October 24, 2017

Public Policy after Utopia

People often ask me how the Niskanen Center’s philosophy differs from standard-issue libertarianism. Usually I say something substantive and policy-related like, “We think the welfare state and free markets work better together, and that hostility to ‘big government’ can actually be counterproductive and leave us with less freedom,” or something in that vein. That’s the sort of contrast people are generally looking for. But I’m never really happy leaving it at that.

Why not? Because this kind of answer is actually pretty superficial. It doesn’t get to the heart of the matter. For example, it doesn’t really get at what I take to be the nature of the intellectual mistake involved in the standard libertarian rejection of the welfare state.  There’s a deeper intellectual issue about how to theorize about politics, and it has nothing in particular to do with libertarianism. It has to do with the utility of something political philosophers call “ideal theory.”

Politics without a compass

Many political philosophers, and most adherents of radical political ideologies, tend to think that an ideal vision of the best social, economic, and political system serves a useful and necessary orienting function. The idea is that reformers need to know what to aim at if they are to make steady incremental progress toward the maximally good and just society. If you don’t know where you’re headed—if you don’t know what utopia looks like—how are you supposed to know which steps to take next?

The idea that a vision of an ideal society can serve as a moral and strategic star to steer by is both intuitive and appealing. But it turns out to be wrong. This sort of political ideal actually can’t help us find our way through the thicket of real-world politics into the clearing of justice. I’ve discussed the problems with ideal theory at length, in the context Gerald Gaus’ tremendous book The Tyranny of the Ideal, in a Vox column. This piece will be easier to understand if you read that first. Jacob Levy’s paper, “There’s No Such Thing as Ideal Theory,” is an outstanding complement. And, on the more technical side, the work of UCSD’s David Wiens is state of the art, and adds texture to Gaus’ critique.

A major paradigm shift in political theory is underway, and it’s all over but the shoutin’ for ideal theory. But it takes a while for the shoutin’ to peter out. New paradigms can take a generation or more to trickle down through the intellectual culture. So we’ve barely begun to grasp what it means to give up on ideal theory, especially in public policy. It’s a bit dramatic to say that the death of ideal theory changes everything, but it changes a lot. It definitely changes what it means to be an ideologically principled think tank.

If you agree with Gaus, as I do, then you will think that there’s a pretty major intellectual mistake lurking within the ideal-theoretic version of libertarianism that the most prominent institutions of the “freedom movement” were built to promote. Again, this has nothing to do with libertarianism, per se. Gaus’ argument is general. It doesn’t matter which normative standard you use to rank possible social systems. It could be the orthodox libertarian conception of freedom as non-coercion, John Rawls’ two principles of justice, or a radically egalitarian conception of material equality. It doesn’t matter. In order to say that any particular system is the best in terms of your chosen normative standard, you’ve got to be able to rank rival systems against that standard. Doing that ranking in a principled, non-arbitrary way requires evidence of what the realization of your favorite possible social world would actually look like. Otherwise you can’t really say that it does better in terms of your chosen standard than competing systems.

Utopia is a guess

The fact that all our evidence about how social systems actually work comes from formerly or presently existing systems is a huge problem for anyone committed to a radically revisionary ideal of the morally best society. The further a possible system is from a historical system, and thus from our base of evidence about how social systems function, the more likely we are to be mistaken about how it would work if it were realized. And the more likely we are to be mistaken about how it would actually work, the more likely we are to be mistaken that it is more free, or more equal, or more socially just than other systems, possible or actual.  

Indeed, there’s basically no way to rationally justify the belief that, say, “anarcho-capitalism” ranks better in terms of libertarian freedom than “Canada 2017,” or the belief  that “economic democracy” ranks better in terms of socialist equality than “Canada 2017.”

You may think you can imagine how anarcho-capitalism or economic democracy would work, but you can’t.  You’re really just guessing—extrapolating way beyond your evidence. You can’t just stipulate that it works the way you want it to work. Rationally speaking, you probably shouldn’t even suspect that your favorite system comes out better than an actual system. Rationally speaking, your favorite probably shouldn’t be your favorite. Utopia is a guess.

Again, this is a general problem. But it does hit especially hard for those who appreciate the unpredictability of complex systems and the inevitability of unintended consequences. It’s no coincidence that Gaus is a Hayekian. As my colleague Jeffrey Friedman argues, expert predictions about the the likely effects of changing a single policy tend to be pretty bad. I’ll use myself as an example. I’ve followed the academic literature about the minimum wage for almost twenty years, and I’m an experienced, professional policy analyst, so I’ve got a weak claim to expertise in the subject. What do I have to show for that? Not much, really. I’ve got strong intuitions about the likely effects of raising minimum wages in various contexts. But all I really know is that the context matters a great deal, that a lot of interrelated factors affect the dynamics of low-wage labor markets, and that I can’t say in advance which margin will adjust when the wage floor is raised. Indeed, whether we should expect increases in the minimum wage to hurt or help low-wage workers is a question Nobel Prize-winning economists disagree about. Labor markets are complicated! Well, the comprehensive political economies of nation-states are vastly more complicated. And that means that our predictions about the outcome of radically changing the entire system are unlikely to be better than random.

