Response to Isaac's inquiry for money wage theory

[Paul Davidson <PB108928%UTKVM1.BITNET@xxxxxxxxxxxxxxxxx>]

     I want to thank Alan Isaac for having read Weintraub's chapter
before attempting to further pursue his demand for a theory of
nominal wage determination. Having steeped himself in Weintraub's
analytical framework, I believe that Isaac and I can have a logical
dialogue rather than merely engage in semantic obfuscation. Isaac
should now be able to see better what I am trying to state and
there is less likelihood that we misinterpret each other's words.
I hope that the following will set out an agreement as to what is
meant by a "theory of nominal wage determination" and why Isaac has
such difficulty in finding it in the GT -- and even fails to see
it in Weintraub where it is explicit.
     Weintraub provides an analytical framework based on
Marshallian microeconomics that are the foundations of Keynes's
aggregate supply and demand functions. This framework makes
the following explicit:
     The necessary and sufficient conditions for a theory of the
determination of a (flexible) nominal wage rate that will always
clear the labor market (for any specified PARAMETERS THAT UNDERLIE
GENERALIZED DEMAND AND SUPPLY FUNCTIONS). On p. 245 of the GT
Keynes explicitly lists the specified parameters to include "The
existing skill and quantity of available labour, the existing
quantity and quality of available equipment, the existing
technique, the degree of competition, the tastes and habits of the
consumer, the disutilities of different intensities of labour and
the activities of superstition and organisation, as well as the
social structure including the forces, other than the variables set
forth below, which determine the distribution of income.  This does
not mean that we assume these factors to be constant; but that, in
this place and context, we are not considering or taking into
account the effects and consequences of changes in them".
     Just as Keynes's analysis provides the basis for showing why
Say's Law requires certain necessary and sufficient conditions
(which could be obtained under classical restrictive axioms) to
shape the aggregate demand and supply curves so that there is no
obstacle to full employment, so Weintraub's analytical framework
provides the necessary framework for showing the logical conditions
for producing a theory of nominal (flexible) wage determination
(given generalized aggregate demand and supply functions) that
would always clears the labor market. (And, as Keynes hinted and
I have demonstrated in MONEY AND THE REAL WORLD, it can be shown
that these necessary and sufficient conditions for a flexible wage
to clear the labor market can hold in a world with a money
"numeraire" but where there is NO asset that possesses, what Keynes
called, the essential properties of money.)
     In other words, Weintraub's analysis permits the analyst to
recognize the necessary and sufficient conditions where flexible
nominal wages will not necessary clear the labor market given a
generalized supply and demand curves. Moreover, Weintraub's
analysis permits one to completely specify how changes in nominal
wages can effect the position of the generalized aggregate supply
and demand functions when they are specified in nominal terms (This
is what I believe Isaac means by "nominal wage dynamics").
     In other words, WEINTRAUB'S ANALYSIS PERMITS ONE TO
ANALYTICALLY DISCOVER THE NECESSARY AND SUFFICIENT CONDITIONS FOR
A THEORY  OF A (flexible ) NOMINAL WAGE RATE THAT ASSURES LABOR
MARKET CLEARING!! A THEORY OF NOMINAL WAGE DETERMINATION WHICH
ASSURES MARKET CLEARING  IS WHAT I BELIEVE ALAN ISAAC IS REQUESTING
WHEN HE KEEPS ASKING FOR A "THEORY OF NOMINAL WAGE DETERMINATION".
(If Isaac does not mean a labor-market clearing money wage
determination theory then what does he mean? In the theory of price
determination in imperfect competition, for example, prices clear
the market in the sense that all that entrepreneurs are willing to
produce at a given (profit maximizing) price are sold, i.e.,
entrepreneur's in their production decision are never off their
short-run flow-supply curve -- as I insisted in my dialogue with
Skott. {Perhaps now some readers will understand why I was so
insistent.})
     Since Keynes's GT argued that one could not solve the
unemployment problem by assuming a perfectly flexible money (or
real) wage, and/or perfectly flexible prices, Weintraub's analysis
makes it obvious why, in Alan Isaac's words the "GT does not
contain a theory of nominal wages". The GT does not contain a
theory of market clearing nominal wage determination -- given
generalized aggregate supply and demand functions.
