My guest this episode is Doug Greenig, CEO and CIO of Florin Court Capital.
Florin Court specializes in delivering an alternative markets CTA, trading over 500 markets ranging from Turkish cross currency swaps to French power markets.
We spend the majority of the conversation discussing what makes these markets unique from traditional markets traded by CTAs. For example, who are the players in these markets, what are the unique considerations for introducing and sunsetting markets, and why we would expect these markets to trend in the first place?
Doug also explains why he thinks these markets tend to behave better than traditional markets, why you don’t need special trend signals to trade them, and the significant diversification potential they can introduce.
Please enjoy my conversation with Doug Greenig.
Corey Hoffstein 00:00
321 Let’s jam.
Corey Hoffstein 00:06
Hello and welcome everyone. I’m Corey Hoffstein. And this is flirting with models the podcast that pulls back the curtain to discover the human factor behind the quantitative strategy.
Corey Hoffstein Is the co founder and chief investment officer of new found research due to industry regulations he will not discuss any of new found researches funds on this podcast all opinions expressed by podcast participants are solely their own opinion and do not reflect the opinion of new found research. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of newfound research may maintain positions and securities discussed in this podcast for more information is it think newfound.com.
Corey Hoffstein 00:50
My guest in this episode is Doug Greenig, CEO and CIO Florin Court Capital. Florin Court specializes in delivering an alternative market. CTA trading over 500 markets ranging from Turkish cross currency swaps to French power. We spend the majority of the conversation discussing what makes these markets unique from traditional markets traded by CTAs. For example, who are the players in these markets? What are the unique considerations for introducing in sunsetting markets? And why would we expect them to trend in the first place? Doug also explains why he thinks these markets tend to be a better than traditional markets, why you don’t need special trend signals to trade them, and the significant diversification potential they can introduce, please enjoy my conversation with Doug Greenig.
Corey Hoffstein 01:40
Doug, welcome to the show. excited to have you here. This has been an exciting time for trend followers, the last year this year, some major market moves, I know we’re gonna get into sort of a distinct element of trend following on this call. So I appreciate you joining me. Thank you for being here.
Doug Greenig 01:59
Hey, thank you for having me. I’ve been looking forward to our conversation.
Corey Hoffstein 02:03
Now, Doug, you are a mathematician by training. And you have done quant finance in several places over the last 30 years. Can you tell us a bit about those experiences, maybe some of the different models you’ve used in your career and lessons learned over the years?
Doug Greenig 02:19
Sure, sure. That’s a big question. So let me let me dig into it a little bit. Well, to begin with, I started out in fixed income and macro trading, Goldman Sachs and later on Greenwich capital. And there, we used a lot of term structure models, volatility models, and prepayment models. In particular, I did a lot of mortgage backed securities, things like inverse Iose, which are actually very, very complicated, fixed income derivatives with curve exposures, complicated optionality ease floors versus essentially swaptions. And indeed, also prepayment exposure and prepayment risk premium. So that’s one kind of thing. It’s a very interesting business. Another area is what you could call cross sectional, systematic macro. And there you’re sort of imitating the style of quantum equity long short, you’re using factors to establish long short positions in various sectors, for example, tenure interest rate swaps, that will be long, some you’ll be short, some in different markets in different countries, and be using factors to sort of determine your relative cross sectional position. It’s a very well established type of trading. Of course, more recently, I’ve been focused on what you might call time series macro, the trend following CTA style, which actually trades a combination of trend and vol, it’s not just about the trend is what volatility does, at various stages of the trend. That matters a tremendous amount. And I guess finally, I should say, from my earlier career, where I’ve been involved in discretionary trading earlier on, one thing you discover is that everybody, in a sense is using an implicit or even explicit model of how the world works. Like, for example, how you look at talking about the President’s, how you look at the US economy right now, depends on whether your economic framework is a post Keynesian framework, or a monetarist framework, you end up with very different answers about what you ought to do, which helps to explain why the Fed is confused. So I’ve seen models used in a whole range of ways, quantitative models and economic models. And most recently, of course, I’ve been focusing very much on trend and vol.
Corey Hoffstein 04:44
You have a really unique perspective, given the breadth of work that you’ve done. I was hoping maybe you could touch a little bit on who has been most influential on you and your development as a quant as a portfolio manager and maybe now as the manager of an asset management firm.
Doug Greenig 05:00
Oh, that’s a great question too. If I leave some people off this list, I hope they will forgive me. Because there are a lot of people I try to learn from everybody, or as many people as I can, and from all these different kinds of experiences that you pick up, but I would say Bob Linderman, at Goldman Sachs, who was just a tremendous econometrician. And he taught me a great deal about how you do modeling correctly, he was actually a pioneer in the use of Bayesian vector auto regressions. And this is where you use a very complicated, loose model and let the data speak, but then regularize the model by imposing Bayesian priors. And by the way, the kinds of techniques that you use there have been used more recently, and things like feed forward and recurrent neural nets, you have the regularization problem. So Bob was very, very influential, is a wonderful person to say I enjoyed working with him. Some of the macro guys I’ve worked with over the years, Mike Novogratz, for example, Jacob Goldfield, I learned a lot by listening to them, and how they look at the world, you need to be suspicious of data. Because there’s a lot of non stationarity in the world, don’t assume that the future is going to be like the past.
