Part 9 in a continuing series on CBA -- Conclusions

There are large areas in the CBO report I haven't yet covered -- for
instance, I didn't even comment on how CBO treated lead and its important 
health effects -- but everyone has probably heard quite enough by now.
What can we learn from CBO's embarassing exercise in mis-calculation?
And, since every CBA I've seen has similar flaws, what does this tell us
about CBA in general?

I.    Economists tend to be scientifically illiterate.

With all due respect to the author of the CBO study, a scientist who worked
in the area of water quality would not have been misled by EPA's non-
mention of the causes of regulation of atrazine in its RIA.  Nor would
a scientist ever report any number as being within a range of 867,000 to
8.7 billion (Usenet readers who don't know what I'm talking about; look
up "significant digits" in any high school Physics text).  Clearly something
is getting lost in translation between the scientists who figure out what
is likely to happen under a certain set of circumstances and the economists
who write CBAs.  This is most apparent in any CBA that deals with issues
of biological science, especially ecosystem science.

II.   CBAs tend to ignore anything that isn't quantifiable.

Most of the problems in the CBA were recognized as potential problems, 
either by CBO or by EPA.  Unfortunately, instead of realizing that these
problems often meant that _no_ reasonable number could be produced, the
CBA writers went ahead and produced cost/benefit numbers anyway.  The result
was that, except for various qualifying statements in the body of the work,
the conclusions bore no relation to reality, since any non-quantifiable
benefits had been thrown out.  In particular, the amazing spectacle of CBO's 
treatment of atrazine and the mediagenic $4 billion per cancer case number
should be considered here.

III.  Because of points I. and II., CBAs are highly biased against the
      prevention of long-term hazards. 

CBAs are notorious for finding that the only cost-effective measures are
those that protect against short-term, obvious effects.  This is partly
because of the effect of discounting (benefits far in the future are 
discounted, while current costs are not) but partially because of points
I. and II. above.  Namely, long-term effects are difficult for those with
a lay knowledge of science to understand and are very hard to quantify.

IV.   Recommendation: Economists should stay out of the science policy 
      business; leave science to scientists and policy to politicians.
      Environmental CBA is a source of disinformation, not information.

My personal recommendation is that analysis of science-based issues should
be left to scientists and that the use of CBA in connection with 
environmental issues should be discontinued.  CBA on non-local issues
offers no useful information and is most often used as a tool of influence by 
anti-environmental political interests, who are quite smart enough to take
advantage of the points I've outlined above.  If these political interests
gain the power through the democratic process to get their way 
legislatively, that's one thing, but it is objectionable that they should
hide behind a screen of pseudo-science to do so.  

This series has by no means exhausted the legitimate objections to the use
of CBA.  In particular, none of the issues of differential cost and benefit
to different groups of people, issues of public valuation of risk by
other than strictly numeric standards, or issues surrounding the economic
valuation of human life by income have been addressed.  I thought it enough
for this example to start by pointing out the strictly factual errors in
the process.

I will expect discussion of this example before any more thoughtless paeans
in favor of CBA are posted to sci.environment.  To aid the memories of
certain sci.environment posters who seem to think that a post can be
ignored as soon as it expires from one's site, I have put this series of
articles on my Web page at http://www.mnsinc.com/richp/sdwa_cba.html

Disclaimer: The opinions expressed in this post and others in this series
are my own, not my employer's.