Providing the necessary information to ensure sound, evidence-based public policy for those with disabilities has been more difficult than for other economically vulnerable populations. To do so, researchers need appropriate data: to identify the population with disabilities, to establish success measures for the public policies targeted on it, and to track outcomes of such policies. The U.S. Bureau of the Census, for instance, has developed criteria based on questions in the Current Population Survey (CPS) to measure the official employment rate, average household income, and poverty rates of the U.S. population by age, sex, race/ethnicity, and education.
The CPS has also been used to track the size, economic success, and government program participation of especially vulnerable populations, such as single mothers. But controversy over the conceptual and operational definition of the working-age population with disabilities has prevented any official counts of this population, its economic vulnerability, and employment outcomes. The failure of the federal government to establish an official measure of the economic outcomes of working-age people with disabilities may to some degree explain why their economic fortunes have languished while the fortunes of single mothers have greatly improved over the last two decades.
Both of these economically vulnerable populations had for many years been treated by policymakers as if they were “not expected to work.” While this meant that they were the recipients of substantial federal funds, usually these funds were conditional on their not working. More important, since policymakers did not expect them to work, little was done to assist them to successfully compete in the labor market. Expectations changed for both groups in the 1990s.
The passage of the Welfare Reforms of 1996, along with a more generous Earned Income Tax Credit, increases in subsidized child care, and better access to subsidized medical insurance and employment services, together with the devolution of greater responsibility for these programs to the states, were all based on the notion that single mothers, like other women, even those with children, were able to work in the marketplace and should not be discouraged from doing so. This change in public policy had a major effect on the behavior of single mothers. It increased their willingness to seek employment by consistently shifting program incentives to favor work over welfare benefits conditioned on not working. The consequences of these policy changes on the employment, income, poverty, and program participation outcomes of this vulnerable population were captured in the CPS and other national data sets. These data have been used to document the major success of these policies as well as what remains to be done to fully integrate single mothers into the labor market and out of poverty.
The Americans with Disabilities Act of 1990 marked a major change in disability policy away from the medical model of disability that stressed the role of a person’s impairment on an “inability to work” to a model that better recognized that the social environment, including the economic, legal, and public policy environment, could have as much to do as the impairment itself with whether a person with a disability was employed. At the insistence of advocates and others, the government launched a multifaceted effort to more fully integrate working-age people with disabilities into society. While the passage of the ADA epitomizes this change, it is also reflected in other legislation, such as the 1998 Individuals with Disabilities Education Act, the 1998 Workforce Investment Act, and the 1999 Ticket to Work and Work Incentive Improvement Act. These pro-work policy initiatives focus on increasing investment in the market skills of individuals with disabilities, breaking down institutional and physical barriers to their use.
Unfortunately, the employment, economic well-being, and poverty rates of working-age people with disabilities appear not to have improved despite these changes in policy, although the evidence is controversial. The controversy is on two levels: quality of the data and interpretation of the data. With respect to the first, data quality, neither the U.S. Bureau of the Census nor any other branch of government has established an official measure of the economic success of this population.
Unlike age, sex, and to a lesser extent race/ethnicity, there is no well-established operational measure of the working-age population with disabilities. However, over the past two decades, a growing consensus has concluded that the old medical-based definition of disability is not satisfactory, that is, a model of disability that assumes that an inability to work is a direct consequence of a medically established pathology. For example, exceeding the medical listings of severity for a given diagnostic group (e.g., infectious and parasitic diseases, neoplasms, mental disorders) is still used by the Social Security Administration as sufficient cause for receiving disability benefits in the absence of work. However, there are people who have such conditions, do not apply for benefits, and continue to work. Hence, while disability begins with a pathology that causes a physical or mental impairment that subsequently limits one or more life activities, such as work, the social environments impaired people face can have as much to do with their work outcomes as does the impairment itself. For example, persons who experience deafness but are accommodated at the workplace with a TTY machine that permits them to use the telephone would not be considered work limited despite their impairment.
Unfortunately, no existing data sets fully capture this more general conceptualization of a work disability. Recent and still controversial research has, however, established that the CPS data, while imperfect, can be used to capture a work-limitation-based measure of the working-age population with disabilities and track its economic success both absolutely and relative to the rest of the working-age population with respect to employment, income and poverty outcomes. The results of this research suggest that in contrast to most vulnerable populations, especially single mothers, people with disabilities have not shared in the fruits of economic growth over the last 25 years, especially during the 1990s.
