Episode #172: Cam Harvey, “This is a Time of Considerable Risk of a Drawdown”

Episode #172: Cam Harvey, “This is a Time of Considerable Risk of a Drawdown”

 

 

 

 

 

 

Guest: Campbell R. Harvey is Professor of Finance at the Fuqua School of Business, Duke University and a Research Associate of the National Bureau of Economic Research in Cambridge, Massachusetts. He served as President of the American Finance Association in 2016. Harvey received the 2016 and 2015 Best Paper Awards from The Journal of Portfolio Management for his research on distinguishing luck from skill. He has also received eight Graham and Dodd Awards/Scrolls for excellence in financial writing from the CFA Institute. He has published over 125 scholarly articles on topics spanning investment finance, emerging markets, corporate finance, behavioral finance, financial econometrics and computer science. He is also a Founding Director of the Duke-CFO Survey. He is Partner and Senior Advisor to Research Affiliates, LLC, who oversees more than $200 billion in client investments. Harvey also serves as the Investment Strategy Advisor to the Man Group plc, the world’s largest, publicly listed, global hedge fund provider. He edited The Journal of Finance – the leading scientific journal in his field and one of the premier journals in the economic profession from 2006-2012.

Date Recorded: 8/5/19

Run-Time: 1:47:13

To listen to Episode #172 on iTunes, click here

To listen to Episode #172 on Stitcher, click here

To listen to Episode #172 on Pocket Casts, click here

To listen to Episode #172 on Google Play, click here

To stream Episode #172, click here

Comments or suggestions? Email us Feedback@TheMebFaberShow.com or call us to leave a voicemail at 323 834 9159

Interested in sponsoring an episode? Email Justin at jb@cambriainvestments.com

Summary: In episode 172, we welcome our guest, professor Cam Harvey. Meb and professor Harvey begin the conversation with professor Harvey’s 1986 dissertation on the yield curve, and his finding that when the yield curve inverts, it precedes a recession. His indicator has yet to provide a false signal. He goes on to explain the model, what it really tells us, and the implications as we move late into the summer of 2019.

Professor Harvey then gets into what an inverted yield curve means for growth, and a study he did that describes the performance of various asset classes before and after yield curve inversions.

He follows up with some background on the Duke CFO survey, and the predictive power it has in foreshadowing recession. As of a recent observation, 85% of respondents believe a recession will begin in 2020 or 2021.

The conversation shifts, and professor Harvey gets into some thoughts on cryptocurrency, and the research that went into the creation of his course on Blockchain.

Next, professor Harvey explains blunders in factor investing, from data mining, to investors not taking correlation of factors into consideration.

As the conversation winds down, professor Harvey discusses what he’s thinking about in his research these days, and disruptions he sees coming in finance.

All this and more in episode 172, including professor Harvey’s most memorable investment.

Links from the Episode:

 

 Transcript of Episode 172:

Welcome Message: Welcome to the “Meb Faber Show” where the focus is on helping you grow and preserve your wealth. Join us as we discuss the craft of investing and uncover new and profitable ideas all to help you grow wealthier and wiser. Better investing starts here.

Disclaimer: Meb Faber is the co-founder and chief investment officer at Cambria Investment Management. Due to industry regulations, he will not discuss any of Cambria’s funds on this podcast. All opinions expressed by podcast participants are solely their own opinions and do not reflect the opinion of Cambria Investment Management or its affiliates. For more information, visit cambriainvestments.com.

Meb: Hey, podcast listeners. Welcome. It’s full-on summertime. We have today an awesome show for you. We got a Dukie here. Professor of finance at Fuqua, right down the road from me. He’s trying to escape the humidity. Flew us to…in office, Los Angeles. Professor Cam Harvey, welcome to the show.

Cam: Great to be on the show.

Meb: Cam, I used to have this fantasy and I’m sure most people do about being a teacher and a professor where you just spend the summer on the beach sipping a Pina Colada. You’re out here hustling, writing papers. I mean, I stopped counting at over a 100 at this point. How’s the summer going?

Cam: Yeah, the summer, actually, is a time…interestingly enough people think that you’re in the beach but no, this is the time to do your research and that’s the main thing you get paid for is research.

Meb: I always say that about taking a summer sabbatical and getting writing done and it seems to never happen. I now have a two-year-old, so he’s kinda becoming the problem I think but writing, for me, I don’t know. I feel like summer’s the hardest but we’re here in Los Angeles at the beach so I don’t know. Well, welcome. It’s great to have you. We gotta start out. You might be sick of talking about this point. I don’t know. But we gotta go way back. Thirty years ago, your dissertation…it’s super timely right now for the first time in maybe a decade. Let’s talk about the yield curve. What’s going on? You wanna give us a little heads up?

Cam: Yeah, it is amazing that my dissertation still lives. Usually, these dissertations are collecting dust on some bookshelf somewhere but I get hundreds of emails about what’s happening with the yield curve and my dissertation in 1986 at the University of Chicago essentially provided a theory and some evidence that the yield curve which is the difference between a long-term interest rate and a short-term interest rate, that was highly correlated with the business cycle and actually was predictive of what was going to happen in terms of real GDP growth. So that was what I published in 1986 so a long time ago and it was actually pretty controversial dissertation at the time because I didn’t have that much data. So I’m working with the data set from 1960s to mid-1980s and I show that when the yield curve inverts which means this weird situation where the long-term interest rate is lower than the short-term rate…when it inverts, it precedes a recession. The problem was that there was only three recessions in my data set and my adviser sitting there was saying, “Well, maybe this is just a lucky finding. Maybe it’s just by chance that you see this. Maybe it’s not real.”

Meb: Which one of the advisers was it?

Cam: Yeah. One of the three Nobel laureates on my committee, Gene Fama. So they were skeptical and for good reason because that’s not that much. Three out of three within sample, it could be data mining and you know how it goes. You publish your paper and then the real world happens. We get the out of sample and if you’re lucky, usually, the effect gets weaker. And if you’re in bad luck, the effect completely goes away. You’ve just data mined some finding. So what happened with me was different. We’ve had four recessions since the publication of my dissertation and the yield curve inverted before every single one.

Now you might say, “Well, that’s no big deal. You could have an indicator that says recession every quarter and it’s gonna hit four out of four because it’s constantly forecasting a recession.” But it’s interesting that there’s been only four inversions, each preceding recession. So what’s extraordinary is the seven episodes that we’ve had, this particular indicator has not yet, and I would like to emphasize the word yet, had a fault signal.

Meb: I would almost want it to have one, just say, “Okay, it’s not perfect. So it’s not a 1,000%.” But it’s summer 2019 and it’s tripped, right?

Cam: Yeah. So I’ve been watching this pretty carefully. In my dissertation, I looked at two yield spreads. One was the five-year yield minus the three months Treasury bill. Then it was the 10 year minus the 3 month Treasury bill and noticed that I’m using the 3 month Treasury bill as the short term yield. Some people use a longer-term rate and I do think that that’s data mining. You might get a slightly better fit using a different piece of the yield curve but I’m just not impressed with that. So what if you get a better fit with the 11-year yield minus the 7.5-year yield?

There’s a million different combinations here. Wanna go with the spread that was published in my dissertation. It’s got a long out-of-sample track record and there’s no indication that it’s broken but as you say, it is fired right now so I’d look for a full quarter on average for an inversion and June 30th, that was the full quarter and both the 5-year spread and the 10-year spread were on average negative for the quarter. The quarter is the measurement point because GDP is measured quarterly. So it kinda makes sense. So the inversion is fairly mild but it did trigger a forecast.

Now I would like to clarify one thing that’s important. My model links the slope of the yield curve to future real economic growth. So empirically, we’ve seen that it predicts each recession but the actual model links the slope of the curve to grow. So it doesn’t matter if the curve is inverted or flat or slightly upward sloping. All of that means lower growth. So if it’s the case that we go into a recession in 2020, or 2021, okay, well, that will be eight for eight. But if it’s the case that we have substantially slower growth in 2020 or 2021, let’s say 1%, then the models worked. The original model is about growth and 1% or half a percent, that’s way below what we’re used to in the high twos or three range.

Meb: And then explain why you think the economic theory works. Why is this something that you mentioned is not just data mined but is actually…it would make sense that this is the way the world should be.

Cam: Yeah, there’s many different ways to explain the basic theory and one of them that I like the best is this model reflects what people believe in the market, so it’s not like a causal model. So yield curve doesn’t necessarily cause a recession with one caveat that we can talk about maybe a little bit later. It reflects what people think. So when people get nervous, they’re worried about what will happen in terms of the economy often there’s a flight to quality. And the safest asset in the world right now is the U.S. 10-year Treasury bond. And people start buying that bond, the yield goes down and that is the action that you need to cause a yield curve inversion.

Meb: I mean, I think not too long ago a lot of people were talking about yields…the 10-year going up to 4, 5, 6 and we’re now back down to I think 1.7 or something, something around there. So for the listeners maybe saying, “Okay. The yield curve forecasting recessions. How might this play out in sort of real-world application of asset management and returns?” And you guys have done some additional work partnering with our good friends, research affiliates and others that it turns out looking at the returns of the stock market before and after an inversion is actually pretty instructive, too.

