Determinants of transnational new product development capability: testing the influence of...

Strategic Management JournalStrat. Mgmt. J.,22: 359–378 (2001)

DOI: 10.1002/SMJ.163

DETERMINANTS OF TRANSNATIONAL NEWPRODUCT DEVELOPMENT CAPABILITY: TESTINGTHE INFLUENCE OF TRANSFERRING ANDDEPLOYING TACIT OVERSEAS KNOWLEDGE

MOHAN SUBRAMANIAM1* and N. VENKATRAMAN2

1Wallace E. Carroll School of Management, Boston College, Chestnut Hill,Massachusetts, U.S.A.2School of Management, Boston University, Boston, Massachusetts, U.S.A.

Based on a survey of 90 transnational product introductions, we find that the transnationalproduct development capabilities of organizations significantly depend upon their ability totransfer and deploy tacit knowledge concerning overseas markets. Specifically, we find thatorganizations which use cross-national teams, teams with members who have prior overseasexperience, or teams whose members communicate frequently with overseas managers in orderto acquire information about tacit differences among countries have greater transnationalproduct development capabilities. This study contributes to our understanding of how organi-zations transfer and deploy knowledge across borders for competitive advantage and makes animportant contribution to the literature on global strategy.Copyright 2001 John Wiley &Sons, Ltd.

INTRODUCTION

In recent years, interorganizational competitionhas not only intensified, its nature has fundamen-tally changed: it has become more knowledgebased, and the sources of competitive advantagehave shifted unmistakably from physical assets tointellectual resources (Prahalad and Hamel, 1990;Quinn, 1992; Stewart, 1997). While this emergentknowledge-based competition has affected a widespectrum of organizations, it has raised someparticularly significant challenges for firms com-peting internationally. The home-based sources ofknowledge and ‘ownership advantages’ that hadlong enabled organizations to compete effectivelyin international markets (Dunning, 1980) are nolonger adequate today. Global rivals now wrest

Key words: tacit knowledge; firm capabilities; globalstrategy; transnational new product development*Correspondence to: Mohan Subramaniam, Wallace E. CarrollSchool of Management, Boston College, Fulton Hall, 140Commonwealth Avenue, Chestnut Hill, MA 02467-3808,U.S.A.

Copyright 2001 John Wiley & Sons, Ltd. Received 23 March 1999Final revision received 13 October 2000

competitive initiative by harnessing knowledgefrom sources in multiple countries in order togenerate new product ideas as well as to buildmanufacturing know-how and technologicalstrength. Indeed, with the increasing globalizationof our economy, the ability to transfer and deployknowledge across borders has become one of thecentral competitive concerns for many organi-zations.

To date, however, we have a limited under-standing of how organizations manage this trans-fer and deployment effectively. Prior studies haveviewed knowledge flows between headquartersand subsidiaries of multinational companies(MNCs) largely as control or administrativemechanisms (e.g., Egelhoff, 1988; Gupta andGovindarajan, 1993; Nobel and Birkinshaw,1998). Consequently, not much is understoodabout the optimization of such internationalknowledge flows as a means for creating competi-tive advantage.

On the other hand, the competitive implicationsof organizations’ knowledge have recentlyattracted a great deal of attention in the strategic

360 M. Subramaniam and N. Venkatraman

management literature (e.g., Cohen and Levinthal,1990; Grant, 1996; Spender, 1996; Zander andKogut, 1995). Of particular interest has been thecompetitive significance of a firm’s tacit know-ledge (Nonaka, 1994). Tacit knowledge largely‘indwells’ in the minds of people as perspectiveson, or images of, reality (Polanyi, 1966). Theinherent difficulties in its codification and com-munication pose significant barriers to the repli-cation of this kind of knowledge by rival organi-zations—which makes tacit knowledge a keydifferentiator and therefore potentially animportant strategic resource (Conner and Pra-halad, 1996). However, the competitive impli-cations of tacit knowledge have not yet beenadequately tested in the international arena. Somestudies have recognized that, as one of the‘sticky’ properties (von Hippel, 1994) of location-specific knowledge, tacitness makes knowledgeespecially difficult to transfer across country bor-ders (Kogut and Zander, 1993; Teece, 1977), butnot much is understood about how firms over-come this difficulty and harness tacit knowledgefrom geographically dispersed sources. Also lack-ing are empirical studies validating the growingbelief that tacit knowledge may have a significantimpact on firm capabilities.

In an effort to fill in some of these gaps inour understanding, this study investigates theimpact of the international transfer and deploy-ment of tacit knowledge on a firm’s capabilitiesfor developing new transnational products. Trans-national products are those that are developedsimultaneously for multiple markets, and whichcontain both features that are standardized acrossmarkets and features that are responsive to indi-vidual local markets. Unlike global products thatare developed with an emphasis on the simi-larities among country markets (Levitt, 1983;Samiee and Roth, 1992), transnational productsreflect an MNC’s concurrent emphasis on simi-larities and differences among countries in theirproduct development efforts (Bartlett and Gho-shal, 1989). The simultaneous development ofproducts for multiple markets might also be calleda sprinkler strategy—in contradistinction to awaterfall strategy, whereby products are firstdeveloped for a single country market and vari-ations are subsequently and independentlydeveloped for other markets (Ganesh and Kumar,1996; Kalish, Mahajan, and Muller, 1995;Ohmae, 1985).

Copyright 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J.,22: 359–378 (2001)

BACKGROUND AND MOTIVATIONFOR THE STUDY

In designing our study of the impact of inter-national transfer and deployment of knowledgeon firm capabilities, we chose to focus on trans-national new product development for tworeasons. First, the transfer and deployment ofoverseas knowledge is a central component of thetransnational new product development process(Takeuchi and Porter, 1986). Because trans-national products are developed with both featuresthat adapt to unique country requirements andthose that are standardized across markets, com-panies must become adept at discerning the differ-ences among countries in order to address whatmay be conflicting requirements. Since knowledgeof unique country conditions typically lies dis-persed in a firm’s overseas subsidiaries (Ohmae,1990), the capability to develop new transnationalproducts fundamentally depends on how pro-ficiently MNCs transfer and deploy knowledgefrom multiple country sources.

Second, transnational product development hassignificant implications for the competitiveness ofMNCs in worldwide markets. A firm’s trans-national approach to product development pro-vides many advantages over competitors whomanage development either as a purely domesticprocess or as a portfolio of independent activitiesin different countries (Ghoshal, 1987; Porter,1986; Yip, 1995). For example, integrating devel-opment activities across countries eliminatesduplication of efforts and saves costs, makingproducts more price competitive (Porter, 1986).It also allows firms to leverage ideas generatedin one geographic market into other geographicmarkets (Bartlett and Ghoshal, 1989). Moreover,launching new products simultaneously in multi-ple country markets reduces the risks of preemp-tive entry by rivals in those markets (Chen andStucker, 1997; Hamel and Prahalad, 1985). Inaddition, a transnational approach enables MNCsto combine the advantages of local responsivenessand the efficiency that comes with centralization(Bartlett and Ghoshal, 1989; Prahalad and Doz,1987). While adaptation of certain product fea-tures to unique country requirements enablesproducts to penetrate individual markets, stan-dardization of other features may enhance theircompetitiveness by providing operationalefficiency. Such compelling advantages have in

Determinants of Transnational New Product Development Capability361

fact made transnational new product developmenta central aspect of some firms’ strategy for com-peting in worldwide markets (Bartlett and Gho-shal, 1989).

