The motivating intuition is that the presence of nonmaximizing agents induces maximizing agents to take advantage of them, and that this might magnify the effect of small deviations from maximizing behavior. This intuition is explored using a simple dynamic model. With an inflexible entry process, small deviations from maximizing behavior may have a substantial impact on the allocation of gains from trade. With a flexible entry process, the effect is dampened by adjustments in entry. Yet these deviations result in a first-order efficiency loss, in contrast to the second-order loss that one would expect from looking at standard static models.

We explore how allowing votes to be traded separately of shares may affect the efficiency of corporate control contests. Our basic set-up and the nature of the questions continue the work of Grossman and Hart (1980), Harris and Raviv (1988), and Blair, Golbe and Gerard (1989). We consider three cases with respect to the allowable price offers (for shares and for votes when they can be traded separately): unrestricted price offers, quantity-restricted price offers, and price offers contingent on winning. Our main results are characterizations of the equilibria and of the circumstances under which vote buying is harmful. We show that allowing votes to be traded separately of shares results in inefficiencies in all the cases we study. Similarly allowing quantity-restricted offers is also harmful, but allowing conditional offers is not in itself detrimental to efficiency. The paper also makes a methodological contribution to the analysis of takeover games with atomless shareholders. It provides a way of dealing with asymmetric equilibria that must be dealt with for a complete analysis and it proves existence of an equilibrium.