Friday, September 3, 2010

In which Mises demonstrates that he does not understand externalities or the arguments of economists who talk about externalities

Commenter Barbarossa asks me for my opinion on this selection from Mises on externalities, as well as examples of cases where I think Mises is vague and confused. Thankfully, I'll be able to kill two birds with one stone.

Mises actually starts off quite well. The whole first half of the piece is fine (with the exception of the first paragraph which offers an odd little romanticist, inverse-Marxist genesis of the state). I was initially thinking my only comment would be to raise a few critiques of the coercion of tort law. Mises leads with a pretty standard explication of externalities, but then it all starts to head south when he lists "two alternatives" facing a market actor considerin a positive externality. This was the tip-off for the degeneration of the piece, and this is precisely where Mises could use some of Tom Sargent's precision. Mises offers two options:

1. The agent enjoys the project in question so much he'll do it anyway even if it means giving away free benefits to the "gullible masses" (Mises's words lower down - not mine), or

2. One person alone can't benefit enough from the project so it's not done unless many people band together.

What could clean this up that math might help with - where to start. Three things come to mind:

a. A little aggregation that recognizes we're never talking about a single investment decision but a family of similar investments

b. A distinction between extensive and intensive margins, perhaps, and

c. Marginalism anyone?!?!?! Menger would be rolling in his grave if he read this selection.

It's not just that there's no math. That's not necessary. I talk about these issues all the time on here without using any math. There's the complete absence of an understanding of any of these concepts in the discussion and it shows as you continue reading.

"A project P is unprofitable when and because consumers prefer the satisfaction expected from the realization of some other projects to the satisfaction expected from the realization of P. The realization of P would withdraw capital and labor from the realization of some other projects for which the demand of the consumers is more urgent. The layman and the pseudoeconomist fail to recognize this fact. They stubbornly refuse to notice the scarcity of the factors of production. As they see it, P could be realized without any cost at all, i.e., without foregoing any other satisfaction. It is merely the wantonness of the profit system that prevents the nation from enjoying gratuitously the pleasures expected from P."

OK, here he has completely abandoned his earlier points about externalized benefits. He says essentially that consumers prefer other projects with higher internalized benefits, and then goes on to blame "the layman and pseudoeconomist" for the alleged crime of ignoring opportunity costs - when the real problem is that Mises himself forgets what he wrote just a few paragraphs earlier about external benefits. He blames other economists (or in his passive-aggressive words, "pseudoeconomists") for something as ridiculous as ignoring opportunity costs (who is guilty of this? of course Mises never says*) when he forgets the problem he identified in the first place with the consumers' choice to forgo projects with externalized benefits. He says the unnamed "laymen and pseudoeconomists" think that the project could be realized "without any cost at all". Excuse me? And you all blame Keynes for erecting strawmen? Who, precisely, thinks this? Can anyone provide an answer? Barbarossa?

He goes on to describe the mainstream response - a little ham-fistedly but more or less accurate, and then continues:

"For every unprofitable project that is realized by the aid of the government there is a corresponding project the realization of which is neglected merely on account of the government's intervention. Yet this nonrealized project would have been profitable, i.e., it would have employed the scarce means of production in accordance with the most urgent needs of the consumers."

Once again, he completely forgets what he himself wrote about externalized benefits at the beginning of this piece! The alternative project is only profitable because of the distortion of the property rights regime. Profit maximizing is not the same thing as welfare maximizing. Profit-maiximizing means that you've made all the tradeoffs necessary to maximize all internalized net benefits. Mises is essentially complaining that "these mainstream economists are wrong when they try to maximize total net benefits because now they're not maximizing internalized net benefits!". Who the hell cares about maximizing internalized benefits? In what universe does that meet any sort of reasonable welfare criteria? It makes zero sense. He goes on:

"The gullible masses, who cannot see beyond the immediate range of their physical eyes, are enraptured by the marvelous accomplishments of their rulers. They fail to see that they themselves foot the bill and must consequently renounce many satisfactions they would have enjoyed if the government had spent less for unprofitable projects. They have not the imagination to think of the possibilities that the government has not allowed to come into existence."

