That’s so, perhaps, because an easy short-term fix to SSDI’s looming insolvency would be to support it with funds from its larger sister trust fund for Social Security’s Old Age and Survivors’ Insurance (OASI) program. However, draining funds from OASI will bring forward that fund’s insolvency and delay the introduction of long-needed structural reforms to SSDI’s rules and operational procedures to improve economic incentives and advance the welfare of millions of Americans. Making such policy changes requires a careful analysis of available information and calls for broader public engagement in debating policy alternatives, both of which take time.

It would be foolish not to acknowledge that a properly crafted wage insurance program to protect workers against the onset of work-disabling health conditions is an important element of the social safety net—that it is important to support the truly disabled who can no longer work and that the provision of such insurance strengthens incentives to remain attached to the workforce. However, SSDI’s current rules appear to achieve the opposite: provide incentives to people with marginal disabling conditions to remain idle and dependent on SSDI benefits. Indeed, SSDI’s rules and operations today almost compel people with temporary disabilities to remain permanently dependent on SSDI benefits.

In the previous article, Jeffrey Wolfe and David Engel discuss shortcomings in the jurisprudence and operation of the SSDI program. This article takes a broader look at the program’s incentive effects on people currently working and people already on the program’s rolls but who appear to possess significant work abilities. It also surveys some of the literature on reform options for SSDI.

SSDI Program Features

The SSDI program was introduced in 1956, following a long period of debate and controversy. Lawmakers’ reluctance to replace state-run programs for the disabled that received federal financial support with a federal disability program was based on uncertainty surrounding how much a federal program would cost and the inherent difficulty of setting administrative rules for determining if a person is disabled—the subjectivity involved in determining whether a person was truly work-disabled or simply unemployed for reasons such as age, lack of skills, or other non-health-related factors.

Today, SSDI is a wage-replacement program for covered workers with a physical or mental condition. The condition must create work impairments sufficiently severe to preclude substantial gainful employment for at least one year or is expected to cause death within one year. Coverage under SSDI depends on prior work history in an occupation covered under Social Security. That is, SSDI is an insurance program for workers, to replace wages lost by the onset of a disabling condition.

The current “definition of disability” under the SSDI program is different from the one that was adopted when the program was first introduced during the mid-1950s. At that time, allowance under the program was conditioned on a work impairment being of “long-continued and indefinite duration”—that is, cases in which the inability to work was permanent. Only workers over age 50 were eligible for SSDI benefits when the program was enacted. Even at that time, the processing of hundreds of thousands of applications to determine whether applicants met then-prevalent standards of proof—objective and subjective—required long and arduous administrative and adjudicating processes.

More than five decades of change, including the extension of benefits to those younger than 50 and shifting eligibility criteria from permanent to temporary (one year) work disabilities, has transformed SSDI from an “early retirement” program for disabled workers age 50 and older, to one that provides wage-replacement insurance to all workers. More importantly, however, today’s SSDI system provides permanent wage replacement even for work-disabling conditions that may be temporary. By itself, this program feature need not be problematic because those who recover from a work-disabling condition would, in general, prefer to reenter the workforce, earn higher wages than SSDI’s monthly benefit, and reap the social and psychological benefits of economic independence, community involvement, social status, and so on, that accompany gainful labor force attachment. That means those remaining on SSDI’s rolls should be genuinely permanently disabled and deserving of long-term wage-replacement benefits.

However, qualifying for SSDI benefits is an extremely lengthy, arduous, and expensive process. This year, for example, expected wait times to get disability cases heard before an administrative law judge (ALJ) range from 224 days in Shreveport, La., to 486 days in St. Louis, Mo. During that time, applicants and their representatives (if any) must document the nature of the disability and provide medical evidence on the applicant’s physical and/​or mental condition. Because SSDI requires applicants to demonstrate their inability to work, they are compelled to leave the workforce during the many months it takes to qualify for SSDI benefits and use up past savings to support themselves while waiting to be qualified. During that time, their skills may depreciate or become obsolete, their health condition may worsen, and their psychological preparedness to return to work may erode. As a result, their likelihood of rejoining the workforce, whether or not they ultimately qualify for SSDI benefits, may be considerably reduced because of the SSDI application and qualification process itself.

