My old employer's blog has a terrific post on "second best economics":
Suppose that I told you that in the absence of the necessary conditions for teleportation, the next best thing is to forget all about the conditions for teleportation and instead fly at near the speed of light. Would you find this helpful if what you actually had was a Toyota and a half-tank of gas? Many "second-best" policy recommendations are a bit like that: the ideal market is a fantasy, so here is an ideal government to fix things. Obviously, this is not very helpful. If we are going to be hard-headed empiricists on the lookout for real-world imperfections, then we must admit imperfections equitably and not blanch in the face of the often appalling failures of government to operate as intended.
The excerpt really doesn't do it justice. Please read the whole thing.
Well, yes, it was an interesting blog post. However, the summary it offers of what "second-best theory" actually means seems to imply something different from what McArdle thinks it does. Here's the interesting quote from Lipsey:
This is an important point since much of the literature that is critical of second-best theory assumes that economists know a distortion when they see one and know that the ideal policy is to remove the distortion directly, something that is necessarily welfare-improving only in the imaginary one-distortion world.
Sounds like sense to me. After all, almost every actually existing market is, as a practical matter, always already (as the Marxists would say) distorted in a zillion ways. Taking the discussion on health care we've been having (interminably) on other threads, the US health care market is so unbelievably complex as to be completely warped -- as is every single other health care market in the entire world. What counts as a "distortion"? On what grounds do we argue that removing a distortion would be beneficial?
The blog argues that we ought to base our public policy decisions not on fictive notions of perfect efficiency models, but on proven real-world examples.
as the moral philosophers like to say, "ought implies can"—we are obliged to strive only for the possible. Likewise, a social ideal is worth aspiring to, and worth using as a standard of policy evaluation, only if we can get there from here. Then why should we care whether the conditions for a fantasy of ideal efficiency are present in the real world? The answer is that we shouldn't care. Furthermore, why should we care whether a given policy is "second-best" relative to our fictive efficiency norm? Again, we shouldn't.
Again, turning to our health policy example: we know that universal coverage is possible because every other advanced country has it. We know that large savings on health spending are possible because every other advanced country spends dramatically less than we do, and gets broadly identical or superior results. We have before at least a dozen examples of systems which are demonstrably superior to our own. It would be ridiculous to reject their actually existing examples because of imaginary ideals of perfect theoretical efficiency.
Posted by brooksfoe | August 26, 2007 12:18 PM