Doug Greenig 06:29
Or to the extent you want to make that assumption, really be explicit in terms of what you’re assuming, for example, in the case of trend following, one of the key elements is that when trends die, or reverse direction, there tends to be a burst of volatility. That’s a fact that I believe will be true in the future to it drives, why it works. So well has been sort of an evergreen strategy. But you really want to think about what assumptions you’re making about the future. On the management side, I really enjoyed working with Ben Carpenter, and Jay Levine, at Greenwich capital. They were very good managers. And one of the lessons I learned there is how you keep a really high performance team, and how you need to work with people if you’re going to retain talent. And the answer is surprisingly simple. I mean, you need to treat them well. But you also need to really share the economics with your team. Don’t be too greedy. If you want to put together a team that’s really going to succeed, the success must be shared. And that was something they did very, very well. And I hope I’m doing too. And of course, I enjoy working with Patrick brahmer That Brummer partners, he has a very interesting and insightful man.
Corey Hoffstein 07:45
You mentioned that your most recent ventures into a focus on time series momentum. Curious what attracted you to the CTA style and what was the ultimate catalyst for launching an alternatives only trend following fund?
Doug Greenig 08:00
Well, as we’ve been discussing, I’ve been around for a long time, and I’ve seen a lot of stuff. And I have an appreciation for how wild and crazy the tails can be. And most people who haven’t experienced it, don’t understand whether you’re talking about the 87 crash, which came just a little before my time, but I was in graduate school at the time and aware of it, where you had like what a 20 standard deviation move if we can use that expression in the s&p 500 some huge thing, or you’re talking about the great financial crisis, the European debt crisis, some of the problems in the mortgage markets over the years, details are thicker than people know. And in a certain sense, too much knowledge can almost paralyze you, if you’re selling the tails. I mean, if you’ve been around long enough, you have a real appreciation of the risks and how unpredictable the world is. Think about the pandemic. Actually, the inflation was fairly predictable. But the pandemic was certainly a shock to markets, even the public health officials had expected this to happen sooner or later. So when you take that orientation that the tails are so pronounced and that the world is very non stationary, and the relationships between assets vary so much over time. When you take that point of view, you start thinking about ways you can own the tails. And one of the few strategies that I’ve seen over the years where you really get to profit from the nonstationary profit from the unexpected. These dislocations is the CTA trend following style. There are other styles where you try to get tail protection buying volatility in the tails certain times, but nothing has proved so evergreen and resilient and consistent through many regimes as trend following. It may not be the highest Sharpe ratio strategies certainly isn’t. But it has such a nice positive skew. And a lot of times the kurtosis is working for you, not against you. And so that’s the attraction of the style. Now, what’s the bad thing about the style? The bad thing about the style is the Sharpe ratio isn’t very high for traditional standard CTAs. Most folks think that the long term Sharpe ratio might be, I don’t know, point 3, point 4, or some number like this. And again, given my comments about non stationarity, I’m not sure you can even attach a stable number to such a thing. But that’s the sort of order of magnitude that sort of makes sense to people. And if only that Sharpe ratio were higher, it would be an amazing strategy. I mean, can you imagine a Sharpe ratio one strategy, where you have a big positive skew? And when dislocations occur, you make a lot of money. That would be a very, very attractive thing, indeed. And that’s what you can get, if it’s done correctly, an alternative markets trend. So that’s the attraction, right there.
Doug Greenig 11:23
Just as a little bit of table setting, before we really dive into these alternative markets, can you maybe provide a couple of examples of what you mean by alternative markets?
Doug Greenig 11:32
Sure. French electricity, Colombian interest rates, Turkish cross currency swaps, Turkish interest rate swaps, CDs, when I use the term alternative markets, of course, I am referring to alternative to standard CTAs standard CTA trade in general, about 100, futures markets, and FX forward markets in total, about 100. Some do 125, some do 75, the European CTAs tend to trade in general, more markets than the American CTAs. But we’re talking about 100 markets, anything that’s not that gets to be called an alternative market. But again, this is a matter of semantics. That’s what I mean, by the way, I use the word but we select our alternative markets to actually be very orthogonal. For example, we do not trade some other people do but we do not trade US interest rate swaps, because that’s too correlated in our view. With the 10 year and five year US Treasuries, which is a mainstay. The futures on US Treasuries are a mainstay of developed markets. CTAs. On the other hand, Colombian interest rates, India, China, those are not mainstay markets for regular CTAs. And we trade these markets through interest rate swaps, for example. So a working definition is orthogonal to the standard or relatively orthogonal to the standard CTA markets.
