Journal of Business Research 60 (2007) 731 – 741

Strategic alliances influence on small and medium firm performance

Cheng-Wen Lee ⁎
Chung Yuan Christian University, Faculty of International Trade Department, 200 Chung Pei Road, Chung Li, 32023 Taiwan, ROC
Received 1 October 2006; received in revised form 2 January 2007; accepted 1 February 2007

Abstract

This study examines whether the new ventures success of small and medium enterprises (SMEs) in the biotech industry relates with the
characteristics of strategic alliances. The study advances a research conceptual framework. Using sampling data gathered from 189 Taiwan biotech
firms through a benchmarking questionnaire, the study tests six hypotheses employing structural equations. The findings are generally consistent
with the literature. The study supports all hypotheses. Consequently, the results show that strategic alliances improve SMEs' new venture success.
This research provides managerial implications for both entrepreneurs and managers and some suggestions for future research.
© 2007 Elsevier Inc. All rights reserved.

Keywords: SMEs; Biotech new venture; Strategic alliance

1. Introduction The biotech industry offers the advantages of powerful market

potentiality and long product life cycle, as well as the
The emergence of biotechnology constitutes a radical disadvantages of huge investment expenditure, time-span R&D,
innovation that breaches the barriers of entry into the pharma- high risk, huge patents fee and complicated regulations/laws. In
ceutical industry, among other industries (Pisano, 1990). Since the light of this balance, strategic alliances may be a faster way to help
early 1970s, about 1600 new biotech firms have emerged to firms successfully enter into specific market domains and acquire
commercialize this technological breakthrough. Extensive inter- complementary resources. As suggested by resource-based
firm cooperation characterizes the commercialization of biotech- theory, the decision to choose alliance partners is predicated on
nology. Indeed, the biotech industry has been identified as the mutual benefit, mutual potential to provide additional resource
industry with the highest alliance frequency among several value of R&D, manufacturing and/or marketing. Inter-firm
industries of high alliance activity (Hagedoorn, 1993). alliances have the ability to alter the opportunities and constraints
In the biotech industry, strategic alliances extend worldwide. faced by potential entrepreneurs (Kogut et al., 1992).
Mowery and Rosenberg (1989) define strategic alliances as a Market access is usually a major motivation for licensors to
vehicle for the diffusion of technological knowledge that can seek strategic alliances. Nevertheless, the need for risk reduction
contribute to firm success. Alliances play increasingly important of commercialization within the biotech industry may be a more
roles in drug discovery and development, not only complement- important reason for strategic alliances. McCutchen and Swami-
ing companies' internal technology and competencies, but also dass (2004) find that the third of the licensees are small firms in
providing partners with access to pipeline products and capital. the biotech industry, not large firms. They also indicate that small
The collaboration between Millennium Pharmaceuticals and biotech firms seeking strategic alliances were more likely to be
Bayer HealthCare AG, together creating one of the world's motivated by R&D time-span reduction than larger firms.
largest pharmacy/biotech alliances exemplifies this cooperation Barley et al. (1992) propose that small and medium
(Ziegelbauer and Farquhar, 2004). enterprises (SMEs) in the biotech industry need the capital
and downstream capabilities that larger firms possess, such as
expertise to deal with regulatory bodies, competence to develop
⁎ Tel.: +886 3 2655215; fax: +886 3 2655299. timely clinical testing, ability to scale up the manufacturing of a
E-mail address: [email protected]. product for large quantities, and established marketing and
0148-2963/$ - see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jbusres.2007.02.018
732 C.-W. Lee / Journal of Business Research 60 (2007) 731–741

distribution capabilities. Sometimes alliances that require small competitive advantage in science-based industries (Grant,
biotech firms yield some of their autonomy and offer the larger 1996). The flow of knowledge among alliance participants
firms an additional source of new products or expertise in a determines their structure/management and in turn, influences
much quicker time frame than in-house product development their success (George et al., 2001). Coombs and Deeds (2000)
efforts (Pisano, 1990). The literature on the biotech industry has find that the firm location is positively related to the amount of
noted that there are many vital differences between small foreign alliance capital received. The hot spot in which a
biotech firms engaged in R&D and larger biotech firms that biotech firm is embedded provides a signal to potential foreign
purchase R&D services from smaller firms (McCutchen and venture partners, and some important research will be done on
Swamidass, 1994). the hot spot. Sambharya (1996) reports that the relative time of
As compared with Japanese, American, and European firms, experience in overseas investment, experience homogeneity
most Taiwan manufacturers are smaller in size. Concurrently, and the proportion of TMT with international experience are
investment in Taiwan biotech industry projects has increased significantly related to the percentage of the firm's foreign sales
yearly and individual companies now make greater efforts to and assets. Here, the term “technical capacities” indicates
develop biotech R&D. Despite this expanded effort, it remains scientific knowledge, firm location, and TMT international
unclear if the commercialization of Taiwan's biotech products experience — key factors of affecting technological innovation
has been successful. Limited resources compel Taiwan's in new ventures. Hence,
biotech firms to choose strategic alliance integrating techno-
logical resources with partners. Hypothesis 1. The technical capacities of alliance partners
This study examines whether the new ventures success of positively and significantly influence the success of biotech new
SMEs in the biotech industry relates with the characteristics of ventures.
strategic alliances. Specifically, it endeavors to ascertain how
alliance characteristics are associated with firm success. This Biggadike (1979) mentions that, in the period of concept
study defines SMEs in the biotech industry as those generating development, evaluation criteria include an index of survival,
under USD 10 million in annual revenue. new product development, and satisfaction in proceeding
business. In the commercialization period, upgrading brand
2. Literature review and hypotheses awareness and sales growth are more important than survival. In
the growth period, market share, cash flow and productivity are
2.1. Partners' technical capacities and new venture success the core indexes for evaluating the success of a new venture.
Finally, in the stability period, profit, market share, productivity,
Reuber and Fischer (1997) find that, in the early phase of internal performance, and future development are the key
entering international markets, firms gather more information factors determining new venture success (Zahra, 1996).
about foreign markets through international strategic partners. In accord with the literature noted above, evaluation of the
Biotech firm's survival depends almost exclusively on success of a new venture uses the two observed variables of new
proprietary knowledge. However, they are generally not product development (NPD) performance and commercial
inclined to provide detailed R&D information to potential success. According to Cooper and Kleinschmidt (1995), this
alliance partners. One consequence is that potential alliance study proposes four criteria to benchmark the firm's NPD
partners may not accurately evaluate the firms' potential for performance: (1) success rate of new product (NP); (2) technical
product commercialization success. At the same time, if biotech success rating of NPD program; (3) the rated profitability of the
firms disclose detailed R&D information, this may create new company's NP program (over the previous 3 years) relative to
potential competitors who have gained specific patent knowl- how much was spent on it; and (4) NP success in meeting
edge from them. This dilemma is called “Arrow's paradox” business objectives. Concurrently, this study proposes four
(Arrow, 1962). Similarly, Lee and Burrill (1994) suggest that criteria to benchmark the firm's commercial success: (1) impact
biotech firms attempt to enhance cutting-edge scientific R&D of NP program on sales; (2) impact of NP program on profits;
with resources from potential partners, but partners may hesitate (3) NP success in meeting sales/profits objectives; and (4)
to cooperate because they lack sufficient information to evaluate profitability of NP program relative to competitors.
the scientific and commercializing capabilities of biotech firms. In addition, firm size also influences strategic alliance
As biotech firms have few commercialized products and little performance (Simonin, 1997). By effectively managing attractive
sales revenue, potential alliance partners have difficultly and generative alliance partnerships, biotech firms can easily
ascertaining their actual strength (Coombs and Deeds, 2000). acquire and utilize new knowledge and upgrade their competitive
Coombs and Deeds (2000) propose that three signaling advantage. This study uses “firm size” as a control variable. Firm
mechanisms – scientific knowledge, firm location, and top size is measured in terms of employees and revenues.
management team's (TMT) international experience – can
overcome information asymmetries, as well as aid the 2.2. Alliances structure and absorptive capacity
evaluation of a potential alliance partner's resource-base and
product commercialization capacity. Three indexes of scientific As many scholars and industry observers note, the
knowledge are R&D expenditures/intensity, patents held and appropriation regime surrounding biotechnological patents is
products in the pipeline. Knowledge is a key foundation of unusually strong because the patented compounds are difficult
 C.-W. Lee / Journal of Business Research 60 (2007) 731–741 733

