Weekly Philo of Economics: A snark too far...an apology to Glen Weyl (and some further critical reflections)

Last week I wrote the following (in response to this):

Third, among economists under sixty the aforementioned Glen Weyl passes for being an intellectual (see here) because he is interested in past economists. And, indeed, he quotes a famous paper by Hayek (incidentally one I mentioned last week.) But without apparently realizing it, Weyl rediscovers Oskar Lange's famous response to Hayek, but Lange could not appeal to netflix. (On Lange, see here.) Weyl also says (I am not making this up) "Others, who value the libertarian tradition that has often been associated with economics, will be forced to articulate other arguments [in favor of the "market economy"], perhaps based on privacy, that are not susceptible to erosion by the increasing power of centralized computation." Now libertarianism is a beautiful American tradition, but for most of its life-time it members have (rightly) complained that economists are ignoring its collective wisdom! (I'll let others reflect on the merits of the argument from privacy in the age of Patriot act.) I apologize for the sarcasm, but it is really scary how utterly ignorant young incrowd economists are of their own past and anything related to associated disciplines.

One of our readers, "Daniel," objected to these remarks (here and here). After a private exchange with Weyl and upon further reflection, I agree that I was not entirely fair  to Weyl.  First, I continue to stand by the claim that Weyl rediscovers Lange's response to Hayek. And this matters because in contemporary economics Lange is a mere footnote to the prehistory of the First fundamental theorem of welfare economics (if he is mentioned at all in textbooks--I checked Mankiw's quickly to no avail). Second, here is how Weyl understands his own claim about what he said about Libertarianism:

I was saying that [A] the libertarian tradition has often been associated with economics, not [B] that every economist is a libertarian or that economists pay attention to what libertarians!  I was also saying [C] many economists would end up being in favor of central planning.  So my main point was that [D] economic arguments classically used to defend libertarian views within economics would get weak.[A-D added by Schliesser.]

It's clear that I failed to convey accurately Weyl's claim. For this I apologize to him and my readers. But below the fold I do want to reflect a bit more (somewhat critically and earnestly) on Weyl's position.

Let's ignore [B] and [C] in what follows. The core of Weyl's original point [D] is that with the cheapness and power of computing, Hayek's original arguments against central planning will lose their force within economics. Now, one point to make is that Hayek's arguments *always* had a funny status within mainstream economics--these arguments  never found a way into the workhorse of economic theory (i.e., the so-called Arrow–Debreu–McKenzie model). [This is not to deny that Hayekian points are alive and well within Austrian economics and are regularly recovered/rediscovered by economists who are interested in market process, complexity, uncertainty, social cognition, etc.] Let me illustrate this.

Now, one of the main aims of Hayek's piece is to argue against a certain image of physics as the appropriate model for economics and, in particular, that (a) statistics are too general ("the sort of knowledge with which I have been concerned is knowledge of the kind which by its nature cannot enter into statistics and therefore cannot be conveyed to any central authority in statistical form") and (b) tools of mathematical economics are inappropriate in helping us think about what the proper role of markets in modern society are (the "character of the fundamental problem has, I am afraid, been obscured rather than illuminated by many of the recent refinements of economic theory, particularly by many of the uses made of mathematics"). 

But even in his brief remarks, Weyl doesn't quite get these features right. For example, he writes, "In his famous 1945 article, “The Use of Knowledge in Society,” F. A. Hayek argued that despite their inequity and inefficiency, free markets were necessary in order to allow the incorporation of information held by dispersed individuals into social decisions." Behind Weyl's terminology of "efficiency"  and "inequity" stand concepts inherited from Pareto, Lange, Lerner, (etc) and put to use in mathematical economics on a daily basis. By contrast, in his 1945 article Hayek never claims about markets that they are inefficient or even inequitable.  What Hayek does allow -- making clear that he dislikes the terminology -- is that markets are not "perfect" in the sense as being developed by mathematical economics at the time. It's only when you assume a certain interpretation of General Equilibrium theorem that that claim becomes identical to a denial of efficiency and inequity. But, of course, GE has a very unrealistic conception of markets, etc. But in the paper Hayek makes clear that he thinks it is in a certain sense nonsensical to talk of markets as efficient (inequitable, etc) or not.

This is not just a case of gotcha. Hayek's arguments are fundamentally epistemic and not economic (in the sense as ordinarily understood). Moreover, Hayek's epistemology relies crucially on the significance  of (a) local knowledge ("the knowledge of the particular circumstances of time and place"), and (b) what we philosophers call tacit knowledge (Hayek quoting Whithead: "Civilization advances by extending the number of important operations which we can perform without thinking about them.") [I think Hayek also has a tendency to conflate tacit knowledge with a kind of craftsmanship.] Now, Weyl, honed in on the first of these claims. (If we grant a premise explicitly stated by Justin Wolfers in the same post that "Today, every interaction we have in our lives leaves behind a trail of data" then) modern technology may (as Lange foresaw) undermine the argument from "dispersed information" against "centralized planning," as Weyl says. That's not insignificant, and I should have appreciated this point more fully. I am a bit skeptical it is true--for information that is withheld from the algorithisms is the genuinely scarce/valuable good (and recognizing its existence becomes ever more crucial [recall this post]), and I don't see that the data-miners have a good grip on this. 

It may be possible to extend Weyl's argument and claim that Netflix's (Google's, the NSA's) algorithms just replace tacit local knowledge and can match our mutual needs. (Weyl is the author of a famous paper on how markets look in the software world.) Of course, among philosophers tacit knowledge has come under fire, so perhaps the foundations behind the Hayekian argument are indeed about to collapse. But we're still a bit removed from software just replacing uncertainty-braving entrepreneurs. (Of course, on Wall Street algorithms have replaced individual arbitrageurs--it's not clear our experiences with these vindicate the futurists!) 

Originally, I wanted to say something about [A] (recall that "the libertarian tradition has often been associated with economics,"). But this post is getting too long as is. I hope it is clear that the association between economics and Libertarian thought (as reflected by Hayek's influence in economics) is always going to be a complicated matter.