Russ Roberts interviews Ed Leamer. If you are not already familiar with Leamer’s views on econometrics, then you should listen.

Leamer has a notion of macro that I think needs more elaboration. He says that in a normal economy, when someone loses a job it does not affect anyone else. However, in a weak economy, one person’s job loss causes other people to lose jobs. This is similar to Leijonhuvfud’s corridor idea. The problem, for those of us who believe that something like this is true, is to explain the relationship between the good economy and the bad economy.

If we think that the economy makes smooth transitions from one state to the other, then the challenge is to tell the difference between the two. If we think that there is an abrupt difference, then the challenge is to explain the discontinuity between the two.

One abrupt-transition story is between a high-confidence state and a low-confidence state. When people have high confidence that, say, house prices will rise or the government of Greece will repay its debt, then borrowers and lenders both think they are well off and can consume a lot both today and in the future. In a low-confidence state, borrowers lose the ability to borrow and lenders take losses. This makes them realize they are not as rich as they thought they were (a Tyler Cowen point), and they suddenly shift down their consumption to reflect this new information.

Leamer suggests briefly that the economy can recover on its own from this sort of change in state. However, government needs to reduce uncertainty, rather than exacerbate it.

I think this is more on track than crude Keynesianism. However, there is much that needs to be filled in.