If your favorite system is quite a bit different from any system that has existed, then even if it were true that it would rank numero uno in terms of your favorite normative standard, you’re not in a position to rationally believe it. Clearly then, it’s not actually useful to aim toward a distant ideal when you don’t really have a good reason to believe that it’s better than actually existing systems in terms of liberty or equality or nationalist solidarity or whatever it is you care about.

This is a hard lesson for ideologues to swallow. I still haven’t totally digested it. But a number of things have become much clearer after giving up on my sinful, ideal-theoretic ways.

Analysis after ideal theory: measurement and comparison

The death of ideal theory implies a non-ideological, empirical, comparative approach to political analysis. That doesn’t mean giving up on, say, the value of freedom. I think I’m more libertarian—more committed to value of liberty—than I’ve ever been. But that doesn’t mean being committed to an eschatology of liberty, a picture of an ideally free society, or a libertarian utopia. We’re not in a position to know what that looks like. The best we can do is to go ahead and try to rank social systems in terms of the values we care about, and then see what we can learn. The Cato Institute’s Human Freedom Index is one such useful measurement attempt. What do we see? Look:


Every highlighted country is some version of the liberal-democratic capitalist welfare state. Evidently, this general regime type is good for freedom. Indeed, it is likely the best we have ever done in terms of freedom.

Moreover, Denmark (#5), Finland (#9), and the Netherlands (#10) are among the world’s “biggest” governments, in terms of government spending as a percentage of GDP. The “economic freedom” side of the index, which embodies a distinctly libertarian conception of economic liberty, hurts their ratings pretty significantly. Still, according to a libertarian Human Freedom Index, some of the freest places in on Earth have some of the“biggest” governments. That’s unexpected.

This is why we need to try to rank social systems in terms of our prized political values. Our guesses about which systems lead to which consequences are likely to be pretty bad.  Suppose we were to poll a bunch of American libertarians, and ask them to tell us which country enjoys more freedom, according the Cato Institute’s metrics. The United States or Sweden? The United States or Germany? The United States or Canada? The United States or Lithuania? I’m pretty sure almost all of them would get it wrong in each of these pairwise comparisons. Why? Because typical libertarians carry an ideal-theoretic picture of the “the free society” around in their heads, and (for some reason!) a minimum of taxation and redistribution is among the most salient aspects of that picture. And that means that Denmark, say, doesn’t seem very free relative to that picture. But there’s a great deal more to freedom than fiscal policy. And we see that, as a matter of fact, the country with the biggest-spending government in the world is among the freest countries in the world, and ranks first in personal freedom.

That is our basic data. It doesn’t necessarily imply that the United States ought to do more redistributive social spending. But when a freedom index, built from libertarian assumptions, shows that freedom thrives in many places with huge welfare states, it should lead us to downgrade our estimate of the probability that liberty and redistribution are antithetical, and upgrade our estimate of the probability that they are consistent, and possibly complementary. That’s the sort of consideration that mainly drives my current views, not ideal-theoretical qualms about neo-Lockean libertarian rights theories.

Though libertarianism is of personal interest to me, I want to emphasize again that my larger point has nothing to do with libertarianism. The same lesson applies to alt-right ethno-nationalists dazzled by a fanciful picture of a homogenous, solidaristic ethno-state. The same lesson applies to progressives and socialists in the grip of utopian pictures of egalitarian social justice. Of course, nobody knows what an ideally equal society would look like. If we stick to the data we do have, and inspect the top ranks of the Social Progress Index, which is based on progressive assumptions about basic needs, the conditions for individual health, well-being, and opportunity, you’ll mostly find the same countries that populate the Freedom Index’s leaderboard. Here:


The overlap is striking. And this highlights some of the pathologies of ideal theory: irrational polarization and the narcissism of small differences.

Some pathologies of ideal theory, both personal and political

Ideal theory can drive political conflict by concealing overlapping consensus. Pretty much any way you slice it, Denmark is an actually-existing utopia. But so is Switzerland. So is New Zealand. The effective difference between the Nordic and Anglo-colonial models, in terms of “human freedom” and “social progress” is surpassingly slight. Yet passionate moral commitment to purist ideals of justice can lead us to see past the fact that the liberal-democratic capitalist welfare state, in whatever iteration, is awesome, and worth defending, from the perspective of multiple, rival political values. We miss the fact that these values fit together more harmoniously than our theories lead us to imagine.

