What's luck have to do with it?

Photgrapher: Carly Triballeau/afp/getty images

School Choice Fails to Make a Difference

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Read More.
a | A

This is the kind of news that school-choice advocates and skeptics alike need to pay attention to: The Economist magazine reports that a team of academic economists found that students who won a lottery in Louisiana to receive vouchers to go to the public or private school of their choice did worse than students who didn't win the lottery.

This outcome flies in the face of the predictions of many economists, who often tout school choice as a way to improve the U.S. educational system while also increasing equality of opportunity. Economists typically assume that people are rational and well-informed, and will make decisions that benefit them. If giving students and their parents more school choice hurts the students academically, then something is seriously wrong with the theory.

The Louisiana finding is just a more extreme version of a fairly typical finding. School-voucher programs rarely actually hurt students, but most produce little if any positive effect. A recent survey paper by economists Dennis Epple, Richard Romano and Miguel Urquiola, covering many randomized studies similar to the Louisiana study, found that the typical voucher program does nothing for test scores. The researchers report:

The evidence does not suggest that awarding students a voucher is a systematically reliable way to improve their educational outcomes. A…large proportion of the best-identified studies suggest that winning a voucher has an effect on achievement that is statistically indistinguishable from zero.

An example of the programs covered in the Epple et al. survey is Washington’s Opportunity Scholarship Program, which began in 2003. This program awarded vouchers to thousands of low-income children in the metropolitan area. Epple et al. report on a 2010 study finding that the program had no detectable impact on test scores in either the short term or the long term. Studies of other programs tell mostly the same story. The Louisiana study is more eye-opening than most, since it found that vouchers can be harmful instead of simply useless. But the basic conclusion -- that vouchers are not doing their job -- is the same.

The evidence isn't entirely bad for vouchers. Test scores are not the only measure of success -- graduation rates are also important. Epple et al. report that many studies find that voucher programs improve graduation rates somewhat for black students. Since one of the purposes of vouchers is to increase equality of opportunity for disadvantaged groups, this is a good sign.

But it is a disappointingly meager result for a program that many conservative politicians -- and a good many economists -- have touted as the solution to the U.S.'s education woes. Vouchers entail a major reorganization of our school system, with uncertain long-term effects -- public schools might be motivated to change in order to compete, but they will also be starved of resources. That they so routinely fail to boost test scores means that they are not living up to expectations.

Why are voucher programs failing to boost academic performance? A pessimistic explanation might be that school quality simply doesn’t matter all that much -- that academic performance is mainly determined by some combination of natural ability, personality and parental influence. Another possibility is that existing public and private schools in the U.S. simply don’t vary that much in terms of quality -- that there really just isn’t that much choice to be had.

But a third possibility is that parents, even when given the chance, simply don’t choose schools that raise their children’s test scores. Parents might be incapable of discerning which schools provide a better education. Or they might simply care about factors other than academic quality. Although evidence shows that parents do care about a school’s overall level of academic performance, that doesn’t necessarily mean that parents care about a school’s ability to boost performance. They might simply want their children to be around a bunch of smart high-achievers -- but this might not actually help their own kids achieve more.

This highlights a big and fundamental problem with the idea of school vouchers. Parents might decide that they prefer schools that let their kid get away with doing little real work. Grade inflation and other perks might become rife, lowering schools’ real educational power while the government foots the bill. This unpleasant possibility is perfectly consistent with the rational-choice theories preferred by most economists. What parents and students want from schools may simply be different from what we want them to get.

Speculation aside, the evidence is clear that vouchers are a policy with underwhelming potential. Chalk this finding up as another success for the empirical economics revolution, and for well-designed policy experiments at the local government level.  The evidence is by now fairly clear -- if the U.S. cares about academic success, policy makers should focus not on turning the school system into a marketplace, but on reforming existing schools to improve their quality.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Noah Smith at nsmith150@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net