Risk of Ruin
Audio documentaries about risk takers.
Risk of Ruin
Zero, by Way of a Hundred
John Hempton runs Bronte Capital, which generates its excess returns primarily through shorting fraudulent companies. John walks through a hypothetical short, explaining how they find it, how they size positions, and the kinds of returns that can be expected. He also discusses his shorts of Valeant Pharmaceutical and Wirecard.
John Hempton on Twitter: https://twitter.com/John_Hempton
John's Blog: http://brontecapital.blogspot.com/
Netflix documentary on Valeant: https://www.imdb.com/title/tt7909184/
Email the show: riskofruinpod@gmail.com
Follow the show on Twitter: https://twitter.com/halfkelly
You have an edge. There's a stock out there that might be trading for$300 million, and it's worth nothing. It's an edge. The question is, is it an exploitable edge, and how do you exploit it without risk of ruin?
SPEAKER_01:Risk of Ruin is a podcast about gambling and life and their intersection, and I'm I'm John Reeder. This is episode 23, Zero by Way of 100. If you follow the financial markets or if you've ever seen the Netflix documentary titled Drug Short, then you know the name Valiant Pharmaceutical. In the summer of 2015, Valiant stock traded for about$250 a share, and then a year later, it was like$25. That's quite a reversal, so let me detail some of the highlights from the ride down. First, Valiant's entire business model, which was like Martin Shkreli on steroids, came under intense public scrutiny. Then, short sellers accused the company of using a shady third party to juice sales numbers. There was an emergency conference call to try to keep the share price up. The CEO was eventually fired. Some prior year earnings were restated. A handful of people either went to jail or paid fines for misconduct. Also, the name Valiant became toxic for investors, Congress, the public, and even other healthcare companies like CVS. So eventually, they changed the name of the company to Bausch Health. Valiant is probably the biggest and best win for short sellers over the course of the bull market. Critics of the company alerted the market to wrongdoing and they made money on their bets. One of the short sellers was John Hempton. He runs Bronte Capital in Australia.
SPEAKER_02:Valiant was one of the seminal events in my life. For those that don't know, Valiant was a very big American pharmaceutical company, peaked out at well over$100 billion market cap, almost$200 billion. And it was a company that bought other companies for their drugs and raised their prices. And in fact, it had no real other business model. The first thing it did when it bought a pharmaceutical company was sack all scientists, get rid of all the cost structure. All they wanted to do was raise prices. And the problem with that is it's a highly decaying model. If you don't keep churning out research papers, you can't expand research. You can't expand the label on your drugs. If you don't keep developing stuff, you won't wind up with the best drugs. And at some point, Valiant started faking its accounts. We're very, very good at determining fake accounts. It just struck us that Valiant was obviously bullshitting. And yet, there were hundreds of well, not hundreds, but dozens of big name US investors that thought that Valiant was the best stock in the world. And the market had proved them right. You know, Valiant went up and then it went up some more and then it went up some more. And as they started faking their accounts or faking the story, it went up further, right? So this was a classic situation where they were faking their accounts. They were faking everything. The stock was going up and everybody on Wall Street was that had picked the stock early thought that they were a genius, and they were marketing themselves as geniuses. And along comes this little fund from Australia that, to be honest, only a few cognoscenti had heard of, and said, the emperor's wearing no clothes. And it turned out, yeah, the emperor's wearing no clothes, and a couple of the people associated with Valiant have now gone to prison. Now, before that, we were a fund in Australia that had interesting results. But we were operating from a beach suburb in Australia, Bondi. We had no Wall Street pedigree. We didn't look like anybody special. And there was a question about the sustainability of it. And after that, it was like, hey, these are the giant killers. So we had a combination of fairly good numbers and we had been very publicly right for the right reasons. And being very publicly right for the right reasons, about the same time I hired a sales guy who was highly competent. And the combination worked exceptionally well. So yes, Valiant was a central event in my life because it helped us raise money for the right reasons. And if you're going to manage money, it does helped to manage a lot more than a little bit, right? So before Valiant, what we had was a nice business, but a small business. And after Valiant, what we had was a very nice business.
SPEAKER_01:The smoking gun in the Valiant case, the revelation that sent the stock price tumbling was the company's relationship with a mail order pharmacy called Philidor. You can think of Philidor as a dedicated sales channel, which allowed Valiant to offload scummy tactics to a third party. For instance, Maybe Valiant had a branded skin cream that sold for hundreds of dollars per tube, but there was also a generic that sold for like 10 bucks? Well, Philidor's role was to push the expensive cream and stick insurers with the tab. So in simple terms, Philidor and Valiant conspired to fleece insurance companies and by extension, fleece people that pay for insurance. But the kicker, and my favorite part of the story, is while Philidor was supposed to be helping Valiant stick it to insurers, The CEO of Philidor couldn't help himself, so he decided he should probably steal at least a little from Valiant. This is Preet Bharara at a press conference when he was U.S. Attorney for the Southern District of New York.
