Cost of Charter Schools

Summit Middle School and Denver-Area Newspapers
Respond to Bogus Allegations by School Districts in 2002

Round 1: School Districts Attempt to Suppress Charters

Reports were commissioned by Boulder Valley, Jefferson County, and other Colorado school districts with the objective of proving that charters impose a financial burden on school districts and cost districts more than regular schools. Obligingly, the reports, issued in February and March, 2002, do just that. The report for Boulder Valley is available here.

On the basis of these reports, several Colorado school districts have imposed an indefinite moratorium on new charter schools, including Boulder Valley, Fort Collins, and Aurora. Jefferson County considered it but backed off.

Round 2: A Denver Post Editorial Board Member Criticizes the Report

Bad Arguments Never Die

by Al Knight
The Denver Post
April 17, 2002

A half-dozen Colorado school districts have commissioned studies clearly intended to undermine support for charter schools by showing they are a drain on limited district resources.

Any participating districts were quite able to determine what they spend on charter schools but apparently wanted a pretense of scholarship and thus hired Augenblick & Myers, a consulting firm, to compile the data. The results, however, were all calculated to make it appear that a dollar spent on a charter school is a dollar stolen from more important work.

Jefferson County, for example, claims it "lost" $6.9 million on charters this year.

How were these "losses" figured? Well, the study's underlying assumption was that if a child enrolls in a charter school the impact is all negative, that is that the district will not only have to fund the student in the charter school but won't reap any offsetting benefit elsewhere. It will never be able to reduce teaching staff, save on supplies or administrative supervision, reduce the need for capital construction or perhaps simply avoid hiring a new teacher who otherwise would be needed.

This pessimistic view was captured perfectly by The Denver Post on Monday with this lead sentence: "Charter schools are costing cash-strapped neighborhood schools millions of dollars by siphoning off classroom funding and school district resources."

What the story didn't say is that Augenblick & Myers is dealing from a stacked deck. In its report for the Jefferson County School District, for example, the firm acknowledges it "accepted the concept that the loss of students to charter schools creates a revenue loss to the district." Under this concept, a loss is created because a charter school attracts a student from this neighborhood school or that, and the district can't immediately eliminate the need for a teacher or an administrator.

This is partly true, but these points also need to be made. Some charter schools were formed eight years ago. Wouldn't you think that maybe the district by now has adjusted its "resources" so that any short-term "cost" of the transfer has been absorbed? Also, isn't it a fact that similar shifts in population occur all of the time? People move. Enrollments rise. Enrollments fall. This teacher may have 22 students. Another may have 30. Yet there has never been a study in Jefferson County schools that whines about the fact that the mobility of the population makes school staffing difficult.

All of these money concerns, by the way, were addressed at the time the Charter Schools Act was passed by the legislature. The lawmakers knew, as did the districts, that charter schools would draw from a wide area and that it was only over time that staff and resources could be adjusted. These are settled matters. Do these districts want to go back and hold the debates over again even though there are now 89 charter schools in the state serving more than 20,000 students?

The thing that makes this rehash of old arguments so galling and so transparently self-serving is that huge factors in public education are utterly ignored so that what amounts to fly specks can be endlessly magnified.

Take the issue of immigration. Tens of thousands of students, many of them here illegally and unable to speak English, have enrolled in the state's public schools, causing a tremendous ripple effect. A recent story indicated that the budget for language instruction had to be stretched to accommodate three times the number of students compared with 10 years ago. Coincidentally, it was announced that the state's graduation rate has dropped, in common with three other states that have high immigrant populations. Yet there hasn't been a sentence or paragraph uttered, let alone a study commissioned, to determine the effect of illegal immigration on the state's school budgets or graduation rates.

Instead we are treated to a rerun of all the old reasons some school districts originally resisted charter schools and still regret having to offer parents and children educational choices they otherwise wouldn't have.

The fact that "limited resources" were spent on this enterprise makes it even more of a disgrace.

