A wide geographic variation in Social Security Disability Insurance (“DI”) and Supplemental Security Income (“SSI”) participation exists across the United States. DI participation ranges from 0.4% in Aleutians West County, AK, to 21% in Buchanan County, VA. SSI participation rates range from 0.1% in Pitkin County, CO, to 21% in Owsley County, KY. What is producing this wide geographic variation?

The Social Security Bulletin (“SSB”) recently published a study examining the discrepancy.  To the surprise of policymakers and the general public, SSB concluded that regional administrative inconsistencies were not the major cause of geographical variation.  Instead, SSB found that nearly all of the variation in program participation is explained by a combination of disability prevalence and certain socioeconomic factors, including employment opportunities and health care access.

SSB approached this study in a unique way. In acquiring these results, SSB used a decomposition method to analyze the geographic variation. Prior research in this area has used regression methods, which determine relationships between an independent and dependent variable, while decomposition methods break down variables into smaller subsets in order to better analyze the data.

For example, SSB “decomposed” the geographic variation in DI/SSI participation into three separate operations. First, the overall participation variation is broken down into disability prevalence and participation rate among persons with disabilities.  Second, the participation rate among persons with disabilities is broken down into several socioeconomic subsets. Finally, the disability prevalence rate is broken down into “within-state” and “between-state” variations. By analyzing three different levels of data, SSB is better able to explain the variation in DI/SSI participation and assess potential policy implications.

The resulting analysis from SSB concluded that nearly all of the geographic variation in program participation is attributable to variation in disability prevalence and socioeconomic factors and that very little of the variation is associated with inconsistencies in program administration. SSB notes that if irregularities existed in SSI/DI program administration, those with disabilities would likely move to an area with a more lenient administration, yet the resulting data did not indicate any such incidence.

SSB highlights some “within-state” and “between-state” variations.  For example in Georgia, DI participation is fairly low in the counties surrounding Atlanta, falling between 2.9 – 3.1%.  Those rates spike to 7.9% in the southeast portion of the state (5.4% is the median DI participation rate among working-age persons).  These same variations hold true for SSI as well.  These range variations are fairly consistent across the country with the lowest participation rates occurring in high-population urban areas, specifically in the northeast corridor, while many of the higher DI/SSI participation rates, some exceeding 10%, occur in the southeastern states.

While SSB has uncovered the causes behind the varying DI/SSI participation rates, disability prevalence and socioeconomic factors, what still needs to be addressed is the cause of the wide geographic variation in disability prevalence.  SSB concludes that if there were no variation in disability prevalence, the geographic variation rates of DI/SSI would be reduced by 80% and 63% respectively.  SSB suggests some influencing factors such as disparities in public health and migration of people with and without disabilities, but further research is needed to examine all possible causes of disability prevalence.

Thanks to Albany Law School summer intern Joseph Abbate for this interesting analysis of United States geographic differences.