Wednesday, October 10, 2012

Repapis on Hayek’s Business Cycle Theory

C. Repapis has the following critical article on Hayek’s Austrian business cycle theory (ABCT):
Repapis, Constatinos. 2011. “Hayek’s Business Cycle Theory during the 1930s: A Critical Account of its Development,” History of Political Economy 43: 699–742.
I provide a summary below, and my own thoughts on the Hayekian ABCT.

I will divide the discussion into two sections: problems with Hayek’s use of (1) general equilibrium theory and the role of expectations, and (2) Austrian capital theory.

I. General Equilibrium Theory and Expectations
Repapis notes that Hayek used general equilibrium theory as the fundamental theoretical framework for his business cycle research (Repapis 2011: 702–703).

Hayek is quite clear that a tendency to general equilibrium in the real world is an assumption of his work:
“… it is my conviction that if we want to explain economic phenomena at all, we have no means available but to build on the foundations given by the concept of a tendency toward an equilibrium. For it is this concept alone which permits us to explain fundamental phenomena like the determination of prices or incomes, an understanding of which is essential to any explanation of fluctuation of production. If we are to proceed systematically, therefore, we must start with a situation which is already sufficiently explained by the general body of economic theory. And the only situation which satisfies this criterion is the situation in which all available resources are employed.” (Hayek 2008: 34–35).
Yet the existence of equilibrium is “not an empirically relevant state of affairs” (Repapis 2011: 703). The idea of equilibrium and a tendency to equilibrium requires that agents have expectations that are fulfilled and nothing is unforeseen (Repapis 2011: 703–704). That is to say, the agents in Hayek’s model of the cycle live in a world of “certain outcomes” (Repapis 2011: 713). But obviously this ignores the reality of fundamental uncertainty in the world and economic life.

Karl Gunnar Myrdal (1898–1987) had already criticised Hayek’s ABCT by drawing attention to the problematic role of expectations in a 1933 paper (Myrdal 1933: 385; Repapis 2011: 713). Importantly, the problematic role of expectations for Hayek’s trade cycle theory was also being raised within the Austrian school in the 1930s, as Ludwig Lachmann related in an Austrian Economics Newsletter (AEN) interview:
AEN: You have talked a number of times about the importance of expectations in business cycle theory. What first drew your interest to expectations as far as the business cycle question was concerned.

Lachmann: Talking to Paul Rosenstein-Rodan, who was then a lecturer at University College, London – not technically in the London School of Economics – but he gave a course on the history of economic thought to which all of us who were research students then went. It was Rosenstein-Rodan who in discussing Austrian trade cycle theory with me said, ‘Ah yes, but whatever happens in the business cycle is in the first place determined by expectations.’ And then he told me of the work that had been done in Sweden.”
Ludwig Lachmann, “An Interview with Ludwig Lachmann,” The Austrian Economics Newsletter, Volume 1, Number 3 (Fall 1978),
Therefore the expectations of entrepreneurs in disequilibrium (always the real state of the economy) do not in reality work in the way Hayek required in his theory.

Although Hayek came to realise that his use of general equilibrium theory and assumptions about expectations were unsatisfactory, nevertheless even in Profits, Interest and Investment (1939) he had still not properly addressed these criticisms or modified his trade cycle theory to deal with them, a charge later levelled against him by G. L. S. Shackle (Repapis 2011: 716; see Shackle in O’Brien and Presley 1981: 241).

To return to the general question of general equilibrium theory, if this theory is false and the assumption of a market tendency to equilibrium is not a description of the real world, as Post Keynesians and even some Austrians in the tradition of Ludwig Lachmann argue, then it follows that the ABCT cannot be considered an accurate theory of actual business cycles. It is a further shortcoming of Austrian economics that certain Austrians can deny the usefulness or truth of general equilibrium theory but continue to use the ABCT (I am thinking of Austrians influenced by Lachmann and even Rizzo and O’Driscoll).

