Thursday, May 23, 2013

Uncomfortable Truths about Charter Schools

I was recently accused, at a public meeting, of having an “intense hatred” of charter schools. I do not. Life is too short to harbor such feelings about anything or anyone.

But I am annoyed and aggravated by the charter school movement in Michigan and by the misinformation spread by its proponents. I have these feelings because I believe deeply in the importance of public schools to the success of the American experiment in representative democracy. Free education is the key to opportunity and social mobility, and an educated citizenry is the best support for an effective elected government. Yet the charter school movement is undermining both financial and public support that public schools need to adapt to and to thrive in the 21st Century. It is also resegregating our schools by race, socioeconomic status, and disability.

Oh, sure, charter academies are classified as public schools, and they do receive public funding. But they do not serve the same students. When my state senator asserts that charters “do a better job for less money,” I strongly object. First, are they doing the same job? Second, do they really do a “better job”? Finally, do they have significantly less money to work with?

I will answer these questions in local terms, where I have rich data collected by the state. There are two* charter academies located within the Van Buren Public Schools boundaries: Keystone and Achieve.

Doing the same job? A different population

I assert that they are not doing the same job: (1) they do not serve students from pre-K through twelfth grade; (2) they serve a much lower percentage of economically disadvantaged children, and (3) they serve a vastly lower proportion of special-needs children.

Demographic data released [Feb 2013] on the Fall 2012 MEAP tests lets me demonstrate that. Of students in grades 3–8 in Van Buren Public Schools, 56% are economically disadvantaged and 10.6% have disabilities. Of students in grades 3–8 at Keystone Academy, 31% are economically disadvantaged and less than 3.9% have disabilities. Of students in grades 3–8 at Achieve Charter Academy, less than 13.3% are economically disadvantaged and less than 5.2% have disabilities. I use the qualifier “less than” because, for many grade levels, only “<10” was given (meaning that no statistically valid percentage could be calculated). In my calculations, I assumed the value of “9” for “<10,” although the actual numbers were certainly lower, based upon the numbers for other grade levels. The proportion of students served by charters who are economically disadvantaged and who have disabilities is, therefore, not just significantly lower than in VBPS, but those percentages are also inflated by the assumption that “<10” equals nine.

Why do differences in student population matter? Because all three categories of students under-served by charter schools are more expensive to educate.

Special education students, for example, typically cost much more to educate than per-pupil and special-education funding amounts to; the difference comes from all the other children’s per-pupil funding. They are also, by state law, educated through age 26, so the cost burden on districts will be greater than on K–8 charters.

Economically disadvantaged students, as a group, arrive at school less prepared, have less stability and support in their home lives, and are more likely to be distracted or handicapped by disruptions and traumas. Simply put, they need more intensive and individualized help than students who enjoy more advantages, and so they cost more to educate to their potential. As with special education students, there is additional funding available for “at risk” students, generally based upon poverty measures, but it does not cover the additional costs — which must come from the per-pupil funding for all other students.

Secondary students, with their science labs, career-tech classes, band and other electives, and extensive athletic programs, cost more to educate than elementary students; the difference comes from elementary per-pupil funding. This is why there are so few charter high schools — not as much profit to be made.

Doing a better job? How to evaluate results

It can be very difficult to fairly evaluate the results forming the basis of what we consider a “good job.” Should we use graduation rate or college attendance and completion rates as measures of success? No. These cannot be measured for K–8 charter schools. How about state achievement test (MEAP) scores? That is problematic, as special education students (over-represented in VBPS) cannot be expected to perform as well as students without disabilities. And there is no characteristic more highly correlated with such scores than socioeconomic status so, again, it is unfair to compare VBPS students as a group with charter school groups which have just one-quarter or half as many economically disadvantaged students.

How about growth from year to year for each cohort? That would make some sense, in that it does not penalize or reward based upon differences in starting place. By that, I mean that it would compare rates of progress, even if some student populations are still harder to educate than others. Student growth should also be our primary objective, which is why state law now ties teacher and administrator evaluation to it. The best measure of student growth would involve valid testing of the same students at the beginning and at the end of a school year, since the cohort does not change as much during the year as it does from one year to the next. We have only recently invested in such testing (using NWEA), but it will be interesting to follow the results as this year concludes.

The MEAP tests, given in the fall, presumably test what students learned the previous year. There is good research data showing, however, that advantaged children continue to learn during the summer, while disadvantaged children lose ground over that time. So, again, tests administered in the fall will show better results for charter schools with many fewer disadvantaged students. The real gold standard would be tests aligned with the Common Core Curriculum, as we phase that in, that will allow national as well as state and local comparisons of progress.

For the past two years, we have administered tests from ACT (EXPLORE in grades 8 and 9 and PLAN in grade 10) that are nationally normed and predictive of how well students will do on the actual ACT test as 11th graders. While such data do not exist for K–8 charters, they can show conclusively what kind of job VBPS is doing in fostering student learning. This year’s testing shows edifying growth for both cohorts.

Our students may have started behind the national average, but they are close to or over it now and — more importantly — their growth rate has far exceeded the national rate in every subject area. I believe this is strong evidence that VBPS is doing a good job. I would love to see similar valid data for the charter academies.

Doing it with less? How finances compare

What revenue do you compare? Different comparisons can be made, often depending upon the politics of those doing the comparison. But there is no single, apples-to-apples comparison we can all agree is fair, either.

