Computing the effect of charter expansion on district budgets

Online tool shows the "severe damage" charter growth can cause

THE MASSACHUSETTS TAXPAYERS FOUNDATION (MTF) report on how charter school openings and expansions impact traditional districts ends with a caveat that should have been its thesis:

“Consolidating, realigning or moving programs – much less closing schools – is likely to be politically fraught. District budgets, heavily weighted toward personnel, may be extraordinarily difficult to cut (at least without severe damage) because of collective bargaining constraints and the fact that so many employees fulfill specialized professional roles. The analysis presented here…makes no attempt to address budgetary issues at the district level.”

We, a husband/wife team from Somerville, have combined our expertise in math and municipal finance to create an online tool that helps residents, local administrators, and voters understand the extent of the “severe damage” that charter expansions can cause to traditional districts. Take a look here:

The model shows what’s likely to happen to local finances if a K-8 charter school opens in a community. The MTF report explains that process accurately. When a student moves from a traditional district to a charter school, the amount of funding spent by the district for that child follows the child in the form of a charter school tuition payment. The state provides some transitional aid to communities as they face these new costs. That funding, however, is temporary and has been significantly underfunded in the state budget.

Thus, the district and city must find ways to cover the new tuition payment expense. Because students leave from different grades and schools, it is not easy to reduce the number of classes in a district. Without closing classes, a district cannot eliminate teacher salaries, which are its biggest expense.

Using the tool, here is the projected financial impact for Somerville if 50 students leave per year to new or expanded charter schools, along with an estimate of how many classes/teachers we can eliminate:

Year Total funding lost Sections closed Additional cuts needed
0 $0 0 $0
1 $335,000 1 $246,000
2 $1 million 3 $763,000
3 $1.7 million 4 $1.4 million
4 $2.4 million 6 $1.9 million
5 $3.1 million 6 $2.6 million
6 $3.8 million 9 $3 million
7 $4.6 million 10 $3.7 million
8 $5.5 million 12 $4.4 million
9 $6.3 million 16 $4.8 million

Given the district’s maximum class size of 25 students, Somerville will be able to eliminate about six classes by year five. Cutting those classes yields a savings of $537,000 through teacher reductions, leaving $2.6 million that must still be cut.  Here is what Somerville can do to close that deficit:

  • Increase maximum class size to 30 students/class (savings of $300,000)
  • Cut other district programs, such as contracted mental health services ($117,000), world languages in the middle grades ($358,000), and assistant principal positions ($896,000),
  • Cut municipal services or table new initiatives and transfer the savings to cover the school deficit, such as eliminate the Parks and Recreation Department ($914,000), and cut branch library hours ($300,000).
  • Raise fees for school and city programming, such as for full-day kindergarten, athletics, and music programming.

Most likely, Somerville will do a combination of the above. When enough students have left – perhaps by year five — Somerville will close a school. However, the remaining 230 students at the target school will need to be dispersed to open seats throughout the district. This will mean splitting up siblings, offering transportation (at an increased cost), and making other tough calls.

The new law will remove any limit on spending on charter tuition for any one district. That means that even if Somerville reaches a new period of stability, the process will begin again as existing charters expand or new charters open.

In any of these deficit closing scenarios, we expect per pupil spending for the students who remain in Somerville to remain stable or increase, even as students experience fewer services and benefits.

The Boston Municipal Research Bureau’s April report similarly found that for Boston, the departure of students for charters has left the traditional district schools under-enrolled. Because of the smaller class sizes, per pupil spending has increased. The City of Boston has spent more on their education budget to cover that expense and has paid more to transport students across town to charters. As with Somerville, any time municipal spending gets cut, it comes at the cost of important city functions. Either programs get cut or new initiatives get put on hold. In all cities and towns, low-income families often depend the most on city services like out-of-school time, mental health services, housing, job training, and transportation. The process of “rightsizing” in Boston has been painful, just as it will be in other districts that will be similarly affected. As with Somerville, even if Boston reaches a new stable state, removing the limit on new schools promises that the new normal will be disrupted again and again.

