US Workforce Investment Act (WIA) – Positive earnings and employment effects for the adult funding stream and mostly negative estimates for the dislocated worker stream research finds

Our examination of the impacts of receiving WIA training rather than solely core and/or intensive services in two anonymous states has yielded a wealth of important findings both substantive and methodological.

We find differences in probabilities of training receipt as a function of race, age and education. Substantial unconditional differences by race largely, but not entirely, disappear with conditioning. Differences by age and schooling comport with standard human capital theory and with the types of training offered by WIA.

Our preferred estimates indicate positive earnings and employment effects for the adult funding stream and mostly negative estimates for the dislocated worker stream. This difference, for which we lack a good explanation, parallels findings in the broader literature and represents an important (and highly policy-relevant) open question in this area. Surprisingly, we do not find statistically different impacts for men and women, though our point estimates generally show large impacts for women, in line with the literature. Using the linked data from the LEHD, we find modest impacts on the characteristics of firms where workers get jobs, in directions consistent with the impacts on earnings. We find modest impacts for most groups on switching industries relative to the last pre-WIA employer. The impacts on firm characteristics illustrate one mechanism through which WIA training affects earnings, but also raises concerns about general equilibrium effects. Our estimates imply that the dislocated worker programs likely do not pass a social cost-benefit test, while for the adult programs the sign of the cost-benefit calculation depends on assumptions regarding discount rates, duration of impacts outside the data, the marginal social cost of public funds and the direct social cost of training. Particularly for State A, most reasonable assumptions yield a passing grade.

Our data cover WIA participants who entered the program between 1999 and 2005, inclusive. To what extent would we expect our findings to generalize to the latter part of the last decade, or to the present day? As the institutional framework of WIA has not changed much in the last decade, external validity will hinge on changes in the participant population and on changes in program impacts with the business cycle.