At last week’s hearing in the RI Senate Education Committee of the Bright Today school choice legislation submitted by Senator Marc Cote (D, Woonsocket, North Smithfield), Senator Ryan Pearson (D, Cumberland, Lincoln) made an assertion about variable cost savings in Cumberland based on the school district’s experience with charter schools:
We’ve lost 286 students to the variability of choice, and the variable cost that has been saved in that district is $462,000. So, that comes out to about $1,615 per student.
Those numbers, like much of the commentary from the committee, derive from a report recently released by the Special Legislative Commission to Study and Assess Rhode Island’s “Fair Funding Formula,” spearheaded by Representative Jeremiah O’Grady (D, Lincoln, Pawtucket). A review of the report, as well as context provided in news reports shows not only that Sen. Pearson was factually incorrect on a technical point, but the conclusion that he draws about variable costs is not valid.
On the technical correction, the $462,000 cited by Pearson is not a measurement of variable costs (or costs that can adjust up and down based on the number of students); it isn’t even a measure of Cumberland’s actual number of charter students. As Jessica Boisclair clarified in the Valley Breeze, that’s actually the dollar amount of additional costs if 286 students attending Blackstone Valley Prep were to return. That’s not applicable either as a comparison or as an assessment of variable costs.
For one thing, Blackstone Valley Prep is entirely elementary and middle school, which creates skewed results. High school is more expensive per student than middle school and elementary school, by nearly 20% in Cumberland. Moreover, high schools arguably have more flexibility to adjust schedules and resources in order to capture variable costs.
For another thing, Blackstone Valley is only one of the nine charter schools to which Cumberland sends students. According to projections from its budget documents, the district actually has around 400 students in charter schools. Variable costs are stepped (meaning they are saved when the change in enrollment reaches a certain threshold), so increasing the number of pupils by 40% could change the picture dramatically.
That leads to the problem with using these numbers as a definitive measure of variable costs. From the 2009-2010 school year to the 2013-2014 school year, Cumberland lost 494 students (using October enrollment numbers). Between the current year and projections for next year, the district expects to add back 57 students. Given those circumstances, it could very well be that the district just happens to be in a period with an unusual amount of excess capacity, meaning that it can bring in students with relatively little cost.
Another reason Cumberland’s budget experience can’t be seen as reflecting a true measure of variable costs is that its budget has kept going up. The following chart shows Cumberland schools’ expenditures (left axis) and enrollment (right axis).
Clearly, in an environment of growing budgets and shrinking student populations, one cannot use a simple comparison of expenditures and enrollment in order to understand variable costs in a district. Although a thorough review is beyond the scope of this analysis, Cumberland’s budget documents do allow for a quick illustration of the kind of adjustments that would have to be made.
For the 2013-2014 school year, Cumberland’s projected number of charter students increased by 110, and its October enrollment decreased by 117. In other words, almost all of the change in the district’s enrollment appears to have been attributable to charter school students. (Incidentally, that’s almost exactly the number of students that the RI Center for Freedom & Prosperity’s RI-DIMES school choice model projects would switch out of Cumberland schools in the first year.)
If we figure out the per-student increase in revenue each year and subtract the increase in per-student spending from year to year (not including the money sent directly to charter schools), we get an approximation of the amount of additional spending per student attributable to the number of students enrolled. Given their incentive structures, largely encouraged by state law, school districts tend to be slow to adjust to reductions in students, so it’s reasonable to look to the year after the large decrease in pupils for the budgetary results. (It also brings the budget and enrollment trends more in line with each other.)
In this case, the increase in per-pupil spending was $70, which is actually $23 more than RI-DIMES projects for Cumberland under the Bright Today legislation for a similar reduction in students. If we take Pearson’s numbers as correct, per-pupil spending would only go up about $4 per student.
Obviously, per student expenditures and variable costs are not the same. If the number of students decreases while the necessary expenses stay the same, the amount spent on each student automatically increases, but without adding any expenditures. However, it isn’t as though there’s no benefit to the students; each student is getting that much more of the teacher’s time and has access to that much more technology, books, gym equipment, and so on.
The whole idea of variable costs is that, at some point, that excess is more than what students need, and the district can stop paying for it. That is why Pearson’s conclusion about variable cost is not valid.
The Center’s analysis suggests that Cumberland’s truly variable costs are somewhere between $7,000 and $9,000 per student. That doesn’t mean that all of that money becomes available the moment a student leaves. The school has to make some decisions. If the administration chooses to spend money on the added benefit of smaller class sizes, then less money will be freed up for other purposes. If a district loses a full building’s worth of students across all of its schools and chooses to keep all of them open to maintain segregated communities, then that is what it is spending its variable cost savings on.
There is no reason families should be deprived of the opportunity for school choice based on the possibility of such decisions, and there’s no reason state taxpayers should be asked to fund the preferences of local districts in that way.