Essays in dynamic uncertainty: Behavioral economics, investment theory and law and economics

The dissertation is composed of three papers on three distinct topics. The papers and their abstracts are:;Exploitable play of believers in the "law of small numbers" in repeated constant-sum games. In repeated fixed-pair constant-sum games with unique equilibria in mixed strategies, rational players avoid exploitable play. Play is exploitable if it deviates systematically from the mixed strategy equilibrium choice probabilities, or if current play fails to be serially independent of past play. Experimental subjects often exhibit exploitable patterns. I develop a model to find patterns of serial dependence, to forecast play and detect players trying to exploit their opponent's patterns.;Investment and cash flow in dynamic firms facing uncertainty and liquidity constraints. Suppose firms facing conventional production functions, constant returns to scale and symmetric convex adjustment costs take random prices as given and choose capital and labor optimally. In a two-period model, a firm with perfect access to capital markets earns more expected cash flow and expects to invest more after a mean-preserving spread in output prices; a similar liquidity-constrained firm may expect less cash flow and investment after such a change.;Ethics, economics and lawyers' conflicts of interest. When a lawyer represents more than one client, the effect of common agency on clients depends upon the nature of the strategic interaction between the clients. Common agency can be synergistic, destructive or neutral. A simple game theory approach to the relationship between the principals distinguishes these three situations and shows when common agency is relatively efficient from the clients' perspective. A reputational dynamic imperfectly implements efficiency. Third party enforcement through the institutions of legal discipline can encourage efficient behavior if the legal rules are efficient. I show that the U.S. law regarding lawyers common agency is usually efficient in that the outcomes of the cases are aligned with efficiency. Further, while the frameworks are very different, legal analysis and economic analysis tend to treat similar cases similarly. Where economics and law differ, the analysis may suggest how to make the positive law of lawyers' conflicts of interest more efficient.