Individual Submission Summary
Share...

Direct link:

In-kind Government Assistance and Crowd-out of Charitable Services: Evidence from Free School Meals

Saturday, November 15, 10:15 to 11:45am, Property: Hyatt Regency Seattle, Floor: 6th Floor, Room: 601 - Hoh

Abstract

Many community organizations operate as an informal safety net by providing necessities to households in need. Although programs offered by the charitable sector have similar objectives to government income assistance programs, there is little empirical work examining how the use of local charitable services responds to changes in the generosity and availability of in-kind government assistance programs. In addition, it is largely unknown how the interaction between the formal and informal safety nets varies with local economic and demographic characteristics.


This paper estimates the degree to which government programs crowd out the use of charitable services, using nutrition assistance programs as an example. Nutrition assistance is a cornerstone of both US government aid and the missions of many community organizations. At the same time, food banks provide food assistance to about 40 million individuals each year (Feeding America, 2014).


We examine how expansions in government nutrition assistance interact with the use and availability of food bank services by leveraging temporal and spatial variation in adoption of schoolwide free meal programs, through which schools provide breakfasts and lunches at no cost to all students, including students whose families do not qualify for free meals based on income. We combine information on the adoption of schoolwide free meal programs with novel administrative data on the amount of food distributed by, donated to, purchased by, and available to distribute for each food bank in the Feeding America network (FA). FA is the largest domestic hunger relief charity in the United States, accounting for approximately 70 percent of the nation’s food banks and distributing 4.2 billion meals to more than 40 million individuals each year (Feeding America, 2020). Accordingly, this study is among the first to leverage data that covers nutrition-based charitable services across the entire country.


Our results are threefold. First, we find evidence of significant, albeit imperfect, substitutability between charitable and public food-related resources. On average, increasing access to free school meals by 10 percent reduces food bank use by 0.87 percent. Second, we find substantial heterogeneity in changes in food bank use. Food bank use falls more in low-poverty areas in which a relatively small share of students qualified for free meals based on their families’ incomes. Students in these areas are also unlikely to qualify for other forms of government assistance and therefore are likely to only have access to charitable services as a safety net. In contrast, there is no significant reduction in food bank use in high-poverty areas.


Third, we examine how expanded free school meals affect the availability of food bank resources by examining donor and organization behavior. We do not find a significant reduction in the amount of food available for distribution, a change in the number of nutrition-focused charities operating in an area, or significant changes in food bank revenue or fundraising activities. Altogether, our findings show that expansions in government assistance can reduce demand for charitable services, even in situations where populations targeted by the formal and informal safety nets are not identical.

Authors