In their latest episode of the VALUE: After Hours Podcast, Cam Harvey, Jake Taylor, and Tobias Carlisle discuss:
- How Cam Harvey Invented The Yield Curve Inversion
- The Fed’s Strategy Is An Own Goal
- How An Inverted Yield Curve Impacts Banks
- Yield Curve Recession Indicator Correctly Picked 8 For 8
- Is The Fed About To Make A Second Mistake To Rectify Their First Mistake?
- Recession Is Imminent
- Flashing Red Yield Curve Could Be Self-Fulfilling Prophecy
- Is The Yield Curve Indicator Giving A False Signal?
- There’s Something Strange Happening In The Employment Sector
- Severe Yield Curve Inversion Causing Real Stress On The Financial System
- SVB Stress-Test Was So Unrealistic
- Cam Harvey – New Book – DeFi and the Future of Finance
You can find out more about the VALUE: After Hours Podcast here – VALUE: After Hours Podcast. You can also listen to the podcast on your favorite podcast platforms here:
Breaker
Overcast
RadioPublic
Stitcher
Full Transcript
Jake: The other day when I was coming home on Friday.
Tobias: We are live. It’s Value: After Hours. I am Tobias Carlisle, joined as always by my cohost, Jake Taylor, and a very special guest today, Cam Harvey. Cam, you’re a Professor of Finance at Duke, and you’re the Director of Research at Research Affiliates. Welcome to the show.
Campbell: Thank you for inviting me.
Tobias: Absolute pleasure. We’ve been using your name in vain and your-
[laughter]Tobias: -and your inversion indicator for a little while. You wrote a PhD dissertation in 1986 on the yield curve inversion, that’s the 10-3, and the implications for recession. What is the 10-3 inversion?
Campbell: So, can I just give you a little bit of background-
Tobias: Sure.
Jake: Absolutely.
—
How Cam Harvey Invented The Yield Curve Inversion
Campbell: -on the whole idea? It’s hilarious now, but in real time, it wasn’t. So, I’m a first-year master student and I applied for internship in Toronto. That’s where I was in between first year and second year. I go into this company that was called Falconbridge, the world’s largest copper miner in the world at the time. And this is in 1982. I walk in as an intern, first-year master student into the corporate development area and they said, “Well, your job is to design a forecasting model for real GDP.”
Tobias: Easy.
Jake: Oh, yeah, no problem, layup.
Campbell: Yeah, that’s exactly my attitude. Like, “Oh, okay, well, this is normal. This is what I should expect in the world of big business.” I just shrugged it off, figured, “Okay, well, I got to do this.” The competition at the time were these companies that were specialists in these giant econometric models. So, they have massive data systems, hundreds of equations, and then you’d have to pay them tens of thousands of dollars to get one number.
So, I’m thinking I’ve got nine weeks, and there’s just no way I can assemble a model like that or the data. I can’t compete against them. What about using some stuff I learned in the intro finance course, that assets actually have information about the future path of whatever, earnings or things like that. So, I started looking at the stock market, I quickly realized that was just all over the place. And the joke at the time was the stock market predicted successfully, like nine of the last five recessions.
Jake: Right.
Tobias: [laughs]
Campbell: So, a huge false positive rate. But I quickly moved to bonds. It just seemed ideal, because a bond has got a fixed coupon versus a dividend that you have no idea what it’s going to be. A bond has got a fixed time to maturity and a stock, again, who knows what the maturity actually is. And then just on the risk angle, if you’re looking at treasury bonds and bills, those have very low risk compared to stocks, where the value of stocks can fluctuate– Even if the cash flows are the same, if risk goes up, then the stock is going to go down.
So, I decided to look at bonds and then decided to look at a spread and yields, and I wanted to do that to take expected inflation out. There was this early paper that I saw published by somebody at the Federal Reserve, 1965. So, it was way back. They noticed a cyclical pattern. It was nothing to do with forecasting, but they just noticed a cyclical pattern. And I said, “Well, I definitely want to look at the yield curve.” I put this model together. It was shocking to me that I could do as well or better than these econometric services. I’m ready to present to the senior people at Falconbridge. It’s my day of presentation. I go in and I’m told that the whole division is laid off, and [Jake laughs] I need to collect my stuff and be shown the door at the bottom of the building.
So, before I could present it to them– To know what’s going to happen in real GDP is so important for copper. It’s like Dr. Copper. You need to know in terms of your exploration budget, opening a mine, closing a mine, all this stuff, very important. I actually delivered something. Well, I didn’t deliver it. So, I’m gone. I’m on the street and I decide, “Well, this idea is pretty cool. Maybe I’ll just spend the next three or four weeks working on it.” Then I went back. My second-year masters, presented the paper, and they said, “Ah, you need to go for a PhD.” And that’s how I ended up at the University of Chicago. So, that’s the story.
The story is very solid economic foundations. Just think of the simplest possible scenario that, if people get nervous about what’s going to happen in the economy, then there’s a flight to safety. And often that safety is the 10-year bond. Just thinking of that alone. Well, if the 10-year bond, a lot of demand for price goes up, yield goes down, and that serves to flatten the yield curve or even inverted. So, the original model in my dissertation at the University of Chicago, 1986, was based upon expectations that financial assets like bonds and stocks but bonds a lot less noisier contain valuable information about the future.
