Analysis of WIMAX/BWA Licensing in India: A real option approach

ANALYSIS OF WIMAX/ BWA LICENSING IN INDIA: A REAL OPTION APPROACH Pankaj Sinha and Akshay Gupta Faculty of Management Studies, University of Delhi Abstract Indian Internet and broadband market has experienced very slow growth and limited penetration till now. The introduction of Broadband Wireless Access (BWA) is expected to aid in increasing the penetration of internet and broadband in India. The report sheds light on the guidelines and procedure used in 4G/BWA spectrum auction and presents comparative analysis of the competing technologies, providing the information about suitability of each technology available. Recently held 4G/ BWA spectrum auction saw enthusiastic participation by the industry and even saw some new entrants in Indian broadband market. Government benefitted by Rs, 385bn that it earned as revenue from the auction of the spectrum and projected it as successful auction. However, the question remains if the auctions were efficient and whether they led to creation of value or will it prove to be burden to the telecom operators and will depress their balance sheet for years to come. The report uses both traditional valuation methods such as Discounted Cash Flow as well as Real Option approach to answer such questions. Using DCF analysis, the broadband subscribers have been forecasted to grow from present 13.77mn to 544mn by the end of 2025. The wireless subscribers are forecasted to be 70% of the total broadband subscribers after 5 years of roll out as it will be difficult to replace all wireline subscribers with wireless subscribers in India due to the high cost of wireless broadband and new technology. WiMAX is expected to increase its presence with time and reach 90mn subscribers from meager 0.35mn subscribers by 2025. Using industry wide cost of capital as 12.05%, the Net Present Value has been found Rs 221bn aggregate with an IRR of 17.1%. Using Real option approach, the value of license has been calculated as Rs 437bn which is 13.5% more than the spectrum fees paid by the operators. This mismatch, between the auction value and the correct value that should have been discovered by supply-demand dynamics, can be due to limited participants in BWA spectrum auctions and companies such as TATA and Reliance opting out of the auction process midway as well as uncertainty about acceptance of new technology with Indian subscribers. 0|Page Table of Contents 1. Introduction ............................................................................................................................. 4 1.1 1.1.1 Introduction ............................................................................................................... 4 1.1.2 India’s growth drivers ............................................................................................... 5 1.2 2. 3. Country Analysis – India.................................................................................................. 4 Industry Analysis – Telecom............................................................................................ 6 1.2.1 Introduction ............................................................................................................... 6 1.2.2 Key growth driver - Mobile services ........................................................................ 7 1.2.3 Telecom services revenues ....................................................................................... 7 1.2.4 Teledensity - International comparison .................................................................... 8 1.2.5 Teledensity – Indian scenario ................................................................................... 9 1.2.6 Telecom – Data revenue share .................................................................................. 9 1.2.7 Non-voice revenue – International comparison ...................................................... 10 1.2.8 Data adoption and usage ......................................................................................... 11 1.2.9 Data application usage pattern ................................................................................ 11 Broadband – Next driver of growth ...................................................................................... 12 2.1 Present scenario .............................................................................................................. 12 2.2 Present problems ............................................................................................................ 14 2.3 Future growth drivers ..................................................................................................... 14 2.4 4G/ BWA spectrum auction ........................................................................................... 16 2.4.1 Auction guidelines .................................................................................................. 16 2.4.2 Auction results ........................................................................................................ 18 2.4.3 Auction price comparison with 3G ......................................................................... 19 2.4.4 Auction price comparison - International ............................................................... 19 Historical Background .......................................................................................................... 21 3.1 Evolution from 0G to 3G ............................................................................................... 21 3.1.1 0G............................................................................................................................ 21 3.1.2 1G............................................................................................................................ 21 3.1.3 2G............................................................................................................................ 21 3.1.4 3G............................................................................................................................ 23 1|Page 3.2 4G ................................................................................................................................... 26 3.2.1 WiMAX .................................................................................................................. 26 3.2.2 LTE ......................................................................................................................... 28 3.2.3 4G versus rest technologies .................................................................................... 29 3.2.4 Comparison with Wi-Fi .......................................................................................... 31 3.3 Typical 4G network ........................................................................................................ 32 3.4 4G success deployment worldwide ................................................................................ 33 3.5 Future Trends ................................................................................................................. 34 4. Literature Review.................................................................................................................. 36 5. Methodology ......................................................................................................................... 45 6. 7. 5.1 Discounted Cash Flow Valuation ................................................................................... 46 5.2 Real Option Valuation .................................................................................................... 47 Data & Assumptions ............................................................................................................. 50 6.1 Population & Population Growth Rate .......................................................................... 50 6.2 Revenue Projections ....................................................................................................... 50 6.3 Internet and Broadband Subscribers .............................................................................. 51 6.4 Broadband Penetration ................................................................................................... 55 6.5 Wireline and wireless broadband ................................................................................... 55 6.6 WiMAX Broadband subscribers .................................................................................... 57 6.7 WiMAX Broadband ARPU ........................................................................................... 59 6.8 Operator’s share in Data ARPU ..................................................................................... 60 6.9 Capital Expenditure ........................................................................................................ 60 6.10 Operating Expenditure ................................................................................................ 61 6.11 Financing of Total Capital Expenditure ..................................................................... 62 6.12 Depreciation & Amortization ..................................................................................... 62 6.13 Cost of Capital ............................................................................................................ 62 Results ................................................................................................................................... 65 7.1 Revenue Forecast ........................................................................................................... 65 7.2 DCF valuation ................................................................................................................ 65 7.2.1 NPV and IRR .......................................................................................................... 65 7.2.2 Sensitivity analysis.................................................................................................. 66 2|Page 7.3 Incremental Revenues for 4G Operators ........................................................................ 68 7.4 Real Option valuation..................................................................................................... 68 7.4.1 Real Option value ................................................................................................... 68 7.4.2 Sensitivity analysis.................................................................................................. 69 8. Conclusion and Recommendation ........................................................................................ 71 9. References ............................................................................................................................. 73 10. APPENDIX ........................................................................................................................ 76 3|Page 1. Introduction 1.1 Country Analysis – India 1.1.1 Introduction India is the largest democracy in the world with its population touching 1.2 bn1. It is also among the fastest growing emerging economy in the world, with a nominal GDP of $1.36 trillion and GDP in terms of PPP of $3.86 trillion, it is 11th largest and 4th largest economy in the world respectively.1 Even during recent global recession Indian economy has shown considerable resilience and has been able to achieve growth rate of 6.7 percent in 2009-10, which is expected to be followed by growth rate of 8.5 percent2 in 2010-11, among highest in the world. The resilience comes from the fact that India’s economy growth is not majorly dependant on exportoriented sectors like other major economies and the economy is mainly driven by the strong domestic consumption. According to a PwC report titled “The World in 2020”, in PPP terms, India's GDP will overtake GDP of Japan in 2011 and of United States by 2045.3 It further mentions that India's annual economic growth rate is expected to average around 8% and will be the world's fastest growing major economy by 2050, surpassing China. 2009 Rank 1 2 3 4 5 6 7 8 9 10 Country US China Japan India Germany Russia UK France Brazil Italy GDP at PPP (constant 2009 US $bn) 14256 8888 4138 3752 2984 2687 2257 2172 2020 1922 2050 Rank 1 2 3 4 5 6 7 8 9 10 Country China India US Brazil Japan Russia Mexico Indonesia Germany UK GDP at PPP (constant 2009 US $bn) 59475 43180 37876 9762 7664 7559 6682 6205 5707 5628 Source: PwC estimates Figure 1: GDP ranking in terms of PPP 1 International Monetary Fund, http://www.imf.org/, accessed on 9/3/2011 Confederation of Indian Industries, http://www.ciionline.org/Economy.aspx?enc=LqAXY5bXIsb2PzUHQxy2iQ==, accessed on 9/3/2011 3 PwC Report, “The World in 2050”, January 2011 2 4|Page The factors in favor of India have been highlighted as increasingly growth friendly economic policies, significantly younger and faster growing working population, start from a relatively lower level of economic development, changing dependence on outsourcing and more on manufacturing exports due to buildup of strong engineering skills and rising levels of education in the general population and attractive growth of consumers in the form of rapidly growing middle class population. 1.1.2 India’s growth drivers The present factors driving India’s growth can be summed as Services, Foreign Direct Investments and Consumption4: • Services – India has major part of its annual GDP coming from services sector. As per CII, service sector accounted for nearly 60% of the overall growth in India’s GDP during the last 10 years. The sector is dynamic and its fastest growing segments are Communications and Banking. • Foreign Direct Investment – India adopted LPG (Liberalization, Privatization and Globalization) policies starting from 1991-1992 and since then has adopted various initiatives such as opening up of new sectors to FDI, raising caps on FDI in sectors already open for investments as well as simplified and changed procedure for investment in order to attract FDI. Sectors that have attractive largest share of the FDI are power, services, telecommunication, computer software & hardware and housing & real estate • Consumption – With around 25-30% population as middle class, there has been a spurt in demand for the goods and services, which is expected to only increase in the coming years. India’s demographic composition also tilts favorably towards consumption with over 210 million people expected to be in 20-29 years age group by 2015. As a result it is expected to further raise domestic demand for services and products, driven by not only numbers but also by higher incomes, increasingly globalised outlook as well as increasing propensity to spend. 