SSDI’s Record of Burgeoning Enrollment

SSDI enrollment has been growing rapidly during the last few decades. The number of people enrolled in SSDI increased from 4.3 million individuals in 1990 to almost 11 million in 2012. Not only has the absolute number of beneficiaries grown larger, the prevalence of disability—SSDI enrollment as a percentage of insured individuals—has also increased. The increase in the SSDI prevalence rate was quite rapid after the onset of the 2007–2009 recession, but the Social Security trustees expect it to plateau in the next few years. A part of the increased prevalence is attributable to the aging of the U.S. population and the entry of the baby boomers into their 50s and 60s during the last two decades. It is well established that the incidence of disability increases with age, so the observed increase in disability prevalence in the population is not surprising.

However, the aging population is only one of the factors underlying the increase in disability prevalence. Micro-data evidence shows that disability rates have increased within age groups. A comparison of the effects of increasing SSDI enrollment rates within age groups with the increase that would occur simply from population aging shows that the latter accounts for about 55 percent of the increase in SSDI enrollment after 1989. One reason for this is the increase in female labor-force participation, which means that more women are covered under SSDI. Even here, however, the age-specific increase in female labor-force participation has lagged far behind the increase in age-specific female SSDI enrollment rates.

It’s noteworthy that SSDI benefits are computed using a formula similar to that used to calculate OASI benefits, but SSDI benefits—unlike OASI benefits—are not reduced if workers qualify for SSDI before their normal retirement age. Thus, workers with a health or mental impairment intending to collect Social Security benefits before attaining their normal retirement age have an incentive to apply for SSDI rather than OASI because qualifying under SSDI yields a larger monthly benefit during their entire retirement.

SSDI has been transformed from an “early retirement” program for disabled workers age 50 and older, to a program that provides wage-replacement insurance to all workers.

SSDI enrollment growth has been especially rapid in the aftermath of the recession of 2007–2009 because of the spike in unemployment. With unemployment insurance being time-limited, many newly unemployed individuals with disabilities tend to use SSDI as a substitute source of support once unemployment benefits run out. Many of those applicants are likely to enroll successfully in SSDI because, under current disability determination rules, SSDI examiners and ALJs award (in some cases must award) benefits for impairments even when objective medical diagnostic criteria are unavailable.

SSDI Enrollment Incentives and Economic Effects

Under the current system, reliance on subjective evaluation and interpretation of disabling conditions—such as for musculoskeletal conditions and mental impairments—implies a large potential for inaccurate outcomes. Individuals with a health impairment who are work-capable but unemployed (or individuals with temporary work-disabling conditions) who successfully enroll in SSDI will face a severe disincentive to rejoin the labor force once the economy improves or their health impairments abate: not only will they lose their disability benefits but, if their condition relapses, they will have to repeat the difficult and lengthy application and eligibility determination process.

Evidence that SSDI is providing very significant work disincentives for those with marginal or temporary work disabilities has accumulated and grown stronger since the 1980s. Micro-data surveys such as the Current Population Survey reveal that, compared to previous decades, working-age individuals today are no less healthy: self-reported measures of health and disability indicate roughly the same frequency of a work-limiting condition among today’s workers as was the case in the 1980s. Figure 1 shows time trends in working and SSDI receipt among working-age adults who self-report a work-limiting disability. Since the mid-1980s, such individuals have increasingly exited the workforce or have chosen to work very few hours and limit earning within the substantial gainful activity (SGA) threshold set under SSDI eligibility rules. As the figure also shows, such individuals have increasingly enrolled into SSDI.

The emergence of such strong anti-work incentives under SSDI’s benefit eligibility rules can be traced to the liberalization of those rules during the mid-1980s. The revised rules have since included guidelines for the evaluation of pain as a disabling factor, clearing the way for disability awards under subjective judgments rather than clear and objective medical criteria. Today, evaluations of musculoskeletal and mental conditions for SSDI eligibility are universally governed by subjective judgments by SSDI examiners and ALJs because, in such cases, objective medical criteria are not available. This is not to say that there aren’t people with these afflictions who deserve SSDI, but rather that allowances based on subjective assessments are likely to involve many errors. Another reason for decision errors is the use of outdated information on the types of jobs that are available in the nation.

It is notable that the system’s response to errors is asymmetric. People with severe disabilities who are incorrectly denied benefits can appeal their decisions, but people with minor or temporary disabilities who are incorrectly awarded benefits remain on the disability roles for many years, until their cases are reevaluated. Unfortunately, the Social Security disability administration’s record of conducting such “continuing disability reviews” to detect medical improvement and revoke disability status has been backlogged for years.