Cam: Yeah, so I guess the issue is what does this mean. So, you know, and I think there’s two broad ways to look at this. So the first thing has to do with this idea of potential causality. And let me elaborate on this before we actually deal with how do you adjust your portfolio. So originally when I wrote my dissertation nobody noticed. And people noticed a little bit in October, 1987. And that was the crash of the stock market. And the consensus forecast for economic growth in 1988 was negative. So people just assumed there would be a recession and it was gonna be a bad one. And at that time I went on the record and forecast 4% real GDP growth in 1988 and that’s essentially what happened. So a few people noticed.

Then we had these two recessions, 1990, ’91. It was pretty mild. And then the 2001 recession was such a minor blip that we didn’t even have a year over year negative real GDP growth rate. But it was the global financial crisis. The yield curve inverting well before the global financial crisis that…

Meb: It was like…it was six, right?

Cam: Yeah, and people really noticed that. So okay. This model’s hot. So now I get remarkable publicity about the model and you wonder if there is the possibility of a self-fulfilling prophecy. So people talking about it. Oh, yield curve’s inverted. We’re going into recession so I’m not gonna pull the trigger on capital investment, I’m not gonna pull the trigger on new hiring, people cut back on their spending and that actually puts us into a recession. So the way to actually look at this is the following. It’s basic risk management.

So given that you’ve got an indicator that is a reliable indicator with no false signals should you ignore it? I don’t think so. So should you borrow a lot of money to build that new plant? Well, why not wait to see how this plays out? And yes, it is possible that this leads to slower growth. However, as a company, there’s two ways to deal with this slower growth. One, you borrow the money to make that big capital investment. We go into recession and you’re out of business and everybody’s laid off or you’re cautious, you don’t go ahead with that investment and you survive. So it’s basic risk management. Consumers, the same thing. This is not a time to borrow on your credit card to go on vacation to some exotic spot. No, this is a time to eliminate your credit card debt and to save.

So there is this aspect that there could be some self-fulfilling prophecy that’s linked to this and the reason that I’m previewing this idea is very much related to what you asked me about…okay, well, what do you do with your portfolio? So it’s not just a matter of capital investment or hiring with the corporation or a retail investor basically getting their house in order. What about their actual portfolio? What do we do? This is a signal about the future growth in the economy. We know that stock returns are linked to growth in the economy. Think about the value of the stock is the expect the cash flows in the future and then we discount those cash flows by whatever appropriate risk.

So the expected cash flows are impacted by the growth in the economy and obviously, the risk is also impacted, risk goes up if you go into a recession. So those two forces push down valuations. So what I did was the following exercise and it’s a LinkedIn article that I just put up and maybe we could link to it.

Meb: Yeah, we’ll put it in the show notes.

Cam: Yeah.

Meb: All the stuff we talk about today, we’ll throw in the show notes.

Cam: Yeah. So this is the idea. It’s a really simple idea. And look, there’s not that many of these recessions and yield curve inversions but it’s a very simple idea. Let’s look at what happens if you invest in the market after an inversion. So I do the quarterly inversion. So in the most recent episode, it would be you invest after June the 30th, 2019. So even though the yield curve was inverted before June 30th, I wait for the full quarter just in consistency with my model.

So I do that and then hold the investment for three years. And then I also look to see what happened in the three years previous. So then I do that for every single yield curve inversion. We sometimes call this in finance as an event study. So time zero is the point where you’ve observed the yield curve inversion and then you take a position. And we look at what happens three years after and three years before. So not surprisingly, if you look at what happens before the market’s doing very well before the yield curve inversion. And then if you look at what happens after, the market does poorly.

So this is on average and it’s also the case that every inversion is different. So obviously what happened during the global financial crisis is way different than let’s say 2001 or 1990, 1991. But on average, it suggests that this is a time of considerable risk of a drawdown in just a plain old long-only investment.

Meb: And if I remember correctly from the article there was two pieces that stood out. One is that…I mean, the real return over the next year…this might’ve been that of treasuries but it was something like minus 9% and there was an additional piece you went to the end where you also looked at the effective value where value is an additional tailwind which is an interesting point of view because almost any way you measure it, value’s been fairly terrible this cycle for the past 10 years. So who knows? It could be the signal to say that value may reclaim some of its former glory.

Cam: Yeah, so it’s interesting to actually think about that. And we’ve seen that play out. You’re exactly right. Over the last 10 years, value has done poorly. Do we believe that that factor will always do poorly? I don’t think so. It’s got a very long history and it’s also got a very long history of having like a bad run. So think about the 1990s where value was underperforming year after year after year, and then in the late 90s people eventually said, “I had enough. I’m gonna bail out.” Exactly at the wrong time.

So you’re correct. I did the same study where instead of looking at the fully funded market return. So market return minus the borrowing cost. I looked at the value factor and how it actually performed in the same sort of analysis and it was striking. So it’s the complete opposite pattern. So before the inversion, value’s doing poorly and it’s kind of consistent with what we’ve lived through for the last 10 years. And then after the inversion, it does well. So the way I look at this is that it’s very important to have a sense of timing. So it’s one thing to be exposed to factors but it’s another thing to move your exposure in a way that’s consistent with some notion of timing. So there are certain times in the business cycle where it’s great just to be long-only the market or in growth. There’s other times that it’s great to be in value or some other factors.

So this yield curve indicator is a piece of the puzzle. I think it’s a very reliable way to forecast economic growth and we know that all of these factors are impacted by economic growth. So why not use that information? And it’s not just value but there are other factors. I’ve got another LinkedIn article I’ve not posted yet but I will preview that I’ve done the same exercise for other factors and again, there is a variety of different performances across the individual recessions, but this is the time to have some of your portfolio tilted towards things that will do well when the market goes down.

Meb: And I would think as an extension that would probably, if I was a betting man, guess it applies to probably the safer sectors. Maybe like consumer staples and also potentially some other asset classes. I would guess that maybe gold or commodities would do well during an inversion. I’m not sure. I feel like I’m recalling some research there somewhere. I don’t know.

Cam: Yeah. So you’re correct. The quality…anything that’s quality does well after an inversion on average. And again, I’m talking inversion and just assuming that it’s associated with the recession because that’s what’s happened historically. I’ve not looked at the commodities, but I do know for my research on gold that gold’s just all over the place. It’s not a reliable hedge. So it’s only a reliable hedge over extreme long term.

Meb: We had Claude on the podcast a while back so we spent a lot of time talking about gold and some other ideas. So this is interesting. I think there’s a lot of areas for research on the yield curve ideas most of which…the thing I like about it is that it has grounding in theory and it somewhat makes sense. Why don’t we sort of transition a little bit away from the curve? Oh, actually, there’s one more thing I wanna talk about on this before we transition. You mentioned a fair amount of ideas and concepts around it being people’s emotions and beliefs and whether it becomes a self-sustaining prophecy. One of the things you started at Duke was starting to survey…is it CEOs or CFOs? I can’t remember.

Cam: Mainly CFOs but some CEOs.

Meb: Why don’t you talk to us a little bit about that and what they’re all saying today as a potential bridge of kinda what we were just talking about too?

Cam: Yeah, so if your listeners don’t know the survey, they should. It’s called the Duke CFO survey and it’s been going on for 25 years. And it’s somewhat under the radar because we don’t have like a PR agency promoting it but it is a survey of CFOs and the CFOs are the ideal people to ask about things like future economic growth. The survey of economists that are done, those are the wrong people to ask. They don’t know what hiring plans are over the next year. They don’t know what the CapEx plans are over the next year. Indeed, they might know a very limited amount outside of their institution. So these CFOs are on the ground. They, again, know the plans before the plans are actually executed and they give us advanced knowledge of what’s happening. And again, an interesting validation of our survey. Go take a look at the Fed minutes that are online and search for Duke and you can see how often the Fed actually talks about our survey.

So this survey again has got a long history. We usually have about 400 U.S. CFOs but it’s a worldwide survey so we’ve got well over a 1,000 CFOs around the world that provide information for us. And we ask them about their hiring plans. We ask them about their employment, their pay out plans, all this stuff but we also ask them about a recession. And the responses recently had been striking. So you’ve got 85% of these CFOs believe that a recession will begin in 2020 or the beginning of 2021. And that’s consistent with the message from the yield curve because the yield curve…it inverts. It doesn’t mean a recession starts immediately. There’s a lead time and the lead time could be 9 months, it could be 18 months. That’s not consistent in terms of the lead. So the CFOs are essentially saying the same thing, that they believe that there is great risk that we will go into a recession or at least a slower growth.

Meb: What’s the normal sort of response? Is it like half of them would say it’s in the next couple of years or 20% or is it all over the place? What’s the…

Cam: It’s not all over the place. Usually, they’re quite optimistic and we’ve got all sorts of things that we track in terms of their optimism and it isn’t all over the place. So again, we’ve got 400 of them. If it was one or two, it would be all over the place. But 400 of these CFOs and we measure this fairly carefully and it is very predictive. So we’ve done the analysis. We’ve got enough data. So early in the survey, we couldn’t do this but we’ve had 25 years of data. So does the CFO survey predict CapeEx spending growth? Yes, it does. So does the CFO survey predict the consumer confidence index? Yeah, it does. So again, bad index…let’s say the consumer confidence index or index of leading indicators. Those are all supposed to be leading indices. So we actually predict those leading index.

