Knowledge that Transforms

The keiretsu structuring of assembler-supplier relations historically enabled Japanese auto assemblers to remain lean and flexible while enjoying a level of control over supply akin to that of vertical integration. Yet currently there is much talk of breakdowns in keiretsu networks. This paper examines some recent developments in Japanese parts-supply keiretsu. We argue that keiretsu relationships are drifting from “hybrid” or “network” (i.e., keiretsu) governance modes toward the extremes of arms-length contracting and top-down administration. These changes are best understood through a combination of transaction cost and learning perspectives on alliance. Consistent with transaction-cost economics, the shift in purchase-supply relationships can be traced to changes in the nature of parts transactions and keiretsu-governance structures. A learning perspective on alliance complements and extends transaction-cost theory, providing additional explanation of the sources of change and the specific governance choices being made. Our first two cases document a drift in Toyota's keiretsu supply network toward a hierarchical form in the management of parts-supply transactions. Toyota has effectively internalized its transactions with Daihatsu by taking a controlling interest. Toyota's strategy toward long-term partner Denso, on the other hand, was very different. Toyota built, from the ground up, an in-house capability in electronic components, thus scaling down its dependence on Denso. A third case considers a general trend in the Japanese auto industry toward greater standardization of parts. With the routinization of quality, reliability, and speed in supply management, the need for keiretsu-style governance has declined. The withering of keiretsu obligations is also traceable to globalization and the continuing weakness of the Japanese economy, which have prompted Japanese firms to question received business practice.

Several prior articles have emphasized the importance of alignment between business and information system (IS) strategies, and between business and IS structures. Seeking to advance our understanding of alignment, we examine the dynamics of changes in alignment through strategy/structure interactions in the business and IS domains. More specifically, we address the following question: In what ways does alignment evolve over time? Changes in the strategic IS management profile (which includes business strategy, IS strategy, business structure, and IS structure) over time are examined using a punctuated equilibrium model, involving long periods of relative stability, or evolutionary change, interrupted by short periods of quick and extensive, or revolutionary, change. Case studies of changes in business and IS strategies and structure over long time periods in three organizations suggest that the punctuated equilibrium model provides a valuable perspective for viewing these dynamics. The cases suggest that a pattern of alignment may continue over a long period, because either the level of alignment is high or the managers do not recognize the low alignment as a problem. Revolutions, involving changes in most or all dimensions of the strategic IS management profile, interrupt the evolutionary changes. However, organizations hesitate to make such revolutionary changes in strategic IS management profiles. Complete revolutions apparently require a combination of strong triggers. Finally, post-revolution adjustments to one dimension of the strategic IS management profile seem to follow revolutionary changes.

Through two exploratory studies, we develop and test an introductory framework of “organizational disidentification.” Our first study explores the concept of organizational disidentification through a qualitative investigation of cognitive relationships with the National Rifle Association (NRA). Findings suggest that organizational disidentification is a self-perception based on: (1) a cognitive separation between one's identity and the organization's identity, and (2) a negative relational categorization of oneself and the organization (e.g., categorizations such as “rivals” or “enemies”). Organizational disidentification appears to be motivated by individuals' desires to both affirm positive distinctiveness and avoid negative distinctiveness by distancing themselves from incongruent values and negative stereotypes attributed to an organization. Our findings also suggest that organizational disidentification can lead individuals to take action (either volunteer work or voicing their opinion) as a result of their perceived separation from the organization's identity. Results of our second study”a large-scale survey of public attitudes about the NRA”provide support for this framework.

In this essay we argue that organization theory's effort to make sense of postbureaucratic organizing is hampered by a dearth of detailed studies of work. We review the history of organization theory to show that, in the past, studies of work provided an empirical foundation for theories of bureaucracy, and explain how such research became marginalized or ignored. We then discuss methodological requirements for reintegrating work studies into organization theory and indicate what the conceptual payoffs of such integration might be. These payoffs include breaking new conceptual ground, resolving theoretical puzzles, envisioning organizing processes, and revitalizing old concepts.

Individuals and social systems are often portrayed as risk averse and resistant to change. Such propensities are characteristically attributed to individual, organizational, and cultural traits such as risk aversion, uncertainty-avoidance, discounting, and an unwillingness to change. This paper explores an alternative interpretation of such phenomena. We show how the reproduction of successful actions inherent in adaptive processes, such as learning and competitive selection and reproduction, results in a bias against alternatives that initially may appear to be worse than they actually are. In particular, learning and selection are biased against both risky and novel alternatives. Because the biases are products of the tendency to reproduce success that is inherent in the sequential sampling of adaptation, they are reduced whenever the reproduction of success is attenuated. In particular, when adaptation is slowed, made imprecise, or recalled less reliably, the propensity to engage in risky and new activities is increased. These protections against the error of rejecting potentially good alternatives on inadequate experiential evidence are costly, however. They increase the likelihood of persisting with alternatives that are poor in the long run as well as in the short run.

