Will the Rising Social Security Full Retirement Age Affect the Disability Insurance Rolls?

10/18/2011
Featured in print Bulletin on Aging & Health

The number of adults receiving Disability Insurance (DI) benefits has risen dramatically in recent years, from 2.3% of adults aged 25 to 64 in 1985 to 4.3% in 2005. Today there are 6.5 million disabled workers on the DI rolls. Benefit payments to these beneficiaries and their dependents are nearly $80 Billion per year, with an additional $45 Billion in annual spending on their medical care through the Medicare program.

Many factors have been explored as possible contributors to the rise in the DI rolls, including less stringent medical examination, the aging of the baby boom, increases in female labor force participation, and recent recessions. Yet there is another hypothesis that has been suggested but not fully explored that a rise in the Normal Retirement Age (NRA) for Social Security benefits might be encouraging more workers to apply for DI.

To understand why this might be the case, a brief review of how benefits are calculated is helpful. When a worker claims Social Security benefits before the NRA, the monthly benefit amount is reduced, since the worker will receive benefits for more years than someone who claims later. As the NRA rises from 65 for workers born before 1938 to 67 for workers born in 1960 and later the benefit reduction for early claiming rises also. For example, a worker claiming at age 62 receives 80% of his full benefit if the NRA is 65 but only 70% if the NRA is 67. By contrast, a worker who applies for and receives DI benefits is not penalized for claiming before the NRA. Thus for early claimants, DI benefits are larger than Social Security benefits, and importantly, this difference is growing over time as the NRA rises.

The effect of the rising NRA on the DI program is the subject of a new study by Mark Duggan, Perry Singleton, and Jae Song, "Aching to Retire? The Rise in the Full Retirement Age and its Impact on the Disability Rolls" (NBER Working Paper 11811). In their analysis, the authors make use of the fact that the relative generosity of DI and Social Security benefits varies across individuals depending on their year of birth, due to the ongoing increase in the NRA. They conduct their analysis using Social Security administrative records for a 1 percent sample of all U.S. workers born between 1935 and 1945, giving them a very large sample and allowing them to accurately calculate benefit entitlements and track claiming behavior.

The ratio of Social Security benefits to DI benefits in their sample varies between 75% and 80%. It is expected that a higher ratio will be associated with a lower probability of DI enrollment, since it is most advantageous to claim DI when the difference between the two benefits is greatest.

In their simplest specification, the authors find a negative relationship as expected, but it is not statistically significant. However, the authors point out that older workers would likely be more responsive to changes in this ratio, since DI applicants are required to spend five months out of the labor force and this requirement is less costly for older workers who may be planning to retire in any event. When the authors allow the effect to vary with age, they find that it is much stronger and highly significant for older workers, particularly those aged 63 and 64.

To test the validity of their findings, the authors conduct a "placebo test" that assumes the increase in the NRA took place eight years earlier than it actually did and assigns workers the ratios that would have occurred in that case. Reassuringly, they find no effect of the ratio.

The authors also explore whether the importance of the ratio varies with income and find that it is stronger for low-income workers, as might be expected since Social Security benefits replace a larger fraction of preretirement earnings for these workers and they are more likely to be in poor health and able to qualify for DI.

Finally, the authors use their estimates to forecast the aggregate effect of the rise in the NRA on DI enrollment. They project that once fully phased in, the NRA increase will raise the probability that men aged 63 and 64 are on DI by 1.6 percentage points, or 13 percent. However, the effect on overall DI enrollment would be modest the authors project that an additional 42,000 men would be on DI in the long run, which represents only a 1.3 percent increase in total DI enrollment and would offset only 4 percent of the financial savings gained by raising the NRA.


This research was supported by a grant from the Alfred P. Sloan Foundation.