Essays in Applied Microeconomics

Public Deposited

This dissertation consists of three essays in applied microeconomics. In the first chapter, I introduce a new statistical test for identifying prejudice from empirical data. In the second chapter, I (joint with James Schummer) consider the revenue maximization problem for a two-sided, one-to-one matching platform. In the third chapter, I (joint with Tomer Hamami) consider the dynamic incentives of news firms to polarize consumers in order to increase rent extraction. Disparities along racial and ethnic lines persist across domains, and distinguishing among the possible sources of such disparities matters. Chapter 1 introduces the first absolute test for identifying prejudice in the common setting where the returns to treatment intensity are weakly diminishing for each treator. As examples, law enforcement officers engage in stops and searches, employers hire, judges deny bail, doctors administer procedures and medical tests, and creditors extend loans and mortgages. The test additionally unifies the existing literature and can be partially extended to the important setting where treators vary in the quality of their information. Empirically, the test finds evidence of prejudice in a dataset where the existing test does not. Methodologically, the test is the first in the literature to jointly use actions and outcomes in the test statistic. Chapter 2 considers the revenue maximization problem for one-to-one matching platforms that choose among stable matching mechanisms. In this sense the work bridges an existing gap between two literatures in economics: matching and two-sided markets. In seeming contrast with recent results in the matching literature, a platform does not want to price discriminate across the two sides of the market based on their relative sizes when preferences are independent. The analysis introduces a new class of matching mechanisms and an approximation for the stable platform's expected revenue using an exact expression for revenue under a constrained serial dictatorship mechanism. The chapter concludes by considering how correlation in agents' preferences affects results and how this contrasts with classic models of pricing in two-sided markets. Chapter 3 addresses a gap in the literature on political slant in news media regarding the means and incentives news firms have to affect their consumers' beliefs over time. Specifically, competitive tensions generate an incentive for media firms to intentionally polarize consumers in order to increase product differentiation, and this can occur even if consumers are initially homogeneous and unbiased. Computational analysis and simulations corroborate this analysis

Last modified
  • 01/16/2019
Date created
Resource type
Rights statement