Also note, assuming current patterns continue, the students who remain in the traditional district will disproportionately be higher-needs children. The current and proposed law creates a disincentive for charters to serve high-needs children. In Somerville, for example, the largest charter school serving Somerville and Cambridge students enrolls 14.1 percent special education students, while our two districts serve 21.6 percent and 21.9 percent special education students. A similar pattern exists in Boston. Most (17 out of 21) of the charter schools that enroll at least 50 Boston students serve fewer students with disabilities than the Boston district does. The most vulnerable students will disproportionately be affected by this long period of pain and instability.

MTF’s chief finding, which the Globe editorial board embraces as resolving questions related to municipal finance, is that per-pupil spending has not been negatively affected by charter expansions. They conclude, thus, that there’s no evidence that district finances have been harmed.

A closer look at the dynamics of municipal financing and operations shows quite the opposite. Because of the all-too-real challenges mentioned in the report’s caveat and described in the BMRB report, we expect families and students will struggle with higher fees, bigger classes, displacement from neighborhood schools, and lost programming even as per pupil spending remains stable or increases.

Meet the Author
Meet the Author
The Globe editorial board hopes that the MTF report and their recent editorial will “put to rest” discussion of the fiscal impact of the charter school funding process. We disagree. We hope that the MTF report, instead of dampening discussion, launches a lively debate so that we can increase our understanding of what is at stake in this complicated and high-stakes decision for our Commonwealth.

Stephanie Hirsch and Joe Calzaretta are parents of three children in the Somerville Public Schools. Stephanie Hirsch has spent almost 20 years working in the analysis of finance and operations data at the state and local level. Joe Calzaretta is an MIT-trained software engineer and mathematician. Contact them at

  • sbrsb

    A very impressive and potentially useful project. However…

    You write this: “Also note, assuming current patterns continue, the students who remain in the traditional district will disproportionately be higher-needs children. The current and proposed law creates a disincentive for charters to serve high-needs children.”

    But the link you provide to “current patterns” is flawed. It ignores both highly differing classification practices and also trends which show steadily increasing enrollment at charter school of kids with special needs. In respect to ELLs, look at the statewide trends here, where, incidentally, you’ll find statewide “New Charter Student Enrollment” at charter schools at 15.9% and the statewide average for non-charter public schools: 8.4%

    How can you properly make projections without taking into account strikingly clear trends like what you see in the charts in that document? At the same time, you’d need to take into consideration the fact that charter school tuition payments are not set at the average rate for sending districts, but instead based on how many students at the receiving charter school are classified as ELL, or from low-income, ecomomically disadvantaged families.

    Even just within a single district, classification practices reportedly vary widely. Quoting from the Boston Public Schools Operational Review Powerpoint:
    “BPS schools are classifying students [as needing SPED services] at widely variable rates”
    “There are 14 schools who classified 100% of the students referred and 35 who referred 0%”
    “- Variable rates likely reflect differing school cultures around classification”
    “- A centralized auditing process can narrow the band of variability”

    And Elizabeth Setren’s research “documents that special needs students are now proportionally represented in charter lotteries. Even those with the highest need are close to proportional representation in charter lotteries. Furthermore, charters remove special needs classifications at a higher rate than traditional public schools and move special education students to more inclusive classrooms.
    These differences in classification practices make the proportion of special needs students in charters appear smaller.”

    One of my takeaway’s from Setren’s research is that English Language Learners seem to successfully learn English faster at Boston charter schools than at Boston traditional district schools. Would you agree? Aside from the intrinsic value to the child, family, and community, there are almost immediate cost savings for the tax payer there… The state’s tuition costs decline when students become expert in English, are no longer classified as ELL, if I understand correctly. That’s a concrete example of the significant value that your model may be missing if it fails to recognize relatively rapid academic gains for charter school students.

    • Mhmjjj2012

      Correct me if I’m wrong. Your argument is charter schools for years and years did not have a student body reflective of the sending districts in English Language Learners, special education and high needs students but because some charter schools are increasing the percentages in those categories after a 2010 law was passed all studies and reports or any kind of analysis has to assume Massachusetts charter schools will reflect those percentages. Right now, the Collegiate Charter School of Lowell has 13% English Language Learners, 6% students with disabilities and 39.1% economically disadvantaged… percentages well below the Lowell Public School District’s 25% ELL, 15.5% students with disabilities and 50.4% economically disadvantaged. Honestly, what are the chances that’s going to change? And why should researchers assume such changes when it hasn’t happened in the real world? VOTE NO on Question 2.