SPEAKER_00:Specifically today, we have arrested and charged two men, Gary Tanner, a former senior executive at Valiant, and Andrew Davenport, the founder and CEO of Philidor. We allege that these two men participated in a fraudulent scheme to illegally use Philidor as essentially a vehicle for personal profit and self-dealing. As alleged, these two men purported to be arm's-length business counterparts. Tanner with Valiant and Davenport with Philidor. But they were, in fact, we allege, partners in crime.
SPEAKER_01:Philidor is only important for our purposes because it shows the fundamental reliability of people. And that reliability is core to John's strategy. If you find some folks running one scheme, the odds of finding them in another scheme increase by a lot. People are amazingly dependable. Anyway, before October 15, 2015, almost no one had heard of Philidor. Even though more and more Valiant sales were flowing through Philidor, they'd never been mentioned in a Valiant filing. But John found them.
SPEAKER_02:People don't have 600-person staffed entities that they don't disclose at all. You couldn't find anything out about it online. And when we started digging, it was pretty clear that they didn't want us digging.
SPEAKER_01:John mentioned Philidor to the journalist Roddy Boyd and the short seller Andrew Left, which set in motion a chain of events that would eventually tank Valiant's share price. On October 15th, John wrote a cryptic blog post that didn't say much more than just the word Philidor. Then, on October 19th, Roddy Boyd published an article linking Valiant and Philidor. Finally, On October 21st, Andrew Left issued a report calling Valiant the next Enron. The rest is history.
SPEAKER_02:The usual problem we have with shorts is you've found them lying three or four times and now you don't know when they're telling the truth. It's a general rule actually in the stock market that if you talk to the management of a company and you believe what they say, almost every company looks like a buy. Now, Every now and again, you catch the management team lying. And once you've caught them lying, if they tell you X, Y, Z, you don't know whether to believe them anyway. So we've caught Valiant on three or four small lies. And then we found Philidor. And Philidor was obviously a big lie, but we didn't have any particular understanding of it. We knew it was key in some respect because it was so large and so well hidden. Now, I actually did pass the information on to Roddy Boyd and Andrew Left. I regret passing it on to Andrew Left. I got nothing back from him and I was hoping to because he was, you know, a fellow short seller. But he also gazumped me a little bit on the short. And I was actually quite upset with it in respect. Roddy Boyd, I passed it on to as a journalist. And journalists, you do not expect them to give you anything back. I mean, the whole thing about journalists is that they're meant to do their own research and publish it. and there have been several insider trading cases where journalists have told investors what they're going to write. When I passed it on to Roddy Boyd, because he was a journalist, I didn't expect anything back and I didn't get it. Roddy acted with utter integrity. Now, when I passed it on, I knew it was big, I knew it was important, and I didn't understand it at all. It's like, imagine you had stumbled on this massive industrial plant in Norway where the Nazis had huge amounts of security around it, where vast amounts of money are being spent, and you can't work out what it's doing. In that case, which is a very famous case, it turned out to be a heavy water plant and was the beginning of maybe a Nazi atomic bomb program. If you had stumbled on it, you would know it was big and important, but realistically, there were probably only 200 people in the world that could work out what it was. I wasn't one of those 200 people. But to give Roddy credit, Roddy worked out about 85% of it and he worked it out himself.
SPEAKER_01:I framed the Valiant episode to make short selling sound easy. Just find a shady company, dig up some dirt, watch the price collapse, profit. But reality is always more complicated. Let's look at this another way. On May 21st of 2015, Valiant was still marching towards its all-time high and it would be another six months before Philidor became public. On that day, John tweeted, quote, Okay, I confess. Valiant is one of many shorts, mostly pharma, that I wish I'd never heard of. Still short, but hard to profit from here. Unquote.
SPEAKER_02:We entered it at about 130. There was a little bit more added at 160 or 70, but never enough that we were squeezed out. The stock went to 260. So at one stage... Valiant was one of the biggest losers in my portfolio, but we had it absolutely nailed and we were very sure we were right. And I eventually exited the shorts at prices between about 15 and 30. It worked out okay, but it still had that characteristic of fraudulent companies, which is, you know, fraudulent companies go from 10 to zero, or in this case, 130 to 30 via a number that you don't understand.
SPEAKER_01:We started with the story of Valiant because it illustrates the promise and the pitfalls involved in shorting stocks. It was life-changing for John Hempton, and it was also a roller coaster ride. He's been at this a long time, and he's seen a lot of these things, so he's going to walk us through what these short ideas look like, and we'll also see how Valiant fits the broader type.