Despite its significance, however, the develop-ment of new transnational products has not beenexamined sufficiently in empirical studies. Priorstudies have focused primarily on the feasibilityof standardizing global products across markets(e.g., Boddewyn, Soehl, and Picard, 1986; Cavus-gil, Zou, and Naidu, 1993; Samiee and Roth,1992), as opposed to the development processesof transnational products. Similarly, studies ofglobal R&D have centered their attention on thelocation, coordination, or evolution of worldwideR&D activities (see Cheng and Bolon, 1993, fora review) rather than on the issues involvedin product development for worldwide markets.Research on the diffusion of knowledge withinMNCs (e.g., Bartlett and Ghoshal, 1989; Guptaand Govindarajan, 2000) has also failed to high-light the ways in which various approaches totransferring and deploying knowledge across bor-ders specifically affect MNCs’ capabilities fordeveloping transnational products. We believethat transnational new product development pro-vides an apt and useful context in which toinvestigate the impact on firm capabilities ofinternational knowledge transfer and deployment.

RESEARCH FRAMEWORK

Our research framework was designed to examinehow the transfer and deployment of tacit overseasknowledge influences the transnational new prod-uct development capabilities of MNCs. The studywas built on the premise that tacit knowledge isan important aspect of new knowledge creationand provides a strong impetus for innovation(Madhavan and Grover, 1998; Nonaka, 1994;Nonaka and Takeuchi, 1995).

Definitions

We define tacit overseas knowledge as ‘theknowledge of the differences among overseasmarkets that is difficult to codify and transfer ina systematic way.’ Note thatdifficulty in codifi-cation and transferis a central attribute of tacitknowledge (Grant, 1996; Nonaka, 1994; von Hip-pel, 1994; Zander and Kogut, 1995). We focus on


knowledge about thedifferences1 among overseasmarkets because addressing unique requirementsof multiple countries is a fundamental issue facingfirms developing transnational products (Cavusgilet al., 1993; Jain, 1989).

We define transnational new product develop-ment capability as ‘the ability to consistently andsuccessfully introduce new products si-multaneously in multiple country markets.’ Wefocus on consistencyof new product introduc-tions, since random or sporadic new product suc-cesses will not enable firms to achieve or main-tain market dominance (Leonard-Barton, 1995;Nobeoka and Cusumano, 1997). We qualify theseintroductions assuccessful, since market successis a key measure of any business process capa-bility (Amit and Schoemaker, 1993; Barney,1991). Finally, we also includesimultaneity ofintroductions, since the integration of productdevelopment activities across countries and thedevelopment of products that concurrently meetthe requirements of several countries are essentialto the concept of transnational new product devel-opment (Bartlett and Ghoshal, 1989; Porter,1986).

Tacit overseas knowledge and transnationalproduct development capabilities

As they strive to meet the requirements of multi-ple country markets simultaneously, productdevelopers must address the differences amongcountries (Jain, 1989). Knowledge about thesedifferences may be tacit or explicit(Subramaniam, Rosenthal, and Hatten, 1998).Many country requirements entail explicit knowl-edge, since they are based on universallyaccepted and objective criteria. For instance,transmission systems for televisions differ fromcountry to country based on universally acceptedengineering specifications such as PAL, SECAM,or NTSC. Similarly, cordless telephones mustrespond to individual countries’ regulatory lawsregarding frequency ranges. Such differences arenot subject to individual interpretations or per-spectives, can easily be codified and communi-

1 While a broader view of tacit overseas knowledge couldinclude tacit similarities as well as differences, we chose tofocus our analysis on tacit differences among countries. Wepresent the implications of this choice subsequently when wediscuss our results.


cated internationally, and are consequently likelyto be understood in the same way by most com-petitors.

On the other hand, many differences amongcountry markets involve tacit knowledge—anunspoken and often subtle understanding of dif-ferences in cultures, tastes, habits, or customs(Jain, 1989; Subramaniamet al., 1998). Takeuchiand Porter (1986), for example, describe the cul-tural expectations that made Campbell’s soupunpopular in Brazil. Brazilian housewives appar-ently felt they were not fulfilling their ‘proper’role if they served soup they could not call theirown, and hence preferred using dehydrated soupstarters (made by Campbell’s competitors) towhich they could add their own ingredients. Theunderstanding of such unique country require-ments is tacit, as it is based upon the personalperspectives and interpretations of the individualsobserving consumer behavior.

Since such cultural quirks lack objective cri-teria for consistent description, they can havemultiple interpretations—with each interpretation,or version of tacit knowledge, having differentimplications for the design trade-offs of the newtransnational product being developed. Forinstance, preferences for the shapes of televisioncabinets vary with cultures and may be interpretedin several ways, making it very difficult to codifyknowledge about them and transfer it across bor-ders (von Hippel, 1994). Unlike the ability tograsp explicit differences, the insight needed toarrive at these interpretations evolves as onebuilds one’s personal experiences with the culture.Even then, the knowledge will reside primarilyin the minds of subsidiary managers located inoverseas markets, who are, moreover, unlikely toperceive and understand it in exactly the sameway as their firm’s global competitors do.

Consequently, such tacit insights provide richpotential for identifying market opportunities thatare not easily detected by competitors. Forexample, Sanyo spotted a unique innovationopportunity for its cordless telephones by inter-preting a tacit difference in habit between its U.S.and home consumers. Facing frequent complaintsabout cordless telephone batteries failing torecharge in the U.S. market, they discovered theproblem to be associated with a cultural traitpeculiar to the United States: American con-sumers tended to snack while using the telephone,and hence soiled the physical contacts for battery


recharging. The design for Sanyo’s new trans-national cordless telephone harnessed this insightby incorporating an electromagnetic battery-recharging feature as opposed to the conventionalmeans of physical contact terminals. Because tacitknowledge can work in this way to turn idiosyn-cratic experiences and interpretations into uniquemarket opportunities, the transfer and deploymentof such knowledge can be a key driver of inno-vation (Nonaka and Takeuchi, 1995).

Transferring and deploying tacit overseasknowledge

Of course, while tacit knowledge may originate asindividual experiences and perceptions (Polanyi,1966), for organizations to harness this knowledgeand turn it into innovative capabilities, theseexperiences of individual subsidiary managersmust be processed into a collective understandingamong the MNC’s innovation team, and thentransformed into explicit transnational productfeature options (Madhavan and Grover, 1998;Nonaka and Takeuchi, 1995). From an infor-mation-processing theory perspective, the extentof tacit overseas knowledge assimilated into thetransnational new product development processcould be seen as a function of a ‘fit’ betweenthe tacitness in the overseas information acquiredand the richness of the information-processingmechanisms employed (Galbraith, 1977; Tushmanand Nadler, 1978).

A ‘fit’ indicates how well organizations makesense of and apply the information they acquirein a specific context (Daft and Lengel, 1986).The concept of ‘fit’ is based on the premise thatorganizations need to match ambiguous or diffi-cult to acquire information with rich information-processing mechanisms (Gresov, 1989; Van deVen and Delbecq, 1974). Because tacit infor-mation is difficult to codify and transfer, assimi-lating such information effectively requires theuse of rich information-processing mechanisms(Subramaniamet al., 1998).