Al Knight (aknight[at]mindspring.com) is a member of The Denver Post editorial board. His column appears Wednesday and Sunday.

Round 3: The Rocky Mountain News Questions the Report's Analysis

Unfair to Charter Schools

Editorial by Rocky Mountain News Staff
April 19, 2002

Charter schools are public schools, and their students are public school students. They typically receive less per-student money for operating costs than other district schools, and often have the additional financial burden of building or renting their own facilities, while non-charter schools operate in buildings funded by the taxpayers.

Given that fiscal reality, the $7.8 million proposed for charter-school construction in the $4 billion school-finance act is pitifully small; charter-school children are 3 percent (likely to be close to 4 percent next year) of the school population. That Senate Democrats in the Education Committee stripped even that construction money from the school-finance bill is disgraceful.

But legislators aren't the only charter school opponents who forget that these schools and their students are every bit as much a part of the public schools as their counterparts in traditional schools.

The evidence is found in tendentious reports commissioned by local school districts from the firm Augenblick & Myers Inc. The one from Jefferson County Schools, for example, purports to demonstrate that Jeffco's 2,960 charter-school students "cost" the district almost $7 million a year.

It's nonsense, as a matter of fact. By the district's own estimates, 40 percent of the charter students wouldn't even be attending Jeffco schools if they weren't in the district's charter schools; at $5,404 in per-student revenue, that's $5.8 million in additional revenue.

Augenblick & Myers blithely ignore this fact. As they say, "We accepted the concept that the loss of students (to charter schools or for other reasons) creates a revenue loss to the district." The district argues that when a handful of students transfers out of a school, the school loses the revenue for them but doesn't save on expenses. But if that argument were valid, it would apply equally if the students left for another neighborhood school. But of course that isn't true; in Jeffco, thousands of students move from school to school every year, and it can't be losing money on all of them.

The flaw in the argument is its assumption that the number of teachers can be reduced only if the number of children who leave in a given grade in a given school is more than half the average class size. But depending on the size of the class, losing even one student may make it unnecessary to split a big class and hire another teacher for that grade.

When children transfer, some schools need fewer teachers and some need more, but on average, across the district, the numbers balance. When they transfer to charter schools, however, regular schools save money on the staff reductions but don't have to pay for any of the staff increases.

The report's argument that fixed costs don't decrease is equally fallacious. Yes, if a single student leaves the district that has no effect on fixed costs. But in Jeffco, the number of students in charter rather than neighborhood schools has grown to 1,800 over eight years — that's nearly six whole elementary schools. You can't build and staff six schools without raising fixed costs. The question isn't how much they are spending now; it's how much they would be spending if traditional enrollment were 1,800 larger.

Or are they telling us that they are operating the district as if all those students had never transferred? If so, why?

Jeffco also claims to be spending $360,000 on administrative support for its charter schools, necessary because it's the chartering authority. If that's the problem, the solution may be to to introduce competition into the system by allowing charter schools to seek their authorization elsewhere. Preferably somewhere where their contribution is fairly and genuinely appreciated.

Round 4: Summit Weighs In

Don't Blame Charter Schools or Choice

by Tom Mahowald and John Jacus
Boulder Daily Camera Guest Opinion
May 18, 2002

A budget shortfall for the Boulder Valley School District has given rise to repeated claims that growing public charter school enrollments and the ability to "open enroll" at any school have somehow imposed a significant cost on the district. Such claims are inaccurate and misleading, and have no place in a debate about making our schools more efficient in times of financial stress.

The real reasons for the budget shortfall are readily apparent, and do not involve BVSD's public charter schools, which routinely do more to improve student achievement with less money, facilities and other resources. Moreover, the recently proposed moratorium on charter school approvals in the BVSD, if adopted, would do nothing to address the root causes of BVSD's budget woes, and it would seriously harm the efforts of many highly motivated people working to improve educational performance and choice within the BVSD at its public charter schools and other schools of choice.

