Skip to main content

Social Security Administration Proposal To Revise Disability Determinations Is Not Justified

The Social Security Administration (SSA) is proposing to change how it evaluates age as a factor in establishing eligibility for disability benefits. Under the change, SSA would raise by two years the ages at which key rules used to determine eligibility for Social Security disability insurance (SSDI) or Supplemental Security Income (SSI) are applied.[1] SSA estimates that the proposal would eliminate benefits for some workers aged 45 and above with certain kinds of disabilities (as well as low education and few work skills) by close to $6 billion over ten years.[2]

In justifying this proposal, SSA has noted that older people are living longer than they did in the past, on average, and many are able to work longer. Studies finding that Americans on average are healthier at a given age than Americans were a few decades ago are not relevant, however, to those individuals who are not healthy at these ages.

The SSA proposal does not appear to be justified by recent data on employment trends of people with disabilities or by data on the skills currently demanded by employers in the national economy, both of which suggest that employment opportunities for many people with disabilities — particularly those of advanced age and with limited education and skills — remain very limited. Moreover, the proposed change would have a highly disproportionate adverse impact on African Americans, who are more likely than other people to have disabilities and lower education and skill levels.

Given this evidence of likely and disproportionate harm, SSA should withdraw the proposal. It should focus instead on developing research to learn more about the ability of older workers with disabilities who have little education and few skills to perform substantial gainful activity on a sustained basis, which is the standard established by the Social Security Act. SSA also should collect and publish data that shows, by race and age of onset of disability, the impact that such a change will have upon various subcategories of the population with disabilities.

Proposed Change Would Raise Eligibility Age for Disability Benefits

The SSA definition of disability is rigorous: applicants for disability benefits must show that they have a severe health condition that prevents them not only from doing their prior job but also from doing any other substantial gainful activity, given their age, education, and work experience.[3]

Because age is a factor in the definition of disability, SSA concluded that in evaluating eligibility for disability benefits, special emphasis should be given to people at higher ages. Accordingly, in 1978, SSA issued regulatory guidelines setting forth specific ages at which individuals with certain types of disabilities who also have low levels of education or skills are considered unable to find other jobs that they can perform, and hence are considered to be eligible for disability benefits. Many older individuals with disabilities and limited education and skills are unable to meet even modest job requirements, and the SSA regulations protect these individuals by enabling them to qualify for disability benefits.

The current SSA guidelines include a grid that specifies the ages at which people with disabilities who have various education and skill levels can qualify for benefits. For example, a person with a severe impairment who has less than a twelfth-grade education, cannot do his or her previous work and has no transferable job skills (or has no prior work experience) can be classified as disabled starting at age 50. If the same individual is aged 45-49, however, he or she would not be classified as disabled. SSA rules regard being younger than 50 as advantageous to securing employment.[4]

The SSA proposal would change these rules by adding two years to each age requirement. In other words, the individuals in the example above would be found not to be disabled until they reached age 52, rather than age 50.[5]

Obtaining Employment Is Not Becoming Easier for People with Disabilities Who are Older and Have Limited Education and Skills

In its proposal to change these rules, SSA states that, “[n]ot only are Americans living longer, but there is clear and overwhelming evidence that the average health of the elderly population is improving” and that “[t]his increase in healthy, active years has already translated into a shift among older adults who are working past 65.”[6] The key question, however, is whether those Americans who would be affected by SSA’s proposed changes — those who have disabilities and limited education and skills, and are aged 45-61 — are more able to find work today than in the past.

Available evidence on this question is limited. But the evidence that does exist suggests there is no reason to think that employment opportunities have grown significantly for people with work-limiting disabilities who are older and have limited education and skills.

One of the relevant sources of data is the Census Bureau’s Current Population Survey (CPS), which since 1988 has asked Americans whether they have a “health problem or disability which prevents [them] from working or which limits the kind or amount of work [they] can do.” This question is not designed to identify either the narrow range of individuals with severe disabilities who would be affected by SSA’s proposal or the broader population with work disabilities. The Census Bureau also cautions that the question has not been validated as an accurate measure of work disability.[7] The CPS data do suggest, however, that employment is not increasing among people who respond yes to this question.

