A Labor Force Profile of Persons with Disabilities - Executive Summary


Alberto Martini

Mathematica Policy Research, Inc.

This report was prepared under contract #HHS-88-0047 between the U.S. Department of Health and Human Services (HHS) and Mathematica Policy Research. SysteMetrics/McGraw-Hill was a subcontractor for the project. For additional information about the study, you may visit the DALTCP home page at http://aspe.hhs.gov/daltcp/home.htm or contact the office at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, SW, Washington, DC 20201. The e-mail address is: webmaster.DALTCP@hhs.gov. The DALTCP Project Officer was Michele Adler.

This report uses data from the Survey of Income and Program Participation (SIPP) to explore the labor market situation of persons with disabilities. During the last decade, the employment of disabled persons has come to the forefront of Federal disability policy. A new consensus has emerged that the problems of disabled persons cannot be addressed exclusively in terms of income support policies, and increased attention has been given to the promotion of civil rights, with the goal of removing discrimination and other barriers to employment. At the same time, changes have been introduced into the system of transfer and social insurance programs in order to reduce the disincentives to return to gainful employment on the part of the recipients of disability benefits.

This study is part of a series of four reports prepared for the Department of Health and Human Services in response to the growing demand for information on the population of persons with disabilities. Other reports in the series include the following:

  • A profile which examines the prevalence of disabilities among the U.S. population, and the demographic and economic characteristics of disabled persons.

  • A profile of the rules and benefit structures of the Federal programs which serve disabled persons.

  • A profile of the participation of disabled persons in Federal income transfer programs.



Researchers in this field have come to recognize that there is no single concept of disability appropriate for all concerns and policies issues. In this report, we adopt the three classifications of disability developed in the first report of the series, which distinguish between limitations in functioning, limitations in working ability, and receipt of disability benefits.

Repeated use is made in this report of a five-level scale of limitations in functioning which is based on the presence of sensory and functional limitations that restrict the individual's ability to perform age-appropriate physical activities. The five levels range from more severe to less severe limitations: (1) need for another person's assistance with daily life tasks; (2) inability to perform one or more sensory or physical functions; (3) experiencing difficulties in multiple sensory or physical functions; (4) experiencing difficulty in just one function; and (5) no limitations in functioning. This five-level scale is simplified at times to a two-level scale, based on the presence of substantial limitations in functioning (SLF), defined by aggregating the first three levels of the scale described above.

The report makes use of a second multi-level scale based on limitations in working ability. This scale captures the inability to work, limitations in the amount of work, such as the inability to work full time or regularly, and limitations in the type of work, such as the inability to work in the same occupation as before the onset of the limitation.

Under the third, more restrictive classification, only persons of working age who receive benefits for reason of disability from Supplemental Security Income, Social Security Disability Insurance, or Veteran Administration programs, are counted as disabled.



Four major issues pertaining to the labor market situation of persons with disabilities are addressed. For ease of presentation, the findings reported below pertain to the limitations in functioning classification only.

  1. To what extent do persons with disabilities participate in the labor market? How does their participation vary by the level of disability, and how does it compare to that of persons without disabilities?

    • Data collected at one point in time (April 1984) show that the employment rate is low among persons of working age who report the most severe levels of limitations in functioning, and rises steadily with less severe levels of limitations. The employment rate is 21 percent among persons who need the assistance of another person with daily life tasks; 33 percent among persons who report inabilities in one or more sensory or physical functions; 44 percent among those who report multiple difficulties in these functions, but not inabilities; 64 percent among those reporting only one such difficulty; and 74 percent among persons with no limitations in functioning.

    • Labor force data that span the entire 1984 calendar year show that the percentage of persons employed at some point during the year by level of limitations follows a pattern similar to the employment rate in April, ranging from 29 percent among persons needing assistance with daily life tasks to 85 percent among persons with no limitations. However, annual weeks and hours worked by those employed during the year vary to a lesser extent. Persons needing assistance work 39 weeks on average, and approximately 1,360 hours, while persons with no limitations work 45 weeks on average, for a total of 1,800 hours.

    • The presence of less severe limitations in functioning, such as the difficulty in only one sensory or physical function, appears to have a small impact on employment. This suggests that the individuals in this group should not be considered "disabled" from the perspective of employment policy. In addition, we find that over 10 million working-age persons are classified in this category, representing almost half of all persons reporting any limitation in functioning.

