ReSolve Riffs with Chris Schindler on Alpha as a Service, and the Future of Systematic Investing
“As the discount rate drives towards zero, it inflates every asset on the planet, including their correlations. Today’s dominant factor is discount rate risk.”
Fasten your seatbelt and dial down the speed on your podcast player, because you are in for an epic ride. With our good friend Corey Hoffstein (CIO of Newfound Research) as co-host, we were once again joined by one of the most interesting and thought-provoking guests we’ve had on the show: Chris Schindler (Co-founder, CEO and CIO of Castlefield Associates). This incredibly rich and nuanced conversation touched on themes that included:
- Who is “on the other side” of trades and investment strategies – persistent wealth transfers that lead to positive utility for both parties, with no “willing losers”
- Crowded and anti-crowded trades – drilling into the reflexive nature of markets
- What leads to crowding – structural and behavioral reasons, along with naturally-long players
- Why alternative risk premia (aka investment factors) have become shockingly crowded, and where investors might find blue ocean in the space
- Behavior doesn’t manifest in a vacuum – why the efficient market hypothesis is naturally at odds with human synchronization
- Alpha in traditional investments implies being early and/or different
- Social media, riot theory and the emergent phenomenon of crowds
- The ultimate CIO’s dilemma – where do we go from here?
- The moral hazard trade – governments, central banks, and self-fulfilling prophecies
- A basket of call options is always preferable to a call option on a basket
We also covered some of the underlying reasons for market synchronization, how many steps ahead of the crowds are likely to lead to profitable trades, and much more. This episode contains more nuggets of wisdom than we could count and is certainly worth saving for future reference.
Thank you for watching and listening. See you next week.
Chris Schindler is the Co-Founder & CEO of Castlefield Associates, a systematic relative value futures hedge fund that launched in May 2021.
Previously, Chris Schindler was responsible for a variety of roles during his 18 years at Ontario Teachers’ Pension Plan. He started in the Research and Economics team where he worked on the Asset Liability Model before transitioning to the asset management side of the firm as a founding member of the newly formed Tactical Asset Allocation group. Over the next 12 years, he was responsible for researching and managing a wide variety of systematic programs as head of the Global Systematic Investing group. Notable programs included an internal CTA, risk parity portfolio, alternative risk premium program, quantitative cash equities and the enhanced beta strategy, all of which were launched between 2005 and 2007, as well as a large number of pure alpha strategies (models not highly correlated to traditional alternative risk premiums). He and his team were also responsible for evaluating and hiring external managers in the systematic space.
In 2016 Chris joined the newly formed Portfolio Construction Group (PCG) as head of Asset Allocation and Portfolio Management. PCG was responsible for making recommendations to the CIO on Total Fund asset allocation decisions, including risk factor balance, asset class composition and weightings, FX hedging policy, tail mitigation strategies and constrained resource allocation.
Chris Schindler is the 2018 ITU Masters Duathlon World Champion.
At Newfound, Corey is responsible for portfolio management, investment research, strategy development, and communication of the firm’s views to clients. Prior to offering asset management services, Newfound licensed research from the quantitative investment models developed by Corey. At peak, this research helped steer the tactical allocation decisions for upwards of $10bn.
Corey holds a Master of Science in Computational Finance from Carnegie Mellon University and a Bachelor of Science in Computer Science, cum laude, from Cornell University. You can connect with Corey on LinkedIn or Twitter.
Rodrigo:00:01:30 You didn’t get a drink, what’s going on Butler?
Adam:00:01:32 I don’t know. I got…..
Rodrigo:00:01:34 That may be the first time I’ve ever had that come out of your mouth.
Corey:00:01:37 I want to say I was sitting here looking at my drink thinking this looks like a lot more liquor than I actually poured. It’s just all the ice melted.
Adam:00:01:43 Lots of ice. I was like, wow, that’s a lot of booze.
Rodrigo:00:01:51 Remind everybody we are live on YouTube, we’re also live on Twitter. Ask questions if you’re interested in hearing Chris’s perspective. And then Corey, Adam, me. I don’t know where you’re at Chris with your social media, but we should retweet that we are live now on our Twitter stream.
Chris:00:02:10 I’m on LinkedIn, does that count?
Adam:00:02:12 It doesn’t.
Rodrigo:00:02:13 LinkedIn, nope. You’re running a business, aren’t you in 2021?
Rodrigo:00:02:21 Now that we can actually say that you are running an actual live strategy.
Chris:00:02:24I need Twitter now? Ok
Adam:00:02:28 It’s a prerequisite. So speaking of you running a live strategy and having a live hedge fund, why don’t you tell us a little bit about Castlefield and what you do over there.
Chris:00:02:40 Yeah, sure. We are a relative value systematic macro shop, and it’s very hard to put yourself in a box because you’re a CTA like, oh no, we’re not a CTA. But we’re not. There’s always boxes everyone tries to put you in, but we kind of go relative value systematic macro, we’re trading futures, but we really like to think that we’re creating a product that’s distinctly different than anything else we’ve seen out there at the moment. So we think we’re really trying to do something unique and different. We have 20, 21, I guess at the moment independent models that are all working together, all kind of betting on the same assets to try and create some ensemble approach to trying to predict which way the market is going to go. Great team, super excited about the team. We’ve been up and running for six weeks, very excited how things are going. I’m really happy to talk about what we’re up to. But yeah, it’s been a long road, so it’s great to be here.
Adam:00:03:37 We were chatting earlier about the…yeah, go ahead.
Corey:00:03:40I was really curious. He said he just launched six weeks ago. When did you start the process of trying to get this launched?
Chris:00:03:46 Well, I resigned from Teacher’s in September of 2018. And so it’s super easy to start a hedge fund. I suggest everyone try it at least once in their life. But yeah, it’s a tough road and I think, I gonna be facetious here, I think like a private equity guy could probably leave, rent an office, set up a computer, do a web page and maybe buy a fax machine I’m not sure, and they’re good to go. It is systematic investing. There’s a couple years like I said, maybe at least a year of research, a lot of operating systems and architecture. There’s a lot of work, But it’s never dull, that’s for sure.
Adam:00:04:30 We were talking in advance about the fact that it’s often really hard for quants to tell their story, right? Maybe quants are not natural storytellers in general, more sort of classically left brain than right brains, maybe traditional discretionary managers. You can tell stories about the stocks you like and that gives him some insight into your process, your framework. With quants it’s a little bit more ephemeral. I know you’ve actually given some thought to how you describe your origin story and the origin story for your strategy and how you think about it. Why don’t you go for it man? Let us hearit.
Chris:00:05:11 Alright. There’s a long form version. So I warn you.
Adam:00:05:18 Before you do, I just want to make sure we follow Mike’s advice and make sure we let you know that whatever we talk about today is not advice. This is for entertainment and illustrative purposes only, and you should not act on this information in markets in any way. And with that, Chris, tell us your story.
Chris:00:05:38 Yeah, absolutely. And you’re absolutely right. I do think when you hire a discretionary manager, you talk to them, you get to know them, you get to know their thought process because they describe their thought process and their investment thesis. And I think it’s very natural, it resonates with a lot of investors. I think it’s really, really important for a manager to be able to tell a good story so that the investor understands how they think. And I think I said in systematic investing, the story is way harder. It’s way harder. And not only that, quants are not good storytellers. Everyone’s been to that conference where some guy stands up with formulas on a PowerPoint slide talking unintelligibly and no one has a clue what he’s talking about. And I find those really bad for the industry. I really think that when you start to invest, you have to start with a story for yourself, but you have to know what you’re looking for. Because there’s lots different types of stories.
There’s the what am I looking for story which is like a thesis that you start with. And that’s very important. There’s also the story of what did I find? I think it’s a less useful story because you can data mine, find some relationships, find some patterns, and then try and explain to yourself why that may have happened, which can be helpful but can also be dangerous. And then I think where the good, the star discretionary investors are extremely good, is the after the fact, here’s why I did what I did story. And the really good ones are the ones who, even when they lose money, can convince you that they made the right trade.
Rodrigo:00:07:02 Hashtag global macro.
Chris:00:07:06 And it’s hard, right? It’s so important that your investor understands your story because they have to hire your thought process, and you as people still. It doesn’t matter who. If you’re going to hire someone to bring them in house, you get to know them well. And when you hire a discretionary manager, you get to know them well, when you hire a systematic manager, I think some of the time you aare just hiring a return stream, or marketing. And that’s the perception as well, because I think a lot of times the allocators don’t know systematic all that well and so it’s hard for them to get to dig into the thought process, and that’s definitely a challenge. And so the onus is on the systematic guys to explain how they think and their thought process in a way that that resonates. So I tried really hard to think about different ways to do it.
Corey:00:07:49 How much of that is self-imposed by quants though? I mean, I look back to post 2008 and over the last decade, quants have done nothing but say, look how great we are. When you buy a quant process, it’s systematic, it’s disciplined, it’s not going to change and we backed ourselves into a corner because the reality is, being a quant doesn’t mean your process doesn’t change. It just means the way that you view the world is through a quantitative lens. We would hope there’s an evolution of process. But the entire smart beta ramp up was this philosophical push for consistency of process. And I think we did this to ourselves in a large way. I honestly think quants are very responsible for the current conundrum they’re in.
Chris:00:08:34 I don’t disagree at all. And there’s a lot to unpack in what you just said there. Because there are certain things where you can say I understand why I have a defined thought process and that thought process should have some stability to it, but the way you capture it, and even the different things you try and capture that, you have to be constantly evolving. If you are standing still you’re falling behind, there’s no doubt about it. And so your entire day to day, and if you don’t love research, and you don’t love coming with new ideas, you don’t love building new stuff, it’s not the right space for you, because you can’t just settle in and lock down and sit back and go retire. But I do understand that, does low vol or low minimum variance or quality or value, they can be done better and differently. I think some of the things have kind of unfortunately locked down a little bit. People can fiddle around at the margin, but the majority of that is probably locked down. I think that’s part of the problem as well for quant investing in general is that when something gets locked down, and becomes commonplace and well known, it gets crowded.
And, if I have a story, I have a thesis, and it’s across the board, it’s not just systematic investing but it’s, crowding is the enemy. There’s just no way around it. Whenever any trade gets too crowded, any idea gets too crowded, any position, any asset class gets too crowded, just by definition, its prices get bid up for the same cash flows you’re buying, the prices are higher, the cash flows are the same, your expected return becomes lower, your risks go up as it gets crowded, your correlations rise as all sorts of people are trading the same sorts of things. And so crowding drives down risk adjusted returns, it drives up correlations, it drives tail correlations, and then in systematic investing in alpha strategies, it’s even worse because the crowding is not just like hey, buy and hold, and you get what you get down the road and there’s other people coming behind you, at least you benefit from that, because everyone’s competing at the margin for the next trade. So crowding gets attacked at the actual time of execution as well for those guys and it’s a real challenge. When I was trying to go back to this, I’m going to repeat myself because I was like, what did we talk about in the first podcast again? I was trying to remember, and I read through the transcript, yeah, we talked about like almost everything.
And so we covered a lot of stuff. And then in the second conversation we really kind of dug into portfolio construction and betas and alt betas and putting the portfolio together and the CIO was challenging the major factors. And the main takeaway being it’s going to be hard to be a CIO because everything is crowded and I think it’s very, very hard for people to look around in the world today and go, where is my value buy? There’s some relative value, but where’s my absolute value buy right now. Are equities cheap? Are bonds cheap, its credit default cheap, like where’s the undiscovered country. And there aren’t many. And then here’s the other problem, not to get too bearish for a second here, but not only is everything expensive, but like your major sources of risk are always growth risk and inflation risk. We talked about this in the last podcasts. And those risks were the defining risks of the world for a very long time, because those risks, they dictated almost everything. And obviously, there’s lots of idiosyncratic things, but those are the major risks that kind of change everything.
Again, in the last 10 to 15 years we’ve had a new one show up on the scene, which is discount rate risk. And discount rates are supposed to move with expectations of change and growth and change in inflation. And in discovering everything that connects the cash flows to the price, the IRR that connects those things, it’s supposed to move with them. Let’s say like a third party comes in and completely breaks the connection and to a certain extent distorts that signal and discount rates are all driven towards zero and the price of everything rises, that becomes a very common source of risk at that point as well.
And I would say the major driver right now in our marketplace is the occasional inflation scare, the occasional growth scare, but mostly, what is the Fed going to do about those two things? Because that’s actually the major source of risk, is this third party response to those two things but the major driver now and the problem with that is every single asset class on the planet, is the present value of future cash flows. Every asset shares that same list of risks. There is no, stocks and bonds respond to growth and inflation differently. They both get hammered in a discount rate that rises unexpectedly, and also does the dollar. And so does everything. And commodities is a dragon, but in emerging, my first developed, man, it’s the wrong time. And this is like we’re very late stage in this process, you start to reduce yourself from two or three factors to one dominant factor and that becomes very scary. Stuff happens like yesterday.
So it’s an interesting day for anyone who doesn’t trade commodities. I think grains had their biggest down day since 2009 or something like that. Like these are these days that are very exciting.
Adam:00:13:22 Some of them had down days that were larger than any day since 1988. Like, it was a very long time.
Chris:00:13:29 I’ve never seen every commodity go limit down. I mean, not quite, but it was quite something. And so that why do we all move together the exact same time? It’s like…
Adam:00:13:40 Except for energies.
Chris:00:13:42 Except for energies.
Adam:00:13:46 So we’ve got this dilemma and it is a common dilemma. It’s a dilemma that’s common to every investor, and I think that was a big motivation for you to sort of start this group 10 years ago. And the problem has only become more acute since then. So Castlefield was/is really designed to be orthogonal to all of the major beta risks along with a couple of other alternative risks. Like you try to hedge out trend following as a factor. But you did a good job of sort of describing the meta story of quant, but not your own story.
Chris:00:14:27 Basically say like, here we are, as a portfolio constructor you’ve got these risks and it’s very hard to diversify discount rate risk, and so the one way…because all your assets are exposed to discount rate risk. And so the one way you might have a chance, that you always think might have a chance, is through alpha and active management and long/short. And again, there is a predominant form of alpha in the world today which I’m going to call buy high, buy higher and mark up, and call it alpha or buy high, lever up, buy higher, mark up and call it alpha. And I think we’ll find in the crash that that is not alpha. That’s beta, but like proper long/short alpha, buy and sell stuff.
If you have a basket of stuff that you’re long, and a basket of short, you have a fighting chance I think discount rate neutral. If you’re long equities and short equities, they both get closed, at discount rates I think you have a chance. The betas of the two, the discount rate shock who knows? And so you look at and you go, it’s a fighting chance, I think it’s got to come as a very, very important extra arrow in the quiver of investors going forward. It’s very important. The problem is, the vast majority of long/short alpha processes out there have really struggled over the last five to 10 years. And so this is like the conundrum, because unless you have literally just playing this, I wouldn’t call it the greater fool theory, that’s totally unfair. But I’d get there early and hope enough people come in behind you to push the price up for investing. Unless you play that game, it’s been very hard in the long/short space. And I’m going to take like a five second, five minutes and talk about…and I know I did this two years ago, but just to go through it again.
Like how have investors traditionally made money? I think this is really important. And so I will do it again and I apologize for people who are going to hear this for the third or fourth time, but like AQR wrote an excellent paper and I think it’s just very important to kind of back into it for a second. Warren Buffett, Bill Gross, George Soros, three of the most famous distressed investors on the planet. And in this paper, they kind of went through, they reviewed their returns. They said, how do they make their alpha over the years? And they were able to explain the vast majority of the alpha from those discretionary players through what’s now known as the modern alternative experience. And so, okay, that doesn’t tell you all that much just yet, except to say that the discretionary guys are doing what the systematic guys are doing and everyone’s kind of doing the same thing which makes some sense. It’s hard to make money being short things that tend to make money over time. It’s hard to make money being short equities, it’s hard to make money being short, like credit spreads…
Rodrigo:00:16:51 Anything with persistent positive drift.
Chris:00:16:54 Anything with a persistent positive drift which is all the things we just mentioned. But now the onus is on you, and this is where the storytelling starts is like, why do these things have persistent positive drifts? No one has to justify why the equity risk premium is expected to make money over time, or the fixed income risk period, and lending. I know when I lend to someone, that I expect to get paid back more than you lend them. And so we expect that there’s a flow of wealth from you to me when I’m lending to you. and no one debates that. And you expect to lend me more than I lend you. They go, so what about these other things? Like, why would there be a low vol risk premium? Like what could possibly explain that? Why would there be a value risk premium? Why is there credit default risk? Why is there a vol selling risk premium? Why is there commodity back … premium? Why is there trend following? And so instead of saying, hey, these things made money and they worked, and let’s just accept that that’s a fact. I think you got to come back to why? What is the driver?
