In reply to Jim Devine's interesting comments:

Say, economics is whatever economists do.  They provide ideological
ammunition to the class enemy.  That should be obvious on this list.  But,
why is their ideology so effective?  Under certain conditions, some ideas
become material forces.  Not any ideas.  Some ideas.  A Hollywood movie may
be trash, but not all trash becomes a Hollywood blockbuster.  Perhaps the
*content* of economic theory (and the way economic theories are constructed)
has something to do with its effectiveness.  But let's see.

It turns out that fundamental uncertainty is crucial to fixed investment
decisions. Further, if one does not acknowledge the existence of
uncertainty, money is neutral, which empirically it is not.

Why does the "recognition of fundamental uncertainty" has not gotten much traction in economic analysis while, say, "rational expectations" and "informational imperfections" have? Is it that such "recognition" is potentially subversive, while the hypothesis of "rational expectations" or "imperfect information" foster acceptance and adherence to the status quo? Is it that the powers that be are afraid that the notion of "fundamental uncertainty" be used by the workers' movement to hit the system on the head? I don't think so.

Here's a straightforward conjecture about why some economic ideas don't
travel far, and others do:

http://web.mit.edu/krugman/www/dishpan.html

Back to Jim Devine's remarks, "fundamental uncertainty" is "crucial to fixed
investment decisions."  Okay.  Assume "fundamental uncertainty."  Now, what
kind of definite, unambiguous long-run investment behavior follows from that
assumption?  None in particular.  All types of behavior would be consistent
with that kind of uncertainty.  People "need conventions."  So how do you
decide what conventions are best?  How are those conventions formed?  How
does that sharpen your analysis?

Say that, "empirically," money is not neutral even in the very long run.
Now, say, you need to conduct a policy intervention in the real world.  Say,
you need to provide union leaders or the leaders of a popular movement with
a good rationale to oppose or push some monetary policy reform or measure.
So you need to quantify the non-neutrality of monetary policy in the long
run to argue the impact.  In other words, you need to estimate the linkage
between, say, interest rate and employment.  So you need a theory (a model)
that implies non-neutrality in the long run.  Do you assume "fundamental
uncertainty" in your model?  If so, how do you derive logically the
non-neutrality from your assumption, to the exclusion of -- say --
neutrality?  You can't!  Actually, from "fundamental uncertainty" you can
derive *anything*, including neutrality in the short run!  On what grounds
do you discard short-run neutrality then?

The question here is: How does the "recognition of fundamental uncertainty"
help an economist discipline her analysis?  The answer is: It's not clear
how.

[Actually, in the last few days, I've come across a series of paper that
claim to model "Knightian uncertainty."  I'm moving slowly with this stuff.
It comes from the decision theory literature applied to economics.  Here are
some references: L. Epstein & T. Wang, 1994, "Intertemporal Asset Pricing
under Knightian Uncertainty," Econometrica, 63; W. Brock & S. Darlauf, 2001,
"Growth Empirics and Reality," The World Bank Economic Review 15-2;  and W.
Brock & Anastasios Xepapadeas, 2003, "Regulating Nonlinear Environmental
Systems under Knightian Uncertainty" in R. Arnott et al. (eds) Economics for
an Imperfect World: Essays in Honor of Joseph E. Stiglitz, MIT Press.  It
seems to be a Bayesian approach where you try to derive robust choices (a
"precautionary rule") when there's uncertainty, not only about the sample
from which your parameters are estimated, but about the underlying theory
itself.  It's not clear to me that this would be enough to satisfy a
hardcore, nihilistic Knightian either.  Because, what's the "scientific
basis" to estimate theory uncertainty if not subjective belief and/or some
assessment of previous historical experience?  Jim Devine says that
according to Keynes, in the face of fundamental uncertainty, people need
conventions.  Well, precisely.  Think of "rational expectations" and
hypothesis of the sort as Keynesian conventions, except that they're
justified on the basis of their contribution to the tractability and sharp
implications of the theory.]

