Limits of Evidence from Abroad on Vouchers

Andrew Coulson responds to my response to his argument that foreign voucher programs show that vouchers will expand the supply of high-quality schools serving poor kids. His arguments still aren't convincing.

Coulson admits that Chile's voucher program, while expanding the number of private schools, has done so much more for upper-income and middle-class, rather than poor, children. But he blames this on details of the program and promises that, in time, as the number of private schools grows, they will eventually serve more poor children. Leave aside that it's not clear this would be desirable, since poor students in Chile's private schools perform less well than those in its public schools. Chile's voucher program has been in place since the early 1980s; at some point believing Coulson's promise that, if we just wait a little longer for the market to work, we'll finally start to see quality private schools serving poor kids in Chile, becomes an awful lot like believing the Bush administration when they say things will eventually turn around in Iraq. There's another novel aspect to this argument, in that voucher proponents often argue--and I think that this is one of their most compelling claims--that we need to give kids more choice NOW because they can't afford to waste time in failing schools while we try to fix the existing system. But Coulson is arguing that, well, maybe the kids need to wait a little longer for the market to work. Doesn't this "wait a little longer and maybe it will work" argument sound oddly familiar?

Coulson doesn't even engage with my argument that the situation of the Netherlands is fundamentally different from that of the United States in ways that make it unhelpful as an example here. Unlike the U.S., several European countries have systems of separate public schooling for children from different religious groups, because of the history of religious strife in this country. The way the Netherlands happened to structure this system in the early 20th century resembles what we would call a voucher. But that does not mean that introducing vouchers into the American system today will produce a significant number of new, high-quality choices for poor kids.

Coulson argues that my unfamiliarity of voucher programs in Denmark and Sweden should somehow disqualify me from talking about school choice policy in the United States, despite the fact that I've spent a significant amount of my professional career studying voucher, charter and other choice issues in this country. This is silly on its face: Exactly how many countries should one know about choice in to be qualified to comment on it? Moreover, it's disingenous: Coulson knows a lot about choice in Denmark, Sweden, etc. because he's spent a lot of time trying to find examples that will support his ideological support for vouchers, not because he's seeking a comrehensive understanding of the world's experience with educational choice from some neutral scholarly position.

I don't mean this as an attack: Questions about the role of choice and diverse delivery models in public education are both empirical and ideological questions. The ideological questions are just as important as the empirical ones, and we shouldn't shy away from them in the name of being evidence-based. But that ideological component also means that Coulson and I will never be able to come to an agreement, based on evidence, about certain issues related to vouchers, because we have very different views about the ultimate goals of educational policy. He's much more interested in expanding choice, whereas I'm much more focused on expanding the supply of high-quality schools serving poor kids. Sometimes those goals support each other. But they are not the same thing and it does not do to pretend that existing evidence shows choice alone will create significant numbers of high-quality new school choice options for poor kids.