For example, Sanyo says it positions itscordless telephone as a product with a ‘softappearance,’ but the company is fully consciousof the vagueness of the term, recognizing that itmust acquire information about what ‘soft appear-ance’ means to its customers in different countrymarkets. As there are numerous possible interpre-tations of ‘soft appearance’ even within a single


country, a comprehensive understanding of all thelocal nuances of such a quality exists only in theminds of the relevant overseas managers, whomust base their definitions on their own experi-ences and beliefs. A product development teamat Sanyo’s headquarters is unlikely to grasp allthe nuances of its diverse overseas managers’knowledge about ‘soft appearance’ if the teamrelies upon lean2 information-processing mecha-nisms such as written memos or reports in orderto acquire that information. The depth and breadthof this kind of impressionistic understanding arevery difficult to document comprehensively inmemos and reports. Such a tacit knowledge baseis more likely to be harnessed effectively by thedevelopment team if it employs richer mecha-nisms such as face-to-face contact with overseasmanagers (Daft and Lengel, 1986; Krauss andFussell, 1991). Using these richer modes of infor-mation exchange, the developers can review vari-ous design alternatives in considerably greaterdetail, and they can utilize the overseas managers’knowledge more effectively in the product designprocess. Hence, the effectiveness of transferringand deploying tacit overseas knowledge is a func-tion of a ‘fit’ between tacit information and richinformation-processing mechanisms. Moreover,since MNCs which achieve such a ‘fit’ will bemore effective at transferring and deploying tacitoverseas knowledge, they are also more likely tohave greater transnational product developmentcapabilities (see Figure 1).

Hypotheses

Our hypotheses build upon the three approachesby which MNCs engaged in transnational newproduct development projects might effectivelytransfer and deploy tacit knowledge through a‘fit’ between the extent of tacitness in the infor-mation they acquire and the richness of themechanisms they use to process this information.

One approach is to use cross-national teams toprocess tacit information about overseas markets.Cross-national teams co-locate the managers whopossess the requisite tacit overseas information

2 Based on Daft and Lengel (1986), we view richness ofinformation-processing mechanisms as a continuum, withleanmechanisms connoting the low end of that continuum. Hence,while face-to-face contact is an example of a rich media ofinformation of exchange, written memos are an example ofa lean media of information exchange.


with other managers who can assist in interpretingthat information and applying it in product design.Co-location enhances the media richness of infor-mation transfer (Clark and Marshall, 1981; Daftand Lengel, 1986), and facilitates the effectivesharing of deep-rooted perspectives, beliefs, andexperiences. Consequently, cross-national teamsare considered one of the primary mechanisms forsharing tacit information (Madhavan and Grover,1998; Nonaka, 1994). Sanyo’s use of face-to-face contact between team members to captureknowledge about country-specific perceptions of‘soft appearance’ is a case in point. Co-locationalso enables groups to bring the sources and usersof ‘sticky’ information together and so to enhancetheir problem-solving abilities (von Hippel, 1994).Enhancing the communication and application oftacit knowledge in this way should ultimatelystrengthen transnational product developmentcapabilities. Hence:

Hypothesis 1: A ‘fit’ between the extent oftacitness in overseas information acquired andthe use of cross-national teams will enhancetransnational new product development capa-bility.

A second approach is to include on the productdevelopment team that must process this kind oftacit information members who have had prioroverseas experience. Such experience providesthe team with a greater ability to make sense ofnew country-specific information, to absorb iteffectively, and to apply it to the creation of newknowledge (Bower and Hilgard, 1981; Cohen andLevinthal, 1990). Dierickx and Cool (1989)describe this ability as ‘asset mass efficiencies’—the snowball effect by which prior knowledgeenables its holders to add more effectively tonew knowledge. Nonaka and Takeuchi (1995)provide some examples of Nissan’s use, for theirtransnational car design, of engineers who hadhad European market experience and thereforehad some tacit knowledge about European drivingconditions. This prior experience and the conse-quent tacit knowledge about the relevant overseasmarkets enhanced the team’s ability to absorbrelated new knowledge about those markets. Thisleads to our second hypothesis:

Hypothesis 2: A ‘fit’ between the extent oftacitness in overseas information acquired and


Figure 1. Transferring and deploying tacit overseas knowledge for transnational new product developmentcapability

the use of teams with domestic members whohave prior overseas experience will enhancetransnational new product development capa-bility.

A third approach to processing tacit overseasinformation is to increase the communicationintensity of project teams (Allen, 1977; Tushmanand Nadler, 1978). Intensive communicationenables individuals to build strong ties(Krackhardt, 1992) and share their beliefs andexperiences so as to transfer tacit informationmore effectively (Madhavan and Grover, 1998;Nonaka, 1994). Frequent communication betweenMNC headquarters and subsidiary managers isalso known as an effective diffuser of knowledgeacross borders (Egelhoff, 1988; Ghoshal, Korine,and Szulanski, 1994; Nohria and Ghoshal, 1994).Thus, the greater the frequency of communicationbetween project team members and overseas man-agers, the more effectively will tacit informationabout the differences among countries be proc-essed, resulting in greater deployment of suchknowledge. Hence:


Hypothesis 3: A ‘fit’ between the extent oftacitness in overseas information acquired bythe project team and the frequency of com-munication between team members and over-seas managers will enhance transnational newproduct development capability.

METHODOLOGY

In conducting our research, we used a cross-sectional survey of key informants to collect pri-mary data from a set of companies competing inworldwide markets. The survey was administeredto key members of transnational new productdevelopment teams. A single informant speakingon behalf of the team provided information abouteach transnational new product development proc-ess. The survey was administered across multipleindustries in the consumer packaged goods, con-sumer durables and industrial product sectors. Itsscope was restricted to manufactured products.We selected for the survey only products that hadbeen on the market for between 1 and 5 years.


Sampling

Sampling of companies and their transnationalnew product development processes presentedsome unique problems. Not all MNCs developtransnational products, nor do all product devel-opment projects within a single MNC involvetransnational products. Hence we concluded thatsending questionnaires randomly to a set ofMNCs was not likely to result in adequate mean-ingful responses. We first created a target samplebase of companies that were likely developers oftransnational products. Then we identified man-agers who were involved with transnational newproduct development projects and could serve askey informants for this study.

Through multiple approaches, we created a tar-get sample base of 152 companies (see AppendixA). First, we solicited the help of ACNielsen, amajor market research company in the UnitedStates, in identifying 90 of their client companiesinvolved in transnational new product develop-ment. Based on discussions with their representa-tives, we selected a list of marketing and R&Dmanagers of these companies who were likely tobe involved in transnational new product develop-ment. Second, we identified 22 more MNCs thatwere corporate sponsors of six research centersof the university where this research was conduc-ted. From these corporate sponsors we obtaineda list of managers who were likely to be involvedin transnational new product development. Third,we added 40 more MNCs that were known fromprior academic and business articles to bedeveloping transnational products. For these com-panies we created a mailing list of informantsfrom the Directory of Corporate Affiliations.

We then invited these 152 MNCs to participatein our study. Fifty-seven of them participated,giving us a response rate of 37.5 percent. Anindependent t-test comparing mean revenues(obtained from secondary sources) of theresponding and nonresponding companiesrevealed no significant differences (atp < 0.05),ruling out any significant response bias. Sincemany of these MNCs were involved indeveloping several distinct categories of trans-national products, we encouraged them to giveus multiple responses. We received a total of 102responses from different divisions and productcategories within these 57 MNCs. We discarded12 of these responses: three pertained to non-


manufactured products (software and franchiseexpansion); five responses were about new prod-ucts not yet launched; and four more were incom-plete. This left a total of 90 usable responsesfrom 45 multinational companies. Of these 90responses, 43 were from the consumer packagedgoods sector, 23 from the consumer durablessector, and 24 from the industrial products sector(see Appendix B for details on the companiesand product categories).

