The US Disability Explosion

Last week, the CBO released a set of projections for the future size of the US labor force.  A key graph was this one, showing the past and projected labor force participation rates for working-age men:

(Aside: I think their projections are a bit optimistic: I expect the ratios for the 25-54 groups to continue to decline as they have been for decades.)

In discussing this figure, they say:

In 2007, 90.9 percent of men ages 25 to 54 participated in the labor force, down half a percentage point from 2000 and more than 3 points below that group’s participation rate in the late 1970s. By 2010, participation had fallen even more, to 89.3 percent, as demand for labor weakened substantially and the male unemployment rate moved sharply higher. The long-term decline in part reflects a shifting age mix, as the number of 45- to 54-year-olds—the age group in which participation falls off as workers begin to retire—has grown as a share of the entire category. Nonetheless, the downward trend has occurred within each group among people between the ages of 25 and 54 (see Figure A-2), and it appears to be associated with worsening job opportunities for men with fewer work skills, particularly in manufacturing, and the interaction of those worsening opportunities with policies that have increased the availability of Disability Insurance benefits.²

Footnote 2 is this CBO report on disability insurance.  That report describes the program thus:

The DI program provides income to nonelderly adults— those younger than the full retirement age—who have worked in the past but who are deemed unable to work now because of a medical condition that is expected to last more than a year or to result in death.5 Disabled beneficiaries receive monthly payments based on their past earnings for as long as they remain in the program. Some family members of disabled beneficiaries, including certain spouses, minor children, and disabled adult children, are also eligible for benefits. If disabled beneficiaries remain disabled and live to the full retirement age, they then transfer to the Social Security retirement program.

To be eligible for the DI program, workers must have a sufficient record of work. Generally, people over the age of 30 must have worked during one-quarter of the years since they were 21 and during 5 of the past 10 years; younger workers are subject to slightly different rules. Eligibility for benefits also requires that workers have monthly earnings below a threshold known as the “substantial gainful activity amount”—currently $1,000 a month, or $1,640 for blind beneficiaries—for at least the past five months. The rules governing the program place no limit on the nonwage income or assets of DI beneficiaries.

The DI program is intended to provide income to people who can no longer perform substantial work because of a disability. But the dividing line between those who can and cannot perform such work is not always clear. Some people who are employed have medical conditions that would allow them to qualify for the program if they stopped working (for instance, people who are deaf ). By the same token, some disabled beneficiaries could probably work and would leave the DI rolls if they found a suitable job. The specific laws, regulations, and administrative procedures used to determine a worker’s medical eligibility for the program have a big effect on the number of DI beneficiaries, as do workers’ decisions about whether to seek DI benefits.

Typically, applicants are considered to be disabled if they have a condition that appears on SSA’s “listings of impairments” or if they are judged to be unable to perform substantial work for medical reasons that are not included in those listings.7 Initially, determining whether an applicant’s disability qualifies him or her for benefits is the job of the Disability Determination Services (DDSs), which are agencies funded by SSA but administered by the states. Of the applications for benefits that the DI program received in 2005 (the most recent year for which nearly complete data are available), 39 percent were approved at the DDS level.

Even with explicit rules about the medical conditions that make a person eligible for DI benefits, determining whether an individual applicant has such a condition is often difficult and necessarily subjective. Moreover, differences in administrative policies among the DDSs and in the circumstances of their applicants appear to have a large effect on decisionmaking. For instance, in 2004, the percentage of all initial claims for disability benefits approved at the DDS level varied from about 25 percent in Tennessee and Mississippi to more than 50 percent in Hawaii, New Jersey, and New Hampshire.

If a DDS rejects an application, the applicant may appeal the decision and request a hearing before an administra-tive law judge. Of the applications received in 2005 and rejected by DDSs, appeals were filed in one-third of the cases, and in three-quarters of the appealed cases, the ini- tial decisions were reversed.10 In recent years, the number of appeals has greatly outstripped the ability of adminis- trative law judges to process cases, leading to large back- logs and long delays.11 For example, in fiscal year 2000, judges issued decisions an average of 300 days after an appeal was filed; processing time reached 514 days at the end of fiscal year 2008.

Over the past few years, SSA has received significant increases in funding to tackle the backlogs of DI claims. As a result, the average processing time for an appeal has fallen to 442 days. Maintaining that improvement will be difficult, though, because the number of applica- tions for DI benefits increased by 21 percent between 2008 and 2009 and applications so far this year are run- ning 4 percent higher than in 2009. The surge in applica- tions, propelled in part by a shortage of job opportunities attributable to the weak economy, will probably result in longer delays in processing future appeals unless further increases in funding are provided.

The report included this graph, showing the total number of persons in the program (as well as it's cost):

I extracted the numbers from this graph, and then divided them by the BLS civilian noninstitutional population, aged 25-65, to get an approximate fraction of the population that are claiming disability benefits:

You can see that the expansion of this program is indeed very substantial, and presumably provides a means of sustenance for at least some of the men of lower educational attainment who increasingly don't participate in the labor force in recent decades.