Monday, June 21, 2010

Keynes, Capital, and some History of Economic Thought

Brad DeLong has a couple history of thought posts up that are worth looking at. In the first he suggests that there are at least seven theories of depression out there: monetarism, Wicksellianism, Minskyism, Austrianism, vulgar Keynesianism, Hicksianism, and Post-Keynesianism. DeLong concludes "If you aren't an economist--hell, if you are an economist--the relationship between all seven of these theories is obscure and confusing". I have a few thoughts on these. First, his description of Austrianism is fair enough as far as it goes, but it obviously needs to include the credit market. Malinvestments aren't simply accidents, of course - they come from credit expansion (of course, with different explanation of what sort of credit expansions produce malinvestments). Second, the absence of real business cycle theories from the list is telling, and probably appropriate. Third, I find the "vulgar Keynesianism" definition a little odd. I guess I'd call that "very oversimplified Keynesianism", but there's nothing particularly vulgar about it. When people talk about "vulgar Keynesianism" I think of "if you increase G then Y has to increase". Fourth, with the exception of Minskyism this completely excludes anything that can be thought of as a "cycle" theory. So we can add in inventory cycles, accelerator-oscillator models, credit cycles, etc. - because all of these can contribute to depressions as well. I personally think all of these are pretty decent explanations of certain economic processes - some more relevant than others depending on the specific downturn that we're interested in. Obviously I think what DeLong calls "Hicksianism" (which I often pass off as Keynesianism) is the most significant in our current downturn.

Which brings me to DeLong's next post, in which he argues that Hicks isn't really explaining Keynes - he's channeling Kahn, Wicksell and Fisher. To a certain extent, he's probably right. Hicks was working on Value and Capital while Keynes was working on the General Theory, so you certainly can't argue that he got his economics from Keynes (and to be clear, no one does argue that). But Hicks himself suggested the IS-LM model was an attempt to use his (Hicks's) framework to elucidate Keynes. I had always taken him at his word on that. I first read the General Theory at a time when my recollection of IS-LM type analysis was very rusty, so while reading it I wasn't really thinking in terms of "does this really align with Hicks". Since then, Lawrence Klein has helped me understand how Keynes works together with IS-LM much better and he made a quite convincing case that Hicks (and Modigliani and Hansen) all really were formalizing Keynes. Anyway, DeLong argues a somewhat contrary position and now I feel like I have to revisit the General Theory in more detail and as a slightly older, slightly wiser reader.

Bill C, of Twenty Cent Paradigms, has a colorful anecdote from Keynes.

Finally, I was rereading Keynes's chapter titled Sundry Observations on the Nature of Capital, where he talks about the roundaboutness of production and the capital structure. He provides the standard Garrison-Hayek explanation that as the interest rate lowers, the capital structure becomes lengthened or more "roundabout". I noticed something interesting in Keynes's discussion, though - he not only refered to the structure of capital, but the structure of demand. For example, he wrote:

"Given the optimum amount of roundaboutness, we shall, of course, select the
most efficient roundabout process which we can find up to the required
aggregate. But the optimum amount itself should be such as to provide at
the appropriate dates for that part of consumers' demand which it is to be
Apparently, the time-structure of demand was an important part of Bohm-Bawerk's capital theory too. It seems to me that you don't see that as much in modern Austrianism. Is that true? It seems essential to me, because it's the time structure of demand that determines the return to capital.


  1. "It seems to me that you don't see that as much in modern Austrianism."

    In every lecture on Austrianism I have ever heard that dealt with capital theory this issue was discussed.

  2. You really like these superlatives, don't you? What about lectures on Austrian epistemology? Have you heard it discussed there? I have a hard time believing that.

    I should clarify - it's obviously discussed in terms of saving and the intertemporal coordination of consumption. Is it discussed with respect to the capital structure, though? That's more what I'm curious about. Keynes is saying that certain capital structures are going to be preferable for providing consumption goods at the time that they are demanded.

    To think of it in a simple way - Garrison's macro-model has a time structure for capital, right? It's the basic Hayekian triangle model. But his modeling of consumption is static - the basic PPF. Now - does Garrison go into more detail elsewhere? Perhaps. Does Garrison faithfully represent Hayek? Maybe not and maybe Hayek goes over it.

