BREAKING: New CTA Index being revealed ft. Andrew Beer

[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 week’s edition of the Systematic Investor Series with Andrew Beer and I, Niels Kaastrup-Larsen, where each week we take the pulse of the global markets through the lens of a rules-based investor. Andrew, it is absolutely wonderful to have you back on the podcast this week.

How are you keeping and where do we find you today?

[00:00:40] Andrew

Thank you, Niels, for having me back. It’s always great to join you on these. Everything’s good for the tail end of the summer, looking forward to the fall.

[00:00:48] Niels

Absolutely, absolutely. Now, I am very much looking forward to this conversation today. I’m looking forward to all of our conversations, of course, but this one in particular, because you are going to share something, it’s kind of like a world premiere, I think, with our audience.

And as you told me before we hit record, it’s like a real scoop in the journalistic sense. So, I’m sure the audience will be on their edge to hear what that is. But of course, we’ll have to wait a little bit because before we get to all of that good stuff, I’m going to be asking you the usual question, and that is, what are the things that have kind of been on your radar lately since we last spoke?

[00:01:28] Andrew

Well, so outside of work, I read an editorial in the Wall Street Journal a little while ago. It was actually an editorial about how is it possible that the Democrat strategy, with Kamala Harris, is basically to avoid any details and any relevant policy decisions? This is a guy from the Wall Street Journal who’s basically challenging that.

But there was a paragraph in it that I found really, really interesting. (You and I have talked a little bit about social media and it’s impact on an entire generation of people, including us, I guess, in a certain way. We were developing our minds during that.) But he made an observation that a lot of the relentless clicking is about finding something new.

And he said, what’s been lost (the second part) is trying to look for quality content. And it sort of brought me back to about almost 25 years ago, when I started a school in Connecticut, and one of the ideas of the school was that instead of having a child read somebody’s interpretation of Beowulf or the classics or something, wouldn’t it be great if they could actually just read it themselves?

And I think with all of these different layers that we get, by the time information gets to us it has been filtered and translated, et cetera, so many different times and synthesized. And AI is actually an extension of that. AI takes everything and boils it down to a paragraph, so you don’t really have to read the whole thing.

Anyway, for me, it just got me thinking that when I see something in the news that I think is important, and I hear about a new indictment or a Supreme Court decision or something, I’m going to go back to what I used to do. I’m going to just download the original document and read it and see what I see differently and get a feel for the fact that people are writing and producing this material. Anyway, I’m going to further remove myself from Selma. I’ll become even less able to engage in a lot of conversations about current events.

[00:03:39] Niels

I think it’s a really important conversation. It’s a really important topic. And I think it does have a massive impact on the current generation.

I don’t know about the future generation. At some point, maybe they’re realizing that we all now have our own individual truth. Nobody really knows what the truth is anymore. And I think that’s a major, major problem.

So yeah, I think it’s very interesting. I can’t really match that in terms of original thought this week.

I was thinking about what’s really been on my radar this week, but I’ve been too busy doing practical chores all week. So, I haven’t really thought much about this. But we are getting close to month end. Of course, August, the month of August has been kind of interesting, to say the least, as we’ve seen in the past.

[00:04:26] Andrew

Good riddance.

[00:04:28] Niels

I mean, as we’ve seen in the past, August always throws a surprise once in a while. And so, it turns out, it’s going to be a down month for CTAs. I think there’s no big surprise for that.

But it’s nothing catastrophic. I looked at the SocGen Trend Index. It’s in a 10.5% drawdown since its high in March. That’s about 50% of its worst drawdown. So, I don’t see that as a big issue.

So, then I was looking at, where do trend followers probably fall this month? What’s been difficult? What’s not been so difficult? But, of course, most things have been somewhat difficult, with a few exceptions.

But, you know, yeah, equities. Clearly, it’s been a challenge to navigate this kind of V-shaped trip we’ve had. But of course, some markets have not fully recovered.

