Absence of conflict of interest: This study was conducted by staff from Mathematica Policy Research, which administers CLEAR. Therefore, the review of this study was conducted by an independent consultant trained in applying the CLEAR causal evidence guidelines.
Citation
Corson, W., Decker, P., Dunstan, S., & Kerachsky, S. (1992). Pennsylvania Reemployment Bonus Demonstration final report. Unemployment Insurance Occasional Paper 92-1. Washington, DC: Employment and Training Administration, U.S. Department of Labor.
Highlights
- The study’s objective was to examine the impacts of the Pennsylvania Reemployment Bonus Demonstration Project on the unemployment insurance (UI) receipt, employment, and earnings of UI claimants approximately one year after program enrollment.
- In this demonstration, about 15,000 eligible claimants were randomly assigned to one of six treatment groups with varying financial incentives and qualification periods or a control group. For all participants, the study team collected administrative data, which provided information on UI benefits receipt, employment, and earnings. The study team also surveyed a subsample of approximately 5,000 participants about their experiences in the year following random assignment.
- The study found that two of the bonus treatments reduced the weeks of UI benefits and total UI benefit dollars received, but the others did not. The study found very few impacts of the bonus treatments on claimants’ reemployment rates or earnings.
- The quality of causal evidence is high for UI benefits receipt outcomes because it is based on a well-implemented randomized controlled trial. This means we are confident that the estimated effects are attributable to reemployment bonus incentives, and not to other factors. However, the quality of causal evidence is moderate for the employment and earnings outcomes because study attrition could not be determined for those outcomes.
Intervention Examined
The Pennsylvania Reemployment Bonus Demonstration
Features of the Intervention
Implemented in 12 randomly selected offices across Pennsylvania, the demonstration restricted eligibility to UI claimants who had filed within the past two weeks, did not have a definite recall date to their former employer, and did not routinely obtain employment through a union hiring hall. The demonstration had six different bonus offerings. The first four entailed different combinations of reemployment bonus amounts—either three or six times the UI weekly benefit amount—and periods within which participants were required to find a job in order to claim the bonus, known as a qualification period. The qualification period was either 6 or 12 weeks. The fifth type of bonus declined from the high amount (six times the weekly benefit amount) to zero over the course of the benefit period, and the sixth was essentially the same as the fourth, except that it excluded the offer of a job search workshop, which was available to all five other bonus groups. To claim any bonus, claimants needed to maintain employment for four months.
Features of the Study
About 15,000 eligible UI claimants from 1988 and 1989 were randomly assigned to either receive the offer of one of the six reemployment bonuses or to an existing services control group, which was not offered a reemployment bonus and continued to receive weekly UI benefits payments subject to existing requirements. The study estimated the average difference between treatment and control group members on measures of UI benefits receipt, reemployment, and earnings using data from UI administrative records and a survey administered to about 5,000 participants. Because the only difference between the fourth and sixth bonus groups was that the sixth group was not offered a job search workshop, the two were combined in the analysis.
The authors estimated impacts by comparing the outcomes of each treatment group to the outcomes of the control group. Regression models controlled for weekly UI benefits amount, base period earnings, potential UI duration, gender, age, race, office, and recall expectations.
Findings
Public benefits receipt
- The study found that the claimants offered the lower of the two bonuses (three times the weekly benefit amount) and the shorter three-week qualification received UI benefits for 0.7 fewer weeks than the control group, for a reduction in UI benefits received of $103. These differences were statistically significant.
- Claimants offered the higher of the two bonuses (six times the weekly benefit amount) and the longer six-week qualification period received UI benefits for 0.8 fewer weeks and received $130 less than control group members. These differences were statistically significant.
- None of the other treatment groups saw statistically significant impacts on UI benefits receipt outcomes.
Employment
- Two of 80 short-term reemployment impacts estimated were statistically significant. This is roughly what would be expected by chance.
Earnings and wages
- Two of 40 short-term earnings impacts estimated were statistically significant. This is roughly what would be expected by chance.
Considerations for Interpreting the Findings
The UI benefits receipt outcomes are based on a well-implemented randomized controlled trial and so receive a high evidence rating. However, attrition could not be determined for the reemployment and earnings outcomes based on either administrative or survey data. The authors included sufficient control variables in their regression models to receive a moderate evidence rating for these outcomes.
In addition, most of the statistically significant findings reported are based on one-tailed tests; CLEAR’s standard is to use a two-tailed test. Because actual p-values are not reported, we cannot determine how many of these results would remain significant after adjusting for a two-tailed test, but presumably it is fewer.
The labor market and UI program have changed considerably since this study was conducted. Therefore, a similar study conducted in the current environment might find different results.
Causal Evidence Rating
The quality of causal evidence is high for UI benefits receipt outcomes because it is based on a well-implemented randomized controlled trial. This means we are confident that the estimated effects are attributable to reemployment bonus incentives, and not to other factors. However, the quality of causal evidence is moderate for the employment and earnings outcomes because study attrition could not be determined for those outcomes.