement explain consumer behavior. 4 - Partner Selection in Brand Alliances Ralf van der Lans, Associate Professor, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong PRC, [email protected], Bram Van den Bergh, Evelien Dieleman ■ FB10 Founders IV We investigate whether partners in a brand alliance should be similar or complementary in brand personality to foster favorable perceptions of brand fit. Using a Bayesian non-linear structural equation model and evaluations of 1,200 hypothetical brand alliances, we find that the conceptual coherence between brand personality profiles significantly affects consumers’ attitudes towards a brand alliance. More specifically, we find that similarity in Sophistication and Ruggedness increases perceived brand fit. The similarity (complementarity) effects for Sincerity, Excitement and Competence are non-linear, indicating that the importance of similarity (complementarity) in partner selection depends on the corresponding levels of the brand on these personality dimensions. Bayesian Applications Contributed Session Chair: Ralf van der Lans, Associate Professor, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong - PRC, [email protected] 1 - Variety Seeking in Movie Choice: The Role of Ratings Joon Ro, University of Texas at Austin, 2501 Lake Austin Blvd. F208, Austin, TX, 78703, United States of America, [email protected], Romana Khan In this paper, we study variety seeking across genres in consumers’ choices at movie theaters. While variety seeking encompasses an array of behaviors that promote diversity in choices made, we focus on two components: the tendency to engage in exploratory behavior, and the tendency to seek sequentially varied experiences. Although movies are a hedonic good for which we expect consumers to engage in variety seeking, several factors, uncertainty about movie quality in particular, mitigate this tendency. Online ratings provide signals of movie quality and serve as a mechanism to alleviate this uncertainty. We investigate the extent of variety seeking in movie choices, and the impact of online ratings on variety seeking. Using a unique consumer level panel data of movie-going at theaters, we estimate a movie choice model that accounts for consumers’ intrinsic preferences for movie attributes, demographics, state dependence, and online movie ratings. Surprisingly, consumers exhibit positive state dependence (inertia) over genres in their choice of movies. However, higher online ratings diminish positive state dependence and induce consumers to seek more variety. We find considerable heterogeneity in exploratory behavior and sensitivity to online ratings across consumers. Demographic factors account for some heterogeneity, as older consumers show more inertia and less sensitivity to online ratings. Theoretical and managerial implications are discussed. ■ FB11 Champions Center I Salesforce I Contributed Session Chair: James Hess, Professor, University of Houston, 4800 Calhoun Road, Houston, TX, 77204, United States of America, [email protected] 1 - DEA with Econometrically Estimated Individual Coefficients: A Pharmaceutical Sales Force Application Soenke Albers, Professor of Marketing and Innovation, Kühne Logistics University, Brooktorkai 20, Hamburg, 20457, Germany, [email protected], Andre Bielecki In business practice, the efficiency benchmarking technique DEA has been met with high approval. However, its mathematical programming and cross-sectional data framework have several drawbacks. DEA weights can sometimes be technically valid but practically ill-specified with low predictive validity. Parameter significance or model fit indicators such as R2 cannot be obtained. Another drawback is that an increasing number of variables leads to an increasing number of efficient units. This may lead to wrong conclusions. We propose a model based on multiple observations per unit that combines econometric estimation of individual coefficients with DEA evaluation techniques. Individually estimated coefficients are used as factor weights for efficiency evaluation operations comparable to the DEA method. The model maintains core DEA features and provides valid individual weights, a reduced number of efficient units, parameter significance, and statistical fit as additional advantages. An application for the sales force of a pharmaceutical company illustrates how this method changes the benchmarking results and increases the potential for efficiency improvement. 2 - Inferring Competition in Search Engine Advertising with Limited Information Sha Yang, The University of Southern California, 701 Exposition Blvd., Hoffman Hall 803, Los Angeles, CA, 90089, United States of America, [email protected] A challenge facing search engine advertisers is how to infer competition with limited competitive information. However, a good understanding of competition is crucial for advertisers to improve their profitability in the generalized second-price auction implemented by most search engines today, in which ad positions are determined based on ad rank. In this paper, we develop a model to help address this challenge. Our model takes into account the key aspects of the generalized second-price auction, and predicts the expected ad rank of competing ads at different positions for a given keyword. The novel aspect of our model is to assume that ad ranks of all competing ads on a given keyword follow a distribution. We then estimate the key parameters of the distribution using the incomplete ordered ad rank data drawn from one advertiser: (1) ad rank of the ad placed at the bottom of the first page on paid listings, (2) ad rank of the focal advertiser at its current position, and (3) ad rank of the ad positioned right below the focal advertiser. We develop a Bayesian approach for estimating the proposed model. We also perform an external validation, which shows that the proposed model predicts the ad rank of one competitor reasonably well. Our empirical result suggests that both the mean and the variance of the ad-rank distribution are heterogeneous across keywords, suggesting different patterns of competition. Finally, we conduct two counter-factual analyses to illustrate how our proposed model can help the focal advertiser improve its profitability by choosing the appropriate keyword management strategies. We find that improving quality score for one point is three times as effective as changing maximum bids to increase the expected profit, holding everything else constant. 2 - Assessing Salesforce Performance: An Empirical Approach Wei Zhang, Assistant Professor of Marketing, Long Island University, College of Management, 720 Northern Blvd, Brookville NY 11548, United States of America, [email protected], Ajay Kalra We propose a new approach to measuring sales people performances. We model the sales agents’ learning about their selling ability through interactions with prospective customers. We develop a Bayesian learning model where sales agents learn about their inherent selling aptitude as well as selling skill development. As skill development evolves, our model is a more general case as compared to consumers’ learning of product quality which remains unchanging. We gauge selling aptitude and development using only sales data, while controlling for customer characteristics and territorial differences. The model allows distinguishing between salespeople who are truly capable and those who are fortunate in getting a customer mix with a high disposition to buy. We also parcel out the sales agent’s ability that accounts for the sales versus the predisposition of the customers as related to their characteristics. We also construct indices that compare a sales agents’ performance to their reference group and assesses their ability for specific customer types. 49 FB12 MARKETING SCIENCE CONFERENCE – 2011 3 - Sales Contests and Quotas with Imbalanced Territories - A Model and Experiments James Hess, Professor, University of Houston, 4800 Calhoun Road, Houston, TX, 77204, United States of America, [email protected], Niladri Syam, Ying Yang 3 - A Study of Consumer Interest in Innovative Products Across Developed and Emerging Markets Gauri Kulkarni, Assistant Professor of Marketing, Loyola University Maryland, 4501 N Charles St, Baltimore, MD, 21210, United States of America, [email protected] This paper studies the design of sales contests and quota-based compensation systems when territories have imbalanced sales potential. We ask, (1) how do the optimal sales, efforts of salespeople, and profits vary with territory imbalance in an optimal sales contest? (2) How do these factors vary with territory imbalance in a quota system? (3) Are sales, efforts and profits larger under a contest than a quota system when territories have different sales potential? The main managerial insight that we offer in this paper is that with imbalanced territories, the firm’s profit with the quota system is higher than its profit with a contest. Furthermore, the advantage of the quota system increases with degree of imbalance. We also find that for a sales contest, the agents in the strong and weak territories exert the same effort which decreases with the degree of imbalance in sales potential to the point of complete shirking. In a quota system the firm’s profit and agents’ efforts also decrease when there is imbalance, but not as much as in a contest. Further, unlike the contest, with a quota system the salesperson in the territory with weaker sales potential works harder than the salesperson in the stronger territory. Three laboratory experiments support the findings of the analytical model. The diffusion of new product sales in both domestic and foreign markets has been a widely researched area in marketing for quite some time. Recent technological advancements have allowed for the measurement of consumer interest in new products prior to the availability of sales data. One example of such measures is the volume of terms submitted to search engines. In other words, the search volume of a particular term can serve as an aggregate indicator of consumer interest in that particular term. This measure is available prior to the launch of a product, since a product need not be available to consumers for them to search for it. Since this online measure of consumer interest is not restricted by geographic boundaries, the levels and patterns of consumer interest across international markets can also be investigated. Therefore, this research aims to study the levels and patterns of consumer interest, as measured by online search term volume, in innovative products across developed and emerging markets. Findings from the study can offer insight on the future diffusion of product sales in these markets and can also aid in managerial decision-making on the timing of international product launches. 4 - Application of Case Study on the Quality of Public Transport in European Cities with a Tool for Digital Ethnography Agustí Casas-Romeo, Professor, Department of Economics and Business Organization. Universitat de Barcelona, Main Building, Tower 2,3rd fl.Diagonal690, Barcelona, 08034, Spain, [email protected], Rubén Huertas-García, Juan Carlos Gázquez-Abad ■ FB12 Champions Center II Internet: Unique Topics The architecture of online communities and social networks provide a constant flow of information, some redundant, some significant, but all of them fed by individual users in the online community. We can use digital ethnography to investigate the ways in which social media get significant results. A key consideration of virtual ethnographic approach is portability ahead of the generalization and this is an important evaluation criterion. Shadish (1995: 419). There is still no complete scientific method for absolute filtering of the sources that the virtual environment offers, however the mass of information has found a way scales that discriminate in a revealing way information “noisy” leading researchers at a similar stage virtual traditional investigators in terms of credibility and reliability, Kozinets (2000:61) leading to comparable results of research. Through Epsilon technology and methodology we present an exploratory ethnographic case study of the quality of service of public transport (Metro, Bus, Commuter Car, Tram, etc...). In various European cities (London, Rome, Paris, Berlin, Madrid, Lisbon). The degree of reliability and speed with which it has conducted an exploratory analysis revealed the severity and cost reduction makes this technique is effective and efficient consolidation. Contributed Session Chair: Agustí Casas-Romeo, Professor, Department of Economics and Business Organization. Universitat de Barcelona, Main Building,Tower 2, 3rd fl. Diagonal 690, Barcelona, 08034, Spain, [email protected] 1 - Quantifying Transaction Costs in Online/Offline Grocery Channel Choice Junhong Chu, Assistant Professor, NUS Business School, 15 Kent Ridge Drive, Singapore, 119245, Singapore, [email protected], Pradeep Chintagunta, Javier Cebollada Households incur a number of transaction costs when choosing stores to make grocery purchases. When the online channel is available as an alternative to physicalstore shopping, they may need to incur additional transaction costs. In this paper, we empirically quantify relative transaction costs when households choose between the online and offline channels of the same grocery chain. A key challenge to quantifying these costs is that several of them, such as picking items from the store & carrying them home, depend upon the items the household expects to buy in the store; and unobserved factors that influence channel choice also likely influence the items bought. Our econometric specification for channel choice accounts for observed and unobserved household heterogeneity, and the endogeneity of the items bought via the “plausibly exogenous” approach in an HB framework. We find the average value to a household of avoiding 1km of travel and ordering online instead is €.47; the relative value of shopping online vis-á-vis offline on a weekday (in bad weather) compared to a weekend (good weather day) is €1.26 (€1.06). For every 10 items bought, the time savings in the online channel over the offline channel are equivalent to €.55; the costs of picking and putting 10 heavy/ bulky items into the shopping cart are €.42; and the costs of carrying 10 heavy/ bulky items 1km are about €.83. On their online visits, households value the net transaction costs avoided by online shopping at €10.92, which exceeds the retailer’s delivery fees. We find considerable heterogeneity in these costs across households and characterize their distributions. We discuss the implications of our findings for the retailer in terms of product offerings, promotion and positioning for the two channels. ■ FB13 Champions Center III Marketing Finance Interface II Contributed Session Chair: Michael Sorell, Research Associate, IMD, Chemin de Bellerive 23, P.O. Box 915, Lausanne, Switzerland, [email protected] 1 - Preventing Raised Voices from Echoing: Advertising as Response to Shareholder Activism Simone Wies, Maastricht University, Tongersestraat 53, Maastricht, 6211 LM, Netherlands, [email protected], Arvid O. I. Hoffmann, Jaakko Aspara, Joost M. E. Pennings Research in marketing as well as finance shows that advertising can positively influence both consumer and investor behavior. However, the inverse logic of investor behavior impacting firms’ advertising expenditures has not yet been studied to date. We examine the extent to which firms exploit spillover mechanisms between consumer and financial markets and decide on product advertising as a strategic marketing instrument targeted at both consumers and investors. In doing so, we focus on an important aspect of investor behavior that governs investment practice but has so far been neglected in marketing research: shareholder activism. We rely on a unique database on shareholder activism proposals in the United States for the period from 1997 until 2009, and show that firms exposed to increased shareholder activism boost their advertising spending in the subsequent year. Individually examining the proposal data rules out the alternative explanation of shareholders explicitly demanding a change in advertising expenditure. Instead, managers appear to strategically use advertising as a reputation management tool to influence stakeholder perceptions. In addition, we find that this relationship is negatively moderated by the firm’s recent stock market performance, implying that firms intensify their advertising reaction to shareholder activism when they are under increased financial market pressure. Overall, this study’s results offer new insights into the marketing-finance interface, and in particular the strategic value of marketbased assets like advertising. 2 - The Effect of Banner Exposures on Memory for Established Brands Titah Yudhistira, University of Groningen, Department of Marketing, Nettelbosje 2, Groningen, 9747AE, Netherlands, [email protected], Eelko Huizingh, Tammo Bijmolt Although the focus of banner advertising has moved from click-through to brand building, there is little evidence of the effectiveness of banner advertising on brand memory for established brands. The purpose of our study is to provide new evidence on how banner exposures affect ad and brand memory as well as to propose and test a mechanism on how online exposures benefit established brands. Our study is unique since we (i) focus on established brands, (ii) emphasize the difference between ad memory and brand memory, and (iii) study the differential effects of three exposure variables, i.e frequency of exposures, time difference between the last exposure and measurement, and average time between exposures (spacing). We base our analyses on massive empirical data, i.e surveys and exposure data from 59.370 individuals in ten countries that were exposed to actual internet campaigns of 26 well-known brands. We find that exposure frequency and spacing have a beneficial effect and that time difference between the last exposure and measurement has a detrimental effect on ad memory. On the contrary, no effect of the three exposure variables on brand memory is found. However, the same effects of exposure frequency and time difference on brand memory are found by including mediating effects of ad recall and recognition. As a practical implication, this study shows that established brands can benefit from banner advertising by advertising more continuously, while assuring enough space between subsequent exposures. 50 MARKETING SCIENCE CONFERENCE – 2011 FB14 2 - The Impact of Consulting on Buying Behavior - The Case of Attention Behavior Nicolas Bourbonus, Frankfurt School of Finance & Management, Sonnemannstrafle 9-11, Frankfurt am Main, 60314, Germany, [email protected], Dominik Georgi, Olaf Stotz ■ FB14 For many customers, the service of consulting is a main criterion for selecting the right bank and the optimization of its investment decisions. By employing a consulting service, private investors hope to be able to make decisions on a more rational basis and to cut down on behavioral inefficiencies, such as ‘attention’. The attention simplifies the buying behavior of the investor such that the investor doesn’t have to select the right share from thousands, but rather takes a targeted look at shares for purchases that have earned his attention shortly before the purchase. For the shares that have earned his attention, the investor then selects those that are fitting for him. From an economical standpoint, these are attention-conducted buying decisions for the investor and are not generally of advantage and have often a negative effect on the performance. The goal of this study is to enhance the previous state of research systematically by first examining for which of the three private investors groups (“without consulting influence,” “mid-range consulting influence” and “strong consulting influence”) the behavior inefficiency attention occurs and by then determining for which of these three investor groups the attention behavior is especially strong. We empirically examine our hypotheses with regressions. Our examination is based on customer and transaction and market data. The research contribution of the study is as follows: The study confirms the influence of attention for the three investor groups “without consulting influence,” “mid-range consulting influence” and “strong consulting influence”. The study present that the attention behavior increases as the consulting influence increases. Chair: R. Mohan Pisharodi, Associate Professor of Marketing, Oakland University, 414 Elliott Hall, School of Business Administration, Rochester, MI, 48309, United States of America, [email protected] 1 - Service Refund as a Price Discrimination Mechanism Zelin Zhang, PhD, University of Kansas, Lawrence, KS, 66045, United States of America, [email protected], Weishi Lim Champions Center VI Pricing Research Contributed Session The refund policy is commonly used in the service industry (such as airline, hotel and ticket reservation) where the service provider adopts advance selling strategy. Consumers can reserve the service during the advance selling period or make a purchase during the spot selling period. For the consumers who reserve the service in advance but finally cannot redeem it, the service provider may offer them an opportunity to return the service by paying a refund charge. In this paper, we show that in addition to the commonly perceived function of the return policy as a profit driver, the return policy also serves as a price discrimination mechanism. More specifically, we show that when there is no capacity constraint, an appropriately chosen refund policy can help the seller maximize the profit by serving the dual functions as a profit driver and as a price discrimination mechanism. However, as the seller faces a more rigid capacity constraint, the role of the refund policy as a price discrimination mechanism diminishes. 2 - Determinants of Gain and Loss Parameters in Store-level Data: A Cross-category Analysis Sebastian Oetzel, Goethe-University Frankfurt, Grüneburgplatz 1, Frankfurt, 60323, Germany, [email protected], Daniel Klapper 3 - Wedded Bliss or Tainted Love?: Stock Market Reactions to the Introduction of Co-branded Products Zixia(Summer) Cao, Doctoral Student, Texas A&M University, 220 Wehner Building, Department of Marketing, College Station, TX, 77840-4112, United States of America, [email protected], Alina Sorescu Reference price models have experienced broad empirical support, and researchers have proposed many methods to infer the unobserved reference price. Their results indicate that demand for a brand does not only depend on the brand’s price, but also depends on the consumers’ behavior if they do more respond to losses (reference price < price) or to gains (reference price > price). Typically the responses to gains and losses are asymmetric. Under certain conditions researchers have found out, if the effect of gains is higher than that of corresponding losses, the optimal pricing policy is cyclical. In contrast, if the gain parameter is less than or equal to the loss parameter, a constant price is optimal. Furthermore empirical results indicate that gain and loss parameters seem to differ across brands, product categories, and retail outlets. However, there is little research which examines the factors associated with these observed differences. Using weekly store-level scanner data representing 34 product categories, the authors estimate store-specific gain and loss effects on the basis of a structural sales response model and relate these parameters to a broad set of category characteristics and store environment moderators (e. g. promotion depth and frequency, number of brand varieties or private-label share). The results add additional insights into the nature and structure of reference price effects and also offer guidelines to retail and brand managers for the planning and evaluation of optimal pricing policies. Co-branding, or the practice of using two established brand names on the same product, is a commonly used marketing tool. However, there is little, if any evidence in academic research supporting the view that co-branded products are wise investments for their parent firms. Whether financial rewards accrue to the manufacturer of co-branded products (e.g., the primary brand parent) or to the partner firm that lends its brand to the co-branded product (e.g., the secondary brand parent), and how these rewards may differ depending on the characteristics of the cobranded product itself are yet unanswered questions. Using a large dataset of cobranded products from the consumer packaged goods industry, we find that announcements of co-branding products are indeed greeted, on average, with positive abnormal returns, above and beyond what the same firms garner from non cobranded innovations, but the magnitude of these returns does vary by the position of the firm in the alliance as well as the characteristics of the co-branded products. We also find that in the short term, the manufacturer of the primary brand gains significantly more from endorsement than from other types of co-branding, but the parent firm of the secondary brand gains most from composite branding. Both partner firms benefit from prior co-branding experience and co-branding with complementary brands. In contrast, innovative co-branded products and products relying on exclusive provisions of secondary brands only benefit the parent firms of the primary brand. The insights from this research offer clear and actionable managerial guidelines for selecting co-branding partnerships that can lead to positive financial returns in both the short and long term. 3 - The Timing and Speed of New Product Price Landings Carlos Hernandez Mireles, Erasmus University, Burg. Oudlaan 50, Rotterdam, Netherlands, [email protected], Dennis Fok, Philip Hans Franses Many high-tech products and durable goods exhibit exactly one significant price cut some time after their launch. We call this sudden transition from high to low prices the price landing. In this chapter we present a new model that describes two important features of price landings: their timing and their speed. Prior literature suggests that prices might be driven by sales, product line pricing, competitor’s sales or simply by time. We propose a model using mixture components that identifies which of these explanations is the most likely trigger of price landings. We define triggers as thresholds after which prices are significantly cut. In addition, price landings might differ across products and therefore we model their heterogeneity with a hierarchical structure that depends mainly on firm, product type and seasonal effects. We estimate our model parameters applying Bayesian methodology and we use a rich data containing the sales and prices of 1195 newly released video games. In contrast with previous literature, we find that competition and time itself are the main triggers of price landings while past sales and product line are less likely triggers. Moreover, we find substantial heterogeneity in the timing and speed of price landing across firms and product types. 4 - The Value of a Global Brand: Is Perception Reality? Michael Sorell, Research Associate, IMD, Chemin de Bellerive 23, P.O.Box 915, Lausanne, Switzerland, [email protected], Arturo Bris, Willem Smit The Global Brand possesses an aura of excellence and is perceived as superior in the eyes of many consumers. Does the Global Brand uphold that promise in financial terms to firms as well? This paper analyzes the operating, financial, and market performance of firms included in Interbrand’s 100 Global Brands during the period 2000-2008. The market valuation of intangibles – in particular of brands – has been extensively studied in the literature. These intangibles (brand, intellectual property, employee satisfaction) are inherently valuable. However, only certain firms in an industry implement a Global Brand strategy, so our hypothesis is that any advantage of such strategy has to be eliminated in equilibrium. Our results, based on pairs of Global Brand firms matched with non-global brand peers, confirm that Global Brands do not earn significantly higher stock returns. They have larger marketing and R&D expenses. However, their EBIT margins are overall higher, suggesting that global brand firms are priced higher because of their better acceptance among consumers. Additionally, Global Brands sell more per unit of capital (asset turnover), thus resulting in significantly higher return on operating assets. Global Brands take on less debt than other firms because they base their performance on a highly valuable, yet intangible asset, and we indeed confirm that the market-to-book ratio of GB is significantly higher. However, we also show that GB do not display significant riskadjusted excess returns. 51 FB15 MARKETING SCIENCE CONFERENCE – 2011 4 - Price Pressure and Supplier Relations: Industry-Specific Findings R. Mohan Pisharodi, Associate Professor of Marketing, Oakland University, 414 Elliott Hall, School of Business Administration, Rochester, MI, 48309, United States of America, [email protected], Ravi Parameswaran, John Henke, Jr. why certain attributes may directly impact attitudinal loyalty (i.e. repurchase and/or recommendation intentions). Our empirical analysis in an intercontinental aviation setting demonstrates that direct effects of attribute performance on loyalty intentions are the rule rather than the exception. Moreover, we highlight the detrimental effects for resource allocation of failing to control for these direct effects. Lastly, we open avenues for future research by exploring additional possibly omitted mediators (Zhao et al. 2010). Original Equipment Manufacturers (OEMs) in a number of industries across the world are known to frequently follow the practice of using adversarial price reduction efforts to extract lower prices from their suppliers. Several other OEMs, while pursuing price reduction, have adopted more cooperative and less antagonistic approaches driven by the belief that adversarial price pressure on suppliers will be detrimental to good working relationships. This research probes the relationship between OEM price pressure on suppliers and the nature of OEM-supplier working relationships. A research model, founded on literature from multiple disciplines, was developed to examine the above relationship. The model has one outcome variable (Overall Relationship), two initial variables (Price Pressure and Other Pressures) and five mediating behavioral variables. Responses covering nine manufacturing industries were collected from a diverse sample of supplier respondents of global OEMs with procurement operations in North America, Asia, and Europe using an Internet-based survey questionnaire. The structural relations in the research model were analyzed through structural equation modeling using LISREL, with the complete data set as well as with industry-specific data sets. The results of statistical analysis reveal a consistent pattern of relationships with industry-specific variations. The overall pattern of relationships indicates that price pressure need not result in poor supplier-OEM relationships, and can exist along with good supplier-OEM relationships if the pressure is administered in a supportive manner. Inter-industry similarities and differences are assessed and their implications for research in marketing as well as for marketing management are discussed. 3 - How do E-Commerce Interfaces Affect Customer Satisfaction and Loyalty? Hsiu-Wen Liu, Assistant Professor, Soochow University, 56, Kueiyang St., Sec. 1, Taipei, Taiwan - ROC, [email protected], Yu-Li Lin This article presents an empirical test of user interfaces in the context of online retailer. The model posits that user interfaces lead to consumer satisfaction and loyalty. The sample includes 600 customer data. Results from the empirical test indicated that user interfaces affects customer satisfaction. Customer satisfaction affects customer loyalty. Further, customer satisfaction serves as a fully mediator of the effects of user interfaces and customer loyalty. Finally, theoretical, managerial and future research implications are included. 4 - Investigating Multipurpose Customers Radu Dimitriu, Lecturer in Strategic Marketing, Cranfield School of Management, 1 Wynyard Court, Oldbrook, Milton Keynes, MK6 2SZ, United Kingdom, [email protected], Fred Selnes Whereas the bulk of the sales of many companies comes from product categories that are typical of their business model and activity, considerable sales opportunities arise from offerings in ancillary categories. For instance, the last decades have seen food retailers extending their product range to include non-food items, such as electrical appliances, clothing, banking solutions or even petrol. We define those customers buying across categories as “multipurpose customers” (e.g., customers buying both food and non-food items from the same retailer). We analyzed the customer purchase data for an online bookshop specializing in selling academic books. We computed repurchase likelihood scores for the customers in the database based on an RFM approach. Controlling for purchase frequency, recency and amount spent, we found that being or not a multipurpose customer (i.e., buying both academic and nonacademic books such as fiction) was a strong predictor of customers’ repurchase likelihood. Multipurpose customers seem therefore to be extremely attractive. Several important questions arise however about multipurpose customers. First, is their higher repurchase likelihood based on affective commitment, or rather on calculative commitment or simply inertia? Second, would a company benefit from initiating campaigns to migrate the other (usually largest) part of their customer portfolio toward becoming multipurpose, or would such a marketing effort lead to a poor return on investment? Overall, our study documents the importance of multipurpose customers and presents a series of propositions meant to guide further research on the topic. ■ FB15 Champions Center V CRM III: Customer Loyalty Contributed Session Chair: Radu Dimitriu, Lecturer in Strategic Marketing, Cranfield School of Management, 1 Wynyard Court, Oldbrook, Milton Keynes, MK6 2SZ, United Kingdom, [email protected] 1 - Do Reward Programs Affect Consumer Behavior? Ricardo Montoya, University of Chile, Republica 701, Santiago, Chile, [email protected], Oded Netzer, Ran Kivetz Reward programs have become ubiquitous in the marketplace and a key tool companies use in the hope of shaping the behaviors of customers, salespeople, and employees. For example, many retailer reward programs attempt to motivate customers to visit the store more often and spend more during each visit. In the present research, we test a series of existing and new hypotheses regarding the effects of reward programs on consumer behavior. We analyze a large transactional dataset from a major retailer’s reward program; the dataset includes individual-level purchases and reward redemptions. We augment our modeling of this secondary dataset with controlled laboratory experiments. Among other predictions, we examine the “goal gradient” hypothesis, the “post-reward pause” hypothesis, and the notion that customers “earn the right to indulge” (in luxury rewards) by exerting more effort in the program. To the best of our knowledge, our research is the first to model transactional data from a large retailer reward program in order to test a diverse set of behavioral hypotheses. For example, we find that as customers approach the program’s reward goals, they accelerate the rate at which they purchase in the chain’s stores (i.e., a goal gradient effect). We also observe that customers who need to exert greater effort to reach a reward threshold are more likely to redeem a luxury reward. Customers also exhibit a “post-reward pause,” whereby after redeeming a reward they temporarily reduce their purchase frequency. Our study of these and other phenomena leverages the richness of the secondary transactional dataset available from this retailer’s reward program. We empirically model these data to better understand the effects of rewards programs on consumer behavior. Friday, 1:30pm - 3:00pm ■ FC01 Legends Ballroom I Choice V: Empirical Results Contributed Session Chair: Christian Schlereth, Goethe University Frankfurt, Grueneburgplatz 1, Frankfurt, 60323, Germany, [email protected] 1 - Measuring Scale Attraction Effects in Charitable Donations: An Application to Optimal ‘Laddering’ Kee Yeun Lee, Doctoral Student, University of Michigan, 620 Hidden Valley Club #201, Ann Arbor, MI, 48104, United States of America, [email protected], Fred M. Feinberg 2 - Shortcuts to Glory? Exploring When and Why Attribute Performance Can Directly Drive Loyalty Johannes Boegershausen, Grenoble Ecole de Management, 12 rue Pierre Sèmard, Grenoble, France, [email protected], Christophe Haon, Daniel Ray When seeking donations, charities nearly universally use an appeals scale: a set of specific monetary values from which potential donors can choose. However, little is known about how to appropriately select the scale points themselves, which are intended to serve as referents or anchors. Choosing them well is crucial: if charities select very high scale points (e.g., to try to increase donation amounts), they may risk alienating donors and receiving nothing; low scale points may encourage more people to donate, but less overall from each. Using unique data from a 3.5 year quasiexperiment, we employ a Tobit-II type model to account for both donation incidence and frequency. The model allows for tests of several distinct (latent) reference-price operationalizations, as well as (heterogeneous) pulling-up and pulling-down scale point attraction effects. Results show that the appeals scale really matters: even though donors can give what they wish (or not at all), the scale impacts both the “whether” and the “how much?” of donations. We find a strong negative correlation between donation incidence and amount, suggesting that asking for too much might raise average donation, but at the cost of lowering the proportion who do donate. Intriguingly, although we did not find significant heterogeneity on when people give (seasonality), scale attraction effect strength does vary across donors. This variation provides tangible information to help charities with “laddering”: deciding how much to increase the amount requested of individual donors based on past history. Providing customers with high satisfaction has been advocated as one of the primary means to enhance their loyalty intentions (Johnson et al. 2006; Gupta and Zeithaml 2006). In order to achieve high levels of overall customer satisfaction, many firms invest substantial resources into enhancing performance on the key service attributes. Over the last decade, there has been a substantial interest in chain frameworks such as the satisfaction-profit chain (Anderson and Mittal 2000), which in essence links attribute performance, customer satisfaction, customer retention, and profit. Surprisingly, despite numerous investigations and extensions of this and related chain frameworks, the occurrence and consequences of a direct effect of attribute performance on loyalty intentions has been largely neglected. Yet, several studies (e.g., Mittal et al. 1998; Kumar 2002; Lariviére 2008) report such unexpected direct effects. However, a critical re-assessment of the (full) mediating role of customer satisfaction in the attribute performance – loyalty intentions relationship is nonexistent. We address this void by drawing from the multi-attribute model, postpurchase thought, and service quality literature to provide a theoretical explanation 52 MARKETING SCIENCE CONFERENCE – 2011 2 - Data or Structure? Using a Field Experiment to Assess the Determinants of Counterfactual Demand Predictive Performance Manuel Hermosilla, PhD Student, Kellogg School of Managment, Northwestern University, 2001 Sheridan Road, Room 474, Evanston, IL, 60208, United States of America, [email protected], Yi Qian, Eric Anderson FC02 emotional content toward negative extreme ratings, falling into a J-shaped distribution (self-selection bias). Lastly, we find that uncertainty related to product quality before consumption is associated with more emotional words across different product categories (uncertainty bias). However, in this context the amount of positive emotional content is higher than that of negative content, suggesting a J-shaped distribution of emotional content. Furthermore, we find that search goods are associated with the lowest amount of expressed emotion, followed by experience goods, and then credence goods. Our findings enhance our understanding of the motivation behind WOM and related consumer behavior in the context of product sales. We evaluate the roles of microeconomic/statistical structure and experimental data as determinants of counterfactual demand predictive performance. The specific structures under evaluation are those developed by Berry (1994) and Berry et al. (1995). Experimental demand data is obtained from a large-scale field experiment that specified several price conditions for various products of a retail category. Counterfactual demand scenarios arose in some of these conditions because experimental prices substantially departed from those observed in a subsample with non-experimental historical data. By re-estimating structural and structure-free econometric models with varying amounts of the experimental data, we are able to isolate the contribution of each source of identification on counterfactual demand predictive performance. Our results show that both experimental data and microeconomic/statistical assumptions (as given by the considered models), provide advantages in counterfactual demand prediction. These results are robust to different measures, and in general, suggest that the key driver of predictive performance in counterfactual demand scenarios is the quality of the data used for estimation. Results thus suggest that (i) if the goal is to generate predictions of counterfactual demand scenarios, it might not be worth the while to undergo the costly process of specification and estimation of a structural econometric model of demand, and (ii) if the goal is to understand the fundamentals of consumers’ choice, parameter estimates of a structural demand model can be reliably used for counterfactual prediction. 2 - Ad Revenue and Content Commercialization: Evidence from Blogs Monic Sun, Assistant Professor, Stanford University, 518 Memorial Way, Stanford, CA, 94305, United States of America, [email protected], Feng Zhu Many scholars are concerned about the impact of ad-sponsored business models on content providers. They argue that content providers, when incentivized by ad revenue, are more likely to tailor their content to attract “eyeballs,” and as a result, popular content may be excessively supplied. We empirically test this prediction by taking advantage of the launch of an ad revenue-sharing program initiated by a major Chinese portal site in September 2007. Participating bloggers allow the site to run ads on their blogs and receive 50% of the revenue generated by these ads. After analyzing 4.4 million blog posts, we find that compared to nonparticipants, the percentage of popular content increases by about 13% on the participants’ blogs after the program takes effect. More than 50% of this increase can be attributed to topics shifting towards three domains: stock market, salacious content, and celebrities. We also find evidence that, relative to nonparticipants, the participants’ content quality increases after the program takes effect. 3 - Estimation of Willingness to Pay Intervals by Discrete Choice Experiments Christian Schlereth, Goethe University Frankfurt, Grueneburgplatz 1, Frankfurt, 60323, Germany, [email protected], Christine Eckert, Bernd Skiera 3 - Does Advertising Affect Chatter? - Assessing the Dynamics of Advertising on Online Word-of-mouth Seshadri Tirunillai, University of Southern California, Los Angeles, CA, United States of America, [email protected], Gerard J. Tellis Knowledge about consumers’ willingness to pay (WTP) is essential for a profitmaximizing price management. This willingness to pay has always been regarded as a point estimate, typically as the price that makes the consumer indifferent between buying and not buying the product. In contrast, this paper uses discrete choice experiments and a scale adjusted latent class model to estimate willingness-to-pay as an interval. The mid value of this interval corresponds to the traditional WTP point estimate and depends on the deterministic utility, while the range of the interval is influenced by the price sensitivity as well as the error variance (scale) that determines the random utility for a product. This error variance, i.e. the degree of uncertainty in consumers’ choices has strong implications for firms competing in the market. Those firms with more fa-vorable products should target consumers with low scale (high certainty), while the other firms should target those with high scale (low certainty). The results of our empirical study demonstrate that such knowledge about individual intervals of willingness to pay enables better segmenting customers. The results further show that the sizes of the will-ingness-to-pay intervals can be large and that ignoring these sizes may lead to non-optimal pricing decisions. Despite the increased importance of consumer media, the factors influencing the User-Generated Content (UGC) have seen limited research. In this study, we investigate the effect of corporate advertising on UGC using a natural experiment in a time series setting. We seek to answer the following questions: 1) If there does exist a relation, can we establish the direction of causality? 2) Among the various metrics of UGC, which metrics are influenced by advertising and how are they affected? 3) What are the dynamics of such a relationship in terms of growth, persistence and decays? We use big budget corporate advertising campaigns to assess the impact of the campaign on the different metrics of UGC (e.g. overall volume, negative and positive UGC) of the target firm and the competitors. We try to assess the influence of advertising on firms by assessing the differential impact on the UGC metrics. We also analyze the dynamics of advertisement on the metrics UGC using multivariate time series. We find that after the introduction of the advertising campaign, the chatter of the target firm increased by about 27% relative to the control firms that did not undertake any major brand campaign during this period. While we find that the positive chatter increases markedly as compared to the synthetic control, there was no systematic decrease in negative chatter during the time period. There is also a spillover of advertising across firms in a market. ■ FC02 4 - Social Influence in the Evolution of Online Ratings of Service Firms Raji Srinivasan, Associate Professor, University of Texas-Austin, 1 University Station, Austin, TX, 78712, United States of America, [email protected] Legends Ballroom II UGC-III (Content and Impact) Cluster: Internet and Interactive Marketing Invited Session Online consumer review websites prominently display consumers’ online ratings of service firms, which influence other consumers’ purchase decisions. Yet there are few insights on the factors influencing online ratings of service firms. The authors develop hypotheses of how other consumers’ online ratings of the service firm moderate the effects of a reviewer’s service encounter characteristics – valence of the service encounter, occurrence of service failure, and service recovery effort – on the reviewer’s online rating of the service firm. They test the hypotheses using an ordered probit model with data from 7,499 online reviews of hotels in Boston and Honolulu between 2006 and 2010. The results support the hypotheses of the moderating effects of social influence on a reviewer’s online rating of the service firm. The authors decompose the effects of service encounter characteristics on online ratings of service firms into ‘service encounter’ and ‘social influence’ components, a novel contribution to the marketing literature in services. From a managerial perspective, when a service failure occurs in a service firm with a high online rating, the decrease in the firm’s long-term rating is more than when it has a low online rating. The opposite is true when a service recovery effort is attempted. Chair: Raji Srinivasan, Associate Professor, University of Texas-Austin, 1 University Station, Austin, TX, 78712, United States of America, [email protected] 1 - Bimodal Distribution of Emotional Content in Customer Reviews: Emotional Biases in Online Customer Reviews Wonjoon Kim, Associate Professor, KAIST, Guseong-dong, Yuseonggu, Daejeon, 305701, Korea, Republic of, [email protected] Word-of-mouth (WOM), most visibly encountered in the form of online customer reviews in recent times, has received considerable attention of late by academics and practitioners alike. While a number of studies have examined the phenomenon’s frequency and distribution patterns to understand its characteristics and the motivation behind it, WOM contents across its distribution have remained underexplored. To fill this important gap in the literature, we analyzed WOM contents using Natural Language Processing (NLP) methods, which are used to study the intersection of computers and human language. Using this approach, we find that more extreme reviews have a greater proportion of emotional content than less extreme reviews, revealing a bimodal distribution of emotional content as in the case of WOM distribution; we refer to this as emotional bias. Second, we find that reviews have more positive emotional content toward positive extreme ratings than negative 53 FC03 MARKETING SCIENCE CONFERENCE – 2011 ■ FC03 ■ FC04 Legends Ballroom III Legends Ballroom V Analytic Models of Online Behavior Structural Models I Cluster: Internet and Interactive Marketing Invited Session Contributed Session Chair: Wenbo Wang, PhD Candidate, New York University, 40 West 4th Street, Room 921, New York, NY, 10012, United States of America, [email protected] 1 - A Dynamic Model of Consumers’ Optimal Default on Financial Products: A Case of Subprime Mortgages Minjung Park, Assistant Professor, University of California-Berkeley, Haas, 545 Student Services Bldg #1900, Berkeley, CA, 94720, United States of America, [email protected], Patrick Bajari, Sean Chu, Denis Nekipelov Chair: J. Miguel Villas-Boas, University of California-Berkeley, Berkeley-Haas, Berkeley, CA, 94720-1900, United States of America, [email protected] 1 - The Interplay between Sponsored Search and Display Advertising Kannan Srinivasan, Professor, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected], Kinshuk Jerath, Amin Sayedi An important decision for a firm is how to allocate its online advertising budget between sponsored search advertising and display advertising. An advantage of sponsored search advertising is that, since the firm advertises in response to a consumer-initiated search, the sales-conversion rate is typically higher than in display advertising. However, a disadvantage of sponsored search advertising is that, to “steal” the firm’s customers, its competitors can also bid on its keyword, therefore driving the advertising costs higher. Using a game theory model, we study the implications of these tradeoffs on the advertising decisions of competing firms, and on the design of the sponsored search auction by the search engine. We find that symmetric firms may follow asymmetric budget allocation and bidding strategies. Moreover, the search engine benefits from discouraging competition in sponsored search bidding by shielding firms from competitors’ bids. This explains the practice of employing “keyword relevance” measures, under which search engines such as Google, Yahoo! and Bing under-weight the bids of firms bidding on competitors’ keywords. We also obtain various other interesting insights on the interplay between sponsored search advertising and display advertising. The increase in defaults in the subprime mortgage market is widely held to be one of the causes behind the recent financial turmoil. Key issues of policy concern include identifying the main drivers behind the wave of defaults and predicting the effects of various policy instruments designed to mitigate default. To address these questions, we estimate and simulate a dynamic structural model of subprime borrowers’ default behavior. In each period, a borrower takes one of three possible actions: defaulting, prepaying, or continuing to make regularly scheduled payments. Dynamics are generated by the fact that defaulting results in an immediate payoff as well as the loss of future option value from keeping the mortgage alive into the future. Key determinants of optimal borrower behavior include the level of net equity as well as expectations about future home prices. Our model also allows for defaults to be driven by binding credit constraints. We use our model to simulate how borrowers’ default decisions would change under various counterfactual scenarios. The simulation exercises allow us to quantify the importance of home price declines versus loosened underwriting standards in explaining the recent increase in subprime defaults. Furthermore, we use simulations to assess the effects of foreclosure mitigation policies, such as principal write-downs, on the behavior of various subsets of borrowers. 2 - Optimal “Last-minute” Selling by a Monopolist Facing Forwardlooking and Risk Averse Consumers Ori Marom, Erasmus University, Burgemeester Oudlaan 50, Rotterdam, 3062 PA, Netherlands, [email protected], Abraham Seidmann 2 - Modeling Consumer Learning of Attribute-specific Preferences Jihong Min, PhD Student, Purdue University, Rm 460 Krannert Building, W. State Street, West Lafayette, IN, 47909-2065, United States of America, [email protected], Subramanian Balachander We characterize profit-maximizing strategies for a monopolist who sells to risk-averse buyers in two time periods and may deliberately randomize the occurrence of clearance sales. Our theoretical investigation shows that such a randomized strategy is optimal whenever the seller’s capacity exceeds a threshold value that is a decreasing function of the degree of the buyers’ risk-aversion. We show that with no loss of optimality the seller can restrict herself to strategies with a deterministic “sale” price. Contrary to intuition, we find that an increase in the buyers’ risk-aversion shifts supply toward randomized “last-minute” selling. Most extant consumer learning models allow for learning about alternative-specific preference such as brand preference in a brand choice model. However, there exist product categories which offer many varieties of products with different attributes. In such cases, it is important to model consumer learning of attribute-specific preferences rather than simply modeling alternative-specific preferences in order to more accurately and parsimoniously describe consumer behavior. In this paper, we propose a structural model of consumer’s Bayesian learning of attribute-specific preference, using scanner panel data. We show that the proposed model and empirical results can enable marketing researchers to identify whether a certain attribute is characterized by significant consumer learning, and to capture the possible separate learning processes of preferences for multiple attributes. Moreover, the model results provide a snapshot of the segmentation of consumers’ learning of different attributes and this snapshot can provide insights into market structure. We conduct policy evaluations such as the introduction of a new flavor for a brand and obtain interesting implications for managers. 3 - Sampling Paid Content Florian Stahl, Assistant Professor, University of Zurich, Plattenstrasse 34, Zurich, 8032, Switzerland, [email protected], Don Lehmann, Oded Koenigsberg, Daniel Halbheer This paper studies profit-maximizing sampling and pricing of paid content for online news publishers. The key feature of our model is the twofold role of free samples which allows publishers to generate advertising revenues and simultaneously disclose editorial quality to potential subscribers. Taking customers’ prior beliefs about article qualities into account and employing Bayesian updating, we derive the subscription demand and characterize the optimal number of articles offered for free as well as the subscription price for the content behind the paywall. Considering two cases where free sampling aims to persuade either all consumers to subscribe or only those with the highest willingness-to-pay, we find that is optimal for the publisher to offer a larger number of free samples when consumers underestimate product quality. 3 - Studying the Switching Behavior of Electricians: Assessing the Impact of a Loyalty Program Madhu Viswanathan, University of Minnesota, 321 Nineteenth Avenue South, Suite 3-150, Minneapolis, MN, 55455, United States of America, [email protected], Ranjan Banerjee, Om Narasimhan Despite the popularity of reward programs, the ability of these programs to create brand loyalty has been questionable. Existing empirical work has found mixed results and focused largely on a single “lock-in” parameter that rationalizes the observed response to a reward (Lewis 2004, Rossi 2008, Hartmann and Viard 2008). In this paper, we identify and estimate the impact of reward programs when these programs create both financial and non-financial costs for the consumer. By financial costs, we refer to all those costs that involve the loss of financially quantifiable resources, like not meeting the target for a reward program. Non-financial costs could be procedural (related to learning about the features and appropriate usage) or emotional (psychological or emotional discomfort due to loss of identity/breaking of bonds). We estimate these two kinds of costs by using a unique dataset that tracks electricians’ purchase behavior during a reward program offered by a switchgear manufacturer in India. In general, estimation of such costs is made difficult by the need to account for forward-looking behavior of consumers, and separating the “lock-in” effect from consumer heterogeneity. Accordingly, we develop and estimate a structural dynamic model of consumer choice and incidence across different consumer segments. We find that behavior differs across segments based on the intensity of usage; financial rewards are the main drivers of purchase behavior for frequent and infrequent consumers of the brand, while both non-financial and financial rewards matter for average users of the brand. 4 - Optimal Search for Product Information J. Miguel Villas-Boas, University of California-Berkeley, Berkeley-Haas, Berkeley, CA, 94720-1900, United States of America, [email protected], Monic Sun, Fernando Branco Consumers often need to search for product information before making purchase decisions. We consider a parsimonious model in which consumers incur search costs to learn further product information, and update their expected utility of the product at each search occasion. We characterize the optimal stopping rules to either purchase, or not purchase, as a function of search costs and the informativeness of each attribute. The paper also characterizes how the likelihood of purchase changes with the ex-ante expected utility, search costs, and informativeness of each attribute. We discuss optimal pricing, the impact of consumer search on profits and social welfare, and how the seller chooses its price to strategically affect the extent of the consumers’ search behavior. We show that lower search costs can hurt the consumer because the seller may choose to then charge higher prices. The paper also considers the impact of searching for signals of the value of the product, of discounting, and of endogenizing the intensity of search. 