Quasi-Experimental Evidence on the Effects of Unemployment Insurance from New York State
This paper examines unemployment duration and the incidence of claims following a 36 percent increase in the maximum weekly benefit in New York State. This benefit increase sharply increased benefits for a large group of claimants, while leaving them unchanged for a large share of claimants who provide a natural comparison group. The New York benefit increase has the special features that it was unexpected and applied to in-progress spells. These features allow the effects on duration to be convincingly separated from effects on incidence. The results show a sharp fall in the hazard of leaving UI that coincides with the increase in benefits. The evidence is also consistent with a substantial effect of the benefit level on the incidence of claims and with this change in incidence biasing duration estimates. The evidence further suggests that, at least in this case, standard methods that identify duration effects through nonlinearities in the benefit schedule are not badly biased.
*Contact Bruce D. Meyer, email@example.com, Harris School of Public Policy Studies, University of Chicago, 1155 E. 60th Street, Chicago, IL 60637. We would like to thank Joyce Burnette and Uri Kogan for outstanding research assistance, John Comiskey of the New York State Department of Labor for supplying the data, and Joseph Altonji, Raj Chetty, John Ham, Robert Fairlie, Paula Worthington and seminar participants at the Institute for Research on Poverty, Northwestern University, the NBER Summer Institute, the University of Chicago and the Federal Reserve Bank of Chicago for their comments. We are grateful to the National Science Foundation, the Searle Fund and the Stigler Center for their support. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.