     But in Chapter 19 Keynes provided the clues for the economic
analysis necessary to let Weintraub see how such a theory had to
be constructed.
     In effect, Weintraub analytical framework permits the analyst
to demonstrate why Keynes believed that, as a general case, a
flexible nominal wage system "is the right and proper adjunct of
a system which on the whole is one of laissez-faire, is the
opposite of the truth" (Keynes, p. 269). This is spelled out in
greater clarity in 1964 in Davidson and Smolensky's AGGREGATE
DEMAND AND SUPPLY ANALYSIS, chapter 11,  and even better in my POST
KEYNESIAN MACROECONOMIC THEORY. (Hopefully I have sharpened my
powers of exposition and insight over the years.)
     If you have learned Weintraub's analytical framework, then it
is possible to work out if and how the point of effective demand
changes as the nominal wage changes (i.e., what Isaac calls "if
effective demand depends on nominal wage dynamics"). Weintraub's
analysis of the three labor market demand curves: The downward
sloping classical demand curve, the vertical Keynes demand curve,
and the upward sloping underconsumptionist demand curve, indicates
the necessary and sufficient conditions for achieving any one of
these three "dynamic paths".
     Based on Weintraub's pioneering analysis and my further
thinking about the problem has permitted me to develop a chapter
that deals in greater depth with these various dynamic paths in
Chapter 11 of my POST KEYNESIAN MACROECONOMIC THEORY BOOK. It is
this chapter that Goodhart cites (as part of the quoted regarding
Goodhart's splendid reaction in my earlier posting) as "where he
{Davidson} explains the core of his differences with mainstream
economists... Davidson contrasts an economy which does so
automatically revert to an equilibrium, an 'ergodic' system, with
his own contention that actual economies are 'nonergodic' and
...THIS IS THE KEY ELEMENT IN KEYNES' OWN APPROACH...Paul makes a
good case that the efficacy of wage/price flexibility in restoring
full employment has been too uncritically accepted. As one of those
whose analysis and teaching Paul is critising, I found the
experience stimulating and provoking". (Emphasis added)
     Only after one has developed the logical conditions for a
theory of (flexible) nominal (or real) wage determination that
automatically clears labor markets, can one engage in a meaningful
logical dialogue as to what the conditions for such a theory
require in terms of underlying axioms. etc. And the same is true
whether we are talking about wage "determination" or wage" dynamics
-- we must explicitly lay out the necessary and sufficient
conditions for any wage "dynamic" paths. This is the 'difficult in
the trenches' work of any good theorist.
     On the other hand, anyone can rush to write a "dynamical"
equation(s) that describe what a real world dynamic wage path looks
like, i.e., that fits the empirical data points of a specific
period of calendar history! (That is what Devine finds so
attractive about Soskice and Carlin's "empirical approach".  But
the problem is if we have any time series of observations, there
are always higher degrees of polynomials (and/or larger
multivariable relationships) that better "fit" the empirical
relationship than equations lower degrees (or less multivariables)
-- until we get a polynomial that has as many degrees as empirical
turning points (or until we get zero degrees of freedom). Given the
dynamic equational relationship that "fits", anyone can "invent"
a narrative that is consistent with the polynomial (multivariate)
that he/she thinks best describes the empirical data points.  But
one does not have a general theory in terms of axioms and the
related necessary and sufficient conditions -- which, in my view,
is the only way to go to provide a logical analytical framework for
"theory" development; questions of existence, stability, uniqueness
or multiple equilibria, etc.
          I hope that the above eliminates one of Isaac's doubt
about my claims. Weintraub has presented a theory of nominal wage
determination that is fully compatible with a general theory -- and
has shown why, although Weintraub in the 1950s did not stress the
ergodic vs. nonergodic dichotomy, the classical theory where
flexible wages can be determined that clear the labor market place
requires special axioms and  do not match "the facts of
experience".
