THE ADOPTION OF AGENCY BUSINESS ACTIVITY, …haiyang/SMJ-AGency.pdf · THE ADOPTION OF AGENCY...

Strategic Management JournalStrat. Mgmt. J., 23: 469–490 (2002)

Published online 1 February 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/smj.233

THE ADOPTION OF AGENCY BUSINESS ACTIVITY,PRODUCT INNOVATION, AND PERFORMANCE INCHINESE TECHNOLOGY VENTURES

HAIYANG LI*1 and KWAKU ATUAHENE-GIMA2

1 Lowry Mays College and Graduate School of Business, Texas A&M University,College Station, Texas, U.S.A.2 Department of Management, City University of Hong Kong, Kowloon, Hong Kong

This study examines the roles of firm characteristics and environmental factors in the formationof interfirm alliances. Specifically, we examine the dual role of these groups of factors asinducements and opportunities for Chinese high-technology new ventures (HTNVs) in theiradoption of agency business activity, a downstream type of alliance involving marketing anddistribution of the products of foreign firms. Results suggest that both internal and external factorsare related to the adoption of agency business activity but the inducement and opportunity valueof environmental uncertainty may be dampened by institutional support provided to HTNVs.Further, we find that successful agency business activity is positively related to new ventureperformance but negatively related to its product innovation efforts. Theoretical and managerialimplications are discussed. Copyright 2002 John Wiley & Sons, Ltd.

INTRODUCTION

Interfirm linkages or cooperative alliances havereceived considerable attention from strategic man-agement researchers in recent years. Studies indi-cate that such alliances provide several benefitsto firms including helping them to develop newtechnology and improve technical skills (Cohenand Levinthal, 1990; Dowling and McGee, 1994;Eisenhardt and Schoonhoven, 1996; Feeser andWillard, 1990; Zahra and Covin, 1993), learn newmanagement skills (Ahuja, 2000; Kraatz, 1998)and develop innovative products (Grenadier andWeiss, 1997).

In parallel with the increased interest ininterfirm alliances, scholars are paying attentionto strategies of high-technology new ventures(HTNVs). Studies indicate that these firms facesignificant problems aptly termed ‘liability of

Key words: agency business activity; product innovation;new technology ventures; China*Correspondence to: H. Li, Lowry Mays College and GraduateSchool of Business, Texas A&M University, College Station,TX 77843-4221 U.S.A.

newness’ by Stinchcombe (1965), including lackof adequate knowledge of their environments,new product development experience, as well asmanagerial and financial resources (Feeser andWillard, 1990; Shan, 1990; Zahra and Covin,1993). Thus, they are highly vulnerable toenvironmental selection and have high failurerates with less than half of these firms lastingfor 5 years (O’Shea and Stevens, 1998). Severalstudies have suggested that interfirm alliance is apotential strategy that HTNVs use to offset thisliability of newness (Dowling and McGee, 1994;Eisenhardt and Schoonhoven, 1990). Empiricalanalysis of HTNVs in emerging technologies hasfound that alliances play an important role in theirdevelopment of new products and technologies(Shan, 1990).

In general, two broad categories of explanationshave been offered for firms’ entry into interfirmalliances (Ahuja, 2000). The first suggests thatfirms enter into alliances to obtain access to neededresources, to learn new skills, and to enhancecompetitive parity (Eisenhardt and Schoonhoven,1996; Kogut, 1988; Powell and Brantley, 1992;

Copyright 2002 John Wiley & Sons, Ltd. Received 1 July 1999Final revision received 25 October 2001

470 H. Li and K. Atuahene-Gima

Varadarajan and Cunningham, 1995). Drawn onresource-based theory (Barney, 1991), this dom-inant view of alliance formation suggests thatthe need for resources acts as an inducement orincentive for firms to collaborate with other firms.The second category of explanation is the oppor-tunity to collaborate perspective (Ahuja, 2000).This perspective suggests that the firm’s abilityto enter into alliances with other firms is deter-mined by its ability to attract potential partners.Such ability is reflected in the resources and otherpotential advantages that the firm is capable ofoffering to potential alliance partners (Granovetter,1985; Gulati, 1998; Kraatz, 1998; Shan, Walker,and Kogut, 1994; Stuart, 2000). In particular,researchers have emphasized the role of social cap-ital and resources available through social networkand institutional ties. For example, Eisenhardt andSchoonhoven (1996) found that large managementteams have greater opportunity to form alliancesbecause of their extensive social connections andrelationships which offer opportunities for find-ing alliance partners or being found as an alliancepartner. Ahuja (2000) suggested that alliance expe-rience offers a core opportunity for firms to enterinto new alliances.

While prior studies have advanced our under-standing of the formation of interfirm linkages,several gaps remain. First, few studies have adop-ted the dual inducement–opportunity frameworknoted above in examining the formation of inter-firm alliances. Moreover, they only provide a lim-ited view of alliance formation. As noted by Ahuja(2000 : 318), ‘any explanation of linkage formationbehavior must account for the actor’s inducementto form linkages and his/her opportunities to col-laborate.’ Second, most prior research has focusedon established firms with only a few focusingattention on new ventures. Yet, new ventures playa significant role in economic and social develop-ment. In particular, the development of HTNVs isviewed as a revitalization tool for developed mar-ket economies and as a driver of economic trans-formation in transitional economies (Bruton andRubanik, 1997; McDougall and Robinson, 1990;Zahra and Covin, 1993; Zhao and Aram, 1995).Yet, little research has examined the inducementsand opportunities that these firms have in forminginterfirm alliances. Indeed, this shortcoming in theliterature is particularly true for new ventures intransitional economies.

Third, previous research on alliance formationamong new ventures has focused primarily onthe upstream activities of the value chain involv-ing technology and new product development(Eisenhardt and Schoonhoven, 1996; Shan, 1990).Alliances that involve downstream activities ofthe value chain such as the adoption of agencybusiness activity involving marketing and distribu-tion of products of alliance partners has receivedlimited attention. Compared to technology andproduct development alliances such downstreamalliances involve lower risk. HTNVs in both devel-oped and transitional economies adopt this strategyas a complement to their business activities. Forexample, companies such as Dell Computers serveas agents for others’ products, in order to providea more complete product line and accessories tocustomers. Wintak Technology, one of the fast-growing Chinese computer companies, helps ASTsell networking products and computers in China.Bruton and Rubanik (1997) observe that it is com-mon for Russian HTNVs to import and wholesalethe products of foreign firms.

Finally, we could not locate any study thathas linked this type of alliance strategy with theproduct innovation and performance of HTNVs.Interfirm alliances are thought to influence productinnovation and performance because they provideopportunities for resource acquisition, learning,and legitimacy (Dyer and Singh, 1998; Kraatz,1998). However, empirical research evidence pro-vides conflicting results (Kotabe and Swan, 1995).We believe that given the resource limitations ofHTNVs and the potential attractiveness of agencybusiness activity to these firms, the relationshipbetween this strategy and product innovation andperformance of these firms warrants systematicempirical research.

In this study, we contribute to the literature byexamining the above gaps. Specifically, we followthe recent works of Ahuja (2000) and Eisenhardtand Schoonhoven (1996) by using the induce-ment–opportunities framework to examine bothexternal and internal factors which drive the adop-tion of agency business activity by a sample ofHTNVs in China. Formally, we define the adoptionof agency business activity as an alliance strategyby which a firm markets and distributes a foreignfirm’s products and services. We further exam-ine the relationship between the success of theagency business (defined as the proportion of aHTNV’s sales attributable to the agency business

Copyright 2002 John Wiley & Sons, Ltd. Strat. Mgmt. J., 23: 469–490 (2002)

Agency Business Activity, Product Innovation, and New Ventures 471

activity) and new venture performance and prod-uct innovation efforts. In the next section, wedevelop hypotheses based on previous theory andresearch. This is followed by a discussion of theresearch design and methods. Next, research find-ings are presented and evaluated. Finally, manage-rial implications and directions for future researchare explored.

THEORETICAL BACKGROUND ANDHYPOTHESES

Several theoretical frameworks have been advan-ced to explain the motives underlying the entry offirms into alliances (Kogut, 1988; Varadarajan andCunningham, 1995). The inducement–opportuni-ties framework, which we adopt here, is informedby resource-based and social capital theories. Theresource-based view suggests that firms are en-dowed with resources but do also need additionalspecific resources in order to compete effectivelyin particular markets (Barney, 1991). This needfor resources is thought to be a major rationalebehind alliance formation. Alliance formation pro-vides tangible resources such as financial capitaland technical skills, and intangible resources suchas social position and market reputation (Dyer andSingh, 1998; Eisenhardt and Schoonhoven, 1996;Nahapiet and Ghoshal, 1998). On the other hand,firms’ alliance formation is also related to theirpossession of resources, particularly those that aredifficult for the partners to create on their ownor obtain from the market (Ahuja, 2000). Forexample, the social capital resources held by firmsmay increase their attractiveness to potential part-ners and open opportunities for them to enter intoalliances (Gulati, 1998).

In alliances between firms from transitional anddeveloped economies, partners differ with respectto the specific resource needs that motivate theformation of the alliance. In general, resourcesof particular interest to firms from transitionaleconomies include financial capital, technical andmanagerial capabilities and reputation, particu-larly that for quality (Hitt et al., 2000). This the-oretical argument may be particularly relevantamong HTNVs. Compared with large and estab-lished firms HTNVs tend to have relatively lim-ited resources (McDougall and Robinson, 1990;Stinchcombe, 1965). Thus, an HTNV’s induce-ment to enter into agency business activity can be

ascribed to its need for resources (Shan, 1990).Through adopting such an activity, an HTNVmay effectively manage its dependence on externalenvironments (Pfeffer and Salancik, 1978). Likenew ventures in other economies those in Chinahave great difficulty in obtaining critical resourcesfrom the domestic market, particularly from banksand factor markets which are severely underde-veloped. For instance, only 3 percent of the 901firms listed on the China stock exchange as ofJune 1999 were non-governmental firms such asHTNVs. Further, of the 300 billion RMB (approx-imately U.S. $38 billion) loans granted by statebanks to Chinese firms in 1998 only 5 percentwere to non-governmental firms such as HTNVs(China Economic News, 2000). In addition, lim-itations with regard to the infrastructure (e.g.,transportation and communication) in China cre-ate challenges for HTNVs’ resource acquisition.This situation suggests that lack of resources couldbe related to the formation of agency businessalliances by HTNVs in China.

