“Cost per student” – Again

At the end of November, I wrote a post about “unit pricing” when it comes to education.   You see it brought up a lot when people start talking about the cost of education.  There’s an idea that somehow a school district should have a set cost per student.  That is, if School District A is spending $10,000 per student, it’s doing better budgeting and cost control than District B which is spending $20,000 per student.  It also creeps into “quality” measures, where you see people claiming that the higher cost is an indicator of a better education.  Which makes it a convenient target for complaints when it comes to tax money.   As I said back then, it’s not the entire story – or even correct.   In one of the local  papers, there was an story about the school costs, and the price per student.  Interestingly, the same points came up.

As I said back then, certain things are requirements.   The state requires that you have a certain base curriculum.  There’s no getting around it.  If I’m required to have 4 years of English, I have to have that whether I have one or one thousand students per year.   The state also requires that teachers have certain qualifications and certifications, so I need to have teachers certified to teach said courses.   All of which means that there’s a certain number of teachers your school district has to have, no matter what.   In New York, each district is required to have a certified district superintendent, and yes, they get paid a lot – they’re in short supply.  That’s in addition to a principal, janitors, secretaries,  etc.  Buildings have to be heated, school buses have to run, supplies have to be bought, and so on.    That adds up to a “fixed cost.”   Which is the point made here:

At Minerva, the school district has nearly 140 students and a $5.3 million budget. The cost per student is nearly $39,000.

Superintendent Timothy Farrell said the low enrollment drives up the cost per student, especially at small school districts.

These districts can cut only so much from the budget because districts must have a certain number of teachers and programs to meet state requirements, he said.

Minerva’s enrollment is low enough that it could double the population of students without an impact to the budget, he said

In other words, if they went from approximately 12 students per class to 24, the budget would still be mostly the same.   They wouldn’t have to hire more teachers or more staff, although the supply and transportation budgets might go up a little.   But what they have is 140 students,  which is why their “cost per student” is so high.  Not because they’re spending profligately, but because there’s a core minimum you have to have for a school district, a “floor” on spending.

Why are there so many small school districts?  Why not merge them?  The answer is that it’s not quite as simple as it seems.  Most of these are “central school districts – that is, they’re mergers of other school districts.  The school I graduated from was the result of a merger of 3 school districts, and a 4’th was merged into it after I left.  Where I live now, the school is the result of 4 schools merging.   In fact, as an elementary student, I went to one of them.  But there are geographic limits.  The nearest school districts now are least 25 miles away,  and there are mountain roads between them which are seriously scary in winter.

The other reason is that at one time, they weren’t small.   One of the factors mentioned in the article is something many of us have seen – the population exodus.  School districts which 20 years ago had a relatively sizable student population no longer do.  One school district mentioned has lost 25% of its student population in the last 5 years.  Even before the recession struck the region was seeing this.  Young families, and people starting families, were leaving the area, while those who remained has much smaller families.  The school districts which in the 60’s and 70’s had to expand their facilities to handle the expanded student population now have facilities which hold half the students.

The solution to it is one that most people wish  would happen.   It relies on having a stable, relatively vibrant year-round economy that offers jobs and opportunities to young families, along with housing that’s affordable.   That, unfortunately, is not what is.  Many of the industries have left, farms are having their own problems, and in some areas the land bubble pushed property prices beyond the realm of “affordable.”  It hasn’t been a quick process, although the recent recession increased it, but people have been leaving the region steadily, and the population that remains has been trending older – which means fewer children to be educated.

This is why the use of the “cost per student” metric is misleading when we’re talking about education costs.  It’s easy to come up with,  and convenient, but it doesn’t necessarily tell the entire story.  It can end up being an “apples and oranges” comparison, and isn’t a reliable measure of “quality” or “out-of-control spending.”   Unfortunately, that’s just how it is used, and people don’t stop to think that unit pricing education may not be the right way to go when it comes to making policy decisions.



Filed under Politics

9 responses to ““Cost per student” – Again

  1. Very well-written and thoughtful piece. Thank you. It highlights how we can never talk about education in isolation. The things that factor into a school’s processes–and its success and failure–are both external and internal.