So Japanese equities is a good guess that that’s probably where most of the damage has been done. I would say JGBs, I think was also tricky for those who trade that. Currencies, sure, but not all of them, though. And maybe not the yen as much as some of the other stuff that’s going on, I would imagine.

Fixed income, a very divided story in August. I think US fixed income has been okay. I think non-US fixed income has been a challenge for trend followers. Energies, not a lot of love from energies this month, frankly, across the whole complex.

Softs and grains though, that’s a different story. I think trends have been decent in those areas. Of course, it’s not the biggest sectors of many CTAs, and therefore they’re not going to be offsetting all of the losses, of course.

Meats, for those who trade those markets, have been a little bit tricky, as far as I can tell. And then metals, gold has been great, but not so great in some of the other metal markets.

So, you know, a down month, but in my view, nothing catastrophic. And then my own trend barometer, even though I should say, again, I repeat this from last week, I have some challenges with my software, so I’m not posting at the moment, but you can always get it from the podcast every week. So, for those who have built replication strategies based on my trend barometer, ooh, that’s a bit of a challenge, but luckily I don’t think there are any who have done that.

It finished at 59, confirming that we have had some better sailing, some tailwind in the last couple of weeks, at least a few weeks. CTA indices as of Wednesday night, so still without the last two days of trading, although yesterday not a lot happened, maybe a little bit of a down day, I would imagine. BTOP50 right now, down 1.68% in August, up 3.43% this year. SocGen CTA Index down 2.58% for the month, up 2% and change for the year. SocGen Trend down 3.63% and up 2.22% for the year. And the Short-Term Traders Index is up 19 basis points, both for the month and for the year.

And then finally, in the traditional world, probably many people wouldn’t have believed that after three days of trading in August, but the MSCI World is actually up 1.7% for the month, up 14.6%. The S&P 500 total return up 1.39% for the month and up 18.3% or so for the year. And the S&P Global Developed Sovereign Bond Index is up about a percent for the month and up one and a half percent for the year.

Before we dive into all of your great content, we have a question, by the way, also from one of your investors. But do you want to put a few words to August specifically in terms of how you saw it? Was it just another August or was it just another month? Was it anything to be excited about?

[00:08:25] Andrew

Well, I think so. I just wrote one of our investors letters a little while ago. So, I think when we’re looking at these big market moves, these big whipsaws, you know, the question that you can’t answer at the time is how much of it is rational repricing of assets based upon changes in underlying fundamental economic or macroeconomic conditions and how much of it is the technical stuff?

And in the midst of it, when things start to move, if you have liquidations or whatever, there’s a feedback loop. I used to always see this in the commodity markets, years ago, when I built a commodity hedge fund business and as things would start to move, the fundamental guys would start changing their fundamental view. People always think somebody else knows something – somebody knows something, the economy really is hitting the windshield.

From a trend following perspective, in early August, my perspective of it was that there were three really, really, really big shifts. If you go to late June, and actually go to mid-July, you had three big things going on.

So, one is that it looked like Trump was going to come in, in the US. He had survived an assassination attempt, and he looked like he was going to be a shoo-in against Biden. It was going to be a massive red wave, the pressures off Biden. You had that economic growth was still okay. People didn’t think the Fed was going to cut rates right into an election. That seemed too political.

In general, the Bank of Japan started making noises, but they looked more like whimpers. There is always this view that, of course, the Bank of Japan should do something, except with a small issue that they risked sort of torching their domestic market, which is actually doing kind of quite well given a lower yen in exports. So those three things changed in like two weeks.

No one on planet earth predicted that Biden would drop out, Harris would come in (Harris was believed to be more unpopular than Biden). And now she’s in this kind of weird, you know, information-free bubble. (And, by the way, I’m a lifelong Democrat, so it’s not saying this as somebody who hates it, but well, I used to be a lifelong Democrat.) Then look in the Bank of Japan, and then you’ve got a couple of economic data points that were different than people expected.

Then the Bank of Japan threw some kerosene into the mix. And so, within a couple of days… Look, there was a period in early August where people were like, it’s the economic flight.

