Three Essays on the US Labor Market: Macroeconomic Trends and Cycles.

Abstract

This dissertation consists of three essays on the US labor market. The first two chapters examine the dynamics of the labor market in search models of unemployment. The third chapter concerns the rising nonemployment of low-skilled males. This research contributes to our theoretical and empirical understanding of how public policies and economic frictions affect the labor market in the short and long term.

Chapter 1 studies the implications of the tax system used to finance unemployment insurance, known as "experience rating." Experience-rated taxes impose higher payroll tax rates on firms that have laid off more workers in the past. In empirical analysis, higher experience rating is found to reduce both job creation and job destruction since firms are reluctant to both lay off and hire workers due to potentially higher tax burdens. A general equilibrium model with realistic UI financing shows that higher experience rating reduces the depth of recessions but can prolong their length.

Chapter 2 examines the sluggishness of the labor market after recessions. In macroeconomic data, labor market tightness---the number of job vacancies per unemployed worker---takes four quarters to react fully to a negative economic shock. A search model of unemployment is extended to include costly capital adjustment. Intuitively, costs of investment can cause a delay in capital adjustment which, in turn, delays adjustment of the labor market. The model is able to account for the slow recovery of the labor market during recessions.

Finally, chapter 3 (with Michael W.L. Elsby and Matthew D. Shapiro) studies the secular increase in nonemployment of low-skilled males over the past thirty years. A simple model of family labor supply predicts that the "replacement rate" of family income during nonemployment relative to employment is an important determinant of male joblessness. Detailed income data show that families buffer income loss from male nonemployment through substantial government transfers and spousal income. The empirical estimates of the replacement rate are falling while nonemployment is rising. This suggest a limited relationship between the relative value of nonemployment, and the role of more generous government transfers, in determining male joblessness.