Potential partners from developed market econo-mies have particular interest in resources thatinclude local market knowledge and access, andthe social connections that would enable them toleverage their own resources. Hence, they tend toselect alliance partners that have complementaryand unique competences, local market knowledgeand access as well as willingness to share suchexpertise (Hitt et al., 2000). It follows that aHTNV enters into agency business activity becauseit has resources needed by the foreign partners andcan act proactively to respond to the resource needsof the partners. As Stuart (2000 : 793) argued, notall potential alliance partners are of equal valueand those well-endowed firms ‘are types of alliancepartners that can produce the best ex post resultsfor their associates.’ For example, Eisenhardt andSchoonhoven (1996) found that alliances formwhen firms are in strong social positions such aswhen they are led by large, experienced, and well-connected top management teams. Gulati (1995)found that previously allied firms were likely toengage in further alliances because they werefamiliar with each other in terms of goals, needs,skills, and capabilities. The possession of localknowledge and social capital may be of greatsignificance in China, where social connections arecritical facilitators of business activities.

With the above considerations in mind, weintegrate the two theoretical perspectives and



examine the relationship between external factors(environmental uncertainty, institutional support,and perceived industry growth) and internal factors(venture size, R&D expenditure, and productdevelopment alliance) and the HTNV’s adoptionof agency business activity.

External factors and the adoption of agencybusiness activity

Environmental uncertainty refers to the degree ofuncertainty in terms of products, markets, andcompetitive behavior perceived by management ofHTNVs in China. It has been viewed as the mostrelevant environmental characteristic that affectsa firm’s decision making (Dess and Beard, 1984;Downey, Hellriegel, and Slocum, 1975; Miller,1987). From a resource-based perspective, whenthe market is uncertain and competitive, HTNVsbecome vulnerable to the external environmentsince they have difficulties in raising capital andother resources. In addition, an uncertain andcompetitive environment presents increased per-formance risk for HTNVs. Forming alliances withother firms provide a potentially viable means ofdealing with such situations for several reasons.First, by increasing communication and informa-tion sharing agency business activity with for-eign firms may mitigate the competitive intensitythat HTNVs face. These alliances may enhancethe learning and adaptive response of the HTNV(Kraatz, 1998) through handling foreign firms’products, customer interaction and feedback pro-cesses. Second, agency business activity focuseson the downstream activities (e.g., marketing andsales) in the value chain. Consequently, it mayhelp HTNVs build their competencies and skillsin marketing and distribution to support their ownproducts.

Third, given environmental uncertainty, theadoption of agency business activity may not onlyprovide Chinese HTNVs with tangible resourcesbut also intangible resources such as reputationfor quality and legitimacy in the Chinese market(cf. Nahapiet and Ghoshal, 1998). For example,given the relatively low level of technologicalsophistication in China, customers equate highquality with products of foreign firms. Thus,agency business activity may not only enhance therevenues of new ventures, but also their reputationfor high-quality products. These arguments suggestthat adoption of agency business activity could

be a risk reduction strategy for HTNVs incomparison with product innovation. In the contextof firms in transitional economies, Peng and Heath(1996) following North (1990) have advancedsimilar arguments for entry into interorganizationalrelationships. Eisenhardt and Schoonhoven (1996)argued that new ventures enter into alliancesto enhance their legitimacy and therefore seekpartners with strong intangible assets such asreputation. This motivation is likely to be strongfor new ventures when the environment isuncertain. In a study of entrepreneurial firms inthe biotechnology industry, Shan (1990) foundthat high environmental uncertainty was associatedwith alliance formation.

Foreign firms entering China may have partic-ular concerns about the level of uncertainty inwhat is generally regarded as a highly complexand difficult to understand marketplace. Therefore,the uncertain Chinese market environment may notonly act as an inducement for alliance formationbut also as an opportunity for collaboration withthe Chinese HTNVs. The logic for this propositionis that the high uncertainty of the Chinese environ-ment and its cultural impediments offer significantchallenges to would-be foreign market entrants.Hence, HTNVs in China become more valuableto potential foreign partners because they havealready developed relationships with key stake-holders in the market and with key governmen-tal agencies. These social relationships representresources and capabilities that may be underuti-lized and therefore can be exploited through entryinto agency business activity (Kraatz, 1998). Thus,for the foreign firm, the Chinese HTNVs haveopportunity to leverage these social resources tolower the level of uncertainty and risk in market-ing and distributing their products. Thus from boththe inducement and opportunity perspectives, weposit that:

Hypothesis 1: Environmental uncertainty isrelated positively to the adoption of agencybusiness activity by HTNVs in China.

Moderating role of institutional support

The above line of reasoning suggests that fac-tors that ameliorate the resource constraints of theHTNVs may influence managerial perceptions ofrisks from environmental uncertainty such that itsrelationship with the firm’s adoption of agency



business activity may be weakened. One suchfactor is institutional support, which refers to theextent to which administrative institutions (e.g.,government departments) provide initial and con-tinuing financial, management, and technical sup-port for HTNVs. The support provided by gov-ernment institutions aims to help HTNVs conductproduct and technology innovations because thesefirms are viewed as the window of the developmentin Chinese high-technology industries. Such sup-port is important for HTNVs because they face sig-nificant problems in raising capital and other finan-cial resources. This problem could be exacerbatedfor Chinese HTNVs in an uncertain environmentbecause, as Simerly and Li (2000) have found,firms have limited potential for raising funds inuncertain environments. Hence, for these firmsgovernment financial support becomes an impor-tant, if not a prerequisite, means of obtaining cap-ital and other resources (Nee, 1989). For example,a report by Beijing High-Technology Experimen-tal Zone (BHEZ) Office (BHEZ Office, 1995) inChina indicated that 76 percent of HTNVs withinthe Zone obtained initial funding from their admin-istrative agencies.

The preceding arguments suggest that institu-tional support may alleviate the liability of new-ness of HTNVs and thus has impact on how man-agers of these firms frame the risks associatedwith environmental uncertainty. With higher insti-tutional support these firms are likely to perceivethe environment as less threatening to their firms’success and survival. This is likely to lower theinducement and opportunity value of environmen-tal uncertainty in the adoption of agency businessactivity. Therefore, we posit that:

Hypothesis 2: The positive relationship betweenenvironmental uncertainty and the HTNVs’adoption of agency business activity is weakerwhen institutional support is higher.

Perceived industry growth is another environ-mental characteristic that may be related to theHTNVs’ adoption of agency business activity. Itrefers to the degree of management perceivedgrowth of their principal industry within the last3 years or since founding (if the venture is lessthan 3 years old). Industry growth is an importantindicator of industry structure, representing a keycomponent of market attractiveness for new ven-tures (McDougall et al., 1994; Porter, 1980). As

Porter (1980) argues, in a rapid growth industrythe entry of new ventures will provoke less retal-iation by incumbent firms. Similarly, Miller andCamp (1985) suggest that high market growth canpotentially reduce the effect of competitive pres-sures on new ventures. Thus, relative to the man-agers of HTNVs in low growth industries, thosein high growth industries may not feel compelledto allocate resources into such activities as agencybusinesses given the significant potential for suc-cess in their core business of product innovation.Instead, they would rather like to reinvest theirresources in long-term in-house product innova-tion to keep pace with changes in the industry(McDougall et al., 1994).

Industry growth also indicates environmentalmunificence: the extent to which the resourcesrequired by the firm are available in the marketenvironment (Pfeffer and Salancik, 1978). Indeed,empirical research has found that venture capital-ists prefer to invest in new ventures in high growthindustries (MacMillan, Siegel, and Narasimha,1985; Sandberg and Hofer, 1987). Thus, fromthe resource-based perspective, HTNVs are morelikely to obtain resources from high growth indus-tries than from low growth industries (Hannan andFreeman, 1977). When managers face high marketgrowth at founding and expect this to continue,their ability to predict and obtain successful out-comes increases, thus reducing the incentive toenter into alliance relationships. This discussionsuggests that higher perceived industry growth hasless power in inducing the adoption of agencybusiness activity. In support of these argumentsDickson and Weaver (1997) found that there is agreater tendency for firms to enter into alliancerelationships when faced with low growth marketsbecause managers become less certain of the futuresuccess of their firms in such environments. Eisen-hardt and Schoonhoven (1996) also suggested thatgiven the munificent conditions in high growthindustries firms are less likely to form alliancesbecause they have little need for resources thatsuch alliances might bring.

Competing arguments to the above one can alsobe made. Firms in high growth industries mighthave greater need for resources and greater incen-tives to find new resources or to lever existingresources (e.g., lever links with customers, oppor-tunities to learn related activities from a tech-nologically superior partner rather than develop-ing on its own). Thus, these firms might have a



greater need to pursue alliances such as agencybusiness activity. This would be particularly thecase in China, where there are limited availableresources for HTNVs. From the opportunity tocollaborate perspective, high perceived industrygrowth might be highly attractive for undertakingagency business activities because foreign firmsare attracted to high growth markets. For example,recent research suggests that industry attractive-ness is a critical variable considered by firms fromdeveloped countries in choosing partners fromemerging economies (Hitt et al., 2000). Thus per-ceived industry growth could be particularly fruit-ful for an HTNV with knowledge and networks inChina, because the knowledge and networks pro-vide a foundation for exploiting the opportunityopen to foreign firms. Clearly, theoretical argu-ments can be made in support of both negative andpositive relationships between perceived industrygrowth and the adoption of agency business activ-ity. Given the lack of prior research in the contextof our study, we offer the following competinghypotheses:

Hypothesis 3a: Perceived industry growth isrelated negatively to the adoption of agencybusiness activity by HTNVs in China.

Hypothesis 3b: Perceived industry growth isrelated positively to the adoption of agency busi-ness activity by HTNVs in China.