The consequence of adopting more liberal and subjective evaluation criteria is expanded eligibility for SSDI. And because people with mental and musculoskeletal ailments are expected to live longer than people with traditional disabilities, and because more of them are women, today’s cohort of disabled individuals on SSDI have a longer expected lifespan than those of previous decades—another reason for higher age-specific SSDI prevalence rates today compared to earlier decades. This has obviously negative implications for the projected balance of SSDI expenditures relative to payroll tax revenues.

Attempts to measure labor losses to the economy from entry by disabled but work-capable individuals into SSDI rolls find significant effects. A 2012 RAND Center for the Study of Aging paper by Nicole Maestas, Kathleen J. Mullen, and Alexander Strand examines a sample of program applicants who are identical to each other with regard to the severity of their impairment and prior labor-force attachment. The employment rate of applicants “at the margin of entry” (23 percent of all applicants in 2005 and 2006) would have been 28 percentage points higher, on average, two years later if they had been denied SSDI benefits. Employment above the SGA threshold would have been 19 percentage points higher and annual earnings would have been higher by $3,781 (including those with zero earnings) after two years.

This research shows that many work-capable individuals with marginal impairments are deciding to apply for SSDI benefits, exiting the workforce, and eliminating earnings above the SGA level during the SSDI application process, and will likely remain unemployed upon qualifying for benefits, despite 28 percent of them being able to work and earn well above SSDI’s SGA level. A 2012 Federal Reserve Bank of Chicago paper by Eric French and Jae Song finds a similar (26 percent) decline in labor-force participation after three years on the part of those who were ultimately awarded SSDI benefits by an ALJ. French and Song also find that if there were no option to appeal the denial of benefits at the initial stage, the employment rate among unsuccessful SSDI applicants would be 35 percent higher.

Reform Approaches: Early Interventions and Changes to Eligibility and Benefits

Empirical studies suggest there is potential to prevent a significant fraction of those experiencing an onset of functional health impairments from ever applying for SSDI benefits, or at least to delay their SSDI applications through positive interventions. Those interventions must be made soon after the onset of a potentially work-disabling condition—if such individuals could be located sufficiently early in the process. The passage in 1990 of the Americans with Disabilities Act provided protections against discrimination for people with such conditions who wish to remain employed. Under the ADA, employers who deny jobs or do not provide appropriate accommodations and work conditions to the disabled face potential damages through the courts. Despite passage of the ADA, however, those among the disabled who could work have increasingly chosen to enroll into SSDI rather than seek supports under the ADA to remain employed.

There is potential for positive interventions to reduce the number of people who apply for SSDI benefits, or at least to delay their SSDI applications.

Since SSDI eligibility is contingent on prior employment, the most natural place to find people who are beginning to acquire a work-disabling condition is in the workplace. This requires employers’ involvement in detecting such cases early and providing disability management services, workplace accommodations, counseling, and work-adjustments in a timely manner.

Some experts and practitioners therefore suggest that the government should mandate the provision of such services by employers. A key issue, then, is whether such a policy would be cost-effective. For employers, such early intervention programs must reduce worker turnover and enhance overall productivity. For the government, costs of existing disability programs must be reduced by limiting or delaying entry into SSDI by workers retained in the workplace through additional employer services. In particular, for a policy that requires greater employer involvement, any higher net outlays by employers must be exceeded by sum of savings in the government’s budget and higher overall productivity from the additional employment. Unless this condition is met, a mandate of this type would not be economically sensible.

International evidence from nations that faced unsustainable increases in disability program enrollments and expenditures appears to be mixed. These nations have successfully implemented programs of employer-provided workplace accommodations for people who acquire work-disabling conditions. In some countries, such as the Netherlands, this was achieved through an employer mandate that required the provision of support services to workers who acquire a disability. Employer demand for case-management services created a demand for insurance against disability onset among workers—a demand that was met through the emergence of firms that specialize in disability case management services and insurance provision. Although Dutch enrollment and expenditures in public disability support programs have declined since the reforms were implemented, those outcomes were also supported by significant new eligibility restrictions and benefit limitations to discourage individuals from seeking a direct path onto the public disability program’s rolls. Disability benefit seekers are first routed through an evaluation process to assess residual functional and performance abilities. Those with such abilities are offered jobs with corresponding functional and performance requirements, even if they were in vastly different occupational, vocational, and education categories compared to the applicant’s job history. Disability reevaluations were also extended to current program enrollees younger than age 45. However, although Dutch disability rolls and public disability expenditures declined since those policies were introduced in the late 1990s, it remains unclear whether total costs (private employer plus government) of providing disability supports and services to retain disabled individuals on employer rolls have been reduced or whether increased costs dominated incremental productivity gains.