The sampling process was by design nonran-dom, since transnational new product develop-ment is not widely practiced but is rather astrategic choice adopted by some MNCs. How-ever, our sample represented a broad cross-sectionof product categories and firms. Several of thefirms represented are leading multinational com-panies who account for a significant portion ofthe global economy. Twenty-two of these com-panies were ranked in the 1997Fortune 1000 listin terms of revenue. Of these 22 companies, eightwere ranked in the top 100, of which four werein the top 50. Many of these companies have beencited in prior studies as exemplary competitors inworldwide markets (for example, see Yip, 1995).A majority of the companies in the sample wereU.S. based. Three companies in the sample werebased in Europe, one in Japan and one inSouth Korea.

Key informants

We used key informants to provide data on trans-national new product development processes. Se-lecting appropriate informants is critical, since thereliability of information about activities related toa group or an organization depends on whetherthe selected persons have the requisite knowledgeor are in a position to have it. All managerswho responded to our survey were members oftransnational new product development teams andwere also in the upper middle management levelof their companies. A majority of the informants(62%) were senior managers carrying titles ofDirector, Vice President or President. They thushad the appropriate vantage point from which toprovide information on the complex and multilevelactivities involved in transnational new productdevelopment. Also, 92 percent of the informantsrepresented the functions of R&D/Engineering,Marketing and General Management, all of whichare critical functions for new product development.


Survey instrumentation

We developed the survey instrument in severalphases. The first phase consisted of several hoursof interviews with managers involved in the prac-tice of transnational new product development.These interviews gave us an appreciation of howthis practice was actually conducted in multi-national companies. Using this understanding asa background, we then systematically searchedthe literature on new product development, inter-national business, strategic management, andorganization theory for relevant scales. Thissearch resulted in only a few scales that wereapplicable to the conceptual framework andhypotheses of this study. For a majority of theconstructs, therefore, we developed new scales.

Measures

The measure for the dependent variable—trans-national new product development capability—was developed by integrating key indicators fromprevious related studies. These indicators include:(a) frequency of new product introductions(Nobeoka and Cusumano, 1997), (b) order ofmarket entry (Banbury and Mitchell, 1995), (c)simultaneous entry in multiple markets (Porter,1986), (d) the ability to be responsive to marketrequirements, (e) the ability to be competitive interms of price (Bartlett and Ghoshal, 1989), and(f) the ability to penetrate new overseas markets(Yip, 1995). (See Appendix C, #1.)

The measure for the degree of tacitness wasadapted from Zander and Kogut (1995). Threedimensions from their study relevant to this con-text were complexity, codifiability, and observ-ability of information about the differences amongcountries. That is, when the information aboutthe differences among countries was more com-plex, less codifiable, and more difficult toobserve, that information was considered moretacit and less explicit. Based on these concepts,we developed some fine-grained measures to cap-ture the construct (see Appendix C, #2).

We used a dummy variable denoting thepresence/absence of overseas managers in theproject teams in order to measure co-location orthe existence of cross-national teams (see AppendixC, #3). Similarly, we used a dummy variable denot-ing the presence/absence on the project team ofdomestic managers with prior overseas experience


in order to measure the use of teams with suchmembers (see Appendix C, #4). Following Ghoshalet al. (1994), we measured communication intensityas the frequency of interactions between projectmembers and overseas managers via telephone, fax,and e-mail (see Appendix C, #5).

Pilot testing

We pilot tested the face validity of these scaleswith 16 managers involved in transnational newproduct development. We spent a minimum of 2hours with each of these managers, discussingevery question and indicator with them so as toascertain that they understood the constructs andwere interpreting the questions accurately. Basedon their feedback, we reworded some of the ques-tions for greater clarity and comprehensibility.

Measurement properties

All constructs using multiple indicators weretested for their reliability. Cronbach alphas for allthe constructs were well above the recommendedvalue of 0.7 (see Appendix C). Furthermore, theconstructs were tested for their convergent anddiscriminant validity. Each construct was pairedwith another construct, and all the pair combi-nations were factor analyzed using varimaxrotation. The indicators of each construct loadedonly on their own construct for all the pairsof constructs. Hence, convergent and discriminantvalidity requirements were satisfied for these con-structs (detailed results available on request). Inaddition, we developed a measure for businessperformance as a means to test the predictivevalidity of our dependent variable. Following Ven-katraman and Ramanujam (1987), this measurecaptured multiple dimensions of business perfor-mance—namely, how satisfied managers were,vis-a-vis their competitors, with their division’sgrowth in sales, growth in market share, returnon investment and return on sales. A significantzero-order correlation of 0.49 (p < 0.01) betweentransnational new product development capabilityand business performance confirmed the predictivevalidity of the dependent variable.

Control factors

We used seven variables to control for factors,other than the tacit overseas knowledge leveraged


by the project team, which might have an influ-ence on transnational new product developmentcapability. The first two variables—brand imagein overseas markets and overseas market share—were chosen because, as factors influenced bysome of the same ‘other’ variables that affecttransnational new product development capability,they covary with that capability. Because of thiscovariance, using these control variables enablesus to control for several of these ‘other’ factorsthat are impossible to enumerate comprehensivelyand to incorporate into the framework. We con-trolled for industry by using two dummy variablesto indicate whether the product was a consumerpackaged good, a consumer durable, or an indus-trial product.

The remaining control variables were chosenbecause their influence on transnational new prod-uct development capability can be inferred fromprior studies. The stronger a firm’s worldwidemarketing infrastructure is, the better its structurewill be for competing in worldwide markets(Porter, 1986), and hence the more likely it willbe to have greater transnational new productdevelopment capabilities. Similarly, the greaterthe number of countries targeted for transnationalnew product development, the greater the firm’schances for simultaneous introductions in multiplemarkets—and hence the greater its likelihood forhigher transnational new product developmentcapability. Next, the higher the market concen-tration, the lower the likely market rivalry, andhence the greater the odds of the new trans-national product’s being successful in the marketplace (Porter, 1980). Finally, we used per-meability of managers across borders to representan organizational factor believed to enhance theinternational flow of knowledge and therefore thechances of innovation. Based on Bartlett andGhoshal (1989), this construct represents the fre-quency with which managers from headquartersand overseas subsidiaries of the focal organizationvisit one another and interact informally with oneanother (see Appendix C, #6 to #11).

RESULTS

Table 1 summarizes the means, standard devi-ations, and zero-order correlations among the con-structs used for testing the hypotheses.

We used multiple regression analysis to test


our hypotheses.3 Table 2 presents the results ofthe regression analysis.

Model 1 is the test for the control variables.Together the control variables contribute to anadjustedR2 of 0.564, and theF-statistic is highlysignificant (p < 0.001). This confirms oura prioriexpectations of the influence of the selected con-trol variables on transnational new product devel-opment capability. Model 2 is the test of themain effects of the extent of tacitness in overseasinformation and cross-national teams. The resultsreveal that the main effects are not significant,and taken individually neither the extent of tacit-ness in overseas information nor the use of cross-national teams is significant. Model 3, which is atest of Hypothesis 1, confirms that the interactionbetween the degree of tacitness in overseas infor-mation and the use of cross-national teams issignificant at p < 0.001. This result supportsHypothesis 1.

Model 4 tests the influence of the main effectsof the extent of tacitness in overseas informationand the use of teams with members who haveprior overseas experience. The results reveal thatthe main effects are not significant. Model 5,which tests Hypothesis 2, confirms that the inter-action between the extent of tacitness and the useof teams with domestic members who have prioroverseas experience is significant atp < 0.05.This result supports Hypothesis 2.