So, I think that from a trend following perspective, this is what I’ve always said about short-term models, is that short-term models sound great but they’re much more at risk of getting ferociously whipsawed in three days like that. If the technical drivers and people are getting ahead of themselves on the market repricing, they have a potential to overreact, either get wrong footed or get out of positions, and then miss the mild recovery.

Now, for guys who pick up on more long-term trends stuff, like us, the risk is that August turns into what it was in November of last year. And it’s an inflection point that then continues in a particular direction and a replication model or a long-term trend follower will be slower to de-risk during that period of time.

I never looked at what was happening in August as anything catastrophic. I think it’s just part of the business. Actually, it does get into part of the struggles that the typical allocator has in getting behind the space. And we can talk about that more a little bit later. The idea that the market is going to tell you when you’re wrong and you’re going to get out of a trade is not something that fits into the standard ethos of an alpha generating hedge fund investment.

[00:12:33] Niels

Yeah, absolutely. Well said.

All right, so, we had a question from Martin before we kick into your topics. Martin writes, as a holder of DBMF, I would characterize the performance of the past couple of years as a series of fairly smooth uptrends punctuated or punctured, I think it says by sharp and disappointing reversals. Let’s say two steps forward and one and a half steps back.

It seems to point to the possibility of certain holdings having an outsized weight in the fund. To be fair, this type of performance has been playing out in other recent entrants to the managed futures ETF space. What might you do differently in the future, if anything, to reduce such drawdowns?

[00:13:16] Andrew

Sure. So, it depends on what period you look at, right? So, 2023 was our worst year from a replication perspective.

There are two known limitations of replication. One is that sometimes we’re going to miss trades. And in January and February of last year, we missed trades.

And then, as we say, if those inflection points then continue, we’ll be slow. So, both those things happened last year. It was the only year where we’ve sort of underperformed the index.

If you look at from the peak, in sort of end of the third quarter of 2022, we are down from that period of time, but we’re still ahead of the Hedge Fund Index over that period of time. So, the issue that Martin’s describing is, I think it’s an industry wide issue. It’s not as though the industry is up 20% and we’re up three.

The challenge (and this is going to lead to the next part of it) is that the way we think about replication is that you aim for pretty few returns and you hope to have this efficiency tailwind, where things can go wrong over the course of the year and you can still match performance because of this fee and expenses and intermediation. But the problem is that when people are comparing us to a net-of-fee fully loaded hedge fund number, which is what the underlying index is, we can look more volatile because we don’t charge incentive fees. Incentive fees, smooth returns.

That’s, in a sense, why we spend a lot of time trying to figure out how do we solve that? Which will probably lead into our next part of the conversation.

[00:14:54] Niels

So, we’re going to be talking about some of these things, actually, because you brought along a number of topics, in addition to the world premiere, of course. We’re going to start out in a slightly different place. And I want you to set the scene because it’s kind of a conversation that you and I have touched on a number of times, I would say. So let’s start with that. Why don’t you take it from here and, and let’s dive into all of this.

[00:15:21] Andrew

Sure. Well, do you want to talk about the launch of the index?

[00:15:24] Niels

No, no, I was going to talk about that in, say, in January and February of 2023, multiple CTAs came out proclaiming that replication only worked during sort of 2019 to 2022. And that was specifically due to certain macroeconomic conditions. And it was going to fail going forward. That was the contention, I think.

[00:15:50] Andrew

Yeah. I’m used to having arrows shot at me and rocks slung at me from the hedge fund industry. So, the way you can think about replication (just to make sure everybody understands), the way a replication model works is you look at very recent history of something like the SocGen CTA index, and you try to figure out what’s driving returns with a simple portfolio and you implement it.

So, in our models, we use 10 factors – big, obvious factors – two tens and 30 years in treasuries, three equity markets, three equity indices, Euro/yen, and gold, and oil.

