Files in this item



application/pdfPollak_Micah.pdf (741kB)
(no description provided)PDF


Title:Essays in political economy
Author(s):Pollak, Micah
Director of Research:Polborn, Mattias K.
Doctoral Committee Chair(s):Polborn, Mattias
Doctoral Committee Member(s):Bernhardt, Daniel; Krasa, Stefan; Gahvari, Firouz
Department / Program:Economics
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Dynamic elections
Endogenous issues
Public choice
Rational addiction
Video games
Abstract:In this dissertation I consider the application of an economic framework to situations outside of traditional economics. I look at two areas: the first is the behavior of politicians in dynamic elections and the second is addiction in online video games. I show how the power of economic incentives shapes phenomena that are outside the realm of traditional economics. Chapter 1 analyzes the incentives of politicians when their behavior in office affects the severity of issues upon which their re-election may depend. Elections are often about which candidate can best deal with the most pressing national issues. The severity of issues, however, is endogenous and depends on the actions taken by earlier politicians. If a politician has a comparative advantage in dealing with a particular issue, then this endogeneity creates an agency problem. An incumbent has an incentive to manipulate the severity of issues to create a more favorable environment for re-election. To analyze this incentive problem I develop a dynamic elections model in which candidates from differentiated parties have a comparative advantage in dealing with a specific issue. The incumbent chooses how much to invest in each issue, which endogenously determines the severity and relative importance of these issues to the voters in the future. I show that when politicians care about re-election they invest inefficiently. An incumbent invests less in the issue in which he has an advantage and more in the issue of his opponent. Because of this behavior, issues improve slower over time, are more severe in the stationary state, and parties remain in control of an office longer than is socially optimal. Chapter 2 uses a unique panel dataset to analyze rational addiction in an online video game. As playing video games becomes more common among both children and adults, the extent to which video games can and should be considered addictive has become controversial. To look at this question from an economics perspective, I develop a model of rational addiction for video games. Using a unique and very large panel dataset collected from the online video game \textit{Team Fortress 2} I show evidence of rational addiction: past and future consumption of this game play a significant role in determining how much an individual plays today. My data are rich enough that by estimating the model separately for each individual, I am able to identify and characterize potential addicts in a way that is consistent with rational addiction. The individuals identified as addicts in this way are very different than from using simple metrics, for instance by only looking at how much an individual plays, to define addiction. Estimating the model separately for each individual also provides evidence of significant heterogeneity among individuals, and suggests an individual-specific approach to analyzing addiction. Finally, I modify the model to allow for learning or endogenous skill development and provide evidence of a skill-playtime feedback loop: by playing today an individual improves his skill which reinforces his decision to play in the future and feeds back into addiction.
Issue Date:2012-09-18
Rights Information:Copyright 2012 Micah R. Pollak
Date Available in IDEALS:2012-09-18
Date Deposited:2012-08

This item appears in the following Collection(s)

Item Statistics