The effect of negative income shocks on pensioners

Abstract

This paper provides the first evidence on the labor supply response to negative income shocks among full-time retirees, exploiting an institutional feature that caused differential and unexpected income losses among otherwise identical individuals in a sharp regression discontinuity design. We conclude that full-time retirees do not return to work despite losing a meaningful share of their annual income. Specifically, we can rule out response elasticities larger than 0.051. This precisely estimated null effect also extends to retirees who have limited savings, who face little demand-side obstacles to reentering the labor force, and to younger individuals who just recently entered retirement. The paper further shows that the negative income shock had no impact on the health of pensioners as measured by their utilization of the health care system. The lack of an employment and health care utilization response suggests that a reduction in benefit levels may have little impact on individuals in our context. At the height of an ongoing global crisis in which public pension funds are rapidly losing value, these results are particularly important.