54 MARKETING SCIENCE CONFERENCE – 2011 4 - Green Lifestyle Adoption: Shopping Without Plastic Bags Wenbo Wang, PhD Candidate, New York University, 40 West 4th Street, Room 921, New York, NY, 10012, United States of America, [email protected], Yuxin Chen ■ FC06 This research develops a structural model of consumer lifestyle adoption. The model captures two behavioral features —- forward looking and accessibility contingency — - that are shared by many lifestyles. Lifestyle adoption often involves a deliberate trade-off between long-term benefit and extra effort at the initial stage, and therefore motivates consumers to look forward in adoption decisions. Moreover, the forward looking adoption is contingent on accessibility of the optimal lifestyle choice. For instance, even though one intends to lives a healthy lifestyle, he may accidentally forget to bring sneakers from time to time. In this case, the gym exercise lifestyle choice is not accessible to him at the decision moment. Inaccessibility can occur due to, for example, memory limitations or cognitive constraints. We estimate the model with a panel data of a green lifestyle adoption of urban shoppers under the plastic bag ban in China. The ban is lifestyle-changing to the consumers because they now have to either bring own shopping bags or pay for plastic bags. This research sheds light on the following questions: 1) How does accessibility contingency interplay with forward looking in the green lifestyle adoption? 2) What is the role of state dependence on the lifestyle adoption? 3) How long does it take a shopper to stabilize on the new green lifestyle and where is the no-return point for this lifestyle? 4) How can marketing mix decisions facilitate the adoption? Contributed Session FC06 Legends Ballroom VII Channels III: Competition Chair: Jaime Romero, Associate Professor, University Autonoma Madrid, Fac. CC. Económicas, Avda. Tomas y Valiente, 5, Madrid, 28049, Spain, [email protected] 1 - When and How Do Coordinating Contracts Improve Channel Efficiency? Ernan Haruvy, Associate Professor, Universiy of Texas at Dallas, 800 West Campbell Rd, Richardson, TX, 75080, United States of America, [email protected] A growing literature shows that coordinating contracts may not improve efficiency in the laboratory to the extent prescribed by theory. We show that this result is largely due to offer rejection where the bargaining procedure is structured as an ultimatum offer, which is the only structure studied in the lab hitherto in relation to coordinating contracts. We show that a less restrictive procedure does not involve this feature and allows coordinating contracts to coordinate. Specifically, we look at three contract formats – wholesale price, two-part-tariff and minimum order quantity. The wholesale price leads to loss of firm surplus because of double marginalization. The other two contracts – the coordinating contracts – allow the manufacturer to coordinate the channel, either by pricing at cost and extracting surplus through a lump sum payment (two part tariff) or through announcing a minimum order quantity which is equal to the efficient quantity and extracting the surplus through price (minimum order quantity contract). Even though for fully rational players these two coordinating contracts are equivalent, these two contracts are different under mild bounded rationality assumptions. Proposals in the minimum order quantity treatment are far more efficient than two-part-tariff proposals in terms of the overall surplus they imply. But in the ultimatum context such efficient proposals tend to get rejected, leading to lower ex-post efficiency. With structured negotiations bargaining, however, rejection rate drastically falls leading to a more direct relationship between proposal efficiency and ex-post efficiency. ■ FC05 Legends Ballroom VI Game Theory II: Market Entry Contributed Session Chair: Matthew Selove, Assistant Professor of Marketing, USC Marshall School of Business, 3660 Trousdale Parkway, ACC 306E, Los Angeles, CA, 90089, United States of America, [email protected] 1 - Cross-market Experience and Market Entry Dai Yao, PhD Student, INSEAD, PhD Room (2nd floor), 1 Ayer Rajah Avenue, INSEAD, Singapore, 138676, Singapore, [email protected], Yakov Bart 2 - Channel Structure and Performance under Co-marketing Alliance Xiao Zuhui, The Hong Kong Polytechnic University, Room 409, New East Ocean Centre, 9 Science Museum Road, Honghum, Hong Kong, Hong Kong - PRC, [email protected], Liu Liming, Zhang Xubing Cross-market experience externality arises when a firm may choose to provide products or services in one market in order to gain experience that could benefit launching products or services in another market. Examples of industries with such externality include IT outsourcing market vs. IT consulting market and PC components market vs. PC market.Multi-market environments that exhibit such cross-market experience externalities are structurally different from industries in which multiple markets are related to each other through characteristics of goods. We propose a game-theoretic model toprovide explanations to different patterns in regards to optimal entry time for market entrants and optimal defending strategies for market incumbents in such multi-market environments.Our analysis characterizes the impact of cross-market experience externalities on firms’ optimal market entry decisions and competitive interactions. A growing number of companies are seeking cooperation to better serve their clients. Among various types of cooperative strategies, lateral relationships between firms at the same level in the value chain, are popular in both good and service markets. In these relations, downstream firms jointly market the products or provide value add service for upstream suppliers to increase profit and promote sales. Although introducing marketing effort from a third party has become more common in business life, this type of cooperation receives little continuing attention in the literature. Motivated by real business cases, we aim to provide a better understanding on how these cooperative activities create values for partnering firms and their common clients. We construct models for marketing and channel decisions of alliance partners under different scenarios. Our modeling analysis focuses on the following issues: (1) how the firms involved make their individual decisions in equilibrium; (2) how different cooperation structures influence the performance of firms involved; (3) how market conditions, i.e., demand uncertainty and asymmetric information, influence the firms’ performance. We characterize the different scenarios in these settings, providing conditions under which the cooperation structure will benefit the partnering firms. We show that, under certain market conditions, the partnering firms and even total channel may be better off under these cooperative activities. 2 - The Benefit of Increased Competition David Soberman, Canadian National Chair in Strategic Marketing, Rotman School of Management, University of Toronto, 105 St. George Street, Toronto, Ontario, Canada, [email protected], Amit Pazgal We consider a fundamental question regarding the need for incumbents to protect their turf and foreclose entry to their industry. This has been a fundamental issue of industrial organization for more than 50 years (Bain 1956) and numerous studies have demonstrated how barriers to entry are used to protect and increase profits. But are there times when a new entrant can be beneficial for incumbent firms? Sometimes late entrants learn from incumbents and are able to improve on existing offerings. However, late entrants are also known to offer stripped down versions of products that have reduced functionality. Using a spatial model, we demonstrate that a stripped down new entrant can lead to higher (and not lower) incumbent profits by capturing a quality insensitive/price sensitive segment of consumers that are the source of intense competition between incumbents. This can create a win-win situation for all competitors including the entrant. Our analysis identifies the precise conditions where this occurs and also conditions where incumbents should seek to blockade the potential entrant. 3 - Should be Close to or Away from Your Competitors? Store Location Choice by Gravity Model Wei-Jhih Yang, PhD Candidate, National Taiwan University, No.1, Sec. 4, Roosevelt Rd., Taipei City, Taiwan - ROC, [email protected], Jesheng Huang, Lichung Jen Agglomeration among competing firms in store location cluster may enlarge consumer demand, but may also create more brand competitive intensity in consumer’s mind spontaneously. From managerial perspective, it is hard to identify who made much more efforts and who should gain more if the economies of agglomeration exist; likewise, it’s also difficult to examine who would threat others and who won’t if the diseconomies of agglomeration exist. More specifically, for channel strategy, a firm’s store should be close to or away from the agglomeration of its competitors? And for brand positioning strategy, how a firm can get a good location in consumer’s mind to avoid the points of parity. Therefore, how to address this kind of dual competitive intensities is the focus of this article. Based on the concept of gravity model from physics, the authors propose a location gravity model to analyze why a firm should close to or away from its competitors. The authors firstly use consumers’ brand perceptual mapping to measure the cognitive distance of the competing brands in an agglomeration through MDS; then, the results of cognitive distance among competing brands will be applied to location gravity model, in order to calculate distribution intensity in the spatial cluster formed by a firm and the agglomeration of its competitors. Based on the proposed model which combining both considerations of dual competition intensities , the authors suggest how a firm can choose the optimal store location which may avoid the exiting competition from the cluster of store locations and get the best position in consumers’ mind. Finally the authors discuss the managerial implications of this model for the blue ocean strategy. 3 - A Dynamic Model of Competitive Entry Response Matthew Selove, Assistant Professor of Marketing, USC Marshall School of Business, 3660 Trousdale Parkway, ACC 306E, Los Angeles, CA, 90089, United States of America, [email protected] This paper develops a model in which competing firms invest in various business formats. For example, they could implement full service or “no-frills” airline models. I derive conditions in which each firm specializes in a single format, and conditions in which both firms adopt both formats. In some cases, increased difficulty of implementing a format increases the likelihood that firms attack each other by investing in each other’s format. This is because barriers to entry also serve as barriers to retaliation, allowing firms to attack without fear of swift punishment. 55 FC07 MARKETING SCIENCE CONFERENCE – 2011 3 - Promotional Reviews Yaniv Dover, Yale University, New Haven, CT, 06520, United States of America, [email protected], Dina Mayzlin 4 - Price Competition in the Spanish Nondurable Retail Industry Jaime Romero, Associate Professor, University Autonoma Madrid, Fac. CC. Economicas, Avda. Tomas y Valiente, 5, Madrid, 28049, Spain, [email protected], Daniel Klapper, Martin Natter In recent years, we have witnessed the proliferation of online consumer reviews of products and services. Researchers have studied the impact of reviews on sales as well as the dynamics of reviews that arise from social factors. While reviews are clearly popular and impactful, there have always been concerns about the authenticity of reviews since firms can manufacture positive reviews for their products (see Mayzlin 2006 and Delarocas 2006). Here we propose an empirical method for detecting the existence of manipulation of reviews that differs from current work (see, for example, Kornish 2009). We then apply this method to investigate under which conditions we expect to see the greatest amount of manipulation. In particular, we show that the amount of competition affects the rate of the manipulation behavior as well as the dynamics of the manipulation process. In this paper we develop a model of spatial competition between retailers. In particular we study price competition between a very large and representative sample of Spanish retailers. These retailers offer a wide variety of products including packed food, fresh meat, seafood, fruits and vegetables and drugstore articles. They all provide non durable products. As many retail markets across the world the Spanish retail market for food products has experienced the exit of small and the entry of large retail corporations. However there still exists a mixture of different store types, ranking from small independently and family owned operated stores to large chain stores. Based on the model of retail competition and an application of the difference in difference estimator we study the extent of price competition that is affected by different store types and regional effects. Regional effects are driven by local economic effects such as real estate prices or discretionary income and the density of competition between retailers which is measured e.g. by the distance between retailers or assortments. Our results show that the differences in prices across retailers and markets are strongly affected by regional effects. Local differentiation between retailers limits the competitive threat and allows smaller retailers to survive even in the presence of large and chain wide operated retail stores. 4 - Multichannel Word of Mouth: The Effect of Brand Characteristics Renana Peres, Assistant Professor, Hebrew University of Jerusalem, Jerusalem, Israel, [email protected], Ron Shachar Although word-of-mouth has been receiving an increasing attention, our understanding of it is still too one-dimensional and does not recognize the richness of the phenomenon. Specifically: (i) it is based on a single conversation channel; (ii) it is based almost exclusively on online measures, in spite of evidence suggesting that offline WOM might have a higher effect on consumption, and (iii) it does not account for the role of brands characteristics which is a central ingredient of marketing. The only way to address these concerns is by creating a comprehensive data set and analyzing it. We conducted a massive data search on 700 US brands from 16 different categories. For each of these brands we collected data on the perceived brand characteristics (collaborating with Decipher Inc. and the Brand Asset Valuator of Young and Rubicam), the offline word of mouth through face-to-face and phone conversations (from the Keller Fay Group), and the online word of mouth through blogs, user forums, and Twitter messages (using the Buzzmetrics tool of Nielsen online). Our preliminary analysis finds that while different online channels are similar to each other, online and offline WOM are quite different. Thus, online WOM cannot act as a good proxy for offline activities. We also find that brand characteristics play an important role in generating WOM and that their role is different between the online and offline channels. While newness of the brand, the brand equity and its visibility/observability are important for both online and offline WOM, the effect of the Aaker’s brand personality variables (excitement and competence) on WOM is significant for the online channels, but they are not significant for the offline channel. ■ FC07 Founders I New Directions in Word of Mouth Cluster: Special Sessions Invited Session Chair: Jonah Berger, University of Pennsylvania, PA, United States of America, [email protected] Co-Chair: Andrew Stephen, INSEAD, Fontainebleau, France, [email protected] 1 - How the Frequency and Pattern of Social Influence Over Time Shape Product Adoption Raghu Iyengar, University of Pennsylvania, 3730 Walnut Street, 700 Jon M Huntsman Hall, Philadelphia, PA, United States of America, [email protected], Jeffrey Cai, Jonah Berger ■ FC08 Social influence shapes diffusion and new product adoption. But how does the quantity of social influence (e.g., number of doses) and its distribution over time impact new product adoption? Does hearing about a product from multiple social contacts increase adoption, and if so, is it better to have such doses concentrated over a short period or spread out over time? We test whether multiple doses of social influence boost adoption, and further, whether influence decays over time and is increased by concentration. Analysis of the adoption of both a new website and hundreds of academic papers indicates that (1) people are more likely to adopt a product if they have received multiple doses of social influence. Importantly, however, (2) the impact of influence decays over time and (3) there is little evidence of concentration. While hearing about a product today has a stronger impact on behavior than a dose a month ago, the relationship between dose concentration and adoption is concave such that multiple doses in a short period has little added effect. That said, the relative importance of decay versus concentration suggests that concentrating two doses in a given period will increase adoption rather than spreading them out. Overall these results provide insight into the mechanisms behind new product adoption and have important managerial implications for getting new products to catch on. Founders II Dynamic Models in Marketing Cluster: Special Sessions Invited Session Chair: Paul Ellickson, University of Rochester, Simon School of Business, Rochester, NY, United States of America, [email protected] 1 - Learning About Entertainment Products: A Dynamic Consumer Decision Model with Learning About Changing Match-Values Mitch Lovett, University of Rochester, Simon School of Business, Rochester, NY, United States of America, [email protected], William Boulding, Richard Staelin How do consumers decide whether to continue watching a TV show week after week? We study consumer learning and decisions when engaging in an entertainment product repeatedly and seek to understand both how consumers use new information to learn about the product as well as what trades are most important in the decision to continue engaging. We develop a model for how consumers learn about products when the match-value for those products could be changing from one experience to the next. For entertainment products, the plot and characters may change with each new experience, potentially altering how much the consumer likes the show in a persistent way. We model this changing nature of the product and the way consumers adjust their beliefs about the product. These beliefs, in turn, are central to our explanation of how consumers decide whether to continue engaging in the product. To test and calibrate this theory, we both conduct a laboratory experiment and collected data in which consumers have multiple experiences with products. We collect measures of viewing behaviors, experienced liking, and subjective expectations for the next experience. We present an approach to estimate how each individual learns about the product as the product itself is changing and incorporate this estimation into a dynamic consumer choice model. Through this research, we extend understanding of how to model consumer learning as well as how consumers behave in repeated interactions with entertainment products. 2 - The Complementary Roles of Traditional and Social Media Publicity in Driving Marketing Performance Andrew Stephen, INSEAD, Fontainebleau, France, [email protected], Jeff Galak The media landscape has dramatically changed, with traditional media (e.g., newspapers, television) now supplemented by social media (e.g., blogs, online communities). This situation is not well understood with respect to the relative impacts of these media types on marketing performance (e.g., sales), and how they influence each other. These issues are examined using 14 months of daily count data for sales and media activity for a micro-lending website. Multivariate time series count data pose a number of statistical challenges, which are overcome using a copula-based multivariate autoregressive count model. The authors find that both traditional and social media affect sales, directly and indirectly through effects on each other. While the unit sales impact for traditional media is larger than for social media, the greater frequency of social media activity results in it having a comparable effect to traditional media in the case of blogs, and a larger effect in the case of online communities. Overall, the results emphasize the critical role that interactive, conversational online social media plays in driving sales. 56 MARKETING SCIENCE CONFERENCE – 2011 FC09 2 - Determining Consumers’ Discount Rates with Field Studies Song Yao, Northwestern University, Kellogg School of Management, Evanston, IL, United States of America, [email protected], Carl Mela, Jeongwen Chiang, Yuxin Chen ■ FC09 Because utility/profits, state transitions and discount rates are confounded in dynamic models, discount rates are typically fixed to estimate the other two factors. Yet these rate choices, if misspecified, generate poor forecasts and policy prescriptions. Using a field study wherein cellphone users transitioned from a linear to three-part tariff pricing plan, we estimate a dynamic structural model of minute usage and obtain discount factors that would normally be unidentifiable. The identification rests upon imputing the utility under linear pricing plans without dynamic structure; then using these utilities to identify discount rates when consumers were switched to a threepart tariff where dynamics became material. We find that the estimated segment-level weekly discount factors (0.86 and 0.91) are much lower than the value typically assumed in empirical research (0.995). When using a standard 0.995 discount rate, we find the price coefficients are underestimated by 23%. Moreover, the predicted intertemporal substitution pattern and demand elasticities are biased, leading to a 27% deterioration in model fit; and sub-optimal pricing recommendations that would lower potential revenue gains by 74-88%. Chair: Aharon Hibshoosh, Professor, San Jose State University and Lincoln University, One Washington Square, San Jose CA 95192, 401 15th Street, Oakland, CA 94612, Oakland, CA, 94612, United States of America, [email protected] 1 - Product Variety Decision: When Specialty Stores Meet with Big-box Retailers Jiong Sun, Assistant Professor, Illinois Institute of Technology, 565 W Adams St., Chicago, IL, 60661, United States of America, [email protected], Tao Chen Founders III Retailing IV: Competition Contributed Session Specialty stores carry more varieties in a particular category, while big-box retailers carry fewer varieties with price advantages. When consumers are unsure about their taste about different varieties, visiting specialty stores offers consumers an opportunity to inspect the varieties and resolve their uncertainty. Offering more varieties may encourage more low-valuation consumers to evaluate and buy the product from the specialty store, but may also intensify the price competition with the big box retailer which enjoys high cost efficiency. In this paper, we examine product variety decisions of specialty stores when facing competition from big-box retailers. We find that consumers’ transaction cost and valuation heterogeneity play an important role. 3 - Does AMD Spur Intel to Innovate More? Ronald Goettler, University of Chicago, Booth School of Business, Chicago, IL, United States of America, [email protected], Brett Gordon We propose and estimate a model of dynamic oligopoly with durable goods and endogenous innovation to examine the relationship between market structure and the evolution of quality. Firms make dynamic pricing and investment decisions while taking into account the dynamic behavior of consumers who anticipate the product improvements and price declines. The distribution of currently owned products is a state variable that affects current demand and evolves endogenously as consumers make replacement purchases. Our work extends the dynamic oligopoly framework of Ericson and Pakes~(1995) to incorporate durable goods. We propose an alternative approach to bounding the state space that is less restrictive of frontier firms and yields an endogenous long-run rate of innovation. We estimate the model for the PC microprocessor industry and perform counterfactual simulations to measure the benefits of competition. Consumer surplus is 2.8 percent higher ($2.8 billion per year) with AMD than if Intel were a monopolist. Innovation, however, would be higher without AMD present. Counterfactuals reveal that consumer surplus can actually increase as the market moves toward monopoly, which suggests policymakers ought to consider the dynamic trade-off lower current consumer surplus from higher prices for higher future surplus from more innovation. Comparative statics reveal that competition does induce higher innovation if consumers have su¢ ciently high preferences for quality and low price sensitivity or if depreciation of the durable good reduces the degree to which a monopolist competes with itself over time. 2 - Variety and Cost Pass-through Among Supermarket Retailers Timothy Richards, Morrison Professor, Arizona State University, 7171 E Sonoran Arroyo Mall, Mesa, AZ, 85212, United States of America, [email protected], Stephen Hamilton, William Allender The extent to which wholesale price changes are reflected in retail prices is an issue of critical importance to manufacturers. Theoretical models of cost pass-through in both the marketing (Tyagi, 1999; Moorthy, 2005) and economics (Bulow and Pfeiderer, 1983; Nakamura and Zerom, 2010) literatures show that competitive, single-product retailers will pass a rise in costs along to consumers on a one-to-one basis if they face perfectly elastic demand. Imperfectly competitive retailers, however, will absorb some of the change in costs depending upon the curvature of demand, the competitiveness of local markets, and the structure of retailing costs. Hamilton (2009) shows that over-shifting, or passing through costs on a more than one-to-one basis, can occur among imperfectly competitive, multi-product retailers because rising input costs cause retailers to reduce the number of products they sell, which softens price competition. We fail to reject this hypothesis using store-level data on breakfast cereal sales from a large sample of U.S. grocery retailers. Our contribution lies in explaining the phenomenon of over-shifting in response to changes in wholesale prices, extending the empirical literature on retail pass-through to multi-product environments, and documenting the extent to which changes in wholesale prices are passed through to consumer food prices. 4 - Dynamics of Pricing Strategy and Repositioning Costs Paul Ellickson, University of Rochester, Simon School of Business, Rochester, NY, United States of America, [email protected], Sanjog Misra, Harikesh Nair 3 - Pricing, Package Size, Advertising and Trade Areas in Spatial Competition of Retail Warehouse Clubs Aharon Hibshoosh, Professor, San Jose State University and Lincoln University, One Washington Square, San Jose CA 95192, 401 15th Street, Oakland, CA 94612, Oakland, CA, 94612, United States of America, [email protected] We measure the cost to retailers of changing pricing formats in oligopolistic retail markets. The choice of pricing strategy is a signficant, but to our knowledge, unquantified, determinant of price-stickiness. For example, adopting an Everyday Low Pricing (EDLP) strategy could have significant effects on long-run market structure, as these firms tend to stick to an initially chosen pricing format for years, if not decades, to avoid incurring adjustment costs. This could be an important source of price rigidity. We exploit a unique dataset containing the pricing-format decisions of all supermarkets in the U.S to measure format-switching costs. The data contain the format-change decisions of supermarkets in response to a large shock to their local market positions: the entry of WalMart. We exploit the price-format responses of supermarkets to WalMart’s entry and infer the repositioning cost using a “revealedpreference” argument similar to the spirit of Breshnahan & Reiss (1990). The interaction between players in a market is modeled as a dynamic discrete game. We find evidence that suggests that the entry patterns of WalMart had a significant impact on the costs and incidence of switching pricing strategy. Our results add to the marketing literature on the organization of retail markets, and to a new literature that discusses implications of marketing pricing decisions for macroeconomic policy. Using game theoretical framework, we model intrinsic short run spatial competition among profit-maximizing Retail Warehouse Clubs RWCs). Traditional modeling of spatial competition has location and price as the key decision variables of the marketing mix of each competitor and typically assumes an instantaneously consumed single product. However, our models assumes that, in the short run store locations are fixed, and more realistically that besides price, RWCs distinctively compete on package size and store advertising while offering many products. We further assume that package size is directly related to RWC’s cost savings due to economics of scale in purchasing, transportation and inventory management. Consumer patronage and purchase decisions are complex, as every consumer minimizes his total purchasing, transportation and holding cost of a portfolio of products. Usually RWCs do not locate near consumers, as convenience stores do, so that consumers encounter time constraints and overall transportation costs. Hence, consumer store visit frequency and average product quantity demanded per visit are often institutionally dependent. Distinctively, the package size offered by an RWC is typically large and the consumer incurs high inventory carrying cost when purchasing non-instantaneously consumed products. The indivisibility of package size raises non-trivial technical and substantive consequences. Specifically, for inventory path valuation, we employ continued fractions and the classical Euclidean Algorithm. The competition is modeled as a sequential game where package size equilibrium is determined prior to price equilibrium. We derive explicit parametric game equilibria solutions for price, package size, store advertising, trading area and profit. 57 FC10 MARKETING SCIENCE CONFERENCE – 2011 ■ FC10 4 - One Size Fits Others: The Role of Label Ambiguity in Targeting Diverse Consumer Segments David Norton, University of South Carolina, 1705 College St., Columbia, SC, 29205, United States of America, [email protected], Randy Rose, Caglar Irmak Founders IV Segmentation Contributed Session Marketers often use product labels to target multiple segments of consumers with a single product. For instance, Calvin Klein’s fragrance CK1 is marketed as “unisex”, targeting both men and women. APG’s Snuggie is labeled as “one size fits all” where the same blanket can accommodate a range of adult body types. Such all-inclusive labels highlight the “all-fitting” aspect of the product, thereby resulting in the expansion of the potential customer pool. However, we argue in this research that marketers may unintentionally be driving customers away with their omnibus labeling practices because all-inclusive labels communicate to customers that the product is likely to fit not only them but also all sorts of other consumers who may not be like them. Since previous research has shown that consumers avoid products or behaviors associated with dissociative reference groups, or groups with whom they want to avoid being confused (Berger and Heath 2007, 2008; Simmel 1904/1957; White and Dahl 2006, 2007), we contend that all-inclusive product labels remind consumers about other consumer groups with whom they do not want to be associated. We focus on this effect of target market specificity or generality on consumer reactions to such products and develop a counter-intuitive argument that leaving the target market of the product ambiguous may actually benefit marketers. A field study helps develop a sensitivity analysis such that we can begin to see the potential financial impact of unintentionally driving consumers away with such labeling practices. Chair: David Norton, University of South Carolina, 1705 College St., Columbia, SC, 29205, United States of America, [email protected] 1 - Loyalty to Service Providers in the Very Short Run and in the Very Long Run: The Impact of Ageing and Cohort Gilles Laurent, HEC Paris, 1, Rue de la Libération, Paris, France, [email protected], Raphaëlle Lambert-Pandraud Older consumers, an important economic target, have been shown to be more brand loyal in various product categories. We investigate whether similarly higher loyalty may be observed for an important service: radio listening, a service with no barriers to switching. We model different aspects of loyalty to radio. We find evidence of higher loyalty in the very short run: Compared to younger users, older persons are more loyal to the service currently provided by a specific radio station, and therefore (i) they listen to it more often and (ii) they listen to it longer on each occasion. We also find evidence of loyalty in the very long run, by comparing radios that existed before 1981 (date of a major liberalization of French radio stations) and radios created after that date. Compared to users belonging to recent cohorts, persons belonging to older cohorts are more likely to listen to the long-established stations that were broadcasting before 1981. Finally, we also find that compared to younger users, older persons are less willing to change for new technologies and therefore tend to use long-established technologies to listen to radio stations, rather than using new ones such as computers, phones, etc. Overall, some of these age differences result from cohort effects, rather than from ageing effects. ■ FC11 Champions Center I 2 - The Dynamics of Brand Preferences Along Consumers’ Life Paths Tingting Fan, New York University, 40 West 4th Street, Room 921, New York, NY, 10012, United States of America, [email protected], Peter Golder Salesforce II Contributed Session Chair: Steven Lu, University of Sydney, CNR Codrington and Rose Streets, Sydney, Australia, [email protected] 1 - Integrated Versus Specialized Salesforce: When Hunting-farming is Harming Ying Yang, University of Houston, 4800 Calhoun Rd, Houston, TX, 77204, United States of America, [email protected], Niladri Syam, James Hess Brands are embedded in our lives. People use brands to build self identities (e.g., Escalas and Bettman 2003), communicate their types to others (e.g., Kuksov 2007), and reflect their social ties to their families and communities (e.g., Berger and Heath 2007; Muniz and O’Guinn 2001). However, current studies only present a static view. Little is known about whether and how brand preferences change as consumers’ lives change. For example, does a young consumer prefer different brands soon after leaving home, starting a relationship, or hunting for a job? If so, what kind of brands does he prefer at each stage of his life path? We track 445 college students’ brand preferences and their life paths from the beginning of their freshman year until the end of their junior year. These three years are interesting and unique because college students experience important changes in their lives, e.g., leaving family, moving to college, beginning to date. It allows us to examine whether these young consumers prefer different brands as their lives change. An analysis of this unique dataset reveals interesting dynamics of brand preferences along consumers’ life paths. We found that these young consumers’ brand preferences diverge at the beginning of college, converge after three months of college, and then diverge again after another three months of college. We also explore how brand preferences differ across four different types of lifestyles, i.e., Workaholic, Social Butterfly, Entertainment Lover, and Sluggard. Our longitudinal study not only fills a gap in the literature of branding but also helps brand managers to understand how consumers build their relationships with brands. An important aspect of salesforce structure, for firms which sell to both new and existing customers, is whether to have two specialized salesforces focusing on new and existing customers separately or an integrated salesforce selling to both. As opposed to the cost-based rationales in the trade press, we advance a novel information-based rationale for the above. Our key insight is that an integrated salesforce allows the firm to better exploit information about the selling environment than a specialized salesforce. Since the customer’s likelihood of purchase involves both its interaction with the specific salesperson and its happiness with the firm’s product, the sales response function has both an agent-specific term and a firmspecific term. Because in an integrated salesforce the same agent sells to both new and existing customers, the effect of the agent-specific term can be filtered out, and information about the firm-specific term can be learnt. The known firm-specific information can be profitably exploited in the future. This is not possible with a specialized salesforce. However the integrated salesforce faces a task-conflict problem. The trade off of the information-extracting power with the task-conflict problem determines whether the firm will prefer an integrated or a specialized salesforce. In selling environments with low firm-specific uncertainty, the task-conflict problems outweigh the information extraction benefits and the firm will prefer a specialized salesforce. When the firm-specific uncertainty is high the information extraction benefit dominates the task-conflict problems and the firm prefers an integrated salesforce. We experimentally test our findings in the laboratory. 3 - Limited Editions: When Snobs Behave Like Conformists and Conformists Behave Like Snobs Sergio Moccia, Doctoral Student, University of Mainz, Jakob Welder-Weg 9, Mainz, 55099, Germany, [email protected], Oliver Heil Limited editions have become a household term in marketing. It appears that they amount to an instrument allowing marketers to signal a product’s exclusivity or uniqueness – thus enhancing the product’s perceived value and, subsequently, increasing consumers’ willingness to pay. These days, limited editions can be found in a vast number of product categories, ranging from luxury items to FMCGs. It is taken for granted that limited editions are preferred by consumers with a rather high need for uniqueness, often called snobs. Importantly, however, it can be observed that there is a strong preference for limited editions among people regardless their affection for the unique and special. We hypothesize that limited editions not only signal exclusivity but also popularity, since many consumers have a seemingly increasing liking for unique goods. As a result, even people with rather low need for uniqueness, often called conformists, have a surprisingly sizeable preference for limited editions. Thus, we argue that, although snobs and conformists constitute two different segments; we believe that there are unique circumstances uniting snobs and conformists in one same segment. We conjecture that this is most often the case when consumers face excess demand, since excess demand may signal both exclusivity (preferred by snobs) and popularity (preferred by conformists). To test our conjectures we use an experimental design in which we manipulate uniqueness, popularity and excess demand and control for quality and value perception. Our preliminary data confirm our conjectures. 2 - Individual-based or Group-based Tournaments? An Experimental Study Hua Chen, University of Houston, Department of Marketing & Entrepreneursh, 334 Melcher Hall, Houston, TX, 77204, United States of America, [email protected], Noah Lim, Michael Ahearne This research examines whether sales managers should design sales tournaments where salespeople compete individually or in teams. Our economic model shows that salespeople will expend higher effort when they compete individually. We test this prediction using economic experiments and find that surprisingly, the prediction does not hold for subjects who have engaged in social interactions before the experiment. We show that the pattern of experimental results can be better explained by a behavioral economics model that captures social preferences among salespeople. 58 MARKETING SCIENCE CONFERENCE – 2011 3 - Investigating Salespeople Turnover in a Dynamic Structural Framework Steven Lu, University of Sydney, CNR Codrington and Rose Streets, Sydney, Australia, [email protected], Ranjit Voola FC13 subsequent customers value customer feedback vis-a-vis their own beliefs about the product or service in their purchase decision. In this paper, we explore whether and when should firms under-promise or over-promise their quality to their customers, when customers have imperfect information regarding such quality in the marketplace. We show that it is not always optimal for a firm to under-promise and over-deliver. Specifically, in a market with two equally matched competitors, underpromising can be sustained only if consumers sufficiently weigh both: i) their initial beliefs about the firm’s quality and ii) any word of mouth about the firm that is inconsistent with the initial quality promises. As these weights decrease, we find that no more than one firm can use an underpromising strategy and for sufficiently low weights both firms prefer to overpromise. Finally, as the true quality of the competitors increases, smaller weights on initial beliefs and positive word of mouth are required to sustain an underpromising strategy. Understanding the salespeople’s turnover process is critical in effective sales force management. However, to our knowledge there are no studies structurally modelling salespeople turnover. We develop a dynamic model to capture salespeople’s turnover decisions. Not surprisingly, sales force turnover, has been an important area of study in marketing research. Various antecedents to sales force turnover have been examined. In a meta-analysis of sales force turnover research, Lucas, Parasuraman, Davis and Enis (1987) highlight that sales force turnover is very difficult to forecast and suggest that a more comprehensive understanding of turnovers will allow for understanding cost reduction, better recruiting, job design and general management practices. Our goal in this study is to provide an assessment of the antecedents of sales force turnover through structural dynamically modelling sales force turnover. Through the dynamic modelling, we also intend to contribute to the theoretical understanding of sales force turnover, by generating insights into the process of learning about person-job fit, performance and demographics and its impact on sales force turnover. Over time, sales people will learn about cognitive skills and capabilities required to do the job, which influence who they feel about good fit. ■ FC13 Champions Center III Financial Decision Making Contributed Session Chair: Carlos Lourenco, Assistant Professor, Rotterdam School of Management, Erasmus University Rotterdam, P.O. Box 1738, Rotterdam, 3000DR, Netherlands, [email protected] 1 - What You Know, What You Do or Who You Know? A Model of Individual Investor Returns Thomas Gruca, Professor of Marketing, University of Iowa, S356 John Pappajohn Bus Bldg, The University of Iowa, Iowa City, IA, 52242-1994, United States of America, [email protected], Sheila Goins ■ FC12 Champions Center II Word of Mouth and Marketing Strategy Contributed Session Chair: Yogesh Joshi, Assistant Professor, University of Maryland, 3301 Van Munching Hall, Robert H. Smith School of Business, College Park, MD, 20814, United States of America, [email protected] 1 - Impact of Company Announcements on the Evolution of Online Word-of-mouth Omer Topaloglu, PhD Student of Marketing, Texas Tech University, Rawls College of Business Texas Tech Unv, MS 2101, Lubbock, TX, United States of America, [email protected], Piyush Kumar, Dennis Arnett, Mayukh Dass Marketing and strategy literature present several studies on the relationship between corporate communications and market response. However, the interaction between corporate communications and consumers’ reactions manifested as electronic wordof-mouth (eWOM) remains to be studied. Although the importance of eWOM as an antecedent of product sales, product awareness, buying intentions, subsequent reviews, and reliability of retailers has been widely explored, how companies tailor their marketing strategies to influence eWOM continues to be a challenge. In this paper, we investigate the effects of company announcements on the dynamics of eWOM associated with a new product. Specifically, we examine how eWOM evolution in terms of its content, quantity, quality, and valence is affected by different types of announcements by the firms and their competitors. We draw implications from our results for managing the interaction between corporate communications and social media. Our focus is the behavior of non-professional investors. Economic changes mean that individuals have to take more responsibility for saving and investing. Most prior research on investor behavior involved concerns professionals (e.g. fund managers) or Wall Street analysts. A continuing stream of research suggests that social networks affect consumer decision making. The question examined in this study is: How do social ties affect behavior of non-professional investors? In addition to social ties, we examine the influences of a trader’s knowledge and market strategy in two electronic futures markets with different levels of uncertainty. Under low levels of uncertainty, individual forecast accuracy determines outcomes. As well, a trader can leverage the information reflected in market prices. Network structure and the accuracy of network information influence both intent to participate and actual market activity levels. Traders with large redundant networks had more accurate forecasts and traded more actively. However, social networks do not have a significant direct or indirect effect on outcomes. In markets with high levels of uncertainty, we find similar network effects on private knowledge and trading activity. However, market success was determined by an accurate knowledge rather than trading activity. Our unique dataset sheds light on the interplay of private information, market actions and social networks on the success of non-professional investors. Our findings imply that investor activity and knowledge is contingent upon accuracy of information in their network. This can prove hazardous to individual investors making investment decisions where their network informational accuracy can’t be know a priori. 2 - Antecedents and Consequences of Pre-release C2C Buzz Evolution: A Functional Analysis Guiyang Xiong, Assistant Professor, University of Georgia, 148 Brooks Hall, Department of Marketing and Distribution, Athens, GA, 30606, United States of America, [email protected], Sundar Bharadwaj 2 - Investing for Retirement: The Moderating Effect of Fund Assortment Size on the 1/N Heuristic Jeff Inman, Frey Professor of Marketing, University of Pittsburgh, 356 Mervis Hall, Pittsburgh, PA, 15260, United States of America, [email protected], Susan Broniarczyk, Mimi Morrin This study focuses on the evolution pattern of online Customer-to-Customer (C2C) buzz over time prior to the launches of new products, or pre-release C2C buzz. Using functional data analysis method, we observe significant heterogeneity in pre-release C2C buzz evolution pattern across products. We also explore the underlying features and investigate the antecedents of the variability in pre-release C2C buzz evolution patterns, especially firm strategic behaviors such as Business-to-Business (B2B) relationships and advertising. Moreover, we demonstrate how pre-release C2C buzz evolution patterns influence new product success (sales and changes in firm stock market valuations upon new product introductions), as well as how this impact is complemented by traditional media and other firm controllable factors. The findings provide unique insights into how firms can strategically influence pre-release C2C buzz to enhance new product performance. Does the number of funds offered in defined contribution plans such as 401(k)’s affect how many funds people choose to invest in or how they spread their dollars across their chosen funds? In this research, we examine the tendency to engage in the 1/n heuristic – investing one’s dollars evenly across all available investment options (Benartzi and Thaler 2001). We decompose this heuristic into its two underlying behavioral dimensions, the tendency to invest in all available funds and the tendency to spread the invested dollars evenly across the funds. We argue that, because choosing from a larger fund assortment size is cognitively depleting, increasing the fund assortment size will decrease the tendency to invest in all available funds (the first 1/n dimension), but increase the tendency to spread one’s dollars evenly among the chosen alternatives (the second 1/n dimension). Our thesis is supported across three experiments as well as in actual investment data obtained from a large financial services firm for actual mutual fund choices by new investors in defined contribution plans. Specifically, we consistently find that offering a larger fund assortment reduces the proportion of funds invested in but increases the tendency to spread one’s dollars more evenly among the chosen funds. Supporting our cognitive depletion argument, Study 1 shows that number of thoughts mediates this process, while Study 2 shows that time spent per fund chosen for investment mediates the process. Study 3 introduces a cognitive load condition, which attenuates the effects observed in the first two studies. Study 4 shows that the predicted effects obtain among actual investors in 401k plans using data for several thousand plan participants across hundreds of defined contribution plans. 3 - Underpromising and Overdelivering - Competitive Implications of Word of Mouth Yogesh Joshi, Assistant Professor, University of Maryland, 3301 Van Munching Hall, Robert H. Smith School of Business, College Park, MD, 20814, United States of America, [email protected], Andrés Musalem Companies are routinely offered the advice that when it comes to meeting customer expectations, they are better off if they under-promise and over-deliver. By doing so, customers are delighted and spread positive word of mouth about the company. This argument, while sound at times, comes with two caveats. First, by under-promising, companies may actually discourage customers from initially trying out their product or service. Second, the benefit from delight may be moderated by degree to which 59 FC14 MARKETING SCIENCE CONFERENCE – 2011 3 - Individual Investors Risk Behavior in Times of Crisis: A Cross-cultural Study Nikos Kalogeras, Maastricht University, Tongersestraat 53, Maastricht, Netherlands, [email protected], Joost M.E. Pennings, Joost Kuikman, Koert van Ittersum second year, both at the customer and at the customer-salesperson dyad level. These results shed light not only on the effects of pricing, but also on the link between the customer relationship to a firm and the relationship to the salesperson. 2 - Volume Based Discounts and Sequential Choice: Structural Estimation and Determination of Optimal Pricing James Reeder, University of Rochester, William E. Simon School of Business, University of Rochester, Rochester, NY, 14620, United States of America, [email protected], Sanjog Misra Recent research argued that by decoupling the risk behavior of individual market participants into the separate components of risk attitude and risk perception, a more robust conceptualization and prediction of their risk responses is possible. Furthermore, it is argued that the mere fear of losing money during a crisis may trigger a negative spiral that can further deepen a financial crisis. Companies that will not be able to raise enough capital to finance projects, hedge funds that face large redemptions that require them to liquidate large positions, or illiquid markets that make fast and efficient trading difficult, are just a few examples of what deteriorating market conditions may bring on. The financial crisis, which started in the beginning of 2008 globally, allowed us to examine the relationship between risk attitudes, perceptions and behavior of individual investors across many Western economies (US, Germany, the Netherlands, Greece, Australia, Italy, Portugal, Ireland). First, the results show that investor risk behaviors during the financial crisis are inconsistent across segments of the population. Second, our findings show that risk attitudes and perceptions drive, although in varied ways, both the participation and quantity reduction decisions of individual investors across segments. Third, our results demonstrate that risk attitude is driven by investors’ gender, age, and income and risk perception is driven by investors’ involvement, trust in information provided by the government, and susceptibility to informational influence. Developers of financial products may use this knowledge when marketing their products, while public-policy makers may gain ideas for introducing regulatory issues and effective communication programs during a financial crisis. This paper analyzes agent reaction to second degree price discrimination, volumebased discounting, with a product line. The proposed model focuses on a business to business transaction between a manufacturing firm and a retailer; in which, the retailer must balance between the perceived and actual product benefits, product cost, and his customers’ preferences. To achieve this goal, I develop an empirical framework that generates a sequential purchasing decision of a doctor (retailer) based upon the prescriptions he writes for his patients (customers) on a daily basis. Unique to this region, doctors not only prescribe medication, but also act as retailers in selling the medication itself, so a doctor must bear the cost of ordering different medications from different brands. Non-parametric analysis demonstrates strategic behavior by doctors at the partitions in the quantity-based discounting scheme, known as “binning.” Omission of this strategic behavior would lead to biased structural parameter estimates; this model implicitly reflects the binning behavior to overcome this source of bias. The application of this model is on a rich data set of over 1,000 doctors for a full year, which contains the daily transaction history, sales force activities, buying group status, promotional discounts, and contractual pricing arrangements. Results from this estimation, and institutional knowledge of real costs, are utilized in policy experiments to understand optimal business policy in pricing and sales force activities. 