Managerial theories of the firm have a long and controversial history, beginning perhaps with Adolf A. Berle, Jr. and Gardiner C. Means's classic (1932) study. Such theories postulated that managers had considerable discretion running corporations and exercised it to pursue objectives other than maximizing shareholder market value.

Critics later denied that managers really have discretion, arguing that such behavior could not survive in equilibrium. Either management would be voted from office, the firm would be taken over, or the firm would go bankrupt in a competitive product market.

The more recent literature on information economics emasculated these criticisms. Managers are important to corporate decision-making because of their expertise and the information they acquire about the firm and its prospects. Yet the very information asymmetries that create a need for management also limit the discipline that the board of directors can impose on managers. At the same time, neither the shareholder-voting mechanism nor the takeover mechanism provides effective discipline. Small shareholders free ride on the efforts of others, since it does not pay for them to obtain the costly information needed to exercise their proxy vote intelligently. Moreover, if a takeover will enhance the productivity of the firm, it will likewise enhance its market value, so small shareholders are better off holding onto their shares than tendering them. Takeovers can therefore be difficult when ownership is dispersed. Indeed, the theoretical arguments suggest that the takeover mechanism should be even less effective than it seems to be in practice. It has become more of a theoretical puzzle to explain the existence of managerial discipline than of managerial discretion.

These observations argue for a return to the viewpoint of the earlier managerial literature, which presumed that mangers have discretion and asked how they will use it (see James G. March and Simon, 1958; Robin L. Marris, 1964; Oliver E. Williamson, 1964; Harvey Leibenstein, 1966). This paper continues that tradition, examining one route - namely, investment choice - by which managerial discretion affords managers the opportunity to obtain "rents" (payments in excess of their opportunity costs). We thereby expand on the theme of Shleifer and Vishny's (1989) insightful paper on "managerial entrenchment."

A somewhat earlier literature on incomplete contracts emphasized the investment consequences of rent-seeking. That literature explained how concern about opportunism can lead to underinvestment in such relationship-specific investments such as a manager's investment in firm-specific human capital (see Aaron S. Edlin and Stefan Reicheistein [1993] for citations). We argue here for a broader view of the biases in managers' decisions: In general, managerial rent-seeking affects not only the level of investment, but also the form.

Our basic hypothesis is simple: given the now well-established scope for managerial discretion, managers have an incentive to exercise that discretion to enhance their income. Any managerial contract is subject to renegotiation, and a manager's pay is the outcome of an often bewildering bargaining process between management, the board of directors, and rival management teams or takeover artists. Two critical factors in that bargaining process are the incumbent management's productivity relative to rivals and the wages that rivals demand. A manager may increase her compensation either by increasing her own productivity (the aspect generally stressed) or by decreasing the threat from rival managers. Shleifer and Vishny (1989) consider the latter possibility and show that a manager can increase her rents by choosing to invest in projects she can manage better than her rivals.

Here we explore a new avenue for entrenchment. The prototypical example of the Shleifer and Vishny (1989) effect is a secretary who becomes invaluable by rearranging a filing system to suit his own idiosyncratic search patterns. Yet we observe that managers often retain their jobs without making such investments; they are rehired even when they select assets for which their own idiosyncratic talents are not the best match. We suggest that these managers may preserve their jobs by investing in activities for which information asymmetries are particularly large (our arguments are in the same spirit as Laurie S. Bagwell and Josef Zechner [1993]). In our theory, managers invest to create these asymmetries, not to exploit their talents. They thus engage in a form of rent-seeking even less productive than that considered by Shleifer and Vishny (1989).

Our theory predicts, for example, that managers will overvalue acquisitions that promise potential, but not certain, synergies, since such acquisitions can bring rents to incumbent managers as long as they get earlier or more accurate signals than rivals of whether these synergies materialize. This argument suggests that the recent wave of acquisitions in the media and communications industries is a sensible entrenchment strategy, because the potential synergies are large, even if the expected synergies might not be.

Such possibilities become clearer as the paper unfolds. Section I sets up the framework for discouraging rivals. Section II analyzes how to discourage rivals who can observe investment levels. Section III considers discouraging rivals who cannot observe the mix of investment, and the final section outlines some implications of the analysis both for policy and empirical work.