And I think the key feature of that driver is that every single one of these, is an anti-crowded play of some sort or another. And so, what do you mean by anti-crowding? You can look at each one of those risk premiums and say, there is a concentrated set of players who are happily paying wealth to another set of players, because it’s in their best interest to do so. And I know we talked to this a couple of years ago, the key features that happily on expectation, paying money from one set to another, the same way as the equity risk premium, the fixed income risk premium. There’s payers, there’s payees and they’re both happily enduring the transaction. Well, why would anyone persistently pay money away on expectation? Does that seem completely irrational?
Rodrigo:00:18:24 Which flies in the face of most people say they’re willing losers. That you are stealing money from those losers.
Chris:00:18:31 If I buy a chocolate bar from you and I spent $1 and I got a chocolate bar, I’m not a loser, I’m happy. I wanted that chocolate bar. I was willing to pay more than $1. And so this is a marketplace, we can get together and I can give you something that makes you better off, and I can pay you some money for it and we’re both better off, we’re both winners, there’s absolutely not a loser in that game. And that’s the key feature of a risk premium, is that there’s this transfer of utility where both are better off and it can be through money, it can be through transfer of risk, there’s a lot of different ways. If you look at each of these things, what are the drivers? Why do people get caught paying money away? Why do they have to do it?
The Three Drivers of Crowding
This is just my story. I’ve got three kinds of drivers that I think explain why people will do this. And types of crowding, because it’s like I said, it’s always from a crowded space to an uncrowded space. And you go like the first reason that people get crowded is for structural reasons. And I can talk about, this is like regulatory or governance and they’re forced to trade in a subset of the universe, because they’re constrained investors. Now, the second set of crowdedness is what I’m going to call behavioral. And people get crowded because it’s just human nature to do the same things as everyone else, and that like naturally creates a set of crowdedness, and so if you can find the other side of that trade, there’s a flow of wealth. And the third type of crowdedness, I’m going to call natural long positions. And when you’re naturally long something, you demand insurance, and so insurance risk premiums, so people are naturally on one side on something. That explains every single one of the modern alternative risk premiums and every single thing that Buffet and Soros and Bill Gross, everything that they found and kind of came up with intuitively over the 20 or 30 years. They discovered it and it totally went, oh, here’s a pocket of crowded, here’s a pocket of under loves. And there’s a flow of wealth.
And so, like to give you some examples like structural, because it’s just my terms I go, constrained investors and I go okay, investment grade credit. The vast majority of investors who invest in investment grade, or sorry who invest in credit, are forced to only invest in investment grade. The vast majority are constrained investors; they can only do investment grade. In fact, a ton of them can only do or spend all their time in the seven to 10 point on the curve because this is the liability immunizing discount point for pension plans. And you go, so these are very crowded trades for natural reasons, because they’re meeting someone’s utility and if you’re an investment grade investor and a name gets downgraded, you have to sell it. And it goes from being a crowded, from a crowded asset with a high price and low return, and when everyone’s forced to sell at the same time, it falls out and it becomes suddenly this under loved, overlooked low price thing with a higher expected return. And these people happily sell it because it’s in their best interest, because if they don’t, they get fired, because their government says you cannot own this thing. It’s absolutely, they’re constrained investors. And by definition, any constrained investor who bumps up against their constraint would mathematically rather have the thing, but if they can’t, if something comes into constraint, you can only have so much, you’re not if it all falls over, you want expectation, we’d like to have it in your portfolio, it means that you expect it to be accretive to you, which means like taking out, you expect that you’re paying someone else to take it. And so there’s an expectation of flow of wealth, but you happily do it because it’s in your best interest.
Now, if you’re an unconstrained investor who isn’t forced to only invest in investment grade, great, buying the fallen angels is a well known strategy, it significantly outperforms. And so this is like just taking advantage of the effect of a constraint, and it’s kind of the gradient between the constrained and the unconstrained investor and there’s a flow of wealth across that gradient. I think that one’s pretty obvious. You have, other structural might be, low vol is explained, because you have a bunch of investors who cannot use leverage, or who have a very strong negative utility of leverage and you go, okay, in the maximum Sharpe ratio world, I would invest in all things. I use my risk free rate to lever up the low stuff, which by the way is what Warren Buffett did. He bought low vol names and leveled them up 1.7 times and that was a massive source of an alpha, because a lot of people can’t or won’t. And so you get crowded in the naturally volatile asset classes. I know we talked about this last time, but we found this in every asset class we look for it. We first found it in fixed income, because why is the 30 year bonds such a lower Sharpe ratio bond than the 10 year bond? And why is that bond is lower than the five year bond? It’s like, I can tell you what a pension plan does when they want more fixed income duration, They don’t lever up a five year bond to the dollar duration one. They move along the curve like everyone and they get in, the back of the curve becomes very crowded. It has all the volatility and half the return and you go well, that’s a loser trade. But if you can’t use leverage, or you don’t like leverage, that’s what you have to face.
And so in every asset class we looked at, the naturally volatile asset class is a lower risk adjusted return and the lower volatility asset class if you levered up, is a higher risk adjusted return and it’s a superior investment. It’s a flow of wealth between the constrained and the unconstrained. You have others, you have tons of structural players, if you’re an ETF guy and your job is to match the index perfectly. You go like, okay, so if I told you, hey, I know that a name is going to get added to the index and you should buy it early, it’s an alpha bet. And because you bought it before everyone else, and you’ll do better. And then you go, not my job, I don’t want to make alpha, I want minimum tracking error. I just want to buy it when I’m told to like everyone else, want the exact same return as everyone else. So when it gets mentioned, everyone else goes to buy, we buy at the same time, I have no shot here. That’s my job because that’s my utility. And so at the same time I know there’s a flow of wealth but I’m happily doing it because my job and my definition of my utility has been different than maximum Sharpe ratio.
And so, you can argue that anyone following a set of rules, who’s being forced to follow a set of rules, and there’s enough people doing it so that becomes a source of crowdedness, is going to have a flow of wealth away from them. And so that’s structural. Behavioral is like, is much easier. Behavioral it’s just, humans love to do the same thing as other humans. Is naturally, it’s very very hard to go against the crowd. And so this explains trend following, it explains value, at the end of the day what’s value? Value is like this name is getting sold down, it’s getting sold down too far. It’s getting sold down because by definition, everyone hates it. And when everyone hates something, there’s a reason, there’s a story, there’s a scandal, there’s a competitor doing well. Who knows? At the end of the day, there’s got to be some reason why everyone knows this thing, this dog, it’s really hard to go against the flow and go, I kind of like this guy here. And it takes a weird intestinal fortitude to do that, but if you can do that, you should capture a risk premium. Once again, concentrated- unconcentrated flow of wealth. And so the utility that you’re transferring is the sense of safety of being with the crowd. And that’s the utility that they’re buying and if you’re willing to pay that pain, you can capture that risk premium. And there’s lots of examples of those.
And then the last one I would say for crowdedness is like the way I think of it is, natural long positions. And we talked about this as well. But if you’re a farmer and you got a field of the corn, your entire livelihood is tied on what the price of corn is going to be in six months, you don’t want that. Rod you don’t want to sell all your wealth and buy only corn futures and sink or swim on what corn’s gonna do. That’s a garbage trade. If you’re a farmer you want to be in the business of growing corn not speculating your entire life’s wealth on what the price of corn is gonna be in six months. So you need someone to insure you. And like any form of insurance, if you go buy home insurance, you know that you’re insuring an unexpected loss. The insurer is expected to make a profit. So you understand that you’re doing it at a loss, but you’re doing it happily, once again, you’re paying that money away, that premium away, because on expectation you’re better off. If your house burns down, that’s a catastrophic loss, you can’t afford that. And so you pay some small piece of your wealth, but you have a massive tail hedge and you’re better off, you feel happier for that trade. And the insurance company takes what’s for them is a tiny sliver of return, and an even tiny sliver of risk that diversifies away, if your house burns down and it’s one of their 10,000, they’re fine. And so they are net accretive. Their risk adjusted return is higher, your risk adjusted return is higher, you’re both better off, both happy with the transaction, but there’s a flow of wealth.
And there are tons of those risk premiums. And what makes it exciting in the commodity world is it’s like the farmer is long corn, but Kellogg’s is short corn. And so the first thing that happens in the futures market is not that you go to a speculator and pay them insurance just to cover your insurance risk. When you go to the speculator you go…first of all Kellogg’s and the farmer get together, and they insure each other … That’s why the futures market is so amazing, because for free, these guys insure each other, but at the end of the day one of them is left holding the bag. Supply and demand imbalances and all the substitutes available. Either they insure…
Rodrigo:00:26:30 Or they get exactly what they want at the time that they make that deal.
Chris:00:26:35 That’s right. And once that imbalance goes too far, there’s either not enough people to supply the insurance on the other side, or the price gets too high and that’s where the speculators will step in and take the difference. So that’s it. Those are the three major sources of crowdedness. And so the genius of the Warren Buffett and Bill Gross and the Soros is that they realized that within each of these major asset classes, where a ton of long/short portfolios that can be created within them, that were the function of supply and demand imbalances are crowdedness imbalances that you can isolate and that they persistently made money over time.
Corey:00:27:15 I don’t mean to interrupt Chris, I think this has been sort of one of the most well established narratives over the last decade. I think that’s when a lot of the smart beta growth was on. I like your phrasing of crowdedness. I am curious though, I mean again a lot of these trades have now become prepackaged and available in a way that they weren’t 30 or 40 years ago. Even if I knew about fallen angels as an individual, it was very difficult for me to put fallen angels categorically in my portfolio until Vanek launched an ETF, and now there’s hundreds of billions of dollars chasing these structurally, which is going to have a reflexive impact upon whether these premiums can even continue to exist, because there was a consistent set of or not consistent, but there was a supply and demand imbalance and now all of a sudden we’ve eased access to one side of the trade, and I don’t think there’s more people on the other side of the trade. So you would expect the premium to diminish. I’m curious as to where you think blue ocean remains in this space? Because to me, it seems like a lot of the empirical evidence would suggest that the premiums have been dramatically squeezed over the last decade.
Chris:00:28:25 Would you say that those trades are getting crowded?
Chris:00:28:29 Okay, and there are anti-crowded trades that themselves have gotten very crowded. It requires a gradient differential. And what happens if enough people come to here, what happens if more? If you’re going to buy insurance and there’s only two insurance companies, they can probably charge you a pretty hefty insurance premium. They’re going to charge you some … and that amount. You come back and there’s 50 insurance companies offering you insurance, presumably that’s getting squeezed pretty hard and one of them is like crazy and wrong, maybe you can get it for like less than it’s worth. At the moment, these anti-crowded plays themselves have gotten shockingly crowded, and that has really really hurt the space, and that commoditization in the space is the main challenge and the problem.
So my point is, it’s not just hurting the systematic guys, because the discretionary guys also kind of do the same stuff and have for a very very long time. And so in many ways do everyone. Like I said, if it’s not a buy high, sell higher, mark up type of trade, it’s a very hard space to be in. So what we started doing was, I would say, this is even back in 2010, 2011 when I first started like the bones, my first set of ideas around this, was thinking like, oh, my God, the space that we’re in, which has done really well for us. And we had a great narrative back in 2004. Because back in 2004 this narrative, like when we put it out there, the response was…and I had a very very supportive set of bosses like Wayne Kosan and Mike Whittle and Neil Petrov, they gave us unbelievable amounts of time and leash, and we built some really good stuff that went really well. And I would say, at that time I think we put together something pretty, relatively innovative, relatively early for a pension plan. But by 2010, 2011, I was hearing the same story that I just told you, back at me from everyone. And it was other pensions plans.
I talked to tons of pension plans, like you guys should all be doing this. When I talked to other hedge funds and you heard it from banks, like bank products started really like coming out like I think you said, like the ETFs. And so this story, which was itself an anti-crowded story, began to get extremely crowded. What are we gonna do about this? Because this is our space, this is going to cause problems. And so like the first thing, and like I said, you’re always innovating. There wasn’t like a, now we’re on to the next thing. It was a constant like, what can we do about this and how do we take advantage of this? And what are the implications of this? And so what we really started to think about was like, how can we take advantage of this crowdedness in this space? Because like when I said you’re going to talk about those structural, structural guys, ETFs, who’s following a set of rules and doing something, everyone’s doing the same rule at the same time. What was very apparent to us was that everyone’s doing kind of the same thing. So this is like the beauty of systematic investing and the massive challenge of it.
And I say, once again, not just like all systematic investing, like doing something similar. But if you’re a trend follower, and we built a CTA, our CTA had I think at first, five different models. We had a moving average crossover breakout, some regression slopes, like … cost of like steel correlations, and different ideas and very early on we were doing an ensemble approach with all the different parameters we can come up with. And the kind of weird thing was they’re all like 80 to 90% correlated. They all did roughly the same thing. And really, when a big trade came, they all were trading at the same time. And then we went to talk to managers because it was one of the things we’re doing at Teacher’s was we had manager, external manager portfolio, went to talk to managers and we always be very careful and say we’re running on like a prop shop and we have, our filter was typically, send us your daily returns, we were aggressive against our internal model suite and we’ll see who’s doing something different.
And this is, the crazy takeaway is, you got 200 sets of returns. You could fully explain 170 of them with what we were doing. We had a very scorched earth approach and so it was like, oh, our models are saying that they’re all kind of similar. When we talk to managers, they’re all kind of doing the same sorts of things and we were looking for the crazies doing like something like a BitPrint that we couldn’t do. But the takeaway is like, this stuff that we’re doing, if I’m buying crude on a Wednesday morning, so is everyone. I have a pretty good sense of what a good chunk of the world is doing, because I’ve been running it myself, and I have a pretty good sense that that’s pretty indicative of what a lot of other people are doing. So it wasn’t like I can go out and say like, I know what everyone else is doing.
But I have a pretty good indicator that my model suite has gotten crowded. And so can I start staring at my own models and start to look at the implications of how different bits and pieces of it interact, when it’s something gone full and when has an asset got crowded, when has a model got crowded, parameters within assets to the model’s gotten crowded, and start to get a sense of what are the implications of that, because once again when anything gets crowded, you want to get on the other side of that trade. If you can find the relatively uncrowded piece you can build a long/short book, and you can build long/short books because when something gets crowded, either its returns get driven down, or its risk gets driven up. And you can build a long/short book by two things, where I think this guy’s got higher return than this, and I have no view about risk. You can build a long/short book where I go, I have no view about return, but I think this thing is riskier than this thing. So this is higher, this is lower Sharpe ratio, this is higher Sharpe ratio, and I can still build a flow of wealth between the higher Sharpe ratio and lower Sharpe ratio just by focusing on where the sources of risk are.
And we can start to look at that the asset level and look at the parameter level, the model level. So we started to build a ton of anti-crowded models within the alt risk premiums themselves. And the key feature of that, and the kind of the cool part of that is that they are themselves, alt risk premium neutral. And they’re also beta neutral. So you’re equity, fixed income neutral on average over time and commodities over time, but you’re also trying to follow neutral over time and it takes a lot of work, and keeping it that way is challenging. But we have like these 20 models and the key feature of them is they’re properly uncorrelated. Like the average cost correlation term over the last 3500 days is like .03. So they’re properly different from each other. And then of course, like anything, it becomes a portfolio construction and risk question to make sure that you can keep those guys uncorrelated and keep them diversified and size them right and build a portfolio …
Adam:00:34:08 Just to unpack that a little bit Chris, when you say you’re looking for anti-crowded factors, that might be interpreted as you’re looking to arbitrage crowded traditional factor plays, but I don’t think that’s quite what you mean.
