Do Charter Schools Have a Negative Financial Impact on Their Host Districts?

If you ask this of almost any district that hosts a charter school, the answer would be “yes.”  A report by MGT of America, which was commissioned by the United Teachers of Los Angeles, found that charter schools cost LAUSD almost $600 million annually.  To be fair, that report is disputed by the California Charter Schools Association here.

But if you ask Paul Bruno, who authored a recent PACE report, “Charter Competition and District Finances: Evidence from California,” the answer is “no.”  He writes that the fiscal stresses resulting from the presence of charter schools in California is less than that found in other states, because “California’s policy context shields districts to a large degree from fiscal strain.”  Many districts in California would be surprised to hear this.

This conclusion is based on an analysis of the impact of charter schools on per student spending (measured by spending per unit of average daily attendance, or ADA) in what is presumably, but not specifically stated, the host district.  That impact is measured by using a regression equation, in which per student expenditures is the dependent variable and one of the independent variables is charter school enrollment as a percentage of district enrollment.  The data are collected “at least once between the 2003-2004 and 2014-2015 fiscal years.”  

The author deserves credit for a technically sound analysis.  But that technical virtuosity is misapplied.  Production functions and other forms of mathematical modeling of relationships are only as good as their underlying assumptions.  In this case, the assumption is that the financial impact of charter schools on their host districts would be manifested in changes in per student spending.  In fact, that is the hypothesis of this paper.  But that hypothesis doesn’t make sense in the context of California’s school finance system.

In California, school districts and charter schools are funded on the basis of a base grant for all students (which is the same amount for all students within a grade span), and the base grant is increased by a supplemental grant and concentration factor, which are awarded on the basis of a district’s or charter school’s enrollment of students who are English learners, low income, or in foster care. In short, California’s districts and charter schools receive a specific amount of revenue per student, and they must live within that amount.  

The funding that each district receives per student is totally unrelated to the number of charter schools in its boundaries or the size of the enrollment in those schools.  Since spending per student is directly related to funding per student, and because the presence of charter schools has no impact on funding per student, there is no reason to believe that the presence of charter schools has any effect on per student spending.  The report uses sophisticated analytical tools to confirm what we already know—per student spending is driven by per student funding.  Period.

So where should we look for a fiscal impact?  Here’s a thought experiment:  consider a district with 100 students, which is funded at the rate of $10,000 per student. That school receives a total of $1 million in revenue and, after setting aside 3% for a reserve, spends a total of $970,000 (or $9,700 per student).  Now let’s say one of those students transfers to a charter school.  The district loses $10,000 in revenue, but its costs are reduced by only a fraction of that revenue loss.  Let’s be generous and say its costs go down $3,000.   Now the district is spending a total of $967,000, or $9,768 per student.  That looks like a $68 per student windfall, right?  But wait, don’t forget that the district also has lost $10,000 in revenue, which is $101 per the remaining students.  When a district loses students, the marginal savings are smaller than the reduction in funding. In this case, the district is $33 per student in the hole.

A district must adjust to this by deficit spending, reducing spending, or some of both.  It is in these areas that the financial consequences of the presence of charter schools will be felt.  We should be asking:  What are the consequences of deficit spending?  How long is it sustainable?  Where is sending reduced?  Which programs or services are affected?  What is the impact on students?  This report does not raise these questions.

Through deficit spending and/or budget cuts, the district will eventually restore its per student spending to what it otherwise would have been in order to live within its per student revenue.  The presence or absence of a charter school has no effect on this dynamic.

The report acknowledges that “per-pupil aggregate fund balances at the end of the fiscal year are significantly negatively associated with local charter school enrollment shares. This suggests that even when California districts are able to reduce their expenditures in the face of charter school competition this may be insufficient to completely offset contemporaneous declines in revenue.”  

My point exactly!  But it gets tossed aside in the next paragraph, in which the author states, “I do not find evidence that districts facing charter competition are financing their expenditures with debt….This suggests that any strain indicated by districts’ declining fund balances is not so severe as to force districts to take on additional debt.”  In truth, this suggests nothing more than that districts are obeying the legal prohibition against borrowing to pay for current expenses of education.  (Districts engage in short-term borrowing in order to manage cash flow, but such loans are repayable within the fiscal year and do not constitute deficit financing of current expenses.)

The author also argues that the 3% fee that districts may charge charter schools “may reduce, or even reverse, the fiscal strain districts might otherwise experience from charter school expansion.”  Far from being a money-maker, most districts report that the costs of oversight (especially the level of oversight that the public deserves) exceeds the revenue from the 3% fee.  This possibility is implicitly dismissed without comment.

Finally, the author refers to charter schools that are “affiliated” with a school district.  This term is not defined.  It could mean either the district in which the charter school is located (let’s call it the host district) or the authorizing district, but the explanation is less that clear:  “I treat any charter school not formally affiliated with a TPS district as though it is affiliated with the nearest TPS district.”  Maybe this refers to charter school that is “affiliated” with (authorized by?) a non-district entity, such as a county office of education, and the “nearest” district is the one in which the charter school is located, but this is not clear.

The distinction between a host district and an authorizing district is considered unimportant by the author, because “this reflects technicalities of oversight and administration, not differences of enrollment policy.”  This is not true, because, for the time period during which data for this report were collected, the host district and authorizing district were not necessarily the same.  Many districts authorized charter schools that were located outside of their own boundaries.  This allowed them to collect the oversight fee (and perform little actual oversight) while the host district experienced the effects of declining enrollment.