4 - Improving Investment Advice Using Preferred Outcome Distributions Carlos Lourenco, Assistant Professor, Rotterdam School of Management, Erasmus University Rotterdam, P.O. Box 1738, Rotterdam, 3000DR, Netherlands, [email protected], Bas Donkers, Benedict Dellaert, Dan Goldstein 3 - A Conjoint Model of Quantity Discounts Kamel Jedidi, John A. Howard Professor of Marketing, Columbia University, 518 Uris Hall, 3022 Broadway, New York, NY, 10027, United States of America, [email protected], Raghu Iyengar Quantity discount pricing is a common practice used by business-to-business and business-to-consumer companies. A key characteristic of quantity discount pricing is that the marginal price declines with higher purchase quantities. In this paper, we propose a choice-based conjoint model for estimating consumer-level willingness-topay (WTP) for varying quantities of a product and for designing optimal quantity discount pricing schemes. We use a novel and tractable utility function which depends on both product attributes and product quantity and which captures diminishing marginal utility. We show how to use such a specification to estimate the WTP function and consumer value potential. We also propose an experimental design approach for implementation. We illustrate the model using data from a conjoint study concerning online movie rental services. The empirical results show that the proposed model has good fit and predictive validity. In addition, we find that marginal WTP in this category decays rapidly with quantity. Finally, we identify four segments of consumers that differ in terms of WTP and volume potential and derive optimal quantity discount schemes for a monopolist and a new entrant in a competitive market. In this paper, we propose and demonstrate how investment advice can be improved by using individuals’ preferred outcome distributions that overcome common measurement hurdles impairing accurate assessment of individuals’ biases in risky decisions. Specifically, we propose the Distribution Builder (DB; Goldstein et al. 2008) as a method of elicitation of risk preferences that does not suffer from the problems that plague existing methods, and that allows teasing out biases in the valuation and probability weighting functions by enabling their estimation simultaneously. We derive the conditions that are required to use the DB to estimate prospect theory’s utility and probability weighting functions in parametric and non-parametric ways. We then apply the DB to measure the shapes of the utility and probability weighting functions in the domain of investments, and illustrate the impact of probability weighting on investment decisions. Specifically, we compare the optimal decisions when probability weighting is not accounted for and the decisions arising after debiasing the utility measurement. For our empirical application, we will use multiple tasks where respondents construct their preferred outcome distributions using the DB. These tasks are constructed independent of the preferred distributions elicited in earlier tasks, avoiding error propagation across questions. Moreover, when building a preferred outcome distribution through the DB, individuals manipulate (i.e. answer the question using) both the outcome levels and the probabilities. A more balanced attention towards both is expected to improve the validity of the results. ■ FC15 Champions Center V CRM V: Customer Satisfaction Contributed Session ■ FC14 Chair: Nima Jalali, PhD Student, University of Wisconsin-Milwaukee, Lubar School of Business, 3202 N. Maryland Ave. Suite S466, Milwaukee, WI, 53201, United States of America, [email protected] 1 - The Utility of DLF Binary Ratings in Customer Satisfaction Measurement and Modeling Keith Chrzan, Chief Research Officer, Maritz Research, 996N 250E, Chesterton, IN, 46304, United States of America, [email protected], Jeremy Loscheider Champions Center VI Price Discounting Contributed Session Chair: Kamel Jedidi, John A. Howard Professor of Marketing, Columbia University, 518 Uris Hall, 3022 Broadway, New York, NY, 10027, United States of America, [email protected] 1 - An Empirical Investigation of the Long-term Effects of Price Discrimination in Business Markets Hernan Bruno, Assistant Professor, INSEAD, Boulevard de Constance, Fontainebleau, 77305 CEDE, France, [email protected], Shantanu Dutta In a well-known, if not influential, paper, Wittink and Bayer (1994) describe some advantages of using binary attribute rating scales in customer satisfaction studies. More recently Rossiter, Dolnicar and Grün (2010) describe the use of a specific type (doubly level free or DLF) of binary rating scales for attribute performance measurement. While they note that no brand attribute belief rating measurements provide valid measures, Rossiter, Dolcinar and Grün also suggest that one type of binary measurement, doubly level-free (DLF), performs better than standard five- or seven-point attribute scaling. To practitioners for whom refusing to measure brand beliefs is not a viable commercial option, finding better brand belief measures will obviously be of interest. Though it proved harder than expected to operationalize DLF attribute wording, we were able to field seven empirical studies in which we can compare test cells of respondents completing DLF binary attribute ratings to control cells of respondents using standard five-point attribute rating scales. Our presentation compares the performance of DLF binary attribute ratings and standard five-point attribute ratings in terms of Correlation properties (R-squared, multicollinearity); Discriminant validity; Model coefficient quality (face validity, number of significant parameters, number of reversed parameters). The practice of charging different prices for identical products to different customers in order to extract higher economic surplus occurs in a broad range of business contexts. This form of price discrimination is typically found in situations where the price is set through private interactions between buyers and sellers. While the short term implications of price discriminations are understood, few papers have investigated its long-term implications. In this paper we use a transactional database from an industrial company to empirically explore the relationship between the price paid for a product and the subsequent behavior of customers. We use two years of transactional data to study the evolution of the transactional behavior of customers and the transactional behavior of individual customer-salesperson dyads. In particular, we test the hypothesis that prices paid by a customer affect the long-term profitability of a buyer-seller relationship. Our results show that the price level paid during the first year significantly affect the transactional frequency observed in the 60 MARKETING SCIENCE CONFERENCE – 2011 2 - One-stop Shopping: A Double Edged Sword? Xiaojing Dong, Assistant Professor, Santa Clara University, Lucas Hall, Marketing, 500 El Camino Real, Santa Clara, CA, 95053, United States of America, [email protected], Pradeep Chintagunta FD01 parameters at more disaggregated levels. In this context, this demand study estimates taste parameters for product characteristics such as the form, the type, and the package size of the product. For example, unbreaded tilapia fillet, sold in 48 oz. pack is treated as a different product than unbreaded tilapia, fillet, sold in 16 oz. pack. In doing so, we are able to identify consumers’ preferences for the product (tilapia vs. catfish, for example), the coating (breaded vs. unbreaded), the form (fillet vs. nugget), and the package size. The concept of one-stop shopping has been embraced in the service sector especially in areas such as financial services, telecommunications, etc. The literature in Marketing appears to highlight the benefits to firms of providing a suite of services to their customers. At the same time, a question that arises is: are there any “side effects” of doing this? Specifically, if a consumer is getting multiple services from the same provider and if that consumer is dissatisfied with one service, would there be negative spillovers into other categories as well? There could be a potential “dark side” to one-stop shopping where subscription to a service that a customer is satisfied with might be imperiled if that consumer faces poor service in a different service offered by the provider - the so called “double-edged sword”. Using a service satisfaction survey data of seven companies providing both banking and credit card services, we study the spillover effects of customer satisfaction and dis-satisfaction across these two categories, and their influences on purchase and recommendation intentions. A multidimensional ordered probit model is estimated using Hierarchical Bayesian method. We find that the spillover effects exist, and these effects are asymmetric between positive and negative experiences and across categories, as well as heterogeneous across individuals. 2 - Determinants of Complement Exclusivity in Platform Markets: A Study of the U.S. Videogame Market Srabana Dasgupta, UBC, 210-1238 Melville Street, Vancouver, V6E 4N2, Canada, [email protected], Souvik Datta, Nilesh Saraf This research explores the determinants of exclusive contracting between vendors of platforms (such as videogame consoles) and vendors of complement products (i.e., video game titles). Utilizing market research data from the U.S. video game market, we use a binary measure of complement exclusivity to develop a discrete choice model with measures such as the characteristics of the platform (e.g., it’s installed base), the variety of games offered on a platform and characteristics of the complement provider as dependent variable. We predict, consistent with the previous theoretical literature, that (i) the exclusivity of complements is determined by the stage the corresponding platform market is in with more exclusivity in the early and latter stages of the platform while there exists less exclusivity in the middle stage, and (ii) by the size of the platform, the vendor preferring to be exclusive on a smaller platform rather than a larger one. 3 - Dynamics of Satisfaction: A Regime Switching Ordinal Model for Affective and Cognitive Factors Nima Jalali, PhD Student, University of Wisconsin-Milwaukee, Lubar School of Business, 3202 N. Maryland Ave. Suite S466, Milwaukee, WI, 53201, United States of America, [email protected], Purushottam Papatla 3 - Manufacturers’ e-B2B Platform Choices - Relational Risk Threshold Chen-Han Yang, National Chung Hsing University, Department of Marketing, 250 Kuo Kuang Rd., Taichung, 402, Taiwan - ROC, [email protected], Ming-Chih Tsai, Chieh-Hua Wen Customer satisfaction has been investigated extensively in the marketing literature both as an outcome and as an antecedent to other outcomes. Antecedents such as quality of a product or service and consequences such as word-of-mouth and customer’s loyalty have been explored. Findings suggest that satisfaction with a product or service is dynamic and evolves as customers use a product or service over a period of time. Additionally, the roles of cognitive and affective factors such as product knowledge and product experience, respectively, have been investigated. Recent research by Homburg et al. (2006), for instance, suggests that affective factors have larger effect on customer satisfaction in the early stages of the consumption while cognitive factors become dominant in later usages. Such findings imply that, in the case of products or services that are used over a long duration, managers should pay greater attention to affective factors in the earlier stages of the relationship with the customer but turn to accentuate cognitive elements in the later stages. There are few findings, however, regarding how and when the roles of cognitive and affective factors in influencing the word-of-mouth resulting from satisfaction (or, dissatisfaction) change. It is this issue that we investigate in this research in the context of a range of health care services used over many years. We use a regimeswitching ordinal model for our analysis and calibrate the model on patient satisfaction data using MCMC methods. This research aims to develop a disaggregate approach segmenting the industrial sellers’ e-B2B platform choice under relational risk threshold. Sellers’ three relational bonds with buyers and two uncertain behaviors are identified to serve as sellers’ demand attributes for their focus choices of either public or private e-platform. A latent class model involving demand thresholds is developed to identify and cluster the seller groups. Reciprocal effects between e-platform focus choice and segmented market can be analyzed to help understand industrial relational risk perception by sellers and their choice behaviors. A total of 162 valid data were collected from the Taiwanese manufactures on their e-business with global channel customers. They are tested and calibrated through the exploration factor analysis and LCM using statistical program of NLOGIT 4.0. The results separate manufactures into three clustered as transaction, trust, and relational structure groups by three buyer demographics, including consumption frequency, channel level, and import market. The results exhibit the effects of both asset specificity and uncertainty upon buyer and seller relationship in industrial e-market business. Investments in buyers present a higher relational risk threshold than opportunism for the choice of e-platform. The empirical results help e-platform operators target their customer and justify service strategies. 4 - Contractual Choices and their Consequences in a Time Inconsistent World Paola Mallucci, PhD Candidate, University of Minnesota, 321 19th Ave S, 3-150, Minneapolis, MN, 55455, United States of America, [email protected], George John, Om Narasimhan Friday, 3:30pm - 5:00pm ■ FD01 Using historical, individual level, panel data from an outdoor activity, we show that participants to the activity are time inconsistent. Specifically we show evidence of under-consumption among subscribers to the activity and we find that this underconsumption persists over time, even when subscribers attend the activity for a number of years, which is consistent with theories of naÔf time inconsistency. Additionally, using the unique features of our dataset and additional studies we rule out several alternative explanations for the observed under-consumption. Specifically, our activity is an investment good with a finite number of consumption occasions, hence, in our context the insurance effect proposed by Lambrecht and Skiera (2006), or the cognitive bias proposed by Nunes (2000) cannot fully explain the observed under-consumption. Additionally, using a conjoint study, we show that consumers do not have an intrinsic preference for either contractual format. The observed behavior of subscribers given the features of the dataset and the additional conjoint study provides strong evidence that time inconsistent preferences drive persistent underconsumption. Legends Ballroom I Choice VI: Applications Contributed Session Chair: Paola Mallucci, PhD Candidate, University of Minnesota, 321 19th Ave S, 3-150, Minneapolis, MN, 55455, United States of America, [email protected] 1 - Modeling Consumer Demand for Type, Form, and Package Size in the Seafood and Fish Industry Benaissa Chidmi, Assistant Professor, Texas Tech University, Mail Stop 42132, Lubbock, TX, 79409, United States of America, [email protected] Manufacturers often offer food products in more than one package size, allowing consumers more choices on a given shopping experience. However, due to the increasing number of items offered in any supermarket store for any product category, researchers in applied demand analysis have devised techniques to overcome the dimensionality problem. One of these techniques relies on the assumption of weak separability and multistage budgeting, allowing the researchers to concentrate on a single group, and sidestepping the effect of important factors such as the type, the form, and the size. The development of discrete choice methods offers a very practical alternative to solve the dimensionality problem and estimate demand 61 FD02 MARKETING SCIENCE CONFERENCE – 2011 ■ FD02 4 - Failed Diffusion on Weak Tie Bridges Janghyuk Lee, Associate Professor, Korea University, Business School, Anam-dong, Seongbuk-gu, Seoul, 136701, Korea, Republic of, [email protected], Seok-Chul Baek, Jonghoon Bae, Sukwon Kang, Hyung Noh Legends Ballroom II UGC-IV (Content and Impact) Cluster: Internet and Interactive Marketing Invited Session In this research we examine whether successful diffusion draws on weak-tied bridges, i.e., the weak ties hypothesis (Granovetter 1973). By analyzing 411,078 Web blog posts created by 61,980 bloggers in June, 2006, we show that strong-tied bridges facilitate successful diffusion in terms of distance measured as maximum path length and volume measured as total number of scraps. Our findings suggest three different patterns of diffusion in blog sphere: public announcement, weak-tied non-bridges, and strong-tied bridges. When diffusion via public and direct announcement is excluded, the findings of this paper demonstrate that information will successfully diffuse via strong-tied dyads and that the sender with bridge position will diffuse his or her information more successfully when he or she is embedded in strong-tied dyads. Chair: Janghyuk Lee, Associate Professor, Korea University, Business School, Anam-dong, Seongbuk-gu, Seoul, 136701, Korea, Republic of, [email protected] 1 - Understanding the Dynamic Process of Online WOM: A HB Choice Model for Online Response Behavior Luping Sun, Doctoral Student of Marketing, Peking University, Guanghua School of Management, Beijing, China, [email protected], Ping Wang, Meng Su Online word-of-mouth (WOM) plays a critical role in reshaping consumers’ attitudes toward new products. Consumers consult online reviews to obtain others’ opinions of the new product, and then form their own ones. In a standard review, one main message initiates the communication, followed by responses from many reviewers to express their attitudes. This dynamic interaction process may change consumers’ original attitudes, and should be thoroughly studied to understand the dynamics. This research investigates how prior responses in a review and the characteristics of the current reviewer per se influence his or her attitude toward the new product. We collected 41 new product reviews from various websites and kept the first 40 to 50 responses for each review, which resulted in 1824 observations (i.e., responses) in total. In order to capture the main message heterogeneity, we apply a Hierarchical Bayesian Ordinal Choice Model with MCMC method to estimate the parameters. We find that prior responses and the current reviewer’ characteristics significantly influence his or her attitude to the product, and the impacts of the two differ greatly across main messages. Another intriguing finding is that positive responses matter more than negative ones in reshaping the following reviewers’ attitude. Finally, theoretical and managerial implications are discussed. ■ FD03 Legends Ballroom III Online Search Cluster: Internet and Interactive Marketing Invited Session Chair: Alan Montgomery, Associate Professor of Marketing, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected] 1 - Consumer Search and Propensity to Buy Ofer Mintz, University of California, Paul Merage School of Business, Irvine, CA, 92697-3125, United States of America, [email protected] 2 - User-generated Content in News Media T. Pinar Yildirim, University of Pittsburgh, Katz School of Business, 249 Mervis Hall, Pittsburgh, PA, 15260, United States of America, [email protected], Esther Gal-Or, Tansev Geylani This article investigates the association between consumers’ pattern of information search and their propensity to buy in a field setting. We expect that a consumer whose information search pattern is skewed towards alternative-based search will have a greater propensity to buy than a consumer whose search pattern is skewed towards attribute-based search. In addition, we expect that the price category selected by a consumer influences their subsequent pattern of search. To test these expectations, we develop a conceptual framework and consider several choice models that allow us to account for endogeneity and simultaneity in the relationship between pattern of information search and propensity to buy. The models are fit to a unique database which recorded the search and propensity to buy of 920 shoppers at a global computer manufacturer’s website. The results confirm our expectations. The implication is that a manager can now identify a consumer who has a higher propensity to buy while that consumer engages in information search prior to a purchase commitment, an important first step in targeting decisions. In this study, we investigate a newspaper’s decision to expand its product line by adding an online edition that incorporates user-generated content (UGC), and the impact of this decision on its slanting of news. We assume that the capability of online editions to offer UGC is especially appreciated by readers who have extreme political opinions. We find that extension of the product mix is a dominant strategy for newspapers resulting in reduced bias of the print editions of the newspapers. When UGC is added by readers to the online editions, each newspaper is indirectly forced by subscribers to offer two differentiated versions of its product. This extension of the product mix leads to reduced product differentiation and intensified competition on subscription fees. In addition, since the extent of slant of the online editions is partly determined by subscribers, the ability of the newspapers to extract consumer surplus via price discrimination is restricted. Due to these two effects, newspapers face reduced profitability in comparison to an environment where they offer only print editions or where they have the exclusive right to choose the bias of both editions. 2 - Return on Quality Improvements in Search Engine Marketing Nadia Abou Nabout, Goethe University, Gruneburgplat 7 1, Frankfurt, 60629, Germany, [email protected], Bernd Skiera In search engine marketing, such as on Google, advertisements’ ranking and prices paid per click result from generalized, second-price, sealed bid auctions that weight the submitted bids for each keyword by the quality of an advertisement. Conventional wisdom and statements by Google suggest that advertisers can only benefit from improving their advertisement’s quality. With an empirical study, this article shows the fallacy of this statement; 25 % of all quality improvements to an advertisement lead to higher prices (measured by price per click) per keyword, 70 % to higher costs for search engine marketing, and 47 % to lower profits. Quality improvements lead to higher weighted bids, which only lower prices if they do not improve the ranking of the advertisement. Otherwise, better ranks likely to lead to higher prices. A decomposition method can disentangle these effects and explain their joint effect on search engine marketing costs and profits. Finally, the results indicate that advertisers benefit if they adjust their bids after improvements to advertising quality. 3 - Online Reviews and Consumers’ Willingness-to-pay: The Role of Uncertainty Yinglu Wu, PhD Candidate of Marketing, E. J. Ourso College of Business, Louisiana State University, 3121 Patrick F. Taylor Hall, Baton Rouge, LA, 70803, United States of America, [email protected], Jianan Wu Previous research studying the impact of online reviews on consumers’ willingnessto-pay (WTP) has focused on the valence and volume of online reviews. These studies theorized that both valence and volume took direct and independent roles in consumers’ uncertainty assessment of online purchases, thus their WTP. Results of these studies showed that the valence of online reviews exhibited consistent positive impact on consumers’ WTP, but the volume did not. The literature rationalized these findings on the heterogeneity of risk attitudes between consumers within the expected utility framework. Based on prospect theory and venture theory, we propose a new conceptual framework to study the relationship between online reviews and consumers’ WTP, in which we argue that the valence and volume of online reviews play different roles in consumers’ uncertainty assessment. We posit that the valence of reviews, which estimates the probability of an outcome (i.e., risk), plays a direct role while the volume of reviews, which assesses the uncertainty about such probability (i.e., ambiguity), plays an indirect role. A consumer may tend to 1) overweight small probabilities and underweight large probabilities, and 2) overweight/underweight more when ambiguity is high, rendering the consumer’s preference of smaller review volume when valence is low, but larger review volume when valence is high. As such, the impact of review volume on consumers’ WTP may vary within consumers, depending upon the review valence. We test our framework using data from a controlled experiment and from an online market. 62 MARKETING SCIENCE CONFERENCE – 2011 FD04 2 - Economic Value of Celebrity Endorsement: Tiger Woods’ Impact on Sales of Nike Golf Balls Kevin Chung, Doctoral Student, Carnegie Mellon University Tepper, 5000 Forbes Avenue, Office 317B GSIA, Pittsburgh, PA, 15221, United States of America, [email protected], Timothy Derdenger, Kannan Srinivasan 3 - Which Link to Click – Sponsored or Organic? An Empirical Investigation on Consumer’s ‘Clickability’ Amalesh Sharma, Teaching Associate, Indian School of Business, Gachibowli, Hyderabad, 500032, India, [email protected], Sourav Borah Sponsored search advertising has arguably become the most predominant form of advertising in the online marketing strategy. Sponsored links are paid links and organic links are natural or non-paid links. Prior research has studied the relation between organic and sponsored search advertisement, click process and clicking time (Chatterjee, Hoffman, Novak 2003), browsing behavior of customers in multi-site context (Park and Fader 2004) and online media selection (Danahar, Lee and Kerbache 2009). Montgomery et al (2004) model online browsing behavior to predict buying behavior of customers. In this research we study which specific link in a webpage is clicked on in a consumer search algorithm and why. An extended understanding on the clickability can open up advanced research area in online marketing literature and help to capture the dynamics in the market. We also try to understand the drivers for such click-ability. For this, we build an integrated model using a hierarchical Bayesian Framework to develop a relationship among factors such as keywords findings, key-word appropriateness to the search, nature of search (information/general), display of the link, rank of the link and previous history if any. Our data comes from a controlled field experiments conducted in India. Our results indicate that browsers click relatively more on the organic link (a counter finding from the earlier researches on clickability). We also predict that our findings will provide critical insight to the practitioners in determining where to place the name of the organization in web page and whether to go for paid advertisements. In this paper, we study the economic value of celebrity endorsements. Despite the size and the long history of the industry, few have attempted to quantify the economic worth of celebrity endorsers because it is terribly difficult to identify an endorser’s effect on a firm’s profit. By developing and estimating the consumer demand model for the golf ball market, we find that after controlling for brand advertisement level and taking into account the inherent quality of the endorser, there is a significant endorsement effect as a result of the extra utility attached to the endorsed product. This extra utility leads not only to a significant number of existing customers switching toward the endorsed products but also has a primary demand effect where there is an overall increase in the number of consumers in the market. By sponsoring Tiger Woods for 10 years, we find that the Nike golf ball division reaped additional profit of $60 million through the acquisition of 4.5 million customers who switched as a result of the endorsement. We also find that the recent scandal regarding Tiger Woods’ infidelity had a negative impact which resulted in Nike losing approximately $1.3 million in profit with a loss of 105,000 customers. However, we conclude that Nike’s decision to stand by Tiger Woods was the right decision because even in the midst of the negative impact of the scandal, had they terminated its contract with the golfer, Nike would have lost an additional $1.6 million in profit. 4 - Predicting Purchase Conversion Rates for Online Search Advertisements Using Text Mining Alan Montgomery, Associate Professor of Marketing, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected], Kinshuk Jerath, Qihang Lin 3 - Determination of Brand Assortment: An Empirical Entry Game with Post-choice Outcome Li Wang, PhD Student, Washington University in St Louis, Olin Business School, Campus Box 1133, Saint Louis, MO, 63130, United States of America, [email protected], Tat Y. Chan, Alvin Murphy Many consumers begin their purchase process at search engines such as Google, Yahoo, or MSN instead of traditional retailers. Consumers rely upon the search results provided by these engines along with paid advertising to make decisions about what sites to visit and subsequently which products to purchase. In this study we propose a statistical model that predicts consumer search and the probability of purchase using clickstream data collected from an online sample of consumers. A challenge in analyzing this data is the textual nature of the search strings and the scarcity of many search terms. We also consider how consumers will search based upon the specificity of the search term. This model is cast in the context of a hierarchical Bayesian model to overcome the limited information for many search strings and consumers. We illustrate how this model can be used to aid advertisers in making decisions about how much to bid, what phrase to bid upon, and the appropriate landing page for the consumer once they enter the web site. This paper studies the two-sided decision of brands’ entry into a large department store. This is a two-sided decision process because an observed brand entry has to be agreed by both department store and the brand. By contrast, in standard entry model the entry decision is only one-sided as each agent can make such decision alone. We assume that the store offers each potential entering brand an optimal contract (represented by a transfer from store to brand) to maximize its expected category value. Given the contract, each potential brand makes a take-it-or-leave-it decision. With post-entry sales and transfer data available, we propose a structural model to incorporate post-entry outcome data into a static entry game of incomplete information where the outcome regression is selection-corrected by the inclusion of entry game. This model is estimated using a direct constrained optimization approach which is robust to multiple equilibria and circumvents the heavy computation burden from traditional nested fixed-point algorithm. The model is applied to women’s clothing category, the largest category of department store, and capable of quantifying the magnitude of inter-brand spillovers. We find evidence of significant competition and complementarity (within and out of category) across different brand types and the selection bias from ignoring the entry decision. The estimation result helps to understand how inter-brand spillovers, brand selling cost and brand demand are interwined to yield observed category brand assortment and shed lights on retail category management. ■ FD04 Legends Ballroom V Structural Models II Contributed Session 4 - An Empirical Model of Dynamic Re-entry, Advertising and Pricing Strategies in the Wake of Product Yi Zhao, Georgia State University, Suite 1300, 35 Broad Street, Atlanta, GA, 30303, United States of America, [email protected], Ying Zhao, Yuxin Chen Chair: Yi Zhao, Georgia State University, Suite 1300, 35 Broad Street, Atlanta, GA, 30303, United States of America, [email protected] 1 - The Impact of the Marketing Mix on Durable Product Replacement Decisions Dinakar Jayarajan, Doctoral Candidate, USC Marshall, 3660 Trousdale Pkwy, ACC 306E, Los Angeles, CA, 90089, United States of America, [email protected], S. Siddarth, Jorge Silva-Risso Product harm-crisis may become the worst nightmare for any firm at some point in time. It may affect both the demand side, such as consumer’s intrinsic preference and sensitivities to marketing mix, and the supply side such as costs and competition structures. Product-harm crisis always triggers a product recall, and in most cases the affected brand will return to store shelf after the crisis is solved. The timing of reentry can be a strategic decision for the firms. In this paper, we empirically study firms’ re-entry strategy after a major product-harm crisis, taking into account the demand side of market. Two other interactive strategies, pricing and advertising, are also modeled through Markov perfect equilibrium framework. The model is applied to product-harm crisis event that hits the peanut butter division of Kraft Food Australia in June 1996. Substantively we find that (1) the observed “large advertising expenditure” strategy for the unaffected brand before the affected brand re-enters the market is optimal, since a “large advertising expenditure” in this period plays an important role in postponing the time of re-entry for the affected brand; (2) the competition after the occurrence of the product-harm crisis becomes more intensive. The direct consequence of this is that both the affected and unaffected firms are forced to adopt “low margin” and “large advertising expenditure” (comparing with the pre-crisis advertising expenditure) strategies. As a result, the long-term profitabilities decrease, especially for the affected brand whose profitability decreases by 38%. Finally, the counterfactual experiments show that the increase in the effects of the state dependence is one of the major reasons for a more intensive competition after the product harm crisis. Product replacements account for more than 65% of sales in durable goods categories (Fernandez, 2000). Therefore, buying a new durable product effectively involves two consumer decisions: a) the decision to replace an existing product, and b) the choice of a specific new alternative. The prior literature has analyzed these decisions separately (e.g., Rust (1987); Berry, Levinsohn, and Pakes (1995)), but not jointly, and has typically ignored the impact of marketing activities on these decisions. We develop a dis-aggregate dynamic structural model of the replacement and brand choice decisions with a forward-looking consumer who forms expectations of future new product prices and promotions. On each purchase opportunity the consumer trades off retaining the current product with buying a new product based on the current and expected future utilities of the existing and new vehicles available in the market. We estimate the model using the Nested Fixed Point (NFXP) approach on automobile transaction data for the entry-level SUV category. We start with a sample of consumers who purchased a new car in 2006 and also traded-in an old one. We work backwards from this point and reconstruct the monthly marketing environment for the new products and the monthly depreciation for the traded-in vehicle for up to eight years. We perform counter-factual analysis to a) examine the extent to which consumers anticipate promotions b) decompose the impact of a promotion on replacement acceleration and brand switching and c) gain insights into the differences in the depreciation rate of different vehicles. 63 FD05 MARKETING SCIENCE CONFERENCE – 2011 ■ FD05 4 - Production Networks in Co-creation Niladri Syam, Associate Professor, University of Houston, 4800 Calhoun Road, Houston, TX, 77204, United States of America, [email protected], Amit Pazgal Legends Ballroom VI Game Theory III: General Co-Creation, the participation of the consumer in the production of goods and services, has gained increased popularity of late due to advances in information technology and flexible manufacturing. In this research we distinguish between two types of co-creation: proprietary and non-proprietary. The former involves the price mechanism mediating the exchange between firm and customer while the latter does not. We study, (1) production externalities between firm and customers, (2) production externalities between the customers themselves, and (3) pricing by firms. The main result in the non-proprietary setting is that the firm’s effort decreases in the number of customers (free-riding effect). We also find that the firm’s effort decreases in the degree of substitutability between its efforts and customers’ efforts. In a proprietary setting free-riding is eliminated. This is a major distinction between cocreation, which is often proprietary, and the extant public goods literature. Also, the firm’s effort increases in the degree of substitutability between its effort and the customers’ efforts. Again this stands in contrast to the non-proprietary case. Overall, we contribute not only by investigating the increasingly important phenomenon of co-creation, but also by developing a novel general method to analyze production externalities in the presence of the price mechanism. Contributed Session Chair: Niladri Syam, Associate Professor, University of Houston, 4800 Calhoun Road, Houston, TX, 77204, United States of America, [email protected] 1 - A Model of the “It” Products in Fashion Kangkang Wang, Washington University in St. Louis, Campus Box 1133, Olin Business School, St. Louis, MO, 63130, United States of America, [email protected], Dmitri Kuksov One of the characteristics of fashion is unpredictability and apparent randomness of fashion hits. Another one is the strong influence of the fashion editor recommendations on consumer demand. This paper proposes an analytical model of fashion hits in the presence of competition and a fashion editor acting on behalf of high-type consumers, in which we consider fashion as a means used by consumers to signal belonging to the high class in a matching game. We show that consistently with the observed market phenomenon, in equilibrium, the editor randomizes between available products. Furthermore, this randomization is essential for the market for fashion to exist when low-type consumers’ valuation of meeting high-type consumers is high enough. Whenever the low-type consumer demand for a product is positive, an increase in price results in higher probability of the product being chosen by the editor but lower low-type consumer demand. We also show that in equilibrium, firms always price as to attract strictly positive demand from the lowtype consumers. The equilibrium price and profits are non-monotone in the low-type consumer valuation, with the equilibrium profit first increasing and then decreasing. ■ FD06 Legends Ballroom VII Channels V: Strategy Contributed Session 2 - Would ”False” Promotions be Profitable? Evidence from Experimental Data Yiting Deng, Duke University, Fuqua School of Business, Durham, NC, United States of America, [email protected], William Boulding, Richard Staelin Chair: Volker Trauzettel, Professor, Pforzheim University of Applied Sciences, Tiefenbronner Str, 65, Pforzheim, 75175, Germany, [email protected] 1 - Name-your-own-price as a Competitive Distribution Channel in the Presence of Posted Prices Xiao Huang, Assistant Professor, John Molson School of Business, Concordia University, 1455 de Maisonneuve Blvd West, MB 011-323, Montreal, QC, H3G1M8, Canada, [email protected], Greys Sosic Retailers use many different kinds of consumer promotions to attract customers to their stores. However, not all promotions are real. For instance, one issue of the ShopSmart Magazine pointed out that around last Thanksgiving, a famous retailer had a coffeemaker “on sale” for $61.99, a discount from the retailer’s posted $69.99 “regular” price. However, the manufacturer’s suggested retail price (MSRP) is only $59.99. Given the fact that retailers commonly price items below MSRP, the retailer’s promotion regarding the coffeemaker is actually “false”. In this study, we examined the key factors that affect a firm’s decision of whether to apply real or false promotion from a profitability perspective. Our study is comprised of two major parts: an analytical model on firm decision process and an empirical model based on data collected from a controlled experiment. In the analytical model, we solved for regions of relevant parameters that support different promotional strategies, including offering “false” promotions. We then used the experimental data to identify the parameter values and in turn investigate whether such false promotions can be profitable. In the experiment, we asked the subjects to make two consecutive product choices among two stores and provided them information about the truthfulness of the store promotions. We then applied structural equation models to the experimental data to identify the relative significance and magnitude of different factors in the consumer choice process. Primary results indicate that while the undetected false promotion can positively affect sales, the detected false promotion has negative effects on store image and in turn lowers consumers’ willingness to purchase from the store. Priceline.com patents the innovative marketing strategy, Name-Your-Own-Price (NYOP), that sells opaque products through customer-driven pricing. In this paper, we study how competitive suppliers with substitutable, non-replenishable goods may sell their products (1) as regular goods through a direct channel at posted prices, and/or (2) as opaque goods through a third-party channel, which allows for the NYOP approach. We model the third-party channel as an intermediary firm that collects the difference between the customers’ bids and reservation prices set by the suppliers, and discuss different channel strategies and customers’ bidding strategies. We show that high-end customers may demonstrate low-end behavior (that is, name their prices prior to attending the direct channel, making an even lower bid than the low-end customers), and that the intermediary firm benefits more from horizontally differentiated goods than from vertically differentiated ones. We also use dynamic programming approach to analyze how should suppliers competitively determine channel prices for given initial inventory levels with the goal of maximizing the average expected profit, and show that time and inventory levels have very different impact in dual-channel versus single-channel settings. Our results suggest that the suppliers may not benefit from the existence of a profit-maximizing NYOP channel. In particular, a monopolist would opt out of the NYOP channel and sell at posted prices only, which implies that NYOP is not appropriate for customer discrimination in the regular-goods market. Numerical results show that suppliers are able to generate higher expected profits in the absence of the NYOP channel. 3 - Facts and Slant in News Production Yi Zhu, University of Southern California, 3660 Trousdale Parkway, Los Angeles, CA, 90089, United States of America, [email protected], Anthony Dukes, Kenneth Wilbur 2 - The Optimal Online Common Agency Strategy in the Presence of Instore Display Advertising Hao-An Hung, Graduate Institute of International Affairs and Global Strategy, National Taiwan Normal University, 5F, No. 637, Bei-an Road, Taipei, Taiwan - ROC, [email protected], I-Huei Wu More and more consumers regularly read the news through the internet. Following the shift of readers’ preferences, newspapers, magazines, and TV networks have also been broadcasting their news on the internet. The increasing use of blogs and the emergence of online social network platforms, like Facebook, YouTube and Twitter, have greatly lowered barriers to the production and delivery of news. This new technology intensifies the competition for “eyeballs” among news providers. In this paper, we ask how increased competition affects the presentation of facts and slant in news production. To answer this question, we examine a model that incorporates the following features: a) the number of facts contained in the news is bounded by the choice of media slant. If a news report is more slanted from the truth, there are fewer facts that can be reported truthfully; b) consumers are not only looking for news that is consistent with their opinions, but they also value the facts in the news. Therefore, news producers must optimally balance consumers’ desire for facts and their taste for slant. Under the monopoly case, we find that the news provider slants the news less so that it can deliver a higher number of facts. As competition arises, however, we find that the news producers report more slanted news to avoid the head-to-head fight. As a result, news reporting becomes more polarized with fewer facts. In the tourism and some retail industries, the online common agency such as Priceline has attracted much attention from both marketers and researchers. This kind of new channel sells opaque products which contains limited product information and may intend to serve only those price-sensitive segments. In order to compete with the others in the industry, service providers (i.e., airlines) tend to redesign their in-store display advertising to fit consumer’s preference much better and to create the image differentiation. In this paper, we try to not only capture the price discrimination effect behind the online common agency but also consider the impact of in-store display advertising invested by service providers. We build a gametheoretic model where service providers can decide to invest on either the online common agency or their own websites and consider whether to launch in-store display advertising. By analyzing this model, we attempt to answer the following questions: (i) In the case where in-store display advertising can enhance only the 64 MARKETING SCIENCE CONFERENCE – 2011 FD08 ■ FD08 valuation of the high-end consumers who have brand loyalty, when and why should service providers invest on in-store display advertising? (ii) In the emergence of instore display advertising, are service providers more likely to join the online common agency instead of their own websites? (iii) Could in-store display advertising mitigate the price competition among airlines, the online common agency, and traditional travel agencies? We hope this study can provide some suggestions for marketers who are interested in the impact of the online common agency and in-store display advertising. Founders II Managerial Myopia and Real Activity Mis-Management: Consequences for Marketing and Firm Performance Cluster: Special Sessions Invited Session 3 - Slotting Allowance and Marketing Channel Strategy: An Empirical Analysis Using Quantile Regression Joo Hwan Seo, PhD Candidate, The George Washington University, 2201 G Street, NW, Funger Hall, Suite 301, Washington, DC, 20052, United States of America, [email protected], Ravi Achrol Chair: Natalie Mizik, Massachusetts Institute of Technology, Sloan School of Management, Cambridge, MA, United States of America, [email protected] Co-Chair: Anindita Chakravarty, University of Georgia, Terry College of Business, Atlanta, GA, United States of America, [email protected] 1 - Performance Benchmarks as Drivers of Marketing: The Role of Analyst Forecasts Anindita Chakravarty, University of Georgia, Terry College of Business, Atlanta, GA, United States of America, [email protected], Rajdeep Grewal A considerable literature has developed in marketing to study the phenomenon of slotting fees and related trade promotion expenditures. The expenditures run into billions of dollars annually, and the practices themselves are controversial. The debate has focused on whether they are the product of market power or an efficient market process, and is largely driven by economic theory perspectives. The empirical literature is also limited because direct measures of slotting-like payments, or slotting allowances, have not been available. This study advances the literature in three important ways. It adds a fresh theoretical perspective – that of marketing channel theory. Second, it uses actual dollar value measures of slotting payments. Third, for data analyses it employs panel quantile regression, a new class of regression methodology that permits the analysis of heterogeneity in firm behaviors. The findings suggest that prevailing distinctions between power and efficiency effects may be simplistic. Channel power is a factor in determining slotting allowance but appears to work to enhance channel efficiency. There is also the suggestion that the payments are a response to shifting channel functions. Further, contrary to conventional belief, trade allowances and advertising are not alternative channel strategies but work in conjunction. In contrast, firms that emphasize a product innovation strategy pay less in slotting. As investors reward organizations for meeting or beating short-term analyst earnings forecasts as well as penalize firms for not being able to do so, firms are pressured to engage in activities that boost short-term earnings. We study whether and how the incentive to meet or beat annual analyst forecasts drive unscheduled cuts in marketing and R&D expenses. Results from a multivariate random effect Bayesian model show that top management annual bonuses intensify the extent to which firms engage in such unscheduled cuts in response to analyst forecasts. Furthermore, well performing firms are as likely as poorly performing firms to react to analyst forecasts in this manner. However, organizations that manage high stock of intangible assets and have considerable marketing related experience within the top management team are less likely to make unscheduled marketing and R&D budget cuts in response to analyst forecasts than firms with low stock of intangible assets and negligible marketing related experience within the top management team. The results also show that unscheduled cuts in marketing and R&D budgets can increase long term downside systematic risk. 4 - Price-matching and Retailing Strategies Volker Trauzettel, Professor, Pforzheim University of Applied Sciences, Tiefenbronner Str, 65, Pforzheim, 75175, Germany, [email protected] 2 - Dynamics of Marketing Effort Valuation: High-Frequency Stock Market Data Analysis Isaac Dinner, IE Business School, Calle de Serrano 105, Madrid, Spain, [email protected], Natalie Mizik, Don Lehmann Many retailers apply price-matching strategies to compete with each other. We consider this phenomenom and extend it to broader business strategies. We show that price matching is not a stand-alone strategic tool as it has to be augmented by the other marketing instruments. In the broader sense, price matching extends to a business matching strategy, i.e. retailers do not only match prices, but also assortment, promotions, store design, store location, etc. How rapidly and in what manner information about marketing activities is reflected in firm valuation is unclear. We use high frequency data to examine the financial market’s ability to fully and timely value marketing and R&D-related spending. We find evidence consistent with the market initially under-valuing both marketing and R&D effort. Specifically, while we find differences in the immediate market response to earnings announcements for firms expanding versus reducing their marketing and R&D effort, we also observe a systematic long-term stock price adjustment. We find that a disproportionate amount of the future stock price adjustment occurs around future earnings announcements. This finding suggests that investors update their beliefs about firm performance only after the outcomes of marketing strategies are realized and actual performance signals are sent to the market. This study contributes to the marketing literature by documenting the dynamic patterns in the stock market response to marketing-related information. ■ FD07 Founders I Meet the Editors Marketing Science/ Management Science/Journal of Marketing Research Cluster: Meet the Editors Invited Session 3 - Changing the Rules of the Game: The Impact on Firm Value of Adopting an Aggressive Marketing Strategy Following Equity Offerings Didem Kurt, University of Pittsburgh, Katz Graduate School of Business, Pittsburgh, PA, United States of America, [email protected], John Hulland Chair: Amit Pazgal, Rice University, Houston, TX, 77005, United States of America, [email protected] 1 - Meet the Editors Editors of leading journals for marketing academics will present their editorial policies and perspectives.The following editors are represented: Marketing Science: Preyas Desai; Journal of Marketing Research: Tulin Erdem; Management Science: Pradeep Chintagunta & Miguel Villas-Boas This paper examines how changes in firms’ marketing strategies following initial public offerings (IPOs) and seasoned equity offerings (SEOs) impact firm value. We first show that both IPO and SEO firms adopt a more aggressive marketing strategy during the two years following their offering. However, we then contend that not all issuers benefit equally from this increase in marketing spending, and identify a boundary condition for the link between marketing expenditure and firm value: relative financial leverage of industry rivals. Our prediction is rooted in the theoretical and empirical literature examining the connection between financial leverage and product market competition. We find that the stock market reacts favorably to an aggressive marketing strategy initiated by issuers competing against relatively highly leveraged rivals, whereas increased marketing expenditures do not translate into higher firm value when rivals are less leveraged. Furthermore, we show that marketing expenditures create value within context: the role of marketing expenditures in enhancing shareholder value and the moderating effect of relative financial leverage of rivals are more pronounced in the two-year window following an offering than at any other time. Overall, this paper contributes to the nascent literature examining how marketing and finance resources interact around equity offerings, and provides evidence for a potential “contingency theory of marketingfinance interface” (Luo 2008) by documenting that the impact of marketing on firm value is heterogeneous across firms and market conditions. 65 FD09 MARKETING SCIENCE CONFERENCE – 2011 2 - Accounting for Unobserved Heterogeneity in Models with Strategic Interactions Zheng Li, PhD Student, University of Pennsylvania, 727.6 Jon M. Huntsman Hall, 3730 Walnut Street, Philadelphia PA 19104, United States of America, [email protected], Maria Ana Vitorino 4 - Customer Satisfaction and the CEO’s Long-term Equity Incentives Don O’Sullivan, University of Melbourne, Melbourne Business School, Melbourne, Australia, [email protected], Vincent O’Connell The task of strengthening top management’s attention to marketing is an enduring challenge. Motivated in part by this challenge, researchers have provided extensive evidence of marketing’s impact on firm value. However, demonstrating marketing’s influence on firm value may not be sufficient to secure senior executive attention. Research and anecdotal evidence suggests that executives may prioritize short-term financial performance – even where this has a negative effect on the firm. Linking long-term executive incentives (stock options and restricted stock grants) to indicators of marketing performance has been proposed as a means of counterbalancing top management’s tendency to prioritize current financial performance. Yet, long-term executive incentives have not featured prominently within the marketing literature. Building on the customer satisfaction and incentive literatures in marketing and on the executive compensation literature in accounting finance and economics, we develop a theoretical rationale for tying CEO long-term equity awards to customer satisfaction. We also hypothesize that volatility attenuates satisfaction’s incentive relevance. We draw on data from the American Customer Satisfaction Index (ACSI), ExecuComp, COMPUSTAT, and the Investor Responsibility Research Centre to test our hypotheses. We find that customer satisfaction has a significant impact on the equity awards paid to CEOs. In addition, we find that the sensitivity of long-term equity incentive awards to customer satisfaction is a decreasing function of the volatility in satisfaction. We discuss the implications of our findings for the incentive and compensation literature, practicing marketers, senior executives and boards of directors. This paper develops a model for the estimation of firms’ strategic-interaction games. Identification of strategic interaction parameters in this type of models can be especially problematic if there are unobserved variables that drive firms’ decisions. Intuitively, if there is a market characteristic which is not included in the model but that shifts the profits of the players in the same direction, we may be inclined to think that there are strategic effects between firms when in fact this is not what is driving the correlation between firms’ entry decisions. Using Monte Carlo simulations we show the biases that occur in the strategic interaction parameters when unobserved heterogeneity is not properly accounted for. We also use the model to study shopping centers’ location decisions. We extend previous work by allowing competition of neighboring malls to be a determinant of entry and location decisions, and allow for negative and positive spillovers among different mall types. By quantifying the importance of inter- and intra- mall competition in developers’ entry decisions, the model allows us to assess claims made in the industry that certain types of shopping centers benefit more from co-location than others. 3 - Demand Growth Patterns of Individual Consumers in a Geographically Expanded Retail Market Jungki Kim, KAIST, Dongdaemun-gu, Seoul, Korea, Republic of, [email protected], Duk Bin Jun, Myoung Hwan Park In a retail market where market demand starts to grow after an enhanced retail institution was introduced, major retail firms have kept searching sites for the same type of new outlets to attract more people. Predicting how much demand will grow due to a new retail outlet and identifying consumers who more strongly respond to the new one is so important to successfully implement the geographic market expansion strategy. This study develops a purchase frequency model of individual consumers with different response characteristics to the geographic market expansion to investigate consumer-level demand growth patterns. The response characteristics of our interest are the likelihood of a consumer going to a new retail outlet located in her neighboring region, the spatial response characteristic, and an upper bound that her demand cannot get over even when retail outlets are oversupplied, the temporal response characteristic. The proposed model is applied to the motion picture exhibition industry. Our empirical study finds that the spatial response characteristic is quite similar across moviegoers regardless of their demographic characteristics, but the temporal response characteristic is dissimilar across demographic characteristics. This study also discusses about managerial implications on the location that is the best for a new outlet in terms of growth in market demand and consumer segments that are more promising to the new outlet. 5 - Managing for the Moment: Role of Real Activity Manipulation Versus Accruals in SEO Over-valuation Natalie Mizik, Massachusetts Institute of Technology, Sloan School of Management, Cambridge, MA, United States of America, [email protected], Sugata Roychowdhury, S. P. Kothari Earnings management literature suggests that either under pressure to satisfy performance expectations or in an attempt to influence valuation, managers manipulate reported performance. Accruals manipulation (AM) and real activities manipulation (RAM) are two means of achieving earnings management. We examine the relative prevalence and interrelation of these two strategies and their influence on future performance. Our findings suggest that firms tend to coordinate their RAM and AM activities to artificially inflate current earnings (i.e., these strategies are complements rather than substitutes). Importantly, we find that RAM is more closely and predictably linked with post-SEO stock market under-performance than AM, indicating that firms engaging in real activity manipulation are more over-valued at the time of equity issuance. Our findings suggest that prior research examining the influence of abnormal accruals on post-SEO under-performance suffers from an omitted-variables bias, and incorrectly attributes mis-pricing to accruals, while in fact it is driven by earnings inflation via real activity manipulation. Stakeholders and regulators would benefit from recognizing and appreciating the negative economic consequences of different forms of earnings management. ■ FD10 Founders IV Survey Research ■ FD09 Contributed Session Founders III Chair: Songting Dong, Lecturer, Australian National University, MMIB, LF Crisp Bldg 26, Australian National University, Canberra, 0200, Australia, [email protected] 1 - The Impact of Different Scaling Techniques on Dropout Rates in Online Surveys Petra Wilczynski, Institute for Market-Based Management, Kaulbachstrafle 45, Munich, 80539, Germany, [email protected], Marko Sarstedt Retailing V: Location Decisions Contributed Session Chair: Jungki Kim, KAIST, Dongdaemun-gu, Seoul, Korea, Republic of, [email protected] 1 - The Effect of in-Store Travel Distance on Unplanned Purchase with Applications to Shopper Marketing Sam Hui, New York University, Tisch Hall, New York, NY, 10012, United States of America, [email protected], Yanliu Huang, Jeff Inman, Jacob Suher Surveys are the most important data collection method in empirical research. As nonresponse and dropout challenge the validity of the collected data, achieving high response rates and reducing dropout rates are crucial in all empirical surveys. Especially in online surveys, high non-response and dropout rates have been reported. Nonetheless, recommendations how to reduce those rates are limited to rather general claims about factors that influence dropout behavior, such as questionnaire length, design as well as layout, and incentives. Specific suggestions regarding which scaling techniques are most accepted by respondents have only been given based on theoretical considerations or secondary data analyses. Particularly, the results of secondary data analyses have to be treated with caution as dropout behaviour might considerably differ due to uncontrolled influencing factors, such as the topic of the survey, the questionnaire design, or the promised incentives. To the best of the authors’ knowledge, to date, there is no study that empirically compares the dropout behavior in online surveys associated with commonly applied scaling techniques. The study at hand aims at closing this gap in research by empirically comparing the dropout rates associated with common scaling techniques, such as semantic differential scales, Likert scales, or constant sum scales. Based on our results, we provide researchers as well as practitioners with guidance on the choice of scaling techniques that are associated with lower dropout and non-response rates. Our results provide the necessary grounds for enhancing the efficiency and data quality of online surveys. Retailers have traditionally located frequently purchased products in strategic locations that encourage shoppers to pass by unplanned categories. In addition, recent advances in location-based mobile marketing have made it possible to integrate shoppers’ location with their loyalty card information and offer targeted promotions to increase distance traveled. The success of such strategies is contingent on a relationship between distance traveled and unplanned purchases. In this research, we study the relationship between in-store trip length and amount spent on unplanned purchases. In-store trip length is endogenous because of omitted in-store and out-of-store variables, simultaneity/reverse causality, and measurement error. To combat these issues, we construct a novel instrument based on the length of a “reference path,” which is determined by only the store layout, a shopper’s planned purchases, and an assumption about shoppers’ search strategies. Using instrumental variable regression, we estimate that the elasticity of unplanned purchase on travel distance is 1.44, which is 53% higher than the (uncorrected) OLS estimate. Based on our econometric framework, we explore the potential of using location-based mobile app strategies and product placement strategies to increase unplanned purchases. We find that by strategically promoting a single additional product category to each shopper, unplanned spending could be increased by as much as 28%. In contrast, changing product location only has a limited effect on increasing unplanned spending. 