to circumvent (Lerner and Merges, 1998). Baum et al. (2000) Part of this process lies in the specific abilities of a firm,
find that incumbents' horizontal alliances depress start-ups whereas part is accumulated from network embeddedness that is
whereas vertical alliances stimulate start-ups in the Canadian difficult to buy or acquire from the market. Innovation and
biotech industry. They also advance that an incumbent biotech productivity (Mowery et al., 1998), cooperative inter-organiza-
firm's patenting (an index of property rights) and alliance tional learning (Lane and Lubatkin, 1998), and practical intra-
building (both a firm-level resource acquisition and industry- firm knowledge transfer (Szulanski, 1996) all widely utilize the
level integration mechanism) can deflect competition and concept of absorptive capacity.
influence rates of founding in the Canadian biotech industry. Cohen and Levinthal (1990) identify absorptive capacity as a
George et al. (2001) address two characteristics of the firm's ability to recognize the value of new external information
alliance portfolio: structure and knowledge flow. The “struc- and then assimilate this value in commercialization. They
ture” refers to whether an alliance is completed with a firm at the propose three constructs (evaluation, internalization, applica-
same level of the value chain (horizontal) or at a different level tion) to evaluate absorptive capacity in researching 1719 SBUs
(vertical). Alliance structures, whether horizontal or vertical, are of manufacturers. Kim (1998) indicates that absorptive capacity
associated with varying degrees of innovativeness (Hagedoorn, is formed from a prior knowledge base and intensity of effort.
1993; Kotabe and Swan, 1995). There are several classes of Lane et al. (2001) also argue that absorptive capacity is related
alliances — horizontal ones with other biotech firms operating to understanding, internalization, and application based on the
in the same sector(s) as the local firm, vertical (downstream) data of 78 international joint ventures. Following to Zahra and
ones with pharmaceutical firms, chemical firms and marketing George (2002), this analysis classifies the absorptive capacity
firms, and vertical (upstream) ones with universities, research into the four dimensions of acquisition, assimilation, conversion
institutes, government labs, hospitals and industry associations and exploitation; the former two imply potential absorptive
(Baum et al., 2000). George et al. (2001) also suggest that capacity and the later two represent realized absorptive capacity.
alliance characteristics designated by structure (horizontal or
vertical alliances) and knowledge flow (generative or attractive) 2.3. Alliances type and new ventures success
are related to the firm's absorptive capacity (ability to value and
apply knowledge). Rothaermel (2001) investigating 889 strategic alliances of
Lane and Lubatkin (1998) attempt to apply the concept of pharmaceutical companies and new biotech firms, finds that an
absorptive capacity to the learning process between alliance incumbent's alliances with providers of new technology were
partners. They indicate that the relationship between absorptive positively associated with the incumbent's NPD, as well as that
capacity and the characteristic of strategic alliance was a dyadic NPD was also significantly related to firm performance.
construct called “relative absorptive capacity”. It includes the Exploration alliances launch with the intent to discover
ability to (1) recognize and value new external knowledge, (2) something new, focused on the “R” in the R&D process
internalize new external knowledge, and (3) commercialize new (Koza and Lewin, 1998a,b). Rothaermel and Deeds (2004) link
external knowledge. George et al. (2001) recognize absorptive the exploration–exploitation framework of organizational
capacity as a combination of the ability to value knowledge and learning to a technology venture's strategic alliances and
the ability to apply knowledge. They also indicate that alliance argue that the causal relationship between the venture's alliance
portfolio characteristics and absorptive capacity jointly influ- and its new product development depends on the type of the
ence performance, as indicated by a sample of 2456 alliances alliance. They also propose that a product development path
formed by 143 biopharmaceutical firms. begins with exploration alliances predicting products in
However, the structure of alliances can influence the development, which in turn predicted exploitation alliances,
development of the performance-enhancing capacities that and concluded with exploitation alliances leading to products
exceed the development of a firm's absorptive capacity (Powell, on the market.
1998). Lane and Lubatkin (1998) indicate that the structure of
alliances have a greater effect on firm performance than can be Hypothesis 4. The type of strategic alliance positively and
explained by absorptive capacity. They also suggest that significantly influences the success of biotech new ventures.
horizontal alliances are likely to give the firm access to multiple
types of knowledge. On the other hand, vertical alliance may 2.4. Alliances partners' relationship and new ventures success
provide access for biotech firms to commercialize their products
or innovations, and meanwhile, increase speed to market as In his theory of social exchange, Blau (1964) proposes two
biotech firm frequently ally with larger pharmaceutical firms. main concepts of trust and commitment; the consequence of the
This study attempts to observe such findings. social exchange process based on mutual interest was to
produce appreciation, responsibility, and trust. Powell et al.
Hypothesis 2. The structure of strategic alliances positively (1996) indicate that the most basic factor of inter-organizational
and significantly influences the counterparts' absorptive cooperation and innovation in the network of biotech industry
capacity. was mutual interest. Doney and Cannon (1997) suggest that the
most important factor for establishing a partners relationship
Hypothesis 3. The structure of strategic alliances positively and was “trust”; the higher the ability and confidence of alliance
significantly influences the success of biotech new ventures. partners, the higher the reliability of alliance partners in the
734 C.-W. Lee / Journal of Business Research 60 (2007) 731–741

