Bruce Bartlett considers the title of John M. Keynes’ magnum opus, The General Theory of Employment, Interest, and Money, an “unfortunate error.” According to Bartlett, Keynes’ core insight is the liquidity trap, which he defines as a situation where both inflation and interest rates are low, making bonds and money close substitutes. Thus, Keynes’ economics are mostly applicable only when an economy is in a liquidity trap. I think Bartlett has it mostly wrong. The liquidity trap only plays a small role in The General Theory, and the book’s major contribution — at least, as intended by Keynes — is its business cycle theory.

If you’re wondering what the liquidity trap is, I give an overview of the development of the theory in a June 2010 Mises Daily, “Krugman contra Hayek.” Most of my overview is based on a paper by Mauro Boianovsky, “The IS-LM Model and the Liquidity Trap Concept: from Hicks to Krugman.” A modern interpretation of the liquidity trap theory is provided by Paul Krugman, in his 1998 paper on Japan. Krugman’s definition is a bit more general: when conventional monetary policy no longer stimulates, otherwise known as the zero lower bound (ZLB).

How big of a role does the liquidity trap play in The General Theory? In Keynes’ 1936 book, the liquidity trap is mentioned, in passing, in chapter 15,

There is the possibility, for the reasons discussed above, that, after the rate of interest has fallen to a certain level, liquidity-preference may become virtually absolute in the sense that almost everyone prefers cash to holding a debt which yields so low a rate of interest.

— p. 207.

Keynes suggested, with healthy skepticism, that the early 1930s may be an example of a liquidity trap, but that these cases are indeed very rare. It certainly is not the centerpiece of Keynes’ theoretical exposition. In fact, the liquidity trap is probably better identified with John R. Hicks (who later repudiated much of his work from “Mr. Keynes and the Classics“), who also introduced the IS/LM diagram taught in intermediate macroeconomics. In any case, none of the theories that Keynes developed in his book were specific to the liquidity trap, nor require the liquidity trap as a precondition for their validity.

Keynes’ main argument, in my opinion, consists of an integration of R.F. Kahn’s multiplier with the macroeconomic framework Keynes began to develop in his A Treatise on Money (and volume II). The main purpose of The General Theory is to explain how an underemployment equilibrium may arise, and Keynes’ theory is that cyclical fluctuations are caused by increases in the stock of savings which cannot be met with greater investment. The theory is similar to monetary disequilibrium, but rather than an increase in the demand for money and sticky prices, the shock in aggregate demand is caused by a sudden reversal of entrepreneurs’ expectations (more on the differences between the two theories here).

The most well known term associated with Keynes is “animal spirits,” but at first he actually frames his theory as a secular outcome of economic growth. As an economy becomes more productive and incomes grow, the propensity to save tends to grow at a faster rate than the marginal propensity to consume. In other words, the proportion of saving to consumption increases over time. A tenet of the Keynes–Kahn multiplier is that present investment is directly derived, by and large, from present consumption. As consumption falls, the scope of investment falls, and vice versa. It follows that at some point savings is bound to increase beyond the point it can be profitably invested, causing an aggregate demand shock. Drawing on “animal spirits” helps with the application of this theory, since recurring waves of optimism and pessimism can cause the sudden changes in expectations that leads to a shortage of investment. But, “animal spirits” is not a central component of the “general theory.”

In the course of explaining his theory, and its many components, Keynes offered two main challenges to what he termed Classical economics. The first, early on, is that wages aren’t sticky, rather they may not be able to fall in real terms, at all. He posited that since labor makes up a significant portion of the costs of production, a nominal reduction in wages will lead to a proprotional nominal reduction in the price of output, leaving real wages the same. By doing this, he circumvented the typical explanation for mass unemployment: the artificial rigidities created by interventionism. Second, he engaged the believe that savings and investment is well equilibrated by the rate of interest. He argued that the interest rate is not only determined by time preference, but also by liquidity preference — interest on non-cash assets have to include a premium to make up for their relative illiquidity. If a high liquidity preference increases the rate of interest well beyond its equilibrium, or natural, level, there will be some discoordination between savings and investment, leading to or aggravating a demand shock.