Internal factors and the adoption of agencybusiness activity

As discussed earlier, to enter into interfirm alli-ances HTNVs not only must have need for resour-ces, they must also possess resources that areattractive to potential alliance partners (Ahuja,2000; Eisenhardt and Schoonhoven, 1996). In thisstudy we focus on internal factors such as venturesize, R&D expenditure, and product developmentalliance, because they reflect HTNVs’ potentialability to provide value to potential alliance part-ners thus offering opportunities to collaborate.

Venture size

A venture’s size reflects the availability of resour-ces and therefore the firm’s capacity to providebenefits to potential alliance partners. Social capital

perspective suggests that large ventures have rel-atively sufficient resources, scale, and the socialnetworks to provide value to their alliance partners(Hagedoorn and Schakenraad, 1994). It followsthat larger HTNVs have more opportunity to attractand engage foreign firms for agency business activ-ities than smaller ones. Some empirical findingssupport our argument. For example, Powell andBrantley (1992) and Gulati (1999) found that thefrequency of interfirm cooperation is positivelyrelated to firm size. From the foreign firms’ pointof view, larger HTNV size may reflect greatersocial legitimacy and connections with the rightadministrative agencies as well as larger coverageof the Chinese market. Such social capital benefitsassociated with HTNVs’ local connections are crit-ical for foreign firms’ success because connections(e.g., guanxi) are thought to be important sub-stitutes for formal institutional support in China.Therefore, from an opportunity to collaborate per-spective, venture size is probably related positivelyto HTNVs’ adoption of agency business activity.

However, conflicting findings also exist. Forexample, Shan and his colleagues (Shan, 1990;Shan, Walker, and Kogut, 1994) found that largerfirms are less likely to form cooperative relation-ships in commercializing new technology. Fromthe resource-based theory viewpoint, this findingindicates that larger size reflects the availabilityof sufficient resources and therefore lowers theinducement to enter into alliances (Ahuja, 2000). Itfollows that larger HTNVs might be perceived tohave less need for an agency alliance than smallerfirms due to the possession of additional resources.Although both positions are plausible, in the cur-rent study there is substantial reason to believethat larger size demotivates alliance formation. Thelogic is that the prime objective of firms in a transi-tional economy for alliance formation is to acquireneeded resources (Nee, 1989; Peng and Heath,1996). Consequently, if large size reflects the avail-ability of sufficient resources, it will be negativelyrelated to the adoption of agency business activity.

Hypothesis 4: Venture size is related negativelyto the adoption of agency business activity byHTNVs in China.

R&D expenditure

This construct is defined as HTNVs’ investment inR&D as percentage of sales over the last 3 years



or since founding (if it is less than 3 years old).R&D expenditure indicates the venture’s ability tocollaborate because it reflects its commitment toproduct innovation (Miller, 1987) and its techni-cal competence and learning capacity (Cohen andLevinthal, 1990; Lane and Lubatkin, 1998). Forforeign firms which attempt to enter into alliances,R&D expenditure represents a simple but manifestindication for them to assess the Chinese HTNVs’technical competence. Because of the technolog-ical complexity of the products for distribution,the HTNVs’ technical competence is important inproviding necessary technological and other sup-port services for products involved in the agencyactivity. Foreign firms are likely to select as agentsHTNVs which have commitment to R&D and arecapable of providing value-added distribution ser-vices. Consistent with these arguments, Hitt et al.(2000) found that firms from developed coun-tries consider the capability for quality of poten-tial alliance partners from emerging economies asa critical selection variable. Similarly, our inter-views with several Chinese HTNVs suggest thatto be successful agents HTNVs have to cooper-ate closely with the foreign product suppliers inproduct application solutions, technical training,and maintenance services. The lack of ability tocooperate may increase the probability that the newventure will be unable to find alliance partners. Forthis reason, technical skills and expertise devel-oped through R&D investments could induce thefirm to enter into an alliance formation in orderto leverage the technical capabilities of the foreignalliance partner (Hitt et al., 2000).

Thus, high R&D expenditures suggest both aninducement and opportunity for these firms to enterinto agency business activities. These argumentsare consistent with Stuart’s (1998) finding thathigh-technology firms tend to form alliances whenthey have a track record of developing new prod-ucts. Hence:

Hypothesis 5: The level of R&D expenditureis related positively to the adoption of agencybusiness activity by HTNVs.

Product development alliance

A firm’s involvement/experience in alliances and/or its accumulated learning from its past alliancesare expected to relate to its alliance formation.

For example, Gulati (1999) found that participa-tion in alliances is influential in the firm’s deci-sion to enter into new alliances. Although thereare many types of alliances, in this study wefocus on product development alliance, definedas HTNVs’ involvement/experience in joint devel-opment, manufacturing, and marketing new prod-ucts with other firms. Unlike the adoption ofagency business activities which concern only thedownstream activities (e.g., importing and sell-ing of products already developed by foreignfirms), product development alliance involves jointactions by the HTNV and its partners to developand sell a new product. Thus, it involves both theupstream and downstream activities of the valuechain (cf. Bucklin and Sengupta, 1993; Magrath,1992). Such an alliance is not part of the agencybusiness activity. Further, product developmentalliance irrespective of the type of product, tech-nology, market, or the firm involved may haveimplications for the HNTV’s decision to adoptagency business activity.

HTNVs’ involvement in a product developmentalliance provides an opportunity for entry intoagency business activity because it signals theirexperience and reliability in collaborative behav-ior to potential alliance partners (Ahuja, 2000;Gulati, 1999). Product development alliances maytake on a more substantive role in alliance for-mation in China, given the lower level of tech-nological capability of most firms in comparisonwith those in the West. It may provide ChineseHTNVs the necessary legitimacy, reputation, andexpertise for successful engagement in alliancesto foreign firms. Thus, HTNVs’ experience ofproduct development alliance serves as differen-tiating capability that is likely to be of value toforeign firms when they assess potential agencybusiness partners. From a social network perspec-tive, the HTNVs’ involvement in such an allianceprovides evaluative information on their collabora-tion opportunities, capabilities, and likely behaviorto potential foreign alliance partners, thus loweringtheir perceived risk associated with alliance forma-tion with the HTNVs (cf. Ahuja, 2000). Further,we argue that Chinese HTNVs’ product develop-ment alliance with another firm provides them withan enhanced opportunity to enter into agency busi-ness activity because it provides them with accessto information about a larger number of firms whomay be potential alliance partners (Eisenhardt andSchoonhoven, 1996).



From an inducement perspective, product devel-opment alliance may also be related positively tothe adoption of agency business activity becausethe knowledge and experience HTNVs gain in suchalliances open their eyes to the need for accessingthe knowledge and information of other firms. Forexample, Powell, Koput and Smith-Doerr (1996)found that R&D alliance is related positively tonew alliance formation with suppliers. Hence:

Hypothesis 6: Product development alliance isrelated positively to the adoption of agency busi-ness activity by HTNVs.

Successful agency business activity, newventure performance and product innovation

The above discussion focuses on both external andinternal factors as inducements and opportunitiesfor HTNVs’ adoption of agency business activ-ity. In this section we examine the relationshipbetween successful agency business activity (asmeasured by the proportion of an HTNV’s salesattributable to the agency business activity) andnew venture performance and product innovationefforts. Given the highly competitive and uncer-tain market environment in China, we contendthat successful agency business activity is posi-tively related to the overall performance of the newventure for three reasons. First, the adoption ofagency business activity represents a risk reductionstrategy. Marketing and distributing the productsand services of foreign firms entail less risk thaninternal product development. Risk can be furtherreduced via the enhanced reputation and legiti-macy effects associated with success in sellingthe products of well-known foreign firms. As Stu-art, Hoang, and Hybels (1999) empirically demon-strated, new ventures can benefit from prominentpartners because of the transfer of status throughinterorganizational networks. The transfer of sta-tus from foreign firms may reduce the risk per-ceived by the market in dealing with the HTNVs.Second, the adoption of agency business providesopportunities for learning from alliance partners(Granovetter, 1985; Kraatz, 1998). Thus, by suc-cessfully distributing products for foreign firms,HTNVs in China obtain access to management andmarketing skills that can contribute to the marketperformance of their own products. Third, empir-ical evidence has shown that distributing prod-ucts for international firms is a profitable business

activity for new ventures (Bruton and Rubanik,1997). Therefore, we argue that success withagency business activity will be positively relatedto the overall performance of the HTNVs measuredby both subjective and objective measures.

Hypothesis 7: Successful agency business activ-ity is related positively to HTNVs’ performance.

Successful agency business activity may also berelated to the HTNVs’ product innovation efforts.From an organizational learning perspective it canbe argued that the adoption of agency businessactivity provides an avenue for HTNVs to learnabout new technology and management skills thatcould improve their technical capabilities (Cohenand Levinthal, 1990; Granovetter, 1985; Kraatz,1998; Lane and Lubatkin, 1998). The productsdistributed by HTNVs for the foreign firms areusually considered innovative in the Chinese mar-ket. Though these products are not developed byHTNVs, they nevertheless provide them with alearning platform for their own innovative efforts(Grenadier and Weiss, 1997). Technical experienceand skills gained through marketing and servic-ing technology products of foreign firms may alsoenhance the absorptive capacity of HTNVs (Cohenand Levinthal, 1990; Kogut and Zander, 1992).In addition, agency business activity may provideextra resources that may encourage experimen-tation and learning (Bourgeois, 1981; Cyert andMarch, 1963). Therefore, we posit that HTNVswith higher proportion of agency business activ-ity are more likely to pursue product innovationwith increased vigor.

Hypothesis 8: Successful agency business activ-ity is related positively to HTNVs’ product inno-vation.

RESEARCH METHODOLOGY

Sample and data collection

This study examined ventures from high-tech-nology industries located in BHEZ, one of thelargest high-tech experimental zones in China.Firms in this zone were chosen because they areamong the fastest-growing high-technology firmsin China (BHEZ Office, 1995). We randomlyselected 300 ventures from a list provided by



BHEZ Office. All selected ventures were no morethan 8 years old. This was consistent with priorresearch that chose the 8 years cut-off for definingnew ventures (e.g., McDougall et al., 1994; Millerand Camp, 1985).