exchange process. Mohr and Spekman (1994) also consider Success in science-based industries requires that a firm
trust as an important attribute, as well as a determinant factor in develop and hone its capabilities (Bogner et al., 1996).
affecting the success of a partner relationship. As for the Absorptive capacity is considered to be a set of capabilities,
measure of trust, Young-Ybarra and Wiersema (1999), in such as the ability to value, assimilate, and apply knowledge
research on information technology alliances, propose that the (Grandstrand and Sjolander, 1990). Gassel and Pascha (2000)
overall extent of trust might influence cooperative behavior. find that cooperation in the biotech field primarily aims at
Homans (1958) argues that communication may establish a gaining access to scientific resources and commercial success; it
better relationship for exchange between partners. Lee and Kim is an interactive “give and take” in order to combine both
(1999) also indicate that efficient communication between partners partners' strengths. Hence,
was a crucial factor in accomplishing a common goal. Dwyer et al.
(1987) suggest that a good working relationship might not form if Hypothesis 6. The counterparts' absorptive capacity of strate-
there was no good communication between buyer and seller. gic alliance positively and significantly influences the success
Regarding the correlation of communication and trust, Etgar of biotech new ventures.
(1979) finds that immediate communication can increase the
degree of trust and decrease the conflict between the alliance 3. Method
partners. Anderson et al. (1994) and Anderson and Narus (1990)
also find that good communication can produce closer trust. 3.1. Factor analysis and reliability measures
Morgan and Hunt's (1994) relationship marketing study proposes
that communication will significantly and positively influence This study uses factor analysis to better define the dimen-
“trust”. Lane and Lubatkin (1998) indicate that the extent of sions of six constructs: alliance partner characteristics, alliances
alliance members' reactions will affect the performance of structure, alliances type, alliance partner relationship, and ab-
knowledge/technology transfer. As for the criteria of communica- sorptive capacity, and new ventures success. Hair et al. (1992)
tion, Young-Ybarra and Wiersema (1999) and Sivadas and Dwyer demonstrate that the general purpose of factor analytic tech-
(2000) use the communication quality of cooperative partners. niques is to condense (summarize) the information contained in
Kauser and Shaw (2001) find that in international alliance a number of original variables into a smaller set of new, com-
coordination, commitment, trust and communication were posite dimensions or variates (factors) with minimum informa-
predictable factors of strategic alliance success. Dwyer et al. tion loss — a process that involves searching for and defining
(1987) argues that establishing a relationship exchange needs a the fundamental constructs or dimensions assumed to underlie
combination of communication and coordination. Sivadas and the original variables. To identify the latent dimensions or
Dwyer (2000) propose that the measure of “coordination” constructs represented in the original variables and eliminate
reveals the extent of co-working toward the alliance goal. variance, this study uses factor analysis to extract these factors.
Morgan and Hunt (1994) identify shared value as the Specifically, this study adopts Varimax rotated component
combination of common behavior, goal, and policy between analysis. Factor loadings are the correlation of each variable and
partners. Heide and John (1992) recognize that shared value was a given factor. Loadings indicate the degree of correspondence
the norm. Dwyer et al. (1987) propose that shared value would between the variable and the factor, with higher loadings
play an importance in contributing to organizational commit- indicating that the variable is representative of the factor.
ment and trust. Therefore, in organizational theory the shared These factors are named appropriately based on the subjective
value was a variable of organizational commitment (Chatman, opinion of our research and hypotheses. Variables with higher
1991). Morgan and Hunt (1994) verify that shared value will loadings have a greater influence on the name or label that
influence trust and commitment. Accordingly, this study represents a factor. This analysis identifies these using eigen-
proposes the following hypotheses. values larger than one and factor loading of each item is more
than 0.5. Table 3 shows results.
Hypothesis 5. The relationship of alliance partners positively In this study, the method of reliability measurement is
and significantly influences the success of biotech new ventures. Cronbach's coefficient α. Cronbach's coefficient α is calculated
for each of these factors to assess the internal consistency of the
2.5. Absorptive capacity and new ventures success model constructs. According to Price and Mueller (1986), a
standard coefficient α of 0.60 or higher generally is considered
Van den Bosch et al. (1999) argue that the key factors of acceptable when using a measure. If statistical significance is
absorptive capacity not only include relative pre-knowledge not achieved, the research may need to eliminate the indicator or
level, but also cover organizational form and combinative attempt to transform it for better fit with the construct. However,
capabilities. Absorptive capacity is recognized as a new creative the values of Cronbach's coefficient α of latent variables and
value from the combination of pre-knowledge base and external observed variables all exceed 0.80 and that of some constructs
knowledge from outside. Baum et al. (2000) researching the even exceed 0.90. This indicates that the research has good
relationship between biotech firms' new venture success and consistency and stability (see Table 1). In addition, content
inter-organizational alliances, find that strategic alliances are an validity should be relatively acceptable because this part of the
important means of accessing or transferring new knowledge question items is adapted from existing literature and also is
and improve new innovative performance. reviewed closely by practitioners.
 C.-W. Lee / Journal of Business Research 60 (2007) 731–741 735

