Re: Piorot on Madrick - Winslow's blast

["John M. Legge" <jlegge@xxxxxxxxxxxxxx>]

Two issues have been conflated.  One is the nature of empirical evidence:
Galbraith used employment and wage data collected by US government agencies
and subjected it to statistical analysis.  The data was consistent with the
hypothesis that wages varied according the location of the employer in the
K/C/S cluster scenario; and in the case of the C group, with the degree of
inter-worker cooperation required.  Within the conventional use of the
English language, the wage differentials and the relative uniformity of
wages within  most (not all) groups can be explained by employers and
managers adjusting wages to achieve the best outcome possible as they see
it.  "Efficiency wage" arguments as far as I am aware rely on a narrower
research base and fail to account for differences between industries.

Arguments complementary to Galbraith's come from Manove (cited in my
previous post) who suggested a "responsibility" premium in wages and offered
a formula which gives results consistent with the observed data; and Dixit,
who showed that the existence of training and recruitment costs made it
reasonable for employers to retain and pay workers even when there was
insufficient work to keep them fully occupied, as long as there was some
prospect of demand recovering.

Since I didn't use the term "rational maximisation" I feel no need to defend
it.

Complexity theory states that agents making choices are constrained in that
decisions must be made in a finite time using finite information gathering
and processing resources.  It then demonstrates that that the trajectory of
a system comprised of such agents in the general case (and in practice, in
all cases) will not be the same as it would if choices were made by agents
with hyper-infinite information gathering and processing capability.

The assumption of orthodox economics that the economy is at, or is moving
towards, a unique, optimal general equilibrium is shown, by complexity
theory, to include EITHER the prior assumption that all economic agents have
hyper-infinite information gathering and processing capability OR a
benevolent neoliberal God (equipped with hyper-hyper information gathering
and processing capability) is guiding the decisions of economic agents to as
to bring about such an optimal equilibrium.  The first leg of this argument
summarises the orthodox economic position; the second, the Austrian and US
televangelist view.  PKers, whether they accept complexity theory or don't,
reject the possibility that agents have the information gathering and
processing capability or divine guidance required to make accurate long term
predictions about the future.

Within the environment in which human agents exist I see no problem in the
argument that they make the best decisions that they can using the
information and computing resources available to them with a view to
progressing their objectives.  The evidence that the majority of consumers
maximise "utility" is zilch, given that there is no consistent definition of
utility or way of measuring it.  The evidence that managers attempt to
maximise profits is also extremely weak; with little supporting evidence and
quite a lot of directly contradictory evidence.

The evidence I have seen collected by Hunt and others is that people as
consumers or managers attempt to keep up with their peers: in Hunt's term,
to achieve "superior", not maximal performance.  This is consistent with
complexity theory: it is a relatively easy matter to compare the performance
of one business with that of another, or the consumption patterns of one
household against another, and does not require hyper-infinite information
gathering and processing capabilities.

In a complex world view abrupt discontinuities are unusual, although not
impossible, and so a business (i.e. a firm or reporting unit within a firm)
that made a profit of x last period will probably report a profit of x plus
or minus d next period, x plus or minus 2d the following period, and plus or
minus d*2^(n-1) in the nth period from now. d/x is of the order of 2^7 when
the unit period is one year, i.e. the uncertainty exceeds the expected value
after seven or so years.

I suppose that Ted's fundamental problem is that he sees complexity theory
as a (flawed) tool for predicting the future, where it is in fact a way of
quantifying the unpredictability of the future,

JML







>
> Ted Winslow blasted:
>
>
> How do you reach the conclusion that the hypothesis that all economic
> behaviour is the outcome of "rational maximization" is
> "solidly grounded
> in empirical evidence"?
>
> There is first of all the problem that the "complexity
> theory" which you
> take as the realistic elaboration of this hypothesis
> constructs reality
> as "a fully deterministic system" having no logical space for
> the notion
> of future outcomes being affected by present choices i.e. no logical
> space for the idea of "rational maximization."
> Self-contradictory ideas
> can't be "solidly grounded in empirical evidence," can they?
>
> I gather that what you mean by empirical grounding is that some
> empirical phenomenon, e.g.  a uniform wage, liquidity preference, the
> domination of current profits in the formation of expectations about
> future profits, etc. can be shown to be consistent with the
> hypothesis
> that the phenomenon in question, e.g. this method of expectation
> formation, is the outcome of "rational maximization" elaborated as
> complexity theory, e.g. since using current profits to predict future
> profits (though not, I gather, future profits in other than the short
> run) can be shown to be "rational" for agents who believe reasonably
> that complexity theory provides the true account of the
> reality they are
> attempting to understand and predict and since agents are
> using current
> profits to predict future profits, it therefore follows that the
> particular economic reality in question is accurately described by
> complexity theory and that agents know this and base their
> expectations
> on this knowledge.
>
> Have you empirically investigated how and why individuals do
> in fact use
> current profits to predict future profits?  Can you point to evidence
> showing they are using complexity theory to do this and that they are
> not, as Keynes claims, using current profits to predict what
> is knowably
> going to be different (even in the very short run e.g.
> Keynes's example,
> taken from Marshall, of the irrational influence of knowably
> temporary
> changes in railway revenues) from current profits and , in addition,
> within a psychological context of "mob psychology"
> irrationally treating
> as relevant many other kinds of irrelevant "news"?
>
> With respect to "uncertainty" you say both that complexity theory
> explains why the future is unknown and unknowable by a finite
> agent and
> that any other way of understanding the phenomenon is
> "religious" in the
> sense of not "scientific" because it's inconsistent with the
> ontological
> premises of orthodox natural science, namely those underpinning its
> conception of "determinism."  This is a religious way of defining
> science.  It dogmatically insists on the adoption of ideas
> that, among
> other problems (e.g. the "determinism" associated with the
> ontological
> idea of "internal relations" can equally well explain
> uncertainty), are
> self-contradictory i.e. against reason.  It implicitly makes
> its own the
> credo of anti-rationalist religion - credo quia absurdum.
>
> By the way, as I've pointed out to you before, treating what, for
> ontological reasons, is necessarily and essentially "vague" "as if it
> were precise and trying to fit it into an exact logical category" is
> another form of scholasticism.  If you believe that
> Whitehead's argument
> about the requirements material must meet for forms of reasoning that
> make use of the logical concept of the "variable" to be
> applicable to it
> is mistaken, you should (given that you understand yourself to be a
> rationalist rather than a dogmatic adherent of religion) be able to
> demonstrate by means of argument how it is mistaken, no?
>
> Ted
>