I am no expert on this...but I agree with Gar that it would seem that "the theory of second best" is not enough of a formal - or rather binding - proposition that it could be "disproved" (plus I have an allergy to such language in economics, anyway).
Like Michael, my understanding of the "theory" is: 'if one condition for optimality has been violated then it *may* be possible to get closer to optimality by violating additional conditions'. Obviously this puts us in some form of General Equilibrium mentality, lack of true historical time, etc. So, overall, rather than focusing people on how the big processes that affect their lives are truly motivated and decided, such frameworks create an illusionary world of rational choice in the hunt for optimality. To me, one could get a lot more by saying 'don't let the perfect be the enemy of the good'. AFAIR, the textbook reference is from a 1950's article "The Economic Theory of Second Best" by Lipset and Lancaster in the Review of Economic Studies. In applied fields, I find that the proposition is often used in two contexts: assessing market failure (skewing a bit left??) and assessing government failure (skewing right). A good and common example of the former might be Stiglitz's article on the use of cost benefit analysis for public projects (an NBER paper from the early '80s) which was dragged out a lot during the Clinton administration and is used in Public Economics courses. As Gar says, there are many examples in International Economics (I find most try to prove that govt intervention fails and that "free trade" is preferred as the second best). FWIW Paul At 10:05 PM 11/18/2005 -0800, you wrote:
Ah - the version I heard was in regard to trade was that tariffs are always a second best solution to market inefficiencies. That is it acknowledges the existence of market imperfections, but says there is always a better way that tariffs of solvings them. Been a long time: Ok, one example I was given was a foreign monoply. That is, even under neo-classical assumptions, a market imperfection in which tariffs would be better than free trade. But the claim was that price controls and regulation would be a better response than tariffs - because you are preventing monoply rents rather than taxing them. Similarly there are infant industries - tariffs would be better than unmitigated free trade, because you develop an industrial base. But subsidies are better than tariffs, because where tariffs are a regressive tax you can get at least some of the taxes for subsidies from the rich, and then your subsidized industry still have foreign competition to keep prices down. You get the idea.One thing that always bothered me about it, was wondering if it really was true that there were not market imperfections (under neo-classical assumptions) where tariffs were the best remedy, rather than the second best remedy. On 11/18/05, Michael Perelman <[EMAIL PROTECTED]> wrote: > Here is the basic idea. Suppose you have a situation like rent > control. Free marketeers say that abolishing rent control adds to > efficiency. The theory of the second best says that it might not > work that way unless all of the other conditions of perfect competition > are met. In short, following market logic might not work, even > according to the sort thinking that free market absolutists promote. > > > On Fri, Nov 18, 2005 at 09:01:47PM -0800, Gar Lipow wrote: > > On 11/18/05, Michael Perelman <[EMAIL PROTECTED]> wrote: > > > One of my students told me that he read that someone disproved the theory of the 2d best. > > > Has anybody heard of such a proof? > > > -- > > > Michael Perelman > > > > Is the theory of the Second Best a general enough theory that it could > > be conclusively refuted? I've always run into as a series of ad-hoc > > rebuttals, comparing trade policy to other rememdies to market > > inperfections, and showing that in a series of particular cases, trade > > policy is a "second best" policy. Is it actually a general welfare > > theorem with general proof or evidence? > > -- > Michael Perelman > Economics Department > California State University > Chico, CA 95929 > > Tel. 530-898-5321 > E-Mail michael at ecst.csuchico.edu >