66 MARKETING SCIENCE CONFERENCE – 2011 FD11 2 - A Machine Learning Approach to Analyzing Multi-attribute Data: The OrdEval Algorithm Sandra Streukens, Hasselt University, Agoralaan-Building D, BE-3590, Diepenbeek, Belgium, [email protected], Koen Vanhoof, Marko Robnik-Sikonja ■ FD11 Typically, multi-attribute data are analyzed using a multiple regression approach. This imposes restrictions that may conflict with current marketing theory. First, the appropriate functional form in terms of possible nonlinearity needs to be specified in advance. A challenging task as the literature shows disagrees about the optimal functional form of the relationships among attribute perceptions and higher order evaluations. Second, several properties cannot be adequately modeled by this class of techniques. For instance, several models (e.g. Herzberg’s (1966) dual-factor theory) have been put forward in which a flat relationship between attribute performance and global judgments is hypothesized for some part of the attribute performance range. Third, multiple linear regression models implicitly assume symmetric relationships, meaning that the consequences on the outcome variable of positive (e.g. increase in performance from 3 to 5) and negative (e.g. decrease in performance from 5 to 3) performance changes are equal. Research by, for example, Boulding et al. (1993) demonstrates that this assumption may not reflect reality. These drawbacks underscore the need for alternative analysis methods that offer enhanced possibilities and flexibility in modeling functional forms that have been put forward in recent studies. Therefore, the aim of this research is to develop and demonstrate an approach (i.e. the OrdEval algorithm) based on machine learning principles for analyzing multi-attribute models that does not require an a-priori specification of the functional form in terms of possible nonlinearity, is capable of adequately capturing a wider set of functional forms, and implicitly takes asymmetry into account. Chair: Hema Yoganarasimhan, Graduate School of Management, UC Davis, 3204 Gallagher Hall, Davis, 95616, United States of America, [email protected] 1 - An Empirical Investigation of Sports Sponsorship Yupin Yang, Assistant Professor, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada, [email protected], Avi Goldfarb Champions Center I Sports and Fashion Contributed Session The sponsorship of sports, arts, culture, and charity events has become a popular promotional tool for organizations of all sizes and across many industry sectors. According to the International Events Group Report, global sponsorship expenditures reached $44 billion in 2009, with the majority of corporate sponsorships being sportsrelated. Although sports sponsorship is a marketing strategy involving a large economy, relatively little is known about how sponsors and sponsored organizations address the challenge of evaluating the fitness of partners or how they leverage their strengths in the negotiation process. Research on the formation and departure of sponsorships remains scant in the marketing literature. In this research, we empirically investigate the formation, renewal, and departure of sponsorships in order to shed light on the decision processes involved. Since sponsorships involve the mutual agreement of two partners (the sponsoring company and the sponsored organization), the proposed research will use a two-sided matching model to analyze a unique dataset—the shirt sponsorships of English Football Leagues. In the sponsorship literature, researchers use either a case study approach or else work with data from surveys or experiments. The research is the first empirical work to use historical data to study the phenomenon of sports sponsorship. Thus, it has the potential to make an important contribution to the literature. In addition, it is expected to be of significant interest to industry practitioners (sports teams, sports leagues, sponsoring companies) as well as policy makers. 3 - Voice Analysis for Measuring Consumer Preferences Hye-jin Kim, Pennsylvania State University, 421A Business Building, University Park, PA, 16802, United States of America, [email protected], Min Ding Marketing research methods, such as conjoint analysis and customer satisfaction, usually collect consumer responses in text formats (choice, rating, ranking, etc.). The estimation, in turn, tend to focus on the response itself (which is one-dimensional) and often fail to recognize that the response sometimes may not be reflecting what a subject really thinks. For example, the subject may experience confusion during the task or give socially desirable answers, which is not possible to identify using text based response data. Alternatively, marketers have adopted physiological or neurological measurements (e.g., fMRI) to gain deeper insights into consumers’ true preferences. However, these methods use expensive equipment and require the subject to be present in a lab environment. To retrieve richer information from the respondent while bearing reasonable cost on the marketer and less restriction on the respondent, we introduce a method of using the consumer’s voice to measure emotional and cognitive constructs such as excitement, engagement, and uncertainty, which in turn will be used to supplement responses to allow researchers and practitioners to obtain more accurate insights. While using the human voice for emotion recognition purposes has been prevalent in other fields such as computer science, its use in marketing has been negligible. We conduct an empirical study to test the validity and implementability of this method. The results and implications for marketers and academic researchers are discussed. 2 - The Consumption of Live Sporting Events: Satisfaction of Very Important Fans Dennis Ahrholdt, University of Hamburg - Institute for Operations Reserach, Von-Melle-Park 5, Hamburg, Germany, [email protected], Claudia Höck, Christian Ringle Sporting events are an international multi-billion dollar business and economic aspects have a greater than ever influence on the activities of sports clubs, since competition is no longer restricted to the sports field, but extends into the competition for revenues from selling broadcasting and sponsoring rights, tickets, and merchandise. Hence, the orientation towards the customer (the visitor), who consumes the product “live sports event”, plays a central role. Creating favorable experiences for fans and thereby fan satisfaction represent a key success factor for sports organizations. Particularly customers of business seats and VIP boxes are important, because revenue is mainly driven by those “very important fans” (VIFs). In addition to revenues from VIFs through ticket sales and catering, VIFs positively influence the image of a sports club which in turn has a positive effect on revenues from merchandising and sponsoring rights. Moreover, VIFs are often sponsors or potential sponsors themselves. For these key reasons, VIF satisfaction is fundamental for any sports club’s long-term financial success and, as a consequence, clubs, as sports businesses, must manage VIF satisfaction proactively by paying increasing attention to the range and quality of the services they offer. An instrument for measuring VIF satisfaction is developed innovatively with this research. Structural equations modeling allows us to empirically test the instrument for VIFs from a major German soccer club. The partial least squares path modeling analysis reveals deeper insights of VIFs’ overall satisfaction and its key drivers. Thereby, the instrument identifies and points towards key areas that require managerial attention to maintain and further improve VIFs’ overall satisfaction. 4 - Estimating Nonresponse Bias in Survey Data Songting Dong, Lecturer, Australian National University, MMIB, LF Crisp Bldg 26, Australian National University, Canberra, 0200, Australia, [email protected], Ujwal Kayande Detecting nonresponse bias in survey data is important for determining the representativeness of the sample, which can in turn affect the validity of the survey findings. Extant methods focus on detecting whether the mean of a variable might be different for non-respondents relative to those who did respond. Yet, most surveys used in academic research are not for estimating variable means, but for estimating the relationship between variables. Therefore, we propose methods that may be able to detect non-response bias not only in variable means, but also in the variancecovariance matrix of variables. We use simulations to compare the performance of the methods. We find that the efficacy of the methods in detecting bias varies, particularly in terms of the propensity for false negatives and false positives. We propose a set of guidelines to use when conducting surveys so that researchers are aware of the possibility of non-response bias in the magnitude of the relationship between variables. 3 - Testing Firms’ Conditional Differentiation Behaviour: Quantitative Evidence in Fashion Advertising Kitty Wang, Rotman School of Management, University of Toronto, 105 St. George Street, Toronto, Canada, [email protected] In this paper, I study strategic interactions between firms in their advertising decisionmaking process. I expand on existing literature by incorporating advertising content into a structural discrete game model; this approach allows for an empirical examination of competitive advertising along multiple dimensions. I construct a novel dataset of print advertising in leading US fashion magazines for a five-year span. For each advertisement, it contains information on brands, products, and various other characteristics of the advertisement. Building on the static discrete entry game framework, I propose a two-stage model to capture the impact that rival firms’ (expected) actions have on own firm’s strategy in terms of when and what to advertise. The two-stage model allows firms to construct expectations based on what they learn through rival firms’ past strategies. It also replaces the commonly used rational expectation assumption with a more general structure for the information set. While previous studies focus on advertising quantity, I find that firms strategically coordinate or differentiate from their rivals’ (expected) actions in terms of timing, location, and content. Furthermore, there appears to be no dominant strategy along all dimensions – firms’ actions depend crucially on firm characteristics, the product category, as well as other advertising content. 67 FD12 MARKETING SCIENCE CONFERENCE – 2011 4 - Identifying the Presence and Cause of Fashion Cycles in the Choice of Given Names Hema Yoganarasimhan, Graduate School of Management, UC Davis, 3204 Gallagher Hall, Davis, 95616, United States of America, [email protected] ■ FD13 Fashion denotes the changing tastes in conspicuously consumed products and choices. There are two classic, but competing theories of fashion – (1) Veblen’s theory of fashion as a signal of wealth, and (2) Bourdieu’s theory of fashion as a signal of cultural capital. Theoretically, both theories can give rise to fashion cycles, and there is considerable debate in the literature as to whether fashions are driven by consumers’ desire to signal wealth or by their need to signal cultural capital. However, there exists no systematic empirical investigation of the phenomenon of fashion. So far, even the existence of fashion cycles hasn’t been empirically established. In this paper, we bridge this significant gap in the literature. We examine the presence and cause of fashion cycles in the choice of given names. Using a rigorous set of panel unit root tests, we first empirically establish the existence of fashion cycles in the choice of given names. Second, by exploiting state-level variation in measures of economic and cultural capital, we test the validity of the two competing theories of fashion. Our result support Bourdieu’s cultural capital theory. Chair: Harikesh Nair, Associate Professor, Stanford University, 518 Memorial Way, Stanford, CA, 94305, Uganda, [email protected] 1 - Online Consumer-to-consumer Communication and Marketing Strategy Ganesh Iyer, University of California-Berkeley, Haas School of Business, Berkeley, CA, 94720, United States of America, [email protected], Zsolt Katona Champions Center III Network Effects Contributed Session We investigate consumers’ incentives to communicate with their peers in the context of the emergence of Internet social media technologies. In particular, we examine the effects of different types of communication cost structures through which new technologies change the way consumers send - potentially product related messages. These new social networking technologies make it possible to send the same message to many receivers for the same fixed cost and for negligible (often zero) marginal cost. This is distinct from traditional word-of-mouth which involves additional marginal cost for each intended recipient. Our results show that the new social networking technologies may reduce incentives of some individuals to send messages leading to less diverse communication. The number of senders may shrink but each sender may potentially send messages to the entire network of receivers. Our results also suggest that the online social communication phenomenon could lead to reduced consumer heterogeneity if the source of messages is a small homogeneous group. We also study the interaction of the decision to send messages and the choice of quality. ■ FD12 Champions Center II Models of Word of Mouth Processes Contributed Session Ingmar Nolte, Assistant Professor, Warwick Business School, Gibbet Hill Road, Coventry CV4 7AL, United Kingdom, [email protected] 1 - Modeling Promotional Word-of-mouth Backhun Lee, PhD Candidate, KAIST Business School, 87 Hqeiro Dongdaemoon-Gu, Seoul, 130-722, Korea, Republic of, [email protected], Minhi Hahn 2 - Identifying High Value Customers in a Network: Individual Characteristics Versus Social Influence Sang-Uk Jung, Doctoral Candidate, University of Iowa, 108 Pappajohn Business Building, Iowa City, IA, 52242, United States of America, [email protected], Qin Zhang, Gary J. Russell Internet communication websites allow consumers to share usage experiences and consumer product reviews. They promote word-of-mouth (WOM) activities of consumers overcoming geographic boundaries. However, firms may take advantage of the anonymity of the internet to disguise their promotion as consumer reviews. Even though most marketers and consumers perceive this “promotional” WOM, there are few studies regarding this issue. Mayzlin (2006) studied this phenomenon based on a game theory and found interesting results. Yet, there is no empirical evidence of promotional WOM so far. Our study contributes to the literature as the first empirical analysis to study promotional WOM and its effects. We collected online user review data from three different movie review websites for 732 movies released in South Korea during a 46-month period. The final data set includes about 2,600,000 user reviews and 950,000 user IDs for each of websites. Individual consumers post a review after they purchase a product. The valence of review depends on the product quality, consumer preference, and random errors. At the aggregate level, the volume of WOM is the number of individual reviews. The valence of WOM is the average of valences of all reviews. Therefore, the volume depends on the previous product sales and the valence depends on the product quality. On the other hand, the firms generate promotional WOM to increase both a volume and a valence of online WOM arbitrarily. Therefore the volume and valence of online WOM shares correlated error terms due to promotional WOM. We develop a mathematical model based on consumer behavior. The estimated results support our model and existence of promotional WOM. The effects of actual WOM and promotional WOM on product sales are investigated. Firms are interested in identifying customers who generate the highest revenues. Typically, customers are regarded as isolated individuals whose buying behavior depends solely on their own characteristics (e.g., previous purchase behavior, demographics etc.). In a social network setting, however, customer interactions can play an important role in purchase behavior. Previous work on social networks has focused most attention on modeling the interaction between individuals and understanding the positions of individuals in a network (e.g., measuring the influence of an individual based on his/her degree of network centrality). Little is known about how network influence directly translates into the benefits to the firm. In this paper, we argue that it is important to take into account both individual and network effects when measuring customer value. Drawing upon the statistics literature, we construct a conditional autoregressive (CAR) spatial model that explicitly shows how these effects interact in generating firm revenue. We apply our model to a unique userlevel dataset from a popular online gaming company in Korea. The data contain information about individual gamer demographics, interaction between gamers, behavior within the game environment, and revenues generated by each individual. Our model outperforms benchmark models in predicting revenues that gamers generate. It also allows us to quantify the relative impact of individual and network effects on revenues. We show that individuals who are most influential in a network sense may not necessarily be individuals who have the highest customer value. Implications of the model for targeted marketing policies are discussed. 3 - Brand Value and Indirect Network Effects in a Two-sided Platform Yutec Sun, University of Toronto, 105 St. George St, Toronto, ON, M4V 1W1, Canada, [email protected] 2 - Where Do the Joneses Go on Vacation? Social Comparison and the Weighting of Information Ingmar Nolte, Assistant Professor, Warwick Business School, Gibbet Hill Road, Coventry, CV4 7AL, United Kingdom, [email protected], Sandra Nolte, Leif Brandes This proposal aims to evaluate the impact of endogenous software supply on brand value in the U.S. smartphone market during 2007-2009. With the introduction of Apple’s App Store, the smartphone was transformed to a two-sided software platform, where consumers and third-party software developers buy and sell smartphone applications, creating positive indirect network effects to the other sides. The indirect network effects are found in many studies as a critical factor in marketing tipping toward one dominant platform, regardless of the intrinsic qualities of the platforms. Yet the increasing reliance of the smartphones on software attributes implies a bigger role for brands, since software qualities are essentially experience attributes. While the relative strength of indirect network effects and brand values can determine the eventual market structure, their strategic relation has remained unclear in the relevant empirical literature. I propose to fill this gap by measuring the contributions of software and brands to the equilibrium profits, combining the approaches of Goldfarb, Lu, and Moorthy (2009), and Nair, Chintagunta, and DubÈ (2004). The preliminary results suggest that without the indirect network effects, iPhone and Anroid’s brand values will decrease by more than 60%. Does social distance between senders and receivers of information impact social learning? This paper aims to answer this question empirically with data on hotel bookings and customer reviews. Specifically, we analyze the effect of group membership differences between reviewers and customers on customers’ preferences for hotels. The data come from the largest travel and holiday booking portal on the web for German speaking countries, and include five years of hotel bookings. Our empirical analysis uses two distinctive features of the data. First, customer reviews include information on whether the reviewer travelled as single, couple, family, or with friends. This classifies reviewers into categories. Second, the data includes full information on each booking, including timestamp, price, travel time, destination, and travelers’ characteristics. This allows us to classify bookers into the same categories, and to reconstruct a hotel’s state of reviews for the time of each booking. We use bookers’ willingness to spend to measure hotel preference. Our findings are twofold. First, we show a positive link between reviewer ratings and customers’ willingness to spend. This result extends previous literature on customer reviews and aggregate product sales. Second, customers from one group weight information from other groups’ reviewers differently leading to asymmetric effects across groups. We contribute to the literature on social distance and opinion formation in two ways: despite the existence of a wide range of theoretical models, and a number of experimental studies, this is the first empirical evidence based on microlevel data. Moreover, on the basis of the documented asymmetry in information weighting we conclude that further theoretical development is required. 68 MARKETING SCIENCE CONFERENCE – 2011 4 - Social Ties and User Generated Content: Evidence from an Online Social Network Harikesh Nair, Associate Professor, Stanford University, 518 Memorial Way, Stanford, CA, 94305, Uganda, [email protected], Reto Hofstetter, Scott Shriver FD14 heterogeneity of both consumers and variable categories using categorical data. The main objective is to obtain a positioning map that also accounts for the heterogeneity in consumer perceptions by identifying joint clusters, each consisting of a distinct subset of consumers and variable categories. A comprehensive view of repositioning through a joint mapping of consumers and attribute categories at each abstraction level adds row-dimension projections. Thus, repositioning strategy can be evaluated based on consumer demographic and/or psychographic characteristics in addition to attribute-level changes. We illustrate the efficacy of the proposed theoretical framework and empirical model with consumer panel data for mid-sized sedan automobiles. We use variation in wind speeds at surfing locations in Switzerland as exogenous shifters of usersÌpropensity to post content about their surfing activity onto an online social network. We exploit this variation to test whether users social ties on the network have a causal effect on their content generation, and whether content generation in turn has a causal effect on the users’ ability to form social ties. Economically significant causal effects of this kind can produce positive feedback that generate multiplier effects to interventions that subsidize tie formation. We argue these interventions can therefore be the basis of a strategy by the from to indirectly facilitate content generation on the site. The exogenous variation provided by wind speeds enable us to measure this feedback empirically and to assess the return on investment from such policies. We use a detailed dataset from an online social network that comprises the complete history of social tie formation and content generation on the site. The richness of the data enable us to control for several spurious confounds that have typically plagued empirical analysis of social interactions. Our results show evidence of significant positive feedback in user content generation. We discuss the implications of the estimates for the management of the content and the growth of the network. Finally, we also provide bounds for robust identification of causal social effects under weak identification conditions. 3 - Business is in My Blood: Do Family Firms Outperform Non-family Firms During Economic Recessions? Saim Kashmiri, PhD Candidate, University of Texas at Austin, 1 University Station B6200, Austin, TX, 78712, United States of America, [email protected], Vijay Mahajan Family firms – firms owned and/or managed by founders or their families – play a critical role in the U.S. economy, making up about 35 percent of firms listed on the S&P 500 or Fortune 500 indices and contributing about 65 percent to the U.S. GDP. This research explores whether family firms exhibit unique strategic behavior during economic recessions and whether this behavior in turn helps them outperform nonfamily firms. Findings based on a sample of 428 U.S. publicly listed firms, over seven U.S. recessions between the years 1970 and 2008, reveals that family firms consistently outperform non-family firms during recessions. This superior performance is partially driven by founding families’ longer horizons while investing in such market-based assets as brand capital, social capital and human capital. Family firms do not decrease advertising as much as non-family firms do during recessions, thereby exhibiting higher levels of advertising intensity. Family firms also get involved in fewer social and employee-related unethical actions. Furthermore, family firms’ lower financial leverage helps minimize bankruptcy threat, also boosting these firms’ performance during recessions. These results underscore the benefits of a maintaining a conservative capital structure, and making consistent investments in market-based assets. These results also add to the scant family-firm literature, demonstrating family firm ownership to be an effective organizational structure during recessions. ■ FD14 Champions Center VI Marketing Strategy I: General Contributed Session Chair: Neil Bendle, Assistant Professor, University of Western Ontario, Ivey School of Business, 1151 Richmond Street N, London, ON, N6A 3K7, Canada, [email protected] 1 - Does Market Potential Always Attract New Market Entry? A Contingency View Namwoon Kim, Professor, Hong Kong Polytechnic University, Department of Management and Marketing, Hung Hom, Kowloon, Hong Kong, Hong Kong - PRC, [email protected], Ge Zhan, Sungwook Min 4 - Are Your Customers Crazy? Neil Bendle, Assistant Professor, University of Western Ontario, Ivey School of Business, 1151 Richmond Street N, London, ON, N6A 3K7, Canada, [email protected] Contrary to Drucker’s (2007) advice marketing has enthusiastically embraced irrationality (Godin 2009, Cusick 2009). Writers draw upon behavioral economics (Kahneman 1994, Ariely 2008, 2010) where academics argue persuasively against Homo Economicus. Shugan (2006) however notes that conflating Homo Economicus - “adept optimizers” with “perfect foresight” who “never tire” - with rational is a marketing ploy by certain economists. Homo Economicus is but one, rather limited, view of rationality (Binmore 2009). Marketers can question the premise that Homo Economicus equals rational. Irrational is a pejorative term for customers while Homo Economicus diminishes our discipline by implying that marketing to “rational” customers is pointless. Furthermore given it is a reaction to the confused notion of Homo Economicus irrational is itself ill defined. Instead of adopting Homo Economicus or embracing irrationality I recommend marketers articulate their own view of consumer rationality accepting consumers have, sometimes surprising, reasons for their actions. A marketer’s view of consumer rationality should answer simple questions such as: Are mistakes irrational? Choices based upon social preferences? I examine the connection between rationality assumptions and marketing strategy to show that a view of rationality should be the basis of a behavioral marketing strategy. For instance if needs are inherently unpredictable seeking to find and satisfy consumers’ current and future needs – a central pillar of market orientation (Kohli & Jaworski 1990) – becomes a futile endeavor. Similarly advertising (including informative advertising) is pointless for those who accept Homo Economicus. A marketer’s conception of customer rationality is critical to determining marketing strategy. Market potential, defined as the anticipated industry sales/customer population in the marketplace, has long been viewed in the literature as being related to market entry motivation (Fuentelsaz and Gomez 2006; Pennings 1982). While numerous findings identify market potential as the main incentive driving firms’ decisions to enter new markets (e.g., Baum and Korn 1996; Swaminathan 1998; Damar 2009), more than a few studies report a weak or even negative impact of market potential on such decisions (e.g., Bronnenberg and Mela 2004; Chesbrough 2003). In this regard, the emerging literature notes that entry into a new market is a complicated and multifaceted event that may be more influenced by other contextual factors such as industry background and the firm’s industry experience, as well as choice of targetmarket strategy (niche versus mass marketing). Based on a theoretical and integrative review of the existing literature on the relationship between market potential and market entry, this study provides a contingency framework to explain the conditions that strengthen or weaken this relationship. Using 187 parameter estimates from 38 previous studies, our meta-analytic regression models have validated significant moderating effects for the market potential – market entry relationship. The moderating contexts are whether: (1) the new market is categorized as a high-tech or non-high-tech industry; (2) the entry is made by an established industry-incumbent or by a start-up; (3) the new market is a new-product market for the firm or a new geographic market; and (4) the new market entry targets a niche market or a mass market. 2 - Repositioning via Abstraction Using Categorical Data Jonathan Lee, Associate Professor of Marketing, California State University-Long Beach, CBA 351 Dept of Marketing, 1250 Bellflower Blvd, Long Beach, CA, 90840-8501, United States of America, [email protected], Heungsun Hwang Positioning is an essential strategic decision that represents a unique challenge for many marketers. While previous empirical models for positioning focus primarily on product features/attributes, we propose an abstraction framework (attributes-benefitsvalues) for the repositioning problem, arguing that consumer benefits and values can offer more effective strategic options compared to an attribute-based perspective. We present a simultaneous two-way clustering approach using a mixture of Bayesian multiple correspondence analysis for repositioning to account for cluster-level 69 FD15 MARKETING SCIENCE CONFERENCE – 2011 Saturday, 8:30am - 10:00am ■ FD15 Champions Center V CRM VI: Customer Satisfaction ■ SA01 Contributed Session Legends Ballroom I Chair: Jiana-Fu Wang, Assistant Professor, National Chung Hsing University, 250 Kuo Kuang Rd., Taichung, Taiwan - ROC, [email protected] 1 - Modeling Determinants of the Satisfaction-loyalty Relationship: Theoretical and Empirical Evidence Young Han Bae, Doctoral Student, University of Iowa, Tippie College of Business, 108 PBB S252, Iowa City, IA, 52242-1994, United States of America, [email protected], Gary J. Russell, Lopo Rego Conjoint Analysis: Improving the Process Contributed Session Chair: Dan Horsky, Simon Graduate School of Business, University of Rochester, Rochester, NY, 14627, United States of America, [email protected] 1 - Best-worst Conjoint Analysis as a Remedy for Lexicographic Choosers Joseph White, Director, Marketing Sciences, Maritz Research, 1815 S. Meyers Rd., Suite 600, Oakbrook Terrace, IL, 60181, United States of America, [email protected], Keith Chrzan Customer satisfaction and loyalty are central constructs to marketing research and practice since they reflect how effectively firms deliver value to their customers, and because they are important determinants of current and future product-marketplace and financial performance. In this study, we develop a comprehensive and flexible theoretical framework for analyzing the association between customer satisfaction and customer loyalty, which also incorporates competitive setting differences. This theoretical framework is grounded in more than 40 years of academic and practitioner research on the association between these two constructs and allows us to more precisely examine the true nature of the association between satisfaction and loyalty. Additionally, we test our theoretical framework by estimating an empirical hierarchical linear model, using American Customer Satisfaction Index (ACSI) data and several customer, firm and industry characteristics. Our findings indicate that the true nature of the association between satisfaction and loyalty is significantly influenced by context. Controlling for such differences allows firms and managers to significantly increase their ability to effectively convert satisfaction investments into loyalty. Additionally, we identify significant decreasing marginal returns for customer satisfaction investments, as well as important trade-offs between intercept and slope on the association between the two metrics. Our study provides important theoretical, managerial and regulatory insights, and broadens our understanding of the essential features of the satisfaction-loyalty association. Designed choice experiments often rely on strategies that seek to maximize efficiency by minimizing level overlap (Huber and Zwernia 1996). In the absence of overlapping levels, however, respondents making lexicographic choices or choices dominated by a single attribute may provide no information on attributes other than the most important one (Orme 2009). A large proportion of respondents make lexicographic or dominated choices (Killi, Nossum and Veisten 2007; Kohli and Jedidi 2007; Campbell, Hutchinson and Scarpa 2006), resulting in poor predictions in highly competitive holdout choices (Chrzan, Zepp and White 2010). Experimental designs that incorporate level overlap exist (Sawtooth Software 2007, Liu and Arora 2010, Chrzan, Zepp and White 2010). Plausibly, Best-Worst conjoint analysis (Swait, Louviere and Anderson 1995) as it does not rely on choice sets with or without level overlap, may provide another type of RUM experiment that avoids the problem lexicographic/dominant choosers pose to minimal overlap choice set designs. In addition to replicating tests of the comparability of Best-Worst conjoint analysis and discrete choice experiments (Swait, Louviere and Anderson 1995, Chrzan and Skrapits 1996), our presentation reports an experiment that compares the ability of a standard minimum overlap discrete choice experiment and a Best-Worst conjoint experiment to predict the choices of holdout respondents (some portion of whom hopefully make dominated choices). 2 - Does the Variance in Customer Satisfaction Matter for Firm Performance? Eun Young Lee, Doctoral Student, Korea University Business School, Anam-dong Seongbuk-gu, Seoul, 136-701, Seoul, 136-701, Korea, Republic of, [email protected], Shijin Yoo, Dong Wook Lee, Sundar Bharadwaj 2 - Using Additional Data Collection and Analysis Steps to Improve the Validity of Online-based Conjoint Sebastian Selka, Scientific Assistant, Brandenburg University of Technology, Erich-Weinert-Strafle 1, Cottbus, 03046, Germany, [email protected], Daniel Baier Although much attention has been paid by both academics and corporate managers to customer satisfaction as a leading indicator of firm performance for many years, relatively little is known about the role of its variance. We investigate the relationship between the variance of customer satisfaction and three different dimensions of firm performance: accounting performance (i.e., revenue and profit), Tobin’s Q ratio, and stock return. Based on National Customer Satisfaction Index (NCSI) data collected by Korea Productivity Center in a last decade, we obtain three main findings. First, we confirm the findings of extant literature – mostly based on US data – that the average customer satisfaction is positively related to the firm performance. Second, we find that the variance of customer satisfaction negatively moderates the relationship between the mean of customer satisfaction and firm performance, i.e., the average customer satisfaction level is more strongly related to firm performance when the variance of customer satisfaction is low. Finally, the variance of customer satisfaction is found to directly affect firm performance as well. More specifically, the variance increases the sales and decreases the stock return. Academic and managerial implications are also discussed. Conjoint experiments are tending more and more to end up with low internal and external validity of the estimated part-worth function (see, e.g., Green et al. 2001), because of (missing) temporal stability and structural reliability of respondents’ partworth functions (see, McCullogh, Best 1979 or DeSarbo et al. 2005). Also respondents’ (missing) attentiveness during conjoint experiments is an important source of noisy data and more or less caused by uncontrolled data collection environments, e.g. many parallel web applications (e.g., social networks, electronic mail, newspapers or web site browsing, ) during CASI. Here, additional data collection and analysis steps have been proposed as solution (see, e.g., Netzer et al. 2008 for an overview). Examples of internal sources of data are response latencies, eye movements, or mouse movements, examples of external sources are sales and market data. The authors suggest alternative procedures for conjoint data collection that deal with these potential sources of internal and external validity by using additional calculations and analysis steps. A comparison in an adaptive conjoint analysis setting shows, that the new procedures lead to a higher internal and external validity. 3 - The Impact of Online Railway Ticket Cancellation Policy on Revenue and Customer Satisfaction Jiana-Fu Wang, Assistant Professor, National Chung Hsing University, 250 Kuo Kuang Rd., Taichung, Taiwan - ROC, [email protected] 3 - Estimation of Individual Level Multi-attribute Utility from Ordinal Paired Preference Comparisons Dan Horsky, Simon Graduate School of Business, University of Rochester, Rochester, NY, 14627, United States of America, [email protected], Paul Nelson, Sangwoo Shin In some railway companies, booking limit is used to allocate the number of tickets among multiple legs in a train. Due to the ticket reservation feature of “first-comefirst-served” and the allowance of cancellations or no-shows, company revenue and customer loyalty might be jeopardized. This study examines a railway company’s online reservation system. When a reservation is made, the company holds it for at most two days before the reservationist makes the real purchase. The reservation might turn out to be a cancellation in the end, while during the above period, other customers might be rejected due to the policy of booking limit. In our June, 2010 database, 54% of the online reservations are either cancelled or give-up without notice. This incurs other customers’ unsatisfaction and a customer may take different strategies to tackle this situation: try again later, try an alternative train, order two or more sections to compose his/her original target section, and use other transport mode. In order to consider the above dynamic and nested customer behavior, a detailed simulation model is constructed according to the information extracted from the ticket reservation database. We use this model to estimate how much profit could have been lost, and how many “loyal” customers are lost due to denied bookings. We also suggest several strategies to improve revenues and reduce rejections for future research. Our work suggests that linear programming analysis of ordinal paired preference comparisons (or such comparisons inferred from interval-level ratings) may lead to better individual utility estimates than alternative methods which use interval-level ratings data directly. In this paper we: (a) outline a theoretical foundation for estimating a cardinal scaled utility function from ordinal preference data, in particular, pairs of pairs or ordered paired comparisons; (b) forward linear programming procedures and their HB versions, designed to estimate individual level attribute weights from such data; (c) evaluate the statistical properties of these estimators and develop statistical significance tests for them; and (d) evaluate the ability of these estimators to predict hold out sample preferences for two real world datasets. Simulations show that our ordinal preference-based weight estimates are more robust to data quality issues than either regression or ordered logit based estimates. Our real world results also show superior predictive performance. Our findings indicate that the higher potential for measurement error and scale usage heterogeneity that resides in cardinal scaled data is an issue. Correspondingly, ordinal preferences in conjunction with our linear programming estimation methodology provide individual level attribute weight estimates that are worthy of academic and managerial attention. 70 MARKETING SCIENCE CONFERENCE – 2011 ■ SA02 SA03 2010). However, research has not investigated the relationship between Tweets about brands and the stock market value of firms that own those brands. The authors collect more than 9 million tweets from Twitter for eight brands to address this issue. The authors use text mining techniques to create measures of volume, valence and word of mouth from the Twitter data. Using data at a daily level for a period of 6 months from October 2009, they evaluate whether the Twitter metrics have an impact on the stock market performance of firms using Vector AutoRegressive models. They control for several exogenous variables such as analyst coverage, advertising expenditures and key firm developments such as innovations, mergers, announcement of earnings, client contracts, strategic alliances, lawsuits and changes in key executives. The authors find that digital conversations in social media have a relation with stock market performance measures (7 of 8 brands have effects on abnormal returns). However, most of the effects are short term supporting the notion of market efficiency. Retweets (word of mouth) is the metric of Twitter with the most influence on abnormal returns. Volume of tweets is the metric with the most influence on trading volume. The authors discuss managerial implications of these results. Legends Ballroom II Twitter and Social Media Cluster: Internet and Interactive Marketing Invited Session Chair: Abishek Borah, University of Southern California, Los Angeles, CA, United States of America, [email protected] 1 - Methodology for Codifying Qualitative Twitter Content into Categorical Data Stephen Dann, Australian National University, MMIB, CBE, ANU, Acton, 00200, Australia, [email protected] Twitter provides an immense wealth of qualitative data both in aggregate and at the individual account level. This paper proposes a method of classification of individual Twitter account content across a two level multiple domain framework. Whilst a range of prior studies have emphasized the quantification and classification of the Twitter public timeline, this method is purpose designed to operate within the account specific level for the purposes of benchmarking, analysis and classifying the use of Twitter at the individual level. This paper expands the existing application of Twitter as a data source for understanding marketing communications performance in social media by offering a split level classification scheme that can codify performance of an account holder in the primary domains of conversation, status reporting, information relay, news creation, and phatic communications. The method allows for a richer classification schema by expanding the five categories with further refinement to explore how the account is used for interaction in the social media space. The novelty of the method is the focal point of providing data to assess the performance of an account against the account author’s intended use of the service. The paper concludes with recommendations for the use of the methodology for market segmentation, social media communication objective setting, and as a rich social media metric. ■ SA03 Legends Ballroom III Effects of Online Medium on Consumer Behavior Contributed Session Chair: Jie Zhang, Associate Professor, University of Maryland, 3311 Van Munching Hall, College Park, MD, 20742, United States of America, [email protected] 1 - Disentangling the Effects of Online Shopping Decision Time on Website Conversion Dimitrios Tsekouras, PhD Candidate, Erasmus University Rotterdam, Burg. Oudlaan 50 (room H15-20), Rotterdam, 3062 PA, Netherlands, [email protected], Benedict Dellaert 2 - Gossip: Can It Kill a Giant? Liwu Hsu, PhD Candidate, Boston University, [email protected], Shuba Srinivasan, Susan Fournier Research on total time spent on pre-purchase information search on retail websites has shown contradictory effects on consumer purchase behavior. While some research has emphasized the positive effects of decision time online on conversion due to the fact that it increases the probability that the consumer selects an attractive product as well as decreases choice uncertainty, other studies showed that decision time represents higher consumer effort and an increased opportunity cost and, hence, has a negative effect on conversion. In order to explain these conflicting findings, we disentangle total decision time spent on a retail website into product comparison and product inspection time. We hypothesize that these two aspects have opposing effects on website conversion. First, we expect that spending more time comparing alternatives decreases the chances of conversion. The underlying reason is that comparison time is an indication of the difficulty of the choice set and of choice uncertainty which increases the risk associated with making a decision. Second, we expect that spending time inspecting each given alternative increases the chance of conversion, because it represents a higher expected utility of searching for information on the alternative. Using clickstream data from 16600 consumers that visited a mortgage recommendations website, we find support for the proposed relationships. This underlines the theoretical importance of disentangling total decision time. Online firms can also benefit from our findings to improve the composition of the offered product choice sets by taking into consideration the allocation of time spent by consumers on a given session and by balancing consumers’ time comparing alternatives with time spent inspecting alternatives’ details. Companies are increasingly confronted with new challenges in the era of Web 2.0. In this study, we focus on the role of brand equity (i.e., strong brand vs. weak brand) in an environment characterized by the proliferation of social media and the interactive interface between consumers and brands. Two powerful features of this environment are transparency and criticism (Fournier and Avery 2011). Once a crisis happens (e.g., product-harm recall), negative buzz is typically generated not only in traditional media but also in social media. It is difficult for a company to bury the mistake, and even worse, the social media environment makes consumers more critical of the company and its brand after a crisis. Conventional branding wisdom suggests strong brands can increase the firm’s profits and reduce the vulnerability of cash flows. Yet in the current environment, the bigger the brand the more vulnerable it is. Our objectives are to examine whether a strong brand alleviates the negative impact of a crisis event and to assess how the different metrics of social media moderate the impact of brand equity on the components of shareholder value: the levels of abnormal returns and stock risks. Using the event study method, we examine product recall announcements during a three-year period from 2008 to 2010 and incorporate daily user-generated content metrics such as volume and valence within the event window. We conclude with a discussion that the role of brand in Web 2.0 in risk management terms. 3 - Structural Dynamic Factor Analysis for Quantitative Trendspotting Rex Du, Associate Professor of Marketing, University of Houston, 375E Melcher Hall, Bauer College of Business, Houston, TX, 77204, United States of America, [email protected], Wagner Kamakura 2 - Clicks to Conversion: The Impact of Product and Price Information Vandana Ramachandran, Assistant Professor, University of Utah, KDGB 319, Salt Lake City, UT, 84102, United States of America, [email protected], Siva Viswanathan, Hank Lucas Trendspotting has been an important marketing intelligence tool for identifying and tracking major movements in consumer interest and behavior. Currently, trendspotting is done either qualitatively by “trend hunters” who comb through everyday life in search of signs indicating new trends in consumer needs and wants, or quantitatively by analysts who monitor individual indicators, such as how many times a keyword has been searched, blogged or “twitted” online. In this study, we demonstrate how the latter can be improved by uncovering common temporal patterns hidden behind the co-evolution of a large array of indicators. We propose a structural dynamic factor-analytic model that can be applied for simultaneously analyzing tens or even hundreds of time series, distilling them into a few latent dynamic factors that isolate seasonal cyclic movements from non-seasonal nonstationary trend lines. We demonstrate this novel multivariate approach to quantitative trendspotting in one application involving a promising new source of marketing intelligence – online keyword search data from Google Insights for Search, wherein search volume patterns across 38 major makes of light vehicles were analyzed over a 81-month period to identify key latent trends in consumer vehicle shopping interest. This study seeks to examine how different types of information – product-related and price-related information provided by retailers – impact purchase-related outcomes for consumers belonging to different states of shopping. Using mixture-modeling techniques on clickstream data obtained from a large online durable goods retailer, we find that a three-state model comprised of directed shoppers, deliberating researchers and browsers, best describes the latent differences across customers. In examining the impacts of information on purchase outcomes, we find that product and price-related information impacts consumers in these three shopping states differently. While product information highlighting features of product alternatives in a category has the strongest impact on deliberating researchers, specific price information related to category-level discounts increases the likelihood of purchase for both directed shoppers as well as browsers. Price information relating to site-wide free shipping has a positive impact on purchase for all consumers. Surprisingly, category-level discounts have a negative impact on deliberating researchers, while rich product information hampers the purchase process of directed shoppers. We discuss the managerial implications of our findings and the role of clickstream analytics in designing dynamic targeting and information provisioning strategies for online retailers. 4 - Is All That Twitters Gold? Market Value of Digital Conversations in Social Media Abishek Borah, University of Southern California, Los Angeles, CA, United States of America, [email protected], Gerard J. Tellis Social media are growing to be highly influential. Consumers and marketers are increasingly adopting social media such as Facebook, Twitter, YouTube. Even the president of the United States of America has a Twitter account. Some studies have shown that Tweets forecast movie revenues (Asur and Huberman 2010; Rui et al. 71 SA04 MARKETING SCIENCE CONFERENCE – 2011 3 - Retargeting - Investigating the Influence of Personalized Advertising on Online Purchase Behavior Alexander Bleier, PhD Student, University of Cologne, Albertus Magnus Platz 1, Cologne, 50939, Germany, [email protected], Maik Eisenbeiss 2 - Uncovering the Dynamics of Product and Process Innovation: An Analysis of Dynamic Discrete Games Xi Chen, Hong Kong Unviersity of Science and Technology, Clear Water Bay, Kowloon, Hong Kong - PRC, [email protected], John Dong In an average online shop, the conversion rate of visitors to buyers is about 2%. However, even as most customers leave the site without purchasing, their sales are far from lost. Having traced a customer’s shopping behavior in the online store, a retailer can employ a special form of online advertising, called retargeting, to induce the prospective buyer’s return and purchase completion. With retargeting, the retailer makes use of the previously collected information to provide the customer with personalized banners at various web sites she subsequently visits. Nevertheless, even though this technique offers several advantages over conventional banner advertising and is currently employed by an increasing number of online retailers, it also poses several risks. For example, since personalized banners are more explicitly recognized than conventional banners, the customer may end up feeling annoyed or pursued, especially when the exposure frequency is perceived as too high. As a result, she may be even less likely to return to the retailer’s store than had she received no advertising at all. Analyzing individual level online sales of a major German retailer matched with clickstream data of its web shop and an affiliated advertising agency, we develop a model to explain the effects of retargeting on customers with respect to their click-through and purchase behavior. Accounting for heterogeneity among individuals, we model their immediate and subsequent reactions to targeted banners. The results of our study are of interest to retailers and advertisers alike, as this dynamic form of advertising is still largely unexplored. Innovation is arguably vital to today’s industry competition, as innovative firms sustain competitive advantage in the market. While it has been long to suggest that innovation activity is a process of interactions among firms, a paucity of evidence was found to uncover this dynamic decision-making across time from R&D investment to profit gain. In addition, past research omitted the importance of forward-looking in firms’ making these decisions. How is firms’ R&D investment converted to profit gain through different types of innovation is also unclear in the literature. We develop a dynamic game-theoretic model to formally analyze firms’ forward-looking decisions of R&D investment over time by considering interactions among firms inoligopoly settings.We separate the effects of product and process innovation on profit as these two types of innovation help sales promotion and costs reduction, respectively. Using a unique panel data set fromGermany, we are able to estimate our model and find empirical evidence.This paper contributes to marketing literature by more realistic modeling and examining the dynamics of how firms decide R&D investmentby considering their competitors’ strategic actions and gain competitive advantage through product and process innovation in a framework of dynamic discrete games. 3 - Market Size, Quality, and Competition in Portuguese Driving Schools David Muir, PhD Student, The Wharton School/University of Pennsylvania, 3000 Steinberg Hall-Dietrich Hall, 3620 Locust Walk, Philadelphia, PA, 19104-6302, United States of America, [email protected], Maria Ana Vitorino, Katja Seim 4 - Usage Experience with Decision Aids and Evolution of Online Purchase Behavior Jie Zhang, Associate Professor, University of Maryland, 3311 Van Munching Hall, College Park, MD, 20742, United States of America, [email protected], Savannah Wei Shi Using a novel data set of Portuguese driving exam information from the first half of 2009, we explore the equilibrium market structure of Portuguese driving schools and provide evidence as to whether exogenous or endogenous sunk costs play a role in determining the equilibrium structure of the driving school industry in Portugal. The data reveal that larger markets exhibit less variable mean school pass rates and greater concentration than correspondingly smaller markets, where we define and measure per-municipality market size by the total number of learner’s permits issued, population, or population density. Specifically, we explore the extent to which larger markets have a tendency to be more concentrated due to greater fixed cost investments in quality, which would impose greater barriers to entry to potential entrants in these markets. With greater investments in school quality, we ex-ante expect the mean school pass rates to be higher with lower variance. Here we measure quality as the per-school mean pass rate: schools with higher and less variable pass rates are likened to be of greater quality. Hence, using a variety of concentration and market size measures for robustness, we test Sutton’s theory of endogenous sunk costs (1991) as it relates to investments in quality in the Portuguese driving school market. We further explore alternative theoretical explanations for these and other empirical regularities in the data. A distinct feature of online stores is that they offer a wide range of interactive decision aids which can facilitate consumers’ shopping processes. The objective of this study is to conduct an empirical investigation on how the usage experience with these various decision aids may affect the evolution of consumers’ purchase behavior in the Internet shopping environment. In the context of online grocery stores, we categorize four types of decision aids that are commonly available, namely, those 1) for nutritional needs, 2) for brand preference, 3) for economic needs, and 4) personalized shopping lists. We construct a Non-homogeneous Hidden Markov Model (NHMM) of category purchase incidence and purchase quantity, in which purchase behavior may vary over time across hidden states as driven by usage experience with different decision aids. Our data are provided by a leading Internet grocery retailer which was among the very first to sell groceries online. The dataset was collected during the period when the retailer first launched its web business, which makes it particularly suited to study the evolution of online purchase behavior. Our preliminary results indicate that online consumers evolve through distinct states of purchase behavior over time, and that usage experiences with different decision aids indeed play different roles in the process. Findings from this study will enrich the understanding of how purchase behavior may evolve over time on the Internet, and will provide valuable insights for marketers to improvement the design of online store environments, as well as to modify promotion messages adaptively according to consumers’ evolving purchase behavior. 