BVSD bases its claims of hidden "costs" of public charter schools on a biased analysis prepared by the firm of Augenblick & Myers for BVSD and several other school districts in Colorado. The A&M reports all suffer from (1) ignoring school district revenue generated by charter schools (state funds for students attracted to charters from private schools or home-schooling, as well as revenue from charter buy-backs of district services); (2) ignoring the forseeability of increased charter enrollments and the district's obligation to manage public schools accordingly; and (3) assuming that every child leaving a non-charter school for a public charter program has a negative impact on the district's allocation of teachers and administrators at the schools losing enrollment (the impact is just as likely to be positive for any given child, and when aggregated across the district, largely results in a wash). Incredibly, even after these obvious flaws of the A&M reports were recently exposed by the Rocky Mountain News and a columnist for the Denver Post, the district has persisted in referring to the BVSD version of A&M's reports as immutable fact, and as support for a moratorium on approving new charter schools.

Instead of developing a theoretical "cost" to the district associated with public charter schools, perhaps we should look at what they accomplish with the money they receive, a far more straightforward approach to determining value in most endeavors, including public education.

BVSD's charter schools have provided innovation, specialization, and consistently high levels of student achievement that raise the bar for all BVSD schools, and they do it with less money and less adequate facilities. A good example is Summit Middle Charter School, in south Boulder, the district's first charter school. Summit has twice earned the John J. Irwin School of Excellence award and consistently ranks among the highest performing schools in annual CSAP testing. Summit also has developed an excellent music program and competitive sports teams, despite its serious lack of adequate facilities at the former elementary school it now occupies. If the current budget debate were truly about efficiency, we would be heaping praise on our public charter schools, like Summit, for doing so much with so little.

The real causes of BVSD's budget woes include a severe shortfall in projected revenues from unequalized Specific Ownership (vehicle) Tax (the district began to rely on such revenue for operating expenses in the late 1990s); disproportionate growth in centralized support functions (administration) and special education over the past four years, during which overall enrollment in the district has risen only slightly; the BVSD's unwillingness to fully address the consequences of declining enrollments within the city of Boulder; and a labor agreement with the BVEA (teachers' union) last year which the district can scarcely afford given all of these other factors. Charters have nothing to do with these problems, and a moratorium would not address them in the least.

Another serious problem with a moratorium is the harm it would do to school choice and educational innovation in general. The A&M report was focused specifically on charter schools, but its flawed logic could be and has been aimed at non-charter focus schools and all open enrollment. With fully 30 percent of students in the BVSD currently open-enrolled, it is clear that charter schools are but one of the many factors which cause year-to-year enrollment changes at surrounding schools, and they are a relatively minor one at that.

The bottom line in all of this is that parents make enrollment choices based on what is best for their children, rather than on what is convenient or efficient for public school administrators. Public charter schools in the BVSD compose a small fraction of open-enrolled students, and a very small fraction of total enrollment.

Additionally, student enrollment in charters has been well established for years and subject to caps. How long can BVSD's administrators claim they are still adapting district operations to the fact of open enrollment and the small portion that constitutes charter school enrollment?

Apparently, as long as we are willing to believe them.

Tom Mahowald and John Jacus are members of the Summit Middle Charter School board of directors.

Round 5: Boulder's Daily Camera Questions District's Logic

Time Out? Not Now

Editorial by Boulder Daily Camera Staff
May 19, 2002

The proposed moratorium on new charter schools in the Boulder Valley School District is a response to real problems — but it's a needless and misdirected response.

The school board is exploring the possibility of a "time out" on charter schools, which are operated under contract with the district. No applications are in the pipeline right now, so the short-term issue before the board is whether to impose a moratorium on something that isn't happening. It's not even clear that the moratorium would serve its intended purpose of blocking future applications: State law provides that a school district "may reasonably limit" the number of charter schools, but that ambiguous language wouldn't deter a highly motivated parent group from challenging the district's policy.

