Three Essays on the Impact of Unemployment Insurance on Employment, Benefit Takeup and Other Labor Market Outcomes

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Jeffrey D. Kubik


Employment, Layoff, Subsidies, Takeup, Unemployment insurance

Subject Categories



This dissertation consists of three chapters on the impact of unemployment insurance on employment, benefit takeup and other labor market outcomes. Chapter one studies the short run employment effect of unemployment insurance policies in the United States by examining how the variations in average benefits, taxes, or subsidies per worker at major industry level in a state/year impact the industry employment level across states and over time. I build simulated instruments for the state UI policy measures to avoid endogeneity issue arise from using UI policy measures derived from administrative data on wages and taxes. I find that benefits have a positive effect while taxes have a negative effect on industry employment before I add the state year interaction term to the regressions. After adding the interaction term to account for the fact that states change their tax schedules often, annually if necessary to meet the need of balancing state UI trust fund, I find the effect of UI taxes and benefits goes away. I investigate further by regressing state average UI measures on lagged state unemployment rate and find clear evidence for policy endogeneity in state UI programs.

The second chapter studies the effect of UI policy on benefit takeup rates. Although the Unemployment Insurance program in the United States provides almost universal coverage to wage and salary workers, the estimated takeup rates have been surprisingly low. Earlier research either uses state level data that do not explore the industry level differences in UI benefits and taxes, or uses administrative data to derive UI policy measures that are endogenous with takeup rates. This study applies instrumental variables of UI measures to analyze the takeup rates at industry level in a state year. I find the benefit elasticity of takeup is 0.81, very close to what Anderson and Meyer found in their 1997 study after adding the earnings spline to control for endogeneity. I also look at the effect of UI taxes on takeup rates and find very small and insignificant effect.

Chapter three studies the effect of unemployment insurance benefits on temporary layoffs waiting to be recalled. Earlier work has used endogenous UI measures derived from individual data on benefits and wages. The main contribution of this paper is to use instrumental variables of UI benefits and taxes at industry-year-state level from all 50 states and D.C. to analyze temporary layoffs at the industry level. I use CPS Basic Monthly data from 1990 - 2002 for this study. I find an increase of average UI benefits by a standard deviation, about $100, leads to a 29.5% increase in temporary layoffs. I use the other unemployment - those unemployed that are looking for a job instead of waiting to be recalled as the "control group" and find higher benefits have no decisive or statistically significant effect on the other unemployment. I further analyze whether the increased observations of temporary layoffs come from higher incidence or longer durations and find higher benefits do no impact incidence of temporary layoffs in a statistically significant way, indicating that most of the increase of observations come from longer durations of temporary layoffs.


Open Access

This document is currently not available here.