Essays in search theory

The dissertation analyzes several important applications of the search-theoretic approach to monetary economics and intermediation theory. After presenting in the first chapter (joint with Peter Rupert, Martin Schindler and Randall Wright) a simple version of the basic monetary models used in the search-theoretic literature we argue in the second chapter (joint with Randall Wright) that it is not possible to construct robust equilibria where some agents use money and other agents do not in a search-theoretic model with homogenous agents. Hence, we examine monetary equilibria where agents can be heterogeneous in terms of utility functions, discount rates, production and storage costs, and so on. Equilibrium is defined in terms of the sets of agents who use money and those who do not. Existence and multiplicity are discussed for a variety of fairly general versions. For some cases one is able to construct a statistic that summarizes all of an agent\u27s characteristics by one number, with the property that agents use money if and only if their value for this number exceeds a certain threshold. However, even in these simple cases, there typically exist multiple equilibrium values for this threshold. The last chapter is devoted to a theory of intermediation in a search-theoretic environment. It provides an equilibrium model, in which intermediaries, like retailers, buy goods from producers and sell them to consumers. A key decision emphasized here is the size of the intermediary: a larger inventory makes it more likely that a given customer will find something that he likes, but it is also more costly to store. The distribution of inventories across intermediaries, and the distribution of prices, which are determined using bargaining theory are characterized. It is shown how to solve for the equilibrium size and number of middlemen in the economy, and this is compared to the solution to the social planner\u27s problem. It is possible to characterize conditions under which there are too many or too few intermediaries, and under which intermediaries will be too big or too small, in equilibrium