    • When combined, these two pieces of information have implications for the possible impact of the Americans with Disabilities Act (ADA) of 1989, which is intended to establish a comprehensive prohibition on discrimination on the basis of disability. In fact, estimates quoted in the premise of the Act imply that there are 21.5 million disabled adults of working age. The evidence from SIPP suggests that this figure, and thus the number of persons potentially affected by the ADA, might have been substantially overestimated.

  2. What is the extent of unemployment among persons with disabilities? Also, does the official definition of unemployment capture the reality of lack of employment opportunities among persons with disabilities?

    • Persons with SLF are significantly more likely to be unemployed when in the labor force than persons with no limitations. The unemployment rate among those with SLF is approximately 12 percent (April 1984) versus 7 percent for persons with no limitations. Also, the incidence of unemployment over the course of a year varies with severity of limitation. Over 30 percent of persons with SLF experience some unemployment during the year, versus less than 20 percent of persons with no substantial limitations.

    • For every person with SLF who is reported as unemployed, there is another person with SLF not looking for work who nevertheless wants to work and would accept a job if one were offered. Among persons with no limitations in functioning these proportions are quite different; for every three unemployed individuals there is only one person who wants a job but is not seeking work. Therefore, the official definition of unemployment, based on active search for a job, "misses" almost half of the disabled persons who want to work.

    • SIPP data indicate that, in April 1984, there were about 2 million persons with any limitation in functioning who reported they wanted to work but did not have a job. This number includes persons actively seeking work (1.1 million), as well as persons who wanted a job but were not actively seeking one (0.9 million). Previous research has estimated that 8 million disabled persons want a job but do not have one.

  3. How do earnings differ between workers with and without disabilities? How do wages differ?

    • Annual earnings and hourly wages do not decrease steadily with the severity of the limitation, as did the extent of employment. In fact, the most severely impaired group, persons who need assistance, report significantly higher earnings and wages on average than all other persons with substantial limitations. The lowest earnings and wages are reported by workers with intermediate levels of limitations in functioning.

    • A possible explanation for the result cited above is that persons needing assistance who are employed have higher levels of education, are younger, and are more likely to be male than other workers with limitations in functioning. The necessity to overcome the difficulties connected with severe impairments might trigger a self-selection process, so that working persons with severe impairments have a demographic composition substantially different from that of the overall disabled population. This different composition might lead to higher average wages for this group than for workers with less severe impairments.

    • The statistical effect of this self-selection process is largely eliminated when a regression analysis of the relationship between limitations in functioning and wages is performed. When demographic characteristics are held constant, the previously observed U-shaped pattern of wages disappears. In fact, the regression-adjusted wages of workers with severe limitations are lower than the corresponding measures for workers with less severe limitations. This implies that the net effect of impairments on wages is underestimated when simple sample averages are used.

  4. Are there differences in health insurance coverage between workers with and without disabilities, and between disabled persons with and without jobs?

    • The percentage of workers covered by employer-provided health insurance exhibits relatively little variation by level of limitations in functioning, ranging from 55 percent among workers needing assistance to 61 among workers with no limitations. On the other hand, coverage by government insurance programs only (Medicare or Medicaid) ranges from 5.2 percent among workers needing assistance to 0.7 percent among workers with no limitations. Because of the combined effect of a low rate of employer provided insurance (57 percent) and a low rate of coverage by government insurance programs (2.3 percent), workers with inabilities or multiple difficulties have the highest level of non-coverage, approximately 15 percent. Almost 9 percent of workers needing assistance and 12 of those with no limitations are not covered by any health insurance.

    • The extent of employer-provided health insurance varies by full-time work status. For full-time workers, the overall coverage from employer-provided insurance is over 70 percent; among them, the lowest rate is again among persons needing assistance (63 percent). The overall picture changes dramatically when we examine persons employed part time. The overall rate of employer-provided coverage is a meager 20 percent, but, surprisingly, persons needing assistance have the highest rate (34 percent), and persons with no limitations have the lowest rate (19 percent).

    • When not employed, 53 percent of persons needing assistance are covered by some combination of government insurance. The corresponding figure for persons with no limitations is 13 percent. In contrast, 50 percent of persons with no limitations are covered by an insurance plan in somebody else's name, while only 19 percent of persons needing assistance are. The rate of noncoverage is, once again, lower for persons needing assistance (14 percent) than for any other group.