The “great recession” of 2007-2008 resulted in substantial economic insecurity among U.S. households, with poverty, unemployment and food insecurity reaching near-record highs. The American Recovery and Reinvestment Act (2009) (ARRA) was perhaps the key piece of legislation enacted in the wake of the crisis. The ARRA included two important changes to the Supplemental Nutrition Assistance Program (SNAP), the largest U.S. food assistance program. (1) An unprecedented increase in the amount of SNAP benefits to households and (2) a one year waiver of program use restrictions for Able Bodied Adults Without Dependents (ABAWDs). The effects of these kinds of changes are generally poorly understood. On the one hand, large rapid increases in program participation and the speed with which increases in SNAP generosity were implemented speak to the program being a flexible part of the social safety net. On the other hand, policymakers may worry that increased generosity may lead to welfare dependence and counter-productive incentives. 

 This study uses household level data from the Panel Survey of Income Dynamics (PSID) and the American Community Survey (ACS) to examine the impact of increases in benefits and ABAWD restriction waivers on program participation and labor force decisions.

Identification of the effects of SNAP benefits uses several such changes (typically in October of each year), followed by the unprecedented ARRA increase in April 2009. The PSID collects monthly program participation and labor force status for households over the periods in which benefits increased, allowing for identification of behavior changes around the ‘sharp’ policy changes. 

To identify the impacts of ABAWD restrictions the authors use program rules as well as data on state and sub-state variation in restriction waivers since 2005. ABAWDs of age 18-50 can use SNAP for 3 months in every 3 years, unless they comply with certain work restrictions. These restrictions were waived for all adults from April 2009 to September 2010. State-wide waivers were issued for most states for 2011, and city, county, and state-wide waivers had been in place in different states since 1997.

The authors conducted an email survey to agencies responsible for administering SNAP in several states to collect waiver histories. The timing of ABAWD waivers is used to identify the impact of ABAWD restrictions on program participation and labor force decisions. Specifically, changes in program participation and workforce decisions of ABAWDs in localities that become subject to waivers are compared to those of ABAWDs in similar counties (with high unemployment rates, but less than the cutoff for waivers). Families with children and households with no adult members below age 50 (cutoff age for ABAWD restrictions) are also used to construct falsification tests.