4 PwC Report, “Mobile Broadband Outlook-2015”, 2010 5|Page 1.2 Industry Analysis – Te Telecom 1.2.1 Introduction Indian telecom industry has sho shown a successful growth story so far growing ing at a CAGR of approximately 30% since 1995. IIn wireless, India is world’s fastest growing mar arket5 with CAGR of more than 117% in last 100 years and reaching 752 million mobile phone one subscribers by February, 20116. In terms of wire ireless connection network, India is at second posi osition, just behind China.7 With a significant contri ntribution of more than 2% of the GDP since 2008-2009, 20 Indian telecom service sector’s contribu ibution is expected to rise even further. Accordin ing to latest TRAI report on subscriptions in India6, total telephone subscriber base has reached 787.28 78 million by Dec-2010 an increase of 22.62 m million compared to Nov-2010. Indian telecom industry in is highly competitive at present with over er 10 service providers vying for the pie. This has ensured intense competition and continuously fal falling ARPU with the introduction of innovative ve schemes such as Re 20 per minute calling rates,, R Re30 per month for unlimited SMSes etc, all of which w makes the rates lowest in the world. Source: TR TRAI Figu igure 2: Service provider - Market share 5 http://www.ibef.org/industry/telecomm mmunications.aspx, accessed on 9/3/2011 6 TRAI December Report, http://www.tr .trai.gov.in/WriteReadData/trai/upload/PressReleases/798/ 8/prerdiv9feb11.pdf, accessed on 1/3/2011 7 http://indiabudget.nic.in/es2008-09/inf /infra.htm, accessed on 9/3/2011 6|Page 1.2.2 Key growth driver - Mo Mobile services 95.5% of the subscribers are wi wireless compared to 4.5% wireline subscribers, rs, with increasing percentages going in favor of w wireless. Overall Tele-density reached 66.16% % in Dec-2010, an increase from 64.34% in Nov-20 2010 with overall urban and rural tele-densities ies being 147.88% and 31.18% respectively. Thiss co constitutes wireless tele-density of 63.22% with ith urban and rural tele-densities being 140.53% and nd 30.11% respectively, wireline tele-density off 2.95% 2 with urban and rural tele-densities being 7.35 .35% and 1.07% respectively and broadband subs bscription of 10.92 Million in Dec-2010 with an inc increase from 10.71 Million in Nov-2010. Accor cording to Informa Telecoms & Media's updated for forecasts8, mobile subscribers in India will exceed eed 1.16 billion by the end of 2013, making it the he world's largest mobile market surpassing Chi hina. This will be achieved on basis of CAGR of m more than 128% expected since end of 2009. Source: TR TRAI F Figure 3: Telecom subscribers growth 1.2.3 Telecom services reve evenues India’s telecom services industry try revenues are forecasted to rise from Rs. 1,13,00 ,000 crore in 200809 (growing at CAGR of 20 perce rcent) to about Rs. 344,921 crore (US$76.57 billio llion) by 2012 with a rate of growth of around 26%, %, which will lead to generation of employment nt opportunities o for about 10 million people duringg th the same period.9 The sector will create this emp mployment for 2.8 8 http://www.informatm.com/itmgconte tent/icoms/s/press-releases/20017778395.html, accessed on 9/3/2011 9 http://economictimes.indiatimes.com/n m/news/news-by-industry/telecom/Indian-telecom-market-to to-be-at-Rs-344921- crore-by-2012/articleshow/2563062.cms cms, accessed on 9/3/2011 7|Page million people directly and for 7 million indirectly. Contribution to the revenue has been mainly form mobile services (up to 90%) with decreasing contribution from the wireline not only in percentage but also absolute terms. Services such as Long Distance (National and International), and VSAT etc have also shown increase over the years. Source: PwC Report/ TRAI Figure 4: Telecom Services Revenue (Rs '00 crore) 1.2.4 Teledensity - International comparison A comparison of India’s telecom industry with other developing countries shows huge untapped population when compared to countries with similar GDP per capita (in PPP terms). In absolute numbers nearly 500 million population is still without a connection and with increasing penetration in rural areas representing about 70% of the population, this market is waiting to be tapped. Source: PwC Report Figure 5: International precedent - Mobile teledensity 8|Page 1.2.5 Teledensity – Indian ndian sc scenario The growth in mobile tele-densit nsity is due to exponential growth in last 5 years ars in both rural as well as urban areas with urbann aareas leading above 100% tele-density and rura ural areas lingering around 21% only. Considering th that most of the unserved 500 million strong popu opulation is in rural areas (combined with saturationn llevels reached in Urban areas) and with ability ity to pay moderate to low charges, time is not far wh when intense competition will drove operators to look for further subscriber growth in rural areass oonly.4 Source: TRA RAI Figure 6: Teledensity 1.2.6 Telecom – Data revenu revenue share Along with the steady increasee in subscribers, another trend that has been seen en is falling ARPU due to mainly decrease in tariff ff aaccompanied with decrease in marginal additio tion of subscribers. As mentioned before, due to in intense competition which has necessitated innovation in in the industry has led to significant low lowering of the ARPUs over the time. This has in turn necessitated for operators to push Value Add dded Services (VAS) and other Data Servicess in i order to keep profit margins from falling furth rther. With time Indian consumers are also becom oming more savvy and thus these special servicess ccan be instrumental in differentiating between n companies c in the long run. Data revenue has beenn stagnant at 11-12% for last few years. 9|Page Source: CMIE Datab tabase10 Figure 77: Data revenue share: Indian GSM Industry 1.2.7 Non-voice revenue – International comparison When compared to other majorr eeconomies and their pattern of usage, it can be clearly seen that as population gets exposed to tec technology, with time the movement is towards ds more data usage and lesser percentage of voice. T This change in percentage of reveue in favor of data comes with more and more use of data applic lications and in turn leads to increase in ARPU. A similar trend is expected to come in India andd tthe data revenue percentage in overall revenue ue should increase further. Source: Reliance re report15 F Figure 8: Non voice revenue mix (%) 10 CMIE Database, accessed on 9/3/201 011 10 | P a g e 1.2.8 Data adoption and usage At present although the revenue from data has been nearly same over the past 4-5 years, the pattern of usage has been changing in favor of the non-SMS data services. According to PwC analysis, currently non-SMS data usage in India is primarily from music related services (such as caller tunes, ringtones and music downloads etc), voice mail services, Multimedia messaging services, data application such as Blackberry, itemized billing, m-commerce applications and download on content applications. Another report by BDA/FICCI sheds light on the usage pattern of smart phone users and overall mobile users in India as shown below: Source: FICCI report Figure 9: Data service adoption and usage 1.2.9 Data application usage pattern The type of applications used by internet users on mobile differs in developing countries when compared to developed nations. As shown below, in countries such as US and Europe, the main usage is of email services followed by search and news/politics and then sports whereas in countries such as China and India, entertainment, games and music takes more percentage usage. This will be instrumental in type of applications that can be provided in future and as applications used in developing countries are more data intensive and thus it will suit easy adoption of 3G/4G technologies in future and will be a key driving factor. 11 | P a g e Source: FICCI report Figure 10: Application usage pattern 2. Broadband – Next driver of growth 2.1 Present scenario India has very low broadband penetration at present. This is due to the wired infrastructure that cannot afford broadband speed in most of the places especially in rural areas and tier 2/3 cities and thus putting an impediment to the process of providing access through broadband. Since 2005 when wireline broadband connections were made available in India, only 10.9 million subscribers have been added so far which looks insignificant compared to nearly 23 million new mobile subscribers added from November to December 2010. Total internet subscribers (mn) 12 10 8 6 4 Non-broadband wireline broadband 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2 0 wireless broadband Source: CMIE database/ TRAI Figure 11: Internet subscriber’s breakup 12 | P a g e As of March 2010, India had 15.9 million broadband subscribers which included 10.9 million wireline broadband subscribers and 5 million wireless subscribers (mainly TATA and MTNL subscribers using un-licensed spectrum). This is far less than the target of 20 million set by Indian government for the broadband users. Thus Indian market with 66.6% mobile penetration is penetrated even less than 1% in terms of broadband and thus presents a huge market untapped. Source: Reliance report15 Figure 12: Global broadband penetration The Indian broadband scenario at present is nearly same as it was there in case of the cellular voice by 1995, which is India ranks almost at the bottom with less than 1% penetration. It is expected that with growth in GDP over next decade, the broadband penetration will increase tremendously as seen in other countries. Source: Reliance report15 Figure 13: Correlation between Broadband penetration and GDP per capita 13 | P a g e 2.2 Present problems According to PwC analysis the main reasons for this under-penetration are: 1) Low Personal Computers penetration along with limitations in terms of the wireline infrastructure 2) Technical limitations such as right of way challenges and relatively high tariffs 3) The existing infrastructure is mainly limited to urban and semi-urban areas with very less presence in rural 4) Further, the existing wireline infrastructure cannot be used for high speed broadband 5) According to present rules, local loop unbundling cannot be done that would have enabled new payers to use the already present infrastructure All this has limited the growth of the broadband services. Even though around 20% of the existing mobile users have handset capable of accessing internet but the access has been restricted due to lack of high speed using the existing technologies in wireless namely EDGE, CDMA 1X and GPRS technologies. Recently CDMA providers have launched EVDO technology to provide wireless broadband access. 2.3 Future growth drivers According the PwC analysis, there are numerous drivers that will be critical in order to reach the expected subscriber base for the broadband that will make the 4G auction affordable to the telecom providers as mentioned below: • Demographic Profile: India’s population in the 20 – 29 age group is expected to reach above 210 million by 2015. This will have direct implication on the change in the taste and preference of the general population and the propensity to try the latest and enhanced services as provided by 4G technology should improve and this will in turn make them early adopters of the technology thus expected to drive the 4G market. • Quality of Service (QoS): Increase demand for streaming videos, online games, improved applications and high speed internet will demand better quality of service than 14 | P a g e is possible at present. The quality can be only improved by better density of the towers, and better coverage, all of which will depend in turn on the density of the subscribers per base station • Applications and Content: With falling ARPU and services being given at present becoming more or less a commodity, applications are expected to be the real differentiator between operators and the quality and variation in applications will depend on the capabilities that can be provided by 4G • Access Device: The adoption will also depend on the switching costs from DSL/ cable to wireless using 4G technology as well as backward compatibility. The technology is still in initial stages and as the scale hasn’t been reached till now, it will require time for the access devices to be cost effective to price sensitive customers in India. • Enterprise: Enterprise require high bandwidths for more productivity and they can easily afford the same at higher ARPUs and hence will be one of the early adopters to better speeds provided by 4G technologies • Declining ARPU: With declining ARPU, operators will have added incentive to look for and push 4G technologies and stop falling ARPUs and hence their margins. • Better Business Case: Considering the higher cost of rolling out wireline broadband in the existing conditions in country it is better business sense to go wireless. • Low Broadband Penetration: There is huge potential in the under-penetrated market such as India where broadband penetration is less than 1%. 15 | P a g e 2.4 4G/ BWA spectrum auction 2.4.1 Auction guidelines Considering the necessity of broadband penetration and the expected benefits of the same to the economy, Government recently auctioned spectrum to be used to provide mobile services and broadband using 3G as well as broadband using 4G. DoT in consultation with TRAI conducted auction and they used auction process similar to the ones used in other countries such as UK, Germany to auction off 2 bands of 20MHz in each of the 22 circles. The auction was conducted in 2010-11 and the process adopted during the 4G spectrum auction and the associated details are given below: Auction Rules • Only one entity to be allowed to bid from the same Group Entity: A group entity is defined as existing licensees with common parents having at least 26% stake in these companies • Invitation Stage (Applications): 19th March, 2010 • Final draft for application: 23rd March, 2010 Auction Timelines • Pre-qualification of bidders: 30th March, 2010 • Auction stage: 9th March for 3G (For BWA it will start 2 days after 3G auction closes) • Payment stage: 10 days after close of auction • Grant Stage: 15 days after completion of auction and full payment • Processing fee of $2200 Application Requirements • Ernest money in terms of bank guarantee – Eligibility points for the auction would be based on amount of earnest money deposited (Pan-India earnest money for BWA is $56mn) • Nomination of authorized signatory Auction is to be carried out in 2 stages, with all circle auctions to start and end at the same time Auction Process • Clock stage to determine the spectrum lot winners • Assignment stage to assign specific frequency to the winners • Negative demand: 0% increase Bid Price Increment • Zero excess demand: 2% increase • Demand in excess by 1 or 2 bidders: 5% increase • Demand in excess by 3 or more bidders: 10% increase 16 | P a g e • Successful bidders are required to deposit 25% of the auction amount within 5 days and remaining 75% within next 10 days Allocation of Spectrum • On full payment, DoT will issue LoI and would allocate the spectrum within 15 days of full payment of auction amount • New spectrum winners will get 90 days from auction closure to fulfill license requirements (UAS License Application) Source: DoT11/ FICCI report Figure 14: Summary BWA spectrum auction process • Any service provider who holds a UASL/ CMTS license or fulfils UASL criteria and will Eligibility acquire one before commencing operations; or • Holds Internet Service Provider (ISP) license category ‘A’ or ‘B’ Frequency Bands • 4 Blocks of 20MHz of spectrum (TDD mode) in the 2.3 GHz and 2.5 GHz band • Reserve price for the auction is designated as: Entry Fee - USD 17.8 mn – Mumbai, Delhi and Category A circles - USD 8.9 mn – Kolkata and Category B circles - USD 3.3 mn – Category C circles Rollout Obligations M$A Guidelines • At the end of five years from date of spectrum allocation, the service provider needs to cover 90% of metro areas and 50% of rural SDCAs • M&A between UASL holder in the same service area is allowed only after 3 years from the date of license • Spectrum charges for existing UASL/ CMTS players to be the same