Why Forge a Consensus? – G.C.

GC: Why do we need to forge a consensus? At this stage of the school choice effort, I see forging a consensus as making the same error of hubris that the public schools make, which is to believe that we can now start ranking the currently available options for school choice because we now have access to all the possible ideas and proposals on school choice that could possibly be developed. Believing this, we then think we should forge a consensus on the best proposal and promote that one. But I don't believe we're there yet and so I am not in favor of forging a consensus.

I agree that we lack perfect knowledge of the relative merits of alternative school reforms. Education policy isn’t like pure mathematics. We will never achieve absolute certainty. The real question is: do we have enough information available to reasonably gravitate toward some school governance and financing structures, and away from others. I contend that the answer is yes. We can consult 2,500 years of historical precedents, and hundreds if not thousands of relevant and well-designed studies of contemporary school systems from around the globe.

The biggest problem standing in the way of a consensus among school choice supporters is not that there is insufficient information to determine the best policies. The problem is the information that is available is largely unknown to policy makers and would-be reformers.

Momentous decisions have been made on far less evidence than we already have at our disposal. The Founding Fathers of the United States had much less to go on when they unanimously declared independence and later decided to establish the first democratic republic in modern times.

Having sifted through some of the historical and international evidence, I find what seem to be several clear and repeated patterns. So, before abandoning the idea of trying to forge consensus within the school choice movement, I hope readers will have a look at this data and decide for themselves if they see the same patterns, different patterns, or no patterns at all.

My contention is that if everybody in the movement examines and reflects on the evidence that is available, we will be inexorably pushed closer to a shared, realistic understanding of education markets. I also argue that making education policy without having seriously examined this evidence is tantamount to getting behind the wheel of a packed school bus and speeding down the freeway with our eyes closed.

Why Forge a Consensus? – J.P.G.

JPG: [I]t is not clear to me that everyone in the school choice movement will agree that there is really a problem here that requires a solution. Andrew’s description of "the current gridlock" is a bit overdone. Different people within the choice movement favor different strategies, but these differences are rarely acrimonious or impediments to progress. The preferred strategies by activists in each state tend to reflect local values and assessments of political viability.

There may be no need to forge a consensus because different policies can and have been adopted in different jurisdictions. This is the beauty of federalism. It may even be that there is no one "ideal" policy but rather ones that are better suited for different circumstances. And each group of state activists can view the success or failures of other state policies to shape which policies they may wish to pursue.

I agree that consensus-building, for its own sake, is relatively unimportant. But the fundamental problem I’m trying to address is not our disagreement itself, but rather the very serious ramifications of that disagreement. I want us to hash out our differences not for the sake of more convivial relations within the choice movement, but because:

  1. Some market-inspired education systems arguably work much better than others

  2. The weak performance of badly-designed pseudo-market programs will make it harder to pass well-designed reforms.

What I’m arguing is that the available international and historical evidence falsifies the hypothesis that there is no one ideal policy toward which we should work. Instead of having U.S. education reformers waiting around trying to learn from the limited body of domestic evidence, I’m advocating that we learn from the huge mass of already existing international and historical evidence. Failing to do so will have serious and unnecessarily dangerous consequences for the quality of education available to current and subsequent generations of American children.

This isn’t about agreement for agreement’s sake.

Reasons not to Forge a Consensus 1 – G.C.

GC: If we were breaking out of a POW camp, I think it would be important to conserve resources and decide on a single plan. We're not in that situation. We're fighting on many different fronts. We are fighting formal legislative battles for school choice, marketing battles for school choice to increase awareness, and lots of people in different states have different constitutional/legislative/political obstacles to overcome even if a majority of parents agree that school choice is a good idea (Illinois vs. California for example). So a choice proposal that might be a real possibility for one state could easily be a non-starter for another, and vice versa. Bottom line: One size doesn't fit all.

But these are two separate questions: 1) What sort of education system would best serve our personal needs and shared ideals? And 2) What is the closest approximation to that system that is viable in a given state at a given time? Just because the answer to the second question might not be the same as the first, or might vary from one state to the next, does not mean that we should not do our best to answer the first question. More than that, unless we seriously try to answer the first question we’ll have little hope of intelligently answering the second.

This is precisely analogous to economic systems. Just because we know we can’t get an ideal economic system established in the United States does not mean that we shouldn’t try to understand what that ideal system would look like and then try to get as close to it as we possibly can, both at the federal level and in every state in the nation. In fact, if we don’t first have an understanding of what an ideal economic system might look like, what rational basis could we possibly have for forming our realpolitik economic policy compromises?

The point of Forging Consensus is that we have to try to figure out what actually works, and why, before we can know what to compromise on and what to settle for.

Reasons not to Forge a Consensus 2 – G.C.

GC: Having seen what happened to LBJ's consensus-driven decision-making in the 1960s, I am not a big fan of his idea of a consensus: Cajoling everyone to support what he wanted. Also, having worked for 10 years in the 1970s for a company (Quaker Oats) that prized consensus-driven decisions. I know that forging a real consensus is very time-consuming and requires give and take on both sides. However, I'm not in favor of forging a consensus on school choice. I think a much better model is the one BAEO has developed, which is to be in favor of parental choice in all its forms so that parents are offered a menu of choices, not just a single dish. What a consensus says is: Here's the one item we're going to put on your menu. The consensus is, you'll like it.

