The Compound and Friends - The Yield Curve Indicator (with Josh, Michael, and Campbell Harvey)

Episode Date: October 7, 2019

Campbell Harvey is a Professor at Duke University and a partner as Research Affiliates. He sat down with Michael Batnick and Josh Brown of Ritholtz Wealth Management to discuss the meaning of the inverted yield curve indicator, which he discovered in 1986 while working on a dissertation. Campbell cites the fact that 7 out of the last 7 recessions had been presaged by a yield curve inversion - which is what happens when it longer term bond yields fall below shorter term bond yields in the Treasury market. He believes that this phenomenon occurs when the market participants begin to grow more pessimistic about the economic outlook. The behavior of executives, lenders, borrowers and investors can change enough during these times to actually become a self-fulfilling prophecy - producing a negative feedback loop that drives a weakening economy into a full-blown recession. Recessions are a normal part of the business cycle, although they can be painful to live and invest through. They can also vary greatly be degree. Campbell fields questions from Michael and Josh about all of the ways in which this time might be different. He acknowledges that it is always possible that the yield curve indicator might stop working as a recession signal, but he believes that because it hasn't yet, this might be a good time for people and corporations to rethink the risks they're taking. 1-click play or subscribe on your favorite podcast app   Subscribe to the mini podcast on iTunes or Spotify    Enable our Alexa skill here - "Alexa, play the Compound show!"   Talk to us about your portfolio or financial plan here:  http://ritholtzwealth.com/   Obviously nothing on this channel should be considered as personalized financial advice just for you or a solicitation to buy or sell any securities. Please see this 3,000 word terms & conditions disclaimer: https://thereformedbroker.com/terms-and-conditions/ Hosted on Acast. See acast.com/privacy for more information. Learn more about your ad choices. Visit megaphone.fm/adchoices

Hey, I'm Josh Brown. I'm here with Michael Batnick and our special guest, Campbell Harvey. Campbell Harvey is the godfather of the yield curve indicator. There's a lot of talk about the yield curve. It's been inverted for a few weeks now. People are concerned that recession is coming. We literally are going to get it from the horse's mouth. Everything you need to know about how the yield curve might affect the economy, the stock market, your investments, your savings, your career, possibly even your life. Stick around. Okay. Campbell, was that like an over-the-top buildup or would you say the yield curve is pretty important right now? It is very important. I think way more important than in the past where people didn't notice. So it is inverted

before the last seven recessions. And it hasn't rendered a false signal. So right now, people are looking at it very seriously. The track record's impressive. It's flashing code red since June the 30th. So I want to back up because this is primarily going to be seen on YouTube and there'll be a lot of people watching who have heard the term yield curve, but I think an explanation is in order just so that everyone understands exactly what we mean by yield curve before we get into the ramifications of inversion. So how would you explain that to people? Sure. So the idea, the yield curve has to do with the difference in interest rates at different maturities.

So you go into a bank and invest in a certificate of deposit. You notice that the rate for, let's say, five years is a lot higher than the rate for 90 days. So that's what we kind of called a positively sloped yield curve. So you're getting paid more interest for putting your money away for longer than for shorter. Exactly. That's a normal situation. And the situation we're in today, and I look at not CDs, but government treasury bonds, that you've got this weird situation where the interest rate on the 10-year treasury bond is lower than the 90-day treasury bill. And that's called an inverted

yield curve. It's not normal, and it is an omen of really bad news. So in 1986, you did a dissertation where you found that in the four recessions from the 60s through the 80s, before each recession started, the yield curve had inverted. And so you said this is a sign of a lack of positive sentiment or the onset of negative sentiment. If people are willing to continue to buy treasuries at lower yields, it signifies that people are less confident. And therefore, a negative economic event is probably not far behind. Am I explaining that right? Yeah, the intuition is really straightforward. is probably not far behind.

