Recent Award

This research will study the effects of losing access to welfare (cash assistance) by taking advantage of a unique welfare reform policy in Michigan that swiftly and unexpectedly removed eligibility from tens of thousands of families from the Temporary Assistance for Needy Families (TANF) program. This project will gather detailed administrative data from the Michigan Department of Health and Human Services (MDHHS) on welfare recipients and their families and link it to a large credit bureau data for the investigation. The focus in this research is how losing assistance affect employment, education, health, and other socioeconomic outcomes. The empirical evidence on the effects of direct assistance on socioeconomic, educational, and health outcomes remain inconclusive. The results of this research will provide more data-driven evidence on the effects of direct cash assistance on several outcomes, an important subject in the face of calls for similar reforms to other social welfare programs, such as SNAP. This research has direct policy relevance, as it provides inputs into policies to improve social support programs. The results of this research project could also help establish the US as the global leader in policies to support poor families.

The sudden and arbitrary nature of the 2011 Michigan TANF policy change provides a natural experiment to identify the causal effects of losing cash welfare. This proposal uses three strategies to investigate the causal effects of this policy change. First, the new rules affected counties to varying degrees, so the PI will use a difference-in-difference regression to study the broad labor-market consequences of this policy change. Second, the policy imposed a kinked relationship between the months accumulated on TANF and the months remaining. Using a regression kink design, the researchers will test for a similar kink in the outcomes. Finally, the policy had differential effects based on the age of the family's youngest child at 2011, which provides another source of variation for causal inference. This project contributes directly to the labor economics literature; in particular to research studying the relationship between government assistance and labor supply. The results of this research provide two key contributions to the field---reassessment of cash assistance using new evidence and under a new (more recent) setting, and this paper looks at the effects of losing - as opposed to gaining - assistance. The results of this research project could help establish the US as the global leader in policies to support poor families.