Vouchers: Building New Opportunities or Just Shuffling Students?

Andrew Coulson responds to my post below about the limitations of the vouchers-based market approach for improving the education of disadvantaged students. Coulson asserts I'm overlooking important evidence that vouchers can substantially expand the supply of quality schools serving disadvantaged students. I'm not unaware of this research, but I do not think it shows what Coulson believes it does. He points primarily to the experience with vouchers in Chile and the Netherlands.

Chile's experience does seem to show that the creation of vouchers can, under certain conditions, lead to the creation of significant numbers of private schools. But it doesn't show that vouchers lead to the creation of a significant numbers of new, high-quality schools serving poor students. The use of vouchers and growth of private school enrollment in Chile is disproportionately concentrated among high- and mid-range SES children and does much less to help poorer children: A higher percentage of children from the top SES quintile use vouchers than from the bottom two quintiles. And, the implementation of these vouchers did not improve educational outcomes in Chile, while there is some evidence suggesting that achievement of low-income students in both public and private schools declined in the 1980s following implementation of vouchers.

The example of the Netherlands is even less relevant here. In the early 20th century, when a lot of Western nations were expanding their public education systems, the Netherlands decided to deal with religious differences by creating a system that included non-government run religious schools for children of different religions. These religions schools are run by independent non-profit boards and much comply with a variety of regulations. The fact that a significant number of Netherlands kids go to privately-run schools today says nothing about the probability that implementing a voucher-like system in the U.S. today will expand high-quality educational options for low-income kids.

I will admit that I'm less familiar with the more recent programs in Denmark and Sweden, although given their age there obviously is not "decades of research" on them. If Mr. Coulson sends me a copy of his book and any other relevant materials, I would be happy to learn more about this. As regular readers of this blog know, I'm far from opposing choice, but the existing evidence I'm aware of does not bear out assertions that an unrestricted market, alone, will create significant new high-quality options for poor kids.

It's telling that Mr. Coulson had to look so far and wide to find examples here. Sure, experiments with vouchers in the U.S. have been small and few, but the evidence that is available from existing programs--including privately-funded scholarship programs--is pretty unconvincing here. The existing programs, by and large, just change where pupils are allocated among existing schools, and create modest amounts of additional space in existing schools on the margins. In particular, existing evidence suggests that, even with a huge expansion in vouchers, the profit motive is just not going to drive a lot of high-quality new schools for poor kids, because there's just not a lot of profit to be made educating poor students in the U.S., particularly if you do it well. In contrast, the charter school movement has created 4,000 schools in the past 15 years that now serve more than 1 million kids, who are disproportionately low-income and children of color.