Re: poly: Anarchy and Empiricism

Perry writes:
>I'm very strongly *inclined* towards anarchy. I favor the system and
>believe that there is good evidence for it, but I don't think the
>evidence is by any means good enough -- I will fully admit that I'd
>prefer to have better evidence for it. Given that the evidence for
>anarchy is better than that for political control, I choose anarchy, ...

This would be a fine place for you to contrast your high standards with
the unscientific sloppy academic economics you complain of. Please
point us to this evidence and a careful analysis of it you respect.
 
Given your complaints, I'm sure this analysis won't have fallen prey to
measuring what's easy to measure, rather than what's important, to using
anything less than careful statistical tests, or to implicitly making
inferences via less-than-completely-uncontestable background theories.

>Part of my disappointment with the economics field is the fact that,
>in spite of the fact that many fundamental questions about the impact
>of political control on the economy could have been definitively
>settled by now, there is still vast amounts of fundamental
>disagreement in the field and much of the public discourse on
>important topics grinds down to competing groups of economists making
>diametrically opposing claims -- something that one does not see in
>more empirically solid areas of study.

If Congress were considering a bill whose merits depended on whether the
universe was open or closed, you can bet that even though the weight of
evidence clearly favors an open universe, public discourse would include
confident claims by academics (like Tipler) that the universe is closed.
This would happen for a wide variety of methods of cosmologists.

Competing groups of economists appear in public discourse mostly because
competing economic groups out there care so much about which economic view
is considered right; economic methodology is largely irrelevant.

Peter C. McCluskey comments:
>>It seems to me that you are imposing different standards on economics than
>>are usually applied elsewhere.
>
>He appears to be imposing standards that are needed for some areas of
>economics where the important theories are fragile enough that similarity
>to well-tested ideas is not enough.

I fully grant that one's confidence in the conclusions of a theoretical analysis
should depend on the fragility of the theories one uses, as well as how far
the situation is from where those theories have been tested. And I grant that
direct tests of such conclusions can greatly improve one's confidence.

Robin Hanson
hanson@econ.berkeley.edu http://hanson.berkeley.edu/
RWJF Health Policy Scholar, Sch. of Public Health 510-643-1884
140 Warren Hall, UC Berkeley, CA 94720-7360 FAX: 510-643-8614
Received on Wed May 6 18:11:37 1998