Can the yield curve still predict recessions?

The 'Fearless Girl' statue stands in front of the New York Stock Exchange.

Angela Weiss / Getty Images

Two years ago, the yield curve inverted. That means short-term interest rates on Treasury bonds were unusually higher than long-term interest rates. When that's happened in the past, a recession has come. In fact, the inverted yield curve has predicted every recession since 1969 ... until now. Today, are we saying goodbye to the inverted yield curve's flawless record?

Related episodes:
The inverted yield curve is screaming RECESSION (Apple / Spotify)
Yield curve jitters
Two Yield Curve Indicators

For sponsor-free episodes of The Indicator from Planet Money, subscribe to Planet Money+ via Apple Podcasts or at plus.npr.org.

Music by Drop Electric. Find us: TikTok, Instagram, Facebook, Newsletter.

Transcript

SYLVIE DOUGLIS, BYLINE: NPR.

(SOUNDBITE OF DROP ELECTRIC SONG, "WAKING UP TO THE FIRE")

DARIAN WOODS, HOST:

Predicting the economy has often been compared to the finance version of astrology or tarot. On news channels, you've got the high priests of finance brought in to prognosticate and speculate and bloviate.

ADRIAN MA, HOST:

And pontificate.

WOODS: And extrapolate.

MA: And gesticulate.

WOODS: And they're often wrong.

MA: But you know, one high priest has been right.

CAMPBELL HARVEY: My name is Campbell Harvey. I'm professor of finance at Duke University.

MA: And Campbell's instrument for divination, the inverted yield curve.

WOODS: We've been covering the inverted yield curve for years on THE INDICATOR. Former co-host Cardiff Garcia was particularly enamored.

(SOUNDBITE OF ARCHIVED NPR BROADCAST)

CARDIFF GARCIA: I am kind of obsessed with the yield curve. I find it fascinating. I find it mysterious. It is one of my favorite indicators.

MA: An inverted yield curve just means that short-term interest rates are higher than long-term interest rates. And historically, that's meant economic pain is coming. Campbell Harvey discovered this powerful indicator decades ago, and it's predicted every recession since 1969.

HARVEY: Importantly, it also does not have a false signal.

WOODS: Until possibly now.

(SOUNDBITE OF MUSIC)

WOODS: The yield curve inverted almost two years ago. We were on the lookout for a possible crash in the economy, but still, no recession has emerged.

MA: And that is obviously great for America, and it also raises the question of whether the yield curve is still a useful prediction tool. There are not a lot of great ways to see if a recession is coming at us, so if the yield curve is truly dead, then our economic alert system just got a whole lot murkier. This is THE INDICATOR FROM PLANET MONEY. I'm Adrian Ma.

WOODS: And I'm Darian Woods. Today on the show, rest in peace, inverted yield curve, question mark.

(SOUNDBITE OF MUSIC)

WOODS: The pioneer of the inverted yield curve, Campbell Harvey, explains that yields basically just mean interest rates. And the yield curve represents the set of different interest rates for locking your money away for different periods of time.

HARVEY: The usual situation is that if you lock your money up for a longer period of time, you get rewarded for that with a higher interest rate, and that's the way it is usually in normal situations. But every so often, it goes the other way, and the short rates increase, and we call this an inverted yield curve.

WOODS: The yield curve we're most interested in is the difference between three-month and 10-year Treasury bonds. In normal times, three-month Treasurys offer a lower rate of interest than the ones locking up the investment for 10 years. But about two years ago, that flipped. It inverted.

MA: The key indicator Campbell looks for is, does the yield curve stay inverted for a full quarter? And yes, that happened last year.

WOODS: And so, going by history, this should have meant a recession was around the corner. The longest gap between the first inversion and a recession was in the lead-up to the Great Recession of 2008, 2009. That was a 23-month lead-up period. We recently passed that milestone.

HARVEY: Yeah, right now it's officially at the maximum. This is something that we haven't seen historically, and we don't have a recession.