    I suppose what I'm saying, Xenophon, is that you don't see it as AS MUCH as the time structure of capital in Austrian models. I don't know if you got that distinction - not that you don't see it at all, but you don't see it as much.

    Could you point me to some discussion of it - not just intertemporal consumption stuff - but with respect to how the time structure of demand informs the capital structure? I appreciate it.

  3. "What about lectures on Austrian epistemology? Have you heard it discussed there?"

    But then again, in an earlier post you called a basic Austrian epistemological point "bombastic", so perhaps you just haven't heard any lectures on Austrian epistemology.

  4. "You really like these superlatives, don't you?"

    I am just relating to you my anecdotal, personal experience.

    No, I called your comment bombastic.

  5. The time structure of production is related to the time structure of consumer-good demand. A lengthening of the structure of production requires a change in time preference. Consumer-goods may be a bit of a generalization. A change in time-preference at the end of their production schedule (I am convinced that distinguishing between consumer and capital goods is impossible, given that what pass off as consumer goods are often capital goods [we do eat food to work; distinguishing between capital and luxury is not worthwhile, really].

    So, this idea of a time structure of demand is implicit in Austrian capital theory.

  6. By the way, DeLong's depiction of Austrian theory is not fair at all. I am writing an article, "Krugman Contra Hayek", which I hope will address all of this. I'm trying to make it as fair as possible (properly depict Keynesian theory), so it's taking me a while.

  7. And time preference I do get - that's obviously in there.

    My concern is this - it seems to me (and I may be wrong) that consumer time preference enters discussion of the capital structure in two ways:

    1. In determining the interest rate that capitalists respond to, and
    2. In coordinating consumption over time for consumers.

    Keynes seems to be saying that different production structures are going to be appropriate for different time structures of demand (for example, you don't want an extremely long production structure for fast food, or for specialized parts that will be needed on short notice). This is not #1 - the role that time preferences play in coordinating savings and investments. It is #2 - the way that consumers order consumption over time. But I don't know if there's anything in Austrian theory that specifically ties that to the productive technology.

    It seems to be in Bohm-Bawerk, and briefly in Keynes.

  8. And to be honest, I'm probably not thinking about this or writing about it clearly.

    But I'm definitely aware about time preference and intertemporal consumption and time preference and the coordination of savings and investment. Those things are well established across the board in economics.

  9. RE: "By the way, DeLong's depiction of Austrian theory is not fair at all."

    Even if you add in credit cycles (which as I noted is a huge omission)?

    How would you summarize it in two or three sentences? At some point a summary is never going to satisfy people (do you think ANY of the others were complete renditions?). What would you have said? What's wrong with it?

  10. While much of Bohm-Bawerk's analysis is useful and correct, the height of Austrian economics did not reach the third-generation Austrians (Hayek and Mises) for a reason. Bawerk is a little outdated, and much of the newer Misesian models on interest and monetary theory were fought by the die-hard Bawerkians.

    The time structure of demand, as you called it, is already implied in Austrian/Misesian capital theory. Xenophon and Jonathan make that point. This was one of Bawerk's contributions, but I would refrain from associating all of Bawerk's theories with Austrian economics as we understand it from Mises.

  11. All I've heard about the time structure of demand is time preference as it (presumably) relates to the interest rate. This does not seem to be what Keynes is talking about. No matter.

  12. Note that Keynes self-evidently never comprehended the trade off between time and output at the heart of Bohm-Bawerk's marginalist understanding of choice in the domain of production goods -- an epic fail on Keynes part, a failure at the very heart of Keynes inability to understand Hayek's work (if he ever tried, I don't think he did.)

  13. Anonymous - could you explain that in more detail? I'm not personally too familiar with Bohm-Bawerk.

    People often claim that Keynes didn't understand that there interest rate coordinates consumption across time. Is this what you're getting at? Because that's ridiculous and I don't know how that meme got started. It might be because of the prominence of liquidity preference that people think he's ignoring intertemporal coordination... perhaps that's where they get it. I think it's pretty funny that people consistently resort to "X didn't understand Y" because they couldn't possibly countenance the idea that "X understood Y perfectly and wasn't convinced by it".

    Anyway - what do you mean by this exactly?


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