It’s also the case that in contrast to the, let’s say, the stock market, the economy is a lot easier to forecast. The intuition for that is pretty clear also that things are sticky, that you make an investment that takes a while to actually pay off in the economy. It’s not like a stock investment. You’re buying equipment or a plant or employees. So, there is predictability in the business cycle and you just need to come up with a model for that. My model has been, I would say, I’m trying not to be immodest here, successful.
Jake: [chuckles] So, you knew what you were going to write for your thesis before you even got into grad school?
Campbell: Yeah, this is– [crosstalk]
Jake: That’s pretty wild.
Campbell: I now evaluate these applications.
Jake: Yeah.
Campbell: Again, I didn’t know anything. Back then, PhD, how long is that going to take? I told my parents and they are shaking their head, and they said, “Well-
Jake: “Who’s going to pay for that?”
Campbell: Masters was excessive, given that [Tobias laughs] neither of them had undergrad degrees. “So, what is a PhD? Like, another year?” I said, “I don’t think so, but I don’t really know.” Actually, the first day there, somebody gave me a tour and I asked the person, “Well, how long you’ve been in the program?” because he looked rather ragged-
[laughter]Campbell: -and a lot older than I expected. He said, “Well, it’s my 9th year.”
Jake: Oh.
Campbell: I said, “You graduating this year?” “I don’t know.”
Jake: And his name was Cliff Asness. [laughs]
Campbell: No. I actually did overlap with Cliff. I came back as a visiting professor. It was actually hilarious because I graduated after three years. Given that I came in with my topic that saves a huge amount of time. A number of years, of coursework, then you start thinking about research. No, the first day, I’m working on my project. So, I was looking for a job after my second year and accept a job after three. And then, I get invited back for a visiting professorship. In the finance seminar, I would just sit with the students in the student area, because I knew them. They’re my colleagues.
Jake: Yeah, your friends.
Campbell: But it was really confusing to the faculty because they thought I was still in the program, “Oh, he’s taking a long time.” But I did overlap with Cliff. He was a brilliant student and actually had the pleasure of reading one of his papers and commenting on it. It was fun to do. At that time, there were so many great students, including Cliff at Chicago.
—
Tobias: Let me just give some shoutouts, and then, I’ve got a few questions for you. Santo Domingo, Dominican Republic, how are you? Nashville. Bendigo, Victoria. Austin, Texas. Mississippi. Fort Lauderdale, Sweden, Massachusetts, Dubai, what’s up? Santa Monica. Gothenburg, Sweden. Dallas, Toronto, Del Boca Vista, Florida. Bermuda, Tallahassee, Greece, Kent, WA, what’s up, guys? Glad everybody could join us.
Jake: And that [crosstalk] spin the globe segment?
—
Yield Curve Recession Indicator Correctly Picked 8 For 8
Tobias: For those guys who’ve just come in late, it’s Cam Harvey, the creator, the inventor, the discoverer of the 10-3 inversion. Cam, a couple of things that you mentioned while you were going through that, that I just want to dig into a little bit. I think that I saw you appear on a Research Affiliates seminar, this is a few years ago now, and you were discussing the 10-3. Before I saw you present it, I didn’t actually realize that– Actually, it was the Meb Faber podcast from a few years ago. I didn’t actually realize that it was the 10-3, because it’s often quoted in the press as being they look at the 10-2. Do you know how that happened or why the 10-2 became substituted in collective mind over the 10-3?
Campbell: Yeah, it’s a very strange situation. So, my dissertation in 1986 uses– actually, I look at the 10 to 3 month and also the 5 to 3 month, and the 5 and the 10, highly correlated. But it’s really important to anchor with a short-term instrument like the three month. I chose the three month because it’s liquid and we measure GDP quarterly. So, let’s do something over a quarter. So, that indicator since 1986, we’ve had– So, this is the out of sample period. In the out of sample period, it is four out of four with no false signals. So, the way I’m looking at it– [crosstalk]
Jake: Which makes it eight for eight then now? Is that accurate?
Campbell: Yeah. So, think of it from the late 1960s. So, that was in my dissertation. Indeed, this is like another story. Most of the key analysis in my dissertation is over this period from the mid-1960s to 1985. In that period, there’s four recessions and there’s four inversions. It looks like four out of four perfect indicator. But my committee is saying, “Well, that could be lucky. You can’t get four out of four.”
It turned out that even though it could be lucky, they liked the idea, were willing to go forward with it for a few reasons. One of the overwhelming reasons was theory was sound. So, theory predicts this. Then if we see it in the data, then well, maybe it’s still lucky but still you’ve got some basis for it that some good strong intuition. That was number one reason. Number two reason, they were fascinated that my indicator got the double dip recession in the early 1980s. So, we had a recession, then a strong recovery, then another recession. These macroeconomic forecasting services, nobody got that. And then the third thing and people of Chicago especially like this that the alternative to paying tens of thousands of dollars a year to these econometric services is a very simple model that is as good or better and it costs the price of a Wall Street Journal, which at the time was $0.25. Yeah, this is good.
Jake: I thought you were going to say because it shows the efficiency of the bond market, the information that’s contained within there.
Campbell: Yeah, the $0.25 will drive the price down to what the efficient price should be.
Jake: Yeah. [laughs]
Campbell: Yeah, that’s where we’re going.
Jake: So, four for four in research, four for four cents in Post, and now recently another inversion has happened.
