NB19-11: Social Security Claiming and Retirement Responses to FRA Changes

The increase in Social Security’s full retirement age (FRA) from the 1983 Social Security Act Amendments is one of the biggest changes to retirement policy in recent decades, and recent Social Security reform proposals suggest further increases in the FRA in the future (CBO 2015). This project analyzes the effect of FRA increases on labor supply and benefit claiming decisions, the interaction between these decisions, and the mechanisms through which the FRA affects these decisions. There is limited evidence on the effect of FRA increases on labor supply and benefit claiming and, in particular, on whether these behaviors move in tandem or independently. If, as our Year 1 results indicate, FRA increases affect claiming behavior more than retirement behavior, then FRA increases will achieve the goal of reducing benefits but not the goal of delaying retirement, which matters for both positive and normative analysis of FRA changes.
In Year 1 of this project, we studied how workers respond to the increase in the Social Security FRA, focusing on the gap between labor supply and benefit claiming. We exploited the phased-in FRA increase from the Social Security Act Amendments of 1983 to study differences in bunching in the age of labor force exit and retirement benefits claiming across cohorts using SSA administrative data. In Year 2, we will continue this analysis and test for “learning” across cohorts, as the data include several cohorts who have already reached their new full-retirement ages. In addition, we will explore mechanisms underlying workers’ responses. We will focus on two potential mechanisms: employer-level norms or policies and geographic effects. Identifying these mechanisms is key for understanding the effects of current a future SSA policy features and reforms. To accomplish these goals, we will do the following:
• Request from SSA the Detailed Earnings File and Numident for CWHS 10% sample
• Use the DER and Numident to investigate employer-level effects and geographic effects
• Develop a model of expected retirement and claiming behavior