Meb: They tend to be helpful on the flip side, too, when times are dire. So I think back to the financial crisis. Are they pretty good at starting to see on the horizon that times might be getting better or they just become super dower and depressed as well then?

Cam: So let me talk about the yield curve on that as well as the CFO survey. So the yield curve is really good at predicting recessions. It’s good at predicting the duration of the recession. So for example, the yield curve was inverted in the global financial crisis for exactly the same amount of quarters as the length of the global financial crisis. So I got the timing of the crisis, I got the duration. What the model isn’t that good at is the severity of the crisis. Indeed, it was comical to me that somebody wrote an article that was highly critical of my yield curve forecasting and pointed out something I wrote in 2006 predicting the global financial crisis, predicting the duration of the crisis but what I didn’t predict was the severity of the crisis. So I did a poor job. I only got two out of three. Two out of three with a single variable. This isn’t like a 1,000 equations and stuff like that. It’s a single variable.

So the duration you can get from this particular model but I would also say that turning points are obviously very difficult to forecast. And the yield curve appears to be a fairly good predictor of both types of turning points. So one turning point is where you go into a recession and the other is where you actually get out of the recession. But, and this is very important, the turning point where you go into a recession is way easier to get wrong. So it is very difficult to call that particular turning point whereas the turning point where you leave a recession, that’s fairly easy to do because recessions are short. So the recession might be three quarters, it might be four quarters, it might be a little longer. So in the recessionary period, you can be wrong by a little bit getting back in but you’re not gonna be wrong by that much. Whereas forecasting the turning point down, you could be out of the market five years too early or you dump your value stocks exactly at the wrong time. So there is an asymmetry that’s really important here and I guess it’s a long way of saying that the turning point where things recovered that’s a lot easier to forecast. Indeed, you don’t even need a model for that. We just know that these drawdowns are not long in duration.

Meb: Yeah. While we’re on the topic of interest rates I had a quick question before we move on in yield curves. A lot of talk in the media today about negative-yielding bonds. As professor with your ivory tower hat on, is this something in the development over the past 15 years that you’ve seen as a surprise? Is it something you see as a normal state of affairs? What’s your general just kinda thoughts on the way the world looks today with…I think it’s not just now sovereigns but also even some corporates, etc.

Cam: Yeah, so it is a different world that we live in but we’ve seen negative yields before, and in my time, the classic negative yield was investing in a Swiss bank account or a bond. So you would have to pay for that and it kinda made sense because Switzerland was the safest place in the world and you’re willing to pay to keep your money there. So these negative yields are not historically unprecedented. What is historically unprecedented is that close to 60% of the sovereign bond yields in developed markets and some fringe developed markets have a negative yield including, and it’s striking, last week the German 30-year bond went zero and slightly negative. So to explain this is complex but the other thing that’s important just in the background is that yield on the U.S. 10 year, that’s looking pretty good.

Meb: Believe me we have a high yield sovereign bond fund. We had to change the name. Excuse me. It’s something like that but it’s basically a carry bond and it’s been bunch of emerging markets and then the U.S. is one of the highest yielders or was, I said before this massive rally. I don’t know if it would still make it in there but it’s funny to see.

Cam: Think about it. That…so the German 10 year negative 0.5% approximately. And again, that has to do with quality. So people believe that with all the stress that the EU is going under that if they’re gonna make a bet it’s going to be in Germany, the strongest economy amongst all of them. So part of it is a flight to quality. The other thing that can’t be ignored is the interventions of these central banks which has been extraordinary in trying to kind of micromanage what’s happening in the economy. And I think that sort of distortions that have been induced we’re going to pay the price for later.

Indeed, as you said, the people…oh, well, I’m not getting any yield on my government bond, so I will increase the risk. And that’s fine for certain people and I’m certainly an advocate of investing in a diversified portfolio with emergent market bonds and things like that. They should definitely be in that diversified portfolio. That’s not what I’m talking about. I’m talking about the person that socks some money away for their pension and they dramatically re-tilt their portfolio towards high yield EM debt and then when the episode happens, they’re walloped with a massive drawdown that they didn’t expect. So they are essentially mismatching the risk that they should take and it’s induced to a large degree by this distortionary policy.

Meb: And speaking of policy, yeah, we’re taking another left turn. I mean, one of the struggles we have a lot here is thinking about how to align long-term investing success for the vast majority of people in this country, incentivise people to be investors, incentivise people to participate in the growth of not just domestic but world economies. And we really transition from a world of defined benefit to…you’re kind of on your own in many cases.

Cam: That terrible mistake in my [inaudible 00:32:01]

Meb: Yeah, and so as you talk to your students, as you talk to…whether it’s policymakers and everything else…I mean, we spend a lot of time trying to think of…I mean, we manage public funds and we see people behave poorly over and over and over again in our funds and I said…I’m trying to think of behavioural ways to say, “Is there a way we can lock people?” And then not from a selfish point of view, but is there a way to get people to behave better? And so in some cases it was like…almost like the concept of an annuity or target date fund but without the massive fees. I mean, I think the fee on average annuity is like 2.25. So it like destroys all the benefits of it. Do you have any general thoughts before…I’m sure you do. Any thoughts on sort of this…if you were giving advice to Congress or the president or anyone else? Any general thoughts?

Cam: There’s a lot of research on this in finance and these defined contribution plans have been a disaster. And that kinda makes sense, right. That people aren’t necessarily experts in portfolio management. They’re working their job. They don’t have the time to invest into learning about portfolio theory. I’m sure they could do it but they don’t have the time.

So you give them this option…you do the 401(k) and it’s usually a disaster in terms of the outcome. The research in finance has looked at other countries and what they do and one model…there is a diversified portfolio that is the default and that people strongly believe that that is not the portfolio they should invest in fine. They had the option and if they lose their money it’s their fault. So I do think that we need to help the general public on this and it makes sense to do something like that but overall it has not been a good experience and I’m afraid that everybody’s gonna pay the price because these portfolios, in the future, will not deliver what was expected at the time of retirement and we’ll have to pay directly or indirectly. So the cost is very heavy.

Meb: And we don’t even have time to get into all the pension funds that still expect 8% returns every year to infinity. But yeah. It’s something we struggle with and think a lot about. Listeners, if you got any good solutions, send it my way.

Cam: It’s also the case that there are certain assets that if you’re young you should be investing in that have the high expect of returns. Indeed, ideally, you’re borrowing at that age and investing in high expect of returns.

Meb: Well, the young people get the first part of that right. They get the borrowing, not necessarily the investing.

Cam: Exactly. They borrow to spend rather borrow to invest. And it’s kind of also interesting to me that a lot of the flexibility to invest in growth companies has been removed given that just the number of stocks has gone down dramatically. So I do believe that that will change, that in the future we’ll have the ability…just the average investor to take a piece of like a venture fund or something like that. And there will be, like, new assets that will emerge in the next five years which will help because you really need to get some of your portfolio into that really high expect of return tranche and right now it’s just not available to the average investor. It’s not even available to some of the larger investors but it’s definitely available to the institutional investor and the very rich that are the first in line to get a piece of Facebook before it went public.

Meb: You know, and the accreditation rules are somewhat odd and I sort of oscillate on my belief here and there’s talk of removing them. It’s almost like I would prefer instead of it just being a…you qualify because you’re rich. It’s like you almost have to take like a DMV test. Like you take this test, fine. You can go throw all your money down a hole but given the fact that the state of sort of personal finance education in this country of people investing so much in lotteries versus investing in their retirement versus…it’s almost, because we don’t teach personal finance in high school or even in many colleges. The ability to say, “Look. Let’s not make it an accreditation requirement. Just not benefit the rich. But let’s make it almost like here, take a test and if you pass, you can light your money on fire if you choose to.”

And you’re starting to see a lot of the blurring of private and public markets I think where a lot of the…what used to be more IPOs are staying private longer but it does seem to be opening a little bit to becoming a little more liquid. Who knows?

Cam: Yeah, no. we’ll see a lot of liquidity. In terms of what will happen in the future is that we’ve got this ancient system that’s lasted for over a 100 years where to buy a stock today and get official custody of the stock takes two days. That’s ridiculous. So it shouldn’t be a settlement time of two days. And the number of middle people is enormous. Everything will be tokenised and these tokens will be available to all investors. And you’re right. Some investors might abuse that but if it’s within a bucket of 10% to 20% where you can invest in your favourite start-ups and yes, that is kind of like a lottery but it’s over a long period. So you’re young. You’ve got 40, 50 years to ride out a lot of these investments going to 0. So I do think that it’s important.

I think it’s important for the economy also. We need investment in innovation. Right now, it’s really hard to do. Many of the entrepreneurs, their first source of financing is borrowing under a credit card. So that is looking enormously expensive. At 22%, 24% interest rate and you think about…they’re looking in a project that looks pretty good. They can make 15% a year, but they don’t do it because the credit card is gonna cost them 24% a year. So they let it go. So all these projects that are really exciting projects are not pursued because the financing is just not available right now. In the future, it won’t be the same. You won’t have to IPO. You won’t have to go to your bank and go through all the hoops at the bank. It’ll be a more peer to peer…it will allow for more innovation. And I see that not in the long, long term. I see that in the next five years.