Campbell: Yeah. So, let’s go through that. That’s going to require some unpacking. I don’t want to leave this hanging, your previous question, well, why is it 10-3 versus 10-2?
Jake: Oh, yeah. Sorry.
Campbell: So, given that the indicator is four for four out of sample and eight for eight to the 1960s, I don’t see a particularly good reason to switch the model to something else that doesn’t have the same theoretical underpinning. So, if it was the case that out of sample, let’s say, I got two out of four, whereas this other indicator got four out of four, then okay, it seems like the model is broken, so let’s fix it. But that’s not what happened. So, the Fed in particular started talking about the 10-2. And by some metrics, it might fit the data a little better. However, it’s given a false signal in 1998.
Tobias: Right.
Jake: Mm.
Campbell: So, again, it’s just not really a good reason to abandon something that’s working.
Jake: Somebody else wanted their name on the 10-2? [laughs]
Campbell: Yeah, maybe. That should be suspicious of that just on its own. So, what I often say is, well, if we want to get the best possible fit in sample,” so calibrated to the data, it’s not the 10-2. It could be the eight-year, 2 month minus the 16 month.
Tobias: [laughs]
Campbell: So, there’s tens of thousands of combinations you could try to get the best possible fit, but we all know that when you do that, that model will most likely fail out of sample because it’s been overfit.
Jake: Right.
Campbell: So, my model is not overfit and it’s done well. So, eight out of eight, but the big question is, “Okay, great. That’s a fabulous historical record. What about today?” So, at the end of December, we had a full quarter inversion. So, my model is about quarters, right? Not months or weeks or days. So, that a 10-year minus 3 month, the spread was negative. So, short rates higher than long rates on average over that quarter. And at that point, I usually put a post up saying code red, and the track record is pretty impressive.
So, this time around, I did put the post up and I said, “The model is giving code red, but let me explain to you why I believe that my model is giving a false signal.” I guess that got the attention of a number of people.
Jake: Yeah.
Campbell: So, Harvey going against his model. Many people could go against my model. Many pundits are against my model already, but it’s a little different when you’re the whatever–
Jake: [crosstalk].
Tobias: Discoverer.
—
Is The Yield Curve Indicator Giving A False Signal?
Campbell: OG, whatever. So, let me make my case and then let me talk about the qualifier that I had. This is a post on LinkedIn, January 4th, 2023. So, this is the way I approached it. First, this is a very simple model. It is a model with one variable. It’s a lot to ask for a one variable model to have a perfect track record forever, and especially given the complexities of our economy. I also make it clear that it’s not in my nature just to promote the model because it’s mine. I totally understand. My training at Chicago is very helpful for this. You understand the limitations of the model. And indeed, I haven’t done research in this area in a long time, in 30 years.
If I was hired by that same firm, which by the way, didn’t make it, [laughter] they were acquired at a significant discount. So, if I was doing that job again, I would do it differently. I would definitely look at the yield curve and that might be 50% of what I look at, but there’d be other things that I look at. So, let me go through the other things that I was considering and why I decided that on January 4th that this was likely a false signal.
—
There’s Something Strange Happening In The Employment Sector
So, the first thing I noticed was the strange situation in the employment sector. Think about unemployment in general. So, unemployment is always low before a recession. It is at best a coincidence indicator or a lagging indicator. So, just to say, “Oh, we’re not going to have a recession because unemployment is so low.” Well, it’s always low before a recession. That’s not what I was interested in. What I was interested in was the ratio of job openings to unemployment. That was very weird. It was almost 2. Right now, I think it’s 1.7. Even that is really high.
What that does is that it allows for some economic slowing. So, just to be clear, a flat yield curve or inverted yield curve means the economy will slow, and that slow could actually manifest itself in a recession, because that’s like the worst scenario is a recession. So, the idea here is that you could be slowing and some people laid off, but their duration of unemployment is low.
Jake: With all the openings?
Campbell: Yeah, with all the openings, you get another job. Then the media was focusing on all of these tech layoffs. I’m really shaking my head thinking, well, the duration for these people who were fortunate enough to win the competition to get to one of these top firms like Twitter, those people are very valuable elsewhere. Then somebody said to me, “Oh, well, looking at the data, they aren’t placed that quickly.” I said, “Well, I totally get that. You take a vacation and take a break and when you get back, you choose where you want to go.” So, that is so different than if you got laid off by Lehman Brothers. Where are you going to go? To another bank, to Bear Stearns? No, you’re facing a long period of unemployment and many people suffer greatly in the Great Recession or global financial crisis.
This duration, both the number of openings to the number of unemployed, plus just the nature of the layoffs, that just didn’t really check the boxes. It was something very different than, let’s say, the global financial crisis. And looked at other things like housing. So, if you look at the ratio of equity to debt just before the global financial crisis, the amount of debt was very large compared to equity. We know that the global financial crisis was in part triggered by what was happening in the housing market. If you look at that today, it looks completely different that the equity is so much larger than the debt that even if housing went significantly down in price, we wouldn’t have the same sort of contagion issues.
—
Flashing Red Yield Curve Could Be Self-Fulfilling Prophecy
So, there are other issues. One other issue is the idea of self-fulfilling prophecy. This is something different than my dissertation. So, this is outside of my dissertation and let me tell you the story.