SPEAKER_02:In Perth in West Australia, and also in Vancouver, there is a large industry doing fake gold stocks. The idea is that you peg a whole lot of land, you tell people that it's got gold, and you're not really mining for gold, you're really mining for money on the floor of the stock exchange. These people bring to mind Mark Twain's famous definition of a gold mine, which is a hole in the ground with a liar on top. Now, imagine you know for sure that one of these gold companies that claims to have a million ounces in the ground has no gold at all. You can actually know that for sure in certain ways. If there's gold in them, their hills, there's gold in the creek below the hill. If you hire a professional panner and you go and pan the creek below the hill and you don't find any gold, then you know for sure that there's no gold in the hill. And in fact, you don't even need to hire a professional panner. All you need to do is contact the local regional, you know, country town librarian. And there'll always be history of the area, you know, that somebody's written. And the librarian would love to talk to you. And they'll give you a history of the gold panning in the area. There's always one. And if there's no history of successful gold panning in the area, then you know there's no gold in the hill. Okay, so now you know there's no gold in the hill. You have an edge. There's a stock out there that might be trading for$300 million and it's worth nothing. It's an edge. The question is, is it an exploitable edge and how do you exploit it without risk of ruin? Suppose you have done this and you decide, okay, now that I've got this edge, I'm going to short 5% of my, well, use the gambling term, my role, you know, my gambling role. or 5% of my portfolio in this stock. Most of the time, the stock's going to go to zero. And over two to three years, you'll add 5% to your portfolio. And if you can do that four or five times a year, you're Warren Buffett, right? Because index plus 15 is better than Warren Buffett. Sounds pretty good, except it doesn't work that way. And the reason it doesn't work that way is that the one thing you know about this guy is he's a liar. he's already lied and told you that he has a million ounces. And there's a fairly good chance that he's going to lie and tell you he has 10 million ounces. You know, finding a million ounces of gold is hard. Finding 10 million ounces of gold is even harder. Lying about a million ounces isn't that hard, and lying about 10 million ounces is no harder than lying about one. Now, if the market believes him, this stock that you've decided to short 5% of your wealth in or 5% of your roll in if you wish to be a gambler, is going to go up tenfold. And so instead of being 5% short, you're now 50% short. Except you've got a problem, which is your capital's gone from 100 to 55, because you've just lost 45% of your bankroll. And so you're now over 90% short. And if you're 90% short a single name, your good friends at Robinhood or interactive brokers, or in our case, Morgan Stanley and JP Morgan are going to put you out of business. And they're going to put you out of business on a stock on which you're right. And so what you have is an edge, but it's an edge that's exceptionally dangerous to exploit.
SPEAKER_01:So that's the problem. And it's not much different than the gambling problems we cover. The issues are find an edge, figure out how much to bet and stay in the game. Addressing those issues also means solving a bunch of related puzzles that we'll get into. But the first question for any gambler or investor is always, should I play at all? Because you don't have to play this game. You can always go find something else. Maybe there's something with a better edge or less volatility. And to be fair, lots of managers actually have thrown in the towel on shorting stocks. They've decided the pleasure just isn't worth the pain. But I think the simplest way to illustrate the potential of shorting is that if you want returns that beat the market and you want returns that are uncorrelated to the market, one way to get there is to make some bets which are actually negatively correlated to the market. So if the whole market dumps, your shorts go up and you can buy discounted quality companies. There are entire books written about this kind of thing, but John recommends the book Expected Returns by Antti Elmanen of AQR.
SPEAKER_02:One of the attractions of the stock market is that you can get an enormous number of bits per week. Although, you know, take my gold mining example. The typical path to zero for that gold mine is five to 10 years. That's actually not a very good return, right? The only functional part of that return is that it's a negatively correlated return. You know, if you have an asset that actually, you know, breaks even over a cycle, but is negatively correlated and you pair it with an index, you'll do just fine. Thank you. There's a lovely book on that, incidentally, one of the best books I've ever read, which is called Expected Returns. Now, the Expected Returns book makes it absolutely clear what the value of a even slightly negative return, but very strongly negatively correlated asset is. And the answer is a lot, right? And it's quite a good book on how you might think about the portfolio construction.
SPEAKER_01:You've probably gotten an email at some point hyping, I can't miss stock that you've never heard of. You look at the email for less than a second and then hit delete. Even if the email doesn't look overly scammy, which usually they do, the very idea of marketing stocks randomly via email is a red flag. Well, John gets those same emails, but he doesn't delete them.
SPEAKER_02:But very early on in the piece, I discovered, I guess what you'd call early 2000s penny stock frauds. And the early 2000 penny stock frauds were, well, you know, it was really ridiculous. It was stuff that was sent to you in spam emails, right? Really crappy stuff. And it turned out that you could short$5,000 worth of these things, but you couldn't short 50,000. And I started a little PA account, personal account, where I systematically started shorting these things. And I realized that you could turn$30,000 into$500,000 pretty quickly doing this. But you couldn't turn$500,000 into$5 million because you became scaleless. Now, but what this taught me was that there was a massive sort of underlying scummy part of the stock market. When we started looking for the bigger caps, it was interesting, but originally it was hard to do. Again, I thought that we could probably work out how to scale it to$30 or$50 or$100 million, which was an interesting number. But it wasn't going to make a very big fund. And then two things happened. One thing that happened was we just got much, much better at finding these things with computers. And that turned our scale limit from, say,$30 million to$300 million. But that required an enormous effort. It required that we go completely global. the big, hairy, audacious goal of trying to find every fraudster in the world in stock markets. The second thing that happened was the market went up, and then it went up some more, and then it went up some more, and then it went up some more. And eventually, vast numbers of retail investors with absolutely no edge joined the market. So at the moment, well, during the crisis, something like 40 million people in the United States opened retail banking broking accounts. Now, I remember the last time that happened, it was sort of dot com. And in that time, it was sort of two million people opened retail banking accounts, of which I was one. And when I look back at what I did when I first opened my retail broking account, I look you in the eye and say I was a goddamn idiot. And the only reason that I didn't blow up was that I did it in 2004, not in 1994, rather than in, say, 1999. And I had read enough books by 2000 to realize that it was all going to end in tears. And I had reject appropriately. But when I look at the people that opened in 1999, they all looked like they were geniuses for a couple of years. And then they all looked like they were zeros. Now, that's what happens when 2 million people open retail broking accounts. This time it's 40 million people. The amount of people that are gambling on crappy stocks where they have no edge is is extraordinarily large, larger than any time in human history. The way I describe the stock market at the moment is it's a market in which people invest in companies in which they've never heard of, recommended on the internet by people that they've never met, sometimes with money that they don't have. So yeah, what we thought was going to be a small opportunity because of history and because of computers became a big opportunity.
