COMING SOON! PQDT Open is getting a new home!

ProQuest Open Access Dissertations & Theses will remain freely available as part of a new and enhanced search experience at

Questions? Please refer to this FAQ.

Dissertation/Thesis Abstract

Unemployment Insurance Policies and Employment
by Maravi Meneses, Cristian A., Ph.D., University of Rochester, 2021, 136; 28315037
Abstract (Summary)

This dissertation consists of three essays concerning the way in which the unemployment insurance (UI) system in the United States affects the dynamics of labor markets, through the financing and the provision of unemployment benefits.

The first chapter provides background on the financing of the UI system. In the U.S. unemployment benefits are financed with employer payroll taxes using experience rating methods, which consist of assigning a tax rate based on the firm's history of unemployed workers who collected unemployment benefits. This creates a cost to the firm when changing employment downward: the expected UI marginal tax cost (EMTC). However, these methods are imperfect in that not every additional separation increases UI taxes (there are minimum and maximum tax rates). I implement a novel empirical measure of the EMTC assuming firms have rational expectations. I find that, on average, firms expect to pay about 56\% of the additional unemployment benefit, with the remaining subsidized. This subsidy has potential implications for the dynamics of labor markets, which I explore in subsequent chapters.

The second chapter studies the effect of increasing unemployment benefits on employment under an imperfect experience rating UI system. An endogeneity problem arises because both subsidized and total unemployment benefits respond endogenously to employment shocks. Thus, I construct instruments to overcome endogeneity by interacting systematic differences across states with aggregate shocks. I find that subsidized unemployment benefits reduced employment during the Great Recession. Also, using job flows data, I show that subsidized unemployment benefits have a positive effect on job separations, consistent with the negative effect on employment, while having little effect on job creations.

The last chapter studies the effect of unemployment subsidies on the composition of unemployment by UI claims. I show that unemployment subsidies increase the incidence of insured unemployment, but reduce that of uninsured unemployment. While the first result is predicted by the existing theory, the second is not. I rationalize this result by proposing a relabeling hypothesis: firms profitably split the non-negative subsidy with workers by relabeling the job separation as UI eligible. The relabeling hypothesis suggests that if the UI subsidy is high, workers receiving benefits should better resemble the characteristics of workers separating via quits. I in fact see evidence of this hypothesis in terms of the impact of the UI subsidy on unemployment duration, recall rate, and wage growth of separated workers.

Indexing (document details)
Advisor: Bils, Mark
Commitee: Bai, Yan, Alexander, Dan, Nelson, Paul
School: University of Rochester
Department: School of Arts and Sciences
School Location: United States -- New York
Source: DAI-A 82/8(E), Dissertation Abstracts International
Subjects: Economics, Labor economics
Keywords: Unemployment insurance, Employment
Publication Number: 28315037
ISBN: 9798582542537
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy