NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Optimal Unemployment Insurance When Income Effects are Large

Raj Chetty

NBER Working Paper No. 10500
Issued in May 2004
NBER Program(s):   LS   PE

Studies of the consumption-smoothing benefits of unemployment insurance (UI) have found that the optimal benefit level is very small, perhaps even 0, for conventional levels of risk aversion. In this paper, I derive a formula for the optimal benefit rate in terms of income and price elasticities of unemployment durations, directly inferring risk aversion for the unemployed from their behavioral responses to UI benefits. The optimal rate of social insurance is shown to depend positively on the size of the income elasticity and negatively on the size of the substitution elasticity. I estimate these elasticities using semi-parametric hazard models and variation in UI laws across states and over time. The estimates indicate that income effects account for 70% of the effect of UI on unemployment durations, and yield an optimal replacement rate around 50% of pre-unemployment wages. These results challenge the prevailing view that social safety nets provide minimal welfare gains at a large efficiency cost.

download in pdf format
   (728 K)

email paper

This paper is available as PDF (728 K) or via email.

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w10500

Users who downloaded this paper also downloaded these:
Landais, Michaillat, and Saez w16526 Optimal Unemployment Insurance over the Business Cycle
Chetty w13967 Moral Hazard vs. Liquidity and Optimal Unemployment Insurance
Feldstein and Liebman w8451 Social Security
Gruber w4750 The Consumption Smoothing Benefits of Unemployment Insurance
Blanchard and Tirole w10443 The Optimal Design of Unemployment Insurance and Employment Protection. A First Pass
 
Publications
Activities
Meetings
Data
People
About

Support
National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us