Allowing an entire school to eat for free, instead of restricting free lunch to students whose families fill out forms, can reduce the number of students who get suspended multiple times, according to a new study.

It’s the latest evidence that universal meal programs, which have also been linked to higher test scores and better health in other research, help students.

“There are many potential benefits to providing universal free meals in high-poverty schools, including achievement impacts … and of course whatever reduction in kids going hungry comes with it,” said Nora Gordon of Georgetown University, who wrote the paper along with Krista Ruffini at the University of California at Berkeley.

The study, which was released last week by the National Bureau of Economic Research and has not been formally peer reviewed, focuses on the federal free lunch program’s “community eligibility” initiative, which allowed schools where many students qualified for free or reduced price lunch to provide the free meal to all students. This was designed to reduce the stigma of receiving the meals among low-income students, streamline paperwork, and ensure no student went hungry. (Previous research has shown that in California, for instance, 13 percent of students who were eligible for subsidized lunch didn’t receive it for one reason or another.)

Gordon and Ruffini took advantage of the fact the the program was rolled out slowly, starting in 2012. This allowed the researchers to compare the suspension rates of the initial schools that took up the program to those in states that adopted it later. The paper also tries to account for the fact that at this time, many states and districts were already making efforts to reduce exclusionary discipline.

The study estimates that in elementary school, the chances of being suspended multiple times fell by about a third of a percentage point in elementary school and half a percentage point in middle school. Those aren’t big changes, but only a small share of students receive multiple suspensions in the first place.

Gordon and Ruffini say community eligibility may have had broader effects because it helped students nutritionally and also because it improved “the social climate of the school by reducing the stigma associated with free meals.”

There was some evidence that making entire schools eligible for free lunch reduced in-school suspensions, too, but the program didn’t seem to reduce the number of students who were suspended just once or have any effect on suspensions in high school.

One limit of the study is that it relies on federal civil rights data, which only reports the share of a school’s students suspended once and the share suspended more than once. This data has also been shown to be inaccurate in some cases, as schools may input wrong information. The results also vary somewhat based on the group of schools that the researchers focus on.

But the finding is consistent with a handful of other studies pointing to benefits of community eligibility, including the preliminary data from other papers that have yet to be published. It’s also in line with other research showing that the timing of receiving food stamp benefits — and therefore, when families may be most able to eat well — affects poor students’ test scores and behavior.  

The community eligibility program has continued to grow. Still, in the 2016–17 school year nearly half of eligible schools didn’t participate.