Jan, commenting on my recent post about Krugman, Minsky and IS-LM, quoted the penultimate paragraph of J. R. Hicks’s 1980 paper on IS-LM in the Journal of Post-Keynesian Economics, a brand of economics not particularly sympathetic to Hicks’s invention. Hicks explained that in the mid-1930s he had been thinking along lines similar to Keynes’s even before the General Theory was published, and had the basic idea of IS-LM in his mind even before he had read the General Theory, while also acknowledging that his enthusiasm for the IS-LM construct had waned considerably over the years.
Hicks discussed both the similarities and the differences between his model and IS-LM. But as the discussion proceeds, it becomes clear that what he is thinking of as his model is what became his model of temporary equilibrium in Value and Capital. So it really is important to understand what Hicks felt were the similarities as well as the key differences between the temporary- equilibrium model, and the IS-LM model. Here is how Hicks put it:
I recognized immediately, as soon as I read The General Theory, that my model and Keynes’ had some things in common. Both of us fixed our attention on the behavior of an economy during a period—a period that had a past, which nothing that was done during the period could alter, and a future, which during the period was unknown. Expectations of the future would nevertheless affect what happened during the period. Neither of us made any assumption about “rational expectations” ; expectations, in our models, were strictly exogenous.3 (Keynes made much more fuss over that than I did, but there is the same implication in my model also.) Subject to these data— the given equipment carried over from the past, the production possibilities within the period, the preference schedules, and the given expectations— the actual performance of the economy within the period was supposed to be determined, or determinable. It would be determined as an equilibrium performance, with respect to these data.
There was all this in common between my model and Keynes’; it was enough to make me recognize, as soon as I saw The General Theory, that his model was a relation of mine and, as such, one which I could warmly welcome. There were, however, two differences, on which (as we shall see) much depends. The more obvious difference was that mine was a flexprice model, a perfect competition model, in which all prices were flexible, while in Keynes’ the level of money wages (at least) was exogenously determined. So Keynes’ was a model that was consistent with unemployment, while mine, in his terms, was a full employment model. I shall have much to say about this difference, but I may as well note, at the start, that I do not think it matters much. I did not think, even in 1936, that it mattered much. IS-LM was in fact a translation of Keynes’ nonflexprice model into my terms. It seemed to me already that that could be done; but how it is done requires explanation.
The other difference is more fundamental; it concerns the length of the period. Keynes’ (he said) was a “short-period,” a term with connotations derived from Marshall; we shall not go far wrong if we think of it as a year. Mine was an “ultra-short-period” ; I called it a week. Much more can happen in a year than in a week; Keynes has to allow for quite a lot of things to happen. I wanted to avoid so much happening, so that my (flexprice) markets could reflect propensities (and expectations) as they are at a moment. So it was that I made my markets open only on a Monday; what actually happened during the ensuing week was not to affect them. This was a very artificial device, not (I would think now) much to be recommended. But the point of it was to exclude the things which might happen, and must disturb the markets, during a period of finite length; and this, as we shall see, is a very real trouble in Keynes. (pp. 139-40)
Hicks then explained how the specific idea of the IS-LM model came to him as a result of working on a three-good Walrasian system in which the solution could be described in terms of equilibrium in two markets, the third market necessarily being in equilibrium if the other two were in equilibrium. That’s an interesting historical tidbit, but the point that I want to discuss is what I think is Hicks’s failure to fully understand the significance of his own model, whose importance, regrettably, he consistently underestimated in later work (e.g., in Capital and Growth and in this paper).
The point that I want to focus on is in the second paragraph quoted above where Hicks says “mine [i.e. temporary equilibrium] was a flexprice model, a perfect competition model, in which all prices were flexible, while in Keynes’ the level of money wages (at least) was exogenously determined. So Keynes’ was a model that was consistent with unemployment, while mine, in his terms, was a full employment model.” This, it seems to me, is all wrong, because Hicks, is taking a very naïve and misguided view of what perfect competition and flexible prices mean. Those terms are often mistakenly assumed to meant that if prices are simply allowed to adjust freely, all markets will clear and all resources will be utilized.
I think that is a total misconception, and the significance of the temporary-equilibrium construct is in helping us understand why an economy can operate sub-optimally with idle resources even when there is perfect competition and markets “clear.” What prevents optimality and allows resources to remain idle despite freely adjustming prices and perfect competition is that the expectations held by agents are not consistent. If expectations are not consistent, the plans based on those expectations are not consistent. If plans are not consistent, then how can one expect resources to be used optimally or even at all? Thus, for Hicks to assert, casually without explicit qualification, that his temporary-equilibrium model was a full-employment model, indicates to me that Hicks was unaware of the deeper significance of his own model.