SPEAKER_01:Remember from John's gold mining example that one of his goals is to make sure that no single name can hurt too much. So his position sizes are very small, less than 1% per company. But that also means he needs a lot of companies to short.
SPEAKER_02:I want to find every fraudulent gold mining company in the world. And I actually don't care if I'm wrong a little bit, right? And the sort of example is if I can short, say, one fifth of 1% of that, hypothetical gold mining company I talked about. But I go and find all his friends, all the people he hangs out with, all of the people he does, you know, all the people he's involved with. You know, Joe is a fraudster. It's likely the five best friends that Joe has are also fraudsters. Rather than short, you know, 5% in Joe's fraudulent gold mining company, I'd probably prefer short 20 bps. in Joe's fraudulent gold mining company and 25 of his best mates.
SPEAKER_01:The process of generating short ideas is part pattern matching from years of experience and part technology.
SPEAKER_02:Solution one for us is just find all of these people. And the way that we find them is with computers, right? The job is there are three underlying assumptions. The first assumption is once a scammer, always a scammer. And that's mostly true, but I can assure you that I can show you successful businessmen that are completely honest that started life as scammers. But again, it's true on a probability basis. The second assumption is if you hang out with scammers, you're probably a scammer. And again, that's statistically mostly true, but it's not admissible in a court of law, right? And it's not admissible in a court of law for a very good reason, which is that it doesn't constitute proof in any particular way. But as I said, it's pretty identifiably the case. In some sense, my standard for following a person is the Terry Thompson, US Supreme Court standard for when it was acceptable for a policeman to pull over a car, which is, do you have a reasonably articulable suspicion? And the third assumption is, if you invest in scammers, you'll mostly lose money. And I invert that, which is if you short scammers, you'll mostly make money.
SPEAKER_01:That makes sense in theory. But how does Bronte actually do this in practice?
SPEAKER_02:We have some software vendors who come from inside the Beltway and sell to government agencies in the US with three letter acronyms. And their problem is roughly our problem, which is that the way that they might track terrorists or the way that they might track criminals is by relationships between people. And so we're actually the first financial institution to be clients of one of these companies. Now, I'm going to tell you that at least part of what we have is vaporware, meaning we can do maybe 25%, 30% of the job that way. But if you come back in two years, it might be 65% of the job that way. And in three years' time, it might be 85%. And when it gets there, we will get more computerized. But there's... Multiple tools. A very simple tool is that I, for instance, signed up to every penny stock newsletter in the world in about 20 years ago. And I have a database of about one and a half million now penny stock newsletters. It's certainly over a million. I haven't looked at the number. And these are letters sent by people we knew to be scumbags in, say, 2005. That database is getting more and more organized and it's getting more and more organized with computers. But I'm going to confess that if the penny stock newsletter came to us in the year 2002, it's in a disorganized structure. And if it came to us last year, it's in a very organized structure. The computer programs just get better over time. And I wish I had been more organized in 2002 about how to do this, but I wasn't.
SPEAKER_01:The other day, I was reading an article about a crypto scam where the perpetrators ran off with a million dollars. And by the time they were charged with fraud for that incident, you know what they were doing? They were running another crypto scam. Again, people are amazingly reliable.
SPEAKER_02:If you know enough geology to work out X is a fraud, our assumption is that if the person's done a fraud, the next stock is also a fraud. And we can be mistaken on that. We will put a short on based on a person alone. But the person alone we have fairly heftily determined was a fraud the first time he was around, which might have been 15 years ago.
SPEAKER_01:No one is going to win every bet. So when you lose, it's important to be able to figure out if you were wrong or just unlucky. Basically, should you make the same bet the next time around?
