The Recession Predictor, Still Predictive? : The Indicator from Planet Money Every time the yield curve has inverted since 1970, the economy has fallen into recession. It's getting close to inverting now, but it may no longer be the recession predictor it once was.

The Recession Predictor, Still Predictive?

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


Hey, everybody, you are listening to THE INDICATOR, where every day we tell you a short story about the economy. And today it is basically Cardiff's birthday. Happy birthday-ish, Cardiff.


Yes. Right. Yeah.

VANEK SMITH: It's not really your birthday. But...

GARCIA: No, it's not. But...

VANEK SMITH: ...Basically could be because we are doing a mid-year update about your favorite indicator, the yield curve.

GARCIA: Yes, the yield curve. We first - (laughter).

VANEK SMITH: It's basically like your birthday but better.

GARCIA: We first talked about the yield curve in January. Listeners of that episode might remember that when the yield curve does something very specific, when it inverts - don't worry; we'll explain that again - when the yield curve inverts, it tends to mean that a recession will start within roughly the next year. And back in January, the yield curve had not yet inverted, but it was getting closer and closer. I'm Cardiff Garcia.

VANEK SMITH: And I'm Stacey Vanek Smith. Today on the show, what has the yield curve done in the last six months since we talked about it?

GARCIA: Plus, Stacey's going to give me a hard time about why the yield curve may no longer be the useful recession predictor that it once was.

VANEK SMITH: I mean, you had it coming.


GARCIA: OK, a quick explainer - or a re-explainer, I guess - on the yield curve. There are different kinds of U.S. government debt or just Treasurys, as they're known. So for example, there is a 10-year Treasury. And if you buy a 10-year Treasury, it means that you basically lent the government money. And then 10 years later, the government pays you back the money, plus it pays you interest according to whatever interest rate it had when you bought the Treasury.

VANEK SMITH: And the yield curve shows all the interest rates for Treasurys with different maturities. So for example, there is a three-month Treasury, a two-year Treasury, a 10-year Treasury. And when the yield curve is normal, when the economy is doing well, in a good place, the yield curve slopes up. And what that means is that long-term interest rates are higher than short-term interest rates.

GARCIA: And there are basically two reasons that long-term interest rates are normally higher than short-term rates. The first reason is that people want to be compensated for locking up their money for a longer period of time. And this makes sense. Like, for example, we know that inflation makes money less valuable over time because the prices of things are going up. So if you're going to lock up your money for a while, you're also going to demand a higher interest rate to make sure that your money keeps up with inflation.

VANEK SMITH: Yeah, that's a lot of delayed gratification - got to get paid for that. The second reason that long-term interest rates are normally higher than short-term interest rates is that people expect the economy to keep growing. So they're not as interested in buying long-term Treasurys. Instead, they might invest their money in the economy itself, like maybe the stock market or a business or something because they think that those things will keep growing. So the government has to pay higher interest rates on those longer-term Treasurys to attract buyers.

GARCIA: Yeah. And conversely, when people are worried about the economy, the yield curve starts to flatten, which...

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

GARCIA: ...Means that the difference between short-term interest rates and those higher longer-term interest rates starts to shrink. And sometimes the yield curve even inverts and actually starts to slope down.

VANEK SMITH: Ree-ree-ree-ree (ph).

GARCIA: Yeah, exactly.

VANEK SMITH: Right? That's the...

GARCIA: Psycho music, please.

VANEK SMITH: (Laughter).

GARCIA: That is when long-term interest rates fall so much that they are lower than short-term interest rates.

VANEK SMITH: And that is where the real trouble happens. That is where you should really start paying attention. In the past, when the yield curve has inverted, it has meant that a recession was coming soon. It's like this - like a canary-in-the-coal-mine situation. In fact, every single time the yield curve has inverted since 1970, the U.S. economy has fallen into a recession within about a year. And that is why so many people pay attention to the yield curve when it looks like it might invert. And that is why Cardiff has been paying so much attention to the yield curve.

GARCIA: While mysteriously, Stacey's just been ignoring it.

