SNAP Purchasing Power and Child Health Care Utilization: Estimating a Causal Relationship

The Supplemental Nutrition Assistant Program (SNAP), formerly known as the Food Stamp Program, is the largest and most impactful anti-hunger program in the United States. The program targets low-income households, as eligibility is based only on household income. In 2018, it provided assistance to an average of 40 million people per month, with an average four-person household receiving approximately $448 in benefits per month. In 2016, the program prevented 7.3 million Americans from slipping into poverty; nearly half of those people were children. It is not only an anti-poverty intervention, but also an economic one: Every dollar in SNAP benefits generates a return of approximately $1.84 to the larger economy. SNAP is one of the most effective, timely, and targeted federal safety net programs.

Many studies have supported the efficacy of SNAP, but none have yet estimated a causal relationship between SNAP and child health based on variations in SNAP purchasing power. Although SNAP benefits are standardized at the national level and therefore do not vary state to state, local food prices, and therefore SNAP purchasing power, do vary state to state. In a 2019 study, Erin T. Bronchetti, Garret Christensen, and Hilary W. Hoynes designed an analysis of the efficacy of SNAP by comparing SNAP purchasing power among 30 geographic regions. The study asks: Do the children of SNAP beneficiaries have improved health care utilization or outcomes in regions where SNAP purchasing power is relatively greater?

The researchers compare the ratio of the maximum amount of SNAP benefits allotted (which does not vary) to the cost of the Thrifty Food Plan (TFP) within each of 30 different geographic regions. TFP is a market basket of foods designed by the USDA to specify the amount and type of food necessary for households to meet adequate nutritional requirements. They then compare these calculated ratios to child health care utilization and outcomes over the 30 geographic regions. They measure child health care utilization in four ways: If the child had, in the past 12 months, any check-up, any doctor’s visit, or any ER visit; and if they delayed or didn’t receive any care. They measure child health care indicators and outcomes in six ways: health status, hospitalizations, missed school days due to illness, if the child missed five or more school days, obesity, and emotional problems.

Their results suggest that children in regions with lower SNAP purchasing power have not only lower health care utilization but also worse health indicators. A 10 percent increase in SNAP purchasing power was associated with an 8.1 percent increase in the likelihood that a child received an annual check-up with their doctor and a 22 percent decrease in missed school days due to illness. Importantly, their study controls for effects due to the costs of other goods, such as housing and transportation, as well as for SNAP participation and health insurance coverage. Furthermore, the authors do not find statistically significant effects of SNAP purchasing power on measures of poverty. By isolating the effects of SNAP purchasing power on child heath, these measures provide stronger support for the authors’ conclusions.

The researchers posit three reasons for improved health among children who live in regions with greater SNAP purchasing power: 1) More food means more nutrition; 2) more food via SNAP means parents can spend more money on health care-related expenses; and 3) when there is enough food, there is less stress in the household and therefore less stress on children.

One limitation of these conclusions is that the study does not find effects on health outcomes of children — only on health care utilization. The authors provide an answer for this: Effects on health outcomes in children will likely be seen in the long term and not over merely 12 months. Another limitation is that by grouping geographic regions, the study ignores the effect of living in a city versus living in a rural area. Additionally, this is not a randomized controlled trial; indeed, the selection into SNAP benefits is not random at all.

Besides these limitations, Bronchetti and colleagues provide robust evidence that greater SNAP purchasing power is associated with greater child health care utilization and indicators. According to Healthy People 2020, a program by the U.S. Department of Health and Human Services, low child health care utilization likely leads to greater long-term health care needs. This may result in a greater future taxpayer burden in terms of public health care expenditures on Medicare and Medicaid. Given the implications, this study lends credence to the notion that any proposed policy to cut SNAP benefits must take into account the negative consequences of low child health care utilization.

Article source: Bronchetti, Erin T., Garret Christensen, and Hilary W. Hoynes, “Local Food Prices, SNAP Purchasing Power, and Child Health,” Journal of Health Economics 68 (2019): 102231.

Featured photo: cc/(William Mullan, by New York Times)