Model 6 is the test of the main effects of theextent of tacitness in overseas information andthe frequency of communication between projectteam members and overseas managers. The resultsreveal that taken individually the extent of tacit-ness and the frequency of communication of teammembers with overseas managers are not signifi-cant explanators of the variance in transnationalnew product development capability. Model 7,which tests Hypothesis 3, reveals that their inter-action is significant. This result supports Hypoth-esis 3.

Overall, all the hypotheses are strongly sup-

3 Factor analysis with varimax rotation confirmed that allmultiple indicator variables loaded on a single factor. Con-sidering the factor scores as weights, the respective weightedaverages of the indicators were used to compute the value ofeach multiple indicator variable. Furthermore, as suggestedby Venkatraman (1989), interactions were computed by multi-plying respective independent variables after transformingthem by centering them around their mean to avoid multi-collinearity problems.

368M

.S

ub

ram

an

iam

an

dN

.V

en

katra

ma

n

Table 1. Means, standard deviations and zero-order correlations

1 2 3 4 5 6 7 8 9 10 11 12 13

Mean 5.15 0.7 0.81 5.14 0.47 0.29 4.9 4.53 4.07 6.89 3.64 4.71 5.89Standard deviation 1.49 0.46 0.40 1.62 0.5 0.46 1.61 1.52 0.96 5.03 1.85 1.53 1.381. Extent of tacitness 1.002. Cross-national teams –0.019 1.003. Teams with members 0.066 0.097 1.00with prior overseasexperience4. Frequency of 0.086 0.421∗∗ 0.111 1.00communication5. Industry represented (A) –0.115 0.104 –0.027 0.131 1.006. Industry represented (B) 0.091 0.068 0.084 –0.164 –0.60∗∗ 1.007. Brand image –0.048 0.167 –0.038 0.109 0.279∗∗ 0.035 1.008. Overseas market share 0.04 0.012 –0.067 0.107 0.060 0.035 0.662∗∗ 1.009. Worldwide marketing –0.016 0.147 0.344∗∗ –0.001 0.050 0.046 0.212∗ 0.135 1.00infrastructure10. Number of countries –0.092 –0.073 0.070 –0.063 0.099 –0.001 –0.070 –0.047 –0.133 1.00targeted11. Market concentration 0.167 0.244∗ 0.057 0.176 0.005 0.081 0.083 0.118 0.015 –0.066 1.0012. Permeability of 0.035 0.243∗ 0.312∗∗ 0.197 0.067 –0.080 0.099 0.169 0.015 –0.068 –0.005 1.00managers across borders13. Transnational new 0.039 0.253∗ 0.151 0.128 0.092 0.078 0.656∗∗ 0.616∗∗ 0.206∗ 0.038 0.244∗ 0.333∗∗ 1.00product developmentcapability

∗p < 0.05, ∗∗p< 0.01

Copyright

2001John

Wiley

&S

ons,Ltd.

Stra

t.M

gm

t.J.,22:

359–378(2001)

De

term

ina

nts

of

Tra

nsn

atio

na

lN

ew

Pro

du

ctD

eve

lop

me

nt

Ca

pa

bility369

Table 2. Regression results

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Control variablesGeographic scope 0.039∗ 0.042∗ 0.047∗ 0.020+ 0.038∗ 0.039∗ 0.039∗

Industry represented (A) –0.199 –0.247 –0.231 –0.106 –0.142 –0.195 –0.273Industry represented (B) –0.045 –0.019 –0.022 –0.122 –0.052 0.052 0.018Worldwide infrastructure 0.129 0.106 0.09 0.065 0.129 0.129 0.149Market concentration 0.127∗ 0.09+ 0.05 0.122∗ 0.122∗ 0.122∗ 0.117∗

Overseas market share 0.226∗∗ 0.249∗∗ 0.299∗∗∗ 0.237∗∗ 0.223∗∗ 0.223∗∗ 0.194∗∗

Brand image 0.410∗∗∗ 0.383∗∗∗ 0.341∗∗∗ 0.426∗∗∗ 0.41∗∗∗ 0.41∗∗∗ 0.425∗∗∗

Permeability of managers 0.235∗∗∗ 0.207∗∗ 0.232∗∗∗ 0.233∗∗∗ 0.285∗∗∗ 0.232∗∗∗ 0.230∗∗∗

across borders

Independent variables(A) Extent of tacitness 0.024 0.034 0.03 0.026 0.018 0.037(B) Cross-national teams 0.389 0.475+

(C) Teams with members 0.214 0.185having prior overseasexperience(D) Frequency of 0.060 0.018communicationInteraction between A & B 0.587***

Interaction between A & C 0.379*

Interaction between A & D 0.092*

Model statisticsR2 0.604 0.619 0.685 0.625 0.644 0.61 0.64Adjusted R2 0.564 0.570 0.639 0.575 0.592 0.553 0.577F 15.09∗∗∗ 12.51∗∗∗ 15.03∗∗∗ 12.64∗∗∗ 12.34∗∗∗ 11.78∗∗∗ 11.79∗∗∗

Change inR2 0.066 0.019 0.030F for change inR2 5.44*** 1.55 2.25+

+p < 0.1; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; N = 90

Copyright

2001John

Wiley

&S

ons,Ltd.

Stra

t.M

gm

t.J.,22:

359–378(2001)


ported. Also the difference inR2 between theinteraction models and the main effects modelsare significant for two out of the three hypotheses(for Hypothesis 1 atp < 0.001 and for Hypothesis3 at p < 0.1).

Sensitivity to firm effects

Several of the firms that participated in the studygave us data for multiple projects, and we wantedto test the robustness of our results after con-trolling for firm effects. However, controlling forall the firms in the sample was not feasiblebecause of limited degrees of freedom to test theresults. Hence, we created a subsample of 40projects from 11 MNCs that had each contributedinformation on three or more projects. Table 3presents the results after controlling for the firmeffects of these MNCs through ten dummy vari-ables.

Models 11, 12, and 13 reveal that two out ofthe three hypotheses are supported in the trun-cated sample test. The interactions between theextent of tacitness in overseas information andthe use of cross-national teams (Hypothesis 1)

Table 3. Regression results controlling for firm effects

Model 11 Model 12 Model 13

Control variablesTen dummy variables for 11 companies (not reporteda) (not reported) (not reported)Geographic scope –0.0016 –0.018 –0.032Worldwide infrastructure 0.130 0.016 0.003Market concentration 0.228∗ 0.273∗ 0.223∗

Overseas market share 0.387∗ 0.372∗ 0.157Brand image 0.185 0.388∗ 0.405∗∗

Permeability of managers across borders 0.099 0.113 0.113

Independent variables(1) Extent of tacitness –0.078 –0.002 0.062(2) Cross-national teams 0.838(3) Teams with members having prior overseas 0.392experience(4) Frequency of communication 0.112Interaction between 1 & 2 0.670*

Interaction between 1 & 3 0.350Interaction between 1 & 4 0.18*

Model statisticsR2 0.82 0.825 0.823Adjusted R2 0.657 0.659 0.663F 5.041*** 4.96*** 5.14***

aThe beta coefficients of the 10 dummy variables were nonsignificant and are not reported for sake of brevity.+p < 0.1; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; N = 40


and the frequency of communication betweenproject team members and overseas managers(Hypothesis 3), respectively, remain significant.However, the interaction between the extent oftacitness in overseas information and the use ofteams with prior overseas experience loses itssignificance. Our interpretation of these truncatedsample test results is that the results of thebroader study are fairly robust.