A letter was sent to the general managers of thesampled ventures. It explained the purpose of thestudy and invited their participation in the study.We contacted by phone those firms that agreed toparticipate to ensure that the potential respondentswere qualified and were committed to cooperatein the research (Feeser and Willard, 1990). As aresult of these efforts, 202 of 300 ventures weredeemed suitable to participate in the study.

We adopted a face-to-face interview method fordata collection in order to improve the responserate (Pearce, 1983). Based on our past researchexperience and the pilot study, we were convincedthat respondents were more likely to provide validinformation with this method. Indeed, this methodof data collection is thought to be better thantraditional mail surveys in emerging economiesbecause it offers a chance for the researchers toclarify questions, and to check and probe aspectsof behavior (Hoskisson et al., 2000). Further, theinterviews offered respondents an opportunity toask for clarification to enhance their understandingof the issues under study.

The questionnaire for the interview survey wasoriginally prepared in English and then trans-lated into Chinese by two management researcherscompetent in both languages and with substantialresearch experience in the subject area in China.To avoid cultural bias and ensure validity, the Chi-nese version was then back-translated into English.We paid special attention to any misunderstandinglikely to result from the translations. The instru-ment was then pretested with 8 founders and 15senior managers from 10 ventures. Using inputfrom the pretest we revised the survey instrumentto improve the clarity and relevance of the ques-tionnaire, and to ensure that questions were inter-preted as expected. To further test the reliabilityand validity of the measures, the questionnaire wasthen subjected to a pilot study with a sample of80 new ventures from a city located in the south-east of China. Results of this pilot study indicatedthat measures loaded strongly on their correspond-ing constructs and showed an acceptable level ofreliability. Our detailed work on back-translationand in-depth discussion with founders and seniormanagers, coupled with the results of the pilot test,

assure our confidence in the general appropriate-ness of the instrument and data collection method.

For data collection, we adopted the key infor-mant approach. The underlying assumption of thekey informant approach is that the person, byvirtue of his/her position in the organization’s hier-archy, is able to provide opinions and perceptionsthat are valid reflections of those of other keydecision-makers in the firm (Phillips, 1981). Whilethe ideal may be to use multiple informants, asingle informant approach could be the source ofneeded information for new ventures of relativelysmall size (Phillips, 1981). In addition, our pilottest suggested that a multiple-informant approachcould be very expensive in China. Nonetheless,in 45 cases, we obtained data from two informantsas a validation sample. We categorized these infor-mants into two groups: CEO/general manager andsenior managers (such as marketing or businessdevelopment managers). We conducted a series oft-tests to determine if there were any differencesamong the responses of the two groups on each ofour constructs and found no statistically significantdifferences.

We further checked the validity of the infor-mants’ responses in two ways. First, the informantswere asked to indicate the level of knowledge andthe extent of their involvement in making strate-gic decisions of their firms (Conant, Mokwa, andVaradarajan, 1990). The means of the level ofknowledge and the extent of involvement were7.18 and 7.09 respectively on a 9-point Likertscale, indicating that the informants were likelyto be knowledgeable about the issues under study.Second, the informants were asked to indicate theirindustrial working experience (Phillips, 1981). Theaverage working experience was 7.8 years, sug-gesting that the selected informants were expe-rienced and knowledgeable about their ventures’industry. The results from the subsample with mul-tiple informants along with high involvement andknowledge of our key informants increased ourconfidence in the quality and accuracy of our data.

We received 202 completed questionnaires but18 were excluded from the analysis because ofmissing values. Thus, the effective participationrate was 61.3 percent (184/300). Of the participat-ing ventures, 50.5 percent were in the electronicinformation industry (such as computers, softwaredevelopment, information technology, consumerelectronics), 17 percent in integrated optical-mechanical and electric products, 12.6 percent in



new energy and new materials, 10.4 percent innew pharmaceutical and bioengineering, and 9.3percent were classified as others (e.g., scientificinstruments, aerospace technology). This distribu-tion is consistent with that published by the BHEZOffice (1995), which indicates an industry distribu-tion of 47.9 percent, 20.1 percent, 13.1 percent, 9.3percent, and 9.6 percent, respectively. We viewedthis consistency as evidence that the sample wasrepresentative of the types of the firms that were inthe BHEZ. In addition, we compared the respond-ing firms with those of nonrespondents and foundno statistically significant differences in terms ofsize (F = 0.98, p > 0.10) and age (F = 0.23,p > 0.10). The mean age of the sampled firms was4.83 years (S.D. 2.3) with 29 percent of the ven-tures being 3 years old or less, 38 percent between4 and 5 years old and 33 percent between 6 and 8years old.

Measures

We followed guidelines generally associated withproper retrospective data collection in order toensure the accuracy and validity of the data col-lected (Miller, Cardinal, and Glick, 1997). Werestricted the recall time frame to 3 years (or sincetheir founding for those firms less than 3 yearsold). A 3-year or less time frame was chosen notonly to ensure we do not place an undue recall bur-den on respondents but also because it seemed anappropriate time frame for negotiation and imple-mentation of agency business activity. Previousstudies of technology development alliances whichinvolve greater technical knowledge developmentthan agency business activity use a 3- to 5-yeartime frame (e.g., Ahuja, 2000; Stuart, 1998). Weinterviewed top/senior managers who were directlyinvolved in the firms’ strategic decision making.We assured all respondents of confidentiality. Wealso provided a full explanation of the useful-ness of the project for the respondents’ organi-zations, and offered an incentive (i.e., summaryof the results) to foster a sense that the respon-dents would benefit from involvement in the study.Adler, Campbell, and Laurent (1989) noted thatrespondents in China are likely to answer questionsbased on the way they want their firms ideally tobe, not the way they actually are. Thus, we askedthe respondents to base their responses on the realsituation in their firms. Finally, prior research hasraised concerns about social desirability aspects

of responses in the Chinese context (Adler et al.,1989). This concern seems likely to be exacerbatedby the on-site interview approach to administeringthe questionnaire because of the great importanceattached to face in the Chinese context. To mini-mize social desirability bias, the respondents wererepeatedly reminded that there were no right orwrong answers to the questions being asked ofthem (e.g., Miller et al., 1997). Also, we deleted 18cases from the analysis due to missing data. These,coupled with the fact that we obtained objectiveperformance data from only 59 ventures, suggestthat problems of social desirability as well as anypressure on the participants associated with ourinterview method could be considered as minimal.

Dependent variables

To measure the adoption of agency business activ-ity we asked the respondent to indicate whetherhis/her firm was currently involved in any busi-ness as an agent for a foreign firm. The con-struct was dummy coded (0 = No and 1 = Yes).We validated this measure using the sample withtwo informants as discussed previously. A cross-check showed that in all 45 cases the two infor-mants were consistent in indicating their firms’adoption of agency business activity. Successfulagency business activity was measured by the aver-age percentage of the firm’s total sales attributableto the agency business for the past 3 years orsince its inception if the venture was less than3 years old.

New venture performance was measured withboth subjective and objective items. Subjectivemeasures were used because prior research hasprovided substantial evidence supporting the relia-bility and validity of perceptual performance mea-sures (Geringer and Hebert, 1991; Dess and Robin-son, 1984). Indeed, McDougall et al. (1994) sug-gest that in the new venture setting a more com-plete picture of performance may be obtained byusing subjective measures of performance. Weused four market-related indicators to measurethis construct: sales growth rate, market sharegrowth, cash flow from market operations, and thefirm’s overall reputation. The respondents wereasked to indicate how successfully their venturehas achieved these goals relative to its establishedgoals over the past 3 years on 5-point Likert scales.To obtain an objective measure of new ventureperformance, respondents were asked to indicate



the sales growth rate (%) of their firms for each ofthe past 3 years or since its founding if the firmwas less than 3 years old. We obtained data from59 ventures. The average sales growth was cal-culated for analysis. This more objective measurewas significantly related to the subjective measureof new venture performance (r = 0.36, p < 0.01).Product innovation was measured with four itemsdrawn from Miller (1987) and Zahra and Covin(1993). The respondent was asked to comparehis/her venture with its major competitors and thenindicate the extent to which the venture engaged indeveloping and launching its own new products.

Independent variables

We measured environmental uncertainty by fouritems drawn from the studies by Miller (1987),reflecting the degree of price, product, techno-logical, and competitive change as perceived byHTNVs in their principal industries. To measureinstitutional support we used five newly developeditems reflecting the role of government in provid-ing financial, technology information, and manu-facturing support to the ventures (BHEZ Office,1995). We measured perceived industry growth bythree items drawn from the work of McDougallet al. (1994) reflecting the growth in demand andattractiveness of the industry. Following previousresearch (e.g., Shan, 1990) we used the number offull-time employees to measure venture size. Wetransformed this measure by natural log because itis expected that the effect of size would increase ata diminishing rate. R&D expenditure as percent ofsales was measured by the average percentage ofsales spent on R&D over the past 3 years or sincefounding if the venture was less than 3 years old.We measured product development alliance by sixitems based on the work of Bucklin and Sengupta(1993) and Magrath (1992) and our field inter-views. These measures assessed the focal venture’scooperation with other firms in the new productdevelopment process including designing, manu-facturing and marketing of new products.

Control variables

In testing the hypotheses, several demographicvariables were controlled for because they mayaffect new ventures’ strategy making and perfor-mance (McCann, 1991; Zahra, 1993). For ventureownership, state- and collectively owned ventures

form one category while privately owned and jointventures form a second category. The reason forbinary categorization is that state- and collectivelyowned firms in China operate under different reg-ulatory conditions from privately owned and jointventures (Nee, 1989). Venture age was used asa continuous variable. We measured the numberof years since its founding, with the upper limitbeing 8 years. Venture origin was classified intotwo categories: independent ventures (which areestablished by independent entrepreneurs) and cor-porate ventures (which are founded and supportedby established companies, e.g., joint ventures).They were dummy coded as 0 and 1 respectively.Industry type was also controlled for with elec-tronic information industry coded as 0 and all oth-ers coded as 1. This binary classification followsthe practice of the BEZ, which lumps computermanufacturing, information technology, softwaredevelopment, and electronics into one categoryas electronic information industry, separate fromall other ventures. Because of the limited samplesize, we did not use the multiple industry cate-gories approach (e.g., coded as dummies for each).Also because over half of the sample is from theelectronic information industry, it is meaningful tocontrol for this potential industry effect.

