Monday, January 14, 2013

Paul Davidson savages the mainstream

Below Peter Radford's critique of economic models (the topic of yesterday's four o'clock) one finds a comment from Paul Davidson:

Keynes drew an accurate portrait of an entrepreneurial economy that uses money contracts (not real contracts) to organize all market production and exchange transactions.(Does this not sound like the world of experience?)

Unfortunately Samuelson, Hicks and others mistook Keynes’s analysis for nothing more than another neoclassical model with sticky wages and prices even though one of the chapters in Keynes’s GENERAL THEORY is entitled “Changes in Money wages”. This chapter shows that even with perfectly flexible wages, there is no automatic mechanism to restore full employment if the economy should suddenly experience a recession.

I finally convinced Sir John Hicks that his ISLM system was not Keynes — and published an article by Hicks entitled “ISLM: An Explanation” in the Journal of Post Keynesian Economics . In this article Hicks admitted that ISLM was not Keynes!! and Hicks ultimately signed on to my argument that the Keynes analysis was based on the assumption that uncertainty meant a nonergodic system.

I have written a textbook entitled POST KEYNESIAN MACROECONOMIC THEORY which places Keynes’s money contract, nonergodic portrait of the real world of experience against the mainstream fictional world of Samuelson, Friedman, Rational expectations, New Keynesianism, etc...

Echoing Radford's terminology, Davidson contrasts the portrait Keynes created to the caricatures created by others. Exactly so.

"...even with perfectly flexible wages, there is no automatic mechanism to restore full employment..."

That's the line that got me. Paul Davidson knows his Keynes.

I've heard before, of Hicks rejecting his own ISLM thing, which I understand is kind of a big deal (even though I don't understand the ISLM).

Now it turns out that Davidson convinced Hicks to reject the ISLM, and that Davidson published Hick's rejection article in his (Davidson's) magazine journal.

Wow. I never heard of Paul Davidson before, but I've heard of him now.

I got stuck on "nonergodic". Wikipedia didn't help:

In mathematics, the term ergodic is used to describe a dynamical system which, broadly speaking, has the same behavior averaged over time as averaged over the space of all the system's states (phase space).

Time and space? I Googled Paul Davidson and found an article at Naked Capitalism that was useful. Before Davidson's article is an introduction by Philip Pilkington (someone I have heard of). Pilkington writes:

In a recent interview I asked the US’s leading post-Keynesian economist and founder of the Journal of Post-Keynesian Economics, Paul Davidson to discuss what is known as the ‘ergodic axiom’ in economics. This is a particularly important axiom as it allows mainstream economists (including left-wing Keynesians like Paul Krugman and Joseph Stiglitz) to claim that they can essentially know the future in a very tangible way. It does this by assuming that the future can be known by examining the past.

Without this axiom the whole edifice of mainstream theory rests on very shaky grounds. Yet, it should be clear to anyone that given that the theory is supposed to explain human behaviour it is unlikely that the future will correlate with the past because people and institutions tend to change and evolve over a given period of time.

Okay. An "ergodic" economy is predictable, because the future is similar to the past. So a nonergodic system is unpredictable.

This must be similar to the "unit root" thing. A system either has or does not have a unit root. In the one case, there is every reason to expect GDP to return to trend after a major setback. In the other case, there is no reason to expect it.

The economy is either ergodic or it is not. In the one case, there is every reason to assume tomorrow will be similar to today. In the other case, there is no reason to assume such a thing.

Pilkington continues:

Yes, often past behaviour will help us understand future behaviour – apply this in a simple psychological way to anyone you know and you will find it to be true – however, it should be quite clear that all future behaviour cannot be wholly explained by past behaviour. Clearly it should be quite obvious that the same should apply when we consider large aggregates of individuals and yet mainstream economics steadfastly refuses to accept this.

Now, broadening the topic, I think of the world in terms of business cycles, large and small. Overlapping cycles, such as Schumpeter described.

Lots of economists have described lots of cycles. I don't remember most of them, but I accept the idea of cycles generally. I think in terms of four or five:

1. The business cycle, where we get a recession every nine or ten years on average, give or take. It doesn't matter exactly how long the cycle is. It doesn't matter that it varies in length. People have recognized this cycle since the time of Alexis de Tocqueville, if not before. So, since 1935-1840 or before.

2. The Kondratieff wave, or the "long wave". I use the words "wave" and "cycle" interchangeably. The long wave, I think that's the same one Greenspan spoke of when he said you get one of these major financial crises every 80 to 100 years. (Maybe not. Maybe these are two separate cycles. That would make five, then.)

