I was pretty impressed with much of Krugman's NYT Magazine magnum opus. Macro is a mess. Now, this isn't what Krugman was saying, but I think his account of the disagreements on fundamental questions exposes macro as a proto-science at best.
Economics, as a field, got in trouble because economists were seduced by the vision of a perfect, frictionless market system. If the profession is to redeem itself, it will have to reconcile itself to a less alluring vision — that of a market economy that has many virtues but that is also shot through with flaws and frictions. The good news is that we don’t have to start from scratch. Even during the heyday of perfect-market economics, there was a lot of work done on the ways in which the real economy deviated from the theoretical ideal. What’s probably going to happen now — in fact, it’s already happening — is that flaws-and-frictions economics will move from the periphery of economic analysis to its center.
There’s already a fairly well developed example of the kind of economics I have in mind: the school of thought known as behavioral finance.
One might have thought Krugman was going to do something like acknowledge the immensely important point, associated with economists such as Ronald Coase and Douglass North, that market institutions in which “frictions” or transactions costs are relatively low are the exception rather than the rule. Markets are not only not frictionless, frictions generally keep markets from getting off the ground at all. When frictions are managed sufficiently to get markets up and going, that's because they are embodied in a complexly interlocking set of institutions and organizations which make this possible. A scientific economics might seek to explain how it is that embodied markets achieve otherwise impossible feats of social coordination.
The “flaw” part of “flaws and frictions” is a little loaded. It's an annoying habit of economists to hold on to homo economicus as a standard for rationality even after they have conceded that homo economicus is a more or less useless over-idealization. That we don't live up the standard of more or less useless over-idealizations obviously does not imply that we are somehow defective. Be that as it may, one might have thought the recognition that a useless over-idealization of rationality does not apply to us might lead one toward a more sophisticated idea of the way minds and markets work together. There are, for example, the profound Hayekian points that individuals are computationally bounded, that expertise is local, that markets enable coordination by conveying otherwise inaccessible information, that epistemic and practical norms are both cause and effect of institutional structure, etc. Maybe we could look at experimental work, such as Vernon Smith's, that explores how real people operate in different kinds of market structures.
It's not like Hayek, Coase, North, and Smith don't have Nobel prizes! But Krugman ignores the best of existing “flaws and frictions” economics and jumps straight to “behavioral finance,” which I'm fairly sure is the same old shit Krugman is complaining about — elegant models of counterfactual worlds — with ad hoc emendations to improve fit with the history the actual world.
Krugman should go further, but he won't. He should say that beginning without “flaws and frictions” — assuming at the start unbounded perfectly rational agents and zero transactions costs — has all the virtues of theft over honest labor. An economics based on those assumption is ipso facto unscientific. The same goes for ad hoc variations on these assumptions. What sciences do is explain. (Sorry Milton.) And scientific explanation is largely a matter of detailing the causal mechanisms underpinning observed regularities.
“Freshwater” economics is not a science. It is a sometimes illuminating exercise in modeling counterfactual worlds. Insofar as “saltwater” economics recognizes that the need for a better account of human psychology and transactions costs in embodied institutions, it is better. But, so far, it isn't. So far, “behavioral” macro is mere aspiration. It's not something anyone is actually doing in a systematic way.
Maybe the most important conclusion I drew from Krugman's piece is the politics of the freshwater/saltwater divide is complete nonsense. To seriously acknowledge “flaws and frictions” is to acknowledge that some institutions create friction while others reduce it; that some institutions enhance the salience of certain “flaws” while others work around them; etc. Having recently read a bunch of “Keynes was right” pieces, it seems pretty clear that lots of left-leaning economists are mistaking flawlessness and frictionlessness as necessary premises in the argument for limited government intervention in market institutions. But the upshot of flaws and frictions could very well be that we shouldn't expect very much from government intervention. It seems pretty clear to me that Keynes' characterization of the role of not-exactly-rational “animal spirits” in recessions is a very small part of an adequate general account of the way the quirks of human psychology tend to scale up to the macro level. The inference from flaws and frictions to Keynesian technocracy tends to be embarrassingly hasty.
The fact is, macro isn't close to resembling a real science. (“The economy,” nationalistically construed, isn't even close to resembling a subject of scientific investigation!) But we can't count on elite economists to admit it, since their claim to authority on matters of public policy stands or falls with their claim to scientific expertise.