Association for Public Policy Analysis & Management

This paper investigates the relationship between the Supplemental Nutrition Assistance Program (SNAP) and college enrollment for ages 19-24, for those near or below poverty. SNAP, formerly and still colloquially known as food stamps, provides food assistance to low or no income individuals and households with the stated purpose of alleviating food insecurity for the poor and needy. It is the largest (in terms of the number of recipients) public assistance program in the United States. For the fiscal year 2017 food stamps benefits worth 60.5 billion were distributed to about 40.3 million individuals (about 12 percent of the population) or 20 million households (Food and Nutrition Service SNAP Tables). Given the scale of the program, there is little research on its impact on higher education outcomes of participants.The information or data source for our study is the Current Population Survey (CPS), it is a monthly survey of U.S. households conducted by the United States Census Bureau for the Bureau of Labor Statistics (BLS). Our study encompasses CPS observations for the years 2000-2017. The CPS uses a 4-8-4-month rotation pattern for surveying households. The chosen household is surveyed for four consecutive months, exits the sample for eight months and re-enters the sample for another four months. A result of the CPS rotation pattern is that half of the respondents in any month will be resampled in the same in month in the following year, albeit less than half are actually resampled given sample attrition. The food stamp survey questions are asked in the Annual Social and Economic Supplement (ASEC), colloquially known as the ‘March Supplement’. For our analysis, the longitudinal component of the March Supplement of the Current Population Survey (CPS) is used to examine the impact of Food Stamp Program participation on college enrollment by the use of a difference-in-difference identification strategy. We find a negative relationship between SNAP and higher education. After further analysis and robustness tests we discover that the negative relationship is likely driven by work requirements associated with SNAP.