Does anyone really understand inflation?
FOR ECONOMICAL the writing – a genre stylistically closer to computer textbooks than to literature – a discussion paper recently released on the Federal Reserve’s website is a blessed relief. Jeremy Rudd, a Fed researcher, includes quotes from William Butler Yeats and Dashiell Hammett. He uses phrases such as “cranky nonsense” and “senseless nonsense.” And, as if to channel David Foster Wallace, he amuses himself in his footnotes, notably one in which he casually observes that the mainstream economy can serve as “an excuse for a criminally oppressive, unsustainable and unjust social order.” “. No wonder his newspaper has become, by central bank standards, a social media sensation.
But it is the substance, not the style, of Mr. Rudd’s article that is most provocative. He directs his arguments towards an axiomatic idea in economics: that expectations determine inflation. The conventional story is simple. When workers expect prices to rise, they demand higher wages. When companies expect costs to increase, they set higher prices. Either way, inflation becomes a self-fulfilling prophecy. The task of central bankers is to set expectations low and stable. If they are successful, they can control inflation.
This idea also seems to have had remarkable success. Inflation in the rich world has been quiet for the past three decades. Every time he shot over the target, he quickly backed up. Expectations are, in the jargon, well established. Indeed, this is why many economists are optimistic about the current surge in inflation: Supply disruptions will eventually subside and price pressures will ease. It’s a heartwarming thought.
Enter Mr. Rudd, the author of over a dozen articles on inflation over the past two decades. The idea of inflation expectations “rests on extremely fragile foundations,” he writes. First, he says, the theory is wrong. Inflation models primarily include expectations as a short-term variable (i.e. what prices will be over the next month or two). To the extent that expectations matter, however, central bankers and analysts see them as a longer-term force, an underlying trend impervious to cyclical ups and downs. However, empirically this is difficult to document. And what expectations matter? There are ordinary people, businesses, forecasters and investors. None, he argues, are very good at predicting prices.
Nonetheless, it is true that the past three decades have seen both subdued inflation and low expectations by any measure. But Mr. Rudd’s claim is that causation has been poorly established. It is not that low expectations led to low inflation, but rather that the observed low inflation led to low expectations. As he notes, it wasn’t until after a recession in the early 1990s, when inflation fell sharply and then remained low, that expectations were lowered. Mr Rudd concludes that obsessing over inflation expectations is unnecessary and dangerous. Useless, because it is the observed prices that count. Dangerous, because central bankers could become unduly confident in their powers of mind control.
The reaction to Mr. Rudd’s provocation was fierce. Tyler Cowen, a prolific economist at George Mason University, cites the extreme example of hyperinflation as proof that expectations matter. When people think their currency will be worth a lot less tomorrow, they leave it. Ricardo Reis of the London School of Economics, who has studied inflation as deeply as Mr. Rudd, notes that no single variable – neither expectations nor money supply, unemployment or interest rates – is perfect for predict inflation. Yet everyone contributes to the image. Plus, he adds, there’s a lot of evidence regarding expectations. Studies show, for example, that companies that think costs will go up tend to set higher prices.
Despite all the heat of the debate, there may be some common ground. Adam Posen of the Peterson Institute for International Economics, a think tank, makes a sane distinction. In the long run, inflation expectations and, in particular, whether people think the central bank will suppress the price spike, are important. Mr Rudd concedes this implicitly, writing that it is better if inflation “is not on people’s radar screens.” In other words, it is good to expect that inflation is and will be a non-problem.
Yet “it’s different to say that fine differences in inflation expectations are either well measured or policy-relevant over shorter time horizons,” says Posen. There is little evidence that communication by central banks alone can control inflation without policy measures. And their credibility is due more to responses to crises than to attempts to manage expectations.
Janet Yellen made a similar distinction when she was Fed chairman. Long-term stable inflation expectations appear to be linked to long-term stable inflation, she said in a speech in 2015, in line with conventional wisdom. But she then said she was “somewhat skeptical” that central banks could influence expectations simply by announcing an inflation target. Instead, she said, expectations could only materialize after a central bank kept inflation close to its target, a process that could take years.
Inspect the unexpected
Some commentators have concluded from Mr. Rudd’s article that the economy is a mess and no one understands inflation. Yet, sifting through the arguments, there are shared ideas of deep importance. First, the credibility of the central bank is valuable. Second, inflation expectations are formed by experiences of inflation. And third, in the long run, such expectations probably matter.
What does this mean for the mounting inflationary pressures now facing much of the world? The hawkish point of view is that central banks need to curb prices before it’s too late and expectations lose their grip. Yet there is also a dove catch. The central banks of the rich world remain credible. By letting inflation climb a little higher now, they could help reset expectations. After all, before covid-19, the big worry in rich countries was too low inflation. Or, at least, that was what some economists and investors expected from the future, however high their expectations might be. ■
For a more expert analysis of the biggest stories in economics, business and markets, sign up for Money Talks, our weekly newsletter.
This article appeared in the Finance & Economics section of the print edition under the title “Grated Expectations”