This proposed project consists of two parts. Part A is an examination of the factors determining transitions by less-educated women into and out of (i) employment and (ii) welfare participation, with particular attention paid to the important seam bias issue in the Survey of Income and Program Participation (SIPP). Seam bias refers to the problem that in the SIPP's four-month retrospective data collection, too many transitions are reported to occur in the interview month as opposed to the preceding three months. A solution to this problem is proposed that allows all of the data to be used, rather than only one-quarter of the data as in the previous literature. Factors to be considered in this examination are the welfare system, the unemployment insurance system, the Earned Income Tax Credit, the expansion of health insurance for low-income individuals and the minimum wage. The sample is restricted to females with a high school degree or less and is stratified by marital status and by the presence of children to allow for the probability that married and unmarried women and women with and without children may behave very differently with respect to labor force and welfare participation. In the case of married women, their husband's (endogenous) income is also considered. Short-run, medium-run, and long-run effects of changes in the policy variables will be estimated. Preliminary estimates show that the proposed estimation approach is feasible.
In part B of the proposal, intertemporal labor supply parameters for men are estimated under the assumption that an implicit contract model describes the economy. The implicit contract model has performed much better than the standard model in empirical testing, but it is difficult to estimate the intertemporal elasticity in such a model because the wage is unobserved. The project shows how to obtain labor supply estimates in this case using a variety of approaches. Preliminary work indicates that the labor supply elasticity for men is likely to be considerably larger than previously thought.
The intellectual merit of the proposal is as follows. Currently, researchers throw out three-quarters of the data in SIPP to deal with seam bias. The approach proposed here allows all of the data to be used, and is readily applicable by other researchers studying many other issues in the SIPP, e.g. employment durations of prime-aged males, or predicting the probability of labor force participation or health insurance coverage. Further, the proposed work differs from previous policy research in this area by focusing on women with relatively low education, considering four groups of women without constraining the policy effects to be equal across groups, and considering many policy variables simultaneously. The role of minimum wages and unemployment insurance has not been considered in the literature on employment and welfare transitions for less-educated women. The proposed estimation of the intertemporal elasticity in an implicit contract framework also fills an important gap in the literature. This model has performed well but an estimate of the intertemporal elasticity for this framework has been missing. The elasticity is useful for assessing models of the business cycle and for calculating optimal taxes and dead-weight losses of particular programs.
Both proposed projects have substantial broader impacts. The first project will help identify factors that lead women with relatively low education out of non-employment or welfare participation and therefore help welfare and employment policy. The methodology developed to address seam bias can also be used by other researchers to estimate important parameters in other policy debates, such as the sensitivity of male unemployment duration to unemployment benefits. Further, the estimated results from the proposed correction will be compared to results from the standard approach in the literature, which will provide important information on data quality to the Census Bureau and collectors of other longitudinal data sets.
The second project will help shed light on what is the appropriate model of the economy for macro purposes, and therefore has the potential to affect policy debates such as those over the appropriate monetary policy. The intertemporal labor supply elasticity for men is also a key parameter in the design of tax systems, so an accurate estimate of this elasticity is essential for tax policy.
