The economist Thomas Sowell turned 90 on June 30. He has made a career out of judging public policies by their actual results rather than their stated intentions and wished-for effects. As befits one of the great minds of our era, he celebrated his 90th birthday by doing what anyone would do on such an occasion, namely, by publishing a book—Charter Schools and Their Enemies, from Basic Books—analyzing charter schools, their effects, and those who oppose them.

Charter schools are publicly-funded schools that play by different rules. Importantly, they aren’t bogged down by cumbersome union contracts like the ones that make it prohibitively expensive to fire teachers. They are also accountable to the people who matter most: students and those students’ parents. If they don’t deliver results, they don’t succeed.

The first half of the book is commentary and analysis. The second is a set of data appendices that would allow readers to reconstruct the quantitative basis for his argument and, possibly, show that he is wrong.

Measuring charters’ effect is more complicated than just comparing simple averages because the pool of students who go to charter schools is very different from the pool of students who stay behind—they are chosen by lottery, but differences might still be contaminated by selection bias (the kinds of families that opt for charters might be very different from the kinds of families that don’t). To get around this, Sowell addresses each of the potential confounding factors that could explain seemingly-superior charter performance. He works to match like schools with like schools, preferably those where charters and traditionals operate out of the same building.

Sowell’s quantitative analysis is a few notches below what you would expect from a technical journal like the Economics of Education Review, but he still makes a pretty convincing case for a book published with a trade press. The well-organized data appendices are gifts to undergraduate statistics and econometrics students looking for a simple data playground. Sowell’s critics who might not trust his analysis can start with the data on which he bases his case.

Charter school critics argue that charters drain traditional public schools of the most motivated students and families. Sowell summarizes research by the Stanford University economist Caroline Hoxby showing that students who lose the charter school admission lotteries don’t do as well as those who win, suggesting that at least some of what we observe is a treatment effect of charter schools. Moreover, a 2008 paper in the Journal of Urban Economics shows that charters improve outcomes for students left behind.

After looking at the data, Sowell concludes that charter schools look like a very effective weapon against the “achievement gap” between white students and black students, pointing to the example of one predominantly-black charter school with average household income of $49,000 had higher test scores than wealthier schools with average household incomes some five times higher.

Sowell makes much of different kinds of “accountability.” There is accountability to inputs, procedures, and byzantine rules—the bureaucratic vision of monopoly education that Sowell says is based on wishful thinking. Then there is accountability that counts—or that should, at least—accountability to families for actually educating children. The proof of the pudding is in the eating, and people are overwhelmingly trying to get their children out of traditional schools and into charters where they are available.

So why don’t charters meet universal acclaim? Curiously, why is it that so many people who typically think of themselves as squarely on the side of “the children” have lined up against them when it comes to quality schooling? Sowell advises readers to follow the money, quoting former teachers’ union official Albert Shanker (1928-1997): “I’ll put it this way: I’ll start representing schoolchildren when schoolchildren start paying union dues.” It’s an ugly quote, but it summarizes an even uglier reality.

Critics will argue that charters and other school choice initiatives are not magic bullets, but this doesn’t strike me as much of an argument because there are essentially never magic bullets in just about any context. Considered in isolation, just walking for thirty minutes every other day isn’t a “magic bullet” for weight loss. This doesn’t mean it isn’t a good idea. Similarly, if charters use fewer resources to deliver better—or even similar—results, expanding access to them is still a good idea.

The achievement gap gets a lot of press in education research circles, and its causes are complex. Thomas Sowell has celebrated his 90th birthday by pointing his prodigious analytical skills at an elephant in the room. He’s fond of saying that there are no solutions, only trade-offs—but the evidence he brings to bear on the charter question suggests that expanding charter school access is a trade-off worth making.