If we take a full equilibrium as our benchmark, and look at how one of the markets in that full equilibrium clears, we can imagine the equilibrium as the intersection of a supply curve and a demand curve, whose positions in the standard price/quantity space depend on the price expectations of suppliers and of demanders. Different, i.e, inconsistent, price expectations would imply shifts in both the demand and supply curves from those corresponding to full intertemporal equilibrium. Overall, the price expectations consistent with a full intertemporal equilibrium will in some sense maximize total output and employment, so when price expectations are inconsistent with full intertemporal equilibrium, the shifts of the demand and supply curves will be such that they will intersect at points corresponding to less output and less employment than would have been the case in full intertemporal equilibrium. In fact, it is possible to imagine that expectations on the supply side and the demand side are so inconsistent that the point of intersection between the demand and supply curves corresponds to an output (and hence employment) that is way less than it would have been in full intertemporal equilibrium. The problem is not that the price in the market doesn’t allow the market to clear. Rather, given the positions of the demand and supply curves, their point of intersection implies a low output, because inconsistent price expectations are such that potentially advantageous trading opportunities are not being recognized.
So for Hicks to assert that his flexprice temporary-equilibrium model was (in Keynes’s terms) a full-employment model without noting the possibility of a significant contraction of output (and employment) in a perfectly competitive flexprice temporary-equilibrium model when there are significant inconsistencies in expectations suggests strongly that Hicks somehow did not fully comprehend what his own creation was all about. His failure to comprehend his own model also explains why he felt the need to abandon the flexprice temporary-equilibrium model in his later work for a fixprice model.
There is, of course, a lot more to be said about all this, and Hicks’s comments concerning the choice of a length of the period are also of interest, but the clear (or so it seems to me) misunderstanding by Hicks of what is entailed by a flexprice temporary equilibrium is an important point to recognize in evaluating both Hicks’s work and his commentary on that work and its relation to Keynes.
Uneasy Money 
David,
You said:
“Overall, the price expectations consistent with a full intertemporal equilibrium will in some sense maximize total output and employment, so when price expectations are inconsistent with full intertemporal equilibrium, the shifts of the demand and supply curves will be such that they will intersect at points corresponding to less output and less employment than would have been the case in full intertemporal equilibrium.”
You may be interested in this picture (graph at link) where, under conditions of imperfect information flow, the price falls somewhere in the triangle below the supply and demand curve:
http://informationtransfereconomics.blogspot.com/2014/05/the-effect-of-expectations-in-economics.html
I associate this deviation with the “information loss” (KL divergence, to be precise) between the actual distribution of the future prices and the expected distribution of the future prices.
LikeLike
The point should not be missed that IS-LM is not Keynesian, but Hicksian, and that Hicks later described it as not applicable in conditions of uncertainty.
LikeLike
David
Great Article. What you might not know is that late in life Hicks himself recanted on the strict equilibrium assumptions of his original IS-LM model in n 1980 – IS-LM: an explanation – in Journal of Post Keynesian Economics.
He said
‘I accordingly conclude that the only way in which IS-LM analysis usefully survives — as anything more than a classroom gadget, to be superseded, later on, by something better – is in application to a particular kind of causal analysis, where the use of equilibrium methods, even a drastic use of equilibrium methods, is not inappropriate. …
I have paid no attention, in this article, to another weakness of IS-LM analysis, of which I am fully aware; for it is a weakness which it shares with General Theory itself. It is well known that in later developments of Keynesian theory, the long-term rate of interest (which does figure, excessively, in Keynes’ own presentation and is presumably represented by the r of the diagram) has been taken down a peg from the position it appeared to occupy in Keynes. We now know that it is not enough to think of the rate of interest as the single link between the financial and industrial sectors of the economy; for that really implies that a borrower can borrow as much as he likes at the rate of interest charged, no attention being paid to the security offered. As soon as one attends to questions of security, and to the financial intermediation that arises out of them, it becomes apparent that the dichotomy between the two curves of the IS-LM diagram must not be pressed too hard.’
Davidson the editor stated ‘ When I first started to think about the ergodic vs. nonergodic dischotomy, I sent to Hicks some preliminary drafts of articles I would be writing about nonergodic processes. Then John and I met several times to discuss this matter further and I finally convinced him to write the article — which I published in the Journal of Post Keynesian Economics– in which he renounces the IS-LM apparatus. Hicks then wrote me a letter in which he thought the word nonergodic was wonderful and said he wanted to lable his approach to macroeconomics as nonergodic!’