The mathematical rationale for using a few… So, the way we built our models is to evaluate what we thought were the positions of the broader managed futures space implemented on Monday afternoon, wait a week, and do it again. There are a lot of reasons why we decided on that, but what you’re really making is a one week bet, right?

You’re making a one week bet that you’re sufficiently accurate there that you’re going to get close to it. You’re not betting they’re going to be close for the next three months. So, if you’re a bit wrong, it’s self-correcting. That’s the whole point of it. It’s not immediately self-correcting, but it’ll be self-correcting over time.

But the simplicity of the portfolio is in opposition to what has been increasingly complicated portfolios among actual hedge funds. And in a space that has a proliferation of products between much simpler and much more straightforward to more complicated, where there has been a long-term degradation in fees generally across the space, going back to where you were 20 or 30 years ago. A lot of the times there’s this conflation of complexity, therefore justifying higher fees.

And so, it’s complexity and lots of instruments types will necessarily lead to better diversification benefits, more alpha generation, and therefore you should pay me more to do that. It’s a very, very good argument without a lot of data behind it. So, we started to show up on people’s maps, in early 2023. Just to put it in context, we started doing this as a business in July 2016. Well, we actually started in 2015, but it was a slightly different way that we do it. So, from July 2016 to the present, we’ve been running as replication models that haven’t changed, So, that’s very unusual for a quant-based strategy.

Back to Martin’s point, are we going to do anything to change it? We may have to make a change in our model at some point, but we have an incredibly high bar for it. And our bias is always not to change it. That’s, again, very, very different from the broader quant world.

The reason it can work… So, if you look at the numbers since 2016, broadly speaking, we’ve done about 300 basis points better than the broad industry with a 90% correlation. So why would you do better than the broad industry and index? Because you’re aiming for pre-fee and we have an estimate of trading costs. Our trading costs are 10 basis points a year. If you start doing more complicated things, your trading costs go up.

So, if you look at it over that period of time, most of the time we outperform fee and expense disintermediation. But there’s plenty of noise.

Now, the argument of why 10 factors works is that over that one week window, which is what matters, if the 10 year treasury is moving, it’s got a gravitational pull that it’s pulling a lot of different things. If oil is going up a lot that week, probably heating oil is not going down. It may for a short period of time, but again, if that’s a 1% position in the portfolio, you’re not talking about a lot of tracking error.

So, the first idea is that basically most factors, millions of different factors; equities, et cetera, over a short period of time can be boiled down to a handful of large liquid instruments, but it’s not perfect. So, you can think about it.

Let’s say in most weeks, we’re, we’re capturing 90% of the instruments.
Most of the time, if you’ve got 10 instruments that are really idiosyncratic, like cocoa, we would say is really idiosyncratic, if we’re not picking up on it, there might be a several week period of time where cocoa is moving far more than we can pick up. But most of the time it’s not moving that much. Or if cocoa is doing well, another contract that we wouldn’t pick up very well is one moving against it.

But occasionally, a bunch of those things will go like gangbusters, and we’ll underperform like we did in January and February. So, the question for us is, is it predictable and does this suggest that anything’s broken? And it’s frustrating when you’re in the middle of it.

You want to change it, and you spend a lot of time worrying about, is there something structural that’s changed in the market? Which is actually the question. And our conclusion was, it wasn’t. And so, we didn’t change it.

Because there are other times, like this year, where the replication models are outperforming a lot this year. There are other times when some of the positions that we don’t get are going to move against you at the same time. So, we run much more diversified trend following models more as risk management tools. And you can see there’s been a ton of whipsaws, you know, idiosyncratic whipsaws spread across a lot of these other instruments.

So, what people in this space were saying, in early 2022, one of the arguments that they were making was that 2016 through the end of 2022 was some bizarre period where the Fed was dominating markets, markets were tightly controlled, et cetera, et cetera. But in this new world in 2023 onward, there would be much more idiosyncratic dispersion between all these different instruments. And hence, that was going to be the driver of a meaningful amount of alpha, which therefore would be unreplicable.

