Today’s show is about what I consider the biggest mystery of the U.S. economy. Last year, economic experts predicted a recession in 2023 with more confidence than they’ve predicted any recession in decades. We ended up with what some people are calling immaculate disinflation: an economy with low unemployment, falling inflation, rising real wages, and narrowing inequality. Harvard economist Jason Furman joins the show to talk about why so many economists were so wrong and what their wrongness teaches us about how the economy works.
If you have questions, observations, or ideas for future episodes, email us at PlainEnglish@Spotify.com.
In the following excerpt, Derek and Jason Furman evaluate why so many economists incorrectly predicted that a recession would occur in 2023 and discuss what actually happened.
Derek Thompson: Jason, economists are never really sure about anything. It’s always, “On the one hand, this; on the other hand, that,” but to the extent that they’re ever sure about anything, they seemed very, very certain that we would have a recession this year. We are not in a recession. We do not, at the moment, seem to be even close to a recession. What happened? Why did all of the experts get 2023 so wrong?
Jason Furman: Yeah, and just for context, forecasters basically never forecast a recession, and then sometimes they happen. This time, they forecast a recession, and one didn’t happen. In terms of my own views, my recession probabilities were lower than a lot of other forecasters’ but way higher than they would normally be. I too am surprised by the economy. Part of it is that there are just these massive movements. The fiscal policy we did was huge, and it has long and variable lags helping the economy. And then we did some more helpings of it with things like infrastructure, [the Inflation Reduction Act], and [the CHIPS Act].
The monetary policy move was also huge, but interestingly, and I think this might be in part of the story, rates rose really, really fast, but they didn’t actually rise that high. If you look at long-term interest rates, they’ve spent most of the last year and a half around 3.5 percent. They’ve gotten a bit higher lately. That just isn’t a high interest rate by historical standards.
And then there was the unsnarling of supply chains. Then you throw in the Russia shock. That was a huge shock to energy prices, but it actually went away relatively quickly. Another part of the forecast error was people thought oil might go to $150, $175 a barrel. Instead, it went all the way back to or even below where it was prior to COVID or anything like this. I think there are a number of different things, but for me, the most important, our fiscal policy continued to prop the economy up. Monetary policy just wasn’t that tight, even if it got a lot tighter. And third, the oil shock went away.
Thompson: So we’ve got the fiscal policy, the monetary policy, the unsnarling of the supply chains, and then the energy shock from the Russia invasion. It reminds me that Michael Cembalest at J.P. Morgan called what’s happening in the U.S. economy in 2023 the “Rasputin effect”—Rasputin, the Russian mystic who was shot, poisoned, attempted to be drowned and somehow survived just an astonishing number of threats to his life before he was eventually murdered. That’s the U.S. right now, just pre-Rasputin murder. It’s surviving the proverbial shot and poisoning and attempted drowning. I want to go back to the first question though, because they were really, really smart economists, like Larry Summers and all these macroeconomists, who accurately predicted that the Biden stimulus would lead to inflation, who then went on to predict rather confidently that the only way to cure that inflation would be for the U.S. economy to have a recession. Again, we’re not anywhere close to a recession. What does it mean for the macroeconomic field that a lot of people seem to have got it wrong? How are you not surprised here?
Furman: I’m a little bit surprised but not shocked. First of all, I’ve at various points put out my sense of what underlying inflation is. The highest number I’ve ever had at any part in the whole inflationary cycle was 4.5 percent. Even when actual inflation was way above that, my view was a bunch of that was transitory. Things like energy prices going up, but the energy prices were going down. In some sense, all of the people who thought inflation was going to be quite stubborn agreed that some of it was transitory. The debate was over how much was transitory. Right now, my reading of the inflation data is that underlying inflation is about 3.5 percent, so it’s drifted down by 1 percentage point, as opposed to headline inflation, which has drifted down, depending on which way you measure it, more like 4 or 5 percentage points. In that sense, the surprise is a little bit smaller.
The second is I think the evidence—and Paul Krugman has this same view—is that what you’d call the Phillips curve, the relationship between labor market tightness and inflation, is nonlinear. That small changes in unemployment or other measures of labor market tightness—and it’s going to be important, I’m going to get into those other measures—can lead to big changes in inflation. But then as the labor market gets looser, it starts having a smaller change on inflation, and that’s part of what explains inflation on the way up and part of what explains it on the way down.
In fact, in September of last year, I put out an analysis based on someone else’s model, by using their model, and I listed a set of unemployment rates and inflation rates, and I think I said unemployment needed to be about 4 [percent] to get inflation to 3.5 [percent]. We’re a touch below that, but not in any dramatically shocking way.
Now, what we have seen, and this I fully expected and am not at all surprised by, is some immaculate cooling of the labor market. And the form that immaculate cooling has taken is not the unemployment rate rising, but job openings falling. And for me, if I had just one variable to assess the tightness of the labor market, it would be the number of job openings for every unemployed worker. That peaked at 2.0 last year. It’s fallen down to 1.6, and the entire fall has been with openings declining rather than the unemployment rate rising.
And a year ago, I said I thought we’d get two-thirds of the way back to our normal relationship for job openings to unemployed. I think in summary, the labor market has cooled, job growth has slowed, hours have fallen, job openings have fallen, the number of job openings per unemployed has fallen. A lot of different measures of the labor market have cooled, and inflation has also cooled probably a little bit more than I would’ve thought it would cool, but not dramatically more.
Now, if inflation falls to 2.5 percent with the unemployment rate at 3.5 [percent], then I would be very surprised, and it would cause me to revise my views. But I just don’t think we’re there yet.
This excerpt was edited for clarity. Listen to the rest of the episode here and follow the Plain English feed on Spotify.
Host: Derek Thompson
Guest: Jason Furman
Producer: Devon Manze
Subscribe: Spotify