DISCUSSION

Transnational new product development capabilityis a key driver in establishing a firm’s competitiveadvantage in worldwide markets. Prior research,while clearly emphasizing the importance of atransnational approach to new product develop-ment if one is to compete effectively in today’sglobal business environment (e.g., Bartlett andGhoshal, 1989; Takeuchi and Porter, 1986), hasnot focused on how this capability could beeffectively developed. Our study highlights waysin which organizations can create this criticalcapability by transferring and deploying tacit


overseas knowledge. The study finds significantpositive effects of transferring and deploying tacitoverseas knowledge after controlling for theeffects of several other factors known to influenceorganizational performance. Our findings providecompelling evidence that organizations effectivelytransfer and deploy tacit overseas knowledge forgreater transnational new product developmentcapabilities by acquiring information about tacitdifferences among countries through rich infor-mation-processing mechanisms.

Differences among countries have long beenunderstood to complicate the process ofdeveloping products simultaneously for multiplecountry markets (e.g., Boddewynet al., 1986;Buzzell, 1968). Moreover, cultural differencesamong countries have been singled out as primeculprits in creating difficulty (Jain, 1989). How-ever, prior studies have neither examined noreven noticed the competitive implications of thechoices firms make about which kinds of differ-ences they will address. Our study suggests thatit is useful to focus on tacit differences amongcountries despite the intrinsic difficulties in doingso, since tacit information could provide firmswith unique insights leading to greater innovationopportunities. Indeed, our results confirm thatfirms which harness tacit insights about overseasmarkets are more likely to have greater trans-national new product development capabilities.

Focus on tacit differences among countries

In our study we chose to limit our considerationof tacit overseas knowledge to knowledge abouttacit differences, as opposed to tacit similarities,among country markets. We did so because webelieve understanding the differences amongcountries is a critical issue for managersdeveloping transnational products. Indeed, evenin focusing solely on tacit differences, our studyfinds unequivocal support for the positive effectsof tacit overseas knowledge on transnational prod-uct development capability. However, it isimportant to recognize that knowledge about tacitsimilarities among countries may also be signifi-cant and might well complement a firm’s knowl-edge about tacit differences. Future studies couldexamine whether including similarities in tacitoverseas knowledge might refine our understand-ing of the variance in transnational product devel-opment capabilities. It is possible that the effects


would be greater simply because deploying moretacit overseas knowledge of any kind might leadto greater capability. It is also possible that theadditional effects would not be significant, oreven that researchers would find that tacit simi-larities had already been accounted for: whenoverseas managers provide information aboutwhat they believe to be unique characteristics oftheir own markets, centralized transnational prod-uct development teams may be sifting throughthat information and picking up cues about simi-larities as well. In addition to examining theeffects of understanding tacit similarities, futurestudies could also broaden the analysis to includeother aspects of tacit knowledge in order to testtheir impact on firm capabilities.

CONTRIBUTIONS

This study makes several important contributionsto theory, methodology, and business practice.From a theoretical standpoint, it contributes tothe global strategy literature by providing newinsights into a previously unaddressed but criticalaspect of competing in worldwide markets. Byconsidering knowledge as a primary asset withinorganizations, this study also contributes to theemerging knowledge-based view of the firm. Thisview contends that the competitive advantage ofa firm lies in its ability to create, store, and applyknowledge (Conner and Prahalad, 1996; Spender,1996; Zander and Kogut, 1995). Of particularinterest to this research stream has been the sig-nificance of tacit knowledge within organizations.Tacit knowledge is now widely believed to bean important and untapped source of competitiveadvantage. This study is one of the first to provideempirical evidence of the critical role of tacitknowledge in influencing firm capabilities.

From a methodological perspective, this studydevelops new measures for key constructs suchas tacit knowledge and firm capabilities. Themeasures exhibit acceptable properties and couldtherefore provide valid approaches for future stud-ies, which might refine as well as apply them.From a managerial perspective, this study pro-vides guidelines for managers who must makesome critical choices in the process of developingnew transnational products. It emphasizes theimportance of designing transnational productswith features geared toward differences among


countries that are not apparent to all competitors.A few possible limitations of this study may

be apparent. One is our use of a single informantfor each project. However, our focus on veryspecific project-team-level activities (as vs.broader issues such as organizational culturewhere one might find considerable heterogeneityacross different subunits) and our collection ofinformation from a knowledgeable project teammember mitigate the general weakness associatedwith the use of a single informant (Venkatramanand Grant, 1986). Another related problem mightbe a common method bias. However, consideringthat all the hypotheses were based upon inter-action effects rather than main effects, it isunlikely that the common method bias wouldhave produced our results. In other words, it isunlikely that managers would have an ‘inter-action-based theory’ in their minds that could besystematically biasing their responses and theseresults.

Finally, it should also be noted that theincrementalR2 of the effects of tacit knowledgeare small compared to the cumulative effects ofall the control variables put together. Theseeffects are nevertheless important, since theyunderscore a relatively unconventional approachmanagers might take to enhance transnational newproduct development capabilities. Discerning suchan atypical approach to enhancing their capabili-ties provides companies with fresh avenues andmore opportunities to maintain or widen theirlead over their competitors. This is particularlytrue as multinational rivals are fast reaching parityin how they compete, and companies can nolonger afford to ignore approaches that createdifferential capabilities—even if their effects aresmall. We believe that our study has an importantrole to play in disseminating to practitioners suchnew ideas and approaches.

CONCLUSIONS

Our study sheds light on some important butunaddressed aspects of innovation processes offirms competing in worldwide markets. We exam-ined the ways in which firms effectively transferand deploy overseas knowledge in order to createsuperior capabilities for transnational new productdevelopment. Our evidence confirms that thetransfer and deployment of tacit overseas knowl-


edge is associated with greater transnational newproduct development capabilities. Our results alsoreveal that the effective deployment of suchknowledge entails its acquisition by productdevelopment teams through rich information proc-essing mechanisms. These mechanisms includecross-national teams, teams with members whohave prior overseas experience, and frequent com-munication among project team members andoverseas managers.

ACKNOWLEDGEMENTS

This study has been supported by the CarnegieBosch Institute through a grant on the topic‘Knowledge in International Corporations’ andthe Boston University Systems Research Center.This material is also based on the work supportedby the National Science Foundation under grantnumber SBR-9422284 to the second author. Anyopinions, findings, conclusions, or recommen-dations expressed in this material are those of theauthors and do not necessarily reflect the viewsof the National Science Foundation.

We wish to thank Luis Martins, Michael Lubat-kin, Steve Floyd, Michael Hitt, Robert Nixon,William Schulze, and Mark Youndt for their com-ments on earlier drafts of this paper. We alsothank Steve Rosenthal, Sushil Vachani, and Ken-neth Hatten for their inputs. We are grateful totwo anonymous reviewers for their helpfulsuggestions.

REFERENCES

Allen TJ. 1977. Managing The Flow of Technology:Technology Transfer and the Dissemination of Tech-nological Information within the R&D Organization.MIT Press: Cambridge, MA.

Amit R, Schoemaker P. 1993. Strategic assets andorganizational rents.Strategic Management Journal14(1): 33–46.

Banbury CM, Mitchell W. 1995. The effect of introduc-ing important incremental innovations on marketshare and business survival.Strategic ManagementJournal, Summer Special Issue16: 161–182.