      • sbrsb

        “but because some charter schools are increasing the percentages in those categories after a 2010 law was passed ”

        I think that somewhat minimizes the circumstance… Have you had a chance to look at the graphs on page 27 here?

        • Mhmjjj2012

          Your link was to the Massachusetts Department of Elementary and Secondary Education’s Charter School Enrollment Data Annual Report dated January 2016. On page 4 under Requirements for Enrollment is the following: “The subgroup composition of a charter school is not required to be a mirror image of the schools in its sending districts and region. Such a requirement would contradict the statutory requirement that a lottery determine admissions when the number of applications exceeds available seats. However, in 2010, the charter school statute was amended to require charter schools to develop and implement student recruitment and retention plans that include deliberate, specific strategies to attract, enroll, and retain a student population that is demographically comparable to similar grades in schools from which the charter school enrolls students. The Department must approve recruitment and retention plans and charter schools must report on and update these plans annually. When deciding on charter renewal, the Commissioner and the Board must consider the extent to which the school has implemented its recruitment and retention plan, whether the school has enhanced its plan as necessary, and the annual attrition rate of students.” So, the “strikingly clear trends” don’t amount to a hill of beans. VOTE NO on Question 2.

          • sbrsb

            “So, the ‘strikingly clear trends’ don’t amount to a hill of beans.”

            Usually you are able to convey your opinions very clearly. In this instance, however, I do not understand what your point is.

          • Mhmjjj2012

            Elizabeth Setren is a “PhD stu­dent in the Department of Economics at the Massachusetts Institute of Technology” so she isn’t a well respected researcher or well respected professor in public policy or well respected professor in education. She’s a PhD student. Ms. Setren refers to her “Discussion Paper” on “Special Education and English Language Learner Students in Boston Charter Schools: Impact and Classification” as a “paper.” But more important on page 3, “2.3 Data and Sample “Schools are excluded from the study if they closed, declined to participate, had insufficient records, did not have any oversubscribed lotteries, or serve alternative students.” How is that even a credible approach? And who cares what a PhD student came up with? I don’t care. I really don’t care. And as far as the graph on page 27 is concerned, just below it is “*The percent of ELLs enrolled in Massachusetts charter schools has steadily increased and now surpasses statewide average enrollment. The enrollment of new students who are ELLs has increased at a greater rate over time when compared to the total enrollment of ELLs at charter schools.” The DESE compared charter schools ELL enrollment to “statewide average enrollment” not to the urban public schools from which they draw their students! That’s nonsensical. Seriously, please explain how that makes any sense.

          • sbrsb

            Setren is not inventing abstruse theories in heretofore unexplored realms of quantum physics; she is gathering and analyzing data obtained from our state department of education. And seems more than qualified to do so, expertly. Just as you may well be in respect to that particular charter school in Lowell. It appears that Professors Joshua Angrist, Parag Pathak and Amy Finkelsteins are among those willing to provide a reference if you’re looking to hire, or make her a keynote speaker at an upcoming event..

            In respect to the page 27 charts provided by DESE, please feel free to erase the graph lines from your mind in respect to the traditional public schools (though they’re perhaps not entirely irrelevant when one considers that charter schools have 3.9% of the students statewide and receive 3.9% of the tuition). But keep in full focus the trend line in respect to new charter school ELL enrollments statewide. And let us know if you think that is somehow irrelevant to the Hirsch/Calzaretta assumption that “current patterns continue” which is itself linked to current static enrollment data that reveals neither highly significant trends, nor tendencies of public district schools to classify children far more often as being ELL or needing special education than do charter schools evaluating youngsters with the same characteristics.