Jake: It’s kind of a Heisenberg principle issue there.
Campbell: Exactly. But let me go through the intuition of it, because it’s very, very important to understand. So, before the global financial crisis, nobody really took note of my model, even though it had done very well. This is important also. Not just in hitting the two recessions afterwards, but not giving false signals. So, in 1987, when stock market crashed, the consensus was a recession in 1988. The strong consensus was a recession in 1988. And my model actually had a number. So, I said, “Listen, the model says 4.2% real economic growth.” That was a joke. That was such an outlier compared to what people were thinking. We had plus 4% growth in 1988. So, that sort of thing. So, you’re not counting a recession correct prediction. It’s a non-recession correct prediction, but nobody really took it seriously. There are pockets here and there, but think of– [crosstalk]
Jake: They’re all wealthy and retired now. [laughs]
Campbell: Yeah, I should be wealthy and retired. [unintelligible [00:26:02]
Jake: [laughs]
Campbell: So, think of the CEO going in front of shareholders in 2009. They’ve been hammered by this recession and they say, “Look, we’re blindsided. We had no idea. If we had known what was going to happen, we wouldn’t have pulled the trigger on this major investment that has put the firm at risk.” Oh, by the way, it’s not just me. All of my fellow CEOs are in the same shape as our company– and we were all blindsided. So, let’s go forward today. After the global financial crisis, people started to realize, “Oh, well, this indicator is six out of six, no false signals.” So, things actually changed.
So, now, suppose we go into a recession in, let’s say, late 2023 or early 2024, and then let’s imagine the same CEO going before shareholders at the annual general meeting saying, “Oh, well, if we had known a recession was coming, we would have never pulled the trigger on this bet your firm capital investment.” The general meeting would erupt in laughter. There’s no blind sighting, unless you’re like an ostrich. This is in your face. It’s all over the place. We’re talking about it today. This is no excuse. So, you see this, “Okay. Oh, yield curve inverted. It’s eight out of eight. So, I’m not going to take the risk. I’m going to delay that capital investment. I’m going to delay hiring. Indeed, let’s do a 5% layoff proactively just so if we go into recession, we won’t have to do the slashing that is so painful for both the firm and the employees, obviously, that are laid off.”
All of this, think about what happens here because the yield curve is flashing code red, these firms are taking positions that actually decrease economic growth. That’s the self-fulfilling prophecy. So, the cutting back investment, cutting back unemployment, and on other things, all of that just drops to the bottom line for GDP. It slows GDP growth. This is crucial. It slows, but it decreases the chance of a hard landing. Does that make sense that you take these actions, you do the 5% layoff, you defer that project that you need to borrow a lot of money for. If we go into a recession, you’re going to survive. If we don’t go into recession, well, we just had some slower growth and we’ll pick it up later when we retrigger some of these projects. So, the yield curve inverting itself is causal now in terms of lowering economic growth, but I view it as kind of risk management.
Jake: Yeah, hit the brakes a little bit. There’s a curve in the road ahead.
—
How An Inverted Yield Curve Impacts Banks
Campbell: Right. So, let me go to another factor that I mentioned and that was the financial system. I said that it was relatively strong compared to what happened in the global financial crisis when the large banks were acting as hedge funds. So, taking extreme leverage and acting as hedge funds, which had like a Fed put option. We’ve made a lot of changes since then. My read of the financial system was that it also was not going to cause a contagion, if went into a mild recession.
So, I went through all of these factors. There’s more of them, but I had a major caveat and I said, “We can avoid a hard landing recession.” But on January 4th, I said, “The Fed needed to stand down, that if the Fed continued to increase rates, then that will cause unnecessary stress, in my opinion, and would lead us to a recession.” So, the Fed, as you know, has chosen to decrease the size of the hikes, but has not stopped. The Fed has not stood down in any way. This creates a second channel of causality from the yield curve to the economy. We can go through that channel. We’re living that channel right now and it’s a channel directly through the financial system. Let me explain what I mean here.
So, let’s think of just a simple model of a bank. Deposits come in and you pay the depositors the short-term interest rate. And then you take those deposits and you lend them out. So, you lend them out to companies and that induces some credit risk, but you’re careful in your due diligence, hopefully. You can also lend to the government, which means just buying government bonds. That’s the revenue that you’re getting. So, the revenue you’re getting are from the payments from the loans and the coupons on the bonds. The cost is what you’re paying the depositors.
This works great almost all the time, because almost all the time, short-term rates are lower than long-term rates. So, now let’s flatten the yield curve and potentially invert it. We’ve got a severe inversion rate now where it’s like 1.5%.
Tobias: 1.68 yesterday.
Campbell: Yeah, it is remarkable. It’s also remarkable given the size of rates. So, if you look at the percentage inversion, it is massive and historically unprecedented. But let’s go back to the bank. So, as that short rate is going up, you are paying more to your depositors. And given that you’re locked into longer-term investments like these loans to companies and the bonds that you bought, that’s not moving. Your business model is being upended. So, the business model works great if those long-term cash flows that are coming in are exceeding the short-term cash needs. So, when you invert the yield curve, you stress that model. Indeed, it could come to the point where it causes big problems. This is interesting to think about that we talk about the inverted yield curve, but it really matters the way that it inverts.