And plus, also the other issue with replication was that we’re a little bit slower, right? We’re always going to adjust more slowly. So, the shorter-term models were great on November 1st, last year, and they were great on March 13th of last year. There were two periods last year where you had major extended prolonged violent whipsaws. But then it hurts you when the reversal happened. I think for most people who’ve looked at it, who don’t have skin in the game, the replication debate is over. It works. The only question is just also being open minded about what its limitations are. How well does it work?

How well should you expect it to work over time? Should you expect it to continue to do 300 basis points of excess returns over time? If you look at the past couple of years, it hasn’t been 300 basis points of outperformance.

[00:23:02] Niels

Yeah, absolutely. These are good points and very interesting. And, of course, it’s only you and I on the podcast. So, I’m going to have to volunteer to take the role of a CTA to think sort of out loud. Reluctantly, I’m taking the role of the CTA, thinking out loud what they may say when they hear this.

The first point, I would say, is that I completely agree with you on this diversification point. I mean, the voices about trading hundreds of markets being better than trading 60 markets. I don’t think that, and I’ve said for years, I don’t see it in the numbers. There’s no evidence for that and so on and so forth. However, I will also be transparent to say that personally, I don’t know, I don’t have enough knowledge or data or anything to say, well, 10 or 15 markets is actually the same. It’s going to be, it’s going to be fine.

It may well be. You’ve obviously proven that you can get quite far with that. But I’m certainly in the place where you don’t have to have hundreds of markets, for sure.

[00:24:10] Andrew

But to be clear though, I’m not advocating doing a 10 factor trend following model. The data is there, right? You add more instruments, you get more diversification.

But remember what diversification also does is it also pulls you from… So, in the broader hedge fund industry, this has been a debate. Bill Ackman’s got eight positions or something like that. Do I want him to have 50 for diversification? No, I want him to have his favorite positions.

So, there is something, but managed futures is different. It’s about a whole series of bets with a 52% probability of being right or something like that, or 55% probability of being right, or making $2 and giving back 50 cents, or making $2 and giving back a dollar when things work.

So, I agree with you that I don’t think that a very, very simple trend following model is the ideal way to approach the space, but as a one week bet, which is what replication is, on the broad exposures of the space, I feel confident with 10.

[00:25:12] Niels

Yeah, I think that’s fine. I think that’s fair. Now, I wanted to ask you something, and that is when you talk about replication, and obviously part of the replication aim is to track a certain benchmark. In your case it’s the SocGen CTA index, and you said you have a correlation of 0.9 or 0.88 or something like that. Do you have like any guidelines to say, well, hang on, if our correlation drops to 75, is something wrong?

Meaning because there are two things that people will judge you on. One is the correlation, so to speak, because that’s part of it. The other one is the amplitude, which is the performance, right? That’s two things that are important here.

And I was just curious, when you talked about it, whether there is something where you say, well, actually correlation wise… People can say, okay, well, if they can’t outperform by 300 basis points, okay, we’ll question that. But they may also question, well, what if the correlation is only 0.75 or something like that? Is that something you’ve thought about?

[00:26:15] Andrew

We think about it. I’m not sure what you can do about it, right? Because, I mean, let’s say, and look, and correlations will go up and down over periods of time.

[00:26:23] Niels

I’m talking about, you know, obviously over a longer period of time.

[00:26:24] Andrew

If you look at rolling periods, it doesn’t drop to 20% because it’s the self-correcting nature of it, right? If you’re flipping coins every week, there will be periods of time where you flip five tails in a row. It’s not likely, but it happens.

There are times, like this year, in a sense, first half of this year, I would say we’re flipping a lot of heads. But this is sort of leading to the second part of the discussion, which is that the central issue is that we’re trying to track something that we cannot invest in directly. And that the index that we’re tracking, we believe is, and will continue to be, the industry standard hedge fund benchmark.

