===================== HES POSTING ==================== 
 
I've been reading the exchanges stemming from Buchanan's request for 
citations to articles about the "sorry state of economics." I know this is 
in bad taste, but the exchanges have provoked me enough to offer a 
citation to some tangential work of mine. I do this because I think the 
topic I have tried to address is of general interest. 
 
I have an article that has been accepted by the Journal of Economic 
Methodology entitled "Now You See It, Now You Don't: Emerging Contrary 
Results in Economics." The article focusses on empirical literatures in 
which an empirical result is first established, but later on conflicting 
evidence seems to emerge. I have 26 examples in the article. I then 
try to hypothesize competing explanations for this "contrary results" 
phenomenon; nine are set forward in the article. I then try to "assign" 
the nine explanations to the 26 examples. 
 
>From my point of view, a major reason these empirical flips are of 
interest has to do with how economists come to believe what they believe. 
One (Friedmanesque and very naive) idea is that we believe what we 
believe because the believed propositions have been subject to repeated 
empirical "tests," and have survived. My examples suggest that empirical 
results in economics sometimes (often???) flip-flop. This makes testing a 
much "ifier" proposition. It also poses immense difficulties for the 
conscientious economist who is trying to make policy recommendations based 
on the empirical findings in the literature--he or she faces the 
difficulty that established results may be contradicted "tomorrow." 
 
I consider all this tangential to discussions about "the sorry state of 
economics" in the following sense. I DO NOT offer this view about 
"reversing empirical work" as a testimonial to some alleged "sorry 
state." I offer it instead in the following spirit: we (as economists) 
need to do the best we can to bring "the empirical facts" to bear on what 
we believe. My work, if it does indeed contain some accurate findings, 
testifies to how difficult it is to figure out in a compelling and useful 
way what it is " the empirical facts" really are. So I view it as 
"attempted consciousness-raising" about empirical difficulties, NOT as 
some attempt to belittle or dismiss the entire enterprise. 
 
Robert S. Goldfarb 
Department of Economics, George Washington Univ 
============ FOOTER TO HES POSTING ============ 
For information, send the message "info HES" to [log in to unmask]