To test the hypotheses regarding how the pro-portion of agency business activity relates to newventure performance and product innovation, theabove demographic variables were also controlledfor. Further, environmental uncertainty and per-ceived industry growth were included as controlsbecause of their potential impact on product inno-vation and new venture performance (Eisenhardtand Schoonhoven, 1990; McDougall et al., 1994).Given the potential confounding effect betweenthe agency business activity and product develop-ment alliance, the latter was controlled for. We alsocontrolled for relatedness of agency business activ-ity with the venture’s own business because suchrelatedness may affect the ease of transferabilityof core skills and economies of scale and scope(Rumelt, 1974). We measured this construct witha single item by asking respondents to indicate theextent to which the products they are marketingand distributing for foreign firms are related to theventure’s own products.

Measurement validation

We used a multistage process to assess constructvalidity with the guidelines outlined by Anderson



and Gerbing (1988). We first examined item-to-total correlations and performed an exploratoryfactor analysis for each scale since the measureshad been modified from previous studies. Then,we conducted confirmatory factor analysis to testfor the unidimensionality and convergent validityof the constructs. Because of the sample size andthe number of items, we divided the constructsinto two subsets of theoretically related groups.The scale items, along with factor loadings and fitstatistics, are provided in Table 1. As the resultsshow, the standardized loadings are highly sig-nificant for all the items, suggesting that all ofthese indicators are similarly responsive to changesin the underlying constructs they are purportedto measure (one item from institutional supportwas deleted because of its lower level of factorloading). Moreover, the fit indices for each modelsuggest that the models fit the data very well, pro-viding further evidence of convergent validity.

Next, we assessed discriminant validity of theconstructs by testing if correlations between anytwo constructs were significantly different fromunity. This required a comparison between twomodels in which one was constrained with thecorrelation equal to one and another was not. Ineach case discriminant validity was evidenced bythe statistically significant chi-square differencebetween the models. For example, the comparisoninvolving product innovation and product devel-opment alliance provided a χ 2(1) = 179.02 (p <

0.01), suggesting that these measures are distinct(results of this analysis are available upon request).

Finally, we assessed the reliability of theconstructs with Cronbach’s coefficient alpha. Allscales except one have reliabilities greater than0.70. The alpha for competitive intensity is 0.60,which is generally acceptable for questionnairescales. Considering the face validity of thisfactor and the strong factor loadings, we believedthat it was reasonable to use this factor inthe subsequent analyses. Table 2 presents means,standard deviations, and correlations for theconstructs.

ANALYSIS AND RESULTS

The adoption of agency business activity

We used logistic regression to test the relationshipbetween external and internal factors and the

adoption of agency business activity by HTNVs.Logistic regression restricts the range of the depen-dent variable to a value between 0 and 1, whichis appropriate for investigating the likelihood orprobability of the adoption of agency businessactivity. The general specification is as follows:log{P(Y = 1)/(1 − P(Y = 1)} = B(X, M), whereP(Y = 1) is the probability that a venture adoptsagency business activity and X and M are vectorsof independent and control variables.

Note that to test the interaction effect as indi-cated in Hypothesis 2 we entered the cross-productof institutional support and environmental uncer-tainty into the logistic model. Table 3 presents theresults of logistic regression analysis that tests therelationship between environmental variables andthe probability of agency business adoption byHTNVs. The model χ 2 is statistically significant,indicating that the predictor variables are relatedsignificantly to the dependent variable, the adop-tion of agency business activity.

Hypothesis 1 states that environmental uncer-tainty is related positively to the adoption ofagency business activity. The results presentedin Table 3 support this hypothesis. The signif-icantly negative relationship between the cross-product of environmental uncertainty and institu-tional support and the adoption of agency busi-ness activity lends support to Hypothesis 2. Thissuggests that institutional support weakens thepower of environmental uncertainty as an induce-ment or opportunity for new ventures to enterinto agency business activity. Perceived indus-try growth has a significantly negative relation-ship with the adoption of agency business activity,which provides support for Hypothesis 3a. In otherwords, our results do not support the alternativeargument in Hypothesis 3b that perceived indus-try growth may act as an opportunity for HTNVsto enter into agency business activity to lever-age their resources to exploit the available growthopportunities.

The results suggest a negative relationship bet-ween venture size and the adoption of agencybusiness activity. This finding supports Hypothe-sis 4 stating that new ventures may need resourcesand may therefore enter into alliances to obtainsuch resources. In other words, our results do notsupport the alternative viewpoint that alliance for-mation offers opportunities for larger new venturesto leverage available resources. Hypothesis 5 stat-ing that R&D expenditure as percent of sales is



Table 1. Construct measurement and confirmatory factor analysis by LISREL

Item description summary Standardizedloading

t-value

Constructs in Model 1Environmental uncertainty (α = 0.60)Rate the degree to which each of these statements describes your principal

industry over the last 3 years:a

1. the competitive intensity has been very high and uncertain 0.65 8.902. severe price competition has been a characteristic of my industry 0.45 4.683. our firm must change its marketing practices frequently to keep up with

the market and competitors0.42 4.62

4. the rate at which products or services become obsolete has dramaticallyincreased

0.85 10.76

Perceived industry growth (α = 0.75)Rate the degree to which each of these statements describes your principal

industry over the last 3 years:1. there has been high growth in demand in this industry 0.67 8.502. this industry offered many attractive opportunities for future growth 0.82 12.443. growth opportunities in this industry have been abundant 0.78 11.62

Institutional support (α = 0.77)Please indicate the extent to which in the last 3 years government and its

agencies have:1. implemented policies and programs that have been beneficial to your

venture’s operations0.64 8.23

2. provided needed technology information and other technical support foryour venture

0.88 11.20

3. played a significant role in providing financial support for your venture 0.53 6.864. helped your venture to obtain licenses for import of technology,

manufacturing and raw material, and other equipment0.46 6.02

5. seldom interfered in the operations of your ventureb

Model Fit Indexχ 2 = 101.26 (p = 0.00), GFI = 0.90, RMSEA = 0.08, NNFI = 0.92,

CFI = 0.92

Constructs in Model 2Product innovation (α = 0.83)Rate your venture relative to its major competitors over the last 3 years the

extent to which it has:1. placed significant emphasis on new product development through

allocation of substantial financial resources0.83 13.04

2. developed a large variety of new products or made dramatic changes inexisting products

0.65 9.32

3. increased the rate of new product introductions to the market 0.80 12.424. increased its overall commitment to develop and market new products 0.78 12.51

Product development alliance (α = 0.86)To what extent do these statements describe your firm over the last 3 years

relative to your competitors?1. marketed complementary new products with other firms 0.73 10.162. introduced new products jointly with other firms 0.83 11.783. promoted new product lines jointly with other firms 0.70 10.144. jointly distributed and provided new product support services with other

firms0.75 10.28

5. jointly designed and manufactured new products with other firms 0.73 10.646. established cooperative pricing agreements with other firms for R&D 0.59 8.20

(continued overleaf )



Table 1. (Continued )

New venture performance (α = 0.78)Indicate how successfully your venture has achieved these goals in the last

three years:1. sales growth rate 0.74 10.942. market share growth 0.81 12.063. cash flow from market operations 0.57 7.474. firm reputation 0.70 10.20

Model fit Indexχ 2 = 156.73 (p = 0.00), GFI = 0.90, RMSEA = 0.06, NNFI = 0.92,

CFI = 0.93

a For new ventures that were less than 3 years old the response time frame was since their founding.b Items were deleted because of low item-to-total correlations.

positively related to the adoption of agency busi-ness activity is not supported, although the sign isin the predicted direction. In support of Hypothe-sis 6, product development alliance has a signif-icant positive relationship with the adoption ofagency business activity indicating the potentialsocial capital opportunities for alliance formationderived from such an alliance.

Successful agency business activity, newventure performance, and product innovation

We used ordinary least squares regression analy-sis to examine the relationship between success-ful agency business activity, new venture perfor-mance, and product innovation. These analysesinvolved only those ventures that had adoptedagency business activity (n = 106). Table 4 pre-sents the results.

In Model 1a, we tested the relationship betweensuccessful agency business activity and new ven-ture performance measured with subjective mea-sures. The results show that successful agencybusiness activity has a significantly positive rela-tionship with new venture performance, therebysupporting Hypothesis 7. Note that product inno-vation is also related positively to new ven-ture performance. As a sensitivity analysis, inModel 1b we replaced the subjective measure withthe objective measure of new venture performance.Although the sample is small, as the data inTable 4 indicate, with this alternative measure, westill obtained a similar pattern of results describedabove with respect to subjective measure of newventure performance. These findings appear to lendsome support for the assertion that the adoption ofagency business activity is related to the perfor-mance of HTNVs (Bruton and Rubanik, 1997).

In Model 2, we tested the relationship betweenthe success of the agency business activity andproduct innovation. Results show that successfulagency business activity has a significantly neg-ative relationship with product innovation. ThusHypothesis 8 is refuted. A possible reason is thata greater success with the agency business activ-ity could lure managers of HTNVs into failingto develop in-house product innovation programsbecause competences in the downstream activitiesinvolved in agency business activity as well ascompetencies in developing the associated networkties with foreign firms may be perceived as eas-ier to negotiate and leverage than in-house prod-uct innovation activity. Successful agency businessactivity could also lead to changes in the goalsand aspirations of managers such that a consciousstrategic choice is made to focus on this activ-ity rather than on product innovation (Cyert andMarch, 1963). It may also be argued that failurein achieving strong product innovation inhibits afirm’s ability to attract foreign firms for whom theHTNV could serve as an agent.

Two control variables—perceived industrygrowth and product development alliance—arepositively related to product innovation. The latterfinding suggests that an interfirm alliance thatspans the entire value chain ranging from productdesign to marketing is more likely to be related toproduct innovation than agency business activitythat involves only the downstream aspects ofthe value chain. Interestingly, we found thatrelatedness of agency business activity with theHTNV’s markets and products is positively relatedto new venture performance measured by objectiveitems. This finding is in keeping with the argumentthat related businesses may be more beneficial tothe firm than is unrelated business because of the



Tabl

e2.