Am I explaining that right? Yeah, the intuition is really straightforward. Indeed, the economic theory behind this indicator is quite straightforward, not controversial at all. That rates have an expectation of real growth embedded in them. And yeah, in 1986, my dissertation, I presented to my committee. They were pretty skeptical, frankly,

because I've got four observations. And there was a known skeptic on that committee. What was Fama's reaction? Fama's reaction was interesting. He was most impressed that this simple indicator outperformed very complex econometric services that were available at that day. And for the price of the Wall Street Journal, you got the forecast for my model. So the economic theory helped, and they passed me. And usually what happens after you publish your dissertation, there's two possibilities. One, the effect gets weaker. And number two,

it goes away completely. So you might have got lucky, four for four. But in my case, in the next three recessions, the yield curve inverted before each one. So you have a seven-time track record of yield curve inversion followed by recession. Not immediate, but it's worth each time. No, no. It's actually good. It's not immediate. It's a leading indicator. It gives us like six months to 18- month lead time. So when you discovered this, there were three previous instances when this happened. Four. Okay. And now there's been three additional ones.

That's correct. And you're 100%. Out of sample. So Mandelbrot said something like the trend has vanished, killed by its discovery. Why did that not happen with the yield curve indicator? managed, killed by its discovery. Why did that not happen with the yield curve indicator? So again, it didn't get that much publicity until after the global financial crisis. So it's possible, given the spike in the popularity of the indicator, that it takes on a different role. So today, it actually might have a feedback mechanism. So people

actually see the inversion. It becomes causal. Yeah, exactly. The so-called self-fulfilling prophecy. So can you get back to just the theory? Why does this lead to lower growth? Yeah, okay. So let me give you the simplest possible view of this.

And that is that when risk increases, when economic growth expectations decrease, the so-called insurance trade, when risk is on, is the safest asset in the world. And that's a U.S. Treasury bond. And the 10-year bond is the bellwether bond. People pile into that, the price goes up, the yield goes down, and you get this weird situation where the long-term yield is higher than the short-term yield. So that's one very simple way to look at it. It's not normal. It is something that foreshadows bad times. One of the theories that people float is that because banks borrow short and lend long, that this is sort of choking off credit in the economy. Is there any truth to that?

Yeah. So that story was much more effective in 1970s, maybe 1960s. So banks are pretty careful at hedging the bigger banks. Maybe some of the smaller banks have this exposure where they really like a positively sloped yield curve. I think it's much more complex than that today. But of course, when they're lending to companies, they're not following the yield curve. There is all sorts of credit risk involved. So it's probably not that simple. It's not that simple.

And again, the banks have instruments today to hedge out that duration exposure. Someone asked you over the summer, just as the yield curve had gone through its first full quarter of inversion. And by the way, that's the trigger, right? For my model, I look at quarters because my model is about forecasting economic growth and GDP is measured quarterly. So I look at a quarter. Okay. So as of June 30th, that was the first full quarter during which the yield curve had been inverted. Is that right? That is correct. And when we say the yield curve being inverted, you're referring specifically to the two-year

treasury and the 10-year treasury? No. So actually in my dissertation, I look at the 10-year minus the treasury bill, which is three months, which kind of makes sense. Three months is the quarter. And I also looked at the five years, both inverted. Is that semantics or does it actually matter? So yeah, because these are highly correlated. So the five-year and the 10-year are highly correlated. But when you start going to different pieces of the curve, like the Fed likes the 10-year minus the two-year.

It didn't invert June 30th but that indicator came way past my dissertation. It doesn't have like the track record as a sample. Okay so somebody asked speaking of being causal it's a leading indicator but then also it's like a yellow light because it signals to CFOs or people that make decisions on large amounts of spending. It signals that, okay, now is the time to pull the horns in. So that's where the self-fulfilling prophecy part is really interesting to me. If enough people are aware of it, is it now a fait accompli where we're just going to have a recession because everyone knows it's inverted?

There's a totally different way to look at it. That is not inconsistent with what you're saying. So think of the CFO. They're pulling a trigger on a major spend. They need to borrow for it. They're going to build a plant. Yeah. So the yield curve is inverted.

