Externalities and the Case for Government Intervention

I had a post yesterday commenting on what I thought is the difference between a smart, honest conservative and an idiot. Namely, as I put it:

the former recognizes that externalities are a problem and there’s some sort of a role for government in correcting them. The latter talks only about the beauty of the free market and property rights.

I mentioned as examples a quote by Milton Friedman, and I also mentioned Arnold Harberger, whose work I have discussed a few times on this blog.

In comments, Dan in EuroLand writes:

Cactus, the point is that MF never mentioned that Gov’t is the best answer to externalities and market failures. A smart “conservative” may recognize externalities exist but that gov’ts may lack the ability to “solve” the externality problem.

Off the top of my head, I can’t recall enough of what MF wrote to say whether is true or not. And its easy to see examples where government makes things worse, not better. (I remember being in a class taught by Arnold Harberger when he said something to the effect that one thing the government should definitely have is a department that goes back and examines the claims made of various programs ex-ante and see whether they worked out ex-post. He felt it would be a good way to diminish the number lousy programs that cropped up over the long run, and of course, he was right.)

But let me try my hand at making the argument one more time…

Assume that every government action has a cost, and that no government action to correct an externality suceeds perfectly, and may even fail utterly. Now, say two parties have a contract, and that contract imposes some sort of a cost on third parties.

To say there is no role for government in correcting externalities is the same as saying that in every instances, the cost of that externality is less than combined cost of government intervention plus whatever the added benefits are of that government intervention. Alternatively, the aggregate sum of the cost of all externalities exceed the aggregate sum of the cost of all government intervention plus the added sum of whatever the added benefits are of that government intervention, and that in individual cases, it is impossible to tell what those components might be.

I’ll leave aside the whole question of whether we should include in that whole equation the issue of the legal system itself, and who ensures that those property rights can be maintained without violence. I would, instead, point out that we live in a world in which externalities can wipe out a very large number of people. And to go all Coasean, this includes externalities for which the imposer of these externalities, or even that the externalities are occurring at all, are unknown until that very large number of people are wiped out. In a world of limited liability caused by imperfect information, at some point, someone will have a business plan in which the expected private gains to engaging in such acts exceed the expected private costs. This is not an assumption – someone made decisions at 3 Mile Island, and Bhopal, and there’s always someone deciding to produce and sell fake medication or tainted food. Now completely reduce the probability that those making such decisions will face any penalties except those imposed by the market. Is that where you want to live?

Anyway, this is all I have time to write this morning. Your thoughts?