Citation
State of Connecticut (2009). Benefit Offset Pilot Demonstration: Connecticut final report.
Highlights
- This study’s objective was to examine 24-month impacts for Connecticut’s Benefit Offset Pilot Demonstration (BOPD), which provided a more-generous work incentive for Social Security Disability Insurance (SSDI) recipients, on recipients’ employment status, earnings, and reliance on benefits.
- Participants were randomly assigned to a treatment group, which was eligible for the more generous work incentive, or a control group that was not. The authors examined impacts on employment and earnings using state unemployment insurance (UI) wage records.
- The study found that, although BOPD achieved no statistically significant impacts on being employed, it did generate some positive impacts on earnings at the first year after random assignment.
- The quality of the causal evidence presented in this report is high for the year two impacts. This means we are confident that those estimated effects are attributable to the BOPD. However, the quality of the causal evidence for the year one impacts is moderate. This means we have confidence that the effects are attributable at least in part to the BOPD, but other factors might also have contributed.
Intervention Examined
The Benefit Offset Pilot Demonstration
Features of the Intervention
Administered by the Social Security Administration (SSA), the BOPD was a pilot test of the later Benefit Offset National Demonstration. The primary intervention was a benefit offset that replaced the so-called cash cliff SSDI recipients who work eventually experience. The cash cliff refers to the fact that SSDI recipients who have completed a nine-month trial work period followed by a three-month grace period have all their SSDI benefits suspended or terminated if they earn more than a threshold amount known as substantial gainful activity (SGA). The benefit offset replaces the complete loss of all benefits for working SSDI recipients, instead gradually withdrawing the SSDI benefit by $1 for every $2 earned above the SGA amount.
BOPD was implemented in four states: Connecticut, Utah, Vermont, and Wisconsin. Apart from recruiting SSDI recipients who received benefits solely based on their own earnings records, had completed a trial work period within the past 72 months, and were not concurrently receiving Supplemental Security Income benefits, the states were free to select their own program eligibility criteria. Connecticut recruited potential participants statewide from the state Medicaid Buy-In program, SSA benefits counseling program, and vocational rehabilitation program. Of the 6,726 Connecticut residents contacted to participate in the BOPD, 887 responded to the initial solicitation. Ultimately, only 265 were found eligible, consented to participate, and were enrolled in the demonstration.
Features of the Study
The BOPD in Connecticut was evaluated using a randomized controlled trial. Connecticut SSDI recipients deemed eligible for the pilot were randomly assigned to either the treatment group, which received the benefit offset, or the control group, which received services as usual. Both groups remained eligible to receive benefits counseling and other standard services.
The authors analyzed data from state UI wage records. The key outcome variables were quarterly employment rates, earnings, and an indicator of earnings in excess of the SGA level. The authors estimated impacts on employment and earnings outcomes for the two research groups for eight quarters (two years) after enrollment.
Findings
- The study found no statistically significant impacts on employment rates one and two years after random assignment.
- The study found statistically significant, positive impacts on average quarterly earnings at the first year following random assignment. However, this impact was not present at the end of year two.
- Treatment group members were more likely than control group members to have earnings in excess of the SGA amount at one (37 versus 16 percent) and two years (28 versus 16 percent) after random assignment.
Considerations for Interpreting the Findings
The volunteers who participated in the BOPD are not likely to be representative of the larger SSDI recipient population because they were typically recruited from organizations that helped clients obtain employment, and are therefore likely to be representative of the 20 percent of SSDI recipients who report an expectation of returning to work. In addition, the decision by SSA to restrict eligibility to SSDI recipients who completed their trial work period within the past 72 months disqualified potential participants with longer work histories; these people might have benefited the most from the program.
Causal Evidence Rating
The quality of the causal evidence presented in this report is high for the year two impacts. This means we are confident that those estimated effects are attributable to the BOPD. However, the quality of the causal evidence for the year one impacts is moderate. This means we have confidence that the effects are attributable at least in part to the BOPD, but other factors might also have contributed.