Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

We argue that the debate regarding the performance implications of demographic diversity can be usefully reframed in terms of the network variables that reflect distinct forms of social capital. Scholars who are pessimistic about the performance of diverse teams base their view on the hypothesis that decreased network density—the average strength of the relationship among team members—lowers a team's capacity for coordination. The optimistic view is founded on the hypothesis that teams that are characterized by high network heterogeneity, whereby relationships on the team cut across salient demographic boundaries, enjoy an enhanced learning capability. We test each of these hypotheses directly and thereby avoid the problematic assumption that they contradict one another. Our analysis of data on the social networks, organizational tenure, and productivity of 224 corporate R…D teams indicates that both network variables help account for team productivity. These findings support a recasting of the diversity-performance debate in terms of the network processes that are more proximate to outcomes of interest.

This article aims to reestablish the long-standing conjecture that conformity is high at the middle and low at either end of a status order. On a theoretical level, the article clarifies the basis for expecting such an inverted U-shaped curve, taking care to specify key scope conditions on the social-psychological orientations of the actors, the characteristics of the status structure, and the nature of the relevant actions. It also validates the conjecture in two settings that both meet such conditions and allow for the elimination of confounding effects: the Silicon Valley legal services market and the market for investment advice. These results inform our understanding of how an actor's status interacts with her role incumbency to produce differential conformity in settings that meet the specified scope conditions.

seminar participants at the Columbia and University of Chicago business schools for guidance and feedback. We take full responsibility for remaining errors. Robust Identities or Non-Entities?Typecasting in the Feature Film Labor Market AbstractWe provide a framework for reconciling two seemingly incompatible claims regarding identity in social and economic arenas: (a) that complex, multivalent identities are advantageous because they afford greater flexibility; and (b) that simple, generic identities are advantageous because they facilitate interpretation by key audiences. Following Faulkner (1983), we argue that these claims do not conflict with one another but that they apply to different contexts. A generic identity is helpful in gaining the recognition necessary for sustained participation in a social arena. However, as one becomes better established, the limitations entailed by a simple, "typecast" identity increasingly rival the benefits. We test these hypotheses in an analysis of the labor market for actors in the feature film industry. Interviews with key informants and analysis of comprehensive data from the Internet Movie Database support the proposed theoretical framework. In addition, the evidence supports the salience of the hypothesized typecasting processes even in the presence of related processes based on underlying skill differences and social networks. Our results have important implications for research on identity formation in various social arenas, categorical boundaries in external labor markets, and more generally, the interplay between actor and position inherent in market dynamics. Robust Identities or Non-Entities?

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. ences at the University of Chicago, Stanford, Harvard, Northwestern, and MIT for their feedback and advice. In addition, I greatly appreciate the assistance provided by Don Palmer, Linda Johanson, and the ASQ reviewers in shaping the current version of the paper. Katherine Schipper and J. Douglas Hanna provided assistance with the Zacks data. All mistakes are my own, of course. The issue of corporate control is examined through an analysis of the de-diversification activity of publicly held American firms from 1985 to 1994. Prominent accounts of such behavior depict newly powerful shareholders as having demanded a dismantling of the inefficient, highly diversified corporate strategies that arose in the late 1950s and the 1960s. This paper highlights an additional factor that spurred such divestiture: the need to present a coherent product identity in the stock market. It is argued that because they straddle the industry categories that investors-and securities analysts, who specialize by industry-use to compare like assets, diversified firms hinder efforts at valuing their shares. As a result, managers of such firms face pressure from analysts to dediversify so that their stock is more easily understood. Results indicate that, in addition to such factors as weak economic performance, de-diversification is more likely when a firm's stock price is low and there is a significant mismatch between its corporate strategy and the identity attributed to the firm by analysts. I ... I realized that analysts are like the rest of us. Give them something easy to understand, and they will go with it. [Before the spin-off,] we had made it tough for them to figure us out." -A.H. Stromberg, chief executive officer of URS Corporation (Brown, 1983: 72). The issue of corporate control has traditionally been pursued by ascertaining who holds power over the public corporation and what their interests are. Accordingly, scholars have debated the degree of control exerted by managers (Berle and Means, 1933), banks (e.g., Kotz, 1978; Mintz and Schwartz, 1985), and founding families (Zeitlin, 1974) as well as the implications of a firm's position in the board interlock structure (e.g., Burt, 1983; Palmer et al., 1995) and its relationship to the capitalist class (e.g., Domhoff, 1967; Useem, 1984). Each of these stakeholders is presumed to direct the firm to act according to its interests, which may not be consistent with those of its shareholders. That shareholders themselves have not been viewed as agents of corporate control reflects the original framing of the issue by Berle and Means (1933), who worried that the diffusion of ownership would empower managers to divert the firm from the pursuit of profit...