This paper reports the results of experiments which examine attributions of leadership quality. Subjects played an abstract coordination game which is like many organizational problems. Previous research showed that when larger groups play the game, they rarely coordinate on the Pareto-optimal (efficient) outcome, but small groups almost always coordinate on the efficient outcome. After two or three periods of playing the game, one subject who was randomly selected from among the participants to be the “leader” for the experiment was instructed to make a speech exhorting others to choose the efficient action. Based on previous studies, we predicted that small groups would succeed in achieving efficiency but that large groups would fail. Based on social psychological studies of the fundamental attribution error, we predicted that the subjects would underestimate the strength of the situational effect (group size) and attribute cause to personal traits of the leaders instead—leaders would be credited for the success of the small groups, and blamed for the failure of the large groups. This hypothesis proved true: Subjects attributed differences in outcomes between conditions to differences in the effectiveness of leaders. In a second experiment, subjects voted to replace the leaders more frequently in the large-group condition (at a small cost to themselves), showing that misattributions of leadership ability also affect actual behavior by subjects. Previous research has demonstrated a tendency to credit or blame leaders for unusual performance. The difference in our study is that subjects should be blaming a structural condition—the size of the group—but they blame the leaders instead. Thus, our experiment is the first to establish a mistaken illusion of leadership.

The real options approach is frequently advocated as an approach that offers a positive and radical reassessment of the value of risk and exploration. We examine a recent case where Merck used the real options approach to justify an investment in an R&D project. This case is used to highlight some of the problems associated with using real options. We note that the assumptions incorporated in most standard option valuation models can conflict with the conclusions reached by strategic analysis. As a result, users of real options models should understand the quantitative aspects of these models, and may often need to create a customized model for each situation. The difficulty of developing customized models may explain, in part, the limited use of the real options approach in strategic analysis.

This paper examines how a collaborative effort between the private and public sectors, called the Derivatives Policy Group (DPG), helped shape current regulation of financial innovation. In 1994 and 1995, this group of six large financial firms developed procedures for risk management, internal controls, and reporting for largely unregulated areas of finance, in cooperation with the United States Securities and Exchange Commission and Commodity Futures Trading Commission. The process succeeded despite strong competition among the firms themselves and incentives for both the public and private sectors to resort to adversarial lobbying and legal challenges. The Derivatives Policy Group was a path-setting event in the development of flexible regulation of financial innovation that is now the norm for related policy making. The case is important in and of itself--the financial markets are a major concern of national and international economic policy--but here we treat it as an instance of a larger class of problems. Organizational science constantly encounters settings that involve numerous participants who compete or have histories of conflicts; who are interdependent, and collectively would gain (and even individually gain long term) by cooperating rather than competing on an issue; who fall under different governance systems; and who try as a group to design rules and principles governing their behavior. Four factors appear repeatedly in the research on the success or failure of such arrangements. These are (1) the initial dispositions toward cooperation, (2) the extant issues and incentives, (3) leadership, and (4) the number and variety of organizations involved. This paper focuses on how these factors shaped the development and consequences of the Derivatives Policy Group, and the general implications of this process for interorganizational cooperation.

The literature on the management of R&D professionals strongly advocates managing R&D projects on a project-by-project basis. This literature suggests that projects should be managed differently depending upon project characteristics such as risk, ambiguity, and nonroutineness. While the primary emphasis of the R&D professional literature has been on project teams, the purpose of this study is to examine the impact of organization-wide controls on innovativeness at the firm level. In a sample of 57 pharmaceutical firms, this study investigates the influence of organizational controls on the research and development activities of R&D professionals. This study is one of a handful of studies that simultaneously explores the use of input, behavior, and output controls. Two categories of innovation are considered as dependent variables: incremental innovations in the form of drug enhancements and radical innovations in the form of new drugs. Contrary to existing theory and hypotheses developed in this study, the results show that input, behavior, and output control enhanced radical innovation, and input and output controls enhanced incremental innovation. These results challenge several important features of existing models of R&D management and diverge from common beliefs about R&D management at the project level. While it is commonly accepted that incremental and radical innovation should be managed differently, the results of this study suggest otherwise. In this instance, the management of R&D activities may be considered more similar than previously thought.