But when you’re talking about a 90% correlation, that kind of outperformance, the question is, what are your alternatives? So, you could get $10 billion from a sovereign wealth fund and go to every one of the 20 funds in it and give them each a 5% allocation and agree to pay full fees. You’re going to match that allocation perfectly. Is that what you want? Then nobody does that, right?

So, then the question becomes that you could hire Abbey to do it, right? You know, Abbey is efficient. These guys are presumably the best in class on the fund the fund side in this space, or you can pick three or four different funds, or you can pick five funds. You can pick six ones.

The whole question is that I see this return stream in the SocGen CTA index and, and Man, we’ll tell you that it should be 40% of your portfolio. And a lot of guys on Fintwit will say it should be a minimum of 25% of your portfolio.

Winton, as you pointed out with that paper, you know, says there are a lot of other considerations. I agree with Winton on this. So, people say, I want that return stream, but you can’t access it directly. And that’s the breakdown between deciding you want something and how do you actually invest in it?

And that’s the issue that we were facing in 2015. I love the return stream. Can we find a different way to get exposure to it given our own constraints? We’ve yet to find a sovereign wealth fund that will give us $10 billion to invest with you and everybody else.

[00:28:33] Niels

Good stuff. All right. So, another thought that I have, so just again, sort of from, from maybe from a manager’s side. Yeah. I mean, I think it makes sense that you guys would outperform in 2024. Part of it is that a lot of managers made new all-time highs in the early part of 2024. So, their performance fees would have kicked in. Obviously, that helps you. Unfortunately, a lot of managers also have quarterly crystallization. So, it didn’t help them oddly enough.

In Q2, when things went the other way, there was no performance fee coming back. So, I think, for me, it makes perfect sense that you would outperform in a year like this.

But then I thought of something else, and this is kind of new territory for me. So, I’m not certain about this, but I am curious about it. And I don’t think any of us will know the right answer, but it is relevant, I think, for the discussion.

Because you’re essentially replicating an index, an index that I think is made up by managers’ composite return track record. That’s different from a specific fund. It’s the composite of all accounts they manage in that strategy, right?

And I was actually thinking that maybe we’re not comparing apples to apples completely. Maybe it’s kind of red apples to green apples. So, I’ll explain that.

Because not all managers include interest income in their composite returns. For example, I know a manager that I work for. We stopped doing this in 2007.

So essentially our composite returns might be four or five percent lower than the fund people can invest in. It’s just because we don’t think it’s our alpha, because it’s not, so we don’t include it in our composite returns. And I was just thinking, maybe that plays a role as well.

Maybe the index that SocGen is creating is an index that doesn’t include very much interest income. And of course, your product will include a lot of interest income these days. So maybe it’s just a factor to be aware of.

I don’t know the answer because I don’t know how everyone calculates their composite returns. But I know you don’t have to. If you can’t identify the interest income, you can’t include it.

So just something to throw out there that I have not heard as part of the discussion on these things.

[00:30:56] Andrew

I would be surprised if, in the SocGen CTA index, that a manager is reporting excess returns and not total returns. That would be…

[00:31:07] Niels

As I said, I don’t know the answer. I just say it’s a factor people should be aware of, of course.

Now, you said also in your notes, you sent to me, you said, candid conversation of two known limitations, replication speed and simplicity.

If you feel you’ve covered that already, that’s fine.

[00 31:27] Andrew

I think so. Yeah, yeah. That’s fine.

I mean, look, it’s not… I mean, look, that’s what I always say about replication. It’s not perfect.

[00:31:31] Niels

No, no.

[00: 31:32] Andrew

If you start with the assumption that I really want to get exposure to the space and reduce idiosyncratic manager risk. And I want liquidity and ideally put in some sort of a wrapper that my clients can invest in. I haven’t found anything better.

[00:31:56] Niels

No, that’s fine.

You mentioned right there, and I think that was part of the early arguments from our years ago when we started having these conversations, was… And maybe you changed your view a little bit. I know you changed your view on a few things. But I remember early on, you very much focused on this thing about manager risk. People would have manager risk if they invested in a single manager. And I accept that completely.

