Tuesday, November 9, 2010

Lecture 1: Theories of the Influence of Money on Prices

Hayek uses the first lecture in Prices and Production to lay out what he sees as the developmental trajectory of monetary theory, as well as an introduction to the advances he proposes to make in subsequent lectures. He identifies four stages of theory, the last of which is said to be in its infancy in 1931, when Hayek wrote the lectures.

The first stage is more or less what most people know of as the quantity theory of money. Hayek has two primary critiques of this approach: it is analytically problematic, and even if it weren't problematic it is practically useless. I think Hayek errs on both of these points (which is not to say that I disagree with him that monetary theory could be developed beyond the quantity theory). First, Hayek suggests that "none of these magnitudes as such ever exerts an influence on the decisions of individuals; yet it is on the assumption of a knowledge of the decisions of individuals that the main propositions of non-monetary theory are based". In this sense, Hayek sees a divorce between monetary theory and what we would know as microeconomics. He continues, "if, therefore, monetary theory still attempts to establish causal relations between aggregates or general averages, this means that monetary theory lags behind the development of economics in general. In fact, neither aggregates nor averages do act upon one another, and it will never be possible to establish necessary connections of cause and effect between them as we can between individual phenomena".

Towards the end of his discussion of the first stage, Hayek writes that "you are all sufficiently familiar with this type of theory to supply these [examples] for yourselves and to correct any exaggerations which I may have committed", and so I'll take him up on this and correct some exaggerations. What Hayek appears to miss here is that the quantity theory is useful not as a behavioral law or a statement of causal relations, but simply as a statement of accounting. If you take all the prices charged in an economy and add them up, and then take all the instances that a particular piece of money in the economy was spent and add all those up, the two sums have to be equal to each other. This is a definitional constraint on the system. Think of it as analogous to the "conservation of energy" in physics. On its own, the conservation of energy doesn't tell you anything about cause and effect, and it doesn't tell you about the dynamics of a particular system. Its power is in parameterizing the system so that when action does occur as the effect of some other cause, that action is forced to conform to the conservation of energy, and so the conservation of energy is a useful principle in understanding how cause and effect play out in physical systems. The same is true of the quantity theory of money. Hayek fundamentally misunderstands the very purpose of these theories if he thinks they imply some sort of causal relationship.

The second thing that Hayek critiques about this first stage is his claim that anything that the quantity theory can tell us is relatively useless. He writes "for none of these magnitudes as such ever exerts an influence on the decisions of individuals; yet it is on the assumption of the knowledge of individuals that the main proposition of non-monetary economic theory are based". This seems odd. The various problems associated with inflation and deflation are very well understood, and were most definitely understood when Hayek wrote these lectures. Fisher had not yet written his famous article on debt-deflation theory, but I don't think anyone has claimed that these ideas were especially original to Fisher. Keynes wrote extensively on the differential impact of inflation and deflation in 1923, and I know Hayek read that book. Debt-deflation, the differential impact of deflation on quits and layoffs, the differential impact on the distribution of wealth and the implications for demand, etc., can't be investigated by asking "what does the price do in this particular market - how do relative prices change?". These things can only be investigated by asking "what happens to the general price level?". So on both counts, I think Hayek is quite misguided in his critique of the "first stage" of monetary theory, although he's certainly right that there is more to be said.

The second stage Hayek identifies is the attempt by Locke, Montanari, Cantillon, and others to trace the path through which an increase in the money supply affects prices. The modern equivalent is something like the "monetary transmission mechanism", which has been so central to Ben Bernanke's own research (back when he did research!). He has a few other things on his plate right now, but even now this transmission mechanism problem has played a substantial role in how he has handled his tenure as Fed chairman relative to Greenspan. I was very interested in this section of Hayek's lecture to see some vague allusions to a long-run vs. short-run Phillip's Curve relationship: "Hume, however, makes it clear that, in his opinion, "it is only in this interval or intermediate situation, between the acquisition of money and the rise of prices, that the increasing quantity of gold and silver is favourable to industry".

The third stage of the development of monetary theory was the link between money and the interest rate. Hayek attributes this insight to many people, including Malthus, Mill, Bentham, etc. - but of course it is found in its most developed form in the hands of Knut Wicksell. Hayek muses that "by a curious irony of fate, Wicksell has become famous, not for his real improvements on the old doctrine, but for... his attempt to establish a rigid connection between the rate of interest and the changes in the general price level." This was interesting to read because I know Wicksell best for his "natural rate of interest" and loanable funds market theory of interest work, and not for whatever this other point is that Hayek is refering to. Clearly, the discipline has recognized in Wicksell as important exactly what Hayek recognized as important in 1931. That's encouraging! In 1931, Hayek felt that Wicksell was being acknowledged for the wrong thing, but now in 2010 we all acknowledge Wicksell for the work that Hayek thought was important.