Indeed, if the CPS data suggest anything, it is that employment may have become more difficult to find for people who have such self-identified work-limiting disabilities and are 45 or older. The following analysis examines the employment status of people with self-reported disabilities aged 45 through 61.[8]

  • Between March 1988 and March 2005 (the first and last months for which these data are available), the share of people aged 45-61 with such disabilities who were working declined from 22 percent to 17 percent, even as the employment rate for all adults — as well as for people aged 45-61 without disabilities — rose slightly.[9]
  • If one compares the most recent peak years of the last two economic cycles, the findings show little change or perhaps a slight decline. Between March 1990 and March 2001, the share of people aged 45-61 with disabilities who were working edged downward from 23 percent to about 21 percent,[10] even as the employment rates for all adults and for people aged 45-61 without disabilities increased.
  • Among people aged 45-61 with disabilities who have no college education, a group particularly likely to be affected by SSA’s proposal, the employment rate declined from 19 percent in March 1990 to 16 percent in March 2001.[11]

Because these and other survey data on disability status suffer from technical limitations, these employment trends are subject to varying interpretations and should be understood only as identifying important questions that require further research by SSA, if SSA wishes to reconsider its current rules on the intersection of disability, age, and employability. As one study notes, “some [researchers] believe current work limitation data cannot be used to determine the overall employment rate for persons with disabilities, others believe the overall rate did not change during the 1990s, and still others believe that the overall employment rate of persons with disabilities decreased during this period.”[12] None of these interpretations, however, suggests that employment prospects improved for people with work disabilities who are age 45 and older and have limited education and skills.In other words, available research either says little about employment prospects for people with disabilities who are age 45 and older and have limited education and skills or suggests that no improvement has occurred. The data provide no evidence that employment prospects for persons with disabilities in this age group who have limited education and skills are getting better, and certainly not to a degree that justifies increasing the age factors in the disability evaluation process, as SSA has proposed.

African Americans with Disabilities Would Be Disproportionately Affected

African Americans make up a disproportionate share of the people who would be adversely affected by the SSA proposal: African Americans constitute 22 percent of Americans aged 45-61 who are receiving SSI or Social Security disability payments, although they constitute only 11 percent of all Americans in this age group, according to a Center analysis of the March 2004 Current Population Survey.[13] (These estimates closely match SSA figures based on pairing Social Security Disability Insurance records with Census survey data.[14]) This analysis uses Census data because the Social Security Administration does not collect data related to race that allow it to answer the question of how the changes SSA is proposing will affect people of different races.The relatively large share of black beneficiaries in disability programs reflects African Americans’ higher rate of disability. Twenty percent of African Americans aged 45-61 reported having a work-limiting disability in March 2004, compared with 11 percent of other Americans in this age range.[15] The high proportion of African Americans with disabilities, in turn, reflects a combination of factors. African Americans are more likely to be employed in physically demanding jobs. They also are more likely to have a history of less adequate health care, greater poverty, and more stressful living conditions.

Not only do African Americans have a higher rate of disability than other groups, but African Americans with work-limiting disabilities also have lower employment rates than the overall population with such disabilities. The CPS data show that:

  • Eleven percent of African Americans aged 45-61 with self-reported work-limiting disabilities were employed in March 2005, less than two-thirds the 17 percent rate for all Americans aged 45-61 with such disabilities.
  • The share of African Americans aged 45-61 with disabilities who were employed did not improve during the course of the last full economic cycle. It stood at 15.0 percent in March 1990 and 14.7 percent in March 2001.
  • For less-educated blacks with disabilities in this age range, employment rates are especially low. Currently, just 8 percent of blacks aged 45-61 who report work-limiting disabilities and do not have a college degree are employed.[16] (In this context, it is worth noting that African Americans with disabilities are more likely than other races to have low education levels. Sixty-seven percent of African Americans aged 45-61 with work-limiting disabilities lack any college education, compared with 62 percent for all races. Thirty percent of African Americans aged 45-61 with work-limiting disabilities lack a high school diploma, compared with 26 percent for all races.)

Given the fact that such a small share of African Americans with work-limiting disabilities in this age category have jobs, the SSA proposal to scale back disability benefits by revising eligibility rules would cause significant hardship among African Americans.

Very Few Job Opportunities Exist for Those with Disabilities and Limited Skills

A main reason for the low employment rates among individuals aged 45-61 with disabilities and limited education and skills appears to be that employers offer few jobs they can perform. Recent Labor Department efforts to quantify the minimal requirements for specific occupations make clear that, for people who have both physical impairments and limited cognitive skills, job options are scarce.[17] For example, of 336 occupations for which the Department has rated the requirements:

  • Only 5 occupations require minimal stamina (only enough to walk one-fourth of a mile) and no more than moderate reading skills (enough to read a memo from management describing new personnel policies).[18] The other 331 occupations require either more stamina, higher reading skills, or both.
  • Only 23 occupations require minimal stamina and no more than moderate speaking skills (enough to interview applicants to obtain personal and work histories).[19] The other 313 occupations require more stamina, higher speaking skills, or both.
  • Only 7 occupations require minimal manual dexterity (only enough to screw a light bulb into a light socket) and no more than moderate speaking skills.[20] The other 329 occupations require more dexterity, higher speaking skills, or both.
  • Only 1 occupation — “shuttle car operators”[21] — requires minimal trunk strength (only enough to sit up in an office chair) and minimal reading skills (only enough to read step-by-step instructions for completing a form).[22] The other 335 occupations require more trunk strength, higher reading skills, or both.

