Transitional Medical Assistance: Does “Welfare to Work” Work for Medicaid?

The Transitional Medical Assistance program was established in the 1980s to encourage transitions from welfare to work by allowing extended Medicaid benefits for a fixed period of time to certain enrollees. Under Section 1925 of the Family Support Act of 1988, states were required to provide TMA to families who lost Medicaid eligibility for work-related reasons for at least six months, and up to twelve months. This program, which had required continued reauthorization, was recently made permanent with the passage of the Medicare Access and CHIP Reauthorization Act of 2015. The Congressional Research Service estimates that TMA accounts for as much as 18% of total US Medicaid enrollment, although states are not required to separately report on TMA. Despite its empirical importance and substantial interest in the labor market effects of both health insurance and transfer programs, there has been essentially no academic research on the program in its 30 years of existence.  TMA has historically been very difficult to study because those families that make it in to the program are likely to be very different from other types of Medicaid-enrolled family, making it hard to establish a comparison group, and enrollees are hard to identify in survey data (and may even be interpreted as misclassified by researchers).

In this paper, I will contribute two central pieces to our understanding of the interaction between the health insurance safety net and the labor market. First, I will provide an in-depth descriptive analysis of the types of Medicaid enrollees who use TMA benefits, how long they use TMA benefits, and an overview of their historical transition rates to work, which has never previously been documented. Second, I will use the TMA program to study the effect of a relative increase in the marginal cost of returning to work on the probability of returning to work as well as the level of earnings in the job that the worker transitions to. In particular, I will take advantage of a new program in Wisconsin, implemented in 2012, that charges substantial premiums to some TMA beneficiaries, to study to what extent Medicaid enrollees are responsive to changes in the marginal cost of transitioning back to work. I will use a set of matched administrative data from the state of Wisconsin to estimate regression discontinuity models that rely on an enrollee’s precise family income, which determines both whether they qualify for TMA and what exactly the premium will be.  This research will help to resolve the question of whether policies designed to transition beneficiaries from `welfare to work’ are effective in the Medicaid context.