Barney JB. 1991. Firm resources and sustained competi-tive advantage.Journal of Management17(1): 99–120.

Bartlett CA, Ghoshal S. 1989.Managing Across Bor-ders. Harvard Business School Press: Boston, MA.

Boddewyn JJ, Soehl R, Picard J. 1986. Standardization


in international marketing: is Ted Levitt in factright? Business Horizons29(6): 69–75.

Bower GH, Hilgard ER. 1981.Theories of Learning.Prentice-Hall: Englewood Cliffs, NJ.

Buzzell R. 1968. Can you standardize multinationalmarketing?Harvard Business Review46(6): 102–113.

Cavusgil ST, Zou S, Naidu GM. 1993. Product andpromotion adaptation in export ventures: an empiri-cal investigation.Journal of International BusinessStudies24(3): 479–506.

Chen M-J, Stucker K. 1997. Multinational managementand multimarket rivalry: toward a theoretical devel-opment of global competition.Academy of Manage-ment Proceedings, Boston, 2–6.

Cheng JLC, Bolon DS. 1993. The management ofmultinational R&D: a neglected topic in internationalbusiness research.Journal of International BusinessStudies24(1): 1–18.

Clark HH, Marshall CR. 1981. Definite reference andmutual knowledge. InElements of Discourse Under-standing, Aravind J, Webber BL, Sag IA (eds).Cambridge University Press: Cambridge, UK; 10–63.

Cohen WM, Levinthal DA. 1990. Absorptive capacity:a new perspective on learning and innovation.Administrative Science Quarterly35: 128–152.

Conner KR, Prahalad CK. 1996. A resource-basedtheory of the firm: knowledge versus opportunism.Organization Science7(5): 477–501.

Daft RL, Lengel RH. 1986. Organizational informationrequirements, media richness and structural design.Management Science32: 554–571.

Dierickx I, Cool K. 1989. Asset stock accumulation andsustainability of competitive advantage.ManagementScience35: 1504–1511.

Dunning JH. 1980. Toward an eclectic theory of inter-national production.Journal of International Busi-ness Studies, Spring–Summer: 9–31.

Egelhoff W. 1988. Organizing the MultinationalEnterprise: An Information Processing Perspective.Ballinger: Cambridge, MA.

Galbraith JR. 1977.Organization Design. Addison-Wesley: Reading MA.

Ganesh J, Kumar V. 1996. Capturing the cross-nationallearning effect: an analysis of an industrial tech-nology diffusion.Journal of the Academy of Market-ing Science, 24(Fall): 328–337.

Ghoshal S. 1987. Global strategy: an organizing frame-work. Strategic Management Journal8(5): 424–440.

Ghoshal S, Korine H, Szulanski G. 1994. Interunitcommunication in multinational corporations.Man-agement Science40(1): 111–122.

Grant RM. 1996. Prospering in dynamically-competitiveenvironments: organizational capability as knowl-edge integration.Organization Science7(4): 375–387.

Gresov C. 1989. Exploring fit and misfit in multiplecontingencies. Administrative Science Quarterly34(3): 431–453.

Gupta AK, Govindarajan V. 1993. Coalignmentbetween knowledge flow patterns and the strategicsystems and processes within MNCs. In


Implementing Strategic Processes: Change, Learningand Cooperation, Lorange P, Chakravarthy B, RoosJ, Van de Ven A (eds). Basil Blackwell: London;329–346.

Gupta AK, Govindarajan V. 2000. Knowledge flowswithin multinational corporations.Strategic Manage-ment Journal21(4): 473–496.

Hamel G, Prahalad CK. 1985. Do you really have aglobal strategy?Harvard Business Review63(4):139–148.

Jain S. 1989. Standardization of international marketingstrategy.Journal of Marketing53(1): 70–79.

Kalish S, Mahajan V, Muller E. 1995. Waterfall andsprinkler new product strategies in competitiveglobal markets.Journal of International Marketing2(1): 29–52.

Kogut B, Zander U. 1993. Knowledge of the firm andthe evolutionary theory of the MNC.Journal ofInternational Business Studies24(4): 625–645.

Krackhardt D. 1992. The strength of strong ties. InNetworks and Organizations, Nohria N, Eccles B(eds). Harvard Business School Press: Boston, MA;216–239.

Krauss RM, Fussell SR. 1991. Constructing sharedcommunicative environments. InPerspectives onSocially Shared Cognition, Resnick LB, Levine JM,Teasley SD (eds). American Sociological Associ-ation: Washington DC; 172–200.

Leonard-Barton D. 1995.Well-Springs of Knowledge:Building and Sustaining the Sources of Innovation.Harvard Business School Press: Boston, MA.

Levitt T. 1983. The globalization of markets.HarvardBusiness Review61(3): 92–102.

Madhavan R, Grover R. 1998. From embedded knowl-edge to embodied knowledge: new product develop-ment as knowledge management.Journal of Market-ing 62(4): 1–12.

Nobel R, Birkinshaw J. 1998. Innovation in multi-national corporations: control and communicationpatterns in international R&D operations.StrategicManagement Journal19(5): 479–496.

Nobeoka K, Cusumano MA. 1997. Multi-project strat-egy and sales growth: the benefits of rapid designtransfer in new product development.Strategic Man-agement Journal18(3): 169–186.

Nohria N, Ghoshal S. 1994. Differentiated fit andshared values: alternatives for managing head-quarters–subsidiary relations.Strategic ManagementJournal 15(6): 491–502.

Nonaka I. 1994. A dynamic theory of organizationalknowledge creation.Organization Science5(1):14–37.

Nonaka I, Takeuchi H. 1995.The Knowledge CreatingCompany. Oxford University Press: New York.

Ohmae K. 1985.Triad Power: The Coming Shape ofGlobal Competition. Free Press: New York.

Ohmae K. 1990.The Borderless World: Power andStrategy in the Interlinked Economy. Harper Peren-nial: New York.

Porter ME. 1980.Competitive Strategy: Techniques forAnalyzing Industries and Competitors. Free Press:New York.

Porter ME. 1986. Competition in global industries: a


conceptual framework. InCompetition in GlobalIndustries, Porter ME (ed). Harvard Business SchoolPress: Boston, MA; 16–60.

Polanyi M. 1966.The Tacit Dimension. Routledge &Kegan Paul: London.

Prahalad CK, Doz Y. 1987.The Multinational Mission:Balancing Local Demands and Global Vision. FreePress: New York.

Prahalad CK, Hamel G. 1990. The core competence ofthe corporation.Harvard Business Review68(3):79–91.

Quinn JB. 1992.Intelligent Enterprise: A Knowledgeand Service Based Paradigm for Industry. FreePress: New York.

Samiee S, Roth K. 1992. The influence of globalmarketing standardization on performance.Journalof Marketing 56(2): 1–17.

Spender JC. 1996. Making knowledge the basis of adynamic theory of the firm.Strategic ManagementJournal, Winter Special Issue17: 45–62.

Stewart TA. 1997.Intellectual Capital: The New Wealthof Organizations. Currency Doubleday: New York.

Subramaniam M, Rosenthal SR, Hatten KJ. 1998. Glo-bal new product development: preliminary findingsand research propositions.Journal of ManagementStudies35(6): 773–796.

Takeuchi H, Porter ME. 1986. Three roles of marketingin global strategy. InCompetition in Global Indus-tries, Porter ME (ed.). Harvard Business SchoolPress: Boston, MA; 111–146.