Defining Anti-crowded Factors
Chris:00:34:27 Yeah, arbitrage is the wrong word. But look, how do you make money in this world? How do investors make money? Very high level, and this is going to be crazy, simplistic and probably wrong, but I’m going to go with you have to be early and or you have to be different. Because look at it the other way, if you do the same thing as everyone else, if you literally have the exact same portfolio as the average person, you should expect median returns. You cannot expect to outperform doing the same thing as everyone else. If you do the same thing as everyone else and you get there late, you should expect to underperform. And I think the key to alpha in this world, the key to outperformance, is early and different ,or early or different, if you can’t do both. And so what we are trying to say is there’s two sources of alpha for an allocator. There’s your managers alpha, and so I can say as a manager, what are we trying to do, is like we were trying to be early and or different across the board or we’re trying to stay away from crowded, but you have to stay away from the crowded trade, we have to find the uncrowded trade. And so what we’re really trying to do is that you can either anticipate flows, or avoid crowded positions.
And in both of those cases, and once again, crowded positions can be imputed lots of different ways. Like we do it from looking at our own internal model suite. We can go like where is this fully crowded, we can anticipate it from…we have one model that’s just literally has just a bunch of trading rules. And they’re kind of nonsense on their own. There’s a whole bunch of trading rules and a whole bunch of utilities and a whole bunch of dueling like agents that we simulate doing their thing, and if you just follow the trading rules, it doesn’t really do anything. But the real key is like, as those trading rules interact, when do they get crowded and what are the implications of that crowdedness? How can I lean against the crowdedness when it happens?
And so we’re not just going after the systematic crowdedness, trying to capture to the best of our ability, like lots of different sources of crowdedness. And then, of course, the key feature in all these is, every single one of these models starts with a story to myself, a thesis. And then once you have that thesis, we start going to look for it, and there always is a dynamic part to it, just like the bones of the idea can be very very stable. But you have to make sure you’re constantly adapting, adjusting because the market is not stable and the instability of the market of course, those volatilities you have to always be adapting volatilities daily, but just really asset change, instability, those betas. If something has changed, if you talk about what has changed in the last 10 years? Well, vol of vol is way higher, right? Like loss … with the crash, take 87 out for a second, when they get 2000-2003 was like, three years, 2008 was six months, 2020 was I don’t know, two weeks, very fast, very ferocious, vol of vol has gone through the roof. But also like the betas of interactions of things, things that used to be uncorrelated have suddenly become far more correlated. And the state of that change has changed as well. And not only has it drifted higher, but the movement of those betas are much quicker. And so one of the major sets of efforts and transformations is like, how do I keep these ideas uncorrelated? How do I keep my assets within my models uncorrelated? Which is obviously huge, but then how do you keep the models themselves uncorrelated from each other, because if pockets have a bunch of models, they’re interacting, and what happens if nine of them go long S&P? What happens if 14 go long S&P? What happens if 20 go long S&P, and how do we handle that?
Rodrigo:00:37:41I was, got to say there’s only a certain number of markets. You’re talking about 20,25 different models but they’re all interacting with a small universe of markets. You’re able to look at the overcrowded trades.
Chris:00:37:55 If there was markets on the on the left side, you’ve got stocks, bonds, commodities, FX, vol. I know you can play with a couple of, but let’s just say roughly speaking those are your liquid assets. I’m going to call that two and a half asset classes, like or two and a half risk premiums because stocks are a risk premium, bonds are a risk premium, commodities. I’m not convinced the long side to commodities is a risk premium. Maybe a bit, but I’m going to say not necessarily, there’s that like, the risk premium to the farmer or the risk of the Kellogg’s, who knows it can be on the side. And FX doesn’t even have a long side, you can’t talk about the long side of FX. So there’s obviously risk premiums within these things but there’s only two and a half assets. I think commodities is a complete misnomer. It’s like commodities like saying stocks and bonds, it’s got at least five kind of different things inside it. So maybe commodities. If you look at that, let’s say got seven assets. Because it’s a bit of a stretch. But let’s say you got seven kinds of things there.
The interesting thing is once you start to transform things through models, they become more diversified, you get more effectively independent assets. If you think of trend following, let’s say US and Canada equities are 80% correlated. I can tell you with certainty that after you trend follow US and Canadian equities, their correlation is less than 80. I don’t know how low it is, but I can tell you it is less than 80 with 100% certainty, because there’s only two cases, if you’re trying to play models, like exactly every time your long S&P you’re long in Canada, and every time your short S&P, you are short Canada, the exact same signal for both of them, then you’re going to be poorly correlated. If you’re ever long S&P and short Canada or even like half long S&P and then full long, anytime your signals are a little bit disparate,.The diversification, your signals kick in, and the correlation between trend followed S&P and trend followed Canada is below 0.8.
And so what you see is as you run through these transformations, these models, you end up with a more effectively independent asset. So maybe you go from like a really what’s more like four or five, to maybe like 10. Now, if we’re doing a second version of that. Now I’m going long/short trend following, which is going long/short S&P 500 and you got a second generation, what you’ll see is the average cost correlation term of your assets has transformed through models transforms or your model comes down again. So you get more and more effectively independent things to play around with. And so that’s a huge boost to diversification benefit.
Corey:00:40:07Chris, can I play a little bit of devil’s advocate to that. Couldn’t that implicitly be true whether your signals have value or not, because you could just say I’m going to introduce a random variable, that random variable that is your signal is going to… So to your trend following example, it doesn’t even have to be trend. If you just had a random variable, you’re drawing a random normal and that’s how you were position sizing your US equities, and you’re drawing another random variable to position size your Canadian equities. By definition, you’ve introduced the new asset but it doesn’t mean that asset, it is an independent bet but it doesn’t mean it’s an independent bet with value. Doesn’t mean it’s actually a new asset that’s additive to your process in any way.
Chris:00:40:54 100% agree, it does not necessarily say that at all. But you hope that your trend following transformation also has some alpha in it, that does create some of that value. So now I have some value created, in a more diversified basket. And so if your transformations are random, yes, your correlation comes down, but I’m not sure that’s going to help you in any way. What you’re looking for is additive accretive signals, but just know that as you add those accretive signals, as you go layers deep, that you’re effective diversification, the number of effectively independent assets you have after those transformations rises, which is what I’m getting at.
Corey:00:41:29 But I think that’s somewhat, what’s the word I’m looking for? Like it’s by definition, it has to be.
Corey:00:41:38 So yes, I agree. It’s introducing more diversification by definition, it’s a question of whether that diversification is valuable.
Chris:00:41:47 Well, yes, and so if you look at it you say, if I have things that have a lower correlation that also have a positive Sharpe ratio that will create a portfolio, yes, that’s valuable. Like, if you gave me 10 uncorrelated things, zero expected return, no. At the end of the day when you’re portfolio constructing, when you’re building ideas, when you’re putting models together, you’re trying to find positive expected return things that have relatively low correlations. That’s how you build a portfolio of ideas.
Rodrigo:00:42:15 May I put an asterix on that? If you’re rebalancing at a certain frequency, you might actually be able to provide a positive expectancy portfolio because of the rebalancing premium, Shannon’s Demon and all that. So diversification on its own with proper rebalancing might actually be accretive, for zero expected returning assets.
Chris:00:42:33 Yeah, but then I would argue that’s a source of alpha. I would just say it’s the source of alpha. I would say, you could take a non-rebalanced portfolio and a balanced portfolio, find the long/short book that is your rebalancing premium, call that alpha, extract out. Maybe, blow it up and put it back in if you want to and say that’s brand new, that’s a source of alpha for sure. And that noise capture? There’s a whole bunch of quant factors where people didn’t realize that you could go long the quant factor or short it. It was the equal weight of the factor itself that was outperforming the underlying benchmark that they were fighting against. And so this is one of those interesting things about equal weight. Part of the outperformance of equal weight is that volatility capture. For sure. And so you’ll see that was part of the outperformance of value weighted. Therefore is there some of that noise capture.
Adam:00:43:22 I want to get into routing as a bit of a social phenomenon. You described three reasons why things can get crowded, one of them is behavioral. And I want to get into whether you think the advent of social media has to an extent exacerbated an emergent social phenomenon. And we talked on a previous occasion about Riot Theory, and how Riot Theory may help to explain the emergence of crowding. How do you put those pieces together or think about that?
Riot Theory and Crowding
Chris:00:44:07 So I took an urban economics class in third year for fun. It was a pretty neat class. And so it had a couple of cool things. One of them was there’s two lessons, like two classes on Riot Theory. And so let me back up for one second, because human synchronization is such an important topic, because the entire efficient market hypothesis assumes a bunch of independent rational agents. And it doesn’t assume, what happens if everyone does the same thing at the same time? Like what are the implications, and it turns out they’re ferocious. And if you look at it you go like, humans at times will naturally synchronize and it just naturally, like organically and like I think there’s some really cool videos you can look at on YouTube. It was like, there’s an audience in like Prague or something at the end of a concert, and this is all doing their independent clapping. And then at some point somehow some tipping point is reached and there’s enough of people clapping at the same time or the whole audience goes like, it’s a tipping point. And suddenly everyone’s clapping at the exact same time. And you can see the audience going, what’s going on? This is weird. But it’s like suddenly you’ve got 5000 people just naturally organically synchronizing. I think that’s cool, no problem with that, right? That’s kind of a neat story.
And then you have other kind of synchronizations, which maybe can be more destructive. Like in 2000, London built their Millennial Bridge and it was like this brand new very exciting, they opened it to the public for the first time and they closed it two days later because it nearly tore itself apart in the first two days. And the issue was like, everyone knew that if you put soldiers across the bridge marching at the same time, that could be a natural frequency and if you get that wrong the bridge will oscillate to it and you can destroy it. So engineers know that. They make sure they engineer bridges without the natural frequency of human footsteps. What they got wrong here was that this bridge had enough flexibility to it that when people started walking, it started to sway a little bit, and as it started to sway, a tipping point was reached, and suddenly everyone had to start walking at the same time because they’re all counterbalancing the sway, and it started to make it sway more and more and more. And so they had forced a synchronization on a group of people and that forced synchronization was extremely destructive because these things become naturally self-reinforcing.
And that’s what the Riot Theory was actually about as well, because the Riot Theory was like there’s a continuum of people. And this is Riot as in like, who’s willing to get out there and throw a rock and smash a window, Riot Theory. And if you think about it, if you’ve got certain percentage of the population is just willing to go out and throw a rock and smash a window at any time, it goes why not? And the why not that most people answer is because, I can get arrested or go to jail, or get caught on YouTube and Facebook and get fired, because there’s repercussions. And the fear of that repercussion is what stops some people from doing it. Now, there’s some people who would never ,doesn’t matter how, if you could be guaranteed you never got caught, some people still never get involved in a riot. But there are plenty of people who will go, might be fun, why not? And as more people get involved this starts to get safer. It’s the same thing like, there’s a great video on YouTube of people dancing. And the first guy out there dancing by himself on the field, I think he’s high as a kite, is doing his thing and he’s by himself like 10 minutes, and then the second dancer comes in. The most important dancer is the second dancer, the person who joins and someone goes, oh that’s cool. And the next thing, everyone joins and it happens very quickly, there’s a tipping point that turns over.
Rodrigo:00:47:23 That’s my specialty.
Chris:00:47:25 That’s right, exactly. So Riot Theory is that, it’s enough people get involved, you got five people walking down the street smashing windows, the police are going to arrest them. But if 10,000 people go down the street, the police have to step back and try and clean up the mess the next day.
Adam:00:47:41 Indya Moore from Trans Friends has a really good invocation or metaphor of this as well with flocks of birds. So think about migrating birds and they’re flying down the coastline and only a very small fraction of the birds knows where they are at any given time. And it turns out, they’ve modeled this and they’ve been able to confirm it empirically that so long as a number of birds equal to the square root of the total size of the flock knows where they are at any given time, then there’s enough whatever it is, being sort of gravity between the flock that it can keep the flock flowing in the right direction. There is a criticality of information that then allows the entire group to rally in the right direction, all kind of the same phenomenon.
Chris:00:48:38 So that’s what I was thinking. I watched that video on clapping actually, the one on clapping like, Oh, that is so crazy, how people were going to get together and suddenly they’re all working together and thought, that’s a lot like what’s happening in investing right now. Because at the end of the day, the efficient market hypothesis goes like, we’re all independent beasts. And he goes, what happens? And we always had investment clubs and you had hedge funds who controlled a lot of money and pension plans do like collecting more money, but what happens when you have like a Reddit army or three or 4 million people who get together and they go, here’s what we’ve got to buy today. And we’ve got to try and it’s sophisticated, like we’re got to understand what the pressure point’s on the short sellers are, what a gamma squeezes but hey, if we all get together and no one sells, everyone buys what will happen? That giant collusion.
If five people go walking down the street smashing windows, you can arrest them. If 10,000 people go down the streets smashing windows that’s a riot, you got to step out of the way, and that can happen, if enough people get into it, it can go across the entire country. If 2 million people are walking down the street that’s a revolution, and that can overthrow a system and who knows what the system is going to look like the next day. And I feel like we’re in a very interesting time right now because there is a bit of a revolution going on where suddenly we’ve got millions of people going like, we can all in concert attack the market with a stated goal of disrupting it.
Rodrigo:00:50:01 Yeah, in a way that didn’t exist even 10 years ago. Whether it’s Reddit, Twitter, whatever the kids are using this days, Twitch, who knows?
Chris:00:50:10 Super interesting to think about synchronicity and how humans synchronize. And whether it’s organic or whether there’s something forcing people or allowing people to synchronize and then understand that synchronization can be extremely destructive. And so what you see is like changes in the marketplace, dramatic changes in the last couple of years. The amount of short like, the fact that the typical investor was an involved seller has maybe a bunch of money at the daily level right now, like massively long vol, you have to understand what are the implications of people who are just massive amounts of long optionality and long volatility. What are the implications of the marketplace and what are the implications of the marketplace now that a bunch of pension plans can no longer sell volatility, because like AIMCO had a blowup, and the boards won’t let them. Suddenly these are short term events that can transform things dramatically. And so as an investor you have to keep an eye on them and understand what are they and start to try to model? What are the implications and how do I need to react to the fact that this market day today is dramatically different than it was even two years ago. And the key feature being vol of vol, correlations, extreme moves, correlations and the tail, big fast changes in beta as a result of all of those, to me it’s been such a difference in the markets over the last two or three years.
Rodrigo:00:51:28 Especially when you’re trying to keep it neutral. When you’re trying take all of that off every single day.
Chris:00:51:33 When you’re trying to take that off, yes. Very hard.
Corey:00:51:35 I’m curious as to your thoughts about what I would almost consider to be regulatorily promoted synchronization. So there’s been a push, I’ll use target date funds in the US as an example. It’s a $5 billion dollar exposure in the early 2000s. Now north of two and a half trillion being used as the primary savings vehicle for most white collar employees, operating in a very synchronized manner for the most part. There’s small deviations in how the glide paths are defined. But, how does that play into your view? Is it sort of play back to the way indices are constructed, and there’s some forced synchronization there as to how capital is deployed?
Chris:00:52:14 Yeah. I know you guys have had Mike Green onto and he’s awesome. His story of hey, that’s some great synchronization when everyone, when they buy equities, buys the exact same package with the exact same weights, and what are the implications of that. Obviously I love it, like you’re trying to buy Microsoft, and Bill Gates isn’t selling his Microsoft, and so that market cap is wrong and the synchronicity of that is creating massive distortions. Any regulatory, like static constraints is going to cause inefficiencies and so if you’re aware of them and you aren’t forced to take advantage of them, you should be able to take advantage of them. If you’re not forced to follow them, you should be able take advantage. And I would say, the great synchronizer and we’re not going to talk about the Fed because I’m not a global macro guy, but the great synchronizer has got to be the Fed. Because like I said, we went from a whole bunch of idiosyncratic factors to not even like ,was growth, and go to growth. I mean, the job pump has been like this for a long time, but the job numbers come out bad and the market loves it. That is good news, why? Because the Fed is going to respond to it in a way that’s favorable for equities.
And so that’s the entire story right now, is what is this one institution going to do? And have they ever created a massive instability? And we’re all riding a tiger, we all got on this tiger in I always say ’97, some people might say 2008, but like whatever you think that the Fed really started messing with stuff. We’ve all been on this Tiger for a very long time and I just don’t see that there’s any way off and so this is…
Adam:00:53:45 We chatted about this on other occasions but as systematic investors who are students of history and the relationships between discount rates and valuations and expected forward risk premia, etc. And who understand or are observing these phenomena, these macro memes playing out over long horizons now, where everything can get pushed to extremes that we’ve almost never seen historically. As systematic investors, how do you think about navigating some of these? We talked about, a microcosm is kind of the Toronto housing market, but this applies to a wide variety of other markets around the world. I would argue it applies very directly to US equities as an example for reasons partly described by or explained by what Corey described. But as quants, how do we navigate markets that are so clearly dislocated because of these types of emergent social phenomena are coordinated or like social synchronization, or survive.
