Retirement Pensions and Disability Insurance for the 21st Century

Authors

Abstract

This project estimates the degree of disability risk among the United States and English over-50 populations. Using a disability measure that closely matches the criteria used by English and U.S. disability systems, we find both higher levels of disability in the U.S. and higher levels of disability risk in the U.S. Furthermore, we estimate spillovers between disability insurance and retirement pension program in the context of the increase in the United Kingdom retirement age for women in the years 2010 to 2019. We document that, despite a significant increase in disability benefit take-up among disabled individuals who would have been otherwise retired, these individuals experienced large losses in household income. Healthy individuals, in contrast, experienced much smaller losses in income, as they responded to the increase in the retirement age by increasing their labor-force participation. Finally, we develop a dynamic model of labor supply, social security benefits, and savings to evaluate the U.K. disability benefit system in the context of the U.K. retirement reform. This model uses as its inputs the parameters of the disability process we have estimated, and is estimated to match the responses to the reform. It can then be used to evaluate other joint reforms of disability and retirement program.

Key Findings

  • Our comparative exercise requires the construction of a comparative measure of disability between the United.States and England. We therefore estimate the degree of disability risk among in the U.S. and English over-50 populations using a state-of-the-art factor model. We find that individuals in the U.S. have significantly poorer disability outcomes on average than individuals in England. For instance, 50-year-old U.S. women have the same disability scores as 67-year-old women in England. Furthermore, we find that individuals in the U.S. face higher risks of a persistent disability shock than in the United Kingdom.
  • We estimate significant spillovers between disability insurance and retirement pension programs in the context of the increase in the U.K. retirement age for women in the years 2010 to 2019. This was concentrated among the disabled, who significantly increased their take-up of disability benefits in the face of increases in the state pension age. The fraction of women ages 60 to 65 receiving disability benefits rose from nearly zero to over 7% as a result of the state pension age increasing from 60 to 66.
  • We find that the increase in disability benefit take-up was concentrated among individuals who had a disability indicator above the minimum disability score (i.e., had some disability). Specifically, the fraction of disabled individuals receiving disability benefits increased by 26 percentage points, whereas the fraction of individuals with the minimum disability score receiving disability increased by only 2 percentage points.
  • Healthy individuals, in contrast, responded to the increase in the retirement age by increasing their labor-force participation. We documented that, despite a significant increase in take-up of disability benefits among disabled individuals who would have been otherwise retired, these individuals experienced large losses in household income compared to healthy individuals. Specifically, individuals with disabilities experienced a fall in household incomes of 29 percent, against a 14% fall for healthy individuals.
  • Although this is consistent with the U.K. disability system being either restrictive (i.e., it was difficult for disabled individuals to obtain benefits) or not generous (thereby being an imperfect substitute to pension benefits), we cannot disentangle whether the large decline in household incomes for the disabled was the result of these individuals being unable to work, or choosing to work.
  • Therefore we cannot draw conclusions about possible improvements to the disability insurance program. To do so, we instead outline a theoretical model in which individuals make decisions to work, save, and apply for disability benefits. This model can match our estimated responses to the reform and can be used to evaluate changes to the disability insurance program.
  • We develop a dynamic model of labor supply, social security benefits, and savings to evaluate the U.K. disability benefit system in the context of the U.K. retirement reform. This model, based on the life-cycle model in Low and Pistaferri (2015), uses as its inputs the parameters of the disability process we have estimated. Once calibrated to match our estimated responses to the U.K. reform, it will be capable of evaluating the performance of alternative disability insurance programs, including changing it to resemble U.S. Social Security Disability Insurance (SSDI).

Citation

Zawisza, Tom. 2023. “Retirement Pensions and Disability Insurance for the 21st Century.” Ann Arbor, MI. University of Michigan Retirement and Disability Research Center (MRDRC) Working Paper; MRDRC WP 2023-455. https://mrdrc.isr.umich.edu/publications/papers/pdf/wp455.pdf

Full Text

Download PDF

Project

Paper ID

WP 2023-455

Publication Type

Working Paper

Publication Year

2023