Health, Income, and Retirement: Evidence from Nineteenth Century America

I investigate the factors that fostered rising retirement rates prior to social security and private-sector pensions by estimating the income effect of a large government transfer, the first major pension program in the United States, covering Union Army veterans of the American Civil War. The pension, because of the program's rules, had only an income effect and these rules create a natural experiment to identify the effects of pensions and health on labor supply. Pensions exerted a large impact on retirement rates. The elasticity of non-participation with respect to pension income was at least 0.66, exceeding even the most conservative estimates of that elasticity with respect to social security payments. Union Army pensions were a much larger fraction of retirement income than social security payments today and this accounts for some of the difference in estimated elasticities. My findings suggest that secular increases in income can explain a substantial part of the rise in retirement rates, although the elasticity of labor force non-participation with respect to transfer income may have fallen over time, perhaps because of the increasing attractiveness of leisure.