Navigating Dislocated Markets
Chris:00:54:54 I got a story for this.
Adam:00:54:55 Yeah, let’s do it.
Chris:00:54:57 When I first started at Teacher’s I was in a group called R&E, and we played on our very first off-site. I was just out of school and we were like, these are people who were…this was like the tactical asset allocation team. There was some groups who are responsible for portfolio construction. There were some risk people, there was a vol seller, fixed income guys, some people who were librarians, it was a hodgepodge. And so in our off-site as a sort of get to know, you we played a game. And if you guys have heard this game, the don’t interrupt, I’d love to see your thoughts on it. And the game was this. Everyone, we’re going to write down a number on a piece of paper between zero and 100. I want you to write down two thirds of what you think the average person is going to write down.
Rodrigo:00:55:44 I know Butler too well for this one I think.
Chris:00:55:46 All right. What’s your answer Adam?
Adam:00:55:49 Well, I know it. So I’m gonna refrain.
Chris:00:55:54 So yeah. You sit for one second go, it may take you five seconds, it may take you 10 seconds. But clearly it’s zero. The answer is zero,. You put two thirds of what everyone else puts, and you put two thirds and you put two thirds, and it’s got to go very quickly down to zero.
Adam:00:56:09 If you’re in a room of economists it is zero. If you’re not in a room of economists, it’s not.
Chris:00:56:14 Of course it’s zero. But that wasn’t the question. The question wasn’t the right answer. The question is, what is this room going to put down? Put down two thirds of it. And so I put down zero. Of course I’m right. And I was so wrong, I was so wrong. I think the average was like 30. And instantly ,after they walked around the house, someone put 75, they just didn’t understand the question. Someone else very proudly, everyone’s got to put 50 on average so I put 33. And someone else would. Well, I thought someone else was gonna put 33, so I put two thirds of that, like 22. One person went, I looked around the room and I went, I think some of these people are gonna go one generation deep, some are gonna go to 25. That was the right answer.
Rodrigo:00:57:00 You go down to maybe three derivatives, three levels and you stop there. Depends how much you respect the crowd. Adam: 00:57:13If you go into more than one level deep, then you’ll go all the way.
Chris:00:57:17 That’s the problem. Some people only went two. So anyway, just to say you think so, and this is the problem, you are not investing against the right answer. There is no right answer, you’re investing on what other people think the right answer is, or what they are going to, at some point. You are investing against other people and they’re not all as smart as you. And so you look at it you say like this is like…In 2007 I was like, this housing market is insane, who still buys it? And the answer is a lot of people, a lot of people. And when nine out of 10 housing markets in the world crashed and Canada didn’t, and everyone’s see, it was the right thing to do. And this is a very important thing. There’s a bunch of takeaways from this. But the fact is first of all, I’m terrible. I’m terrible at trying to figure out what people are going to do, I completely get it wrong, and so like just, as a quant. At the end of the end of the day you have to think about it, you have to put yourself in the shoes of what is the average person going to do, and the average person is going to go…Because I would say, if nine out of 10 real estate markets in the world crashed, what’s the riskiest real estate market in the world still? And I’ll say it’s the one that hasn’t yet crashed. But the vast majority of people go, look, this market’s never crashed. It’s safer.
And in fact, you have risk people, you have these people, prestigious pension plans going like, real estate has no risk because yours hasn’t crashed. They go well, I put my returns into this, somebody throw it in the optimizer go, we should do more real estate. And it’s like, ouch. But at the same time, if enough people do that, this is the moral hazard trade. I get it wrong every time. The moral hazard trade is that enough people do it, at some point the government goes like, if I owe the government a million dollars it’s my problem. If I owe the government or the bank to $10 billion, like the government they have a problem. They cannot raise interest rates now. So that was the right trade. It was the right trade. It turns out there was an institutional put, put in behind it, to protect it. I cannot answer from a discretionary perspective, what do you do in this world because I always go stay awake, stay awake, and I know how it ends, it has to end badly. And I think that’s the genius of George Soros, he goes, anytime I see a bubble started, I just want to get on it. I want to get on it early because that’s how you make money. And so I’m a run away from this because it’s going to end badly and I’ve done that so many and it’s clearly not the right answer.
Corey:00:59:36 Chris, is it fair to say that it’s almost a game of just stay one derivative ahead of the crowd. So if the crowd is doing zeroeth level thinking you need to do the first derivative. If everyone is done now with the first derivative, you need to go to the second derivative. And so you don’t need to go to the limit. It’s just always being that derivative further than the crowd.
Chris:00:59:58 Right. And you can call it the derivative further, you can say that you want to be one step, you want to predict where the crowds gonna to be. You want to find out where the crowds gonna go, you want to get there early. That’s obviously that’s the best. It sounds easy. It’s obviously the challenge.
Corey:01:00:09 It’s a big challenge for me this year because I’ve said, every time I say this is dumb, when I look at what’s going on in the markets, that was actually a great time to invest in whatever that is. Because I think there’s a lot of zero level thinking going on. That’s not an insult to what’s happening, it’s just, I’m the one missing out on making money.
Chris:01:00:26 Yeah, you’re wrong.
Rodrigo:01:00:28 You’re muted Adam.
Adam:01:00:29 I was wondering like, is it zero level or is it Nth level. It’s like the person who guessed 20 when the answer was 20 was able to read the right level of the crowd….I know that’s what you’re saying.
Chris:01:00:45 But he has, his thinking of Nth level was like what is this group of people going to do and how do I get in front of that?
Rodrigo:01:00:53 Right. It’s not what a rational player should do, it’s what this group of people is likely to do…
Rodrigo:01:00:59 …and get there beforehand.
Corey:01:01:00 At GameStop mania I had a friend text me who bought GameStop at 420. And he said, I’m sure I’ll lose money, but this is my middle finger to Wall Street. And I said Citadel is loving this. So it’s not a middle finger to Wall Street, you’re just giving up your money. But to me, there’s I’m not saying zeroeth level in a derogatory sense. But there’s zero investment logic going on there.
Rodrigo:01:01:29 Well, there’s no investment logic but there is a human logic. I just listened to this podcast series from the CBC on what was that exchange that went under, Quadriga. And the two guys who founded that exchange, started their money making endeavors by creating Ponzi schemes in these online forums where other people participated in these Ponzi schemes where it’s like, send me this money and I will return X amount of money for you next week. I can’t tell you my formula, but you’ve got to get that money. Everybody involved knew they were Ponzi schemes, but they were playing the human game. It was like gambling of I think I can figure out when I’m not going to be the guy holding the bag. … gaming that, right?
Adam:01:02:19 Remember Vernon Smith? I remember he’s got these videos for many…yeah, exactly. So he ran a lab for many years, I think it was at USC or one of the universities in California. And he ran this lab where he performed experiments on originally students, but eventually, like economists and floor traders and the CME and all kinds of things. And he had them play this market game, and we knew that the value of the asset was known at each point in time. But you could still trade around the value of the asset. So it was this sort of trading game and there was a terminus, and you didn’t know when the terminus was. But anyways, he played this game once and this happened with every single group of traders. See, he played the game the first time, the value of the asset declines over time but the participants create a massive bubble and then the bubble crashes at some point between 1/3 and 2/3, until the terminus. Then they play a game with the same players. Have the players learn the lesson? And they do exactly the same thing and they ask the players afterwards, you observed what happened the first time, why did you do this again? They said, well, because I thought I could get in and out ahead of everybody else. And then on the third time they effectively began to moderate.
Chris:01:03:10 That’s why I feel a little bit sad for the GameStop crew. But at the end of the day it’s like this whole holder. Obviously the risk of this is that you get a bunch of people who don’t really know what they’re doing and they’ve been told by everyone that if you never sell, if we all stay together forever, if you never sell. But of course you have to sell at some point .This isn’t got to pay dividends. I was, got to make it, but you know how this game ends. You have to sell. And so really, it’s the sucker left holding the bag at the end of the day that goes, I thought we’re all in this together guys. Of course, you need those people, but this whole standing up and say, hey, guys, we all in this together. It’s like, it is a giant, giant, giant game of chicken, no doubt, because if you’re ever gonna make money out of it, you’ve created a liquidity squeeze. It is going to come down very fast when it comes down. It’s a giant game of chicken, it’s a great fun game right now but at the end of the day it doesn’t end well.
Adam:01:04:30 So I would argue the entire market is an example of this phenomenon at the moment. So this is the CIO’s dilemma. So where do we go from here? How to put a guy on the spot.
Rodrigo:01:04:49 Look, just send them the wire instructions to your fund there Chris. That’s when you start saying Bank of America, SWIFT Code.
Where From Here?
Chris:01:05:00It’s going to be hard. And I don’t know because I look at this equity market and I go, really? Because I don’t think it ends well from here. But no one wants bonds anymore, and I still believe a diversified risk parity process is absolutely the foundation of how you start investing. But no one wants bonds here. That’s a hard sell. So what people might be looking for a little bit is, what do I replace the bonds in my portfolio with? And the answer is like, they should be levered bonds, they should be in there, you should have some, you might want to have stops on it, you want to have some protections in place to make sure that you protect yourself. At the end of the day, you probably want some in there because there are still growth and deflation risks in the market. But you probably, and this is what I was trying to get to last year, as you probably want a little bit less beta in general, because beta is pretty scary here. If you got less beta, you still want to make some returns, what do you do?
Now, if I’m an investor, if I’m an allocator, I got a couple different stories here. But basically, if I’m an allocator, I’ve got two different sources of alpha. I got my managers alpha, I’ve also got my alpha. And what’s my alpha as an allocator? My alpha as an allocator is finding good managers. So I’m going to call that like manager selection alpha, portfolio construction alpha, putting them together well. Some managers might have edges over others and their edges might be you can trade quickly or that you have a good mandate, or you have more capital, it might be that you’re less constrained because in any sort of crowdedness, it’s like who’s the constrained and the unconstrained investor?
There are some investors out there who have a mandate, who can go into emerging managers and some who can’t. Emerging managers academically proven, tend to outperform, and they tend to buy returns, and they tend to be more diversified, because they tend to be a bit different. So that sounds like what you want for your portfolio, but at the same time, they’re way harder to invest in because they’re way more risky, personally. It’s so much easier to invest in the big name than it is to invest in the small guy that takes more risk. It takes more management diligence, operate through due diligence, there’s a lot of extra work that goes into it. But if you’re willing to get your hands dirty doing the extra work, you should really get some premium for it and there’s a small number of investors who can. And I would say whenever you have a competitive advantage over the rest of the universe, it behooves you to take advantage of it. So I think that’s an opportunity for some alpha. I could talk about that a fair amount more. I also think that as an allocator putting managers together is a massively difficult problem, and an also potentially large source of alpha.
So I’m looking at it going like, if you’re a pension plan you go, I built my beta well, and then I think about alpha like that. And whether that’s like internal alt risk premium or internal active management, some people may still consider private equity and credit and infrastructure active, but I think there’s a little, you have to make sure you separate out what are the betas and what are the liquidity premiums, and how much of it is active, what’s the value added from active management minus the cost, and so you have to be careful with that. But let’s say there’s some value as well, and so where do we get our liquid active management? Like I said, if you want to get that discovery neutrality, you have to put a portfolio managers together? And man, is that ever hard? It’s hard to select good managers. It’s hard to select good discretionary managers, because you don’t really know what they’re gonna do. You’re buying into a thought process. It’s very hard for many people to select good systematic managers because (a), most allocators aren’t familiar with the space and (b), the systematic guys, they all tend to sell and tell the same stories.
Adam:01:08:15 It’s a really interesting conundrum in systematic. Obviously, who’s been the most successful systematic manager? I would argue, maybe AQR.
Corey: 01:08:28 Successful in raising money or successful in performance?
Adam:01:08:31 No, I mean successful as a business.
Rodrigo:01:08:34 We’re pretty happy, successful and like happiness, ReSolve Asset Management.
Adam:01:08:40 Okay, put a pin in that. But I think there’s a really interesting tension, because one of the reasons why AQR I think, grew to such a huge size is because they gave away their secret sauce. They published in very fine detail all of the techniques that they use to generate alpha and it allowed allocators to wrap their head around it, to present rigor to their investment committees, to look at back tests and say, this is what I’m extrapolating into the future. But by virtue of doing that, it also meant that it (a), you allow every other systematic manager in the world to use exactly the same techniques to generate exactly the same return streams including to a very large degree institutions like Teacher’s. Or a wide variety of other institutions who decided that they were going to build their own internal risk premia desks.
Chris:01:09:43 And so I’m a huge fan of the guys AQR. I think they’re awesome. I think that they are still doing some cool stuff, I think they did some really cool stuff. I’m not sure that they shared as much as you think they shared, I think the stuff they were doing, but they weren’t like, there’s basically different types of alpha generators. It was the guys who were the grand collectors and diversifiers, and they were doing some cool stuff themselves, but they were more like gamers like, let’s go academically, let’s click all the different ideas out there, let’s try and portfolio construct all of them together. And that grand collection of ideas is an absolutely viable way to create alpha. The problem is that the things that you’re collecting may be well known. I don’t think AQR invented any of them. He can talk a bit about like, some of the factors he worked on. But at the end of the day, I think what they did is they collected and put together a very good large and diversified basket of stuff. But that stuff itself was already known. We didn’t get our ideas from AQR, we were already running that basket when we….We were already well built, and so I think at the end of day, not like we invented them either, a lot of those things.
Adam:01:10:40 No, it’s not the invention of them, it’s the publication and mass distribution of them. I guess we’re I’m going is, to what extent did that engender crowding?
Chris:01:10:51 I wouldn’t blame AQR for the crowding at all. I would say you can blame me; I may have told as many people as AQR what I think is…obviously not as influential.
Rodrigo:01:11:02 It’s like synchronous discovery, everybody seems to have the amount of access to data, amount of white paper distribution.
Chris:01:11:11 I would say, many of the things the AQRs were doing we’re like collections of well-known risk premiums, and they wrote books about them. And so they may have helped a bit, but I think like their success may have driven rapid like competition, and because what they were doing was less a bunch of independent secret sauces and more of a grand collective, that can be probably replicated to a certain extent. I would say, look, every single strategy has its optimal AUM size. And so, one of the challenges is… I get this question all the time, which is, why would we invest in you when I can invest in D.E Shaw or Two Sigma? They have 1500 employees, they have 500 PhDs? And so, how do you compete with them? And I think the question is, you can do the alpha of collecting ideas, you can do the alpha of creating your own new ideas. And what we really focused on is we tried to build things that we think are truly different. And to that extent, they go I’m not competing with Bridgewater, with a D. E Shaw, with an AQR, with a Two Sigma. I’m not even close of course, I’d be crazy to compete directly head to head with them. Because when you say what’s competition, the competition is if they make money, I lose money. If I lose money, they make money. We’re taking each other’s money, and I go, we’re not even close, we’re doing completely different things. We’re not competing in any way.
But what I would say is that we’re both partners in helping our institutional investors meet their return requirements, we’re working together, we’re almost partners working together because we are in no way competing. But that doesn’t take away from the fact that I think we can be competitive. I think we can be a useful additional arrow in the quiver. But from that perspective, how do you get there, and you say like, you have to be different. And so different is hard. Like when I said the challenge of portfolio construction, I got to throw you guys another quiz here. I have two independent managers, or I’ve got a manager who’s uncorrelated to me. And we know that if the correlation is zero, the diversification benefit is like 1.4 times, it’s the square root of two. So my expected Sharpe ratio is 40% higher if I get two independent managers. How many .5 correlated managers do I need to have to get the same diversification benefit?
Chris:01:13:31 So four says, .5 is halfway between zero and one. And that’s an answer.
Corey:01:13:39 That’s an answer.
Rodrigo:01:13:42 You said some words.
Chris:01:13:43 Anyone else?
Corey:01:13:45 Well, not with a response like that.
Rodrigo:01:13:47 Yeah, that’s right, I’m terrified.
Adam:01:13:49 Now, I’ve anchored.
Chris:01:13:53 Seven, okay good.
Chris:01:13:55 Yep. Keep going.
Chris:01:13:56 No, keep going higher.
Corey:01:13:59 Well, it’s got to be a lot. It’s got to be a tremendous amount.
