Alex Tabarrok and Dani Rodrik have moved from arguing about industrial policy into arguing about the blinkers (or otherwise) of libertarianism.
It is in that spirit that I have been mulling about the derision and incredulity with which my recent post on industrial policy was met among some libertarian bloggers. … The real revolutionaries here are the libertarians. They envisage a real good world out there that looks like nothing we have now (or have ever had), and they want us to get there. Second, there are really deep philosophical differences here that have nothing to do with economics per se. Most importantly, I believe government can be a force for good; they do not. But third, libertarians hold on to their priors so strongly that they seem impervious to evidence. They shrug off the fact that there is more freedom and more wealth in those parts of the world where the government is stronger, not weaker. With respect to industrial policy proper, they refuse to engage with the fact that every nation that has grown rapidly has made use of it. I look at the world and see some government programs that work and others that fail. I want to understand what determines these outcomes, and to know how we can improve the ratio of the first to the second. When libertarians look at the same programs, they see one wreck after another.
Dani Rodrik responds here to my pointed remarks on his argument for industrial policy. Rodrik’s response, however, is along the same lines of his earlier – “I’m sophisticated, your simplistic” – post on why economists disagree. In this case, it’s ‘libertarians are ideologues who are immune to evidence.’ Rodrik, however, has painted himself into a corner because he cannot at the same time say that the “systematic empirical evidence” for market imperfections in education, health, social insurance and Keynesian stabilization policy is “sketchy, to say the least” (also “difficult to pin down” and ‘unsystematic’) and also claim that libertarians are ideologues who are immune to evidence. Say rather that libertarian economists are immune to sketchy, unsystematic, difficult to pin down evidence. Rodrik is thus right that he is “not as unconventional as I sometimes think I am. The real revolutionaries here are the libertarians.” The libertarian economists are revolutionaries, however, not because they are immune to evidence but because they respect evidence so much that they are unwilling to accept “conventional wisdom” simply because it is conventional.
My tuppence worth: I think that Rodrik is too quick here to lump all libertarians together (one of the lessons that I’ve had to learn over the last few years in the blogosphere is that you can’t and shouldn’t do this). Some libertarians, most notably Tabarrok’s co-blogger Tyler Cowen, have been willing to argue that government can be a force for good, and that other libertarians need to accept this (n.b. however that Tyler argues that the “vice” of “retreat[ing] to a mental model where the quality of government is fixed and we compare government to market” is widespread among libertarians). On the other hand, contra Tabarrok, I think it’s nearly inarguable that libertarians are far from ‘immune’ to “sketchy, unsystematic, difficult to pin down evidence” when that evidence seems to support the generic claim that markets are better than governments. And this results in exactly the kinds of unwillingness to re-examine priors that Rodrik is talking about; the superiority of markets to other modes of social organization is taken as a given. The best and most pointed take on this that I know of was written by Cosma Shalizi, natch; it’s worth quoting in extenso :
Libertarian capitalism … is a curious ideology in many ways. The one which concerns us today has to do with its advocacy of capitalism. On the one hand, the sanctity of private property and private contracts is held to be a matter of inalienable natural right, guaranteed by the fundamental facts of morality, if not a basic part of Objective Reality; capitalism is the Right Thing to Do. On the other hand, much effort is devoted to arguing that unfettered laissez-faire capitalism is also the economic system which will produce the greatest benefit for the greatest number, indeed for all, if only people would just see it. Natural right therefore coincides exactly with personal interest. A clearer example of wishful thinking could hardly be asked for. It’s not hard to see what function this plays, rhetorically. Many people who are not persuaded by the natural right argument can be lead to go along with libertarian proposals by considerations of economic efficiency. (I imagine the number of people who are unpersuaded of the economics, but buy the sanctity of property, is much smaller.) …
Now, I am the last person to deny that the invisible hand is a very powerful and valuable concept, and I’m certainly not going to deny the fundamental theorems of welfare economics; Debreu’s Theory of Value is one of my favorite books. Under certain precisely specified mathematical conditions, perfectly competitive markets inhabited by perfectly rational agents will allocate scarce resources in ways which cannot be altered without making some people worse off. Whether those conditions are satisfied by any economic system in the real world is an empirical question, and the answer is of course No. Given that those theorems do not apply, the efficiency of markets is another empirical question, or rather a whole series of questions, with answers depending on the market and the tasks they are being asked to perform. There are many situations where markets are a very valuable and powerful social technology, a useful way of coordinating actions, allocating resources, and eliciting valuable efforts. … There are other situations where they produce awful, even perverse results, and still others where they’d never begin to get off the ground, like funding basic research or national defense. …
Now, if the empirical track-record of what are conventionally called free markets is decidedly mixed, there are three courses of action open to the libertarian. (1) Embrace the natural-liberty argument wholeheartedly, and say that we should adopt laissez-faire even when it hurts us, because it’s the right thing to do. Unsurprisingly, moral austerity in defense of liberty finds few takers, though it has some. (2) Argue that the empirical track-record of alternative economic arrangements is actually no better than that of free markets (that, e.g., every instance of market failure is at least matched by an instance of “government failure”), so that’s a wash, and accordingly we should go with the market solution, since that respects natural liberty. (3) Argue that, appearances to the contrary, free markets really are optimal. This option, unlike the other two, is incompatible with intellectual honesty; it is also by the far the most popular, perhaps because it can be well-paid.
This seems to me to set out the intellectual problem pretty clearly. If your starting point is that markets are necessarily optimal, you never even get to the question that Cosma identifies of their relative efficiency for specific tasks, and thus can’t begin to participate in the debates that Dani would like to see happening. There’s no doubt in my mind that some libertarians (of the Economist leader writing variety) believe this most of the time, or purport to believe this in public. I don’t think that Alex is this unsophisticated – I imagine that he would agree in principle that markets are likely to be inefficient in the real world. However, I’m not sure what kinds of empirical evidence might lead him to concede that government may be better than the market at solving a particular problem in the real world, apart perhaps from the very minimal problems of protecting private property etc where most libertarians agree that the state can play an important role. This isn’t to say that there mightn’t be be such evidence; rather that after a few years of reading his blog posts, I’m not sure what it would be (I would be very interested in finding out if he’s willing to respond), and that I think that an explicit openness to having one’s mind changed, even on very important priors, by evidence, is an important pre-requisite to real debate.