Abstract:

In 1995, the Social Security Administration started sending the annual Social Security Statements to workers. It contains information about the worker's estimated benefits at the ages 62, 65, and 70. In the first chapter, I use this unique natural experiment to analyze retirement and claiming decision making. Despite the previous availability of this information, the statement has a significant impact on workers' knowledge about their benefits. These findings are consistent with a model of costly information. I use this exogenous variation in knowledge to analyze the optimality of workers' retirement decisions. Before the Statement was introduced to uninformed workers, who are more likely to be low-educated and black, made, on average, worse retirement decisions. The estimated Social Security benefits contained in the Statement appears to have helped low-educated workers, but not black workers.

In response to an earlier "crisis" in Social Security financing two decades ago, the Congress implemented an increase in the normal retirement age (NRA) of two months per year for cohorts born in 1938 and afterward. These cohorts reached retirement age in 2000, and in the second chapter, I study the effects of these benefit cuts on recent retirement behavior. The evidence strongly suggests that the mean retirement age of the affected cohorts has increased by around half as much as the increase in the NRA. If these increases in work effort by older workers continue, it will have extremely important implications for the estimates of Social Security trust fund exhaustion that have played such a major role in recent discussions of Social Security reform.

Beneficiaries of Social Security face restrictions on how much they can earn without incurring the earnings test (ET). In 2000, President Clinton eliminated the ET between age 65 and 70. In the last chapter I evaluate how this removal impacts the long-term finances of the Trust Fund. I find that the Social Security Administration in the long run is actually saving money and that the removal appears to be Pareto-efficient. A removal of the remaining part of the ET is likely to be even less costly and to produce larger increases in labor supply and contributions.