neo- vs. post-Keynesians, etc.

[Jim Devine <JNDF%LMUACAD.BITNET@xxxxxxxxxxxxxxxxx>]

On Fri, 28 Jan 1994, Paul Davidson said (with some typos corrected
and my comments added, labelled "JD"):
PD: I was not trying to be patronizing. I really am trying to see if
I can make you understand my position. If you want to defend
'quantity constraint' models, then you must explain what is the
cause of quantity constraint. Most explain quantity constraint
systems (e.g., Tobin, Clower, New Keynesians) as due to dQ/dT >
dP/dT. All quantity constraint people that I know admit that if the
price speed of response was instantaneous, full employment would
always occur -- there would be not quantity restraint.
(Incidentally, this is not true in Keynes's system -- but it is true
in the classical system where additional restrictive axioms are
introduced to assure this outcome.) Do you have some logical
mechanism to explain quantity demand restraint that does not involve
this faster price speed of reaction?

JD: I, for one, see nothing wrong with the assumption that prices do
not adjust instantaneously. If Keynes didn't see this
less-than-instantaneous adjustment of prices as necessary to
understanding the phenomenon of involuntary unemployment, that's
great. But the real world as I see it does not have instantaneous
adjustment of prices, so assumptions of non-instantaneous adjustment
are o.k. Rather, what happens is a combination of price adjustment
and quantity adjustment, a phenomenon that mucks up the Walrasian
story in a very simple but fundamental way. Any factor that mucks up
the Walrasian story is important, as a strike against the orthodoxy
and in favor of reality.

To my mind, the quantity constraints occur because of slow-downs in
the rate of accumulation due to various crisis tendencies (which
tendency dominates depends on the historical era). In our era, the
accumulation process has been politicized, making things
complicated, but not changing the nature of capitalism
fundamentally. Now it's possible that one result of quantity
constraints would be falling prices is recession. But as I. Fisher,
J.M. Keynes, J. Tobin, and others (including, I believe, Professor
Davidson) point out, deflation is likely to make things worse rather
than better under realistic assumptions.

(BTW, am I right to assume that the issue is not dQ/dt > dP/dt but
rather dP/dt > 0? After all, if prices do not adjust, then
quantities must adjust, leading to pecuniary externalities linking
markets in a multiplier-type process.)

PD: In my quoting of Leijonhufvud and Hahn I was not appealing to
authority. I was trying to show that the originator of this speed of
response -- quantity restraint analysis, as well as a logically
thoughtful GE theorist, recognized that you can not hang the
quantity constained apporach on Keynes. This is simply a question of
the History of Economic Thought.

JD: I found the references to Leijonhufvud and Hahn to be
interesting, especially in that I thought the former's
interpretation of Keynes, though useful at times, was rather off
base. However, the Authority that I was referring to was none other
than J.M. Keynes. Sometimes I get the impression is that you,
Professor Davidson, think that everything that Keynes said (at least
in the GT, the article by the same name, and the Treatise on Money)
was true and that the main source of truth is Keynes. Material about
nonegodicity is important because it complements Keynes' truth,
while the distinction between effective and notional demands is not
important, because it does not complement Keynes' truth. I hope that
my superficial impression is wrong.

I find Keynes to be very interesting and suggestive but sometimes
extremely vague.  And I see nothing wrong with developing new ideas
or old ideas that do not stem from Keynes (or from whomever). Put
another way, I feel that it's useful to separate the History of
Economic Thought from economics _per se_, and then study the
interrelations between these fields.

PD: I appreciate your desire on empirical references 'the Facts' as
the basis of the argument. Unfortunately, in macroeconomics, the
facts are not independent of the theory. The 'facts' reported in the
NIPA, for example, depends on the theory of aggregate output you
use, not the other way around. ... when is a money payment a
consumption expenditure and when it is an investment expenditure
depends in large part on your theory of consumption vs. your theory
of investment...

JD: I totally agree with this point. I think it's sad that most
economists take the NIPA as gospel. However, my point was that all
economists should be open to what old Lenin called 'empirico-
criticism.'  (I bet no one expected his name here on pkt; I'm not a
fan of his either.) We have to get beyond pretentions of ideological
purity which drive us to read and re-read the Holy Books (whatever
they are).  After all, the Holy Books are subject to as much
interpretation as the 'empirical facts.'

PD: ...I find your desire to create a 'larger tapestry' by weaving
bits and pieces of Marxian, New Keynesian, Post Keynesian, etc.
strands 'to fill the gaps' an artistic endeavor -- that may provide
you with a feeling of great beauty in my view, one which cannot hang
together because these strands are spun on logically incompatible
looms.

JD: The tapestry metaphor is not meant to be aesthetic (though I
think someone could make money selling tapestries to apartment-
dwellers who live within thin walls :-) ). It's rather a reference
to the concept of a 'totality,' in which different parts fit
together into a whole that is different from (and greater than) the
sum of the parts and the nature of this whole affects the character
of the parts. The distinction between notional and effective demands
for labor-power (or for whatever) can have a different character and
different implications in the totality of an alternative theory than
in the _ad hoc_ Walrasianism of Clower _et al_. Wool threads can be
useful both in tapestries and in clothing. (Of course, it's possible
that Clower _et al_ have produced nothing but polyester threads that
are dreadful *anywhere*, but I'm not convinced.)