Meb: And two quick comments here. One, what was sort of the origins on you becoming interested in sort of the blockchain, smart contract space? Was it student-driven? Was it just reading the academic literature? And second…this is a two-part question. Totally unrelated. Would love to hear your thoughts, if any, on if you love, hate the idea of income sharing agreements where a lot of these colleges are starting to do it for a set of students taking on a bunch of debt where they’ll essentially be able to give back a little bit of their income. There’s some companies like Lambda School that’s doing it as well as other companies outside of the education system doing it. So anyway, that’s complicated two questions.

Cam: So the origins of me being interested in blockchain and that’s what I teach at Duke University, it’s a course called innovation and crypto ventures. Six years ago I retired as editor of the “Journal of Finance.” I hadn’t taught in six years. And I decided to revamp my investment course and it was an international investment course. There was a section on FX and I figured, “Well, I’m doing the euro, the dollar, the yen, dot, dot, dot. What about this thing called Bitcoin?” And I heard a little bit about it. So I decided to add a section on cryptocurrency and at the time that was the only blockchain application. And I started to prepare this two-hour lecture and immediately realized that this was way more sophisticated than I thought.

So I heard a lot of misinformation. I started reading the background and about the structure of the basic idea. I read about all of the failures of digital currencies in the past. We’ve been researching digital currencies since 1981 and every single one failed and they failed for a very simple reason that if you think of a…like a photograph or a music file, a video file. You can make a perfect digital copy. So for currency, you can make them perfect digital copies. So you need some sort of ledger that keeps track of the serial number. Think of it that way. And that’s what this technology actually does. It makes it possible to have a digital currency. So very exciting. I did that two-hour lecture. My students…I thought I bombed in the lecture because they just sat there. Usually, they’re out of there as soon as we hit the hour but they’re just kinda sitting there. And then they came up to me and said, “That should be a course, not a lecture.”

And I reshaped it into a course and the course is a blockchain course, so I talk about cryptocurrency but it’s not a huge part of the course. It’s on the other applications. So contracting, eliminating much of the back office, supply chain applications. So there’s hundreds of application to this technology that are emerging, that are really exciting, that I think you really have to be in the space. So that’s the origin of my teaching and now it’s a 100% blockchain teaching.

Meb: That’s awesome. Is enrolment in 2019 like versus 2018?

Cam: Yeah, so very interesting. Yeah, you know. The enrolment was highly [inaudible 00:41:59] price of the cryptocurrency. So in 2018, I had 231 students. Half of the class at the Fuqua School of Business at Duke. However, unfortunately, many of those students were there for me to teach them how to become a cryptocurrency millionaire or a billionaire, right. They wanted that Lambo. And so my course is not just a cryptocurrency course. Again, it’s a blockchain course and it’s very broad. It brings students not just from finance but from strategy, from marketing, from accounting and it’s very diversified in terms of the application both for-profit applications, non-profit applications and it grows every year. But let me get to your second question.

Meb: Which by the way again, totally unrelated. It was in my brain so I had to spit it out before I forgot.

Cam: Yeah. So we actually do in my course. In my course, the grade is determined by a 15-page pitch deck for a company that is using blockchain technology. So a number of groups have presented ideas that are related to having a register for a particular asset. So it might be a painting, a very valuable painting that there’s no way that the investor could actually own that but if you tokenize the painting then you’re gonna have many people actually owning it. There are many other applications that we look at but one of the applications that constantly comes up is human capital. So human capital is the most important asset in investors’ portfolios. Yet we often completely ignore it. And we ignore it because the data is hard to come by.

For example, we think of having a diversified portfolio and let’s say that you work in the mining sector to give an example. And I give that example because it’s a very pro-cyclical sector. Your portfolio shouldn’t have any mining stocks because you’re already invested in mining so…and this holds for individuals. It holds for large institutional investors. The giant one trillion dollar Norges Norway Pension Plan, it basically is funded by oil and gas revenue and they shouldn’t be invested in oil and gas. So this idea that you need to look at your portfolio as a whole, an important aspect of your portfolio is your human capital and that needs to be taken into account when you actually design your portfolio. It almost never is.

Meb: So you’re saying you shouldn’t put all your money in your company’s stock in your retirement plan?

Cam: Okay, so that, indeed, you could argue that you should short but we really can’t do that because it would create a perverse incentive. So yes. You’d need to have…one of the most basic things in finance is diversification and this is diversification but what I’m saying is that you need to look at all of your assets. So when we talk about the market portfolio people think of that as, “Well, it’s the equity portfolio.” No, no. It’s not just equity. It’s credit. It’s real assets like real estate. It’s human capital. It’s just really hard to measure that human capital. So there had been some experiments in the past and the most famous experiment of monetizing human capital is the so called…so it’s David Bowie. He issued a bond that was collateralized with royalty payments from his library and that bond…I forget the exact maturity but it sold well and he was essentially able to get the money upfront from the future royalties and the bond matured and everything went fine.

So for David Bowie to do this as already like an established performer, that’s one thing but it’s another thing for somebody just starting their career to say, “I’m gonna sign off 10% of my income to pay for my college.” When that 10% of the income…if they do really well could be millions of dollars. So we need to be like really careful in actually designing this but I do believe…and again, it’s linked to my course because we’ve got many more applications. Again these applications are usually people like performers whether it’s entertainment or sports that have the ability to monetize their human capital. So we don’t do a good job of this in two aspects. The first aspect is people ignore this for their portfolio and make a fundamental mistake. And the second is that we don’t have a very good mechanism to monetize the future value of that human capital but we will and that’s where this tokenization comes in. it proves a way to do this at a fairly low cost.

Meb: Yeah. One of the papers we had written was kind of along these same lines of specifically to the asset management industry making the argument that most financial advisers are like four times leverage the stock market. So they own stocks in their own portfolio which is most of their investments because stocks are more volatile than bonds. Their clients are all invested in stocks. So the company’s revenue stream is exposed to that. Often clients will panic when times are really bad and so withdraw during a recession or depression. Shouldn’t be if the financial advisers [inaudible 00:48:08]. And then lastly, if you work for a company that’s not your own company like at Wire House, like Morgan Stanley or Lehman, you’re exposed to that company being exposed to the revenues. And so I said theoretically like an airline would hedge fuel, the cereal company would hedge wheat. Theoretically financial advisory…you can make the argument that they shouldn’t own any or even be short and I think it’s our least-read paper and no one has any interest but that’s something we’ve thought about here quite a bit.

Cam: It’s so basic. That’s the thing that it’s remarkable to me that the idea that you need to be fully diversified and look at all of your assets, that’s a basic idea. Yet it’s like totally orthogonal what happens in the market.

Meb: I think that if you applied it to like housing, people can kinda get. Like if you’re, “Hey, you’re a real-estate guy. You shouldn’t necessarily be buying a 100 houses.” That’s a little more tangible than I think the investment market. But whatever. It’s a…we’ll post it in the show links. I don’t think anyone will [inaudible 00:49:09]… We heard of a couple of funds that reached out and said they’ve implemented some ideas there but…

Cam: Yeah, it’s a basic mistake. That’s the thing. And what is remarkable to me that there are so many basic mistakes that are made by investors…and again, I’m not just talking about a small investor. I’m talking about large investors thinking about things just really incorrectly.

Meb: What’s your favourite? This’ll be a good transition to the three blunders, but do you have a favourite mistake that…I have a long list but chasing performance is probably number one for me. That’s up there.

Cam: That’s pretty up there. But how about this one? I get a phone call from a former Duke student saying that they’ve got this fund and it’s really simple idea. They’ve outperformed 10 years in a row and wonder if I wanted to become an adviser. They’ve raised a huge amount of money. So I said, “Well, what’s your strategy?” He’s like, “Well, what we do is we just roll the S&P 500 futures and that gets us beta of one and then we write some put options and collect the premium. So we collect about 200 basis points of premium every year and we outperform the market by 200 basis points.” And I said to this former Duke student, I said, “You’re a former Duke student but I don’t actually remember that you were in my course.” And he said, “Oh, you’re right. I didn’t have time to take your course.”

So you would’ve failed my course because that isn’t outperformance. That’s taking risk and yes, you make a premium over the benchmark of the market but that’s not the right benchmark. You’ve got this tail risk in writing that…if the market goes down severely you’re gonna take a massive loss. So you are not looking at this properly in terms of risk adjustment. You are not outperforming. You are basically just getting paid for taking the risk that is involved with writing of the option. So things like that are just routine. We see this all the time that option writing gets mixed in just before the end of the quarter so it doesn’t show up on the filing. So people basically taking this downside risk that isn’t that visible until it happens.

Meb: And option selling is like…it pops its head up like every three to five years. You see a fund going around marketing itself with a sharp ratio of two or three that just writes a bunch of options. Then the markets change. Sometimes [inaudible 00:51:49], sometimes S&P and every like three to five years we write an article about it but it’s funny. The nice thing about having a blog is you can go back and it’s like a diary from 10…over 10 years ago now at this point and we…all the option sellers…we had a huge chart. This is just waiting to blow up fast forward, none of them exist anymore. And I think there’s ways to do it absolutely that…

Cam: Or you could do it tactically. So I have no problem with that. So you look at particular options that might be underprized and you buy them or overpriced and you sell them or you put on an option sell or buy at a particular point in time tactically. So there are plenty of ways to actually do this but what I’m talking about are the people that just don’t get it that this is just taking risk. And when you take that risk you get a reward. That is not alpha. The alpha in…what I described as zero and it’s not outperformance and it’s striking that you see this so often.