SPEAKER_02:The first way we're wrong is that we're not wrong. It just goes from 10 to 0 via 100. And the second way we're wrong is that we're actually just wrong. Like, you know, this goldmine really does have gold in it, right? Or this drug that we're short really does work. And if I break down the sort of long history of the firm, we have far more type A errors than type B errors. Moreover, the type A errors hurt us far more. And the reason for this is, imagine I'm short this gold mine. I think it's a complete fraud. It has a market cap of 2 million, 500 million, and it has 2 million ounces in it. Well, if it's a complete fraud, then it should go to zero. But the problem with it going is that it could go to zero via 10X because if they've faked 2 million ounces, they might as well fake 10 million. If it's a real gold mine, it's not going to go to zero. In fact, it's going to double, right? And it will double as they take the gold out of the ground and distribute it in cash to you. But it's not likely to go up 10x. And the reason it's not likely to go up 10x is that it's really, really, really hard to find 2 million ounces of gold. And it's extremely unlikely that that the guy that found 2 million ounces of gold and is completely real is going to turn around and find 10 million ounces because it's really, really, really, really hard to find 10 million ounces of gold. That sort of stuff doesn't happen twice to the same person, or not very often, right? So the net effect of which is I'm far more likely to lose money on a stock if I'm right than if I'm wrong.
SPEAKER_01:A real-world example of this idea is the bankrupt German company Wirecard. You can think of Wirecard as roughly twice the size of Theranos and about 20 times as malicious. By the time Wirecard collapsed, they had a former Libyan intelligence officer stalking journalists, and they also got the German regulator to file criminal complaints against those same journalists.
SPEAKER_02:If a stock goes from$10 to$0 via$100, I'm going to lose money on it. And the reason I'm going to lose money is when it's$60, the position is now too big. So I'm forced to cover something. And when I cover things like that, I make a loss. And you might think this is a moot problem, but it isn't. The biggest loser in the history of my firm was a very big German fraud called Wirecard. It was probably the biggest fraud in European history. It blew up spectacularly in 2020. And everybody seems to think that I was very clever for working out that Wirecard was a fraud, but I promise you I wasn't. I worked it out in 2009. At that point, the stock was nine euro and it peaked at 191 euro. And last time I looked, it was seven euro cents.
SPEAKER_01:Wirecard went to the moon and then it came back to earth. On the way up, there was only one thing that could be done about it.
SPEAKER_02:It wasn't that bad. And the reason was that we covered a little bit every year on the way up. And so whilst we started 1% short and the stock went up, from nine to 190, which sounds like we should have lost 20% of the fund. It turns out we lost about four and a half percent of the fund over a decade. That's 40 bips a year. And at some point or other, your clients are going to ask you and the ones that did ask us to explain it. And we actually went through it. And later that just makes us look smart, right? Because we spent quite a lot of time with the most sophisticated of our clients explaining the position. And all of that makes me look very smart in retrospect, but thanks, I'd prefer a refund. And I knew it was going to blow up sometime. I just thought, you know, in 2010, I thought, well, that's the year. In 2011, I thought, well, that's the year. And by 2019, I was scared to add to it on the way down. And the reason I was scared to add to it on the way down is that it had hurt me so much on the way up, right? Because By this stage, this was something that had beaten me around for years. And the end was nigh. They had investigations into them. They had audit reports. There was$1.8 billion that nobody could seem to find in the Philippines. Turns out the$1.8 billion was never there. But nonetheless, it was getting near to the end. And I should have just been throwing more and more and more money at the shore. And I was scared.
SPEAKER_01:Some of the people I approach for this podcast bristle at the idea that what they are doing is gambling. They think of gamblers as people that buy lottery tickets, not people that make careers off repeatable edges. But John agreed to come on the podcast precisely because it's about gambling. He's curious if any of the gamblers that listen might have ideas about how to better manage his position sizings or how he could remove emotion from the process. He's open to the idea that gamblers might have some understanding of math, that could help him. Although, he's already considered and discarded some of the obvious ideas. For instance, he's thought about whether he can make bets in proportion to his edge, i.e. Kelly bets. But the Wirecard example shows why full Kelly bets would be a tough road.
SPEAKER_02:But it means that somebody in my position can't bet Kelly. And the reason they can't bet Kelly is that their business will not survive a Kelly betting process. I'm not even sure they can bet half Kelly. If you've got an edge, in order to exploit it properly, at least in size, you would need to be able to quantify it, and I find that very hard, but you'd also need clients that can accept what you're doing and are prepared to come along for a very wild ride.
SPEAKER_01:A reasonable question might be, if Bronte can stay short of stock for a full decade when it's gone 20x in the wrong direction, are they even capable of giving up?