A twofold confusion on Mises's part here. First, he assumes strict crowding out. Of course you can never assume that. But let's say we're at full employment of all factors of production and you can assume that. Again, he's essentially saying "these mainstream economists are wrong when they try to maximize total net benefits because now they're not maximizing internalized net benefits". If he had any rigor in his writing. If he at all applied what he should have learned from Menger, that is how this paragraph would have read. That is what he is saying - and that is absurd on its face. Any undergraduate that pays attention in micro would laugh at that. The concluding three paragraphs just work through the crowding out theme, and doesn't require much treatment. If crowding out is going on and there's no legitimate externality then of course Mises is right - it's elementary.

Mises demonstrates that he (1.) doesn't understand marginalism, (2.) doesn't understand or at least can't apply the concept of externalized costs and benefits, (3.) doesn't understand what other economists argue, and (4.) makes extremely cavalier assumptions about things like crowding out. This is very sloppy economics.

I have read reams of Hayek that I've found quite enlightening. Some Hayek I thought he put the emphasis in the wrong place, and in a few points (mostly methodological, as I've said before) I think he's actually wrong. The same with Rothbard, actually. I've read a lot that I like from him - he most definitely puts the emphasis in the wrong place in many instances and he has Mises's flare of being entirely ignorant of what the opposition thinks (or at least if he's cognizant of what his opposition thinks, he strawmans it for rhetorical purposes - which may be the case). I've never had this feeling with Mises. I've read a couple chapters of Human Action, a chapter or two of Liberalism, and the odd essay of his that gets thrown up on Mises. So as Mattheus regularly points out, I really don't know Mises. But in what I have read I've come across nothing of value yet, which has not been my experience at all with Hayek, with Rothbard, with Garrison, with Ropke, with Lachmann, or with any of the GMU crowd (Don Boudreaux is slim pickings, I must admit though). I sincerely do not understand why Mises has the following that he does.

I have no inclination at all to read one of Mises's books - but please, pass on selections from him and I'm happy to read and review them like this. I like to like people. It would be great to say "hey that's a good point" to something other than his ability to start an essay by correctly defining "externality" or explain something as basic as opportunity cost. This is not a promise to read whatever you suggest - so if you suggest something, give me a good reason why.

And while I did not find much of value in this piece, I do want to express my gratitude to Barbarossa for sharing it with me.

*Mises is not the only Austrian guilty of this bizarre, uncited accusation regarding the opportunity cost of factors of production. A couple weeks back Steve Horwitz at Coordination Problem, asked "Other than EWOT [Economic Way of Thinking, an Austrian textbook], how many intro books derive the supply curve using an approach that emphasizes the rising marginal opportunity cost of alternative uses of the inputs?" My immediate reaction was "how the hell do you get a supply curve without talking about the opportunity cost of alternative uses of inputs?" I couldn't even conceive of what an alternative explanation would be. I promptly found ten textbooks that talked about it. It wasn't hard because... well... every textbook on the market talks about it. Why do Austrians think other people ignore this stuff? Why do Mises and Horwitz accuse others of ignoring it without any evidence that they do? When they spread untruths like this, you get younger Austrians running around spouting complete fabrications about "mainstream economics".

6 comments:

  1. I'd like to read the original Mises piece before I try and comment. (Hopefully get some time tonight...)

    However, as a general observation about (some) Austrians' approach to externalities, I'll quickly offer the following parallel: http://madeinpain.files.wordpress.com/2009/05/56_the_first_rule.jpg
    (Substitute in "externalities" where appropriate...)

    ReplyDelete
  2. "I have no inclination at all to read one of Mises's books..."

    You could always listen to one of them on your commute, morning run, whatever. FEE has a number of his books available for free download as I recall. Not everyone is into that sort of thing, but it is easier than trying to pick through Mises' less than clear prose.