          • Mhmjjj2012

            You keep making the case to VOTE NO on Question 2. The data PhD student Setren gathered excludes charter schools most likely to bring charter schools results down such as: “if they closed, declined to participate, had insufficient records, did not have any oversubscribed lotteries, or serve alternative students.” That’s the point. Pro-charter schools studies, reports, and papers have to keep their focus narrow. The 3.9% of the charter schools students statewide receive 3.9% of the tuition but do not reflect the sending public schools districts costs in ELL and special needs. That’s a drain on public schools funding. Leaving public schools with a higher percentage of students with higher costs but less funding. I’m pretty sure PhD students with one pro-charter schools “paper” under their belt would only be keynote speakers at pro-charter schools events. And if “current patterns continue” does that mean charter schools will have in a few years only 100% ELL, low income and special education students? That’s not even possible because that’s not how charter schools operate. VOTE NO on Question 2.

          • sbrsb

            Setren indicates that the schools that were included: “account for over 86.1 percent of Boston charter enrollment in the 2013-14 entry”. You are critical of her research because it didn’t also include charter schools that had closed as of 2005 and 2009?

            Please keep in mind that, according to the paper’s footnotes, the only school excluded for not having an oversubscribed lottery was Up Academy Dorchester. And the only school excluded for serving alternative students was Boston Day and Evening Academy Charter. And neither of those schools is a Commonwealth Charter schools the type that is the subject of Question 2. They both are in-district Horace Mann charter schools.

            “3.9% of the charter schools students statewide receive 3.9% of the tuition but do not reflect the sending public schools districts costs in ELL and special needs.” When comparing tuition statewide, it’s worth noting that, statewide, charter schools are enrolling a disproportionately low-income, high ELL population.

            In respect to making local comparisons of tuition and special needs between sending and receiving school systems, it is important to recognize that charter school tuitions are adjusted higher or lower in direct proportion to the number of low-income/economically disadvantaged and ELL student that they have enrolled each year.

            Within the eight cities with the highest charter enrollment, on average according to the recent MTF study, the tuition per student is very slightly lower in the charter schools (7.1% vs 7.0%), presumably reflecting a marginally less high needs population (though grade levels also affect tuition rates).

            In respect to children with disabilities (where tuition rates could be less reflective of populations served), that DESE Enrollment Data document that we’ve been referring to on page 10 cites highly reputable research:

            “Because identification practices can vary between schools and districts, particularly with respect to special education, it can be helpful to examine the proportions of applicants to charter schools that are classified as English language learners and students with disabilities at the time of application to a charter lottery. A recent MIT discussion paper by researcher Elizabeth Setren, Special Education and English Language Learner Students in Boston Charter Schools: Impact and Classification, does just that for charter schools in Boston. Setren finds:
            “By Spring 2014, students across the pre-lottery levels of special education classroom inclusion and English language proficiency are, for the most part, similarly represented in charter lotteries and BPS. Small gaps remain for substantially separate special education students in middle school and high school and for beginning English speakers in high school.”

            So small gaps remain, but given earlier, wider gaps, we would anticipate that enrollment overall will continue with somewhat larger gaps for a few years.

            Given all that, what would you suggest as most fair? Charter school tuition should perhaps be two or three tenths of a percent lower rather than one tenth of a percent lower, in those eight cities with the highest charter enrollment? At least for a few years?

            Or would it perhaps make sense to keep in mind that according to Stanford CREDO research, the results for the typical student in a Boston charter equated “to more than twelve months of additional learning per year in reading and thirteen months greater progress in math. At the school level, 83 percent of Boston charter schools have significantly more positive learning gains than their district school peers in reading and math, and no Boston charter schools were found to have significantly lower learning gains.” Should we keep in mind that that’s in substantial part due to longer school days and/or more school days? And perhaps there should be an upward tuition adjustment to reflect that?

          • Mhmjjj2012

            Something else to keep in mind on the PhD student’s footnotes is #5 “Uphams Corner Charter School closed in 2009. Fredrick Douglas Charter School and Roxbury Charter High School both closed in 2005.” Why did those charter schools close? Those charter schools were excluded from the study because they closed but there was absolutely no mention of why those charter schools closed. Don’t you find that more than a little odd? So I looked up those charter schools. Uphams Corner Charter School didn’t just close…according to a Massachusetts Department of Elementary and Secondary Education press release dated January 27, 2009: “The Massachusetts Board of Elementary and Secondary Education voted 6-2 on Tuesday to revoke the charter of Boston’s Uphams Corner Charter School, citing unmet conditions and a steady decline in performance.” And guess what other charter schools are cited at the bottom of that press release for having their charters “revoked or not renewed?” Frederick Douglass Charter School and Roxbury Charter High School in 2005. For that PhD student to state those charter schools simply “closed” with no further explanation is the understatement of the year. This is the kind of deceit in pro-charter schools “studies,” “reports,” and “papers” that is deplorable… downright deplorable. Stay tuned. I intend to respond to your comments in more detail.