Cor

rela

tion

mat

rix

and

sum

mar

yst

atis

ticsa

Var

iabl

eM

ean

S.D

.1

23

45

67

89

1011

1213

14

1.In

dust

ryty

pe0.

490.

50—

2.V

entu

reow

ners

hip

0.78

0.42

0.09

—3.

Ven

ture

age

4.83

2.03

−0.0

4−0

.10

—4.

Ven

ture

orig

in0.

340.

480.

02−0

.27∗∗

−0.0

6—

5.V

entu

resi

ze16

0.9%

347.

3%0.

06−0

.23∗∗

0.25

∗∗−0

.14

—6.

R&

Dex

pens

e0.

080.

05−0

.03

−0.0

00.

04−0

.02

−0.0

7—

7.Pr

oduc

tde

velo

pmen

t3.

110.

92−0

.10

0.14

−0.0

4−0

.22∗∗

−0.0

7−0

.22∗∗

0.86

allia

nce

8.E

nvir

onm

enta

l3.

880.

680.

06−0

.09

0.10

0.05

−0.0

20.

050.

040.

60un

cert

aint

y9.

Inst

itutio

nal

supp

ort

2.76

0.87

0.16

∗0.

13−0

.01

0.07

−0.0

10.

070.

24∗

−0.1

20.

7710

.Pe

rcei

ved

indu

stry

3.85

0.77

−0.0

10.

20∗∗

0.09

−0.1

2−0

.04

0.12

0.12

0.06

0.07

0.75

grow

th11

.A

dopt

ion

ofag

ency

0.59

0.49

−0.1

4−0

.06

0.09

0.09

−0.1

5∗0.

090.

19∗

0.16

∗−0

.04

−0.0

9—

busi

ness

acti

vity

12.

Succ

essf

ulag

ency

busi

ness

acti

vity

0.43

0.31

−0.2

7∗∗0.

15−0

.10

0.10

−0.0

60.

100.

04−0

.29∗∗

0.02

−0.0

10.

06—

13.

Prod

uct

inno

vatio

n3.

950.

84−0

.08

0.15

0.02

−0.1

6∗−0

.05

0.16

∗0.

37∗∗

0.19

∗0.

100.

51∗∗

−0.0

3−0

.18

0.83

14.

New

vent

ure

perf

orm

ance

3.58

0.70

0.09

0.21

∗−0

.02

−0.1

7∗−0

.02

−0.1

10.

19∗

0.05

0.12

0.38

∗∗−0

.07

−0.0

20.

44∗∗

0.78

aN

umbe

rsin

diag

onal

are

relia

bilit

ies.

Sign

ifica

nce

leve

ls:

∗ p<

0.05

;∗∗

p<

0.01



Table 3. The adoption of agency business activity:logistic regression analysis

Variables B Wald χ 2

Control variablesIndustry type 0.16 0.13Venture ownership −0.70 1.27Venture age 0.12 1.17Venture origin 0.35 0.47

Independent variablesEnvironmental

uncertainty0.75∗ 4.70

Institutional support −0.19 0.48Environmental

uncertainty ×institutional support

−0.96∗ 4.10

Perceived industrygrowth

−0.71∗ 4.87

Venture size −0.38∗ 4.26R&D expenditure as

percentage of salesturnover

1.24 1.21

Product developmentalliance

0.53∗ 4.27

Constant 0.48 0.06

Model χ 2 24.06∗

�χ 2with the entry ofinteraction term

5.40∗

d.f. 11N 184

Significance level (two-tailed): ∗p < 0.05

ease of transferability of core skills and economiesof scale and scope (Rumelt, 1974; Teece, 1980).

DISCUSSION AND IMPLICATIONS

Resource-based and social network theories haveconfirmed the importance of interfirm relationshipsfor individual firms’ capabilities and success inuncertain environments. There has been a signif-icant amount of research on interfirm alliancesbased on these theories, particularly those involv-ing technology development alliances and betweenestablished firms. In this study we contribute to thisstream of research by examining the factors thatact as inducements or opportunities for technologynew ventures to enter into downstream alliancesinvolving marketing and distribution. Particularly,we focused on the adoption of agency businessactivity as an interfirm alliance between HTNVsin China and foreign firms and examined itslinks with internal firm characteristics and external

environmental factors. We also explored its perfor-mance and product innovation implications.

Our findings provide some support of our the-oretical arguments that external and internal fac-tors may act as inducements and opportunities forHTNVs in China to enter into agency businessactivity with foreign firms. First, our results sug-gest that environmental uncertainty is positivelyrelated to the adoption of agency business activity.This finding provides support for the resource-based view that HTNVs have problems raisingcapital and other resources in an uncertain envi-ronment and therefore may enter into alliances insuch an environment. In other words, when thereis a need for resources an uncertain environmentappears to act as an inducement for alliance for-mation. The same finding appears to support theopportunity to collaborate perspective, if viewedfrom the viewpoint of the foreign firms enter-ing China. These firms place significant emphasison unique competencies, market knowledge, andmarket access in selecting partners from emergingmarkets (Hitt et al., 2000). These selection crite-ria may become even more important when theenvironment is uncertain. Hence, environmentaluncertainty appears to provide an alliance forma-tion opportunity for HTNVs in China because theirmarket knowledge and other unique competenciesin the Chinese market are prime resources requiredby foreign firms to reduce their risks in enteringthe market.

However, the relationship between environmen-tal uncertainty and the adoption of agency businessactivity is moderated by institutional support suchthat when institutional support is high environmen-tal uncertainty is negatively related to the venture’sadoption of agency business activity. It seems thatinstitutional support may lower managerial percep-tions of environmental risk facing HTNVs. Thisreduces the inducement and opportunity to enterinto alliances as a coping mechanism against envi-ronmental uncertainty.

The second group of findings is that perceivedindustry growth and venture size are negativelyrelated to the adoption of agency business activity.The former finding is similar to Eisenhardt andSchoonhoven’s (1996) argument that munificentmarket conditions are related negatively to allianceformation. The latter finding is consistent with pre-vious research results that the linkage betweenalliance formation and firm financial assets is weakat best (Gulati, 1999). From the resource-based



Table 4. Successful agency business activity, product innovation and new ventureperformance (standardized regression coefficients)

Variables Model 1aNew ventureperformance

Model 1bNew ventureperformance

Model 2Product

innovation

Control variablesIndustry type 0.20∗ 0.09 −0.15Venture ownership −0.14 −0.27∗∗ 0.11Venture age 0.01 −0.09 −0.10Venture size 0.08 0.12 −0.01Venture origin −0.09 0.04 0.08Environmental

uncertainty0.05 0.05 0.12

Perceived industrygrowth

0.16∗ 0.02 0.37∗∗∗

Product developmentalliance

−0.12 −0.44∗∗∗ 0.42∗∗∗

Relatedness of agencybusiness

0.09 0.25∗∗ −0.04

Independent variablesSuccessful agency

business activity0.29∗∗ 0.32∗∗∗ −0.25∗∗

Product innovation 0.51∗∗∗ 0.49∗∗∗

R2 0.31 0.75 0.44Adjusted R2 0.18 0.65 0.37d.f. 11/55 11/27 10/73F -value 2.30∗∗ 7.46∗∗∗ 5.84∗∗∗

Significance levels (two-tailed): ∗p < 0.10, ∗∗p < 0.05, ∗∗∗p < 0.01.

view of the firm, these findings suggest that givenmunificent environment and strong financial assets,larger Chinese HTNVs are unlikely to enter intoagency business activity. It may be that smallHTNVs have relatively fewer internal resourcesand thus have greater propensity to engage inagency business activity in order to spread theirrisks (Teece, 1980). In addition small firms mayadopt agency business activity in order to offera fuller line of products—something difficult todo through internal offerings—with economic effi-ciency. In contrast, larger firms may have lessneed to do so because they are more likely tohave a broad enough line of their own. Theimplication of these results is that the opportu-nities for alliance formation provided by industrygrowth and availability of resources from largersize are not utilized by Chinese HTNVs. In otherwords, when alternative sources of resource arenot lacking, HTNVs tend not to seek networkrelationships with foreign firms (e.g., MNCs).In this sense, by not using the growth mar-ket and size to attract alliance partners ChineseHTNVs in our sample are not tapping all the

available opportunities for growth through allianceformation.

The third major finding of our study concernsthe positive relationship between the adoption ofagency business activity and product developmentalliance. The finding supports the argument thatsocial capital is important for HTNVs’ alliance for-mation, reinforcing the recent finding that foreignfirms consider alliance experience as a critical fac-tor in selecting alliance partners in emerging coun-tries (Hitt et al., 2000). As argued previously, fromthe social capital perspective the HTNV’s attrac-tiveness to potential foreign partners depends onwhat it can offer to them in return. In this sense ourfinding suggests that product development allianceformation may not only provide opportunities forthe HTNVs to discover potential partners, but per-haps more importantly it indicates their reliabilityas alliance partners by providing signals as to thetechnical and marketing capabilities they have tooffer.

Clearly, our findings indicate that the logic forHTNVs’ adoption of agency business activity ismultifaceted. Such an adoption is related to both



the firm-specific and environmental inducementsand opportunities available to it (Ahuja, 2000).Our study extends the literature into a new con-text: technology new ventures in a transitionaleconomy.

In respect of the outcomes of agency businessactivity, our findings suggest a positive relation-ship between successful agency business activityand new venture performance. Given the cross-sectional nature of our study, an alternative expla-nation that cannot be completely ruled out is thatbetter performing HTNVs in China may be morelikely to have a greater capability to be successfulwith agency business activity.