And look, people look at sharp ratios. That’s another example where, as you mentioned, somebody’s got the sharp ratio of two, looks great. And then another strategy has got a sharp ratio of 0.5. Well, if you look at it more carefully their sharp ratio looks at the expected excess return divided by the volatility. Volatility is a symmetric measure. What I care about is not just volatility. I care about the downside risk also. And it’s no surprise. The style that’s got the highest sharp ratio has also got the biggest downside exposure. And the one with the smallest sharp ratio has got the biggest upside exposure. So given that people like the upside exposure, they don’t like the downside exposure, that will skew the sharp ratio. So it’s a basic thing that again it is so prevalent not just small investors but large institutional investors going down the sheet, looking at the sharp ratios and so…oh, I like that strategy. Without thinking about what does the tail property of their portfolio…

Meb: It was like the old Texas hedge. You buy the futures, short the puts and buy a ticket to Argentina. In case it doesn’t work out you’ve got a ticket to somewhere. That’s funny. I could probably go talk about my pet peeves. A lot of it particularly at this point in the cycle of the U.S. stock market romping and stomping and really just creaming everything else in the world. So much we spend time on this podcast talking about expectations or you have people talking about expecting somehow 20% return, sharp ratios. You basically can’t find the best performing hedge fund investment managers of all time or sharp ratio of even above one is largely unsustainable over very long periods. It’s just really hard to do. And as are 20% returns of course.

Let’s talk a few more blunders because you wrote a paper about it with our friends at Research Affiliates, “Alice’s Adventure in Factorland.” You wanna tell us what that’s all about?

Cam: Yeah. So the idea of the paper…we look at three mistakes that investors make with respect to factor investing and let me go through some of these mistakes. The first one is really basic. There are so many factors out there. There are thousands of ETFs offering factor exposure. In my own research, I’ve documented 500 factors that have been published in the academic literature and make the point that that understates the number that have been discovered because many of them weren’t published because they didn’t work and let’s put work in quotations. So with all of these factors, it is obvious that some of them are just purely data mined. So investors need to be especially careful that you get somebody that is proposing a new factor, offering this great premium, it looks great in the backtest but at a sample, it will likely fail because so many things have actually been tried.

And it even…we see the same sort of phenomena within kind of a no one factor. So somebody’s got a better value factor. So how do they come up with the better value factor? They looked at 50 or 60 different combinations of things that represent value. They put weighted average of certain things that work together and it outperforms a plain vanilla value factor. Do I have confidence that it will outperform in the out-of-sample period, in the live trading period? No, it has been data mined. So the amount of data mining is enormous. There’s organisations. They don’t care. They will put out 300 different factor ETFs fully knowing that a good chunk of them have nothing. So they’re essentially faults but they hope that a number of them will look good. Those will be continued, the other ones dropped and you know how it goes.

So basically, the loser here is the investor. You need to go head on realize that the data mining is severe. Some of these more recent factors that have emerged are probably faults.

Meb: Do you have a favourite nonsensical one?

Cam: Yeah. Yeah. I’ve got…

Meb: It’s funny when we talk about this because you say, “Look. Almost everyone is working from the same CRSP data set. Almost everyone has a lot of PhDs.” And so if you go type in even in some of the multifactor funds, the ones that…you know, if you’re using value composites legit you’ll see AQR, LSV, [inaudible 00:57:30] the same thing. So to find a factor that’s really, honestly unique it means it has to be either…probably you have an entirely data set that no one else has which is pretty hard. Otherwise, everyone is trying to do it or you have a particular insight which let’s be honest is pretty hard and rare to find something that is also not gonna be like just co-correlated to another factor. But were you talking about the letters of the alphabet? Was that your [inaudible 00:58:00]?

Cam: Yeah, let me tell that story. So this is my favourite factor that I again reduced in my presidential address to the American Finance Association. So I showed this exercise. So to use the so called CRSP data. So all the equities in the U.S. from 1963 and it showed cumulative returns for a strategy that I said was a new factor and I said, “Look. I was really careful in doing this because I split the data. So I used half the data to engineer this factor and then the other half to validate.” And if you look at the cumulative returns it’s a straight line going up. There is a few blips but every year looks really good. It’s got a great sharp ratio. The global financial crisis? You can’t even see it in the data. You can’t eyeball it in this particular strategy return.

And then, on top of that, this strategy has no correlation with the market return. It’s got no correlation with value, momentum, size. All it’s got zero beta across all of the well-known factors so essentially this is pure alpha and people looking at it…and even at the end, often at the very end you see a training off a performance. No, not this factor. So people very excited about this, that I’ve introduced something that will revolutionize perhaps finance and then I told them what it actually was.

And what I did is I looked at all of the stocks in the universe and basically formed portfolios based upon ticker symbols. So think of the A portfolio is all the stocks that have A as the first letter of the ticker. And I also did this with the second letter and the third letter and this great portfolio was short the stocks that have S as the third letter of their ticker and go long the stocks that have U as the third letter of the ticker. It’s great. It just knocks the ball out of the park. It is completely bogus. And this is exactly what’s happening. This is just an example of data mining where there’s no foundation whatsoever. And look, I used tickers but there’s another paper out there that looks at 2.1 million different combinations of things to try to see what works best in terms of a factor. And the numerator of this factor kinda makes sense. That has to do with earnings per share but the denominator is rental payments that the firm owes four years out.

So it’s a fundamental variable. It’s not like a ticker symbol and it just does amazingly well. Again, it’s useless. Nobody would put money on something like that. You can’t tell an economic story but we’re faced with this problem that people are selling stuff. Often, there’s no economic foundation to the particular investment or the foundation is…it comes after the fact. So they find something by data mining and then come up with a story. That’s not the way research is done. You come up with a plausible economic story. Then you go to the data to see if it’s supported in the data. No, people are basically data mining. They’re torturing the data to try to find something that they can sell. They will wrap it with some story and it’s the investor that’s the loser. They’re almost surely disappointed.

Meb: What’s not surprising about a lot of those funds is they tend to also be high fee. Right? Is they almost universally tend to be expensive. You don’t see too many really…look, all the…excuse me. This is changing a little bit in the ETF world today as there are some really low-cost funds but you tend not to see a lot of the slick ideas come out as a 10 basis point fund. They tend to be one and a half, 2% often.

Cam: And that pricing could be set because the people that have developed the idea are very skeptical of the idea. So you might as well get the money while you can and the fees are set high initially and it’s kind of a faults sort of portrayal of quality because often you set fees high. That support your story. Oh, well, this is really special. And yeah. We can set fees high because we can. And if you don’t want in, fine. And so there’s a lot of game theory going on also but I do think that the actual people running the fund are themselves skeptical. They’ve got a short sort of life and [inaudible 01:02:51]

Meb: You led into my favourite stuff which is percentage of mutual fund managers that own zero dollars of their own fund and in some categories it’s like 50% to 80%. It’s the most…and I used to joke. I said, “I don’t understand why this is.” Then I said, “Actually…” One day I woke up and I said, “Oh, it’s because they’re smart. They know they shouldn’t be investing in this terrible tax, inefficient, expensive vehicle that they themselves don’t believe in.” So I think that’s usually a big red flag with a lot of funds there.

Cam: The second blunder that investors make is kind of related to what you’re talking about and that is that investors have a very naïve view of risk management when it comes to these factors. So we often see the factors kind of described in terms of so called normal, bell-shaped distributions when these factors are highly non-normal. So the tail behaviour in these factors is way more severe than what you would expect with a so called normal distribution. And in the paper, we go through factor by factor and we do this really interesting exercise that focuses on momentum.

So what we do is in kind of the early sample of the momentum factor, we look at the returns of the momentum. This is cross-sectional momentum, a factor. And then we match that with a normal distribution. So if the historical return is in the fifth percentile then we also draw the fifth percentile of the normal. So we’ve got this fake portfolio. As if momentum was normal. And it tracks fairly closely to momentum until momentum crashes. And then there’s just a dramatic difference between the two.

So the idea here is that many of these factors are subject to crash risk that’s not taken into account and people are surprised by this. And they should not be surprised by this. And so I think it’s a big mistake that investors make and kind of related to this mistake is this idea that they think that some of these problems can go away if you diversify across the factors. So it goes something like this. Well, we know some of these factors are probably fake and data mined, but we’ve got a factor portfolio and we’ve got enough good ones that we can ride out the bad ones. So that’s one logic which I don’t particularly like. Why would you invest in a fake factor? But the second one is, “Oh, yeah. Well, we know that these factors have some non-normalities that…we know that they’ve got tails and that’s exactly the reason that we’re not invested in one factor. We’re invested in a factor portfolio. So that turns out to be very naïve also because that kind of assumes that when you put many of these assets together you get back to the so called normal distribution and indeed there’s a theorem in mathematics called the central limit theorem that actually predicts this but it turns out that in the particular finance application it’s not a very good application of this theorem.

