- The formation, governance and performance of inter-organizational relationships, including alliances, buyer-supplier relationships, etc.
- Knowledge diffusion across firms, industries and regions
- The impact of formal and informal institutions on the topics above, with a special interest in the economy and institutions of Japan.
In my research, I encountered methodological limitations of extant research. A smaller stream of papers contributes to the field by highlighting common methodological errors, their consequences for theory testing and errors in inferences, and potential solutions.
The formation, governance and performance of inter-organizational relationships
I've sought to contribute to recent work integrating formerly isolated theoretical approaches to understanding inter-firm relationships. For example, my dissertation built and tested a model integrating transaction cost economics, the literature on inter-firm relationships, and the firm capabilities literature to understand how buyers select suppliers for technically innovative components.
While adherents of each literature have criticized the others for what they omit, this paper moves beyond mutual recrimination and provides an integrated theoretical model by incorporating the key concerns of each literature, recognizing that the relative importance of each will vary according to the situation under consideration. Implicitly or explicitly, this model is at the heart of almost all of my research in this stream as I examine the contingencies that make the concerns of the three literatures more or less relevant.
In "How much you know versus how well I know you" (SMJ 2005), the level of technical uncertainty was the key contingency. In "Do modular products lead to modular organizations" (SMJ 2006) and "Modularity and the impact of buyer-supplier relationships on the survival of suppliers" (Mgt Sci 2007), the contingency of interest is modularity, that is, the degree to which one component of a product can be altered without requiring compensating changes in other components. In "Do modular products...," I provided a stronger empirical test for the widely held, but largely tacit assumption that firms with modular products can and will adopt modular organization forms. Using a distinctive empirical setting, I tested this assumption more cleanly than has been done previously, finding that the assumed relationship was overly simplistic and that organizational modularity is a more multiplex phenomenon than previously recognized. The results contributed to an emerging stream of research focusing on the previously under-appreciated costs of designing and maintaining a modular organization.
In "Modularity and the impact of buyer-suppler relationships..," with Will Mitchell (Duke) and Anand Swaminathan (UC Davis), I provided one of the first studies how modularity affects suppliers (prior work focused on buyers). Constructing a unique dataset on the U.S. auto industry from 1918 to 1942, we studied modularity as a contingency affecting how a supplier's relationships with its buyers influence its survival. The results informed the broader literature on inter-organizational relationships by refining the concept of autonomy to include current autonomy, generated by existing relationships, and potential autonomy, generated by relationships that could be formed, but do not currently exist.
With Thomas Mellewigt (Freie Universitšt Berlin) and others, I am contributing to the nascent literature focusing on the underlying governance mechanisms used in alliances. For example, recent work has shown that formal/contractual and relational governance mechanisms (e.g., contracts and exchange of employees, respectively) are not mutually exclusive, but says little about when one approach is superior to the other. We showed that the optimal configuration of governance mechanisms is contingent on the type of assets (property or knowledge) involved in an alliance ("Choice and performance of governance mechanisms: Matching alliance governance to asset type" SMJ 2009). By introducing the critical contingency of alliance content, our findings contribute to transaction cost economics, the literature on relational governance, and recent work studying their interaction.
Plural sourcing, sometimes called “make-and-buy” is an important variation on inter-firm relationships. Recent papers exploring such plural sourcing modes have yielded rich, but inconsistent, theoretical and empirical insights. Thomas, Anna Krzeminska and I (“Reconceptualizing plural sourcing” forthcoming, SMJ) suggest that resolving these inconsistencies and setting the foundation for future work requires reconceptualizing two aspects of plural sourcing: what and how. “What” refers to a surprising lack of clarity of what is meant by “same” inputs. We re-conceptualize “same” as a spectrum of degrees of similarity and propose how similarity might be measured. “How” refers to the governance modes combined in concurrent sourcing. Extending the literature’s predominant focus on make/buy, we re-conceptualize plural sourcing as a set of combined governance modes—make/buy, make/ally and buy/ally—distinguished from single modes of governance by certain shared characteristics, but differing from each other in their capabilities.
Within the alliance literature, it is often assumed, often implicitly, the firms with power will apply that power to structure alliance terms to their benefit. However, there is considerable empirical evidence to suggest that this is not always the case. Rather, firms may accommodate the concerns of their partners and accept exchange conditions more relevant to their partner’s concerns than their own. In the first of several papers examining this phenomenon of accommodation, Jongkuk Lee, Bill Qualls and I (“Alliance experience and accommodation in the choice of alliance governance structure” working paper) examine accommodation in R&D alliances within the bio-pharma industry. We find that more experienced firms are more willing to accommodate their partner’s concerns in the choice of governance structure (equity versus non-equity).
The literature on inter-firm relationships has focused at length on the impact of repeated ties between buyers and suppliers. However, many technological products are made up of a number of subcomponents whose completion requires coordinated efforts from different suppliers. Therefore, selecting a team of suppliers that can coordinate effectively is critical to success, a challenge that has been ignored in the literature. Marko Madunic, Joe Mahoney and I (“Choosing teams over star players: Strategic selection of supplier teams in customized projects” working paper) posit that prior experience amongst the suppliers of those subcomponents can be as, or even more, important as prior experience between the buyer and the individual suppliers. Using data on the selection of teams to construct FPSO (i.e., floating, production, storage and offloading) vessels, we examine how buyers select of team of subcomponent suppliers, balancing each supplier’s individual characteristics and the experience the suppliers have in working with each other.