Chris:01:14:04 So this is crazy. I would say it’s over 100. It takes more than 100 .5 correlated manages to get to 1.4 times diversification, then the answer is, it’s infinitely many because it asymptotically approaches.
Rodrigo: 01:14:18 That’s right, it goes a flat line.
Chris:01:14:20 So quite literally you could find a thousand .5 correlated managers and not have gotten the benefit of one uncorrelated manager. And this is not intuitive. It’s not obvious. Uncorrelated is very, very important. And so how do you keep your managers uncorrelated though, because they don’t want to be, and this is the other major challenge. Imagine you can find 10 uncorrelated managers which is super, super, super hard. And the fact is that they’re uncorrelated, they each have like that one cool little Sharpe ratio .5 alpha of process that no one else has. That’s any good, that’s what I want you to run. If I can find 10 uncorrelated managers, then those 10 guys are uncorrelated, the diversification benefit is going to be square root of 10 like, like a bit more than three times.
Adam:01:15:06 Now try holding 10 uncorrelated managers on your book as independent line items.
Chris:01:15:13 Yes, that’s your first problem. So there you go. Now I go like okay, I can have a Sharpe ratio of one and a half, if I can keep these guys uncorrelated. The problem is no manager wants to run a Sharpe ratio .5 process. It’s bad business risk. You’re gonna to blow up at some point. And so they want a higher Sharpe ratio. Well, how do I get a higher Sharpe ratio? I put some other stuff in there, there’s well known stuff. There’s lots of well known stuff but even if I just put, and this is very interesting because you have a problem with your managers who go like, if your managers just put a little bit of one beta, say the S&P 500, bonds, it’s bonds, it’s vol selling, its credit, its trend following, when you put like a little bit of that in there, as a manager your Sharpe ratio rises, or you put half your risk in something that’s .5 your Sharpe ratio goes up .7, you get paid more. Your business increases, you’re happier.
If each manager just put a little bit of beta, like a tiny amount, .2, .2 here 20% of your risk, at the portfolio level, my Sharpe ratio has come down significantly and my beta has gone up significantly. In fact, if each manager just was 20% of their risk in one beta only, I’m going to be 75% correlated to that beta at the portfolio level.
Rodrigo:01:16:22 Which is what we saw nearly every active manager do over the last 10 years.
Chris:01:16:27 Exactly because they’re getting incentivized too because (a), investors don’t understand how important uncorrelated is, and they don’t properly reward it and defend it. It’s because at the end of the day if I’ve got 10 managers, I’m going to fire the one who does worst and keep the one who does best. I want to win when everyone wins. I want to lose when everyone loses, I don’t want to lose when they’re winning, I’m going to get fired. And so different is valuable and important, but it has to get rewarded right, and it has to get defended. But at the end of the day, why does this is math work? I like to say, the beta bio accumulates in your portfolio like mercury in a tuna. And the issue is like alpha is by definition uncorrelated. So when you put them together, the risk decreases. And so the amount of risk in alpha decreases. But if you put that beta in, it’s the same beta. And so the beta adds, and so if you look at your portfolio of 10 managers, the beta stays as the beta adds up and the alpha disappears, and the portfolio becomes more and more beta and say, I’m paying fees, my Sharpe ratio of my portfolio is like almost cut in half. I’m paying way more in fees, I have less alpha and I’m like 80% correlated by beta. What the hell went wrong? All it took was just a little bit.
And by the way, it’s never a little bit. And so it’s hard, you have a massive conflict of interest. And so, uncorrelated for managers is really hard. And you have another big problem with portfolio managers is that you never get as much vol as you want, because you go like, imagine you could do it. Imagine you could find 10 uncorrelated managers and convince them to only run their Sharpe ratio half, which is by the way, got to be very hard to sell, but you do it, you convince them. I’m never gonna fire you or whatever, off we go. And you put them together. And you go, okay, and they’re all running 10 to 15 vol, let’s just say 10 vol to make the math easy. That at the portfolio level, you only have 3% volatility.
Rodrigo:01:18:07 That’s right, with a Sharpe ratio of what?
Chris:01:18:10 One and a half.
Rodrigo:01:18:11 One and half. So low single digit returns.
Chris:01:18:15 You’re making your 5% return and you’re, you cut your risk legally, I was trying to replace equities in my portfolio. Why don’t you guys take more risk? And you’re like, I don’t want to take, I don’t want to take more than 50% of the Sharpe ratio of half, but that’s the problem. So now you’ve got this issue where your volatility is too low, your return on cash is bad and you’ve got a massive deep problem, because it turns out when you do that as well you have two major issues. When you put 10 of these managers together and you’re paying them each, let’s just make the math easy once again, you’re paying him two and 20 because like I’m convincing you to just run the Sharpe ratio half, I pay you two and 20. And at the portfolio level, you end up paying 65% of the returns away in fees. Well, why is that? Well, because, I can run through the math at some point, but it’s like, just take my word for it for a second. And so that’s a massive problem. I go like, man, my fees are too high. Even if I’ve solved the portfolio construction problem, I’m paying way more in fees than I wanted to, my return on cash, my volatility is too low, what’s going wrong?
And so if you think about the other side of it, if you had 10 uncorrelated managers, each with a Sharpe ratio of a half. Imagine they’re all in your one shop, they’re all under one hood. Because those first 10, you have a basket of call options instead of a call option on a basket. If the managers had no skill at all, you screwed up, they weren’t Sharpe ratio half, they had Sharpe ratio of zero, they had no skill. At the end of the year, five up and five are down and you’re paying performance fee to these guys and you don’t get it back from these guys. So you’re ending up paying 80 basis points a year on expectation and performance fees for no skill. Whereas, when you put them all together, the five go up the five go down, there’s nothing there. You pay no performance fee. And so that’s a big, big difference, but that’s actually not the main difference.
The other big difference is that 2% management fee, because you’re paying management fee per unit of vol. I’m buying 10% vol with 2% management fee. Okay, I put those guys together at my portfolio level and think 2% management fee for 3% vol, well that’s not great. If you had those 10 managers under one hood and you’re running that thing at 10% vol, that’s like each of those managers running at 31% vol, because that 31% vol diversifies down to 10, and you’re charging two and 20 for that. One sixth the fees. It’s one sixth, the return on cash is half the fees. And so it goes without saying that a multi strat, where you put the fees together, is an extremely fee and balance sheet efficient way of doing it, if you can get managers to cross, but discretionary managers never want to do that. If I make money, I want to get paid. If I lose money, I want to… Have a systematic models, they don’t care. Like 10 or 20 independent systematic processes under one hood is extraordinarily efficient for a whole ton of reasons. And it also solves those portfolio construction problems, because the manager who wants to cheap beta in, if you’ve got 10 models under one hood, you’re not going to put a little beta in each one of them and drive the Sharpe ratio down. Why the hell would you do that? You’ve taken where you were against your manager, where you had the opposite utility of them and selling your line because your managers portfolio constructed on your behalf, not on their behalf. And so those beta games disappear, a whole bunch of other utilities disappear. So I would say, portfolio construction manager is hard. But the extent that you can, find managers with multiple processes under one hood, that is …
Rodrigo: 01:21:36 That is an incredible point when talking about multi strategy quantitative managers, is I don’t even like to say that we have a strategy. I like to say we have a bunch of virtual managers under the hood. And we have a bunch of silos that seem to be similar. But even within those silos, we have a bunch of managers that are slightly different. Trend and momentum managers, or whatever. And you’re using the efficiency where we’re not even trading those independently, we are creating signals, we’re aggregating the signals and making one trade for all of them, which is another thing that is really difficult to do when dealing with discretionary managers or independent managers, even if you’re using them under a single platform. Because you can’t necessarily, or you won’t get cooperation. Each one of those independent quantitative virtual managers are not running, let’s say we’re targeting 10 vol, we’re not running them a 10 vol each, we’re running on a 30, 40 vol, but when put together they target 10, and it goes on and on and on and on like that.
Chris:01:22:38 There’s all sorts of other, I wrote a little white paper on this and it was one of those little toss ups I gave to presentations at the end. I thought it was fairly obvious, but it turns out like people, it was pretty well received. But, I forgot what I was going to say, I lost my train of thought.
Corey:01:23:02 And this is an endlessly fascinating thought that the path to go down. I love this idea of a basket of call options versus a call on a basket. I wonder, it makes in a tremendous amount of intuitive sense, I can see why hedge funds wouldn’t want to do it from a business perspective. But I also go back to the knockout risk of leverage itself. That asking a manager to operate at a higher degree of leverage, which might be good for you, might be inherently bad for the manager. So does it make sense say at the allocator level to then say, okay, we’re going to lever our book and just allocate more capital to the managers rather than asking the managers to take that leverage risk.
Chris:01:23:45 It totally would, mathematically identical, like I can give you $50 million at 20 vol or $100 million at 10 vol, obviously as a geometric drain difference, but call it very similar. If I want more exposure, I can just give you more.
Adam:01:23:58 Well, you’re paying more fees though then, too right?
Chris:01:24:00 Well, depends, like a technically incorrect, the correct fee adjusts for the volatility. Like I would say you should always adjust your fees for volatility. But if someone’s charging two and 20 for 20 vol they should be charging one and 10 for 10 vol, they should be charging a half and 20.
Adam:01:24:14 A half and 20, that’s exactly how we price but I’ve yet to see anyone else price the same way.
Chris:01:24:19 Well, I wish, I went to conferences where I had managers with they had two processes with the same fee structure 1.5 to 2 vol. Who in the right mind would buy the second one? It’s twice as expensive?
Adam:01:24:27 The answer is everyone.
Rodrigo:01:24:29 The answer is everyone.
Chris:01:24:30 People don’t want 20 vol, they want 10 vol because they can go to bed at night and not worry about it.
Adam:01:24:34 I’ve had the exact same conversation with Cliff. Why don’t you run this hotter? Because nobody wants it hotter? I know, but people don’t want efficiency.
Chris:01:24:43 But you’ve got to be clear, people have like their own personal sense of loss. They also have like their board’s. And so look, if you’re a manager and I know no matter what happens if I go down 15% you’re gonna fire me. I’m never going to let myself go down 15%, and there’s a significant cost to that. There’s this massive cost to fire you. Once again, let’s talk about our Sharpe ratio of half managers, it for a second manager at 10 Sharpe ratio for half managers. And you had a rule that says when they lose 15%, I’m going to fire them, and they don’t do it and they don’t play games, they just run their constant 10% vol the entire time, after 10 years you’re going to have to fire 14 managers, just on noise. That deal hasn’t changed, nothing has changed, and it’s because you should expect peak to troughs of 15% frequently, if they’re not hiding their risk and higher moment of risk, what you want a manager to do. You don’t want a manager to make money nine years out of 10, and lose eight year’s worth of money. You want proper random walk and if they do that you fire them. They go.
So what they will do, that’s a terrible game, because when you fire managers that costs a ton of money, it costs you like 50 to 80 basis points to fire a manager early. Why? Because nothing has changed, there’s still a Sharpe ratio of half, you’re gonna replace it with another Sharpe ratio of a half, but what you have is when you fire those, you left the underwater curve behind, you left the high watermark. And that on average is a drain and this is assuming you can with zero friction. Fire that manager and replace him with an equally good one, equally independent .5 that you have apparently a stable of them the next day, which you can’t. But if you could even, the fact that you leave 50 to 80 basis points. So it’s like, should I pay two and 20 or 1.2 and 20? You argue that endlessly, but if you fire him early, you’re costing yourself that.
Rodrigo:01:26:23 You can’t ask the new manager to reset his high water mark to what you just had?
Chris:01:26:27 No, of course not. So you leave that behind, that’s your first one. Your second one, if your manager knows he’s going to get fired and he does play games, he’ll start self-optionalizing. This is also very costly for an investor, for an allocator. The self-optionalizing is, look, it’s very hard to force a discretionary manager to take risk when they don’t want to. We had this in our group at Teacher’s, at TAA, because managers will play games and they’ll say, I don’t see the opportunity right now, you cannot make me take risk where I don’t see the opportunity. And so there’s a utility conflict because it may be December, and they want to shut this down. Maybe they’re underwater a bit and they don’t want to take risk. But like if a good trading day comes along, as an allocator, I want you to take that trade whether you’re up 10% or down 10%, whether it’s January, whether it’s December, that’s what’s right for my portfolio. And so that’s a real challenge, you cannot make a discretionary guy take risk.
And so when a manager self optionalizes, well, I’m paying you two and 20 for your Sharpe ratio of a half at 10% vol, and you cut your risk in half because you’re down 7%. Well, now I’m like I’m paying two and 2o for 5% vol and so if you think of it, like a 10% vol with let’s say there’s a Sharpe ratio of half that’s making 5%, and 2% goes way to management fees, when they cut the risk in half, Sharpe ratio half, suddenly, there’s nothing, there’s only 50 basis points left after management fee. So when manager’s self-optionalize, not only do you lose your diversification benefit, I had 10 independent point fives and now I don’t, that cost you, but also I’m paying double the management fee at the time. And so that cost another 50 to 60 basis points a year. And so when you look at it and go, like these are little small things, and you start to see how they how they add up in the portfolio and I haven’t even talked about how hard it is to find independent point fives. But to understand there are a bunch of behavioral challenges as an allocator that introduce these incredible headwinds. And oftentimes they’re not you, your board member goes, well, you have to fire someone that’s up 50%. Long story short to bring a full circle, there’s very few managers to go I’ll run 30 vol. Because of course, unless you go, we’re cool with you losing 50…
Rodrigo:01:28:31 Well, what is it at a .5 Sharpe, what is the probability of a 50% drawdown? It’s massive.
Adam:01:28:41 At what vol?
Rodrigo:01:28:43 I play at 10 vol. 30 vol, sorry.
Chris:01:28:48 I wouldn’t say it’s that high. So you think of the S&P is a .5 and it does 50 all the time. But the key feature of the S&P is, it’s got non constant volatility so that it gets to the 50% drawdowns because it does it at 30 or 40% vol, when it’s going around this. We call it a five sigma bull over the course of a year. Not super likely, if they’re probably normal random walk, but you’re right. The 1.5 sigma bulls will happen every two to three years, and that’s 15%. And so you’re firing people on what you…So this is the other thing, and like we are now in this…we’re six weeks in and we have entered the roller coaster, and it’s unfortunate because you wish this isn’t the way it goes.
Rodrigo:01:29:31 It takes into your strategy.
Chris:01:29:32 It takes into your strategy of live trading and you go, it’s tricky because you’re processed. If you’re trying to get a Sharpe ratio of one, you’re expecting to make four basis points a day. Four basis points a day is like, is nothing. But your market volatility is 70 basis points a day, one standard deviation or on bad days, it’s two times that, like once a month and so that’s like 140, 150 basis points you should expect to lose like once a month at least. And so you get a couple of those, it is not a statement of the alpha process, it is simply a statement about risk. Like market risk comes, it’s ferocious, it’s noisy and it takes a very long time for the process to prove itself. And so it’s just as hard to create a minus Sharpe one process as it is to create a Sharpe one process. If you have no alpha, you should expect to decay towards zero, unless you’re turning. But if your turnover is massively crazy, like massively crazy, like a minus one is just as hard as a plus one, minus two is just as hard as plus two. Zero alpha is zero returns. Anything like negative Sharpe is a market event, is market risk, is liquidity risk, has nothing to do with the process. But it’s very hard as an allocator to extract yourself in that statement because it’s hard to see past the noise.
Adam:01:30:49 You were acting in an allocator role for many years. Did you introduce any decision making frameworks to (a), make the best of your managers and maximize the chances you are going to get the most out of your allocations? And then at the broader portfolio level, any sort of decision making frameworks to get the best out of your team, internally?
Getting the Best Out of Managers and Teams
Chris:01:31:12 Oh, yeah. So other managers, first of all. Look, if you tell your managers you’re going to fire them when they’re down 10%, you’ve introduced so many moral hazards against yourself, like you have just punished yourself with that statement. And you should expect to say that at 10% vol, maybe it’s 20, maybe it’s 30. But at what point do you challenge the alpha statement? I’m not sure because once again, really big draw downs had nothing to do with alpha. They are a risk statement. And whether that’s a no. If your manager screwed up the risk statement somehow, we are going to do risk this way and they did that, that is a strong conversation. If it happened because something happened in the market. Like in March of last year, no one screwed up the risk. An incredibly almost impossible to predict and like very hard to respond to market event happen very quickly. You either had a great two weeks or terrible two weeks, there was very little in between.