And they say, well, let's wait. So they don't do that spending. Why do they say let's wait? Because the possibility of recession, going to recession, they are caught in a situation where they could be insolvent. So if they ignore the yield curve, they could basically go out of business. So it makes sense to delay. And you're right that that delay could lead to slower growth. But it's possible to look at this as risk management. So you cut back your spending somewhat, but you avoid that very sharp hard landing.

So it's possible that this self-fulfilling prophecy injects a different level of risk management and indeed we could avoid a recession, just have slower growth. So, okay, so following that train of thought, if more people were aware of your work on the yield curve in 1998, 1999. So we had a yield curve inversion in the 2000 year. So right now people talk about a buyback bubble.

But back then you had the opposite. You had a CapEx bubble. Corporations could not stop spending. Now part of that was because of Y2K. They thought they had to update all their machines and all their software. And then part of that was you had 4% economic growth and everything looked great and people felt really confident because of stock prices. But whatever. Corporations did not heed the yield curve and they spent and spent through the year 2000.

And then in 2001, we're mired in a pretty bad recession if you think that if more people had been aware of um what you were talking about then possibly enough cfos would have pulled back on all their spending that it wouldn't have been quite as bad as it ended up being so actually i do believe that And I think like a sharper example is the global financial crisis. Okay. So we had plenty of lead time on the inversion. That crisis- When did it invert?

July 2006. And when did the recession officially start? December 17? Yeah, it was like 20, 21 months later. So it was a very long sort of warning. And that recession was unexpected. So people were caught totally surprised. I think this time is different in that there's plenty of information out there that suggests we're slowing.

And it's not just the yield curve. There are other forward-looking indicators that are suggesting slower growth. So this is a single variable indicator. Have you tried to add other indicators to maybe get the timing of the stock market right? What have you done with the model aside from just the yield curve indicator? Yeah. So one thing I make very clear, this is a single indicator. It's got a great track record, but it's foolish just to look at one indicator. You need to look at a spectrum of different things. And the sort of indicators that I look looking at are forward looking like the yield curve.

At Duke, you guys follow a CFO sentiment. Can you talk a little bit about that? Yeah. So for, you know, over 20 years, we've been surveying CFOs around the world, including maybe 400 US CFOs. And they're basically telling us that 50% of those CFOs expect a recession to start in 2020. And it goes to 85% if you spill over into the first part of 21. What's the average number ordinarily? Like it's 50% higher? Oh, yeah. This is way out of the park in terms of... When is the last time that many CFOs expected recession statistically and we didn't have one? Did we have a false signal like that in 15-16?

Yeah, so it's really interesting. We've done this analysis of the survey. It's a leading indicator of the leading indicators. And the most striking sort of number from the survey was that the CapEx spending that's expected is the second lowest since December 2009. So that's the sort of scenario we're looking at, that it is bad in terms of the history. Sorry, the CapEx year-over-year growth yeah the total dollar spending growth but don't you think that there should be an asterisk because the year before we got this massive tax stimulus and now you're lapping those really great numbers and it's harder to do that and so if you look at the at the time series this is not just a one-off okay that we've actually seen over uh the

past three quarters uh a decrease and and it's not a good omen. What about the level of rates? I mean, obviously, if something is different this time, it's that we have an inversion with rates that's at a historically low level. 2%, 3%, versus 7% on average. Can you talk about what the ramifications of that might be? Yeah, so— Or is it any—I mean, that's tough. Low rates are really, like, abnormally low rates. Again, are consistent with this sort of insurance type of motive

that the safest thing to be in is treasuries. So the rates are really low. The central banks have distorted the interest rates, and I think we're going to pay for it in the future. It just doesn't make a lot of sense. I was talking to somebody last week from Denmark who gets paid for their mortgage. So that is just highly distortionary. And what it does is it props up firms that shouldn't exist. They're basically kept alive because the rates are so low. It stifles innovation. It favors very large firms. All of this stuff will come back to

bite us in the long term. What about the argument that it already is? I know it's a little bit further outside of your lane, but I'm sure you read all the stuff that we do about Uber and Airbnb and Lyft. And obviously we work recently. It's hard to argue that innovation has been stifled when these companies that start off as little more than an idea are able to attain valuations of 5 billion, then 15, then 50. Why would, obviously that's an artifact of low rates also. So it's a little bit paradoxical, is it not? Okay, so this idea of innovation being stifled, we don't really know what the counterfactual is. So people seem to be impressed that US growth is close to 3%.