So, in the case that we’ve got today, both short-term rates went up and long-term rates went up. But short rates went up more than long rates. So, why does that matter? It matters because these banks have these longer-term investments. I guess Silicon Valley Bank is a great example of that, where they’ve got their commercial loans that had no problems, that failure had nothing to do with the quality of their loan book. It was very high-quality, but it was the loan book to government. So, the bonds that they were holding, and those bonds, even if you’re holding them to maturity, when the long rates go up, when interest rates go up, the value plummets. Maybe you can’t hold them to maturity. Maybe you have to sell. That’s where that bank went insolvent.
—
SVB Stress-Test Was So Unrealistic
Now, I’m not saying that SVB demise is purely a result of regulators. I’m making a simple point here that when you invert the yield curve, that creates stress in the banking system. Silicon Valley Bank is a prime example. So, if we think about what they did, they said, “Well, rates are really low–” This is, let’s say, three years ago, “Rates are really low, but we can significantly increase our profit just by switching to higher duration bonds.” So, let’s buy long duration bonds rather than short duration and we can get more profit. This is the so-called reach for yield. In doing that, you also increase risk and that risk was realized.
Now, of course, you could hedge. You could do some swaps. They had some swaps, but nothing really that meaningful. Why? Well, I think they unloaded their swaps mainly because it was profitable to do another source of profit. This is also a regulatory failure. I know we’re veering a little bit, but it’s important to understand this, because even though SVB wasn’t subject to the so-called stress test, if they had, they would have passed. That’s because the stress test adverse scenario was so unrealistic. The policymakers took us far away from the adverse scenario.
—
Severe Yield Curve Inversion Causing Real Stress On The Financial System
So, the reason I’m going on this thread is that the inverted yield curve in the severe inversion has created stress in the financial system, something that I thought originally could be mitigated. This stress in the financial system is causing uncertainty. And right now, we’re in a lull. We saw a few banks get into trouble. We saw SVB go down, we saw another bank have $100 billion in withdrawals, somehow remain in business, and now, we’re in this lull period. The Fed has said, “Well, our banking system is sound and secure.” Okay, fine. Want to show me some evidence?” That’s the case.
Jake: Subprime is contained.
[laughter]Campbell: Yeah. You just say, “Oh, everything’s okay,” then people just assume it’s not the case. But people want to believe it’s the case. So, we’re at this stage of the business cycle where we see some problems in the banking system and just say, “Oh, well, that’s just a one-off. This is just idiosyncratic.”
Jake: Is that like Bear Stearns hedge fund failing in 2007?
Campbell: Exactly. “It’s a one up, no big deal.” But we need to look under the hood. I’d wish that the Fed would do its analysis. I would go and ask for another stress test with a realistic scenario at minimum or take a look. This is pretty simple exercise. Let’s say the Fed is thinking of another 25-basis point hike. Then go, do the math, and figure out the duration exposure of all these banks, and figure out how many additional banks go negative equity when you do this. So, it’s a simple piece of information. You’re doing this, you know that flattening the yield curve is going to stress the system. So, wanting to measure how much stress you’re going to induce.
These banks, frankly, are not that difficult to value. One of the key things is just looking at the bonds that these banks hold. Again, this is not a difficult exercise to do. I think it’s incumbent upon our policymakers to take action that’s data driven. We won’t know exactly what they see, what they’re looking at for five years, because the minutes are embargoed for five years. We only get a summary. But I certainly hope they’re doing this. This is related to what we’re talking about that I fear–
—
Is The Fed About To Make A Second Mistake To Rectify Their First Mistake?
Campbell: I’m not an advisor to the Fed or anything like that, but I fear that the Fed now realizes that they made a mistake keeping rates low for so long and having zero rates effectively, when we’ve got robust economic growth, low unemployment, record high stock prices, what were they thinking? [crosstalk]
Jake: [crosstalk] budget deficits too in the fiscal side?
Campbell: Yeah. So, now they’re thinking, “Okay, well, we were late.” They tried to talk this temporary thing for a long time until it was a joke. Anytime a Fed person said temporary, the audience would start to laugh. I fear that they don’t want to be early in exiting. What they’re doing is thinking that they can solve the first mistake by a second mistake. Two negatives make a positive. It doesn’t work that way. Two negatives make a bigger negative. It is unfortunate. I do believe that we should look at the data. But again, looping back, given that the Fed has not stood down, nor have they given an indication that they will stand down the next meeting, this has created significant risk in the financial system.
The other thing that I worry about that I’ve mentioned– Again, we don’t know the extent of the damage in the financial system right now. Indeed, we’ve got this incredibly dysfunctional situation. It’s so weird. If you look at the average savings rates, they’re very low. They’re like 50 basis points. And then the average money market rate is like 450. So, what’s happening is, people just sweeping their money out of these banks and going elsewhere. Why is the savings rate so low? That’s all they can afford. They’ve got the market power to actually do it. Hopefully, some people stick around at 50 basis points, even though they could get 4% more in a very simple way. When I see that, that to me just screams risk. But it’s not just the banking system. I worry also about commercial real estate.
Let me give you my logic on this. So, remember I said that if I was doing my job at Falconbridge as a student, I’d be looking at other indicators in addition to the slope of the yield curve. Some of those indicators would be credit indicators. So, think of like high yield minus treasury, stuff like that. If you look at that today, it’s not really a problem. But there’s other things that could be looked at. If you look at, let’s say, CMBS, like mortgage-backed securities that are commercial, the spread over Treasuries has been climbing. It’s been climbing for a while. I think that is a fragile market. There’s another reason for the fragility, and that is that the demand for commercial space has suffered, what I consider a structural change given the pandemic. So, it’s a lot more likely that some of your workforce can work from home.
