@article {13495, title = {Job Demands and Social Security Disability Insurance Applications}, number = {WP 2023-461}, year = {2023}, institution = {Michigan Retirement and Disability Research Center, University of Michigan}, address = {Ann Arbor, MI}, abstract = {We use data from the Health and Retirement Study to identify the effect of job demands on applications for Social Security Disability Insurance and Supplemental Security Income benefits and to assess whether these job demands have been changing among older (ages 51 to 61) workers. We find that workers in jobs with physical demands {\textemdash} physical effort, stooping, heavy lifting {\textemdash} are more likely to apply for disability benefits, controlling for workers{\textquoteright} age, education, marital status, and health. We find that other job characteristics that we can measure {\textemdash} requiring good eyesight, concentration, and dealing with people; and being stressful and becoming more difficult {\textemdash} have little effect on disability benefit applications. We do not find a reduction in the physical demands of jobs held by older workers over our 1992 to 2016 sample period. When we control for workers{\textquoteright} education, they have increased. More in line with expectations, we find older workers{\textquoteright} jobs increasingly require good eyesight, concentration, and dealing with people, and weaker trend increases in stressfulness or increasing difficulty of the job. Together, these findings suggest that changing job requirements are unlikely to be an important driver of changing disability benefits applications in the foreseeable future. }, keywords = {Job demands, Social Security Disability Insurance, Supplemental Security Income}, url = {https://mrdrc.isr.umich.edu/publications/papers/pdf/wp461.pdf }, author = {Charles Brown and John Bound and Fang, Chichun} } @article {13494, title = {Using Administrative Data to Validate HRS Survey Responses on Application for DI and SSI Disability Benefits}, number = {WP 2023-462}, year = {2023}, institution = {Michigan Retirement and Disability Research Center, University of Michigan}, address = {Ann Arbor, MI}, abstract = {In this paper, we use administrative data from the Social Security Administration to validate survey responses for the Health and Retirement Study (HRS) regarding the application for disability benefits from Social Security Disability Insurance (DI) or Supplemental Security Income (SSI), focusing on applications that occurred after individuals entered the HRS. In our samples, amongst those that the administrative data identifies as having applied for DI or SSI, over 40\% either do not report having applied or inaccurately identify whether or not the application was successful. We find some evidence that the less well educated, those with cognitive limitations, and those experiencing a health limitation on their capacity for work are more likely to misreport applications. We also explore the effect that reporting errors have on parameter estimates in a simple model of the application for DI benefits. Parameter estimates are qualitatively similar regardless of whether we use survey or administrative data to identify the application for DI benefits in our model.}, keywords = {administrative data, disability insurance, Social Security, Supplemental Security Income}, url = {https://mrdrc.isr.umich.edu/pubs/using-administrative-data-to-validate-hrs-survey-responses-on-application-for-di-and-ssi-disability-benefits/}, author = {John Bound and Charles Brown and Fang, Chichun} } @article {12029, title = {Health and Retirement Study Imputation of Lifetime Earnings: Data Description and Technical Documentation}, number = {Version 2}, year = {2021}, institution = {Institute for Social Research, University of Michigan }, address = {Ann Arbor, MI}, abstract = {This data product, Imputation of Lifetime Earnings, provides lifetime earnings estimates of respondents in the Health and Retirement Study (HRS). It uses information from earnings records provided by the Social Security Administration (SSA), imputation, as well as projection, to estimate cumulative lifetime earnings through various ages. This documentation details the types of SSA earnings records available to the HRS, how the earnings records are utilized in this data product, and what imputation and projection methods are used when SSA earnings records are not available. Users who only need to know the definitions of variables included in this data product may proceed to Section 4 directly. This restricted data product is intended for exclusive use by you and the persons specified in the Agreement for Use of Restricted Data from the Health and Retirement Study and/or the Supplemental Agreement with Research Staff for Use of Restricted Data from the Health and Retirement Study. The HRS gratefully acknowledges the special assistance of the SSA{\textquoteright}s Office of Research and Statistics for their assistance in retrieving the administrative records of HRS respondents who gave consent for those records to be used for research purposes.