CPR Professors Receive 2012 Grant Funding
Four Professors in The Center For Policy Research Have Received Funding for 2012 for Three Projects from The Center For Retirement Research at Boston College/SSA
“The Impact of Vietnam-Era Military Service on Social Security Benefits” -Gary Engelhardt
Abstract: Vietnam-Era veterans are approaching traditional ages for the take-up of DI and early- and normal-age OASI retired-worker benefits. A series of highly influential labor-market studies have found that Vietnam-Era military service has resulted in lower annual earnings for veterans compared to non-veterans. However, these impacts seem to have dissipated over time: Vietnam-Era veterans now earn roughly as much as non-veterans. Somewhat overlooked has been the impact on lifetime earnings. The key metric of real lifetime earnings in the OASDI program is Average Indexed Monthly Earnings (AIME), a primary input into the calculation of the Primary Insurance Amount (PIA). Lower lifetime earnings for veterans (relative to non-veterans) would lead to lower AIME, lower PIA, and ultimately lower OASDI benefits and retirement income. Therefore, this project moves beyond annual earnings and examines the impact of Vietnam-Era military service on AIME, PIA, and OASDI benefits. We match Vietnam Draft Lottery numbers to earnings histories from SSA’s Continuous Work History Sample (CWHS) for men from Vietnam-Era birth cohorts using SSA administrative data on date of birth. To account for the fact that selection into the military is potentially non-random, we use the randomly assigned lottery numbers to construct an exogenous measure of draft eligibility. We use this as an instrument for military service and perform instrumental variable (IV) estimation of the causal impact of veteran status on AIME, PIA, and DI take-up and benefits. We use our estimates to assess the overall impact of Vietnam-Era military service on OASDI benefits and retirement income.
“The Intersection of Veteran’s Benefit Programs and Disability Insurance Among Veterans” -Janet Wilmoth and Andrew London
Abstract: Although there is substantial functional limitation and disability in the veteran population, relatively little is known about how veterans’ uptake of Department of Veteran’s Affairs (VA) disability compensation programs is related to use of Social Security Disability Insurance (DI). Not all veterans are eligible for full VA disability benefits, which could increase reliance on DI among the portion of the veteran population that does not qualify for the VA program. In addition, the extent to which the use of VA disability benefits supplements or supplants DI use among veterans may depend on the age of the veteran and vary across cohorts who served in the military during different historical time periods. This project will use data from the 1992, 1993, 2001, and 2004 Survey of Income and Program Participation (SIPP) to examine the relationship between veterans’ receipt of VA service-connected disability compensation and participation in DI. Participation in, and income received from, VA and DI programs will be considered.
“How Much Do Changes in Housing Wealth Influence the Decision to Retire” -Jan Ondrich
Abstract: As a result of declines in private wealth and pension plan values during the Great Recession, many workers at or close to retirement age who still had jobs planned to delay their retirement. Those who lost their jobs, on the other hand, were more likely to collect Social Security. Since the start of the Great Recession, financial asset values have largely recovered, but housing values have not. Although there are a number of studies of the impact of the stock market on pension values and retirement behavior, there has been significantly less work on the impact of the housing market on retirement, in general, and the continuing housing impact of the Great Recession, in particular. This project examines the impact of the decline in housing values in the Great Recession on retirement using detailed panel data on older Americans drawn from the 2004-2010 waves of the Health and Retirement Study (HRS). Information on housing values, foreclosures, and market conditions will be matched to HRS households using restricted-access geo-code data at the MSA and zip code level. This will allow the impact of housing markets on retirement behavior to be identified by across-MSA and across-zip code variation in the depth of the housing market downturn, which has varied substantially over time and geographically during the recession.