Monday, February 26, 2007

Richard Chase On Reswitching

"Thus the reswitching anomaly, along with its theoretical developments and implications, has been placed in abeyance. And so it must be, for if this criticism were taken as being no less applicable to the real world than the theoretical, then it follows, as already noted, that orthodox economics is unable to make any reliable statements concerning the relationship of production to the various input markets. That is, the neoclassical vision of a market-coordinated production system, along with derivative growth and distribution theories, are all invalidated. As a consequence, the nature of the entire traditional circular flow conception is called into question...

...It is one thing to say that this conception of indirect economic management does not satisfactorily achieve its goals because of the existence of such real-world problems as bottlenecks, power, premature inflation, inflationary expectations, random shocks, ratchet and spillover effects, and the like. In such situations, an economically coherent and consistent market-based system of production and distribution is still assumed to exist, though it is overlaid with political, institutional, and psychological factors that affect economic adjustments and performance. The basic strategy, in this case, would be to maintain the general neoclassical-synthetic emphasis on fiscal and monetary management (with perhaps somewhat greater stress on the monetary tool, if the monetarists were to have their way), and supplement these tools with finely targeted direct and specific devices - for example, stricter antitrust enforcement, more sharply focused incentive (and disincentive) taxes, expanded job training and subsidization programs - so as to allow and encourage the effective functioning of centerpiece fiscal and monetary devices.

It is quite another thing to argue that key markets in the system, particularly those in the resource or input sector, do not possess the fundamental economic characteristics necessary to the orderly systematic functioning that is postulated by mainstream theory..." -- Richard X. Chase, "Production Theory," in A Guide to Post-Keynesian Economics, (edited by Alfred S. Eichner), M. E. Sharpe, 1978, p. 79-80

12 comments:

Anonymous said...

To make an understatement, there are quite a few missing logical steps between claiming that capital reswitching exists and claiming that the neoclassical vision of markets-coordinated production system failed.

In fact, the claim is unjustified bullshit. That is, there is no necessary logical connection between the existence of reswitching and the coordonation or lack thereof of markets.

I hope that Post-Keynsian economics does have higher standards of argument that those exhibited in the claims made by Richard Chase.

Alex

(Yes, I know that I did not make an argument either -- but there is first a need to specify a shadow of an argument for Richard Chase's position before one can even begin to argue)

Robert Vienneau said...

I don’t know what “capital reswitching” is.

Obviously, the Chase quotation is from a longer article, published in a semi-popular publication. I don't know why, unless he has read that article, Alex would think he has anything to say about whether Chase fairly summarizes a view developed over decades in thousands of articles.

Anonymous said...

So, what is the argument?

Alex

Robert Vienneau said...

If Alex were to reply, I would have expected an apology for his previous deeply non-serious, near-libelous, and ignorant balderdash. He must have met a bug in blogger that stripped the apology out of his comment.

Anonymous said...

Well, I guess no argument now either.

The notion that there is something I should apologize for --to whom? to Richard Chase? -- is just strange; as is your answer to my request for an argument.

Your insults are amuzing --if somewhat unflatering towards you-- but they only lower the expectations about what one will read in your blog. Which is fine.

Alex

Blissex said...

«real-world problems as bottlenecks, power, premature inflation, inflationary expectations, random shocks, ratchet and spillover effects, and the like. In such situations, an economically coherent and consistent market-based system of production and distribution is still assumed to exist,»

The latter assumption is quite absurd, because of the ''second best'' theorem anyhow.

Reswitching just explodes the internal contradictions of the theory, the ''second best'' theorem is perfectly sufficient to do that to its relevance...

Blissex said...

«few missing logical steps between claiming that capital reswitching exists and claiming that the neoclassical vision of markets-coordinated production system failed.»

The above and this:

«no necessary logical connection between the existence of reswitching and the coordonation or lack thereof of markets»

are subtly and dishonestly non-equivalent, because the «neoclassical vision of markets-coordinated production systems» and «the coordonation or lack thereof of markets» are rather different concepts...

The «classical vision» is that not just markets are coordinated, but that markets are coordinated in the best possible way, that is they achieve maximum value added, and in a way that also makes the distribution of income Pareto optimal.

That the resulting distribution of income is both optimal and determined solely by marginal productivity is the central verity of neoclassical economics.

Reswitching quite obviously is incompatible with that (it is so shockingly obvious that there is no need of an argument here); reswitching is not all all incompatible with the notion of ''coordinated markets'', indeed in Sraffian economics markets are well coordinated too, and precisely this, in particular the coordination between the ''money'' and the ''capital goods'' markets, that gives rise to reswitching (and tyhat coordination is wilfully missing in the neoclassical models).

BTW, most/all neoclassical models have only *sale* markets, because they assume an initial stock of goods/services that gets auctioned once and for all time (with laughable tricks as lucasian expectations).

It takes Sraffian (or other non-neoclassical) modeling to account for production, which is inherently inter-temporal, and then reswitching happens because interest rates create inter-temporal links...

Anonymous said...

Blissex,

I assure you that any difference of meaning between the quotes that you selected from my reply are entirely imaginary as far as I’m concerned – and as a result any perceived dishonesty is entirely in your mind. In both sentences I was talking about the same thing, about the coordination of markets where to an approximation there is utility maximization, profit maximization and all markets clear. This is what I take to be the neoclassical vision of market coordinated production (and consumption) – or rather, that departures from these kind of states are bound to be temporary and short lived and there will be equilibrating tendencies to bring about such states. But for our present purpose we can ignore this qualification.