Table 1
Factor analysis results
Items Eigenvalue Proportion (%) Accumulated proportion (%) Naming Cronbach's α
Partners' technical capacities 12 0.92
5 4.653 38.775 38.775 Scientific knowledge (y1) 0.83
4 2.814 23.450 62.225 Firm's location (y2) 0.85
3 1.829 15.241 77.466 TMT international experience (y3) 0.82
Alliances structure 8 0.89
5 3.564 44.550 44.550 Horizontal integration (y4) 0.85
3 2.463 30.788 75.388 Vertical integration (y5) 0.80
Alliances type 8 0.93
5 3.426 42.825 42.825 Exploitation alliance (y6) 0.89
3 2.737 34.213 77.038 Exploration alliance (y7) 0.89
Alliance partners' relationship 11 0.95
5 4.545 41.318 41.318 Trust (y8) 0.92
2 2.332 21.200 62.518 Communication (y9) 0.90
2 1.053 9.573 72.091 Coordination (y10) 0.87
2 1.003 9.118 81.209 Shared value (y11) 0.94
Absorptive capacity 8 0.94
5 4.221 52.763 52.763 Potential absorptive capacity (y12) 0.87
3 2.058 28.298 81.061 Realized absorptive capacity (y13) 0.89
New ventures success 8 0.93
5 4.286 53.575 53.575 NPD performance (y14) 0.91
3 1.978 24.725 77.607 Commercial success (y15) 0.94
Firm size 4 0.96
2 2.458 61.250 61.250 Employees (x1) 0.90
2 1.008 25.200 83.450 Revenues (x2) 0.94

3.2. Conceptual framework Linear Structural Relations Model (LISREL). It uses factor
analysis to derive the observed variables.
In accord with literature, this study establishes a conceptual In this framework (Fig. 1), latent variables (and observed
framework and proposes six hypotheses to be tested using a variables) include the following: partners' technical capacities

Fig. 1. Conceptual framework.

736 C.-W. Lee / Journal of Business Research 60 (2007) 731–741

Table 2 for international affairs receive inquires. The result is 189 valid
Characteristics of individuals and firms in the sample responses (37.8%). The 189 strategic alliances in our sample
Characteristics Frequency split into 37 exploration and 159 exploitation alliances. Only 17
Job category CEOs 78 (41%) alliances are targeted towards both. All 159 exploitation
International manager 54 (29%) alliances are non-equity alliances, while the 37 exploration
Technical manager 27 (14%) alliances split into 24 non-equity and 13 equity alliances. The
Marketing manager 30 (16%)
average age of an alliance is more than 3 years old. Respondents
Annual sales of firm Less than USD 1 million 2 (1%)
USD 1–2 million 20 (11%) discuss whether their companies have ever engaged in strategic
USD 2–4 million 34 (18%) alliances. More specifically, respondents indicate the degree of
USD 4–6 million 32 (17%) their involvement by answering the above-mentioned questions
USD 6–8 million 54 (29%) on strategic alliances.
USD 8–10 million 42 (22%)
To evaluate the relative importance of different absorptive
More than USD 10 million 5 (3%)
Partner's home Europe (German, Italy, UK, etc.) 58 (31%) capacity, each respondent specifies the relative importance of
country USA 43 (23%) characteristics of alliance partners, structure of strategic
Japan 32 (17%) alliances, type of strategic alliances and relationship of alliance
Others (Korea, Taiwan, India, 56 (30%) partners on a 7-point Likert scale, indicating whether relative
Australia, etc.)
importance is very low (1) or very high (7). The question items
are, for example, (i) You can require new medicines/products
(scientific knowledge, firm's location, and TMT international because of alliance partners' patents. (ii) Owing to partners'
experience); alliances structure (horizontal and vertical integra- involvement in NPD, your new products are commercialized
tion); alliances type (exploitation and exploration alliances); more successfully. (iii) You can improve the level of self-
alliance partners' relationship (trust, communication, coordina- procurement through a horizontal integration (partners/compe-
tion and shared value); absorptive capacity (potential and realized titors). (iv) You can improve the R&D capacity through a
absorptive capacity) and new ventures success (NPD perfor- vertical integration (suppliers/buyers). Characteristics of the
mance and commercial success), and control variable (firm size). respondents are illustrated in Table 2, which shows a wide size
distribution among responding firms.
3.3. Data collection and sample
4. Results
This paper attempts to survey biotech firms located in
Taiwan (Taiwanese and non-Taiwanese), as well as a wide range 4.1. Descriptive statistics
of biotech applications involving biomedicine, bio-agriculture,
bio-food, bio-chemical, bio-environmental, and bio-service. In This study simultaneously employs correlation analysis to
order to grasp the comprehensive configuration of this industry, assess the relationships between the variables. The correlation
our research conducts in-depth interviews with three larger matrix indicates the relationship patterns of partner technical
biotech firms prior to the questionnaire investigation, and capacity, alliances structure, alliances type, alliance partners'
thereupon, sends the questionnaires to 500 companies listed in relationship, absorptive capacity, new venture success, and firm
“2002 Taiwan Bio Industry”. CEOs or the managers responsible size (see Table 3).