School officials want to move forward anyway, in the belief that charter schools have contributed to financial problems and worrisome enrollment trends. We've heard that lament often in recent months — so often that it's beginning to have the feel of a campaign to single out and stigmatize charter schools, which for all their distinctive qualities are public schools within Boulder Valley. There are costs and benefits to the operation of charters. The district tends to overstate the costs and minimize the benefits.

Relying on a study by the firm Augenblick & Myers Inc., district officials estimate that charter schools are draining $3.5 million this year from the district's general fund, because revenue decreases but costs don't when students migrate to charter schools. We won't deny, although some do, that the migration imposes substantial costs on the district. But the calculations don't take into account revenue generated by charter schools, which buy services from the school district (and attract many students who otherwise wouldn't have attended the Boulder Valley schools at all). To blame the district's budget shortfall on charter schools is, at the least, to magnify one problem while ignoring many others.

Studies also can't measure — and school officials rarely emphasize — the benefits of charter schools. First of all, they're sources of innovation (and the district would do well to explore whether their success stories might be transferred to other settings). They attract parents who otherwise might have taken their children out of the district and into private schools. They provide a distinctive choice within the public school system, relieving the pressure for more drastic solutions such as vouchers.

One concern underlying the proposed moratorium is that local schools have become more segregated along racial and economic lines as more parents take advantage of choice programs. But when 30 percent of students now choose to attend a school outside their neighborhood, why point the finger at charter schools, which currently attract 5 percent of all students?

For that matter, why point the finger at anyone? Parents across Boulder Valley are making individual decisions in the best interests of their children. If changing enrollment patterns are one by-product of those decisions, school officials should be working creatively within that new environment instead of trying to wish it out of existence.

When school officials complain that the budget would look healthier if they didn't have to contend with charter schools, they're fighting the battles of 1995 in 2002. Charters aren't some unexpected addition to the educational landscape. They've been around for years, and they're here to stay. The school board shouldn't give even the appearance of treating them as a scapegoat.

Round 6: Rocky Mountain News Opposes Jefferson County Plans to Restrict Charters

Jeffco Schools Put Budgets First

Editorial by Rocky Mountain News Staff
August 24, 2002

EDUCATORS PRATE about their devotion to children's education, but when budgets get tight their real interests emerge. Children are the means to preserving educators' jobs. Providing the kind of education the children's parents believe is best for them is a distant second.

This week Jefferson County said it was considering limiting or suspending new applications for charter schools in order to protect its budget. They're not ready to make a decision, said board President Debra Oberbeck — thank goodness for that — but that they would even give serious consideration to such a step is distressing, as is the fact that several other districts have already taken it.

The idea that charter schools somehow "cost" the district money is simply wrong. In the first place charter schools are district schools, and the tendency of school officials to talk as if they aren't reveals a deep ambivalence about the charter school movement.

But it's wrong from an accounting standpoint as well. The district relies on an analysis earlier this year by Augenblick & Myers, which purported to discover that each charter school student "cost" the district $2,565. But obviously a student's departure from a regular district school to a charter school removes both revenues and costs from the central budget.

If a company outsources one of its functions, for instance — say the cafeteria — its accountants won't make the error of saying the loss of lunch receipts "caused" budget cuts.

For school districts, transfers in and out of charter schools, like transfers between regular schools, are pretty much a wash, with regard to spending on teachers, operations and maintenance. The totals of income and expense may go up or down, but the per-student tally doesn't change much. There's one notable exception, though, and that's administration.

Jeffco spends much more on administration than other districts — 8 percent of its budget, which is 40 percent more than Cherry Creek spends, for example. The loss of that overhead money, we suspect, is what really rankles. It's true Jeffco gets less per student than some other districts, but that fact is unrelated to how many children are in charter schools.

