Working Paper: NBER ID: w29433
Authors: Courtney Coile
Abstract: As Americans work longer in response to a changing retirement landscape, it is important to ask whether there are groups being left out of this trend. Geography is a natural lens through which to examine this question, given regional disparities in the employment of prime-age individuals. In this study, we explore the geography of retirement using data from the U.S. Census/American Community Survey and other sources. We find large differences across U.S. commuting zones in employment rates at older ages, with a gap of about 20 percentage points between areas at the 90th and 10th percentiles of employment. Low-employment areas are systematically different, with a less educated and more diverse population, more low-wage jobs and import competition from China, poorer health outcomes and health care access, lower government spending, and more income inequality. Although these correlations are not necessarily causal, these factors collectively can explain about four-fifths of the geographic variation in employment at older ages.
Keywords: retirement; employment; geography; social security disability insurance; regional disparities
JEL Codes: J22; J26; R23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Low-employment areas (J68) | Less educated population (I25) |
| Low-employment areas (J68) | More low-wage jobs (F66) |
| Low-employment areas (J68) | Poorer health outcomes (I14) |
| Less educated population, More low-wage jobs, Poorer health outcomes (I32) | Geographic variation in employment at older ages (J26) |
| Geographic characteristics (R12) | Employment outcomes (J68) |
| Low employment at older ages (J26) | Low employment at younger ages (J69) |
| Low employment at older ages (J26) | Higher rates of SSDI receipt (H53) |