Table 3
Correlation matrix
Mean S.D. y1 y2 y3 y4 y5 y6 y7 y8 y9 y10 y11 y12 y13 y14 y15 x1 x2
y1 5.80 0.82 1
y2 5.65 0.81 0.55 1
y3 5.69 0.75 0.33 0.21 1
y4 5.63 0.78 0.29 0.23 0.57 1
y5 5.89 0.75 0.19 0.21 0.16 0.18 1
y6 5.68 0.74 0.31 0.33 0.25 0.19 0.44 1
y7 5.34 0.90 0.19 0.24 0.13 0.12 0.14 0.22 1
y8 5.31 0.86 0.20 0.23 0.13 0.16 0.19 0.30 0.39 1
y9 5.62 0.77 0.42 0.31 0.25 0.31 0.44 0.26 0.23 0.27 1
y10 5.62 0.70 0.26 0.24 027 0.31 0.45 0.36 0.16 0.32 0.59 1
y11 5.75 0.82 0.24 0.27 0.18 0.23 0.31 0.26 0.17 0.24 0.49 0.55 1
y12 5.50 0.82 0.42 0.38 0.32 0.20 0.24 0.24 0.32 0.22 0.44 0.41 0.34 1
y13 5.75 0.85 0.33 0.42 0.24 0.32 0.32 0.22 0.27 0.35 0.37 0.32 0.28 0.41 1
y14 5.62 0.78 0.34 0.25 0.19 0.27 0.27 0.37 0.34 0.41 0.35 0.28 0.29 0.39 0.51 1
y15 5.47 0.72 0.30 0.41 0.30 0.19 0.31 0.35 0.45 0.19 0.24 0.22 0.31 0.35 0.27 0.38 1
x1 5.13 0.71 0.36 0.33 0.12 0.14 0.30 0.10 0.14 0.08 0.27 0.14 0.05 0.22 0.24 0.21 0.23 1
x2 4.42 0.68 0.34 0.32 0.35 0.33 0.28 0.11 0.09 0.16 0.13 0.23 0.15 0.16 0.21 0.08 0.30 0.87 1
Level of significance: p b 0.001; p b 0.01; p b 0.05.
 C.-W. Lee / Journal of Business Research 60 (2007) 731–741 737

4.2. Measurement model These results may conclude that each of the six latent factors is
well defined.
In order to specify the measurement model, this study makes
the transition from factor analysis, in which the researcher has 4.3. Structural model results
no control over which variables describe each factor, to a
confirmatory mode, in which the researcher specifies which As Fig. 2 shows, path analysis indicates that six hypothesized
variables define each construct (factor). The manifest variables paths are supported at a significant level (less than 0.05) —
the respondents provide are “indicators” for the measurement partners' technical capacities, alliances structure, alliances type,
model, because they provide a means to measure, or indicate the alliance partners' relationship, absorptive capacity and new
latent constructs (factors). The most obvious difference between ventures success. In the initial hypothesis, this study finds that
the measurement model and factor analysis is that the former partners' technical capacity positively and significantly influ-
has a much smaller number of loadings and resembles the ences new ventures success (path coefficient = 0.67, t = 5.83).
exploratory mode of factor analysis. Researchers can specify a There is support for H1. Moreover, alliances structure is
measurement model for both exogenous constructs and positively related to absorptive capacity (path coefficient = 0.58,
endogenous constructs. t = 4.72), and new ventures success (path coefficient = 0.72,
Table 4 presents λx and λy coefficients from a LISREL t = 7.06). The results support H2 and H3. The alliances type
analysis of a hypothesized causal model of new venture success. positively influences new ventures success (path coeffi-
All the coefficients are moderately high, approximately 0.58 or cient = 0.63, t = 5.68). The findings support H4. Alliance
more. In addition, all the loadings are statistically significant. partners' relationship positively affects new ventures success
The strong statistically significant correlation between the factor (path coefficient = 0.54, t = 4.65). The findings support H5.
and their measures suggests the presence of convergent validity. Finally, a significant and positive relationship exists between
absorptive capacity and new ventures success (path coeffi-
cient = 0.35, t = 2.28). Hence, the findings support H6.
Table 4
Measurement model parameter estimates 4.4. Overall model fit
Latent variable Observed variable λx and λy Error Latent
variance variance Various measures evaluate the fit of the model (Bentler,
Partners' technical 0.279 1995; Sörbom and Jöreskog, 1982). Kelloway (1998) suggests
capacities that the use of chi-square test is reasonable when the study
Scientific knowledge (y1) 0.5877 0.516 involves a large sample. The first measure is the likelihood ratio
chi-square statistic. While the value has a statistical significance
Firm's location (y2) 1.0488 0.179
level above the minimum level of 0.05, the statistics support the
TMT international 0.5920 0.509
experience (y3) argument that the differences of the predicted and actual
Alliances structure 0.338 matrices are insignificant, indicative of an acceptable fit.
Horizontal integration (y4) 0.7165 0.392 However, as the chi-square was very sensitive to sample size,
the degree of freedom could be used as an adjusting standard to
Vertical integration (y5) 0.7265 0.355
judge whether chi-square was large or small (Sörbom and
Alliances type 0.246 Jöreskog, 1982). Therefore, in this study, when the chi-square
Exploitation alliance (y6) 0.8918 0.071 per degree of freedom is below six, this shows a reasonable fit,
Exploration alliance (y7) 0.8118 0.305 while a ratio between one and two is an excellent fit. The ratio of
Alliance partners' 0.117 the model is 2.029, indicating a fairly good fit. The model fit
relationship
assessment approach is involved, using several diagnostics to
Trust (y8) 0.8485 0.385
Communication (y9) 0.7657 0.475 judge the simultaneous fit of the measurement and structural
Coordination (y10) 0.8234 0.366 models to data collected for this study. The goodness-of-fit
Shared value (y11) 1.0064 0.071 index (GFI) is another measure LISREL provides. The adjusted
Absorptive 0.392 goodness-of-fit index (AGFI) is an extension of the GFI,
capacity
adjusted by the ratio of degrees of freedom for the proposed
Potential absorptive 0.8122 0.366
capacity (y12) model to the degrees of freedom for the null model. The GFI for
Realized absorptive 0.8863 0.266 the overall model is 0.96 and the AGFI is 0.93.
capacity (y13) Other types of fit measures include the Comparative-Fit
New ventures 0.316 Index (CFI), the Root Mean Square Residual (RMR), and the
success
Standardized Root Mean Square Residual (SRMR). This study
NPD performance (y14) 0.7781 0.397
Commercial success (y15) 0.7017 0.522 uses CFI to explain the difference between the model and the
Firm size – independent model without co-variables. The closer the value is
Employees (x1) 0.9518 0.105 to 1, the better the model fit. The RMR is the square root of the
Revenues (x2) 0.9882 0.342 mean of the squared residuals — an average of the residuals
Level of significance: p b 0.001. between observed and estimated input matrices. Owing to
738 C.-W. Lee / Journal of Business Research 60 (2007) 731–741

Fig. 2. Structural model results.