That is less a statement about you and your risk statement than it is a statement about man, every now and then that the markets are insane. But if you feel like the manager screwed up their risk, that is a very significant conversation for me. So risk portfolio construction, monetarily philosophy style, I’m very serious about that because I’m really focused on what you’re contributing to my portfolio. The way that you can stick with a manager when they do badly is, if you actually get to know them as people, and get confident and comfortable with their intellectual thought process. Once again, very hard with systematic investing because systematic investors cannot give away their code and their secret sauce, because then it becomes useless. So at the end of the day, it’s like you have to get comfortable with them as people and their thought processes and so we would dig into model building philosophy, portfolio construction philosophy, thoughts about parameterization, how do you think about building the model and how do you think about doing it, and then I’ve got your correlations to know if you’re doing something different for me. What you’re willing to tell me about that thought process so I can get more comfortable with it. But it’s the managers who you never really, there’s a couple of yellow flags that when they lose money you go, I don’t know what to do now. You feel conflicted.
And this is once again inherently a challenge with systematic investing because the discretionary guy who told you a story about why they do what they do, it’s like if it didn’t work out, you’re like, but I got the story. And in fact, if they’re still holding on to it, like oftentimes they go, well, this thing got cheaper, it’s even better now. I lost money and I like it more. You’re never gonna get to see someone go, hey, the systematic investor lost money and I like it more. And that’s a difficulty in the space, but at the same time…Look, quant has had a really rough couple of years, a lot of factors that have been off for a couple years because … are talking to like that, and balance was going this way. The key feature of these guys is they have to be here, they don’t get to leave this process. They don’t get to leave their constrained box. It’s just these guys who get to decide if they want to be there or not. And if too many of them come to here, they’ve got to go no, this is not a game I want to play, it will come back down, it has to. These guys will always be here, these guys will self equal at some point, to the point where there’s a decent return on risk. Otherwise, they’ll keep leaving.
And so, the grand diversifier here who does this most efficiently, the best is still got to, it will return. Quant will like, absolutely make a comeback. It may never be as good as it was in 2000, it will never be as good as it was in 2005. But it will be good. It’ll be useful.
Rodrigo:01:34:16 The Sharpe ratio will be lower than what you saw when it was less crowded but there will be a positive Sharpe ratio given the inefficiencies that we see.
Chris:01:34:23 And I would say look, nothing right now is going to be as good as it was in 2005. Good luck buying infrastructure today at 2005 prices, or real estate, or equities or anything, everything is expensive and crowded. And so I think it’s everyone’s problem.
Rodrigo:01:34:45 Because I want to go back to continue on that thread of decision making because the way you were answering that question is manager’s decisions that they make, but we had a conversation offline about, we just had Annie Duke. We’re just talking about how do you create within a large organization, a better decision making process. And I would love for you to tell a story about the framework that you and the team came up with and then what the consequences of putting that framework together was.
Better Decision Making Processes
Chris:01:35:13 Sure. The end of 2015 when they created this new group, portfolio construction group at Teacher’s and I was head of asset allocation and portfolio management, which ultimately I was responsible for, and we’re trying to figure out, we think total fund portfolio construction. Everything to do with asset mix, stocks, bonds, commodities, factors exposure, hedging, alpha, beta, liquid, illiquid, and active budgets to the departments, we have to think about this and the other day we went like, and this is just my thought process was like, there’s so many incredibly smart people with Teacher’s and right now I got my very first committee meeting and they had all these committee meetings.
And this is just a way a bunch of investors will tend to get together. And here’s how our typical committee meeting goes, that goes into it. It’s an FX committee or an EM committee, and they get together, and someone promotes, pitches something and there’s 15 people in the room, because a lot of people want to get involved in the total fund which is fantastic, because it’s great that people are like wanting to get involved and want to contribute. And so you have 15 people in a room and someone pitches something, there’s a bit of discussions and questions are asked, and then there’s a vote. And I just couldn’t help but think, I read Tetlock’s book, I thought about it a bit, I couldn’t help but think there’s gotta be a better way. Because there’s a lot of, there’s a lot of problems with this. Because the first time we put something on, 15 people voted, we all approved it, but who owns the risk? It’s like, that was me. I was the throat to choke on that decision at the end of the day. And I went, like think about this a little bit. Because, here’s a bunch of the challenges. I got 15 people in the room, but I maybe only heard from two or three people. And whenever you have a bunch of people together, it’s the noisiest person, like me most of the time, but like it’s the noisiest person, it’s the loudest person, it’s seniority, what did the boss say? And so there’s a real art to collecting the independent view, so they don’t contend anymore. So that’s the first thing.
The second thing is accountability and ownership. Who owned it? People put this trade on, but at the end of the day if it went south, there was a whole bunch of duck and cover, because they didn’t actually own it, and didn’t feel a sense of ownership and accountability. And then the third thing is, how do you get off it? When do you end it? What’s the process for that? How timely is that? Do you have these meetings every two weeks and every two weeks and go, are we still doing this or not? And so I was thinking there’s gotta be a better way. And so the way I thought about it was can we capture this information in a timely manner with accountability and oversight and alignment? And so we spent a lot of time thinking about and working towards was it was a framework. but what happened? We put like, a trading game on people’s PC. In this trading game, you have, you had a bunch of assets that you could buy or sell. And if someone pitched a trade, you could add that as an asset and basically everyone, they could say, do I want to buy this or not? So you own it. But here’s what had to happen for it to work You have to have some sense of responsibility and accountability, and you have to care.
So as people have to get paid on this, you’re asking people to take a risk and to put their hand up and contribute to the whole fund, but if they went, we should buy stocks versus bonds, or we should buy IG versus high yield or like those trades, like can we wait to put them into a system where someone owns that, as a bet. But we’re not actually putting the bet on. Because this is the problem, because if a bunch of people like credit and a bunch of people like equities, and maybe they’ve got too much equity factor at the fund, I don’t necessarily want that exposure. I want to put that into portfolio construction framework. I’m thinking about what, but I have the information captured. But the key feature is this, if one and a half days in, someone who’s obviously long energy here, but I don’t like crude anymore, we sell it. I have real-time instantaneous feedback of what the person actually provided…
Rodrigo:01:38:41 High quality information, you may not use that, cost you some money.
Chris:01:38:48 Timely, independent. And so this is the way to capture like all of this, like market knowledge and fund. And who wants to put their hand up and play, and a bunch of people did, and it was like juniors and a couple of problems came up though, which was, we couldn’t comp people differently. That’s an entire HR discussion. But hopefully they get involved anyway just because it’s their job. But a lot of people went mm-hmm, that’s risky and no benefit. Maybe not. But a bunch of people still did. The second question was is this anonymous or not? What happens if I look like an idiot because I got it completely wrong. What happens I went against my boss, and I got it right and he got it wrong or vice versa. So the question is, do you make it anonymous or not? And that becomes, yes you could and then maybe not. Maybe anyone who wants to complain, you let people in back office and finance fully participate? I guess because we’re taking these singles and we can weight them however we want. You’re capturing information like maybe ideas come from anywhere. But really…
Corey:01:39:42 Aren’t you also risking just giving them a free call option if you’re going to compensate them on the upside but not penalize them for making bad calls, unless their job is going to be on the line. It’s like why not just gamble on high risk plays?
Chris:01:39:54 Isn’t that just invest to get a pension plan though. I’m going to get in trouble for saying that.
Corey:01:39:58 You tell me.
Chris:01:40:00 Anyway. So you take the high risk, that was an exact part of it. It’s, wouldn’t I just take the most volatile thing? It’s like, well, of course, I don’t want to do that. So I’m going to give you the unit that you’re buying is a vol targeted unit. Because I don’t want you just to play high vol versus low. Because yes, you would then take the highest vol thing you could and expect, an expectation get paid out. So vol unit targeted which is a little bit difficult to explain, I don’t want you to bet on 100 things, I’m looking for your portfolio construction, I’m looking for your ideas to limit the number trades. I don’t want the high frequency in and out everyday signals, I want the long term hold. So we had to charge like a significant update ask, to make it until we thought through a lot of these a challenges and all these problems.
And ultimately, it didn’t really fly and I think that the main challenge, there’s a bunch of challenges, like I said, there’s compensation challenges, getting people involved challenges, but there’s also a lot of pushback from the senior team going like, this is my job. I’m the guy, we’re the people making these calls. I don’t want to have to do what they tell me to do. This is my job. And I think it’s funny because like I heard Annie Duke say the exact same… I was sitting there nodding, she was like, it turns out these betting processes, if you can make them independent, anonymous, you can collect the information. And I think this one’s even better if it was real time and properly accountable. But if you can do that, it does tend to do very well. But the issue is challenges, and org structure, and a sense of authority and a sense of delegated responsibility. And if some people are going like, well, why don’t you just give us the money. We’ll trade it in our book. And it’s like, well, because I don’t know what we want to do at the portfolio level. We’re capturing information, not necessarily responding like…Anyway, it was a very, I thought it was really cool. I believe the software is still somewhere over there. But I’m not sure it’s being used.
Rodrigo:01:41:40 What’s odd about it is that it is a tool to help the people higher up to make better decisions that will ultimately help their careers. It’s just so short sighted to be like, no, it’s not about proper decision making, it’s about ego. That’s all I see. Because you wanted it, you were up there, you were going to use it. I. Well, I know they’re probably here watching I’m sorry, Chris.
Chris:01:42:09 There were a lot of difficulties with it that were institutionally challenging. I think it’s the way it’s been. I think it’s a very useful idea. If anyone said like, this is something we could do in our institution that captures information, there was also literally, why not put it on the dealer’s desks and let them contribute as well. But people who only kind of raise, what about the conflict of interest? You’re paying me and I could be long credit here, can I build a long/short arb in my book and it was like, it gets very complicated to get smart people thinking their way through. Like it depends on…
Adam:01:42:40 The best thing is that you’ve got this process which hits all of the major, ticks all the boxes in terms of optimal decision making. You’re taking the best ideas from Tetlock, the best ideas from Annie Duke and all the best decision making, best practices. You put it all together, you’ve got a really good system and those in charge hate it because it compromises their position at the firm. And it’s not at all unique to Teacher’s. I’m sure they…Sure. But if you apply this anywhere, you’d run into the same organizational frictions. You’ve got these fiefdoms, everyone needs to justify their jobs and justify their incomes and so it’s not just that you’ve got to have a great decision making process, you gotta have a great decision making process that also is aligned with the incentives and the structure of the organization. So all of these pieces need to fit together.
Institutional Decision Making
Chris:01:43:38 Look, there’s basically two types of institutional decision making pretty much. There’s literally like, can I click information and pull it up, or is it command down? Help me make a decision. I’ve got to make a decision and you’re going to implement it. Or, I’m trying to draw advice up. I’m a big draw advice up fan, but I totally respect that there’s another side to that. And I’ve heard it argued very well many times as well, which is the command down.
Adam:01:44:06 So let me make sure that we sort of wrap this conversation up with something. There’s been a huge amount of really interesting content, but like something super practical. So circling back to the CIOs dilemma, and we recognize you stated we certainly agree the most…
Chris:01:44:26 I’m sorry you’re talking. Can you give me one minute, I’m totally sorry. I’ll be back.
Adam:01:44:30 Of course.
Rodrigo:01:44:32 I just want to say…he should have gone to the bathroom before he started the podcast. I told him five times. It’s like my child. So the one thing that I think is important to hear because we get this question all the time as to why, your quants, I want to know the exact rules that you that you use and then when you use it, I want to make sure that you’re going to continue to use those rules, which is what all these ETFs are. Rules based decision making, index, create an index way to back test. Promise the back test, execute, you go flatline for a bunch of years. Those are seen in the same light as quantitative investors like us that are constantly trying to evolve and improve and move away from the crowd, and they’re not the same thing. But I think that’s an important distinction and we’ve been grouped into the same class. But we’re really active managers using quantum rules in order to be able to provide pure alpha over time.
Adam:01:45:33 Everything in the investment distribution process makes it difficult for systematic investors to innovate. You’ve got to describe your strategy in great detail in the prospectus and in the fact sheets. Any time you add new markets or new information sources, you need to amend this prospectus. Like it’s everything at the index ETF level if you make changes or you’re jeopardizing your passive index status, and it’s got the potential tax consequences, like there’s all of these incredible obstacles which is why there’s a trade off between access and the likelihood of success. If you have access to it and you’re not an accredited investor with a high level of sophistication, it is much more challenging for the manager of that product to deliver consistent innovation and consistently improve their product, because of all the headaches and major regulatory obstacles that there are in front of them…
Chris:01:46:47 It’s difficult for sure.
Corey:01:46:49I was just going to say, it almost raises the question of why bother calling yourself a systematic investor anymore? Just call yourself a quantitative investor.
Chris:01:47:00 I don’t know what’s a dirtier word. I got to be honest. Because people think equity quant , systematic models. Or I would just say like, at some point it wouldn’t the best thing to be like have some signals and make some decisions and just like literally think of discretionarily.
Rodrigo:01:47:13 You’re an active manager that uses coding in order to be able to execute every aspect of that process.
Chris:01:47:22 And I hate this term but there’s this continental term which is a mixture of…
Rodrigo:01:47:26 Oh my God. Our head trader just threw up on the screen right now.
Corey:01:47:31 But I do see that a lot in like the volatility space, where they’re just they’re quantitative investors, but they’re not systematic. And they do draw that distinction. And again, I think we put ourselves in the corner.
Chris:01:47:44 I believe in the discipline of systematic investing so much, I believe in the discipline of the vol targeting of the portfolio construction, I could not come close to doing what we’re doing. Twenty ideas, each of which has like 1000’s of sub parameters and you cannot do that.
Corey:01:47:59 I would argue it’s not systematic, because you’re going to evolve it over time. It’s quantitative, you are quantitatively enabled. Yes, you’re…
Adam:01:48:07 Not like a day to day decision making .
Rodrigo:01:48:10 There are boundaries between correlation, what you can do with correlations and volatilities though. And you want to understand correlations, you can’t do it by gut. It is so complex, it’s so multi dimensional theory, like it’s three dimensional in a way that you…
Corey:01:48:24 So you use quantitative models.
Chris:01:48:28 We’ve had this exact discussion, because at the end of the day there’s a taint right now on some terminology, and it is very interesting because people ask me, they still want to put you in a box. So what are you what box are you in like, are you trying to play? Are you convex or you’re mean reversion or, well, no. They go, but you’ve got to be one or the other. It’s like, no. Then you must have switching. Yes, of course we do. It’s like it becomes a very challenging set of question because they want to know what box you’re in, and I go like, look at my correlations. If I’m correlated to anything, put me in that box. But I’m pretty sure everyone I’ve ever sent my correlations to are going like, my returns are going, this doesn’t look like anything else that we have. I’m pretty sure it’s pretty unique and pretty different and from that perspective it’s very hard to think what box I belong in. Because a box shouldn’t be a name.
How do you compete with an AQR or Two Sigma? Why am I? Is it because you both have like a quant or systematic in the name because there’s a universe of difference in the way different people will approach problems. Like at the end of the day, correlations are where the rubber hits the road, on whether you’re doing the same thing or not. And so, at the end of it, different is different, and whatever tools are at your disposal to get there, different comes, not on the technology you use, but from the creativity that you have to get there.
I got another story. I know you guys are gonna to get tired of me very soon, but I love this stuff. He said, what do you need? What do you need in terms of creativity, because I love this analogy. He goes like, ants. Ants, when they wander off and they do this, they wander around and when they find food, they follow that pheromone trail back and they lay down more pheromones and they harden this trail on the other ants start following the trail, and there’s a line of ants. And they go, it’s a very efficient process. It’s a very efficient search process because like, once the food is found, like all the ants have a straight line to that food. And the crazy thing is, about 10% of ants, they don’t follow that line. They at some point they go, I’m going to go this way, and they wander off. And the thing is, it’s so important to have the ant who’s wandering off the line, because if you don’t, they all follow the line and at some point, the food runs out. It turns out, it’s a very efficient hairy search process where like a lot of those ants who wander off, they don’t find anything, but you need the ant who is non- aligned. Like I don’t want to follow this line anymore. I want to do something different because that’s how you stop following away, and the worst case scenario, it’s called a death spiral. The ants will start following each other in a circle. And like literally all the ants follow each other, they go on the circle that goes really in a circle until they literally starve to death. Corey: 01:51:13I’ll add a little color to your story there. It’s not just 10%. The more uncertain the terrain, the greater the number of ants that peel off.