What’s the general theory, then? All of this is explained within the context of a novel macroeconomic framework. Keynes was advancing a theory of the coordination of several macroeconomic aggregates: investment, savings, consumption, interest, et cetera. He argued that capitalist economies are prone to demand shocks — not under special circumstances, but generally. And, actually, referring to Keynes’ theory as a business cycle theory may be somewhat misleading, because the business cycle theory is really only secondary to the macro framework Keynes was attempting to construct. That is, demand shortages are only a part of the broader theory of Keynes’ vision of how economies work on aggregate: investment, and therefore employment, is determined by the expected demand for final output. Keynes relegated supply-side considerations to the back burner.

Maybe by “core insight” Bartlett means the key concept that economists took from Keynes. But, even then, I think he’s wrong. The key, and in my opinion erroneous, belief is that present demand for consumers’ goods output determines the scope of present investment, especially without considering supply side qualifiers that would radically change the implications of Keynes’ general theory. It’s this idea that informs the opinion that consumption drives the economy (and that to restore aggregate demand we must stimulate consumption). It’s this relationship which is one of the most important facets of Keynes’ general theory of macroeconomic coordination.

Also, briefly, I’m not sure just how skeptical of monetary policy Keynes was. I recommend two papers on the topic: D. Moggridge and S. Howson, “Keynes on Monetary Policy;” and E. Dickens, “Keynes’s Theory of Monetary Policy.” Also, Hicks discussed some differences on monetary policy between Keynes and Ralph Hawtrey, in his book Economic Perspectives.

Pete Boettke writes on Keynesianism and how it changed the “language” of economics, arguing that this was a turn for the worst. Ryan Murphy argues that this can’t be true, since the evidence suggests that differences in language plays only a minor role in changing the way we think. In defense of Boettke, I think Murphy is is taking his criticism too far and Boettke’s use of the word “language” too literally.

Much of post-war economics was very different to the economics of the interwar years. This is one of the themes in Kim Kyun’s Equilibrium Business Cycle Theory in Historical Perspective. According to Kyun, interwar econometricians sacrificed much of the richness of interwar (business cycle) theory in exchange for more formal, but much more simple models which they could use to test against data. These models suffered from very similar problems that modern econometric models do, and so it can be argued that the decades following the Second World War represent, to one extent or another, theoretical stagnation. Some of this was reversed thanks to Robert Lucas, but not completely.

I think Murphy is right in that it’s always possible to use any language to describe your ideas. For example, RBC and DSGE models can be re-written to include things that Austrians are concerned about (this is actually a research program I’d like to explore, once I know the tools well enough). In fact, in ways much of this work has already been accomplished. In other words, models or a “language” don’t necessarily constrain the economist to a relatively limited number of research paths.

Nevertheless, it’s also true that existing models emphasize some concepts more than others. When learning these models, students will be exposed to these areas. They won’t be as exposed to the areas left unstudied. Someone looking to modify the existing models with different emphases, therefore, has to be exposed to these different emphases in other ways. This oftentimes happens, but I’m not so sure that it happens to everyone (and often, when it does happen, it goes either ignored or it becomes something of a paradigm shift — e.g. the microfoundations revolution). Those that don’t receive exposure to, or develop on their own, ideas not focused on in existing models will have a stronger impetus to continue research in the areas she knows best, which is why economists like Boettke argue that certain programs are left understudied.

So, while “language” might not make it any harder to think about one thing than another language, the concepts that a set of models looks at are the concepts that students of these models will get to know. This certainly does affect the beliefs and ideas that economists will hold. This isn’t true only of the mainstream. It’s something I’ve noticed within Austrian circles, as well, which is why I spend so much time arguing that we should opt for open-mind approaches as much as possible. There are ideas that aren’t emphasized in much of the (older) Austrian literature, so you have to go elsewhere to find them.

