What the research says about school closures like the ones St. Louis is weighing

In 2012 and 2013, the School District of Philadelphia closed 30 of its 247 schools to address declining enrollment, long-term fiscal deficits, and lagging academic achievement. A few years later, two researchers at the University of Pennsylvania compared the changes in academic and behavioral outcomes between third through eighth grade students who were displaced by a school closure and their peers who weren’t affected by one.

What they found could be relevant for Saint Louis Public Schools, which is weighing how to overhaul its school portfolio this summer after decades of enrollment decline and three years of a growing deficit.

The district is currently holding events to get feedback from the public on the three models it’s considering: one that keeps the traditional structure of separate elementary and middle schools, one that has a mix of separate schools and PK–8 schools, and one that moves most elementary and middle schools to the PK–8 model. At least 17 elementary schools under all three models would close or be repurposed, with the most disruptive model closing several more.

Research on school closures hasn’t reached a consensus. Studies across different cities have found displaced students losing ground, gaining ground, and showing no change at all. This pattern has led Matthew Steinberg, lead author of the 2019 Philly study, to conclude that a closure is not inherently good or bad for students academically. “It depends on how the closure policy is designed and implemented,” he says.

The schools that closed in Philly were in high-poverty neighborhoods where enrollment decline had been significant and academic achievement was low, Steinberg says. He and his co-author found that, on average, school closures had no effect on displaced students’ achievement in English and math.

But that average concealed two different realities. Displaced students who moved to higher-performing schools than the ones they’d left saw academic gains in English, while students who landed in lower-performing schools lost ground. “When kids move to schools serving higher-performing kids, that insulated those kids from an adverse achievement effect,” Steinberg says. That’s important in St. Louis, he says, or anywhere a district is deciding which schools to close.

Displaced students missed 19 percent more school days in the first year after closure, equal to about 3.3 additional days, and that effect on absences continued into the second year. When displaced students had to travel farther for school, they had more absences and suspensions.

Students in schools that took in displaced students, which the researchers called receiving schools, were also affected. They lost ground in English and math, with declines that persisted into the second year.

How much ground they lost, and how much ground displaced students lost, tracked with how many displaced students ended up in a single receiving school. Both groups did worse as the share of displaced students rose.

To Steinberg, that presents a clear warning. 

“As St. Louis thinks about closing schools, it is not a good policy to move all kids from a closed school together to a new school,” he says. (“Less than 10 percent is probably the sweet spot,” he says.)

He adds, “It’s not just about whether a district chooses to close a school building or not, but rather how does the district’s policy approach intentionally move and shift students and resources in a way that protects the achievement of kids in the schools that receive displaced students and helps support the achievement of students that are entering those new schools.”

If the 17 or more school closures across the three models SLPS has shared for public consideration become a reality, 2,566 students would be displaced, according to 2026–27 enrollment projections. The plan also aims to establish feeder school patterns, turn several magnet schools into neighborhood schools, and repurpose shuttered schools as affordable housing, community hubs, and health clinics.

Superintendent Myra Berry tells SLM the district has developed some preliminary enrollment projections based on capacity, but final school-by-school assessments won’t be determined until there’s a “final” restructuring plan after the board votes. And even then, the work will just be starting. “I know we keep saying this is a final plan,” she says. “This is a live document, and we are not so far in the clouds that if something happens, we’re going to say, ‘Oh no, we’re not willing to change.’ No, if it needs to be changed, we’re going to change it.”)

That’s because the final portfolio will determine feeder patterns, attendance boundaries, specialized programming—details that will ultimately affect where students displaced by school closures end up, in addition to family choice.

Berry says the district doesn’t currently anticipate any fixed caps other than a school’s capacity. The district will evaluate each school independently, taking into account not just a building’s total enrollment, but also classroom capacity, grade-level distribution, staffing and support needs, and transportation.

While the district has looked at academic performance data across schools, Berry says it’s taking a more comprehensive approach, not just looking at how schools have historically and currently performed, but what it would take for schools to meet students’ needs—academic and otherwise—moving forward under each of the three models.

“When looking at this entire plan, the purpose of the Future Ready Saint Louis Public Schools is not simply to move our students from one building to another building,” Berry says. “But it’s to create more of a smaller and sustainable network to make sure that every school … is a well-resourced school for staffing, programming—existing right there in their individual neighborhoods.”

So if a school that has been meeting state standards ends up being repurposed, Berry says the district will look closely at what that school was doing right and emulate that at the schools where students end up so that they “can still have that same experience and still produce those same results.”

The board will vote on a final plan later this month, but there are calls from the Board of Aldermen and the public to delay it. The plan wouldn’t take effect until the beginning of the 2027–28 school year.

Lessons from the past

From 2003 through 2007, SLPS closed 16 schools for much the same reasons it’s now considering closing more: declining enrollment and finances. Back then, though, the pressure was $95 million of debt rather than an operating deficit; now, it’s a $41 million deficit.

Years later, Vanessa Garry, an associate professor at the University of Missouri-St. Louis, and Susan Uchitelle, the former director of the Voluntary Interdistrict Transfer Program and later a charter school founder, examined that earlier round of closures to see whether African American children displaced by them were better off at their replacement schools.

They found that the replacement schools were not academically or economically stronger than the closed schools, and that there wasn’t much difference between the schools the district closed and the ones it kept open. They also found that African American students in income-balanced county schools through the transfer program had higher English proficiency rates than their peers in the city’s income-imbalanced district and charter schools. (Students at the SLPS magnet schools, also created as part of the 1980s desegregation effort, were an exception. Garry and Uchitelle found that African American children at the magnet schools performed significantly better on the state’s English test than their peers in other SLPS schools.) The authors couldn’t track individual displaced students, so the comparison is between school populations, not the same children over time.

Their research points to a conclusion similar to Steinberg’s: outcomes depend on where students land. And the Voluntary Interdistrict Transfer Program, which sent St. Louis city students to county schools after they were chosen by lottery, is now winding down—eliminating what had been a strong option for city kids. 

“Closing schools is necessary when there are too few students enrolled, but it’s not a reform,” says Garry, speaking about the early 2000s closures. They made sense from operational and financial perspectives, but reform takes more: “It’s a concerted effort of engaging the community, engaging the teachers, school leaders, the city,” she says, adding that students should have a say also. Back then, that wasn’t the case. The district had hired an outside firm to oversee the closures and the community was not kept abreast of developments.

That engagement is where the current district administration says it’s standing firm. Berry says that even after the board votes on a plan, regular community feedback sessions will continue.

“This is going to be a lot of hand holding, and not hand holding in a bad way, but hand holding to make sure that we are hearing and understanding the voices,” she says.