275 Systematic Investor Series, Group Conversation Part 1 – December 23rd, 2023

[00:00:00] Intro

You’re about to join Niels Kaastrup-Larsen on a raw and honest journey into the world of systematic investing and learn about the most dependable and consistent yet often overlooked investment strategy. Welcome to the Systematic Investor series.

[00:00:23] Niels

Welcome and welcome back to this edition of the Systematic Investor series with Katy Kaminski, Cem Karsan, Rob Carver, Mark Rzepczynski, Richard Brennan, Alan Dunne, Nick Baltas, Andrew Beer and I, Niels Kaastrup-Larsen. As you can tell from this introduction, today and next week will be very special episodes because it’s the first time that all nine of us are together for one big conversation and debate.

Firstly, let me just say a big thanks to all of you for making the time for this extended recording, which I really have been looking forward to. We’re recording on December 14th, and this conversation will be split into two parts and published on December 23rd and December 30th. We’ve got a great lineup of topics for our listeners over the next two episodes.

Let me just mention a few of the themes that we’re going to be covering today and next week. We’ll be talking about 2023’s performance and the outlook for trend and volatility into 2024. We’re going to be talking about the shifting market microstructure and the impact of zero-day options as well as the Fed pause, as well as all the things. We’re going to be discussing classification. Are we using the right names for trend following volatility? Should it be a classic trend or quant systematic? We’ll debate this.

We’re going to be discussing backtesting and rebalancing, and how to deal with this. We’re also going to be diving into non-trend strategies. Are they worth your time and your money? How much diversification is really needed? We’re going to be debating research, whether it can stand still and how AI fits in. And of course, I’m sure we’re going to be talking about whether to replicate or not to replicate. So, as you can tell, we really have a packed agenda. Let’s just dive into it right now.

Now, the format for our year-end conversation will be that we will pick what we think are the best and most relevant topics from a long list of ideas that we shared with each other beforehand. Generally, as we always do on the podcast, we’re going to keep it pretty fluid and see how we get on.

On a small programming note, let me just mention that we are having some technical issues with Rich’s connection, so he may not last the full conversation, but we will try to get as much from Rich as we can.

Since it’s polite to let the ladies go first, why don’t you, Katy, share your first topic? Feel free to direct it to anyone you would like to comment on it. Of course, afterwards, if anyone else has something to say, a strong view, feel free to jump in.

[00:03:07] Katy

Well, I thought, starting off, that we should start by talking about what happened in 2023 and some of the things that have been a theme this year. I, myself, have been writing my year-end review of what went on this year and trying to reflect upon what has been, what I call a period of turbulence and some consolidation for trend following strategies. I was looking at these themes and there were a couple of things that we noted in our yearly note, and one of them was the impact of extreme events.

The second, is the persistent increase in volatility in the fixed income market and the focus on the fixed income market. The third, was also a focus on drawdown recovery for strategies like trend and the impact of these extreme events and consolidation. So, for me, it’s really been about thinking about what themes drove returns and investor behavior this year.

I think all of us know a very interesting point is that if you take just two days in the year 2023, and that’s those two really extreme days in March, you had two days that actually shifted the index for SG Trend from positive to negative. Those were some of the most extreme days for managed futures that we’ve seen in my career and for those of us that have been in the field longer as well.

A few years ago, we wrote a paper called “Measuring Turbulence”. This is an interesting paper. It uses the Mahalanobis distance to measure how big a move is relative to time periods. What is interesting is that what happened to us in March were some of the biggest events for managed futures that we’ve seen historically. For me, it was really the bottom of the short end of the curve.

I think what we’ve been looking for the rest of the year, was when are we going to see the bottom of the long end or when are we going to see that pivot in the hiking cycle? We really saw two of those bottom-type events. I always remind investors that as trend followers, we don’t pick tops and bottoms, we follow the main part of the trend. So, a year like this year was really about hitting that change, finding the next trend and where markets are moving next.

So, I think today, a lot of people are going to talk about these themes and what the next theme is. But for me, it was really about looking at what this year really meant to us, and it was really the consolidation of the fixed income trend, the extreme events and the impact of those. But also, just another key point is that trends actually did work. Yields did go up on average this year.

One other thing that we found in our annual review; I like to study CTA style factors. Last year, and we’ll talk about this later probably too, we studied about bond bias and how everybody is so used to being long bonds. An interesting point, in my yearly review, was noting that our long bond bias factor is actually negative this year. So even if you had taken some of that long bond bias and put it back in, because you thought I just want to go back to being long-biased, it actually had a negative impact on trend.

The reason for this is that there were a lot of periods this year where the short bond trend worked very well, particularly after March going into around September, October. So, I think for me, this is a year of consolidation, a year of moving towards the next trend, following fixed income, seeing where the next phase of that trade goes.

I’m excited to hear what my fellow panelists have to think about that because those are the three things that came up for me this year. Extreme events, end of the bond trend, next phase of the hiking cycle, and a year of consolidation after one of the best years in trend that we’ve seen in many, many decades. So, with that, I’d love to hear what some of my fellow panelists thought for their key themes for 2023 if they were thinking the same.

[00:07:22] Niels

Exactly. Katy, I’m going to just jump in here and I’m going to try and direct it to at least two that I think would be interesting to comment on that from their perspective. Actually, I thought of Andrew and Cem, because obviously, Andrew, from the industry as a whole, and Cem from a, kind of, extreme vol situation may be interesting.

Anyone else who has a strong view should also chime in. I don’t know Cem or Andrew if anyone would start off, but I thought it was some interesting observations, Katy.

[00:07:55] Cem

Yeah, I’ll dive in. I think the inflation question and the bond story are the biggest big-picture question. I think if you get that right, much like maybe you got it right for 30 years and the monetary policy right; or 30, 40 years, the monetary policy driven a decrease in interest rates. I think that one will be critical to broader multi-annual performance.