Teece DJ. 1977. Technology transfer by multinationalfirms: the resource cost of transferring technologicalknow-how. The Economic Journal, 87(346): 242–261.

Tushman ML, Nadler DA. 1978. Information processingas an integrating concept in organization design.Academy of Management Review, 3(3): 613–624.

Van de Ven AH, Delbecq AL. 1974. A task contingentmodel of work-unit structure.Administrative ScienceQuarterly 19(2): 183–197.

Venkatraman N. 1989. The concept of fit in strategyresearch: toward verbal and statistical correspon-dence. Academy of Management Review14(3):423–444.

Venkatraman N, Grant JH. 1986. Construct measure-ment in organizational strategy research: a critiqueand proposal. Academy of Management Review11(1): 71–87.

Venkatraman N, Ramanujam V. 1987. Measurement ofbusiness economic performance: an examination ofmethod convergence.Journal of Management13(1):109–122.

von Hippel E. 1994. Sticky information and the locusof problem solving. Management Science40(4):429–439.

Yip GS. 1995. Total Global Strategy. Prentice-Hall:Englewood Cliffs, NJ.

Zander U, Kogut B. 1995. Knowledge and the speedof transfer and imitation of organizational abilities:an empirical test.Organization Science6(1): 76–92.


APPENDIX A: OUR SAMPLING APPROACH

Note: Appendix B provides the list of MNCs and product categories surveyed. The list presents abroad description of the transnational product categories surveyed. We however got responses fromprojects that were associated with a finer distinction among these product categories. For example,among refrigerators there were different responses for refrigerators designed for tropical climates,and large size refrigerators. Similarly among baby foods, our responses reflected products that differedin terms of price points or the geographic regions they were targeted for. We took care to ensurethat each of our responses reflected an independent transnational new product development projectand a distinct associated transnational product category.



APPENDIX B: COMPANIES AND PRODUCT CATEGORIES REPRESENTED INTHE SAMPLE

Sr.No. Company Product Category

1. 3M Sponges, Cosmetics2. Abbot Laboratories Baby food3. Alberto Culver Shampoo4. Bausch & Lomb Contact lenses5. Becton Dickinson Digital thermometers, Insulin injectors6. Black & Decker Cordless drills7. British-American Tobacco Cigarettes8. Brown Group Children’s footwear9. Caterpillar Company Hydraulic excavators10. Cummins Diesel engines11. Daewoo Dishwashers, Hydraulic excavators, Forklift

Trucks, Machine tools12. Duracell Batteries13. Eastman Kodak Films, Digital cameras14. Fuji Xerox Copiers15. GE Appliances Refrigerators16. Gillette Writing instruments, Razors17. Hallmark Cards Greeting cards18. Helene Curtis Cosmetics19. Hershey Foods Chocolate confectioneries20. Hewlett Packard Medical imaging devices, Printers, Digital cameras21. Honeywell Temperature control systems22. Jafra Cosmetics Cosmetics23. Johnson & Johnson Toothbrushes24. Liptons Food products25. L’Oreal Cosmetics26. Lucent Technologies PBX systems27. Maytag Dishwashers, Refrigerators28. Mead Johnson Baby foods29. Motorola Cellular telephones30. Nortel PBX Systems31. Ocean Spray Cranberries Food products32. Oral-B Toothbrushes33. Osram Sylvania Fluorescent lamps34. Parker Pen Company Writing instruments35. Philips Digital cameras36. Polaroid Digital cameras, Films37. Reckitt & Colman Air freshners38. Reebok Athletic shoes39. Siemens Induction motors, Generators40. Smith Kline Beecham Toothpaste41. Tambrands Tampons42. Tennant Machine tools43. Trane Company Room air conditioners44. Whirlpool Refrigerators, Washing machines45. Wyeth-Ayerst International Baby food



APPENDIX C: CONSTRUCTS AND THEIR MEASURES

1. Transnational New Product Development Capability (Cronbach Alpha: 0.84)

With respect to your key competitors, please rate how yourproduct category currently fares, onthe following dimensions:

I Frequency of new global product introductionsI Being first in the market with new product introductionsI Ability to introduce new versions simultaneously in several marketsI Ability to respond to unique requirements of different countriesI Ability to price competitivelyI Ability to penetrate new overseas markets

Measured on a 1 to 7 scale: Much worse than competition — About the same — Much betterthan Competition

2. Extent of tacitness in overseas information (Cronbach Alpha: 0.81)

Please indicate the characteristics of the information acquiredfrom overseas locations(aboutdifferences among overseas markets, manufacturability in overseas plants etc.) The informationyour project acquired:

Was simple Was complexWas easy to comprehensively document in Was difficult to comprehensively document inmanuals or reports manuals or reportsWas easy to comprehensively understand from Was difficult to comprehensively understandwritten documents from written documentsWas easy to precisely communicate through Was difficult to precisely communicate throughwritten documents written documentsWas obvious to all competitors Had subtle nuances known only to a few

competitorsWas easy to identify without personal experience Was difficult to identify without personalin the overseas locations experience in the overseas locations

EMeasured on a 1–7 scale

Note: This question was preceded by two other questions that provided the survey respondentsthe background for this construct. These questions were:

1. How significant for your project was the following information to develop the new global product?(on a 1–7 scale: Was of no significance — Was of great significance)

2. How new or novel was the following information used by your project for developing the newglobal product? (on a 1–7 scale: Not at all new — Substantially new)

3. For each of these questions, the indicators were:I Information on overseas market preferencesI Information on the feasibility of manufacturing various design alternatives in overseas plantsI Information on differences among overseas markets



3. Cross-national teams

Dummy variable — measured as 1 if there were overseas members in the team, 0 otherwise.

4. Teams with domestic members having prior overseas experience

Dummy variable — measured as 1 if the team had domestic members who were posted inoverseas offices for at least a year, 0 otherwise.

5. Frequency of communication (Cronbach Alpha: 0.86)

How frequently did the following activities with regard to your project occur?I Telephone conversations with overseas managersI Faxes to and from overseas managersI Email to and from overseas managers

Measured on a 1–7 scale: Very rarely (once a year) — Moderately (monthly) — Very frequently (daily)

6 & 7. Brand image in overseas markets and overseas market share (Control Variables)

With respect to your key competitors, please rate how your product category currently fares, onthe following dimensions:

I Brand image in key overseas marketsI Market share in key overseas markets

Measured on a 1–7 scale: Much worse than competition — About the same — Much better thancompetition

8. Worldwide Marketing Infrastructure (Control Variable)

I Marketing/Sales offices that supported your product categoryMeasured with respect to ‘presence in important country markets’ of this indicator, on a 1–5scale: In none 0% — In very few < 10% — In some (10–50%) — In most (50–90%) — Inalmost all (90–100%)

9. Market Concentration (Control Variable)

The market served by the global product:Had many competitors Had few competitors

Measured on a 1–7 scale.

10. Number of countries targeted (Control Variable)

Requirements of how many countries were important considerations for the new global product’sdesign?

11. Permeability of managers across borders (Control Variable) (Cronbach Alpha: 0.88)

How prevalent were the following practices in your Division (at the time of the project)?I Regular visits to HQ by overseas managers (middle management)I Regular visits to overseas subsidiaries by HQ managers (middle management)I Informal interactions between HQ and overseas subsidiary middle managers

Measured on a 1–7 scale: Very rare — Moderately prevalent — Extremely prevalent.


Determinants of transnational new product development capability: testing the influence of...

Documents

Transcript of Determinants of transnational new product development capability: testing the influence of...