Corey Hoffstein 13:11
Well, given your use of the word orthogonal, I suspect my next question here might be a bit redundant, but I want to ask it regardless, which is, how different are these alternative markets from traditional markets? Are they truly idiosyncratic return streams? Or do you find a significant degree of sort of common macro economic drivers, particularly in the tails, like I might imagine, liquidity would be a common macroeconomic driver that could affect all of these markets.
Doug Greenig 13:40
So you have both elements, all markets to a greater or lesser extent, or at least the vast majority do get affected to some degree by the major global themes. For example, if the Fed is tightening, monetary policy, and liquidity in general, and monetary liquidity is declining around the world, that will actually ripple through a lot of markets. Likewise, if China is slowing down, or if China is providing additional stimulus to grow faster, that will have ripple effects. But these markets do have some exposure to the common themes, but they also have considerable idiosyncratic return and risk elements. For example, Turkey has gone through, I don’t know, four or five mini crises in the last five years, and we’re all familiar with the inflation problem that Turkey has and the controversies within the Turkish political and financial world about how best to deal with it. The President President Erawan has the view that higher interest rates are not what Turkey needs. The markets in general have the view that the conventional higher interest rate response is exactly what Turkey needs. And it needs a lot more of it. And so Turkey has gone through some financial booms and busts. And but in general, I’m referring to how the lira has been performing what’s going on with two year interest rates in Turkey. It’s gone through some cycles there, but the general trend toward lira weakness is quite pronounced. And so that sort of stuff really is idiosyncratic. And we’ve done very well in Turkey. And there are many other examples in the power markets, electricity, natural gas emissions, those really don’t have a lot to do with, for example, the Fed, or what’s happening in the stock market. The idiosyncratic component is very large. We have a lot of exposure in emerging markets, we cast our net quite widely. So if you think there are going to be some interesting developments, divergences, dislocations, trends, any of the stuff in emerging markets, we will capture that better, we think, than regular CTAs, you did bring up the subject of liquidity. And it’s interesting, many of the markets that we trade are actually not that illiquid. It’s just that they’re operationally complex. It’s a pain in the butt to trade some of these markets. And, you know, whether you’re trading some individual freight routes, or you’re trying to trade on short Chinese commodities, with offshore money, there are significant hurdles for a standard seat, you know, that people face if they want to trade these markets, can these hurdles be overcome? You bet that’s what we do. But having that specialist experience and knowledge, and resolving these operational issues, modeling issues, data issues, execution issues, that’s really key. So it’s not about trading, just trading illiquid markets. If you want to trade a liquid markets, you can do what some other people have done, but we reject, which is just pick out some really crappy futures in Chicago, or in the other developed market futures exchanges, and say, I want to trade this, I don’t know sour milk, or something, some future that sort of trades by appointment, and do that instead, what we’re doing is doing the work to figure out how we can access China’s very liquid. On short commodity markets, we were among the first to do this to crack that nut, or how we can participate in the market volatility being generated by the policy lurches in Turkey. So that would be my comment that our portfolio is not extremely illiquid at all. It’s just a harder or operationally difficult portfolio of assets.
Corey Hoffstein 17:58
So far, all the examples you’ve provided of alternative markets have been actually truly unique futures markets. But one of the other examples I often hear when talking to CTAs, who are including alternatives and their managed futures programs is the idea of synthetic markets. So for example, they might not trade US bonds, but they might construct a 2-5-10 Butterfly trade and follow the trend in that. Can you talk a little bit about whether you incorporate some of these synthetics? And if not, why not?
Doug Greenig 18:29
We do to a very limited degree. But here’s the challenge. A twos fives 10 Butterfly on a given yield curve should not in general be expected to trend in the way a directional market is directionally being factor one, or even curve slope. That would be factor two. So whatever it is, you’re trying to trade trend on, you should sort of make sure that it makes sense to trade trend on it. I mean, there’s a whole industry of people who like to trade mean reversion in the curvature of the yield curve. Indeed, that used to be a basic stock trade on the Salomon arbitrage desk. A lot of people in the business would trade the curvature between the 10 year and the 30 year sector were sort of 20 year bonds, which used to be 30 year bonds, but are very off the run would get very cheap, the boat trade the stuff around. So you got to be careful about trying to trade trend on things that are not really directional effects. On the other hand, you can sometimes find combinations of assets that trend better than unitary single term assets. And there can be very good reasons why. And by the way, at times the yield curve slope will trend better than the level. Okay, so we have a relatively small percentage of our risk, you know, a few percent spread across some of these synthetic assets that are actually chosen and weighed, so that they have very good trending properties. But our real focus, again, is on operationally difficult markets and trading standard, competent, properly implemented trend on those markets. We think the low hanging fruit is having more markets and better markets, and less correlated markets, as opposed to trying to get super cute with models or creating synthetic assets. I mean, the basic orientation is there’s very low hanging fruit, if you’re willing to do the grungy operational work of onboarding these markets, and building the portfolio from them.
