Skip to content
Breaking News Alert House Speaker Kills Effort To Stop The Feds From Spying On Americans Without A Warrant

Why Megan McArdle Shouldn’t Lose Faith In School Choice Just Yet


Megan McArdle is the latest right-ish person to cast shade on school choice after a few studies found initially negative effects on kids’ math achievement in Louisiana’s voucher program. Although after a few years they leveled up again, these were the first negative school voucher effects found in any high-quality studies. Her latest column asserts “We libertarians were really wrong about school vouchers.” I partially agree with that statement, but for much different reasons.

McArdle’s reasons essentially boil down to “We’ve got a bunch of voucher programs now, but U.S. public education still stinks and it looks like many parents have different education priorities than I do.” Here are a few key paragraphs — feel free to click over and read the whole thing:

Some studies suggest that voucher programs do modest good; others suggest that they do very little; and a few suggest that the impacts are actually negative. My overall takeaway from the literature is that voucher programs probably do a little bit of good. But the emphasis is on the word ‘little’; they are not a cure-all, or even much of a cure for anything. It was reasonable to think, in 1997, that voucher programs could change the world. Now we have two decades of evidence…

McArdle then says maybe the modest academic effects studies keep finding for school vouchers happen because a new study finds that parents care more about environment:

The socioeconomic status of the students in a school is somewhat easier for parents to observe than the quality of the pedagogy. It’s not then, all that surprising that when researchers sat down to analyze parental decision-making in New York City public school, peer group seemed to be what parents were looking at. And peer group matters a great deal.

McArdle thus concludes that school choice is less legitimate because many families use that opportunity to choose a specific school culture and relationship networks for their kids. Nope. This is in fact more evidence that school choice is a success.

Are Parents’ School Preferences Legitimate?

Let’s start here by taking a look at a National Bureau of Economic Research working paper, the only research McArdle directly cites in the column. Notably, it studies not vouchers but how parents choose among public high schools in New York City. They rank their choices among those available, and a centralized mechanism uses those rankings plus schools’ own criteria to assign kids. So this system is not especially comparable to access to private schools, although the paper says parents’ real preferences are “weakly” in line with how they rank schools on the application on average.

The paper finds that parents’ top two preferences for the schools they rank on the application are home proximity and academic performance. The researchers then control for the part of achievement that comes from kids’ natural ability rather than superior teaching. This results in the conclusion that parents actually prefer schools with smarter kids (“peer quality”) rather than ones that are better institutionally at moving kids’ test scores up.

The authors say this is troubling and negates one of the promises of school choice, of improving instructional and curricular quality. Instead, they suggest choice more efficiently sorts smarter kids into better schools.

our findings imply that parents’ choices tend to penalize schools that enroll low achievers rather than schools that offer poor instruction. As a result, school choice programs may generate stronger incentives for screening than for improved school productivity…

If parents respond to peer quality but not causal effectiveness, a school’s easiest path to boosting its popularity is to improve the average ability of its student population. Since peer quality is a fixed resource, this creates the potential for socially costly zero-sum competition as schools invest in mechanisms to attract the best students.

Actually, This Shows How Smart Parents Are

Several responses. First, it seems reasonable that rating systems that look for how much a school adds to students’ test scores penalize schools that have higher starting scores, because their students have less of a gap to close. Schools with more smart kids may thus also have great instruction that shows up less on value-added measures. The reverse of this argument has been used with some success to complain about judging teachers in low-achieving schools based on student test score growth year after year.

Second, the idea that the general public knows the difference between high natural ability that produces good academic performance versus exemplary teaching that produces good academic performance is widely disproven by a number of scholarly and popular works on IQ and testing. In other words, just because parents send their kids to schools that perform better more because of the student body than the teachers doesn’t mean that is parents’ intent. They look and see high achievement and say I want my kids to be part of that. That’s not bad.

In fact, this intuition is backed up by research that finds that one’s academic output is a little like one’s gene expression: Environment can’t change the underlying ability students have, but it can either help a student reach the top of his or her ability or shunt him or her down to its lower bound.

