Are you referring to the Arrow impossibility theorem? Samuelson says that
the theorem does not refer to a social-welfare function but to a voting
rule. If you go back to the original Bergson article (written under the
pseudonym Burke), you find the social welfare function formulated in a way
that makes it virtually impossible to refute.

Or to measure.

The main point about GNP or GDP is that it is supposed to be a measure of
how much junk is produced. Like the horizontal axis in the widget market,
its units of measure are things. For happiness or well-being -- or survival
-- you need another variable or two. Or ten.

At 17:39 9/05/05, you wrote:
One of the problems with a GPI-type measure is that it assumes that
there's a social welfare function, something that economists have
shown cannot exist (though they then go on to use it or similar,
implicitly or explicitly). The only way I know to aggregate
"preferences" is through democracy, not by adding things up using
fixed weights (or whatever).

> >But if the government started using the GPI (or similar) to measure
> >the economy's success, it wouldn't last -- unless there were a mass
> >movement to back it up.
>
> It would also be really really hard to calculate. There are enough
> imputations already in the NIPAs....
>
> Doug
--
Jim Devine
[EMAIL PROTECTED]
http://myweb.lmu.edu/jdevine

Robert Scott Gassler Professor of Economics Vesalius College of the Vrije Universiteit Brussel Pleinlaan 2 B-1050 Brussels Belgium

32.2.629.27.15

Reply via email to