Behind The Curve When the yield curve inverts, people worry that it's a sign we're headed for recession. But its predictive reliability also depends on the way it inverts.
NPR logo

Behind The Curve

  • Download
  • <iframe src="https://www.npr.org/player/embed/673425427/673430050" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Behind The Curve

CARDIFF GARCIA, HOST:

Hey, everyone. Welcome to THE INDICATOR FROM PLANET MONEY. Cardiff and Stacey here. The topic of today's episode is by popular request.

STACEY VANEK SMITH, HOST:

Cardiff, you have only yourself to blame in the topic for the show today.

GARCIA: Or to give credit to.

VANEK SMITH: You could also put it like that. Your brand is now inextricably linked to this topic, which is...

GARCIA: The yield curve.

VANEK SMITH: One part of the yield curve has inverted. Let that sink in (laughter).

GARCIA: (Laughter).

VANEK SMITH: So a lot of people got in touch to ask, isn't this supposed to be a bad thing? What does this mean for the U.S. economy? Does this mean we're headed into a recession? Like, inverted yield curve equals sirens?

GARCIA: Yeah. Let's take a step back for a second, though. The yield curve shows the interest rates for all the different government bonds, or just Treasuries, because these Treasuries have different expiration dates. Like, there's a Treasury that expires in three months, one that expires in two years, one that expires in 10 years.

VANEK SMITH: So roughly speaking, when longer-term Treasuries have higher interest rates than short-term Treasuries, that's good. You can expect the economy to keep growing. For example, if the 10-year Treasury note is paying a higher interest rate than the three-month Treasury.

GARCIA: But when longer-term interest rates fall below short-term interest rates, it might suggest that the economy is headed for a recession. And in fact, ever since the 1960s, every time the yield curve has inverted, a recession has followed. It's a remarkable track record. And guess what? Yesterday, a part of the yield curve inverted.

VANEK SMITH: Dun-dun-dun (ph).

GARCIA: Yeah. The interest rate on five-year Treasuries fell below the interest rate on three-year Treasuries. Does that in fact mean we're headed for a recession?

VANEK SMITH: So to explain exactly what happened yesterday and what it means, we called the Duke finance professor who first discovered that the yield curve could help us predict recessions. His name is Campbell Harvey.

GARCIA: Today on the show, what you should take away from yesterday's yield curve inversion and, just as crucially, what you should not take away.

(SOUNDBITE OF MUSIC)

GARCIA: Campbell Harvey, welcome back to the show.

CAMPBELL HARVEY: Great to be back.

GARCIA: Remind us - what are the specific parts of the yield curve that we should be looking at when it comes to predicting a recession?

HARVEY: So my research focused on the five-year yield minus a very short-term yield.

GARCIA: In other words, the interest rate on five-year Treasuries versus the interest rate on very short-term Treasuries, like three months?

HARVEY: Exactly. So my research focused on the interest rate or yield on five-year Treasury bonds and relative to three-month Treasury bills, something very short-term. So there's a long-term component, a short-term component, and the difference between the long and the short is called the yield curve.

GARCIA: OK. And so supposedly what happens is that when the three-month interest rate, the three-month yield, is higher than the five-year interest rate, that suggests that a recession is coming. But that's not all, right?

HARVEY: The idea is historically it's a very unusual situation when the short-term interest rate is greater than the long-term interest rate. And that, at least since the 1960s, has reliably predicted a recession, let's say a year, a year and - ahead in terms of what happens to GDP growth. So it has been a reliable indicator.

GARCIA: OK. But it's not just enough for, very temporarily, a longer-term interest rate to fall below a shorter-term interest rate. Like, if it happens over the course of a few days, that still doesn't say anything. Right?

HARVEY: That is correct. So it's obviously not good news in terms of for the overall situation. But in my dissertation, I was looking for a situation where we have an inversion where the short rate is higher than the long rate for a full quarter. So GDP is measured over a quarter, so it made sense to take a look at the relative bond yields over a quarter - not a day or a few hours, but a full three months.

GARCIA: OK. So I want to now discuss specifically what happened yesterday. The interest rate on the five-year Treasury fell to 2.83 percent. The interest rate on the three-year Treasury was slightly higher at 2.84 percent, so that is an inverted part of the yield curve. And so we got a lot of email and people wondering, is this meaningful? What does this say about the likelihood of a recession? What do you tell them in light of what you just said of what we should be looking for?

HARVEY: I got a pile of emails...

GARCIA: (Laughter).

HARVEY: ...Congratulating me, saying, happy yield curve inversion day.

GARCIA: Welcome to my world.

HARVEY: (Laughter) So this was - again, this is a piece of the yield curve. And historically, there's been pieces of the yield curve that got misaligned temporarily. I really am hesitant to put too much weight on these little pieces of the curve. I think it's essential to look at a long-term rate relative to a very short-term rate. So I don't want to read too much into the three-year and the five-year yield, but it is an issue that the five-year relative to the three-month is very flat. So the direction, the momentum that we're moving in suggests that there is a likelihood of a full-blown inversion between long and very short.

GARCIA: OK. Actually want to just pause to explain what you just said for our listeners and then get you to elaborate on that a little bit. When you say that there are parts of the yield curve that are in fact becoming flatter, what you're saying is that short-term interest rates have continued to go up recently, and long-term interest rates have been falling, especially in recent days. In other words, there are other parts of the yield curve that are getting closer to inverting where short-term interest rates become higher than long-term interest rates. And so a flatter yield curve, what should we glean from that? What information should we glean from a flattening yield curve?

HARVEY: In my dissertation, I look at the behavior of the entire curve, so it's not just inversions. The inversions are associated with recessions. It's also the case that a flatter yield curve is associated with slower economic growth. It might not be a recession, but it's associated with slower economic growth. And that is worrisome. It is not an alarm bell yet for a recession. However, if it continues to go in this direction, it might well be.

GARCIA: OK. So last question. Listeners know how much I love following the yield curve. I think it captures so much interesting information, not just about the U.S. economy, but also about the U.S. economy's position, like, relative to the rest of the world. But I got to say, I also get kind of nervous about, like, monomaniacally focusing on one indicator for anything, but especially for something as massive and complex as where is the U.S. economy going in the next couple of years. What do you think about this idea that maybe we should be looking at a multitude of things and not just one indicator?

HARVEY: It's a very good idea. Why would you look at just one piece of information? It's true that the yield curve has been very reliable. It hasn't had any false signals since the mid-1960s. So it's been extremely reliable, even predicting the global financial crisis. However, it's just one piece of information, and it makes sense to look at other things.

Look at, for example, risky corporate debt relative to Treasuries. Look at other leading indicators. Look at the level of volatility in the stock market. There are many different things that you could look at to form a view of the probability of a recession. To look at one piece of information essentially assumes that it's the only relevant information for forecasting the economy, and that doesn't make any sense.

GARCIA: Yeah. I guess the other thing too is that just because one thing has a remarkable track record, like the yield curve really does, doesn't mean that that'll be the case for ever. The world could change. And at some point, it might stop being as reliable as it has been in the last half-century or so.

HARVEY: That's correct. So it has been very reliable for the past 50 years. There's no guarantee that it will continue to be reliable in the future.

(SOUNDBITE OF MUSIC)

Copyright © 2018 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.