I have another critique of Hayek here - I think he is far too hard on Wicksell when it comes to equilibrium and the price level. Hayek writes "according to Wicksell, the equilibrium rate of interest was a rate which simultaneously restricted the demand for real capital to the amount of savings available and secured stability of the price level" - taking issue with this position, Hayek counters that "the banks could either keep the demand for real capital within the limits set by the supply of savings, or keep the price level steady; but they cannot perform both functions at once. Except in a society in which there were no additions to the supply of savings, i.e., a stationary society, to keep the money rate of interest at the level of the equilibrium rate would mean that in times of expansion of production the price level would fall." Hayek's analysis here is right, of course, but I think his critique of Wicksell is misguided. Wicksell provides us with a loanable funds market equilibrium, right? The equilibrium position assumes a given demand for and supply of loanable funds. Of course if there is an "expansion of production the price level will fall", as Hayek says - but the whole point is if there is an expansion of production there will be a shift in the demand for loanable funds and there will be a new loanable funds market equilibrium. Imagine if you were describing a market for food and you noted that where supply and demand met you had an equilibrium price where the real price level would be stable. Hayek here is saying "well that doesn't make sense because if you make more bread the price level will fall!". Certainly it will, but that's because the supply curve has shifted. So, just as I would not take his critique of the quantity theory very seriously, I would not take his critique of Wicksell very seriously. There are ways to complicate and add to the loanable funds market, but there's nothing fundamentally wrong with it.

The fourth stage of the development of monetary theory is of course Hayek's turf: the impact of changes in the quantity of money on relative prices, particularly prices of the same good over time. I can't argue with the fact that this is one avenue to develop. I've never accused this version of ABCT of being illogical. Hayek is a little vague on why exactly this is such an important culmination of prior monetary theory. It appears to me to be an interesting side-issue to look into, not a conclusion towards which monetary work was naturally heading. But of course, my bias in this regard is probably obvious. If you asked a group of people reasonably well educated in the history of economic thought "which early twentieth century economist earned his fame for his work bringing monetary economics into mainstream economic theory", the answer would not be Hayek. The answer, in all likelihood, would be Keynes. This is not to say that Keynes invalidates Hayek's work. This was, at it's core, both of their projects. They both sought to make monetary theory relevant for the broader world of economic theory. My feeling is that Hayek took this project down a detour. It is an interesting detour and probably an accurate detour, but a much less significant path than the path that Keynes took. This seems to be the assessment of Hayek and Keynes's peers and descendants as well. Money is important because it is the medium through which demand is translated into effective demand, and the quantity of circulating medium substantially influences the efficacy of demand in stimulating production. I don't want to make this a post about Keynes, but the similarity between their missions was absolutely unmistakable in reading this first lecture.

I guess my lingering question is "why is this the fourth stage, other than because Hayek is writing about it and he wants to put it in the context of earlier stages?" Any answers? Why should I care more about relative prices than about the impact of money on effective demand?
*****
Other discussions of Hayek's first lecture tied into our reading group (graciously organized by Jonathan Catalan) can be found here, here, and here. Please let me know if I've missed one.

11 comments:

  1. I feel as if your critique of Hayek's criticism of the first stage really misses the essence of Hayek's point.

    First, the quantity theory does explain a casual relationship. It explains that if m, v, or q changes so will p. It illustrates a very basic relationship between these variables. What Hayek is critiquing, and what I'm trying to get across in Lee Kelly's blog, is that at this the MV=PQ is practically worthless, because the relationships are many times as complicated as that.

    By the way, I have trouble seeing MV=PQ holding even as an accounting identify under specific circumstances, namely during liquidity traps. Even during periods of healthy growth, the real nature of inflation oftentimes means that the "average price level" (or P) does not oftentimes reflect the actual quantity of money in circulation. Also, there are more specific Austrian criticisms of the concept of velocity (not that there is no such thing as velocity, but that given the nature of price formation changes in velocity form '1' to '3' won't create a proportional rise in prices).

    Second, I think your attempt to link theories like debt-deflation to the general price level is erroneous and misguided. Although, tt probably explains why all of these theories have been subpar themselves. (Acknowledging that these are theories you apparently believe in. However, they are theories which Austrians disagree with, and you should be aware of this when you make your criticism.)

    In any case, I don't think Hayek would say that all theory based on MV=PQ is useless. He's just saying that the focus on the mechanistic quantity theory of money has led monetary theory astray, because it has ignored the effect of money on the real economy (namely, the capital structure).

    Regarding Wicksell, you write,

    "but the whole point is if there is an expansion of production there will be a shift in the demand for loanable funds and there will be a new loanable funds market equilibrium."