So what happens? You put this portfolio together, maybe 5 or 10 factors and it doesn’t look normal either. It’s got the same downside risk. So that I think is a naïve view of diversification that’s kinda related to the third blunder and the third blunder is that people are just not very good at taking the correlation of these factors into account. So there’s a reason that when you put a diversified portfolio of these non-normal factors together. You still have a crash risk and that is that when one factor crashes it’s bad news for the other factors, that the correlation actually increases. So in a drawdown period the correlation actually goes up and it works against you. So we’ve got this really powerful graph in the paper that looks at the beta of a factor portfolio against the market. And right before the global financial crisis, it’s got a bit of about minus one or even below minus one.

So if you think you’ve got a long-only chunk in your portfolio, most of your portfolio is long-only market. And then you layer on factor exposure. You think you’ve got a beta of minus one. The long-only part is got a beta of one. You go into the financial crisis like, “Well, it’s no big deal. Even if we have a correction I’m essentially neutral to the market.” So the crisis unfolds. That beta goes from minus one to zero. So it dramatically changed. So you thought you had this hedge, the hedge vanishes. So this idea that correlations are somehow fixed and constant through time is something I’ve been preaching about for years. And it just is amazing to me that people don’t take this into account.

Initially, the first time that I talked about this was in the context of diversifying your portfolio, in terms of international investment. So you look at the correlations and let’s say the correlation on average is 0.5 for international investment. You add it to your portfolio. You’ve reduced the risk of your portfolio. That’s great. But that correlation is an average correlation. So what happens when the U.S. market goes down dramatically? So what happens is that 0.5 correlation turns into a 0.9. And the international markets go down also so you don’t get the diversification. And then when the market is going up, the U.S. market’s going up you’d like to get that 0.5 also so you get some carry along but no, when the U.S. market’s going up, the correlation goes down. So on both sides, U.S. market going up or going down, you’re a loser in terms of what you expected in terms of the correlation.

So this is a story I’ve told for like 20 years and the same story applies to factor investing. So the correlation is not constant and when you most need that diversification from the factor portfolio, it often vanishes. So what we argue in the paper? The paper is about kinda three blunders that investors make. The next paper that we’re working on right now is how to mitigate some of these mistakes.

Meb: Not just the prescription but now we’re getting the solution.

Cam: The solution. And obviously, to preview some of it, one thing that’s very important is correlation management. So correlations vary through time. I’ve got an old paper with Claude Erb who you said was on recently and what we did was we looked at historical correlations between asset classes over the longest possible period and related them to things that we can observe in the economy. And we actually found that these correlations are predictable, highly predictable.

So you think about the stuff that is predictable in finance. Volatility is very persistent. Correlation is predictable. The returns are much more difficult. Very difficult to predict returns. But what we’re saying…even if you can’t predict the returns, to have some sense of the timing of the correlation is very valuable. And people do this for volatility also. So people do volatility timing or risk parity or stuff like that. It’s an easy way to get some extra return in certain asset classes like equity and credit, but this idea of actively managing correlations is very important. Indeed, the idea is a basic idea and that is that every person that is an asset owner should be able to answer the following question. What happens in 2020 if the S&P has a drawdown of 45%? What is going to happen to your portfolio? If you don’t have a pretty good answer to that then you’re not doing your job as an asset manager. Indeed, it’s really easy to be an asset manager over the last 10 years with the markets just going up.

Meb: I have a solution for you. It’s…you have a private equity fund that only discloses NAV once a year.

Cam: Yeah.

Meb: That’s how you afford the drawdown. Or you’re an endowment. You ought to disclose it once a year.

Cam: That’s on my list. People think, “Oh, well, I’ve diversified my portfolio with private equity because it has very low correlation, very low volatility. Well, that’s just false. The volatility’s no different than comparables in their portfolio and people just get misled by that spin. It’s awful mistake. But yeah. This is really important to do that correlation management. These correlations are predictable. The risk management of knowing what’s gonna happen in a drawdown are important because as the drawdown unfolds then you know how to readjust your portfolio, that you say, “Wow. We planned for this and this is what we’re going to do.” And you’ve got a strategy.

The worst thing that you can do as an asset manager is to improvise at the worst possible time where there’s chaos in the market. You need to have these different scenarios, understand your exposure, your factor exposures. So these three blunders, it doesn’t mean that you shouldn’t invest in factors. That’s totally not the message of our paper. You need to be wise. You need to be a skilled manager of the actual factors. You need to take basic things into account, that volatility is persistent and that is going to be an important driver of your weighing of factors. The correlation management needs to be very carefully taken into account and doing that you maximize the chance that your portfolio will outperform others in a time of stress. You can’t completely crisis-proof anything unless you do something at very high cost like by put options which is very, very expensive. But there are many things you can do with the factor portfolio that can manage a drawdown in a way that you end up as a winner rather than a loser.

Meb: Well, I think so many people in the industry…if you’re a student of history I think it helps a lot. One of our favourite books is “Triumph of The Optimist”. It looks back at a lot of the historical returns of markets throughout history and the problem I think with 2008 for so many people was they always…I don’t know. Dependent…hoped is probably the best word that these correlations would help them. And 2008, 2009 was one where you hadn’t seen a market like that in a while where I think the 2000 real-estate and commodities did fine and so everyone started piling into other things like the bricks and all sort of other ideas and then 2008, 2009 happened. Of course then what did everyone become interested in? Tactical or maybe CTAs after that but then you’ve had this romping, stomping bull. But it’s interesting that in some of the cases the asset classes…you know, the ones that people particularly expect to hedge some of the time…you mentioned gold, you know. That works decently but not always. I think one of the worst ones in the S&P gold was down like 16% or something. But puts you say is almost one near guaranteed but that’s the way markets work. You gotta pay for it. Right?

Cam: That’s right. If you want insurance on your house, fire insurance, you’ve gotta pay for it. It’s got a negative expect of return but if your house burns you get the payoff. So yeah. We just routinely pay for that and that’s…again, it depends upon your preference. Like how much you wanna avoid that downside and that will determine how much of the upside you give up. And that’s what this is all about, figuring out what the risk preference is for the investor and then designing a portfolio that optimally suits that risk preference. So in other words, gets the highest expect of return for that level of risk.

Meb: And the risk topic is so hard I think particularly for individuals and financial planners where a lot of people…they may not really know their risk until they’ve been through it. I mean, we’ll often hear people say, “Yeah, I can handle 50% drawdown. No sweat. Put all my money in stocks.” And then don’t realize that…going back to our earlier part of the conversation that often market drawdowns occur…there also happens to be a recession. They’ve lost their job. They got excited about three houses. It kinda all happens at once and then they say, “Wait, I can’t handle 50% drawdown. Like I sell it because I don’t have any money elsewhere.” And so…

Cam: And you sell exactly at the wrong time. So you sell low. And then you end up buying high. So it’s the opposite of what you should be doing but that happens where unfortunately people don’t do the planning that they do. They have to liquidate some of their portfolio at the worst possible time and they lose the rebound.

Meb: So it’s summertime. You’ve got a curious mind. What else is on your brain these days? We didn’t get the chance to talk about the great manure crisis of…what was it? 1894?

Cam: Yeah. There’s a lot of things on my mind and I’ve got a number of research papers that are ongoing and one of them is called “Faults and Missed Discoveries in Financial Economics” and it really has to do with not as much investing in a bad manager. When you think about the mistakes that investors make, so one mistake is you invest in a bad manager and the manager disappoints. Another mistake that’s not as talked about is passing on a manager that turns out to be good. And there’s kind of two related mistakes on top of that. You’ve got a manager in your portfolio. They’ve done poorly. And you decide not to redeem them because, “Oh, well, I think they’re smart. Maybe they’ve just had bad luck.” But it turns out that it’s not bad luck. It’s lack of skill.

Another mistake is you’ve got a manager in your portfolio. Manager’s gone through kind of a rough time and you decide that that manager is unskilled. You redeem but that was a mistake because they were skilled. So all of these…these are different types of mistakes that really aren’t talked about that much in academic finance. Usually, we focus on investing in the manager that turned out to be unskilled. So that’s the so called faults positive error.

My research is looking at this other error and developing a new framework which is I think intriguing. I know it’s different but it gives investors different type of decision criteria so that you could have a decision criteria that I’m willing to miss five skilled managers to avoid investing in one unskilled manager. So that sort of trade-off we haven’t really thought about before and that is the topic of some ongoing research. I’ve got research looking at investment horizons that’s brand new and that again is just very poorly researched in my field.

Meb: This is an area of particular interest to me and I don’t know where you’re going with this but we spend a lot of time talking about this.

Cam: Yeah. So just even think about how we think about the measurement of risk. So if you’re able to hold for a longer horizon the actual risk exposure that you face is different. So this whole idea…we think of beta as the sensitivity to whatever factor or market return. It’s usually measured over a monthly horizon. But that’s not necessarily relevant for the investor that’s holding over a longer horizon. And different investors have different horizons. Needs to be taken into account. So it’s very early but it’s very exciting. The theory is exciting and we have some very preliminary empirical results that are exciting also.