3. The Price waves described by David Hackett Fischer in his book The Great Wave: Price Revolutions and the Rhythm of History. Fischer describes

four very long waves of rising prices, punctuated by long periods of comparative price equilibrium. This is not a cyclical pattern. Price revolutions have no fixed and regular periodicity. Some were as short as eighty years; others as long as 180 years. They differed in duration, velocity, magnitude, and momentum.

(Excerpt from a very large PDF, a powerpoint type presentation that highlights key points from Fischer's book.)

4. The Cycle of Civilization, pretty much as described by Arnold Toynbee -- but driven by economic forces. Specifically: driven by the human desire to accumulate wealth. Call it the "economic security" motive. Call it "self interest".

Assume a system at equilibrium. Posit a disturbance to the system. Feedback arises from the disturbance. There are two kinds of feedback. One kind helps the disturbance dissipate, so the system returns to equilibrium. The other kind makes the disturbance bigger, so the system moves away from equilibrium.

Both disturbances may come into play in the same system. If they alternate, the effect may be to create a cyclical pattern. Wikipedia provides an example of stock prices, which I summarize:

1. A disturbance: Stock prices begin to rise.
2. People react by trying to get in early.
3. People getting into the market makes the disturbance bigger. Prices rise more.
4. But there is "the knowledge that there must be a peak". People get out early.
5. People getting out slows the increase of stock prices and creates a peak.
6. Now the disturbance is different: Stock prices begin to fall.

The process continues, driving prices to a bottom. Then the disturbance toggles again, and prices begin to rise.

The pattern thus created is a cyclical pattern, a repeating rise and fall of (in this case) stock prices. The cyclical pattern is generated by the plans and preferences of the people taking part in the process.

Cycles arise naturally from economic activity.

Davidson's article (it would be a 3 or 4-page printout) is a bit long for me, but very readable. He writes, for example:

Logically, to make statistically reliable probabilistic forecasts about future economic events, today’s decision-makers should obtain and analyze sample data from the future.

It's funny because it's true.

Davidson writes:

In simple language, the ergodic presumption assures that economic outcomes on any specific future date can be reliably predicted by a statistical probability analysis of existing market data.

Okay. I'm not ergodic, I guess. I don't make predictions. (Or maybe I'm a monkey wrench that brings the whole ergodic system down: If somebody can't foretell the future, the ergodic system fails!)

Wow, the whole "in simple language..." paragraph is great. Here's the rest of it:

By assumption, New Classical economic theory imposes the condition that economic relationships are timeless or ahistoric ‘natural’ laws. The historical dates when observations are collected do not affect the estimates of the statistical time and space averages. Accordingly, the mainstream presumption (utilized by both New Classical economists and New Keynesian economists) that decision-makers possess rational expectations imply that people in one’s model process information embedded in past and present market data to form statistical averages (or decision weights) that reliably forecast the future. Or as 2011 Nobel Prize winner Thomas Sargent [1993, p. 3], one of the leaders of the rational expectations school, states “rational expectations models impute much more knowledge to the agents within the model (who use the equilibrium probability distributions)… than is possessed by an econometrician, who faces estimation and inference problems that the agents in the model have somehow solved”.

If capitalism is the high point of the cycle of civilization -- the 150-year "limiting point" Keynes described as "the greatest age of the inducement to invest" -- and if Toynbee's picture of history is generally correct, then yes: It is unsafe to assume that tomorrow will be like today. I'm reminded of something Robert Heilbroner wrote. Heilbroner did not expect business civilization to survive another 500 years.

Today, it seems, nobody does.

Paul Davidson:

Keynes’s uncertainty concept implies that the future is transmutable or creative in the sense that future economic outcomes may be permanently changed in nature and substance by today’s actions of individuals, groups (e.g., unions, cartels) and/or governments, often in ways not even perceived by the creators of change.

In other words, as Sarah Connor put it in the picnic table: no fate. The cycle of civilization is not set in stone. It is a pattern that tends to emerge from the aggregate behavior of individuals. I would add that Toynbee was right: the dynamic is "challenge and response", and civilizations die by suicide.

The economy is challenging us, and if we fail to find the right response, it's over.

So if you want to preserve capitalism, you might want to start by reconsidering your assumptions. And even if you don't.

Okay. "Ergodic" means the outcome is known, or that you can calculate all the probabilities for all the possible outcomes. Uncertainty means you can't.