    Hayek addresses this in the lecture. This doesn't maintain equilibrium. Hayek argues (partly in what you quote) that you can either *try* maintain a stable price level or you can allow the market to form an equilibrium rate of interest. The new rate of interest after the incrase in Q is the natural rate of interest, and so stable price level and maintaing the natural rate of interest are mutually exclusive.

    ReplyDelete
  2. If M, V, or P change then Q doesn't have to change. The change in P could offset the entire change in M and V. What I think you mean is that if M changes and V and P stay constant, then Q has to change. This is true, but (1.) this isn't causal - so far it's simply a correlation, and (2.) you can't really just say "holding V and P constant", because you can't hold V and P constant. They each have their own causal determinants, so it's silly to try to derive action on these variables from the quantity theory (that's akin to deriving fiscal policy from Y=C+I+G). What it does provide is a constraint on the system that's useful for modeling.

    Could you explain why this wouldn't hold in a liquidity trap or why this wouldn't hold in other circumstances? I haven't read your conversation with Lee Kelly yet (I will), but I'm not sure why you think this doesn't hold true.

    However, they are theories which Austrians disagree with, and you should be aware of this when you make your criticism.

    Right! My criticism is that he disagrees with them (and disagrees without comment here, for that matter). I wouldn't take this too far, though I was always under the impression Hayek was cocerned about deflation for precisely the reasons that Fisher, Keynes, and others lay out. It just didn't seem to make it into this lecture.

    Hayek addresses this in the lecture. This doesn't maintain equilibrium. Hayek argues (partly in what you quote) that you can either *try* maintain a stable price level or you can allow the market to form an equilibrium rate of interest. The new rate of interest after the incrase in Q is the natural rate of interest, and so stable price level and maintaing the natural rate of interest are mutually exclusive.

    Right, but the new price level and natural rate of interest will be stable. Clearly both will have to change, but they are both stabilized at the same point. My concern is that Hayek is critiquing a static model for not being dynamic. Fair enough - dynamic models are nice. But Wicksell's static model is right as a static model.

    ReplyDelete
  3. "If M, V, or P change then Q doesn't have to change. The change in P could offset the entire change in M and V. What I think you mean is that if M changes and V and P stay constant, then Q has to change."

    I actually didn't say anything about changes in Q as a result of changes in M, V, or P. Please re-read my comment.

    Did you purposely misconstrue my point? Or that there is a causal relationship between changes in m (or v, or q) and p.

    "Could you explain why this wouldn't hold in a liquidity trap or why this wouldn't hold in other circumstances?"

    Because there are times where there is an increase in M, but not a proportional rise in P. MV=PQ doesn't account for these situations.

    "Right! My criticism is that he disagrees with them..."

    Okay, then maybe you're better off making cases for them, instead of saying, "Hayek is wrong, because these theories are right!"

    "Right, but the new price level and natural rate of interest will be stable."

    Um, this doesn't address the idea of price stability. Price stability infers that prices remain the same. Hayek is arguing that with an increase in quantity either prices will fall, and there will still be intertemporal equilibrium, or you can increase m in an effort to maintain prices, but at the same time disrupt that intertemporal equilibrium.

    ReplyDelete
  4. I actually didn't say anything about changes in Q as a result of changes in M, V, or P. Please re-read my comment.

    OK. Is there a substantive difference between talking about m, v, and p vs. q or m, and q vs. p? I misread p and q but I'm a little unclear on how this matters.

    "Because there are times where there is an increase in M, but not a proportional rise in P. MV=PQ doesn't account for these situations."

    In a liquidity trap, liquidity preference increases which means V decreases (V is the reciprocal of liquidity preference as it's usually conceived). So M goes up, V goes down, and because the interest rate can't go down (the interest rate simply being a very important "p" to consider), the burden is bourne by Q. Is that not what happens in a liquidity trap? Lots of M, rigidity in P, a decrease in V, and a drop in Q to make up the difference? How does the quantity theory not communicate these dynamics?

    "Okay, then maybe you're better off making cases for them"

    I suppose I would feel more obligated if Hayek spent any time on it. He didn't, so I'm just noting that while on page 200 Hayek said "everybody agrees that a change of prices would be of no consequence whatever if all prices in the widest sense of the term were affected equally and simultaneously". He actually says this. My point is that's not true at all. Everybody clearly does not agree on this, and Hayek should have known that. Now perhaps Hayek disagrees with "everybody" but he cannot claim that "everybody agrees" here. I should have quoted that line from page 200 in the post. Maybe I'll write another short post elaborating.

    "Hayek is arguing that with an increase in quantity either prices will fall, and there will still be intertemporal equilibrium, or you can increase m in an effort to maintain prices, but at the same time disrupt that intertemporal equilibrium."