Corey Hoffstein 20:58
Given the operational difficulty in accessing some of these markets, I suspect that it might actually change the nature of the players in those markets. So here I’m thinking about sort of the percentage of speculators versus producers versus consumers of those goods. Can you talk a little bit about how that composition changes in these alternative markets versus traditional markets? And how that may create opportunities or challenges?
Doug Greenig 21:23
Yeah, that’s another good question. So you can see right away studying the data or just thinking about the matter, that alternative markets offer superior diversification. We have 500 markets of things like steel rebar in China, South African maize, Malaysian palm oil, French electricity, California carbon emissions, you go through the list, you know, Turkish interest rates, and so forth. And you can see that there, a lot of them have very little correlation to one another, a lot of them, so you get that diversification. But the other thing that you get with these markets, is you tend to get a better ratio of directional movement to meaningless chop. And you can measure that with various statistics, you can look at various autocorrelation numbers, or you can simply look at the results of applying standard trend models to this. Because what a trend models tell you, they tell you the amount of directional movement relative to the amount of choppiness right? That’s when trend models do well, and they also tell you a lot about what happens around trend reversals? Do you get a burst of volatility? That’s good? Or do you not? It’s not so good. So the alternative markets have a pretty good ratio of directional movement to chop, as a general statement, and it has a lot to do with the market participants, as you were raising this question. When the market participants are not just a bunch of speculators, there tends to be better market behavior. That’s my experience. That’s what seems to be the case. And I would say that, over the years, a stylized fact is that these alternative markets that we trade have a lot less pure speculative, financial participation, relative to the size of the markets. You think about historically, electricity markets, where the big producers, and the big consumers are very large, relative to a small number of CTAs involved with these markets. Now, this situation can change indeed, we kind of expect to change, more people have gotten involved with some of the markets we’re involved with. And it raises questions, will the behavior deteriorate? Will you start to see more of the big chaotic reversals that you see frequently and may I call it over speculated developed markets? So it raises that question, now, our solution to this is to keep adding more markets and pushing the frontier further out. So we’ve been adding 50 to 75 new markets a year. And I think we’ve added over 50 so far this year, and we continue to have our eyes on new markets, and that automatically dilutes the old ones. And I’m certainly scrutinizing some of the markets where we think there’s been an increase in speculative participation by CTAs and others for whether or not we might want to drop them. So I’m keeping an eye on things but the number one thing is to keep pushing into greener pastures, and improving the diversification is so important to the program.
Corey Hoffstein 24:56
Well, you led me really nicely into the next question I wanted to ask what are some of the considerations, you know, maybe the boxes that need to be checked for when you’re going to introduce a new alternative market. And then conversely, when you’re going to sunset, a market from the program?
Doug Greenig 25:14
Well, in terms of introducing them, you need to have some data that you need to have figured out operationally how you’re going to trade this thing. You need to know enough that it’s not a pegged or ultra regulated market, where you might not expect trend to work, we don’t apply trend, for example, to pegged regimes, because then when the peg breaks, you get this chaotic move. So that doesn’t work for us. So you need those basics, you don’t necessarily need a vast amount of historical data, because we apply relatively similar models, across markets have similar liquidity. But you do need a very deep understanding of the liquidity of the market, how it varies over time, you know, the depths of the market. And you also want to begin by trading small, so that you confirm that your assumptions about transactions costs are borne out when you actually begin executing in those markets. So you’re gonna be trading them small for a while, and gradually ramping up and making sure that your the models that you’re using for your transactions costs are right. You also want to be aware of idiosyncrasies that may influence the modeling credit, for example, for I mean, you could use a Merton model to understand this point. Credit widens quickly and tightened slowly as a rule. And therefore, if you’re trading credit spreads, we do that for a reasonable number of indices. Your model needs to take that dynamic into account. So you want to understand enough about the market. You also want to understand if there are counterparty risks that matter, what are the risks of some kind of huge tail event in one direction or another, although usually, the tail is our friend, usually. So I think that would be a pretty comprehensive list, hope haven’t left anything out on the stuff that we look at an onboarding market. We’re quite efficient about this, because this is all we do. I’m not trying to run, despite my background, having done a lot of different things, this business is very focused. We’re an alternative markets trend follower. And our job is onboarding alternative markets that diversify the portfolio are different from what we have. That is an obviously a very important consideration, the diversification effect of the markets and to trade them effectively manage the portfolio in a very precise way. That’s all we do. So you know, we’re really on the lookout all the time for new markets to add things like Japanese power and Indian commodities are high on our to do list for coming years. How do you think about quantifying or maybe exploring from a qualitative perspective, the diversification potential of adding a new market, when there’s limited access and data availability? Well, sometimes you just use logic, you know, for example, electricity markets, or local markets, really, because an electron in the northeast of the US can’t find its way to the power grid in the UK. Okay, so this is the reason that electricity prices are very, very different in the US versus the UK, the UK is more closely linked to Europe, for example. So you just understand logically that certain markets are really disconnected, you understand that emerging market interest rates, and the stories around different countries are so much more diverse than the story in Western Europe, where the economic linkages, so some of it is just thinking and using common sense. In other cases where we have data, you can actually just model it out. Now you need to remember that our portfolio was so big at this point, we cover all the main sectors that you can cover. And we have over 500 macro markets that were trading around the world. Any one market that we add, is a small incremental change. But that was also true when we only had 300. And the last 200 have been 200 incremental positive changes, for example, adding inflation swaps during the pandemic that really paid off when inflation exploded. By the way, that is a market that you don’t normally see a lot of good historical results trend following. However, it was perfectly clear with the amount of monetary stimulus that was occurring, that introducing those markets and allowing the models to make up their own mind about future trends would be a smart thing to do. In the worst case, trends don’t emerge. And the positions are small, the models tell you have small positions. But you you wanted to have that option, that if things were moving, the models could pick up on it. That’s how I look at it. And again, when you have 500, markets going to 501 502, the marginal effect is small. But hopefully in three years or something, we’ll be having a conversation, I’ll be talking to you about the 750 markets we have. And there’s a difference probably between 750 and 500, you know, the 750, we might have in three years versus the 500 and change we have right now, given the breadth of markets that you trade.
