Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: Social security benefits, income from assets and retirement savings are the three most important sources of retirement income (SSA, 2000). However, participation in retirement savings plans, either through employer-sponsored pensions or individual retirement accounts (IRAs), is not uniform, with only 43% of elderly households receiving income from such plans (SSA, 2000). While others have studied the impact of race (Blau & Graham, 1990) and health (Lum & Lightfoot, 2003) on retirement savings separately, this study explores the interplay of human capital, health capital (Grossman, 1972) and race on retirement savings.

Methods: This study used the first wave of the Health and Retirement Study, a nationally representative sample of 12,652 individuals between the ages of 50 and 60 in 1991. The final sample included 7,350 households: 5,233 non-Hispanic white households, 1,400 non-Hispanic black households, and 717 Hispanic households. Controlling for demographics, income, wealth and work status, we independently estimated the effects of human capital, as measured by education, and health capital, as measured by health status, on the probability of having employer-sponsored pensions using logistic regression, and on the amount of money saved in IRAs using Tobit regression. We also measured the interaction effects of human capital, health capital and race/ethnicity on participation in retirement savings plans.

Findings: Among the 7,350 households, 43.6% had an employer-sponsored pension, and 42.1% had an IRA. Human capital was positively associated with both employer-sponsored pensions and IRA participation. Interestingly, increased human capital had a stronger effect for African Americans in receiving both employer-sponsored pensions (p<.01) and participation in IRAs (p<.001) than for non-Hispanic whites. On the other hand, health capital had a significant impact on both employer pensions and IRA participation only among white respondents. After controlling for the effects of human capital, health capital, income, wealth, work status, and other demographic characteristics, there was still a significant disparity among retirement savings between whites and both African Americans and Hispanic Americans. Overall, African Americans are 71% less likely to have an IRA (p<.001) and Hispanics 73% less likely to have an IRA (p<.001).

Policy Implications: While current policy debates focus on increasing the return rates for retirement investments, more focus is needed on increasing participation in retirement savings, especially among people of color. As increased human capital results in increased participation in employer and IRA savings regardless of race, and actually reduces some of disparities between African Americans and non-Hispanic whites, there should be more investment in human capital. In addition, we should consider removing the health disincentives on private retirement savings by allowing IRA withdrawals for medical reasons.

Blau, F. & Graham, J (1990). Black-White Differences in Wealth and Asset Composition. The Quarterly Journal of Economics, 105(2), 321-39.

Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political Economy, 80, 223-225.

Lum, Y. & Lightfoot, E. (2003). The impact of health on retirement saving among older workers. Social Work Research, 27(1), 31-44.

Social Security Administration. (2000). Income of the aged chartbook 1998. Washington DC: author.