    • Thank you. 🙂 I started really looking at this when a co-worker was complaining about the cost of the local school. There seems to be an idea that schools should only cost X dollars per student, and if a school has 100 students it should cost 100X, while a school with 1000 students should cost 1000X. On the flip side, there is the idea that if one is spending Y dollars per student, and Y > X, then the school is better. Neither is necessarily the case, you have to look at the specifics for a state and the locality.

  2. Any thoughts on what role having students who need extra services plays in the higher average per pupil cost for public schools versus private schools?

    I often hear conservatives argue that public schools are inefficient compared to private schools because the public ones spend far more per pupil. But my assumption is that public schools also provide important and needed educational opportunities for students with learning or other disabilities, for whom English is a second language, etc., while private schools tend to have much lower numbers of such students. So, sure, private schools appear to be more cost effective, but only because they are offloading important educational services that we should provide onto the public schools.

    Any thoughts? Is my assumption correct, or am I missing something here?


    • Absolutely it’s a factor. Private schools get to select the students they’ll accept, while public schools must accept everyone. That means that the private schools don’t have to offer special education, vocational training, or even transportation. When you can select a population of generally healthy, non-disabled students, etc., well then you don’t have those costs.

      • Thanks, Norbrook. Do you know of any estimates of how much of a role such services play in the higher per pupil costs for public schools versus private schools? For example, do we know how much public schools pay on average for such services versus how much private schools pay?

        • As I recall, in general private schools don’t pay for them – if they offer them at all. Most of the “special needs” children either attend public schools, or have some subsidization for that – in other words, the local school district is picking up the tab. That’s one of the reasons pointing to “private” versus “public” is a mistake. It’s really an apples and oranges comparison.

          My point here is that this state (NY), and really, most states, have a set of requirements that any of the public schools must meet, in terms of curriculum, teacher certification, and services. That, and the infrastructure costs, sets a hard floor on what any school district is going cost. Even if I have just one student per grade, I still have to meet those requirements. The school district in my town, 30 years ago, had more than double the number of students it has now. But the number of teachers, services, etc. needed hasn’t changed, so the cost per student “doubled.”

  3. Aquagranny911

    You are spot on WP. Most private schools won’t take children with special needs and a lot of public schools try to funnel their ‘problem’ students to other schools to keep their test scores up. I tell you this from both a professional and a personal perspective. I worked in education and social services for 30 years. I will share this personal story. A public school actually tried to hold on to one of my grandsons, despite the fact they were not meeting his needs because he test at genius level and his test scores contributed to their “Excelling” rating. He has severe ADHD, Bipolar Disorder and a learning disability but still tests in the 99th percentile on the State exams. It took a full blown manic episode and a hospitalization on the psych unit to finally get them to transfer him to a school that is meeting his special needs.

    All schools play the numbers game because of standardized testing. They try to rid themselves of non-English speakers, children with LD problems or any child who can’t keep their percentiles UP. Some schools, like the one of my Grandkiddo, will hold on to a “high performer” even if he causes other problems they can’t deal with appropriately.

    Our children are being turned into “units of performance” rather than being truly educated. I’ve watched our educational system become so broken over the years that I don’t know how we can ever fix it on the grand scale and I’m an optimist! Schools and school systems here and there are applying innovative solutions to these problems but there just aren’t enough of them.

    There has been so much of the “blame game” being applied. Blame parents, environment, teachers…find a simple immediate solution…schools spend entirely too much time “teaching to the TEST” rather than educating our children. We really can’t blame the teachers. This has been given to them as a mandate and sometimes their jobs depend on how well their students do on standardized tests. Meanwhile, many children are marginalized or just fall through the cracks of our current system

    The standardized testing “industry” has a big stake in our children’s education. They make millions maybe billions, I don’t have all the numbers, annually on producing tests and scoring those tests nation wide. They have a huge financial stake in how we educate our children.