Chris:01:51:21 I read a stat the other day because I was trying to catch up on it. It’s like different ants have a different propensity to peeling off. But yeah, it’s so cool. Because yes, this is exactly what you need. Like you need the annoying idiots who challenge status quo. As frustrating and annoying as they are, but you have to. And so this is where it’s like, cross discipline is so helpful from this perspective because I think you can challenge the status quo a lot if you come from the outside in. If you’re trained in a certain way of thinking, like I think already you’re caught in the rut to a certain degree. And I have that trend following.
The CIO’s Dilemma
Adam:01:52:00 I was coming out of the trend following story, I was coming out of circuitously. I asked you before about the CIOs dilemma. So I was gonna go into, you sort of start with global risk parity as kind of the core, and then you scale it to the appropriate level of vol which has its own problems which we’ve discussed, and then you bolt on some alts. So what are the alts that have the greatest propensity to scale well? What can virtually every institution bolt on? And then let’s sort of go down the capacity spiral, let’s start bolting on different potential sleeves of alts and then see if we can put a viable kind of portfolio framework together for CIO’s here.
Chris:01:52:57 Sure. So I feel like we led each other by the nose to this one, but like I got this I think. The first and most important one is trend following. Like it’s got a ton of features over time, they get very advantageous and one of them is like, when we first launched our CTA at Teacher’s in 2004, 2005, trend following was like out of fashion. By 2007, I think we fired all of our trend followers like externally, and we were still arguing. This is important, and 2008 comes along. We go like man, I you wish you had more trend following.
Adam:01:53:28 Hold on, I just want to make sure that we stop and describe what you mean by trend following because there’s a meaningful proportion of the people that are listening who think trend following is timing the S&P, right? So that’s not what we’re describing. We’re talking about 60, 70, 80 markets long/short, trend following with risk management, and appropriate portfolio construction, like what I call the diversified CTA type strategy
Chris:01:53:52 That’s right, when I say CTA, and probably like now it’s like pretty well understood that you have to risk target properly, you have to have an ensemble approach, you probably have to have multiple parameters, probably multiple techniques, and so I think trend following is probably done pretty well right now as a thing. And it got crowded too, like everything got kind of crowded but at the same time it didn’t do well for quite a while, people started to pull out of it and it’s literally in the middle of it’s like best six months. But 2014 was amazing because it got crowded, but every now and then, man you wish you had some more trend following. But there’s like other aspects to it that I really like. Trend following you can think of it as you take trend following, and slap on the risk parity, and now you have enhanced risk parity, you have risk parity with stocks. Well that man that looks good. If you ever want to build a nice back test, go long and out stocks and bonds, and then get long/short commodities, and that makes you something better, because trend following like has been around for 30 or 40 years, since the original risk parity.
Risk parity is a pretty new concept, but like trend following was equal weight by volatility, stocks, bonds, commodities, FX, and you put that thing together and hey, the other thing about trend following is it tends to sit long the risk premium side of things. Like things that tend to go up, trends tend to follow, and you stop that when it’s falling. Like call it a risk reduction if you want. Not even an alpha call, it’s like pull it when it’s not doing well, you can scale the amount that you want. Now you have a long process with vol targeting and some stops. That actually helps a ton. And so, trend following is so naturally complementary to the typical institutional portfolio. And man, if anyone’s lacking something in their portfolio right now it’s probably commodities. I know that commodities have been the terrible child of the last few days, but just to say over time…
Adam:01:55:35 Draw on this piece of paper where commodities hurt you.
Chris:01:55:40 They hurt me and then they helped me but at the end of the day…But if you look at trend following you go like, what is it? Like inflation risk is a very significant risk to everyone’s institutional portfolio because like inflation still hits stocks and bonds, and everyone’s got stocks and bonds. And so trend following doesn’t cover all sorts of… Inflation, I know we talked about this last time. You said we kind of went what is inflation? Is it monetary inflation? Is it CPI? Is a commodity price inflation? Is it demand side? Is it war time supply inflation? At the end of the day, is it COVID supply chain inflation? We have lots of different factors of inflation and it’s not really a silly, one way to capture it. And so we went like we don’t know how to cover inflation so we’re going to build a basket of things to try and catch it. So it’s gonna be a mixture of break-evens, which is the kind of real return bond over nominal bond, commodities, gold, different ways that you think about trying to capture this because actually, we don’t care about inflation per se. What you care about is that you know that unexpected changes of inflation is something good. I keep getting this wrong. Changes in expectation of inflation is what moves the price of assets. And this is key. Once again, it’s what people think, it’s how people change their views relative what their views were before, that moves prices of things. And so when people have an expectation of inflation, they change their expectation of inflation due to whatever, the news of the Fed or whatever is going on, they will respond. And the problem is when you expect inflation to be higher, the implications are pretty bad for most of the assets in your portfolio. And whether it’s because the assets themselves, then everyone goes, what’s a real price? It’s like everything on this planet is real priced.
If something didn’t have a real pricing power it would go to zero, because inflation goes up over time. Every single asset on the planet is real pricing. The question is, does it tend to make more money or less money when you have an inflationary shock? And in that case, it’s very few things that actually, are actually like inflation hedges. And so a TIP, a one year bond. A one year bond is an incredible inflation hedge from one perspective, because you go like inflation comes in higher than expected, next what does it do? It does its absolute best to price itself at exactly the place it needs to be to respond to that inflation. It just happened, like nothing responds faster to inflation than a one year bond. It is like literally the best. An equity is not even that. An equity is a series of cash flows. An equity is like, imagine you have a series of one year cash flows, like a series of one year bonds, a one year bond forward, and usually what happens in inflation? Well, in inflation, you lose money in that first year like you do with a one year bond. But then because it turns out you had a cost shock and it takes a while for you can pass it on, and you try and raise the price, but people are going buy less, and so you take a bit of a hit. But the expectation’s you can get it back over time.
Well guess what? If you do a rolled process of one year bonds, you take a hit your first year, but your next year’s pay you back. It’s a very similar concept, like everything is real priced, real estate, everything is. And so the real question is how quickly does it respond to an inflation shock? The problem is most things respond negatively at first, and then over time, we’ll catch up. Commodities, if the inflation shock comes from the stocks, commodities are obviously a very strong natural hedge.
Adam:01:58:45 Also, with the financialization of commodities, it ends up being a preemptive hedge. Capital flows to commodities in anticipation of the potential of an inflation shock. And actually they end up accretive.
Chris:01:59:00 Yeah, absolutely. And then it gets ahead of itself, and has to come back a bit, because obviously we’ve had some big moves in commodities recently. But like I said, it’s like trend following captures that.
Adam:01:59:12 Why do commodities trend? Like why should we expect them to trend?
Chris:01:59:15Goddamnit …Look, behaviorally speaking you can say lots of reasons why people track. A lot of synchronicity, or like a flow of information. I want to do what other people do, because there’s a lot are using major trend following, but I think there’s one, and this is like when I first…and I think many economists just get this wrong, and they get it wrong so badly. And you hear it, and you hear this all the time because they will look at this world in a deterministic sense. They look at it and they say like, what’s our supply? What’s our demand? We have this much crude, we have this much demand for crude the price should be here. Here’s the market clearing price. It should be this, and anything other than that is like speculators causing volatility. You’re like that’s maybe, but the thing is, like when I like I say once again, coming from cross discipline, when I first started, I did actuarial science and so one of the courses I took was operational research, and they had this, I thought it was a really cool course on the theory behind how lines are formed. It’s called queueing theory. And there’s some takeaways. Like it’s more efficient to have four lines to forecast. It’s like you have four lines to forecast here so you have one long line that that splits off into four. It’s always better to have one line that splits off to the four because the expectation is the same, but there’s way less standard deviation. You can get in the slow line or the fast line and wait longer or wait less. Same expected return, but less risk. You should always have one queue, and that’s a takeaway of queueing theory.
But the one that really struck me was, you can model the distribution of risk. And I know it seems to be, but he talked about all the times that people always think everything is a normal distribution, and in many cases everyone always thinks normal distribution. Especially when we’re pricing volatility in commodities we think normal distribution. Where, like where can it go? What is the range of possible outcomes, and you go I don’t know, 20% vol. And it gets it completely wrong. And it gets it completely wrong because it misses like a very important dynamic in the market, which is if you think about a cashier, this is like a one queue, one line up. A cashier can handle one person a minute, that’s great. Like one person shows up every minute and it takes me a minute to deal with them, I’ll never have a line. So people show up every minute and I take a minute to burn through them and I never have a line. And that’s true. So you call it like equilibrium, no line. And so they call it the clearing price, and if anyone shows up and it takes longer than a minute to show up there will never be a line. If they show up anything less than a minute, the line will go infinitely long. But you say like there’s a clearing price. And now you go, okay. Let’s say we clear. Let’s say we go like, we can handle one person per minute and one person per minute shows up and now you can say, let’s say we introduced a little bit of stochastic nature.
So it’s a little bit of randomness to the arrival priceSo it’s still onaverage, one person per minute. But let’s see how that changes the system. And the answer is it completely changes the distribution of how long you should expect to wait in a line. Because what you find out is okay, so you show up, and maybe someone shows up at a minute and a half later, someone shows up in a minute and a half later someone shows up, no line still. And then suddenly, someone shows up, and 30 seconds later somebody shows, up and 30 seconds later somebody shows. And now the second person is waiting 30 seconds, the next person’s waiting a minute and a half. And someone else shows, because it only burns down at one person per minute. And so this line, it builds up quickly and it burns down. And so every now and then it goes, it gets really fat tailed, and then it burns down and then it sits empty. And most of the time it’s empty.
And this is a system that completely clears. But suddenly, it’s got a line like 10% of the time. And so sometimes that line is like it’s quite long. I think, okay, let’s introduce another source of stochastic noise. The cashier on average takes a minute, but sometimes takes 30 seconds, sometimes it takes a minute and a half. And I see you model this whole thing with possible distributions, with exponential videos for arrival times, and you’ve modeled on this and there’s a distribution that gets created. Because now sometimes people show up, like three people show up in a minute and sometimes it takes three minutes to do someone, and suddenly this line can get really long. And that’s like a system that clears, it clears perfectly, supply and demand match. But at the end of the day, just do some randomness of supply and demand like noise, you get these incredibly long lines. And you say like, clearly the analogy to commodities is you never get a line, you get a price. You get like, how much do I have to pay, and if I really need this thing how much am I going to pay to get there? And you’ll see that most of the time it clears and sometimes it’s empty, and the price gets very low, and occasionally goes way further than you think it does, because it’s not normal. There’s nothing normal about that distribution. It’s a very fat tail distribution. Why does the price of sugar go from two to 20 and then back to two. It’s because there’s a stochastic noise to this thing and there’s no way to introduce new queues quickly. And so if you could introduce as many queues as you need to and take away as quickly as possible, then there’d be no problem. But the fact is, there’s a stickiness that it takes a while to build a new mine or to grow a new field, or sometimes there’s nothing you can do about it. And at the same time, all the vagaries of demand cyber, it’s like the price of this gets cheaper, I’m got to like move over here, and so at the end of the day the main takeaway is commodities have to trend. They have to occasionally, because the underlying process is so completely nonlinear.
Rodrigo:02:04:13 Yeah, the markets can be efficient, long term, while also creating these non normal distributions that allow for trend following or many other phenomena, I mean.
Chris:02:04:26 Yeah. And queueing theory goes like, it goes in operations research, which is like now you’ve got five machines, this queue goes into this queue, goes into this queue, goes into this queue, and how do you make that thing efficient and it gets very big. And of course, supply chain is that. It’s not one into one into one. It’s a whole series of these things and you realize that it can be right, and it can be right most of the time, and it’s a very fragile system and occasionally goes off a little bit because of randomness. And then you can get these really big price moves, because people need that thing. And if you need that thing, you wait a long time in line or you’ll pay a high price for it. And the thing is the key features is only commodities that have that effect as far as I can tell, because everything else you can perfectly fungibly trade the future for today. I can trade a future S&P for a today S&P, and the only difference is the risk free rate. I can infinitely create new queues, whereas commodities, if I need it right now there’s nothing else I can do but to pay up for.
So I think it’s a very particular specific, commodities specific dynamic until I’d say, there’s only one thing, like trend follow commodities. But the problem is, as a CIO, is that there’s not that much liquidity in commodities. There’s only five or six that have enough liquidity, but I would say it’s like the 20-25 most liquid commodities. If you’re like a $200 billion pension plan, there’s just not enough for you. You’re out of luck, you’re out of luck, you cannot hedge your commodities. And so you come out of different ways. You come at it with natural resources, infrastructure that might have peripheral exposure, you might try and buy mining stocks, but then got that equity risk. It’s like, clean, clean, clean, commodity exposure is very hard to come by if you’re a big plan. This is one of those areas, I always talk about this. When you have smaller pension plans or smaller institutions, if your high net worth, you have a very, very significant advantage over the big plans. And you don’t have many advantages over the big plans. But this is absolutely one of them. But there’s a bunch of things that you can do that they can’t. And like I said, always take advantage of your competitive advantages.
Rodrigo:02:06:20 And yet everybody wants to mimic the big plans. I want to run my portfolio, like the big pension plans. Why? Why would you do that? They’re full of constraints.
Adam:02:06:30 That’s why we ran our Master Class and specifically targeted solutions for those with less than $10 billion to allocate. Because once you get above that level, you are so limited in the solutions that you can bring to bear, that there’s not really a lot of differentiation that you can put…
Chris:02:06:46 In liquid capital markets it’s very hard to suck enough money out of liquid capital markets to feed a $200 billion machine. That being said, and this is why you can see like these big plans going towards privates, like there’s massive capacity. I’m still struggling with the buy high /sell higher, mark up, is that alpha or not, question. That thing that you bought five years ago and the cash flows haven’t changed, you just pay more for it. Yes, you’ve made money, you got there early to someone else was willing to pay more for the same thing and as long as, but if you went and bought another thing at that new price I don’t think you’re ahead necessarily, but if you can sell and get out of the market, then the question is like, did you exit completely and try and come back in later.
Adam:02:07:25 It’s like it’s like owning a home in a certain market. Your family’s here, your friends are there. The home price goes up in value. You think you’re wealthier but you’re not really unless you’re going to move to someplace that’s substantially cheaper.
Chris:02:07:37 This did, not to get back the moral housing play but like this is the problem with the Fed doing the discount rate, doing with the inflation, is that it’s completely disproportional. It’s uneven because yes, you don’t feel any wealthier, because you’re not. You’ve got the same house you had, but the person entering the market is screwed because…and that’s where you go and this is where the Reddit Army, Bitcoin, I know you guys love Bitcoin so I’m not going to get too deep, or so but like I feel these are intergenerational wealth transfers, trying to bring money back. And I think they should because at a certain extent the boomers have made this very unfair for most investors. But anyway that’s an aside.
Adam:02:08:19 Anyone else want to go anywhere else? We’re 10 minutes past two hours.
Rodrigo:02:08:24 Or we need to have a part two.
Corey:02:08:26 In conclusion what you’re saying is I should trend follow the most illiquid crypto. Yeah, that’s my takeaway. That’s where I’m going with this.
Rodrigo:02:08:35 It’s a market, that shows us.
Adam:02:08:39 Not advice.
Crypto and Books
Rodrigo:02:08:40 Not advice. Generally, from a structural barriers to entry perspective, there is a lot more opportunity for alpha in crypto in low AUM amounts than you do have in liquid markets and traditional…
Chris:02:08:55 Sure. The only thing I would say about crypto is it’s just very very high volatility, you could also probably go after 100% vol in crude.
Rodrigo:02:09:03 Yeah, but the leverage issue comes in there. Nobody’s gonna give you that much leverage.
Chris:02:09:07 But then it shows to be very very careful with that leverage. Crypto, not my thing. And the only thing is, not like I’m negative. Just more like I don’t know it.
Corey:02:09:22 There’s a new question in the chat here for you Chris. Book recommendations.
Chris:02:09:28 Interesting. I am going to circle back and say I haven’t read a good investing book in probably two or three years. Like I said, I love them like, I think there about10 or so books that I really love. I’ll send that out.
Rodrigo:02:09:47 Will send you Corey’s book later.