RMR's lack of standardization, however, comparison between H1 indicates that the impact of alliances upon the rate of new
coefficients is more difficult than with the standardized venture success might vary according to partners' technical
coefficients. Other diagnostics for this model include capacities. For example, partners with less international experi-
NFI = 0.95, CFI = 0.97, RMR = 0.022, SRMR = 0.018, and ence are less likely to act opportunistically than those with
RMSEA = 0.033. This measure is quite similar to the parsimo- abundant experience in biotechnology. In addition, developing
nious Normed Fit Index (NFI) and has a recommended the biotechnology industry relies on advances in scientific
acceptance level value greater than or equal to 0.90. Further, knowledge and its commercial exploitation, as Liyanage and
the model has a NFI value of 0.95, which means that 95% of the Gluckman (2004) mention. Concurrently, this suggests that the
observed measure covariance is explained by the composition future geography of biotechnology innovation will be limited to
model. The structural model results in Fig. 2 show that overall those locations with existing biotechnology research assets. This
model fit is within an acceptable level. study emphasizes partner' technical capacity as an important
factor in determining new ventures success.
5. Discussion Building from the results of H2 and H3, this paper argues that the
successful development of biotech new ventures requires effective
5.1. Conclusions organizational structures and the co-evolution of two types of
knowledge systems – scientific and innovative – for continuous
The rate of new ventures success is becoming an important growth of biotechnology firms. In addition, the research
competitive dimension in many industries (Stalk and Hout, illuminates the strategic alliances of biotech firms harnessing the
1990). This study finds reasonably strong support for all potential economic benefits of biotechnology networking, whether
hypotheses presented. Our research focuses are not only on horizontal or vertical integration. Our investigation indicates that
product development, but also on the speed of market respondents highly agree with this item: “You can improve the
penetration as a result of new venture success. This research R&D capacity through vertical integration.” Accordingly, firms
analysis looks at — the impact of strategic alliances on tend to withdraw from the product development path to discover,
absorptive capacity and the impact of new venture success with develop, and commercialize promising projects through vertical
different extents of absorptive capacity. integration, as a technology venture grows. As Lee (2003)
 C.-W. Lee / Journal of Business Research 60 (2007) 731–741 739

indicates, where M&A has technological motivation, post-M&A Although Taiwan's biotechnology still lags behind the USA,
technological performance of the acquiring firm improves. Europe, or Japan, the Taiwanese have been carrying out an
Strategic alliances are complex events with various motivations. extremely well coordinated effort to excel in this field. Taiwan's
While SMEs decision-makers choose an appropriate alliance biotech firms will intensify their successful approaches, learned
structure in compliance with their motivations, their new ventures from other alliances partners, to advance the development of
success will be upgraded. biotechnology, as well as to build strong international networks,
With regard to H4, our results find that exploration alliances improve strategic alliances efficiencies, and maximize global
predict products in development, which in turn predict competitiveness.
exploitation alliances. In addition, exploitation alliances predict In contrast to research institutes and universities focusing on
products on the market, indicating not only that there are basic research, biotech firms tend to specialize in particular
different motivations and goals for different types of alliances, research areas. Thus, specialization within the industry creates
but also that different alliance types proceed different outcomes incentives for biotech firms to enter into alliances with other
(i.e., new venture success or failure). Rothaermel's (2001) study firms to gain access to complementary technological know-
also indicates that incumbents exhibit a preference towards how. Likewise, specialization motivates pharmaceutical com-
alliances that leverage complementary assets (exploitation panies to gain access to financial resources, as well as
alliances) over alliances that focus on building new technolog- technological, market and manufacturing know-how (Mitchell
ical competencies (exploration alliances). and Singh, 1996). This is also the reason why Taiwan's biotech
Concerning H5, our conclusions find that turning good science SMEs enter alliances with biotechnology companies, rather
into commercially valuable biotechnology products not only than develop their own biotechnology operations.
requires specialized skills and knowledge, but also needs effective A large body of research on inter-organizational linkages
communication, coordination, shared values, and mutual-trust examines the benefits and costs that these linkages confer on
with counterparts. Marketplace realities are driven by a number of incumbents (Baum and Oliver, 1991; Powell, Koput and Smith-
human, cultural, and political factors that work within the domain Doerr, 1996). Baum et al. (2000) find that new ventures
of the business of biotechnology (Kung et al., 2000). Market share integrating potential competitors' innovative capacity may
of new ventures is important to investors and venture capitalists, perform more successfully than ventures relying on individual
people who keep the biotech industry alive and dynamic. effort. Powell et al. (1996) indicate that an innovative unit was a
Meanwhile, in developing a supply chain management via network rather than simply a specific firm in the biotech industry.
alliance, counterparts maintain close relationships in the first If a firm is unable to develop a learning network, it will be at a
place. They will obtain a larger market share by improving product competitive disadvantage. Strategic alliances provide an essential
quality and reducing costs through cooperation and collaboration. knowledge platform. In general, large biotech companies do not
The result of H6 suggests that a counterpart's absorptive readily transfer their technology or know-how to other (poten-
capacity in a strategic alliance will influence new ventures tially rival) firms. While SMEs are acquiring new technological
success. As noted, biotech firms usually join multiple alliances knowledge, strategic alliances provide a good choice.
to gain access to knowledge and skills across the various phases
of their value chain. Building upon the dynamic capabilities 5.3. Limitations
view of the firm, the absorptive capacity is an important factor
of SMEs' potential and realized capacities. The firm's potential Interpret these results with caution because of the study's
and realized capacities can differentially influence the creation, limitations. While several important benefits occur for single
sustenance, and advancement of its competitive advantage industry studies, the fact that this research's focus on the
(Zahra and George, 2002). biotechnology industry may limit the generalizability of the
findings. The current results reflect the nature of the measures
5.2. Managerial implications used, especially those related to new ventures success. This
study uses “firm size” as a control variable. However, factors of
The contribution of this paper lies in creating links between patents held and R&D intensity are also measures of the
inter-firm cooperation as a mechanism of new ventures to adapt observed variable “scientific knowledge”. Future studies might
to radical technological change, firm innovative output, and benefit from exploring other indicators of new ventures success.
industry and firm performance in the post-innovation period. For instance, patents held, R&D intensity, partner size, and the
The results support the notion of incumbent survival through origin of the partner's home country could be used as control
complementary assets (Tripsas, 1997), and the importance of variables. The use of alternative control variables should help to
differentiating between technological and market-related capa- establish the robustness of the current findings.
bilities when adapting to a new technology (Mitchell, 1992).
Insights into the importance of inter-firm collaboration for 5.4. Future research directions
research are provided based on extensive empirical analyses of
Taiwan's biotech new ventures. Establishing inter-firm collab- Taiwan's biotech firms are much smaller, and often refrain
orative relationships is considered to be vital as commercial from providing access to internal scientific resources, as well as
biotechnology research becomes more independent from frequently regard joint government-sponsored R&D as unavoid-
academic research. able. As a consequence, it is necessary to overcome SMEs'
740 C.-W. Lee / Journal of Business Research 60 (2007) 731–741

