John C. Ham, Robert J. Lalonde, The Effect of Sample Selection and Initial Conditions in Duration Models: Evidence from Experimental Data on Training, Econometrica, Vol. 64, No. 1 (Jan., 1996), pp. 175-205

The past 20 years have seen important changes in public policy with the potential for significant impacts on health insurance for children. These changes included both those explicitly intended to expand access to public insurance for children, including expansions in eligibility for Medicaid and the introduction of the State Children’s Health Insurance Program (SCHIP), and other changes in antipoverty policy. Since health insurance coverage among children is entwined with parental welfare participation and employment, shifts in policy designed to encourage work in place of welfare participation--such as welfare reform and the expansion of the Earned Income Tax Credit (EITC)--may have secondary impacts on children's health insurance coverage. As parents leave welfare, with its guaranteed health insurance through Medicaid, for jobs that may or may not have health insurance coverage offered as a benefit, children may experience a change in the source of their health insurance coverage or may become uninsured. Similarly, changes in health care markets and economic conditions such as rising health care prices and cyclicality in the availability of employment may also affect children's coverage. Despite the potential importance of these factors for coverage, the fraction of children who are uninsured has remained largely constant, particularly through the 1990s. However, this relatively constant level of uninsurance may mask changes in the underlying dynamics of health insurance among children. In this paper, we use monthly data from the 1986–1996 panels of the Survey of Income and Program Participation (SIPP) to examine patterns of health insurance coverage among children during the period 1986–1999, focusing on transitions between public coverage, private coverage, and no coverage. Using these data, we find that over the 1990s the rate of transitions among all three insurance states--public insurance, private insurance, and no insurance--increased, with a particular increase in transitions involving public coverage. We investigate whether there is evidence of a relationship between insurance transitions and various policy and economic variables, focusing on the impacts of expansions in public coverage availability, the effects of other policies directed at the poor that affect employment and insurance coverage, and economic conditions. We find that several of the policy changes that took place over the 1990s had important effects on health insurance transitions for children.

Understanding the employment dynamics of disadvantaged families is increasingly important. We estimate duration models describing these dynamics for disadvantaged single mothers and use them to conduct a rich set of counterfactual analyses. We use a misreporting model to correct for “seam bias,” the problem that too many transitions are reported between reference periods in panel data. We find effects of demographics, minimum wages, unemployment rates, and maximum welfare benefits, but not policy changes introduced through state welfare waivers, on employment dynamics. We find that two commonly used ad hoc methods of addressing seam bias perform substantially worse than our approach.

The authors address two questions using experimental data on disadvantaged women. First, what is the impact of being offered Job Training Partnership Act classroom training on the duration of unemployment and employment? Second, what is the effect of actually participating in this training on the length of such spells? Belonging to the treatment group shortens unemployment spells but has no effect on employment spells. Actually participating in training has a larger positive effect on the exit rate from unemployment than the effect of simply being a member of the treatment group. Ignoring the endogeneity of actual training in estimation substantially underestimates its effect.

Inexperienced women, along with economics and business majors, are much more susceptible to the winner’s curse, as are subjects with lower SAT/ACT scores. There are strong selection effects in bid function estimates for inexperienced and experienced subjects due to bankruptcies and bidders who have lower earnings returning less frequently as experienced subjects. These selection effects are not identified using standard econometric techniques but are identified through experimental treatment effects. Ignoring these selection effects leads to misleading estimates of learning.

We estimate the effect of endogenous training participation on transitions in and out of employment for disadvantaged women in the Job Training Partnership Act (JTPA) study. Decomposing the effect of training on employment into its effects on transitions in and out of employment has the potential to develop more effective programs. We also consider a potentially serious identification problem that arises when individuals do not undertake training immediately, and we propose a test to shed light on this problem. We find that this problem is not important in our context. JTPA classroom training substantially reduced unemployment durations, and thus it complements programs that increase employment durations.