by Mario Rizzo
There has been some important discussion emanating from Paul Krugman’s unoriginal question implicitly about the Austrian Business Cycle Theory (as well as other sectoral theories of employment shifts both during and outside of business cycles). (See Econbrowser, Marginal Revolution, Econlog, Angry Bear, for examples.)
His question, as Tyler Cowen states it: “…[W]hy, say, a housing boom – which requires shifting resources into housing – doesn’t produce the same kind of unemployment as a housing bust that shifts resources out of housing.”
The question was asked, in more general form, in 1937 in Gottfried Haberler’s Prosperity and Depression in his analysis of “monetary overinvestment theories.” This was pointed out by Roger Garrison in 1999.
I suggest an answer to the general form of the question: Why does the boom — which misallocates resources to certain sectors — not cause unemployment as labor moves into those sectors, and yet the bust causes unemployment as those resources are re-allocated?
In Hayek’s theory, the phenomenon hinges on (a) distinguishing between relative and absolute movements in the demand for labor; (b) the capital-complementarity of some labor inputs and (c) recognizing that the boom involves a temporary, but unsustainable, increase in aggregate output (that is, beyond the production possibilities frontier).
(1) In the boom there is a relative expansion of capital goods production as the interest rate falls. This is relative to intermediate stages of production closer to consumption. But there is also increase in aggregate demand for labor. Therefore labor which “should” be unemployed (searching and the like) is absorbed into stages both far from consumption, as well as in consumption industries themselves. (The latter is due to the derived demand effect of interest rate reductions as savings decrease and consumption rises.)
(2) In the bust there is a relative decline in production far from consumption. But there is also a decrease in the aggregate demand for labor.
(3) Therefore in the first case labor is not so much thrown out of certain sectors and there are more opportunities for employment overall. While in the second case there are fewer opportunities overall and labor is thrown out, so to speak.
So now why can the unemployment be prolonged?
Labor can be specific in at least two ways. First, there can be skill-specificity. Certain workers can earn higher wages in certain industries because their skills are finely-tuned to the tasks in that industry. They will wait for things to sort themselves out before accepting lower-paid employment elsewhere. Second, there can be “capital-specificity” of labor. This is the idea that labor inputs are complementary to capital inputs. Now when the bust comes and capital goods production experiences a contraction, the relevant complementary factors are in short-supply relative to labor inputs previously employed.
There may also be “secondary” problems as demand falls more generally due to unemployed factors of production and possible monetary contraction.
If labor (and economists!) knew how all of these shifts would ultimately work out in detail, wages and prices would instantly adjust and there would be no business cycle as it is usually understood. But labor, capital owners, the Fed, Paul Krugman and I do not. And so we are where we are.
(Acknowledgement: I have had some discussion of this issue with Roger Garrison of Auburn University, but he is not responsible for my errors.)