Issue Brief: Volume 1, Number 8

The Urgency of Reforming Entitlement Programs: The Case of Social Security Disability Insurance

Author: Mark Duggan

Enrollment in the Social Security Disability Insurance program has risen significantly since the late 1980s; consequently, program expenditures have far outpaced revenues and the SSDI trust fund is projected to hit zero in 2016. Moreover, the SSDI program, as currently administered, discourages applicants and recipients of benefits from seeking and returning to work, thereby reducing federal tax revenues at a time when deficit reduction is critical. The SSDI program can and must be reformed.


  • Enrollment in the Social Security Disability Insurance program has risen significantly since the late 1980s; consequently, program expenditures have far outpaced revenues and the SSDI trust fund is projected to hit zero in 2016.
  • Moreover, the SSDI program, as currently administered, discourages applicants and recipients of benefits from seeking and returning to work, thereby reducing federal tax revenues at a time when deficit reduction is critical.
  • The SSDI program can and must be reformed. Reconsidering the medical eligibility criteria for benefits, promoting earlier medical interventions, allowing for SSA representation at appeal hearings on benefit decisions, enacting time limits for some benefit awards, and increasing the frequency of continuing disability reviews can help make the program more efficient, encourage individuals to return to work, and enhance economic growth.

The Social Security Disability Insurance (SSDI) program represents an extremely important part of our nation’s safety net, as it protects workers and their families from the risk of a disability that prevents or greatly inhibits a person’s ability to work. It currently provides insurance to more than 150 million American adults.[1]

The future sustainability of this program, however, is at risk. Nearly 9 million adults received SSDI disabled worker benefits in July 2013. This number represents a significant increase in SSDI enrollment over the past 25 years. As shown in Figure 1, enrollment in the SSDI program has grown steadily since the late 1980s, from 2.3 percent of adults aged 25-64 in 1989 to 5.0 percent by 2012.[2] As the number of people receiving SSDI benefits has swelled, so too have total program expenditures, which exceeded $140 billion in the 2012 calendar year. Since SSDI recipients also receive health insurance through the Medicare program (after two years from onset of disability), total Medicare expenditures for SSDI recipients topped $80 billion in 2011. All told, SSDI expenditures currently exceed program revenues by almost 30 percent, and as a result, the program’s trust fund is rapidly being depleted, having fallen from $216 billion at the end of 2008 to $107 billion in July 2013. Current projections from the OASDI Trustees (under the intermediate scenario) suggest that the SSDI trust fund will hit zero in 2016.

Figure 1: Percentage of Adults 25-64 Receiving SSDI Benefits

Furthermore, this increase in SSDI enrollment has coincided with a reduction in employment rates among individuals with disabilities.[3] Academic studies have shown that the SSDI program, as currently administered, not only reduces the incentive to work, but also creates obstacles to re-entering the workforce. By inhibiting labor force participation, SSDI in turn reduces tax revenues—which is particularly troubling today, when revenue generation and deficit reduction are sorely needed.

In this issue brief, I summarize the factors that are responsible for the growth in SSDI enrollment and then outline the implications of this growth for the U.S. labor market. I conclude by discussing the potential for changes to SSDI that could increase employment and improve economic well-being among individuals with disabilities while also reducing the fiscal burden of the program.

Why Has SSCI Enrollment Increased?

Some of the growth in SSDI enrollment reflects demographic changes—in particular, the aging of the baby boom generation. Individuals in their fifties and early sixties are significantly more likely to receive SSDI benefits than their counterparts in their thirties and forties.

However, as Table 1 demonstrates, the percentage of adults receiving SSDI has also risen sharply within age groups.

