Reducing Unemployment Insurance Improper Payments

The failure of claimants to meet work search requirements is a major source of improper payments in the Unemployment Insurance program, and one that has been challenging to address. DOL's Office of Unemployment Insurance (OUI) commissioned a study to identify drivers of work search improper payments (errors) and provide evidence on potential ways to reduce the work search error rate.

Mathematica's study examined several questions to help identify ways to reduce work search improper payments. They included:

• How do state workforce agencies (SWAs) vary in their policies and procedures related to work search?
• Which state policies and practices are correlated with higher or lower rates of work search improper payments? 
• How do states vary in how they measure improper payment rates and how does that affect cross-state differences in observed improper payment rates?   

To answer those questions and identify options for reducing work search improper payments, the study team collected and analyzed data from multiple sources, including (1) Benefit Accuracy Measurement (BAM) data, a federal measure of UI improper payment rates; (2)  the State Policies and Practices Dataset created by Mathematica from existing sources of published information; and (3) direct outreach to a subset of SWAs. 

Findings

We found substantial variation across states in their work search policies and practices, including the amount of work search effort that UI claimants are required to undertake, the claimants' requirements to record and report their efforts, and the SWAs’ strategies to verify and enforce adherence to requirements. These differences can affect the amount of work search undertaken by claimants and whether their efforts meet UI eligibility requirements. In the BAM data, we found large variations across states in work search error rates, with many states having zero or few work search errors. Much of the observed variation in improper payment rates was due to difference in whether SWAs deemed a payment improper when a claimant had not complied with work search requirements or had not responded to the request for documentation, rather than to variation in rates of claimant non-compliance. As a result, BAM-based improper payment rates do not provide meaningful cross-state comparisons of the amount of work search compliance. Our quantitative analyses of predictors of work search improper payments found little robust evidence of particular policies likely to cause higher rates, but that is not surprising given the poor quality of the data on those errors. The project team produced alternative payment rates that corrected some of the weaknesses in the original BAM rates. If good measures of work search error do not exist, it is difficult to empirically identify which policies are most important in driving those errors or for reducing them. The project team presented DOL with a range of recommendations for improving measurement of work search improper payments, in order to permit future work to identify ways to increase work search activity and reduce improper payments.