How do school choice programs affect budgets and performance of school districts?

Opponents of school choice programs argue the programs harm school districts, both financially and in their ability to serve their remaining students. Evidence does not support this position.

If school choice programs — charter schools, vouchers, or tax credit scholarships — harmed the existing public schools, it would be a reasonable argument against school choice. Especially if the students who remain in public schools had less of an opportunity to learn.

The prevalent argument is that charter schools and other public school alternatives drain funds from public schools. That is, if a public school student chooses a charter or private school, and if the money follows the student to the other school, the public school district loses money that it otherwise would have received. Therefore, the public school district is worse off, and so too are its students.

A rebuttal is that since a public school has shed the responsibility for schooling the student, its costs should fall correspondingly. This would be true if all the costs of a public school are variable. Some costs are fixed, however, meaning they can’t be adjusted quickly — in the short run, that is. An example is the cost to maintain a classroom. If a school has one less student than the year before, it still requires the same support for utilities. One or several fewer students doesn’t mean that fewer teachers are needed.

Public schools and their lobbyists, therefore, argue that school choice programs are a financial burden to public schools. Under school choice programs, they say, public schools lose students and their accompanying funding, but the public schools retain their fixed costs.

The Fiscal Effects of School Choice Programs on Public School Districts (cover)The question, then, is what portion of a school’s costs are variable, meaning costs that schools can adjust quickly, and what portion are fixed, meaning they can’t be adjusted quickly? Benjamin Scafidi, professor of economics at Kennesaw State University, has examined schools looking for the answer to this question. His paper The Fiscal Effects of School Choice Programs on Public School Districts, published by The Friedman Foundation for Educational Choice, holds answers to these questions.

The first question is this: What is the relation of school choice programs to school districts’ variable costs? Scafidi has endeavored to determine the breakdown between variable and fixed costs in each state. In Kansas, for the 2008 – 2009 school year, total spending per student was $11,441. Of that, Scafidi estimates $3,749, or 32.8 percent, were fixed costs. Variable costs were $7,692, or 67.2 percent. Since then spending has risen, but there’s no reason to think the allocation of costs between fixed and variable has changed materially. For the school year ending in 2014 total spending per student was $12,960. That implies fixed costs per student of $4,251 and variable costs per student of $8,709.

Now, how much money would a public school lose if a student chose, say, a private voucher school under a voucher program? In Kansas we don’t have vouchers for school choice, so we can’t answer the question directly. We do know that base state aid per pupil in Kansas is $3,852. That is the starting point for state spending per student.

In a recent presentation on this topic, Scafidi said: “Any school choice program where about $8,000 per student or less, on average, follows the child to the school of his or her choice, improves the fiscal situation of the public school district, on average, and students who remain in public schools have more resources available for their education.”

A typical Kansas school district, therefore, with variable costs of $8,709 per student, has its fiscal situation improved when it loses a student and its $3,852 in state funding.

Kansas School Finance Formula, from Kansas Policy Institute, August 2014
Kansas School Finance Formula, from Kansas Policy Institute, August 2014
Many Kansas students, however, trigger much more funding due to weightings that compensate for the purported higher costs of some situations. The largest weighting in Kansas, based on numeric magnitude, is the at-risk weighting. It adds 45.6 percent to base state aid. So if a Kansas public school loses such a student and weighting, it loses $5,608 in funding. That is far less than its variable costs of $8,709. State funding for Kansas schools in the 2013 to 2014 school year was $7,088 per student, still less than school districts’ variable costs.

I asked Scafidi what is the dividing line between variable and fixed costs? The answer is that within two or three years, schools should be able to adjust their fixed costs to be in line with their needs. This is in line with the economic and accounting reality that says in the long run, all costs are variable.

Can school districts adjust their costs quickly in response to changing enrollments? This may be a problem for the very smallest districts, those with just one or two teachers per grade, Scadifi concedes. In his paper, Scafidi illustrates two examples of districts in Georgia with just over 1,000 students making adjustments. In Kansas, there are 286 school districts. Of these, 207 have enrollment of less than 1,000 students, but only 20 percent if the state’s students are in these small districts.

School districts often dispute the contention that they are able to reduce their variable costs rapidly in response to enrollment changes. Scafidi notes that if school districts say they cannot reduce costs when they lose students, the implication is that all of their costs are fixed. If true, then schools should not receive additional funding when enrollment rises. After all, if all their costs are fixed, costs do not change with enrollment — either up or down.

We have seen that school choice programs do not harm the finances of local school districts. The second question concerns the quality of education for the students who remain in public schools.

To answer this question, we must recognize the wide variation of teacher efficacy. Some are very good, and some very poor. Further, the difference between good and bad is large. Eric A. Hanushek and others have found that very good teachers routinely produce 1.5 years of gain in achievement during an academic year. Bad teachers produce 0.5 years of gain. If a student is unfortunate enough to experience ineffective teachers two or three years in a row, the student may be so far behind as to never catch up.

What does this have to do with school choice programs? If public schools have to downsize due to students lost for any reason — including school choice programs — this gives public schools an opportunity to shed their least effective teachers. This means that students who remain in public schools have a higher likelihood of experiencing the most effective teachers.