If falling enrollments are causing a budget squeeze, why don't Jeffco officials take a long hard look at what charter schools offer that they don't? What do parents want?

Where are the waiting lists? If officials really believe that having more children in charter schools is harmful to the district, why not do something to bring them back?

That would be a constructive response. Foreclosing the possibility of parents' choices by denying new applications is not.

Colorado law allows districts to "reasonably limit" the number of charters, but what exactly that means hasn't been tested in court.

Saving bureaucratic jobs isn't a good enough reason, we hope. But a long-drawn-out court battle, which could last the length of a child's whole elementary-school career, isn't the best way to solve the problem.

Rep. Keith King, R-Colorado Springs, has said he may introduce a bill to give charter school applicants an alternative to school districts. Many states already do so, in a variety of ways; some have a state chartering agency, some allow universities or municipalities to issue charters. If districts turn hostile to their own charter possibilities, the legislature should step in.

Round 7: Economist Defends Jefferson County School District

Speakout: School Budgets and Faulty Reasoning

by Rudy Andras
Rocky Mountain News
August 30, 2002

Who would agree with all three of these statements?

4 is an even number. 1 and 3 added together equal 4. 1 and 3 are even numbers.

No one of course! This line of reasoning clearly illustrates the Fallacy of Division. Errors in reasoning like this were identified by Aristotle 2,400 years ago, and catalogued in his book On Sophistical Refutations.

On Aug. 24 the Rocky Mountain News editorial, "Jeffco schools put budgets first," concerning a possible moratorium on approving new charter schools, presented a line of reasoning that essentially came to the conclusion that "1 and 3 are even numbers."

In Economics 101 we were taught how to correct these errors in reasoning.

Consider these three statements on which the editorial was based:

1. Funding declines (for other district schools) result from pupil enrollment declines as students enroll in charter schools.

2. Total cost declines should equal the funding declines.

3. Total cost declines (both fixed and variable) result from pupil enrollment declines as students enroll in charter schools.

Statement 1 is obviously true. Economics tells us that statement 2 might be true. It depends on the variable-versus fixed-cost structure of total costs. Economics teaches us that, in the short-run, statement 3 is false.

The following example, based on economic principles as well as real-world school budgeting, exposes the editorial's error in reasoning. In addition, the example should also cause "right" reasoning people to reject the policy prescriptions put forward in the editorial.

A K-6 school in Jeffco has 462 students (22 students in 21 classrooms — a pupil-to-teacher ratio of roughly 22-to-1).

Each classroom loses one student to a newly opened charter school. (This is a reasonable assumption of how students exit traditional neighborhood schools for charter schools.) Based on this exodus, funding for the district's other schools is cut by about $119,700 ($6,000 times 21 less 5 percent — charter school funding is set by law at 95 percent of district per-pupil revenue, recognizing the administrative costs of charter schools).

Here's where variable- and fixed-cost differences are important. The first cost-cutting response will be to cut one teacher, since 21 fewer students now attend the school. Most would agree this is a reasonable "variable cost"-cutting response. Let's say that this reduces salary/benefit costs $50,000. That leaves $69,700 of cost cutting to go.

But wait! Implementing this variable-cost cut might not save the whole $50,000. Although the schoolwide pupil-to-teacher ratio is still 22 (441 divided by 20 teachers), those 441 students must be assigned to 20 teachers, not 21 as before.

Which class sizes will go up and which won't? Heterogeneous children (different age and skill levels) are not homogeneous production inputs like in industry! Second-graders just can't be mixed with sixth- graders.

Cutting one variable-cost teacher will result in most classrooms still having roughly 21 students per teacher, while in other classrooms the pupil-to-teacher ratio will increase in order to keep the schoolwide pupil-to-teacher ratio at 22. Those few classrooms where the pupil-to-teacher ratio increases will likely need extra staffing to keep learning from suffering. As it turns out, the initial variable-cost-cutting response has a fixed-cost component that can't be cut.