I suspect this has something to do with the fact that utopia-dwellers around the world seem to be losing faith in liberal democracy, and the fact that  “neoliberalism” can’t get no love, despite the fact that they measurably deliver the goods like crazy. Yet ideologues interpret this loss of faith as evidence of objective failure, which they diagnose as a lack of satisfactory progress toward their version of utopia, and push ever more passionately for an agenda they have no rational reason to believe would actually leave anyone better off.        

It is intellectually corrupt and corrupting to define liberty or equality or you-name-it in terms of an idealized, counter-factual social system that may or may not do especially well in delivering the goods. Commitment to a vision of the perfect society is more likely than not to lead you astray. Consider how unlikely it is for a typical libertarian to correctly predict more than a couple of the top-ten freest countries on the libertarian freedom index. The fact that ideological radicals are pretty unreliable at ranking existing social systems in terms of their favored values ought to make us skeptical of claims that highly counterfactual systems would rank first. And it ought to lead us to suspect that ideal-theoretical political theorizing leads us to see the actual world less clearly than we might, due to cherry-picking and confirmation bias.

If you’ve already irrationally ranked a fanciful social system tops in terms your favored value, you’ve effectively committed to the idea that the world works in a certain way without sufficient evidence that it actually does. This is almost always a commitment of identity and group membership rather than a judgment of reason. And it leads you to cast about for evidence that the world does work the way it would need to work in order to vindicate your ranking. You end up lending a great deal of credibility to comforting evidence, while ignoring and dismissing evidence that the world doesn’t work that way you’d like it to work. The result is that your ideal-theoretic commitment ends up driving your model of the world.

But if your ideal theory is likely to be wrong in the first place, using it as a filter for evaluating evidence is going to leave you with a disastrously distorted picture of the way the world actually works. And that means you’re going to make systematically terrible predictions about the likely consequences of this or that policy change. You may want to identify reforms most likely to promote  liberty or equality, or whatever, but you’ll end up really bad at this because your distorted ideological model of the world will leave you unable to evaluate evidence objectively.

Progress in policy requires idealistic moral passion without preconceived ideals  

For me, the death of ideal theory has meant adopting a non-speculative, non-utopian perspective on freedom-enhancing institutions. If you know that you can’t know in advance what the freest social system looks will look like, you’re unlikely to see evidence that suggests that policy A (social insurance, e.g.) is freedom-enhancing, or that policy B (heroin legalization, e.g.) isn’t, as threats to your identity as a freedom lover. Uncertainty about the details of the freest feasible social scheme opens you up to looking at evidence in a genuinely curious, non-biased way. And it frees you from the anxiety that genuine experts, people with merited epistemic authority, will say things you don’t want to hear. This in turn frees you from the urge to wage quixotic campaigns against the authority of legitimate experts. You can start acting like a rational person! You can simply defer to the consensus of experts on empirical questions, or accept that you bear an extraordinary burden of proof when you disagree.

I think the reign of ideal theory in political philosophy turned lots of incredibly smart, principled, morally motivated people into unreliable, untrustworthy ideologues. This has left the field of rational policy analysis to utilitarian technocrats, who have their own serious problems. Long story short, we ended up with a sort of divide in public policy between morally passionate advocates trapped in epistemic bubbles and technicians capable of objective analysis but devoid of guiding vision.

What we need are folks who are passionate about freedom, or social justice (or what have you) who actively seek solutions to domination and injustice, but who also don’t think they already know exactly what ideal liberation or social justice look like, and are therefore motivated to identify our real alternatives and to evaluate them objectively. The space of possibility is infinite, and it takes energy and enthusiasm to want to explore it. Imaginative hypothesis generation is the great intangible without which progress is impossible or maddeningly slow. The technicians, the quants, the lab rats tend to be awful at dreaming up hypotheses. Ideological moral passion is the perfect wild horse to harness; it could power the exploration of the near frontier of the feasible. But thanks to the tyranny of the ideal, it’s a source of intellectual energy more often wasted hooked up to a wagon train headed off the map to Shangri-La.

Nobody will fight for what works if the people who will fight are blind to what works, or if the people capable of seeing what works don’t have the imagination to look for it, or won’t fight for it if they see it.

I think this is what makes Niskanen different. We’ve put misguided, utopian idealism behind us, but retain the moral passion that once attracted most us to radicalism, and have channeled it into discovering and fighting for what is most likely to actually work to make our society freer, more prosperous, and more just.

Will Wilkinson is Vice President for Policy at the Niskanen Center