Noah Smith has a nice summation of his critique of macroeconomics, which mainly comes down, as I read it, as an appeal for researchers to stay close to the ground. That’s definitely good advice for young researchers.
But what about economists trying to provide useful advice, directly or indirectly, to policy makers, who need to make decisions based on educated guesses about the whole system? Smith says, “go slow, allow central bankers to use judgment and simple models in the meantime.” That would be better than a lot of what academic macroeconomists do in practice, which is to castigate central bankers and other policymakers for not using elaborate models that don’t work. But is there really no role for smart academics to help out in this process? And if so, what does this say about the utility of what the profession does?
The thing is, those simple models have done pretty darn well since 2008 — and central bankers who used them, like Bernanke, did a lot better than central bankers like Trichet who based their judgements on something else. So surely at least part of the training of macroeconomists should prepare them to be helpful in applying simple models, maybe even in making those simple models better.
Reading Smith, I found myself remembering an old line from Robert Solow in defense of “fancy” economic theorizing:
In economics I like a man to have mastered the fancy theory before I trust him with simple theory … because high-powered economics seems to be such an excellent school for the skillful use of low-powered economics.
OK, can anyone make that case about modern macroeconomics? With a straight face? In practice, it has often seemed that expertise in high-powered macroeconomics — mainly meaning DSGE — positively incapacitates its possessors from the use of low-powered macroeconomics, largely IS-LM and its derivatives.
I don’t want to make a crude functional argument here: research that advances knowledge doesn’t have to provide an immediate practical payoff. But the experience since 2008 has strongly suggested that the research program that dominated macro for the previous generation actually impaired the ability of economists to provide useful advice in the moment. Mastering the fancy stuff made economists useless at the simple stuff.
A more modest program would, in part, help diminish this harm. But it would also be really helpful if macroeconomists relearned the idea that simple aggregate models can, in fact, be useful.