Meb: Well, you mentioned earlier and to tie this into a practical…I think for the younger people especially listening to this about…you said young people should actually be borrowing to invest and I was giving a talk in Dublin and talking to a younger audience and someone had asked something like…along the lines of, “Meb, what’s your best advice to some undergrads?” and I said, you know, “The decision for you guys is to focus on…I think it trumps what you actually invest in specifically or the fact that you choose to save and invest in the first place.” And I said, “But here is the challenge you guys will face. Let’s say you will be coming up on spring break. I assume you guys do that in Ireland and you have the choice. You either have the ability to go to Ibiza.” I said Cancun but that was more of a U.S.-focused example. I said, “And you’re gonna spend $2,000. However, the example I’m giving you is that you could put that $2,000 in an equity account and in 25 years you compound at 10%. That $2,000 is now worth 20. If you hold it for 50 years…so by the time you’re retiring age that $2,000 is worth $200,000 assuming no taxes.” And I said, “So can the young 20-year-old have some empathy with the you 70-year-old and say that $2,000 is better spent saving it or having $200,000 in retirement and thinking about how to invest in compound.”

And I said, “Actually my advice is you should probably go to Ibiza because you have a lifetime of memories and maybe that’ll be better spent.” But I said, “It’s a way to think about your life being young is the biggest thing you have is the ability to compound a long time horizon. And the example you give of where the volatility of stocks on a one year time rising versus the volatility of say stocks on 20 or 50-year-time horizon and the spread of percent how performance versus other assets. It’s things that I feel like people talk about theoretically but I don’t know how it makes them the actual implementation or policy for most investors.

Cam: People are not very good at compounding in general. So people are not very good at exponents. And even people that you might think are skilled that should be able to do it, people that might have master’s or PhDs in quantitative fields, they’re just not good at that. And it leads investors to make a number of mistakes in terms of their saving where they are basically undersaving. These young people should have some leverage at this point. Most of the students in the business school at Duke are highly levered and it’s no problem getting that student loan because they make it back in a few years given the bump in income that they get from their master’s but it’s really hard to basically say to these investors, especially when you’re young that you need to give something up to invest for a time when you’re 70 years old. That just doesn’t go over well. The trade-off in terms of the time horizon, they just don’t get.

So the alternative is using leverage or tilting the portfolio…you think about it. They’ve got a small amount of savings. Often, it’s in a certificate, a deposit or a bank account that makes nothing. So if there was a way to invest in a different way and that’s what we’re talking about earlier, that investing in a broad range of different assets where there’s no round lots…you buy fraction of a share of IBM at very low fee and have a portfolio like that where you’re diversified across different types of assets. Some of these assets are just not available today but will be available…like you own a piece of a self-driving Uber, 10% of a self-driving Uber, or you own 10% of a cell phone tower and you’re paid in real-time dividends. So any time somebody uses it you’re paid in real-time dividends.

So there are all these assets that will be available. It’s very exciting in terms of looking into the future and how asset management will change. And I’ve given this speech many times that this is actually a great period for financial advisers because on one hand, you see the sort of robo investment tools maybe eating into their business a little bit but really not. It’s more…they’re using the robo investment tools and fine-tuning it for a client. But there’ll be so many of these new assets that come on to the market that people will need some help and that’s where they come in and for the medium term at least there’s just so many opportunities.

But again this is an issue. If you think it’s a problem with these young people in Dublin and the lack of saving you need not look very far towards, let’s say the unfunded liabilities of the U.S. government in terms of social security and other obligations. It’s hundreds of trillions of dollars and that doesn’t appear on the national debt. So there’s no saving going on there either. It’s a problem that has been a problem for generations. I don’t see it going away but I do see a possibility of these new assets coming online that will have high expect of returns. Some of them will be risky in terms of…they will have explosive upsides where you have many people that might be able to invest in a piece of the new Google or the new Facebook. Not just the select few.

So right now our system essentially reinforces income inequality. The best investments go to the people that have very, very large portfolios. So to get in line for our private equity, for example, you look at the performance of private equity. It’s nothing special but it’s diverse. So the people that get the best deals are the people with the biggest portfolios. So I see sort of it’s gonna become much more democratic in terms of the playing field for investment. It’s also the case as I mentioned earlier that we’ll see a much bigger peer to peer sector. Indeed, I remember when my grandfather died going through his portfolio I noticed that he held a mortgage which…that…I didn’t know about that. So that meant that he had given somebody a loan to buy a house and he held the actual mortgage. So he was playing the role of the bank. So he was undiversified and doing that though maybe he knew the person and that helped mitigate the risk. But think of buying a fraction of a percent of thousands of mortgages diversified across the U.S. or even across the world. So that way you’re gonna get a much higher rate of return. There’s no middle person. You’re investing directly. The small business can go to the crowd for funding. So instead of that bank loan, the bank is asking like 12% for they could be funded at 6% or 7%. And that’s a good deal for them. It’s a good deal for their investors because they’re getting a much better rate of return than they could get buying a commercial bond or something equivalent to that.

It also allows for small ideas to actually come to fruition. So there’s so much stifled innovation in our economy and around the world because of so called financial frictions. So it might be that the bank is charging a rate that’s too high. It might be that the bank isn’t even interested in you because the loan amount is too small. So if it’s too small it doesn’t cover the fixed cost of the bank. So now all these microloans can be efficiently provided and surfaced in terms of the credit function of the bank, in terms of scoring. Well, that’s a joke right now because just off the shelf credit, scoring algorithms that are free, available on the internet are very competitive with what the banks are doing. So all this stuff is available. I do see the upside here and the upside is the following, that I believe that we sometimes learn too much from history and we think about economic growth and the world, the U.S. growth declining. We’re lucky to get that 3% and who knows when we’ll ever get the 4%. Europe declined even further. They’re lucky to get two and more likely to get one and kind of the narrative is that as you become more developed you expect the real GDP growth rate to go down and indeed just look to China. So it used to be they’ve got 12% growth, then a 10% and then 9% and then now 7% but still very good but you kinda see well, they’re developing. Therefore the rate of growth goes down.

So I think that just extrapolating from the past on this is a mistake and we know that GDP growth is very important for equity valuation. That’s the numerator. So that’s the cash flows of the firm highly correlated with economic growth. But my narrative is different and that is with this technological change that’s going on right now we will dramatically reduce the frictions. So these small business that weren’t small businesses because they couldn’t get the funding, they get the funding. And they invest in their idea and some of those ideas turn out to be really big ideas.

We also allow for financial inclusion. So even in the U.S., an extraordinary number of people are unbanked. And around the world, you’ve got 2.1 billion people that are unbanked. So obviously they can’t get a credit card and they can’t do business because they’re unbanked unless it’s a cash type of business. But with this new technology, everybody will be banked. You need a smartphone and the smartphone is your bank. You carry your vault on you in the smartphone. So this allows a large group of people to join the modern economy to transact on the internet both buying things and selling things. And that’s a big idea. So if you think of two billion people that human capital coming onto the world market, that basically takes that declining growth rate in emerging markets and jacks it up with the discontinuity, the size of the industrial revolution and actually bigger if you look at what happened there. And what happens in emerging markets? Anything that’s positive is good for developed markets also.

So I actually am very especially bullish on emerging markets and I believe that people have too quickly written off this very slow growth that we’re gonna be stuck at 2% or 1%. No, no, no, they’re not figuring in the positive benefit of the technology that we’re dealing with today and when that happens we’re operating in a world…just you think of financial institutions. They’re the same today than they were a 100 years ago. Nothing’s really changed. The stock market is essentially the same. Yeah, we’ve got some electronic stuff but it’s still very, very clunky. Payments, clunky. Think of this. You swipe with your Visa or MasterCard. That’s 3%, 300 basis points. Right? That’s a massive fee. And the gas station even more. That is super inefficient. And that sort of thing will go away. Those transaction costs will become very, very small. That’s all good. That’s a shock to the economy. That 3% fee has been the same for 40 years. It doesn’t make any sense. And we’re seeing some credit cards get nibbled away where they give some points away, stuff like that. That does nothing compared to what I’m talking about. So when we can transact more efficiently, when all of these people join the modern world, the internet we will see a surge in growth which will be good for everybody on average at least and the stock markets.

Meb: So you’re saying this is like a 10, 20-year sort of transition? Do you think there’s any scenario where a lot of this…there’s like a dam breaking catalyst? Because we talk about this in the investment management industry and I say, “Look. You can basically buy the global market portfolio with ETFs for essentially free.” I mean, it’s like five basis points. You include short lending, it essentially has a negative expense ratio. You don’t have to pay commissions if you don’t want to. Vanguard, Robin Hood, all these other brokers. So I’m like, “You can invest for free.” But we live in a world where there still exists a 2% S&P 500 mutual fund. God forbid you go to Canada or Japan. Someone was talking about the average fees that someone pays once you include all the intermediaries. It’s like 4%. And then people always ask me. They say, “Meb, you know, when’s this disruption happening?” Because every day in the media you see well, ETFs are low-cost funds, excuse me, are eating high-cost fees’ lunch. I’m like, “Yeah, but it’s like…the high fee stuff is still like multiples of size bigger despite the fact that most of the evidence shows that you don’t need to be paying these things.”

