Disability and employmentReevaluating the evidence in light of reporting errors

Disability and employment

Reevaluating the evidence in light of reporting errors

Kreider

Brent

Kreider, Brent

Author

Pepper

John V.

Pepper, John V.

Author

text

working paper

Chestnut Hill, Mass. Center for Retirement Research at Boston College20022002monographic

Chestnut Hill, Mass.

Center for Retirement Research at Boston College

2002

monographic

Englisheng

English

eng

electronicapplication/pdfborn digital

electronic

application/pdf

born digital

Long-standing debates about relationships between labor supply behavior and health status among persons nearing retirement age have centered largely on disagreements about the reliability of self-reported health indicators. In light of reporting errors in work capacity, this paper considers the problem of predicting how employment rates vary with disability status when "true" disability is unobserved. Rather than imposing the strong assumptions required to obtain point identification, we take a step back to evaluate what can be inferred under a variety of assumptions that are weaker but arguably more credible than those imposed in the existing literature. Although these assumptions do not identify the conditional employment rates except in special cases, nonparametric bounds for these parameters can be obtained. Using data from the Health and Retirement Study, we estimate a set of bounds that formalize the identifying power of a number of different assumptions that appear to have broad consensus in the literature. Our results suggest that models estimated under the assumption of fully accurate reporting lead to biased inferences. In particular, it appears that nonworkers tend to overreport disabilities.

Brent Kreider and John V. Pepper.

CRR WP2002-6

CRR WP

2002-6

http://crr.bc.edu/images/stories/Working_Papers/wp_2002-06.pdf

MChBEnglisheng

MChB

Englisheng

English

eng