With regard to Boettke, my only complaint is that I think he has the culprit wrong. Keynesianism is relevant to the extent that it was the dominant school of thought during the immediate post-war era. But, there are different ways to think about Keynesianism. Kyun argues that it was the formalization and econometric drive of post-war economics which reduced the richness of theory. Keynes was actually skeptical of this program. Many of Keynes’ followers also went down different paths, many of them towards research programs that are now relatively obscure (e.g. Hicks’ later work, Minsky, et cetera), but thick and dynamic. More recently, Keynes has been re-interpreted (e.g. Shackle, Leijonhufvud, and Steele) in “Austrian” terms, with an emphasis on coordination. This is analogous, I think, to what Murphy has in mind.

Nevertheless, I do think there is something to the belief that there exists in economics “modes of thinking,” determined by the emphases of the models that economics students learn. This is why schools that close themselves off tend to focus too little on the very things their models speak little of, and why larger, more open schools tend to have a more versatile, richer tradition. It’s also a very good reason why Austrians should consider themselves only as compliments to the rest of the Neoclassical tradition — not just because of what we can teach them, but because of what they can teach us.

I’m reading required segments of George A. Akerlof’s and Robert J. Shiller’s Animal Spirits for a Keynes–Hayek seminar held, here, in San Diego, over this weekend. They write the following,

Within a year of the publication of Keynes’ General Theory, John R. Hicks published a quantitative interpretation of Keynes that emphasized a rigid multiplier and the interaction of its effects with interest rates. Hicks’ version soon superseded Keynes’ original as the authoritative embodiment of Keynesian theory. Keynes was ruminating, discursive, disjoint, impenetrable, but nevertheless provocative and amusing; Hicks was orderly, efficient, and logically complete. Hicks’ version won the day. He is not as famous as Keynes, for he is often viewed as a mere interpreter of Keynes’ genius. But in terms of the history of thought, the “Keynesian revolution” was just as much a “Hicksian revolution.”

— George A. Akerlof and Robert J. Shiller, Animal Spirits (Princeton: Princeton University Press, 2009), p. 14.

Not to mention, in many ways Hicks’ work was original, or at least based on previous theoretical research of his.

I’ve been thinking about how much Keynes contributed to the economist of the post-war, and I’m beginning to convince myself that the answer is “not much.” Many of the ideas in The General Theory are so far out there that they were bound to be rejected by the brunt of the profession. A lot of the overarching ideas were already out there, including Keynes’ earlier work. Maybe The General Theory‘s role in the history of thought was as an impetus. But, what seems most influential about Keynes is this widely held belief that he is the “father of macroeconomics,” despite the fact that Keynes’ direct influence on economics is probably overstated.

We must look at the price system as such a mechanism for communicating information if we want to understand its real function — a function which, of course, it fulfills less perfectly as prices grow more rigid. (Even when quoted prices have become quite rigid, however, the forces which would operate through changes in price still operate to a considerable extent through changes in the other terms of the contract.) The most significant fact about this system is the economy of knowledge with which it operates, or how little the individual participants need to know in order to be able to take the right action.

— F.A. Hayek, “The Use of Knowledge in Society,” Individualism and Economic Order (Auburn: Ludwig von Mises Institute, 2009), p. 86 (emphasis mine).

Rigid prices make the pricing process less effective than it would otherwise be, but even in the case of rigid prices there are other means of conveying the same information. We also should step away from interpreting Hayek as arguing that relevant knowledge in inclusive in prices per sé. While prices do contain, or reflect, abstract information, looking at a price is not the only method of attaining relevant knowledge. Profit and loss, which are related to prices, are also a means of conveying information (and a good disciplining process). In fact, in this article and others (e.g. “The Facts of the Social Sciences”) Hayek mentions other means of conveying information in society, including language.

Of course, Hayek’s main intention is not to write about how awesome the pricing process is. In “Economics and Knowledge,” for instance, I get the sense that his purpose is to draw his readers towards the relevance of studying the market process. There he makes the point that if we always assume equilibrium then we assume away some of the main problems of economics. Not only this, but he argues that by focusing too much on the “pure logic of choice” we reduce the scope of economic analysis much too narrowly. Hayek wants to clarify the importance of looking at how society moves towards equilibrium, or, stated more broadly, how individuals coordinate with one another. He was pleading this case to the economics profession.