weakness in basic R&D using international alliances as a strategy. George G, Zahra SA, Wheatley KK, Kan R. The effects of alliance portfolio
However, clinical studies are currently being conducted in characteristics and absorptive capacity on performance: a study of
biotechnology firms. J High Technol Managem Res 2001;12:205–26.
collaboration with universities and hospitals. In Taiwan's case, Grandstrand O, Sjolander SE. Managing innovation in multi-technology
biotech SMEs sometime ally with hospitals such as the National corporations. Res Policy 1990;19(1):35–60.
Taiwan University Hospital and Veterans General Hospital in Grant RM. Toward a knowledge-based theory of the firm. Strateg Manage J
Taipei. Several joint research programs are also in progress with 1996;17(Special Issue):109–23.
Hagedoorn J. Understanding the rationale of strategic technology partnering:
international institutes in the United States, Canada, Australia and
interorganizational modes of cooperation and sectoral differences. Strateg
other regions. Manage J 1993;14:371–85.
This study has focused on inter-firm collaborative method. Hair J, Anderson R, Tatham R, Black W. Multivariate data analysis. 2nd ed. New
Yet the general outline of this research could be applied to other York: Macmillan; 1992.
forms of strategic alliances. For example, this research could be Heide JB, John G. Do norms matter in marketing relationships? J Mark
extended to examine inter-organization alliances such as 1992;56(2):32–44.
Homans G. Social behavior as exchange. Am J Sociol 1958;62:597–606.
governmental research institutes, universities, and non-profit Kauser S, Shaw V. International strategic alliances: the impact of behavioral
organizations (e.g. hospitals). As a second suggestion, if biotech characteristics on success. J Euro-Mark 2001;10(1):71–98.
SMEs become to larger firms or multinational enterprises, they Kelloway EK. Using LISREL for structural equation modeling: a researcher's
may use a merger and acquisition (M&A) method to exploit guide. Thousand Oaks, CA: Sage Publications; 1998.
partners' technical capacity. Much could be learned from the Kim L. Crisis construction and organizational learning: capability building in
catching-up at Hyundai Motor. Organ Sci 1998;9:506–21.
formulation of M&A know-how construct and its antecedents Kogut B, Shan W, Walter G. The make-or-cooperate decision in the context of
and effects on firm performance in light of previous researches an industry network. In: Nohria N, Eccles RG, editors. Networks and
(Pennings et al., 1994; Simonin, 1997). organizations: structure, form and action. Boston: HBS Press; 1992.
Kotabe M, Swan KS. The role of strategic alliances in high-technology new
Reference product development. Strateg Manage J 1995;16:621–36.
Koza MP, Lewin AY. The co-evolution of strategic alliances. Organ Sci
Anderson EW, Fornell C, Donald RL. Customer satisfaction, market share and 1998a;9:255–64.
profitability: findings from Sweden. J Mark 1994;58(3):53–66. Koza MP, Lewin AY. The co-evolution of strategic alliances. Organ Sci
Anderson JC, Narus JA. A model of distributor firm and manufacturer firm 1998b;9:255–64.
working partnerships. J Mark 1990;24(1):42–58. Kung SD, Wong JT, Ip NY. Biotechnology finds its way to Hong Kong. Int J
Arrow KJ. Economic welfare and the allocation of resources for invention. In: Biotechnol 2000;2(4):355–63.
Nelson RR, editor. The rate and direction of inventive activity: economic Lane PJ, Lubatkin M. Relative absorptive capacity and interorganizational
and social factors. NJ: Princeton University Press; 1962. learning. Strateg Manage J 1998;19:461–77.
Barley SR, Freeman J, Hybels RC. Strategic alliance in commercial biotechnology. Lane PJ, Salk JE, Lyles MA. Absorptive capacity, learning, and performance in
In: Nohria N, Eccles RG, editors. Networks and organizations. Cambridge, international joint ventures. Strateg Manage J 2001;22:1139–61.
MA: Harvard Business School Press; 1992. Lee CW. M&A strategy and technological innovation performance of high-
Baum JAC, Calabrese T, Silverman BS. Don't go it alone: alliance network technology firms. Int J Bus Strateg 2003;4(2):54–75.
composition and startups' performance in Canadian biotechnology. Strateg Lee Jr KB, Burrill GS. Biotech '95 reform, restructure, renewal. Palo Alto, CA:
Manage J 2000;21(3):267–94. Ernst & Young; 1994.
Baum JAC, Oliver C. Institutional linkages and organizational mortality. Adm Lee YH, Kim D. Development of application execution environment in
Sci Q 1991;36(2):187–218. partitioned systems. Final research report to Honeywell International;
Bentler PM. EQS structural equation program manual. Los Angles: BMDP 1999. Dec.
Statistical Software; 1995. Lerner J, Merges RP. The control of technology alliances: an empirical analysis
Biggadike RE. Corporate diversification: entry, strategy and performance. of the biotechnology industry. J Ind Econ 1998;46:125–56.
Boston, Mass.: Harvard University Press; 1979. Liyanage S, Gluckman P. The determinants of biotechnology innovative
Blau PM. Exchange and power in social life. New York: John Wiley and capability: the dynamics of knowledge and marketplace. Int J Biotechnol
Sons; 1964. 2004;6(2–3):281–300.
Bogner WC, Thomas H, McGee J. A longitudinal study of the competitive Morgan RM, Hunt SD. The commitment–trust theory of relationship marketing.
positions and entry paths of European firms in the U.S. pharmaceutical J Mark 1994;58:20–38.
market. Strateg Manage J 1996;17(2):85–107. McCutchen Jr WW, Swamidass PM. The popular strategic configuration of
Chatman EA. Channels to a larger social world: older women staying in contact R&D aggressiveness and research funding in emerging biotech firms: a
with the great society. Libr Inf Sci Res 1991;13(3):281–300. cluster analysis. J High Technol Managem Res 1994;5(2):213–32.
Cohen WM, Levinthal DA. Absorptive capacity: a new perspective on learning McCutchen Jr WW, Swamidass PM. Motivations for strategic alliances in the
and innovation. Adm Sci Q 1990;35:128–52. pharmaceutical/biotech industry: some new findings. J High Technol Managem
Coombs JE, Deeds DL. International alliances as sources of capital: evidence Res 2004;15:197–214.
from the biotechnology industry. J High Technol Managem Res 2000;11 Mitchell W. Are more good things better, or will technical and market capabilities
(2):235–53. conflict when a firm expands? Ind Corp Change 1992;1: 327–46.
Cooper RG, Kleinschmidt EJ. New product performance: keys to success, Mitchell W, Singh K. Survival of businesses using collaborative relationships to
profitability and cycle time reduction. J Market Manag 1995;11:315–37. commercialize complex goods. Strateg Manage J 1996;17(3):169–95.
Doney PM, Cannon JP. An examination of the nature of trust in buyer-seller Mohr J, Spekman RE. Characteristics of partnership success: partnership attributes,
relationships. J Mark 1997;61:31–55. communication behavior, and conflict resolution techniques. Strateg Manage J
Dwyer FR, Schun PH, Oh S. Developing buyer-seller relationships. J Mark 1994;15(2):135–52.
1987;51(2):11–27. Mowery DC, Oxley JE, Silverman BS. Technological overlap and interfirm
Etgar M. Sources and types of intrachannel conflict. J Retail 1979;55(1):61–78. cooperation: implications for the resource-base view of the firm. Res Policy
Gassel K, Pascha W. Milking partners or symbiotic know-how enhancement? 1998;27:507–23.
International versus national alliances in Japan's biotech industry. Int Bus Mowery DC, Rosenberg N. Technology and the pursuit of economic growth.
Rev 2000;9:625–40. Cambridge: Cambridge University Press; 1989.
 C.-W. Lee / Journal of Business Research 60 (2007) 731–741 741