Corey Hoffstein 31:01
This might be an impossibly large question. But I’m curious from a theoretical perspective, whether you think the reason these markets trend is fundamentally different than the reason we might see trends in more traditional markets. So for example, one of the common arguments is that in traditional markets, we see behavioral hurting effects, or we might see an anchoring bias or a number of things that lead to the creation of a trend. Is it possible in some of these smaller markets, that it’s actually the operational burden creates limits of arbitrage that slows down price discovery? Could it be a totally different theoretical argument as to why prices trend in these markets?
Doug Greenig 31:41
If these markets were more illiquid than they are, that would be an interesting hypothesis. But these markets in some places are as liquid or more liquid. But none of the markets that we’re trading really count as illiquid markets, where there’s some sort of slow moving price discovery. The reason these markets trend is exactly the same reason that the developed markets trend. I think the question we’ve got to ask ourselves is why did the developed markets trend less well, than they used to, from the standpoint of trend following now, the premise of my question, is the trend following performance in general has declined over the last? Oh, I don’t know. 13 years or something. In the developed markets? I think it has, even though trend followers had a great year last year. And I think the reason that developed market trend following is less attractive than it used to be, is, for one thing, correlations among the assets they trade got higher, particularly during quantitative easing. At various points in time, the main theme for markets was what are the central bank’s doing in terms of loosening monetary policy? Or what are the central bank’s doing in terms of tightening monetary policy, there wasn’t enough idiosyncratic disconnected stuff to create the diversification that you need. Because remember, roughly, your Sharpe ratio goes up with the square root of the number of independent bets you have as a kind of rule of thumb. So I think developed market trend following got to correlated, perhaps the number of independent bets got quite low, as we measure it, between 2010 in 1415, so that I think reduced its performance. Another big element is the central bank, given the debt burdens that we see in the developed economies, there are very, very high levels of sovereign and corporate debt. And central bankers are very keen not to have a deflationary collapse. And so policymakers try extremely hard to prevent severe depressions, recessions, and financial collapse. We saw that in the great financial crisis, we saw even more of that the first two years of the pandemic. But what this means is they are trying to stop out trends using monetary and fiscal policy. So I think that her trend followers in the 2009 10 period, and things have become more interesting again, now that central bankers have less freedom because of the inflation problem. So alternative market trend works better, because you have greater diversification. You’re getting more of those idiosyncratic trades, and you’re not really the target of central bank market control, if I can call it that, or attempts at market control, given the operational complexity required to access to a number of these markets, which might preclude less sophisticated players, is it possible that there’s actually more sharks swimming in the same pond, and that it’s actually harder to extract alpha in some of these markets, because it’s only sophisticated players that’s operating within them. It’s an interesting thought it hasn’t proved to be the case. I mean, we made around 30%, in 2021. This is a 10, Vol. 20%, about 20%. Last year, we’re not seeing it. We’re seeing very good performance of our models in these markets. I think that perhaps the reason is that the other participants in the market are not primarily speculators working on the same time scale, as we are the example of the power markets historically, being a good one. So you can have two sophisticated people trading in a market with different motives and different time horizons. And both of them sort of making money. Of course, you need to have other people in the market as well. But the point is, I haven’t seen that because the performance of alternative markets trend has actually held up very well. I mean, you know, an interesting question is, are the markets behaving better or worse than they did three or four years ago. And to the extent you can measure it, it looks like they’ve been performing better for the past two years, at least as well. And you haven’t seen developed markets holding up developed markets trend signals holding up as well, over the past decade. Although we’ve had a couple of really good years in that period.
Corey Hoffstein 36:45
We’ve been talking about trend following fairly generically, but there are, of course, so many different ways that trend following program can be implemented, which is why you see such large dispersion, among trend following managers, even if they’re all trading the same markets. Is there anything you have to change about your trend following programs to trade these markets versus the way traditional markets might be traded?
