Place, Space, and Flyover States: The Geography of Poverty and the Nonprofit Social Safety Net in America

Abstract

Although it is an important component of overall federal anti-poverty policy, the in-kind human services safety net has been under-researched. Nationally, our social safety net has two parts: direct transfer programs to individuals and local in-kind human services programs that give goods and services such as food assistance, housing assistance, transportation assistance, to buffer against poverty. National analyses of equity of human services provision have been limited due to the locally devolved nature of nonprofit human services policy. In this dissertation, I use national administrative data and mixed methods to analyze equity in the provision of nonprofit human services. I analyze national data at the county level to measure equity in human services provision for rural Americans, Americans experiencing poverty, and Black, Hispanic, and Latino Americans. I also conduct the first primary data collection in counties that have indicators of being human services deserts, to learn more about the landscape of service provision in the places most at risk of underprovision.

The first project uses IRS-990 data to assesses differences in county expenditures on nonprofit human services, by level of county rurality, in American counties. It also examines the relationship between nonprofit human services expenditures and county-level poverty rates. The findings suggest that rural counties have a negative association with nonprofit human services expenditures as compared to urban counties after controlling for relevant covariates. A negative association between county-level poverty and nonprofit human services expenditures is explained by the combination of county-level rurality and regional effects, underscoring the importance of policy attention to the under-provision of human services in rural counties. In the second chapter, I identified around 300 counties that appear to have no nonprofit human services, many of which are small and rural. I then interviewed local officials in these counties in two states, Georgia and Kansas, to better understand the service landscape in counties they serve. This project contributes the first primary data collection in places that have indicators of being human services deserts. I find that transportation and distance barriers to accessing human services were the most frequently discussed theme in both sites. A majority of interviewees in Georgia raised concerns about their county having very few services overall, while about half of Kansas interviewees voiced that concern. These interviews add another indicator suggesting that many of these counties may be human services deserts. They provide information about human services provision in this set of counties and directions for future research. For the third project, I use IRS-990 data to analyze racial and ethnic disparities in nonprofit human services expenditures at the county level in the United States. The findings suggest that the county percentage population of Black and Hispanic or Latino residents is negatively associated with nonprofit human services expenditures net of other predictors.

Taken together, this dissertation shows that the nonprofit human services safety net underserves vulnerable populations, including rural Americans and Black, Hispanic, and Latino Americans. By understanding the geographic distribution of nonprofit human services to vulnerable groups, I show how and where this anti-poverty policy is falling short of the intended goals of buffering against poverty and hardship. Empirically demonstrating the spatial mismatch of how this area of poverty policy underserves vulnerable communities is the first step in addressing that problem through policy change.