4 - Investigating Income Dynamics Using the BLP Market Share Model Andre Bonfrer, Professor of Marketing, Australian National University, LF Crisp Building, College of Business and Economics, Acton, 2601, Australia, [email protected], Anirban Mukherjee We examine how changes in household demographics (income dynamics) alter preferences for white goods and the implications for marketers’ pricing and assortment decisions. Prior studies of differentiated products have used a single measure of current income and abstracted from the role of income dynamics on consumer preferences. We use a novel dataset from Household Income and Labor Dynamics in Australia (HILDA) survey to study a number of different ways that income dynamics may affect consumer preferences, including income changes due to exogenous shocks (e.g. arising from public policy decisions such as tax-rebate incentives), permanent versus transitory income changes, and expected versus unexpected income changes. In the empirical work, we build on the BerryLevinsohn-Pakes market share model to allow for a richer representation of the demographic drivers of preferences. We estimate the model on sales data for washing machines and refrigerators in Australia. We find that income dynamics affect both price elasticity and consumer preferences for other attributes (such as energy efficiency). ■ SA04 Legends Ballroom V Structural Models III Contributed Session Chair: Andre Bonfrer, Professor of Marketing, Australian National University, LF Crisp Building, College of Business and Economics, Acton, 2601, Australia, [email protected] 1 - An Equilibrium Analysis of Online Social Content-sharing Websites Tony Bao, PhD Candidate, Cornell University, 360 W 34th St, Apt. 7M, New York, NY, 10001, United States of America, [email protected], David Crandall User-generated content sharing sites have become very popular in the last few years, but very little work has addressed how to model the dynamics of user interactions on these sites. Two distinguishing features of user-generated content – being free and non-rival – preclude application of the celebrated market equilibrium theory. We develop a content equilibrium from first principles. Consumers searching content take the sampling probability distribution as given in deciding consumption, and producers are motivated by attracting endorsements. Sampling probability is a key policy instrument. Endorsement may explain why a small number of producers generate most content. Individual behaviors alone cannot explain genesis and persistence of sampling probability and endorsement. Consumers and producers can be compatible, and their interaction gives rise to endogenous sampling probability and endorsement. Inequality arises: higher quality producers always earn more endorsements and produce more content, and higher quality content is easier to find. We show that despite this inequality, content systems are optimal for consumer welfare. We use this framework to show how content website operators can adjust policies in order to achieve the operator’s objectives. . 72 MARKETING SCIENCE CONFERENCE – 2011 SA06 ■ SA05 ■ SA06 Legends Ballroom VI Legends Ballroom VII Game Theory IV: Signaling Channels VI: General Contributed Session Contributed Session Chair: Yuanfang Lin, University of Alberta, 3-23L Business Building, Edmonton, AB, T6G 2R6, Canada, [email protected] 1 - The Green Monoploist Kyung Jin Lim, University of North Carolina at Chapel Hill, KenanFlagler Business School, Chapel Hill, NC, United States of America, [email protected], Sridhar Balasubramanian, Pradeep Bhardwaj Chair: Joseph Lajos, Assistant Professor of Marketing, HEC Paris School of Management, Groupe HEC, 1 Rue de la Libération, Jouy en Josas, 78351, France, [email protected] 1 - Impact of Consumer Returns on the Manufacturer’s Optimal Returns Policy Thanh Tran, Assistant Professor of Marketing, University of Central Oklahoma, 100 N. University Drive, Department of Marketing, Edmond, OK, 73034, United States of America, [email protected], Ramarao Desiraju Consumers are now increasingly sensitive to the environmental impact of their product choices. In response, firms have adopted multiple approaches to becoming “green.” In this paper, we examine how a monopolist can design a portfolio of carbon footprint reduction strategies. We develop an analytical model of a market with two consumer segments – receptive consumers and skeptic consumers – that differ in their acceptance toward firm’s carbon emission management claims. We demonstrate how a monopolist serving this market can maximize profits by balancing decisions related to carbon offsetting, the degree of processing – which can impact product quality, and price. We delineate key interdependencies between these decisions and other production parameters in terms of their effects on outcomes of interest to consumers, the environment, and the firm. For example, we demonstrate that, surprisingly, improvements in quality efficiency – the quality level associated a certain amount of processing – can increase profits but harm the environment. We also demonstrate that, under certain conditions, an increase in process inefficiency – defined as the amount of emissions associated with a unit of processing – can lead to lower product-level and aggregate emissions. Paradoxically, ‘dirtier’ technologies can lead to greener outcomes. In practice, most retailers accept returns from consumers. Interestingly, much of the extant literature on a manufacturer’s optimal returns policy—-that is offered to the retailers—-does not take into account the returns that retailers accept from consumers. The decision to accept returns from consumers clearly affects the retailer’s stocking and pricing choices; since these choices also impact the manufacturer’s profit, it is important to assess how the manufacturer’s optimal returns policy may be modified. Therefore, we add to the existing literature by developing and analyzing a mathematical model to characterize the inter-relationship between the returns policies of the retailers and the manufacturer. Our analysis reveals that when consumer returns are accounted for, the optimality of accepting returns from retailers depends critically on the magnitudes of the prevailing rate of consumer returns, the degree of consumers’ risk aversion, production cost and the marginal cost of stockouts. In particular, we show that even when demand may be anticipated without error, the manufacturer can sometimes induce its retailers to compete more intensely. We explain the intuition underlying these results and suggest directions for further work. 2 - Mass Behavior in a World of Connected Strangers Jurui Zhang, University of Arizona, 1130 E Helen Street, Room 320C, Tucson, AZ, 85721-0050, United States of America, [email protected], Yong Liu, Yubo Chen 2 - The Effects of Asymmetric Interdependence on Asymmetric Conflict - Using Response Surface Analysis Hyang Mi Kim, Korea University, LG-POSCO Buidling 507, Seoul, Korea, Republic of, [email protected], Jae Wook Kim Networks and the relationships embedded in them are critical determinants of how people communicate, form beliefs, and behave. The emergence of eCommerce platforms such as Amazon and eBay and social media websites such as Facebook and Twitter has made the action of “strangers” more observable to others. At the same time, the actions of “friends” also have become more visible than ever before. Since observation learning and herding behavior are important cognitive and behavioral phenomenon of consumers, it is interesting to examine how different types of networks (e.g., strong vs. weak tie networks) may carry information, both observed and through private signals, and be influential in shaping beliefs and actions. This study investigates how information accumulates and how mass behavior could form in different networks. A theoretical model is developed to examine Bayesian equilibrium of mass behavior through observational learning over weak vs. strong tie networks. We also test the key theoretical results using lab experiments to demonstrate validity and enhance insights. The role of interdependence has been regarded by researchers as an antecedent of many outcomes of various relationships. By reviewing the results of the previous research, we are able to observe mixed results with conflict issues and questioned why. We think several reasons, the first, it is a measurement problem. Even though interdependence involves relative concepts, previous researchers have measured dependence by one party in dyads. Furthermore, operational definitions are differently identified in each research. The second, it is related to theoretical approach; previous studies have its own limitation to explain every situation and condition clearly, and assume a linear relationship between interdependence and outcomes. Therefore, we examine the relationship between interdependence asymmetry and the phases of conflict by using Response Surface Analysis. And we measured dependence directly from bilateral parties and composed various combinations (i.e., bilateral self-perception, unilateral self-perception, and bilateral partner-perception of dependence). Our research gains 96-perfect paired data from the fire-fighting equipment industry. As a result of analysis, the relationship between interdependence asymmetry and the phases of conflict are a nonlinear relationship (i.e., inverted U shape). In the case of bilateral self perception, the results are all significant. But in other data, the results are not. In this study, there is no difference between interdependence asymmetry and phases of conflict, which is not consistent with prior research. Also, the measurement issues have been pointed out prior research, and our results support. Finally, the case of bilateral self perception is the most properly measured composition to identify the role of interdependence. 3 - Informational Effect of Soldout Products on Consumer Search Behavior and Product Evaluation Yuanfang Lin, University of Alberta, 3-23L Business Building, Edmonton, AB, T6G 2R6, Canada, [email protected], Paul Messinger, Xin Ge Consumers often encounter soldout products during the search and purchase decision-making process. The objective of this paper is to explore how the presence of a soldout product provides an informative signal to consumers about the value of the soldout and related products and thereby influences consumer shopping behavior. In particular, we consider the following research questions: (1) Will observing a soldout product option make consumers more willing to search for information about related products in the category? (2) Will the presence of a soldout product increase consumers’ perceived desirability of an available option that has similar attributes to the soldout product? (3) Will consumer preference towards an available product option increase when there is a soldout product in the environment with superior/inferior attributes? To develop hypotheses concerning these questions, we build an analytical model where a consumer’s utility is defined in two-dimensional attribute space for product options in a given category. Given uncertainty about their tastes, consumers are assumed to hold some prior beliefs concerning the exact attribute levels of a product option, and of the corresponding utility weight on each product attribute in the utility function. Through Bayesian updating, the presence of a soldout product option induces consumers to rationally form posterior beliefs regarding those utility parameters for the soldout product and for available options remaining in their choice set. Implications from such a Bayesian updating process thus generate hypotheses concerning the above research questions. We test these hypotheses through a series of laboratory studies to provide behavioral evidence supporting our analytical predictions. 3 - Do Channel Roles and the Sales-distribution Relationship Differ Between Countries? Joseph Lajos, Assistant Professor of Marketing, HEC Paris School of Management, Groupe HEC, 1 Rue de la Liberation, Jouy en Josas, 78351, France, [email protected], Hubert Gatignon, Erin Anderson Although it is intuitive that retailers should increase coverage of products that are selling well, it is somewhat less intuitive that retailers should take the risk of increasing coverage before sales increases have materialized in an effort to push these products on consumers. This is especially true for innovative new durable products which can be both risky and expensive for retailers to hold in inventory. In this research note, we use a simultaneous equation model to analyze sales and distribution coverage of two brands of an innovative new consumer durable in competing types of distribution channels in four European countries in order to examine whether retailers in different countries make their coverage decisions for a new durable product in the same ways, and if not, to explore the conditions under which they are likely to engage in market-making. 73 SA07 MARKETING SCIENCE CONFERENCE – 2011 ■ SA07 4 - An Empirical Study of the Effects of Production Timing Decisions on Movie Financial Performance Tom Fangyun Tan, PhD Student, Wharton Business School, University of Pennsylvania, 3901 Locust Walk, MB 253, Philadelphia, PA, 19104, United States of America, [email protected], Kartik Hosanagar, Jehoshua (Josh) Eliashberg Founders I Entertainment Marketing I: Movies Contributed Session Chair: Tom Fangyun Tan, PhD Student, Wharton Business School, University of Pennsylvania, 3901 Locust Walk, MB 253, Philadelphia, PA, 19104, United States of America, [email protected] 1 - Demand Lifting through Pre-launch Marketing Activities Shijin Yoo, Assistant Professor of Marketing, Korea University Business School, Anam-dong Seongbuk-gu, Seoul, 136-701, Seoul, Korea, Republic of, [email protected], Tae Ho Song, Janghyuk Lee We study the impact of pre-release decisions on the box office success of movies. Specifically, we focus on the impact of the timing of production and marketing activities. We find that the total production time for a movie is negatively associated with its box-office revenues. Further, we find that the duration of different production stages, namely pre-production, filming, post-production and distribution, have different impacts on the box-office revenues. For example, the length of the distribution stage has the most negative impact. Finally, we find that coordination between the timing of marketing and production activities plays an important role in determining the box office success of movies. Our findings underscore the value of streamlining and coordinating the operational and marketing activities for movies. A successful new product launch is one of key missions for marketers. However, a shortened product life cycle brings challenges to marketers in many industries (e.g., electronics) in developing an optimal launching strategy. Research has been relatively limited to understand the role of pre-launch marketing activities such as ads to secure the minimum demand level required for generating upward spiral between demand and word of mouth (WOM). We propose that there exists a minimum level of latent demand at the pre-launch stage that will generate positive WOM to accelerate demand realization, which is directly influenced by pre-launch marketing activities. We introduce a two-stage demand model: (1) a pre-launch stage where the latent demand is generated by pre-launch marketing activities, and (2) a post-launch stage where the lagged demand is realized with temporal and confirmatory decay due to the discrepancy between consumers’ expectation and the revealed quality of a product, which is also affected by post-launch marketing activities and WOM. Therefore the marketer should decide the intensity of pre-launch marketing activities considering this complex demand dynamics. We apply the proposed model to the dynamics of movie demand in Korea by collecting data of weekly box-office sales, offline advertisement, on-line advertisement, and on-line word of mouth – rating, valence, and volume – from two major internet movie sites. The current research discloses how pre-launch marketing activities affect the creation of and the relationship between WOM and demand after launch. This finding will provide marketers with new insights to find an optimal level of pre-launch marketing activities. ■ SA08 Founders II Continous-Time Marketing Cluster: Special Sessions Invited Session Chair: Olivier Rubel, University of California-Davis, Davis, CA, United States of America, [email protected] 1 - Life-cycle Channel Coordination Issues in Launching and Innovative Durable Product Xiuli He, University of North Carolina-Charlotte, Belk College of Business, Charlotte, NC, 28223, United States of America, [email protected], Gutierrez J. Gutierrez We analyze the dynamic strategic interactions between a manufacturer and a retailer in a decentralized distribution channel used to launch an innovative durable product (IDP). The underlying retail demand for the IDP is influenced by word-of-mouth from past adopters and follows a Bass-type diffusion process.The word-of-mouth influence creates a trade-off between immediate and future sales and profits, resulting in a multi-period dynamic supply chain coordination problem. Our analysis shows that while in some environments, the manufacturer is better off with a myopic retailer. We characterize equilibrium dynamic pricing strategies and the coordinate the IDP supply chain with both far-sighted and myopic retailers throughout the entire planning horizon and arbitrarily allocate the channel profit. 2 - Awareness and Preference-based Consumer Segmentation in Forecasting Movie Box-office Performance Sangkil Moon, Associate Professor, North Carolina State University, College of Management, Raleigh, NC, 27695, United States of America, [email protected], Barry Bayus, Youjae Yi, Junhee Kim Forecasting consumers’ new product adoption has been investigated by numerous innovation and marketing researchers primarily targeting durables and repeatedly purchased products. By comparison, forecasting consumers’ adoption of entertainment products such as movies and books has been scarce because of those products’ properties of one-time purchase and hedonic experience consumption. The unique properties make it ineffective applying common adoption models developed for durables or repeatedly purchased products. Based on the industry practice that movie studios use weekly survey data containing consumers’ awareness and preference (AP) measures of new upcoming movies to forecast their box-office performance, we develop a theory-driven forecasting model based on the AP measures of such entertainment products. Specifically, our forecasting model captures four distinct AP-based consumer segments that can influence the new product sales performance in different manners. In other words, our forecasting model is based on our assumption that not only the nature of preference (positive vs. negative preference) but also new product awareness timing (early vs. late awareness) influences the sales differently. Since awareness and preference take place in two successive steps before new product adoption, this two (early vs. late awareness) by two (positive and negative preference) classification results in four distinct consumer groups in sales forecasting. Our movie-level forecasting model reveals that these four groups have distinctively different impacts on new product sales. In our empirical application, we demonstrate the distinct existence of the four consumer segments using recent data from the Korean movie market. 2 - An Exact Method for Estimating Structural Continuous-time Models with Discrete-time Data Prasad A. Naik, Professor, University of California-Davis, Graduate School of Management, Davis, CA, United States of America, [email protected] In continuous-time models, the time index t is continuously differentiable on any interval However, market data arrive at discrete points in time (e.g., weeks). That is, in the data series, the time parameter tk is not continuously differentiable; rather it takes discrete values in the integer set k={1, 2, 3, …,T}. Extant dynamic models ignore this distinction, apply the first-order difference approximation, and then estimate model parameters using regression analysis. In contrast, this paper develops an exact method to estimate simultaneously not only the continuoustime sales dynamics, but also the implied Nash equilibrium feedback advertising strategies using discrete-time data. Thus the proposed method, first, extends the previous work in marketing on estimating continuous-time univariate sales models (Rao 1986) and multivariate sales models (Naik, Raman and Winer 2005) by incorporating feedback strategies in a structural manner. Second, we show how to make robust inferences to safeguard against unknown misspecification errors via Huber-White sandwich estimator. Third, we gain conceptual insights that the sales decay is not restricted to the unit interval as in discrete-time estimation; that the usual carryover effects depend on the sales decay, ad effectiveness and brand profitability. Finally, we illustrate the application of the proposed method by analyzing data from the triopoly cars market. 3 - Can Star Actors and Directors Reduce the Risk of Box Office Failure? An Analysis of Risk Effects Alexa Burmester, University of Hamburg, Welckerstr. 8, Hamburg, 20354, Germany, [email protected], Michel Clement, Steven Wu 3 - Advertising Investments under Competitive Clutter Olivier Rubel, University of California-Davis, Davis, CA, United States of America, [email protected] The movie industry invests heavily in star actors and directors, using them as quality signals to consumers. Star actors and directors know that they signal value to the audience and ask for high salaries, resulting in substantial financial risks for movie studios. However, previous literature has drawn very little consideration to the box office risk effects of actors and directors. This study provides (1) a methodological framework to measure the risk of input factors on any outcome variable and (2) presents an empirical analysis of the risk-shifting power of star actors and directors with respect to the box office, including moderating effects using quantile regressions. Using a sample of 2,109 movies, the authors find that higher actor power increases the risk of the box office outcome, while the overall effect on the box office risk is constant for the director power. The results further indicate that the effect size of director power is higher than the effect size of actor power. Thus, investing in director power is less risky and more effective. Behavioral and empirical studies document that competitive clutter erodes firms’ advertising effectiveness in stimulating sales. When a brand increases its spending to generate more sales, it also adds to the clutter endured by its competitors. Despite the vast literature on advertising competition, the question of how firms should optimally manage their advertising investments under competitive clutter remains open. To fill this gap, we propose a dynamic competitive model where firms’ ad spending tarnish the effectiveness of competing brands’ advertising. We analytically characterize the optimal strategies and empirically validate the dynamic sales response function. Doing so, we gain three insights. First, firms de-escalate their advertising investments under competitive clutter rather than escalate because their ad spending becomes strategic substitutes. Second, we show that firms should react to competitive advertising attacks only if these attacks impact own advertising effectiveness, and not if they change the “drift” of the sales dynamics. Finally, firms’ reactions in face of the clutter 74 MARKETING SCIENCE CONFERENCE – 2011 SA10 ■ SA10 differ depending on their respective clout and vulnerability, i.e. some firms increase their ad spending when they become more vulnerable to competitive clutter whereas other decrease their ad spending. Founders IV Health Care Marketing I Contributed Session ■ SA09 Chair: Yansong Hu, Assistant Professor, Warwick University, Warwick Business School, MSM Group, Coventry, CV4 7AL, United Kingdom, [email protected] 1 - Future Challenges for eHealth Concept Based on Market Analysis Lenka Jakubuv, PhD Student, Czech Technical University in Prague, Faculty of Biomedical Engineering, Czech Republic, Sìtná sq. 3105, Kladno 2, Kladno, 272 01, Czech Republic, [email protected], Juraj Borovsky, Karel Hana Founders III Retailing VI: Auto Industry Contributed Session Chair: Tae-kyun Kim, Assistant Professor, Rutgers Business School, 1 Washington Park, Room 993, Newark, NJ, United States of America, [email protected] 1 - Financial Incentives and Adoption of Hybrid Cars Sriram Venkataraman, Emory University, Goizueta Business School, 1300 Clifton Road NE, Atlanta, GA, 30322, United States of America, [email protected], Anindya Ghose In recent years the healthcare costs in the Czech Republic rapidly increased. In 2006 there were expenditures related to the patients suffering cardiovascular diseases of about 1,097 mld. € (EU 110 mld. €), which is approx. 13% (EU 10%) of the overall costs for health care. A high potential for future can be seen especially in certain sophisticated technology applications, where the eHealth concept represents new opportunities of this product idea. The main aim is to present results of market analysis, which was focused on implementation of Telemonitoring system in the Czech Republic. On the basis of situation mapping and analysis regarding questionnaires and surveys of target market players there were defined goals and suggested appropriate marketing strategy. The service concept and characteristics, placement, pricing politics and other marketing aspects were discussed. For the purpose of further research are assumed next methods of economic evaluation (CBA) as well as industrial engineering tools (ABC) and approach of Health Technology Assessment. Several clinical trials already showed benefits of Telemonitoring in comparison with conventional care in the term of prevention and reduced mortality. The application of eHealth system may have a valuable role both in quality and economic aspects for healthcare provided in Czech Republic. The research was supported by HEDF grant nr. 123/2011 and realized also within the marketing courses at FBME CTU. In the United States, federal, state and local governments offer financial incentives to spur adoption of hybrid vehicles. These incentives vary across auto models, geographic markets, and over time. We examine the extent to which these financial incentives impact hybrid-car adoption rates after controlling for local retail marketstructure, local preferences, market-specific time-varying factors and so on. Using a rich new dataset containing sales of new-vehicle sales (hybrids and non-hybrids) across all states in the U.S. between years 2005 and 2010, we examine the synergistic effects of these incentives and show to what extent coordination incentives can accelerate hybrid technology adoption. Because the auto industry has had a massive impact on the environment and hybrid cars are playing a critical role in the current era of the green transportation revolution, our study can provide valuable insights for auto manufacturers and policy-makers alike. 2 - Auto Industry Crisis and Firm Outcomes O. Cem Ozturk, PhD Candidate, Emory University, 1300 Clifton Rd. NE Suite W462, Atlanta, GA, 30322, United States of America, [email protected], Sriram Venkataraman, Pradeep Chintagunta 2 - Exploring Relationships Among Marketing Effort, Customer’s Personality and Hospital Brand Experience Ravi Kumar, Indian Institute of Management, Lucknow, FPM Office, Prabandh Nagar, Off Sitapur Road, Lucknow, 226013, India, [email protected], Shailendra Singh, Prem Purwar, Satyabhushan Dash The U.S. automobile industry has been severely impacted by the current global economic crisis. So much so that the U.S. Treasury Department had to step in and rescue industry giants like Chrysler and GM. This study formally examines the impact of large-scale market-structure changes triggered by these aforementioned events on both consumer demand and the retailer marketing-mix decisions. Our empirical analyses show that these market-structure changes have indeed had an economically significant impact on both consumer demand and retailers’ marketing mix. Furthermore, these effects vary across brands (e.g. Ford vs. GM) and within the same brand across product categories (e.g. sedan vs. truck). Taken together, our findings provide valuable insights for auto manufacturers, auto dealers and policy-makers alike. Brand experience is a multi-dimensional phenomenon that encompasses sensory, affective, behavioral and intellectual experiences of customer and is evoked by functional, mechanic and humanic clues provided by firm. It is the subjective internal and behavioral responses to firm generated stimuli. The present study aims to conceptually analyze brand experience and its antecedents for the healthcare industry. Specifically, the study intends to assess the effects of firm’s marketing effort in creating long-lasting hospital service brand experience and explores the moderating role of patient’s big five personality trait between marketing effort and brand experience. Based on the existing literature on marketing mix elements, customer experience and big five personality trait, and qualitative in-depth interviews with Indian patients, the study proposes a causal framework of hospital brand experience. The framework links marketing effort (e.g. technical process, administrative procedure, interpersonal care, healthscapes management, medical service, service charges, communication, access convenience, social responsibility) to hospital brand experience. Patient’s big five personality trait is presented as moderating variable between marketing effort and brand experience. Testable hypotheses have been developed for empirical validation. This paper provides vital insights about the relevance of different marketing efforts and experiential marketing in creating unique, memorable and sustainable brand experience. The proposed framework can be used in gauging the effectiveness of firm’s marketing decisions and strategy for different customer segment. This work is an offshoot of ongoing recent research on customer experience and enhances current understanding of brand experience. 3 - Variation in Retailer Competition in Durable Goods Markets: An Empirical Study Tae-kyun Kim, Assistant Professor, Rutgers Business School, 1 Washington Park, Room 993, Newark, NJ, United States of America, [email protected], S. Siddarth, Jorge Silva-Risso There is considerable evidence that retailers directly compete with each other at a local level. For example, retailers make strategic choices about the overall pricing formats (Lal and Rao 1997; Bell and Lattin 1998) and retail competition is shown to be one of the most important factors in explaining price variations across stores (Shankar and Bolton 2004). Recent research on the automobile market shows that the extent to which consumer demand depends on a brand’s local distribution intensity (Bucklin, Siddarth, and Silva-Risso, 2008) and dealer location plays an important role in a consumer’s vehicle choice decisions (Albuquerque and Bronnenberg, 2006). Because there is significant heterogeneity in the intensity of distribution among different brands and, across different geographical areas of the same brand, it stands to reason that the nature of competition among retailers will also depend upon local conditions. Recent actions by auto manufacturers like GM and Chrysler to selectively reduce the number of dealers in specific areas also suggests that dealer competition is likely to vary by market. We define two different types of horizontal interaction (HI) as follows. Intra-channel HI refers to the competition between dealers of the same nameplate (e.g. between different Toyota dealers) and Inter-channel HI to the competition between dealers selling cars of different nameplates (e.g. Toyota Dealers versus Honda Dealers). A structural model of retailer competition is proposed to infer both types of HI and estimated using consumer automobile purchase data. We find that (i) inter- and intra-channel competition in the same market can be quite different, and (ii) the nature of intra-channel competition itself varies from one market to the other. 3 - Offering Pharmaceutical Samples: The Role of Physician Learning and Patient Payment Ability Ram Bala, Indian School of Business, Gachibowli, Hyderabad, India, [email protected], Pradeep Bhardwaj, Yuxin Chen Physicians learn about drug effectiveness either through detailing or free drug samples. They administer these samples to patients as follows: first, as a subsidy to those who may not have the ability to pay and second, to better match a drug with a patient when detailing information alone proves inadequate. However, a profit maximizing firm prefers the opposite sequence. We characterize the sampling decision in both monopoly and competitive settings and find that the optimal or equilibrium sampling level as a function of patient payment ability displays an inverted-U shape. We also study the social welfare implications of sampling. In particular, we find that the use of samples enhances welfare only at intermediate levels of patient lifetime value to the pharmaceutical firm. 75 SA11 MARKETING SCIENCE CONFERENCE – 2011 3 - You May Have Influenced My Next Purchase: Social Influence in Food Purchase Behavior Jayati Sinha, PhD Student, University of Iowa, S252 John Pappajohn Business Building, Iowa City, IA, 52242-1000, United States of America, [email protected], Gary J. Russell, Dhananjay Nayakankuppam 4 - From Invention to Innovation: Technology Licensing by New Ventures in the Biopharmaceutical Industry Yansong Hu, Assistant Professor, Warwick University, Warwick Business School, MSM Group, Coventry, CV4 7AL, United Kingdom, [email protected] Licensing out technologies is a major source of revenue for new technology ventures. While technology licensing is economically important and poses unique challenges, it has yet received little attention in past marketing literature. We analyze how alliances with established firms, structural positions within scientific research, patent and commercial networks, and the nature of technology portfolios (i.e., radical and incremental innovations) of new ventures influence the number of successful licensing deals secured by the new ventures. We draw on theories of structure and legitimacy of new ventures to conjecture how the number and nature of alliances, centrality positions within scientific research, patent and commercial networks, and the composition of technology portfolios can affect a new venture’s number of licensing deals, and how these effects may vary over time. We study these key issues by examining all public firms in the UK biopharmaceutical industry from 1989 to 2009. Because we observe overdispersion in the data, we specify a negative binomial model to investigate the conjectured effects as drivers of successful deals, and provide a sharper understanding of the relations between the collaboration and co-creating with innovation communities, structural positions within innovation networks, the trade-off between radical and incremental innovation portfolios, and time of the specific innovation. Social influence is an important aspect of consumer decision making. Social influence can manifest itself in two ways: social interaction effects (interactions among neighbors) and neighborhood effects (influence of spatially proximate others). In this research, we argue that, beyond commonly evoked effects of the socio-physical environment, neighborhood social interaction patterns may have a decisive influence on food consumption behavior. Moreover, due to the nature of these social processes, this phenomenon should be more likely to emerge for product categories that are socially salient. We test these hypotheses by fitting a conditional autoregressive (CAR) spatial model to purchase data collected in two product categories: organic foods (socially salient) and non-organic meat and poultry (not socially salient). Although both categories provide evidence for the existence of neighborhood effects, only organic foods exhibit clear evidence that the level of social interaction drives purchases. We discuss the implications of these results for the effectiveness of social media in marketing. In particular, we argue that socially expressive product categories (such as green products) are better candidates for marketing strategies incorporating social media. 4 - Intra and Cross-household Influences as Predictors of Individual Consumption Jose-Domingo Mora, Assistant Professor of Marketing, University of Massachusetts Dartmouth, 285 Old Westport Road, CCB 219, North Dartmouth, MA, 02747, United States of America, [email protected] ■ SA11 Champions Center I Social Influence I There is clear evidence in marketing scholarship on the networked nature of the demand side of markets. There have been few attempts though at incorporating such influences in models predicting individual consumption. Major difficulties in this regard are posed by data sets reporting individual behaviors, not one-on-one interactions, and by the endogenous relationships between the amounts of consumption of interacting individuals. We present a random coefficients model of individual consumption where intra-household and cross-household influences are captured by proxy variables. This model is estimated on a cross-section of peoplemeter data reporting individual consumption of television programs. Program consumption of wives is defined as the DV. Building on Lull (1980, 1982); Yang, Narayan & Assael (2006) and Yang, Zhao, Erdem & Zhao (2010) intra-household influences between wives and family members are captured as co-viewing of television programs. Two alternate operationalizations of co-viewing are tested. Consistent with research in social networks, levels of socio-economic status by city are assumed as independent “cells” where social exchanges take place among wives. The social multipliers for each of these cells, subscripted by program genre, are estimated using Graham’s (2008) linear-in-means model of social interactions. The estimated social multipliers enter the RCM of individual consumption of wives as proxy variables for cross-household influences. A non-hierarchical specification of this RCM is compared with a hierarchical specification where social influences mediate the effects of demographics and program characteristics. Contributed Session Chair: Jose-Domingo Mora, Assistant Professor of Marketing, University of Massachusetts Dartmouth, 285 Old Westport Road, CCB 219, North Darthmouth, MA, 02747, United States of America, [email protected] 1 - The Silent Signals: Implicit User Generated Content and Implications for Consumer Decision Making Sunil Wattal, Assistant Professor, Temple University, 1810 N. 13th Street, Philadelphia, PA, 19122, United States of America, [email protected], Anindya Ghose, Gordon Burtch The body of work on user-generated content (UGC) has burgeoned of late. However, much research on UGC focuses on explicit contributions (e.g. discussion boards and online product reviews). There is a notable lack of consideration for implicit forms of UGC in this body of work, where statements of approval or disapproval are implied by users’ actions. Much of this work also lacks a holistic consideration of the process by which users consume such information and render their decisions. To examine these phenomena, we analyze panel data drawn from a crowd-funded marketplace that enables authors to pitch their news article ideas to the crowd, and to subsequently obtain investment from them to pursue the article’s research and publication. We propose two novel measures of implicit UGC: investment frequency (the number of supporters per day of funding) and density (the number of supporters per dollar of funding). We find that users’ likelihood of investing in a project decreases with the frequency of prior support, suggesting anti-herding or contrarian behavior manifests in this context. Further, we find that users’ likelihood of investing increases with density of support, implying that users are not swayed by abnormally large contributions, perhaps because they are presumed to be contributions by the author’s friends and family, or by the author themselves. We discuss the implications of our findings for practitioners and scholars dealing with implicit UGC, and identify avenues for subsequent research. ■ SA12 Champions Center II Online Word of Mouth Research Contributed Session Chair: Mounir Kehal, Research Faculty and Professor of BIS and Computing, ESC Rennes School of Business, 2, rue Robert d’Arbrissel, CS 76522 Cedex, Rennes, 35065, France, [email protected] 1 - Get Something for Nothing: Designing Optimal Free Sampling Strategy for Online Communities Shuojia Guo, Rutgers University, 1 Washington Park, Newark, NJ, 07102, United States of America, [email protected], Lei Wang, Yao Zhao 2 - A Model of Social Dependence and Intra-group Interaction Youngju Kim, Doctoral Student, Korea University, Business Main Building, Seoul, 136-701, Korea, Republic of, [email protected], Jaehwan Kim, Neeraj Arora To better explain group decision outcomes, we attempt to capture both social influence on each member and the intra-group interaction among members when they are engaged in choice decision. Since various purchases are made by groups in market place, and their choices are affected by individual member’s preference, which is also affected by other people around each members through her/his own network structure, capturing both effects is necessary. It is expected that ignoring these sources will result in misinterpretation of the individual and group behaviors. Through simulation study and empirical estimation against real data, we found that the model neglecting either of the two influences gives rise to biases in three ways: First, the model systematically underestimates the importance of factors driving consumer choices. Second, the model systematically underestimates the heterogeneity among decision units. Third, the group-level parameters, power of the group members on their group decision, are also distorted. Firms often offer free samples to attract customers. Free samples may affect sales in three ways. First, the sampling event itself raises product and brand awareness among customers. Second, the samples reveal true product quality and attract repeated purchases from satisfied customers. Third, customers who have the chance to try the free samples may spread positive or negative word-of-mouth to other potential customers. In this paper, we empirically investigate the causal effect of a free sampling event on a product’s sales. Specifically, we decompose the sampling effect into three effects: awareness effect, experience effect, and word-of-mouth effect. Using data from the largest Chinese e-commerce website taobao.com, we identify factors that affect these three effects and discuss optimal sampling strategies in terms of optimal size of free samples and optimal product pick for multi-product retailers. 76 MARKETING SCIENCE CONFERENCE – 2011 SA13 2 - The Impact of Online Referrals on Consumer Choice in the Context of Charity Donations Kyuseop Kwak, University of Technology Sydney, P.O. Box 123, Broadway, 2007, Australia, [email protected], Luke Greenacre, Valeria Noguti, Alicia Tan 2 - Measuring Cross-category Spillover Effects of Private Label Branding in U.S. Supermarket Retailing Sophie Theron, Graduate Student, Arizona State University, 7171 E. Sonoran Arroyo Mall, Mesa, AZ, 85212, United States of America, [email protected], Timothy Richards, Geoffrey Pofahl The growth of online communications has encouraged consumers to base their purchase choices on the opinions of others (Mayzlin and Godes 2004; Pitta and Fowler 2005). Using the views and experiences of others, consumers are able to learn about product offerings (Chen and Xie 2008).Such online referrals assist consumers during their information search process and ultimately with their purchase decisions (Smith, Menon et al. 2005). This research examines the effect of online referrals on consumer decision making. A general model is developed using theories of social influence. This model decomposes an online referral into two parts: the recommendation of the referrer, and the actual purchase behavior of that referrer. The perceived credibility of the referral is also examined through the manipulation of the referrer’s popularity as an information source, and the rated usefulness of their referral by other consumers. The effect of the online referral attributes and credibility indicators on choice behavior is observed using a Discrete Choice Experiment in the context of charity donations. The results show that recommendations and the prior donation behavior of the referrer, as the key characteristics of an online referral, influence the decisions of respondents. It is found that respondents typically follow the recommendation of the referrer. Interestingly though, respondents are less likely to donate to the charity that was donated to by the referrer. However, when the referral is deemed credible through a high message usefulness rating from other consumers, respondents invert their behavior and follow the ‘herd’; donating to the charity the referrer donated to. This presents a rich account of the role of online referrals in consumer choice. One of the most important developments in food retailing over the past 20 years has been the expansion of private labels. According to The Nielsen Company, private labels reached an all-time high of 23.7% of total grocery units sold in 2009. Numerous studies have sought to explain this growth, looking at both consumer attributes and category features as potential causal factors. However, while many of these studies examine private labels in multiple categories they do not consider intercategory relationships or potential spillover effects that may contribute to the overall growth of store brand sales. The objective of our research is to estimate the extent of demand/pricing interrelationships among private label products across categories and, thereby, identify incentives for private label proliferation that have previously been ignored. Our goal is to describe the composition of an entire shopping basket, and why so much of it is increasingly occupied by private labels. Our model consists of two components, which are linked through the common utility maximization process: (1) a brand choice probability, and (2) a joint category-choice probability. We extend the models of Song and Chintagunta (2007) and Mehta (2007) with respect to brand choice and category incidence to capture the effect of private label pricing and promotion across multiple categories. Our model estimates interrelationships among categories within a chosen store and among brands within these categories as functions of the “umbrella branding” power of a retailer’s private label strategy. We estimate the model described above using several food categories from the 2008 Nielsen Homescan data. 3 - What Drives Private-label Margins? Anne ter Braak, Tilburg University, Warandelaan 2, Tilburg, Netherlands, [email protected], Inge Geyskens, Marnik G. Dekimpe 3 - eWom Conducing Text-based Knowledge Diffusion through the Social Web: An Empirical Study Mounir Kehal, Research Faculty and Professor of BIS and Computing, ESC Rennes School of Business, 2, rue Robert d’Arbrissel, CS 76522 Cedex, Rennes, 35065, France, [email protected] Extant research shows that retailers earn higher percentage margins on private labels (PLs) than on national brands (NBs), and that PL margins vary across categories. However, retailers’ PL margins may also differ within categories, since (i) multiple PL suppliers often co-exist within a category and (ii) multi-tiered PLs are increasingly being introduced. Using a unique dataset containing PL supplier information, wholesale, and net selling prices for the complete PL assortment of a major Dutch retailer, combined with GfK scanner panel data, we show that retailers’ percentage PL margins differ within categories based on characteristics of both the PL supplier and the PL quality tier. Specifically, we find that a retailer obtains lower PL margins from dual branders - NB manufacturers that produce significant amounts of NBs and PLs for the retailer - as compared to dedicated PL suppliers or NB manufacturers that produce only very little PLs. Moreover, a retailer’s PL margin is affected by a supplier’s power and the length of the retailer’s relationship with that supplier. However, this effect is not monotonic, but displays a U-shaped form. Finally, a retailer obtains lower margins for the lower-quality economy PL than for the standard PL. While being a low-cost provider is important, our findings suggest that price is not the only criterion taken into account by the retailer when selecting a PL supplier. Retailers value the quality control offered by NB manufacturers that also produce a significant amount of PLs, and are also aware of the benefits obtained through longer-term relationships with PL suppliers. Finally, when a retailer is more dependent on a PL supplier (i.e. the supplier is more powerful), the PL supplier is less likely to be ‘squeezed’ by the retailer. eWOM (Electronic Word of Mouth) is a growing discipline of investigation, concording principles and methodologies from international marketing, to virtual consumer behaviour, computational sciences and to text mining, to mention a few. As it permeates a feel and diversity vis a vis its traditional counterpart that of WOM. Management practices and information technologies to handle knowledge of retail (in particular eRetail) organizations may prove to be complex. As such knowledge (with its explicit and tacit constituents) is assumed to be one of the main variables whilst a distinguishing factor of such organizations; amidst those specialist in nature, to survive within a marketplace. Their main asset is the knowledge of certain highly imaginative individuals that appear to share a common vision for the continuity of the organization. Systematic (corpus-based) studies – through analysis of specialist text, can support research in specialist knowledge management, the marketing of knowledge and the knowledge of marketing. Since text could be assumed to portray a trace of knowledge. In this paper we are to show how knowledge diffuses in a specific environment, and thus could be modelled by specialist text. That is dealing with the customers’ feedback and corporate communication, and having embedded within the specialist domain knowledge (formalized marketing know-how) and eWOM (informal marketing know-what) about the business sector and brands – based on a listing of sectors to be covered and outlined accordingly. 4 - The Dynamic Impact of Increasing Price-gap And Assortmentimitation on Private Label Performance Murali Mantrala, Professor of Marketing, University of Missouri, Trulaske College of Business, 438 Cornell Hall, Columbia, MO, 65211, United States of America, [email protected], Elina Tang, Srinath Beldona, Shrihari Sridhar, Suman Basuroy ■ SA13 Champions Center III Private Labels I: General With their private labels (PL) retailers often imitate the assortment of the national brand/s (NBs). For example, they leave little or no “assortment gap” between the NB and PL product assortment. While the rationale is to signal comparable quality at lower prices, empirical research on the efficacy of assortment imitation is sparse. Considering the significant costs (e.g., inventory & shelf space usage) associated with assortment decisions, understanding how the degree of imitation impacts the PL’s performance is important. Extant research has mainly focused on NB-PL price gaps, e.g., whether increasing price gaps always leads to increased PL performance. Additionally, research to date has examined short-run impacts of PL strategies. There is no research documenting the long-run or dynamic impact of either price gap or assortment imitation strategies, which can provide richer assessments. Accordingly, the authors empirically investigate the relative long-term effectiveness of NB-PL price-gaps vs. assortment-imitation gaps in generating PL sales, with weekly data from two major product categories over 95 weeks obtained from a national supermarket chain. They find that increasing assortment-imitation leads to positive (negative) long-run effects on PL sales when category knowledge & involvement are low (high), but negative accumulated long-run effects when category knowledge & involvement are low. Interestingly, the opposite result is true of increasing price-gaps. The authors subsequently discuss the implications for marketing theory and practice. Contributed Session Chair: Murali Mantrala, Professor of Marketing, University of Missouri, Trulaske College of Business, 438 Cornell Hall, Columbia, MO, 65211, United States of America, [email protected] 1 - Implementing Online Store for National Brand When Competing Against Private Label Naoual Amrouche, Assistant Professor of Marketing, Long Island University, 1 University Plaza, Brooklyn Campus H700, NY, 11201, United States of America, [email protected], Ruiliang Yan We propose a game-theoretic model in three contexts. First, only the national brand (NB) is offered through a traditional retailer. Second, the private label (PL) is introduced by the traditional retailer. Finally, the NB manufacturer opens an online store. We reassess the benefit of introducing the PL and investigate the profitability of implementing an online store. We found that the retailer is not always enjoying the introduction of the PL. Also, the manufacturer could benefit from such strategy. The quality differential between the NB and the PL, the PL’s potential and the cross-price competitions are all important factors to determine the result of such strategy. Hence, the manufacturer opens the online store either to counter the threat of such strategy or to expand his market. 77 SA14 MARKETING SCIENCE CONFERENCE – 2011 ■ SA14 ■ SA15 Champions Center VI Champions Center V Marketing Strategy II: Firm Performance CRM VII: Customer Equity Contributed Session Contributed Session Chair: Soumya Sarkar, Indian Institute of Management Calcutta, D H Road, Joka, Kolkata, 700104, India, [email protected] 1 - Various Strategic Orientations: Theoretical Comparison, Construct Refinement and Empirical Analyses Christian Hoops, Research Associate, TU Dortmund University, Department of Marketing, Vogelpothsweg 87, Dortmund, Germany, [email protected], Michael Bücker Chair: Anita Basalingappa, Associate Professor, Mudra Institute of Communications (MICA), Shela, Ahmedabad, 380058, India, [email protected] 1 - The Formation of Impulse Buying: A Perspective on Self-control Failure of Consumer Behavior in CRM Kok Wei Khong, Associate Professor, University of Nottingham Malaysia Campus, Jalan Broga, 43500 Semenyih, Selangor Darul Ehsan, Malaysia, [email protected], Hui-I Yao Many researchers argue that market, entrepreneurial and learning orientation are the three major strategic orientations which affect organizational performance (Hurley/ Hult 1998, Lumpkin/ Dess 1996). However, only a few studies have investigated the antecedents, moderators and consequences of interaction orientation. This study fills the gap and compares various orientations with each other. According to Ramani/ Kumar (2008), interaction orientation reflects the ability of a company to interact with the individual customer and to gather information from successful interactions. The authors identify four dimensions of the construct: customer concept, interaction response capacity, customer empowerment and customer value management. Exploratory depth interviews concluded that there is another sub-construct conceptualized as the understanding of the customer’s problem. Our factor analysis shows that indeed a fifth dimension exists. Furthermore, this paper analyzes the influence of all verified and many new antecedents of interaction orientation, categorized into seven groups. Different regression analyses show that the organizational culture has the greatest impact on this orientation. The first notable finding is that B2B firms exhibit a greater degree of interaction orientation than B2C firms. Ramani and Kumar hypothesized that in their study. We show that there are B2C industries such as financial services, whose companies also have a greater interaction orientation. This could be the reason why the authors could not prove their hypothesis. Finally, this study examines the influence of strategic orientations on organizational performances such as product innovation or effectiveness and rejects synergetic effects between interaction and market orientation. Customer Relationship Management (CRM) has become the principal modern marketing approach in both research and practice. One of the goals of CRM is to manage customer relationships proactively in the long run. However, evidence shows that impulse buying is a hefty chuck of spending. This study reviews the formation of impulse buying in consumer behavior. Three major ingredients, standards, a monitoring process, and the capability to modify one’s behavior, contribute to selfcontrol failure, which leads to impulse buying that probably generates purchase regret as well as harming a firm’s reputation. Implications for CRM of insights into impulse buying are discussed. 