Do Financial Incentives Induce Disability Insurance Recipients to Return to Work?

Disability Insurance (DI) programs in the U.S. and elsewhere have historically provided little incentive for recipients to experiment with returning to work, since recipients typically lose eligibility for benefits if they earn more than a modest amount. In recent years, however, some countries have adopted return-to-work policies that allow recipients to keep some of their benefits if they return to work. Under a proposed change to the U.S. DI system known as the "$1 for $2 offset," benefits would be reduced by $1 for each $2 of earnings above an allowed amount rather than revoked.

Advocates of these policies suggest that they may encourage DI recipients to return to work and ultimately lead some recipients to exit the DI program, increasing the income of DI recipients and reducing program costs. Opponents counter that by making the program more generous such policies may induce DI recipients to work less and may encourage more people to apply for DI. The effects of these policies in practice are not well understood, due to a lack of empirical evidence.

In How Financial Incentives Induce Disability Insurance Recipients to Return to Work (NBER Working Paper 19016), researchers Andreas Kostol and Magne Mogstad explore the effectiveness of programs that incentivize the return to work by DI recipients.

To do so, the authors make use of a natural experiment resulting from a change in DI policy in Norway. There are many similarities between the DI systems of the U.S. and Norway, not only in program design but in trends - in both countries, the DI rolls have expanded in recent years, due in large part to a liberalization of the screening process, resulting in a rapid increase in the share of DI recipients suffering from difficult-to-verify disorders such as mental illness and musculoskeletal disease. However, the U.S. DI program is smaller, offers less generous benefits, and has younger, lower-income beneficiaries.

In 2005, the Norwegian government introduced a return-to-work program. Prior to the program's introduction, beneficiaries faced a discontinuous drop in benefits if they earned a dollar above the substantial gainful activity (SGA) threshold. Under the new program, the discontinuous drop was eliminated and DI benefits were reduced gradually, by $0.60 for every $1 in earnings above the SGA, so that the level of earnings permissible before DI benefits were completely eliminated was increased. Importantly, only recipients who were awarded DI before January 2004 were eligible for the new program, allowing the researchers to compare the labor market outcomes of DI recipients facing the new scheme to those facing the status quo.

Turning to the results, the authors find that the return-to-work program increased labor force participation by three percentage points in the year of its introduction, with the effect rising to five to nine percentage points by the program's third year. The effect may strengthen with time due to the time it takes DI recipients to find jobs. The authors estimate that a ten percent reduction in the tax rate on labor force participation reduced labor force non-participation by one to three percent. The program also doubled participants' average earnings, with the largest effect on earnings near the SGA threshold, where the implicit tax on earnings decreased the most.

Next, the authors explore the effect of the program on DI benefits received, taxes paid, and program costs. They estimate that the program reduced costs by 3.5 to 5 percent, reflecting a significant decrease in benefits and a small increase in taxes paid by DI recipients.

The analysis is not informative about the level of induced entry that might occur if all new DI applicants were made eligible for the return-to-work program, since the program applied only to DI recipients who had been awarded benefits prior to its introduction. However, the authors calculate the amount of induced entry that would be necessary in order for the return-to-work program to lead to an increase in DI program costs, and find that the response to DI benefits would need to be much higher than what the existing literature suggests.

Finally, the authors explore the response to the program among different groups of beneficiaries. They find that increases in labor force participation are concentrated among younger DI recipients, who may experience greater gains from returning to the labor force or lower costs from working. The response is greater among DI recipients living in areas with low unemployment rates, suggesting an important role for demand side factors. The response is also greater among those with more education or labor market experience, and greater for males than females.

As the authors point out, the liberalization of the screening process and the shift away from physically demanding work have blurred the line between the totally and permanently disabled and those who are disabled but retain some work capacity. Their results suggest "that many DI recipients have considerable capacity to work that can be effectively induced by providing financial work incentives" and that providing such incentives both raises the income of DI recipients and reduces DI program costs. Although they encourage readers to exercise caution in applying their findings to other countries, their work suggests "that the work capacity and labor supply elasticity of DI recipients in Norway are comparable to those of DI recipients in the U.S., which lends some support to the external validity of our analysis of the return-to-work program."

The authors acknowledge funding from the Social Security Administration through grant #DRC12000002-01-00 to the NBER as part of the SSA Disability Research Consortium.


National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us