I'm shocked he's posted something on this topic that I agree with so much, so I ought to link it.
Don quotes Carl Dahlman who writes:
"This is not science; it is metaphysics: value judgments and political goals will enter into the determination of whether externalities occur in our world. You cannot show analytically that the government, in principle and in all cases, handles externalities better than the market; nor can you prove the opposite: it all depends on what point of reference you choose. And that is not a question of positive economics…. It is doubtful whether the term “externality” has any meaningful interpretation, except as an indicator of the political beliefs and value judgments of the person who uses (or avoids using) the term."
The only thing I disagree with is the first sentence that it's not science - it's metaphysics. This was my comment on the post, which should explain what I mean:
"Exactly right - I'm shocked you've posted something I agree with on this topic! Although I'm not sure this means it's "metaphysics". It means its subjective - and all values in economics are subjective.
In a market transaction, each person acknowledges the standing of the other person in dealing with these subjective values. That's the strength of market transactions. But of course not all actions occur in the market, and when it doesn't occur in a market there are very real questions about a person's standing.
Do I have standing when it comes to the way I subjectively value pollution? That's a much tougher question that people disagree on. It's not metaphysics. Nobody can contest that I bear a subjective value about those things any more than you can contest that I bear a subjective value about the market transactions I make. The question is, what do we do about my subjective values regarding these externalities. But that was never a positive question anyway - that was always a normative question."
Last weekend I read an article by Pete Leeson on the theory of clubs and public choice - but it touched on externalities and dismissed them. The article was good as a whole, but I thought he did a really atrocious job handling the externality question. That and Don's post really make me want to write more about this now.