Nonegodic = path dependent temporary equilibrium – voila
The islm ‘gardget’ can be rescued. It assumes loanable funds, it assumes no limites to borrowing. Im my own work I have modeled the demand for money curve s a demand for leverage using Ole peters non ergodic theory of rational leverage, and the IS curve using a stock flow consistent model whereby banks lending is limited by their profits, equity and collatorol. This has produced fascinating results as it still produces a liquidity trap but also a hicksian ;credit deadlock’ and secular stagnation following a financial crash. I would withdraw the debunking of liquidity preference theory in that article though, it can still survive as a liquidity markup.
http://andrewlainton.wordpress.com/2013/07/26/a-simple-post-keynesian-alternative-to-is-lm/
LikeLike
Is Lucas 1972 a temporary equilibrium model in just that sense?
LikeLike
Interesting, the point of models and the point of reading dead (or otherwise) economists is roughly the same: illuminate an inquiry into how the economy works.
Its always a fine line between learning the model and learning prior wisdom as tools for ones own inquiry and having that learning become the end and goal itself.
I don’t think your post reflects this but I do get the uncomfortable impression that the antecedent and some comments do skirt over the line to making Hicks views the goal rather than a guide post towards a destination (shared as a destination by Hicks as well, naturally).
If HIcks did have a murky grasp of expectations he’d be in crowded company and it would go a long way to explaining why Keynes wrote the Book, and Hicks the commentary.
Was it Mark Thoma who said the only good economist is a dead economist? I paraphrase, perhaps it was that the good economics came from dead economics. Either way,
Long live Keynes!
Who whatever he actually wrote, always managed to convey that the ideas written were merely the tip of an iceberg, the smaller part of a profound understanding.
LikeLike
Hicks compares his flex-price model to Keynes’s model where at least the price of labor is endogenously set.
I’m struggling a bit to see how if one assumes perfectly flexible wages (and prices) one could get an equilibrium at less than full employment.
I can see how suppliers may be pessimistic about the future and produce less than is required to produce full employment. But this would lead to falling wages. Falling wages would lead to falling prices , and a Pigou effect that would lead to increased demand and (presumably) increased optimism among suppliers that would (with perfectly flexible prices) quickly lead to a full employment equilibrium.
I suppose if you assume pessimism among suppliers that is not changed by falling prices and increased money-wealth then you could get stuck in a low expectations equilibrium – but with flex-price assumed these assumption seems rather heroic to me.
LikeLike
David Glasner: Did you notice the dejected and depressed tone in that article by Sir John R. Hicks?
LikeLike
Jason, Thank for the link. From a quick look at your diagram, I am not sure that we are making the same argument. My view is that supply tends to be elastic at the expected price, so that if demand turns out to be less than expected so that the equilibrium price turns out to be less than expected and the amount supplied falls a lot compared to what would have happened if the expectation of the price fall had been expected.
demandsideforecast , Well, it’s not strictly Keynesian, but it is Hicks’s take on Keynes, so I don’t think one should assume that it has nothing to do with Keynes. The uncertainty part is not in the diagram, but there is still an equilibrium concept that Keynes was describing and IS-LM does capture the determinants of that equilibrium state.
Andrew, Thanks for the link to the Hicks article. Very interesting. As you might have noticed, I wrote a further post discussing one aspect of the article. Perhaps I will have more to say about it down the road.
Nick, Sorry, not sure which Lucas article you are referring to? Do you mean Lucas’s paper on monetary neutrality?
Dan, Well, I think before we can make progress in applying the ideas of a dead economist to current problems or using the ideas to generate new ideas of our own, we need to understand what the dead economist was actually saying. So if I did cross the line, I’m sorry, but I make no apologies.
Rob, If we define full employment as the intersection of a demand and supply curve, then it is necessarily true that whatever level of employment corresponds to that intersection is full employment. If we define full employment as some “normal” level of employment that the economy usually generates, or some potential high level of employment that we believe the economy is capable of generating, then it is very possible that the level of employment corresponding to the intersection of the demand and supply curve could under some circumstances (such as I discuss in this post) could be less than normal or potential full employment.
Blue Aurora, Not really, but I wasn’t thinking about his state of mind. You are probably right that Hicks was not entirely happy with how IS-LM ended up.
LikeLike
David,
Thanks for the reply,
You say “then it is very possible that the level of employment corresponding to the intersection of the demand and supply curve could under some circumstances (such as I discuss in this post) could be less than normal or potential full employment”
I can see how that would happen. Oddly enough I can envisage it more easily in a barter economy: A fruit producer cuts back his production due to lowered expectations of future sales. He fires some workers. They would be willing work for less, but as they get paid directly in the fruit they produce lowering wages will have only a small effect on employment levels compared to the initial fall in expectations (especially if consumption of fruit by workers is a large part of total consumption).
However I find it harder to see how this could happen in a commodity money world. Here when workers accept lower wages this will lead to a lowering of the money price level and an increase in the value of money, which will cause people to spend more , which will increase expectations of future sales and address the fall in expectations that drove the initial slowdown.
Still trying to think thru how this would work in a credit-money world.
LikeLike