Doug Greenig 37:09
We have to be mindful of the liquidity. But as I’ve said, it’s not as if we’re trading terribly illiquid assets or anything like that. The answer is basically No, there’s nothing you need to change. But there’s some things that you need to think about. You see, develop market trend followers struggle with the issue of their relatively low Sharpe ratio. And so there are things that they do to try to mitigate that. One thing that they sometimes do is they trade very, very slowly, which allows them to try to collect more carry or something. Another thing that they do is they introduce a lot of many of them introduce large carry signals into their models in fixed income, in currencies, sometimes in commodities, as well, or introduce other ways of either selling volatility, or collecting risk premia in a way trying to improve their performance in between those occasional dislocations that occur every five years in the developed markets. We on the other hand, are starting out with a more diverse, better set of markets, in my opinion. And because you’re starting out with that better set of markets and with the Sharpe ratio, that’s one or higher. Because you’re doing that you don’t need to compromise, you can keep in other words, you don’t introduce a bunch of crappy all to risk premia signals, you don’t do much in carry, you keep things very pure, with an overwhelming weight on trend, and try to keep that convexity. Now you can do the same thing and develop markets. But those long periods of pretty stagnant performance, discourage investors and discourage the manager. The great thing about alternative markets is with a diversification level, four times higher in terms of the number of independent bets, for example, that’s about the right number, in my view, with that you’re starting in a much better place. And so you don’t need to be giving away any of that tail performance and positive skew that make CTA is such a compelling strategy distribution.
Corey Hoffstein 39:30
I would imagine that one of the challenges in a program like this is actually figuring out how to construct a portfolio from these 500, potentially 750 markets in the future. You know, for example, thinking about things like balancing statistical risk, operational risk, regulatory risk, and maybe capacity constraints in certain markets. Curious how you think about that actual portfolio construction problem?
Doug Greenig 39:56
It’s a tricky problem. It’s a tricky problem and we all know that the correlations among markets can be unstable. And so a highly optimized process playing off one market against the next, because of a transiently high magnitude correlation is not a good way to proceed. Instead, we go for robustness. So what we do is we use a tree structure for constructing a portfolio. And the starting point, intellectually, is the idea of having a similar amount of risk, and all of the branches and all of the twigs. However, you need to take into account the fact that markets like butane and propane, are going to be very correlated. So we take that into account to kind of group things in a tree structure. In addition, we use a method of dynamic gearing, where the changing correlations are detected as changing levels of delivered volatility from different twigs, branches and limbs of the tree. And so the system automatically adjusts so that the volatility that we’re delivering moves toward the desired level, whether that’s from changing volatility in the single temps, or whether it’s from changing more often from changing correlations.
Corey Hoffstein 41:29
How do you measure capacity or trading costs in some of these markets that are more difficult to access? In particular, I’m thinking about markets that might trade bilateral OTC.
Doug Greenig 41:41
You begin with talking to the dealers and discussing their market share how big they are, how big the overall market is, typical bid offer spreads. And by the way, whenever we, we go into a market, we actually always request a two way market. So we’re continuously gathering data as we execute trades, on the width of bid offer spreads, so forth. And by the way, our trading isn’t always as predictable as you would think. So we get a lot of data on this, I would say that we have regular meetings, we have a dedicated trading team of three guys, who are all experienced execution traders. And so those trades that we don’t execute electronically, we have, you know, a lot of contact with the dealers and are sort of leaning what conditions are. And even on the trades that we execute electronically, we’re measuring slippage, we have meetings at regular intervals, measuring the slippage of each kind of asset with each kind of Counterparty and each individual named Counterparty. And so we’re gathering all of this data and keeping very, very careful track. But before we enter a market, we engage in these discussions with the counterparties. And then when we begin in the market, we’re trading in small size confirming our impressions about what market conditions are like. And occasionally we’ll change our mind and drop a market. We dropped Australian power at one point early on, simply because the liquidity conditions were too variable. In our view, it just wasn’t worth the trouble. It helps having a lot of experience with stuff because, as I said, we started with about 200 markets. We now about 500 markets, onboarding markets, assessing liquidity conditions, talking with dealers and brokers, estimating the size of an over the counter market. That’s sort of what we do for a living. And that’s very key to our success with this program. Obviously, the demand for trend following strategies has picked up fairly dramatically in recent years.
Corey Hoffstein 43:54
How do you think allocators should think about introducing an alternatives only trend following program to their portfolio at large and then maybe in particular, in combination with a more traditional program?
Doug Greenig 44:10
Well, allocators need in general traders in general people managing capital in general, all of us need to think a lot about dislocations, the tails of the distribution. I’ve been around for a number of decades now. And I can tell you, I’ve never seen a more unsettled and potentially unstable macro environment than what we have right now. It’s a very interesting one. But dramatic things are likely to happen in the coming years as they have over the last several years. I don’t think we’re just settling down into some nice equilibrium. And in that context, there’s absolutely a place for long convexity strategies. Now developed market CTAs have their place.But we delivered the same positive skew in some cases even more, because we are not. We don’t need to delude what we do down with non trend elements. So, an alternative market, CTA can be your primary trend allocation, or you can put it together with a developed market trend. Now, what’s the right ratio? I’m not sure opinions differ, I would say that the alternative market side offers so much better performance over stretches of time about, you know, in Sharpe ratio terms, twice as much or more, that it should be a significant part of the allocation. If you agree with my thesis, that owning details through trend following programs, and long convexity strategies makes sense.