MA: And this is worth repeating. Even though some people in some parts of the economy might be feeling the pinch, the U.S. economy is not in a recession.

WOODS: And the way that economists assess whether we are in a recession is by using the three Ds - depth, duration and diffusion. Depth, as in it has to be fairly bad - incomes crashing, factories shutting, stores selling a lot less stuff. And then the duration means the problems have to last for a while. And finally, diffusion means that several sectors should be affected, not just one - say, for example, a crash in tech that may not affect the wider economy. And looking now at jobs growth, at industrial output, at consumer spending, it's just hard to see any evidence of a recession.

HARVEY: I strongly agree that we're not in a recession right now.

WOODS: So is this the end of the yield curve as a recession indicator?

HARVEY: Recession, if it occurs - it might be different in that the lead time is longer, so not 23 months but 27 months. But at some point, I would have to declare, if there's no slowdown in economic growth, a false signal. At this point, it seems premature to do that.

WOODS: It's understandable that you don't want to declare the death of the inverted yield curve just because it's a little bit longer than the last longest time, but at some point, you do wonder whether it's just so long that of course there are going to be recessions at some point, of course, because economies go into recession. At what point do we think that the indicator isn't as useful as it once appeared to be?

HARVEY: It's a matter of looking at this scientifically. So you've got an indicator that has been accurate, 8 out of 8, no false signals. The lead time to a recession is variable, and the lead time varies from six months, historically, to 23 months. And it would just be unscientific for me to say, oh, well, we've reached 23 or 24, therefore this indicator isn't useful anymore.

MA: The inverted yield curve does have solid economic theory behind it. When the economy is likely to have a lot of growth in the future, investors want higher interest rates on their U.S. Treasury bonds. When there's concern there's a recession coming, they don't demand as high interest rates for that predicted future.

WOODS: Trillions of dollars are at stake in the Treasury market, so people are really motivated to find the best forecast. The yield curve sums up all of those concerns like this gigantic financial brain.

MA: But according to Campbell, this giant brain might not be as accurate as it once was. I mean, the very fact that people like us at THE INDICATOR are talking about the yield curve so much might change its predictive power.

HARVEY: So think about that change of behavior where people more cautious, companies not willing to make large investments because the yield curve's inverted, and it might put their company at risk if a recession actually happened.

MA: Think about early last year, when it seemed like every tech company was downsizing.

HARVEY: People see that the yield curve is inverted. They change their behavior. This decreases investment. It decreases economic growth. But it's also risk management, and it decreases the chance that we actually go into a serious recession.

WOODS: Business managers maybe didn't get caught up in as much hype this time around, which means less overinvestment. In short, if people didn't jump on that rocket ship in the first place, there's less far to fall. And if true, Campbell says this is a good thing.

HARVEY: I also want to make it very clear that I hope the indicator is providing a false signal. Nobody wants a recession. A recession is very damaging. The impact on families - it is something that nobody actually wants.

MA: And Campbell also adds we should never rely on just one indicator.

HARVEY: The inverted yield curve is just one piece of this giant puzzle of the U.S. economy, and it's naive to think that that is sufficient information to forecast what's going to happen in the U.S. economy. So obviously, the yield curve will give a false signal. Maybe it's this time. Maybe it's next time. But it's inevitable that there will be false signals from a simple model. So yes, you need to look at other data in the economy.

WOODS: So whether it's jobs numbers, consumer confidence or even vibes on the street, we will be keeping our microphones out, listening for changes in the winds of the swirling U.S. economy.

(SOUNDBITE OF MUSIC)

MA: And in the meantime, it might be premature to declare the yield curve dead, but we could say it's sort of in a - taking a nap right now.

WOODS: I would say a coma, and we're not sure if it can be revived.

MA: Fair enough.

WOODS: We're monitoring the vital signs very carefully.

MA: This episode was produced by Cooper Katz McKim with engineering by Cena Loffredo. It was fact-checked by Sierra Juarez. Kate Concannon edits the show, and THE INDICATOR is a production of NPR. Transcript provided by NPR, Copyright NPR.