My view, broadly, is that what we’re getting here in terms of inflation. Even though we’re seeing a slowdown in demand and cyclical slowdowns, and the headline number is coming down, there’s a lot of clues underneath the hood that structural inflation is still very strong. Services inflation, particularly the labor component x real estate is very, very strong and accelerating right now, amidst what is otherwise a cyclical decline. There’s a lot of clues that say this is not the end for inflation.

I think it’s very interesting that we’re recording this the day after the Fed meeting where Powell came out and basically put a mission accomplished banner on the tanker, when I think history has not judged other chairmen that have pivoted as quickly very well. The causes of the inflation, much like in the 1970s and other periods before, is a function of populism and rebalancing. I think that’s a critical thing for people to understand.

It’s not just the fiscal that comes with that. It’s the protectionism, it’s the deglobalization that leads to competition, war globally, whether that’s economic or actual hot wars, commodity scarcity and other entities competing and guarding their resources. We saw this again in the ’70s and other periods as well. That stuff has not ended. If anything, underneath the hood, we’re seeing more of that this year.

You’re just not seeing that in price. It’s being combated by the cyclical effects, which are generally short-term, of monetary policy and taking real REITs positive. So, I think that’s something to keep an eye on going forward. I think the early pivot there, in my view, will not be judged well by history, but that’s my opinion. I’d love to hear other people’s thoughts. But I’ll dive into that one.

I think this next year will be, particularly going into an election here in the US with more fiscal spending, a very interesting one to watch.

[00:10:39] Andrew

That’s much deeper, much more insightful than what I’ll say. I’ve been calling this the year of the whipsaw, basically. I think that to me, the observation has just been total and utter confidence, month to month, that the market is one month pointing that we’re going to have a taper in January, and then by February it’s higher for longer, and then by March half the month was higher for longer and then it’s, “Oh, my God, we’re going into a global banking crisis, we’re going to have a recession by June.”

I think what’s underlying it is, and we’ll talk about this more later, but basically, the alpha generation (however you define it in the space), is when it’s contrarian. Every institutional investor out there, every wealth management portfolio is designed to move slowly. And sometimes, there are these huge tankers pointing in a direction, and every now and then the world changes faster and it takes them a long time to pivot and change.

I think, back to last fall, I started talking to people about the inflation trade in early 2021. If I told half the people that I spoke to that I know with 100% certainty, God tapped me on the shoulder and he told me that inflation is coming back, they still wouldn’t have changed their portfolios because they can’t. They have 10-year asset allocation assumptions. They change things very much at the margin.

So, for about 18 months, the whole inflation trade and being short rates was a contrarian trade. We’ve also called this the one trade market because that trade plays out through every other market. By last fall, it was crowded. And so, what you’ve seen, since last fall, are these incredibly sharp moves where every time there’s a hint of a taper, people are rushing to reload on duration and then it looks higher for longer and they dump it.

So, I think that’s just underlying it, the behavior culture that’s driving the volatility. Which is hard particularly when you’re a long-term trend follower and you feel like you bounce back and forth from propeller to propeller, which certainly felt like the case this year.

[00:12:34] Katy

I love this because our prognostication for 2024 fits with both of you. We expect higher volatility of inflation and growth next year, which is kind of in the same theme, that we’re going to have a lot more undulations similar to the fact that we’ve moved to a new phase of that interest rate trade. So, I think we’re all on board. The cycles of those and how to trade those is going to be the challenge, but it definitely seems like you both have a similar view coming from different perspectives.

[00:13:05] Niels

Great. Super.

Cem, why don’t you bring up another topic that you’ve thought of this year?

[00:13:13] Cem

Yeah, I guess I’d be remiss if I didn’t lead in with a conversation about volatility. I think one of the biggest things that’s happened, broadly, across markets is the acceleration of volumes in the options market, particularly zero DTE, and also the effects that I’ve highlighted that are becoming much more prominent and important. I think it will play very importantly into the coming year of structured product issuance.

I think that it, in my world, is the most important thing and I think it is critical to not only trend but all investing right now. Very few people really have their thumb on this and it’s so critical. I’ll start with what I think is the most important structured element, which is the structured product issuance.

Because of higher interest rates throughout the last several years, and poor market performance, probably, over the last two years (even though this year has been quite good), and concerns by investors. There has been a dramatic move towards investments that now yield 8%, 8.5% that are broadly selling vol and that are stacking maybe a 2.5%, 3%, 3.5% yield on top of that risk-free 5.5%.

These structured products take many different forms, I’m not going to go into all of them, but essentially are issued by banks and are very desirable. These are structured products that maybe yielded only 4.5% when interest rates were 4% lower. Now you’re getting a significantly higher yield on those. Importantly, there’s less appeal to investing in the equity market, broadly, for all the reasons we’ve covered.

You look at the last two years of performance, obviously, it hasn’t done very well as opposed to the 15 years before where we saw, on average, 12% annual returns. So, the value proposition to investors of a non-correlated higher yield has really moved dramatic increases in structured product issuance. What effect has that had? Why is that important?

Because ultimately what that means is that dealers, the banks in particular, are taking on dramatic amounts of implied volatility. If customers are selling it to them and these banks are issuing it, they’re taking on long volatility. So, this long volatility is then being pushed on to the market. Counterintuitively, historically, when interest rates have gone higher, historically, people move to the bond market.

You’re getting some of that, but you’re getting bond plus, which is generally structured products. These did not exist in the 1970s. This, ironically, had a feedback loop that dampened volatility at the core, at the center of markets which is the index itself, which is where most of these structured products are issued. Not only has that driven record, really low VIX. If you look at the chart of implied volatility, it’s been a plane crash almost.

Not only has it driven up, but it’s driven a very important dynamic that other people are thinking about, which is dispersion. The vol supply is coming right into the middle, kind of like a donut of markets, right into the S&P and the indexes broadly. Meanwhile, interest rates, as they go higher, are increasing volatility on the edges and around, broadly, in the market.