as the revised spectrum Spectrum Usage Charges charges for 2G spectrum. BWA spectrum is not to be counted to arrive at the relevant slab of spectrum • Standalone BWA operators or ISP license holders will be charged 3% of AGR, after the first year of allocation of spectrum Source: DoT/ FICCI Report12 Figure 15: BWA spectrum auction guidelines 11 DoT, “Auction of 3G and BWA spectrum, February 2010 12 FICCI, “3G and BWA: The Next Frontier”, January 2009 17 | P a g e 2.4.2 Auction results BWA auction was successful and Government got overwhelming response from the industry13. The auction process went for 16 days and 117 rounds of bidding and ended with total revenue of Rs 12848 crore for the Government with the pan-India spectrum auction. The pan India bid amount closed at around 7.3x the base price of Rs 1750 crore for pan India spectrum. The only company to take pan-India license turns out to be Infotel Broadband Services (all 22 circles) followed by Aircel (8 circles), Tikona Digital (5 circles), Bharti (4 circles) and Qualcomm (4 circles) and Augere (Mauritius) (1 circle). Many companies like TATA and Vodafone decided to stay away from the auction and this helped in lowering the auction amount as competition remained contained within limits unless what is being said about 3G licenses. Still, the auction is expected to put intense pressure on the balance sheet of all the companies involved in the auction As the BWA revenue will take some time to show and the roll out will take time as it has to be done progressively in all the circles and will be possible only after 4-5 months of allotment of the spectrum. Source: DoT/ HSBC14 Figure 16: BWA Spectrum Winners 13 ICICI direct research report, “BWA spectrum auction”, June 2010 14 HSBC Global Research, “Indian Telecoms”, June 2010 18 | P a g e 2.4.3 Auction price comparison with 3G Although the pan-India multiple in 4G is 7.3x it is still cheaper when compared to the 3G auction that took place before 4G auction and saw intense competition from all the major players in Indian telecom industry. Pan India cost in terms of US$/ MHz/Population for BWA comes around one-fifth of the 3G thus significantly improving chances of better returns from the 4G services to be launched within one year timeframe along with 3G and will be competing with 3G in broadband area. Source: Reliance Report15 Figure 17: 3G and BWA Spectrum Auction 2.4.4 Auction price comparison - International Internationally auction method is being used since 2000 when Switzerland auctioned its 3G license. Since then various countries have adopted the route of the auction as it is expected to maximize the value of the spectrum and help respective Governments get the much needed revenue. When considered with global auction data, the prices paid for 3G as well as 4G are still very much competitive in India and so higher/ better IRR is expected from the license in India compared to most of other countries. 15 Reliance Industries Limited, “BWA – Analyst report”, June 2010 19 | P a g e 15 Source: Reliance Report Figure 18: Global trend in Spectrum auction prices 20 | P a g e 3. Historical Background 3.1 Evolution from 0G to 3G 3.1.1 0G 0G is the era of Mobile radio telephone systems which preceded the modern cellular mobile telephony technology we see nowadays. These systems are sometimes called pre-cellular and included technologies such as Push to Talk (PTT or manual), Mobile Telephone System (MTS), Improved Mobile Telephone Service (IMTS), and Advanced Mobile Telephone System (AMTS) systems.16 These mobile telephones were available as mountable on cars or trucks as well as briefcase models. The full unit consisted of a transceiver (transmitter-receiver) mounted in the vehicle trunk and attached to the "head" (dial, display, and handset) mounted near the driver seat. 3.1.2 1G 1G refers to the start of the wireless telephony technology, which is mobile technology as we know today17. They constitute the analog telecommunications standards that were introduced in the 1980s and replaced by 2G digital communications. The main differentiating factor between 1G and 2G is that the radio signals that 1G networks use are analog, while 2G networks are digital. Although both systems use digital signaling to connect the radio towers (which listen to the handsets) to the rest of the telephone system, the voice itself during a call is encoded to digital signals in 2G whereas 1G is only modulated to higher frequency, typically 150 MHz and up. 3.1.3 2G Three primary things that separated second generation 2G cellular telecom networks from previous technologies were18: Digitally encrypted phone conversations, Efficient systems that allowed greater mobile phone penetration levels and introduction of data services for mobile, starting with SMS text messages. 2G technologies can be divided into TDMA-based and 16 http://en.wikipedia.org/wiki/0G, accessed on 10/3/2011 17 http://en.wikipedia.org/wiki/1G, accessed on 10/3/2011 18 http://en.wikipedia.org/wiki/2G, accessed on 10/3/2011 21 | P a g e CDMA-based standards depending on the type of multiplexing used. TDMA based GSM and CDMA based cdmaOne is being used in India. Digital signals helped in increasing efficiency in two ways: 1) Digital voice data could be easily compressed and multiplexed and thus allowed more calls to be packed into same amount of radio bandwidth. 2) Less radio power emission from digital systems ensured that cells could be smaller and thus more could be packed in same space accompanied by less expensive equipments 2G technology provided many advantages such as Digitization which ensured that digital data services such as SMS could be started, reduced fraud as cloning of sets was not possible and enhanced privacy. Health concerns were lowered due to relatively low energy level signals used in emission. But the technology has its limitations such as: 1) Weaker digital signals in a less populated signal might not reach cell tower due to relatively low frequency at which 2G works 2) Digital signals although good overall but their performance worsens when conditions gets worse and experience drop outs. Overall 2G networks are suitable for voice services and slow data transmission and thus the technology evolved and we had GPRS (2.5G) and EDGE (2.75G): • GPRS: General Packet Radio Service, GPRS can provide data rates from 56 kbit/s up to 115 kbit/s. It can be used for services like Wireless Application Protocol access (WAP), Multimedia Messaging Data Service (MMS) as well as for the Internet for using basic services such as email and access World Wide Web. One big difference is that in GPRS, charges are decided on the basis of megabytes used instead of per minute conversation as in 2G. • EDGE: EDGE provides potential to increase capacity of GSM/ GPRS networks by 3 times. The specification can achieve data-rates up to 236.8 kbit/s. 22 | P a g e 3.1.4 3G 3G standards comprises of the standards set for mobile phones and mobile telecommunications services fulfilling specifications by the International Telecommunication Union (ITU) and are known as International Mobile Telecommunications – 2000 (IMT-2000)19. The standards include application services such as wide-area wireless voice telephone, mobile internet access, video calls and mobile TV, all in a mobile environment. The difference between 3G and the previous standards is that atleast 200 Kbits/s speed must be provided in 3G as per IMT-2000 specifications. The latest 3G standards namely 3.5G and 3.75G can be used to provide mobile broadband access of several Mbit/s to laptop computers and smart phones. Two main technologies under 3G are: • UMTS: Universal Mobile Telecommunications System, started in 2001 and standardized by 3GPP is used in today’s cell phones as hybrid between UMTS and GSM. Unlike 2G technologies such as EDGE and CDMA2000, UMTS requires setting up of new base transmission stations (BTS) and new frequency allocations. But in spite of this, it is still closely related to GSM/ EDGE as it builds upon concepts from GSM technology. Also, most UMTS handsets support GSM, allowing seamless dual-mode operation. Therefore, UMTS is sometimes marketed as 3GSM, emphasizing the close relationship with GSM and differentiating it from competing technologies. Several radio interfaces are offered, sharing the same infrastructure such as: - W-CDMA: Wideband Code Division Multiple Access, the original and most widespread radio interface. It utilizes the DS-CDMA channel access method and the FDD duplexing method to achieve higher speeds and support more users compared to most time division multiple access (TDMA) schemes used today. - HSPA+: Evolved High-Speed Packet Access is the latest UMTS technology variant, it can provide peak data rates up to 56 Mbit/s in the downlink in theory (28 Mbit/s in existing services) and 22 Mbit/s in the uplink. 19 http://en.wikipedia.org/wiki/3G, accessed on 11/3/2011 23 | P a g e • CDMA2000 system: Started in 2002 and standardized by 3GPP2.Its most used variant is the latest release EVDO that offers downstream peak rates of 14.7 Mbit/s: EVDO: Evolution-Data Optimized or Evolution-Data is a telecommunications - standard for the wireless transmission of data using radio signals, and is used typically for broadband Internet access. It uses multiplexing techniques including code division multiple access (CDMA) as well as time division multiple access (TDMA) to maximize both individual users' throughput and the overall system throughput. It is also standardized by 3GPP2 as part of the CDMA2000 family of standards and has been adopted by many mobile phone service providers around the world – particularly those previously employing CDMA networks. In India it is being provided by Reliance and TATA. Although Mobile WiMAX standards formally also fulfill the IMT-2000 requirements and are approved as 3G standards by ITU, these are typically not branded 3G, and are based on completely different technologies. The bandwidth and location information can be leveraged using 3G devices and some of the applications are: • Mobile TV – a provider redirects a TV channel directly to the subscriber's phone where it can be watched. • Video on demand – a provider sends a movie to the subscriber's phone. • Video conferencing – subscribers can see as well as talk to each other. • Tele-medicine – a medical provider monitors or provides advice to the potentially isolated subscriber. • Location-based services – a provider sends localized weather or traffic conditions to the phone, or the phone allows the subscriber to find nearby businesses or friends. Further Evolution19 Both 3GPP and 3GPP2 are currently working on extensions to 3G standard that are based on an all-IP network infrastructure and using advanced wireless technologies such as MIMO, these 24 | P a g e specifications already display features characteristic for IMT-Advanced (4G), the successor of 3G. However, falling short of the bandwidth requirements for 4G (which is 1 Gbit/s for stationary and 100 Mbit/s for mobile operation), these standards are classified as 3.9G or Pre-4G. 3GPP plans to meet the 4G goals with LTE Advanced, whereas Qualcomm has halted development of UMB in favour of the LTE family. ITU Common Data IMT-2000 name(s) Bandwidth 4G TDMA SingleCarrier EDGE EDGE Evolution (IMT-SC) Pre- CDMA2000 EVDO Channel none TDMA UMB Evolutionary FDD upgrade to cdmaOne CDMA Direct (IMT-DS) CDMA TDD upgrade to GSM/GPRS (IMT-MC) Spread Description Evolutionary CDMA MultiCarrier Duplex CDMA U W-CDMA M TD-CDMA T TD- S SCDMA HSPA Areas Worldwide except Japan and South Korea Americas, Asia Family of revolutionary worldwide standards LTE Europe (IMT-TC) China Short-range, FDMA/ TDMA (IMT-FT) DECT none TDD FDMA/ standard for Europe TDMA cordless USA phones IP-OFDMA WiMAX (IEEE 802.16) OFDMA Source: ITU 20/ WIkipedia Figure 19: Overview of 3G/IMT-2000 standards 20 http://www.itu.int/dms_pub/itu-d/opb/stg/D-STG-SG02.18-1-2006-PDF-E.pdf, accessed on 11/3/2011 25 | P a g e worldwide 3.2 4G 4G is the fourth and latest generation of telecommunications cellular wireless standards21. It is first end-to-end total IP based network and the speed requirements for 4G service set the peak download speed at 100 Mbit/s for high mobility communication (such as from trains and cars) and 1 Gbit/s for low mobility communication (such as pedestrians and stationary users).22 With 4G systems, comprehensive and all secure IP-based mobile broadband solutions are possible for mobile devices such as laptop computer wireless modems, smart phones and other mobile devices. Further access services such as ultra-broadband Internet access, IP telephony, gaming services, and streamed multimedia can be easily provided to the users. At present technologies such as Mobile WiMAX and LTE are counted among 4G technologies. 3.2.1 WiMAX WiMAX (Worldwide Interoperability for Microwave Access) is a telecommunications protocol that has been designed to enable pervasive, high-speed mobile Internet access in both fixed and mobile medium. The current WiMAX revision provides up to 40 Mbit/s with the IEEE 802.16m update expected to offer up to 1 Gbit/s fixed speeds. WiMAX is also described as "a standardsbased technology enabling the delivery of last mile wireless broadband access as an alternative to cable and DSL". WiMAX operates on the same principles as used in Wi-Fi23. It involves providing around 14 wireless transmission of data using a variety of transmission modes, from point-to-point links to portable Internet access. According to reports, WiMAX provides broadband wireless access (BWA) up to 30 miles (50 km) for fixed stations and up to 5 - 15 km for mobile stations. As the delivery mode doesn’t need line of sight presence and thus this technology is also called “Non Line of Sight Transmission”. 