This project combines empirical evidence and economic theory to study the optimal design of unemployment insurance. Empirical studies have shown that the probability that an unemployed worker finds a job depends on how long she has been unemployed (hereafter: duration dependence). Economic research recognizes two reasons why, in particular, the long-term unemployed are less likely to find jobs: unemployed workers lose skills during unemployment; and those that reach long-term unemployment may have been, on average, relatively unskilled when they originally lost their job. However, the existing literature on optimal unemployment insurance does not typically take these features into account. The principal investigators' preliminary research indicates that duration dependence in the hazard rate of finding a job is critical for the design of unemployment insurance. Whether unemployment subsidies should decrease or increase during an unemployment spell depends on the exact reason why the long-term unemployed are less likely to find jobs. This project studies the optimal design of the unemployment insurance schedule by creating and simulating a model that matches empirical evidence on duration dependence. The project also explores new data sources to study duration dependence and its causes.
Broader Impact: This project will have a broader impact along a number of dimensions. First, the U.S. unemployment insurance system is large. According to the U.S. Department of Labor, in 2005 some 8 million workers received benefits, costing states and the federal government about $31 billion. In many other countries, particularly in continental Europe, the unemployment insurance system is several times larger on a per capita basis. Thus, a properly designed system may significantly improve the welfare of a great number of households, either by reducing taxes or improving the efficiency of unemployment subsidies. This project may have implications for other social insurance programs, including Social Security, Disability Insurance, Medicare, and Medicaid. Third, this project integrates methods and empirical evidence from a variety of fields in economics, including public finance, labor economics, macroeconomics, and incentive theory. Fourth, the principal investigators will use graduate research assistants to help with all aspects of this project, in the process training those graduate students. Finally, this project will improve the quality and discourse of academic study by the broad dissemination of the resulting research, through scientific papers, as well as in presentations at conferences, university seminars and courses.