Look! -- Look! -- Look at this:

In the classical (ergodic) theory, where all outcomes are conceptually calculable, there is never a need to keep options open. People will therefore spend all they earn on the products of industry (Say’s Law) and there can never be a lack of effective demand to prevent the system from reaching full employment.

See where Davidson's going now?

You know, I know Paul Davidson is right, that he has his Keynes right, because Davidson boils his argument down to Say's law. And Keynes said

The classical theorists resemble Euclidean geometers in a non-Euclidean world who, discovering that in experience straight lines apparently parallel often meet, rebuke the lines for not keeping straight as the only remedy for the unfortunate collisions which are occurring. Yet, in truth, there is no remedy except to throw over the axiom of parallels and to work out a non-Euclidean geometry. Something similar is required today in economics.

And, writing of Say's law, Keynes said

It is, then, the assumption of equality between the demand price of output as a whole and its supply price which is to be regarded as the classical theory’s ‘axiom of parallels’.

Paul Davidson is saying that, too.


Tom Hickey said...

Art, you may be interested in this Amazon review of Davidson's PKE macro text:

Keynes's model of his theory of effective demand is in chapter 20,not chapter 3 June 1, 2006

By Michael Emmett Brady VINE™ VOICE

In mid 1935, Keynes wrote an extensive reply to Dennis Robertson's mathematically illiterate critique of Keynes's theory of effective demand. After correcting the many errors in Robertson's extensive letter, Keynes told Robertson that the development of his theory of effective demand, based on the Employment Function, was not contained in chapter 3 of the General Theory (GT;Keynes had sent Robertson a draft copy of what turned out to be the first 17 chapters of the GT) but in a later chapter. It is obvious from the title of chapter 20 of the GT "The Employment Function, "that chapter 20 of the GT is the chapter Keynes was referring to in his letter to Robertson.

Some 20 years later,in a series of articles published in the Economic Journal between 1954 and 1956,purporting to deal with the Keynes's theory of effective demand, Robertson simply repeated all of the previous errors contained in that letter of 20 years ago. Robertson also enlisted the aid of Harry Johnson to write a mathematical appendix of " What Keynes could have Meant", based only on pp.24-30 of the GT. Robertson even mentioned the 1935 exchange of letters between himself and Keynes but, in a deliberately dishonest manner, never made any mention of the point by point rebuttal made by Keynes, especially Keynes's point that chapter 3 (which contains pp.24-30) did not contain Keynes's worked out analysis.Unfortunately, Paul Davidson, following in the footsteps of his mentor, Sydney Weintraub, fell for Robertson's intellectual fraud hook,line, and sinker.

All of Paul Davidson's work on Keynes's theory of effective demand follows precisely from the articles published in 1954-1956. Davidson,like S.Weintraub, who passed all of the mathematical errors in Robertson's articles down to his son, E Roy Weintraub, has passed all of the Robertson-Johnson errors down to a host of Post Keynesians over a nearly 50 year period. The heart of this book is an interpretation of pp.24-30 of chapter 3 of the GT in which Davidson repeats the many confusions and errors that Robertson published in 1954-1956.

The exact same errors appear,for just one example, in Davidson's bungling of the clear differences between the D-Z model of chapters 3, 20, and 21 of the GT and the Y- multiplier model of chapter 10. Keynes spent an entire page explaining the differences to Robertson. However, Robertson,who could not pass a pre algebra course, knew that he could never understand. (For Davidson's error filled analysis,see Davidson,1994,pp.19-31,pp.164-174. Not a single shred of this analysis appears in chapter 20 of the GT).

The only reason for buying this book is to compare the errors of Davidson in 1994, who uses a different notation, with the errors of Robertson-Johnson, committed in 1954-1956(and by Robertson initially in 1935). Davidson's claim, that Post Keynesian economics represents a logical counterweight to neoclassical economics, is simply a bad joke, since Davidson's many errors have been discovered by mathematically trained neoclassical economists.

Luke Smith said...

I've been into post-Kenyesian thought since reading Henry C.K. Liu, post-Keynesian essayist, in college. I have read up on Hyman Minsky and kept current with some of Steve Keen's blog. Certainly PK is not the end-all-be-all.

Economists like Schumpter and Friedman, and historians like Rothbard, helped to define the dichotomy of the business cycle. Though Rothbard seems to have been against the notion of such Kondratief cycles or any economic wave for that matter - I still find his historical analysis useful.

