Knowledge transfer and partial equity ownership

• Author: Arghya Ghosh,Hodaka Morita
• Date: 01 December 2017
• Published date: 01 December 2017
• DOI: http://doi.org/10.1111/1756-2171.12213

RAND Journal of Economics

Vol.48, No. 4, Winter 2017

pp. 1044–1067

Knowledge transfer and partial equity

ownership

Arghya Ghosh∗

and

Hodaka Morita∗

An alliance often involves one firm acquiring an equity stake in its alliance partner. We explore

oligopoly models that capture the link between knowledge transfer and partial equity ownership

(PEO), where alliance partners can choose the level of PEO. PEO can increase the alliance

partners’ profitability by inducing knowledge transfer, but the PEO itself reduces their joint

profit because it induces other firms to take more aggressive actions. This trade-off endogenously

determines the level of PEO, which can benefit consumers and/or improve welfare. Given the

growing antitrust interest in PEO, we explore the antitrust implications of our analysis.

1. Introduction

A strategic alliance exists when two or more independent organizations cooperate in the

development, manufacturing, or sale of products or services (Barney, 2002). In recent years, the

incidence and importance of interfirm collaborations have substantially increased (Caloghirou,

Ioannides, and Vonortas, 2003). It has been widely recognized in the literature that one of the

most fundamental objectives of strategic alliances is the transfer of knowledge between partner

firms (see Hamel, 1991; Mowery,Oxley, and Silverman, 1996; Gomes-Casseres, Hagedoorn, and

Jaffe, 2006; Oxley and Wada, 2009).1

Modern economies are becoming increasingly knowledge intensive. Drucker (1993), for

example, has argued that in the new economy, knowledge is not just another resource alongside

∗University of New South Wales; a.ghosh@unsw.edu.au, h.morita@unsw.edu.au.

Wewould like to thank the Editor, Mark Armstrong, and two anonymous referees for valuable and insightful comments.

We thank Jay Pil Choi, Cl´

emence Christian, Curtis Eaton, John Panzar, Mike Waldman, and seminar participants at

Cheung Kong Graduate School of Business, Fair Trade Commission Japan, Hitotsubashi University, Kyoto University,

La Trobe University, Massey University, Michigan State University, University of Arkansas, University of Calgary,

University of New South Wales, University of Tokyo, Vanderbilt University, and University of Western Australia for

helpful comments and discussions, and Tomohiro Ara, Shuai Niu, Xuan Nguyen, Bora Buth, and Cheng-Tao Tang for

able research assistance. Financial support from the Australian Research Council is gratefully acknowledged.

1Gomes-Casseres, Hagedoorn, and Jaffe (2006), for example, hypothesized that knowledgeflows between alliance

partners would be greater than flows betweenpairs of nonallied fir ms, and found empirical results that are consistent with

this hypothesis.

1044 C2017, The RAND Corporation.

GHOSH AND MORITA / 1045

the traditional factors of production—labor, capital, and land—but the only meaningful resource

today (see Quinn, 1992, and Toffler, 1990, for similar arguments). Knowledge is often classified

into tacit and explicit knowledge.2Explicit (or codified) knowledge can be transmitted in formal,

systematic language, whereas tacit knowledge is nonverbalizable, intuitive, and unarticulated

(Polanyi, 1958, 1966). Because tacit knowledge is difficultto convey, its transfer requires greater

effort (Reagans and McEvily,2003). Tacit knowledge can be transferred through up-close obser-

vation, demonstration, hands-on experience, a sharing of feelings and emotions, and face-to-face

contact (Hamel, 1991; Nonaka, 1994; Nonaka and Takeuchi, 1995; Cavusgil, Calantone, and

Zhao, 2003).

Explicit knowledge can be transferred through licensing and contracting, because it is verifi-

able. However, because tacit knowledgecannot be codified and hence cannot be verified, licensing

and contracting play, at best, a limited role in the transfer of tacit knowledge (Mowery, 1983;

Pisano, 1990). Here, however, equity ownership plays a critical role in facilitating the transfer

of tacit knowledge (Mowery, Oxley, and Silverman, 1996). Using patent citations as a proxy for

knowledge flows, Mowery, Oxley, and Silverman (1996) and Gomes-Casseres, Hagedoorn, and

Jaffe (2006) empirically explored the effectsof equity ownership between alliance partners on the

extent of knowledge flow. Empirical results of both studies supported the hypothesis that equity

ownership enhances the extent of knowledge flow between alliance partners.3As a real-world

example of the connection between equity ownership and knowledgetransfer, Okasan Securities

announced in March 2015 that they were going to form an equity strategic alliance with Marukuni

Securities, in which Okasan, after acquiring around 20% stake in Marukuni, would help Marukuni

strengthen its sales capabilities and back office services.4

Partial equity ownership (PEO) induces knowledge transfer between alliance partners. This

article explores an oligopoly model that captures this important link between PEO and knowledge

transfer. We consider an industry consisting of several (three or more) firms (firms 1, 2, . . . ) that

produce a homogeneous product and compete in quantities. Compared to firm 2, firm 1 has a cost

advantage due to its superior knowledge that firm 2 does not have. Constant marginal costs of

firms 1 and 2 are c−xand c, respectively. If firm 1 transfers its superior knowledge to firm 2,

the knowledge transfer increases firm 2’s competitive position by decreasing firm 2’s cost by x,

and this in turn reduces firm 1’s profitability. Firm 1 therefore would not transfer its knowledge

without monetary return. We assume that the knowledge is tacit and not verifiable, and hence

contracts or licensing agreements cannot guarantee monetary return for firm 1.5

In this context, PEO can play an important role in facilitating the transfer of knowledge.

Firms 1 and 2 havean option of for ming an equity strategic alliance in whichfir m 1 ownsa fraction

θ∈[0,1] of firm 2’s shares. At the beginning of the model, firms 1 and 2 jointly determine the

level of θto maximize their joint profit in the subsequent equilibrium, and the monetary terms of

the equity transfer through bargaining. Then, once θis chosen, firm 1 determines, based on its

2See Polanyi (1958, 1966) for a seminal work on tacitness of knowledge. See also Nonaka (1994), Nonaka and

Takeuchi (1995), Grant (1996),Inkpen and Dinur (1998),and Adler (2001), among others.

3Both studies used the Cooperate Agreements and Technology Indicators (CATI) database developed by the

Maastricht Economic Research Institute in Technology (MERIT) to identify alliances of firms. Mowery, Oxley, and

Silverman (1996) focused on bilateral alliances that involved at least one US firm and which were established during

1985 and 1986, and the patent data were drawn from the Micropatent database, which contains all information recorded

on the front page of every patent granted in the United States since 1975. Gomes-Casseres, Hagedoorn, and Jaffe (2006)

matched the firms in the CATI database to the NBER Patent Citations Data File. They used only alliance and patent

data from information technology sectors, and analyzed citations on an annual basis from 1975 to 1999. See also Ono,

Nakazato, Davis, and Alley (2004) for empirical evidence from the Japanese automobile industry.They also put forth a

theoretical framework incorporating a partial ownership arrangement and technology transfer,but they did not derive the

equilibrium of the model by solving it.

4www.okasan.jp/english/news/osg/news.php?cat_No=01&year_No=2015&id_No=150330_1. See also the Web

Appendix (sites.google.com/site/hodakamorita/home), where we list three other examples of the connection between

equity ownership and knowledge transfer.

5Our model also applies to explicit knowledge that is hard to transfer through contracts or licensing agreements.

C

The RAND Corporation 2017.