VANEK SMITH: You know, there's a lot of good stuff on Netflix right now, Cardiff. OK. So now that we have re-explained the yield curve, Cardiff, I know you are dying to tell us what it has been doing since we last talked about it in January. And, you know...

GARCIA: Damn right.

VANEK SMITH: ...Has it inverted?

GARCIA: Here's the deal.

VANEK SMITH: Is it psycho time?

GARCIA: Back then, when we last talked about it, the 10-year Treasury rate was about six-tenths of a percentage point higher than the two-year Treasury rate. Right now it's only about three-tenths higher. So in other words, the yield curve...

VANEK SMITH: Like, and?

GARCIA: ...Has continued to flatten, but it has not yet inverted. It's just gotten a lot closer to inverting, close enough that policymakers at the Federal Reserve and also just economists everywhere have been debating just how much it matters that the yield curve is so flat. And this is a debate that I got to say has been going on for a little while, but it has really intensified lately. And it centers on one question - is the yield curve as useful for predicting a recession as it used to be?

VANEK SMITH: So it actually might not be, and here's why.

GARCIA: Damn it.

VANEK SMITH: (Laughter) So remember that there are two reasons that historically long-term interest rates are higher than short-term interest rates during the good times, why the yield curve is not inverted. And one of those reasons is that people demand to be compensated for locking up their money - the delayed gratification thing. Economists call this term premium because they have to have a really boring term for everything.

GARCIA: (Laughter).

VANEK SMITH: And the idea is that you should get paid more for buying something that is long-term. And some economists who try to estimate the term premium believe it has gone away, or at least that it's gotten smaller. And these economists say that people who buy Treasurys do not care as much as they used to about being compensated for locking up their money for the long term.

Now, this might be because people who buy Treasurys are no longer as worried that inflation will erode away the value of their money while it is locked up in the Treasurys, so they don't demand higher interest rates to make up for it. Whatever the case, the theory is that Treasury buyers just less nervous about the whole locking up their money for a long time thing.

GARCIA: And if that's true, then you would expect the yield curve to be flatter than it used to be just because this term premium has gone away but not because the economy is expected to slow down, which is what we've been worried about here. So again, if these economists are right, then the yield curve has flattened for reasons that don't suggest a recession is getting closer. So...


GARCIA: ...Where does that all leave us? Should we ditch the yield curve?

VANEK SMITH: In flat land?



GARCIA: ...Is the right response (laughter). For one thing, economists still disagree on whether the term premium has gone away. We just don't know yet. And in part that's because it cannot be directly observed. The term premium can only be estimated using economic models. And that's important work, but those models are not the last word even among economists. Some things just can't be settled definitively.

VANEK SMITH: So in the end, we are still left with an indicator that has had an amazing track record for at least the last half century. It might be a crude indicator now. It might be less useful than it was before. But, you know, we don't actually know that for sure. And, you know, Cardiff, it is important to remember that everything has a peak. You know, it's like a fine wine, right? It will get better and better and better and better, and then at a certain point it starts to get worse. And that might be just a thing to think about.

GARCIA: You're saying that that might be the yield curve now. It might have peaked.

VANEK SMITH: I mean, I'm not...

GARCIA: You worry. You worry.


GARCIA: You suspect.

VANEK SMITH: I suspect.

GARCIA: Yes. And I think that you actually might be right. But given its track record...


GARCIA: ...OK? - we also know that it's dangerous to assume that this time is different. And that doesn't mean that this time is the same. But we should have learned enough lessons over the last couple of decades, especially when it comes to the financial crisis, that when there's, like, this flashing light going off...


GARCIA: ...You should not ignore it.


GARCIA: And all I'm saying is that if the yield curve inverts and starts to send out that warning signal, it would be a mistake to ignore it.


VANEK SMITH: If you were able to sit through that whole discussion about the yield curve, and if you are currently in school or a recent graduate, you should apply to be our intern. We are currently taking applications for our fall internship. You can go to to find the application. It is a paid internship. And you just basically get to hang out and talk with us about the yield curve all day. The deadline is July 15.


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

NPR transcripts are created on a rush deadline by an NPR contractor. 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.