In a recent post, I wrote against the temptation to seek out universal solutions to problems, particularly when the approach in question is, “just look at however they’re doing things over there, and then do the same thing over here.” This was inspired by a commenter who, in reply to one of my posts, suggested “if the goal was better education we would just do what Massachusetts and New Jersey do and avoid what Oklahoma does”, and then repeated the same claim almost word for word in reply to another post a few days later. I argued that this kind of approach is unlikely to be successful because there’s no intrinsic reason to think whatever works in Massachusetts must therefore work equally well everywhere. There’s simply too much variation from state to state, district to district, or even student to student to assume that there is a one-sized-fits-all approach that just needs to be universally imitated.

When I wrote that post, there was a particular response I was anticipating, and as predicted it appeared as a question in the comments. If we ought to be cautious about universal solutions, and shouldn’t expect them to work everywhere, then “would this not also apply to school choice/vouchers?” After all, isn’t school choice itself offered as a universal solution? Indeed, we can expand this question even further – does that not also apply to the market? Economists are known to approach issues like population density and housing development by saying the “free market is the best way of determining what sort of housing density is appropriate.” Doesn’t that suggest that the market is being treated as a universal solution?

No, and here’s the difference – “the market” is not in itself a solution for anything. Instead, markets are set of conditions allowing solutions to be created, tried, and voluntarily adopted. In a similar way, “science” is not a solution to any particular problem. If a new disease begins to spread, the solution to that particular problem is not “science.” The solution will be something more specific – a new kind of vaccine or new antibiotic that deals with the particular issue. Science per se isn’t the solution – science is a method used to find solutions. Markets, too, are not a solution – markets are about allowing the emergence of various different solutions, as well as a means by which those solutions are vetted, through competition for the voluntary adoption of a proposed solution by people who, through direct experience and based on their specific circumstances, prefer it against various alternatives. 

In the same way, school choice is not a “solution” to the various problems of education, because school choice doesn’t mean implementing any particular approach in any particular place. Advocating for a greater role of school choice is not at all like saying “just do what New Jersey does,” or assuming that particular aspects of different curriculums will be equally effective if integrated into different districts.   

Another difference between the universal-solution mindset and maximizing the scope for varied, competing solutions was revealed in another question that came up in the comments, where it was asked, “If your goal was to improve your child’s education why not improve the quality of all the schools?” One problem with this question, as Chris Freiman points out, is that even if you agree that improving the quality of all schools is the goal, opponents of school choice would still need to provide a “good reason to effectively trap students in underperforming public schools *before* they get fixed.” But more fundamentally, this question only makes sense in a mindset described by the late Jeffrey Friedman as “naïve realism,” which I discussed in detail in two different posts as part of a series reviewing his final book.

To ask “why not improve the quality of all schools” makes sense only if you assume that improving the quality of all schools is a solved problem with a known solution. Friedman described polling data showing that many citizens believe “the reason social problems persist is that elected officials have ‘the ability but not the will to take care of the nation’s problems.’ The ability was, for them, the easy part, or so it seems; the hard part was the will.” As a consequence of this belief, those same citizens naturally concluded that unsolved social problems were proof that “officials had bad intentions, not inadequate knowledge, such that they deliberately, willfully declined to solve problems they knew how to solve.” The possibility that policymakers simply don’t know how to solve an issue is not given serious consideration in the mindset of naïve realism. The same applies here. If you don’t think that improving the quality of all schools is a solved problem with a known solution, then it’s pointless to ask why it’s not being done. It’s not being done because nobody knows how to do it. 

I don’t believe there is anybody who possesses the four kinds of technocratic knowledge needed to improve the quality of all schools. Nor do I think the answer is anything as simplistic as saying “Look at systems that are performing well. Are there aspects you could integrate into your system?” Maximizing school choice is grounded in the idea that there isn’t a single, universal solution to how to best run a school system, or even that aspects of one school’s system that will work well in another system, or with different student populations. The advocate of school choice says “This is a complex issue. There are many different solutions that can be tried, and no one approach will be the best for any system or even any student. So the best approach isn’t to try to universalize some particular school system, or to try to mimic different parts of different programs piecemeal expecting them to work the same way everywhere. The best approach is to let a thousand flowers bloom. Let’s try to maximize the different kinds of programs that are made available to students and maximize the ability of students and families to try these different programs, rather than just assigning students to a single, predetermined program based on their zip code. The more different approaches are available, and the more students have the ability to try these different approaches, the greater the odds that any given student will be able to find a system that best suits their own needs and learning style.”  

This is also true of markets. Markets are not a universal solution – markets are beneficial precisely because there is so rarely a universal solution. The market is a system that allows a great multiplicity of solutions to be offered, and the success of any given solution depends on it being voluntarily adopted by people who through direct experience find it most satisfactory compared to the alternatives. To use a government system rather than market system is to intrinsically reduce the scope of options that may be tried. Sometimes that’s just fine, like in cases where there really is an essentially universally best approach that can be reliably known. For example, a system that universally assigns things like “murder” and “arson” to the “not allowed” category will get no complaints from me. But most of life doesn’t work that way, including for issues like education or population density. This why I advocate for a system that maximizes the ability to offer different, competing solutions for voluntary adoption.