In two new papers, I bring into consideration the extensive literature on the effect of pre-entry experience (de alio versus de novo firms). Lihong Qian, Rajshree Agarwal and I (“Configuration of value chain activities: The effect of pre-entry capabilities, transactions hazard and industry evolution on the decision to internalize” forthcoming, Organization Science) are examining firm’s choices of which value chain activities to internalize when entering a new industry. Using data on U.S. bioethanol producers, we find that pre-entry experience in a given activity makes it more likely that a firm will also internalize that activity in the focal industry. However, pre-entry experience also allows firms to better manage external relationships, meaning their decision is less affected by the transaction hazards associated with that value chain activity.
Paolien Chen, Rajshree and I (“Incumbents’ relationships with technology suppliers and ability in managing the relationships in facing a disruptive technology: De alio versus de novo” working paper) are examining the impact of an incumbent’s relationships with existing suppliers on an incumbent’s likelihood of moving into the new market created by a disruptive technology and compares the impacts between de alio and de novo incumbents. Using the decision to adopt digital switching technology by U.S. wireless telecommunications service operators, we find that pre-entry experience (de alio firms) allows firms to benefit more from the advantages of having a large network. They gain more information from the network, while being less affected by network inertia. However, the network inertia generated by long-duration ties to suppliers disadvantaged de alio and de novo firms equally.
Knowledge diffusion across firms, industries and regions
This research stream with Rajshree Agarwal examines the transfer of knowledge between organizations. “Death hurts, but it isn’t fatal” (AMJ 2007) began with the observation that the innovative knowledge created by firms that ultimately exit their industry represents a source of technology upon which surviving firms can build. However, no empirical work had examined whether the knowledge dies with the innovating firm or subsequently diffuses and the existing literature on inter-firm knowledge transfer leads to contradictory predictions. Building on work on private and public knowledge and using the disk drive industry as our empirical setting, we found that death hurts knowledge diffusion: firm failure reduces the subsequent use of a firm’s technology by others, even controlling for the quality of the technology. However, it isn’t fatal to knowledge diffusion—there is clear evidence of significant post exit diffusion of knowledge. Theoretically, the paper showed that the ability to use a firm as a “template” plays a critical role in successfully replicating its knowledge, particularly when the private aspect of the knowledge is significant. Beyond the obvious managerial implications, our findings give guidance regarding what specific innovations are most amenable to incorporation. There are also policy implications for the evaluation and organization of public investment in technology development. Because the phenomenon is broadly relevant, but under-explored, we believe we can make a real impact in both the short- and long-run.
Publications, like patents, represent codified knowledge that can diffuse across disciplinary “organizations.” High status publication outlets are particularly suited for enabling knowledge transfer. In a paper examining the position of the Academy of Management Journal in management’s scholarly conversation ("A faustian bargain? The growth of management and its relationship with related disciplines" AMJ 2007), Rajshree and I studied knowledge transfer within and across disciplinary contexts. One of the most striking findings is the importance of the dual role that leading journals can play in introducing concepts from related disciplines ( e.g., sociology) to management research, and feeding back into those disciplines.
The impact of formal and informal institutions on business strategy
The structure, governance and performance of inter-organizational relationships are contingent on the institutional environment that surrounds them. My research partnership with Thomas Ginsburg (Law, University of Chicago) examines the impact of institutional change on inter-firm relationships, focusing primarily on Japan. Beyond Japan’s economic importance, recent changes in its institutions and our access to previously unavailable data will allow us to draw from and contribute to both the legal and strategy literature. Our first paper ("The unreluctant litigant? An empirical analysis of Japan’s turn to litigation” JLS 2006) showed that the Japanese are turning from informal to formal dispute resolution mechanisms (e.g., the courts) as formal mechanisms improve, despite culturalist assertions of a litigation-averse Japanese society. Our next three planned papers study (a) the impact of improvements in Japan’s formal institutions on firm’s supplier relationships, (b) changes in firms’ relationships with government agencies, and (c) foreign firms in Japan as exemplars of favoring formal over informal institutions. Our research generalizes beyond Japan by allowing us to compare inter-organizational relationships in institutional regimes of differing. This research stream complements earlier work in which I demonstrated the variety underlying the previously monolithic model of ‘Japanese-style’ buyer-supply relationships, which was based almost entirely on the automobile industry ("Same rules, different games: Variation in the outcomes of ‘Japanese-style’ supply relationships" AIM 2004; "Do 'Japanese-style' supplier relationships exist? How industry characteristics shape the impact of national institutions" working paper).
Methodological issues in management research
This research stream comes from my efforts to address methodological issues I encountered during my dissertation research. Traditional means of comparing logit or probit coefficients across groups, e.g., foreign and U.S. firms, are valid only under specific conditions unlikely to exist in practice. Recent work emphasized the theoretical potential for this problem and proposed a more robust test. In "Confounded coefficients: Extending recent advances in the accurate comparison of logit and probit coefficients across groups" (working paper), I used Monte Carlo simulations to determine the empirical significance of this theoretical problem, tested the power of the suggested test, and offered two alternative approaches. My findings indicate that a change in practice is necessary to ensure valid statistical inference. Additionally, I wrote a package for the Stata statistical program, complogit, that automates the tests discussed in the article.
This research raised my interest in the state of logit/probit modeling in the strategic management research. The resulting study "The use of logit and probit models in strategic management research: Critical issues" (SMJ 2007) reviewed the methodological literature and leading strategy journals, revealing four areas in the use and interpretation of logit/probit models where current practice fall short of ideal. For each issue, the paper provides a background, a review of current practice, and recommendations for best practice. I discovered that insights on logit/probit models are scattered across multiple literature (e.g., political science, sociology, econometrics, and transportation economics) and no one field has addressed the full range of issues. One of the paper’s primary contributions is to integrate information and best practices from diverse fields in a framework unified by consistent terminology and notation.
I have papers in early stages that take a similar approach to two other methodological issues I've encountered in recent research.