    I understand this. I'm saying that Hayek is looking at Wicksell's static equilibrium in the loanable funds market and critiquing it for not being an intertemporal equilibrium. It's fine to say "the intertemporal coordination is important to consider". I think it's wrong to say "his position on static equilibrium is wrong because it is not intertemporal".

    ReplyDelete
  5. My point is that MV=PQ does imply a causal relationship. It was the entire purpose behind the quantity theory of money - to show how changes in the supply of money could affect prices.

    In regards to the liquidity trap, are we seeing in a fall in Q or V today, proportional to the increase in M? My argument is not that these variables don't exist, or don't change, or don't hold the relationship implied by the quantity theory of money, rather that the quantity theory of money is an insufficient model to accurately illustrate the relationship.

    Also, as I mentioned in a previous comment, we're not only discussing the liquidity trap, but also the boom period between 2002 and 2007, with a very large increase in M, but not necessarily with a proportional changes in V, P, or Q.

    The point is that the formula masquerades relative changes, that might not reflect on what are essentially averages, such as V and P.

    Now, moving on to changes in prices, obviously Hayek is referring the price of debt, as well, so his point holds.

    Finally, regarding Wicksell's theory, Hayek is pointing out a weakness in Wicksell's thinking. I'm not sure what's so difficult to understand.

    ReplyDelete
  6. "Debt-deflation, the differential impact of deflation on quits and layoffs, the differential impact on the distribution of wealth and the implications for demand, etc., can't be investigated by asking "what does the price do in this particular market - how do relative prices change?". These things can only be investigated by asking "what happens to the general price level?""

    This is incorrect.

    Deflation, like inflation, does not affect all industries and all prices equally nor simultaneously. Certain industries profit from deflation, and some (definitely) lose by it. Asking about impact differentials with regard to deflation can only make sense in Hayek's sense, because the general price level tells you nothing about the shift of relative prices.

    You cannot look at the general price level to discern the impact of deflation.

    ReplyDelete
  7. 1. You and I seem to have a different understanding of the quantity theory of money. As I've said, I don't see it as a causal relationship and I don't think it was intended to be that. I think it was intended to be an accounting relationship.

    2. Yes, the change in M is exactly proportional to the change in PQ/V. It has to be. How couldn't it be? If it's not, then we are not measuring the same thing described in the theory.

    I'm deeply confused about why you think the quantity hasn't help either between 2002 and 2007 or between 2007 and today. Why are you saying the changes aren't proportional? I understand you think they aren't, but I don't see how they couldn't be. Are you saying that CPI might not exactly capture P? I could agree with you on that.

    ReplyDelete
  8. Mattheus, you're simply wrong. The only reason why inflation or deflation are considered harmful is because it is impacting all prices at once. The generality of the deflation is precisely the problem. You can assume that all relative prices stay precisely the same. So let's say all prices decline by 10%. No change in relative prices at all. Hayek is saying nobody cares about this. But lots of people - lots of contemporaries of Hayek - have demonstrated that these sorts of general price movements do matter and affect people differently.

    Now - you are right in the sense that the way in which inflation and deflation actually plays out doesn't fall equally on all sectors. Sure. And there will be relative price effects there. But general movements will still matter. Think of a change in average prices. Some of it will be due to changes in relative prices, and Hayek has useful things to say about these price changes. But some of it will be borne by all prices and that general price change does matter.

    ReplyDelete
  9. I honestly did not start this day expecting that I would be arguing over whether the quantity theory of money has held true in the last decade.

    I always thought the quantity theory was the one thing that ALL OF US agreed on!

    ReplyDelete
  10. Why is consideration of a general price level important?

    No acting man pays heed to a general price level, but only to relative prices that govern his ability to acquire goods that satisfy his wants.

    You can assume that all relative prices stay precisely the same.

    If all relative price relationships stay the same, but all prices decline by 10%, where's the harm? Your income declines by just as much as your cost of living.

    I always thought the quantity theory was the one thing that ALL OF US agreed on!

    Nonsense. We all agree on praxeology, duh.

    ReplyDelete
  11. Interesting:
    "Hayek: You know, about forty years ago, I once wrote a sentence something like this: Almost the worst thing which could happen would be if mankind ever forgot the quantity theory of money—except they should ever take it literally. While I still believe that it is true that the price level is determined from the quantity of money, we never know what the quantity of money in this sense is. I think the rule ought to be that whoever issues the money must adapt the quantity so that the price level will remain stable. But to believe that there is a measurable magnitude which you can keep constant, with beneficial effects, I regard as completely wrong."

    http://hayekcenter.org/?p=354

    ReplyDelete

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