So you’re getting rate volatility, like you’ve discussed and others have discussed. You’re getting FX volatility. You’re getting precious metals volatility. You’re getting commodity volatility. You’re getting all these things because interest rates are going higher and liquidity, broadly, is decreasing. But as that happens and yields go higher, you’re getting structured product issuance which is providing massive supply to the middle of everything, which is driving a massive dispersion and difference between how index vol is performing, in the equity land, relative to everything else. I think that’s the biggest takeaway.

Zero DTE, because of that vol pinning on the index, has been a very desirable place to sell volatility after initially getting a lot of interest from people last year because realized volatility was outperforming implied. So, from the long side, people were speculating, kind of like they were in YOLO calls and other ways, really, directionally, and trying to take advantage of realized vol. Now it’s really turned into a vol selling.

What zero DTE does is it allows people not to worry about this more complicated implied vol performance issue and just bet on the realized outcome. So that’s what’s really driven the appeal of zero DTE. Zero DTE has been made out to be something that’s going to destroy the world, that it’s going to blow up everything. I do think it has an incredibly dangerous function in the sense that it’s a magnifier.

Right now, that magnification is actually creating even more vol compression because of all the vol selling, but it’s zero DTE and that is so fickle. It can change in a day what that positioning is. If we were to get an event, that could very much turn the other way and become very much making dealers short vol and making customers long, which itself could be very vol accelerating as opposed to dampening.

So, it’s an accelerant, and I think the massive amount of volume there is dangerous. There’s not enough regulation that’s caught up with it. There’s no proper margin for zero DTE. It is 40% of the market now. Believe it or not, margining is done on a daily basis, so zero DTE options are actually not margined by regulatory authorities. It’s insane. There’s this issue, one, and two, there’s a simple fact that zero DTE can only be hedged by zero DTE.

If you go ask market makers how they’re hedging it, they’re hedging it with the only thing that they can because the deltas move so quickly and there’s nothing that has as much gamma or can be hedged in that way. Which is creating a reflexive loop where if, at one point, instantaneously, the markets get swept or there’s a dramatic long vol trade because of an event, that can force everybody into a cascade and can be very dangerous as well.

So, I think those are the things I’d highlight around vol in the current market, all very interesting, different new things. I will give a little teaser here to predictions going forward. What happens when interest rates come down to this trade? This has happened, all of the structured vol supply, which is really the edge that is the core dampener here, has been there because interest rates are higher.

Ironically, if interest rates go down, you’re actually going to see the opposite coming. So, I’ll leave that for everybody to think about.

[00:19:41] Niels

Now, I’m going to just call out a couple of names here, but I’m thinking we have only one who really works at a bank among us, and that’s Nick. So, Nick, any thoughts on this topic?

[00:19:55] Nick

Where do I start? Without necessarily claiming that I have the volatility expertise that Cem has. The one thing I would say, is that zero DTE is certainly, possibly, the keyword that has been played out more than any other in the last year and a half. Obviously, we’re observing the market in a variety of ways. We look at it, specifically where I sit on the systematic investing side of the business, as to how we can allow our investors and our clients to benefit from it but obviously in a very careful manner.

A few things that were mentioned, I think where the danger lies is that this is still a derivative. So that gives you magnification. I think that was precisely well-put by Cem. The way we look at it is, first we do some sort of analysis on net demand and net supply of those. We’re typically look at the grade between maturities and the deltas. So, the further out you go out of the out-of-the-money region, the more demand you have had from the market.

That can be easily understood because on the one hand, on the out-of-the-money call options, these are like lottery tickets. You come in, in the morning, you get the call option. If the market skyrockets for whatever reason, you basically get a gain at extremely little cost. On the opposite front, you have the so-called crash puts, and this can also be utilized by broker-dealers to hedge their gap risk.

Here you have some sort of natural demand, which serves specific purposes. Which is different from just systematically harvesting a particular premium, which obviously makes those much richer. Therefore, that’s a natural place to do selling. One of the things we’re trying to operate on is identifying possibilities of creating a yield-enhancing profile, effectively selling those options, harvesting the implied realized, but doing so obviously in a very careful manner.

We have seen this greed that I mentioned, maturity and strike shifting in the recent years. The classical weekly/monthly space that investors have been selling volatility has now been shifting into those shorter-dated more out-of-the-money options. It is a place that we’re very carefully monitoring. I’ll possibly leave it at that point before it becomes a bit more contentious, but this is certainly a keyword for the days.

[00:21:35] Nick

Okay. Anyone else? Mark?

[00:22:24] Mark

I wanted to ask you the question because I put this all under the broad topic of market structural change. We’ve got a change because of zero DTE, we’ve got a change because of the flows that are going into structured products. So as systematic investors, we’re constantly trying to say, “How do I adjust my models because of this? What is going to be different? Is it systemic and repeatable so it’s going to last for a while or is it not going to last?”

So I’ve ask the question, given all of this information in flows, how do you adjust a model to account for this? Or don’t you? It’s interesting, but if, let’s say, I got lower vol in equities, I could say my models might take into account different vol regimes, so therefore, I’m not worried about it. So, is this an issue that can be modeled? How would you handle this market structure?

[00:23:26] Niels

Rob, you’ve written more books than anyone else put together. So how do you deal with this?

[00:23:35] Rob

I don’t know. It’s been a long time since I traded vol. I haven’t traded it at all.

[00:23:38] Niels

But it could have an impact on other models, I think, on other markets. I think that’s what Mark is, as a model trader, so to speak, do we need to take into account these types of changes or do we just put it and say, well, this will be just an anomaly for a period of time, and then it goes back to normal?

[00:23:55] Rob

I guess it depends on your timescale. Obviously, the sort of stuff that the guys have talked about is very short-dated gamma. So intraday volatility dynamics or trend dynamics would probably be quite heavily affected. I guess for the timescale that most of us trade at, at the margin, all that’s going to do is probably have a quite smallish effect on our volatility estimate.