That may be too abstract, so I'll change gears: my methodology
involves eclecticism of interests, but also an attempt to avoid
eclecticism in results, by synthesizing. This process also involves
a constant dialectic (if you'll excuse a dirty word) between
empirical-inductivist research and logical-deductive analysis. The
'gaps' that I'm trying to fill are in Marx's theory of crisis, and
in those of his followers. The essentially empirical proposition
that prices do not adjust instantaneously to clear markets (along
with its New Keynesian rationalizations) and the theoretical
apparatus of the distinction between notional and effective demands
may fill some gaps.

PD: When you speak of 'for limited purposes' you can make use of the
assumption that the MPP of labor curve is the demand for labor
curve, while 'you agree...that the marginal physical product of
labor curve is not a demand curve' don't you have the slightest
twinge of conscience that you are not being logically consistent
with your students.

JD: of course I do. Abstraction from what I believe to be elements
of empirical reality is painful. But unfortunately, abstraction is
necessary to understand the complexity of reality. It is also needed
in teaching.

PD: If the MPPL curve is sometimes used 'as simplification of
empirical reality', then other times it is not useful as a
simplification of empirical reality. Have you got any general
principle that we can all use to decide when it is an applicable
simplification of a REAL WORLD MONETARY ECONOMY?

JD: absolutely, the quantity and quality of one's assumptions --
i.e., abstractions -- depend on exactly what questions one wants to
answer. For example, different assumptions are needed to discuss (1)
financial crises and (2) the economic impact of the minimum wage. Of
course, it is absolutely crucial to keep the nature of one's
assumptions (abstractions) in mind rather than taking them for
granted.

BTW, I'd like to stress that my view of assumptions is different
from the common one of the deductive-axiomatic approach. While the
latter sees an assumption as an addition to a pre-existing set of
assumptions (axioms), I see an assumption as a simplification of
perceived empirical reality.

PD: If not, then you must have a whole tool kit of specific case
explanations of specific episodes in history...

JD: I try to have a general and abstract view (theory) that explains
as much as possible concerning the laws of motion of the capitalist
mode of production. However, I realize that since different episodes
in history not only have similarities but differences, we cannot
rely simply of the general theory. We also need specific
institutional and historical information (as with the Social
Structures of Accumulation school or the French Regulation school,
though I have my differences with them).

PD: You indicate you that Patinkin's analysis helps clarify.... But
Patinkin draws a MPPL curve and a supply curve of labor in real
terms and then discusses situations where either the buyers of labor
or the sellers of labor (or both) are not on their respective demand
and supply curves. In fact, in Patinkin's book it possible for labor
market participants to be at any possible set of coordinates in the
positive quadrant. But if they can be anywhere --- especially
anywhere off the real demand and supply curves, what do we mean by
the forces of demand and supply in the labor market?

JD: We have to answer this by going outside of Patinkin. As any good
news reporter knows, it's a mistake to rely too much on one source.

PD: Finally you want to talk about class conflict in terms of
workers vs. capitalists rather than workers vs. workers, or
capitalists vs, rentiers....

JD: For me, it's not "rather than" but instead "in addition to."
Otherwise, I agree totally with your empirical points about Reagan,
Thatcher, _et al_.   BTW, for the interested, I have a long ms.
(available via snail-mail) that tries to contrast the liberal and
the Marxian viewpoints, starting with a liberal view and using
mostly neoclassical or Keynesian language. I see class conflict in
terms of the existence of two distinct and, to a large extent,
mutually-exclusive collective goods. In the simplest possible terms,
each class has to deal with free riders, people who work against the
class interest (i.e., horizontal intra-class conflict).

PD: Which theorists are 'unaware that we live in a dynamic
economy?'...

JD: I wasn't as clear as I should be. To my mind, Keynesians are
usually not aware that the development of capitalism is a
dynamic-disequilibrium process, with not just "fundamental
uncertainty" (nonergodicity), but also what might be called "surplus
uncertainty." The latter is uncertainty that arises not from the
nature of things but from the fact that capitalism is more than
simply a market economy (what Marx termed "simple commodity
production"), but also an economy in which capitalists are pushed to
"accumulate or die" (and also one with an undercurrent of class
tension, even when class conflict is not overt). Thus, Barkley
Rosser's discussion of uncertainty that is the implication of the
way idealized models of the economy work, rather than simply an
assumption, is very interesting.

Further, capitalism should be seen as a system that has some
internal tendencies to self-destruct, setting the stage for some new
non-capitalist economic system. Unfortunately, the tendencies toward
creating something (e.g., socialism) to replace capitalism are
weaker, so we see situations like the Great Depression, where
capitalism falls apart but survives.

sincerely,

Jim Devine   BITNET: jndf@lmuacad    INTERNET: jdevine@xxxxxxxxxxxxxxx
Econ. Dept., Loyola Marymount Univ., Los Angeles, CA 90045-2699 USA
310/338-2948 (off); 310/202-6546 (hm); FAX: 310/338-1950