Contrary to our hypothesis, results indicate thatsuccessful agency business activity has a nega-tive relationship with product innovation effort ofthe HTNVs. This finding appears contrary to theorganizational learning perspective of alliance for-mation (Cohen and Levinthal, 1990; Kogut andZander, 1992) and the organization slack view(Bourgeois, 1981; Cyert and March, 1963) that byadopting this strategy HTNVs may learn technicalskills, thus contributing to their product innova-tions. Several reasons may account for this find-ing. First, successful agency business activity maydivert the limited resources of the HTNVs awayfrom their own technological innovation as theyneed to provide technological services to supportthe selling of the products of their alliance partners.As Rumelt (1974 : 1) argues, when a firm decidesto enter another business, ‘. . . it is making a strate-gic decision whose consequences may alter thefundamental nature of the firm and may involve aswell a substantial redeployment of resources and aredirection of human energy.’ Second, a profitableagency business may lull managers of HTNVsinto a false sense of security to the neglect ofproduct innovation (Eisenhardt and Schoonhoven,1996). Managers may consider that this activityis a less risky alternative growth strategy. Forexample, competences in the downstream activi-ties involved in agency business activity as wellas competencies in developing the associated net-work ties with foreign firms may be perceived aseasier to negotiate and leverage than engaging inin-house product innovation. Third, it is possiblethat what HTNVs have learnt through this kindof downstream activities are marketing and salestechniques rather than technical skills, which maystifle product innovation. Finally, like all interorga-nizational strategies agency business activity may

also be plagued by conflicts and other relation-ship problems between the parties in the alliance.Indeed, a study of technology ventures by Kotabeand Swan (1995) found no evidence to supporta positive relationship between interorganizationalrelationships and product innovation.

Though not hypothesized, an interesting resultfrom the study is that product development alliancehas a significantly positive relationship with theHNTV’s product innovation effort but a negativerelationship with new venture performance. Thesefindings are clearly in contrast to those obtained forsuccessful agency business activity, indicating thedifference between these two constructs. Unlikeagency business activity, a product developmentalliance spans the entire product innovation pro-cess from design, development, to marketing. Aplausible explanation for the positive relationshipcould be that a product development alliance mayinvolve the transfer of technical as well as com-mercial skills from partners to the HTNV. Theseoffer differential advantages to HTNVs in carry-ing out product innovation activities (Calantone,Schmidt, and Song, 1996). It is also possible thattechnical skills resident in a firm allow productinnovation success, which permits greater successby the firm in attracting other companies for aproduct development alliance. However, a prod-uct development alliance may be negatively relatedto new venture performance because HTNVs mayhave a weaker position in the relationship. Thusit is difficult for HTNVs to generate benefitsfrom such alliances. Admittedly, this explanationis speculative and warrants future study.

Previous research has argued that alliances withforeign firms appear to be an important route forfirms to grow in a transitional economy (Peng andHeath, 1996; Zhao and Aram, 1995). This studyadvances this research stream by examining thefactors that may be related to the adoption ofagency business activity by HTNVs in China. Theresearch presented here is part of a growing streamof strategic management literature investigatingfirm behavior in economic transitions (Bruton andRubanik, 1997; Khanna and Palepu, 1997; Pengand Heath, 1996). Prior research has claimed thatHTNVs should commit to a focus on productinnovation (Eisenhardt and Schoonhoven, 1990;Feeser and Willard, 1990). Our study suggests thatthough product innovation is an important activityfor HTNVs, agency business activity is an equallyimportant strategy.



Managerial implications

These findings have implications for managers.First, they suggest the need for managers to takea dual perspective of the internal and external fac-tors facing the firms in alliance formation deci-sions. For example, firm resources could be asource of inducement as well as opportunity forHTNVs to enter into alliances. Yet, our resultsindicate that managers may not be taking fulladvantage of the resources available to competein the alliance formation market. Whereas productdevelopment alliance appears to be a resource thatprovides opportunity for entry into alliance, thesame opportunity that could be afforded by venturesize (because of its implications of market pres-ence and social networks) appears not to have beentaken by firms in our sample. Further, managersof HTNVs view institutional support as alleviatingresource limitations to the extent that their percep-tion of the uncertain environment as an inducementand opportunity for alliance formation is reduced.From the opportunity-to-collaborate perspective, itcould be suggested that institutional support couldbe used to build the HTNVs’ internal resources andcapabilities to make themselves attractive partnersto potential alliance partners.

Another important implication is that althoughthe successful agency business activity is posi-tively related to a new venture’s performance, ithas a negative relationship with the new venture’sproduct innovation activity. It may be argued thatgiven the different goals and aspirations of HTNVs(Cyert and March, 1963; Kirchhoff, 1977), a focuson positive relationship of agency business activ-ity with performance and its negative relationshipwith product innovation may be a conscious strate-gic choice made by management (Child, 1972)rather than a withering of skills. Although we areunable to determine which is the case, our resultshighlight the importance of managers being con-scious of the potential dysfunctional implicationsof agency business activity for product innova-tions. This is important not only for effective goalsetting for HTNVs but also for focusing their inten-tions and goals in negotiating and implementingsuch alliances.

Limitations and future research

Despite some contributions to the literature andpractice, this study has limitations that should

be addressed in future research. First, the cross-sectional data used in the study do not allowfor causal interpretations among the variables. Forexample, as mentioned previously, although wefound that a successful agency business activ-ity is positively related to new venture perfor-mance, it is also possible that better performancemay precede successful agency business activity.Ideally the study would have benefited from atime lag between the measurement of the indepen-dent and dependent variables for causal relation-ships to be determined. However, as Kenny (1979)argued, a careful study of cross-sectional relation-ships before attempting to validate the findingsvia more costly time-lagged longitudinal studiesis a commonly accepted approach for establish-ing causal relationships. We hope this study willserve as a foundation for such a study in thefuture.

Second, our theoretical model focuses on only afew internal and external factors. Other potentiallyrelevant variables that merit consideration in theadoption of agency business activity are CEO ortop management team characteristics such as size,previous experience, and social network relation-ships which several scholars suggest may influencealliance formation (Eisenhardt and Schoonhoven,1990, 1996; Kraatz, 1998; Nahapiet and Ghoshal,1998). In addition, other theoretical perspectivesmay also help explain HTNVs’ tendency to adoptagency business activity. For instance, from amimetic isomorphism point of view, HTNVs mayeconomize on search costs and imitate the actionsor strategies of other firms as a means of appearinglegitimate (Schoonhoven, Eisenhardt, and Lyman,1990). Thus, when some HTNVs adopt agencybusiness activity and achieve better performance,this specific course of action becomes taken forgranted and institutionalized. This theoretic per-spective warrants research in the future.

Third, the constructs in the study were mea-sured with self-reports. The use of retrospectivedata may pose such potential problems as lim-ited recall of the respondents and biased per-ceptions of past realities. We took several mea-sures during data collection to improve reliabil-ity and validity of retrospective reporting, includ-ing strengthening our analysis with more objec-tive new venture performance data. However, thisshould not exclude the use of alternative method-ologies in future research. In addition, our study



is limited to the extent that the construct envi-ronmental uncertainty combines the informationaland resource perspectives of the environment thatmay have different influences on firm decisionmaking. However, we do not believe that thisis a problem here as we found that environ-mental uncertainty and perceived industry growth(which clearly captures resource munificence inthe environment) have differential relationshipswith adoption of agency business activity. Nev-ertheless, we call attention to the need for futureresearch to differentiate and examine the resourceand informational dimensions of environmentaluncertainty. Further, the single-item measure ofrelatedness of agency business activity could beimproved.

Finally, our sample has potential bias with halfof the sample being in the electronic informa-tion industry. It may also be biased towards newventures with greater chances of survival. Hence,we caution that our sample is not representativeof new ventures in China or other economies.To enhance the generalizability of the findings,future research ought to build on these resultsand examine the adoption of agency businessactivity in other places in China and in othereconomies. In particular, a comparison of thealliance formation behavior of HTNVs in devel-oped and transitional economies would shed use-ful light on the degree to which peculiaritiesof transitional economies such as underdevelop-ment of institutions have implications for allianceformations. Currently, the literature asserts thatHTNVs in transitional economies face environ-mental conditions and significantly severer prob-lems in obtaining capital and other resources thantheir counterparts in developed economies. Forexample, it has been argued that the underde-velopment of strategic factor markets makes itdifficult for firms to raise capital and transferassets and ownership in transitional economiessuch as China. Similarly, property rights are notclearly defined because of lack of an adequatelegal framework. Government regulations are alsovery unpredictable, which creates more uncer-tainty and high costs for firms’ growth (Peng andHeath, 1996). Scholars assert that these conditionsmake alliance formation a prime strategy for newventures in transitional economies than in devel-oped economies. Yet, there is little comparativeempirical evidence to support such assertions. We

hope future research would take on this line ofinquiry.

ACKNOWLEDGEMENTS

We would like to thank Yan Zhang for her insight-ful comments on the earlier versions of this paper.We are grateful for financial support from theHong Kong Government Competitive EarmarkedResearch Grant (No. 9040122) and Research Grantat Lingnan University in Hong Kong.

REFERENCES

Adler NJ, Campbell N, Laurent A. 1989. In search ofappropriate methodology: from outside the People’sRepublic of China looking in. Journal of InternationalBusiness Studies 20(1): 61–74.

Ahuja G. 2000. The duality of collaboration: inducementsand opportunities in the formation of interfirmlinkages. Strategic Management Journal 21(3):317–343.

Anderson JC, Gerbing DW. 1988. Structural equationmodeling in practice: a review and recommended two-step approach. Psychological Bulletin 103: 411–423.

Barney JB. 1991. Firm resources and sustained competi-tive advantage. Journal of Management 17: 99–120.

BHEZ Office. 1995. Zhongguo Beijing Xinjishu ChanyeKaifa Shiyanqu Yanjiu Baogao [Research Report onBeijing High Technology Experimental Zone (BHEZ)in China]. Beijing.

Bourgeois LJ III. 1981. On the measurement oforganizational slack. Academy of Management Review6(1): 29–39.

Bruton GD, Rubanik Y. 1997. High technology entre-preneurship in transitional economies: the Russianexperience. Journal of High Technology ManagementResearch 8(2): 213–223.

Bucklin LP, Sengupta S. 1993. Organizing successful co-marketing alliances. Journal of Marketing 57(April):32–46.

Calantone RJ, Schmidt JB, Song XM. 1996. Controllablefactors of new product success: a cross-nationalcomparison. Marketing Science 15(4): 341–358.

Child J. 1972. Organizational structure, environment andperformance: the role of strategic choice. Sociology 6:1–22.

