RE: Equilibrium & Market Clearing

[paul davidson <pdavidson@xxxxxxx>]

At 02:18 PM 02/10/2000 -0500, you wrote:
> Peter:  Sraffa's model is designed to expose a particular set of
> relationships, including the relationship between prices and distribution.
>  What we learned about the latter relationship from Production of
> Commodities has led to an important critique of the neoclassical
> distribution theory that grounds the orthodox claim that markets tend to
> generate full employment.  I don't see how you can blow off Sraffa unless
> you think that these insights are irrelevant to how a market economy
> functions.  The burden of proof would be on Post Keynesians to justify that
> assertion.


Justification follows next paragraph

> To say that Sraffa's model "is incapable of incorporating time" seems to me
> to be a non sequitur.  The model incorporates just what it needs to
> incorporate in order to investigate the relationships it wants to clarify.


Simple. The Sraffian model shows that the classical model is NOT a general
theory even on its own terms with all of its axioms. Sraffa demonstrated
that the classical theory had logical deficiencies.  Others then showed,
that as a general theory one could not rule out reswitching IN  A
NONMONETARY  production economy. Sraffa did a wonderful job to show that
the classical theory did not even work the way classical theorists
described it in the never-never world of classical economics.  It was
therefore a devastating negative blow to classical economics.  But unless
you want to be on the council of economic advisors for the President of
Never-Never land, the Sraffian model is of little use for practical problems.

Unless you believe that money is neutral, i.e., has no effect on production
and employment decisions, then the Sraffian model is relevant.

But in 1935 Keynes wrote an article in a Festscrift for Arthur
Spiethoff  entitled "A Monetary Theory of Production" [reprinted in CWK vol
13] in which he argued that in an entrepreneurial market oriented, money
using economy expectations and money were relevant in both the short run
and the long run.  And Keynes noted in this 1935 article this was the
theory he was working on and believed he was bringing forth.

So  Sraffa is irrelevant unless you are assuming that expected prices and
the availability of finance does NOT affect entrepreneurial
production  decisions on WHat? HOW? and FOR WHOM?  In other words, Sraffa
does not answer the important questions of employment of labor and
capital  and the resulting distribution of income in an entrepreneurial
economy where the future is not known in advance, and where therefore
sensible, but not "rational expectations"  matter --  and the availability
of finance matters.

> The characterization of the Sraffa's model as "identically a multi-period
> model in which all future production coefficients are known in the present
> and in which the resulting future prices are equally discounted by all
> participants" projects onto Sraffa views that he never held.  His book
> contains not a word about expectations or future prices or about how agents
> discount future prices.  And why should it?


If you think entrepreneurial expectations and finance do not affect hiring
decision and therefore income distribution then it should not matter that
Sraffa never introduces it into his model.  Once he does,  if the future is
nonergodic , then all the centers of gravity and persistent forcesthat
Sraffaians such as Garagnani  and Eatwell invoke will disappear.





>  To investigate such issues,
> one needs a different sort of model. That certainly does not mean that
> Sraffa's model is not useful for the investigation of the issues it does
> take up;


Yes -- but for practical policy problems of the real world in which we
live, a logically consistent Sraffa model provides no clues.


>  nor does it mean that th
> e issues it investigates are irrelevant,
> as some Post keynesians seem to think.  Must a model explain everything?


No just the important things like unemployment, inflation, financial market
volatility and its affect on real economic activity, economic growth,
should we follow a fixed or flexible exchange rate, a Taylor rule, a
currency board,  should emerging market economies dollarize,  how should we
save social security, etc.



> And anyway, Sraffa's model is stucturally identical to Leontief's, and just
> like the latter it can be modified to incorporate change over time.


But two wrongs do not make one right!

>  But of
> course, imlicit in Sraffa's method is the view that dynamic questions
> require a more historical and institutional approach (see Garegnani's 1984
> OEP paper).  This is a stance that has much in common with Post Keynesian
> views about method.


In the early 1980s Sylos-Labini asked me come to his Indtitute in Rome to
join him, Gragnani, Kregel, and Roncoglia in a private all day session to
see if we could compatibility between the Sraffian and Post Keynesian
system.  It was a futile ndertaking -- for my argument was that until one
introduced money, uncertainty, etc into a Sraffian mpdel there was as much
comparability between Keynes and Sraffa as between Keynes and the classical
system.  Garagnani's position  was money , etc was irrelevant because there
were always PERSISTENTcenters of gravity that the economy was moving
towards and in the long run money did not matter. [Notice the similarity to
current classical arguments bout reversion to the mean.]

> To be clear: I realize that not everyone finds the same sorts of
> theoretical questions "interesting", and I have no quarrel with efforts to
> understand how exprectations influence the performance of the economy.

>  (But I do think that very little in the way of substantive results have
> emerged from such efforts, for reasons that can be found in the Sraffian
> literature on method.)


Fine, if you will introduce money and uncertainty in a logically consistent
manner with rest of the Sraffian methodology then I will try to use a
Sraffian model. In my poor understanding of Sraffa's methodology -- I can
not see how these things can be introduced in a logically consistent
manner.  Please show me that I am wrong!




>  What I don't understand are the undercurrents of
> dismissiveness and even antagonism toward the Sraffian tradition that is
> characteristic of some of the Post Keynesian literature.  Minsky, for
> example, has come very close to arguing that Sraffa's economics is a
> variant of neoclassical theory (see his remarks in Bharadwaj & Schefold,
> Eds: Essays on Piero Sraffa).  It's all very strange to me.


It should not be surprising when Minsky's main contribution is the
financial fragility hypothesis that he finds the Sraffian system where
finance and speculation is not destabilizing to be similar to neoclassical
theory.  You may disagree with Hy, but you should be able to understand why
he is presenting this view.

Paul
Paul Davidson
Holly Chair of Excellence in Political Economy
Editor, JOURNAL OF POST KEYNESIAN ECONOMICS
Economics Department, 523 SMC
University of Tennessee
Knoxville, Tennessee 37996-0550
phone: (865)974-4221
fax: (865) 974-4601
email: pdavidson@xxxxxxx
web page: http://econ.bus.utk.edu/Davidson.html