You write:

“The «classical vision» --do you mean neoclassical vision? Alex---

is that not just markets are coordinated, but that markets are coordinated in the best possible way, that is they achieve maximum value added, and in a way that also makes the distribution of income Pareto optimal.”

I’m not sure what “maximum value added means” but yes, I think that the neoclassical vision does imply that the distribution of income is Pareto Optimal (again, I’ll agree to ignore any qualifications).

But then you go astray:

“That the resulting distribution of income is both optimal and determined solely by marginal productivity is the central verity of neoclassical economics.”

It was already discussed on this blog that the claim that the distribution of income is determined solely by marginal productivity was never part of the neoclassical vision. More precisely, while John Bates Clark was clearly advocating this mistaken theory, and sometimes the claim gets repeated today, neither the early neoclassical theorists nor more modern rigorous neoclassical theorists advocated it.
You can see more on this on

http://cepa.newschool.edu/het/essays/margrev/distrib.htm

You should look at the quotes from part (B) section (ii) on the above webpage. (You’ll have to read or skip a few paragraphs until you get to the relevant quotes)

So on this point you’re mistaken.

Then you write:

“Reswitching quite obviously is incompatible with that (it is so shockingly obvious that there is no need of an argument here)”

What is “that”? If “that” is the complete determination of income distribution by marginal productivity then you’re quite right that reswitching in incompatible with “that”.
But as you saw above, the claim is not something that forms a part of the neoclassical vision.

If by “that” you mean Pareto optimality of income distribution – then by all means, do show me how reswitching invalidates the claims to Pareto Optimality of the distribution.


The Richard Chase argument –if there is any—remains a mystery to me.

I know that there were more recent attempts by B. Schefold and P. Garegnani to show that capital reversing implies some result about the instability of General Equilibrium models – but they don’t seem to have convinced the economic profession. And at any rate, the Richard Chase claim is much earlier than any of the attempts I mentioned.

Also, the fact that General Equilibrium can be criticized on some ground or another is really not the same with what Richard Chase was claiming.

So, how in the world does reswitching invalidate the neoclassical vision of market-coordinated production?

Alex

Blissex said...

But perhaps we are takling about different things, but reswitching shows that the distribution of income depends on the rate of interest, which cannot be determined endogenously. So there is no unique market clearing equilibrium and distribution of income, and this blows Pareto optimality.

Alex, if it were possible to prove Pareto optimality without all the insane assumptions to eliminate money and time in Debreu/Lucas/... so as to eliminate reswitching, it would have been done...

h.e. said...

But perhaps we are takling about different things, but reswitching shows that the distribution of income depends on the rate of interest, which cannot be determined endogenously. So there is no unique market clearing equilibrium and distribution of income, and this blows Pareto optimality.

Why? It seems to me that a model with multiple equilibria can have more than one equilibrium that is Pareto efficient.

Blissex said...

«a model with multiple equilibria can have more than one equilibrium that is Pareto efficient.»

Sure, it might happen, but the central verity of economics is that *all* (actually the single one :->) equilibria are Pareto optimal.

In other words, that ''capitalism'' necessarily results in a just distribution of income, and that ''leverage'' plays no role in it.

Sraffa's argument may be reckoned to be that the exogenous interest rate determines the distributions of income, and while some of those *might* be Pareto-optimal, that is hardly relevant.

Hey, Sraffa's main thrust is that ''capital'' and its price are not so well defined in neoclassical economics, and if one fantasizes that ''capital'' is a ''factor of production'' (whatever those two terms mean!) then it becomes pretty difficult to have a nice ''best of all possible worlds''neoclassical theory of the distribution of income.

Anyhow, again, this aspect of Sraffa's work is just an amusing critique of the inner workings of neoclassical theory, because both the second-best theorem and the absurd aggregation conditions imply that neoclassical theory is pretty much irrelevant anyhow...

As to that, I think that Pareto-optimality is a red herring too, even if it serves a useful propaganda role in neoclassical theory.

h.e. said...

"In other words, that ''capitalism'' necessarily results in a just distribution of income, and that ''leverage'' plays no role in it."

Who equates "Pareto Optimal" with "just"? The argument for why markets might be Pareto Optimal is fairly simple...if there are unexploited gains to trade people will eventually find them. Once all gains to trade have been exhausted, no further welfare gains are possible. There is no claim that this end-state is just.

Now whether the neoclassical model usefully represents capitalism is probably an article of faith, since I doubt the model is even intended to be tested. I have mixed feelings about the potential propaganda function of the welfare theorems. On the one hand, I'm inclined to think that Pareto efficiency is a secondary goal, so the claim that (idealized) markets are efficient doesn't form much of intellectual (or moral) defense for a particular form of social organization. However, I realize that there are arguments formed against policies with regards to distorting the workings of the market, so clearly there is the potential.

For the broader issue of 'leverage' (which I interpret as referring to power relationships, institutions and the like), discussions about the "internal consistencies" of the neoclassical model and the Sraffian critique seem irrelevant. One could simply assert that the model is clearly not a representation of capitalist markets and disregard it.