One more example and then I’ll let you answer. Did I see is one of the major disruptions of the automated space is the fact that we did a poll the other day on Twitter. I said, “What do you earn on your savings and checking account?” It was, “I don’t know, zero to 50 bips, 50 to a 100 or above a 100.” And it was like 60% or 70% were below 1 and the reason I mention this because Betterment just recently came out and Wealthfront all these others are doing it where they offer like 2.5% on FTE, ICE-ensured checking accounts and savings accounts so I said…I mean, I looked at my Bank of America. I’m a preferred client. Mine is five basis points. And to me, like I think like the automated service is really cool but to me, you’re starting to see these massive disruptions where…and going back to the fees it’s like all these things happening. Is there something that causes it to actually like flush, dam break or is it the way that I tend to lean which is as people die, as people get divorced, as assets pass along it never goes back. Do you have any thoughts on predicting the future?

Cam: It’s kinda my job. So my job is to prepare my students to be disruptors in the future, not disruptees. So part of my job is to think about that future. And when I look at kind of traditional financial institutions, there is like a reason that you’re only getting five basis points on your savings account or your checking account. It’s basically all of the fixed costs of the infrastructure, all of the bricks and mortar, all of the money that these banks spend on security. Like cybersecurity is a major, major expense. So you’re paying for all of this and what I’m saying is the technology will change on so many different dimensions. So the cybersecurity will largely become irrelevant with blockchain technology. So that is a massive saving right there. You don’t need the bricks and mortar. They’ll be much more peer to peer so that’ll be good also. And the traditional banks, you talk about five basis points. Try doing like an international wire transfer. And I did one recently to Canada and I said, “Well, gee. What rate are you giving me?” And they quoted a rate that was far away from the market rate. And I said, “Well, I’m a preferred customer also so [inaudible 01:36:39] do a little better.” So you need to ask for that first which is really not fair. And they gave me a marginally better rate. So they’re making just a huge amount of money on a very simple transfer.

So all of that is going to go away. It’s gonna go away sooner than people actually expect. This is not a 20-year sort of expedition. This is more in the 10-year time frame. Different countries will have different horizons. For example, you mentioned Canada. It’s got essentially an oligopolistic, highly concentrated banking sector that would be very difficult to disrupt but in the U.S., it’s wide open. Yes, we do have some really big banks but they are the ones that are most at risk. So these fees basically just take money out of the economy. So just think of that. If 3% was put into everybody’s like wallet today that would lead to more spending. So it basically…this will happen over I think a time frame that’s a lot shorter than people actually expect. Even though you don’t see a lot right now. you will see it. And for the financial institutions, it will be a catastrophe.

Meb: Good. I’m looking forward to it. I’ve got the smores. You bring the matches.

Cam: But fees, in general, are kind of interesting. We talked earlier about…it doesn’t make any sense if somebody is underperforming and they’re charging this really high fee and then trying to spoof potential investors saying that they’ve got this great product. It is also the case that if markets were reasonably efficient, that if you got something good you should be able to charge for it. So you’ve got a good model. It’s got limited capacity and you should be able to charge a higher fee for it. And people doing the due diligence should be able to determine, “Okay, this is good and I’m willing to pay the fee.” Because really all that counts is the after fee return.

So it is foolish to look at the return of any fund before all costs are factored in. So I don’t care if it’s high fee. If they can deliver on a high fee, then that’s great. But we also need to be careful about what deliver actually means. So you’ve got a huge number of asset managers out there that I classify as unskilled. And what they’re doing is they are exceeding a benchmark. Maybe the S&P 500. And all they’re doing is taking on some known factor risk and they get paid for it, which doesn’t make any sense because again on the fees you can go to an ETF to get that exposure in a fairly passive way at very low cost like the stuff that Research Affiliates does. That sort of product is ideal. It is very disruptive.

So these products are designed to disrupt the unskilled managers that are just taking risk and then investors are paying them for that. So instead of doing that you get this very low-cost ETF that delivers performance that’s appropriate for whatever factor at a very low fee. So if you think of the asset management industry, I see two kind of forces in terms of kind of shaking out that industry. So one force is this low fee product that’s designed to outperform a standard benchmark like the S&P 500. And investors can widely get that and that really squeezes these managers that don’t really have any skill.

And then there’s another force having to do with the technology where given there’s just so much information today and it’s almost impossible for the manager to digest all that information, we need technological tools that helps us. And I’m talking about discretionary investors, I’m talking about systematic investors that the very largest firms are heavily investing in machine learning tools to aid their investment. And these are fabulous tools that can be misleading also if you don’t know what you’re doing. They can be misapplied, but I’m talking about firms that are large enough to have skilled people operating these tools to provide timely information for their discretionary managers and what that does is it causes a shakeout. So these smaller asset managers that can’t afford to invest in the human capital that’s related to machine learning, the computing resources that are substantial and the data and data is not free. It’s expensive. Anybody who says is free, they don’t know data science. Just the cleaning of the data takes a lot of time. So these larger firms have the economies of scale to actually do this and what we’ll see is kind of a concentration, which I think is good for investors because when you put these tools to the data, it can really help the manager in terms of what they actually select.

So those two forces are really attacking small and midsized asset managers. So the low fee stuff doing basic ETFs on the factors and the firms that might be able to charge a higher fee that are really squeezing something out that is not obvious just eyeballing the data or using whatever spreadsheet to look at some company’s valuation. So you see that bifurcation.

Meb: A couple of comments. One is that when we think about fees we always think about them as just like a high jump bar. The higher fees don’t mean you can outperform. I mean, look at the famous example of Rentech or some of the most famous hedge funds. It just means it’s a higher bar. And as a default when we…we joke with listeners on this program. I said, “If we ever get to 5, 10 billion…” I used to say 10, now I say 5. I said, “I’m gonna launch a fund and we’re gonna do the old…” Was it Wall Street Journal? The Dart Portfolio? And I said, “We’re gonna have a party every year. The 100 people can come that are shareholders. Throw darts against the wall. We’ll pick our portfolio for the year. We’ll launch it, low fee portfolio just to kind of troll the whole industry.” Because we’ll equate it. We’re not gonna market cap [inaudible 01:43:14] it. So that’ll give us a little bit of a tilt towards a couple of things that should help against market weighed portfolios. But I, for one, am looking forward to all the developments. It’s hard. I mean, Charlie Monger often says [inaudible 01:43:26] he’s in his 90s now. Asked him for an advice and he’s like, “I’d be a fisherman. You go where the fish are.”

When you think about a lot of these funds and data sets it’s probably not gonna be doing research in the U.S. on some large cap stocks and maybe you’re gonna become a frontier market analyst in Nigeria or somewhere else where…or new asset classes in different ones. Cam, we gotta start winding down. We’ve kept you for a long time. We haven’t even talked about the 100 of your papers yet so we’ll have to have you back.

We wanted to add a couple of other quick questions that people love to answer. One is looking back over your career and this can be personal, it can be involved in really anything. We like to ask people what’s been their most memorable investment. It could be good, it could be bad, it could be both, it could be just the one that’s just seared in your brain. Is there anything that comes to mind?

Cam: My best investment was to kinda go against my family’s wishes and go to University of Chicago for a PhD. They totally didn’t understand that. I was the first one with a bachelor’s degree and then a master’s. Why do you need that? So that was probably for me to go to University of Chicago at that time the best investment. And this is consistent with our conversation that we need to include human capital in the investment. So that was an investment of my human capital that turned out to be a pretty good investment. The best trade that I ever did was when the yield curve inverted in mid-2006, I went 100% in cash.

Meb: Uh-oh. Ominous.

Cam: Yeah. And that turned out to be a great one. And I guess recently when I started teaching my blockchain course I bought some Bitcoin very cheaply. I wish I bought more.

Meb: Yeah.

Cam: But that had a spectacular upside.

Meb: That’s funny, yeah. You’ve gotta have some skin in the game when you’re teaching it. I totally whiffed the whole crypto. We actually had it on one of our research portals for a payment source that people could pay, but this was like 2014 or something and no one paid. I’m like, “This isn’t interesting. No one’s using it.” Obviously not the right use case of people paying for things where you could pay with a credit card. Both those are great. You know, it’s funny. I have the exact opposite but probably worked out. When my brother had gone to grad school at University of Chicago doing a PhD and it took him like a decade and when I graduated college, I was originally gonna go directly…I was a biotech guy, engineer. And I was originally gonna go directly to PhD and he’s like, “Meb, you’re young. Take a couple of years. Make some money because it’s gonna be a long slog and then you can always go back.” And then kinda went to the finance route and starting in the investing. This is the year 2000 so a lot of bubbles.

Cam: But it’s worked out perfectly.

Meb: It’s worked out so far. It did in a different way. You know, my hobbies and career and vice versa. You know, where do people follow you if they wanna see everything you’re up to, your writings, your goings ons? What’s the best places?

Cam: So there’s many different things you can do. @camharvey is my Twitter handle and then you can find me on LinkedIn. I’m using that more than I used to because it’s convenient place to put articles. And so I’m using that. Also just Google me. My web page has got all my publications and my teaching material. Everything is online.

Meb: We’ll add all the links to the show notes at mebfaber.com/podcast. Cam, thanks so much for joining us today. It’s been a blast.

Cam: Yeah. Great to be on the show.

Meb: Listeners, again, we’ll add those to show notes. If you like the show, subscribe. We love to hear some feedback@themebfabershow.com. Leave us a review. We love to hear the good ones, the bad ones, everything in between. Subscribe to the show on iTunes, Radio Public, Breaker, anywhere else. Thanks for listening, friends and good investing.