Because if you’re using, say, about a month’s worth of returns of vol, which is roughly what most people do in a kind of, hand waving way, then you’re talking about one day of that, which is like 4% being affected by potentially some weird vol dynamics at the margin because of this stuff.

So, I’m quite relaxed, to be honest. I guess you should probably stay away from things like executing trades around the period when these things are being hedged or when they’re about to close because that could cause some potential big gaps and stuff. But again, you probably would be sensible not to do that regardless of the vol environment.

So yeah, I remember trading vol 20 years ago and being short gamma with an hour before expiry. It seems bizarre to me that people voluntarily put themselves in that position. But we live in crazy times, right?

[00:23:13] Niels

Let’s jump on to the next topic, Alan. Very curious to hear what you brought along.

[00:25:20] Alan

Good stuff. Shifting gears a bit, and I suppose back to a topic I’ve come back to a few times before, and I know it’s a favorite of yours, Niels, as well around trend and non-trend strategies. The question is around the value of non-trend strategies. Very often, you hear from managers, particularly trend followers, who’ll say non-trend doesn’t really add a lot of value. It’s, really, just for smoothing returns.

First of all, is that true? Does non-trend really just smooth returns in a diversified managed futures portfolio? Secondly, if that was actually true, is that not a good thing in itself? What’s the value of smoother returns? I suppose I’m thinking of this from the perspective of if I think of the types of investors I’ve worked with, you get different categories of investors.

Some are more long-term family office type investors, who are very much shooting for the absolute returns and the highest compound average annual growth rate. But often you have a lot of people who have career risk and work in private banks or wealth management firms and multi-asset portfolios, etc. They have to report to an investment committee or a board. Smooth returns are actually, in my opinion, quite valuable for these people. So those two questions.

I suppose firstly, from the perspective of trend and non-trend, can non-trend work? Just to give a little bit more context, we had a research paper from CFM going back a few years, where they highlighted that if you looked at the returns from the SocGen CTA index and applied a simple trend, a six-month moving average model, it explained the majority of the returns of the index. So, their conclusion was CTAs weren’t very good at non-trend.

I guess maybe an ancillary question to this, is it not whether non-trend doesn’t work but maybe CTAs aren’t particularly good at doing non-trend in aggregate? So, probably three questions there in total. Very curious, probably to get Rich’s view on this because I’ve definitely heard Rich talking about trends and that kind of compounding over time, and probably Rob and, maybe, Mark as well.

[00:25:47] Niels

All right, Rich, I think as someone who embraces the trend too, let’s call it a healthy extreme, what do you have to say?

[00:27:53] Richard

Yeah, well it was a great question by Alan. Look, I think we all accept the fact that if we get an uncorrelated mix of strategies together, the natural result is to smooth the overall return profile of the portfolio once we bring them all together. But my question coming back to this would be, are we just contributing to smoothness or are we actually contributing to lifting power, what I’m calling lifting power of the portfolio?

So, one of the concerns I’ve got with the integration of non-trend into trend is that we accept that trend is already very globally diversified. When you’re therefore allocating into non-trend, my counterargument to this would be, are we making the portfolio possibly less diversified than it is under a globally diversified trend treatment?

The other question I’d raise is, what are the new risks we are incorporating into the portfolio with non-trend and how is that contributing overall to the portfolio? Is it increasing negative skew opportunities into the portfolio? Is it bringing in new risk into the portfolio?

We all know about the LTCM blow-up. We recognize that was under a concession, an extended history of what we call incredible smoothness, magnificent Sharpe ratio. There was this hidden risk embedded within LTCM and a negative skew and a crisis event cascaded out of that great result into what it actually became. So, are we, potentially, by incorporating non-trend into trend, are we for a start- can we go further in trend to achieve all of these outcomes like a smoother portfolio plus the lifting power in trend alone?

I tend to think that more effort needs to be placed on system diversification in trend. Market diversification is very well-known in trend, but I think that more effort in trend alone could be put into systemic diversification. Because I do think that we get the best of all worlds in trend, that by cutting losses short, letting profits run, we’re already embedding a positive skew asymmetry into the portfolio process. I think that’s very helpful.

One of the problems we’ve got to face is that typically, performance returns tend to cluster. With greater diversification, particularly in system diversification, I think that we can spread the impacts of this nonlinearity or the nonlinearity, the outliers which are driving a lot of our performance. I think we can spread that better across the time series as opposed to it being necessarily clustered.

So, there are some of the things I’d say. I love the question from Alan. But of course, being 100% trend following as I am, what is this non-trend we speak of?

[00:31:06] Niels

All right. Well, let’s jump to someone who actually does a bit of both, I guess. Rob, you embrace both camps.

[00:31:15] Rob

Yeah, I’m the famously impure trend follower, as everyone knows. Trend follower plus everything, that’s me. Obviously, those criticisms are all interesting. Generally speaking, if you add non-trend you’re going to push up your Sharpe ratio, but obviously, other things in your distribution will change. So, you will get less positive skew, but you won’t necessarily get negative skew. So, this is kind of a straw man here that you’ve got a choice between either running LTCM or pure trend. There’s an awful lot in between those two extremes.

You can quite easily add a fair bit of non-trend to your portfolio. Such that the Sharpe gets bumped up by quite a lot and the skew is reduced, but not massively, and you still end up with something that’s got a pretty positive skew behind it. I do think that the area of diversification is an interesting one because it’s definitely true that the diversification within trend following as a strategy is very high.

You get a very big uplift from trading one instrument, to trading 100 instruments in trend following, and that’s a bigger uplift than you get in, say a non-trend strategy like carry. Whether that means that the combined strategy has less diversification within it, I’m not convinced. I think you need to actually test that empirically. I’ve never actually done that work.