Third, the idea of uncertainty tells us to reject models that assume that
there are perfect futures market (i.e., standard general equilibrium
theory).

This is a misunderstanding. Nobody (but a Republican I met the other day while canvassing for ACT in Montgomery County, PA) believes that "there are" complete or perfect future markets in the real world. That is not what general equilibrium theory is about. Actual future markets are whatever they are in real life. *Assuming* that there are complete markets for contingent commodities is not believing that they exist. The assumption is a simple way to deal with uncertainty in economic theory. That's the Arrow-Debreu approach. But, again, that is *not* an ontological assertion. They couldn't care less about the actual nature of uncertainty in the world. All they want is some simple way to model it so that they can derive testable implications. Would you fault car manufacturers for not using human beings in their car-safety tests? Shouldn't they keep using rubber dummies instead, even though dummies have *only some* (very superficial) but *not all* of the attributes of actual human beings?

We need to make up our mind.  If we are going to criticize Hollywood's
output on the grounds that car chases in real life are not nearly as
spectacular as those we see in Schwarzeneger's movies, I don't think we'll
be able to understand why and how those movies command the audiences they
do.  If you have a better make-believe world, show us.

I didn't say that either simplicity or internal logical consistency was
dispensable.  Rather, the problem is that many if not most neoclassicals
see them as _substitutes_ for an empirical orientation (as do dogmatists of
other schools).

Frankly, I don't understand this statement. I'd think that most economists believe that theoretical and empirical work are complements, not substitutes. In theory, you want parsimonious, tractable "mechanical analogs" that replicate *some* (not all) of the attributes of a real economy. From those devices, you wish to deduct assertions about the behavior of the economy when certain things change, assertions that you can then test against available empirical information.

They are not. This is especially a problem since the theory that emphasizes
deduction over empirical orientation is used to dream up policies which are
then actually put into practice by organizations such as the International
Monetary Fund. (For example, I heard one IMF economist say that the
determinant of the level of employment was the size of the labor force.)
The cookie-cutter/one-size-fits-all approach of the IMF follows from not
only its top-down (dictatorial) approach but also its application of
neoclassical dogma, caring only about the logical consistency and
simplicity of their models.

Are you saying that we should reject Marx's insights because some Marxists exhibit the political behavior of a toddler? So, are you saying that we should reject the implications of an abstract, deductive "mechanical analog," not because the mechanical analog is logically flawed, but because IMF demagogues use it to validate their decisions?

Let me clarify this in case it is not obvious: The ideological role that
economic theory plays is evident.  It's not reasonable for us to expect
professional economists to undertake a comprehensive and radical critique of
capitalism.  As a rule, professional economists (like computer programmers
or poets) respond to the incentives the system has in place.  That's to be
predicted from a straightforward application of their own models.  But for
an *economic* ideology to be of any use, it has to have a modicum of
practical relevance, which is to say that -- one way or another -- it has to
deal with the real world.  Actually, if you push me, it's clear to me that
for the longest time in modern history it's been the advocates of capitalism
-- not its enemies -- who have been more willing (and able) to acknowledge
and deal with the real world as is.  Else we'd have a different world
already.

Of course, we should never take those results at face value.  They should be
criticized.  And critique, in the tradition I'm most familiar with, means
not only rejection, but also a painstaking appropriation of the germs of
rationality contained in a given doctrine.  And in this task, conventional
economists have made (and continue to make) things increasingly *easier* for
us.  By clearly specifying the assumptions of their theories, and by
deriving their results with strict compliance to public-domain mathematics
and formal logic, these guys are doing us a big favor.  It becomes much
easier for the critics of capitalism to find when and where the Man is
sticking the rabbit into the hat before he pulls it out.  Still, we need to
do our homework.

Talking about homework, that's all I can type for now.

Julio

_________________________________________________________________
Visita MSN Latino Noticias: Todo lo que pasa en el mundo y en tu pa�n, �en
tu idioma! http://latino.msn.com/noticias/

Reply via email to