AN ECONOMETRIC ANALYSIS OF UNEMPLOYMENT INSURANCE BENEFIT ADEQUACY (RATIONING CONSTRAINTS, TOBIT MODELS).

Abstract

Explicit parameterizations of labor supply are specified and estimated on a sample of single unattached individuals using data from the Panel Study of Income Dynamics and a generalized Tobit maximum likelihood method which is consistent under the assumption that employed hours are exogenous. Results of these estimations are then used to compute triangle approximation and direct closed form solutions for labor market constraint compensation. Underemployment compensation estimates are generated and compared to actual and hypothetical payments which would accrue under the UI systems of representative states. Certain compensation results for overemployment are also offered. Where they are directly comparable, results from Tobit estimation of the basic labor supply relations are found to strictly dominate ordinary least squares (OLS) results in terms of efficiency. While the OLS and Tobit parameter estimates differ dramatically in most cases, the latter are consistent with the bulk of recent empirical labor supply research. A corollary purpose of estimating the several labor supply specifications is the search for an appropriate structure of preferences to be used in modeling the labor-leisure choice problem. Direct likelihood ratio tests yielded no best form, but suggested that more flexible parameterizations are to be desired. Results on compensation amounts tend to support accepted standards of UI benefit adequacy. For all levels of unemployment the direct compensation results suggested that "one-half gross wage replacement" would slightly overcompensate individuals from a utility based perspective.