I don't think marginal utility theory is circular in this way. It's well-known that people drink more when the drinks are free, while they'll be more restrained if they have to pay for each drink. This fits well with the idea that if MU > price, one continues to drink, so with diminishing MU, the lower the price the more one drinks.
I think that the problem is different. As one drinks alcohol, one's tastes (judgement) change. Neoclassical theory can't deal with endogenously-driven taste changes while maintaining its basic "consumers rule" attitude. Further, there's likely more than one equilibrium, which messes everything up. Of course, tastes can also be changed via advertising and other societal influences. Neoclassical economics, almost by definition, doesn't deal with the determination of tastes. On top of that, the neoclassical vision of psychology is stunningly stupid. Usually not acknowledged is the fact that some people have a (genetic?) predisposition toward alcoholism or addiction in general. The internal conflict between "have another" (the id), "you won't be able to drive" (the ego), and "it's immoral" (the superego) is of course ignored. I'm hoping that someday experimental/behavioral economics provides us with a more sophisticated vision of psychology. It too will likely ignore the sociological aspects, since all economists _know_ that sociology is fuzzy BS. In general, I'd say that the stuff about marginal utility is a theoretically superficial and unsatisfying effort to explain the normal downward slope of demand curves. Perhaps it would be better to simply skip homo economicus and instead simply use demand curves. Of course, as I've said, I'm not a micro-man. Calling Gil Skillman or Julio Huato in here to answer this question! On 5/18/06, Walt Byars <[EMAIL PROTECTED]> wrote:
So, the price of a good is determined by the utility (or some other property) of the marginal unit of the good. But, obviously, which unit is the marginal one consumed depends on the price. This of course could be resolved by postulating a demand curve with all of each price and its corresponding marginal unit, and the corresponding profit maximizing price chosen. This then relies on cost of production. You can regress all the way to a point where there is no circularity. But, for more political economy oriented people, doesn't the cost of production of goods, even when you regress to the point where there are no produced inputs, depend on prices? Say at certain prices of goods, workers will effect a slowdown in protest etc... Is this a circularity or is it resolvable?
-- Jim Devine / "the world still seems stuck in greed-lock, ruled by fossilized fools fueled by fossil fuels." -- Swami Beyondananda