    The thing I like about Mises is the whole sovereignty of the consumer idea.

    Anyway, if you just want to read something by Mises I would recommend this partly because it has the virtue of being rather short: http://www.hacer.org/pdf/Mises09.pdf

    ReplyDelete
  3. I want to like Mises: he is admired by many who I admire; and I appreciate the Mises Institute.

    However, I just don't "get it". I have tried to read Human Action about half a dozen times. Each time I am dissapointed, frustrated, and ultimately bored. The first 100-150 pages are really quite bad, in my opinon. After that it gets better, but there is little novel or insightful in Mises's writings.

    Mises had some good ideas, and I intend to finish Human Action. But I just don’t think he the great economist: I completely fail to understand why anyone would hold him in such high regard.

    ReplyDelete
  4. One things that really bugs me is that Mises's use of technical terms like "proof" and "tautology" is inconsistent and misleading.

    The term "proof has a precise meaning in the context of logic: a sequence of valid inferences from premises to conclusion. It is an abstract thing. A proof may exist that is never discovered, or a proof may be practically impossible to demonstrate.

    An informal use of the term "proof" is to describe an argument that convinces or compels someone to accept the conclusion. The command or challenge to "prove it!" is usually meant in this socio-argumentative context.

    Much of the time, confusing these different types of "proof" is harmless, and it is clear why the two concepts are often entangled in peoples' minds. But there are important differences.

    A logical proof offers no more reason to accept its conclusion than to reject its premises: a proof merely describes logical relations. The matter of accepting or rejecting a conclusion or whether an argument is convincing or compelling are mostly psychological issues which have nothing to do with the proof itself.

    A logical proof is truth-preserving only if it is entirely or partially circular. However, assuming one's conclusion, implicitly or explicitly, when trying to convince someone of an argument's conclusion is considered an error in most socio-argumentative contexts (i.e. begging the question).

    That Mises appears to not make this distinction in places is rather disconcerting. His claims regarding the rigourous, logical, formal, and precise nature of praxeology are not particularly persuasive next to rather elementary confusion regarding the use of terms like "proof".

    ReplyDelete
  5. For example:

    Mises claims that "fundamental logical relations are not subject to proof or disproof" because "every attempt to prove them must presuppose their validity." Yet elsewhere he states that "dedutive [reasoning] ... cannot produce anything else but tautologies and analytic judgments. All its implications are logically derived from the premises and were already contained in them." (italics mine).

    By "are not subject to proof", Mises presumably means that one cannot convince someone who objects to "fundamental logical relations" with with an argument which must necessarily presuppose such relations. The comment that follows certainly suggest that "proof" here means something like "capacity to convince". Mises states: it is "impossible to explain [fundamental logical relations] to a being who would not posses them on his own account."

    However, according to Mises's own account of deductive reasoning, all proofs must pressuppose whatever they seek to prove. So it appears there can be no proof or disproof of antyhing. Mises then cements the confusion: "efforts to define [fundamental logical relations] according to the rules of definition must fail. They are primary propositions antecedent to any nominal or real definition. They are ultimate unanalyzable categories."

    The validity of "fundamental logical relations" is provable, at least in the logical sense, precisely because the validity is already assumed in some premises -- that's just how logical proofs work! Mises's error seems to be a confusion of these two common meanings of "proof," when really they have not all that much to do with each other: identifying them as one and the same to produces many paradoxes.

    I don't know what to make of all this. Mises makes some exceptional claims about praxeology, and attempts to place it on par with pure logical and mathematical analysis. But Mises is confusing and misleading, and he leaves me feeling quite suspicious of his understanding of such matters.

    ReplyDelete
  6. I'm assuming by "classical liberal" and "pro-market" you mean libertarian, right?

    Your comment actually reminds me of one thing I have definitely appreciated about Mises - I'll blog on it tomorrow.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.