The use of general equilibrium theory by Hayek as the foundation of his business cycle research already throws up insuperable problems for the ABCT.

Repapis (2011: 717) contends that by the time of Hayek’s book The Pure Theory of Capital (1941) his analysis and understanding of equilibrium was far from the way he had defined it in Prices and Production, so that his business cycle theory now needed to be thoroughly revised. But Hayek never did this.

I would conclude that this is why it is perfectly legitimate to say that Hayek’s business cycle theory was ultimately a failure, a view essentially also taken by Witt (1997: 46–48) and Caldwell (2004: 228).

II. Austrian Capital Theory
Another problem with the ABCT is certain aspects of its capital theory. As is well known, Austrian capital theory categorises capital goods into higher or lower orders, as removed from the final consumption goods. But the notion that every capital good can be classified into such a higher or lower class is open to question (Repapis 2011: 706, n. 15; Marshall 1961 [1890]: 64–65, n. 3).

Repapis points to another problem with Hayek’s capital theory as assumed in Prices and Production:
“… [sc. there is] a central simplification in Hayek’s capital structure in Prices and Production. This is that capital depreciates completely once it becomes productive, or to put it another way, all capital is ‘circulating’ capital and there is no discussion of ‘fixed’ capital. However, this is a highly unrealistic assumption and Hayek knew it. In a letter to Keynes on 7 January 1932 he writes that ‘I am conscious that I have treated the durability factor lightly too in Prices and Production, but I did so because I hoped to make it less difficult and because I assumed a greater familiarity with Böhm-Bawerk’s concept of the average length of production than I ought obviously have done’ (Keynes 1987, 262). In the same letter he distinguished between two different ways to represent the capital process: one is on the ‘duration of the process of production,’ which is what Prices and Production was based on; the other is on the ‘durability of many instruments of production,’ which his analysis so far ignored. He concludes that ‘the problem becomes, of course, a little more complicated if one combines, as one has to do to come nearer to reality, the two factors determining the existence of capital’.” (Repapis 2011: 720–721).
In the real world, capital is heterogeneous, but also has a degree of durability and substitutability.

While the neoclassical view of capital as homogenous (or as a sort of transformable putty) is wrong, just because capital is heterogeneous and sometimes non-substitutable, it does not mean all throughout the economy one will also find quite durable, adaptable and substitutable capital goods.

Repapis draws the following conclusions:
“Where does this leave Hayek’s business cycle theory? Whereas one of Hayek’s most insightful critiques of contemporary business cycle theories was that they regarded capital as an absolute ‘datum’ that is given for short-period analysis, and therefore capital theory was separated from business cycle theory,
he finds his theory occupying the opposite extreme, one in which capital is a flow, and has nothing to say about the changing uses of fixed capital. However, the changing use, and future use, of durable goods in the business cycle is a central preoccupation of the capitalist and cannot be ignored in any theory of the cycle. As G. L. S. Shackle (1981, 240) writes, ‘Hayek’s argument, viewing “capital goods” as materials which only retain their physical identity through a process of fabrication into consumable form, overlooks the grip that durability has in constraining the business man’s choice of productive methods.’” (Repapis 2011: 723).

“Hayek considers capital a fragile, perishable good, constructed for a particular purpose whose value substantially diminishes once its specific usefulness is surpassed. Minor changes in the interest rate or changes in demand may lay waste machines built over time and dearly paid for.” (Repapis 2011: 724).

“In the other extreme, Hayek underplays the malleability of existing capital and its ability to be transformed into something profitable if the economic conditions of the cycle so demand. His insistence that investment must start again almost from scratch is an extreme position, but again it underlines what other contemporary theories of the 1930s underplay, the real loss of capital in the cycle due to unforeseen demand changes.” (Repapis 2011: 725).
No doubt there is a real loss of capital when demand changes or the interest rate rises, but is it as catastrophic and sweeping as imagined in the ABCT? Hardly. The capital structure also has a degree of adaptability, versatility and durability that means that the Austrian policy prescription of liquidationism in a recession does not logically follow either.