There are limits to Hayek’s work on prices and coordination. I think that Hayek needs to be supplemented with Mises’ emphasis on profit and loss. But, that Hayek isn’t the end all, be all is natural — there’s only so much one person can accomplish. This is why there is burgeoning Austrian scholarship on these same topics, in the tradition of Hayek, Mises, and Rothbard.

The theory of stick prices and monetary disequilibrium is sometimes attributed to New Keynesians. While New Keynesians have put effort into combining sticky prices into modern dynamic equilibrium models, the truth is that monetary disequilibrium theory precedes the post-war era. Neither is it a Keynesian theory generally. While there are vast differences between New Keynesians and the economics of Keynes, I thought it interesting to look at what the exact differences are between Keynes’ general theory of the output gaps and the theory of monetary disequilibrium. Any discussion should make clear that the two are not the same, even if Keynes’ theory, in some sense, borrows from earlier work on money shortages.

I have discussed the theory of monetary gluts before, but remember that its essential facet is that a rise in the demand for money will cause a fall in the demand of other goods. In a world of monetary exchange, where we don’t usually exchange commodities for commodities, a shortage of money is tantamount to a reduction in the volume of exchanges. Price rigidities do matter, because it may be that the market clearing price for a product is zero. Or, since not all prices move simultaneously, it may be that revenue doesn’t match the costs of production (or the firm may not be able to cover fixed costs). In other words, it may be preferable to restrict output. The driving force, in any case, is a rise in desired cash balances.

Keynes’ general theory has a different origin, although the dynamic process may be similar. The catalyst, though, isn’t necessarily a rise in desired cash balances. Society may want to save more and earn a return (interest) based on their time preference. One of the most important gears in his theory is the investment multiplier, where the rate of consumption indirectly decides the rate of investment. Accordingly, in a world of rising savings there must be a decrease in the proportion of income consumed, implying that the profitability of investment must also fall. This is a paradox between savings and the scope of investment, where a mismatch causes a monetary disequilibrium. This, then, leads to a restriction in output, because real wages cannot fall (any fall in nominal wages, according to Keynes, will occur to the same extent as a fall in the price of the goods labor produces).

The differences between the two theories lead to two different recommendations. A traditional monetary disequilibrium can be resolved, in lieu of sufficiently flexible prices, by an increase in the supply of money. Those who have already met their demand for money will use excess money to buy the goods of others’, letting those selling their output fulfill their desired cash balances. In Keynes’ framework this is not an optimal solution. He wanted to increase investment, but unless new money is used specifically to increase consumption, then attempting to restore monetary equilibrium will fail. His policy recommendations were much more specific: the partial socialization of investment and/or a reduction in the rate of interest, to lower the costs of capital.

The major incompatibility of the Austrian perspective and Keynes’ is not monetary disequilibrium. This latter concept should be considered part of the Austrian’s general body of theory, and it probably is for most Austrian economists. Instead, the important incongruity is the theory of the multiplier. According to Austrians, a fall in the proportion of consumption doesn’t lead to a reduction in the scope for investment. Rather, an increase in savings will cause a decrease in the cost of capital (in general) on its own. A rise in savings increases the scope for investment — the exact opposite of Keynes’ conclusion.

I should make clear that Keynes does consider, in part, what would be the Austrian critique of his theory. This is where his theory of interest comes in. In a pure loanable funds theory of interest, an increase in savings will lead to a fall in the market rate of interest due to competitive pressure (and the fact that it implies a fall in time preference — the rate of return to induce people to part with present consumption falls). Keynes didn’t believe this to be true, developing the liquidity preference theory of interest in response. There are various details for and against Keynes’ theory of interest and investment, but long story short he believed that the market rate for loanable funds may end up higher than time preference. I’ve read that even this issue could be solved by reestablishing monetary equilibrium, but I’d add that even if this were the case then the prices of capital goods would simply fall by more than what would otherwise be the case — in other words, there’s more to the cost of capital than just its rate of interest.

I don’t mean to turn this post into an attempted refutation of one or the other. I only want to make clear the differences between traditional monetary disequilibrium and Keynes’ business cycle doctrine. They are two very different animals, and it’s incorrect to the mix the two together.