Fixed Costs and Labor Supply
NBER Working Paper No. 484
This study is a theoretical and empirical analysis of the effects of time and money costs of labor market participation on married women's supply behavior. The existence of fixed costs implies that individuals are not willing to work less than some minimum number of hours, termed reservation hours. The theoretical analysis of the properties of the reservation hours function are derived. The empirical analysis develops and estimates labor supply functions when fixed costs are present, but cannot be observed in the data. The likelihood function developed to estimate the model is an extension of the statistical model of Heckman (1974) that allows the minimum number of hours supplied to be nonzero and differ randomly among individuals. The empirical results indicate that fixed costs of work are of prime importance in determining the labor supply behavior of married women. At the sample means, the minimum number of hours a woman is willing to work is about 1300per year. The estimated fixed costs an average woman incurs upon entry into the labor market is $920 in 1966 dollars. This represents 28 percent of her yearly earnings. Finally, labor supply parameters estimated with the fixed cost model are compared to those estimated under the conventional assumption of no fixed costs. Large differences in estimated parameters are found, suggesting that the conventional model is seriously misspecified.
Document Object Identifier (DOI): 10.3386/w0484
Published: Econometrica, Vol. 49, no. 4 (1981): 945-964.
Users who downloaded this paper also downloaded* these: