Now I get it.

Source: Bettmann/getty images

Answering the Hardest Question in 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

Xavier Gabaix, a New York University economist who gets far less attention than he should, has written what might prove to be the most interesting macroeconomic theory paper in years. The title, “A Behavioral New Keynesian Model,” isn't exactly exciting and the paper is still incomplete, but it might help resolve the most important and difficult macroeconomic debate in academia today -- whether low interest rates cause inflation, deflation or neither. And it might signal a sea change in the way macroeconomic theory gets done.

Traditionally, macroeconomists have believed that low interest rates encourage inflation. But first Japan, and now the U.S. and Europe have kept rates low for years now, and inflation has stayed stubbornly low. A radical group of macroeconomists, including Stephen Williamson of the Federal Reserve Bank of St. Louis and John Cochrane of the Hoover Institution, have introduced a new theory called Neo-Fisherism, which says that a long period of low interest rates actually holds prices down instead of pushing them up. Williamson and Cochrane have both repeatedly stressed that New Keynesian models -- the most mainstream type of macroeconomic theory -- can easily yield the Neo-Fisherian result instead of the traditional view. One problem is that the standard models are often ambiguous -- they offer a number of possible, radically different outcomes for the economy, with no way to tell which will happen.

Gabaix tackles these problems with a simple, intuitive, yet bold step. Instead of assuming that people are perfectly rational, he theorizes that they have limited attention -- what psychologist Herbert Simon called “bounded rationality.” When interest rates or gross domestic product change, people in Gabaix’s model don't quite realize that things are different. Even more importantly, they’re short-sighted -- they don’t think as much about the probability of a recession happening 10 years from now as they do about one occurring in the next six months.

Those ideas probably seem obvious to most people. When events are further in the future, you worry about them less, right? I know I do. But to macroeconomists, this is a pretty radical step. Most macroeconomic researchers are strict adherents to the cult of perfect rationality. If the economy looks like it’s being driven by behavior that isn't quite rational, macroeconomists usually bend over backward to explain it as a failure of economic institutions, rather than a result of human psychology. Christopher House, my own teacher at the University of Michigan, thought behavioral economics was a fad, and would never have a big influence on macro theory.

Gabaix might just prove House wrong. Along with another recent paper by Mariana Garcia-Schmidt and Michael Woodford, Gabaix’s new theory is putting human limitations front and center, and slaughtering the sacred cow of Homo economicus.

And Gabaix gets many eye-catching results. First, the battle between Neo-Fisherism and traditional monetarism is resolved -- low interest rates increase inflation for a while, and then decrease it in the long run. So both theories are right. That result seems to fit the experience of Japan, which has had interest rates at zero for decades, without any sustained rise in consumer prices. Unlike in standard theories, Gabaix’s model has no ambiguity about the economy’s future path -- in econ jargon, the equilibrium is unique. And his ideas might also explain consumption and saving behavior more accurately than the typical approach, which assumes that people decide on how much to save today based on how much they might be making two decades in the future.

Gabaix also resolves another longstanding puzzle. Standard macroeconomic theories say that so-called forward guidance by a central bank should have a huge impact on the economy -- in fact, the farther in the future the bank promises to take action, the bigger the effect should be today. That’s just not what we see happening in real life. It also makes absolutely no sense -- if the Federal Reserve promises to raise interest rates 30 years in the future, it shouldn’t send the economy into a tailspin. But in Gabaix’s model, forward guidance loses the magical powers that it has in standard theories.

Finally, Gabaix’s idea seems to fit our experience with the zero interest rates. Nominal interest rates can never go very far below zero. In typical theories, this sends the economy nose-diving into a deflationary spiral. But in reality, economies that experience very low rates for many years -- first Japan, and now the U.S. -- seem to grow slowly and have low inflation, but not crash. In Gabaix’s model, that’s exactly what happens.

So what does Gabaix’s theory say we should do to boost the economy? Fiscal stimulus. When consumers are short-sighted, they will consume more of any windfall or tax cut they receive, which enhances the power of spending increases and tax cuts to increase growth.

Like all macroeconomic theories, Gabaix’s paper -- which isn't even in its final, published form -- should be regarded as a thought experiment. But it seems to solve a lot of the major puzzles that economists and policy makers argue about. And it does so with a remarkably simple tweak -- one that also achieves the longstanding dream of using behavioral economics to explain recessions and booms. I’m excited to see whether this paper manages to shake up the field of macroeconomics, which is desperate for new ideas.

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