}, keywords = {lifetime earnings, Social Security Administration}, url = {https://hrs.isr.umich.edu/sites/default/files/restricted_data_docs/1635887419/HRS\%20Imputation\%20of\%20Lifetime\%20Earnings.pdf}, author = {Fang, Chichun} } @article {9432, title = {The Effect of Affordable Care Act Medicaid Expansion on Post-Displacement Labor Supply among the Near-Elderly }, number = {WP 2017-370 }, year = {2017}, month = {09/2017}, pages = {1-41}, institution = {Michigan Retirement Research Center, Institute for Social Research, University of Michigan}, address = {Ann Arbor, MI}, abstract = {Expanded health-insurance coverage under the Affordable Care Act (ACA) provides alternative channels to obtain health-insurance coverage outside employment, which in theory may affect whether people want to work, how much they work, and the sorting of individuals into jobs. Although health insurance exchanges are available in all states, ACA Medicaid expansion is only available in states that chose to expand Medicaid coverage. The state-level variation in timing of Medicaid expansion provides a quasi-experiment setting that can be used to examine how healthinsurance coverage affected labor supply. In this paper, I study how Medicaid expansion affects the labor supply and re-employment outcomes of displaced (involuntarily unemployed) workers who are near-elderly, low-income, nonmarried, childless, and nondisabled. Data from 2011-2016 waves of monthly Current Population Survey (CPS) as well as 2010-2016 waves of Displaced Workers Survey (DWS) are used. Results from a discrete-choice model using the CPS suggest that, some displaced workers in expansion states became less likely to exit unemployment to employment while some others became more likely to exit unemployment to not-in-labor-force immediately following Medicaid expansion. While robustness tests suggest this may partly be attributed to state-level idiosyncrasies, my results reject large and persistent effect of Medicaid expansion on unemployment exits. The DWS does not have enough statistical power to identify the difference in re-employment outcomes between displaced workers in expansion and nonexpansion states.}, keywords = {Affordable Care Act, Employment and Labor Force, Medicare/Medicaid/Health Insurance}, url = {http://mrrc.isr.umich.edu/wp-content/uploads/2017/12/wp370.pdf}, author = {Fang, Chichun} } @article {8974, title = {Cohort Changes in Social Security Benefits and Pension Wealth}, number = {WP 2016-350}, year = {2016}, institution = {University of Michigan}, address = {Ann Arbor, MI}, url = {http://www.mrrc.isr.umich.edu/publications/papers/pdf/wp350.pdf}, author = {Fang, Chichun and Charles Brown and David R Weir} } @article {8908, title = {Health and Retirement Study Imputations for Employer-Sponsored Pension Wealth from Current Jobs in 2010: Data Description and Usage}, year = {2016}, month = {03/2016}, institution = {Institute for Social Research, University of Michigan}, address = {Ann Arbor, Michigan}, url = {http://hrsonline.isr.umich.edu/modules/meta/xyear/penswealth2010/desc/PWI10A.pdf}, author = {Fang, Chichun and Amy Butchart and Helena Stolyarova} } @article {5706, title = {Pension Estimation Program Users Guide}, year = {2016}, institution = {University of Michigan}, address = {Ann Arbor, Michigan}, abstract = {Periodically, the HRS obtains pension Summary Plan Descriptions (SPDs) from the employers of study respondents1. These SPDs are then analyzed and coded, and the plan description data, along with specific data from the respondents, are analyzed using a software product known as the Pension Estimation Program. This program is designed to estimate the pension entitlements held by respondents of the HRS, based on the plan formulas and benefit provisions obtained from the linked sample of pension providers. The Pension Estimation Program uses systems of equations to represent each of the pension plans, including all benefit formulas and payment provisions. These equations, in turn, use as input work and income histories of the respondents. For a given set of assumptions, the program calculates the appropriate pension entitlements, and generates output data files for subsequent analysis. This is the documentation for the current (April 2016) version of the pension estimation program which replaces all previous versions. Researchers may wish to examine the Pension Coding Program Users Guide for information on HRS coding procedures.}, keywords = {Methodology, Pensions}, author = {Fang, Chichun and Amy Butchart and Helena Stolyarova and Michael A. Nolte and Robert Peticolas} } @article {8916, title = {Prospective Social Security Wealth Measures of Pre-Retirees Wave 10}, year = {2016}, institution = {Institute for Social Research, University of Michigan}, address = {Ann Arbor, Michigan}, author = {Fang, Chichun and Kandice Kapinos} } @article {5851, title = {Racial Difference in the Use of VA Health Services}, year = {2015}, institution = {Ann Arbor, MI, Michigan Reirement Research Center, University of Michigan}, abstract = {We study the factors that affect the utilization of health care services administered by the Department of Veterans Affairs (VA) and its racial differences. Due to data limitation, previous research in this regard mostly only focuses on veterans who are VA users or at least eligible for VA services. We fill in the gap in literature with a random sample of veterans 51 and older from the Health and Retirement Study. We find that, among all veterans, those who are black and less healthy are more likely to use VA health services. These factors, nevertheless, are no longer statistically significant after the sample is restricted to veterans who are eligible for VA services. We also find that VA health services and services provided through other channels are at least partial substitutes: VA usage drops when a veteran becomes age eligible for Medicare or when a veteran has health insurance coverage through employment. This drop in usage holds not only among all veterans, but also among veterans eligible for VA services. Finally, perception about the quality of services delivered in VA versus non-VA facilities strongly predicts VA services usage. Those who have favorable views toward VA use VA services more, and the results from variance decomposition suggests a majority part of the racial difference in VA usage can be attributed to the racial difference in such perception.}, url = {http://www.mrrc.isr.umich.edu/publications/papers/pdf/wp334.pdf}, author = {Fang, Chichun and Kenneth M. Langa and Helen G Levy and David R Weir} }