I know this firsthand at Duke, because we had all these buildings planned for all of the offices that we needed people in trailers, and stuff like that. Well, all of a sudden, 25% are working from home and we don’t need to construct a new building or two. So, this is just manifesting itself in big cities. So, I do worry that that could be the next source of stress. As soon as that happens, then we revisit the financial system, which is unresolved right now. I know this is terrible to say because it sounds like a conspiracy situation, but maybe– [crosstalk]
Jake: We welcome conspiracies on the show. [laughs]
Campbell: The Feds got the data. I don’t know. Maybe they’ve done the exercise that I’m suggesting. Maybe they know the number of banks that have negative equity right now. Maybe they had that information. They’re sitting on it. So, you can totally break that conspiracy theory by resolving uncertainty and just releasing the data. So, it shouldn’t be that people have to go do this on their own, go through the 10Ks, and bank by bank, try to do the math. This is why we pay the regulators. They’ve got a job to do. So, the job is actually, in one way, it’s constrained because they’re enforcing the regulations that exist, and those regulations come from Congress because that’s a regulatory framework.
So, the other important function is to monitor. To monitor, you need to have data on the health of all of these banks. We have a large number of banks. I know that the top four get most attention, but that second tier could be very important, and we’ve already seen significant weakness in that second tier. So, this is a long way of getting back to the yield curve. This is the second channel of causality. The first channel of causality that I went through is the so-called self-fulfilling prophecy.
When it inversed, people cut back and it slows growth. The second channel of causality goes through the financial sector. I’ve described one of the methods here that the bank’s business model is stressed, because they have to pay out more, they’re receiving less. When those long rates go up, just the value of those loans go down. That’s what happened at SVB. You need to take a write down for your available for sale portfolio. At least your ultimate maturity portfolio should take a similar write down, but it doesn’t according to the accounting rules.
Jake: Can’t you flip those bonds to the Fed though using one of their windows and getting 100 cents on the dollar–? [crosstalk]
Campbell: Right. But think about that, what’s the cost of doing that? So, yeah, the Fed’s new system, I got a bond that’s worth $60. I can send it for collateral and the Fed says, “Well, that’s $100 bond.”
Tobias: At par. Yeah.
Campbell: At par. So, we’ll lend you some money, but we’re going to lend it at Fed funds rate, which is super expensive. So, that is expensive to do. Indeed, I’m watching that very carefully, because that’s telling you something that is like an indicator. I think most banks have figured this out and don’t want to use this facility, because it’s just so obvious that you must be desperate, if you’re going to do this. So, again, we just need the data. We don’t have the data. It’s just unclear how serious the situation is and the Fed, by increasing rates, is playing with fire.
—
The Fed Strategy Is An Own Goal
Let me also say that, if we do go into recession– I’ve called this the Fed strategy, an own goal. [Jake laughs] Then I was told, “Well, Americans don’t really know what that means. Everybody in Europe knows what that means.” Let me explain what I mean by that. So, the reason to increase rates overwhelmingly is inflation. But if you look at the last nine months, inflation is running at 3.2%. That’s not 2, but it’s close. Then more importantly, what is driving that inflation? That inflation is being driven by shelter or housing. So, 70% of that inflation is coming from housing. The Fed made this mistake before that inflation was going up and they were saying, “Oh, well, this is temporary, it’s no big deal.”
I’m looking at the data. Housing and rentals are up double digit, but that was not being reflected in the CPI. So, housing enters with a lag. The intuition for listeners is real simple, think about you’re in a lease and you’re locked in, and then all of a sudden, rental inflation goes up by 15%. Well, you’re locked in. If your lease is a month old and there’s annual lease, then inflation is zero for you for the next 11 months, but then when it comes off, you’re going to suffer the 15% and it goes into the data. So, shelter is this lagging indicator. Again, it was the Fed’s mistake not to see that the inflation was going to be way more permanent because of the housing inflation.
So, today, so we’ve got inflation of 3.2% over the last nine months. That’s annualized 3.2%. But look at the housing market. So, housing prices are going down, rentals are going down. It’s the same story that it will take up to a year for that to work its way into the CPI. In my opinion and also, just by the way, shelter is 33% of the CPI and 40% of the Fed’s favorite indicator of the personal consumption expenditure deflator. So, to me, there’s not a good reason to keep on hiking. All we do is to increase the probability of a problem with the financial system and increase the probability of a recession. Not just a technical recession, but a potential hard landing recession.
—
Tobias: Does the steepness say anything to you that this is–? If you run the data back on the SEC’s web or whatever it is, the Fed’s website, it runs it back to ’80, and there’s nothing like 1.68, which was the close yesterday, which was the record in there. There’s another data series that goes back and it captures ’77 and ’80, which are both– It was steeper and very noisy through there.
