Column

How a federal free meal program affected school poverty stats

Student poverty rate increased only 2 percentage points after schools adopted a new lunch program for all, a Missouri study finds

Photo of Jill Barshay

Proof Points

The Hechinger Report is a national nonprofit newsroom that reports on one topic: education. Sign up for our weekly newsletters to get stories like this delivered directly to your inbox.
student poverty

Participation in the federal lunch program is used to track student poverty rates.

In 2014, schools had a new way to give students free breakfast and lunch, paid for by Uncle Sam. Instead of asking low-income families to apply for the meals, a school district could opt to give everyone free food if at least 40 percent of the student population was already on other forms of public assistance or fell into a needy category, such as being homeless or in foster care.

This new “community eligibility” option was a policy change by the U.S. Department of Agriculture, which administers the school lunch program, and was intended to reduce paperwork and make it easier for schools to feed hungry kids. But counting kids who qualify for free or reduced price lunches had also been the way we tracked student poverty. There was some concern that school districts could mistakenly be reclassified as 100 percent low income overnight. New York City, for example, began offering its 1 million public school students free breakfast and lunch in 2017. More than 60 percent of the city’s students met the public assistance criteria but the children of relatively wealthy parents also attend public schools.  Some school buildings don’t have many poor kids in them.

Many sounded alarms that $16 billion a year in federal aid for low-income students could be misdirected to not-so-poor schools. School ratings and individual teacher evaluations across the country were supposedly at stake because of the poverty assumptions embedded in how much added value a school or teacher is providing. Even the most simple reading and math test scores might be corrupted, misleadingly showing improvements in the achievement of low-income kids.

I wrote a piece about these concerns back in 2015 in the early days of the new free lunch option. But there hasn’t been much data analysis until now, when one Missouri study finds that the fears might be overblown and that our statistics on student poverty rates haven’t changed much.

“We don’t need to panic,” said Cory Koedel, a University of Missouri economist who presented his preliminary findings at a conference of the National Center for the Analysis of Longitudinal Data in Education Research (CALDER) in January 2019. “If you liked the free-meal data five years ago, you probably shouldn’t start hating the free-meal data now.”

Related: In 6 states, school districts with the neediest students get less money than the wealthiest

Specifically, Koedel looked at all the schools in Missouri that had adopted the community eligibility option during the first three years of the program, from 2014 to 2017, and found that it had the effect of raising the percentage of kids in the state who are getting free or reduced price lunch by only 2.3 percentage points — from 51.2 percent under the old system to 53.5 percent under the new system. Achievement gaps between rich and poor on tests were virtually unchanged.

The new data does overstate poverty within some schools; 16 percent of schools in Missouri participated in the new program and their poverty level suddenly jumped to 100 percent. But since students in most of those schools were already quite poor, the size of the jumps was not as drastic as many feared. Koedel also found that the same public assistance data required for the new community eligibility provision is a suitable substitute for calculating a more accurate level of poverty inside a school building when that is important.

The study, “Using Free Meal and Direct Certification Data to Proxy for Student Disadvantage in the Era of the Community Eligibility Provision,” was conducted by Koedel and his colleague Eric Parsons, also at the University of Missouri. It is not yet published in a peer-reviewed journal and is still being revised.

At first glance, the findings seem to defy common sense. How could schools be jumping from, say, 60 percent to 100 percent free lunch, without affecting the statewide figures on the number of low-income students?

Related: No longer counting who’s poor in school

Most schools that chose to participate in the new free-lunch-for-all program tended to have high levels of poverty to start with. In practice, only a small number of students flipped from being labeled “not poor” to “poor.”

Roughly 30 percent of schools in Missouri met the minimum threshold for community eligibility of 40 percent of the students participating in other low-income public assistance programs. Those schools already had an average of 79 percent of their students receiving free or reduced price lunches. The number of students in the lunch program is typically much higher than the number of students on public assistance. Free or reduced price lunches are given to families that make as much as 85 percent above the poverty line, or roughly $47,000 for a family of four.

Yet only half of these eligible schools in Missouri chose to participate because the federal government doesn’t reimburse for the full cost of the meals until more than 62.5 percent of the students are in public assistance programs. Schools below that threshold have to kick in some of the cost themselves. So many decided to continue the old lunch program, only giving lunches to the families who apply and qualify. As a result, schools that participate in the new community eligibility option tended to have even higher rates of students on the lunch program beforehand, often above 85 percent. For a back of the envelope calculation, a jump of 15 percentage points in the poverty rate among 16 percent of the schools computes to only a 2.4 percentage point increase in the number of poor students statewide.  (Multiply .15 x .16  = .024 ) That’s almost identical to the 2.3 percent jump that the economists calculated from the actual data.

Even if every school in Missouri eligible for the new meal program were to participate, the number of students in the lunch program would have risen by 5.3 percentage points to 56.5 percent, the economists calculated.

As in Missouri, the most recent federal data shows that the new community eligibility program is having a modest impact on how we count low-income students nationally. Participation in the school lunch program has remained a flat 52 percent of all U.S. students from 2013-14 before the community eligibility option went into effect nationwide through 2015-16, its second year. During the same period, the U.S. economy was growing so one might have expected the number of students in the lunch program to have fallen as families’ incomes improved.

Still, it’s possible that as more communities participate, the poverty data will shift more. Koedel cautions that Missouri has relatively small school districts.  States with larger school districts could see their poverty numbers shift more, as could states with higher rates of poverty.

Meanwhile, the practical consequences of the new regime seem to be minimal so far. The amount of poverty in a school, in theory, should be factored into many school and teacher ratings systems in order to calculate how much added value they are providing to students. That’s because schools and teachers shouldn’t be penalized for instructing poorer students who tend to have different yearly test score gains. But Koedel explained to me that Missouri and most other states use previous student test scores as a proxy for poverty and the actual federal lunch program statistics aren’t typically entered into formulas.

Measuring poverty in each school remains critical for calculating how to distribute state and federal aid. Some states are avoiding the problem by continuing to using the old percentage of the number of students in the lunch program before the introduction of community eligibility. Koedel calculated that using the same public assistance data that allows schools to qualify for community eligibility, a process called direct certification, is a good substitute. However, he says that states with large numbers of Hispanic students may still need to make further adjustments.  That’s because Hispanic families often don’t participate in public assistance programs even when they’re poor enough to qualify for them — either because they don’t want to call attention to their immigration status or because of cultural opposition to receiving government handouts. Both FutureEd, a think tank at Georgetown University’s McCourt School of Public Policy, and the Food Research & Action Center, an advocacy group, are tracking how states are adjusting their funding formulas to distribute aid to low-income students.

This story about student poverty was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for the Hechinger newsletter.

The Hechinger Report provides in-depth, fact-based, unbiased reporting on education that is free to all readers. But that doesn't mean it's free to produce. Our work keeps educators and the public informed about pressing issues at schools and on campuses throughout the country. We tell the whole story, even when the details are inconvenient. Help us keep doing that.
All donations doubled through the end of the year.

Join us today.

Letters

Jill Barshay

Jill Barshay is a staff writer and editor who writes the weekly “Proof Points” column about education research and data. She taught algebra to ninth… See Archive

Letters to the Editor

Send us your thoughts

At The Hechinger Report, we publish thoughtful letters from readers that contribute to the ongoing discussion about the education topics we cover. Please read our guidelines for more information.

By submitting your name, you grant us permission to publish it with your letter. We will never publish your email. You must fill out all fields to submit a letter.





No letters have been published at this time.