Re: poly: Modeling Economic Singularities

From: Perry E. Metzger <perry@piermont.com>
Date: Mon May 04 1998 - 17:41:01 PDT

"Peter C. McCluskey" writes:
> perry@piermont.com ("Perry E. Metzger") writes:
> >The "overall" approach isn't sufficient. If you have a mathematical
> >model, at the very least you have to try plugging in historical data
> >into the stated equations and see if it comes out with a reasonable
> >match. Even that isn't always enough.
>
> And you'd rather complain about the absence of such testing rather
> than do such tests yourself.

I was under the impression that, in the scientific method we've been
operating with since, oh, Francis Bacon, that most people agreed that
it was the person PROPOSING a hypothesis that was responsible for
proving it, and not the rest of the world.

I've become very disappointed in the field of Economics. Just because
I agree with a large segment of the economic community (the free
market camp) doesn't mean I agree with their methodology. Their
methodology appears to involve experiment only very rarely, and in
general appears to have as much rigor as their pseudoscientific
opponents. Experimental economics *is* a pretty vibrant field, but a
rarely practiced one compared to other more solid "sciences". In the
usual case, no one demands the same levels of rigor from economics
that is expected of, say, a pharmeceuticals researcher.

This, of course, has lead to a rather nasty result. In fields like
physics, someone claiming that objects fall up or that prayer can
impact the path of a particle would be laughed out of the field. In
Economics, though, the standards are low enough that someone claiming
that minimum wage laws won't hurt anything can happily hold a tenured
position at Harvard. Consider the amount of damage this has caused the
world.

I make the (speculative) claim that this is because the field does not
conduct itself with rigor. Sure, Milton Friedman gets to publish
brilliant papers and no one disturbs the grand old man with requests
that he find someone to try to verify his hypotheses, but that very
laxity with the economists we all know and love leads to people like
Robert Reich being able to spew the purest excrement from behind
comfortable full professorships without so much as a second glance.

Because the field demands so little of its practitioners, we are
forced to accept the crud as well as the gold, and people who are
clearly charlatains get to have as much standing as people who are
probably right. In all fairness, why should this not be, given that
both have essentially the same evidence for their positions?

If everyone was held to the same standards as chemists or physicists
had to follow, this would not be possible. The entire phenomenon of "I
see your nobel prizewinner who claims the minimum wage is bad and
raise you two nobel prizewinners who claim the opposite!" would go
away -- we'd have answers, not muck.

In short, it is not enough that I agree with someone like Robin in a
general way -- it is necessary that he climb on out of his goddamned
ivory tower and play by the 398 year old rules in the Novum
Organon. How smart he is has no relevance, and how much he acts like
his fellows in the field has no relevance either. I even tend to
believe what he says in a vague general way, but I rarely believe he
has any solid evidence for what he's saying, so my feeling is as good
as worthless, too.

[As an interesting aside, let me note that decades of "portfolio
theory" have produced no tangible results for people trying to figure
out what to invest in, but simple fact gathering (like Morningstar's
information on relative mutual fund performance) and experiment
regularly do produce interesting results.]

> or are you trying to convince me to put you in my killfile?

I think I'm rapidly losing interest in the vague speculations on the
"Polymath" list. You might as well save yourself the trouble. I'm
unlikely to remain very long.

I don't mind the occassional speculation, but I do get a bit sick of
nothing but speculation after a while.

Perry
Received on Tue May 5 00:44:04 1998

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST