School Choice Does Not Reduce the Cost of Education

Neal McClusky at the Cato Institute tweeted that I was wrong in my recent assertion that a school choice system costs more than a single public system. I asked him for an example of a place with a choice system that had lowered the cost of schooling, and he referred me to a couple of articles, the most thorough of which is a 2005 paper from Cato, "Saving Money and Improving Education: How School Choice Can Help States Reduce Education Costs." 

James Shuls cites some of the same studies in the smackdown he administered to me over at the Friedman Foundation blog. Shuls also proves I am not a Jedi, which strikes me as an easier sell than convincing me that choice saves money.

What May Be The Heart of the Matter

What we're going to learn here is that McClusky, Shuls, and I disagree in part because we are using the same words to mean different things.

The Cato report, for instance, says "reducing the cost of education" when it really means "reducing the amount of money spent on education by government." Can it be that just as some liberals think that government money is basically free and doesn't have to be factored into cost, some conservatives think that only government money counts.

The paper looks at several studies of school district (though on district studied is DC, which is never an example of anything) as examples of how this magic trick works. In an earlier draft I tried to walk you through each report with responses and parallelllllzzzzzzzzzzzzzz.... Yeah, it was like that. So although the following is not exactly a point-by-point response to Cato's research, I think it better captures why several favorite examples of how choice saves money are simply false.


It's About Capacity, Not Output

A school is like an airplane, not a factory. Flying that 747 from PIT to LAX costs almost exactly the same for five passengers as it does for a full flight. The idea that having one less student reduces the school's costs by-- well, by anything at all-- is just insupportable (nor, in fairness, does the Cato paper try to support it). At most, if we move twenty-some students who are all the same grade out of the district, we might be able to lose a teacher.

Cost-per-pupil figures are meaningless. It's a statistical construct, like saying someone dies in a car accident every twelve minutes. If I reduce the number of pupils in my district, my cost per pupil just goes up. There will be occasional break points, where I can shed a teacher, an administrator, or in extreme cases, a building. But if my cost-per-pupil is $10K, that doesn't remotely mean that reducing my pupil population to one student would mean I could run the district for $10K.

So let's say my school district serves 100 students at $1 million total budget. You take ten students to your charter.  My district is still spending, say, $980,000, and your charter is spending, say, $200,000 to educate your ten. Total cost-- the actual amount of money being paid by somebody-- to educate the 100 students has gone up $180,000.

Where Does the Extra Money Come From?

"Not from the taxpayer-by-way-of-the-government" is the important answer that we're looking for.

The real answer varies with situation. In many communities, private school means parochial school. Catholic school tuition is generally way below the cost-per-pupil in a public school. But the tuition cost is also not sufficient to keep the school running. Hence the fund-raising fairs and sports ticket raffles and slice of the collection plates that all help fund the parochial school systems.

If we're talking fancy-shmancy private school, the answer is Mumsy and Dad. If Lillywhite Academy will even accept your voucher, the voucher will only make a small chink in tuition costs. Tuition at the exclusive Ivy Preps like Philips Exeter will cost you four times the public school cost-per-pupil. Onviously not everybody is in the Ivy Prep league, but nobody out there is making a serious attempt to run a top-notch private school on $4 K a head.

No, the OTHER Extra Money

The public school money. Because, in my example, the public district still needed $980,000 to run, but they were down to $900 K because of the lost ten students. The public schools will recoup that same way as always-- increased local taxes.

And If The Extra Money Doesn't Appear

There is a scenario in which the choice set-up does reduce total costs, but that's not truly a function of choice-- it's a function of slashing a school district's budget thereby forcing it to cut programs. So having school choice can have the side-effect of reducing educational offerings for the community as well. 


Concrete Example: Pennsylvania Paves the Road To Hell

Regrettably, I am neither a Jedi nor a thinky tank (just a guy with a blog), so my access to big baskets of facts and data is limited. Given that the Cato Institute's best reading recommendation to me was a paper from ten years ago, I'm not sure anybody else is actually loaded with real data on this issue, either. It would be nice if someone filled that gap, because I can even entertain the notion that there is a combination of numbers and price points that might make these mythical savings actually appear.

But in the meantime, I'm reduced to what I've seen and researched first hand in Pennsylvania.

In PA, we loves us some cyber-charters. And we have a funding formula that sends pretty much the full cost-per-pupil figure to the charters. In two local district, in one school year, a loss of about seventy-some students to charters resulted in a loss of about $800,000 in school revenue. In my own district, with about 1,500 students K-12, that was a brutal chunk of money, and the only way to make up that kind of shortfall (which a school cannot budget or plan for because it does not know how many charter students it must pay for until the students make the move) was to do some massive slashing-- in our case, closing neighborhood elementary schools.

Losing those students did not significantly reduce the costs of running our district at all; it simply forced us to offer fewer educational opportunities to our students.

Bottom Line

There are many many arguments to have about choice, and it's good at times to focus on just parts (I am not even annoyed that the Cato paper is all about choice cutting costs with nary a word to consider the educational effects-- sometimes you just have to focus). But the argument that choice makes education cheaper is a loser, and the fact that some very smart people with access to lots of resources have failed to throw anything convincing at me only makes me more secure in my own Jediless findings.

Running several school districts is more expensive in toto than running just one. The savings that keep being touted are really only about savings of tax dollars, and keeping taxes low. Why would I want to keep my taxes low if that just means that I'm going to be spending more of my own money on my children's education? There's an ugly conclusion at the end of this line of inquiry-- I can afford to pay big bucks for my own children to get a good education, but if we keep public schools low-budget and tax support down, I won't have to spend my money educating the children of Those People. Choice doesn't reduce the total cost that we as a country pay for education-- it just moves the cost around a little, and reduces still more of the requirement for Some of Us to spend our perfectly good money supporting Those People.

So unfortunately, I must concede that from such a point of view, there are certainly some conservatives who can get behind that.