Should we compare General Fund revenue, the “operating” budget for schools and districts? These contain year-to-year anomalies that can distort data. For example, VBPS must take a several-million-dollar loan every spring, since the per-pupil Foundation Allowance monies from the state are doled out to suit the state’s cash-flow needs, not ours. Thirty-three percent of those funds come to us after school is out for the summer, and 22 percent arrive after our fiscal year ends on June 30. So, more than ten percent of General Fund revenues may be borrowed funds with a liability, including interest, that is even greater. Similarly, we borrow to finance new bus fleets. In the year we take out such a loan, our General Fund will show a notably large increase that does not represent actual extra money for educational purposes. In both such cases, it looks as if we have more money to work with than we do; in fact, we will owe more than we have borrowed.

Some people want to compare just state and local revenue totals, but not a total containing federal revenue. This is because the bulk of federal revenue will be extra funding for English language–learning students, those with disabilities, and those considered at risk due to economic disadvantage. While those grants do constitute extra revenue related to extra educational challenges, they do not come close to covering the additional expenditures necessary to deal adequately with those challenges. Most charter schools have much lower percentages of students in those categories and so receive less federal funding — but they also have a correspondingly smaller burden on their expenditures.

Charter school proponents will be quick to note that they are not able to bond for facilities, which is true. This means that they must pay for their buildings out of their per-pupil operating funds. Most deal with this expense by leasing a building erected by a charter system operator, such as National Heritage Academies. This lease money is both a significant cost to the charter academies and a significant profit center for some operators. (Eighty percent of Michigan's charters are for-profit; the national rate is 35 percent. And just wait until all the newly authorized on-line schools get started next fall!)

Charters also pay a small percentage (3%) of their revenue to the organization that authorized their charter, which is intended to cover some costs of supervision and evaluation by that body.

But there are also large expenses that charters do NOT have:

• Retirement. How huge this is cannot be overstated! Traditional public school employees are, by law, participants in the Michigan Public School Employees Retirement System (MPSERS). Charter employees are not required to participate in MPSERS, and almost none of them do. In a sense, charters have the same economic advantage as transplant auto manufacturers had over the Big Three: General Motors, Ford, and Chrysler had a legacy of retirees stretching back to the 1940s whose pensions and health care were supported by current operations. Similarly, traditional public schools pay about 25 percent of payroll (and rising every year) to support current retirees in the system; nearly all charter schools avoid this significant surcharge. For every employee who works enough hours to qualify for benefits — whether an administrator, teacher, custodian, secretary, bus driver, paraprofessional, or cook — traditional public school districts pay more than $25 for every $100 they earn into MPSERS. This surcharge consumed an average of $1305 per pupil in FY2012.

And the State Senate’s budget proposal for the 2013–14 school year would put increased funding into the per-pupil Foundation Allowance rather than the MPSERS contribution proposed by the Governor. That means charter schools would get a real revenue increase, but public schools would have that “increase” more than eaten up by a raised MPSERS contribution rate. Instead of ameliorating the “unlevel playing field” related to MPSERS, the Senate’s proposal would make it significantly worse.

• Transportation. This is a huge expense in for VBPS, the second largest district in Wayne County (only Detroit is larger); it comes from every child’s per-pupil funding. Charter schools are not required or expected to provide transportation, and districts are not legally allowed to charge for it.

• As noted above, traditional public schools have significantly higher percentages of high school, special education, English language–learning, and economically disadvantaged (including homeless) students than do charter academies. The extra, unreimbursed costs of educating these groups is taken from the per-pupil funding of all other students.

The bottom line is that I have seen no convincing data to show that charter schools have less money to spend in the classroom.

One other difference

While this does not have financial ramifications, I think it important to recognize that charters are resegregating our public schools by race. When I moved to Michigan in 1978, I was startled to discover how racially segregated its communities were. In looking for a place to settle, I deliberately chose the Belleville area because it was both racially and economically integrated, and I wanted that environment for my children. But the same MEAP demographic data cited above also shows racial disparities. I cannot get demographic data for the district’s population but have assembled it for the communities which comprise VBPS.

It is striking how much the percentages of Asian, black, and white students vary from one school system to another and compared to the overall community. These large variations should not exist, and I question whether they do not undermine the mission of public schooling to offer equal educational opportunity.


“Choice” is described as the panacea for whatever ails public education, when there is absolutely no evidence that choice alone guarantees or even fosters better results. Choice leads to demonizing of unionized teachers and denigration of traditional public schools — after all, choices are framed in terms of relative merits, so the new choice must portray itself as better than the old ones. Thus, traditional public schools that serve increasingly needy students and reach for ever higher performance bars are portrayed as failing, because that serves the narrative that we must have charter schools to rescue children from being trapped within them. Painting traditional public schools as “failing” and their teachers as incompetent weakens community pride and support while demoralizing staff.

This view of our schools and teachers is unwarranted. My children each spent 13 years in VBPS; they also attended college on merit scholarships, earned engineering graduate degrees, and are gainfully employed and contributing to society. Our schools and teachers did very well by them — and I believe those schools and teachers are both significantly better now. They are better despite the fact that state support for traditional public schools has been cut dramatically in recent years, with no relief in sight. So, yes, I get a little hot under the collar when it is alleged, despite evidence to the contrary, that “charter schools do a better job with less money.”

* This blog entry has been edited to remove references to a third charter school that is no longer located within VBPS boundaries.