21 http://en.wikipedia.org/wiki/4G, accessed on 11/3/2011 22 http://www.itu.int/ITU-R/index.asp?category=information&rlink=imt-advanced&lang=en, accessed on 11/3/2011 23 IMRB, “Mobile Internet in India”, December 2009 26 | P a g e Source: Intel report24 Figure 20: WiMax Profile: Radio frequency spectrum The various types of receivers possible are fixed-wireless type, Nomadic type and Mobile type. Due to the bandwidth and range of the WiMAX products, various applications are possible such as: • Portable mobile broadband connectivity spanning across various cities and countries through a mix of devices: Using WiMAX companies have started providing alternate to incumbent broadband providers and the increased competition will lead to further reduced ARPUs and thus will benefit customers. WiMAX is a possible replacement candidate for cellular phone technologies such as GSM and CDMA, or can be used as an overlay to increase capacity and backhauling. At present copper wire, satellites and microwave links are used for backhauling which can be easily replaced by cost effective WiMAX. One unique use of WiMAX was found during disasters such as Katrina and Tsunami when WiMAX acted as the only viable medium for the communication as other mediums couldn’t be used and thus it is expected to form a backup network in case of emergency in near future • Better wireless alternative to cable and DSL for "last mile" broadband access especially in inaccessible and rural areas: The relatively cost of deploying a WiMAX network (in comparison to GSM, DSL or Fiber-Optic) is low and thus it is now possible to provide broadband in places where it might have been previously economically unviable 24 Intel Corporation, “Internet everywhere – 4G technology”, April 2010 27 | P a g e • Data and telecommunications such as Voice over Internet Protocol (VoIP) and Internet Protocol Television (IPTV services) which is also known as triple play. Availability of option to combine various services in single medium will open path for partnerships between various wireless and cable companies and benefit all partners involved. • As a source of Internet connectivity as part of a business continuity plan 3.2.2 LTE Long Term Evolution (LTE) is the latest standard in the mobile communications technology branch that produced the GSM/ EDGE and UMTS/HSPA network technologies25. The LTE specification provides downlink peak rates of at least 100 Mbps, an uplink of at least 50 Mbps and round-trip times of less than 10 ms. LTE supports scalable carrier bandwidths, from 1.4 MHz to 20 MHz and supports both frequency division duplexing (FDD) and time division duplexing (TDD). LTE is also an IP based network architecture designed to replace the GPRS Core network and ensure support for and the mobility between different networks for example GPRS and WiMAX. The main advantages with LTE are high throughput, low latency, plug and play, combination of FDD and TDD in the same platform, an improved end-user experience and a simple architecture resulting in low operating costs. LTE will also support seamless passing to cell towers with older network technology such as GSM, cdmaOne, UMTS, and CDMA2000. Source: Intel report24 Figure 21: LTE profile: Radio frequency spectrum 25 http://en.wikipedia.org/wiki/3GPP_Long_Term_Evolution, accessed on 11/3/2011 28 | P a g e 3.2.3 4G versus rest technologies 4G technologies are at the frontend of the evolution process that started with 0G and still continuing. With the evolution technologies have become more mobile and flexible allowing for the newer applications such as better backhauling, triple play and faster broadband. One of the important contributing factors has been the increasing bits that can be packed per Hertz and thus deriving more data in the limited spectrum available as can be seen below: Source: Reliance report15 Figure 22: Mobile technology- Bits per Hertz Increasing speed is in turn allowing users to move towards more complex and data intensive applications and unbundling the power of the hardware and software technology. Source: Reliance Report15 Figure 23: Mobile technology evolution 29 | P a g e A more comprehensive comparison is shown below: Standard WiMAX advance LTE - advance Family 802.16m UMTS/4G SM WiMAX LTE 802.16 UMTS/4G SM Wi-Fi 802.11 EDGE UMTS W-CDMA HSDPA+ GSM UMTS/3G SM HSPA+ UMTS-TDD EVDO 1x Rev. 0 EVDO 1x Rev. A Primary Use Download (Mbits/s) Uplink (Mbits/s) Mobile 1296 (in 20MHz 100 (in 20MHz Internet bandwidth) bandwidth) 1296 (in 20MHz 100 (in 20 MHz bandwidth) bandwidth) Mobile 128 (in 20MHz 56 (in 20MHz Internet bandwidth) bandwidth) 100 (in 20MHz 50 (in 20 MHz bandwidth) bandwidth) General 4G General 4G Mobile Internet Mobile Internet General 3G UMTS/3G Mobile SM Internet CDMA20 Mobile 00 Internet EVDO Rev. B 300 (in 4x4 configuration in 20MHz bandwidth) or 600 (in 4x4 configuration in 40MHz bandwidth) 1.6 0.5 0.384 0.384 14.4 5.76 56 22 16 2.45 0.15 3.1 1.8 4.9xN 1.8xN Source: Wikipedia Figure 24: Comparison of Mobile Internet Access methods WiMAX provides lot of flexibility in terms of the speed provided when compared along with the mobility provided. In all the digital technologies, with the mobility the speed decreases as bit error rate starts increasing but WiMAX still provides better output when compared to most of the available technologies. 30 | P a g e Source: Wikipedia, Copyright Benjamin M. A'Lee Figure 25: Speed versus Mobility of wireless technologies 3.2.4 Comparison with Wi-Fi Both technologies are sometimes confused due to the similarity in the function and as both are used to give wireless broadband connections but both have distinctive properties that differentiate them and give them their own unique usage as shown below: WiMAX Range License Users Long range system, covering many kms Uses licensed or unlicensed spectrum Wi-Fi provides access to a local network Unlicensed spectrum only More popular with the business user More popular in end user devices Connection-oriented connectionless and contention based Based on connections between the Contention access based - all subscriber base station and the user device. stations that wish to pass data through a Each connection is based on specific wireless access point (AP) are competing scheduling algorithms for the AP's attention on a random Type of Media Access Control QoS 31 | P a g e interrupt basis. Networks Peer-to-Peer (P2P) and ad hoc Peer-to-Peer (P2P) and ad hoc networks, networks, but end user devices must and direct ad hoc or peer to peer be in range of the base station. networking between end user devices without an access point is also allowed Even though both technologies are adept for different scenarios they are still used together by some providers. WiMAX network operators typically provide a WiMAX Subscriber Unit which connects to the metropolitan WiMAX network and provides Wi-Fi within the home or business for local devices (e.g., Laptops, Wi-Fi Handsets, smart phones) for connectivity. This enables the user to place the WiMAX Subscriber Unit in the best reception area (such as a window), and still be able to use the WiMAX network from any place within their residence Source: Wikipedia Figure 26: WiMAX and Wi-Fi comparison 3.3 Typical 4G network A typical 4G broadband network consists of a tower which is also called BTS, which is Base Transmission Station which houses all the required equipments for the transmission of the frequencies to make WiMAX available. There can be different types of the end users such as Fixed Broadband users where WiMAX usually complements DSL and cable broadband, Mobile laptop broadband users where WiMAX competes with WiFi/ EDGE & 3G, Mobile cellphone broadband users where the usual competitors are EDGE and 3G and the last but rising use is in backhauling to connect to other networks. 32 | P a g e Source: Intel report24 Figure 27: Typical 4G network 3.4 4G success deployment worldwide Since 2008, 4G networks have come to existence in various parts of the world with various degrees of successes and issues. Source: Protiviti Research45/ Reliance report15 Figure 28: BWA success case studies 33 | P a g e 4G is being advertised as unique and history making event with major stress on the speed and flexibility provided by the same. Source: Intel report24 Figure 29: 4G advertised 3.5 Future Trends According to the IMRB Research23 the future trends which are expected to be observed for Mobile internet in India due to advancements in the devices, technologies and the content delivery are as following: • Ubiquitous Connectivity: The rise in the demand for the connectivity from the enterprises and consumers will lead to the connectivity among the various platforms. Ubiquitous connectivity designates a seamlessly integrated platform for interactions and collaboration across diverse and global communication environments. This will enable users to access and share the content over various platforms with complete mobility. • Fixed Mobile Convergence (FMC): Fixed mobile convergence provides way to integrate the wire line and wireless technologies and services to create a single telecommunications network. FMC promises to overcome the physical barriers that prevent the telecom service providers from targeting the potential customers with all types of services. This initiative will 34 | P a g e offer the opportunities to the wire line service providers to not remain tethered to the landline networks and wireless operators will be able to utilize the resources to satisfy the growing demand of mobile subscribers. This initiative will lead to seamlessly connectivity for all devices, access points, applications and underlying networks, to deliver an enhanced user experience. • Quadruple Play: The transformation of telecommunication technologies will enable the triple play service of voice, data, and video with the wireless service i.e. mobility. The implementation of this quadruple play services has restricted to the some of the countries of the world. With the improvement in the technologies, it has been expected that users will be able to access these uninterrupted services even on the move. • Mobile Advertising: Mobile as a device offer lot of opportunities for the low PC but high mobile phone penetrated Indian market. Considering the success of mobile advertising in some of the countries globally, relevant formats of the mobile advertising will be introduced in the Indian market. • Unified Messaging Service (UMS): Traditional communications systems delivered messages into several different types of stores as per the type of message. Unified Messaging Service will help in bringing together all messaging media such as voice messaging, SMS, email into a single interface which is accessible from variety of devices. Based on the functionalities offered by this system, this service it will be more helpful for the enterprises compared to the users. 35 | P a g e 4. Literature Review With time, telecom industry has been overseeing various changes and the industry is becoming more and more dynamic. Ever evolving technology which is gaining pace day by day, worldwide auctions and intense competition has ensured that it is not only difficult but impossible to evaluate future potential of investments in the sector using traditional methods such as Discounted Cash Flow (DCF). With various options available now with the telecom company a more and suitable method is using option method for valuation. In a paper, Alleman (2002)26 describes real option and associated benefits in using the approach for telecommunication projects. He mentions that in traditional DCF method the valuation is static in nature as it doesn’t take into consideration the managerial flexibility such as option to delay, option to expand or option to abandon, option to contract etc. Traditional capital budgeting fails to account for this flexibility and fails to integrate with strategic planning. According to the author, real options are relevant to telecommunications in several areas such as: Strategic evaluation, estimation and cost modeling. The major planning in strategy area has been limited to the budget estimation and scenario analysis based on mainly DCF. This was era of regulatory strategy but with changing times price elasticity, uncertainty and other economic considerations have taken centre stage and thus real option approach is the need of hour. According to author, many behavioral assumptions that are taken as embedded in econometric structures and are necessary for estimations have been invalidated by real option approach and should be adopted. Traditionally forward looking costs methods used around the world have been base on cost models whose foundation are traditionally applied to DCF but can be easily adapted to real option models. But author cautions against adding the real option result linearly to the DCF result. Taking the argument forward and expanding it further is the paper by Mastroeni and Naldi (2009)27. Authors took into consideration, side obligations and commitments that come bundled with licensee and the restriction on the transfer of the license. This gives rise to the 26 Alleman, James, “A new view of telecommunications economics”, Telecommunications Policy, Vol. 26, pp. 87- 92, 2002 27 Mastroeni, Loretta, Maurizio Naldi, “A real option model for the transferability value of telecommunications licenses”, Annals of Telecommunications, Vol. 65, No. 3-4, pp. 201-210, 2009 36 | P a g e transferability option value and add to the existing option value. Authors applied various settings related to the reselling price, expiry time of the option and the variability of the telecommunications market and found that reselling price and the volatility plays the most crucial role and the option’s value increases superlinearly with an increase either in the reselling price or in the volatility. Also option value was found to grow no more than linearly with the expiry time and the growth rate gets lower as the reselling price increased. Authors also found out that to assess the value of the option when the side conditions are known (or at least estimated), the method to negotiate the side conditions should also be used. The two parameters that are in control, that are reselling price and the expiry time should form the part of the license itself or the contact and the third parameter, that is volatility may be left out as it is not in the control. These parameters can then be tuned to reconcile the price of the option with its exercise conditions. Figure 30: Impact of the expiry time on the Figure 31: Impact of the license reselling price Figure 32: Impact of the expiry time on the reselling Figure 33: Impact of the volatility on the reselling option’s value option’s value 37 | P a g e There have been many studies of previous auctions mainly of 3G auction wherein real option approach has been used along with traditional DCF valuation approach. In an article Basili and Fontini (2003)28 tried to evaluate if too much was paid for the 3G license. They calculated aggregate option value of the UK 3G telecom licenses using the expected revenues from the 3G business, capital expenditure to be incurred in setting up of the infrastructure and the operating costs in maintaining the operations for the period of the license. They found the aggregate revenue extracted by the UK Government to be slightly less than the aggregate price of the license paid by the telecom companies who won the licenses. The justification for selecting UK 3G auction for analysis was given that UK auction has been found to be most efficient in the article by Binmore and Klemperer (2002)29. But after two years of the auctions, doubts were casted on the efficiency of auction as the telecom companies saw fall in share prices and unsustainable burden of debt which nearly undermined the other business lines of these companies. The article showed that the value paid by telecom operators is slightly less than the expected total value from the license and thus in the long run any lowering in the performance of the operators should not be due to the high license fees paid but due to incapability of the operators to give “killer applications” to the users in order to attract them, technical difficulties in providing full capability networks at all places combined with world-wide economic downturn resulting in less than forecasted customer base. The article ends with a positive note that is full time horizon equal to the license duration is taken into consideration than the licenses will yield good returns for all telecoms. In another paper by Stille et all (2010)30, authors analyzed the contribution of the real option method to the decision making process in the telecommunications industry and selected 3G auction prices in Brazil. Considering only traditional method of discounted cash flow showed 28 Basili, Marcello, Fontini, Fulvio, "The option value of the UK 3G Telecom licenses: Was too much paid?", info, Vol. 5 Iss: 3, pp.48 – 52, 2003 29 Binmore, K. and Klemperer, P., “The biggest auction ever: the sale of the British 3G telecom licenses”, Economic Journal, 2002 30 Stille, R., Lemme, C., Brand , L, “An application of Options of the Royal Evaluation of the License of the benefits system 3G Mobile Telephony Services in Brazil”, Journal of Finance, Rio de Janeiro, Vol 8, No. 3, 329-349, 2010 38 | P a g e telecoms to be at loss but when coupled with the value due to the option embedded with the license, they found 64% better value and thus making the license profitable in the long run. The value in using real option approach lies in the presence of the high degree of uncertainty in implementation of a project such as 3G, managerial flexibility as well as high degree of exclusivity or barrier to entry. The high degree of uncertainty exists due to various reasons such as uncertain/ untested new technology, uncertainty about services and volume of subscribers/ users, uncertainty about taste and preferences of customers in future as well as socio-economic factors present in the market such as income levels, education levels and the age, all of which are out of control of the company. Managerial flexibility is derived from the option to buy the license or delay it or simply not take license, option to invest in the network for new technology and the option to select the cities or services that manager will like to give. Manager can decide the parameters of the network in terms of the scale and capability according to its expectations about the market. There can be rules and regulations in place that might limit this flexibility to some extent. But flexibility in terms of network capacity such as density in selected areas, choice of technology etc is still in hands of the managers. The third and last value addition to the option comes from the exclusivity in terms of holder of the exclusive spectrum which creates barrier to entry and thus restricts competition. In another paper, Moja and Mkhize (2009)31 examine applicability of real option valuation techniques by cellular telecommunication operators in South Africa when making capital investment decisions in next-generation service orientated architectures. They used both BlackScholes and Binomial models to examine their effectiveness in valuing capital investments within a cellular telecommunication industry in South Africa. Results show that real option valuation techniques are effective in analysing investments in cellular telecommunication industry. Their strengths are mostly demonstrated when determining the value of strategic options that are added to traditional (base-case) net present value. 