Chris:02:09:48 I’ve been reading like sci fi recently because I love this what if question a sci fi as opposed to what was, through someone else’s lens.
Adam:02:09:58 What are you reading?
Chris:02:10:01 Neil Stevenson, Adrian, the …the, like the fifth season I thought that stuff was great. The Three Body Problem.
Rodrigo:02:10:15 That’s right. We finally found somebody that loves The Three Body Problem.
Chris:02:10:19 I don’t think so. I didn’t love it. But I read it.
Rodrigo:02:10:24 We’re gonna delete that later.
Chris:02:10:27 But I thought it was very interesting. Like I said I like podcasts. I love listening to like, I would recommend to anyone to listen to Dan Carlin’s Hardcore Histor., I think that is like the most phenomenal podcast. And I think the coolest thing about that is like, humans are humans are humans are humans are humans. And if we’re talking about behavioral psychology, we’re talking about culture and the implications of culture and how you act. It’s like, the crazy thing is like 2000 years ago, or 1000 years ago, or 3000 years ago, it’s like, humans are humans, but the culture dictates so much of the actions into such incredible consequences. And it’s just really interesting to see how people react and move like in the culture that’s created for them at that time. And the crazy implications of like, what did it really mean for the first voting states in Greece to….when Athens first voted against to go to war with the Persians, it’s like, oh my God, they were like, that was the worst decision ever. And it happened to work out because of a couple of things that saved them, but the Persians would like have absolutely murdered them. And it was just one of those, well, wait, we’re like, we’re a democracy and why not, let’s take a fling on this. You’re Julius Caesar. A lot of his campaign was because he was in debt. He needed to go, the resource at the time was slaves. The motivation is in the factories, in the stores. I think he’s extraordinary. I have to be honest, I am so buried in finance full time, day to day. And I haven’t found a book that…maybe I should write it, that explains my thought processes cleanly. Antti Ilmanen’s book on Expected Returns is obviously absolutely worth reading.
But there’s 10,000 books on portfolio construction. They’re probably all quite good. But I think they’re all missing the most important point which is how much risk to take, and we’ll get into that in a different podcast, because this is a long and important discussion.
Adam:02:12:19 All right, well, let’s close a couple of loops. You got 21 strategies, how do you allocate to them? Is like 1, 20 first on each? How do you think about it?
Allocating to Strategies
Chris:02:12:33 Mostly, so I would say we start with that because if somebody back tests, they all back test about as good as a CTA or a Sharpe of one roughly, or a little bit better. But then you take into account if there’s any kind of big turnover differences, if there’s any breadth differences, most of them play all the assets, what happens when one of them is commodity-only? You take that in consideration. I would say ultimately, turnover breadth are the two that kind of need to pull back on. Some that aren’t as good. A couple of them correlate with each other at a .2, or .3, which bugs me because anything above that I’ll say, one thing, pull them together. Two or three .2s, this is once again the correlation thing. How many .2’s before they’re not independent and you’ve actually got, the answer is it’s a point three, and suddenly three become two. But at the end of the day, there’s a couple, and the bigger challenge is not like, the signals themselves, you can run independently, but they’re not independent things. They’re not hey, here’s 20 managers doing their own thing. What they are is 20 sets of inputs into what do I want to do with the S&P today. So at any given point in time that your signal in the S&P is absolutely continuous, it’s rising and it’s falling as these things are saying I want more or less and they’re voting. And then you’ve got other things coming into account, which is your vol targeting, your beta neutralization, all of those pieces with all the different betas that you have in there, are also adjusting for your S&P. So all that’s working together.
When you say these guys, if you let them run as independents, the real risk is that, even if they’re on average independent, every now and then three or four, they’ll be all leaning the same way,.Or six or seven will lean the same way, suddenly roiling up my portfolio level. I want 20 independent bets, I don’t want one concentrated short bond bet or long equity bet or short trend following, because I want 20 independent things, not one thing. And the secret sauce and probably the most important feature is, how do I respond to that without destroying the alpha signal, which when you want more of that thing, you want more of that thing when one point has become too much of that thing, and you have to start pulling back on because it becomes a contagion. There’s a cancer, and you want you want to kill it out. And so the portfolio construction and risk and beta neutralization is everything I would say, or it’s so much of it.
Adam:02:14:42 And beta neutralizing risk parity, beta and trend following beta.
Chris:02:14:51Sorry. Just got a call from somebody. I have beta neutralizing naturally by single generation, is your absolute best mate. I think they do when you generate, literally you say when you build a model, you want to make sure that the signals themselves on average they are zero. My equity on average is zero, and my bonds on average, every single asset on average is zero, but even then it doesn’t get up to neutral necessarily because you might be long in high vol and short on lower vol regimes. You might end up with some residual beta, but you do your best at the signal generation piece. And you do their your best at for all the beta, for the signal generation piece, and then you look at all of those guys together, you got now what is my residual beta that’s left over, because there’s always some beta left over. When we’re doing equity quant at Teacher’s, we built to the best of our ability beta neutral quant. How do you do it is, if I have a Canadian quant factor, I would beta neutralize to the Canadian market. Because that’s alpha, it’s uncorrelated to the Canadian market. But if I checked that alpha and check this correlation to other markets, I was 40% correlated the S&P 500 and 200% correlated the FTSE. Well, why, because the Canadian market itself is a pretty concentrated and idiosyncratic thing.
And so as they go, I gotta beta neutralize to everything. And the same thing, so I have to beta neutralize to the S&P 500. I still might have some correlation too until, betas, they stick around. And so you’ve got to handle it where you can, and then you look at the entire system and you handle them again. And then you can handle them in lots of different ways. Or you can handle them backwards looking, forward looking and different levels of responsiveness. And a lot of that has to do with am I leaning against a signal, or am I trying to demean a long thing, or trying to lean against when it gets too big. Like I said, there’s some art and science to it all. But at the end of the day, the real goal is to try and be as uncorrelated as possible to everything, while still having your positive returns obviously, because without the positive returns, I would say the lack of correlation is less than useful.
Adam:02:16:37 And then finally, do you have any expectation that your strategies are going to run their course? That there’s either crowding or some other phenomenon is going to drive the expected alpha to zero at some point?
Driving Alpha to Zero
Chris:02:16:55 Absolutely. And I think there’s two parts to this. The first one is, I think the themes may still persist for quite a while, because I think there’s always, like I said, it’s the same thing, these guys are here but the question is, how to capture them? Do you have to get faster? Do you have to get more reactive? Do you have to be more on top of the beta change? And the answer is yes. Because even though we use the beta, the market is constantly changing. And so even if the idea can still be captured, the parameterization to do it has to be adaptive and dynamic. And so we have a bunch of models which are constantly trying to find, not one sample look back at what would it have been, but how do I constantly try and figure out what it should be right now? This is why I’m so excited to do this. I love when someone builds a house and you’re like, I tangibly built a thing. Again, that’s the pleasure of physically building something. I get that pleasure out of building a new model or building. I love researching and discovery. I love the process. Yes, I’d love to keep going on this. And you’re constantly trying to push to the right side of the curve and add new things.
Rodrigo:02:18:02 Then I think we should take Jason Buck’s advice earlier on, that we should just call ourselves discretionary alpha managers. This is exactly what we’re doing. Why are we dealing with these quants systematic BS? Let’s just say we’re discretionary managers trying to constantly find alpha. Maybe we’ll be taken seriously then.
Chris:02:18:25 Yeah, I mean, like I said, the challenge we have is that it’s so replicateable. Someone leaves with your code then someone else can do it, and someone else can do it, and someone else can do it and suddenly that alpha disappears. And so, and alpha disappears really fast. Every good idea.
Rodrigo:02:18:43 But how’s that? Isn’t that the same thing as a discretionary manager? Once he gives away his model it’s the same thing? Think about Tiger Fund, the Baby Cubs, or the Cub Tiger or whatever they’re called.
Chris:02:18:54 They were systematic. If you’re a discretionary manager, and I say here’s the trade I’m doing, I’m buying bunds and selling China. What am I doing next month? Good luck, you have no idea. And that’s the beauty. I can tell you the entire trade, the entire thought process. And you get one for free.
Corey:02:19:14 I think that’s true if it’s an open enough universe for a discretionary manager. The number of discretionary equity managers I look at that have an identical funnel, it’s that upside down triangle that goes here’s our universe and then we’re going to screen on quality and screen on profitability and valuation and then we end up with our portfolio. And by the way, they have to do that, because all the allocators are trying to put them in a very defined box from a risk perspective. Again, that’s why you can explain so many managers with these primary characteristics and very few of them actually deliver any alpha at the end of the day. You’ve got these agency problems and…
Chris:02:19:57 Yep, I don’t disagree at all. Someone’s gonna argue, someone smarter than me is gonna argue that everything is a risk premium, there’s no such thing as alpha. If anything persistently makes money, they’re probably sitting. They’re either systematically capturing or discretionarily or heuristically capturing something persistent. Maybe, maybe not. There is a discretionary point to what we’re doing because every single decision that we make is a discretionary decision that’s just happened to me ahead of time. Every single thing we do is discretionary, which is why you’re investing in the person, because you have to buy into the thought process, the intelligence, the creativity, the integrity. And the integrity is, it’s so easy to cheat yourself in this space. And everyone does it out there, there’s no way not to a little bit. But the question is, I’m I cheating myself? We got to be very careful about that. Eyes wide open. How am I trying to build the best thing possible, and how do I defend myself against, that is the first question.
The second question is, am I cheating you? Which is, one of the things that we had the pleasure of doing over and over again, was getting dealer products produced to us and we’re probably the only team in the world that would be, you’re pitching a trade to us let’s just check it. I’ve already told the stories.
Rodrigo:02:21:18 Maybe to me privately, I can’t remember anymore.
Chris:02:21:20 Yeah. When a dealer said, we would do like hey, here’s a process. We’d test it. And the thing is, the dealers have gotten better with their product I think over time, but they were never that good, because the problem is you have this agent-management mismatch which the…especially challenges early on, is when you’re telling me you’re selling a slot that has no performance fee, just management fee. Obviously all you care about is selling as much as possible, not necessarily how well it does. And if it doesn’t do badly, if it does really terrible, you can roll it around start a new one, and your sales team will sell that one. And so why performance fee is so important is, because you’ve got to eat you’re cooking in some way. And another one in particular, this is one of the biggest probably. We had a Sharpe ratio of 2 presented to us, and we were got to test it. What assets, and then the second we said we’re going to test it, they came back and they said, yeah, it’s only a 1.2. The glitches they walked back from a two to 1.2, the second we said we’re going to test it and they went, oh, okay, that’s cool. Either way, what assets did you use? They went, we’ll use this and this and this and we’re like, it was a currency thing. I said why didn’t you use the Yen, why didn’t you use those ones? They’re like, oh it’s the most liquid. I know, you got Canada and you don’t have the Yen, and that seems weird that you’re missing.
So of course, we tested on all 14 currency pairs that we had and at the end of the day, but they chose the eight that worked and got rid of the six that didn’t, because they’re trying to create a back test. That’s something, so that 1.2 came down to a .6. And then we went okay, so what parameters did you use? And you start to like literally, this thing that was presented to us, when we first started, we think it was like a .4. But then we went okay, there was actually a bit of an idea in there at .4, and then we kind of went how can we turn it into something else. But the fact is, there’s a lot of decisions that you can make that you wouldn’t have known to make at the time. And those can artificially inflate your back test and they can result in a worse process going forward. And so you have to make sure that there’s a proper alignment, to make sure to reduce the probability of that.
Rodrigo:02:23:21 I will just add because you mentioned that to be to be successful at making money in asset management you either have to be first, you have to be smart, or you can cheat. And I think that’s exactly what the movie Margin Call, the character goes on and says, before he decides to be first. You can be first, you can be smart, or you can cheat. And I think a lot of these major banks are cheating knowingly. Not always. But you can see it. They make a lot of money doing that.
Chris:02:23:54 I think they’ve gotten better. Offline, I’ll tell you a story about the East German, and because I rowed a lot when I was younger, we had the East German and National Teams over to coach the Canadian team, and they brought over what was at that time the brand new technology for training. How to train better, and literally I made gains in my performance that were unbelievable by just changing the way that I trained. But it took two years, there was huge push back against it. And I’d say, that’s a mixture of smarts and technology. Once again, the alpha of that disappears over time and it turns out Canada was just behind East Germany.
Now, when you’re talking about technology and smarts, of course they were also cheating because he didn’t bring his drugs over with him. But at the end of the day we were really behind, because we didn’t have a drug program. And then that alpha, in most sports, it’s exactly the same thing happens, and that alpha disappears over time as well because technology starts to spread out, and then all it does is just raise the playing field and that advantage you have, which is I train better than anyone. I raced at the world duathlons championships two years ago, three years ago and time goes by, because I had this technological advantage. I knew how to train better than almost anyone. But bit by bit I’m telling my friends. Oh my god, you’re training wrong, you can do this, and they all get 10% faster. And then all that happens is well, now everyone’s just faster, and that advantage disappears. And of course, over time this advantage disappears, the skill reduces and it becomes more and more of a luck question. And we could talk about the others, that’s a totally separate topic, but that is technology, intelligence, dedication. And then what Canada didn’t have at the time. I’m going to guess still is the is the cheat.
Rodrigo: 02:25:45 Okay, that some good tech.
Chris:02:25:50 Yeah, masters.
Rodrigo:02:25:53 I was, got to go back and shoot myself in the face.
Chris:02:25:55 I was super excited, was, got to train. When I first left Teacher’s I got this year off, I have to take a year off. I’ve got to train for this thing. I gotta go. I want to go win it next year. So the race was in, it was in Denmark in June and in September I got an email going hey Chris, the guy who won your race failed his drug test, he’s been disqualified and you’re now the world champion. I ignored it. I thought it was a bullshit letter. Send us your medal, we’ll send you… two weeks later, hey, just so you know it was exciting. They had a two page spread on him and DNF, etc. And so then I was a World Champion. I never want to do this again. So I’m done.
Rodrigo:02:26:40 That was awesome. That’s amazing. You probably missed some good sponsorship opportunities because…
Chris:02:26:46 Now I’m 25 pounds heavier.
Rodrigo:02:26:52 You’ve reverse engineered the weight training program.
Chris:02:26:55 There’s no weight training.
Rodrigo:02:26:58 Would have there been any sort of accolades or sponsorship available to you for having won that if you were at the time number one, because I know that’s a big issue with Olympians.
Chris:02:27:15 The key in this world is to find an obscure sport and race other 45 year olds. Short answer, I don’t think so. Long answer, I would be happy to sponsor people because the young kids who are getting out there trying to get going, the last thing I need is sponsorship. You got to go find the kid who’s 16 and get up and running. It’s a big issue.
Rodrigo:02:27:37 Yeah, it’s big issue with the Olympians are those that come in second and then a year later they say well, we found that this guy was doped up, so you get the medal, but all the sponsorship and all the accolades, all the travel around the world, all the things, that life path is gone. So you could have been a famous du-athlete.
Chris:02:27:54 Turns out there wasn’t…Yeah, it was one of those things didn’t it, probably never want to do it again, moved on to the next challenge. Except for bragging to you later.
Adam:02:28:17 Good job. We should release poor Corey here. He’s been a gracious co-host. We’ve been holding him hostage.
Corey:02:28:22 This has been fantastic, I got to tell you. I was super excited for this one. I don’t know how I worked my way into co-host, but I’m very happy to be here. I love it. Two and a half hours, I cannot believe there is anyone still watching but I feel very excited to be here.
Rodrigo:02:28:40 But they’ll listen to it later, bye and bye, as it tends to happen. Awesome.
Adam:02:28:46 Thanks Corey, thank you so much for joining us. This was absolutely all that we expected.
Chris:02:28:54 I think I told all my stories. So it’s…
Rodrigo:02:28:59 You wish. All right, good having you guys on. Thanks, Corey. Thanks, Chris.
Adam:02:29:06 Yeah, Like and Follow. By the way…
Chris:02:29:09 Not investment advice.
Rodrigo:02:29:10 No investment advice.
Adam:02:29:13 All right. Cue the music.
Chris:02:29:14 See you guys.
*ReSolve Global refers to ReSolve Asset Management SEZC (Cayman) which is registered with the Commodity Futures Trading Commission as a commodity trading advisor and commodity pool operator. This registration is administered through the National Futures Association (“NFA”). Further, ReSolve Global is a registered person with the Cayman Islands Monetary Authority.