I just read an interesting blog post here by Krugman. It would take too long to summarize, but the essence of it is that heterodox economists [as he refers to them] are wrong to think that he and other orthodox macro theorists should be kicked to the curb because of the failure of the economics profession to predict the crash of 2008 and other things. The problem is not with our theory, Krugman says, but with those of us to paid too little attention to the evidence, preferring instead to elaborate elegant mathematical fantasies of perfect rationality and such.
Krugman alludes to people I have not read [and had not heard of, to be honest], and in any case I am not competent to evaluate his claims. But his remarks raise a question that I find quite interesting, so I will say something about it. [By the way, Krugman seems to be feeling just a tad defensive in the wake of the Piketty tsunami, even though he has been admirably generous in his praise of the book.]
The point is this. Any rich and complex economic model of the economy will almost certainly be flexible enough to allow those who use it to discuss almost anything with its aid, but that does not mean that the models are equally valuable. Each model directs the attention of its users to certain facts and structures in the economy and away from certain others. Hence the choice of model is important, for all that one can, with some effort, take account in one's own model of anything that people are saying who use other models.
The classical model developed by Smith and Ricardo and brought to its height [in my judgment] by Marx made a series of choices in its formal development that had the deliberate effect of highlighting two questions about the economy: How is the social product divided among the three great economic classes of society? and What are the conditions of sustained economic growth? These two issues -- distribution and growth -- are virtually thrust on anyone using the classical model. If you are doing economics a la Smith, Ricardo, and Marx, it is simply impossible not to notice that the interests of workers and capitalists are opposed, that what goes to one class is taken from the other. It is also impossible not to notice that luxury consumption detracts from growth.
The neo-classical model pioneered by Walras, Jevons, and Menger highlights two quite different questions: How are relative prices set in the market, and how do these prices direct producers in their allocation of scarce resources having alternative uses?
The question of price determination can certainly be studied in the classical model, but that study is simply a means to studying the distribution of the social product. By the same token, growth and distribution can be studied in the neo-classical model, but that is not its easiest or most natural use, and the model directs attention away from the fact that the interests of the working class and the capitalist class are necessarily opposed.
Heterodox economists who criticize standard economics [and by implication Krugman, although I would be surprised if he were their principal target] are not, I suspect, saying that the standard model is incapable of finding a place within it for inequality and economic instability. In all likelihood, they are saying [or, at least, they ought to be saying] that the standard model directs attention away from those issues instead of drawing our attention to them. And that is a legitimate criticism of an economic model, which is, after all, nothing more than an intellectual device for focusing our attention and directing it to what is important.
But of course, what counts as important is a matter of evaluative judgment, which means that Economics cannot be a value-free science. And that really is a challenge to Krugman as well as to the rest of the Economics profession.