So, I’ve got some homework to do now before I come back to you on that point. Maybe somebody else has done that work and can comment on that, but I haven’t checked. My gut feeling is that, actually, a combined strategy of trend and non-trend would still be very diversified because it would still have all the original diversification of the trend strategy plus the diversification of the non-trend strategy, and so on.

I think the point about, we can do all this with trends, it’s true to an extent but there are diminishing returns. If you go from trading a single trend strategy on a single market, you can improve that a hell of a lot by adding additional markets up to a point. You obviously get diminishing returns beyond a certain number. We can argue about whether it’s 50 markets, 100 markets, 250 markets. But at some point, that kind of improvement is going to be leveling off and you’re going to struggle to get much more.

The same thing happens if you diversify across different kinds of trend strategies. So, if you go from just using a single trend strategy using different variations of a theme, again, you can get an improvement, but that levels off and that levels are very quick. So, there aren’t that many orthogonal ways of picking up trends out there, is what I would say. I do think you reach this point where you can’t improve, let’s say, the joint characteristics of the portfolio, by which I mean some handwaving combination of Sharpe and skew say.

You can’t improve the joint characteristics of the portfolio that much by just sticking in trend. Adding in some non-trend does improve that quite a lot. Although obviously, as I said, it does shift gears slightly away from being very positive skew, not great Sharpe, to being quite a bit better Sharpe and a little bit less positive skew. As I said, as long as you don’t go crazy and keep pulling that lever.

About LTCM, I think you’ve still got a portfolio that’s got some nice trend-like properties but has got a lot more internal style diversification with it.

[00:34:32] Mark

I was going to ask the question, that when you talk about smoothing, who should do the smoothing? You could do the smoothing at the hedge fund level or you could say the smoothing should be done at the investor level so that he has a better idea of how much smoothing he wants. So that’s one question.

The second is that is smoothing for a trend follower, is that a business decision? If you smooth, then you can reduce drawdowns and you could live another day to continue on your strategy.

The third question is that if we do add smoothing or non-trend characteristics, can a trend follower have a comparative advantage at doing that? Can someone who’s a systematic trend follower, does he have skill sets that could actually then be transferred to non-trend type of modeling? Because we’re all systematic modelers, so you’re just doing a count. You’re counting different events and you’re trying to look for repetition.

So, is that a skill that’s transferable and it just happens that we do it in trend? I’ll throw those out as some questions.

[00:35:45] Rob

I know other people want to join in and I’ll just say very quickly, that it’s quite true, to say on the first two points, that the incentive structure for me, as an individual trader who is doing what I’m doing, is going to be different from someone who’s offering a product out. If you’re offering out a product and you’re offering out a product to institutional investors who want to buy trend following, then clearly you want to give them trend following.

I’ll shut up now because I think the people working in institutions can answer these questions better than I can.

[00:36:10] Cem

Yeah, just real quick. Again, I have got sound off here from the vol side. The question ends up being what is the trend? Again, I’m not going to try and tell you guys what the trend is. But in my opinion, it’s a two-dimensional expression of market movement which is asset-based. Options themselves add the non-linearity and the ability to express some of these things from the beginning, the stops, the moving trend, the speed of the move.

All of these things, I think there is an ability to express and I agree that the expertise that comes from this quantitative modeling on this side can be put back into and expressed in volatility terms. We brought up before in the introduction and increase of using options, how can that affect trend? I would just throw this in here. That nonlinearity that’s been put in the market should be expressed with more nonlinearity itself. So maybe there’s a place out there to do trend but with more nonlinear expression through options as a way to express it.

[00:37:20] Katy

I’ll just make one comment. I tend to be a little more pluralistic. I think that people have different objective functions and different goals, and that goes with what Mark said. Trend is a very divergent strategy, as Mark would also probably say, and these other strategies are more convergent. So, it’s going to depend on what is the objective of an individual investor.

Some investors have different constraints related to being able to hold something, being able to explain it to their board. Then others are really saying, I want something that has the biggest convexity, that has the best complementarity to my overall portfolio as a whole, and I’m trying to optimize my overall Sharpe ratio. So, there’s room for all of those options depending on what the actual investor’s objective function is.

So I think we all probably agree with that, but that means that there’s room for all of these approaches across the space.

[00:38:21] Niels

Yeah. I’ll jump to Nick and then Andrew.

[00:38:22] Nick

I’ll be super quick because I think we have discussed a number of good points. The only thing I would say is the following. Adding a number of strategies together or profiles, we have to be extremely careful not cannibalizing what we already do. When I think about carry, and I completely agree with Katy, these are more like divergent strategies versus convergent in how we put it together.

In an environment where I’m buying bonds because the curves are upward sloping and I have positive momentum, then happy days. Or possibly, I’m concentrating too much risk to go back to some of the points that Rob made. At the time that curves are inverted, but for whatever reason, prices have not yet followed through, I’m now cannibalizing my long exposure with my short exposure and the whole lot just doesn’t really make any sense.

In other words, people say, how about reversion in trend? I’m like, okay, if I do anti-trend, I’m going to add the two together, I’m going to get to zero. So, I think what is extremely careful is trying to add more profiles. I think we should keep in mind what are the principal components we’re achieving and then try to the extent of possible minimize those exposures so that the addition of a new theme is purely additive, at least from a performance and Sharpe ratio perspective.

[00:39:30] Andrew

Of course, I completely agree with what Katy said. A lot of it depends on whether people stare at line items or they look at the whole portfolio. Just from observation of this space, when we got into the space in 2015, non-trend was cool. This was the period of time when people were afraid of the commoditization of trends through bank swap products and other things. There was this idea that there was a high alpha added component of the space and then there was something that was becoming somewhat more commoditized.

That all changed in March of 2020. Trend did better, much to people’s surprise. And then it did better in 2020, and it did better in 2021, and it did better in 2022, knocking the lights out in 2022. This year, it’s doing worse. , the really interesting thing is going to be this, the investor bias.