SPEAKER_02:We have a hard rule, which says that if we've lost 5%, now remember, it's very hard for us to lose 5% on a single name. And the reason it's very hard is that our typical sizing is sort of 50 bits, right? But we have a hard rule that says if you lose more than 5% on a single name, yeah, you just give up on that name forever. We've never hit the hard rule, not once. The closest we've ever come is Wirecard, right? But the hard rule is still annoying. we have given up on names because the most common reason we give up on a name is imagine you know a typical biotech fraud that has a market cap of 250 million and 12 million of cash and it's telling you about this drug that it's going to make and you know that it's not going to work and that the wall street doesn't know and the stock goes from 100 million to 400 million and then Because it's a hot stock, they raised$200 million. So now it's a$600 million company with$200 million of cash. And then the stock halves. It's still a$300 million company with$200 million of cash. And we have to give up, right? Because by being a fraud, the company turned real. So yeah, we've given up on names like that as well. But we don't not watch. We would prefer watch with computers. And the reason we would prefer watch with computers is that that depersonalizes it anything i can do to sort of de-emotionalize it's a good thing but the goal here would be to watch it because you don't know where the people are going and maybe they got away with it this time but that doesn't mean that they'll get away with it next time the other thing is we've occasionally gone back we've had one of those companies then Spend two thirds of the cash. So they're now down to only having, say, 70 million of cash. And most of that cash has been spent on salaries for the senior execs. Right. So they've just lifted 130 million. And then they have another hot drug and they're suddenly a billion dollar company. And I want my computer to flag that. Right. But there are several reasons for keeping the bets small. One is that you don't know the optimal bet. That's why it's half Kelly, the third Kelly. But the second reason for keeping the bet small is it de-emotionalizes it. And if it de-emotionalizes it, you can come back. More and more times, I want to use computers. I want to use math to size positions. And the reason I want to use all of that stuff is so that I don't have to feel like it's me against the world, right? Because the moment I start thinking it's me against the world, I'll get depressed. I'll also make bad decisions.
SPEAKER_01:One name that John actually has given up on is Tesla.
SPEAKER_02:There's an old saying on Wall Street that if you mixed... Originally attributed, I think, to Charlie Munger, which is that if you mix turds with raisins, they're still turds. Elon Musk is the single best example of turds mixed with raisins I've ever seen in my life. Elon Musk bought a company that made... From his brother that made solar roof tiles... and put a whole lot of roof tiles on houses and pretended that they were solar panels and faked the demonstration. I can't even imagine doing anything so fraudulent. But Elon Musk is also the CEO of a company that took a rocket up to the space station and then landed it on a barge in the middle of the Atlantic Ocean. I can't ever imagine having that much achievement. And the problem with Elon Musk is you have... one of the most bizarrely fraudulent promoters I've ever seen, attached to one of the most successful CEOs I've ever seen. And he's what you might call turds mix with raisins. And the problem here is when I see the turds, I short it. And because once you've seen somebody lying once, you know, our slogan is once a scumbag, always a scumbag. Second slogan is, you know, if you hang around with scumbags, you're probably a scumbag. And Elon looks like a scumbag at one side of the mirror, and on the other side of the mirror, he genuinely looks like a genius. And maybe Elon is that sort of once-in-a-generation scammer, real person, genius, developer, fraudster, all rolled into one. And, yeah, when I found Elon lying, I shorted him, and I lost money. I'm not the only one, right? And then I... I had to question my assumption of, you know, once a scumbag, always a scumbag, because the guy is also the CEO of a company that took a rocket to space and landed it on a barge. You know, the guy has more achievement than I can ever have, and yet another breath is as nasty as any scumbag I've ever seen. Quite weird. So I've just given up on him. It's like, you know, my theory is that you should short scumbags and not good people, but he's both.
SPEAKER_01:In some sense, Jon Schaub is to understand and predict human behavior. He has to understand the people that run these companies, and he also has to understand people that invest in the market. That job is made harder because the composition of the investing population has changed.
SPEAKER_02:Once I start looking at shorts, I'm looking at the gamut of human behavior. And humans are just weird. If you had told me three years ago that you could convince four million Americans to buy stock in a movie cinema chain, I'd have told you you were mad. It happened. And there's a very famous example, AMC, which if you look at it, everybody's in it for some kind of financial manipulation reason. But ultimately, it's pretty obvious that every year, since in the 1950s, the average person went to the movies like 40 times a year, and it's now half a dozen times a year. And if you have a large screen TV and good stereo around it, it's probably one time a year because the trend is not your friend. Even more extreme, 10 million Americans bought shares in a company that sells video games on CD-ROM. When I think of yesterday, I think of selling video games on CD-ROM. It's sort of the dead market. I would never have believed that such a mania was possible.
SPEAKER_01:Human behavior is really weird. John looks for frauds so he can make money when they fail, but he also has an academics interest in these schemes. He thinks about what kind of people might be involved, where they might come from, and how culture might have played a role. He's like a fund manager, moonlighting as an anthropologist.
SPEAKER_02:Also, if you look at where frauds are, they come from certain places, and the places are red flags. The number one fraudulent place in the world is Vancouver. And the reason is that Canada has no national securities regulators. It has state-based regulators. And the British Columbia Securities Commission is extremely weak and to some degree has been captured by stock promoters. Second, the Canadians have a view which is that they don't seem to object to Canadians defrauding people south of the border. It's like Americans are fair game. So what you have in Vancouver is an absolutely beautiful city, lovely to live in, and a regulator that doesn't care about you. Toronto is fairly bad as well. Within the United States, the two hotspots are Boca Raton and Salt Lake City. And Boca is pretty obvious. Boca is just a city next to a bunch of rich old people. And rich old people attract people that want to sell things to rich old people, whether they be medical or investment scams or whatever it is, the scam. It's almost in Boca's question with some businessmen is, what's your scam? The other place, which is Salt Lake City, it took me a while to work out. And the answer is actually Mormons. I'm going to say something politically polite here, but in fact, it is fairly accurate. When Mormons are age 20, they go on missions and they have to sell religion door to door. And selling religion door to door is probably the hardest thing you could possibly imagine doing. It's just horrendously difficult. But at the end of it, You don't mind rejection and you're probably a darn good salesperson. Now, Mormons are 2% of the US population. They're about 6% of Fortune 500 CEOs. And a disproportionate number of those CEOs came up through the sales function. Mormons are pound for pound the best salespeople in the world. And they're the best salespeople in the world because they get the best training in the world. Now, that is good if you're selling software. But it's bad if you're selling financial scams. But it's the same skill set. And so it's not that Mormons, I think, are inherently worse than anybody else. It's just that they're better trained to be financial scammers than anybody else. I've observed tartly that about 2% of the American population are Mormons, but about 10% of the financial scammers are Mormons. And that's more a statement about their training than about their morality.