Corey Hoffstein 45:54
You mentioned earlier that the competition in this space has started to grow. When I talked to a lot of the larger trend followers, they all mentioned that they’ve been trying to expand their alternative markets programs over the last couple of years. From your perspective, what characteristics should an allocator be looking for when they do due diligence on an Alternatives program?
Doug Greenig 46:18
Well, the first thing is it needs to be the right size, it’s possible to be too large, because you lose your nimbleness and you have to begin to trade more slowly. If you get too large for the space. Remember, I was saying that our allocation philosophy begins with the idea that would we’d like to allocate an equal amount of risk for the same amount of signal strength in each of the markets. Now, we’re not quite able to do that. But that’s the idea, you’re trying to be very much in that direction. And to a large degree, we do accomplish it. If you go from a size of 2 billion or 3 billion to triple that, all the work that I have done indicates that you’re going to lose a fair bit of that diversification, you’re gonna have to stuff more stuff into the most liquid of those markets losing some diversification, you’re going to need to train more slowly. And you’re going to have a variety of issues. For example, you’ll find it hard to hit your volatility target, according to modeling work that we’ve done, it just becomes ungainly. And I think our results show that we have in comparison with peers that we have a very good size. So our capacity is in the context of about 3 billion, we are not planning to be much larger than that. That’s where you can manage a fund very effectively. So size is important. Next point, you want it to be reasonably orthogonal from the developed markets. I don’t think it’s a great idea to be trading US interest rate swaps. That’s just a bunch of overlap, in effect, with developed market CTAs. You want less overlap with developed market CTAs. I don’t think you want a big equities program. CTAs are not that great at trading equities. They’re not. And I’m referring to cash equities. I’m even referring to equity index futures. That in general is a difficult area for CTAs. And so beware, it’s a place where people can get capacity, they can grow. But if I want to put money into a quant equities framework, particularly in the developed markets, there are quant equity shops that are simply a lot better at trading cash equities in the quant framework than any CTAs are. That’s my opinion. So what you’re really looking for is a large number of markets, you’re looking for focus, you’re looking for experience, you’re looking to add markets for a manager who adds markets and continues to press into the frontier. You want the pure trend style, and you don’t want the manager to be too large, and therefore less nimble, or compromised in terms of the markets they’re training. That’s my philosophy of it. And I really tried to design Florin Court to be the best possible alternative markets trend follower. I mean, I was able to get out a clean sheet of paper in the beginning and really think about what we wanted to do and then implemented.
Corey Hoffstein 49:43
I noticed that missing from your examples of alternative markets was any mention of crypto, which is an area I’ve spoken to other trend following shops as having included in recent years. Is this an area that you’ve explored introducing into your program?
Doug Greenig 50:00
That’s a great question. We were involved with crypto before crypto futures were available. But we had to use Exchange Traded crypto derivatives. Okay, was the GBTC, for example, and there were others that we used to get involved with the market, I think it was in early 2017, we first became involved. And then we’ve looked to expand our activity in the market. But we are very, very careful about things like counterparty risks. And so we could never get ourselves comfortable with crypto exchanges. As counterparties we want to see audited financials, we want to see all the things that a pension fund, because you know, we have some real blue chip investors, very much oriented toward pension funds and endowments and sovereign wealth funds, they expect and we expect that all the counterparties that we deal with are absolutely top class, and that we have conducted a very thorough process of vetting them. And we just couldn’t get there with the crypto exchanges, we could see when the futures became available, we took more risk in those. And the crypto has been a very productive sector for us. And I wish the space were going in a better direction. Now the US seems to be pretty determined to stamp out crypto. But what’s held us back really is the counterparty problem.
Corey Hoffstein 51:34
You’ve spent your career straddling sort of both quant and discretionary in your different roles. Given that perspective. Maybe you could share some thoughts around what do quants typically get wrong? And what do you think non quants get wrong?