Do they look at long-term trend which has gone to a rough whipsaw recently relative to, and do they frame it in the context of the preceding years, or do you have this pivot back to non-trend because it has worked better this year? That’s an open question. We shall see.

[00:40:40] Niels

Okay, great stuff. All right, let’s pivot, so to speak. Nick, what have you brought along for us?

[00:40:50] Nick

I think it’s going to be interesting because, what I want to point out is, and I think, Katy and Alan already used the term SocGen Trend. For all of us here and most likely for the vast majority of our listeners, SocGen Trend has been quote-unquote the “benchmark”, and I’m going to explain myself when I say quote-unquote.

This comes from the so-called Newedge Trend Index where a futures broker, back in the days, acquired by SocGen, and then they continued rolling out that index which, itself, is the largest 20 CTAs or the largest 10 trend following CTAs or two variations of it. Pretty similar in a way because the vast majority of CTAs are doing trend following. Anyhow, that’s not the point here. The point being, this has been the benchmark that every one of us has been monitoring as the proxy of the industry.

As of about a month or so ago, it is no longer available in Bloomberg, at least quote-unquote for “free”. So, it’s now become available under a license and that has, obviously, brought a number of questions with ourselves and our clients and so on. What’s going on with his benchmark? Unless we are able to look at it in a free manner, how are we going to operate from this point onwards? It might be a small thing or not just a small thing.

By all means, I’m not going to make any comment with regard to a business decision I would imagine by the issuer. But it has raised a question. What are we now using to compare our returns against? Obviously, some of our clients said we should use the DBMF. So definitely I’m going to ask Andrew later on what he thinks about this one. Let me tell you my views.

I think in the CTA space and in the active management world, benchmarks themselves are very hard to define because even the constituents themselves are living creatures. The models we run today, or all of you guys are running today, are not the models you were running back then.

So even building a portfolio today with new ideas as to how we build a trend strategy, or whatever strategy, and comparing ourselves against the trend industry performance in 2015, it’s an unfair comparison because it’s now a different model. We’re not just building a rule today that we backtest as a rule and then we try to understand how better or worse we’ve done.

We’re now backtesting ourselves against a universe of managers that haven’t seen COVID, that haven’t seen rate shocks of 6 and 7 and 8 sigmas. So, to me, benchmarks in this space are very hard to define. I’m going back to the point I made on a “benchmark”. That’s the first thing. The other thing, however, which is critical for our discussion, is that human beings, let alone investors, they need comparables.

Somehow, it’s our human nature to look for something that we compare ourselves against. If we outperform, we’re better, we’re worse, we try to make better, we try to get there. So, I don’t think the fact that SocGen Trend Index is no longer available, at least in the form that it used to be, is going to negate the need for a benchmark. Whether that was good or bad, that’s another discussion.

So here’s my question to the group. What’s going to happen with the benchmark?

[00:44:06] Niels

Well, I think you’re absolutely right, Nick, in saying that it will be interesting to hear Andrew’s thoughts, because I guess, Andrew, although you may not be tracking the Trend Index, you are kind of relying on getting data on an index in order to replicate it. So, what are your thoughts? What if they take the SocGen CTA index and do the same? Are there other ways of getting the data?

[00:44:32] Andrew

First thing, Nick, I completely agree with all your points. One of the huge issues with this space is, again, you think about the typical allocator who’s building long-term capital market assumptions, SocGen CTA Index started in 2000. You will see people who will run numbers back to the 1980s, but the farther and farther back you go, the more made up they are. Because somebody in 2000 decided that there was some guy in 1985, who happens to have been alive 15 years later, who they pretend they would have invested in on an equally weighted basis.

The index issue is something that’s not just… When you think about the S&P 500, the S&P 500 is a living organism as well, and the nature of the firms that are inside of it change over time. That was Buffett’s point. When you’re investing in the S&P 500, you have dynamic companies, you have dynamic changes. It’s probably changing over time. This industry needs a better benchmark.

The SocGen CTA Index, itself, was inherently flawed in a lot of ways in that it was never really accessible. It was accessible to everybody here because you guys are industry practitioners, but what the industry needs is an investable version, where you can say it’s an SPY. Where you can say, “Here’s my easy way of getting exposure to the space. It’s not going to be perfect. It’s going to be broad-based exposure. But wow, look at what this guy is doing over here. We think he does much better and we think he’s going to do much better over time, therefore, we’re going to use him.”

So, I think this is something that’s going to evolve a lot over time. There is a circularity in it, whenever you’re talking about a hedge fund benchmark, in that you can’t invest in it. You can’t invest in it. So, you can look at the SocGen CTA Index but you can’t actually invest it unless you want to invest across the funds. So, it’s a huge issue. I am working on it, guys. Hopefully, you’ll hear more about it in the next several months.

[00:46:26] Niels

Okay, cool. If I may jump in myself and just comment on one thing, Nick, in that regard, I remember I was at a conference. I think, actually, it was the EQD Conference where I met you in January of this year, Nick. I went to a launch hosted by SocGen where they talked about the SocGen Trend Index, and how they were building a model to replicate it. So, they clearly were inspired by Andrew to replicate CTAs.

I think they mentioned during that lunch that they were going to launch that as an investable index. I also noticed that on the website of SocGen, when you download the data now for the Trend Index and the other indices, it first of all has a very short lookback now, it only goes back like a year, and it talks about a net asset value. It doesn’t talk about whatever a thousand has come into.

So, I wonder if what you’re seeing on Bloomberg has something to do with the fact that they either have launched it or are about to launch the Trend Index as an investable index where they do a replication of that index and not the index that Andrew is replicating. So that would be my, not my outrageous prediction for 2024. We’ll come to that in the second half. But I’m just throwing it out there. Andrew?