Campbell: Yeah. So, my advice, Treasury bills go back to 1940s. So, just use the series that’s called– and this is advice for everybody, there’s a long series of the three-month bill quoted it on a discount basis. You just need to do a conversion. So, discount basis is this weird quirk of bonds. If you’ve got, let’s say, a one-year Treasury bill and let’s say the price is $90, so in a year you get like $100, they will quote that as a 10% discount yield. But we can easily convert that to a true yield. So, if you buy it $90, you hold it until $100, you’re going to be making 11.1%. So, there you can get– [crosstalk] Look at the Fed series only from the 80s because of this, they should use the discount yield and convert it.
Again, you look at that series, pretty clean from 1968, no false signal. 1998 was close, and that was during the LTCM disaster. Yeah, it was true. There was a lot of uncertainty at that time, and the yield curve flattened correctly. So, that’s exactly what it should have done. But again, today, it’s different because it’s got all these causal influences.
Jake: Does that hold up at international–? Can you replicate with other bond markets outside of the US?
Campbell: Yeah, that’s interesting because when I did the dissertation, I did some other countries and published papers in other countries on this. There seemed to be some predictive ability everywhere but Japan.
[laughter]Jake: Japan’s always got– [crosstalk]
Campbell: Yeah. This is early in my career, I’m thinking, “Oh, well, that’s so weird.” But now looking back, “Yeah, I understand. Everything is weird in Japan.” Indeed, the most interesting paper I did, in my opinion, other than dissertation was I looked at Canada and the US. So, I’m Canadian. I figured I’ve got to do paper in Canada.
Jake: Yeah, home country.
Campbell: In Canada, they got such a high beta with the US. So, their business cycle, very closely tied and interest rates to some degree. So, I fully expected that the US yield curve would predict Canadian GDP, but that wouldn’t have been interesting. What I did was I looked at the difference between Canadian economic growth and US growth, and then looked at the difference between the Canadian yield curve and the US yield curve, and found that had predictability. So, the incremental growth of Canada either above or below the US, that was strongly predicted by the difference in the yield curve slopes. I thought that was interesting evidence. Those markets, fairly liquid markets. When you go offshore to other markets, you’ve got illiquidity issues in Japan. It’s extreme now given BOJ buys everything.
Jake: Yeah. Is there an argument maybe that you might lose some predictive ability if there is a more Japanification of other markets?
Campbell: Yeah, I always think about that. There’s always noise. This simple model I’ve got, it doesn’t even have the Fed in it. So, the Fed is creating noise. Yeah, you always worry about this. I thought about this even in my dissertation with the Fed doing Operation Twist in the 1960s, the first version of it. It’s like, “Oh, well, that’s going to distort the predictive ability of the yield curve.” You could think that– look, I believe that is the case you will distort. In today’s case, it isn’t a distortion because it’s causal. So, given this extreme inversion, they are stressing the financial system, which means deposits are fleeing, going to money market funds, the banks are cutting back on their loan books as a result of the deposits fleeing. So, you tighten credit.
All of these are ingredients of a self-inflicted wound to push us into an unnecessary recession. It is a blunder. The magnitude is very large. I think that, sometimes we just don’t appreciate what sort of stress a recession imposes on not just our economy, but our people. Yet, unemployment is awful. It causes stress within families and all these other problems that are hard to count. In this particular situation, it’s unnecessary.
—
Recession Is Imminent
Tobias: Some of the presentations that I’ve seen you give, you talk about the timing from the inversion. So, you say, October 25, we invert plus 90 days to send a signal. It gets you through to January 25. And you say six months from the inversion is the earliest that a recession has manifested, which would be today, April 25. The average is about 12 months through to October 25, and then as long as 15 months, which would be January 25, 2024. Do you have any view– if we do see a recession, do you need the un-inversion to happen before the recession is declared?
Campbell: No. Though that has happened, it totally depends on duration. So, the yield curve is very good at giving you forewarning of a recession. We’ve already said eight out of eight. The lead time varies. So, it varies in a range mainly, let’s say, 6 to 18 months of lead time. Then there’s another quality of this indicator that is, I think remarkable, and that is that the length of the inversion closely matches the length of the recession.
So, if you do this in months over the last four recessions, so those are the ones out of sample, the difference between the length of the inversion and the length of the recession is only one month. So, it is very accurate in doing that. You’ve got a good lead time, you’ve got a matching of the length. What the indicator is not as good at is not the duration magnitude of the recession. It’s just too much to ask. I give you two out of three with one variable that surely you can deliver a handful of other pieces of information to give the magnitude.
—
Tobias: Well, I wondered if the steepness– It just maybe you don’t have enough examples of– we’ve got eight altogether so far. I’m concerned that the steepness of this one indicates something nastier than ordinary coming down the pike. Can we just talk a little bit about, what is the causal relationship? How do you view the two yields in relation to each other and what that is telling us about what’s coming?
Campbell: Yeah, one simple way to look at it is any interest rate has three components to it. One is expected inflation, the other is expected real growth in the economy, and the third part is the expected risk. So, if we’re looking at Treasuries, let’s ignore the risk. We’ve got inflation and we’ve got the growth. In a very simple way of thinking about it, suppose the inflation cancels out and then you’ve just got the difference between longer-term growth and shorter-term growth, and that’s the mechanics of the growth forecast. Again, this is just a single indicator. One way to think about it is that when the yield curve flattens out or inverts, that just means slower growth.
The causal channels, the second channel is a real channel. So, just think about what’s happening here. Short rates are very high. People moving their money from banks to money market funds, banks having to get a lot tougher on loans. That slows growth right there. And that’s the thing that could push us into recession.