    Please don’t think I am some kind of “anti-testing freak” Testing has its place and provides valuable information to teachers and schools about curricula and about those students who are not “getting it” or who may need more time or help to master the material. However, I do not think testing should be the primary criteria on how we judge a teacher’s performance or that of a school nor should we allow those with vested financial interests drive our educational systems.

    Sorry, if this comment got too long but this is one of my hobby horses I ride when the subject comes up. Thanks to any who take time to read it and ♥ to you, Norbrook for another excellent diary.

  4. John

    As a libertarian, while I think there are better ways of measuring costs, our schools ARE currently too costly, too bloated, and too underperforming. For instance, in my hometown of Shreveport, LA, there are way too many administrative staff, especially at the School Board. You can walk into the School Board at any given moment and see 2/3 of the employees sitting around and chatting. You could run the place identically with 1/3 of the staff, and I doubt my district is unique in that aspect. Our School Board itself has 20+ members, and would be MORE productive with 5 or 10.

    As far as teaching to the test goes, I think it’s definitely valid, but only as long as the test is good. For instance, I have no problem with teaching to the SAT, ACT, AP, IB, etc. Those are, generally speaking, quality tests that test actual intelligence/content mastery. What I have a problem with are tests like the GEE 21, the Louisiana high school proficiency exam, which the state spends many millions in designing, distributing, and administering, yet is a horribly designed and useless exam. Not to mention it’s worthless beyond state borders. I fail to see why high schools couldn’t just administer the ACT. It’s a much better test, much cheaper (since someone else has already designed it), and you can administer it in half a school day instead of collective 5 schooldays like the GEE. Make it mandatory, like Tennessee.

    I also think, and I say this as a public school graduate, that people place too much stock in traditional public schools. They refuse to close ones with critically low enrollments, terrible performance, etc. They denigrate private schools as bastions for the rich and privileged, when most aren’t. The traditional mandatory district system is also a bad idea. One good thing Shreveport has done is gradually move all of its high schools to include a magnet component. This has introduced a large amount of inter-school competition (since students can now choose their school), raising the performance of the entire district.

    And lastly, and I know this is a touchy issue, I think a lot of the blame lies with teachers’ unions. I say this only because I genuinely feel that they are protecting bad teachers at the expense of good and even average teachers. Look at merit pay (which I fully endorse, by the way, as long as the test is good and the evaluations objective and high-quality): it would raise the salaries of great teachers and even the average teacher. It would only hurt bad teachers (and there are definitely plenty of those). Not to mention that it gives a day-to-day incentive for teachers to do their best beyond simple altruism. But unions’ stonewalling on this issue, not to mention their out-of-control benefits in many states (not the case in Louisiana, but in states like New Jersey), have earned them a justifiably bad reputation as of late.

    I know people say that students aren’t products, and that real education can’t be measured, and that schools aren’t businesses, and they may be right. But I really can’t see how things such as smaller bureaucracy, better and objective tests, and higher teaching standards and incentives should be in any way contentious issues.

    • Well, that’s why I keep saying an apples and oranges comparison. The School Boards for most of New York are elected and unpaid positions with no staffs. You’re also showing your urban influence. I live in one of the most sparsely populated parts of the country, so your idea of closing an “underperforming” or sparsely attended one doesn’t work. All the schools in this area could be considered “sparsely attended,” but unless you’re willing to bus students anywhere from 30 to 50 miles to the next one, it’s not an option.

      While teachers unions do have some blame, you should also be looking at the school administrators. Here in New York, a teacher is under contract, and “on probation” – in effect, without much protection – for the first several years of their careers. They can be fired for cause or just not renewed at the end of the year for any reason, including lack of competence. The problem is that school administrators don’t really have the tools or just don’t bother. That’s why there’s a new initiative by Bill Gates to develop those tools for better evaluating teachers.

      One of the points I’ve made here in the comments is that private schools get to deny or refuse to accept students. That’s the nature of a private school, whether it’s an expensive one or a cheap one. With that comes the ability to select students who don’t need various services, thus cutting the costs. Public schools do not have that option.