          • Mhmjjj2012

            The “Stanford CREDO research” relies on comparing real charter school students to “virtual” or not real public school students. CREDO acknowledges this approach is “a quasi-experimental study design.” And guess what the study found? The real charter school students showed more learning gains than the make believe public school students. It’s true. You can’t make this charter school stuff up. Real charter school students vs. pretend public school students. That’s what passes as a charter school “study.” VOTE NO on Question 2.

          • Mhmjjj2012

            “The recent MTF study” is the Massachusetts Taxpayers Foundation’s report “Public Education Funding In Massachusetts: Putting Charter Schools In Context” that was “made possible by a grant from The Boston Foundation” a pro-charter schools nonprofit known for financing “studies” that always make charter schools shine. There didn’t appear to be anyone listed as preparing the report just the “Massachusetts Taxpayers Foundation” which is unusual. It’s a typical pro-charter schools report that glossed over the loss of funding from public schools to charter schools. The public deserves an informed debate on charter schools starting with the fact charter schools drain funding from public schools. Instead charter schools proponents have done an exceptional job of limiting the debate to “choice,” “lift the cap,” “wait list,” etc. There’s no discussion on charter schools in Ohio and Florida just to name two states with serious issues or online charter schools or the pro-charter school agenda.

          • Mhmjjj2012

            Regarding Elizabeth Setren, nowhere in her charter schools “Discussion Paper” does she describe who she is…nowhere. Only her name appears on the front page and again on the abstract page. In fact, there’s a “*” next to her name on the abstract page leading to a note extending appreciation to a list of people for their assistance and at the end of that note is: “This work was supported by a National Science Foundation Graduate Research Fellowship.” So that means “Setren” wasn’t even a “PhD stu­dent in the Department of Economics at the Massachusetts Institute of Technology,” as she currently lists herself, at the time she did the charter school paper. Unbelievable. Shame on the MIT Department of Economics School Effectiveness and Inequality Initiative for not ensuring the author is properly and clearly identified on papers released under its banner. What passes for charter schools studies, reports & papers, who prepares them and who finances them, just takes a little effort to see how ridiculous they are.

    • jeanabeana

      trends are just that “How can you properly make projections without taking into account strikingly clear trends like what you see in the charts in that document? ” they go up and down…. they are often FADs. The charter school proponents want schools like grocery stores or hardware stores that will “go up and down” and will close if they don’t work because they have an ideology of the “free market” model for everything in the world. It is the wrong paradigm to be used in education and human services. They think of it as “gambling” — you start out all the charter horses in the race and then you see which ones stumble and fall, which ones die on the course and which ones “win” and they call that “success”…. It is not the way to operate a school system; hospitals and schools are responsible for all the people in their population/sample NOT just the winners who are supposedly the superior beings. Free market is a euphemism; every corporate lobbyist has a thumb on the scale … look at Climage Change for another example of using the free market (lie) as a description or Chaos theory where you can’t rely on bell curves for everything.

  • Jack Covey

    The latest is that over $21.7 million of out-of-state money from the most
    ruthless capitalists who have ever walked the Earth — Eli Broad, the Walton
    family of Walmart, Wall Street hedge fund managers, etc. — is pouring into
    Massachusetts to pass Question 2.

    Read this well-researched article here for that $21.7 million figure:

    These profit-minded plutocrats who are pouring in this money obviously …

    — do not live in Massachusetts,

    — have no children, grandchildren, or other relatives that attend public
    schools in Massachusetts

    — have never given a sh#% about the education of middle or lower income until
    recently, when they realized they could make a buck off privatizing
    Massachusetts schools via the expansion of privately-run charter schools,.

    They want to these corporate charter schools to replace truly public schools
    — the ones that, for generations, have been accountable and transparent to
    the public via democratically elected school boards, and which are mandated to
    educate ALL of the public… including those hardest or most difficult to
    educate … special ed., English Language Learners, homeless kids, foster care
    kids, kids with difficult behavior arising from distressed home lives.