Many individuals aged 50 or older with disabilities and limited education and skills are unable to meet even these modest job requirements. In recognition of this fact, SSA’s current regulations protect these individuals by enabling them to qualify for disability benefits. For example, individuals who have severe disabilities, cannot do their prior work, have worked only in unskilled jobs or have no prior work experience, and have less than a twelfth-grade education currently can qualify for benefits when they reach age 50. Under SSA’s new proposal, individuals in this group — even those who cannot read or write — would not qualify for disability benefits until they reached age 52.Similarly, SSA has a special rule for people age 45 to 49 with a severe disability who have done only unskilled work and can no longer perform their past work, are now restricted to sedentary work, and are unable to communicate in English. In such cases, SSA will find an individual in this age range to be disabled.[23] Under the SSA proposal, such individuals would be classified as not disabled and be ineligible for disability benefits until they reached age 47.


SSA’s proposal would deny needed disability benefits to many individuals who have little if any chance of finding employment because of their severe disabilities, and limited education and skills. African Americans would be especially hard hit, since they are more likely than other groups to have disabilities and lower levels of education and since African Americans with disabilities in this age category are less likely than other individuals with disabilities to be employed.

SSA has defended its proposal on the basis of the overall improvement in Americans’ health over the past few decades. But that overall improvement is not directly relevant to the matter at hand. The available data suggest that individuals who have disabilities — the people affected by the proposal — are not finding it easier to obtain work and that the number of jobs available to them remains very small.

In short, SSA’s proposal is not supported by the available data and research. As a result, SSA should withdraw the proposal and conduct research on this matter to determine whether changes should be made in this area, and if so, what those changes should be.

End Notes

[1] See SSA notice of proposed rulemaking (NRPM) on age as a factor in evaluating disability, 70 Federal Register 67101-67109 (November 4, 2005).

[2] SSA expects that the proposal would save $5.834 billion from 2006 through 2015, $1.871 billion of which would come in the first five years (2006-2010). The bulk of the savings would come in Social Security ($3.231 billion over ten years), but large savings also would come in SSI ($834 million), Medicare ($1.034 billion), and Medicaid ($735 million). Id. at 67107.

[3] To be found disabled and eligible to receive Social Security disability benefits, a person must be unable “to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months.” Section 223(d)(1)(A) of the Social Security Act, 42 U.S.C. §423(d)(1)(A). “An individual shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy, regardless of whether such work exists in the immediate area in which he lives, or whether a specific job vacancy exists for him, or whether he would be hired if he applied for work. For purposes of the preceding sentence (with respect to any individual), ‘work which exists in the national economy’ means work which exists in significant numbers either in the region where such individual lives or in several regions of the country.” Section 223(d)(2)(A) of the Social Security Act, 42 U.S.C. §423(d)(2)(A). The SSI disability provisions basically parallel this language. See Sections 1614(a)(3)(A) and (B) of the Social Security Act, 42 U.S.C. §§1382c(a)(3)(A) and (B).

[4] SSA generally views being age 49 or younger as advantageous in securing employment. (Rules 201.09, 201.10, and 201.18) There is one exception to this rule. If a person is aged 45-49, has a severe medical impairment, cannot perform his or her past work, is restricted to sedentary work, has done only unskilled work, and is unable to communicate in English (or is able to speak and understand English but is not able to read or write English), SSA will find him or her disabled. Under the proposed regulation raising the age categories by two years, this age category would be changed so that it applied to individuals aged 47-51. (Rules 200.00(h)(1), (2) and Rule 201.17) To capture this group, the employment analyses in this paper use data beginning at age 45; however, the impact of SSA’s proposal is likely to fall most heavily on people aged 50 and older.

[5] If SSA increases the age categories by two years, as it proposes, people aged 50-51 who now qualify for disability benefits under Rule 201.09 would instead be denied benefits under Rule 201.18, which would apply to applicants aged 47-51 instead of those aged 45-49.

[6] 70 Fed. Reg. 67102.

