Congress is expected to vote this week on a new farm bill, which includes changes to the food stamp program. Lawmakers should take the time to read up on recent research about the program’s effects. Innovative research has demonstrated convincingly that young children whose families receive food stamps benefit later in life.
Food stamps, technically known as the Supplemental Nutrition Assistance Program, is the nation’s most important anti-hunger program. Last year, more than 40 million low-income working families, people with disabilities and poor seniors received assistance, averaging about $125 a month. Roughly 70 percent of the participants live in families with children.
The classic problem in teasing out the impact of something like food assistance to poor working families with children is the old adage about correlation versus causation. Because we don’t randomly assign families to the program, it can be hard to tell what impact the program itself had. But the researchers—led in particular by Hilary Hoynes of the University of California at Berkeley and Diane Whitmore Schanzenbach of Northwestern University—have a clever way of getting around that problem.
The food stamp program was established on a pilot basis in eight poor counties by President John F. Kennedy in 1961; the pilot was expanded to more than 40 counties in the early 1960s. The Food Stamp Act of 1964 then authorized the creation of a program in every county in the nation, at the county’s option. More and more counties adopted the program, and legislation in 1973 then mandated that every county participate by 1975. So between 1962 and 1975, the program became available in an increasing number of counties.
It’s this gradual rollout that allows the researchers to study the effects of the program, because children living in otherwise similar families either did or didn’t receive benefits depending on whether their county voluntarily participated at the time. (The researchers show that county choice seems to be unrelated to other factors that may have substantially affected children living there.)
The economists focus on people born between 1956 and 1981, who were young children when the program was expanding, and who grew up in families with a parent with less than a high school education. They find that access to the program as a young child significantly improved economic outcomes and health status as an adult.
In particular, food stamp access as a child was associated with much lower risk of metabolic syndrome as an adult and, especially for women, higher levels of educational attainment and income along with lower participation on means-tested benefit programs. For example, food stamp access during childhood is linked to a 5 percentage point reduction in heart disease and an 18 percentage point increase in high school completion rates, compared to those who lacked access.
This evidence contradicts some critiques of food stamps, which misleadingly argue that it’s an inefficient and ineffective program.
The authors also highlight that access seems to matter most in utero and up until age 5. Gaining access to food stamps after age 5, by contrast, didn’t improve health outcomes as an adult, perhaps because the person had already been put on a particular health trajectory by that age.
The research demonstrates a few key points.
First, finding ways of getting past the “correlation versus causation” challenge is crucial to producing convincing evidence.
Second, the returns to investing in young kids must be examined over long periods. A broader array of evidence suggests that investments during early childhood in other means-tested benefit programs such as Medicaid pay large dividends later, and that’s exactly what the food stamps research shows. If you do the analysis only during the year the benefit is received, you miss this dynamic effect.
As policymakers debate changes to the program this week, they should take into account the evidence that food stamps are a wise investment in the futures of low-income young children.