Re: Debunking Economics

[Steve Keen <keen@xxxxxxxx>]

Some good points here Robert; I'll reply in situ.

Robert Vienneau wrote:

> I have read through "Madness in Their Method" and skimmed some of the
> remaining chapters. I am looking forward to being able to direct people to
> Steve's book. I couldn't get through the chapter on original work on "Size
> Matters" because the graphs took up too much of my memory. I have some
> questions, though.
>

Sorry 'bout that! But with the number of illustrations in the text, there was
no way to avoid large files. Maybe the best trick would be to download the PDFs
and work within Adobe, rather than a (memory hugging) browser.

>
> I don't know that I've ever fully worked through the SMD theory. But I
> wondered if an assumption of homothetic and identical utility functions was
> SUFFICIENT, but not NECESSARY, for well-behaved aggregate demand functions.
> Might it not be the case that some other configuration of utility functions
> might, by weird happenstance, yield well-behaved downward sloping aggregate
> demand functions? Then saying that theory shows demand curves cannot be
> well-behaved would be too strong. It would only be that theory gives no
> reason to expect downward-sloping demand curves and suggests that they
> would rarely appear, at least on utility-maximizing grounds.
>

Good point. Bill Barnett has recently pointed out to me that there are some
other conditions which allow the Slutsky conditions to hold at the aggregate
level, but not the mainstays of neoclassical welfare SARP, WARP, and GARP. I
will probably footnote this though, because I believe that--as I argue in the
brief chapter on mathematics--the failure to prove aggregation under general
conditions is akin to a "proof by contradiction" that society can't be reduced
to the sum of its parts. That it is still passably possible to do so, under
some conditions, to some limited degree, to me doesn't really detract from that
fundamental failure.

>
> I think a lot of economists believe that marginal productivity says
> something about market outcomes being just or rewarding people for what
> they contribute. I find that belief silly. Did you find a written text
> saying that? I think a much more reasonable justification of capitalism
> would be based on Hayek and a discussion of the rules of the game. This is
> sort of Blaug's line in his pamphlet on the CCC. Might you want to mention
> something like this? Of course, such a defense of capitalism does not say
> anything about distribution being just, does not support a hard-right
> neoliberal attack on a mixed economy and policies to mitigate an unequal
> distribution, and permits experimentation with laws defining property
> rights differently.
>

I actually discuss Austrian economics in moderate detail (somewhat more than I
devote to Post Keynesianism!) in the "Alternatives" chapter. While I still
regard the Rothbard-style Austrianism with disdain, reading a book by the
Austrian philosopher Chris Sciabarra ("Marx, Hayek and Utopia", SUNY Press,
1996) has made me less dismissive of Hayek than I once was. It's also true that
the Austrians have a defence of capitalism which is not entirely dependent upon
the welfare concepts of neoclassical economics. But I feel that it is still
heavily dependent on capitalism straying not too far from a neoclassical
equilibrium, and the conditions of disequilibrium being not too different to
those of equilibrium. Since I dispute both hopes, I dispute some of the
normative conclusions the Austrians reach about capitalism.

But generally, I agree that Hayek et al give a much better foundation for
supporting capitalism than do the neoclassicals.

>
> I do not see that Bhaduri's argument that the marginal product of capital
> is generally unequal to the interest rate (because of price Wicksell
> effects) has an analogy for the marginal product of labor and wages. I
> think Hahn's Cambridge Journal of Economics paper on the Neo-Ricardians
> makes the claim for an analogous argument questionable. On the other hand,
> I hadn't thought about the connection of this argument to Sraffa's 1920s
> papers before. Given the stuff on markup pricing and the theory of the
> firm, I don't think you have need of some analogous argument here, anyways.
>

I thought the analogy--that what applied at the "micro" level did not apply at
the aggregate--was reasonable. Is that what you are objecting to? (in which
case I'd better read Hahn's paper).

>
> One aspect of reswitching I like is as follows. If a technique is optimal
> at two or more values of, say, the interest rate, then some wildly
> different distributions will be compatible with the same quantity flows, at
> least when looking at physical marginal products. Very different VALUE
> marginal products can be compatible with the same  relevant (left and right
> hand) derivatives in physical terms. So much for the vulgar belief in
> marginal productivity as rewarding factors according to their contribution.
> That's probably not very clear, but do you make some point like that?
>

I actually took a different tack to explaining Sraffa in this book than I've
taken before--more on the measurement issue, from which you can show that
rather than the rate of profit depending on the amount of capital, the measured
"amount" of capital depends on the interest rate. I then tangentially refer to
the possibility of reswitching, in one aside and a couple of examples. That's
one concept which I thought just too difficult for this book--and if any reader
wants to pursue it from there, they have references to follow now. I hope that
my measurement explanation will assist in coping with re-switching.

>
> I found the transition from the CCC to methodology too abrupt. I know this
> is not your fault. I have never been able to understand how one can respond
> to the logical difficulties highlighted by the CCC by citing the F-twist.
>

Oh well; methodology had to appear at some point! I thought that, by the time
I'd dumped on the preceding issues, it was time to raise the question of "how
can they defend such sloppy theorising?"

>
> Might you want to include something about Steedman and Metcalfe's critique
> of comparative advantage and the HOS theory of international trade?
>

Yes! But I ran out of both space and time. In the end, I decided to leave that
for the hoped-for second edition!

>
> Should there be something about general equilibrium theory here? I
> understand the claim that Sraffa showed distribution is prior to pricing.
> But, given Hahn, I thought that too strong. I think Sraffa showed that
> could be the case, and reconstructions of Classical wage theory and the
> Kaldor-Kahn-Robinson-Pasinetti PK distribution theory show how distribution
> theory and pricing theory can be combined.
>

I actually discuss GE in the section on dynamics, because I think that side of
its weaknesses hasn't been properly critiqued. In other words, Sraffa et al.
give a good reason for doubting its logical consistency, but there is a
question as to just how "general" is a generalisation of Walras which omits the
time issues of which he was so conscious?

>
> Maybe this raises a question of approach. If some ideas were debatable, do
> you want to take a position that a competent mainstream economist could not
> argue with? Or do you want to adopt a harder position consistent with, say,
> Neo-Ricardianism that might go a little beyond what the theory shows. After
> all, why should neoclassical theory be the default?
>

I'm trying to land somewhere between those two stools. I'm not trying to
convert neoclassicals, but I don't want to be easy pickings for them either.
But if my position is less impregnable than it should, but can only be defended
by requiring neoclassicals to "defend the indefensible", as the English
opponents of fox-hunting say, then I'd be delighted!

>
> I assume you're aware that some editing has to be done to make figure
> numbering and footnotes totally consistent with the text. Occasionally you
> use the phrase "saw off" in a way that I assume is Australian slang. I
> don't think that that works in American English.
>

Yes; I tried to rely on the program (Ventura Publisher's) auto numbering, but
it was buggy (despite its otherwise superior performance to Word). I'm probably
going the way of TeX after two less than perfect run-ins with Word "wanna be"
DTP programs (Lotus Word Pro and Ventura). The learning curve may be a real
pain, but from what I've seen, TeX works. I can't honestly say that for Word
Pro or Ventura--let alone Word.

And I replaced "saw off" with "transcended".

>
> I hope you find these comments helpful.

They were indeed--and my list of acknowledgements continues to grow. Many
thanks Robert,

Steve