We now know that the United States experienced two major business cycles over the last two decades of the twentieth century: 1979 through 1989 and 1989 through 2000. Based on CPS data and confirmed with National Health Interview Survey and Survey of Income and Program Participation data, the employment rates of working-age people with and without disabilities were pro-cyclical over the first business cycle, falling over the recession years of the early 1980s and rising over the long period of growth from 1982 to 1989. The rise in employment for those with disabilities, however, was not as great as for those without disabilities, so over the decade, the employment rate of the former fell relative to that of the latter. The case was the same for single mothers.
The relative employment rate of working-age people with disabilities fell even more over the 1990s business cycle. Although employment of those both with and without disabilities fell from the business cycle peak year of 1990 to 1993, during the growth years 1993 to 2000, employment of those without disabilities rose, while employment of those with disabilities continued to fall. The employment rates of those both with and without disabilities fell between 2000 and 2003, but the relative employment rate of those with disabilities continued to fall.
In contrast to working-age people with disabilities, the employment of single mothers grew dramatically over the 1990s, especially after the implementation of the Welfare Reforms of 1996. By the end of the decade, the employment of single mothers was, for the first time, higher than that of married women with children.
Between 1979 and 2000, the household income of working-age people with disabilities remained about the same. Their wage-earning declines were offset by increases in their disability transfers, Social Security Disability Insurance (SSDI), and Supplemental Security Income (SSI) and by the greater earnings of other household members. Their overall household income, however, fell relative to those without disabilities, whose income rose significantly over this period of substantial economic growth. In contrast, the income of single mothers, which fell relative to married mothers in the 1980s, grew both absolutely and relative to married mothers over the 1990s, especially after welfare reform. The overall increase in the average income of single mothers rose because their labor earnings and the government subsidies they received based on their work more than made up for the decline in their welfare benefits.
The ability of the average working-age persons with disabilities to at least maintain the same level of economic well-being while the income of the rest of the population was increasing, however, obscures what happened to the most vulnerable part of the working-age population with disabilities. This group has consistently faced a greater risk of poverty than working-age people without disabilities. In 1979, a working-age person with a disability was 3.2 times more likely than one without a disability to be living in poverty. By 1989, their relative risk had risen to 3.5, and by 2000 to 4.1. Over these two business cycles, the poverty rate of working-age people with disabilities rose from 21.9 percent to 28.1 percent, while the poverty rate of those without disabilities remained about the same, 6.9 to 6.8 percent. In contrast, the poverty rate of single mothers rose during the 1980s but fell both absolutely and relative to married mothers in the 1990s, especially after welfare reform.
But what accounts for the failure of working-age people with disabilities to be more fully integrated into mainstream society? Although there has been controversy over the measures of economic success reported above based on CPS data, the controversy surrounding the interpretation of these data has been greater. To understand what is driving these outcomes, it is first necessary to recognize that most working-age people with disabilities experience the onset of their work limitations after they enter the labor market and remain in the labor force long after this initial onset of a work disability. Research shows that the average length of time before individuals apply for SSDI benefits following onset is 15 years and that only 72 percent of workers with disabilities will ever apply for SSDI benefits. The rest continue in the labor force until they exit onto the Social Security retirement program. The timing of SSDI application is affected (holding the seriousness of the impairment constant) by a worker’s characteristics and the economic environment he or she faces. Younger and better-educated workers will delay application longer than older or less educated workers. But accommodation matters, and employees who are accommodated by their employers stay in the labor force longer than those who aren’t. This is true across a wide range of accommodations. The most frequent firm accommodations, involving changes in job hours and job responsibilities, are not costly from an accounting perspective, although they may be from a work organization perspective.
Economic incentives also matter. Workers whose future earning streams are higher or who are less likely to be accepted for SSDI will delay applying for benefits and continue in the labor force longer than workers with lower future earning streams and higher likelihood of SSDI acceptance.
