Straw Men and Choice Regulation
Neerav Kingsland is smart, honest, and, in a positive development for ed philanthropy, is now leading education efforts at the Arnold Foundation. I enjoyed his response to my series on the dangers of high regulation of school choice (it started here, and then had parts 1, 2, 3, and 4). But the bulk of the critical response from Neerav and others seemed to focus on defeating a straw man rather than what I actually wrote. They depict me as arguing against any regulation when my post was explicitly against “high-regulation” of school choice.
Maybe the mistake was on my part. Maybe I just wasn’t clear enough, so let me succinctly reiterate here my concerns. Some of the big reform-oriented foundations have drifted toward a high-regulation approach to school choice that I think is very dangerous and counter-productive. Many of the features of this high-regulation approach can be seen in Louisiana and New Orleans. They are:
1. Families should have choices, but they should only choose among quality schools. So, a portfolio manager, harbor master, or other type of regulator should use test scores to identify who is and is not a quality school operator and eliminate from the set of options a large number of schools that appear to be sub-par.
2. All choice schools should administer state achievement tests so that regulators and families can more easily make comparisons. Besides, it is normal to expect that government funds bring with them the requirement to demonstrate performance to the government.
3. To ensure equity of access, choice schools should not be allowed to use their own admissions criteria but should be required to take all applicants or admit by lottery. They should also be required to accept the amount of state funding as payment in full. And equity would be further enhanced if we targeted choice programs toward low income students in low performing traditional public schools.
4. If private schools are reluctant to go along with this high-regulation approach, maybe it is best just to concentrate on charter schools which have no alternative but to accept whatever regulations come with state funding. Besides, private schools seem to bring with them other political problems and things uncomfortable to foundations, like religion.
In describing this high-regulation approach increasingly preferred by the big reform foundations I don’t think I am making a straw man out of their views. This seems like a fair summary of what they believe and it describes what they have put into practice in choice programs in New Orleans and Louisiana.
But I think my objections are being turned into a straw man. Neerav asks in the title of his post: “To Regulate or Not?” Joel Klein similarly tweeted: “read you as saying choice per se, no regulation, is solution.” This is not what I argued. My argument was against high-regulation, not all regulation. Here is what I was actually trying to argue:
1. Test scores are useful but are not strong enough predictors of later life outcomes to determine which are the “quality” schools that should be among the options available to families. So, regulators relying on test scores will experience false positives and false negatives if they try to actively manage a portfolio of schools. This doesn’t mean that regulators should never close a school or should ignore test results (both of which are strawman arguments), but they should be significantly more humble about what they really know about school quality. It means that they should give a fair amount of deference to parent preference and should only eliminate schools from choice programs if there are multiple indicators of failure, and exercise some human judgment about those indicators. In the end I think I agree with Ashley Jochim who argued: “It may be that government will be more effective at establishing performance floors – much like they do in the arena of auto safety – rather than driving continuous improvement in school quality.”
2. High regulation does not improve equity of access because it drives away many of the highest quality schools. Heavy regulation caused two-thirds of the private schools in Louisiana to refuse to participate in the state’s voucher program, including most of the best private schools in the state. Neerav is right that I should have addressed how equity could be achieved without these regulations. Rather than trying to compel equity of access through regulations that instead drive schools out of the program, we should incentivize equity by having student-weighted voucher amounts. That is, students who are more expensive to educate (disabled, ELL, low-income) should be provided with larger voucher amounts so that schools are compensated for taking on more expensive to educate students. Most of our concerns about equity of access are likely to be addressed if we simply empower disadvantaged families with higher resource amounts.
3. Focusing only on charter schools because they are more compliant with a high-regulation approach is a serious mistake. The evidence suggests that private school choice programs may have stronger later-life outcomes for students than charters. But more importantly, I argued: “No one knows the ideal political strategy or regulatory scheme, so having a variety of different approaches [vouchers, charters, ESAs, etc…] allows us to learn about how these different methods for expanding choice are doing. We need choice among choice.”
Just as we should be humble about using test scores to identify quality schools, we should be humble about knowing the ideal political or regulatory strategy. While I don’t exactly know the right regulation for choice, I’m pretty confident that the high regulation strategy being pursued in LA and New Orleans is a really bad idea. It deters quality schools from participating, it unnecessarily excludes schools from being options, and it is counter-productive in how it interferes with the operations of choice schools.
— Jay P. Greene