SPEAKER_01:You can probably hear more than a little amusement in John's voice when he talks about these companies. He's at least somewhat entertained by the ridiculousness of it all.
SPEAKER_02:I really don't like them in my own market. And the reason I don't like them in my own market is I'm sort of slightly patriotic about Australia. And Australia has a large compulsory private investment scheme called superannuation. And broadsters love this because Australia has very, very large pools of disengaged money. And I can just see people being ripped off left, right, and center. And our securities regulator is very weak and maybe even a little captured by the broadsters. But to be truthful about it, my business depends on piss-weak regulators. Every day, I would prefer nastier people in stock markets and more incompetent regulators. And so far, the world hasn't disappointed me. Right. You know, some regulators are worse than others. Canada is particularly bad. Canada is the only country in the top hundred in the world that doesn't have a national securities regulator. And that lack of a national securities regulator has become a feature, not a bug of the Canadian stock market. The Canadian stock market exists to rip off Americans to some degree. But bad regulators is a fairly standard feature in financial markets. And nasty people is a fairly standard feature. One of the beauties of the stock market is there's very few places that an ordinary person goes and gives$10,000 to somebody for a piece of paper. And they do that all the time in the stock market. So it's a really nice place for a fraudster to operate. The other thing that's very impersonal from the fraudster's perspective, there are con men who have to get into your heart. They have to touch you. The famous... love cons where you try to convince the person that you're romantically attached to them, a la the Tinder swindler, and you steal from them. But that actually is kind of hard and rather confronting. Stock market cons are very depersonalized. And because they're so depersonalized, they're actually easier to pull off. So there are far more con men in the stock market than you would give credit for.
SPEAKER_01:You might be wondering if the strategy of trying to find every fraud in the world actually produces returns. Does it work in practice?
SPEAKER_02:There's this wonderful software program for portfolio management. If you're actually in the money management business, I couldn't recommend it highly enough. It's a cloud portfolio analysis software called Novus. And the people who run Novus actually looked at our portfolio and they ran our short book and our long book. And they thought, well, your long book has about one and a half percent of alpha. but there's only ever been about 20 names, right? So that 1.5% of alpha, we can't assign any statistical significance to. Now, I happen to think that my 1.5% outperformance on the long side was taking lower risk stocks, but I'm just going to assert that because there's actually no real way of measuring it, right? We've got, you know, 20 stocks over 10 years. It doesn't matter. On the flip side, they said, look, you've had 1,100 different shorts. And depending on how we measure your benchmark, your alpha on the short book is 13% to 17%. And we thought it was more like 14% or 15%, but it was a lot. It was a ridiculous level of outperformance on the short book. Now, annoyingly, that didn't actually make much money. And the reason is in that period, the market was going up a teens percentage the whole time. So our short book was breaking even to making small amounts of money, right? Which was ridiculous outperformance.
SPEAKER_01:If you listen to the long-term capital episode of this podcast, then you heard the world's most famous example of the failure of diversification. Even though LTCM had many varied trades in diverse markets and geographies, Those trades became highly correlated in the fall of 1998. Diversification failed at the very time it was most needed. John went through a much less dramatic version of that experience during the meme stock craze. Somehow, all of the garbage became correlated.
SPEAKER_02:If you look at our short book, it just made money consistently relative to market. It basically broke even in the market rising teens. And if you pair that with a long book a la... the book expected returns and the maths around it, that makes money all day. And then when those 40 million retail investors turned up, they bid up the most crappy stocks you could imagine. And our short book was minus 20% alpha. I mean, my guess is that if it had happened for three or four straight years, our clients would have left us anyway. So the answer to how long could we have put up with it is more a question of how long could the clients have put up with it. I should have been elated because the opportunity set was enormous, right? Here are 40 million dumb retail investors to take the money from. And in fact, I was miserable. It is interesting because, you know, up until the COVID crisis, the shorting frauds, it's a pretty consistent way of making money. And then, you know, the joke on Twitter was that fraud became an asset class. The best performing stocks in the market were the most fraudulent, right? And the reason is that, you know, The average Wall Streeter knows pretty well to be skeptical about a gold mine that has suddenly announced 10 million ounces. But the average Reddit board reader doesn't know to be skeptical because they're new at the game.