2 - Monetizing UGC: A Hybrid Content Approach Theodoros Evgeniou, INSEAD, Blvd de Constance, Fontainebleau, 77300, France, [email protected], Kaifu Zhang, Paddy Padmanabhan, Inyoung Chae The monetization of User-Generated Content (UGC) continues to be a challenge for many websites. Many advertisers are hesitant to embrace UGC-based advertising because of the large uncertainty about content quality. In this paper, we investigate the emergent Hybrid Content Strategy, wherein a firm hosts both UGC and Professional-Generated Content (PGC). We argue that the firm can explore the synergy between two types of content in order to better monetize both. For example, the firm can utilize UGC mainly for consumer retention or mainly for consumer acquisition, while generating advertising revenue from PGC - or vice versa. The firm can in general shift customers between UGC to PGC. Data from a large portal that has both PGC and UGC are used. We also discuss how a firm should balance its UGC and PGC audiences to optimize acquisition, retention, and revenue generation. 2 - Effect of Advertising Capital and R&D Capital on Sales Growth, Profit Growth and Market Value Growth Gautham Vadakkepatt, Assistant Professor, University of Central Florida, Department of Marketing, College of Business Administration, Orlando, FL, 32816, United States of America, [email protected], Venkatesh Shankar, Rajan Varadarajan 3 - Competitiveness of Customer Relationship Management: Does Profitability Really Matter? Tae Ho Song, Visiting Researcher, UCLA Anderson School of Management, 110 Westwood Plaza, Los Angeles, 90095, United States of America, [email protected], Sang Yong Kim There is mounting pressure on firms to exhibit organic (due to internal efforts) growth in sales, profit, and market value. Unfortunately, there is a limited understanding of the drivers of organic growth. In this research, we examine the effects of advertising capital and R&D capital on sales growth, profit growth, and market value growth. We also examine how interactions between these two strategic variables and their interactions with environmental contingency factors of dynamism, munificence, and complexity impact these performance measures. Using dynamic panel data analysis of 185 firms over an eight-year period (2000-2007), we uncover a nuanced understanding of how advertising capital and R&D capital affect firm growth. Our results show that both R&D capital and advertising capital directly affect sales growth, but neither has a direct impact on profit growth. Furthermore, R&D capital has a direct impact on market value growth. We also find that while the interaction of advertising capital and R&D capital does not directly affect sales growth or market value growth, it has a positive direct impact on profit growth. Finally, we find that environmental contingencies matter. For instance, environmental dynamism negatively (positively) moderates the relationship between R&D (advertising) capital and sales growth. Although the vast majority of researchers and practitioners in marketing supports that long-term performance orientation such as customer relationship management has been regarded as the better strategy than short-term performance orientation, some have raised the doubts about optimality of customer relationship management in the competitive environment (e.g. , Boulding et al 2005; Shugan 2005; Villanueva et al 2007; Musalem and Joshi 2009). In this research, we investigated the source of competitiveness of customer relationship management strategy. We set up the competitive market structure between customer equity maximization strategy (for customer relationship management) and short-term profit maximization (for noncustomer relationship management), and did the game theoretical analysis using simulation methods. The results show the long-term profit when both of two competing firms choose the strategy of short-term profit maximization can be greater than when both choose the customer equity maximization strategy in the equilibrium status. Interestingly, it is inconsistent with the conventional wisdom. With the different standpoint of view, our results also mean that the customer equity maximization is better risk-aversive strategy than short-term profit maximization since our results imply that the short-term profit maximization is the payoffdominant equilibrium and the customer equity maximization is the risk-dominant equilibrium. Thus, we propose the long-term performance orientation such as the strategies of customer relationship management or customer equity maximization is the strategy for avoiding the risk and uncertainty caused by market competition rather than the strategy for maximizing the long-term performance itself. 3 - Influence of Market Orientation on Corporate Brand Performance: Evidences from Indian B2B Firms Soumya Sarkar, Indian Institute of Management Calcutta, D H Road, Joka, Kolkata, 700104, India, [email protected], Prashant Mishra Organizational buyers perceive corporate brands, radiating from the firms’ culture, value propositions, and stakeholder relations, as specific cues for information, promises, and quality. Effectively managing the brands acquires a strategic dimension since it leads to a sustainable competitive advantage - brand equity. It is an indicator of the performance of the brand, and in turn, of the firm performance. Studies have revealed positive correlation of market orientation with both financial and marketbased performance parameters, but have not contemplated corporate brand performance as an outcome of market orientation. The absence of corporate brand performance is a fundamental lacuna in market orientation - performance relationship literature. Also not much research has happened on emerging economy firms, their adoption of market orientation, and its impact on firm performance. This study, still in its working stages, is situated in those above-mentioned hiatuses to study the impact market orientation creates on corporate brand performance of business-to-business firms. This is its prime contribution to literature. The other noteworthy contribution it makes emanates from the methodology. Our distinct research design stands out from the common practice of acquiring judgmental performance data from the respondents to market orientation instruments. Instead we cover both ends of individual B2B marketer-customer dyads. The personnel in strategic marketing positions of selling firms provide the market orientation data. At the other end, purchasers/users from the buyer organisations indicate the perceived strength of the marketer brands, assessing the firm performance. 4 - Differential Influences of Market Structures on Cognition and Affect Anita Basalingappa, Associate Professor, Mudra Institute of Communications (MICA), Shela, Ahmedabad, 380058, India, [email protected], M. S. Subhas This paper explores the influences of market structure on Cognition and Affect. This has been done keeping in mind the generalizations that can be applied while developing customer acquisition and customer retention efforts for any firm in a particular market structure. The questions addressed in this paper are: Can we generalize certain kind of consumer attitude across firms in a particular market structure? Does market structure have an influence over consumer attitude? The paper limits its scope to cognition and affect stages of the Tri-component model. Three product markets in the Transportation sector in India were chosen to make possible the comparison of the three market structures from a common base – which is the transportation sector. (Imperfect monopoly – Railways; Oligopolistic market – Airways; and Monopolistic market – Road transport). 78 MARKETING SCIENCE CONFERENCE – 2011 Saturday, 10:30am - 12:00pm SB02 ■ SB02 Legends Ballroom II Social Networks ■ SB01 Cluster: Internet and Interactive Marketing Invited Session Legends Ballroom I Chair: Christian Barrot, Kühne Logistics University, Hamburg, Germany, [email protected] 1 - Stimulus and Mutual Interaction Stochastic Bass Model Tolga Akcura, Ozyegin University, Uskudar, Altunizade, Istanbul, Turkey, [email protected], Kemal Altinkemer Advertising: Strategy Contributed Session Chair: Gangshu Cai, Kansas State University, Calvin 101B, Manhattan, KS, 66502, United States of America, [email protected] 1 - Competing In Hollywood Claudio Panico, Assistant Professor, Bocconi University, Via Roentgen, 1, Milan, 20136, Italy, [email protected], Sebastiano Delre There are an increasing number of people who write blogs and participate in social media. According to the latest statistics, 20 to 25% of Internet users already participate in microblogging and use services from twitter.com. Already, many firms start to capitalize on discussion themes by selling advertisements. Consequently, the use of keywords emerge as an important consideration for the managers. In this research we propose a model to capture the microblogging behavior of individuals. The people who tweet can be represented as members of a network who interact with each other and be influenced by the themes in the society. Accordingly, as the members of the network are stimulated, they adopt a discussion following a diffusion model while beeing influenced by the mutual interactions in the network. The model we propose acknowledges both societal (exogeneous) forces and network related mutual interactions within a diffusion mechanism. The diffusion in our model is based on an individual level stochastic Bass framework. We test our model using data collected from blogscope.com and tweetmeme.com. The estimation results show that our model suits well to model the microblogging behavior observed on the Internet. The big studios spend hefty sums on producing and advertising their movies. How do the studios make their strategic budget choices? To which extent do they differentiate their movies? How do their choices affect viewership over the movie life-cycle? To address these research questions, we build a location model where two studios compete to attract a population of moviegoers with heterogenous preferences. The studios locate simultaneously their movie on the Hotelling segment. They involve in predatory marketing to attract the consumers when the movies are launched. In the post launch the viewership depends instead on the word of mouth. We compare the two cases when the moviegoers exhibit no satiation and variety seeking. Our model captures some of the salient features of the motion picture industry and yields interesting predictions for the competitive positioning and the strategic budget choices of the studios. We contribute to the rich literature on competitive positioning (for example, Tabuchi and Thisse, 1994, and Tyagi, 2000) studying how the studios’ budget decisions relate to their location choices and endogenizing the level of differentiation of the movies. Our research is also linked to the more recent literature that analyzes competition when consumers have a variety seeking behavior (Kim and Serfes, 2006; Seetharaman and Che 2009; Sajeesh and Raju, 2010). Our analysis could also be generalized to other entertainment industries such as video games, concerts, sport events, music, books, etc., where the firms offer experience goods, strongly advertise before the launch and then experience word of mouth effects in the post-launch. 2 - Assessing Value in Product Networks Barak Libai, Professor, Interdiciplinary Center (IDC), POB 167, Herzlyia, 46150, Israel, [email protected], Eyal Carmi, Ohad Yassin, Gal Oestreicher Traditionally, the value of a product has been assessed according to the direct revenues the product creates. However, products do not exist in isolation but rather influence one another’s sales. This issue is especially evident in eCommerce environments, where products are often presented as a collection of webpages linked by recommendation hyperlinks, creating a large-scale product network. Here we present the first attempt to use a systematic approach to estimate products’ true value to the seller in such a product network. Our approach, which is in the spirit of the PageRank algorithm, uses easily available data from large-scale electronic commerce sites and separates a product’s value into its own intrinsic value, the value it receives from the network, and the value it generates to the network. We use this approach on data collected from Amazon.com and from BarnesAndNoble.com. Focusing on this domain of interest, we find that if products are evaluated according to their direct revenue alone, without taking their network value into account, the true value of the “long tail” of electronic commerce may be underestimated, whereas that of bestsellers might be overestimated. 2 - Advertising and Pricing Strategies for Luxury Brands with Social Influence and Brand Maintainance Jin-Hui Zheng, Business Division, Institute of Textiles and Clothing, The Hong Kong Polytechnic University, 912B, New East Ocean Center, East Tsim Sha Tsiu, Kowloon, N.A., Hong Kong - PRC, [email protected], Chun-Hung Chiu, Tsan-Ming Choi Social needs play an important role in the consumer purchase of conspicuous products such as high-end fashion labels. In addition, in order to maintain the brand strength, sufficient advertising is needed for communicating with customers. To capture this feature, this paper considers that there are penalties for insufficient advertising of the brand. This paper analytically studies the optimal advertising and pricing decisions for fashion brands in a market consisting of two segments with opposite social influences on each other. We develop an optimization model for this problem, explore the optimal solution scheme and conduct sensitivity analysis. Our analysis reveals that the optimal strategies follow different scenarios and sometimes counter-intuitively. For instance, it is optimal for a brand of conspicuous product to advertise to both market segments while sell to one market segment only. Managerial insights are reported. 3 - Consumers as Active Participants in Viral Marketing Campaigns Analyzing Forwarding Behavior Christian Pescher, Ludwig-Maximilians-University Munich, Geschwister-Scholl-Platz 1, Munich, 80539, Germany, [email protected], Martin Spann, Philipp Reichhart Tie strength is one of the most important factors when it comes to explain the influence of word-of-mouth related communication. Strong tie referrals have a greater influence on the recipient than referrals obtained from weak ties, because they are perceived to be more trustworthy and credible. In mobile viral marketing campaigns companies send messages to a number of consumers, the so-called seeding points, and hope that they forward the message to their friends as unsolicited referrals. So far, literature has failed to analyze the effects of tie strength on the seeding point’s decision making process in mobile marketing campaigns. In this paper the authors conduct a mobile viral marketing campaign and survey study participants. They build a three stage model of the seeding point’s decision making process from awareness to the number of referrals. They find that the likelihood to make an unsolicited referral increases with decreasing tie strength and that low tie strengths lead to a higher number of unsolicited referrals. Further, they find that psychographic indicators are significant in the early stages of the seeding point’s decision making process, whereas sociometric indicators are significant in its later stages. 3 - Persuading Consumers with Social Attitudes Daniel Halbheer, University of Zurich, Department of Business Administration, Plattenstrasse 14, Zurich, 8032, Switzerland, [email protected], Stefan Buehler This paper studies pricing and advertising in a Hotelling model with consumption externalities. The key feature of this model is that a consumer’s valuation of a product depends both on persuasive advertising and an unobservable individual (dis)utility from other consumers buying the same product. Assuming that consumer valuations follow a bivariate uniform distribution, we derive product demands and characterize equilibrium prices and advertising levels. We find that both quality leaders and cost leaders invest more in advertising and sell at higher prices. Also, we show that firm profits are increasing in the average degree of exclusivity in the population. 4 - Efficacy of Advertising Structures and Cost Sharing Formats in a Competing Channel Gangshu Cai, Kansas State University, Calvin 101B, Manhattan, KS, 66502, United States of America, [email protected], Bin Liu, Zhijian (Zj) Pei This paper evaluates the efficacy of advertising structures in a variety of scenarios through a model with two asymmetric competing channels. We identify the equilibrium areas of manufacturer advertising, retailer advertising, and hybrid structures, and further demonstrate that channel members prefer retailer advertising if channel substitutability is low; otherwise, manufacturer advertising has the edge. We also explore the impact of advertising cost sharing and find that explicit cost sharing does not always help the advertising providers, unless intensive competition can be well contained. 79 SB03 MARKETING SCIENCE CONFERENCE – 2011 3 - Modeling Unobserved Drop-out Rate to Optimize e-Panelist Lifetime Value Arnaud De Bruyn, ESSEC Business School, Cergy, 95000, France, [email protected] 4 - An Empirical Comparison of Seeding Strategies for Viral Marketing Christian Barrot, Kühne Logistics University, Hamburg, Germany, [email protected] Seeding strategies have a major influence on the success of viral marketing, but previous studies using computer simulations and analytical models produced conflicting recommendations about optimal seeding strategies. We therefore compare four different seeding strategies in two complementary small-scale field experiments and one real-life application during a viral marketing campaign that involved more than 200,000 customers of a telecommunications provider. Our results empirically show that seeding strategies that target either well-connected individuals (“high degreeness”) or individuals who connect different parts of the network (“high betweenness”) lead to a success rate twice as high as that of a random seeding strategy. In addition, our study reveals that well-connected individuals are attractive seeding points for viral marketing campaigns because they are more likely to participate. However, they do not exploit their higher reach potential and do not have more influence on their peers than less-connected individuals. These empirical results can help to better calibrate simulation studies and analytical models to produce more realistic results with respect to diffusion processes. Online access panels are becoming of paramount importance in marketing research, and constitutes a great asset for market research firms. In this paper, we show that traditional models fail to quantify the true “cost” of an electronic solicitation, and we demonstrate that each additional solicitation not only decreases the likelihood of future participation, but might even increase the drop-out rate (a mostly unobserved phenomenon that has a dramatic impact on the lifetime value of an e-panelist). Our model estimates the likelihood that an e-panelist will respond positively to a new solicitation, as a function of his past behavior and how many times he has been solicited so far, and integrates a latent “wear out” effect. We fit the model on a sample of more than 700,000 e-mail solicitations sent over a period of 3 years, and demonstrate that each additional solicitation contributes to a long-lasting wear-out effect; the unobserved drop-out rate can reach up to 10% at each additional solicitation. 4 - Why People Use Social Media: How Motivations Influence Goal Pursuit Donna L. Hoffman, University of California-Riverside, Anderson Graduate School, 900 University Avenue, Riverside, CA, United States of America, [email protected] ■ SB03 Legends Ballroom III As social media applications proliferate and the dynamics of online social interaction continue to evolve, researchers are seeking deeper understanding of how and why people use social media so that theoretically consistent models linking user motivations, social media goals, and consumer behavior outcomes can be constructed. In this research we use a multi-level modeling framework to test a conceptual model of the relationship between social media goal pursuit and subjective well-being that is grounded in motivational theory. In the theoretical framework, the broad range of goals people have for using social media is uniquely determined by two broad dimensions that specify the primary focus of the 1) content interaction and the 2) person interaction. The social media goals corresponding to these dimensions are hypothesized to be pursued according to the basic needs that social media satisfy for its users and the motivational orientations supported by those needs. The model states goals, in turn, lead to different subjective outcomes with the relationship between social media goal pursuit and well-being moderated by perceptions of overall well-being in specific life domains, along with constructs related to social identity. Two large-sample studies support the model hypotheses. Researchers may use these results to build upon a common set of constructs for understanding why people use social media and marketing managers may use the results to help focus social media strategic efforts. The results may also assist marketers seeking to incorporate into their content applications those social components that best satisfy consumers’ basic needs and lead to the most positive outcomes. Online Consumer Behavior Cluster: Internet and Interactive Marketing Invited Session Chair: Donna L. Hoffman, University of California-Riverside, Anderson Graduate School, 900 University Avenue, Riverside, CA, United States of America, [email protected] 1 - Post-consumption Satisfaction with Movies: A Multivariate Poisson Analysis of Online Ratings Ruijiao Guo, PhD Student, University of Wisconsin -Milwaukee, 3528 N Oakland Ave, Apt 4, Milwaukee, WI, 53211, United States of America, [email protected], Purushottam Papatla Post-consumption satisfaction with movies, as conveyed through ratings at sites like IMDB.com, rottentomatoes.com, boxofficemojo.com, Netflix.com or blockbuster.com, has an effect on off-theater revenue streams such as DVD sales, movie rentals and, potentially, on the performance of sequels as well. It is therefore important for movie producers to understand which factors affect consumer satisfaction with movies and the relative importance of each factor. Such insights would be useful in producing movies that are likely to lead to high customer satisfaction and, hence, better economic performance. This is the issue that we investigate in this research. Specifically, we examine the relative roles of human capital, hedonic product attributes, utilitarian product attributes and market coverage on post-consumption satisfaction with movies. We operationalize post-consumption satisfaction with a movie as the vector of counts across rating levels 1-10, for that movie, at IMDB.com. Since the counts across rating levels are likely to be correlated both due to observed as well as omitted factors, we model the counts jointly using a Multivariate Poisson Log Normal specification and calibrate the model using MCMC methods. Findings, managerial implications, limitations, and future directions are discussed. ■ SB04 Legends Ballroom V Product Management: General Contributed Session Chair: Yeong Seon Kang, Doctoral Student, University of California, Irvine, CA, The Paul Merage School of Business, Irvine, CA, 92697-3125, United States of America, [email protected] 1 - Voice Banking: An Exploratory Study of the Access to Banking Services Using Natural Speech Mauro Arancibia, CEO, Merlìn Telecommunications, Matilde Salamanca 736. Ofic 301, Providencia, Santiago, 7500657, Chile, [email protected], Claudio Villar, Jorge Marshall, Natalia Arancibia, Sergio Meza 2 - The Role of Trust in the Firm-hosted Virtual Community in Purchase Intentions Formation Illaria Dalla Pozza, Italy, [email protected] The rapid increase in number of virtual communities on the Internet raises important questions on the influence that these communities can have on consumer purchasing intentions and behaviour. Information in virtual communities comes from anonymous and faceless individuals who can increase the perception of risk and uncertainty. In this particular situation, the exigency of trust arises, as the virtual community is an environment abounding in uncertainties and ambiguities (Ridings et al 2002). This study suggests that trust in the virtual community is a key determinant in fostering the influence virtual community can have on consumer purchasing intentions and behaviour. Differently from previous research, that has predominantly focused on trust in dyadic (one-to-one) relationships, we here focus on trust between an individual user and the entire community of participants in a firm-hosted online community (one-to-many relationship). There is a limited body of related prior research analyzing the role of trust in online consumer to consumer marketplaces and in one-to-many relationships (Pavlou and Gefen 2004, Ridings et al 2002 and Wu and Tsang 2008). In analyzing firm-hosted virtual communities, this paper adds and extends the existing literature by answering the following four key questions: what are the drivers of trust in a firm-hosted virtual community? How does the trust in the virtual community influence consumers’ intentions to act on the community advice and intentions to buy? How is trust in the virtual community interrelated with the loyalty for the firm hosting the community? How do product characteristics affect the relationships? One of the factors associated with a country’s economic development is the penetration level of the banking system in the population (Chaia A. et al 2009). In Latin America the average of people without access to banking services is above 65% (McKinsey, 2008). Although in Chile this value is considerably smaller, there is a sizable population ready to take advantage of a variety of traditional and new banking services. Penetration of cell phones in the region (e.g Chile, near 100%) can exhibit levels that are comparable to those of developed countries. This fact has opened the possibility of using wireless devices as primary channels of access to banking services. Some pioneer institutions are already offering basic applications. Since the penetration of smart phones is still low in the focal population, banking applications based on natural speech become interesting. Some key questions are raised: who would be willing to adopt the use of banking services through their wireless devices? What are the main challenges of the adoption? In this study we explore these questions in a field experiment conducted in Chile with 50 potential customers in a pilot product using their own cell phones. The main results of the study are: 92% of the individuals were able to complete the transaction on first trial. 64% of the individuals believed they would adopt the product when available. Perception of security, complexity and transaction length emerged as the three main factors that challenge the adoption of the service. No gender or age effect to explain differences was found, across the pilot sample. The paper concludes with new insights, further elements to research and implications for CRM. 80 MARKETING SCIENCE CONFERENCE – 2011 SB05 2 - R&D Spillover and Product Differentiation in Fully Covered Market Xin Wang, University of Cincinnati, 402 Carl H. Lindner Hall, 2925 Campus Green Drive, Cincinnati, OH, 45221, United States of America, [email protected], Yuying Xie 2 - Bounded Rationality in Dynamic Games: Insights into Strategy Optimization Amid Player Uncertainty Jennifer Cutler, Duke University, Durham, NC, United States of America, [email protected] The literature is limited on the effect of R&D spillover on the strategic choice of product quality. A recent study finds that the externality of R&D spillover unambiguously enhances the strategic development of quality for both the high-end producer and the low-end producer, which is contrary to the traditional viewpoint that spillovers decrease the spillover supplier’s incentive to improve. This paper explores the spillover effects on quality when markets are fully covered. Three scenarios are analyzed by using a popular setting of duopolistic competition in a quality-price two-stage game. Scenario 1 is a simultaneous game without production cost. We find that the firm producing at the low-end has a corner solution, and technology spillovers have no effect on either firm due to the corner solution. Scenario 2 considers asymmetric production cost by assuming a unit cost to the highend producer. We find that the spillovers improve quality of the low-end firm but discourage the high-end firm to innovate. Scenario 3 considers the combination of market leadership and two-way spillovers. Specifically, we assume the low-end firm to be the leader in stage 2. We find that both qualities are improved due to technology spillovers. Overall, these results show that R&D spillovers have diverse effects on product qualities in various circumstances. The question of how technology spillovers influence foreign investments in product quality requires careful examinations of consumer preferences, the nature of products, and the competitiveness of firms. In boundedly rational paradigms, players of lengthy dynamic games are limited in how many levels ahead they can think. When selecting a strategy, players must make several decisions: how far ahead they will choose to think (subject to their cognitive constraints), how far ahead they believe other players will think, and how to define higher order beliefs (e.g. player 1’s beliefs about player 2’s beliefs about player 1). We use analytical and numerical analyses to explore the main effects and interactions of these decisions on payoff in generalizable dynamic game contexts that are independent of payoff structure. First, we demonstrate asymmetry in opponent estimation errors that are too high versus too low, and provide contextual parameters that moderate and reverse the differences. Second, we explore the effects of increasing levels of thinking, and show how they interact with opponent estimation errors. Third, we introduce costly effort, and reevaluate the outcomes for net payoff. The theoretical results are then demonstrated in several well-known dynamic marketing games. The results provide new strategic insights for optimizing payoff amid uncertainty in other players’ abilities. First, contexts in which it is advantageous to underestimate opponents’ abilities are delineated from contexts in which the reverse is true. Second, situations in which it is advantageous to restrict one’s own level of thinking are defined. 3 - Firm Strategies in the “Mid Tail” of Platform-based Retailing Baojun Jiang, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected], Kinshuk Jerath, Kannan Srinivasan 3 - Downsizing or Price Competition Responding to Increasing Input Cost Yeong Seon Kang, Doctoral Student, University of California, Irvine, The Paul Merage School of Business, Irvine, CA, 92697-3125, United States of America, [email protected] While millions of products are sold on its retail platform, Amazon.com itself stocks and sells only a small fraction of them. Most of these products are sold by third-party sellers, who pay Amazon a fee for each unit sold. Empirical evidence clearly suggests that Amazon tends to sell high-demand products and leave long-tail products for independent sellers to offer. We investigate how a platform owner such as Amazon, facing ex ante demand uncertainty, may strategically learn from these sellers’ early sales which of the “mid-tail” products are worthwhile for its direct selling and which are best left for others to sell. The platform owner’s “cherry-picking” of the successful products, however, gives an independent seller the incentive to mask any high demand by lowering his sales with a reduced service level (unobserved by the platform owner). We analyze this strategic interaction between the platform owner and the independent seller using a game-theoretic model with two types of sellers – one with high demand and one with low demand. We show that it may not always be optimal for the platform owner to identify the seller’s demand. Interestingly, the platform owner may be worse off by retaining its option to sell the independent seller’s product whereas both types of sellers may benefit from the platform owner’s threat of entry. The platform owner’s entry option may reduce the consumer surplus in the early period though it increases the consumer surplus in the later period. We also investigate how consumer reviews influence the market outcome. This study is to determine the firm’s optimal strategy in choosing between a price increase or downsizing the package size under conditions of a spatially differentiated market and each consumer’s diminishing marginal utility as product size increases. When a manufacturing firm faces a steep increase in its variable input cost, the firm usually increases the price of its product to share the loss in its profits with consumers. Downsizing can be an alternative strategy for the firm instead of a price increase. We assume that a firm downsizes the product package while maintaining its price. We find that the degree of competition and the amount of input cost increase affect a firm’s optimal strategy. If the rival firm chooses to increase the price when responding to an input cost increase, then the firm’s increasing price is always a dominant strategy. When the degree of competition is sufficiently large and the amount of input cost increase is sufficiently small, both firms choose a price increase in a unique Nash equilibrium. When the degree of competition is sufficiently small or when the amount of the input cost increase is sufficiently large, then both firms choose a price increase or both firms choose downsizing in equilibrium. Both cases are Nash equilibria and the former has higher profits than the latter because both firms pass through the entire amount of the input cost increase to the end consumers by the price increase. Therefore, both firms’ profits can maintain the same value as the benchmark case before the input cost increases. We also show that the asymmetric pairs of a price increase by one firm and downsizing by the other are never Nash equilibra in our framework, and both firms symmetrically choose the same strategy in equilibrium. 4 - Repeated Consumption Pattern under Different Pricing Schemes Ruhai Wu, Assistant Professor, McMaster University, 1280 Main Street West, Hamilton, ON, L9G5C5, Canada, [email protected], Suman Basuroy Consumers with repeated consumptions often sign in long-term pricing contracts, such as monthly membership, prepaid plan, etc. They estimate their future consumptions when choosing among different pricing schemes. Their contractual choices subsequently influence their realized consumptions. In this paper, we develop a repeated game model studying the consumers’ decisions on pricing schemes and on repeated consumption. We study the three pricing schemes, namely, pay per visit, Tperiod (monthly) membership and prepaid K-visit (10-visit) pass. The model shows that the consumers realize different consumption patterns under different pricing schemes, and illustrate the consumers’ optimal choices among the pricing schemes. Interestingly, under certain condition, consumers choose T-period membership even though it is expected to result a “per-usage” cost higher than that under other pricing schemes. This outcome is consistent with the finding in Della Vigna and Malmendier (2006), where the authors explained it as consumers’ overestimation of future consumption. We also show the firm’s price discrimination solution in context of design of pricing schemes, given consumers’ contractual and consumption decisions. ■ SB05 Legends Ballroom VI Game Theory V: General Contributed Session Chair: Ruhai Wu, Assistant Professor, McMaster University, 1280 Main Street West, Hamilton, ON, L9G5C5, Canada, [email protected] 1 - Within-firm and Across-firm Search: The Impact on Firms’ Product Lines and Prices Anthony Dukes, University of Southern California, Marshall School of Business, Los Angeles, CA, 90089, United States of America, [email protected], Lin Liu Many firms offer a related set of products, all of which are intended to satisfy a specific consumer need, but each differing on feature, color, or styling. Furthermore, comparing alternatives within a firm’s product line may be non-trivial for consumers. But most models of consumer search assume that consumers incur search costs only when comparing products of different firms. In this research, we examine the implications when consumers incur costs searching within a firm. Specifically, we decompose consumer search into within-firm and across-firm components and examine a model in which firms compete on prices and product line length. We show that the consideration of within-firm search costs leads to insights not seen in the prior literature on consumer search. Under certain conditions, as firms’ products become closer substitutes, consumers search less across firms but more within each firm, causing firms to expand their product lines. And, unlike across-firm search costs, an increase in within-firm search cost does not necessarily lead to higher prices. 81 SB06 MARKETING SCIENCE CONFERENCE – 2011 ■ SB06 4 - The Role of Team Composition in Cross-cultural Business Negotiations Robert Wilken, Professor, ESCP Europe Business School, Berlin, Germany, [email protected], Frank Jacob, Nathalie Prime Legends Ballroom VII Improving Efficiency in Marketing Negotiations The management of cultural issues in negotiations has significantly gained in importance. On an inter-organizational level, companies negotiate with international partners. From an intra-organizational perspective, a company may ask itself how to compose its negotiation teams in terms of cultural mixture. Particularly, is it beneficial to add a team member who has the same cultural background as the business partner? Such a person could act as a “linking pin”, somebody familiar with the cultural background of the negotiation partner and able to better understand the partner’s communication styles. We investigate cross-cultural business negotiations between German and French students. We observe that in monocultural teams, communication styles are consistent with our expectation that seller teams with high collectivism scores (French) adopt a more integrating and a less attacking strategy than do seller teams with a lower collectivism scores (German). Multicultural teams, however, are not more integrative per se: Adding a French (German) “linking pin” to the German (French) team leads to an increase (decrease) in the use of integrating behavior. Both the behavior of the seller team and its individual profit is substantially driven by the buyer’s behavior: A German buyer who faces a French team that includes a German team member increases (decreases) his or her attacking (integrating) behavior, whereas a French buyer does the opposite (although not to the same extent). Thus, team composition in terms of cultural mixture strongly interacts with the nationalities involved in a negotiation setting, providing important implications for negotiation practice in cross-cultural environments. Cluster: Special Sessions Invited Session Chair: Ralf Wagner, University of Kassel, DMCC – Dialog Marketing Competence Cente, Mönchebergstr. 1, Kassel, 34125, Germany, [email protected] Co-Chair: Katrin Bloch, University of Kassel, DMCC – Dialog Marketing Competence Cente, Mönchebergstr. 1, Kassel, 34125, Germany, [email protected] 1 - Measuring Efficiency of Negotiated Exhanges: An Evaluation, Refinement and Extension P.V. (Sundar) Balakrishnan, University of Washington, WA, United States of America, [email protected], Charles Patton, Robert Wilken Negotiated exchanges are of great importance in consumer markets and even more so in the huge volume of business market exchanges. Despite all the research and training on bargaining, it is a well known problem that many negotiations end in impasse and many more negotiated exchanges fail to reach Pareto efficiency. As a means to understanding the economic loss, we critically examine some of the most relevant bargaining efficiency measures. The most important conceptual shortcoming of extant measures relates to their focus on the economic outcome aspect of negotiation efficiency, implying that the resource-expending dimension and sociopsychological outcomes are neglected. We also examine some computational shortcomings of extant measures of negotiation efficiency. Based on these limitations and the observations concerning extant measures, we derive a set of desired properties that an appropriate measure of bargaining efficiency should have. This list includes the suggestion to go beyond the standard economic viewpoint and extend the discussion to the twin concepts of outcome efficiency (i.e., the proximity of outcomes to the Pareto frontier) and resource efficiency (outcomes obtained relative to the inputs expended). Upon this foundation we briefly introduce Data Envelopment Analysis (DEA) as a technique for measuring efficiency that overcomes a number of the conceptual and computational limitations identified in prior measures. 5 - Facing Bargaining Power Ralf Wagner, University of Kassel, DMCC – Dialog Marketing Competence Cente, Mönchebergstr. 1, Kassel, 34125, Germany, [email protected], Katrin Bloch One of the recurring topics in retailing is the power shift in the supply channel toward the retailers. Several studies focus on the reasons for the power shift rather than highlighting the consequences for the retailer-manufacturer relationship. Despite a considerable amount of literature about gender, the intersection of power and gender is seldom challenged. In this paper, we combine both research fields and analyze the force of gender differences and bargaining power on the success in the domain of retailer-manufacturer negotiations. In our negotiation experiment with 92 negotiators, we consider the issue authority as a proxy of bargaining power. Differences in issue authority allocation turn out to have a significant impact on negotiation success. In scenarios with substantial differences in bargaining power, particularly female and mixed dyads failed to achieve a mutually satisfactory result. We learn that an increase in issue authority for one of the two parties does not necessarily lead to an increase in negotiation efficiency. Female negotiators rely on their bargaining power, rather than systematically improving mutual utilities whereas masculine dyads negotiate on a consistently higher, but still inefficient level. In the light of these differences, we argue that bargaining power in commercial negotiations does not compensate for insufficient negotiation skills or efforts. On the contrary, unbalanced bargaining power decreases the likelihood of success. 2 - Benefits of Mediating Lawyers in Negotiations Olivier Mesly, Professor, UQAT, Canada, Montreal, QC, Canada, [email protected] Negotiation outcome depends not only on strategies but also on the way parties perceive each other and on context; the presence of a lawyer is thought to affect the outcome from the onset due to the negative perception parties have of him. Poitras, Stimec, & Roberge (2010) outline such negative effects as follows: (1) negotiating parties may display less motivation towards conflict resolution; (2) the lawyer is perceived as someone promoting the billing of hours rather than as a person genuinely interested in a prompt resolution to the current conflict; (3) the lawyer is perceived as fighting over issues that have little real relevance, thus generating unnecessary costs; (4) the lawyer is seen as being trained towards confrontation, not mediation; (5) the lawyer is believed to have a limited understanding of the problems at hand, thus reducing his ability to solve them creatively; (6) the lawyer is thought to encourage his client to show little emotion (such as guilt, remorse, etc.), thus limiting the expression of true sentiments, which is essential towards the building of mutual trust; finally (7) the lawyer is seen as filtering the information to his advantage, regardless of the well-being of other parties. The current paper examines the dynamic resulting from the presence of a lawyer in a negotiating forum using the OPERA model (Mesly, 2010), a model that emphasizes the role of perceived predation (perceiving the other as a potential predator) and its impact on trust and cooperation, two elements critical to negotiation outcome. ■ SB07 Founders I Entertainment Marketing II Contributed Session Chair: Erik Bushey, Doctoral Student in Marketing, University of Illinois, 350 Wohlers Hall, 1206 South Sixth Street, Champaign, IL, 61820, United States of America, [email protected] 1 - The Impact of Product Placement on Ad Avoidance Natasha Foutz, Assistant Professor of Commerce, McIntire School of Commerce, University of Virginia, 340 Rouss & Robertson Hall, Charlottesville, VA, 22904, United States of America, [email protected], David Schweidel, Robin Tanner 3 - The Role of Intuition and Deliveration in Negotiations Katrin Bloch, University of Kassel, DMCC - Dialog Marketing Competence Cente, Mönchebergstr. 1, Kassel, 34125, Germany, [email protected], Ralf Wagner In recent years, there has been a considerable rise in the amount of product placement on television. Though research has separately examined the effectiveness of product placement and the extent to which audiences avoid traditional television ads, research has not considered how these techniques may be related. In this research, we explore the impact of product placements in prime time television programs on the audience levels in subsequent commercial breaks. Using second-bysecond data on advertising, product placement, and set top box tuning, we find that product placements can mitigate or exacerbate ad avoidance. Central to the impact of product placements on the extent of ad avoidance is the “match” between the brands and product categories featured in both the product placements and the ads. Focusing on the first advertisement in a commercial break, we find that product placements that do not match the product category or brand of the ad contribute to a slight rise in ad avoidance. When an ad follows product placements from a competing brand in the same product category, we observe larger declines in the audience levels during the ad. In contrast, ads that follow product placements from the same brand exhibit smaller audience declines during the ads. We discuss the resulting implications for both advertisers and television networks and provide a roadmap for future research in the changing television landscape. The role of the intuition within negotiations is a controversial topic. Dijksterhuis and Nordgren (2006) highlighted that intuitive acting persons get better results in complex negotiation tasks. Contrastingly, Bazerman and Malhotra (2006) take the disadvantages of intuitive negotiation behavior for granted and reduce their perspective to opportunities to overcome intuition. Using the Preference for Intuition and Deliberation Scales from Betsch (2004) we asses within a neutral and a highly emotional negotiation task the different preferences for intuition and deliberation in decision making. We learn that deliberative negotiators perform on a constantly high level in both scenarios. Intuitive and situational-depending negotiators have a tendency to be less successful in emotional scenarios, albeit they perform on the same level than deliberative negotiators in the neutral tasks. The results are relevant for both researcher and practitioners. For the former are provided with persuasive arguments for reconsidering the value of deliberation in negotiation research critically. The latter get evidence about circumstances in which intuition supports and in which intuition sabotages the negotiations. 82 MARKETING SCIENCE CONFERENCE – 2011 2 - The Advertising Role of Professional Critics in the Book Industry Michel Clement, University of Hamburg, Welckerstr. 8, Hamburg, 20354, Germany, [email protected], Marco Caliendo SB08 and rule out alternative explanations by showing that information on others’ propensity to disclose affects respondents’ discomfort associated with divulgence (Study 1B), and not their interpretation of the questions (Study 1C). The second set of studies suggests that divulgence is anchored by the initial questions in a survey; people are particularly likely to divulge when questions are presented in decreasing order of intrusiveness. Study 2B suggests that the effect arises by affecting people’s judgments of the intrusiveness of the inquiries; Study 2C goes further by showing that the effect is altered when privacy concerns are primed at the outset of the study. This research helps understand how consumers’ propensity to disclose is affected by continual streams of requests for personal information, and by the equally unavoidable barrage of personal information about others. We focus on the advertiser’s role of a book critic (e.g., Oprah Winfrey). Based on a large data set from the German book market, we are able to estimate the effect of a well-known TV reviewer (Elke Heidenreich) on book sales in the short, mid and long term. Thus, we contribute to the existing literature about critics in measuring the advertising effect of critics in the entertainment industry. Furthermore, we control for the selection bias of critics who might be biased towards potential best-sellers or star authors and prefer to review these books. Consequently, the probability to review a book by the reviewer will strongly correlate with book sales raising the question of causality. Therefore, we reduce this selection bias by using a combination of propensity score-matching and difference-in-differences methods. Our study of the advertiser’s role of an individual critic produces two key findings. First, our analysis shows that mass media critics are influencers – even when they are just predicting well. Our empirical study supports our theoretical reasoning, because we find strong dynamic effects on sales by the reviews of Elke Heidenreich. Second, our probit analysis reveals substantial indications about selection effects of individual critics which allow publishers to better address critics in order to “motivate” them to review a new book. 2 - Privacy as Resistance to Segmentation Luc Wathieu, Georgetown University, McDonough School of Business, 37th and O Streets, Washington, DC, 20057, United States of America, [email protected] When more information about consumers becomes available, competing firms target finer market segments without regard for the overall efficiency of the marketing system. While this generates better value propositions for some newly isolated consumer types, other consumers might in the process experience market exclusion, suffer a price increase, or switch to a new product with less consumer surplus attached. This paper proposes that the fear of such externalities arising from finergrained segmentation is a legitimate (and heretofore overlooked) source of privacy concerns. In the context of a simple model, conditions surrounding the occurrence of privacy as resistance to segmentation are identified, and remedies are discussed. The results explain the “privacy paradox”: consumers routinely claim that privacy is very important to them, but they don’t do much to protect it. It turns out that privacy concerns reveal a business opportunity for marketing intermediaries (e.g., access suppliers, retailers, clubs, consumer communities) capable of organizing cohesion among diverse consumer types to fend off micro-segmentation. 3 - US Holidays in Non-US Markets: Moderating Role of Movie Nationality in Demand Fluctuation Joonhyuk Yang, PhD Student, KAIST, Guseong-dong, Yuseong-gu, Daejeon, 305701, Korea, Republic of, [email protected], Wonjoon Kim Because of the short life cycle of movies, it is important to decide the release timing of motion pictures as well as to consider the movies’ attributes and demand seasonality. In this study, we compare the different patterns of decay effect and demand fluctuation between Hollywood and local movies in a non-US market. The authors find that the decay effect of Hollywood movies in a non-US market is greater than that of local movies. In addition, it is found that the effect of US holidays on seasonal demand for Hollywood movies in a non-US market is positive, while the effect is negative for local movies. This study contributes to a better understanding of the heterogeneous product nationality effect on product life cycle and demand fluctuation, especially for movie-importing countries where local and Hollywood movies competes. 3 - Misplaced Confidences: Privacy and the Control Paradox Laura Brandimarte, Carnegie Mellon University, Pittsburgh, PA, United States of America, [email protected], Alessandro Acquisti, George Loewenstein We introduce and test the hypothesis that control over publication of private information may influence individuals’ privacy concerns and affect their propensity to disclose sensitive information, even when the objective risks associated with such disclosures do not change or worsen. We designed three experiments in the form of online surveys administered to students at a North-American University. In all experiments we manipulated the participants’ control over information publication, but not their control over the actual access to and usage by others of the published information. Our findings suggest, paradoxically, that more control over the publication of their private information decreases individuals’ privacy concerns and increases their willingness to publish sensitive information, even when the probability that strangers will access and use that information stays the same or, in fact, increases. On the other hand, less control over the publication of personal information increases individuals’ privacy concerns and decreases their willingness to publish sensitive information, even when the probability that strangers will access and use that information actually decreases. Our findings have both behavioral and policy implications, as they highlight how technologies that make individuals feel more in control over the publication of personal information may have the paradoxical and unintended consequence of eliciting their disclosure of more sensitive information. 4 - Modeling Head-to-head Competition and Quality Decisions in Television Program Scheduling Erik Bushey, Doctoral Student in Marketing, University of Illinois, 350 Wohlers Hall, 1206 South Sixth Street, Champaign, IL, 61820, United States of America, [email protected], Udatta Palekar In this paper we address the question of when is head-to-head television scheduling optimal. Previous studies that have examined this problem did so when there were fewer television networks who were significantly more robust with respect to their program offerings. Over time as the number of television networks has increased, the breadth of their program offerings has decreased. This presents a problem for the previous findings. Initial studies assumed that when television networks were faced with potential head-to-head competition for a specific type of audience, they would always be able to offer programming that could appeal to a different type of audience, resulting in networks “sharing” all audience types throughout the evening. This is no longer a viable option as modern television networks programming portfolios have become more specialized as they attempt to cater to specific types of audiences. In this paper we design a game theoretical model to help us better understand the justifications for two recent and well publicized instances of head-to-head scheduling: the NHL deciding to compete directly with the NBA and WCW deciding to start the first “Monday Night Wars” with the WWE. We find that the decision to engage in head-to-head competition is affected by the cost to produce programing and the quality of available alternative programs. We also find evidence that the decision to engage in costly head-to-head programming may be the result of a prisoners dilemma, where networks would be better off cooperating but cooperation is not a Nash Equilibrium. We extend our model to account for the “lead-in” effect and find that this popular scheduling strategy only exacerbates the problem. 