Doug Greenig 51:50
Quants often do not appreciate how non stationary the world is. They worship data a little too much. Now, when you’re in a very data rich environment, perhaps doing high frequency trading in equities, that problem is mitigated. But when you’re talking about macro, the data are sparse enough or too sparse. And the issue of non stationarity becomes very, very important and hard to deal with. People do too much unconditional estimation. Let me see if I can come up with an example. An outside reversal is an interesting technical signal. But outside reversals primarily matter in the late stages of certain kinds of trends. And then outside reversals means a lot. But much of the rest of the time, they don’t mean anything. And so if you were to test that signal, you would want to test that signal conditional on certain facts being in place, conditional on a regime that you’re in. And so there’s just way too much unconditional estimation, on one hand, and other quants actually overfit to thanks. So it’s a very, very delicate balance between recognizing that things are non stationary, being too rigid and unconditional, in the things you look at, and then being too loose. So you need some judgment about what’s going on. You also want to understand the meaning of the signals that you’re using and why they ought to work, and why you think they might work in the future. I’ve discussed the fact that one of the reasons that trend works is that when narratives change, there tends to be a burst of volatility. It’s an idea that I picked up from Minsky, you can read about the Minsky framework and Kindleberger spoke, for example. And it happens on many scales and markets, that burst of volatility around narrative shifts and trend reversals is pretty key to the way trend following works. Indeed, if you don’t do that, right, you lose them after sharp. So a lot of quants just don’t think enough about what things mean. Non quants. On the other hand, well, they have their own problems. They have trouble sizing traits, they have trouble oftentimes spotting the factor exposures that they have, because emotions enter into what they do. They end up for example, changing the reason that they have a discretionary trade on as a manager when I was managing some discretionary traders in the past, when a person starts shifting the reason for hanging on to a trade. That’s a good sign that they shouldn’t be. What was your original rationale? And what’s that borne out? That’s the question. If you start shifting the reason that’s a big issue, and you got to remember non quants need to remember this trades are not your friends with quants need to remember that models may not be your friends either. It just kind of depends on whether they’re appropriate for the conditions at hand.
Corey Hoffstein 55:08
Well, Doug, last question for you here. And I want to sort of play into your background in macroeconomics. CTA struggled in the 2000 10s. And you started to touch on this a little bit in a prior question, but I wanted to get your thoughts on the current macro environment. And whether you think the coming years will ultimately be good for trend in long convexity strategies.
Doug Greenig 55:32
Well, of course, what did Yogi Berra say, making predictions is difficult, especially about the future, I want to be a little bit humble about our capacity to know what the future holds. That being said, there’s been a huge buildup of leverage in western economies. And leverage is ultimately destabilizing, it makes things more fragile. We have a situation where the central banks over the past couple of decades have been able to paper over or relief, various macro problems like the great financial crisis, the European sovereign debt crisis, the threat of the pandemic, with large amounts of stimulus, and governments in general, have become accustomed, and the citizenry in various countries, they’re accustomed to large deficits and deficit spending. So we’re at a kind of delicate point. And now that inflation burst into view, in my opinion, in my opinion, because of the huge amount of monetary and fiscal stimulus, now that it has arrived, central banks are considerably more constrained. And so I think that we’re at risk of some very significant dislocations going forward, as a consequence of this setup, then, you have a few other things adding to this very spicy mix. The post World War Two environment that we’ve seen was actually a very, very good period for the world as a whole. It was characterized by relative peace, there was American economic, and then ultimately, military dominance after the Cold War, I think that there was a lot of growth in many emerging market countries, it was a very, very good environment. In addition, for most of the period, fossil fuels were cheap and plentiful. But now we see things changing. China, of course, is rising as a significant economic, political, and ultimately strategic rival to the US, and how the US reacts, because right now, the US is reacting with a great deal of fear and hostility. I think we should not discount the risk of the US and China coming into increasingly intense forms of conflict, already. We’re seeing sort of the global supply chain picture, the semiconductor restrictions being applied to China. All of these things are becoming much more salient for macro economics. And then you have the energy transition. I know people have different opinions. I personally believe that the carbon emissions that we have done, the greenhouse gases are a significant component of climate change. I’m not sure the basic thrust of what the scientists and people like Greta Thunberg are saying, a lot of it is true. A lot of it is true. In my opinion, read vos loves Neal’s book. But here’s the problem. Modernity largely rests on carbonization. The amount of usable energy per person has gone up by a factor of four or five over the last few centuries. It’s all about fossil fuels, it’s everywhere. Whether you’re talking about plastics, nitrogen, fertilizer, cement, and concrete. Okay, steel, so much of what we take for granted around us is a build up from fossil fuels. So we’re going to try to decarbonize in a couple of decades and we have to, but it’s going to be like the most difficult thing ever. And if you think our governments in the western world are up to this challenge, I guess I would be inclined to disagree with you. But I would be very welcome to you’re trying to convince me that the sort of stuff we’ve seen from the UK and Germany in the US will lead the way towards decarbonisation. So I think it’s going to be a very difficult environment from the supply side because of these challenges. So putting it all together, you have the leverage cycle, you have populism, you have decarbonisation, and you have a move toward a multipolar world In which the US is not calling the shots, and where things become considerably more complicated. And so the potential for exploring the tails potential for dislocations and shocks seems to me to be awfully high. So I sort of like what I’m doing for the environment we’re in.
Corey Hoffstein 1:00:21
Well, Doug, thank you so much for joining me. This has been a wonderful exploration of these alternative markets that we don’t often get to talk about.
Doug Greenig 1:00:30
Well, thank you so much, Corey, and I appreciate the excellent questions.