In the United States, disability insurance experts have proposed requiring employers to purchase private disability insurance to pay for support services for workers who acquire a disabling condition. Such a proposal would need careful study for cost effectiveness because employer premium costs would ultimately be shifted to employees. One criticism of such a proposal is that it violates the current public expectation of being able to apply for SSDI if they become disabled—a benefit that workers are already paying for through the SSDI portion of the payroll tax. However, such a proposal meets the fundamental public finance principle of “beneficiaries pay” for additional future (contingent) work accommodations and benefits. If such private disability insurance for wage replacement and support services delays SSDI applications by several months, cumulative public savings could be sizable.

Another alternative is to introduce higher experience-rated payroll tax rates for employers who fail to retain disabled employees (or tax credits for employers who do retain disabled workers) similar to that employed under the U.S. Unemployment Insurance program. Such a policy would induce employers to economize on their share of SSDI payroll taxes by offering work-accommodations and employment supports to workers who acquire health impairments, exhaust benefits from private short- and long-term disability insurance coverage, and eventually become work-disabled and seek SSDI enrollment.

What about reforms targeted toward people who are already enrolled in SSDI? Recognizing the desirability of incentivizing beneficiaries to exit from SSDI rolls and reenter the workforce whenever possible, the Social Security Administration has undertaken several initiatives and is investigating alternative policies through pilot projects. Unfortunately, the results to date of these initiatives have been disappointing. For instance, the Ticket To Work (TTW) program invites SSDI beneficiaries to explore the option of returning to work. For interested SSDI beneficiaries, the TTW program arranges to provide vocational rehabilitation, counseling, job placement, and other services. Unfortunately, the take-up rate for TTW is extremely low, partly because SSDI’s long and cumbersome qualification process discourages beneficiaries from risking termination from SSDI. TTW contacts beneficiaries when they are awarded benefits—usually many months after they first applied for SSDI, and well after applicants’ motivation and ability to return to work has significantly eroded. Another experimental program is the Benefit Offset National Demonstration (BOND), an initiative designed to investigate the effect of reducing the implicit tax rate that beneficiaries face upon returning to work. Under BOND, beneficiaries initially receive a nine-month waiver from any reduction of benefits when they first return to work, followed by a subsequent period in which benefits are reduced $1 for every $2 in beneficiary earnings above the SGA amount. The intent of BOND is to examine whether a slower benefit claw-back rate would induce more SSDI beneficiaries to resume gainful employment and whether such a program would promote “induced entry” into SSDI—that is, more applications by disabled individuals with residual work ability. The results to date from BOND, however, indicate a very small job take-up rate among existing beneficiaries.

Recognizing the structural shortcomings of the existing SSDI program, some advocates propose the creation of a new universal public insurance program—financed by levying a new payroll tax—to extend wage replacement and support services for the disabled. Services, work accommodations, and partial wage replacement for disabled workers would be made available upon the acquisition of a work-disabling condition. The insurance would be portable for the worker and would free employers from having to provide accommodations or services to new hires if they are disabled. Unfortunately, a publicly funded system of this type would mimic the shortcomings associated with all third-party-payer programs—Medicare, for example. With neither employers nor employees responsible for limiting the cost of accommodations, the type and cost of accommodations and the required payroll tax rate to finance them could skyrocket. In addition, it’s unclear how today’s SSDI beneficiaries would be grandfathered into the system. If their SSDI benefits are replaced with benefits from the new public disability system, there may be little overall cost savings and the unfunded obligations of the current SSDI system may simply be transferred to younger and future taxpayers through the new public disability insurance system.

Dutch disability reforms demonstrate the potential for sizable positive labor market effects. For example, the effect of benefit reductions in the Dutch disability system during the early 1990s showed that affected beneficiaries would partially (31 percent) offset benefit losses through other public and private social support programs. In addition, many beneficiaries returned to work and exhibited a large rebound in earnings (62 percent), indicating significant unutilized work potentials among existing disability insurance beneficiaries. Those results apply equally to long-term Dutch beneficiaries as to recent enrollees. Taken together, they show that, on average, beneficiaries make up the entire benefit reduction through other sources—social programs or additional earnings. However, Dutch benefit reductions were implemented when benefit replacement rates were extremely high—80 percent or more during the 1970s and early 1980s. Given the generally considerably lower—50 percent—disability benefit replacement rate in the United States, the adoption of similar disability reforms may not be feasible and may not yield similarly strong positive labor market effects.