I sympathize with those associations, but I don’t see that they have much to do with what I’m asking for. No one, least of all me, has the power to force choice supporters to accept policies they don’t approve of.

I simply want to ensure that "school choice" advocates realize that there is a whole world of evidence out there bearing on the very policies we are advocating, and that we ignore this evidence not at our own peril but at the peril of every child whose education we affect. What I want is a debate, along the lines of the Federalist and Anti-Federalist papers. As you say, this would be time consuming, but that hardly seems a high price to pay for potentially avoiding policy errors affecting millions of children.

Isaac Newton once said that if he saw further than some others it is because he stood on the shoulders of giants – the many great minds and the accumulated research of the generations that preceded him. Well, there are hundreds of excellent international studies out there on the relative merits of alternative school systems, a mountain of real evidence on the workings of market, pseudo-market, and non-market education. It is disturbing that so few would-be education reformers are bothering to climb up that mountain to get a better view of the policy landscape before devising their next piece of choice legislation. We are too often standing behind the ankles of giants, completely blind to what has already been discovered.

As for the idea that there is intrinsic value in pursuing many different policies simultaneously, regardless of what the evidence shows, well, I disagree. While some policies may well be complementary with one another, and while some may be better suited to one state than another, it’s a mistake to blindly embrace all choice policies as if they had equal merit. In the run-up to WWII, Winston Churchill was pretty much right on everything, and Neville Chamberlain was pretty much wrong on everything. What the U.K. needed at the time was not some hodgepodge of Nazi appeasement mixed with military buildup. Britain needed to go 100% military. When the Declaration of Independence was debated, it would not have been better if it read that "all men are created equal, but some are more equal than others, and in fact there might be cases where we want to treat them differently, so, really, whatever you want to do is fine with us."

Some tools are better than others for a given job. Some mechanisms are faster, better, and more efficient than others. Some public policies are simply better than others at achieving the public’s goals and aspirations. American society would not improve if we had a little more anarchy mixed in with our rule of law. It wouldn’t be better if we had a little theocracy mixed in with our separation of church and state. If we figure out that one education policy really is better at achieving the public’s goals and aspirations than any alternative, then telling people to adopt only a little of that policy mixed in with a lot of sub-optimal ideas is counter-productive.

Consensus May not Be Possible – J.P.G.

JPG: [I]t is not clear, as Andrew suggests, that "[our] differences of opinion rest on testable assumptions." While Andrew’s paper engages in a reasoned consideration of different approaches it inevitably rests on a set of implicit values that are not so easily tested empirically. As he admits in the conclusion, "this paper may also highlight irresolvable differences. Positions that are held out of deep ideological, religious, or political conviction are rarely subject to persuasion or compromise." People may oppose universal vouchers because they believe it is unjust to use tax money to subsidize the educational options of people who already can afford those options. This is more of a value judgment than a testable proposition. People may prefer certain regulations on schools that can participate because they are extremely risk averse and are willing to sacrifice efficiency for greater certainty of avoiding certain outcomes. Again this is a value judgment. In other words, there are tradeoffs between competing values when deciding between different choice strategies. Which sets of values people should prefer is not easily amenable to empirical examination.

There are certainly some differences of opinion that are not amenable to reason and evidence, including, as you pointed out, the morality of universality.[1] On the other hand, a great many of our differences patently are empirically resolvable. Even on the subject of universality, much of the opposition comes not from the moral judgment you describe but from the belief that programs narrowly targeted at the poor are actually the best way of serving the poor. If that belief turns out to be false, as I argue it does, at least some current supporters of narrowly targeted programs may be persuaded to abandon them in favor of something better.

Similarly, on the issue of regulation, your example assumes that government regulations necessarily provide the "greater certainty of avoiding certain outcomes" that they promise. If this can be shown to be a mistaken assumption in most cases, as I believe it can, some current supporters of regulatory "solutions" may change their positions, at least on certain regulations. It’s just a question of whether or not they can be persuaded that regulation won’t achieve the outcome(s) they seek.

There are, moreover, many other issues that are easier to resolve empirically, such as the merits of at least some direct payment of tuition by parents.

I’m not suggesting that everyone presented with the relevant evidence will be persuaded on every point discussed in Forging Consensus. But I do think that a substantial fraction of people within the school choice movement are more concerned with actual social and educational outcomes than they are with the process by which we try to arrive at those outcomes. In other words, I think enough people are open-minded on enough of the relevant issues to make a discussion of the evidence worthwhile.

Incidentally, this belief is itself subject to empirical examination. All we have to do is get folks to look at the historical and international evidence and see if anyone changes their mind.

Forging Consensus May not Be Useful – J.P.G.

