From the Field

How Charters Change a School District

Does running a district of charter schools cost more or less than a traditional district? And does this model contribute to school segregation? As director of the Education Research Alliance for New Orleans, Tulane Professor Douglas Harris and his colleagues set out to answer these questions. FutureEd Editorial Director Phyllis Jordan talked with Harris, who sits on our advisory board, about his recent study with colleague Christian Buerger comparing spending levels in New Orleans schools from 2000 to 2014 to those of a group of Louisiana school districts that had spending patterns nearly identical to New Orleans’ before the city’s shift to charter schools. By 2014, New Orleans was spending 13 percent more overall than the comparison group, some of it in surprising ways. He also shared insights from his organization’s latest research on school segregation.

Research has shown that the seismic shift to charter schools in New Orleans has led to an improvement in student achievement. Your new research looks at how this affects the system-wide cost of delivering education. What did you find?

We’ve found a few interesting things. One is that shifting to a charter-based system increased the cost of education in New Orleans by about $1,358 per student relative to similar districts. So there was more money going into the system.

But it was also spent differently.

The first change is that they increased administrative spending. And you can break that into two parts. They paid administrators higher salaries, and they hired more administrators. One reason we think that happened is just economies of scale. There are now more than 40 organizations that run schools in New Orleans. Charter school organizations are smaller than school districts, so instead of one IT system you have to have 40 IT systems. Instead of one accounting system, you need 40 accounting systems. And so on.

What are the trends in instructional spending?

Instructional spending went down, even though total spending went up, which is a somewhat surprising finding, especially given the argument that traditional public schools are not getting enough money in the classroom.

Why do you think instruction spending is lower?

One of the big changes in New Orleans is charter schools’ hiring of less experienced teachers. So they could decrease salaries in an absolute sense while still increasing salaries for teachers relative to their experience.

Do pensions play into that?

Pensions are a big part of it, too. The charter schools are not required to participate in the state pension system, and that system is pretty expensive.

What role does transportation play? When you go to a charter system, students are less likely to use their neighborhood schools.

Right. Transportation costs are separate from either of the two categories we have talked about so far. Transportation also went up by a few hundred dollars per pupil and is a substantial increase, which was the least surprising finding in a choice-based system. We knew from prior research that the average distance to school for students increases by about two miles. They’re traveling a lot farther. When I get to work in the morning, I can see the freeway, and all you see are school buses.

Given what you found, what should New Orleans—or, more broadly, charter management organizations—do to contain costs?

The larger, interesting point here are the ways in which charter schools choose to spend their money. They’re hiring younger teachers and that, by itself, changes almost everything else; it changes how much you pay them, which frees up money for other things. In addition to the economies-of-scale issue, you can imagine that they might think logically that they’re going to need more and more able managers and leaders to be able to deal with that lack of experience. Part of it is just professional development, more classroom observations and feedback, and so on. But it’s just a different model of running schools: less-experienced, lower-paid teachers, more administrators to support them.

Is it a smart strategy?

The charter sector is producing positive results in New Orleans. So it’s hard to say that they’re doing something wrong in how they’re spending money, even though we can’t say exactly what’s driving those academic results. They could be spending their money more effectively. We just don’t know. It’s hard to be that specific about the causes.

But it’s not impossible that this is a good model, at least in some situations. They’re able to hire teachers who seem to be high quality, doing a good job, and they have a lot of flexibility over the hiring, dismissal, compensation, development, and so on.

You talked earlier about economies of scale. Is there a model for a unified administration for many charter schools?

It comes down to a trade-off between autonomy and centralization. There have been many cases in New Orleans where charter management organizations have banded together to try to save money through creating group contracting activities. Food service is a good example.

In other instances, I think the schools prefer to go their own way. As an example, even though it’s more costly, the charter leaders that I talked to fiercely defend their transportation spending for a couple of reasons. One is it gives them control over their school hours. The second is that they actually use the time on the bus for instilling the schools’ values. It might seem surprising, but that time on the bus is even part of community building.

You followed up the spending piece quickly with an analysis of the impact that moving to a charter district has had on racial integration in New Orleans. The city schools were already fairly segregated before Hurricane Katrina. Did the switch to choice make much of a difference?

First, I want to recognize my collaborators on this study: Lindsay Bell Weixler took the lead, and Nate Barrett and Jennifer Jennings (of NYU) were co-authors as well.

To answer your question, no, it didn’t really make a difference. We looked at a wide variety of segregation measures and subgroups, including race/ethnicity, income, special education, English-language learner status, and achievement level. There is no clear pattern across these groups. They were segregated before, and they’re segregated now.

Most prior studies have focused on race and ethnicity. On that count, our results are similar to studies of other cities. The best evidence on this suggests that, on average across cities, there’s a small increase in racial segregation when charter schools are introduced, but the results seem to depend on the context and which groups and measures are used.

How are these results different in the elementary and high school level?

We find almost no evidence of changes in elementary school, but a mix of changes—up and down—in high schools. It’s not surprising that the two levels would be different. At the elementary level, we’ve found that families are more sensitive to the distance of home-to-school, and housing is about as segregated as before. At the high school level, parents and children and willing to travel far and wide, so housing segregation and distance are less of a factor.