31 Moja, N., Mkhize, M., “The application of real option valuation techniques in the cellular telecommunication industry in South Africa”, South Africa Journal of Business Management, Vol. 40, No. 3, 2009 39 | P a g e Source: Mun, 200632 Figure 34: Traditional versus new analytics Authors further suggest using the expanded NPV (extended DCF) method for valuing the optioninclusive value of the project. The expanded NPV method is defined as the sum of the traditional NPV and the expected value of future projects made possible by the initial investment. In other words, the expanded NPV for IT projects comprises the sum of traditional NPV (obtained through DCF) and the value of embedded options provided by the initial infrastructure or technology investment. The embedded options provide management with strategic flexibility for future project expansion, deferral or abandonment. Follow-on projects in the form of compound options (options-on-options) are also possible. The similar analysis is possible for real option approach to the telecommunications industry especially for the spectrum licenses as it provides various flexibilities when it comes to the implementation part and the time and method of use of the spectrum license. 32 Mun, J. 2006. Real options analysis – tools and techniques for valuing strategic investments and decisions. 2nd Edition. New Jersey: Wiley Finance. 40 | P a g e Traditional NPV from direct benefits Possible followon projects Possible followon projects IT Expanded NPV Implementation Option value from future investment opportunities Magerial flexibility Mangerial flexibility in decision-making on implementing follow-on projects Deferral Abandonment Source: Dai, Kauffman & March (2000)33 Figure 35: Option-inclusive value of an IT project A more illustrative approach was given by authors showing various aspects of the real option: Source: Damodaran (2001)34 Figure 36: Illustration of an option pricing model 33 Dai, Q., Kauffman, R. J. & March, S. T., ‘Analyzing investments in object-oriented middleware: An options perspective.’ In Proceedings of MIS Research Center Working Papers Carlson School of Management, Minneapolis, University of Minnesota, p.p.1–27., 2000 34 Damodaran, A., “The dark side of valuation”, New Jersey: Prentice Hall, 2001 41 | P a g e Another attempt to valuate telecommunications technology, specifically UMTS was made by Herbst and Walz (2001)35. Authors adopted real options approach to analyze the value of auctioned UMTS-licenses at that time, focussing on Germany as the largest European market. They developed a real options model with abandonment as well as a growth option. Due to lack of concrete forecast, they suggested and used an indirect approach by assuming a stochastic process for the number of mobile phone users in Germany. They further used numerical analysis rather than on closed form solutions in valuing flexibility inherent in the UMTS investments. On the basis of a sensitivity analysis they found out that the initial customer base of a mobile phone company and the realized net cash flows per user are the two most important and crucial parameters in establishing the value of the option. Also they found that the results will be different for incumbent and new standalone player as it will be difficult for a new player to be competitive to the incumbent players and will have to ramp up the subscribers fast in order to catch up. Real option approach has been used for areas other than telecommunications such as for valuating Intellectual Property rights such as by Chang et al (2005)36. Real option approach to valuate Intellectual Property rights is an acceptable methodology but as pointed out by the authors the assumption of the constant rate-of-return might not be always true and thus there is need to vary volatility with time in either direction that is increasing or decreasing with time. Authors in the paper incorporated a sensitivity variable to account for the volatility of the expected rate of return and modified it with time. Other major findings were that Vega will be negative when option is deep-in-the-money. If the rate-of-return shortfall is variable and increases with volatility, option value would have a negative relation with volatility. 35 Herbst, Patrick, Walz, Uwe, “Real Options Valuation of Highly Uncertain Investments: Are UMTS-Licenses Worth their Money?”, Department of Economics, University of Tuebingen, 2001 36 Chang, Jow-Ran, Hung, Mao-Wei, Tsai,Feng-Tse, “Valuation of Intellectual Property- a real option approach”, Journal of Intellectual Capital, Vol.6, No.3, pp. 339-357, 2005 42 | P a g e Application of real option in yet another field was given by Benaroch and Kauffman (1999)37 in their paper on IT investment decisions making. Authors pointed to the lack of authoritive and concrete application of real option valuation to IT projects and provided three insights: provides a formal theoretical grounding for the validity of the Black-Scholes option pricing model in the context of the spectrum of capital budgeting methods that might be employed to assess IT investments, showed why the assumptions of both the Black-Scholes and the binomial option pricing models place constraints on the range of IT investment situations that one can evaluate that are similar to those implied by traditional capital budgeting methods such as discounted cash flow analysis and presented the first application of the Black-Scholes model that uses a real world business situation involving IT as its test bed. They implemented real option on a timing analysis of the deployment of point-of-sale (POS) debit services by the Yankee 24 shared electronic banking network of New England. They found that projects which involve infrastructure development and wait-and-see deployment opportunities should be evaluated using real option approach as it can handle timing issues, scaling or even abandonment as companies gets to know about the business environment with time only. The main issues that remains with using the real option pricing is the restrictions on using the log-normality of the perceives value of the IT project and the unavailability of the information to calculate the variance of the returns from the project. Authors classified IT projects in various types such as IT infrastructure investments (investments made without expectation of immediate payback but forms the basis of the follow on investments that will convert investment opportunities into the option’s underlying asset, for e.g. intranet and multi-media user interface technologies, financial and operational risk management technologies and security safeguards, data warehousing etc), Emerging technology investments (projects with uncertain cost, adoption and diffusion, the value of the underlying asset is subject to both changing expectations of the future costs on the part of the analyst and the market at large. In this case, the impact of stochastic cost (uncertain exercise price) drives the use of option pricing. For e.g. Internet advertising and selling, migration to an electronic market mechanism for transacting etc. In all these case the future cost attached with the exercising an option to build on a network, a market mechanism or a standard, is unknown today), Application 37 Michel, Benaroch, Kauffman, Robert J., “A case for using real options pricing analysis to evaluate information technology project investments”, Information Systems Research, Vol. 10, No. 1, pp. 70-86, 1999 43 | P a g e design prototyping investments (With prototyping, the firm aims to maximize the value of an application development project whose value will ultimately be determined by how well its functionality can remain in synch with the support needs of a changing business process. The value inherent in the underlying asset is of somewhat less interest to the firm than the ability to react: to both adapt and change the application’s functionality as required to remain competitive. When there is considerable uncertainty in an organization about whether an application will be able to “do the job” when it is delivered, or there is risk aversion on the part of management in making capital investments in IT, efforts to stage or “chunk” such projects, and monitor their payback over time, is an appropriate approach. From this perspective, much of the value of a prototype project will be in the options that it offers the firm in the future) and last being Technology-as-product investments (When the technology is a core part of a product, issues of level of commitment and ramp up, timing and roll out, and delay and abandonment must be considered. Here the benefit is derived from framing such choices in the context of option pricing by focusing on such elements as time remaining to exercise, when the option matures and by tracking the value of the option to change the course of a project. For example, Otis Elevator and the decision to re-capture the after-market for its elevator servicing, Chemical Bank’s failure with the Pronto home banking project, Morgan Bank’s success with RiskMetrics for financial risk management in international commercial banking, and First Boston Corporation’s decision to create products and a new company, Seer Technologies, from what had been a major systems infrastructure building project etc.). Authors have further argued that option pricing models can be applied to capital budgeting decisions involving non-traded information technology assets. 44 | P a g e 5. Methodology This research calculates the aggregate value of the 4G spectrum licenses auctioned in India in financial year 2010-2011. In order to calculate the aggregate value, 4G spectrum license has been treated as a project. The methodology adopted is as defined below: Discounted Cash Flow (DCF) valuation Sensitivity Analysis Real Option valuation Figure 37: Steps in valuation Step1: DCF valuation Discounted Cash Flow method has been used to evaluate the Net Present Value (NPV) of the expected cash flows/ earnings from the deployment of the spectrum auctioned for implementation of 4G/BWA technology. The calculations and the forecasts of various parameters done in DCF will serve as base data for the next steps of valuation, which is using Real Option. The assumptions, projections and forecasts used in the DCF valuation are explained in next chapter on data. Step2: Sensitivity Analysis • Sensitivity analysis has been done on the NPV and IRR obtained from the DCF valuation model in order to analyze the risk associated with the project and obtain the safe range in which the project can be executed profitably. Step3: Real Option valuation • Real option valuation has been used to evaluate the optionality available with the Telecom operators who have won spectrum licenses. The nature of optionality has been identified and Black schools formula used to evaluate the optionality. 45 | P a g e 5.1 Discounted Cash Flow Valuation The two standard calculations covered under DCF valuation are presented below: • Net Present Value Method: Under this method present value of the company is calculated by discounting the cash flows expected from the project at a risk adjusted rate of return. NPV is the difference of the present value of cash inflows from the project and the present value of the investment overlay that will go into the project. Here Investment overlay includes capital expenditure as well as license costs. 1 1 Where, Cin = Cash inflow in period ’n’ Con = Cash outflow in period ’n’ T = Life of the project R = Risk adjusted discount rate of the project The above equation can be re-written as, Where, S = Net present value of the cash inflows from the project K = Net present value of the investment overlay for the project The decision criterion for selecting a project based on NPV method is, NPV > 0: Select NPV <= 0: Reject • Internal Rate of Return: Internal rate of return is that discount rate at which NPV of the project becomes zero. That is it is the rate of return generated by the project and thus gives a clear cut-off rate. The equation for IRR is given by 46 | P a g e 0 1 Where, T = Life of the project Cn = Net Cash flow in period ’n’ inclusive of investment outlays IRR = Internal Rate of Return (Discount rate) The decision criterion for selecting project based on IRR is, IRR > R: Select IRR <= R: Reject Where, IRR = Internal Rate of Return R = Risk adjusted rate of return required from the project Discounted Cash Flow method gives the static value of the project as seen in the Literature review and thus is apt for the project where there is no managerial flexibility but as we discussed earlier, in case of spectrum and subsequent deployment of the 4G technology, managers have ample flexibility such as to delay, abandon or expand and thus DCF alone is not sufficient to calculate the exact value of the spectrum and thus we will proceed to the Real Option 5.2 Real Option Valuation For calculating the value the Black-Scholes model given by Damodaran (2000)38 has been used which is defined as below: ! &' . ln - . &* 38 #$% . /01 &' . &' 3√ 2 #() % . &* 3*4 3√ Damodaran, Aswath, "The option of Real option", Journal of Applied Corporate Finance, Vol. 13, No. 2, pp. 29– 44, 2000 47 | P a g e Where, Parameter S q In context of Real Option Present Value of cash flows expected from the project Opportunity Cost of not Expanding (Explained Below) In context of Financial Option Stock Price Dividend Yield Expected Competitive T Advantage Period/Rights for Time to Expiration Expansion rf Risk free rate of return on 10 year GOI bond Risk free rate of return K Present Value of Capital Costs Strike Price (σ)2 Volatility of Project Value Volatility of Return on Stock Figure 38: Real option parameters N(d1) & N(d2) are normal cumulative distributions function which gives the range of the likelihood of the real option viability before expiration date, T. With regard to 4G spectrum auction, there are two types of players, incumbents who are already providing broadband service either as DSL or cable or other wireless technologies or standalone players who will be starting to provide broadband services in the country. As per the rules and regulations of the license, the players have an option to provide service to upto 90% population in Urban areas and upto 50% Rural by 5th year and later on in modular approach to other areas. Basili & Fontini (2003) in their paper stated that when an operator has the flexibility of implementation of technologies as a segmental process then it has implicit flexibility to launch new services in a sequential and discrete manner. Operator thus also has the flexibility to start off the rollout by a pilot project and then scale up the project according to the need and market response. 