4 - Social Networks, Personalized Advertising, and Privacy Controls Catherine Tucker, Massachusetts Institute of Technology, Sloan School of Business, 100 Main Street, Cambridge, MA, United States of America, [email protected] This paper investigates how internet users’ perception of control over their personal information affects how likely they are to click on online advertising. The paper uses data from a randomized field experiment that examined the relative effectiveness of personalizing ad copy to mesh with existing personal information on a social networking website. The website gave users more control over their personally identifiable information in the middle of the field test. The website did not change how advertisers used anonymous data to target ads. After this policy change, users were twice as likely to click on personalized ads. There was no comparable change in the effectiveness of ads that did not signal that they used private information when targeting. The increase in effectiveness was larger for ads that used less commonly available private information to personalize their message. This suggests that giving users the perception of more control over their private information can be an effective strategy for advertising-supported websites. ■ SB08 Founders II Privacy and Marketing Cluster: Special Sessions Invited Session Chair: Catherine Tucker, Massachusetts Institute of Technology, Sloan School of Business, 100 Main Street, Cambridge, MA, United States of America, [email protected] 1 - The Impact of Relative Standards on the Propensity to Disclose Alessandro Acquisti, Carnegie Mellon University, Heinz College, Pittsburgh, PA, United States of America, [email protected], Leslie John, George Loewenstein Two sets of studies illustrate the comparative nature of disclosure behavior. The first set investigates how divulgence is affected by signals about others’ readiness to divulge. Study 1A shows a “herding” effect, such that survey respondents are more willing to divulge sensitive information when told that previous respondents have made sensitive disclosures. We provide evidence of the process underlying this effect 83 SB09 MARKETING SCIENCE CONFERENCE – 2011 ■ SB09 markets is optimal. Emerging markets have a larger proportion of cash-constrained consumers with limited ability-to-pay, compared to developed markets. In order to sell to such consumers in emerging markets, firms reset the prices of their smaller, comparatively cheaper package sizes. At equilibrium, this distortion in prices of small package sizes causes a reversal in relative quantity discounts offered by the firms. A limited field survey of Indian and American markets lends support to our theoretical findings. Founders III International Marketing I: General/Emerging Markets Contributed Session Chair: Sameer Mathur, Assistant Professor, McGill University, 1001 Sherbrooke West, Montreal, QC, H3A, Canada, [email protected] 1 - Global Expansion to vs. from Emerging Markets: An Empirical Study of Cross-border M & A’s Completion Chenxi Zhou, University of Florida, Marketing Department, P.O. Box 117155, Gainesville, FL, 326117155, United States of America, [email protected], Qi Wang, Jinhong Xie ■ SB10 Founders IV Health Care Marketing II Cluster: Special Sessions Invited Session While multinational companies have long been expanding into emerging markets, in recent years companies based in emerging markets have started to expand abroad. This new trend imposes challenges to global marketers and raises new issues for marketing scholars.This study investigates how the direction of global expansion:TO vs.FROM emerging markets, may affect the determinants of the completion probability of announced cross-border mergers and acquisition. Specifically, we empirically compare two types of cross-border M&As: (1) In-Bound, in which a firm from a developed economy acquires a firm in an emerging economy, and (2) OutBound, in which a firm from an emerging economy acquires a firm in a developed economy. Using data collected from the two largest emerging markets, China and India, from 1980 to 2009, our empirical analyses reveal some fundamental differences between the two types of cross-border M&As. For example, country-level factors, which measure the differences between the two countries involved (e.g., difference in political, trade, and legal environment), are the most influential factors for the successful completion of In-Bound M&As. However, firm-level factors, which measure the acquirer’s capability (e.g., firm size and experience), are the most influential factors for the completion of Out-Bound M&As. Furthermore, deal-level factors, which measure characteristics of the specific transaction (e.g.,percentage of stake sought by the acquirer, whether the deal size is disclosed, and whether the deal is paid by cash), have differential effects on the two types of M&As. These findings provide important managerial implications to global marketers on how to enhance the success of global expansions. Chair: Jaap Wieringa, Associate Professor, University of Groningen, Groningen, Netherlands, [email protected] Chair: Philip Stern, Professor, Loughborough University, Loughborough University, United Kingdom, [email protected] 1 - Product Bundling in Patent-protected Markets Eelco Kappe, Erasmus University Rotterdam, Erasmus School of Economics (ESE), Erasmus, Netherlands, [email protected], Stefan Stremersch What is the effect of product bundles in patent-protected markets? Combination drugs are an example of such a product bundle and have become a very popular method of extending the life cycle of drugs that lose patent protection. We disentangle the effects of introducing a combination drug (a combination of two molecules into one drug) on its components and the market, i.e. cannibalization vs. market expansion. Empirically, we investigate how the price and promotion of the bundle impact its components and vice versa. We also compare this to optimal promotion and pricing strategies under different organizational forms, like product management and category management. Using unique data on over 100 combination drugs in the pharmaceutical industry, we assess how these effects differ over drugs. The model allows for heterogeneous effects and corrects for the price and promotional endogeneity. 2 - Going Global: Why Some Firms from Emerging Markets Internationalize More than Others Sourindra Banerjee, University of Cambridge, 2360 Portland Street, Los Angeles, CA, 90007, United States of America, [email protected], Rajesh Chandy, Jaideep Prabhu 2 - How Generic Drugs Affect Brands Before and After Entry Jaap Wieringa, Associate Professor, University of Groningen, Groningen, Netherlands, [email protected], Peter Leeflang, Ernst Osinga The expiration of patents on branded drugs opens the door for lower priced generic products based on the same molecule. Typically, the combined sales of the branded and the generic product, i.e. total molecule sales, are lower after patent expiration than before. This decline has been hypothesized to follow from a reduction in marketing expenditures for the molecule, starting far ahead of patent expiration. Despite its importance, as far as we are aware of, no study has empirically tested the dynamics that are inherent to the introduction of a generic. We consider dynamics in the time period before and after the introduction of the generic. Relying on a nonlinear state space model in combination with importance-sampling based estimation, we analyze two major brands that went off-patent in 1997 and show that generic sales’ increases fully go at the cost of the corresponding brands. We conclude that the reduction in the brand’s marketing expenditures towards patent expiration is in line with an appropriate allocation of marketing expenditures on the long run. Through a simulation experiment we show that continued spending on marketing leads to higher molecule sales. Depending on the assumption one makes about the amount of marketing expenditures needed to offset competing brands’ marketing expenditures, we find that continued marketing expenditures may create cost savings for society. The recent rapid internationalization of firms from emerging markets has generated a great deal of interest among academics and practitioners alike. How do emerging market firms which lack foreign market knowledge, have inadequate financial resources and have weak institutional support in their home countries achieve such remarkable global expansion? In this paper we argue that a top management team member with international education and international experience play the role of a strategic actor who helps emerging market firms overcome their lack of foreign market knowledge. Further, we propose that emerging market firms surmount their inadequate financial strength by accessing international financial resources and circumvent their weak institutional environment through their affiliation with a business group. We test our hypotheses using uniquely compiled data on 173nonstate owned Indian firms from the Bombay Stock Exchange 500 index. Our results provide broad support for our thesis and offer important implications for research and practice alike. 3 - Multinational Strategic Alliance Models between Taiwan and China Shih-Wei Huang, National Taiwan Normal University, 5F., No.256, Sec. 1, Heping E. Rd., Taipei, 106, Taiwan - ROC, [email protected], Wun-Hwa Chen, Ai-Hsuan Chiang 3 - How, When and to Whom Should Pharmaceutical Innovations be Promoted? Katrin Reber, University of Groningen, Groningen, Netherlands, [email protected], Peter Leeflang, Philip Stern, Jaap Wieringa In the globalization competition environment, it seems hard for firms to use only their own resources to seek breakthrough in technology and marketing. We have found strategic alliance become a strategies for enterprises from multiple countries to share resources, co-develop technologies, learn important skills, enhance their competitive position, and then achieve the synergy effect. With increasing environmental awareness, the green energy industry such as LED attracted much attention from both managers and researchers. Our research will interviews with senior managers in Taiwan LED industries. We discuss how Taiwan and China enterprises form strategic alliance to integrate the supply chain, uniform standards, acquire technology, gain knowledge and then co-develop products and then entry into new markets. Pharmaceutical companies continually launch new drugs into the marketplace, and for these drugs, timely adoption by and diffusion across physicians is crucial to commercial success and the ability to recover the high and risky investments. In this study we investigate what marketing strategies are effective to ensure the successful market introduction of new drugs, and how marketing response varies by productrelated, decision-maker-related, and time-varying factors. Combining these three dimensions simultaneously we provide a broader picture of how these factors jointly influence the effectiveness of marketing communication. This helps understanding why and how new products are accepted differently, and may provide guidelines for managers regarding (1) the introduction of other innovations in the same category, and (2) the improvement of marketing resource allocation across products, physicians, and time. For our empirical analyses we have access to the complete prescription history of a panel of 1505 UK-based physicians over 23 year. We also have information on the detailing history of each of the physicians as well as on various physician characteristics. We base our analyses on a four-segment-trial-repeat model where we consider a time-dependent Marckovian structure for the transition probabilities. To accommodate physician and product heterogeneity in the model a hierarchical Bayesian setup is followed. 4 - Quantity Discounts in Emerging Markets Sameer Mathur, Assistant Professor, McGill University, 1001 Sherbrooke West, Montreal, QC, H3A, Canada, [email protected], Kannan Srinivasan, Preyas Desai Emerging markets like China and India are becoming dominant in the global economy. This paper contrasts the quantity discount decisions make by competing firms in such markets to the quantity decisions made in developed markets such as the United States. Anecdotal evidence indicates that while high-quality firms like Dove offer a larger quantity discount than comparatively low-quality firms like Garnier Fructis in the United States, the trend in opposite in India. This paper uses equilibrium analysis to explain why such a reversal in firm strategy in emerging 84 MARKETING SCIENCE CONFERENCE – 2011 4- Optimal Allocation of Marketing Resources: Employing Spatially Determined Social Multiplier Effects between Physicians Sina Henningsen, Department of Innovation, New Media, and Marketing, Christian-Albrechts-University at Kiel, Germany, [email protected], Soenke Albers, Tammo Bijmolt SB13 humans’ innate drive for uniqueness that leads them to avoid too much similarity. Therefore, we experimentally investigate whether the degree of similarity, manipulated by the first name of the requester, influences compliance with a request. 600 marketing and sales managers were invited to participate in an online survey. They were randomly assigned to one of three conditions: one third of the participants were invited by a person with the same first name (“high similarity”), the second third received the request by a person with same first name initials (“low similarity”), and the remaining third was contacted by a person with a completely different first name (“no similarity”). The conditions were identical with the exception of the name that appeared as the requester of the invitation. The results support our assumption of an inverted u-shaped relationship between the degree of similarity and humans’ willingness to comply. Persons with identical initials were significantly more likely to participate in the survey than both those who received the request by a person with different initials and those with same first name. We find no difference between high and no similarity conditions. A central aim of marketing managers is the optimal allocation of budgets across customer segments and market regions. Often, these allocation decisions are subject to interpersonal contagion. However, sales response models traditionally assume customers’ independence. We present a method for incorporating spatial spillovers into sales response functions and a subsequent optimization procedure that allows for allocating sales/profit optimizing budgets across segments and regions while accounting for spillover effects. Employing hierarchical Bayesian estimation, our results from an empirical application to a pharmaceutical panel dataset indicate that substantive additional sales/profits can be ex-pected due to incorporating spillovers and thereby improving allocation. 4 - User Personality, Perceived Benefits and Usage Intensity of Social Networking Sites: An Indian Study Duraipandian Israel, Associate Professor, School of Business & Human Resources, XLRI Jamshedpur, Circuit House Area (East), Jamshedpur, Jharkhand State, 831035, India, [email protected], Debasis Pradhan ■ SB11 Champions Center I Social Networking Sites (SNS) such as Orkut, Facebook and LinkedIn have attracted millions of users around the globe. While studies have been conducted to explore the user motives to participate in such virtual communities, research linking the impact of the personality and usage intensity is scantily done. Even so, the available studies indicate mixed evidence between user personality traits (such as extraversion, agreeableness, openness to experience, neuroticism and conscientiousness) and the SNS usage. Research linking user personality and SNS usage intensity is expected to have major strategic policy implications for marketers in strengthening the advertisement and promotions of their products as SNS are emerging as attractive medium for marketers to reach out to the target audience. This paper based on empirical research conducted among the youth (n=203) in India explores the impact of user personality traits on the SNS usage intensity, with perceived benefits in using SNS as a mediator. Mediation regression analysis of the data indicates that user personality trait has direct and indirect effect on the usage intensity of SNS. Further, it is observed that the relationship between user personality, perceived SNS benefits and the usage of SNS varies for male and female respondents. Social Influence II Contributed Session Chair: Duraipandian Israel, Associate Professor, School of Business & Human Resources, XLRI Jamshedpur, Circuit House Area (East), Jamshedpur, Jharkhand State, 831035, India, [email protected] 1 - Co-creation of Social Value in an Online Brand Community Kwok Ho Poon, DBA Student, Graduate School of Business/The Hong Kong Polytechnic University, Rm B, 11/F, Chau’s Comm Ctr., 282 Sha Tsui Road, Tsuen Wan, Hong Kong - PRC, [email protected], Leslie S.C. Yip This study explores one of the core features of a brand community, its social aspects which drive value co-creation among members of the community. Value co-creation emphasizes the dual role of customers as consumer and creator of value through interactions with the company and other customers, sharing experience on products and brands. This service-dominant (S-D) logic has not been adopted to examine brand communities which serve as an ideal platform for these interacting processes to take place. In this study, we try to examine empirically how iPhone users who are members of an online iPhone community interact and derive new values from community participation. It provides insights to practitioners on the value co-creation dimension of brand communities and gives future research direction on the social aspect of online marketing. ■ SB13 Champions Center III Private Labels II: Effect on the Distribution Channel Contributed Session 2 - What is There to ‘Like’ About Facebook? K N Rajendran, Associate Professor, University of Northern Iowa, 342 Curris Business Building, Cedar Falls, IA, 50614-0126, United States of America, [email protected], Steven B Corbin, Ciara Pearce, Matthew Bunker Chair: Alexei Alexandrov, Assistant Professor of Economics and Management, University of Rochester (Simon), Univ. of Rochester, Carol Simon Hall 3-110P, Box 270100, Rochester, 14627-0100, United States of America, [email protected] 1 - Retailer Brand: To Keep it Private or Not? Yunchuan Liu, Assistant Professor, University of Illinois at Urbana Champaign, 350 Wohlers Hall 1206 S. 6th St., Champaign, IL, 61801, United States of America, [email protected], Liwen Chen, Steve Gilbert Social media and social networks (like MySpace, Facebook, and LinkedIn) have proliferated and grown exponentially over the past decade or so. Even Hollywood has taken notice with a recent movie (“The Social Network”) tracing the origins of Facebook. There has been a tremendous interest in understanding aspects of the social network phenomenon by all manner of organizations, business in particular. Articles have been published in fields as diverse as educational technology (Baran 2010) , managerial psychology (Klumpfer and Rosen 2009), human resources (Elsweig and Peoples 2009), information technology (McAfee 2010, Palvia and Pancaro 2010, Venkatraman 2010), marketing communication (Zhang 2010), intellectual property (Steinman and Hawkins 2010), academic performance (Kirschner and Karpinski 2010, Yu et al 2010), among many others. Our primary interest is in understanding the behavior of ‘liking’ on Facebook. We collect and analyze survey information from a fairly large sample of respondents who self identify themselves as ‘liking’ a business or organization. In the study being presented, we compare characteristics and behavior of two samples; those that have actually experienced the good or service being provided by the organization they ‘like’ vs. those who have not had such experience. Preliminary results suggest there are interesting differences between the groups in measures pertaining to identity, norms, involvement, and word-of-mouth behavior, some of which appear to be counter-intuitive. A retailer with its own brand can either make its brand private by selling through only the retailer’s own stores or make the brand public by selling through another competing retailer. In the first case, the retailer brand becomes a private label; in the second case, the retailer brand becomes more like a national brand. In this paper, we study when a retailer should keep its brand private and when not, and the consequent implications for distribution channel members. We find conditions under which the retailer should keep its brand private or sell it through another retailer. We show that selling through a competing retailer can help expand a retailer’s market. However, it also has significant implications for the competition between the retailers, the distribution channel, and consumer welfare. 2 - Retailer Brand Introduction with Consumer Evaluation Ying Xiao, University of Illinois at Urbana Champaign, 350 Wohlers Hall 1206 S. 6th St., Champaign, 61801, United States of America, [email protected], Yunchuan Liu In the market place, there is a growing trend for retailers to introduce retailer brands (store brands) to complement manufacturer brands and enrich the in-store product variety. In this paper, we study the effects of a retailer brand on the profitability of a manufacturer and a retailer when consumers do not have full information about the products and have to incur a cost to evaluate the brands. We show that a manufacturer can benefit from a retailer brand, and the benefit may increase with the rising quality of the retailer brand. This happens when the introduction of a retailer brand motivates the retailer to induce consumer evaluation and the manufacturer can take advantage of that to charge a high wholesale price. Depending on the level of consumer evaluation cost, either a decentralized channel or a centralized channel can offer more product varieties. Furthermore, at certain evaluation costs, consumers can be better off in a decentralized channel than in a centralized one. 3 - Too Much or Not Enough - How the Degree of Interpersonal Similarity Forces Compliance with Requests Johannes Hattula, Research Assistant, University of St. Gallen, Institute of Marketing, Dufourstr. 40a, St. Gallen, CH-9000, Switzerland, [email protected], Sven Reinecke, Stefan Hattula Imagine you are asked for participation in a survey. Would your willingness to comply depend on characteristics of the requester? Would you be more likely to answer when there is a similarity between you and the requester? And, particularly, would the degree of similarity affect your willingness? Many studies provide support for the persuasive role of similarity on willingness to comply. When people share similarities, they feel socially connected that is enough to increase compliance with requests. However, contrary to that literature, research on uniqueness empathizes 85 SB14 MARKETING SCIENCE CONFERENCE – 2010 3 - Market Expansion Effort in a Common Retailer Channel with Asymmetric Manufacturers Serdar Sayman, Associate Professor of Marketing, Koç University, Rumeli Feneri Yolu, Sariyer, Istanbul, 34450, Turkey, [email protected], Gangshu Cai time periods while obtaining the revenues from selling the product. However, from a public policy perspective, a recall delay means that a defective product is on the market longer without the defect being remedied. To shed insights into factors that drive the recall decision, we use a unique secondary data set and focus on brandrelated factors that influence a firm’s decision to delay a recall. The findings of our study have important implications for the management of product recalls. This paper evaluates the market expansion effort in a common retailer channel where two manufacturers are selling products through a common retailer. We compare scenarios of no effort, manufacturers’ efforts, retailer’s effort, and hybrid effort – where either manufacturer or the retailer provides effort. Our results indicate that when the channel competition is relatively less intense, the retailer’s effort dominates the manufacturers’ efforts for both the manufacturers and retailers; while the manufacturers’ efforts might be advantageous when channel competition becomes too intense. We also study the case of store brand and one manufacturer brand. 3 - Virtual Events: An Emerging Tactic that Complements the World of Experience Marketing Nipun Agarwal, Global Events Strategist, IBM India Pvt. Ltd., Manyatha Embassy Business Park, Bangalore, India, [email protected] With the challenges and demands of globalization, in a highly competitive environment where there is explosive growth of information, the requirements for businesses, their executives, business partners and customers to come together and exchange knowledge, build networks, and nurture relationships has never been greater. In any organization, marketing plays a major role to own this responsibility. Most marketers achieve this objective through conducting “events”. But in today’s economy, it comes with challenges, when budgets are slashed and partner, customer are not much ready or can travel and on above this, the attendees who do participate, no follow up plan gets executed because of ineffective response lead management system. As a result, marketing are being forced to scrutinize every detail to prove the value of their events. Now, marketers feel that traditional events suffer from high costs, limited audience reach, low flexibility, and inconsistent outcomes as the market is more globalized for any organization. Now with more advanced technological capabilities like high speed internet, streaming video, life-like graphics – virtual events have arrived as a growing tactic which complements the world of experience marketing, supported with other web social media tools which enables marketers to start the conversation. Virtual Events provide both organization and attendees with many advantages which traditional physical failed to achieve. This paper will discuss on how different organization irrespective of their sizes, are using virtual events as an add-on marketing tactic in their overall marketing plan to create awareness, generate leads and collaborate with customers and partners. 4 - Effects of Manufacturers’ Advertising on Volumes, Retail Margins, and Retail Profits Alexei Alexandrov, Assistant Professor of Economics and Management, University of Rochester (Simon), U of Rochester Carol Simon Hall 3-110P, Box 270100, Rochester, 14627-0100, United States of America, [email protected] I show that when a retailer is selling two symmetric products, each produced by an independent manufacturer, higher product differentiation results in higher wholesale and retail prices (as opposed to prior theoretical literature), but lower retailer margin and profit. If one of the products is retailer’s private label, and the private label is not perceived to be much worse than the national brand, then as product differentiation increases, the retailer’s margin decreases, so does the volume sold of the national brand, and so does the volume sold of the private label. If the private label product is perceived to be much worse, then some or all of the effects above might be reversed. My paper also offers another explanation to the empirical finding of advertising’s opposite effects on the retailer margin and the wholesale price. ■ SB14 ■ SB15 Champions Center VI Champions Center V Marketing Strategy III: General CRM VIII: Customer Lifetime Value Contributed Session Contributed Session Chair: Nipun Agarwal, Global Events Strategist, IBM India Pvt. Ltd., Manyatha Embassy Business Park, Bangalore, India, [email protected] 1 - To Research or to Execute? Analysis of the Drivers of Marketing Performances Chiara Saibene, SDA Bocconi School of Management, Via Bocconi 8, Milan, 20100, Italy, [email protected], Fabio Ancarani Chair: Peter Pal Zubcsek, University of Florida, Department of Marketing, 212 Bryan Hall, P.O. Box 117155, Gainesville, FL, 32611-7155, United States of America, [email protected] 1 - Churn Prediction Using Bayesian Ensemble in Telecommunications Market Jaewook Lee, Associate Professor, POSTECH, San 31, Nam-gu, Pohang, Korea, Republic of, [email protected], Namhyong Kim An innovative, broader and integrated approach to marketing research and to customer insight management, leading to more effective marketing strategies, is key for gaining sustainable competitive advantage. In fact, MSI underlines among Top Tier Priorities 2010-2012 the following issues: a) using market information to identify opportunities for profitable growth; b) understanding customer experience and behavior; c) developing marketing capabilities for a customer focused organization; d) leveraging research tools and new sources of data. Are companies really able to transform market information into strategic marketing decisions leading to competitive advantage and superior performances? If not, are there other drivers different from analytic and strategic marketing competences leading to superior performances? We conducted in fall 2010 a research on 300 European Marketing and Sales Managers. We measured companies’ ability to manage marketing and sales competencies, the use of marketing metrics and perceived and objective companies’ performances. As regards market research, strategy and performances, we find counterintuitive results: a) companies’ execution competences are the most important driver of performance; b) competences related to market research and understanding and strategic decision making competences do not have a statistical positive impact on performance. These main findings show that, in our sample, marketing information does not translate into differential marketing strategies and that execution is, at the end, the main driver of performance. We then deepen our research on the financial industry because in this industry marketing and market information competences are playing a more and more critical role. Correct predictions in the field of marketing play an important role in the customer management and maintenance and are essential in direct marketing and micromarketing which focus on a specific group of customer. Recently, churn prediction becomes a major issue of telecommunications market, one of the most competitive industries with 20-40 % of customers leaving their provider in a given year. The objective of churn management for target marketing is to maximize the value of the company by minimizing the churn rate through identifying the customers who are likely to leave and enhancing the management activities for those customers. In this study, we consider the customer data provided by a wireless telecommunications company where the data are large-scaled and highly imbalanced as well as the two class of data highly overlap each other. Due to these reasons, the various conventional data mining techniques are not so successful to get meaningful results. To overcome this problem, we propose a novel principled Bayesian ensemble model that optimally aggregates prediction results and provide generative process based on variational analysis. The results of applying the proposed method to our dataset show significantly better performance than the existing methods in terms of predictions and complexity as well as provide an important predictive property of the generative models that are crucial for supporting managers’ flexible and timely pricipled decisions. 2 - Improved Churn Prediction With More Effective Use of Customer Data Özden Gür Ali, Koc University Rumeli Feneri Yolu, Sariyer, 34450, Istanbul, Turkey, [email protected], Umut Ariturk, Hamdi Ozcelik 2 - Recall Now or Recall Later: Investigating Drivers of a Firm’s Decision to Delay a Recall Meike Eilert, University of South Carolina, 1705 College St, Columbia, SC, 29208, United States of America, [email protected], Kartik Kalaignanam, Satish Jayachandran In this paper we focus on the customer churn prediction problem in the highly dynamic banking industry of emerging markets with ever-changing expectations of consumers, increasing diversity in products and services offered, and abrupt changes in economic conditions. Traditional churn models are typically estimated on cross sectional data pertaining to a particular time period and applied on subsequent periods, which is appropriate under static environments. On the other hand, customer behavior responds to changes in the environment. We show that using longitudinal data along with dynamic variables describing the customer experiences, and the economic environment improves prediction accuracy independent of the model used. Further, we propose the use of ordinal logistic regression to capture the time to churn in non-contractual settings and evaluate its impact on predictive accuracy of customer churn. Every year, numerous products are recalled because they violate safety standards and pose a hazard to consumer well-being. In 2008, the Consumer Product Safety Commission supervised 465 recalls, a 36 percent increase from the number of recalls issued in 2000. To date, most research in the product recall area has focused on the consequences of a recall on stakeholder attitudes and behaviors towards the firm. However, little is known regarding when firms recall defective products. We address this gap in research by investigating the factors that influence the firm’s recall decision, particularly the decision to delay a recall. A recall delay refers to the time period between the firm being aware of a potential safety problem and its decision to issue a recall. By delaying a recall, the firm can move recall-related costs to future 86 MARKETING SCIENCE CONFERENCE – 2011 SC02 decided the optimal strategy in a two-sided market where it collects revenue from advertisers and from subscribers. We use a two-stage model where the demand model is a random coefficient logit system and the supply model is a dynamic programming system. The company makes its decision based on the demand that it observes, i.e., demand model is estimated first and then the demand parameters are used to estimate the supply model. In conclusion, the online gaming content is an area where consumers are willing to purchase. Furthermore, the optimal number of total subscribers to have before moving to a paid subscription is roughly 100,000. If the company is considering a second paid subscription plan, it should be introduced after having 200,000 subscribers and should be a low-end plan compared to the first paid subscription plan. Compared to the myopic model, the dynamic model predicts that the company can achieve earlier introduction of paid subscription plans. 3 - Information Communities: The Network Structure of Communication Peter Pal Zubcsek, University of Florida, Department of Marketing, 212 Bryan Hall, P.O. Box 117155, Gainesville, FL, 32611-7155, United States of America, [email protected], Imran Chowdhury, Zsolt Katona This study puts forward a variable clique overlap model for identifying information communities, or potentially overlapping subgroups of network actors among whom reinforced independent links ensure efficient communication. Using simple network analysis methods, we derive the definition of information communities from a parsimonious information-theoretic model of communication. We posit that the intensity of communication between individuals in information communities is greater than in other areas of the network. Empirical tests on two communication networks show that the variable clique overlap model is more useful for identifying groups of individuals that have strong internal relationships in closed networks than those defined by more general models of network closure. Our findings thus extend the scope of network closure effects proposed by other researchers working with communication networks using social network methods and approaches, a tradition which emphasizes ties between organizations, groups, individuals, and the external environment. Further, our method has potential marketing applications in segmenting networked markets, refining customer lifetime value estimates, and targeting viral campaigns. 4 - Contextual Advertising Kaifu Zhang, INSEAD, Blvd de Constance, Fontainebleau, 77300, France, [email protected], Zsolt Katona This paper studies the strategic aspects of contextual advertising. Contextual advertising entails the display of relevant ads based on the nature of the content that a consumer views and exploits the possibility that consumers’ content browsing preferences are indicative of their product preferences. Typically, a contextual advertising intermediary manages the advertising slots next to the content and sells the advertising space, usually through a second- price auction. Our results show that contextual targeting impacts advertiser profit in two ways: first, advertising through relevant content topics helps advertisers reach consumers who have strong preferences for their products. Second, heterogeneity in consumers’ content preferences can be leveraged to reduce product market competition, even when consumers are homogeneous in their product preferences. These effects lead to higher advertiser revenues. The intermediary profits from the sale of advertising space and may have an interest in increasing second-price bids at the expense of advertiser. Saturday, 1:30pm - 3:00pm ■ SC01 Legends Ballroom I Advertising and Two Sided Markets ■ SC02 Contributed Session Legends Ballroom II Chair: Kaifu Zhang, INSEAD, Blvd de Constance, Fontainebleau, 77300, France, [email protected] 1 - The Impact of Advertising on Media Bias Tansev Geylani, University of Pittsburgh, Katz School of Business, 320 Mervis Hall, Pittsburgh, PA, 15260, United States of America, [email protected], T. Pinar Yildirim, Esther Gal-Or Auctions and Pricing Cluster: Internet and Interactive Marketing Invited Session Chair: Woochoel Shin, Assistant Professor, University of Florida, Department of Marketing, P.O. Box 117155, Gainesville, FL, 32611, United States of America, [email protected] 1 - Lemony Prices: An Online Field Experiment on Price Dispersion Zemin Zhong, Peking University HSBC Business School, N520, PKU, Univ.Town, Xili, Nanshan, Shenzhen, GD, 518055, China, [email protected], David Ong In this study, the authors investigate the role of advertising in affecting the extent of bias in the media. When making advertising choices, advertisers evaluate both the size and the composition of the readership of the different outlets. The profile of the readers matters since advertisers wish to target readers who are likely to be receptive to their advertising messages. It is demonstrated that when advertising supplements subscription fees, it may serve as a polarizing or moderating force, contingent upon the extent of heterogeneity among advertisers in appealing to readers having different political preferences. When heterogeneity is large, each advertiser chooses a single outlet for placing ads (Single-Homing), and greater polarization arises in comparison to the case that media relies on subscription fees only for revenues. In contrast, when heterogeneity is small, each advertiser chooses to place ads in multiple outlets (Multi-Homing), and reduced polarization results. We challenged the law of one price in an online field experiment. In the experiment, prepaid phone cards were sold on taobao.com, China’s largest online market with sales already equaling eBay in 2010. We show that consumers will not always buy the lowest priced product, all else being equal. Our field experiment design has two important advantages. We ruled out many confounds in prior empirical studies, especially unobservable heterogeneity. Our field experiment design eliminated the external validity issue of the few laboratory experiments. The main result is that out of 594 sales, 94 were of the higher priced item. Higher priced items constituted 25% of the sales when the price gap was 0.05%. This decreased nearly linearly to 1% when the gap was 2.5%. Our findings contribute to the “online price dispersion” paradox literature. 2 - Matching Markets for Contextual Advertising: The Tao of Taobao and the Sense of AdSense Chunhua Wu, Washington University in St. Louis, One Brookings Drive, Saint Louis, MO, 63130, United States of America, [email protected], Kaifu Zhang, Tat Y. Chan 2 - Two-dimensional Auctions for Sponsored Search Amin Sayedi, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, [email protected], Kinshuk Jerath Contextual advertising enables the targeted delivery of relevant ads based on the Internet content a consumer views. Critical to the success of contextual advertising is the ability to accurately match the product being advertised with the relevant on-line content. To achieve this goal, traditional contextual ad platforms such as AdSense relies on sophisticated page analysis algorithms to centrally allocate ads to content pages. In contrast, other platforms such as the Chinese Taobao create a two-sided market where advertisers and publishers self select. We provide an empirical comparison of these different mechanisms. Using data from the Taobao.com contextual ad matching market, we estimate the valuation function for advertisers and investigate the determinants of successful matching. Using the estimated valuation function, we perform a counterfactual analysis wherein the matching in Taobao is done in a centralized fashion. Our results show that the market based mechanism is more efficient in terms of total social welfare, yet the central planner mechanism may be more profitable for the platform. As sponsored search becomes increasingly important as an advertising medium for firms, search engines are exploring more advanced bidding and ranking mechanisms to increase their revenues from sponsored search auctions. For instance, MSN, Yahoo! and Google are investigating auction mechanisms in which each advertiser submits two bids: one bid for the current display format in which multiple advertisers are displayed, and one bid for being shown exclusively. If the exclusive-placement bid by an advertiser is high enough then only that advertiser is displayed, otherwise multiple advertisers are displayed and ranked based on their multiple-placement bids. We call such auctions two-dimensional auctions and study two modifications of the GSP mechanism that Yahoo! has recently proposed. We show that allowing the advertisers to bid for exclusivity always generates higher revenues for the search engine; however, it might be better or worse for the advertisers depending on their values for exclusivity as well as the heterogeneity in their values for exclusivity. If the heterogeneity across the advertisers is large enough, we see that both social welfare and search engine revenue increase in two-dimensional auctions. Furthermore, when exclusive bidding is allowed, even if an exclusive display is not the outcome, the search engine still extracts higher revenue because allowing two bids increases the competition between the advertisers. Finally, if the heterogeneity across the advertisers is very large, it is better for the search engine to use an exclusive-only mechanism. 3 - Is Online Content Worth Paying For?: A Two-sided Market Approach Jinsuh Lee, Purdue University, 403 W. State St., W. Lafayette, IN, 47907, United States of America, [email protected], Manohar Kalwani This paper examines whether online companies can charge fees for online information to the consumers. We investigate this issue for an online gaming company which decided to move from free subscription to paid subscription and subsequently introduced a second paid subscription option. In essence, this company 87 SC03 MARKETING SCIENCE CONFERENCE – 2011 3 - Does Higher Transparency Lead to More Search in Online Auctions? Peter T. L. Popkowski Leszczyc, University of Alberta, Edmonton AB, Edmonton, T6G 2R6, Canada, [email protected], Ernan Haruvy explains the role of Government in marketing sustainable development. The study significantly contributes to the non-profit marketing theory and practice. 2 - Linkages between Infrastructure and Consumption Demand in Emerging Markets Puja Agarwal, Doctoral Student, INSEAD, 1, Iyer Rajah Avenue, Singapore, Singapore, [email protected] In a controlled field experiment, we examine pairs of auctions for identical items under different conditions. We find that auction design features that are under the control of the auctioneer – including information transparency, number of simultaneous auctions, and the degree of overlap between simultaneous auctions – affect bidder search and optimization. Clickstream data show that a significant relationship between information transparency and price dispersion can be linked to search. Specifically, information transparency is fully mediated by lookup behavior, while number of concurrent items is partially mediated. Combining these findings, we make auction design recommendations. The recent financial crisis has highlighted the role of emerging economies as the drivers of global growth. The emergence of consumer markets in Asia, Africa, Latin America, and the Mideast provides a new menu of opportunities for manufacturers, marketers and service providers. The conventional wisdom in the business press is that private consumption takes off when GDP/capita crosses a certain threshold (e.g., $3500). But the reality is that this has not happened. This research attempts to shed light on the key drivers of private consumption growth across countries, products and services by going beyond the role of income. Employing historical data on expenditures across different consumption baskets in 78 countries, this paper focuses on the role of infrastructure, both physical and financial, in determining the shares of different product categories in consumers’ budget. We use Deaton and Muellbauer’s (1980) Almost Ideal Demand System (AIDS) as the empirical methodology to analyze the impact of infrastructure on expenditure shares. Not only do we find that infrastructure and depth of credit markets are important in determining the relative shares of various product categories, but also highlight the asymmetric effect at broad category and sub-category levels. We also find heterogeneity in the infrastructure elasticity of demand for same product categories across OECD and Non OECD (emerging) countries. Across emerging markets, a boom in infrastructure building is underway. In this context it is important for marketers to not just identify key infrastructure variables but also its impact in heightening or dampening demand for their products and services. 4 - First-page Bid Estimates and Keyword Search Advertising: A Strategic Analysis Woochoel Shin, Assistant Professor, University of Florida, Department of Marketing, P.O. Box 117155, Gainesville, FL, 32611, United States of America, [email protected], Preyas Desai, Wilfred Amaldoss With the help of the technological advancement, search engines can use advertiserspecific information in devising the mechanism of the keyword search auction. In contrast to the universal minimum bid used in the traditional auction format, search engines now provide advertiser-specific minimum bids (ASMB) or First-Page Bid Estimates (FPBE) in their keyword search auctions. In this paper, we investigate profit implications of these mechanisms in the setting where the per-click valuation of advertisers is uncertain. Since FPBE is only an estimate and is not strictly enforced by the search engine, advertisers might choose not to conform to it and thus, FPBE is expected to generate lower profits to the search engine than ASMB. Contrary to this naÔve intuition, we show that FPBE dominates ASMB in terms of the search engine profit. Moreover, FPBE can achieve higher search engine profits than in the auction without minimum bids, thus proving itself to be the optimal mechanism for the search engine. We further discuss the possibility that the search engine may manipulate the listing order by use of ASMB or FPBE. Finally, we discuss the implication of using FPBE to the search engine’s effort to fight against the click fraud. 3 - Poverty (Tenure) Track Daniel Shapira, Ben-Gurion University, Ben-Gurion University, Beer Sheva, Israel, [email protected], Eran Manes We put forward a general theory that captures the long-term interplay between incentives and performance of individuals in teams, where human-capital externalities and spillovers (peer effect) are present, focusing on the case of research departments within academic institutions. Our model traces the dynamic evolution of two separated regimes; one in which quality dominates, and the other in which quantity substitutes quality. In both regimes, reward structures are endogenously awarded so as to elicit team members to perform in accordance. The existence of a competitive market for academics perpetuates rather than assists the disentanglement from the poverty trap regime. We also provide empirical evidence which lend strong support to the theory. Our theory and findings have far-reaching managerial implications. They imply that even at the expense of a short-term downfall in both performance and ranking, decision makers in academic institutions must provide incentives to encourage high-quality research. ■ SC03 Legends Ballroom III Meet the Editor: Journal of Service Research Cluster: Meet the Editors Invited Session 4 - The Nature of Informal Garments Markets: An Empirical Examination in Emerging Economy Prashant Mishra, Associate Professor, Indian Institute of Management Calcutta, Joka, Calcutta, India, [email protected], Gopal Das Chair: Sharad Borle, Rice University, Houston, TX, 77005, United States of America, [email protected] 1 - Meet the Editors Editors of the leading Journal of Service Research will present their editorial policies and perspectives. The informal markets play a significant role in urban areas. They provide a variety of low-priced goods, generating employment for a large number of people. Many authors in the past have lamented that in most Asian countries the contributions of informal market sellers are hardly ever recognized by the governments and society. Rather, the policy makers and other elements of local governance in developing Asian countries like India attempts to control informal sector activities and elements, such as peddlers, street-side garments and food outlets through various mechanisms including coercion. However, in spite of trying, the local authorities were unable to suppress these markets. That means the consumers are accepting rather purchasing the products from the informal markets, in spite of developments in formal segments. Although the informal business activities have been quite profound and are expanding rapidly in India and other Asian countries, there has been hardly any research done on this age old marketing phenomenon in the Indian context. The objective of this paper is to fill this research vacuum by exploring the factors influencing Indian consumer choices while purchasing garments from informal markets. The study proposes survey 200 consumers frequently participating in purchases from informal garments markets in the city of Kolkata, Mumbai and Delhi and Chennai all metropolis located in different parts of India. The outcome of the study is expected to help develop insights into reasons behind continued consumer patronage of informal bazaars despite the onslaught of organized retail and modern developments of retail front. The outcome is also expected to contribute to the debate on the economic vs social drivers for continued existence of the old and the new formats of the markets in developing economies. ■ SC04 Legends Ballroom V Unique Topics 2 Contributed Session Chair: Nithya Rajamani, Researcher, IBM India Research Labs, Gachibowli, Hyderabad 32, Hyderabad, 32, India, [email protected] 1 - Role of Government in Marketing Sustainable Development: An Exploratory Investigation V. Mukunda Das, Professor and Director, Chandragupt Institute of Management, Phaneeshwarnath Renu Hindi Bhawan, Chajjubagh, Patna, BI, 800001, India, [email protected], Saji K B Sustainable development is concerned with meeting the needs of people today without compromising the ability of future generations to meet their own needs. Sustainable development therefore involves: (i) a broad view of social, environmental, and economic outcomes; (ii) a long-term perspective, concerned with the interests and rights of future generations as well as of people today; and (iii) an inclusive approach to action, which recognizes the need for all people to be involved in the decisions that affect their lives. Marketing the notion of sustainable development has become essential these days as most of the stakeholders to the development processes are often unaware of the undesirable consequences of the short term intent of their actions. The Government can play a very crucial role in this context. Although there are observations and arguments available galore to this direction, the extant literature in marketing is silent on the role of Government in marketing sustainable development. To address this research gap, we conducted an exploratory study that investigated the potential antecedents for explaining the role of Government in marketing sustainable development. For the exploration purpose, we have considered the case of the present Government in the Indian state of Bihar, which ever since its inception is working towards the cause of sustainable development through several carefully crafted social projects like JEEViKA (Bihar Rural Livelihoods Project). The present paper reports a theoretical framework that 88 MARKETING SCIENCE CONFERENCE – 2011 ■ SC06 SC07 4 - Can CRM Create Goal Incongruence Among Salespeople and Their Firms? An Agency Theory Perspective Doug Walker, Assistant Professor, Iowa State University, Department of Marketing, 2350 Gerdin Business Building, Ames, IA, 50011, United States of America, [email protected], Eli Jones, Keith Richards Legends Ballroom VII Consumer Preferences Contributed Session Chair: Doug Walker, Assistant Professor, Iowa State University, Department of Marketing, 2350 Gerdin Business Building, Ames, IA, 50011, United States of America, [email protected] 1 - Awareness and Ability to Express Preferences and its Impact on the Establishment of Causal Relations Rubén Huertas-García, Department of Economics and Business Organization. University of Barcelona, Main Building,Tower 2,3rd f, Diagonal 690, Barcelona, 08034, Spain, [email protected], Paloma Miravitlles-Matamor, Esther Hormiga, Jorge Lengler Numerous publications tout the benefits of Customer Relationship Management (CRM), but generally overlook the fact that obtaining customer value can be a matter of perspective. Stakeholders within the firm may differ in their valuations of prospects and customers, even when analyzing the same data. In a sales setting, the expected value of a sales target to a profit maximizing firm with an infinite time horizon can be far different than the valuation assigned to the same target by a riskaverse, utility maximizing salesperson with a limited time frame. The sales control prescriptions from the agency theory literature result from models that incorporate concave sales curves, indicating sales targets are prioritized by their expected values. However, the implicit assumption is that firms and their salespeople share identical rankings of these targets. In practice, however, salespeople are generally autonomous, and firm and salesperson valuations can diverge due to uncertainty, risk aversion and time horizon. Implementing CRM, through both increased data gathering and more sophisticated data analysis, allows both firms and their salespeople to improve the precision of sales target valuation from each perspective – thus creating goal incongruence between the principal and the agent. In situations where potential targets are heterogeneous in terms of risk and time horizon, CRM-generated information results in the need for costlier sales controls to combat increased levels of goal incongruence between salespeople and the firm. This study’s contributions appeal to academics and practitioners. Studies on consistency of individual preferences have been based on two areas of knowledge: economic analysis of consumer decision-making, and information processing theory (Frank, 2005; Simon, 1990). However, numerous precedents and characteristics contribute to people lacking sufficient insight to express their own preferences (Kramer, 2007). Decision-making processes depend on the extent to which preferences have been formulated. Thus, people who do not have a mental reference have a more complicated task to carry out (Chernev et al., 2003) and are consequently more likely to use a heuristic to simplify the process. This study analysed whether participants’ levels of awareness and ability to express their preferences has an impact on the capacity to establish causal relationships in marketing research. We used an instrument that is based on definitions of stability and equivalence to assess and verify participants’ degree of consistency in a choice experiment on international entrepreneurial intentions. Our hypothesis is: Individuals who express a more consistent system of preferences in their entrepreneurial intentions will generate more significant causal relationships in subsequent evaluations than those who express a less consistent system of preferences. The sample consisted of 127 university students. The results appeared to support the hypothesis, if the level of consistency of the subjects is not taken into account; a large amount of statistical noise is generated. In such cases, the sample size needs to be increased to obtain acceptable results. Such mechanisms could be used to select participants’ degrees of consistency on the topic under study before they are surveyed. ■ SC07 Founders I Entertainment Marketing III Contributed Session Chair: Dominik Papies, University of Hamburg, Welckerstr. 8, Hamburg, 20354, Germany, [email protected] 1 - Influence of Film Adaptation on Motion Picture Performance: Experiences on SF Films in Hollywood Sunghan Ryu, KAIST Business School, S313 Supex Bldg, 87 Hoegiro Dongdaemoon-Gu, Seoul, Korea, Republic of, [email protected], Young-Gul Kim, Jae Kyu Lee 2 - A Bayesian Approach to Estimating Demand for Product Characteristics: An Application to Coffee Purchase in Boston Margil Funtanilla, Graduate Research Assistant, Texas Tech University, MS 42132, Lubbock, TX, 79409, United States of America, [email protected], Benaissa Chidmi In Hollywood, film adaptation is one of the most important sources for producing motion pictures in both of historical and industrial aspects, and SF genre is the most remarkable one of whole film adaptation cases. Even though hundreds of researches about relationship between motion picture performance and motion picture related factors have been executed, there is no research focusing on the special context such as film adaptation. Hence, in the current research, researchers examined the impact of several factors on motion picture performance in the context of SF film adaptation based on 51 cases from 1980 to 2008. The result shows quality factor (critics’ review) and scale factor (budget and number of screen) have significant impact on motion picture performance which is consistent with previous literature, and some production factor such as special effect and MPAA rating also have considerable influence on the performance. In the case of film adaptation related factors, the result verified that title adaptation has the biggest impact on both of short term and long term performance. Author power, experience of director and remake/sequel were declared that they have limited influence on motion picture performance. The current research has implications for both of production and marketing aspects in motion picture industry. Especially for marketing perspective, one of the long held beliefs in the industry is proved that utilizing the title of original novels and author power would be a proper approach because people are attracted by more familiar things than totally new one. Finally, future research areas reflecting limitations of the current research are proposed. As coffee remains to be one of the popular beverages in the United States which translates to intense competition between markets, it is useful to investigate the demand for this product’s characteristic conferred by the production process as a determinant of consumers’ choice. Just like ready-to-eat breakfast cereals, coffee provides an interesting multitude of choices available that will challenge the taste buds but even retailer-level demand estimation on coffee are less common in the economic literature relative to richness of resources on the former as past studies have focused more on the supply side. Ever since Lancaster (1966) presented his product characteristics approach to consumer demand, several studies attempted to challenge some of the assumptions underlying the model and put forward a more simplified approach and some with much relaxed assumptions that are suited with aggregate data common in marketing and economics. Ladd and Suvannunt’s (1976) Consumer Goods Characteristics Model looks upon a product as a collection of characteristics and the utilities that are derived from consuming this product depend upon these characteristics. For instance, some consumers may enjoy a cup of instant French roast coffee relative to instant Columbian roast from the same grocery shelves. In essence, the total amount of utility consumers’ get from coffee consumption depends upon the product characteristics embedded in each brand. Using scanner data from Information Resource, Inc. (IRI) on 14 coffee brands over 156 weekly periods, this paper estimates the demand for these coffee product characteristics by recovering relevant parameters in a Bayesian fashion. This will shed light on the substitution pattern as well as identify consumers taste for product characteristics. 3 - An Anti-ideal Approximation of the Mixed Logit Model Robert Bordley, General Motors, 30500 Mound Road, Warren, MI, United States of America, [email protected] We focus on choosing a product based on its scores on various attributes. Ideal point (and to a lesser extent, anti-ideal point) models provide visually appealing product positioning maps. This can be useful in decision-making. But the mixed logit model is technically more appealing. This paper presents an approximation of the standard mixed logit model which is equivalent to an anti-ideal point model based on the standardized attribute scores of each product. The author is looking for collaborators to help extend and apply this approach. 