The Great Debate
December 14, 1998
"A few weeks ago," began a recent article by MIT economist Paul Krugman in the online magazine Slate, "a journalist devoted a substantial part of a profile of yours truly to my failure to pay due attention to the 'Austrian theory' of the business cycle -- a theory that I regard as being about as worthy of serious study as the phlogiston theory of fire."
Since the Ford International Professor of Economics had in fact been the subject of a recent article of mine ("Not Just Academic," November 9), and since I did scold him for ignoring Austrian business-cycle theory, I sat up and took notice. What follows are excerpts from Krugman's article and from our subsequent correspondence.
Except for calling it a theory of "overinvestment" (malinvestment is more like it) Prof. Krugman did a reasonable job of summarizing the Austrian theory.
"In the beginning," he wrote, "an investment boom gets out of hand. Maybe excessive money creation or reckless bank lending drives it, maybe it is simply a matter of irrational exuberance on the part of entrepreneurs. Whatever the reason, all that investment leads to the creation of too much capacity -- of factories that cannot find markets, of office buildings that cannot find tenants. Since construction projects take time to complete, however, the boom can proceed for a while before its unsoundness becomes apparent. Eventually, however, reality strikes -- investors go bust and investment spending collapses. The result is a slump whose depth is in proportion to the previous excesses."
But, added Krugman, this theory is "disastrously wrongheaded."
Feeling called upon, I wrote Slate the following letter, which they printed, and from which excerpts follow:
"Krugman is probably right that a proper understanding of the causes and cures of recession requires a few Keynesian insights. But it's astonishing that he apparently believes you can completely dispense with the obvious relevance of Austrian theory. Questioning that theory, he asks, 'Why should the ups and downs of investment demand lead to ups and downs in the economy as a whole?' Well, there are at least three reasons.
"To begin with, as he himself suggests, workers who are laid off from the capital goods industries that are contracting do not instantaneously find jobs elsewhere (the "frictional problem"). To that Krugman logically demands to know, 'Why doesn't the investment boom ... also generate mass unemployment?' The answer is that it could, but it's all in the timing. Usually an investment boom builds slowly enough for labor to adjust, but a contraction can sometimes happen too quickly for the adjustment to be made.
"Second, those who do lose their jobs must of course reduce their consumption, which causes layoffs in other areas of the economy.
"And third, a factor that has been commonly remarked upon in these situations, with bank loans going sour, 'bank-dependent' businesses that may even be relatively sound ventures can also be denied credit, thereby causing further layoffs, further cutbacks in consumption, and so on.
"All of this is why the story does indeed resemble what, in Krugman's words, 'actually happens in a recession, when every industry -- not just the investment sector -- normally contracts.'
"... I continue to be a fan of Paul Krugman's writings. In this case, I think his problem is simply this: He honestly seems to believe that no economic theory can be worth its salt unless somebody has expressed it in math. The Austrian economists insisted that all good theory must be put in words; ergo, they aren't worth listening to. That's truly a pity, because Krugman himself has an obvious gift for words, and he'd probably find the Austrians to be kindred spirits, if he'd only make some attempt to overcome the limitations of his training."
Krugman responded by first defending himself against the charge that he had misrepresented Austrian theorist Friederich Hayek (for which I had also faulted him), then wrote:
"As for the rest: If there was any abstruse math either in my latest piece or in my column about the baby-sitting co-op [a parable Krugman used to explain his own view of business cycles], it was well hidden not only from the readers but from the author. It's true that I do believe that an economic theory should 'add up': that it should give a consistent account of who is doing what, and of where the money comes from and where it goes. But that doesn't seem like a lot to ask, even if it turns out that some plausible-sounding theories fail that test.
"I have to admit that it's a bit dismaying, given the amount of effort I have made and continue to make to express economic ideas in plain English, still to find myself charged with being just another one of those math-obsessed economics nerds. I do think that math is a useful tool for economic thought (see my piece 'Two Cheers for Formalism'). But the trouble with Austrian theory, as with many heterodox economic doctrines, is not that it is expressed in words, but the fact that its proponents use a sort of blizzard of fancy words to cover up a fatal lack of intellectual clarity."
I then sent him an E-mail, which read in part:
"My letter praised your writing skills, as did my article, so I'd hardly call you 'one of those math-obsessed economics nerds.' But since your 'formalism' article makes clear your belief that math is the way economists should talk to other economists, I also assume you feel that no theory is worth serious attention until it's put in math.
"But here's my real problem with your response. The point of my letter was to address the questions you raised about how malinvestment can lead to a general recession. You're obviously not persuaded. Couldn't you take some space to explain why?"
He then send me back an E-mail, which read in part:
"About your explanations: Remember that economic slumps in general are not short-lived: certainly the 1930s slump went on and on. So the 'sudden shift' explanation of why the boom doesn't also create unemployment is not plausible -- and anyway, unemployment actually falls (and consumer-goods production increases!) in investment booms. To say that the investment slump depresses output because laid-off workers consume less is already to implicitly accept a Keynesian view of the world... .
"Last point: Everything you say sounds sort of quantitative -- it's about investment, consumption, employment, etc. Why then not create a simple mathematical model to illustrate the ideas? Why hasn't anyone modeled Austrian theory? The reason, I believe, is that it can't be done -- not because the concepts are too subtle, but because in a fundamental way the story does not make sense, and can only sound plausible if you rely on a fuzzy verbal exposition."
I then sent him another E-mail, which read in part:
"As for why the 'Thirties slump went on and on, let me quote from Hayek again: 'Although I do not regard deflation as the original cause of a decline in business activity, such a reaction has unquestionably the tendency to induce a process of deflation -- to cause what more than 40 years ago I called a "secondary deflation" -- the effects of which may be worse, and in the 1930s certainly were worse than the original cause of the reaction made necessary, and which have no steering function to perform.' [italics added]
"But it probably remains a bit of a puzzle that recessions both before and since the Great Depression were relatively short-lived, while that one wasn't. In any case, the inability to fully account for the 'Thirties doesn't discredit Austrian theory. Does your idea that recessions are always and everywhere an excess-demand-for-money phenomenon fully explain why some can last a year or two longer than others?
"My 'sudden shift' argument for why the boom doesn't also create unemployment says only that it doesn't have to do so if the movement of workers into capital goods industries is gradual. It says nothing about why 'economic slumps in general are not short-lived,' so it doesn't stand or fall on that basis.
"The fact that 'unemployment actually falls (and consumer goods production increases!) in investment booms' doesn't negate the theory either. There's a definite tendency for the economy to grow to begin with (call it the 'healthy' part of the boom), so unemployment could be falling for that reason alone. Consumer-goods production will also increase, although not as fast as it would if workers weren't being pulled away to the capital goods industries.
"As for the Keynesian aspects of the theory, in the first quote I sent you, Hayek did in fact write that 'unemployment may itself become the cause of an absolute shrinkage of aggregate demand' and that definitely sounds Keynesian. So there you're quite right. But let's get good ideas wherever we can find them.
"As for whether the theory can be put in math, I spoke with Auburn University Prof. Roger Garrison, who said he's expressed it graphically. He added that this might satisfy your desire to make sure things add up, an attribute he admires in your work.
"A final point: Austrian business cycle theory is a theory of malinvestment (not 'overinvestment'). The idea that malinvestment (possibly in consumer-goods industries as well) can help set the stage for recession makes a lot of sense and seems to have a lot of applicability. What else does crony capitalism mean if not that?"
* * * * *
See also Roger Garrison's compelling critique.
For a broad look at Krugmanism, see his personal website.
Also see Henry Hazlitt on the failure of Keynesianism.