Wednesday, October 20, 2010

Robert Murphy on the Austrian School and Keynesianism

So I've been meaning to write about this new post by Robert Murphy that "puts the Austrian business cycle theory to the test". I wanted to share a few thoughts and get people's reactions.

The post is a decent catalog of some of the problems with Krugman's take on things, a good start for explaining the congruence of the crisis with ABCT, but a very bad attempt to draw conclusions from that.

1. First, Murphy is right to point out indices of decline rather than absolute declines (which Krugman does). Construction has seen a higher rate of job loss than manufacturing, which is of courses consistent with ABCT.

2. Murphy notes that Austrians opposed a suite of government actions. He writes: "Now say what he will about the Austrian economists, surely Dr. Krugman will concede that they opposed the above actions, and that they would have predicted at any time that these policies would lead to economic stagnation". It's true that they opposed these actions, but then again Krugman said many were inadequate or just plain wrong too. It's worth probing what each side predicted the effect would be. Krugman (and I for that matter) suggested we would have continued deflation and low interest rates. Not all, but a sizable portion of Austrians predicted inflation and high interest rates. This makes sense - both of those results are easily derived from Keynesian and Austrian approaches, respectively. So Krugman and the Austrians were both right here - but who was right for the right reasons? Generally speaking, I'd say Krugman.

3. Robert Murphy is being entirely disingenuous when he writes this: "In contrast to the Austrian story, what does the Keynesian view predict? Well, if a recession is really just about a general drop in aggregate demand, then we shouldn't see any particular relationship among individual sectors, and how they respond both in magnitude and across time. If you reread Krugman's commentary on his chart, that's exactly what he himself says the Keynesian story means." What can spark a demand shock? Lots of things, of course - but one major cause is a popped bubble that leaves people with a lot less income and wealth than they previously had. Murphy abstracts away from the bubble completely and acts like this is all a question of the capital structure and roundaboutness. You can't remove the housing crash from this narrative - that was the aggregate demand shock which brought the rest of the economy down. The question shouldn't be "how do Keynesians account for the construction employment" - that's obvious - there was a housing crash after some standard bubble psychology and bad policy. The question should be "how do Austrians account for the breadth of unemployment". Clearly the basic rebalancing of the capital structure can't explain this. You need some story about malinvestments as overinvestments or overproduction in addition to the standard malinvestment story. The case can be made, but the onus doesn't seem to be on the Keynesians.

4. So where do we end up with Murphy? Well he's using the same fallacious logic that I had to correct from him with my 1920-21 depression paper. Is this downturn consistent in some important respects with ABCT? Sure. Of course it is. My feeling is that that is because ABCT is a logical and accurate theory of a real macroeconomic process. So of course you're going to see it in the real world. Does that mean the Keynesian explanation is wrong or not also consistent? Of course it doesn't. These are not mutually exclusive processes that we're describing here, and the theories that you're promoting are only mutually exclusive because you have defined them as such. You don't just assume that oil embargoes can't cause recessions because you've attached the "Keynesian" or "Austrian" label to your lapel, do you? No - because the supply shock of an oil embargo isn't inconsistent with additional demand shocks or capital structure adjustments. Robert Murphy lives in a Manichean world, not a scientific world. He wants economic theories to engage in a battle of the titans. This was how it was with his explanation of the 1920-21 downturn too. For Murphy, if ABCT fit the facts (he didn't make that much of a case that it did - he barely talked at all about the capital structure), then Keynesianism couldn't fit the facts. Part of this was simply because he grossly misunderstood Keynesianism, but a lot was also because he couldn't accept the fact that under some circumstances Keynesianism and ABCT predict much the same thing.

So long story short, I think Murphy makes some decent points and a good beginning of a case for ABCT, but draws some very bad conclusions from it.

20 comments:

  1. Cool, thanks for taking a look at this article.

    I'm a little confused as to how you reconcile the two theories though -- ABCT paints the central bank as a chief cause of the business cycle, but the central bank is essential for Keynesian theory.

    ReplyDelete
  2. Well, I don't reconcile the two theories as the Austrians usually think of them.

    What I think is reconcilable is the impact of central bank activity on the capital structure. The fact that that process is real does not mean it is the dominant driver of the business cycle - I don't think it is.

    My position is that, yes, we may have these adjustments of the capital structure in response to central bank actions and that's interesting and that's real, but the problems associated with that discoordination pale in comparison with the problems of a central bank not providing an adequate money supply.

    So I accept ABCT as a theory of the business cycle, not as "the" theory of the business cycle.

    ReplyDelete
  3. To what extent do you think malinvestments [as a result of credit expansion and interest rate manipulation] contribute to the bubble, if any? Could they at least provide the initial catalyst for a bubble, with bubble psychology taking over afterwords?

    Thanks for your input.

    ReplyDelete
  4. I don't really know and I have yet to find a really solid treatment of this.

    It's plausible, but if it were a major factor you'd think that bubbles would consistently be in the early stages of the production process, or in especially round-about production processes. Do we see this? I don't think we do. Usually the real destructive bubbles (from an AD and from an output perspective) are in assets.