Table 1: Percentage of Adults Receiving SSDI 

Consider individuals between the ages of 50 and 59. In 1989, 1 out of 23 adults in this age group was receiving SSDI benefits. But by 2012, this had almost doubled to 1 in 12. The increase was similarly dramatic for adults in their forties and also substantial for adults in their early sixties and those between 25 and 39. The aging of the population therefore explains just 18 percent of the increase in SSDI enrollment from 2.3 percent to 5.0 percent during the 1989 to 2012 time period.[4]

To be insured for SSDI benefits, a person must have worked in at least five of the ten most recent years. Because employment rates have increased among women since the 1980s, the fraction of women insured under the program has risen as well, from 66 percent to 76 percent between 1989 and 2012. This has also contributed to enrollment growth in the SSDI program and partially explains why SSDI has grown more rapidly among women than men during this time period. But this factor explains just 12 percent of the rise in SSDI enrollment. Taken together, the aging of the baby boom population and changes in the fraction of adults insured for SSDI can explain less than one-third of the growth in the program from 1989 to 2012, depicted in Figure 1.

A much more important determinant of the growth in SSDI enrollment since the 1980s is the liberalization of the program’s medical eligibility criteria that occurred in the mid-1980s.[5] As shown in Figure 2, there has been a dramatic increase in award rates for mental disorders and diseases of the musculoskeletal system (e.g., back pain). In contrast, award rates for neoplasms (cancer) and circulatory conditions (e.g., heart attack, stroke) have remained roughly constant. This shift is important because, as shown in recent research, the employment potential of SSDI applicants with these more subjective conditions remains substantial, and it is often difficult to verify the severity of these conditions (in contrast to cancer or heart conditions).[6] With the liberalization of the medical eligibility criteria, it has become increasingly possible for people who are capable of working to qualify instead for SSDI benefits.

Figure 2: Awards Per 1000 Insured for SSDI by Diagnosis Category in 1983, 1989, 1999, 2009

The reduction in the generosity of OASI retired worker benefits also has had a spillover effect in propelling SSDI enrollment. Individuals born in 1937 or earlier could receive 80 percent of their full retirement benefit if they claimed retired worker benefits at the age of 62. As a result of federal legislation passed in 1983, this has gradually fallen to 75 percent for individuals born from 1943 to 1954 and will soon fall to 70 percent for individuals born in 1960 or later (with an associated increase in the full retirement age from 65 to 67 as well). No corresponding changes were made to SSDI benefits and thus SSDI has become relatively more attractive financially. More specifically, SSDI benefits were 25 percent more generous than retirement benefits at age 62 for those born in 1937 or earlier but will be 43 percent more generous for those born in 1960 or later. Recent research demonstrates that the declining generosity of retired worker benefits has induced a substantial number of adults to apply for and ultimately receive SSDI, and that this explains a substantial fraction of the growth in SSDI enrollment since the late 1980s.[7]

The financial motivation for seeking SSDI benefits is all the greater because replacement rates (potential benefits divided by potential earnings) for the typical low-skilled worker have risen on account of two interrelated factors.[8] First, SSDI (like OASI) uses a progressive 90-32-15 benefit formula with “bend points” that increase each year with average earnings growth. Second, earnings for low-income workers have grown more slowly than the average, and as a result workers replace an increasing fraction of their earnings at a 90 percent rate rather than 32 percent rate. This has increased the financial incentive to apply for SSDI benefits and, once enrolled in the program, to stay enrolled.

The sensitivity of the SSDI program to economic conditions has been another important driver of enrollment growth. As shown in Figure 3, applications to the SSDI program are highly responsive to the unemployment rate, with applications rising substantially during economic downturns and falling when the economy improves. Previous research has shown that the SSDI program has become much more sensitive to economic conditions since the early 1980s (partly due to the program’s less stringent medical eligibility criteria) and that individuals who lose their job or who are unable to find a new job are increasingly likely to exit the labor force and apply for SSDI benefits.[9]> Thus the program is to some extent serving as a form of long-term unemployment insurance for some workers, which is troubling when one considers the very low rate at which SSDI recipients return to the labor force.

Figure 3: SSDI Applications Per 1000 Insured Workers and Unemployment Rate

Still other factors have contributed to the steady rise in SSDI enrollment since the late 1980s. Individuals who are initially rejected when they apply for SSDI have become more likely to appeal those decisions and are increasingly likely to be represented by a lawyer or other professional if/when they ultimately appear before an Administrative Law Judge (ALJ).[10] The fraction of recipients receiving a continuing disability review (CDR) and exiting the program for no longer meeting SSDI’s medical eligibility criteria has also declined due to a decline in authorized spending for CDRs.