[7] U.S. Census Bureau, Uses and Limitations of CPS Data on Work Disability (downloaded March 21, 2006),

[8] This paper analyzes statistics for people ages 45 through 61, rather than through the “normal retirement age,” because SSA’s proposed regulations do not change the rules under which people between the age of 62 and the “normal retirement age” will be evaluated. (In the SSI program, a person can be determined to be disabled at any time prior to age 65. After that date, the person receives benefits as an aged individual. In the Social Security Disability Insurance program, a person can be determined to be disabled at any time up until the normal retirement age. That age used to be set at 65. As a result of Social Security legislation enacted in 1983, the normal retirement age is gradually being raised for people who were born in 1938 or later. The normal retirement age will be set at age 67 for people born after 1959.)

[9] From March 1988 to March 2005, the employment rate of all adults rose slightly, from 61.1 percent to 61.8 percent, while the employment rate of persons aged 45-61 without disabilities increased more substantially, from 69.8 percent to 74.3 percent.

[10] The March 2001 employment estimate is available from two different versions of the March 2001 CPS. One estimate shows an employment rate of 21.1 percent. Another estimate, based on an enlarged sample size and revised sample weights introduced by the Census Bureau in that year, shows an employment rate of 19.9 percent. The change between March 1990 and March 2001 is not statistically significant under the first estimate but is statistically significant under the second estimate. All other changes discussed in this section are statistically significant.

[11] A possible further complication in making these comparisons over time is that the employment estimates shown here rely on a CPS employment question that was changed in 1994. On closer inspection, however, we see no indication that the change in the survey question is driving the results shown here. For example, alternative comparisons for the same demographic group examined here, but based on a separate set of survey questions about annual work experience that were not changed, show a peak-to-peak decrease of close to 2 percentage points in the employment rate, similar to the decrease shown above. Likewise, in the monthly employment data used here, the proportion of people working more than 20 hours a week — a group unlikely to have been affected by the particular wording change in the CPS questionnaire — declined about 2 percentage points from March 1990 to March 2001.

[12] Robert Silverstein, George Julnes, and Renee Nolan, “What Policymakers Need and Must Demand from Research Regarding the Employment Rate of Persons with Disabilities,” Behavioral Sciences and the Law 23:399-488 (2005), page 427. The authors cite evidence that people with disabilities who are not in the labor force have become more likely to report their disability now than in the past.

[13] These tabulations exclude individuals receiving disability payments solely on behalf of children or other family members.

[14] Specifically, our CPS-based estimates show that African Americans make up 26 percent of those receiving SSI disability payments and 19 percent of those receiving Social Security disability payments. SSA’s estimates, based on administrative data, show that 19.7 percent of disabled workers receiving SSDI in 2001 were African American. See Social Security Administration, Annual Statistical Report on the Social Security Disability Insurance Program, 2004 (March 2006),, table 63.

[15] CBPP tabulations of the March 2004 Current Population Survey.

[16] Moreover, the employment rate among this group edged downward during the course of the last full economic cycle, from 13 percent in March 1990 (the peak of one business cycle) to 12 percent in March 2001 (the peak of the next business cycle).

[17] Data from the Labor Department’s Occupational Information Network (O*NET) are available at The authors understand that SSA does not view O*NET data as appropriate for determining disability under the Social Security Act. It is not being used here for that purpose, but rather to suggest that, contrary to the statements in the SSA proposal, few if any jobs may exist for individuals of more advanced age, limited education, and limitations resulting from their disabling conditions. In this context, the O*NET data appear to be the most appropriate, authoritative, and up-to-date data available.

[18] O*NET generally ranks job requirements on a scale from 0 (low) to 7 (high). “Minimal” stamina here refers to an ability level of 1 on the O*NET stamina scale. This is the lowest level for which O*NET documentation provides a concrete illustration of what the stamina scale means. “Moderate” reading skills refers to a reading skills level of 4, the second-lowest level for which a concrete illustration of the meaning of the reading scale is provided.

[19] “Minimal” stamina here refers to a level of 1 on the O*NET stamina scale. “Moderate” speaking skills refers to a level of 4, the second-lowest level for which a concrete illustration of the speaking scale is provided.

[20] “Minimal” dexterity here refers to a level of 1. “Moderate” speaking skills refers to a level of 4.

[21]The Labor Department says that shuttle car operators operate a diesel or electric-powered shuttle car in an underground mine to transport materials to the mine cars or conveyor. It should be noted that a degree of flexibility and dexterity is required for this occupation, as well as the ability to tolerate the dust and breathing conditions in a working mine.

[22] “Minimal” trunk strength refers to a level of 2, the lowest level for which O*NET documentation provides a concrete illustration of the meaning of the trunk strength scale. “Moderate” reading skills refers to a level of 4.

[23] See footnote 5.