We study a dynamic general equilibrium model in which firms choose their investment level and capital structure, trading off the tax advantages of debt against the risk of costly default. Bankruptcy costs are endogenous, as bankrupt firms are forced to liquidate their assets, resulting in a fire sale if the market is illiquid. When the corporate income tax rate is positive, firms have a unique optimal capital structure. In equilibrium, firms default with positive probability and their assets are liquidated at fire-sale prices. The equilibrium features underinvestment and is constrained inefficient. In particular there is too little debt and default.

[It is generally agreed that speculators can make profits from insider trading or from the release of false information. Both forms of stock-price manipulation have now been made illegal. In this article, we ask whether it is possible to make profits from a different kind of manipulation, in which an uninformed speculator simply buys and sells shares. We show that in a rational expectations framework, where all agents maximize expected utility, it is possible for an uninformed manipulator to make a profit, provided investors attach a positive probability to the manipulator being an informed trader.]

[How should new securities be designed? Traditional theories have little to say on this: the literature on capital structure and general equilibrium theories with incomplete markets takes the securities firms issue as exogenous. This article explicitly incorporates the transaction costs of issuing securities and develops a model in which the instruments that are traded are chosen optimally and the economy's market structure is endogenous. Among other things, it is shown that the firm's income stream should be split so that in every state all payoffs are allocated to the security held by the group that values it most.]

We frequently observe that contracts do not include all of the contingencies that would seem to be necessary for optimal risk sharing between the parties to the contract. One reason may be that the possibility of renegotiation makes the contract more contingent than it appears. A simple contracting problem is used to show how even a simple contract may achieve optimal risk sharing if new information arrives slowly relatively to the speed of renegotiation.

Traditional asset-pricing theories assume complete market participation, despite considerable empirical evidence that most investors participate in a limited number of markets. We show that once the participation decision is endogenized, market properties change dramatically. First, limited market participation can amplify the effect of liquidity trading relative to full participation; under certain circumstances, an arbitrarily small aggregate liquidity shock can cause significant price volatility. Second, there exist multiple equilibria with very different participation regimes and levels of asset-price volatility. Third, under plausible conditions the equilibria can be Pareto-ranked; the Pareto-preferred equilibrium is characterized by greater participation and lower volatility.

Empirical evidence suggests that banking panics are related to the business cycle and are not simply the result of “sunspots.” Panics occur when depositors perceive that the returns on bank assets are going to be unusually low. We develop a simple model of this. In this setting, bank runs can be first-best efficient: they allow efficient risk sharing between early and late withdrawing depositors and they allow banks to hold efficient portfolios. However, if costly runs or markets for risky assets are introduced, central bank intervention of the right kind can lead to a Pareto improvement in welfare.

When the profitability of investment depends on the general level of economic activity, entrepreneurs have an incentive to delay investments during a recession. Endogenous delay thus prolongs the recovery from a recession and heightens the effect of the boom. This paper describes a dynamic model that exhibits both delay and cycles and develops methods for analysing the role of delay in propogating business cycles. A number of interesting characteristics of the cycle are revealed. First, the effect of delay is asymmetric: it lengthens the recovery but not the downturn. Second, delay can increase the amplitude and typically reduces the frequency of the cycle. Third, it can reduce the average level of activity, but it achieves this effect by prolonging the recession rather than by reducing the amplitude of the cycle. The welfare effects of delay are ambiguous, however.

The paper describes a Walrasian theory of markets with adverse selection and shows how refinements of equilibrium can be used to characterize uniquely the equilibrium outcome. Equilibrium exists under standard conditions. It is shown that, under certain conditions, a stable set exists and is contained in a connected set of equilibria. For generic models there exists a stable outcome, that is, all the equilibria in the stable set have the same outcome. These ideas are applied to markets with one-sided and two-sided uncertainty. Under standard monotonicity conditions, it is shown that the stable outcome is separating and implies a particular pattern of matches of buyers and sellers.

The cost of gathering information about unfamiliar securities may lead to gains from standardization: firms issue a particular security because it is used by other firms. To support standardization as an equilibrium phenomenon, information must be non-transferable (otherwise it might be revealed by prices or the observation of other agents' decisions) and it must be generic (useful in evaluating a number of securities). A competitive equilibrium in which standard contracts are used may be subject to coordination failure: while there always exists a constrained efficient equilibrium, there may also exist Pareto-ranked equilibria.