For all of these reasons, enrollment in the SSDI program has grown steadily and rapidly while average health has if anything improved among non-elderly adults during this period.[11]

Labor Market Effects of the Rise in SSDI Enrollment

While providing valuable insurance to tens of millions of Americans, the SSDI program reduces the incentive to work both for individuals on the program and also for those applying for SSDI benefits. In order to receive an SSDI award, a beneficiary must be deemed unable to engage in substantial gainful activity (SGA, currently $1,040 per month). Once on the program, an SSDI recipient has little incentive to return to work, as earnings above the SGA threshold lead to a termination of benefits. And given that the present value of the average SSDI award is $270,000 (including Medicare benefits), that is an outcome that many SSDI recipients would be reluctant to seek.

The growth in SSDI enrollment has coincided with a substantial reduction in employment rates among individuals with disabilities. For example, from 1988 to 2008, the employment rate of men in their forties and fifties who reported a work-limiting disability fell from 28 percent to 16 percent while the corresponding rate for men without a disability rose slightly from 87 to 88 percent.[12] As shown in Figure 4, the gap in employment rates grew similarly rapidly for women with and without disabilities in this same age range.

Figure 4: Employment Gap: Men and Women Ages 40-59 without Disabilities Relative to those with Disabilities, 1988-2008

Previous research has shown that workers have become increasingly likely to respond to adverse labor demand shocks by applying for SSDI rather than seeking a new job.[13] This serves to reduce both the unemployment rate and the labor force participation rate below what it otherwise would be. It also reduces the eventual employment rate as SSDI recipients rarely leave the program to return to the workforce. For example in 2010, only 0.7 percent (7 out of 1,000) of SSDI recipients left the program for improving health and/or to return to work.

This responsiveness of the SSDI program to economic conditions can be seen visually in Figure 2, with increases in the unemployment rate leading to large increases in the SSDI application rate. My analysis of this application data reveals that there have been approximately 2.5 million “extra” SSDI applications since 2008 as a result of the economic downturn. Many of these applicants have withdrawn from the labor force, either because they have been awarded SSDI benefits or are still in the process of applying for benefits given the long lags in the process (especially at the appeal stage). Still others have likely withdrawn because their attachment to the labor force has declined during the application process (even if ultimately denied) and thus their potential wages as well.

The steady increase in SSDI enrollment since the late 1980s has contributed to a differential decline in labor force participation among both men and women in the U.S. relative to other industrialized countries. For example, the labor force participation rate declined by 4.7 percentage points (from 93.4% to 88.7%) among men 25-54 in the U.S. during the 1990 to 2011 period while falling just 1.5 percentage points (from 93.6% to 92.1%) among the EU-15.[14] Similarly, while the labor force participation rate was almost unchanged among women 25-54 in the U.S. from 1990 to 2011 (rising slightly from 74.0% to 74.7%), it increased by 14.8 percentage points (from 63.7% to 78.5%) among women in the EU-15 during this same period. Thus labor force participation rates for both men and women in the 25-54 age range were in 2011 substantially higher in the EU-15 than in the U.S. While there are of course many factors that influence both the level and the trend in labor force participation, previous research indicates that the SSDI program is an important factor.

Improving Work Incentives In The SSDI Program

The disability determination process that is currently used by the SSDI program awards benefits to individuals who are deemed unable to engage in substantial gainful activity. This reduces the incentive to work among those who have filed an initial application for SSDI and among those appealing a rejection. According to data from the Social Security Administration, approximately 40 percent of SSDI awards are now made on appeal and the time between the initial application and the ultimate decision is very long for this group. For example, the average lag for an applicant who appeals to an Administrative Law Judge is 27 months.[15] This is problematic because those initially rejected are likely to be in better health on average than those receiving an initial award and thus to have higher employment potential. And the longer that a person remains out of the workforce, the more their earnings potential declines. Thus even if an applicant never receives an SSDI award, the program’s application process can permanently harm his/her employment prospects.[16]

One way to improve incentives in the SSDI program is to provide medical intervention sooner for individuals with work-limiting conditions, so that they can continue working. Many individuals with more subjective disorders—such as back pain—could benefit from such early intervention. In a recent paper, David Autor and I proposed adding a “front end” to the SSDI system that would include early intervention through rehabilitation and related services with the goal of keeping workers with work-limiting disabilities in the labor market.[17] Employers would contract with private insurers to administer this coverage and would have a financial incentive to keep their workers off the SSDI system (much as the Unemployment Insurance and Workers’ Compensation programs provide employers with these types of financial incentives).

