Exploring associations between SNAP Employment and Training programs and Labor Force Participation among ABAWDs with disabilities

• In Event: Place and Opportunity: Human Services and Public Policy

Abstract

Constituting more than 14 million people, low-income working-aged adults without dependents comprise a large portion of the U.S. labor force but face significant barriers to employment (Greenstein et al., 2024). Among this population, individuals who do not have a government-determined disability that prevents them from working are classified as Able-Bodied Adults Without Dependents (ABAWDs) in the Supplemental Nutrition Assistance Program (SNAP) in which their benefits are conditioned on strict work requirements.  Despite being deemed the most capable of work, many ABAWDs self-report disabilities that impede their ability to work (Bauer & Hardy, 2024). Research finds evidence of positive associations between self-reported disability and unemployment rates (O’Brien, 2013); this, combined with the fact that low-income ABAWDS likely lack the resources or connections to obtain sufficient documentation for disability benefits (Rupp & Davies, 2004), suggests that additional support from the SNAP Employment and Training (E&T) programs may improve ABAWDs’ employability. However, there is limited empirical evidence on the associations between the employment support (e.g., the quantity and quality of employment services) offered in the state-designed E&T programs and labor force participation. Even less is known about how these associations differ for participants with self-reported, work-limiting disabilities. 

This study addresses two key questions: (1) Is the level of employment support provided through state SNAP E&T programs associated with labor force participation rates among ABAWDs? (2) Are these associations moderated by the presence of self-reported disabilities? To answer these questions, we create a repeated cross-sectional sample of prospective ABAWDs using the 2001-2023 Survey of Income and Program Participation panels. We identify prospective ABAWDs as 18-54 year-olds with household incomes under 131% of the Federal Poverty Line who do not have dependent children or receive Supplemental Security Income or Social Security Disability Insurance. We merge data from the FNS-583 E&T Program Activity Reports, an administrative dataset, to construct a measure of employment support offered in the E&T programs, and the SNAP Policy Database to construct an instrumental variable (IV). We specify an IV approach with two-stage (2SLS) panel regressions to account for endogeneity that could bias associations between SNAP and labor force participation. To answer our first research question, we estimate 2SLS regressions predicting multiple measures of labor force participation using instrumented SNAP receipt interacted with an index of state E&T program supportiveness. To answer our second research question, we repeat these analyses after stratifying our sample to individuals who self-report a work-limiting disability. All models control for individual and state-level covariates with state, year, and month-fixed effects.

Findings from this study will contribute important policy insights into the effectiveness of SNAP E&T programs. Identifying whether specific employment supports help adults with self-reported disabilities participate in the labor market has important implications; most adults who leave the labor market due to a permanent disability, regardless of disability benefit receipt, do not re-enter the labor market at a later time (Goodman et al., 2017). Evaluating the effectiveness of policy support is crucial to supporting state infrastructure that facilitates labor force participation and economic well-being.

Authors

• Jennifer Renee Daniels, University of Wisconsin-Madison
  Non-Presenting Co-Author

• Jamie Koenig, University of Wisconsin-Madison
  Presenting Author

• Anna Walther, University of Wisconsin-Madison
  Non-Presenting Co-Author