Tobias: Because there’s some research on older– I don’t want to say ancient, but it’s like 1800s, it seems that there was a lot of inversion through the 1800s, because there’s some speculate that hard money, like a gold standard, indicated an expectation of deflation and that was reflected in this constant inversion. Have you seen that before?
Campbell: Yeah. My dissertation actually uses data going back to 1900, but I collected data going back 200 years. When we go back that far, or actually even before 1953, the data are very challenging.
Tobias: The quality of the data?
Campbell: It’s challenging because of illiquidity, challenging because of extreme yield curve control before the Fed-Treasury Accord. Then we go back further, but we don’t have Treasuries. We don’t have it like a Treasury bill. So, people use– [crosstalk] [crosstalk]
Campbell: Yeah. Remember I said, there’s three components to the yield. Well, that risk component becomes much more important. So, some of these countries are risky. It’s not like today where the US is the reserve currency of the world and the safest instrument in the world is the 10-year bond. So, I think that it is much different. Most of my research– [crosstalk]
Jake: That’s what Silicon Valley Bank thought.
[laughter]Campbell: Yeah. Again, we’re talking two different risks.
Jake: I’m just kidding.
Campbell: One is sovereign risk and one is duration risk. They need to manage the duration risk.
Jake: Yeah.
Campbell: Yeah, no, I totally agree. What they were doing, and many other banks just buying those bonds, well, it’s relatively safe in terms of default, but you need to manage the duration risk. They failed to do that. Again, I don’t know how many other banks have failed to do that.
Tobias: People were expecting negative rates. I think it was a somewhat forgivable error, because not that long ago, everybody thought the US was potentially going to negative rates. Most of the rest of the world, the developed world, was in negative rates. So, I think it’s a forgivable error. Not everybody did it though. JPMorgan Chase was aware, M&T Bank was aware, lots of other banks.
Campbell: I totally understand errors, but if you work in finance, you get paid like these people get paid. There’s this concept known as hedging.
Jake: [laughs] Yeah.
Tobias: [crosstalk]
Campbell: There’s a concept that’s related to it called risk management. So, I don’t buy it. This is, again, reach for yield. I do believe that the Fed made this situation far worse when they gave an adverse scenario of, “Okay, well, this is the worst that can happen. The Fed funds rate is 0 to 25 basis points, and the 10-year bond is 0.5%, increasing to 1.5% over a year.” Then, the Fed takes the real world so far out of their adverse scenario, so far away that banks going like this, like, “We thought that we were stress tested and we could survive the worst scenario.” No. So, that is unfair. I’m willing to cut some slack there because of the gross failure of regulatory oversight.
It’s like two things here. So, one is constructing a stress test to construct the mechanism. That’s a regulatory issue. Then there’s a supervisory job that you do to make sure that you give adequate diligence to all of the banks making sure they’re safe. Again, the Fed knew about problems at Silicon Valley Bank well before it went under. They just didn’t do anything about it.
—
Cam Harvey – New Book – DeFi and the Future of Finance
Tobias: Cam, we’re well over time here, but I’d be remiss if I didn’t ask about the book over your shoulder that you’ve got a new book, Defi. Do you want to talk about the book?
Campbell: Yeah. So, it’s DeFi and the Future of Finance. This has been something I’ve been super interested in over the last eight years. I’ve taught this material for eight years. I do research in this area and I’m very excited about the possibility of decentralizing some of the interesting aspects of our financial system and other parts of what we do. It will be good for consumers. It’s a technology of inclusion and of financial democracy. It also challenges all of the monopolies and duopolies that we have today.
This is not just a comment about our financial sector. This goes well beyond, because the concept of Web 3.0 is what we’ve got today with Web 2.0, plus the decentralized finance. So, that reasonable things happen, so that if you get served an ad on social media, then you get paid for it as it should, rather than the advertiser paying the platform and you get nothing. This really changes the way that we interact in many different ways. It has, I think, very interesting implications for finance in general, and that’s my focus. But I do definitely talk about all of these other ideas that should be within Web 3.0. I’ll leave with just one example of the simplicity of this.
Think of the major cloud computing providers. And now think about your laptop or your desktop. How much CPU do you use every day? For most people, it’s not much. It’s maybe an hour, maybe two hours. It’s not that you’re running simulations overnight. It’s just not used. So, why not rent that out? It’s just lying around, so you can generate revenue. It’s just a simple Web 3.0 application for cloud computing. So, many different industries will be shocked by this, even though all the media attention is on the trials and travails of Sam Bankman-Fried, which, by the way, has nothing to do with decentralized finance that’s centralized finance, or the price of Bitcoin or Dogecoin? No, there’s something else happening under the radar that’s not just speculation on cryptos that will affect all of the top names, all of the top companies.
Tobias: That sounds fascinating. Campbell Harvey– [crosstalk]
Jake: There’s an irony too that YouTube is going to serve up 25 ads on this while [Tobias laughs] paying Toby and I absolutely nothing.
[laughter]Campbell: Exactly. Yeah. Not in the future. Not in the future.
Jake: Okay, good.
Tobias: Campbell Harvey, thank you very much for your time.
For all the latest news and podcasts, join our free newsletter here.
Don’t forget to check out our FREE Large Cap 1000 – Stock Screener, here at The Acquirer’s Multiple:
