Empirical Evidence on the Labor Market Impacts of U.S. Social Insurance Programs

Abstract

Social insurance programs exist in the United States to help workers maintain their standard of living across different states of the world. Examples include unemployment insurance, which aids workers through the state of being unemployed, and Social Security, which supports workers through the state of retirement. The three essays in this dissertation study how these types of social insurance programs alter the decisions workers make in the labor market. The first and third essays focus on unemployment insurance, where the first essay focuses on how different types of workers make decisions in the presence of unemployment insurance and the third essay studies how all workers respond to changes in the provision of unemployment insurance. The second essay examines how Social Security retirement income influences the decision of late-career workers to participate in the labor market. All three essays emphasize that the willingness of workers to pursue a job in the labor market relies upon the social insurance available to them outside of employment. Theoretical models of optimal unemployment insurance predict that the job search and savings behavior of unemployed workers will partially be determined by how long a worker expects to remain unemployed. Empirical evidence suggests, however, that workers often underestimate the duration of their unemployment spell. These biased beliefs about the duration of unemployment among unemployed workers should therefore affect their job search and savings behavior. To date, no reliable data have been used to empirically analyze to what degree biased beliefs would change the behavior of unemployment workers. In the first essay, titled 'Biased Beliefs and Job Search: Implications for Optimal Unemployment Insurance,' I use a novel dataset, the Survey of Unemployed Workers in New Jersey, to evaluate how biased beliefs vary across unemployed workers and how they influence the behavior of those workers. I find that overly-optimistic unemployed workers underestimate the duration of their unemployment, leading them to spend 26 percent less time searching for a job each week than those with a pessimistic bias. I also find that overly-optimistic unemployed workers have over $8,500 less saved at any given point during an unemployment spell. These results suggest that unemployed workers with an optimistic bias would benefit from an information "nudge" that encourages increased search effort and could lead to faster reemployment. The first essay demonstrates how workers respond to the presence of social insurance when they are still focused on rejoining the labor market. That is, it provides evidence on the intensive margin. However, it does not say anything about how it would influence a worker's desire to participate in the labor market at all, on the extensive margin. In the second essay, 'Do Late-Career Wages Boost Social Security More for Women than Men?,' Matthew Rutledge and I estimate the incentives for older workers to continue working during their retirement-age years when they could be collecting Social Security. Any worker who delays claiming Social Security receives a larger monthly benefit because of the actuarial adjustment. Some claimants - particularly women, who are more likely to take time out of the labor force early in their careers - can further increase their benefits if the extra years of work raise their career average earnings by displacing lower-earning years. This essay uses the Health and Retirement Study (HRS) linked to earnings records to quantify the impact of women's late-career earnings on Social Security benefits relative to men's. The essay finds that the average gain in Social Security retirement benefits from working one additional year raises women's monthly benefits by 8.6 percent, of which 1.6 percent is from late-career earnings. These results suggest that, especially among women, there are additional benefits to delaying claiming and further increasing the retirement age. Through both of the first two chapters, the parameters outlining the social insurance program were held constant. In reality, the rules of a social insurance program can change over time. Motivated by this possibility, my third chapter, 'The Impact of Unemployment Insurance Extensions on Worker Job-Search Behavior,' explores how reservation wages and job search effort respond to extensions of unemployment insurance. Current economic theory predicts that reservation wages should rise following an extension of potential benefit duration, while search effort should fall. Previous papers in this literature focus on the end result, which is that UI extensions result in prolonged unemployment spells. Using the Survey of Unemployed Workers in New Jersey, and the UI benefit extension in the United States in November 2009, this paper identifies the worker behaviors that lead to prolonged unemployment durations. Employing hypothesis testing and event study analysis, this study shows there are lagged, significant increases in reservation wages and decreases in search effort following the benefit extension. The results suggest that an alternative model of job search is needed.