The payoff to keeping a potential SSDI applicant in the workforce is very high. As noted above, the average present value of an SSDI award (including Medicare expenditures) is approximately $270,000. Additionally, to the extent that the program reduces employment, it reduces tax revenue and GDP. While many who are awarded SSDI benefits are completely unable to work, recent research makes clear that a substantial number of them could work.

Increasing employment among individuals with disabilities could improve their economic well-being and increase their autonomy while also reducing the fiscal strains on Social Security. Past efforts to achieve this goal have unfortunately had little impact. For example, the Ticket to Work and Self-Sufficiency program, which was authorized by Congress in 1999, gives employment networks and vocational rehabilitation agencies a financial incentive to help SSDI recipients return to work. Despite issuing more than 12.3 million tickets over a ten-year period, the program helped less than 17 thousand SSDI recipients (0.1 percent of all tickets issued) return to work. Part of the reason for the low takeup was that the incentives often arrived too late – after SSDI recipients had been out of work for many years. Recent efforts to more directly increase work incentives among disability insurance recipients by eliminating the “cash cliff” have had some success in other countries and similar reforms in the U.S. could increase employment and economic well-being among current SSDI recipients.[18]

There are other potential reforms that could improve the functioning of the SSDI program. For example, currently only the applicant and his/her representative are present at appeal hearings before ALJs. Thus SSA does not have someone present to explain why they rejected the application twice, and this may partially explain why 72 percent of those initial decisions that appeal a second time are overturned by ALJs.[19] Additionally, there has been a substantial decline in recent years in the share of SSDI recipients receiving a continuing disability review (CDR), with this partially explaining the lower exit rate from the program. Careful consideration of the appropriateness of the program’s medical eligibility criteria also seems warranted given the major shift in the conditions with which individuals qualify for SSDI benefits, as shown in Figure 2. And to the extent that economic (rather than only health) factors are considered by a disability examiner or ALJ when making an SSDI award, one could consider a time limit or a mandatory CDR for some awardees.

The lack of progress in improving work incentives in the SSDI program stands in marked contrast to the Temporary Assistance to Needy Families (TANF) program. Reforms introduced in the 1990s (along with changes to the Earned Income Tax Credit) led to substantial gains in employment among past, current, and potential future TANF recipients and to a steady drop in program enrollment and expenditures. Similar progress is possible within the SSDI program if policymakers make reform of this program a priority. The need for such progress is indeed urgent, both because of the pending expiration of the SSDI program’s trust fund and the trends in the U.S. labor market described above.

The negotiations that will occur during the next several weeks regarding the debt ceiling and the possible government shutdown provide an excellent opportunity for policymakers—in a bipartisan fashion—to finally make meaningful changes to this large and rapidly growing program that has essentially been on auto-pilot for almost three decades. SSDI should remain a central part of the nation’s safety net, but with its current design, the program is simply not serving individuals with disabilities or taxpayers well. Reforms to the program that increase the incentive both for SSDI recipients to return to work and for potential SSDI applicants to remain employed would strengthen the program’s long-term financial outlook while enhancing economic growth.

About the Author

Mark Duggan, PhD

Mark Duggan is the Rowan Family Foundation Professor and Chair of Business Economics and Public Policy and Professor of Health Care Management at the Wharton School, as well as the Faculty Director of Penn Wharton PPI. He is also a Research Associate at the National Bureau of Economic Research. He received his B.S. and M.S. degrees in Electrical Engineering from M.I.T. and his Ph.D. in Economics from Harvard University. He currently is Co-Editor at the American Economic Journal: Economic Policy.