JPG: Even if we had to agree on our preferred policy and even if we could use evidence and reason to identify the ideal program, political and budgetary constraints could lead to different, valid choices about compromises that one should make in order to adopt a less than ideal program. There is a large amount of uncertainty around the political wisdom of making various concessions from the ideal, making it impossible to forge a consensus around the right policy proposal. These decisions are a matter of local political judgment rather than empirical examination. They are also a matter of value judgment because different people may be willing to make different compromises to pass a program at all. Some people would prefer to have no program while others think that undesired features could be modified over time or are tolerable since they believe some program is better than none.

There is definitely truth in what you say. As I noted above, the ideal program and the closest approximation to that program that is viable in any particular state are two separate things. Clearly you are right that it is up to the people of each state to choose their own compromises. It does not follow from these observations, however, that the assessment of policy compromises should not be informed by the international and historical evidence. In fact, I argue the only way anyone can rationally determine which compromises are desirable and which are not is to look at that evidence. Cutting political deals on market education reforms without first consulting the relevant evidence on how similar compromise systems have actually worked in practice is simply foolish.

Forging Consensus Is Unnecessary – G.C.

GC: This is not a path that we need to take. For example, I don't hear legislators talking about how to provide aid to higher education by "forging a consensus" on the pursuit of just one of the following menu items: Pell Grants, GI Bill, Stafford student loans, other student loans, Hope Scholarships, tax credits, etc.

And, to make my position clear up front, I'm certainly not in favor of a consensus that I believe picks the second-best option for school choice by promoting tax credits over vouchers. I'm not enthused about tax credits as a long-term solution to the problems of K-12 education, although I can see its value in a) saving religious schools from the short-term unintended consequences of charter schools, b) reducing the impact of double payment for K-12 education, and c) generally advancing understanding of school choice. But I don't see tax credits as having the ability to improve public education and education in general the way vouchers do.

Fair enough. All I’m asking is that people who seek to affect education policy first look at the international evidence seriously, and then seek out an open debate on the relative merits of their preferred policies based on that evidence. My suspicion is that we will come closer to consensus as a result of that process. If not, so be it.

Why Assume There’s Only One Correct Policy Tool? – G.C.

GC: Why do we assume there is only one tool to solve every problem?

I don’t. My recommendations in "Forging Consensus" are not based on that assumption. Some problems certainly have many equally effective solutions. But I have looked at scores of modern and historical school systems and seen consistent flaws in those funded by governments, and to a lesser extent those funded by non-governmental third parties. Because of that, and because non-refundable tax-credits, among other advantages, minimize third-party funding, I favor them. If the evidence were different, my conclusions would be different. If the evidence showed third-party funding was equivalent or superior to direct parental funding, I’d treat vouchers and tax credits equally or favor vouchers (other things being equal).

I’m open to the possibility that I am misreading the evidence I’ve reviewed, or that I have missed some contradictory evidence. As a result, I am more than willing to look at any alternative interpretations of the cases I’ve studied, and at new case studies, that anyone would care to present.

Why Vouchers or Tax Credits? – G.C.

GC: Why EITHER vouchers OR tax credits? Although I prefer universal vouchers — with "rich" included in "universal" — one possible strategy would be to focus vouchers on low-income families, phase them out for middle-income families, and then phase in tax credits for higher-income families. The Minnesota parental tax credit/deduction, which you chose not to discuss, has features that aim the deduction at higher-income families and the credit at lower-income families. Given the demagoguery that Democrats and socialists use against "the rich," it seems this dual approach might be worth discussing.

I don’t separately discuss Minnesota’s credit because that credit is refundable, making it essentially a voucher (and hence it is encompassed for all relevant purposes in my discussion of vouchers). I don’t discuss deductions at all because they are not as potent as credits and there doesn’t seem to be any more opposition to credits than to deductions. Advocating deductions would thus not be an efficient use of our time.

I don’t recommend the hybrid system you describe because I argue that vouchers (whether for the poor alone or for everyone) suffer problems that non-refundable tax credits do not, and I lay these concerns out in Forging Consensus. I also point out in the paper the advantage for low income families of having multiple scholarship providers instead a single payer government system, which also militates in favor of tax-credit programs.

I am of course open to specific disagreement with my conclusions on this subject. In fact, eliciting such a debate is largely the point of Forging Consensus.

On the issue of universality, my understanding of the public’s educational goals is that Americans want to ensure universal access to good schools. Since the rich can already participate in the education marketplace, their access is already assured. The program I describe is thus universal in its assurance that everybody have meaningful access to the educational marketplace.

The idea of extending vouchers or tax-credits to the rich is equivalent to the idea of extending food-stamps to the rich. It isn’t done because it isn’t necessary to achieve the program’s goals. The rich can already participate in the market for foodstuffs, and they are better off doing so without any involvement from the state. It would, furthermore, make no sense for the rich to participate in the food-stamp program because they would still end up paying for their food themselves, through their taxes, and they would have to pay the expanded program’s extra overhead in addition to continuing to pay for their own food (albeit indirectly).


[1] In principle, there is no reason why moral codes cannot be based on reason and evidence, but in practice they seldom are.