China Economic News. 2000. Implications of YuxingInfotech’s listing in Hong Kong. 21 February: 6–7.

Cohen WM, Levinthal DA. 1990. Absorptive capacity:a new perspective on learning and innovation.Administrative Science Quarterly 35: 128–152.

Conant JS, Mokwa MP, Varadarajan PR. 1990. Strategictypes, distinctive marketing competencies and orga-nizational performance: a multiple measures-basedstudy. Strategic Management Journal 11(5): 365–383.

Cyert RM, March JG. 1963. A Behavioral Theory ofFirm . Prentice-Hall: Englewood Cliffs, NJ.



Dess GG, Beard DW. 1984. Dimensions of organi-zational task environments. Administrative ScienceQuarterly 29: 52–73.

Dess GG, Robinson RB. 1984. Measuring organizationalperformance in the absence of objective measures.Strategic Management Journal 5(3): 265–273.

Dickson PH, Weaver KM. 1997. Environmental determi-nants and individual-level moderators of alliance use.Academy of Management Journal 40(2): 404–425.

Dowling MJ, McGee JE. 1994. Business and technologystrategies and new venture performance: a study of thetelecommunications equipment industry. ManagementScience 40(12): 1663–1677.

Downey H, Hellriegel D, Slocum J. 1975. Environmentaluncertainty: the construct and its application.Administrative Science Quarterly 20: 613–629.

Dyer JH, Singh H. 1998. The relational view: cooperativestrategy and sources of international competitiveadvantage. Academy of Management Review 23(4):660–679.

Eisenhardt KM, Schoonhoven CB. 1990. Organizationalgrowth: linking founding team, strategy, environment,and growth among U.S. semiconductor ventures,1978–1988. Administrative Science Quarterly 35(3):504–529.

Eisenhardt KM, Schoonhoven CB. 1996. Resource-basedview of strategic alliance formation: strategic andsocial effects in entrepreneurial firms. OrganizationScience 7(2): 136–150.

Feeser HR, Willard GE. 1990. Founder strategy andperformance: a comparison of high and low growthhigh tech firms. Strategic Management Journal 11(2):87–98.

Geringer JM, Hebert L. 1991. Measuring performance ofinternational joint ventures. Journal of InternationalBusiness Studies 22(2): 249–264.

Granovetter MS. 1985. Economic action and socialstructure: the problem of embeddedness. AmericanJournal of Sociology 78: 1360–1380.

Grenadier SR, Weiss AM. 1997. Investment in technolog-ical innovations: an option pricing approach. Journalof Financial Economics 44: 397–416.

Gulati R. 1995. Does familiarity breed trust? Theimplications of repeated ties for contractual choicein alliances. Academy of Management Journal 38:85–112.

Gulati R. 1998. Alliances and networks. StrategicManagement Journal 19(4): 293–317.

Gulati R. 1999. Network location and learning: theinfluence of network resources and firm capabilitieson alliance formation. Strategic Management Journal20(5): 397–420.

Hagedoorn J, Schakenraad J. 1994. The effect of strategictechnology alliances on company performance.Strategic Management Journal 15(4): 291–309.

Hannan MT, Freeman J. 1977. The population ecology oforganizations. American Journal of Sociology 82(4):929–964.

Hitt MA, Dacin T, Levitas E, Arregle JL, Borza A. 2000.Partner selection in emerging and developed marketcontexts: resource-based and organizational learning

perspectives. Academy of Management Journal 43(3):449–467.

Hoskisson RE, Eden L, Lau CM, Wright M. 2000.Strategy in emerging economies. Academy ofManagement Journal 43(3): 249–267.

Kenny DA. 1979. Correlation and Causality . Wiley: NewYork.

Khanna T, Palepu K. 1997. Why focused strategies maybe wrong for emerging markets. Harvard BusinessReview 75(4): 41–51.

Kirchhoff BA. 1977. Organization effectiveness measure-ment and policy research. Academy of ManagementReview 2(3): 347–355.

Kogut B. 1988. Joint ventures: theoretical and empiricalperspectives. Strategic Management Journal 9(4):319–332.

Kogut B, Zander U. 1992. Knowledge of the firms,combinative capabilities and the replication ofknowledge. Organization Science 3: 383–397.

Kotabe M, Swan KS. 1995. The role of strategic alliancesin high technology new product development.Strategic Management Journal 16(8): 621–636.

Kraatz MS. 1998. Learning by association? Internationalnetworks and adaptation to environmental change.Academy of Management Journal 41(6): 621–643.

Lane PJ, Lubatkin M. 1998. Relative absorptive capac-ity and interorganizational learning. Strategic Manage-ment Journal 19(5): 461–477.

MacMillan IC, Siegel R, Narasimha RNS. 1985. Criteriaused by venture capitalists to evaluate new ventureproposals. Journal of Business Venturing 1(1):87–105.

Magrath AJ. 1992. The 6 Imperatives of Marketing:Lessons from the World’s Best Companies . AmericanManagement Association: New York.

McCann JE. 1991. Patterns of growth, competitivetechnology, and financial strategies in young firms.Journal of Business Venturing 6(3): 189–203.

McDougall PP, Covin JG, Robinson RB Jr, Herron L.1994. The effects of industry growth and strategicbreadth on new venture performance and strategy con-tent. Strategic Management Journal 15(7): 537–554.

McDougall PP, Robinson RB Jr. 1990. New venturestrategies: an empirical identification of eight‘archetypes’ of competitive strategies of entry.Strategic Management Journal 11(6): 447–467.

Miller D. 1987. The structural and environmentalcorrelates of business strategy. Strategic ManagementJournal 7(3): 233–249.

Miller A, Camp B. 1985. Exploring determinants ofsuccess in corporate ventures. Journal of BusinessVenturing 1(1): 87–105.

Miller CC, Cardinal LB, Glick WH. 1997. Retrospectivereports in organizational research: a reexaminationof recent evidence. Academy of Management Journal40(1): 189–204.

Nahapiet J, Ghoshal S. 1998. Social capital, intellectualcapital, and the organizational advantage. Academy ofManagement Review 23(2): 242–266.

Nee V. 1989. A theory of market transition: fromredistribution to markets in state socialism. AmericanSociological Review 54(October): 663–681.



North DC. 1990. Institutions, Institutional Change andEconomic Performance. Harvard University Press:Cambridge, MA.

O’Shea M, Stevens C. 1998. Governments as venturecapitalists. OECD Observer 213(August/September):26–29.

Pearce JA. 1983. The relationship of internal versusexternal orientations to financial measures of strategicperformance. Strategic Management Journal 4(4):297–306.

Peng MW, Heath PS. 1996. The growth of the firmin planned economies in transition: institutions,organizations, and strategic choice. Academy ofManagement Review 21(2): 492–528.

Pfeffer J, Salancik GR. 1978. The External Control ofOrganizations . Harper & Row: New York.

Phillips LW. 1981. Assessing measurement error inkey informant reports: a methodological note onorganizational analysis in marketing. Journal ofMarketing Research 18(November): 395–415.

Porter ME. 1980. Competitive Strategy . Free Press: NewYork.

Powell WW, Brantley P. 1992. Competitive cooperationin biotechnology: learning through networks. InNetworks and Organizations: Structure, Form andAction , Nohria N, Eccles R (eds). Harvard BusinessSchool Press: Boston, MA; 366–394.

Powell WW, Koput K, Smith-Doerr L. 1996. Interorga-nizational collaboration and the locus of innovation:networks of learning in biotechnology. AdministrativeScience Quarterly 41: 116–145.

Rumelt RP. 1974. Strategy, Structure, and EconomicPerformance. Harvard Business School Press: Boston,MA.

Sandberg WR, Hofer CW. 1987. Improving new ventureperformance: the role of strategy, industry structure,and the entrepreneur. Journal of Business Venturing2(1): 5–28.

Schoonhoven CB, Eisenhardt KM, Lyman K. 1990.Speeding products to markets: waiting time to firstproduct introductions in new firms. AdministrativeScience Quarterly 35(1): 177–207.

Shan W. 1990. An empirical analysis of organizationalstrategies by entrepreneurial high-technology firms.Strategic Management Journal 11(2): 129–139.

Shan W, Walker G, Kogut B. 1994. Interfirm cooperationand startup innovation in the biotechnology industry.Strategic Management Journal 15(5): 387–394.

Simerly RL, Li M. 2000. Environmental dynamism, capi-tal structure and performance: a theoretical integrationand an empirical test. Strategic Management Journal21(1): 31–49.

Stinchcombe AL. 1965. Organizations and social struc-ture. In Handbook of Organizations , March JG (ed.).Rand McNally: Chicago, IL; 142–193.

Stuart TE. 1998. Network positions and propensities tocollaborate: an investigation of strategic alliance for-mation in a high-technology industry. AdministrativeScience Quarterly 43: 668–698.

Stuart TE. 2000. Interorganizational alliances and theperformance of firms: a study of growth andinnovation rates in a high-technology firms. StrategicManagement Journal 21(8): 791–811.

Stuart TE, Hoang H, Hybels RC. 1999. Interorgani-zational endorsements and the performance ofentrepreneurial ventures. Administrative Science Quar-terly 44: 315–349.

Teece DJ. 1980. Economics of scope and the scope ofthe enterprise. Journal of Economic Behavior andOrganization 1: 223–247.

Varadarajan PR, Cunningham MH. 1995. Strategic allian-ces: a synthesis of conceptual foundations. Journal ofthe Academy of Marketing Science 23(4): 282–296.

Zahra SA. 1993. Environment, corporate entrepreneur-ship and financial performance: a taxonomic approach.Journal of Business Venturing 8(4): 319–340.

Zahra SA, Covin JG. 1993. Business strategy, technologypolicy, and company performance. Strategic Manage-ment Journal 14(6): 451–478.

Zhao L, Aram JD. 1995. Networking and growth ofyoung technology-intensive ventures in China. Journalof Business Venturing 10(5): 349–370.


Date post:	06-Sep-2018
Category:	Documents
Upload:	vantram
View:	214 times
Download:	0 times

THE ADOPTION OF AGENCY BUSINESS ACTIVITY, …haiyang/SMJ-AGency.pdf · THE ADOPTION OF AGENCY...

Documents