Pennings JM, Barkema H, Douma S. Organizational learning and diversification. Sivadas E, Dwyer FR. An examination of organizational factors influencing
Acad Manage J 1994;37(3):608–40. new product success in internal and alliance-based processes. J Mark
Pisano GP. The R&D boundaries of the firm: an empirical analysis. Adm Sci Q 2000;64(1):31–49.
1990;35:356–82. Sörbom D, Jöreskog KG. Recent developments in LISREL: automatic starting
Powell WW. Learning from collaboration, knowledge and networks in the values. University of Uppsala, Department of Statistics; 1982.
biotechnology and pharmaceutical industries. Calif Manage Rev 1998;40(3): Stalk G, Hout TM. Competing against time. New York: Free Press; 1990.
228–40. Szulanski G. Exploring internal stickiness: impediments to the transfer of best
Powell WW, Koput KW, Smith-Doerr L. Interorganizational collaboration and practice within the firm. Strateg Manage J 1996;17(Winter Special Issue):
the locus of innovation: networks of learning in biotechnology. Adm Sci Q 27–43.
1996;41:116–56. Tripsas M. Unraveling the process of creative destruction: complementary assets
Price JL, Mueller CW. Handbook of organizational measurement. Marshfield, xand incumbent survival in the typesetter industry. Strateg Manage J
MA: Pitman; 1986. 1997;18:119–42.
Reuber AR, Fischer E. The influence of the management team's international Van den Bosch F, Volberda H, de Boer M. Coevolution of firm absorptive
experience on the internationalization behaviors of SMEs. J Int Bus Stud capacity and knowledge environment: organizational forms and combinative
1997;28:807–25. capabilities. Organ Sci 1999;10(5):551–68.
Rothaermel FT. Complementary assets, strategic alliances, and the incumbent's Young-Ybarra C, Wiersema M. Strategic flexibility in information technology
advantage: an empirical study of industry and firm effects in the biopharmaceu- alliances: the influence of transaction cost economics and social exchange
tical industry. Res Policy 2001;30:1235–51. theory. Organ Sci 1999;10(4):439–59.
Rothaermel FT, Deeds DL. Exploration and exploitation alliances in biotechnology: Zahra SA. Technology strategy and new venture performance: a study of
a system of new product development. Strateg Manage J 2004;25(3):201–17. corporate-sponsored and independent biotechnology ventures. J Bus Venturing
Sambharya RB. Foreign experience of top management teams and international 1996;11:289–321.
diversification strategies of U.S. multinational corporations. Strateg Manage Zahra SA, George G. Absorptive capacity: a review, reconceptualization and
J 1996;17:739–46. extension. Acad Manage Rev 2002;27(2):185–203.
Simonin BL. The importance of collaborative know-how: an empirical test of Ziegelbauer K, Farquhar R. Strategic alliance management: lessons learned from
the learning organization. Acad Manage J 1997;40(5):1150–74. the Bayer-Millennium collaboration. Drug Discov Today 2004;9(20):864–8.