I'm not impressed with that at all. The potential of this economy could be double that. But given that we've got this weird environment, it's not just the inverted yield curve, but we've got extremely low, effectively 0% real interest rates or negative. That just is a recipe for future failure. How long would the economy have to continue to expand? Let's use quarters. Given that the yield curve had inverted last quarter, how long would it have to continue to expand for you to say, okay, something now is different.

There are other variables that are possibly more important than the rates and the inversion of those rates. What was the longest lag? The longest was 21 months. So that's the global financial crisis. So we're three years from now absent a statistical recession. Right. And what are we grinding away at? Like 2.5%, 3% growth-ish?

Yeah. So if we continue to do that and muddle through, would you say that maybe conditions have been so distorted by low rates that the yield curve indicator is less powerful? Yeah. So look, this is a very simple model. It's a single variable to describe a very complex economy. So obviously, the economy isn't driven by the yield curve relation. And I think you're asking, could this be a false signal? And the answer is, of course. Maybe seven out of seven is lucky. There is the economic theory behind it.

That's very sound. The intuition is very good. Could this be a false signal? Absolutely. Do I believe that the model is broken? Is there something structural that suggests to me that maybe it's not going to work this time? The answer is no. But what if the inputs are broken by what the Fed has done?

Okay. Well, let's look at that. And many people have said this time is different. We're looking for a lifeline here, Campbell. Yeah. So throw us a lifeline. Sure. So two things. Number one, the Fed was much more powerful in 1960s and 70s in terms of their intervention

in the bond market. Today, the bond market is so large, it's really hard for them to do much, especially at the long end. The short end maybe somewhat, but the long end is out of their control. That's number one. Number two, QE. People point to QE. Well, this time it's different because of their control. That's number one. Number two, QE. People point to QE. Well,

this time it's different because it's QE. Well, frankly, QE since 2017 has been unwinding. So that would actually steepen the yield curve. And it's not happening. They're selling. That's what I want to ask you. I want to follow up on the bond curve is now so large. Because people would say, and I'm not saying this is right, but people would say when you look at U.S. recessions and monetary policy and yield curves in the 60s and 70s and 80s, it's not that there wasn't some globalization. It's that it was nowhere near the degree as intertwined as it is today. And how can we ignore that a lot of the reason for depressed bond yields and high bond prices is demand coming from Europe, from Japan, and those central banks

arguably are way more YOLO than ours is. They are still talking about further negative rates. They are still pursuing new avenues of QE. So don't we have to factor that in that that wasn't happening during the last seven yield curve inversions? So it was happening in some of the inversions, the more recent ones. But it is a fair point that the U.S., while it has like a lower share of world GDP, is the single most important driver of the world GDP. I agree with that. And if you look at the relation between world GDP and a GDP-weighted yield curve across different countries, you see the same predictive power, and it's being driven by the U.S. So all countries need to look to the U.S. because if the U.S. gets slower growth or goes into a recession, that is bad news over the world. The U.S. consumer seems to me to be the foremost engine of economic growth on planet Earth. And I understand that's relatively recent development. Maybe it doesn't

last forever. But arguably, over the last 10 years, if you focused on what was going on with the U.S. consumer, you got things mostly right. Does the U.S. consumer get affected by the yield curve or not until they lose their job? Like is there a transmission mechanism there? So there definitely is. So I gave the example of the CFO not pulling the trigger in a major expenditure and borrowing and things like this. So people don't get hired as a result of that? Yeah. So for the consumer, you're thinking, well, I see this yield curve's inverted. There's

a possibility of a recession next year or the year after. Maybe I need to rethink what I'm doing. And there's many different ways to rethink. So one might be not doing this really expensive vacation where I need to max out my credit card. But there's other subtle things. So if there's a recession coming, it's probably not a good time to look for a new job. Or quit your job. Or quit. That's a disaster period. But you start looking. People figure it out. And you're going to be the first one axed.

