Saturday, March 23, 2013

Why Keynes and not Hicks?

Jonathan considers the question.

The Keynesianism we have is really Hicksianism. I do think economists appreciate this. I think most economists know IS-LM as we use it is not in the General Theory (of course you can yank it out of there if you wish, but be prepared to get nagged by some post-Keynesians). So of course the Keynesianism that grew out of IS-LM great out of Hicks (for the rest of this post though, when I talk about Keynesianism I'll be referring to old Keynesianism/IS-LM not the newer stuff).

So if we know this why do we still praise Keynes?

This is going to sound a little fuzzy, but I think it's because Keynes taught us how to think about things and Hicks taught us how to model things. Those are two very different tasks, and it's the former task that tends to loom large in our psyches. When we want to walk someone through the logic of a Keynesian answer to a question, we always go back to the props that the General Theory furnishes us with. The catchy passages in the book don't hurt - many people have pointed this out. But I don't think it's just that. It's how compelling the argument is. Hicks doesn't give us anything like this sort of intuitive argument that we can whip out when we need it. He gives us the model. The model is important, but it doesn't frame the theory the way Keynes does.

As an example of what I mean, think about the theory of the Phillips Curve or the NAIRU. These are big ideas associated with Phillips and Friedman, respectively. That's because they presented us with the idea and when we think about the idea we're drawn back to their exposition of the idea. But lots and lots of people have modeled the Phillips Curve and the NAIRU - some models explaining the same phenomenon but with strikingly different explanations. Some models are constructed differently but give us basically the same story. A model is just a bit of math to illustrate a concept - it's smaller than a "theory". You can have lots of different models of the same theory.

Now Hicks's model was a big deal. But I think Keynes still looms large because he ultimately gave us the concept of Keynesianism.

I would be careful not to understate Keynes's contribution, though. Liquidity preference theory of the interest rate is essentially original to him (you can find wisps of antecedents in Franklin and Fisher), and it's actually one that Hicks was skeptical of early on (Hicks referred to it as seeming like interest rates were being pulled up by their own bootstraps). I think the fact that we have preferred the Hicksian version of liquidity preference is that his exposition is more Marshallian than Keynes's (which gives us this nebulous idea of lots of potential states of the economy). Keynes wanted to present a picture of multiple potential states - an unclosed model - because he was presenting an alternative to the idea (we can argue about how prevalent it was) that the economy is always going to be operating at full employment or at least tending towards it. That view of things - that model of the bigger idea - didn't last as the model that we use to represent the bigger idea, because most of us are more comfortable with closed models and unique equilibria.

Keynes also has a lot of good substantive contributions that are simply left unexploited by mainstream economists and have been taken up by heterodox economists.

1 comment:

  1. Among the overlooked aspects of Keynes's General Theory are the two references to A Treatise on Probability in Chapter 12 and Chapter 17 of the GT. (Specifically, Page 148, and Page 240...the footnote on Page 240 directs one back to the footnote in Chapter 12.)

    The reference is to Chapter 6 of the TP. And if you read that carefully, on Page 76, you'll be referred to Section 7 of Chapter 26. And a few pages after Section 7, you are introduced to what Keynes calls a "conventional coefficient c of weight and risk". This is essentially a decision rule with properties that reflect non-linearity and non-additivity in probability theory and decision-making.

    It would be forty years after the publication of the TP that Daniel Ellsberg would publish his seminal 1961 article in the Quarterly Journal of Economics. Ellsberg's article, which provides a powerful critique of the axioms of S.E.U. decision theory, and the subsequent doctoral dissertation he completed in 1962 expanding upon his article, can also be seen as experimental and empirical proof supporting Keynes's theory of liquidity preference.

    As for the "picture of multiple potential states" as you so term it, Daniel Kuehn...aren't you better off saying that Keynes was talking about an economy with *multiple* equilibria?

    As I have mentioned a number of times before, Book V (Chapter 19, appendix to Chapter 19, Chapter 20, and Chapter 21) of the GT contain the mathematical expositions that Keynes used for his model. In fact, you could easily say that contrary to a widespread misperception, Keynes's magnum opus *DID* have what would be termed "microeconomic foundations".

    Chapter 19 and its appendix compare and contrast Keynes's theory with A.C. Pigou's mathematical formulation of the classical theory of unemployment (see The Theory of Unemployment [1933]). If you read that part of the GT and Pigou's book, it will be easy to tell who has the more general case and who has the more restrictive case, Daniel.

    Chapters 20 and 21 deal more with liquidity preference and the monetary aspects of the GT in good detail through elasticities.

    For more references regarding the mathematical expositions in the GT (and that mention page in Book V of the GT), see the following articles please...

    And see if you can get this chapter of this book...

    Ferreira, R. S., and Philippe Michel. "Reflections on the Microeconomic Foundations of the Keynesian Aggregate Supply Function." The Foundations of Keynesian Analysis. London: Macmillan (1988): 251-262.

    As for an assessment of Sir John R. Hicks and IS/LM, perhaps you might wish to look into this anthology, Daniel.

    Finally, you might find this review of a book by Sir John R. Hicks of interest.


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