As far as most of these wave theories, I think the integration of energy sources (coal, oil, nuclear, etc) and the engines the power (steam, internal combustion), have a great deal of influence.

The Arthurian said...

Thanks Tom. I read the review here & again at the link. The review points to a lot of closed doors, but opens none of them. It provides a whole genealogy of people it says are wrong, but provides not one example of what they have wrong.

In what the review offers as "just one example" the closest we come to finding out specifics is the "bungling" of something that is not quite identified.

I would have to read three years of the Economic Journal (1954-56) just to get started trying to figure out what the reviewer is talking about. But since I don't care that much, I have to say the review is not very good.

The Arthurian said...

Luke: "As far as most of these wave theories, I think the integration of energy sources (coal, oil, nuclear, etc) and the engines the power (steam, internal combustion), have a great deal of influence."

"Real" factors create the real limits for the economy, and energy must be right at the top of the list of real limits.

I'm pretty sure, however, that monetary problems -- monetary imbalances as the source of problems -- are very often overlooked: Because monetary problems work themselves out in the real economy, and monetary problems look like shortages of oil and Cabbage Patch dolls and toilet paper and such.

I never deny the effect of real limits. I only point out something that all such problems have in common: Money. Money is always involved in the transactions that give us oil and cabbage patch dolls and toilet paper.

If the problems are widespread, it must be because the "root" problem is of critical importance. The root could be oil, because oil is of critical importance. Or the root could be money -- in which case it might still *look* like the root is oil.

So, I look at money.

Jerry said...

I don't know if this helps or not, but what i think "ergodic" means is...

Say that you were to identify all of the "independent variables" in the economy. I don't know what - maybe, number of people, level of technology, prime rate, reserve ratio, etc ... there are probably a lot of them. Say there are 20.

Then you can say that the "state" of the economy is well (completely?) described by a "vector" of 20 numbers. the number of people is this, the prime rate is that, etc. That is the state of the economy.

Then you can imagine that there is some 20-dimensional space, and these variables are the coordinates. Somewhere in this space is a point, which is the current state of the economy. As time progresses, this point moves around into different positions (different states; different combinations of the 20 variables).

(If there were only 2 or 3 variables, this would be easier to think about -- the economy is a spec buzzing around in 3-space, like Descartes and his fly. But it's the same thing with 20.)

Anyway, when you have all of that set up ... you can identify chunks or regions of this 20 dimensional space that mean something. e.g. maybe "this corner over here is a good economy", or "this region over here has low tax rates", or "the top half here is good for the wealthy and bad for everybody else", or whatever.

(In thermodynamics, you might call the individual points "microstates", and the big regions "macrostates", and the volume of the region is like "the number of microstates (point) that correspond to that macrostate (e.g. 'a bad economy'."
The 20 dimensional space might be called a "phase space" or a "configuration space".)

Each of those regions has a "volume" in the space.

If the system is ergodic, that means that the system will wander around in that space, and will spend an even amount of time in each spot/position/set-of-coordinates/state, on average. If you were to wait long enough.

So, if you have mapped out the space and found that 60% of the volume corresponds to a bad economy, and 40% corresponds to a good economy, then if the system is ergodic, the economy will be bad 60% of the time and good 40% of the time.

If the system is not ergodic, then the amount of time that it spends in each region won't directly correspond to the volume of the region.

Maybe the tie-in to this other guy's definition is: if the system is "not ergodic", then maybe once it goes to a certain spot, it "falls down a hole" or something and can not get back out to explore the rest of the space. No matter how long you wait. It's not moving around randomly and evenly.

How this is useful in mathematics is this: if you understand the workings of the system well enough to derive, mathematically, that it is an ergodic system, then you have this whole suite of mathematical conclusions that are provably true, and you get "for free" a great understanding of many aspects of the system.

I'm not convinced that the economists are doing this. It kind of sounds like they are using e.g. "unit root" as a synonym for "i think it won't recover" (or whatever).
Whereas what I think they should be trying to say is, "I think that i understand how the economy works, and it is this: xyz. Then after doing 10 pages of algebra, I found that this xyz system has a unit root. So I predict that it will not recover because that is the mathematical behavior of systems that have unit roots. If it does recover, then in means my understanding, xyz, is incorrect."

I don't know if that's what they're doing or not? It sort of doesn't look like it to me, though, from my casual look. It looks like some kind of cargo cult thing. ( - i love feynman, he's such a brilliant dick! )