Skip to main content

Unemployment insurance programs and the choice to leave the labor force (Conway, 2022)

Review Guidelines

Absence of conflict of interest.

Citation

Conway, P. J. (2022). Unemployment insurance programs and the choice to leave the labor force. Southern Economic Journal, 88(4), 1373-1400. DOI: 10.1002/soej.12557

Highlights

  • The study's objective was to examine the impact of North Carolina’s unemployment insurance policy reforms in 2013 on employment outcomes. 
  • The study used a difference-in-differences design to estimate the impact of North Carolina’s unemployment insurance policy reforms on labor status changes. Using data from the 2000 to 2019 Current Population Survey, the author compared labor status changes over time in North Carolina to the rest of the United States. 
  • The study suggested a positive relationship between North Carolina’s unemployment insurance policy reforms and the share of individuals exiting the labor force.  
  • The study receives a low evidence rating.  This means we are not confident that the estimated effects are attributable to North Carolina’s unemployment insurance policy reforms in 2013; other factors are likely to have contributed.

Intervention Examined

North Carolina’s 2013 Unemployment Insurance Policy Reforms

Features of the Intervention

North Carolina made two major changes to their unemployment insurance policies in 2013 that impacted beneficiaries throughout the state. First, they decreased the maximum weekly benefit from $535 to $350. Second, they reduced the number of weeks recipients were eligible for benefits from 26 to 20 weeks. The weekly eligibility limit was linked with the unemployment rate and fell to 12 weeks in 2015 after unemployment rates fell. As a result of these changes, the U.S. Department of Labor ended access to extended unemployment compensation payments for the state’s residents starting in July 2013.

Features of the Study

The study used a difference-in-differences design to estimate the impact of North Carolina’s unemployment insurance policy changes in 2013 on labor status changes. The sample included all individuals in the U.S. aged 25 to 54 in the 2000 to 2019 Current Population Survey (CPS).  The treatment group included residents of North Carolina whereas the comparison group included individuals from the rest of the United States.  

The author used an event-study design. Using individuals' monthly labor statuses, the author examined the share of individuals transitioning from one labor status (employed, unemployed, or not in the labor force) to another and estimated conditional transition probabilities. The author compared these conditional transition probabilities among North Carolina residents to individuals in the rest of the United States.

Findings

Employment  

  • The study suggested that North Carolina residents had a higher probability of shifting labor status from Unemployed to Not in the Labor Force in the first two quarters following the policy change, compared to the rest of the United States.  
  • The study suggested that North Carolina residents had a higher probability of shifting labor status from Not in the Labor Force to Employed in the first quarter following the policy change, compared to the rest of the United States.

Considerations for Interpreting the Findings

The author demonstrated parallel pre-intervention trends for labor status changes in North Carolina compared to the rest of the United States. However, the author did not control for age, race/ethnicity, or gender, as required in the SASR protocol. Since the author did not account for preexisting differences between the groups before the intervention, these preexisting differences between the intervention and comparison groups—and not North Carolina’s unemployment insurance policy changes —could explain the observed differences in outcomes.

Causal Evidence Rating

The study receives a low evidence rating.  This means we are not confident that the estimated effects are attributable to North Carolina’s unemployment insurance reforms in 2013; other factors are likely to have contributed.

Reviewed by CLEAR

December 2024