Skip to content

Expirations of Pandemic Jobless Programs Caused an Unprecedented Drop in Access to UI

POLICY REPORT: Expiration of Pandemic Jobless Programs Caused Unprecedented Drop in Access to Unemployment Insurance PDF

TECHNICAL APPENDIX: Expiration of Pandemic Jobless Programs Caused Unprecedented Drop in Access to Unemployment Insurance PDF

PRESS RELEASE: Termination of Pandemic Unemployment Insurance Expansion Led to Unprecedented Drop in Access to Jobless Benefits, Especially for More Vulnerable Workers

At the start of the pandemic, Congress temporarily expanded Unemployment Insurance (UI) programs through the federal CARES Act. This report mainly focuses on Pandemic Emergency Unemployment Compensation (PEUC), which expanded the weeks of UI eligibility. This and other emergency provisions were set to expire on September 4, 2021, which is commonly referred to as a “benefits cliff.” However, policymakers in some states chose to end these programs before that date. This policy report evaluates the impact the expansion and expiration of these programs had on the US labor market using data from the U.S. Department of Labor, the Current Population Survey, and California’s Employment Development Department.

The research found that the primary effect of the early turn offs was to lower the share of unemployed that receive unemployment checks, depriving jobless workers of financial assistance and limiting the ability of the UI program to bolster the economy. In contrast, the research found no appreciable increases in employment following the sharp benefit cutoffs. Also, the resulting decrease in the recipiency rate (the share of unemployed workers who are receiving UI benefits) was three times as large as a similar cliff at the end of the Great Recession.

Key Findings

1. Access to UI fell drastically as a result of the benefits cliff. Prior to the turnoffs, about 50% of unemployed workers nationally were receiving UI benefits, but this rate declined to an average of around 20% two months after states’ turnoff dates.

This panel shows the population-weighted mean recipiency rate across all states around the time of withdrawal from UI expansions. This panel pools together data from both early and normal turnoff states, and then the data is normalized so that the month of the turnoff is shown as month 0 for both groups of states. The red vertical line represents the time of the UI expansion turnoff. The graph displays that, across all states, the average recipiency was steady prior to the turnoff at about 50% and then declined to an average of about 20% in one month post turnoff. The dotted lines represent the upper and lower bounds for the 95% confidence interval.

Notes: This figure presents event studies of the recipiency rate around the time of withdrawal from UI expansions. The top panel displays the population-weighted mean recipiency rate for all states (excluding Louisiana) combined. The recipiency rates are combined relative to each state’s turnoff, meaning that -1 on the x-axis is averaging May recipiency rates from early turnoff states and August recipiency rates from normally scheduled turnoff states. Louisiana is excluded from all panels. The dotted lines represent the upper and lower bounds for the 95% confidence interval. (A version of this figure in the report includes two additional panels, breaking out early turnoff and normal turnoff states). 

 

2. The drop in access showed up earlier in states that opted to end UI expansions earlier, underscoring the causal effect of the UI expansion in making benefits accessible. The two groups of states trended similarly in the months prior to early turn-off.

Figure 3: Trends in Recipiency Rates

This line graph displays the difference between the population-weighted average recipiency rates for early and normally scheduled turnoff states between January 2021 and August 2021, relative to what this difference was in June. The solid blue line represents this difference, while the dotted blue lines represent the upper and lower bounds for the 95% confidence interval for this difference. Prior to the June turnoff, the confidence interval contained the value of 0, suggesting that the gap between the recipiency rates for the two groups of states was stable prior to the turnoff. After the turnoff, this difference trends negative, reaching a difference of more than 20 percentage points by August, and zero is no longer within the confidence bounds. This suggests that the gap in rates widened after the turnoff.

Notes: In this figure, the difference between the population-weighted average recipiency for early and normally scheduled turnoff states (early minus normally scheduled turnoff) is shown relative to the difference in June, with the dashed lines representing the 95% confidence interval for this difference. Excludes Louisiana. (A version of this figure in the report includes an additional panel, showing the population-weighted average recipiency for early and normally scheduled turnoff states separately). 

 

3. The expiration of benefits spurred little, if any, immediate return to work. There was no noticeable jump in employment at the benefit cliff. Even accounting for month-to-month variability, employment gains due to the expiration of benefits could not have been greater than half of a percentage point.

This panel shows the population-weighted mean employment population ratio for all states, 3 months before, and 3 months after the turnoff of benefits. Data is normalized so that the month of the turnoff is set as 0 for both groups of states. The dotted lines represent the upper and lower bounds for the 95% confidence interval for this mean. This graph shows that the employment population ratio was rising steadily in the months before and after the turnoff, going from about 60% three months prior to the turnoff to about 61.25% three months after the turnoff.

Notes: This figure presents event studies of the EPOP rate around the time of withdrawal from UI expansions. It displays the population-weighted mean EPOP rate for all states (excluding Louisiana) combined. The EPOP rates are combined relative to each state’s turnoff, meaning that -1 on the x-axis is averaging May EPOP rates from early turnoff states and August recipiency rates from normally scheduled turnoff states (excludes Louisiana). The dotted lines represent the upper and lower bounds for the 95% confidence interval. (A version of this figure in the report includes two additional panels, breaking out early turnoff and normal turnoff states). 

 

4. Already vulnerable workers were the most likely to be impacted when PEUC terminated in California, including women workers, less-educated workers, older workers, Black workers, and food and retail service workers.

This bar chart shows education breakdown for UI claimants in California in the week of the PEUC turnoff. About 40% of PEUC claimants had only a High School Diploma or GED, while this number is 32% for regular claimants. The chart shows that groups of claimants with higher levels of education were more likely to be regular claimants as compared to PEUC claimants.

Notes: Displays the proportion of claimants by highest degree obtained by claimant type. The bars add up to 100% for each claimant type. Educational data is collected by the Employment Development Department and is self-reported.



Stay Informed