The employment experience of working-age people with disabilities is discouraging, especially since the passage of the Americans with Disabilities Act of 1990 was meant to usher in a new generation of pro-work policies that would more fully integrate working-age people with disabilities into the mainstream of American life, including employment. Despite the increased duration in employment that accommodation brings and the likely rise in the percentage of workers with disabilities who were accommodated in the 1990s following the implementation of the ADA, their employment rates fell over the entire 1990s business cycle and has done so, relative to those without disabilities, since the mid-1980s. The causes of this decline in relative employment are controversial, but the preponderance of the evidence suggests that it was not caused by an increase in the severity of workers’ impairments; and neither changes in the nature of work nor the availability of employer-provided health insurance were major factors.
The most likely causes of the decline in relative employment were the substantial reduction in the eligibility standards for entry onto the SSDI rolls that occurred in the mid-1980s and the increase in benefits relative to market wages. The unintended consequence of these well-meaning but anti-work policies was to discourage working-age people with disabilities from taking advantage of the strengthening of their legal right to work.
It has also been argued that the increased costs firms must bear to accommodate workers with disabilities have made them less likely to hire such workers and that this unintended consequence of the ADA offsets its positive impact via increased duration on the job for those who are accommodated. The evidence of a negative ADA effect on employment is more controversial and is sensitive to alternative definitions of the population with disabilities.
What is clear is that, as with those without disabilities, the employment of working-age people with disabilities is influenced by market forces. The skills that people with disabilities bring to the job will affect their chances of employment, their duration on the job, and the age at which they leave the labor market. The wage premium that those with more than a high school education can command has dramatically increased over the last two decades. This is true for those with and without disabilities. The best predictor of job market success is one’s level of education; in the constantly changing American job market, the long-term rewards to well-educated workers demonstrate its value. Hence, those workers with good job skills and high education levels are the most likely to be retained by their employers during bad times and to command higher salaries during both good and bad times.
Government policy can also influence the employment of working-age people with disabilities. Historically, the federal government’s approach to providing economic security for working-age people with disabilities has been dominated by a caretaker approach, reflecting the outdated view that disability is solely a medical issue.
A glance at the federal budget shows that cash transfers and medical care expenditures (SSDI, SSI, Medicare, and Medicaid) dwarf expenditures for efforts to integrate those with disabilities into the labor force. And the eligibility rules for these programs discourage work and more than likely offset any pro-work effect of the ADA during the 1990s.
The dramatic improvement in the employment, average income, and poverty rates of single mothers after the Welfare Reforms of 1996 show that it is possible for government policy to shift from a caretaker to a pro-work priority and to have a significant impact on a population that was “not expected to work.” To successfully integrate working-age people with disabilities into the workforce, however, it will be necessary to first establish an official definition of the working-age population with disabilities based on a currently available data set like the CPS and then use it to monitor the success of policy changes that are consistent with providing support for work, making work pay, and expecting people with disabilities to do what they reasonably can to support themselves. The recent experience of single mothers suggests that reforms with these three features can both improve the lives of many people with disabilities and achieve broad-based political support.
- Blank, R. M. 2002. “Evaluating Welfare Reform,” Journal of Economic Literature 40:1105-1166.
- Bound, J. and Burkhauser, R. V. 1999. “Economic Analysis of Transfer Programs Targeted on People with Disabilities.” Pp. 3417-3528 in Handbook of Labor Economics, 3c, edited by O. C. Ashenfelter and D. Card. Amsterdam, Netherlands: Elsevier Science.
- Burkhauser, R. V., Butler, J. S. and Gumas, G. 2003. “A Dynamic Programming Model of Social Security Disability Insurance Application,” Journal of Applied Econometrics 19:671-685.
- Jette, A. and Bradley, E. 2002. “Conceptual Issues in the Measurement of Work Disabilities.” Pp. 183-210 in The Dynamics of Disability: Measuring and Monitoring Disabilities for Social Security Programs, edited by G. Wonderlich, D. P. Rice and N. L. Amado. Washington DC: National Academy Press.
- Marin, B., Prinz, C. and Queisser, M. 2004. Transforming Disability Welfare Policies: Towards Work and Equal Opportunities. Burlington VT: Ashgate.
- Rupp, K. and Stapleton, D. C., eds. 1998. Growth in Disability Benefits: Explanations and Policy Implications. Kalamazoo, MI: W. E. UpJohn Institute for Employment Research.
- Stapleton, D. C. and Burkhauser, R. V., eds. 2003. The Decline in Employment of People with Disabilities: A Policy Puzzle. Kalamazoo, MI: W. E. UpJohn Institute for Employment Research.