Should Voucher Schools Be Allowed to Use Admission Requirements?
Thanks to Cato's Jason Bedrick for being a good sport in responding to my prior post. While I still strongly disagree with what he's saying, I think this is useful because our disagreements are now getting a bit more precise.
To rehash, our debate is about whether excessive regulation explains the negative effects of the Louisiana voucher program. I still see no evidence, suggestive or otherwise, that this is the case.
Let me reiterate my earlier concerns with how Bedrick is interpreting the Wolf and NBER studies: The fact that private schools with less enrollment growth were more likely to accept vouchers tells us nothing about whether regulation drives out higher-performing schools. Any organization that is losing revenue will try to make that up in other ways, such as accepting voucher students. This is true regardless of regulation. So, no suggestive evidence there.
He is on at least somewhat stronger ground with Wolf's very interesting survey of private schools. First, let's accept his focus on why non-participating schools say they aren't participating. Even in that case, the top concern was the threat of future regulation, which implies that they are more worried about what might happen next, not about the existing regulations that are at issue. Of course, no government can credibly commit not to regulate in the future.
The one thing we may agree on is Bedrick's statement that: "Nearly all of the nonparticipating schools' top concerns related to the voucher program's ban on private schools using their own admissions criteria . . . or requiring schools to administer the state test." To be more precise, 3 of the top 5 concerns, and 6 of the top 11 were related to admissions criteria (again #1 is future regulation). The testing requirement seems secondary as it does not show up anywhere in the top 5 concerns.
As I said at the Cato panel, private schools don't like admission requirements because they don't want to accept the low-income students from low-scoring schools who are eligible for the program. The Wolf survey affirms that (albeit in somewhat nicer terms). Private schools want students who are more like the students they already serve who, not coincidentally, are already doing reasonably well and can already afford private schools.
While we agree on the effect of admission requirements on private school participation, we probably disagree on the implications. The Louisiana voucher program was specifically designed to provide choice and options for high-quality schools to disadvantaged students. If Louisiana were to decide to allow private schools to take voucher and use admission requirements, it would effectively redefine the purpose of the program. The state has the right to do that, but it seems questionable to think about the purpose of the program as a "regulatory burden."
Finally, even if we did count admission requirements as a regulation, there is still no evidence, suggestive or otherwise, that the regulations explain the negative effects. This part isn't really Bedrick's fault. It's just the nature of the beast. We don't have data on the effectiveness of private schools that are not participating--because they are not participating.
See Part II of my response to Bedrick next week.