89 SC09 MARKETING SCIENCE CONFERENCE – 2011 2 - Buy-now Prices at Entertainment Shopping Auctions Jochen Reiner, PhD Student, Goethe-University Frankfurt, Grueneburgplatz 1, Frankfurt am Main, 60323, Germany, [email protected], Martin Natter, Bernd Skiera ■ SC09 Entertainment shopping auctions (ESA), also labeled as pay-to-bid or penny auctions, recently gained popularity on the Internet. ESA require all bidders to pay for their bids. As a consequence, providers of ESA realize most of their profits by the bidding fees, in particular by the ones that are paid by the losers of the auction. As these foregone bidding fees decrease customer satisfaction, providers of ESA introduced Buy-Now Prices (BNPs). BNP means that a bidder can, at any time during the auction, put the full or part of the bidding fees previously placed during the auction toward buying the auction item. At the first glance BNPs are a favorable, risk reducing, option for the bidders. However, due to the reduced risk it is likely that the behavior of the bidder changes, e.g. that they bid more aggressively. As a consequence this “customer friendly” option might cause higher effort and lower chances of bidders to win an auction. Additionally, as BNPs limit losses of bidders, they might also reduce profits of auctioneers. Our study investigates analytically and empirically the consequences of BNPs on auctioneer’s profit and bidder behavior. Therefore, we empirically compare the results of 7,104 auctions, in 12 categories, including 6.508 million bids from 334,626 unique registered bidders. 5.807 of these auctions used BNP and 1.297 auctions did not. We find a positive effect of BNPs on the number of bidders (+7.55 per auction) and the number of bids (+3.39 bids per auction). In total, auctioneers and bidders benefit from BNPs, as we also find that BNP increase loyalty. Contributed Session Founders III International Marketing II Chair: Fareena Sultan, Professor of Marketing/Robert Morrison Fellow, Northeastern University, College of Business Administration, 202 Hayden Hall, Boston, MA, 02115, United States of America, [email protected] 1 - Beyond Globalization: Effectiveness of Technology Strategies of Foreign Firms in China Bennett C. K. Yim, Professor in Marketing, University of Hong Kong School of Business, Meng Wah Complex, 729M, Pokfulam Road, Hong Kong, Hong Kong - PRC, [email protected], Caleb Tse, Eden Yin Current shifts in global market landscape and technology endowments are reshaping the ways firms strategize their technology competence in foreign market operations. Many of them have adopted technology strategies beyond what the globalization and glocalization paradigms would prescribe. Against the prediction of global marketing strategy literature, a number of prominent multinational corporations have established R&D centers in developing countries to develop new products for both local and global markets; a strategy labeled as reverse innovation by Immelt, Govindarajan and Trimble (2009). Others have developed risk-sharing partnerships with local firms to drive collaborative R&D. Despite anecdotal evidences of some early successes, the effectiveness of these technology strategies has not been substantiated. Through the resource-based view, this study conceptualizes the evolution of international marketing paradigms and develops hypotheses to test the effectiveness of foreign (both wholly owned and IJV) firms’ technology strategies within boundary conditions of cumulated knowledge asset and government support in the China market. Results derived from a sample of almost 1,200 firms operating in China support the effectiveness of technology strategies involving local R&D investments and technology staff capability, particularly for firms with endowed knowledge asset. Foreign firms’ technology strategies also benefit from organizational alliances (forming IJVs) through effective use of government support. The results substantiate the criticalness of developing technology strategies in the present global economic landscape. This study also contributes to the international marketing literature and provides managerial implications for foreign firms operating in China. 3 - Testing Strategies in Hollywood: A Duopolistic Game vs. an Agent Based Model Sebastiano Delre, Assistant Professor, Bocconi University, Via Roentgen 1, Milan, Italy, [email protected], Claudio Panico We develop a competition game where two movie producers compete to attract a population of consumers with a different taste parameter. We start from a simple analytical solvable model and from its unique Symmetric Nash Equilibrium (SNE). Then we implement the game into an Agent Based Model (ABM) that can replicate and extend the analytical model. Our extension consists of creating an experimental design that allows us to test four different realistic strategies used by the studios: “the launching strategy,” “the competing on shares strategy,” “the growth strategy” and “the competing on profits strategy”. We found that only 7 out of 576 simulation scenarios Pareto-dominate the SNE and that they cannot substantially increase the profits of the two studios. Moreover we found that “the launching strategy” is usually overestimated and that the most efficient strategy is the “the competing on shares strategy”. However “the competing on shares strategy” has to be adopted only when the competitor either does not use “the launching strategy” or it does use it as a “blockbuster strategy”never when the competitor uses a “sleeper strategy”. 2 - National Influencers on Adoption and Usage of Online Auction Websites: New Zealand, Germany & Korea Tony Garrett, Korea University Business School, Anam-Dong, Seongbuk-Gu, Seoul, 136-701, Korea, Republic of, [email protected], Jong-Ho Lee, Stefan Bodenberg The emergence of online auction websites is a major element in e-commerce yet relatively little is known about national differences and determinants of its adoption and usage. National environment differences could be a factor in the adoption and usage of online auction websites. Differences in the levels of usage have not been examined in previous research. After an extensive review of the literature an extended Technology Acceptance Model (TAM) questionnaire is developed and administered on a sample of users of online auction website in three diverse national cultural environments: New Zealand, Germany and Korea. Additionally attitudes, behavior and usage patterns are examined to determine the factors that influence the adoption and usage of the medium. Another objective is to test the extended TAM’s cross-national robustness. Results suggest that although the core TAM is robust for overall adoption behavior in New Zealand and Korea and mainly for Germany, there are some differences in the extended model between New Zealand and Germany. This is not the case when the levels of usage are examined in detail. The core TAM model is relatively robust in predicting light usage however it has relatively low prediction of heavy usage in all of the national samples. Attitudes and behaviors between heavy and light users and the national samples are examined to provide explanations for the differences. Results for example show that the Korean sample has higher levels of excitement and lower levels of perceived usefulness and positive past experiences to the use of online websites relative to the other samples. Further results along with academic and managerial implications will be given. 4 - An Experimental Analysis of Price Elasticities for Music Downloads Dominik Papies, University of Hamburg, Welckerstr. 8, Hamburg, 20354, Germany, [email protected], Martin Spann, Michel Clement Although the media industry has received considerable attention by marketing scholars, little has been done to analyze price elasticities in digital media markets suffering from piracy. Theoretical arguments do not make clear predictions as to whether low or high elasticities can be expected and it is unclear whether knowledge about price elasticities from traditional shopping environments can be readily generalized to an online setting. To address this research gap, we conducted two studies to analyze consumer price elasticities in the market for music downloads. In Study 1 we use field-generated sales data from the past to estimate price elasticities. Study 2 is a field experiment at a large European download store, which avoids potential endogeneity problems by experimentally manipulating retail prices for music downloads. Our results from both studies show that demand for music downloads is surprisingly inelastic to price. Based on our analyses we derive optimal prices for music downloads that have been implemented by the store. 3 - Unraveling the Internationalization-profitability Paradox Joseph Johnson, Associate Professor, University of Miami, 5250 University Drive, Coral Gables, FL, 33146, United States of America, [email protected], Debanjan Mitra, Eden Yin As firms continue their push into international markets a key question remains unresolved. Do profits increase as a firm enters more countries? Given the scramble by firms to go global, it seems obvious. Yet, research provides scant and mixed support, if any. Moreover, past studies, while making important contributions to the topic, use cross-sectional data on a limited set of firms to address a broad and longitudinal question. Whether these paradoxical findings will generalize across time 90 MARKETING SCIENCE CONFERENCE – 2011 SC14 2 - The Introduction of a Store Brand in a High-quality Market Segment: Analysis of a Natural Experiment Elena Castellari, PhD Student, University of Connecticut, Agriculture and Resource Economics Department, 1376 Storrs Road, Storrs, CT, 06269, United States of America, [email protected], Rui Huang and context is unclear. The current study attempts to address the above limitations by compiling a unique data set on 6405 firms from 52 countries spanning a 27 year period (1980-2007) and then delineating the different internationalization-profit relationships across different firms. Across all firms in our data, we find a positive but insignificant profit impact of internationalization. However, a latent class analysis reveals several distinct patterns of the profit impact. In particular, some firms have an increasing convex internationalization-profit relationship while others have an increasing concave pattern. We argue that those firms that enter international markets to exploit their current stock of resources obtain quick profits that subsequently plateau and those that enter international markets to explore new opportunities obtain low (and sometimes even negative) initial profits which subsequently increase. We discuss a testable implication of our theory in the context of the observed sequence of international market entry. Store brands (SBs) play an increasingly important role in the retail industry. Since the early 90’s we observe an intense expansion in different segments of the market and increasing breadth in the SB product portfolio. Most existing research has focused on SBs that are viewed as lower-quality substitutes for the incumbent national brands (NBs). However, recently retailers have introduced premium-quality SBs. If consumer choices are context-dependent, then the launch of a premium-quality SB and that of a low-quality SB would have drastically different implications. This study examines the impact of a premium SB introduction on competition. We use quarterly IRI InfoScan market-level data from 2004-2008 to analyze effects from the introduction of a premium-quality private label on market prices and shares of NBs. We exploit a natural experiment that occurred during our data – the introduction of a Store Brand in the high-quality milk segment. Different geographic markets saw different SB introduction schedules in our data, which allows us to use a difference-in-differences design to examine the impacts of the high-quality SB introductions on the prices and shares of different incumbent NBs in the high-quality milk segment, as well as on the standard-quality milk segment. Specifically, we compare the prices and shares of the incumbent NBs in markets with and without SB introduction, before and after the introduction. Our findings could shed light on SB introduction and positioning strategies. 4 - Consumers Un-tethered: A Three-market Study of Consumer Acceptance of Mobile Marketing Fareena Sultan, Professor of Marketing/Robert Morrison Fellow, Northeastern University, College of Business Administration, 202 Hayden Hall, Boston, MA, 02115, United States of America, [email protected], Andrew J. Rohm, Tao (Tony) Gao, Margherita Pagani This study examines factors influencing consumers’ acceptance of un-tethered, or mobile, marketing across three influential markets (U.S., China, and Europe). Considering the highly personal and private nature of the mobile device, we draw upon the technology acceptance theory and incorporate three individual-level characteristics, namely attachment, innovativeness, and risk avoidance, as antecedents to attitudes toward mobile marketing. We also investigate how permission-based acceptance influences the link between consumers’ attitude and participation in mobile marketing activities. Our findings show both cross-market similarities and differences. Perceived usefulness, consumer innovativeness, and personal attachment are found to directly influence attitudes toward mobile marketing in all three markets. In China and Europe, risk avoidance also negatively influences attitude toward mobile marketing. Depending on the market, innovativeness, risk avoidance, and attachment also serve, as moderators, to weaken the effect of perceived usefulness on mobile marketing attitude. Furthermore, permission-based acceptance strengthens the relationship between attitude and mobile marketing activities. The results also confirm the uniformly prominent role of ease of use in affecting usefulness perceptions. We draw implications from these findings related to both theory and practice. 3 - Investigation of Determinants of Private Label Success in an Integrated Framework Hyeong-Tak Lee, PhD Student, University of Iowa, S219 John Pappajohn Business Building, The University of Iowa, Iowa City, IA, 52242-1994, United States of America, [email protected], Thomas Gruca Private label products often benefit retailers and can adversely affect consumer packaged goods manufacturers. Due to their strategic importance, private labels have been the subject of a great deal of academic research, much of it fragmented. In this study, we integrate previous research on category determinants of private label performance, the demographic or socioeconomic characteristics of private label buyers, and the effects of competition in the category in a single empirical analysis. Using data from 625 product categories, we attempt to create empirical generalization about what factors drive private label market share. Partial least squares modeling is applied to our data to test new and prior hypotheses on private label market share determinants. Our findings suggest that private label success is positively associated with the level of concentration, the margin potential, and the price gap between private label and the average price of national brands. Financial risk of the category influences the private label market share adversely. In addition to category determinants, private label share in categories where consumers are older, have more education, have larger families and have less knowledge about product quality. ■ SC13 Champions Center III Private Labels III: Effect on Market Shares Contributed Session Chair: Hyeong-Tak Lee, PhD Student, University of Iowa, S219 John Pappajohn Business Building, The University of Iowa, Iowa City, IA, 52242-1994, United States of America, [email protected] 1 - The Long Term Impact of a Recession on Brand Shares Satheeshkumar Seenivasan, PhD Candidate, State University of New York at Buffalo, School of Management, 232 Jacobs Management Center, Amherst, NY, 14260, United States of America, [email protected], Debabrata Talukdar, K. Sudhir ■ SC14 Champions Center VI Marketing Strategy IV: Firm Performance Contributed Session Chair: Sohyoun Shin, Visiting Assistant Professor, Eastern Washington University, 668 N. Riverpoint Blvd., Spokane, WA, United States of America, [email protected] 1 - Drivers of International Growth: Analysis of U.S. Franchisors’ International Growth Strategies Bart Devoldere, Vlerick Leuven Ghent Management School, Reep 1, Ghent, 9000, Belgium, [email protected], Venkatesh Shankar A unique aspect of store brands is the counter-cyclical movement of their market shares with business cycles. Besides gaining market shares during downturns, store brands also manage to retain some of the gained shares post downturns. In this paper, we undertake an in-depth analysis of consumer choice behavior during the recent global recession to understand the drivers of increased store brand share as well as the underlying causes for the persistence in store brand shares post recession. Employing a brand choice model of consumer learning which accounts for inertia and allows for differential marketing mix sensitivities, we study the purchase behavior of 890 households in yogurt category over a period of three and half years. Our results suggest that all three factors – persistence in marketing mix sensitivities, inertia as well as preference updation due to learning contribute to the rise of store brand shares. Consumers are more sensitive to prices and promotions during recession and persist in their sensitivities even after the end of recession. Increased sensitivities account for 62% of gain in market share of store brands while learning accounts for 21% and the remaining 17% is driven by inertia in consumer choices. Implications of this recession time consumer behavior and strategies to cope with this behavior are discussed. Firms are increasingly going global to realize high rates of growth. By some estimates, among the Standard & Poor 500 companies, those that derive over half of their sales revenues overseas are expected to grow about twice the rate of companies focusing on the U.S. Global growth is particularly important in the franchising context. Why do some franchisors grow larger than others internationally? Is it due to effective pricing policy decisions, such as up-front fixed fees and royalty rates that franchisors charge the franchisees for use of the franchise brand? Or is it due to the right decisions related to strategic control, including the number and proportion of outlets owned and operated by the franchisor? Or is it due to strategic selection of the most attractive international markets? We investigate the drivers of international growth for franchisors. Drawing on agency and power relationship theories, we develop hypotheses on the influence of these strategic decisions on the size of international operations. We develop a model of international franchise system size that includes the effects of these strategic decisions as well as those of environmental push and pull factors. We test our hypotheses on panel data relating to 200 U.S. business format franchise systems during 1999-2010. We estimate our model using a Hierarchical Bayesian approach. Our model controls for unobserved firm and industry effects, accounts for endogeneity of decision variables, and corrects for selection effects due to system failure. Our results offer important implications for researchers and practitioners on international growth strategies. 91 SC14 MARKETING SCIENCE CONFERENCE – 2011 2 - Analyzing the Dynamics of Satisfaction, Recommendation and Customer Acquisition Henning Kreis, Freie Universität Berlin, Alte Schönhauser Str. 29, Berlin, 10119, Germany, [email protected], Till Dannewald 3 - Exploring the Components of Marketing Process Capability & Confirming its Relationship w/Performance Sohyoun Shin, Visiting Assistant Professor, Eastern Washington University, 668 N. Riverpoint Blvd., Spokane, WA, United States of America, [email protected] It is often argued, that positive word-of-mouth communication, as a result of customer satisfaction, can help companies to keep up their sales; gain new customers and expand market share (e.g. v.Wangenheim and Bayon 2007). Empirical evidence, however, is still scant regarding the relationship and effectiveness of satisfaction, positive word-of-mouth in increasing customer acquisition over time (e.g., Godes and Mayzlin 2004). In this research, we estimate a model that captures the dynamic relationships among customer acquisition, recommendations, and customer satisfaction on the basis of longitudinal data. Unlike most other recent studies (e.g. Trusov et al. 2009) which are concentrating on analyzing such effects in online settings, we focus on durables (cars), using monthly data of an evaluation period of more than seven years. Therefore our first contribution lays in the linkage of satisfaction with and recommendation of durables to real purchase behavior. Because of the endogeneity among the variables in focus, we apply vector autoregressive modeling techniques with which we also account for possible time trends in the data. This approach allows us to contribute to the literature by incorporating both direct effects as well as indirect effects of recommendation and satisfaction (e.g., satisfaction increases positive word-of-mouth activity, which in turn drives customer acquisition). We use the estimated coefficients to simulate the net impact of customer satisfaction and recommendation on customer acquisition, by setting up the corresponding impulse response functions. Herewith we are able to quantify short-term as well as long-term effects of recommendation and satisfaction and to estimate carryover effects for different types of recommendation in our dataset. In strategic marketing literatures, marketing capability is considered to be an essential driver of business performance. However, this critical construct has seldom been measured through an integrative set of scales. As Day (1994) have argued that capabilities and organizational processes are closely entwined because it is the capability that enables the activities in business processes to be carried out, marketing capability should be examined as a set of processes. The author suggests ‘Marketing Process Capability (MPC)’ as a linear succession of five steps in two subsets: ‘OutsideIn Process Capability (OIPC)’ and ‘Inside-Out Process Capability (IOPC).’ OIPC contains capability of market-sensing and market information management, and IOPC explains marketing program planning, execution, and auditing capability. It is an applied version of comprehensive understanding of ‘architectural capability’ proposed by Vorhies and Morgan (2005). Both OIPC and IOPC have been proven to have a strong impact on firm performance including CS, market efficiency and profitability. The research was carefully designed and completed as following. First, an exploratory qualitative study was conducted to understand how the practitioners define and think ‘capability in marketing area to create customer value.’ This question was generated from the previous study and 42 practitioners from various functions participated in this qualitative survey. 2nd qualitative study was carried with 7 marketing professionals to confirm the results from the 1st study. 3rd study was conducted with 6 academic experts to consolidate the components of MPC through further discussions. Lastly, an empirical study was followed to verify the relationship between MPC and FP with 141 companies. 92 Session Chair Index A H M Agarwal, Nipun SB14 Alexandrov, Alexei SB13 Algesheimer, Rene FA04 Haenlein, Michael TB03 Henderson, Ty TC09 Herzenstein, Michal FA13 Hess, James FB11 Hibshoosh, Aharon FC09 Hoffman, Donna L. SB03 Horsky, Dan SA01 Hu, Yansong SA10 Mahapatra, Sabita TC11 Mak, Vincent TC06 Mallucci, Paola FD01 Mantrala, Murali SA13 Mathur, Sameer SB09 McWilliams, Bruce FB09 Mizik, Natalie FD08 Montgomery, Alan FD03 Mora, Jose-Domingo SA11 B Barrot, Christian SB02 Basalingappa, Anita SA15 Basar, Berna TD10 Bendle, Neil FD14 Berger, Jonah FC07 Biehn, Neil TA14 Billore, Aditya FA03 Bloch, Katrin SB06 Bonfrer, Andre SA04 Borah, Abishek SA02 Borle, Sharad SC03 Braun, Michael TC07 Bushey, Erik SB07 I Jalali, Nima FC15 Jamal, Zainab TC15 Jedidi, Kamel FC14 Joshi, Yogesh FC12 C Cai, Gangshu SB01 Campbell, Merle TC08 Casas-Romeo, AgustÌ FB12 Chakravarty, Anindita FD08 Chan, Kimmy Wa TD07 Cheng, Ming TD12 Choi, S. Chan TD06 Choodamani, Roopa FB04 Cohen, Michael TA05 Cui, Anna S. TD08 Kang, Wooseong FA05 Kang, Yeong Seon SB04 Kehal, Mounir SA12 Khwaja, Ahmed FA08, FB08 Kim, Ho TB11 Kim, Jungki FD09 Kim, Tae-kyun SA09 Konus, Umut TD09 Kornish, Laura TC02 Kunter, Marcus FA14 D L Dehmamy, Keyvan TA04 Diels, Jana TA10 Dimitriu, Radu FB15 Dong, Songting FD10 Lajos, Joseph SA06 Lambrecht, Anja TC14 Lee, Hyeong-Tak SC13 Lee, Janghyuk FA15, FD02 Lee, Julie FA10 Lehmann, Don TA01, TB01 Lenk, Peter J. TA07 Libai, Barak TB03 Lilien, Gary FA07 Lin, Chen TD03 Lin, Yuanfang SA05 Lourenco, Carlos FC13 Lu, Steven FC11 Luo, Xueming TA13, TB13, TC13, TD13 Elberse, Anita TA03 Ellickson, Paul FC08 F Feit, Elea McDonnell FA12 G Nair, Harikesh FD13 Narayanan, Sridhar TC04 Nolte, Ingmar FD12 Norton, David FC10 J K E N Israel, Duraipandian SB11 Tan, Tom Fangyun SA07 Toker-Yildiz, Kamer FB01 Trauzettel, Volker FD06 Tripathi, Manish FB02 Tucker, Catherine SB08 U Umesh, U.N. TC12 Urban, Glen FB07 Orhun, A. Yesim TA06, TB06 V Papies, Dominik SC07 Pazgal, Amit FD07 Pisharodi, R. Mohan FB14 Poncin, Ingrid FB03 Pradhan, Debasis FA11 R Rajamani, Nithya SC04 Rangaswamy, Arvind FA07 Ratchford, Mark TC05 Romero, Jaime FC06 Rooderkerk, Robert TB10 Rubel, Olivier SA08 S Salisbury, Linda Court TD01 Sarkar, Soumya SA14 Schlereth, Christian FC01 Schwartz, Eric TA11, TB02 Selove, Matthew FC05 Shang, Yi-Yun TC10 Shin, Sohyoun SC14 Shin, Woochoel SC02 Shulman, Jeffrey D. FB05 Sinitsyn, Maxim TD14 Sisodiya, Sanjay TA08 Soch, Harmeen TD15 Sood, Ashish FA02 Sorell, Michael FB13 Sotgiu, Francesca TB09 Spann, Martin TD04 93 T O P Gangwar, Manish FA09 Garber, Larry TD11 Gauri, Dinesh TA09 Ghose, Anindya TB07 Gu, Zheyin (Jane) FA01 Gupta, Sudheer FA06 Srinivasan, Raji FC02 Srinivasan, Shuba TA02 Stephen, Andrew FC07 Stern, Philip SB10 Stuettgen, Peter TC01 Sudhir, K. FA08, FB08 Sultan, Fareena SC09 Syam, Niladri FD05 Valentini, Sara FB06 van der Lans, Ralf FB10 van Heerde, Harald TD02 Velu, Chander TB08 Vernik, Dinah TD05 Villas-Boas, J. Miguel FC03 Voleti, Sudhir TB04 W Wagner, Ralf SB06 Walker, Doug SC06 Wang, Jiana-Fu FD15 Wang, Wenbo FC04 Wieringa, Jaap SB10 Worm, Stefan TB12 Wu, Ruhai SB05 Y Yang, Jun TC03 Yoganarasimhan, Hema FD11 Z Zerres, Alfred TA15 Zhang, Jie SA03 Zhang, Jonathan TB14 Zhang, Kaifu SC01 Zhao, Yi FD04 Zheng, Li TB05 Zheng, Xiaoying TA12 Zubcsek, Peter Pal SB15 Author Index A Abou Nabout, Nadia FD03 Achrol, Ravi FD06 Acquisti, Alessandro SB08 Adams, Jeffery TA15 Agarwal, Nipun SB14 Agarwal, Puja SC04 Ahearne, Michael FC11 Ahn, Dae-Yong FB08 Ahrholdt, Dennis FD11 Ailawadi, Kusum TC09, FA09 Akcali, Elif TD09 Akcura, Tolga SB02 Akdeniz, M. Billur TD13 Albers, Soenke FB11, SB10 Albuquerque, Paulo TB02 Alexandrov, Alexei SB13 Algesheimer, Rene FA04 Ali, Özden Gür SB15 Allenby, Greg TC01 Allender, William FC09 Allman, Helena TB12 Alptekinoglu, Aydin TD09 Altinkemer, Kemal SB02 Amaldoss, Wilfred SC02 Amrouche, Naoual SA13 Ancarani, Fabio SB14 Anderson, Eric TA06, TC14, TD14, FC01 Anderson, Erin SA06 Ansari, Asim TB14, TC03 Araña, Jorge TA14 Arancibia, Mauro SB04 Arancibia, Natalia SB04 Ariturk, Umut SB15 Armelini, Guillermo FA15 Arnett, Dennis FC12 Arora, Neeraj TA04, TC09, SA11 Arroniz, Inigo TC10 Aspara, Jaakko FB13 Ater, Itai FA14 B Backhaus, Klaus TA15 Bae, Jonghoon FD02 Bae, Sang Hee TB09 Bae, Young Han FD15 Baek, Seok-Chul FD02 Bagchi, Rajesh TB10 Baier, Daniel FA10, SA01 Bajari, Patrick FC04 Bala, Ram SA10 Balachander, Subramanian TB09, TD12, FC04 Balakrishnan, P.V. (Sundar) SB06 Balasubramanian, Sridhar SA05 Banerjee, Ranjan FC04 Banerjee, Sourindra SB09 Bao, Lin TA04 Bao, Tony SA04 Barrot, Christian SB02 Barth, Joe TC10 Bart, Yakov FC05 Basalingappa, Anita SA15 Basar, Berna TD10 Basuroy, Suman SA13, SB05 Batra, Rishtee TA10 Bau, Raimund TC11, FB01 Bayus, Barry SA07 Becerril-Arreola, Rafael FB10 Beckers, Sander F. M. TC13 Beldona, Srinath SA13 Beltramo, Mark FA12 Bendle, Neil FD14 Berger, Jonah FC07 Bermes, Manuel FA13 Bernstein, Fernando TD05 Bertini, Marco TC14 Bezawada, Ram FB06 Bharadwaj, Sundar FC12, FD15 Bhardawaj, S (Sivkumaran) TD10 Bhardwaj, Pradeep SA05, SA10 Biehn, Neil TA14 Bielecki, Andre FB11 Bijmolt, Tammo TA03, TB03, TC02, FB11, FB12, SB10 Billore, Aditya FA03 Bleier, Alexander SA03 Bloch, Katrin SB06 Boatwright, Peter TC01, FA12 Bodenberg, Stefan SC09 Boegershausen, Johannes FB15 Bohling, Tim TC08 Bonfrer, Andre SA04 Boo, Chanil TD04 Borah, Abishek SA02 Borah, Sourav FD03 Bordley, Robert SC06 Borovsky, Juraj SA10 Boulding, William FC08, FD05 Bourbonus, Nicolas FB13 Bowman, Douglas TD03, FA02 Boya, Unal TD11 Bradlow, Eric FB07 Branco, Fernando FC03 Brandes, Leif TB11, FA04, FD12 Brandimarte, Laura SB08 Braun, Michael TC07 Bremer, Lucas TC03 Bris, Arturo FB13 Broniarczyk, Susan FC13 Bruce, Norris TD12 Bruno, Hernan FC14 Bücker, Michael SA14 Buckinx, Wouter TA11 Bucklin, Randolph TA02, TC02, TC15 Buehler, Stefan SB01 Bulla, Jan FB06 Bunker, Matthew SB11 Burmann, Christoph TA05 Burmester, Alexa SA07 Burtch, Gordon SA11 Bushey, Erik SB07 Carmi, Eyal SB02 Casas-Romeo, Agustí TA12, FB12 Castellari, Elena SC13 Caudill, Helene FA11 Cebollada, Javier FB12 Chae, Inyoung SA15 Chakravarty, Anindita FD08 Chan, Kimmy Wa TD07 Chan, Tat Y. TB14, FD04, SC01 Chan, Tat TC03 Chandrasekaran, Deepa TA08 Chandrasekhar, Suj FA05 Chandukala, Sandeep TA03 Chandy, Rajesh TC06, SB09 Chang, Sue Ryung FA01 Chang, Young Bong TC08 Che, Hai FA08 Cheng, Ming TD12 Chen, Hua FC11 Chen, Li-Wei TD15 Chen, Liwen SB13 Chen, Sixing TB07 Chen, Tao FC09 Chen, Wun-Hwa SB09 Chen, Xi TC10, SA04 Chen, Yubo SA05 Chen, Yuxin FC04, FC08, FD04, SA10 Chiang, Ai-Hsuan SB09 Chiang, Jeongwen FC08 Chiang, Wei-yu Kevin TC06 Chidmi, Benaissa FD01, SC06 Ching, Andrew FB08 Chintagunta, Pradeep TA06, TB04, FB12, FC15, SA09 Chiu, Chun-Hung SB01 Choi, S. Chan TD06, TD12 Choi, Tsan-Ming SB01 Choodamani, Roopa FB04 Chou, Shan-Yu FA06 Chowdhury, Imran SB15 Chrzan, Keith FC15, SA01 Chu, Junhong FB12 Chung, Jaihak FA01 Chung, Kevin FD04 Chu, Sean FC04 Clement, Michel SA07, SB07, SC07 Cohen, Michael TA05 Cole, Cathy TA12 Conrady, Stefan TA03 Corbin, Steven B SB11 Cosguner, Koray TB14 Cotterill, Ronald W. TB07 Coughlan, Anne T. TA09, TA13, TD14 Coussement, Kristof TA11 Crandall, David SA04 Cui, Anna S. TD08 Cui, Geng TD07 Currim, Imran TA03 Cutler, Jennifer SB05 C Cagan, Jonathan FA12 Cai, Gangshu SB01, SB13 Cai, Jeffrey FC07 Caliendo, Marco SB07 Campbell, Arthur FB05 Campbell, Merle TC08 Cao, Zixia(Summer) FB13 94 D Dahan, Ely TB07 Dai, Tinglong FB05 Dannewald, Till SC14 Dann, Stephen SA02 Das, Gopal SC04 Dasgupta, Srabana FD01 Dash, Satyabhushan SA10 Dass, Mayukh TD12, FA02, FC12 Datta, Souvik FD01 Datta, Sumon TB06, TB09 De Bruyn, Arnaud FA07, SB03 Dehmamy, Keyvan TA04 Dekimpe, Marnik G. SA13 Deleersnyder, Barbara FB04 Dellaert, Benedict TD01, FC13, SA03 Dellarocas, Chrysanthos TA02 Delre, Sebastiano SB01, SC07 Deng, Yiting FD05 Derby, Joseph TD12 Derdenger, Timothy FB08, FD04 Desai, Preyas TD05, SB09, SC02 Desiraju, Ramarao SA06 Devoldere, Bart SC14 Dewani, Prem TC12, TD10 Dieleman, Evelien FB10 Diels, Jana TA10 Dimitriu, Radu FB15 Ding, Min FD10, TB11 Dinner, Isaac TD02, FD08 Dong, John SA04 Dong, Songting FD10 Dong, Xiaojing FC15 Donkers, Bas TA11, TD01, FC13 Dotson, Jeffrey TA03, TC05, FA12 Dover, Yaniv FA02, FC07 Drechsler, Wenzel FA05 Driessen, Paul FB03 Du, Rex SA02 Dukes, Anthony FA14, FD05, SB05 Dutta, Shantanu FC14 Duvvuri, Sri Devi FB01 E Ebbes, Peter TB04 Eckert, Christine TA14, FC01 Eckhardt, Giana TB04 Eilert, Meike SB14 Ein-Gar, Danit TC15 Eisenbeiss, Maik FB06, SA03 Elberse, Anita TA03 Eliashberg, Jehoshua (Josh) SA07 Ellickson, Paul FA08, FC08 Elmadag Bas, A. Banu TD10 Elrod, Terry TB04 Elsner, Ralf TA06 Erdem, Tulin TC04, FA01 Erickson, Gary TA12 Evans Jr., Robert TD13 Evgeniou, Theodoros SA15 F Fan, Tingting FC10 Fang, Eric TA08 Fei, Qiang TA12 Feinberg, Fred M. TD01, FC01 Feit, Elea McDonnell FA12 Feldman, Ronen TB07 Feng, Jie TD03 Fine, Monica TD13 Fischer, Marc TB01 Fok, Dennis TB05, FB14 Foubert, Bram TB09 Fournier, Susan SA02 Foutz, Natasha SB07 Franck, Egon TB11, FA04 Franses, Philip Hans TB05, FB14 Freling, Traci TD13 Fresko, Moshe TB07 Freund, Alexander TA15 Funtanilla, Margil SC06 G Gabel, Sebastian FB01 Galak, Jeff FC07 Gal-Or, Esther FD02, SC01 Gangwar, Manish FA09 Gao, Tao (Tony) SC09 Garber, Larry TD11 Garnier, Marion FB03 Garrett, Tony SC09 Gatignon, Hubert SA06 Gauri, Dinesh TA09 Gàzquez-Abad, Juan Carlos TA12, FB12 Ge, Xin SA05 Gedenk, Karen TC09 Geng, Xianjun FB05 Georgi, Dominik FB03, FB13 Geylani, Tansev FD02, SC01 Geyskens, Inge SA13 Ghose, Anindya TA02, TB07, SA09, SA11 Ghosh, Pulak TB04 Ghoshal, Tanuka FA12 Gielens, Katrijn TB09, FB09 Gijsbrechts, Els TB09, FA09 Gilbert, Steve SB13 Gilbride, Timothy J. TA07 Gleason, Kimberly TD13 Goettler, Ronald FC08 Goic, Marcel TD14 Goins, Sheila TA12, FC13 Goldenberg, Jacob TB07, FB02 Golder, Peter TC04, FC10 Goldfarb, Avi TB07, TC02, FD11 Goldstein, Dan FC13 Gomez, Miguel FA04 Gopalakrishna, Srinath FA05 Gopinath, Shyam TB04 Gordon, Brett FC08 Grasas, Alex TD09 Greenacre, Luke SA12 Grégoire, Yany TA08 Grewal, Rajdeep TB04, TB13, TD06, FD08 Groening, Christopher TB13 Gruca, Thomas FC13, SC13 Gu, Zheyin (Jane) FA01 Guan, Wei FA06 Guler, Ali Umut TA14 Guo, Ruijiao SB03 Guo, Shuojia SA12 I Guo, Xiaoning TC10 Gupta, Pola TD11 Gupta, Sachin TC04, FB01 Gupta, Sudheer FA06 Gutierrez, Gutierrez J. SA08 Ingene, Charles TD14 Inman, Jeff FC13, FD09 Inoue, Akihiro FA10 Irmak, Caglar FC10 Ishihara, Masakazu FB08 Ishimaru, Sayaka TA10 Israel, Duraipandian FA11, SB11 Israeli, Ayelet TD14 Iyengar, Raghu FC07, FC14 Iyer, Ganesh FD13 H Ha, Kyoung Nam TA12 Haenlein, Michael TB03 Hahn, Minhi FD12 Halaszovich, Tilo TA05 Halbheer, Daniel FC03, SB01 Hamilton, Stephen FC09 Han, Sangpil TB07 Hana, Karel SA10 Hansen, Karsten TC14 Hanssens, Dominique TA13, TB01, TB11, FA07 Haon, Christophe FB15 Hariharan, Vijay Ganesh TD04 Haruvy, Ernan FC06, SC02 Hattula, Johannes SB11 Hattula, Stefan SB11 He, Xiuli SA08 Heil, Oliver TA09, FC10 Heitmann, Mark TC03 Henderson, Ty TC09 Henke, Jr., John FB14 Henningson, Sing SB10 Hermosilla, Manuel FC01 Hernandez Mireles, Carlos FB14 Herzenstein, Michal FA13 Hess, James FB11, FC11 Hibshoosh, Aharon FC09 Higuchi, Tomoyuki FB04 Hildebrandt, Lutz TA10 Hiziroglu, Abdulkadir TC15 Ho, Candy K. Y. TD11 Ho, Shu-Chun TC08 Ho, Ying TD11 Hoban, Paul TC02 Höck, Claudia FD11 Hoffman, Donna L. SB03 Hoffmann, Arvid O. I. FB13 Hofstetter, Reto FD13 Hoops, Christian SA14 Hormiga, Esther SC06 Horsky, Dan TD01, FA13, SA01 Hosanagar, Kartik SA07 Hsu, Liwu SA02 Hsu, Wei-Che TD09 Hu, Chia Ming FA12 Hu, Yansong SA10 Huang, Dongling TA05 Huang, Guofang FA08 Huang, Jesheng FA12, FC06 Huang, Peng FA03 Huang, Rui TA05, SC13 Huang, Shih-Wei SB09 Huang, Xiao FD06 Huang, Yan TA02 Huang, Yanliu FD09 Huertas-Garcìa, Rubèn TA12, FB12, SC06 Hüffmeier, Joachim TA15 Hui, Sam FD09 Huizingh, Eelko TC02, FB12 Hulland, John FD08 Hung, Hao-An FD06 Hwang, Heungsun FD14 Hwang, Minha TD09 Hyatt, Eva TD11 J Jacob, Arun TB08 Jacob, Frank SB06 Jacobson, Robert TA12 Jakubuv, Lenka SA10 Jalali, Nima FC15 Jamal, Zainab TC07, TC15 Jank, Wolfgang FA02 Jayachandran, Satish SB14 Jayarajan, Dinakar FD04 Jedidi, Kamel TA09, FC14 Jen, Lichung FC06 Jerath, Kinshuk FB05, FC03, FD03, SB05, SC02 Jha, Subhash TD10 Jiang, Baojun SB05 Jindal, Niket TB13 John, George FD01 John, Leslie SB08 Johnson, Jean TA08 Johnson, Joseph SC09 Jones, Eli SC06 Joo, Mingyu TD02 Joshi, Yogesh FC12 Jun, Duk Bin FA04, FD09 Jung, Sang-Uk FD13 K K B, Saji SC04 Kalaignanam, Kartik SB14 Kalogeras, Nikos FC13 Kalra, Ajay TB10, FB11 Kalwani, Manohar SC01 Kalyanam, Kirthi TC04 Kamakura, Wagner SA02 Kanetkar, Vinay TC13 Kang, Sukwon FD02 Kang, Wooseong FA05 Kang, Yeong Seon SB04 Kansal, Anurag TC12, TD10, FA03 Kappe, Eelco SB10 Karaca, Huseyin TA09 Kashmiri, Saim FD14 Katona, Zsolt TA02, FD13, SB15, SC01 Katsumata, Sotaro TA10 Käuferle, Monika FB06 Kauffman, Ralph TA15 Kayande, Ujwal FA07, FD10 Kehal, Mounir SA12 Khan, Romana FB10 Khong, Kok Wei SA15 Khwaja, Ahmed FA08 Kim, Alex TD12 Kim, Dong Soo FA04 Kim, Ho TB11 Kim, Hyang Mi TD15, SA06 Kim, Hye-jin FD10 95 Kim, Jae Wook TD15, SA06 Kim, Jaehwan TC01, FA04, SA11 Kim, Jin-Woo TD13 Kim, Jiyoon FA15 Kim, Jungki FD09 Kim, Junhee SA07 Kim, Min Chung TC13 Kim, Namhyong SB15 Kim, Namwoon FD14 Kim, Sang Yong FA15, SA15 Kim, Sang-Hoon TB09 Kim, Tae-kyun SA09 Kim, Taewan TD05 Kim, Wonjoon FC02, SB07 Kim, Young-Gul SC07 Kim, Youngju SA11 Kim, Youngsoo FA03 Kissan, Joseph TA13 Kivetz, Ran FB15 Klapper, Daniel TA03, FB14, FC06 Knox, George TC07 Koenigsberg, Oded FC03 Konus, Umut TD09 Kopalle, Praveen TB04, TD06, FB01 Kornish, Laura TC02 Korzenny, Felipe TB12 Koshy, Abraham TC12 Kothari, S. P. FD08 Kreis, Henning SC14 Krishnamurthi, Lakshman TA09 Krishnan, Ramayya FA03 Kuikman, Joost FC13 Kuksov, Dmitri FD05 Kulkarni, Gauri FB12 Kumar, Anuj TA02 Kumar, Ashish FB06 Kumar, Piyush FC12 Kumar, Pradeep FB04 Kumar, Ravi SA10 Kumar, V TA06, TC08 Kumar, Vineet FB08 Kunter, Marcus FA14 Kurt, Didem FD08 Kwak, Kyuseop TC01, SA12 L Lajos, Joseph SA06 Lam, Simon TD07 Lambert-Pandraud, Raphaëlle FC10 Lambrecht, Anja TC14 Landsman, Vardit FA14 Langer, Tobias TC09 Lanoue, Christopher FA04 Laurent, Gilles FC10 Lee, Backhun FD12 Lee, Clarence TA03 Lee, Dong Wook FD15 Lee, Eun Young FD15 Lee, Eunkyu TD05 Lee, Hyeong-Tak SC13 Lee, Jae Kyu SC07 Lee, Jaewook SB15 Lee, Janghyuk FA15, FD02, SA07 Lee, Jinsuh SC01 Lee, Jonathan FD14 Lee, Jong-Ho SC09 Lee, Jongkuk TC08 Lee, Julie FA10 Lee, Kee Yeun FC01 Lee, Sanghak TC01 Lee, So Young TD15 Leeflang, Peter TA03, SB10 Lehmann, Don TA01, TD06, FB07, FC03, FD08 Lemon, Katherine N. FB06 Lengler, Jorge SC06 Lenk, Peter J. TA07 Leon, Carmelo J. TA14 Li, Jia FA15 Li, Lianhua TD01 Li, Tieshan TD05 Li, Zheng FD09 Libai, Barak TB03, TC15, SB02 Liberali, Gui FB07 Liechty, John TB04 Lilien, Gary TD04, FB07 Lim, Kyung Jin SA05 Lim, Noah FC11 Lim, Weishi FB14 Liming, Liu FC06 Lin, Chen TD03 Lin, Qihang FD03 Lin, Yuanfang SA05 Lin, Yu-Li TD07, FB15 Liu, Bin SB01 Liu, Hsiu-Wen TD07, FB15 Liu, Lin FA14, SB05 Liu, Qiang FB01 Liu, Qing TA03, TA04 Liu, Yan TB09 Liu, Yong SA05 Liu, Yunchuan SB13 Lo, Desmond (Ho-Fu) TA15 Loewenberg, Margot FA04 Loewenstein, George SB08 Lopes, Hedibert TB04 Loscheider, Jeremy FC15 Lourenco, Carlos FC13 Louviere, Jordan TC01, FA10 Lovett, Mitch FA08, FC08 Lu, Luo TC07 Lu, Steven FC11 Lu, Yina TC07 Lu, Yingda FB08 Luan, Jackie FA09 Lucas, Hank SA03 Luo, Lan TA01 Luo, Xueming TA13 Lurie, Nicholas FA03 M MacDonald, Emma TD09 Mahajan, Vijay FD14 Mahapatra, Sabita TC11 Mak, Vincent TC06 Mallucci, Paola FD01 Malshe, Ashwin TA13 Manes, Eran SC04 Mantrala, Murali SA13 Mark, Tanya FB06 Marom, Ori FC03 Marshall, Jorge SB04 Martìnez-Lòpez, Francisco J. TA12 Maruotti, Antonello FB06 Mason, Charlotte FA07 Mathur, Sameer SB09 Mayzlin, Dina FA02, FB05, FC07 Mazursky, David TD08 McAlister, Leigh TB13, TC13 McKechnie, Sally FA03 McWilliams, Bruce FB09 Mehta, Nitin TB13, FA08 Meierer, Markus FA04 Mela, Carl TA06, FB08, FC08 Mesly, Olivier SB06 Messinger, Paul TC01, TD01, SA05 Meza, Sergio TA09, SB04 Miklòs-Thai, Jeanine FB05 Min, Jihong FC04 Min, Sungwook FD14 Mink, Moritz FB03 Mintz, Ofer TA03, FD03 Miravitlles-Matamor, Paloma SC06 Mishra, Prashant SA14, SC04 Misra, Kanishka TC14 Misra, Sanjog TB02, FC08, FC14 Mitra, Debanjan SC09 Mittal, Vikas TB13 Mizik, Natalie FD08 Moccia, Sergio FC10 Moe, Wendy W. TA07, FA02, FB02 Monga, Ashwani TB10 Monroe, Robert TC01 Montaguti, Elisa FB06 Montgomery, Alan TD14, FD03 Montoya, Mitzi FA05 Montoya, Ricardo FB15 Moon, Sangkil SA07 Moorman, Christine FA13 Mora, Jose-Domingo SA11 Morrin, Mimi FC13 Motohashi, Eiji TA10, FB04 Muir, David SA04 Mukherjee, Anirban SA04 Mukunda Das, V. SC04 Multani, Navneet TD15 Murphy, Alvin FD04 Musalem, Andrès TA06, TC07, FC12 O R O’Connell, Vincent FD08 O’Connor, Gina TD08 O’Sullivan, Don FD08 Oakenfull, Gillian FA11 Oestreicher, Gal FB02, SB02 Oetzel, Sebastian FB14 Olivares, Marcelo TC07 Oliveira, Eduardo TC12 Ong, David SC02 Ordanini, Andrea TA08 Orhun, A. Yesim TA06 Orsborn, Seth FA12 Osinga, Ernst SB10 Otter, Thomas TA04 Ozcelik, Hamdi SB15 Ozsomer, Aysegul TC11 Ozturan, Peren TC11 Ozturk, O. Cem SA09 Raj, Roopika TC12 Rajendran, K N SB11 Ramachandran, Vandana SA03 Rand, William TA02, TB03 Rao, Anita TB14 Rao, Raghunath TB14 Rao, Vithala TB08 Ratchford, Mark TC05 Ray, Daniel FB15 Reber, Katrin SB10 Reeder, James FC14 Rego, Lopo TA12, FA13, FD15 Rehme, Jakob FA06 Reibstein, David TA13 Reichhart, Philipp SB02 Reichman, Shachar FB02 Reinartz, Werner TB11, FB06 Reinecke, Sven SB11 Reiner, Jochen SC07 Reyes, Antonieta TB12 Richards, Keith SC06 Richards, Timothy FA04, FC09, SA13 Ringle, Christian FD11 Risselada, Hans TB03 Ro, Joon FB10 Robnik-Sikonja, Marko FD10 Rohani, Khalil TC10 Rohani, Laila TC10 Rohm, Andrew J. SC09 Rolef, Charlotte TB09 Román, Hernàn FA15 Romero, Jaime FC06 Rongen, Eric FB03 Rooderkerk, Robert TB10 Rose, Randy FC10 Rosenzweig, Stav TD08 Rossi, Frederico TA06 Roy, Sudipt TC09 Roychowdhury, Sugata FD08 Rubel, Olivier SA08 Rubera, Gaia TA08 Rudolph, Thomas TC13 Russell, Gary J. FD13, FD15, SA11 Rutz, Oliver TA02 Ryu, Sunghan SC07 P Padmanabhan, Paddy SA15 Pagani, Margherita SC09 Palekar, Udatta SB07 Pancras, Joseph TA06, TA09, TC09 Panico, Claudio SB01, SC07 Papatla, Purushottam TC09, TD03, FC15, SB03 Papies, Dominik SC07 Parameswaran, Ravi FB14 Park, Chang Hee TB02 Park, Jungkun TC03 Park, Minjung FC04 Park, Myoung Hwan FD09 Park, Sungho TC04 Park, Young-Hoon TB02 Parry, Mark TD14 Patton, Charles SB06 Pauwels, Koen TA02 Pavlidis, Polykarpos TD01 Pazgal, Amit FC05, FD05 Pearce, Ciara SB11 Peers, Yuri TB05 Pei, Zhijian (Zj) SB01 Peng, Siqing TA12 Pennings, Joost M. E. FB13, FC13 Peres, Renana TB03, FC07 Perez, Anthony FA11 Pescher, Christian TD04, SB02 Pisharodi, R. Mohan FB14 Pofahl, Geoffrey SA13 Poncin, Ingrid FB03 Poon, Kwok Ho SB11 Popkowski Leszczyc, Peter T. L. SC02 Pozza, Illaria Dalla SB03 Prabhu, Jaideep TC06, SB09 Pradhan, Debasis FA11, SB11 Prime, Nathalie SB06 Prins, Remco TA11 Purohit, Debu TD05 Purwar, Prem SA10 Putsis, William TB10 N Nagaoka, Toshihiko TC05 Naik, Prasad A. SA08 Nair, Harikesh FC08, FD13 Nam, Hyoryung FB02 Nam, Sungjoon TB05, TB09 Narasimhan, Om FC04, FD01 Narayanan, Sridhar TC04 Nath, Prithwiraj FA03 Natter, Martin FA05, FC06, SC07 Nayakankuppam, Dhananjay SA11 Nekipelov, Denis FC04 Nelson, Paul SA01 Neslin, Scott TC09, TD02, FA09, FB06 Netzer, Oded TB07, TB14, FB15 Ni, Jian FA08 Nijs, Vincent TA09 Nishimoto, Akihiro TA10 Nitzan, Irit TC15 Noguti, Valeria SA12 Noh, Hyung FD02 Nolte, Ingmar FD12 Nolte, Sandra FD12 Norton, David FC10 Nüesch, Stephan TB11 Q Qiang, Lu TC06 Qian, Yi TA04, FC01 Qiu, Chun (Martin) FA09 Qualls, William TA08 96 S Saboo, Alok R. TB13 Saffert, Peter TB11, FB06 Saibene, Chiara SB14 Sainam, Preethika TB10 Salisbury, Linda Court TD01 Sänn, Alexander FA10 Saraf, Nilesh FD01 Sarkar, Soumya SA14 Sarstedt, Marko FD10 Sayedi, Amin FC03, SC02 Sayman, Serdar SB13 Schaefer, Richard TB14 Schilkrut, Ariel TC07 Schiraldi, Pasquale TB06 Schlabohm, Wiebke FB04 Schlereth, Christian FC01 Schöler, Lisa TB08, FA13 Schulze, Christian FA13 Schumacher, Maureen TC08 Schwartz, Eric TA11, TB02 Schweidel, David TC07, FA02, FB02, SB07 Seenivasan, Satheeshkumar SC13 Seetharaman, P. B. TB14, FA09 Seidmann, Abraham FC03 Seiler, Stephan TB06 Seim, Katja SA04 Selka, Sebastian SA01 Selnes, Fred FB15 Selove, Matthew FC05 Seo, Joo Hwan FD06 Seymen, Omer Faruk TC15 Shacham, Rachel TC04 Shachar, Ron FA08, FC07 Shaffer, Monte TC12, TD08 Shang-Jui, Huang TC08 Shang, Yi-Yun TC10 Shankar, Venkatesh SA14, SC14 Shanmugam, Ravi FB09 Shapira, Daniel SC04 Sharma, Amalesh FD03 Shen, Qiaowei TC06 Shi, Savannah Wei SA03 Shi, Yuying FA06 Shin, Jiwoong FA02, FB05 Shin, Sangwoo FA15, SA01 Shin, Sohyoun SC14 Shin, Woochoel SC02 Shriver, Scott TC06, FD13 Shugan, Steve TC04 Shulman, Jeffrey D. FB05 Siddarth, S. TA03, FD04, SA09 Silva-Risso, Jorge FD04, SA09 Simester, Duncan TA06, TC14 Singh, Jagdip FD12 Singh, Shailendra SA10 Singh, Sonika TD03 Sinha, Jayati SA11 Sinitsyn, Maxim TD14 Sisodiya, Sanjay TA08 Skiera, Bernd TB08, FA13, FC01, FD03, SC07 Smith, Howard TB06 Smith, Michael TA02 Smith, Randall FA12 Smit, Willem FB13 Soberman, David FC05 Soch, Harmeen TD15 Sohl, Timo TC13 Song, Minjae TD01 Song, Reo FA13 Song, Tae Ho SA07, SA15 Sood, Ashish FA02, FB02 Sorell, Michael FB13 Sorescu, Alina FB13 Sosic, Greys FD06 Sotgiu, Francesca TB09 Soutar, Geoff FA10 Spann, Martin TD04, SB02, SC07 Spencer, Fredrika TB08 Sridhar, Shrihari TD02, SA13 Srinivasan, Kannan FC03, FD04, SB05, SB09 Srinivasan, Raji FC02 Srinivasan, Shuba TA02, SA02 Srinivasan, V. Seenu” TC06 Sriram, S TA06, TD02 Staelin, Richard TB08, FC08, FD05 Stahl, Florian TC03, FC03 Steenburgh, Thomas FB01 Steenkamp, Jan-Benedict TB01, FB07, FB09 Stephen, Andrew FC07 Stern, Philip SB10 Stonedahl, Forrest TB03 Stotz, Olaf FB13 Stremersch, Stefan FB07, SB10 Streukens, Sandra FD10 Strijnev, Andrei TA05 Stuettgen, Peter TC01 Su, Meng FA01, FD02 Subhas, M. S. SA15 Sudhir, K. TB06, FA08, SC13 Suher, Jacob FD09 Sultan, Fareena SC09 Sun, Baohong FB08 Sun, Jiong TA01, FC09 Sun, Luping FA01, FD02 Sun, Monic FC02, FC03 Sun, Yutec FD13 Swait, Joffre TD01 Syam, Niladri FB11, FC11, FD05 Tsai, Ming-Chih TD09, FD01 Tsao, Hsiu-Yuan TD15 Tse, Caleb SC09 Tsekouras, Dimitrios SA03 Tucker, Catherine TC14, FB07, SB08 U Umesh, U.N. TC12, TD08 Urban, Glen FB07 V Vadakkepatt, Gautham FA13, SA14 Valentini, Sara FB06 Valizade-Funder, Shyda TA09 van Birgelen, Marcel FB03 Van Bruggen, Gerrit TD04 Van den Bergh, Bram FB10 Van den Bulte, Christophe TB03 van der Lans, Ralf FB10 van Doorn, Jenny TC13 van Heerde, Harald TD02 van Ittersum, Koert FC13 van Lin, Arjen FA09 Van Oest, Rutger TC07 van Soest, Arthur TD01 Vandenbosch, Mark FB06 Vanhoof, Koen FD10 Varadarajan, Rajan SA14 Velu, Chander TB08, TC06 Venkataraman, Sriram TA06, TB04, FA03, SA09 Venkatesan, Rajkumar TC09 Verhoef, P.C. (Peter) TB03, TC13 Vernik, Dinah TD05 Villar, Claudio SB04 Villas-Boas, J. Miguel FC03 Vir Singh, Param FB08 Viswanathan, Madhu FC04 Viswanathan, Siva SA03 Vitorino, Maria Ana FD09, SA04 Voleti, Sudhir TB04 Voola, Ranjit FC11 T Talay, M. Berk TD13 Talukdar, Debabrata TA09, TD04, SC13 Tan, Alicia SA12 Tan, Tom Fangyun SA07 Tang, Elina SA13 Tang, Tanya TA08 Tanner, Robin SB07 Taylor, Gail FA09 Teixeira, Thales FB03 Telang, Rahul TA02 Tellis, Gerard J. TA13, TB08, FC02, SA02 ter Braak, Anne SA13 Theron, Sophie SA13 Thomadsen, Raphael TD09 Tian, Yue FA02 Tibbits, Matthew TB04 Tirunillai, Seshadri FC02 Toker-Yildiz, Kamer FB01 Toklu, Kerem Yener TA14 Topaloglu, Omer FC12 Townsend, Janell FA05 Tran, Thanh SA06 Trauzettel, Volker FD06 Tripathi, Manish FA02, FB02 Trivedi, Minakshi FB01, FB06 Tsai, Kuen-Hung TC10 W Wagner, Nils TD04 Wagner, Ralf SB06 Walker, Doug SC06 Wang, Huihui TA06 Wang, Jiana-Fu FD15 Wang, Kangkang FD05 Wang, Kitty FD11 Wang, Lei TD12, SA12 Wang, Li FD04 Wang, Luming TB04 Wang, Paul TC01 Wang, Ping FA01, FD02 Wang, Qi TA13, SB09 Wang, Qiong FA06 97 Wang, Tiffany Ting-Yu TB12 Wang, Wenbo FC04 Wang, Xin SB04 Wang, Yantao TA04 Wang, Yanwen TD02 Wathieu, Luc TC14, SB08 Wattal, Sunil SA11 Watts, Jameson TC02 Wedel, Michel FB11 Wei, Muyu TD07 Weitz, Bart FA06 Wen, Chieh-Hua FD01 White, Joseph SA01 Wiebach, Nicole TA10 Wieringa, Jaap SB10 Wies, Simone FA13, FB13 Wilbur, Kenneth TD02, FD05 Wilczynski, Petra FD10 Wilken, Robert SB06 Wilson, Hugh TD09 Wind, Jerry FB07 Winer, Russ TA01 Wintoki, M. Babajide TA13 Worm, Stefan TB12 Wu, Chunhua SC01 Wu, Fang TC01 Wu, I-Huei FD06 Wu, Jianan FD02 Wu, Ruhai SB05 Wu, Steven SA07 Wu, Tao FA05 Wu, Yinglu FD02 X Xiao, Li TB11 Xiao, Ping TC06 Xiao, Ying SB13 Xie, Jinhong TA13, SB09 Xie, Yi TA12 Xie, Yuying SB04 Xiong, Guiyang FC12 Xu, Zibin TD14 Xubing, Zhang FC06 Y Yamada, Masataka TC05 Yan, Hsiu-Feng TD15 Yan, Ruiliang SA13 Yang, Botao FA08 Yang, Chen-Han FD01 Yang, Geunhye FB09 Yang, Joonhyuk SB07 Yang, Jun TC03 Yang, Sha TA04, FA01, FB10 Yang, Wei-Jhih FC06 Yang, Xiaoqi TB07 Yang, Ying FB11, FC11 Session Index Thursday, 8:30am - 10:00am Thursday, 3:30pm - 5:00pm TA01 TA02 TA03 TA04 TA05 TA06 TD01 TD02 TD03 TD04 TD05 TD06 TD07 TD08 TD09 TD10 TD11 TD12 TD13 TD14 TD15 TA07 TA08 TA09 TA10 TA11 TA12 TA13 TA14 TA15 Marketing Science Institute I Google WPP Award Papers Internet Bayesian Econometrics I: Methods & Application New Product I: Introduction Competition I: Measuring the Impact of Competition in Retail Markets I ASA Special Session on the Marketing-Statistics Interface – I Innovation I: Open Innovation Promotions I Consumer Behavior: Perceptions Direct Marketing Branding Quantifying the Profit Impact of Marketing I Customers’ Willingness-to-pay Research B2B: Relationships Friday, 8:30am - 10:00am Thursday, 10:30am - 12:00pm TB01 TB02 TB03 TB04 TB05 TB06 TB07 TB08 TB09 TB10 TB11 TB12 TB13 TB14 Choice II: Effects on ... Online Advertising - II Internet: Car Buying Econometric Methods II: General New Product IV: Strategy Competition IV: Quality Services Innovation IV Retailing I: General Consumer Behavior: Decision Making Advertising Content Bidding Quantifying the Profit Impact of Marketing IV Pricing and Competition CRM II: Customer Loyalty FA01 FA02 FA03 FA04 FA05 FA06 FA07 Marketing Science Institute II Empirical Modeling Social Networks and Profitability Bayesian Econometrics II: Methods & Application New Product II: Diffusion Competition II : More on Measuring the Impact in Retail Markets ASA Special Session on the Marketing-Statistics Interface – II Innovation II Promotions II Decision Making Response to Advertising Brand Identity Quantifying the Profit Impact of Marketing II Dynamic Pricing Issues FA08 FA09 FA10 FA11 FA12 FA13 FA14 FA15 Choice III: More Effects on … UGC-I (The Evolution and Impact of Online Opinions) Internet: Customer Response Dynamic Models I New Product V: Design & Development Channels I: General Panel Session: Cases? Projects? Simulations? Problem Sets? What’s the Best Way to Teach Marketing Science? The Long Run Consequences of Short Run Decisions I Retailing II: General Measurement Issues Using Endorsers in Advertising Aesthetics Marketing Finance Interface I Pricing and Consumer Behavior CRM IV: Customer Loyalty Thursday, 1:30pm - 3:00pm Friday, 10:30am - 12:00pm TC01 TC02 TC03 TC04 TC05 TC06 TC07 TC08 TC09 TC10 TC11 TC12 TC13 TC14 TC15 FB01 FB02 FB03 FB04 FB05 FB06 FB07 Choice I: New Models of … Online Advertising - I Internet: Social Influence Econometric Methods I: General New Product III: Adoption Competition III: General ASA Special Session on the Marketing-Statistics Interface – III Innovation III Promotions III Consumer Behavior Advertising Strategy Brand Equity Quantifying the Profit Impact of Marketing III Consumer Responses to Pricing CRM I: Customer Lifetime Value FB08 FB09 FB10 FB11 FB12 FB13 FB14 FB15 98 Choice IV: Market Structure & Substitution UGC-II (Quest for Comprehension and Integration) Internet Relationship Dynamic Models II Game Theory I: Decisions Under Limited Information Channels II: Relationship Management Panel Session: Collaborative Research: Reasons Why, Difficulties and Potential Models (Data Base Sharing and Prospective Meta Analysis The Long Run Consequences of Short Run Decisions II Retailing III: Competition Bayesian Applications Salesforce I Internet: Unique Topics Marketing Finance Interface II Pricing Research CRM III: Customer Loyalty OZYEGIN UNIVERSITY (OZU) INVITES YOU TO ATTEND THE 35TH ANNUAL MARKETING SCIENCE CONFERENCE HOSTED IN ISTANBUL, TURKEY June 20-22, 2013 HONORARY CONFERENCE CHAIR – Erhan Erkut, University President CO-ORGANIZERS – Tülin Erdem, [email protected] Koen Pauwels, [email protected] ozyegin.edu.tr Room Legends Ballroom I Legends Ballroom II Legends Ballroom III Legends Ballroom V Legends Ballroom VI Legends Ballroom VII Founders I Founders II Founders III Founders IV Champions Center I Champions Center II Champions Center III Champions Center VI Champions Center V Track 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 Promotions I Consumer Behavior: Perceptions Direct Marketing Branding Quantifying the Profit Impact of Marketing I Customers’ Willingness-toPay Research B2B: Relationships Promotions II Decision Making Response to Advertising Brand Identity Quantifying the Dynamic Pricing Profit Impact of Issues Marketing II Promotions III Consumer Behavior Advertising Strategy Brand Equity Quantifying the Profit Impact of Marketing III Consumer CRM I: Customer Responses to Lifetime Value Pricing Retailing I: General Consumer Behavior: Decision Making Advertising Content Bidding Quantifying the Profit Impact of Marketing IV Pricing and Competition CRM II: Customer Loyalty Retailing II: General Measurement Issues Using Endorsers in Advertising Aesthetics Marketing Finance Interface I Pricing and Consumer Behavior CRM IV: Customer Loyalty Retailing III: Competition Bayesian Applications Salesforce I Marketing Internet: Unique CRM III: Finance Interface Pricing Research Topics Customer Loyalty II Retailing IV: Competition Segmentation Salesforce II Word of Mouth Financial and Marketing Price Discounting Decision Making Stratety CRM V: Customer Satisfaction Retailing V: Location Decisions Survey Research Sports and Fashion Models of Word of Mouth Network Effects Processes CRM VI: Customer Satisfaction Health Care Marketing I Social Influence I Health Care Marketing II Social Influence II 7:30AM 8:30AM T H U R S D A Y Marketing Science Institute I Google WPP Award Papers Internet Bayesian Econometrics I: Methods & Application Social Networks and Profitability Bayesian Econometrics II: Methods & Application New Product I: Introduction Competition I: Measuring the Impact of Competition/… ASA Special Session on Innovation I: the Marketing- Open Innovation Statistics/… New Product II: Diffusion Competition II: More on Measuring the Impact in/… ASA Special Session on the MarketingStatistics/… New Product III: Competition III: Adoption General ASA Special Session on the MarketingStatistics/… 10AM Morning Break Marketing 10:30AM Science Institute II Empirical Modeling Noon 1:30PM Choice I: New Models Of … Online Advertising - I Internet: Social Influence Econometric Methods I: General Internet: Car Buying Econometric Methods II: General Innovation III Choice II: Effects Online on … Advertising - II New Product IV: Competition IV: Strategy Quality Services Innovation IV 5-7PM Reception Continental Breakfast / Registration Choice III: More Effects on … UGC-I (The Evolution and Impact of Online/… Internet: Customer Response Dynamic Models I New Product V: Design & Development Dynamic Models II Game Theory I: Decisions Under Limited Information Channels I: General The Long Run Consequences of Short Run Decisions I Panel on Teaching Marketing Science 10AM Morning Break Choice IV: UGC-II (Quest for 10:30AM Market Structure Comprehension & Substitution and/… Internet Relationship Channels II: Relationship Management The Long Run Consequences of Short Run Decisions II Panel on Collaborative Research Noon Lunch Analytic Models Choice V: UGS-III (Content 1:30PM of Online Empirical Results and Impact) Behavior Structural Models I Game Theory II: Market Entry Channels III: Competition New Directions Dynamic Models in Word of Mouth in Marketing Channels V: Strategy Meet the Editors/ Management Mkg. Sci.; Man. Myopia and Real Sci.; JMR Activity/… 3PM 3:30PM Afternoon Break Choice VI: Applications UGC-IV (Content and Impact) Online Search Structural Models II Game Theory III: General 5-7PM Gala Dinner 7:30AM Continental Breakfast / Registration 8:30AM Conjoint Analysis: Improving the Process Twitter and Social Media Effects of Online Medium on Consumer Behavior Structural Models III Game Theory IV: Signaling Channels VI: General Game Theory V: General Improving Efficiency in Marketing Negotiations Entertainment Continuous-Time Retailing VI: Auto Marketing I: Marketing Industry Movies 10AM 10:30AM 3PM Marketing Strategy I: General Marketing Online Word of Private Labels I: CRM VII: Strategy II: Firm Mouth Research General Customer Equity Performance Morning Break Advertising: Strategy Social Networks Product Online Consumer Management: Behavior General Auctions and Pricing Meet the Editor: Journal of Unique Topics 2 Service Research International Marketing I: GeneralEmerging/... Entertainment Marketing II Noon 1:30PM Privacy and Marketing Afternoon Break 7:30AM 8:30AM S A T U R D A Y Innovation II Lunch 3PM 3:30PM F R I D A Y Continental Breakfast / Registration Private Labels II: Effect on the Distribution Channel Marketing Strategy III: General Lunch Advertising and Two Sided Markets Consumer Preferences Entertainment Marketing III Conference Concludes International Marketing II Private Labels III: Marketing Effect on Market Strategy IV: Firm Shares Performance CRM VIII: Customer Lifetime Value Friday, 1:30pm - 3:00pm Saturday, 8:30am - 10:00am FC01 FC02 FC03 FC04 FC05 FC06 FC07 FC08 FC09 FC10 FC11 FC12 FC13 FC14 FC15 SA01 SA02 SA03 SA04 SA05 SA06 SA07 SA08 SA09 SA10 SA11 SA12 SA13 SA14 SA15 Choice V: Empirical Results UGC-III (Content and Impact) Analytic Models of Online Behavior Structural Models I Game Theory II: Market Entry Channels III: Competition New Directions in Word of Mouth Dynamic Models in Marketing Retailing IV: Competition Segmentation Salesforce II Word of Mouth and Marketing Strategy Financial Decision Making Price Discounting CRM V: Customer Satisfaction Conjoint Analysis: Improving the Process Twitter and Social Media Effects of Online Medium on Consumer Behavior Structural Models III Game Theory IV: Signaling Channels VI: General Entertainment Marketing I: Movies Continous-Time Marketing Retailing VI: Auto Industry Health Care Marketing I Social Influence I Online Word of Mouth Research Private Labels I: General Marketing Strategy II: Firm Performance CRM VII: Customer Equity Friday, 3:30pm - 5:00pm Saturday, 10:30am - 12:00pm FD01 FD02 FD03 FD04 FD05 FD06 FD07 FD08 SB01 SB02 SB03 SB04 SB05 SB06 SB07 SB08 SB09 SB10 SB11 SB13 SB14 SB15 FD09 FD10 FD11 FD12 FD13 FD14 FD15 Choice VI: Applications UGC-IV (Content and Impact) Online Search Structural Models II Game Theory III: General Channels V: Strategy Meet the Editors Marketing Science/Management Science Managerial Myopia and Real Activity Mis-Management: Consequences for Marketing and Firm Performance Retailing V: Location Decisions Survey Research Sports and Fashion Models of Word of Mouth Processes Network Effects Marketing Strategy I: General CRM VI: Customer Satisfaction Advertising: Strategy Social Networks Online Consumer Behavior Product Management: General Game Theory V: General Improving Efficiency in Marketing Negotiations Entertainment Marketing II Privacy and Marketing International Marketing I: General/Emerging Markets Health Care Marketing II Social Influence II Private Labels II: Effect on the Distribution Channel Marketing Strategy III: General CRM VIII: Customer Lifetime Value Saturday, 1:30pm - 3:00pm SC01 SC02 SC03 SC04 SC06 SC07 SC09 SC13 SC14 99 Advertising and Two Sided Markets Auctions and Pricing Meet the Editor: Journal of Service Research Unique Topics 2 Consumer Preferences Entertainment Marketing III International Marketing II Private Labels III: Effect on Market Shares Marketing Strategy IV: Firm Performance