     (Please do not fault either Keynes or Weintraub for not having
spelt out in infinite detail the "full" general theory. After all
look how many mainstream economists'-years of work it has taken
orthodox economists to spin out in greater detail their theories.
If Keynes and Weintraub had solved everything, there would have
been nothing left for me -- or other Post keynesians to develop.
But they were pioneers  -- who like all pioneers -- break a new
trail which leads their followers to a better land.  If the trail
they left has not been completely paved  but still have some rough
ruts, let us seize the opportunity to smooth out these difficulties
-- rather than condemning them for not providing from the outset
a six lane limited access interstate complete highway network.)
     Two, Alan Isaac sees in Weintraub's chapter Only"i. an
exogenous nominal wage or ii. a simple Phillips curve story." As
I have tried to indicate above, Isaac can not see the forest for
the trees. He cannot see that Weintraub has specifically spelled
out the necessary and sufficient conditions for a theory of nominal
wages that clears the market for any given effective demand (Just
as Keynes, on pp. 25-26 illustrates the conditions necessary for
Say's Law.) The fact that Weintraub then goes on to discuss
specific cases such as "the simplest, and reasonably realistic ...
labor supply curve... as perfectly elastic" (Weintraub, p. 125 as
quoted by Isaac) and the Phillips curve (that Weintraub was writing
the manuscript at the time when the Phillip's had just published
his findings and still deals with an idea that was just taking off
in the economics literature shows how up to date he was) just shows
how his general analytical approach can be used to analyze simple
as well as more complex cases.
     Building on Weintraub's framework, I have tried to explain
why, in a nonergodic system, entrepreneurs prefer to deal in a
labor market place where nominal wages are fixed for a period of
time ONCE THEY HIRE WORKERS, rather than renegoitaite money wages
every minute (on the spot)! Entrepreneurs use forward money
contracts to deal with ignorance about the future. Forward nominal
contracts permit entrepreneurs to control net cash flows in a world
in which receipts and expenditures can not otherwise be predicted
and the money-wage contract (in a nonslave economy) is the most
ubiquitous of all forward contracts.  With the use of nominal
contracts to organize production and exchange activities, liquidity
becomes essential.
     With the abolition of slavery, labor can only be purchased on
forward nominal contracts.  The existence of forward contracts for
labor, by definition, requires a wage specified in advance for the
duration of the forward contract, for the life of a contract. (Of
course, in a slave economy, there is never any involuntary
unemployment of slaves -- even though entrepreneurs can control the
real wages of their slaves. Under slavery, spot contracts for the
purchase of labor  -- human capital -- are just as possible as for
"real" capital.)  Are the existence of forward contracts for labor
or for goods, or for the hiring of capital services (HERTZ rent-
a-capital good) to be dismissed as an"unsolved problem" in
economics?
     Neither Weintraub or I  are guilty of what Isaac claims
"ignoring the problem" of nominal wage determination. I certainly
have not ignored this problem in my POST KEYNESIAN MACROECONOMIC
THEORY. (The aforecited quote from Goodhart indicates that he has
recognized that I have paid attention to this problem.) Once we
have a general theory of nominal wages on the table we can lay out
the necessary and sufficient conditions to "explain" any specific
case whether they involve what Isaac refers to as  efficiency wages
or conflict theory or other case. What you will find in my POST
KEYNESIAN MACROECONOMIC THEORY is an analytical framework which
explains why -- even in an open economy context -- we cannot rely
on the law of comparative advantage to provide a real world case
of fixed relative efficiency wages -- even if there are freely
flexible exchange rates (the open economy equivalent of
instantaneously flexible wages and prices). Moreover, if a relative
efficiency wage structure is socially desireable, then once one
knows the necessary and sufficiency conditions for achieving such
a structure, one can develop a policy-- even in an open economy -
- which achieves this result WHILE STILL ASSURING GLOBAL FULL
EMPLOYMENT CONDITIONS!
     Thanks again Alan Isaac for allowing me to discuss these
issues in a manner that I hope is more meaningful to you now that
you have read Weintraub.