Most other cost-cutting proposals that cut "true" fixed costs (i.e., cutting the art teacher, the principal, bus transportation, other administration and administrative support, etc.) can only realistically be addressed in the long run. Fixed-cost-cutting proposals are on Jeffco's budget plate already due to enrollment declines brought about by changing county demographics. The difficulty of achieving "true" fixed-cost cuts is exacerbated by "artificial" fixed-cost cuts precipitated by additional enrollment declines caused by charter schools.

The fixed-cost-cutting dilemma is heightened further if charter schools are underachieving on CSAP tests versus other district non-charter schools with similar demographic characteristics (i.e., 10 percent or less free/reduced lunch students), costing the district additional funding based on the performance promise. Charter school results, as a group, were a whopping 17 percentage points below noncharters on the fifth-grade math CSAP in 2002.

Errors in reasoning — fallacies — are as prevalent today as they were in Aristotle's day. If this error in reasoning is the "light" the Rocky Mountain News editorial page purports to "give," who needs darkness?

Rudy Andras is a Denver economist and vice president at RBC Dain Rauscher Inc., 1200 17th Street, Suite 2200, 303-595-1213

Round 8: The Rocky Explains It Again

Quit Scapegoating Charter Schools

Editorial by Rocky Mountain News Staff
September 10, 2002

Is the loss of students to charter schools a burden to the remaining schools in a district?

Some districts have been making this case for years — in fact, ever since charters became a reality in this state. But the debate has recently taken on newfound meaning because of the stagnating economy and the likelihood of slower revenue growth. Some districts are even talking about imposing a moratorium on new charters, citing their impact on local budgets as an excuse.

Jefferson County is one of those districts looking at a moratorium, which prompted an editorial here not long ago opposing the idea. In response, Rudy Andras, an economist who advises Jeffco, sent us a Speakout column (published Aug. 30) that attempted to explain why the loss of students to charter schools is a financial hardship to regular schools. Instead, his column proves only how the scapegoating of charter schools produces simplistic analysis.

Andras asked readers to imagine an idyllic pre-charter K-6 school, with exactly 22 students in each of 21 classrooms. "Each classroom loses one student to a newly opened charter school," he says, adding "this is a reasonable assumption of how students exit traditional neighborhood schools for charters."

No, it isn't reasonable. In fact, enrollments vary widely by grade and the differences have very little to do with charters. For example, according to the school reports on the Web, Adams Elementary had 52 first-graders and 38 second-graders in 2000-2001, while Bear Creek had 115 in first grade and 135 in second grade.

The district has to adjust to a host of such changes every year, adding, subtracting and reassigning teachers as children come and go. On balance, it may be hard on the district's budget if total enrollment goes down, but one can't say in advance that it is hard on a particular school. Reality is too messy and too lumpy to assert any such thing.

Moreover, it is misleading to assume that all of a charter school's students walked out of regular public schools the same year, if indeed they ever attended a traditional school in the district. Jeffco's 14 charter schools have about 3,500 spaces. The first Jeffco charters opened in 1994, so the charter enrollment has been growing gradually. The difference for a single school in a single year averages two or three students, insignificant compared with the fluctuations that occur for other reasons.

Andras also accuses us of ignoring fixed costs, when his argument does the same at the district level. The charter enrollment is the equivalent of six additional school buildings. Does he really believe that the district could have built and be operating six more schools without increasing the district's fixed costs?

For that matter, if Andras were correct that a traditional school is harmed because it has fewer students than it might otherwise have, shouldn't he be equally alarmed at other forms of school choice — a magnet school, a different neighborhood school, a private school or home-schooling? To be consistent, the district would have to oppose all of these forms of school choice, including open enrollment, and simply assign students to schools in such a way as to maximize the use of facilities and minimize costs.

But children are not sandbags, available to be piled up against the foundations of a slowly emptying school building to keep the budget money from leaking out. Officials who argue as if they were should be ashamed.