Professor Duggan was the 2010 recipient of the ASHEcon Medal, which is awarded every two years by the American Society of Health Economists to the economist aged 40 and under in the U.S. who has made the most significant contributions to the field of health economics. Along with his co-author Fiona Scott Morton, he received the National Institute for Health Care Management’s 2011 Health Care Research Award for their work on Medicare Part D. He was a Fellow of the Alfred P. Sloan Foundation from 2004 to 2006 and a Visiting Fellow at the Brookings Institution from 2006 to 2007.

Professor Duggan served from 2009 to 2010 as the Senior Economist for Health Care Policy at the White House Council of Economic Advisers. He has been a Consultant and Expert Witness for the U.S. Department of Justice, and recently testified before the Social Security Subcommittee of the U.S. House Committee on Ways and Means.


1 This brief draws heavily from recent testimony by Mark Duggan before the hearing on “Encouraging Work Through the Social Security Disability Insurance Program,” held by the House Committee on Ways and Means Subcommittee on Social Security, on June 19, 2013.

2 It should be noted that SSDI is not the only federal disability program that has been growing rapidly and becoming increasingly costly. The fraction of veterans receiving VA disability, for instance, has almost doubled (from 8.8 to 15.8 percent) since 1999, after remaining stable since the early 1970s.

3 David H. Autor and Mark Duggan, “Supporting Work: A Proposal for Modernizing the US Disability Insurance System,” Center for American Progress and the Hamilton Project (2010).

4 Put another way, if age-specific rates of SSDI enrollment had remained unchanged from 1989 to 2012, the percentage of adults 25-64 on SSDI would have increased from 2.3 percent to 2.7 percent.

5 Mark Duggan and Scott A. Imberman, “Why Are the Disability Rolls Skyrocketing? The Contribution of Population Characteristics, Economic Conditions, and Program Generosity.” In Health at Older Ages: The Causes and Consequences of Declining Disability among the Elderly (University of Chicago Press, 2009), pp. 337-379.

6 Till Von Wachter, Jae Song, and Joyce Manchester, “Trends in Employment and Earnings of Allowed and Rejected Applicants to the Social Security Disability Insurance Program,” American Economic Review 101, no. 7 (2011): 3308.

7 Mark Duggan, Perry Singleton, and Jae Song, “Aching to Retire? The Rise in the Full Retirement Age and Its Impact on the Social Security Disability Rolls,” Journal of Public Economics 91, no. 7 (2007): 1327-1350.

8 David H. Autor and Mark Duggan, “The Rise in the Disability Rolls and the Decline in Unemployment,” The Quarterly Journal of Economics 118, no. 1 (2003): 157-206.

9 Autor and Duggan, 2003.

10 In the average year from 2000 to 2008, administrative law judges made awards in 72 percent of their decisions. This is striking when one considers that ALJs consider appeals only among those rejected twice previously by SSA. One potential contributor to the high award rate is that SSA is not represented at the hearing – only the applicant and/or his/her representative are typically present with the ALJ. See Social Security Administration, Annual Statistical Report on the Social Security Disability Insurance Program, 2011, available at (2012); David H. Autor and Mark Duggan, “The Growth in the Social Security Disability Insurance Rolls: A Fiscal Crisis Unfolding,” Journal of Economic Perspectives 20, no 2 (2006): 71-96.

11 Duggan and Imberman, 2009.

12 Autor and Duggan, 2010.

13 Autor and Duggan, 2003.

14 These differences are even larger when focusing on men between the ages of 25 and 64 and are somewhat smaller when restricting attention to the 1990 to 2008 period. Declines in labor force participation among men aged 25-54 were also much lower in Australia, Canada, and Japan than in the U.S. during this same period (OECD, 2013).

15 Social Security Administration, Office of the Inspector General, “Audit Report: Disability Claims Overall Processing Times,” 2008.

16 David H. Autor, Nicole Maestas, Kathleen Mullen, and Alexander Strand, “Does Delay Cause Decay? The Effect of Administrative Decision Time on the Labor Force Participation and Earnings of Disability Applicants,” MIT Working Paper, 2011

17 Autor and Duggan, 2010.

18 For evidence in Norway, see Andreas Kostøl and Magne Mogstad, “How Financial Incentives Induce Disability Insurance Recipients to Return to Work,” forthcoming in the American Economic Review (2013).

19 Social Security Administration, (2012).