This can't be real.

Photographer: Michel Dufour/wireimage/getty images

Breaking the Spell That Grips Economics

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Read More.
a | A

Back in the aftermath of the 2008 crisis, there were a lot of people criticizing macroeconomics, and rightly so. Macro models had failed to include finance, and thus had failed to spot the warning signs in the runup to the crisis. Overconfident macroeconomists had declared that the “central problem of depression-prevention has been solved,” only to be caught flat-footed by a new depression. Central bankers complained that the existing models were too rigid and complex to be of much use in an emergency.

Now, almost a decade later, a new breed of heavyweight macroeconomists has been taking their own field to task. And instead of complaining about the content of the models, they’re upset about the culture of the profession itself.

The most prominent new critic is Paul Romer, the new chief economist of the World Bank, and a former professor at Stanford and New York University. In a blistering new essay entitled “The Trouble with Macroeconomics,” Romer says that the field has become more like a religion than a scientific discipline. Instead of checking assumptions against data, he says, macroeconomists rely on tradition and received wisdom when constructing their models. Romer pulls no punches, calling this approach “post-real” and relating it to the debunked pseudoscience of the Middle Ages. He writes:

When a few talented researchers come to be respected for genuine contributions…Admiration evolves into deference to these leaders…guidance from authority can align the efforts of many researchers, conformity to the facts is no longer needed…if facts disconfirm the officially sanctioned theoretical vision, they are subordinated. Eventually, evidence stops being relevant. Progress in the field is judged by the purity of its mathematical theories, as determined by the authorities.

The chief villains in this process, as Romer sees it, are the Nobel prize-winning crop of macroeconomic theorists who transformed the field in the late 1970s and 1980s -- Robert Lucas, Edward Prescott and Thomas Sargent. Together, these superstars advanced theories holding that monetary policy has little or no effect on the economy.

Romer takes particular issue with this idea. Citing the sharp recessions of the early 1980s, which most people attribute to Federal Reserve Chairman Paul Volcker’s interest rate hikes, Romer says it’s bleedingly obvious that monetary policy can affect real output. He sees much of modern macro as one long attempt to deny this obvious fact.

It’s here that Romer’s critique is slightly lacking. Most of the profession does believe in the power of monetary policy. The dominant form of macroeconomic model for at least a decade has been so-called New Keynesian models, which say that interest rates play a very large role in stabilizing the economy. These are also the dominant form of modern macro model in use at central banks. New Keynesian theorists such as Greg Mankiw, Michael Woodford and Olivier Blanchard haven’t yet received Nobel prizes, but their views and ideas are thriving in the marketplace of ideas.

So the econ profession did come around, albeit too slowly, to the belief that monetary policy is very important. The revolution led by Lucas, Prescott and Sargent in the 1980s caused many methodological changes, but its substantive message -- that government couldn’t stabilize the business cycle -- was eventually rejected by most of the profession.

Now, the big question is whether faith in monetary policy might have been misplaced. The seemingly small effects of quantitative easing, and the difficulty of dealing with very low interest rates, are causing some macroeconomists to cast about for alternatives to the New Keynesian paradigm.

This illustrates the real fundamental problem with macroeconomics -- the lack of good evidence. Business cycles take years, and each one is a bit different from those that came before, so the number of data points is very small, especially relative to the enormous complexity of modern theories. Time-series statistics, one of the core methods of empirical macro, often requires almost as many assumptions as theory. What’s more, natural experiments -- the lucky accidents that allow microeconomists to glean facts about cause and effect -- are very rare in macro. That makes it very hard to test any theory. Evidence does exist in macro, but it’s thin, subtle and ephemeral. It gently nudges theorists gently in the direction of the truth, rather than giving them a hard push.

So Romer’s call for more realism and empiricism in macroeconomics, while welcome, will be hard to implement. Because data is sparser and weaker in macro, theorists have a natural incentive to rely on their own pre-existing beliefs, on generally accepted practice and on the wisdom of authorities. Revolutions in macro are rare. Instead, expect to see gradual evolution, accompanied by fierce debate, fads and fashions, and occasional bad feelings.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Noah Smith at nsmith150@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net