48 | P a g e The various parameters in the Real Option are as below: • S = The Present value of cash flows is calculated from the assumptions a will be explained in the next chapter. This is the output from DCF model. • K = The Present value of capital expenditure required to rollout 4G/ Broadband wireless access services. This value is also explained in the next section. • t = This is the time period over which the option may be exercised lest the telecom operators will lose competitive edge in the form of lost exclusivity or due to the rules and regulation decided by the concerned Government. In case of 4G/BWA auctions in India, 5 years from the time of allotment of the license has been fixed as the deadline for the partial rollout. • rf = This is the Risk free rate on 10 year GOI. The yield on 10 year GOI bond is 7.94% as of March39. • σ = The volatility of project returns. The exact volatility of the expected project returns is usually done using the Monte-Carlo analysis with relevant probability distributions of the input variables. In the absence of information about the relevant probability distribution, annualized standard deviation of returns of Bombay Stock Exchange Technology, Media and Telecom Index (BSE TECK) has been used as proxy which is equal to ~27.5%. It can be argued that the volatility of this project can be different from the one of the Index/ Industry but as standard industry practice this volatility value has been used. • q = This is the opportunity cost of waiting and not rolling out 4G/BWA services. It is difficult to predict the exact pattern of cash flows to the firm as it depends on many external and internal factors that can vary differently from the assumptions made. Thus the exact loss by waiting to roll out 4G/BWA services is difficult to determine and hence it has been assumed that telecom operator plans to extract equal dividend yield = 1/T, that is 6.66% (T=15 years for 4G spectrum license) . That is telecom operator will be extracting 6.66% of the present value of the expected returns each year. This will be then the loss in case operator chooses not to deploy the network and wait. 39 http://www.tradingeconomics.com/Economics/Government-Bond-Yield.aspx?Symbol=INR, accessed on 10-03- 2011 49 | P a g e 6. Data & Assumptions In this chapter various assumptions have been explained along with their source. Also the reason behind the projections/ forecasts has been explained. The data discussed here forms the basis for the Discounted Cash Flow analysis later. 6.1 Population & Population Growth Rate India’s population was 1.15 Bn in 200940 and population growth has been declining from 1.7% in 2000 to 1.3% in 200941. A population growth rate of 1.5% has been assumed for the next 15 years over the base of 2009 population. This growth rate assumption is justified by the planning commission’s working paper which projects India’s population growth from 1.51% to 1.39%42. 6.2 Revenue Projections Revenue projects can be done by separately projecting various drivers of the revenue. In telecommunications industry revenue is given by: 6 ! 7 8 9 !:; 0 < !;=>0 ; 0= In this case subscribers are WiMAX subscribers exclusively but to arrive at the WiMAX subscriber figures, projections related to total broadband users, wireless and wireline mix in broadband and the proportion of WiMAX in wireless needs to be projected as can be seen below: 6 ! 7 8?( @AB@ A 9 !;=C DEF Where, !;=C DEF !;=L (IJIKK %C DEF H !;=C (IJIKK %L (IJIKK H !;=?( @AB@ A Thus we need to project four variables, namely ARPU from broadband, Total Broadband subscribers, % of broadband subscribers who will be wireless and percentage of wireless subscribers who will have WiMAX connection. 40 http://www.google.com/publicdata, accessed, on 10-03-2011 41 http://www.google.com/publicdata, accessed on 10-03-2011 42 http://www.planningcommission.nic.in/reports/wrkpapers/wp_hwpaper.pdf, accessed on 10-03-2011 50 | P a g e 6.3 Internet and Broadband Subscribers As shown before and repeated here for clarity, India is experiencing growth in broadband connections and the broadband connection percentage among the Total Internet subscribers is increasing at an increasing rate. This trend in fall in non-broadband internet and rise in wireline and wireless broadband subscribers is expected to continue as per the trend seen in other nations. Total Internet Subscribers (mn) 12 10 8 6 Non-broadband 4 wireline broadband wireless broadband 2 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 0 Source: CMIS Database Figure 39: Historical Internet Subscribers breakup Now the forecast for the broadband penetration and subscribers from year 2011 onwards was done. For this various research reports and article were consulted. The natural way to forecast a technology in a particular region is to compare the behavior shown to previous technology in the same region when compared to other regions as well as the behavior shown by the new technology in other regions. The forecast was done in two parts: 1) Through economic data and trend analysis 2) Through the forecasts by research reports Reconciliation was done between data from both types of sources and the practical estimates were selected in the end. First the mobile tele-density trend was compared in India with other developed nations and the results are as shown below: 51 | P a g e Source: OECD Datab abase Fig Figure 40: Historical Mobile tele-density It was found that the tele-density ity in India is catching up fast and is taking the same sa trend as was seen in developed countries such ch as US and UK around 8-10 years back. That is the acceptability of the mobile technology has bee been similar even though with a lag of some year ears due to various reasons such as unavailability of affordable connectivity and technology. A view iew of mobile teledensity between US and India is aas shown below: Year 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 US 0.20 0.25 0.31 0.39 0.43 0.51 0.55 0.63 0.72 0.81 0.87 .00 .01 .01 .03 .05 .08 .15 .21 India 2008 2009 2010 .30 .37 .50 Figure 41:: Mobile tele-density comparison: US and India Next the broadband tele-density ty w was compared and the trend was found to be similar, si that is the tele-density of broadband in India dia is catching up fast with that of developed natio ations and the trend is taking similar path as we saw w iin developed nations and on similar lines as mo obile tele-density. The tele-density of broadband is sstill very low < 1% and thus it is expected to increase inc at fast rate considering the auction of 3G and 4G spectrum. 52 | P a g e Source: OECD Da Database Figur gure 42: Historical broadband tele-density Next the rate of change of broadb dband subscribers was considered and we found d a general trend in all countries under analysis andd w with initial high increase in the subscriber the rate rat of addition tampered off. This is as expected ed as with more and more subscribers added, with ith time saturation is reached and unless a breakthrou rough technology comes the net additions keepss falling fa with time. The trend seen is as shown below ow: Source: OECD D Database Figure 43 43: Historical broadband subscribers % change Considering the facts mentioned ed aabove, the rate of growth of the broadband subs bscribers was forecasted on similar rate to thee oone seen in the developed countries and is as giv iven below: 53 | P a g e Year US % 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 100.18 54.96 44.61 34.63 29.78 25.10 15.90 10.17 8 2 2011E 2012E 2013E 2014E 2015E 2016E 2017E 2018E 2019E 2020E 2021E 2022E 2023E 2024E 2025E 125.00 104.52 84.07 42.67 34.77 30.00 25.00 16.00 10.00 8.00 2.00 1.50 1.50 1.50 1.50 (historical) Year India % (Forecast) Figure 44: Broadband subscribers growth rate forecast The rate of growth has been selected similar to the ones seen in developed nations. As the data is available till 2010 only and hence we got rate till 2021, that is for next 10 years and for the rest 5 years it has been assumed equal to the population growth rate expected in that period. Next various Research report were consulted with an authoritative report by Aircel. This report assumes great significance as Aircel participated in Indian auction process and has won spectrum licenses in both 3G as well as 4G. The Aircel report43 also tries to forecast the subscribers based on similar assumptions as shown earlier. In the report they predicted broadband subscribers starting 2009 till 2015. The broadband (wireline) subscribers have been predicted as 7.7mn in 2009 increasing to 10.9mn in 2010 and 69.4mn by 2015. The actual broadband subscribers (wireline) seen has been nearly 80% of the predicted ones, 6.22mn in 2009 and 8.7mn in 2010 and thus the rest of predictions need to be adjusted by 80% factor and the same adjustment factor has been retained on out total forecast of the broadband subscribers till 2025. In another report by TRAI, the estimate has been given for the total broadband users as 75mn and 160mn by year 2012 and 2014 respectively. The figures forecasted does reach the target by 2014 but misses the target by 2012 considering very low penetration at present and the previous forecast by TRAI also missed the actual subscribers in 2010 by high margins. Hence in long run forecasted values are concurrent with TRAI forecasts. Another report by Telcordia44 forecasts total broadband users by 100mn by 2013 which looks achievable by the forecasts. A report by consulting firm Protiviti45 again forecasts 100mn subscribers but by 2012, which definitely is not feasible considering the present scenario and the mark will be missed by at least 2 more years. 43 44 45 Aircel Report, “Recent trends in mobile data needs”, 2010 http://www.telcordia.com/news_events/media_coverage/express-computers-nov2010.pdf, accessed on 1/3/2011 Protiviti, “WIMAX: The Quintessential answer to broadband in India”, 2009 54 | P a g e 6.4 Broadband Penetratio etration The broadband tele-density for forecasted on the basis of the forecasted grow rowth rate of the broadband subscribers comes aagain similar to the developed countries. The he tele-density is forecasted to be around 2% byy 22011 and increase to 29% by 2020 and remain ain nearly constant thereafter due to the constraints ints in terms of the per-capita earnings and demographics. dem The forecasted tele-density of broadb dband subscribers can be seen below: Figu igure 45: Broadband penetration forecast This is again as per the broadband and penetration that has been reached by 2010 in various v countries that varies from 23% to 30%. W With a delay of 8-10 years the same penetration ion will be seen in India in terms of broadband. 6.5 Wireline and wireless ireless broadband Worldwide the majority of bro broadband users are either on DSL or cable, e, that is wireline broadband as they have develope ped and sophisticated network already present.. In India and other developing countries the situation tion is different for example as in case of India most mo of the wiring for wireline telephone cannot su support broadband speed. Only fraction of thee wiring w that does support broadband speed is main ainly in urban areas that’s too in Circle A. More oreover the cost of laying wire and giving access us using wireline broadband is costly compared to providing p access via wireless. As shown in HSBC BC report46, there is not only cost differential between be providing 46 HSBC Report, “Indian Telecom”, 201 010 55 | P a g e new wireline connection and wireless connection but also huge difference in ARPU in favor of wireless subscriber and hence any company will be better off giving wireless connection rather than wireline connection. Hence in future most of the growth in broadband is expected to be via wireless. In India, wireless broadband is being provided only since 2009 and so there is not much historical trend. Also as in developed countries wireless subscribers forms negligible portion and hence there is no historical data to rely on. As we can see since introduction of wireless broadband mainly by TATA and Reliance in un-auctioned spectrum the percentage of wireless broadband subscribers among total broadband subscribers is increasing fast. The data for wireless subscribers has been taken from Reliance report47. Breakup Broadband subscribers 100% 80% 60% Wireless Broadband % 40% Wireline Broadband % 20% 0% Source: Aircel Report Figure 46: Breakup - Broadband subscribers This trend is forecasted to continue and the proportion of the wireless broadband subscribers will increase as the marginal addition of new broadband subscribers will be more in wireless than in wireline. According to Aircel’s projections the percentage of wireline and wireless are projected to change as given below: Year L+2 L+3 L+4 L+5 L+6 Wireline Broadband Subscribers 0.58 0.46 0.37 0.33 0.31 Wireless Broadband subscribers Source: Aircel Report 0.42 0.54 0.63 0.67 0.69 Figure 47: Breakup broadband 47 http://www.ril.com/rportal1/DownloadLibUploads/1276353679981_FPR120610.pdf, accessed on 1/3/2011 56 | P a g e Where ‘L’ is the year when wirel ireless broadband will be made available for subsc scription. Thus the report predicts a steady changee in the composition of the broadband users. Thee same s pattern has been followed and the change ha has been made constant at 30-70% breakup as even ev in long run it will be difficult to exactly have ve everyone on wireless technology due to the he high ARPU for wireless and hence keeping thee eestimates on the conservative side the proportion tion of the wireline and wireless has been freeze att 20 2016 as shown below: Figur gure 48: Forecasted breakup – Broadband 6.6 WiMAX Broadband and sub subscribers Next WiMAX subscribers needd tto be forecasted. Again there is similar problem em of no historical data available either in otherr ccountries to compare to. Very few countries es have auctioned spectrum for 4G technology app pplication and thus there is lack of adequate research res reports on other countries. As mentionedd iin chapter on broadband, 4G technologies are superior to all technologies present till now whe hen it comes to broadband and thus it is naturally lly expected that it will be preferred over other techn chnologies. As per the Reliance report, at present ent there are 0.2mn subscribers of WiMAX in 20099 aand nearly 0.35mn subscribers in 2010. There are ar various reports by leading consulting firm andd banks and an analysis was done of all the reports re to reach a consensus figure for the WiMAX AX users according to the expected wireless subs bscribers. A report by FICCI titled “3G & BWA: Th The Next Frontier, Business Models, Projectionss and a Imperatives” predicts 1mn subscribers in first st year of availability and increases it to 10.8mn n users u by 5th year. If this data is used then it givenn 77% of the total wireless subscribers every year ar according to the wireline subscribers calculated in previous section. This proves that the underly rlying assumptions 57 | P a g e are similar in this report and the he subscribers calculated above. But the WiMAX X subscriber base cannot remain constant over tim time and should change in favor of WiMAX X as most of the companies have taken WiMAX X llicense and so should be launching in order to get returns over the price paid for the license. The he HSBC report predicts 45mn users by the year ar 2016. 