[00:47:47] Andrew

I would add one thing. In our world, the key thing is you need daily data. Fortunately, this is an area where there’s a lot of daily data. A guy named Corey Hoffstein, in the US, as soon as they announced this basically showed you take 10 mutual funds that have some trend somewhere in the first sentence of their prospectus and you combine them together and you get something that’s a 98% correlation to the SocGen CTA Trend sub-index.

I think this is a critical issue that the industry needs to solve because we’re dealing with people. The questions that Nick is asking, in terms of the integrity of the index, the accuracy of it, 99% of people just want something they can point to and they can say this is what we’re plugging into models, this is what it’s done over x period of time, and make some inference, some theatrical inference as to what they think the correlation, risk-return, Sharpe ratio and everything else to the strategy is going to be over time.

This industry, in part, really hasn’t had it. If we solve that, I think you can start to really expand upon it.

[00:48:55] Niels

Before we jump on to the next topic, I wanted to maybe ask you Cem. In terms of vol indices and tracking what the industry for vol managers are doing? I imagine that it’s even worse than it is in our industry, but maybe you can enlighten me here.

[00:49:10] Cem

Yeah. The whole multi-dimensionality part creates lots of problems. When you say vol manager, even if you say long vol or short vol, there is really no one long vol or one short vol. You can be long a one delta put two weeks out or long a straddle a year out. Those things are all long vol. They have almost no relationship to one another except on the far tail.

To look at performance, month over month, for something that has a tail strategy in that regard is almost impossible, never mind the short vol relative value pieces. As I like to say, vol is not an asset class. Vol is the underlying distribution that underlies all assets. I think trying to see it, to look at it as a separate asset class is a problem to begin with.

[00:50:14] Niels

Okay. Fair point. All right. Well, the next in line to bring up an insightful topic is you, Rob.

[00:50:21] Rob

There’s a whole set of issues about what we call ourselves, how we describe ourselves, how we classify ourselves. It’s hilarious that we can’t get our act together with respect to the outside world. But also, we’ve got infighting between people calling themselves classic, people calling themselves other things. So, it’s kind of amusing.

I find it difficult enough to explain to my mother what I do for a living without having to explain to her that, actually, there are these other people who are called classic trend followers and I’m not one of them. By the way, I wouldn’t call myself a quant trader because I’m not higher-frequency. So not really a question there, but a kind of kickoff for a discussion, I suppose.

[00:50:39] Niels

I think that’s fair. Since we have one of the people who are being called out at least, I’m not sure whether he’s doing it himself, but I know others are calling him out. Including him in the group of, apparently, only five in the world now. There are only five classic trend followers in the world left. Luckily, we have one of them here today, and that’s you, Rich.

[00:51:21] Richard

Well, thank you very much. Thank you very much. Now, listen, you’ve all got a chance to come over to our side. But look, I think it’s important to distinguish ourselves as trend followers. I tend to sit in Harold de Beer’s camp in that I don’t want to be considered to be doing exactly the same thing as everyone else in trend following land.

So, in relation to why we position ourselves as classic. We do want to differentiate ourselves from trend followers who are, what we would be considering, adding non-trend into their portfolios. We want to recall the golden days of trend following, of the ’80s and ‘90s where these very simple models, with no complex overlays, were being adopted by the trend followers.

We like to think that that is, actually, a strength, as opposed to traditional old-style weakness. So yeah, look, I’m very much in the camp of the need to distinguish what we do from what others do, I suppose. Already, we’ve got this difference in opinion regarding what is trend for a start. We see that across the investment world, across the trading world.

Some people consider it components of cross-sectional momentum, others think of it as components of absolute momentum. Others are thinking that trends might be segments of a mean reverting cycle. Others think that trends might be components of the tail region, which they refer to as outliers. So, there are all of these different opinions on what is trend.

This classic style we’re talking about is a style of model that, when they look at their histogram, are looking at 5% to 10% of their trades being the things that contribute to the overall profits. They accept that 90% of their trades, which were targeting trends when those trades were being implemented because the models were applying a rigorous process from the backtest, those 90% were effectively not outliers. They were not the things contributing to the overall success of our program.

So, I think it’s important when we’re talking with investors, when we’re talking with allocators, that we can clearly say what we do so that they can appreciate what’s under the hood of our programs rather than a generic sort of, “Oh, we’re trend followers, but we’ve got a bit of trend and non-trend in there. We’ve got a bit of pattern recognition, got a bit of this, got a bit of that, all of those components in there,” because I think that fuzziness doesn’t really help the allocator, it doesn’t really help the consumer.

So yeah, Rob, I’m sorry, you might not meet the classic definition. But look, with a bit of practice next year, I’m sure you’ll be able to get there mate.

[00:54:10] Niels

Good stuff. All right. Actually, I’d love to hear your thoughts, Katy, on this. Since you are also deeply rooted in the trend following space, whether you would define yourself as classic or not. I’m excited to find out.

[00:54:22] Katy

That’s a good question. My general view with trend following is that what’s so interesting about the strategy is really following an approach that adjusts to changing market moves. The reason why trend works very well is that it’s not really about adding your overlay of what you think should happen. It’s really about following what is happening in the markets.

So, from my perspective, it’s really that philosophical approach which makes you a pure trend follower, where you’re letting the data tell you what’s happening. So that really is a distinction for me between a pure trend follower and non-trend follower is that somebody who says, “Where are the markets going, let’s follow them”. You could divide it into classic, non-classic.

For me, it’s really that philosophical approach that you are moving with markets as they change. That’s where that diversification comes in the moments of the biggest stress. Because it’s so hard to make a decision and to actually prognosticate about how I enter this trade, what should I do in those moments? That’s where that diversification from trend following comes from is that philosophical view that I’m going to follow the markets.

So, I think for me, it’s more about a philosophical distinction. I do think that there is value to adding other approaches into trend if you have a different objective function. It’s something I do as well, so maybe I’m not a pure trend. But if I’m honest with myself, I think the fascinating part about being a trend follower is that dedication to the process and this idea of measuring what the market is telling you and letting the market help you to follow the prevailing themes with a systematic approach.