    Are proponents of Question 2 seriously making the argument that out-of-state
    billionaires and Wall Street hedge fund managers are pumping in all this money
    because those folks care so much about the education of kids in Massachusetts?
    You really think they are NOT seeking a big money return on these ($21.7
    million campaign donations?

    Does anyone actually believe such nonsense?

    Can you provide an example of JUST ONE TIME in the past where they poured in
    this kind of cash to something … no strings attached, and with no expectations
    of return?

    If, as Q 2 supporters like Marty Walz claim, the most ruthless capitalists that
    have ever walked the Earth are now kicking in this kind of cash to pass
    Question 2 merely because they care about children’s education —

    … and if they are not about their profiting through the privatization of
    public schools brought about by the expansion of privately-run charter schools,

    … then I’m sure one of you Q 2 supporters could google and find a past example
    where they have done something similar .. .again out of generosity… with no
    expectation of an eventual monetary return…

    Something like …

    “Well, back in 2000-something, or 1900-something, these same folks donated
    $20 million to the (INSERT CHARITABLE CAUSE HERE). Here’s the link that proves

    No, I didn’t think so. When this was brought up in a debate, Mary Walz
    refused to address it, saying, “We need to talk about the kids, not the
    adults.” Well, keeping money-motivated scum from raping and
    pillaging Massachusetts public schools IS CARING ABOUT THE KIDS, Marty!

    So the real question is:

    To whom do the schools of Massachusetts belong? The citizens and parents who
    pay the taxes there?

    Or a bunch of money-motivated out-of-state billionaires and Wall Street hedge
    fund managers who are trying to buy them via Question 2, and the expansions of
    privately-managed charter schools which they control, or also profit from their
    on-line and digital learning products that will be sold to these charter school

    If you believe the former, THEN FOR GOD’S SAKE, VOTE “NO” ON QUESTION 2.

    Send them a message: Massachusetts schools are NOT FOR SALE!!!

  • IdaHakk

    What percentage of the per pupil spending in Charter schools goes to pay for the cost of retirees from the system who may have worked from 1970 to 2014? As much as 4% to 10% !

    What percentage of the per pupil spending in Charter schools goes to pay for the cost of students with severe disabilities? Easily 7% to 12% !

    My point is that charter school funding formulas may significantly overstate the cost of education — if those formulas fail to include a deduction for the amounts cited above. As a result, the amount of money that is “moved with the student” may be much greater than the cost that taxpayers were originally paying to educate that student in traditional public schools.

    Good work pointing out the potential fiscal disaster pending if Question 2 passes for all towns that are likely to significantly increase their charter enrollment.

  • Gary MacDougall

    DATA IS FLAWED. Only showing 2014-2015 and leaving out grades 9-12 is limiting the trends and keeping the data points to support your thesis. This is exactly how good stats people show bad news. Scrub the data.

    Move along, nothing to see here folks.

  • Patricia Nolan

    The enrollment projections are odd to me, as aCAmbridge School Committee member who follows this issue closely. Currently the three Cambridge charters schools do not fill their seats with Cambridge residents. In other words, there are RIGHT NOW seats that Cambridge residents have first dibs on but do not fill – even though we do spend a lot. So why would a new charter be added to our district – when the existing ones can’t fill their seats? The focus should be on outcomes – whether charters do better. They do. So when a district’s student opt out of traditional public schools and goes to a charter – and those students do better in a public charter school, society is better off.
    Vote Yes on Question 2. vote for Equity and to support families

  • jeanabeana

    “My comment in response to the author (wearing one of my very limited number of masks – “sbrsb”):

    you will find this guy everywhere … he has an ego need to prove he is a superior debater. There are some other ideologies he has that are apparent when you see the many comments in different places you can notice.
    he is like the guy who writes everywhere as Virginia/Brian/aspx…. they use these avatars and nom de plume to cover up what they are selling but it mostly comes out to the same thing …. going back to some ancient Viennese economists like Schumpeter Peterson who, because he puts Harvard on his resume, has to be the most superior and knowledgeable person on education (like Checkers Finn et al)

  • jeanabeana