2 The report by Telcordio is again wrong he here as it predicts 60mn WiMAX users by 2013 201 which is not possible as the rollout is happeni ening at present and so it cannot reach this figure re in 3 years time. The report by FICCI predicts 28.5 8.5mn WiMAX and 4.9mn LTE subscribers by 2015 20 and a similar report by Protiviti predicts it as 80-90mn by 2019. Now to forecast WiMAX X subscribers, the broadband subscribers of 3G wil will also need to be considered as 3G is also capa pable of providing high speed broadband and is wir wireless as well as already launched and thus hav ave a first hand in terms of reach ability to subscrib ribers. The only factor that will restrict broadband and users using 3G will be the limited spectrum prov rovided to telecom operators which will ensuree that th much of it is left for the voice subscribers. Al Also 4G technology once established will be more mo cost effective than compared to 3G technology. gy. Hence the forecast of the WiMAX subscribers rs is done using all the data made available by the rep reports and has been forecasted as below: Figu igure 49: Forecasted: WiMAX subscribers The underlying assumption is th that the rate of the penetration of the WiMAX X subscribers will itself increase with a rate of 1--4 % in coming years and will remain constan tant after 10 years, according to the initial assumptio ption of broadband growth being stable after 10 0 years from now. The growth rate n the penetration ion of the WIMAX subscriber is given below: 58 | P a g e 2012E 2013E 2014E 2015E 2016 16E 2017E 2018E 2019E 2020E 2021E 2022E 2023E 2 2024E 2025E 1% 2% 3% 4% 4% 2% 2% 2% 2% 0% 0% 0% 0% 0% Figure 50: Fore orecasted rate of growth in WiMAX penetration rate 6.7 WiMAX Broadband and AR ARPU ARPU for mobile broadband is aaround Rs. 70048. Penetration of broadband iss very v low in India and as there is not much historyy oof broadband and thus the price war as seen in n voice v hasn’t been seen in broadband and it has rema mained more or less same over the years. Source: OECD Da Database Figure 51: DSL Broadband Price Hence it can be safely assumedd tthat broadband charges will not fall much in coming com years. Also the report by FICC also considers ers only a minor decline in 3G modem ARPU fro rom $14.1 in 2009 to $13.3 by the end of 2013 whic hich gives us a rate of -1.45% CAGR. A -1.5% drop d annually till the end of 2020 has been assume med and then the ARPU is kept constant in thee next n decade. For 4G broadband revenue, the prem remium has been kept at zero over 3G broadba dband ARPU. The reason for same being that custo stomer will not be able to differentiate much betw etween 3G and 4G services at its end except for spe peeds and thus arbitrage will be created if theree is i any significant differential between ARPU from m 3G services and 4G services and hence has been een kept same. 48 http://www.bsnlevdo.in/bsnl-evdo-ne news/mts-to-offer-14-7-mbps-speeds-through-cdma-dongle gles, accessed on 10- 03-2011 59 | P a g e 6.8 Operator’s share in Data ARPU The value chain of broadband and data which is provided over mobile mostly has 4 major players: • Device Provider – Manufacturer of the mobile phone and other access devices • Application Provider – Software vendors, game and application developers • Content Provider – Content providers which the application uses • Channel Provider – Telecom operator which facilitates and provides channel or bandwidth Currently the share of data revenue is heavily skewed towards the telecom operators because of the limited uptake of data services in India for reasons mentioned before. Operator’s share of the total data service revenue is expected to go down from the current 70-75% to 63-67% by the end of 2015 as per PwC report. This is bound to happen as when data services will uptake in the country supported by 3G rollout. In the DCF, a decline from 72.5% to 65% from end of 2010 to end of 2015 respectively and constant at 65% from 2016 to 2030 has been considered. 6.9 Capital Expenditure Capital expenditure is Indian telecom industry can be broadly divided into two segments: 1. Investment in Passive Infrastructure 2. Investment in License Fees In India investment in passive infrastructure is not done by telecom operators but by separate entities set up exclusively to maintain and provide passive infrastructure on sharing basis. Companies such as Reliance Infratel and Indus Towers are the two main providers of tower infrastructure on rental basis. The operators have to install their BTS (Base Transmission Station) according to the technology and density of the area. For 4G rollout the Investment in passive infrastructure has been calculated as given below: For estimating the capital expenditure following methodology has been used: 60 | P a g e Parameters Values Source Total number of 4G subscribers 90406932 Forecast 720-840 (Selected 700) Protiviti Report 129153 Calculated $10000-12000 Protiviti Report by 2025 Number of subscribers supported by a BTS Number of BTS to be installed Cost of one BTS (Selected Rs. 495000) Total cost of installing passive Rs. 64576mn Calculated infrastructure Figure 52: Passive infrastructure investment This given per subscriber cost of installing passive infrastructure at around Rs. 700, which combined with Rs. 4277 paid for license, gives total capital expenditure per subscriber as Rs. 4995 which lies in the range Rs. 3000-5000 given by PwC report. The second part of the capital expenditure is in the form of License Fees. Here license fees is the aggregate value paid by all the operators for acquiring 4G/BWA licenses in all the 22 circles combined which is equal to Rs. 385000 Million as per TRAI website. 6.10 Operating Expenditure The breakup of the operating expenses for a typical telecom operator in India has been taken from a report prepared by FICCI in which they had given the industry wide cost structure with various heads. Expense Head Net Interconnection Charges as % of Gross Revenues Percentage Driver 20% Gross Revenue Network Operating Expenses 15% Net Revenue Sales & Distribution Expenses 7% Net Revenue IT Expenses 2% Net Revenue Service Expenses 3% Net Revenue 61 | P a g e Billing, Collection and Bad Debt 2% Net Revenue Marketing Expenses 3% Net Revenue Personnel Administration 5% Net Revenue Total Operating Expenses 37% Net Revenue Spectrum Usage Charge as % of AGR 3% Net Revenue EBITDA as % of Net Revenue 58% Net Revenue EBITDA as % of Gross Revenue 46% Net Revenue Source: FICCI Report Figure 53: Operating Cost Structure in Typical GSM operator in India 6.11 Financing of Total Capital Expenditure The total capital expenditure comes to around Rs 321.5mn and for financing such amount a debt equity ratio of 1 has been assumed. As valuation is being done considering it as a project and hence financing has been kept separate from the cash flows part and it has been assumed that the industry will maintain the debt and equity in the same ratio. The debt equity ratio is taken as the average debt equity ratio of 3 telecom operators for the last 5 years. Year Bharti Airtel Idea Cellular Reliance Communications Average 2006 0.65 4.96 NA 2007 0.47 1.95 0.71 2008 0.33 1.84 0.82 2009 0.28 0.67 0.6 2010 0.14 0.57 0.48 Average 0.374 1.998 0.6525 1.00 Source: Moneycontrol49 Figure 54: Industry Debt Equity ratio 6.12 Depreciation & Amortization For depreciation of capital expenditure and amortization of license fees a period of 15 years has been considered and straight line method has been assumed. 6.13 Cost of Capital The formula to calculate Weighted Average Cost of Capital (WACC) is given as 49 http://www.moneycontrol.com, accessed on 10/03/2011 62 | P a g e M7 I 9 MI A 9 MA 9 1 Where, I = Cost of Equity (Calculated using CAPM model) A = Cost of Debt (Assumed to be 9% based on companies real cost in industry)50 MI = Weight of Equity in Capital Structure (assumed 50%) MA = Weight of Debt in Capital Structure (assumed 50%) T = Tax Rate (Assumed to be 30%, IT act) Cost of Equity: Cost of equity has been calculated using Capital Asset Pricing Model given by I 1 N9 O 1 Where, N – This is measure of systematic risk of the project 1 – Risk free rate of return (Risk free rate of return is considered as yield on 10 Year GOI bonds which is 7.94% in March-2011) O – Rate of return on market (Rate of return on market has been calculated using Nifty 2010-2011) N has been calculated using two methods namely: 1. Median beta of portfolio of the telecom companies 2. Weighted Beta of a portfolio of stocks of telecom companies with weights being given according to the market capitalization of the stock Company (Stock) Bharti Airtel Idea Cellular 50 Beta Market Capitalization (Crore) Weight in Portfolio 0.73 122489 72% 1.06 20031 12% http://articles.economictimes.indiatimes.com/2010-05-25/news/27573892_1_bharti-airtel-3g-third-generation- mobile-spectrum, accessed on 10-03-2011 63 | P a g e Reliance Communications 1.52 MTNL 1.17 18976 2753 11% 2% TATA Communications 6022 4% 0.53 Median Beta 1.06 Portfolio Beta Sector Beta 0.86 0.54 Source: Reuters51 Figure 55: Beta calculation In order to have conservative forecasts, the highest value of 1.06 has been taken for the base case valuation. The WACC calculations performed are as given below: Parameter Debt-Equity Ratio Weight of Debt Weight of Equity Cost of Debt Tax Rate Return on Nifty Monday, January 04, 2010 Friday, December 31, 2010 Return Cost of Equity Risk Free Rate on 10 Year GOI Bond Market Return on Nifty in 2010 CY Beta Cost of Equity Cost of Capital value 1 0.50 0.50 9% 30% 5232.2 6134.5 17.25% 7.94% 17% 1.06 17.80% 12.05% Figure 56: Cost of capital calculations 51 http://in.reuters.com/, accessed on 10/3/2011 64 | P a g e 7. Results The result from DCF and Real op option valuation has been presented in this chapte pter 7.1 Revenue Forecast The forecasted Gross Revenue an and the Free Cash Flow to Firm (FCFF) using g the th base case and the assumptions explained in prev revious chapter are as shown below: Figure re 57: Forecasted Gross Revenue and FCFF The gross revenue for the base ca case are forecasted to be Rs 643bn by year 2025, 5, an increase from Rs. 6.6bn in year 2011. 7.2 DCF valuation 7.2.1 NPV and IRR The NPV of the project after con considering the license fees as well as capital expenditure ex comes out to be Rs. 221bn and IRR com omes out to be 17.1%. The DCF model has been give gi in Appendix. Method Va Value Selection criteria NPV (Bn) 2221 NPV > 0 IRR 17 17.1% IRR > WACC Figure 58: Base case NPV and IRR 65 | P a g e Result Project shoul uld be accepted ed Project shoul uld be accepted ed As shown, using either method as criteria, project consisting of 4G license purchase and network roll out should be undertaken by telecom operators to increase their shareholder value. The project looks attractive by both measures and returns are above the returns expected by market in the base case assumptions. 7.2.2 Sensitivity analysis From the understanding of the DCF model, there are many critical factors which decide the outcome of the DCF and hence they need to be checked for the variance that can be seen in case the base assumptions are not met. These critical parameters has been selected as Subscriber adjustment factor, EBIDTA margin, Beta and Premium in ARPU compared to 3G broadband ARPU. 1. Subscriber adjustment factor: Subscriber adjustment factor is important parameter as it decides how many broadband subscribers will be there and decides if it will exceed or remain lesser than the projected subscriber base. Subscriber base forms one of the important variable in revenue calculation and hence is directly responsible for the DCF calculation. As shown below, ceteris paribus, NPV will remain positive even if the adjustment factor is made 0.5, that is half of the projected subscriber base and thus the project will remain viable even if the forecast of subscriber overshoots original subscribers. If the subscribers turn out to be less by 48% of the forecasted subscribers number (after adjustment value) then the NPV will become negative and the project will become unviable. Adjustment factor Sensitivity Analysis 0.48 0.6 0.7 0.8 0.90 1 NPV -2 82 151 221 291 360 Figure 59: Sensitivity Analysis - Adjustment factor 2. EBIDTA Margin: EBIDTA margin has been assumed to be 47.4% in the base case but as 4G technology is new and there is lack of information among customers and hence in market like India, telecom operators might need to put extra expenditure in promotion 66 | P a g e and marketing and might even need to push customers for high speed broadband. Besides as the technology is still to be scaled and hence telecom operators might need to incur extra expenditure in terms of the subsidizing device cost at the customer end. Thus overall effect will be to decrease EBIDTA margin and this can affect the NPV calculations. A scenario analysis has been done and EBIDTA margin has been varied from 50% to 40% to show how much variance can be seen in NPV resulting from different EBIDTA margin than expected. Here again as can seen form the output below, the NPV is still positive at 40% EBIDTA margin and hence it is robust enough to withstand variation in the margin. The reduction in NPV with reducing EBIDTA margin if not fast and so project can tolerate huge reduction in EBIDTA margins. If EBIDTA margins decrease to 30.3% or below, project will become unviable. Sensitivity Analysis 50% 47.4% NPV 255 221 EBIDTA margin 46% 44% 42% 30.3% 203 151 0 177 Figure 60: Sensitivity Analysis – EBIDTA margin 3. Beta and Premium over 3G – As shown before, the beta of the company and hence the project can vary from 0.54 on the lower side to 1.2 on the upper side and hence this can affect the NPV of the project. The scenario analysis has been done for pessimistic beta value of 1.2 to optimistic value 0.54 and the premium of 0% over 3G to up to 20%. As can be seen, again the NPV is very robust and remains positive for even reduction in the premium over 3G and increase in beta value. Keeping beta at maximum deterioration of 1.5, Premium over 3G should fall by 22.7% in order for project to become unviable. Sensitivity Analysis NPV (Bn) Beta 0.5 0.7 0.9 1.06 1.2 1.5 -22.7% 203 162 116 82 54 0 -10% 299 252 199 160 128 66 Premium of -5% 337 288 232 190 157 92 4G over 3G 0% 375 323 264 221 186 118 5% 412 359 297 252 215 144 10% 450 394 330 283 244 170 Figure 61: Sensitivity Analysis - Beta and premium over 3G 67 | P a g e 4. Beta and EBIDTA margin – The last scenario analysis has been run between beta and EBIDTA margin. Beta can change with time and can put increasing pressure in terms of the return expectations from the project and if combined with reduced EBIDTA margins can turn explosive. From the output shown below, it can be seen that NPV is again robust with simultaneous changes between EBIDTA margin and beta and can tolerate deterioration in both the parameters. Taking maximum deterioration in beta at 1.5, EBIDTA margins can fall till 36.5% before project becomes unviable. Sensitivity Analysis NPV (Bn) EBIDTA margin Beta 36.5% 0.5 212 0.7 160 0.9 114 1.06 80 1.2 52 1.5 -1 42% 301 242 190 151 120 59 44% 333 272 217 177 144 81 46% 365 302 245 203 169 103 47.4% 388 323 264 221 186 118 50% 430 362 300 255 218 147 Figure 62: Sensitivity Analysis - Beta and EBIDTA margin 7.3 Incremental Revenues for 4G Operators As mentioned before, there can be two types of telecom operators who have taken licenses for 4G spectrum: Incumbent operators and standalone operators. For standalone new entrant in broadband services, there is no incremental revenue as it will be entirely new business line for it. For the incumbent operator there can be loss of revenue in its business related to non-broadband subscribers and wireline subscribers as well as 3G broadband subscribers. But as ARPU for 4G broadband is more than any other source of access to internet and hence no fall in final revenue is expected with switching from older technology to newer technology and hence there will be only incremental revenue and never a loss due to switching. 