    Why should housing be especially prone to malinvestment? If anything the production process for housing is SHORTER now than it was before. Pre-fab/McMansions/big developments, etc. This is not an obvious candidate for ABCT malinvestments, is it?

    I would love to do a detailed empirical study on the actual distortions of the capital structure - not just plotting employment changes for three different industries - and see how big these processes really are. Because really I feel like we don't know. I suspect they are there, but they are negligible. It's worth looking into.

    ReplyDelete
  5. "It's plausible, but if it were a major factor you'd think that bubbles would consistently be in the early stages of the production process..."

    I've never found this to be a very convincing attack on the malinvestment theory. The question has more than adequately addressed by a number of Austrian economists.

    "Why should housing be especially prone to malinvestment?"

    Because the government is so heavily involved in its promotion; and that has been the case for all of the 20th century and much of the 19th century.

    ReplyDelete
  6. It's not an attack, Xenophon.

    "Because the government is so heavily involved in its promotion; and that has been the case for all of the 20th century and much of the 19th century."

    OK, but this isn't ABCT anymore. And btw, citing two centuries of intervention doesn't do much to isolate the causes of a bubble that occured in a matter of years in the early twentieth century.

    ReplyDelete
  7. Attack, criticism, whatever.

    "And btw, citing two centuries of intervention doesn't do much to isolate the causes of a bubble that occured in a matter of years in the early twentieth century."

    The bubble began in the mid to late 1990s actually. You can argue with Vernon Smith on the subject: http://online.wsj.com/article/SB119794091743935595.html

    Anyway, it does a lot to isolate the causes of such; the causes always lay with the government trying to promote one or another interest in the name of one cause or another (helping the poor, promoting transportation, paying for the war effort), which naturally attracts a flood of speculative money, which in turn leads to lots of malinvestment.

    ReplyDelete
  8. Lawrence Reed has a great podcast on stimuli, deflation and the Great Recession on a FEE podcast: http://fee.org/media/stimuli-deflation-and-the-great-recession/

    ReplyDelete
  9. At the risk of moving the goalposts, couldn't you consider the increase in people "flipping" houses as a lengthening of the structure of production?

    ReplyDelete
  10. Maybe - explain that more. Houses are an odd-ball thing to talk about to begin with becuase it's sort of mixed consumption and investment.

    The idea being the production process of a house for final sale is lengthened because the flipping process is part of the production process?

    I could see that. But I'm guessing this has more to do with trends in house prices than changes in the interest rate, wouldn't you think?

    True - there would be more flipping under low interest rates than under high interest rates, holding prices constant. But was that really the source of all the flipping? I doubt it. I'm guessing it had more to do with speculation on the value of the house itself.

    ReplyDelete
  11. nickn,

    Well, not just flipping, but government encouragement of home equity loans was also a big part of it. The government did its level best to encourage lots of private debt; it continues to do so, but lots of Americans have opted out of that con.

    ReplyDelete
  12. Xenophon - Nick is talking about changes to the capital structure... I'm not quite sure what the interventions you're talking about has to do with that. Could you explain?

    ReplyDelete
  13. You can't talk about capital structure without discussing the government's efforts to manipulate it.

    ReplyDelete
  14. Why haven't you discussed Wilkinson's take on DeLong's remarks?

    http://www.willwilkinson.net/flybottle/2010/10/09/the-elite-economist-revolving-door-problem-delong-edition/

    ReplyDelete
  15. I still don't understand why you think that we need an overinvestment theory to supplement that of malinvestment in order to explain the "breadth" of unemployment. I have already explained it within the parameters of malinvestment. If you think it's wrong or insufficient then fine, but at least explain why this is so, instead of ignoring it and claiming that we need more.

    Also, you officially win: I am back to responding here. :) The dark side is too appealing!

    ReplyDelete
  16. Didn't we come around to the idea that we agreed on this point when we talked about it before? The point is, the capital structure can't just pivot - the slope of the capital structure can't just be reduced - the level itself has to actually increase. That's all I'm saying which I think is what we eventually determined you were saying too. That's fine, but that's not uniquely Austrian Business Cycle Theory. If ABCT is driven by "low interest rates increase investment" what is it really adding?

    ReplyDelete
  17. Name me an economist that won't tell you that.

    ReplyDelete
  18. "The point is, the capital structure can't just pivot - the slope of the capital structure can't just be reduced - the level itself has to actually increase."

    I'm not sure what you are trying to say here.

    "If ABCT is driven by "low interest rates increase investment" what is it really adding? "

    In what regard? It illustrates the effect of an increase in investment without a previous increase in the volume of real accumulated capital.

    ReplyDelete
  19. Meyer on macro, etc.: http://www.clevelandfed.org/forefront/2010/09/ff_2010_fall_02.cfm

    ReplyDelete
  20. Yep - it was an interesting interview. I mentioned it here:
    http://factsandotherstubbornthings.blogspot.com/2010/10/of-streetlamps-and-macroeconomics.html

    When I discussed the Caballero WP - you might be interested in that one.

    ReplyDelete

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