These NYC parents aren’t stupid or prejudiced. Peer group matters. Research finds that being placed into a classroom with more misbehaving peers is likely to increase a child’s bad behavior; kids of single moms are less likely to become single mothers themselves if they live in a mostly married neighborhood; and disadvantaged students are brought up to higher achievement by being placed into classrooms with higher-achieving peers. What we need in not just schooling but society at large is for more people to aspire towards the behaviors, social networks, and attitudes of high achievers, rather than accept the soft bigotry of low expectations based on background.

All this paper really shows is that test results are highly correlated, not with school quality, but with native intelligence. This is something we’ve known since the 1960s Coleman Report. That was one of its bombshell findings, and it has since been widely and repeatedly replicated. Since the scientific debate over the extent to which achievement is hardwired versus environmental is far from settled, it seems unfair to blame parents so prematurely for making choices to place their kids at high-achieving schools.

It’s even more so when you consider that public schools have become increasingly segregated by race and income largely because richer people — who are disproportionately people of higher IQ — can and do buy their way into better, more expensive, public school districts. Eliminating programs that broaden choices to less-wealthy people does not reduce peer-group selection. In fact, the research finds school choice programs integrate races and classes far better than assigned public schooling does. So, even if we assume both that this is happening and that it is bad, a choice system is less bad on this score than traditional school districts are.

Most Programs Aren’t Designed to Offer Good Choices

If the problem is that there are simply not enough good schools to go around, well, that is something not only not well-addressed by public school systems (hence decades of searching for improvements), but it’s a key problem with existing voucher programs. They are not designed to increase the supply of good schools.

Places like the Wall Street Journal and right-leaning think tanks have wildly celebrated the proliferation in choice programs since Republicans began leading a majority of state legislatures. I think the cheer has outpaced the reality. The number of children using vouchers is miniscule, and by design, since most programs have very low participation caps. Using EdChoice and NCES data, I estimate that less than 1 percent of U.S. K-12 kids use school vouchers today. No wonder it’s not “working miracles.” It’s affecting very few Americans.

These programs are also regulated by the same people whose failures at running public school systems generated pressure for vouchers in the first place. Exhibit No. 1: Many voucher programs require participating schools to administer state-tests, which are notorious for poor quality, mostly connected to Common Core, and well-recognized as putting curriculum in a headlock of failed education-school ideas.

Exhibit No. 2: School-choice initiatives so far have not appropriately deregulated state private-school laws to better fit a choice environment. Most states regulate private schools a lot more than most people would expect, in deeply important ways such as teacher certification and curricula, as well as accreditation. The quick summary is that it’s essentially impossible to get a private school going that can participate in a school choice program from the get-go, like charter schools do. Further, if one does open or even operate an existing private school, it is pushed by regulations to behave more like existing public schools than something superior.

Thus, all school choice programs have done, besides with other counterproductive regulations imposing extra regulatory barriers to participation, is provide a modest way to fill empty seats in existing, already overregulated private schools. This is not a market that will produce hardly any new choices for families. So, again, it’s not surprising that research finds these programs do, on balance, benefit both participants and kids at nearby public schools, but that the effects are rather modest. That’s about the best in academic effects we could hope for such underfed, weed-choked tokens to the concept of parent choice in education.

There’s one area in which vouchers are already a huge, clear success, and it’s an economic area, so I’m surprised McArdle missed it. Existing programs offer slightly better long-term results for kids and society — at dramatically less expense than public schools. The typical choice program costs $3,000-$6,000 per kid. The average annual per pupil spending in public schools is $12,000. The math is not that easy, of course, but even when people smarter than me work it out you cannot ignore the fact that existing school choice structures offer a slightly better product at a deep discount.

Given that states face impending fiscal nightmares, not to mention the federal government’s, this ought to be seen as yet another major reason to dramatically accelerate the growth of genuine parent choices in K-12 education.