7.4 Real Option valuation 7.4.1 Real Option value The value of the Real option was calculated according to the variables as explained in 5th chapter and has been given below: 68 | P a g e Parameters S (Bn) K (Bn) rf Q t (years) Σ Output D1 Value 670.768 64.6 7.94% 6.66% 5 27.5% Value 4.2172589 D2 3.6023402 N(D1) 99.9987% N(D2) License Value (Bn) 99.984% 437 Figure 63: Real Option calculation The value of optionality attached with the 4G spectrum is valued at Rs. 437bn which is more than the price paid by telecom operators that is Rs. 385bn. This shows that telecom operators have been cautious in the auction and have paid such that they still have up to 13.5% more value remaining in the spectrum. This is good news for telecom operators as they have received the spectrum with extra value above optionality value by 13%. This will give them extra space to execute managerial flexibility. From Government perspective, this 13% signifies the value that they could have earned as revenue over and above the amount received in auction. This points towards not so efficient auction process and the need to find better way to allow discovery of efficient price by supply-demand dynamics. 7.4.2 Sensitivity analysis In Real Option valuation two parameters assume great significance, namely time to maturity and volatility. Now in Indian 4G auction process, the time to maturity has been fixed to 5 years by use of rules and regulations by DoT but it might be changed in future and so time to maturity will assume significance. Similarly volatility can change with time and hence can sway the value of option. The sensitivity analysis of both parameters in given below: 69 | P a g e Sensitivity Analysis Real Option (Bn) Capital expenditure Time to maturity 64 3 498 4 467 5 437 6 410 7 384 8 360 70 494 463 434 406 381 357 80 486 456 427 400 375 351 90 478 448 420 394 369 346 120 455 427 400 375 352 331 142 437 411 385 362 340 319 Figure 64: Sensitivity Analysis - capital expenditure and Time to maturity The value of the capital expenditure should increase by 130% in order to make the optionality unviable at 5 years time to maturity and hence the optionality value is not very stable and in case of escalation of cost, telecom operators can go in loss easily and hence the auction has went more in favor of Government rather than telecom operators. Sensitivity Analysis Real Option (Bn) Value Volatility 0.275 437 0.30 437 0.45 438 0.50 438 0.60 440 0.70 442 Figure 65: Sensitivity Analysis - Volatility and Real option value The last sensitivity analysis will be done with respect to the subscribers forecasted and the adjustment factor in case it happens to change against the expectations. Again the same results can be see that is if the adjustment factor turns out to be little lower than forecaste then the value of optionality will decrease and will fall below Rs. 385 paid for the license and the telecom operators will be in loss. Sensitivity Analysis Real Option (Bn) 0.4 0.705 Value 214 384 Adjustment factor 0.8 1 1.2 1.4 437 658 769 548 Figure 66: Sensitivity Analysis - Adjustment factor and Real option value Overall as seen with the various sensitivity tests, it can be easily seen that telecom operators do have some optionality value left after the spectrum fees paid but if the underlying assumptions of the base case varies then it can lead to operators loosing on the margins and thus it can prove disadvantageous in the long run. 70 | P a g e 8. Conclusion and Recommendation In this research, first analysis has been done of the country, here India and then the telecom industry. Indian telecom industry in one of the fastest growing telecom industry and boasts of its second position worldwide. Various reasons attributed to this growth have been explained in the report such as demographic changes, increasing affordability, favorable policies by Government etc. Recently held 4G/ BWA spectrum auction saw enthusiastic participation by the industry and even saw some new entrants in Indian broadband market. Government benefitted by Rs, 385bn that it earned as revenue from the auction of the spectrum and projected it as successful auction, one which generated maximum value for the nation. The report evaluates 4G spectrum, first as a project using traditional DCF valuation and later as spectrum license using Real Option valuation method in order to take into consideration the managerial flexibility and strategic decision making aspects. The report tries to answer questions raised about the efficiency of the auction and the end value creation for all the parties involved. How will the players recover the license costs, along with the equipment and rollout costs? With the falling ARPU every year, will industry be able to earn enough to flourish or just survive or may be vanish under burden of the loan. In first step, DCF valuation has been done. During the analysis, broadband subscribers have been forecasted to grow from present 13.77mn to 544mn by the end of 2025. The wireless subscribers have been forecasted to be 70% of the total broadband subscribers after 5 years of roll out as it will be difficult to replace all wireline subscribers with wireless subscribers in India due to the high cost of wireless broadband and new technology. WiMAX is expected to increase its presence with time and reach 90mn subscribers from meager 0.35mn subscribers by 2025. The base case EBIDTA margins are supposed to be 47% and industry wide cost of capital as 12.05%. This forms the base case for the DCF analysis. The revenue has been forecasted using the subscriber number and the forecasted ARPU as Rs. 6.6bn in first year of roll out which increases to Rs. 643bn by 2025. The Net Present Value has been found to be positive and comes to around Rs 221bn aggregate with an IRR of 17.1%. Thus 4G license as project is attractive proposition when compared to 71 | P a g e cost of capital as cut-off rate but considering that there are many risks associated with a new technology, the returns look nominal. The report further tries to analyse the risks associated with the base case and it has been found that there is less flexibility for the telecom operators and margins might get squeezed in long term. As a project, any telecom operator will like to have returns that justify the risk and time involved in the project. The License price reached during the 16 days of auction leaves 17% return under the base case assumptions. To probe this aspect further, in next step, sensitivity analysis has been done to analyse the risk associated with the assumptions and their effects on the NPV. It has been found out that if the subscribers turn out to be less by 48% of the forecasted subscribers number (after adjustment value) then the NPV will become negative and the project will become unviable. Also telecom operators will lose any positive NPV if EBIDTA margins fall to 30.3% from present 47.5% that is a fall of 33% in EBIDTA margin. If beta is kept at the most unfavourable scenario that is at 1.5 then project becomes unviable either if premium over 3G should fall by 22.7% or if EBIDTA margin falls to 36.5% or below. Thus there is very less space for maneuverability with telecom operators in terms of changes in the underlying assumptions of the base case. Using Real option approach that takes into consideration the managerial flexibility and the value of strategic decision available with the telecom operators such as option to delay, option to choose technology, option to abandon after some time etc, the value of license has been calculated as Rs 437bn which is 13.5% more than the spectrum fees paid by the operators. Thus telecom operators have substantial value left after paying for the license fees and thus it has been bought as undervalued. This leaves telecom operators with excess value that they can use effectively as managerial flexibility. Where telecom operators have gained, Government has lost. This 13% surplus could have been garnered by Govt. if the auction process would have been highly efficient. This points towards the loopholes in the process which needs to be corrected. The gap between actual price that should have been paid and the lesser value finally paid can be due to telecos pulling out from auction process mid-way such as TATA and Reliance (led by Anil), lack of sufficient competitors, different industry view of risk with the new technology roll out with Indian subscribers or any other differing view compared to the base case. I future Govt. should try to make auction process more competitive and extract the maximum value as per the supply-demand dynamics. 72 | P a g e 9. References Aircel Report, “Recent trends in mobile data needs”, 2010 Alleman, James, “A new view of telecommunications economics”, Telecommunications Policy, Vol. 26, pp. 87-92, 2002 Basili, Marcello, Fontini, Fulvio, "The option value of the UK 3G Telecom licenses: Was too much paid?", info, Vol. 5 Iss: 3, pp.48 – 52, 2003 Binmore, K. and Klemperer, P., “The biggest auction ever: the sale of the British 3G telecom licenses”, Economic Journal, 2002 Chang, Jow-Ran, Hung, Mao-Wei, Tsai,Feng-Tse, “Valuation of Intellectual Property- a real option approach”, Journal of Intellectual Capital, Vol.6, No.3, pp. 339-357, 2005 CMIE Database, accessed on 9/3/2011 Confederation of Indian Industries, http://www.ciionline.org/Economy.aspx?enc=LqAXY5bXIsb2PzUHQxy2iQ==, accessed on 9/3/2011 Dai, Q., Kauffman, R. J. & March, S. T., ‘Analyzing investments in object-oriented middleware: An options perspective.’ In Proceedings of MIS Research Center Working Papers Carlson School of Management, Minneapolis, University of Minnesota, p.p.1–27., 2000 Damodaran, A., “The dark side of valuation”, New Jersey: Prentice Hall, 2001 Damodaran, Aswath, "The option of Real option", Journal of Applied Corporate Finance, Vol. 13, No. 2, pp. 29–44, 2000 DoT, "Auction of 3G and BWA spectrum", Feburary 2010 FICCI, "3G and BWA: The Next Frontier", January 2009 Herbst, Patrick, Walz, Uwe, “Real Options Valuation of Highly Uncertain Investments: Are UMTS-Licenses Worth their Money?”, Department of Economics, University of Tuebingen, 2001 HSBC Global Research, “Indian Telecoms”, June 2010 http://articles.economictimes.indiatimes.com/2010-05-25/news/27573892_1_bhartiairtel-3g-third-generation-mobile-spectrum, accessed on 10-03-2011 73 | P a g e http://economictimes.indiatimes.com/news/news-by-industry/telecom/Indian-telecommarket-to-be-at-Rs-344921-crore by 2012/articleshow/2563062.cms, accessed on 9/3/2011 http://en.wikipedia.org/wiki/0G, accessed on 10/3/2011 http://en.wikipedia.org/wiki/1G, accessed on 10/3/2011 http://en.wikipedia.org/wiki/2G, accessed on 10/3/2011 http://en.wikipedia.org/wiki/3G, accessed on 11/3/2011 http://en.wikipedia.org/wiki/3GPP_Long_Term_Evolution, accessed on 11/3/2011 http://en.wikipedia.org/wiki/4G, accessed on 11/3/2011 http://in.reuters.com/, accessed on 10/3/2011 http://indiabudget.nic.in/es2008-09/infra.htm, accessed on 9/3/2011 http://www.bsnlevdo.in/bsnl-evdo-news/mts-to-offer-14-7-mbps-speeds-through-cdmadongles, accessed on 10-03-2011 http://www.google.com/publicdata, accessed on 10-03-2011 http://www.google.com/publicdata, accessed, on 10-03-2011 http://www.ibef.org/industry/telecommunications.aspx, accessed on 9/3/2011 http://www.informatm.com/itmgcontent/icoms/s/press-releases/20017778395.html, accessed on 9/3/2011 http://www.itu.int/dms_pub/itu-d/opb/stg/D-STG-SG02.18-1-2006-PDF-E.pdf, accessed on 11/3/2011 http://www.itu.int/ITU-R/index.asp?category=information&rlink=imtadvanced&lang=en, accessed on 11/3/2011 http://www.moneycontrol.com, accessed on 10/03/2011 http://www.Planningcommission.nic.in/reports/wrkpapers/wp_hwpaper.pdf, accessed on 10/03/2011 http://www.ril.com/rportal1/DownloadLibUploads/1276353679981_FPR120610.pdf, accessed on 1/3/2011 http://www.telcordia.com/news_events/media_coverage/express-computers-nov2010.pdf, accessed on 1/3/2011 http://www.tradingeconomics.com/Economics/Government-BondYield.aspx?Symbol=INR, accessed on 10-03-2011 74 | P a g e ICICI direct research report, “BWA spectrum auction”, June 2010 IMRB, “Mobile Internet in India”, December 2009 Intel Corporation, “Internet everywhere – 4G technology”, April 2010 International Monetary Fund, http://www.imf.org/, accessed on 9/3/2011 Mastroeni, Loretta, Maurizio Naldi, “A real option model for the transferability value of telecommunications licenses”, Annals of Telecommunications, Vol. 65, No. 3-4, pp. 201210, 2009 Michel, Benaroch, Kauffman, Robert J., “A case for using real options pricing analysis to evaluate information technology project investments”, Information Systems Research, Vol. 10, No. 1, pp. 70-86, 1999 Moja, N., Mkhize, M., “The application of real option valuation techniques in the cellular telecommunication industry in South Africa”, South Africa Journal of Business Management, Vol. 40, No. 3, 2009 Mun, J. 2006, “Real options analysis – tools and techniques for valuing strategic investments and decisions”, 2nd Edition, New Jersey, Wiley Finance PressReleases/ 798/prerdiv9feb11.pdf, accessed on 1/3/2011 Protiviti, "WIMAX: The Quintessential answer to broadband in India", 2009 PwC Report, "Mobile Broadband Outlook-2015”, 2010 PwC Report, “The World in 2050”, Jan 2011 Reliance Industries Limited, “BWA – Analyst report”, June 2010 Stille, R., Lemme, C., Brand , L, “An application of Options of the Royal Evaluation of the License of the benefits system 3G Mobile Telephony Services in Brazil”, Journal of Finance, Rio de Janeiro, Vol 8, No. 3, 329-349, 2010 TRAI December Report, http://www.trai.gov.in/WriteReadData/trai/upload/ 75 | P a g e 10. APPENDIX DCF Model Projections 76 | P a g e 77 | P a g e 78 | P a g e DCF Assumptions 79 | P a g e 80 | P a g e