Published online by Cambridge University Press: 25 May 2016
I develop and calibrate an equilibrium search model with endogenous savings and search intensity. The wage is endogenized using Nash bargaining and the number of vacancies is tied down by a free entry condition. This allows me to conduct a counterfactual analysis of the optimal unemployment insurance (UI) level. The provision of UI is motivated by the worker's inability to perfectly insure against income shocks, but at the same time UI introduces distortions into workers' search intensity decisions and firm vacancy creation. I find that equilibrium effects are important. When the UI level is raised 25%, they constitute around one-third of the increase in total unemployment. However, even with limited savings by workers, optimal UI is close to zero. It is further shown that ignoring the possibility of self-insurance greatly affects the optimal UI level.
I would like to thank Rasmus Lentz for getting me started on this project, and for many helpful suggestions. This work was partly done while I was visiting the University of Wisconsin–Madison; their hospitality is greatly acknowledged. I would like to thank Jesper Bagger, Francois Fontaine, Mark Kristoffersen, Jeremy Lise, Dale Mortensen, Michael Svarer, and Chris Taber and participants at the 2012 SOLE meeting for valuable comments. I would also like to thank the anonymous associated editor and two anonymous referees for helping me substantially improve the paper. Any errors are my own.