Chapter 3 of Colin Rogers’ Money, Interest and Capital (1989) deals with neo-Walrasian general equilibrium theory and the role of money in that theory.
Such general equilibrium models do not take the value of capital as a given quantity, but as an array of quantities (Rogers 1989: 45).
Neo-Walrasian models reduce to ones of perfect barter in which perfect knowledge, tâtonnement and recontracting are assumed to coordinate all economic activity before trade commences (Rogers 1989: 46).
Money as a unit of account is added without disrupting or changing any of the perfect barter conditions (Rogers 1989: 46). That is, money is an inessential addition to a real model in which perfect barter is assumed (Rogers 1989: 46).
An economic model that treats money as an unnecessary addition does not adequately describe real world monetary capitalist economies, especially the problems of a modern economy with a credit and banking system (Rogers 1989: 47).
The most sophisticated neo-Walrasian model is that of Arrow-Debreu, which abolishes any problems that arise in the real world from uncertainty or shifting expectations; the Arrow-Debreu model also collapses the future into the present (Rogers 1989: 48).
Alternatively, neo-Walrasian temporary equilibrium models relax the Arrow-Debreu assumption of a complete set of futures markets for all time-dated commodities, but still face the problem of expectations (Rogers 1989: 48–49).
The solution that was adopted by many neoclassical models was the Rational Expectation hypothesis, which assumes that agents correctly predict on average the equilibrium prices of the future (Rogers 1989: 49).
Thus neo-Walrasian temporary equilibrium models are functionally equivalent to Arrow-Debreu models, and once again reduce to models where money is an inessential addition (Rogers 1989: 49).
Rogers considers the neo-Walrasian equilibrium model of Hahn (1982), in order to compare it with Wicksellian monetary theory. In Hahn’s model, equilibrium does not generate an equality of interest rates between the two commodities produced in the model (Rogers 1989: 53). The peculiar feature of the neo-Walrasian interest rate theory is that it does distinguish profit from interest (Rogers 1989: 58).
In short, all neo-Walrasian models must be rejected as irrelevant to real world economies because they ultimately have no real role for money (Rogers 1989: 67).
BIBLIOGRAPHY
Hahn, Frank. 1982. “The Neo-Ricardians,” Cambridge Journal of Economics 6.4: 353–374.
Rogers, C. 1989. Money, Interest and Capital: A Study in the Foundations of Monetary Theory. Cambridge University Press, Cambridge.
"The solution that was adopted by many neoclassical models was the Rational Expectation hypothesis, which assumes that agents correctly predict on average the equilibrium prices of the future (Rogers 1989: 49)."
ReplyDeleteAnd that is the assumption of a strong-form EMH which has been empirically discredited since the late-1980s. Robert Shiller won the Nobel Prize for discrediting this recently.
Lesson? Mainstream macro is an incoherent mess and few mainstream economists have any idea what they're talking about any more. One moment they will make policy proclamations that rely on an implicit strong-form EMH, the next they will deny the existence of strong-form EMH. It's a farce. Most of them don't understand their own theories. Rather they have some cobbled together notions of how economies work and then they make proclamations on policy that are basically just a reflection of whatever fad is in vogue at the time (monetarism in the 1980s; Taylor Rule inflation targeting in the 1990s and early 2000s; market monetarism or NGDP targeting post-2008).
They're jellyfish swaying this way and that with the tides of fashion and fad. And that is why it is less important what is being said in economic debate than who is saying it. Science, this is not.
It's true that money plays no role in the standard general equilibrium model. Which is exactly why economists usually add other elements to those models (usually money in the utility, or cash in advance constraints, plus sticky prices) when they want to study isssues where monetary policy, and money in general, matters. One could argue about whether these approaches are succesful, but it looks like Rogers doesn't do so (perhaps understandable, given the age of the book).
ReplyDeleteAlso the rational expectations bit is somewhat confused. Intertemporal general equilibrium models can be defined either purely in terms of state-time dated goods that are all traded at the beginning, or more realisticaly as sequence of markets where people trade present goods and future financial claims (I guess this is what Rogers calls temporary equilibrium). The latter framework requires rational expectations in the conditional sense - agents know future prices, conditional on future state. But what weight do they give to the likelihood of any future state actually happening can be arbitrary.
In other words, agents knowing that crop prices will be high if the future weather is bad means they have "conditional" rational expectations. Agents knowing actual probability whether the weather will be bad is rational expectations in usual sense. These are two different things (and it's the latter assumption that's usually invoked and criticized in macroeconomics).
Finally, @Philip Pilkington: "strong-form EMH which has been empirically discredited since the late-1980s." No, it hasn't. What has been discredited is EMH plus the assumption of constant expected returns. EMH is not really testable on its own (something that Fama has been always explicit about), so in my opinion usefulness of the term itself is a bit dubious. Anyway, whether observed asset prices are better explained through behavioral biases or rational responses to changing macroeconomic conditions is still an open question. A balanced overview is e.g. here: http://www.institutionalinvestor.com/Article/3315202/Asset-Management-Equities/The-Great-Divide-over-Market-Efficiency.html
"Anyway, whether observed asset prices are better explained through bevavioural biases or rational responses to changing macroeconomic conditions is still an open question."
DeleteCould we just start with a clear definition of "rational"? According to your earlier statements "but what weight they give to the likelihood of any future state can be arbitrary". This appears to categorise all possible future price guesses as fully rational. The strong form EMH seems awful weak at explaining any future price states then. It appears to say absolutely nothing at all about them in fact.
Yes, I've seen Fama's (and Malkiel's) absurd responses. These days they're claiming that all the EMH says is that "you" (who this "you" is is completely left in the dark; is it me or Lord Keynes? or is it Soros or Buffett) cannot beat the market in the "long-run" (again, undefined). The EMH has been in tautology territory since the early 1990s. But if you examine the original statement of the theory closely you will quickly find it to have been a tautology in the first place.
DeleteBut whatever way you phrase it the implication is the same: most mainstream economists reject the idea of perfectly rational, stable financial markets but then make policy advice and provide commentary assuming perfectly rational, stable financial markets. They don't understand their own theories properly. The profession is in a right state. And its reflected at just about every level.