[00:54:39] Niels

Well, you certainly don’t disappoint me, Katy, because you’re always very inclusive. Of course, by that definition, we’re all included in this. But anyways, I’m sure it’s a topic that we’ll also cover in the next year.

[00:56:35] Mark

Niels, I was going to ask the question that we still wanted to get to. I always like to have, for hedge fund strategies or any kind of investment strategies, you always think of it like a taxonomy. We do it in the animal kingdom, we do it in the plant kingdom. So, if we had to do it in the strategy kingdom, I think for example, we’re all systematic. Okay, so that could be top. We’re all disciplined. Okay, that’s next.

Then you can say there’s some that are more market-directional, then there’s those that are market-neutral. Then there’s those that are short-term, long-term. Then there’s the style of how you do it. I always classify it according to style, the timeframe and the markets traded. But if we had that, would we be better off as an industry if we had a taxonomy so that people could then actually be able to view what we do and who we are? I think the answer is yes. But I’d be interested in what other people think.

[00:57:40] Niels

Rob, what are your thoughts?

[00:57:41] Rob

I absolutely agree. I think one thing that shouldn’t be in this taxonomy, necessarily, or that high up is the asset classes we’re trading. That’s what’s the least important thing I think to an extent, and maybe there’s a distinction in vol and non-vol. But does it really matter, for example if I get my equity exposure through buying a bunch of individual equities or indices to a first-order approximation? Not so much.

Does it matter whether I take my positions through futures or CFDs? That’s even less important. But one of the things about that kind of name of what we do, managed futures, is it puts the type of thing we’re trading right up there. I think, in terms of allocators, the first box they want to put you in is the asset class box. As a strategy that occurs across asset classes, that’s always going to make things a bit more awkward.

I think for me, the asset class should be right at the bottom of this taxonomy. I’m going to have more in common with somebody who’s doing what I do but trading individual equities than someone who’s trading futures but in a completely different way.

[00:58:53] Niels

I’m going to come to you in a second, Andrew. But by giving that response, Rob, you’ve actually clarified something for me, because we know that one of our very good friends to the show of course, he talks about mismanaged futures and how he’s not part of the managed futures industry anymore. But of course, by your definition, he wouldn’t be because he trades individual stocks, not futures.

[00:59:16] Rob

But he’s still part of the trend following family.

[00:59:18] Niels

He’s part of the Katy family, yeah.

[00:59:21] Rob

Like Katy, I believe in inclusivity, and I think these silly debates about whether you’re doing one thing slightly different from somebody else, but if you actually look at the output of returns and how they come out, they’re not going to be that different.

[00:59:35] Richard

I love this man, by the way. I just thought I’ve got to say that.

[00:59:41] Niels

Andrew, you had a thought?

[00:59:43] Andrew

My thoughts are twofold. One is that asset classes changed their names to something that has a positive connotation. So, junk bonds became high-yield bonds. Leveraged buyout sounded really scary, it became private equity. Asset-based lending became private credit. Most allocators just need a simple construct or a name. If they like it, they don’t ask a lot of subsequent questions.

My experience is you say managed futures and then a lot of people that I’m talking to don’t know what a futures contract is, but it sounds incredibly scary. You say something managed, there’s no positive connotation to it. Everyone on this call believes that trend following is a good idea. I was surprised, when I went to talk to people, that a lot of people associate themselves with being the guy who gets it the wrong time.

I used to use the story of Eddie Murphy and Dan Ackroyd in Trading Places and I’m thinking they’re going to identify with those guys, and the crowd is foolishly driving around prices to make the money. They associate with the crowd. All of these strategies, whatever they are, they’re all opportunistic and it’s all trading. If you try to get it down and I think when you think about something that’s called opportunistic, that’s positive. It’s not unopportunistic, so opportunistic and it’s trading.

The reason I’ve put it in that context is, going back to this point about where I think the alpha generation from space is, it is not eight other people on this call beating each other up and extracting money from one guy or the other. It’s not a zero-sum game. It’s the eight people on this call with trillions, tens of trillions of dollars that’s designed to move slowly, that has an ethos about mean reversion, not mean aversion.

So, the label for the space, the common thing that picks up trend, non-trend, carry, etc. Everything else, is basically a way for this space to generate excess returns over time that’s complementary to what the rest of things people have in their portfolio and that it’s conveyed in a positive way, where somebody hears and they say, “Oh, that sounds great, I want that.”

[01:02:02] Niels

That is it for part one of our 2023 year-end special. We will, as mentioned, bring you the second part next week which will also include our outrageous predictions for 2024.

We hope that you enjoyed this, and if you did, perhaps as your Christmas gift to us, you could take five minutes of your time and go to Apple podcasts, Spotify, Amazon, or wherever you listen to your podcast and leave us a rating and review. This will mean a lot to us.

While you’re there, make sure you follow the podcast as this also helps us grow the show. If you want to receive my weekly newsletter, just head over to toptradersunplugged.com/TTU-news.

From all of us at Top Traders Unplugged, thank you so much for listening. We look forward to being back with you next week for part two of this conversation. Until then, happy holidays, and take care of yourself and take care of each other.

[01:03:05] Ending

Thanks for listening to the Systematic Investor podcast series. If you enjoyed this series, go on over to iTunes and leave an honest rating and review and be sure to listen to all the other episodes from Top Traders Unplugged. If you have questions about systematic investing, send us an email with the word “question” in the subject line to info@toptradersunplugged.com, and we’ll try to get it on the show.

Remember, all the discussion that we have about investment performance is about the past and past performance does not guarantee or even infer anything about future performance. Also understand that there’s a significant risk of financial loss with all investment strategies and you need to request and understand the specific risks from the investment manager about their products before you make investment decisions.

Thanks for spending some of your valuable time with us and we’ll see you on the next episode of the Systematic Investor.