And even if you do move, you have no seniority. You're going to be the first one axed. So stuff like that I think is really important for consumers. So this is not just for corporations. This is for consumers, retail investors. Everybody should take this into account. You're a partner at Research Affiliates. How does things like the trade war, the inverted yield curve, the state of the economy, how does that figure into the work that Research Affiliates

is doing in asset management? Yeah. So we did a really interesting analysis that's posted on LinkedIn, my LinkedIn page, where we looked at the performance of the stock market after inversions. Okay. So you observe the inversion for a quarter, then what happens to the stock market? What did you find? It's grim. So what we did is we averaged over the last seven recessions.

And you see before the inversion, market does great. After the inversion, it does poorly. After the inversion, it does poorly. So we also did the same analysis for- Valuations, dare I ask? Yeah, exactly. So you're guessing it. So what about different investment styles? And we haven't kind of put this out in a paper yet, but I did show, for example, value investing. What does it look like after recessions? And it does well. So it does poorly before, and it does well after inversions. So if you have a value tilt to the way that you're allocating a portfolio- Keep hanging on.

Well, arguably, that's like your time to shine because you didn't own the highest flying, most economically sensitive stocks. That is correct. Okay. And they're very sensitive to a slowdown. So Mike brought a chart that he made that he'd like to show you. I showed him before we started. Oh, okay.

But we're going to flash this on the screen. So let's get into what you're saying. So just in terms of how abundant information is, and maybe this causes a feedback mechanism. So I did a chart showing the yield curve on one axis and the number of New York time mentions on the other. And looking historically, there was mentions. People spoke about it. But now in the most recent one, it is off the chart.

So does this – and I guess we already spoke about this. That's because you're famous now. So is this going to affect the real economy? So I actually think that this is like a valid sort of self-fulfilling prophecy sort of situation where people are aware of this indicator. They will be more careful. It's risk management. So you need to look ahead and plan. And you need to use all of the information, not just the yield curve, but other information in the economy to make the best possible plans so that you survive a slowdown.

It might not be a recession. that you survive a slowdown. It might not be a recession. It might be a slowdown. My model is about forecasting future economic growth. So if we slow to 1%, I would consider that an accurate forecast. Your message to stock investors is not panic. It's be aware that this probably isn't the best time to be taking the maximum amount of risk. That's right. Or to be risking capital that you might need to live on over the next two years.

So I think this is a sound. A yellow light, not necessarily a bright red one. Yes. So this is the time, in terms of asset management, this is a time where you need to reflect upon your strategy. It's actually easy to manage assets when the economy is booming and it's much more difficult to manage into a turning point. So what I recommend is a defensive posture right now. This warning

sign from the yield curve, the Duke CFO survey, other information is in your face. So it's way better to have a plan, to go by the plan, than to find yourself in a situation where the recession hits and you have to improvise. Improvising is a bad way to do asset management. Just making decisions under duress. Exactly. Rather than having a plan in advance. You want a plan.

I think that's great advice. Campbell, will you come back when the coast is clear? Definitely. For the post-game interview? We really appreciate you being here. Your stuff is great. Where can people read more of your thoughts on an ongoing basis? What's the best place for us to send them to?

So my LinkedIn page. We'll link to that. And all of my research is linked on the research affiliate site. Fantastic. Thank you so much. This has been Campbell Harvey, the godfather of the yield curve indicator. I'm Josh Brown. This is Michael Batnick. Go ahead and give us a like, subscribe to the channel if you are not yet subscribed, and we will see you soon.