Other reform directions include the direct reduction of benefit generosity, scaling benefits according to the severity and effect of health impairments on individuals’ work abilities, or time-limiting benefits to ensure return to work by people with temporary work-disabling conditions. Although such initiatives have yet to be attempted in the United States, experience from similar policies in the Netherlands appears to echo the outcomes of welfare reform legislation in the United States during the mid-1990s. This change replaced the Aid to Families with Dependent Children (AFDC) program, which provided cash benefits and food stamps to single mothers with dependent children under a means test, with the Temporary Assistance for Needy Families (TANF) program, which eliminated AFDC’s “entitlement” and substituted assistance payments for just five years over recipients’ lifetimes and not more than two years without participating in the labor force. Welfare reform that made benefits conditional on job market attachment succeeded in reducing childhood poverty and dependency and increasing workforce engagement by single mothers with children.

It’s tempting to believe that a similar approach to SSDI reforms may also reduce enrollments and dependency among the disabled. Whether such a result materializes, however, may also depend on whether macroeconomic conditions will match the robust conditions of the 1990s (in the United States as well as in the Netherlands) soon after a disability program reform is enacted; whether ancillary supporting policies to make work pay are enhanced—such as the Earned Income Credit; and whether accelerating globalization does not erode labor demand for jobs that disabled workers could perform in the United States. The creation of a positive job-market environment—which appears to be lacking currently—appears to be crucial to the success of disability system reforms.

Conclusion

The relative quiescence in the U.S. Congress about the need for reforming SSDI—which is imminently facing insolvency—is troubling. A temporary financial fix by shifting funds from the OASI trust fund is available but does not appear to be desirable. Apart from the prospect of exhausting its trust fund by 2016, SSDI suffers from several structural flaws, primarily its strong disincentives to workers with disabilities to remain employed. The anti-work and pro-SSDI-dependency gravitational effect of the program has resulted in a ballooning of SSDI enrollees since the mid-1980s. That ballooning has accelerated in the aftermath of the 2007–2009 recession, as unemployed workers with disabilities sought SSDI enrollment once their unemployment benefits ran out.

SSDI’s enrollment boom can be traced to liberalized eligibility polices introduced during the mid-1980s when benefit awards began to be based on subjective judgments about the severity of health conditions such as back pain and mental disability. Other contributory factors are the incentive to enroll in SSDI rather than collect OASI early retirement benefits because the former are not reduced for early retirement, and the bureaucratic use of outdated descriptions of available jobs in the U.S. economy.

The long and arduous process of qualifying for SSDI worsens applicants’ disabilities and resources, and saps physical and psychological capacities to return to work. Hence, SSDI reforms should prioritize a reduction in future SSDI enrollments rather than induce the existing population of enrollees to leave SSDI rolls. Doing so implies a crucial role for employers in early apprehension of the onset of disabling conditions in the population and extending timely accommodation and support services to affected workers. International experience with disability reforms also suggests tying tighter benefit and eligibility rules with encouraging labor-market attachments among the disabled. It also suggests providing work accommodations and support services to improve the environment within which health impairments can be prevented from becoming work-disabilities. Success of such approaches, however, will require ancillary policies to improve the macroeconomic environment and increase the likelihood of job attachments for people with disabilities.

The imminent financial crunch on SSDI and that program’s structural shortcomings urgently call for reforms to prevent skyrocketing enrollments by marginally health-impaired but not work-disabled individuals. Postponing such policy changes by adopting temporary financial transfers from OASI will accelerate broader insolvency of Social Security. It may also compromise the goal of providing needed support for the genuinely permanently disabled.

Readings

• “Does Disability Insurance Receipt Discourage Work?” by Nicole Maestas, Kathleen J. Mullen, and Alexander Strand. RAND Center for the Study of Aging Working Paper No 853.3, June 2012.
• “Social Support Substitution and the Earnings Rebound: Evidence from a Regression Discontinuity in Disability Insurance Reform,” by Lex Borghans, Anne C. Gielen, and Erzo F.P. Lutter. National Bureau of Economic Research Working Paper No. 18261, July 2012.
• “Supporting Work: A Proposal for Modernizing the U.S. Disability Insurance System,” by David H. Autor and Mark Duggan. Hamilton Project and Center for American Progress, December 2010.
• “The Declining Work and Welfare Among People With Disabilities: What Went Wrong and a Strategy for Change,” by Richard V. Burkhauser and Mary Daly. AEI Press, 2011.
• “The Effect of Disability Insurance Receipt on Labor Supply: A Dynamic Analysis,” by Eric French and Jae Song. Federal Reserve Bank of Chicago Working Paper No. WP 2012-12, 2012.