SPEAKER_01:Bronte has also had some uncomfortably high correlation more recently.
SPEAKER_02:We short frauds, and there are frauds in various sectors. Undeveloped uranium mines have a very high propensity to fraud. Undeveloped gold mines have a high propensity to fraud. Speculative oil... Wells have a high propensity to fraud, but also alternative energy has a very high fraud rate. As I often say it, there, you know, there was a very famous fraud done by a friend of the former president of the United States. And what it did was it advertised on Facebook and said, you know, bernanke's going to destroy the dollar and um the union funds are going to bankrupt the states and you've got to put all your money into precious metals and it led you through a bunch of byzantine websites and eventually it got you to invest in a whole lot of gold mining frauds and what they've done is they've used people's ideology against them and the same fraud exists in clean energy where you um convince people, as I happen to believe, that greenhouse gases are a very serious threat to the world and that, you know, it would be very good if people invested in alternative energy. And having convinced them, you then steal their money. I call those the left-wing and the right-wing fraud. And we short both of them. And how you feel about those says more about you than about the frauds because they're the identical fraud. Now, normally, those are not correlated. Left-wing frauds are going up when right-wing frauds are going down and vice versa. And then along came Ukraine. And in the Ukraine, you know, suddenly there was a war and the gold price went up. But there was also a giant energy squeeze in Europe. So all the alternative energy frauds went up. And for that matter, the oil price went vertical. So all the oil frauds went up. And people started speculating that we were going to need uranium rather than oil. So all the uranium frauds went up. And so clusters of frauds that were not normally correlated got correlated at the barrel of Vladimir Putin's gun, right? And I didn't enjoy that.
SPEAKER_01:You can learn a lot about markets by walking into a casino and standing behind the roulette table. Specifically, watch what happens when a number hits twice in a row. On the next spin, the bets will just pile up on that number. Then pay attention to the rest of the world and you'll see that same kind of thing everywhere.
SPEAKER_02:The absolute best thing to do during this bubble or during this wild market is to embrace the things that have been going up very hard. I have an acquaintance with a fund manager who was up 100% about six months ago. He's given it all back. There was no way we would ever be up 100% in a six-month period, but then there's also no way we would give it all back. But the problem is that the clients who are outside can't see whether you are being sensible or not. All they see is the external returns. And yeah, we have some very sophisticated clients. One of the big university endowments in the Northeast has a day-to-day look at our book. In fact, they have a computer feed, which shows them what we're investing in. And they're very comfortable. And when we lose money, it turns out that they're often very comfortable, right? But the average client isn't like that at all. And this is a problem right across The stock market, it's actually one of the reasons why the market's so bubblicious, which is that when you get a really good run, say, in tech stocks or biotech stocks or in mining stocks in Africa, I don't care what the bubble is in. My favorite one was nickel mining stocks in Australia in the 1960s. But when you get one of those bubbles, the people who embraced the bubble have the best returns. And when they have the best returns, the clients come to them. And then they have more money. And, you know, eventually the whole stock market's run by the maddest people, right? Because the maddest people get the most money. And then when it unwinds, it unwinds just as spectacularly because the clients that would chase hot money will leave just as fast when the returns are not good.
SPEAKER_01:The thing I like about the puzzle we've been hearing about is that it is so dense. There are so many layers to it. Before you can realize the potential of market beating negatively correlated returns, There are a million different things to figure out, and John Hempton is so focused on solving it that he agreed to be interviewed for a niche gambling podcast on the very small chance that a listener might help. John doesn't need to do this stuff anymore. He's already a rich guy. But that kind of doesn't matter if you have a puzzle to solve.
SPEAKER_02:I finally passed that point where the money makes actually no difference at all. But you've got to do something. And every now and again, the regulators... You know, it's really nice to be able to say, hey, I understood Wirecard, even though I made no money. Or, hey, I understood Valiant and I made a fortune. That's a really nice thing to do. And, you know, kind of proudly, the number of people that have gone to prison because I've written a letter to regulators and explained the fraud now stands at four. And I kind of like that too, right? You know, there are different ways of keeping count, but that's one of them. But the other thing is human nature is just so goddamn weird. You find new ways of ripping people off, new psychoses all the time. The beauty of this is that I want to de-emotionalize it from my perspective, but I do want to explore. And the financial markets are a really nice place to explore. Now, I don't want to do what long-term capital management did. which is take a absolutely certain edge, exploit it too far to the point that they ceased to be rich. I mean, that's just stupid, right? So, and there's a trade-off there. At some stage, you know, if the edge disappears or the edge can't be exploited, we'll write to the clients and say, look, we're just going to become a long fund. We might even cut bees when that happens, right? And I suspect they'll trust us because... If we say that we did, because by that stage, I hope we have some very good results to demonstrate it. But at the moment, it's just fun.
SPEAKER_01:Risk of Ruin is written and produced by me. Special thanks to John Hempton for walking us through his very unique outlook on the markets. I'll post links in the show notes so that you can follow John on Twitter or check out his blog. If you want to get in touch with the show, you can email us, riskofruinpod at gmail.com, or you can follow us on Twitter, at HalfKelly.