Essays in Economic Theory

Open Access
Author:
Salcedo Gonzalez, Bruno
Graduate Program:
Economics
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
May 11, 2016
Committee Members:
  • Edward James Green, Dissertation Advisor
  • Vijay Krishna, Committee Chair
  • Ruilin Zhou, Committee Member
  • Lisa Lipowski Posey, Committee Member
  • Lisa Lipowski Posey, Outside Member
Keywords:
  • Game Theory
  • Economics
  • Industrial Organization
  • Collusion
  • Pricing Algorithms
  • Repeated Games
  • Prisoners Dilemma
  • Risk Aversion
  • Rationality
Abstract:
The present dissertation consists of three independent essays on Game Theory and its applications to Economics. The first essay compares two standard notions of rational choice under uncertainty: dominance by pure actions, and dominance by pure or mixed actions. I show that these two notions are equivalent for agents that exhibit sufficient risk aversion. Moreover, risk aversion is a cardinal property. Thus, the different forms of dominance considered cannot be distinguished based solely on the ordinal data that can be directly inferred from observed choices. The second essay explores the consequences of relaxing a standard assumption in the analysis of strategic situations: the assumption that choices are made simultaneously and independently. Choice interdependence can have drastic consequences. For example, it can enable cooperation in some single-shot prisoners’ dilemmas without the use of binding contracts or side payments. I consider a class partially specified environments in which the actions available to each agent and the agents’ preferences are fixed, but the sequential and information structures of choices are design variables. I propose a simple and tractable solution concept—interdependent choice equilibrium—that fully characterizes all the outcomes that can be implemented in such environments without contracts or side-payments. The third essay proposes a model of dynamic price competition between firms that use automated pricing algorithms to set their prices. In my model, algorithms are fixed in the short run but can be revised at exogenous dates, and each firms is able to decode its rival’s algorithm. I show that, when the revision opportunities are infrequent relative to the arrival of consumers into the market, every sub-game perfect equilibrium of the game leads in the long run to profits that are arbitrarily close to the Pareto frontier. That is, the use of pricing algorithms might not only facilitate collusion, but might inevitably lead to it. In contrast with the plethora of equilibria that typically arises in models of repeated competition, my model is able to generate sharp predictions about the behavior of the firms in the long run.