Shocks, Institutions, and Secular Changes in Employment of Older Individuals

Abstract

Employment rates of males aged 55-64 have changed dramatically in the OECD over the last 5 decades. The average employment rate decreased by more than 15 percentage points between the mid-1970s and the mid-1990s, only to increase by roughly the same amount subsequently. One proposed explanation in the literature is that spousal non-working times are complements and that older males are working longer as a result of secular increases in labor supply of older females. In the first part of this paper we present evidence against this explanation. We then offer a new narrative to understand the employment rate changes for older individuals. We argue that the dramatic U-shaped pattern for older male employment rates should be understood as reflecting a mean reverting low frequency shock to labor market opportunities for all workers in combination with temporary country specific policy responses that incentivized older individuals to withdraw from market work.