These problems with the Austrian capital theory and its assumptions underlying Prices and Production are yet another severe blow to the ABCT.

The criticisms of Austrian capital theory as raised by Repapis also mirror the comments of David Ramsay Steele (as quoted on Robert Murphy’s blog). I end by reproducing his remarks:
“The distinctive thing about Austrian trade cycle theory is its view of ‘real’ factors in the onset of the slump. Of course, much of what Mises and Hayek say overlaps with the ‘purely monetary’ theories of people like Milton Friedman, and long before that, of people like Hawtrey. So there is no dispute that inflation of credit may create a phoney boom, followed by an uncomfortable period of adjustment. What is distinctive about the Austrian theory is that it says the specific physical form of the capital which is malinvested plays a crucial role in the onset of the slump. So, for example, if lengthening the production structure requires a particular type of big, expensive machine that has no use with a shorter production structure, then that machine will have to be written off as a loss, since it is not suitable to the ‘return to reality’ when the boom is over.

What struck me very early about this (I think it crossed my mind when I read Rothbard’s book on the 1930s depression, around 1971) was that it’s an empirical claim, and at a quick glance, such physical incongruities don’t seem to loom all that large. So, if the production structure lengthens, you change the shape of investment into something more appropriate to a lower time-preference. Fair enough. But what does this really mean? Let’s say you have a factory. You start to use different types of machine tools, let’s say. Still, most of your factors will be just the same, or almost the same, as before: electricity, computers (or in the old days, office stationery), unskilled workers, workers with various types of skill such as accountants, engineers, salespeople, and managers, your factory building itself, your use of trucks to get materials into the factory and products out, and so on. In other words, the overwhelming majority of the factors you employ are not specific to higher or lower orders of production. It’s true that their application to specific tasks will shift a bit, but this goes on all the time, and is an inexact science at best.

Since the claim that physical incongruities are crucial is an empirical claim, I was then struck by the experience of the US at the end of World War II. If ever there was a case of an abrupt, almost overnight, mismatch between prior allocations of capital and today’s applications, we could hardly imagine a more spectacular example. Millions of people left the army and found civilian work. Hundreds of thousands of factories which had been producing military goods had to transform their operations into civilian production. Why was the whole system not seized by a violent slump?

To the purely monetary approach, this is simple and obvious. There was no violent contraction of the money supply, so there was no slump. But to the Austrians, what explanation could there possibly be? Their claim is that once the boom has got going it cannot be ended without a slump, and that this is so because of the need to suddenly re-allocate physical assets to completely new uses. But that re-allocation was obviously thousands of times greater in 1945 than it could ever be as the result of a few years of bank credit expansion, and yet there was no slump! The whole system adapted to the utterly changed conditions with amazing ease and smoothness. ….


Caldwell, B. 2004. Hayek’s Challenge: An Intellectual Biography of F.A. Hayek. University of Chicago Press, Chicago and London.

Hayek, F. A. von, 1967 [1935]. Prices and Production (2nd edn.). Augustus M. Kelly, New York.

O’Brien, D. P. and John R. Presley (eds.). 1981. Pioneers of Modern Economics in Britain. Macmillan, London.

Marshall, Alfred. 1961 [1890]. Principles of Economics (9th edn.). Macmillan, London.

Myrdal, G. 1933. “Der Gleichgewichtsbegriff als Instrument der geld-theoretischen Analyse,” in F. A. Hayek (ed.), Beitrage zur Geldtheorie, Vienna. 361–487.

Repapis, Constatinos. 2011. “Hayek’s Business Cycle Theory during the 1930s: A Critical Account of its Development,” History of Political Economy 43: 699–742.

Witt, U. 1997. “The Hayekian Puzzle: Spontaneous Order and the Business Cycle,” Scottish Journal of Political Economy 44: 44–58.

No comments:

Post a Comment