Author: Shengwu Li

Publisher:

Published: 2016

Total Pages:

ISBN-13:

This dissertation consists of three essays in theoretical and behavioral economics. They all concern decision-making in complex environments. The first chapter is entitled Obviously Strategy-Proof Mechanisms. It is generally held that some strategy-proof mechanisms are easy for non-experts to understand, and others are difficult to understand. However, this distinction is not captured by standard game theory. In this chapter, I define obviously dominant strategies. Whether a strategy is obviously dominant depends (just) on the extensive game form. I characterize this definition in two ways: Obviously dominant strategies are exactly those strategies that a cognitively limited agent can recognize as dominant. Obviously strategy-proof (OSP) mechanisms are those that can be run by a social planner with only partial commitment power. For an environment with one-dimensional types and transfers, I characterize the OSP mechanisms and the OSP-implementable allocation rules. I test and corroborate the theory with a laboratory experiment. The second chapter is entitled Context Effects as Explained by Foraging Theory, and is coauthored with Neil Yu. This chapter reconciles two seemingly competing explanations of context-dependent choice, one invoking psychological mechanisms, and the other Bayesian learning. We prove that standard context effects are features of the optimal solution to a general dynamic stochastic resource- acquisition problem. The model has two key ingredients: inter-temporal substitution and learning about the environment. Interpreted as a description of animal foraging behavior, it explains why context effects might be adaptive in nature. Interpreted as a description of consumer choice problems, it suggests that context effects might result from rational inference. A simple experiment shows that the latter interpretation sometimes holds. The third chapter is entitled Thickness and Information in Dynamic Matching Markets, and is coauthored with Mohammad Akbarpour and Shayan Oveis Gharan. We introduce a simple model of dynamic matching in networked markets, where agents arrive and depart stochastically, and the composition of the trade network depends endogenously on the matching algorithm. We show that if the planner can identify agents who are about to depart, then waiting to thicken the market is highly valuable, and if the planner cannot identify such agents, then matching agents greedily is close to optimal. We characterize the optimal waiting time (in a restricted class of mechanisms) as a function of waiting costs and network sparsity. The planner's decision problem in our model involves a combinatorially complex state space. However, we show that simple local algorithms that choose the right time to match agents, but do not exploit the global network structure, can perform close to complex optimal algorithms. Finally, we consider a setting where agents have private information about their departure times, and design a continuous-time dynamic mechanism to elicit this information.