Unemployment Insurance and Labor Turnover: Estimates from a Multiple-Spell Two-State Competing-Risk Hazard Model with Endogenous Unemployment Insurance Receipt
Speaker
Abstract
| The main objective of this study is to investigate the relationship between the generosity of the unemployment insurance (UI) system and the duration of post-unemployment jobs. A theoretical framework is developed that distinguishes between two opposite effects of UI on subsequent employment durations: a direct positive effect operating through the positive relationship between the reservation wage and job match quality, and an indirect negative effect operating through the negative relationship between the duration of unemployment and job match quality. In addition, a distinction is made between the impact of past UI benefit payments and the impact of future UI benefit entitlements. A critical review of the relevant empirical research is provided, and an improved empirical strategy is presented by developing a multiple-spell two-state competing-risk hazard model with endogenous UI receipt. The proposed model has three components: the equation for the receipt of UI, the equations for the duration of unemployment (which distinguish between unemployment spells that end in recalls and unemployment spells that end in new jobs), and the equations for the duration of subsequent employment (which distinguish between employment spells that end in quits and employment spells that end in layoffs). The model is estimated using 1978-2002 weekly panel data derived from the National Longitudinal Survey of Youth 1979 (NLSY79). The NLSY79 is an ongoing survey sponsored by the Bureau of Labor Statistics of the U.S. Department of Labor that is based on a nationally representative sample of 12,686 individuals who were between 14 and 21 years old as of December 31, 1978, and who were residing in the U.S. on January 1, 1979. Results from policy simulations based on the hazard model estimates indicate that a 10 percent increase in the weekly benefit amount raises the expected duration of post-unemployment jobs by 2 to 4 weeks. |
| Contact information: |
Dirk van Dierendonck |