But do you think people speak enough about replicator risk? And what I think about when I say that there are two things that come to mind. Like any systematic manager, our systems could be, you know, out of sync. Let’s call it that. Or I wouldn’t… I mean, of course, you could have a rogue code. That could happen theoretically as well or whatever. But I’m thinking, well, that must apply to a replicator as well. You could have something in the model that goes out of sync for a while.

So that could be kind of similar.

[00:32:49] Andrew

Absolutely. I mean, the replication space, right, is… I mean, first of all, I mean, the number of smart quants who have made boneheaded decisions on the replication front, right?

So, replication is… I mean, people have been doing this, these kinds of kind of top-down replication really since 2006, 2007, which is when I kind of stumbled into the space.

And there is institutional knowledge that goes into how people decide to build these different kinds of strategies. And it’s very, very hard for most… There’s always been this kind of divide between people who understand hedge funds and quants. Like AQR, if AQR wanted to get into the replication business, they’re probably not going to make stupid mistakes. But you’ve had a lot of people who’ve done things like, I’m just going to replicate this fund, or I’m going to replicate that fund, or I’m going to replicate… One of the banks at one point decided to try to only replicate funds (this is in the late 2000s) with sharp ratios of 2.5 or higher, with a factor model with daily liquidity. That’s predictably stupid. They ended up with every mortgage-backed securities fund in it.

So no, I totally agree. I mean, so our ethos (then this goes back to how long we’ve been doing it for), we just don’t think complexity helps in this.

So, what’s happened, competitively, is that we’ve built the business as the… There was nobody who replicated managed futures like this before we got into it. And we’ve kind of established not only this idea of replication. And if you’d asked eight years ago, if you’d asked a hundred allocators whether they thought this would work, a hundred would have told you absolutely not. And so, we’ve kind of established the space, and then we’ve done it in a very simple way.

So, when there are new entrants that are coming into the space, who are going to have to add something to be somewhat different. It may help, right? So, Katy, Katy and Alpha Simplex have launched an ETF. Since that ETF launched, the numbers look very, very great. They look great and predictable and very much in line, or maybe even a little bit ahead of what we’ve seen in our simpler models.

So, if they’re doing something that we’re doing, they’re adding something onto it. And then they may be adding some value there over time. My general view is that the bottom-up models, which are supposed to add a meaningful amount of value because you’re actually making a call on whether something has crossed over from the 200 versus 20 day or whatever moving averages you’re using, I don’t think those add value to it. But I could be proven wrong, and we’ll see.

[00:35:41] Niels

Yeah, I completely take that on board. And absolutely, of course, there’s always going to be something.

The final thing I was thinking about, in preparation for this, that I thought would be also a difference, and maybe it’s more of curiosity, more of a question, because, as you know, I think what you guys do for the audience that you go for is a great service. People can get exposure to this strategy, which everyone should.

But I was thinking also maybe from a selfish point of view, the hoops we have to go through as a manager when we get in touch and we talk to institutional investors. And a lot of that is actually to do with the organization. It’s not even to do with the model.

It’s, you know, how many people do you have here? How many people do you have there? And who replaces that guy? Who replaces this girl?

And I was just thinking replication strategies, to me, seems to be simpler and therefore requires fewer people. There are simply fewer people doing it. How do you deal with that? I’m just curious when people come and say, okay, we’re interested, but over here we have 100 people.

[00:36:51] Andrew

Where’s the army?

[00:36:53] Niels

We have 100 people over here. Even at our size at DUNN, we’re not the biggest in the business. All I’m saying is it’s something I come across as a topic.

[00:37:06] Andrew

Well, I think you’re pointing to a broader issue, which is the fund to due diligence process, as it is. So, whenever you’re thinking about something like that, you want to take a step back and think like, you know, this is an investment decision for them.

It’s often not really, because the hedge fund selection, as a practical matter, is an utterly efficient process. You don’t have investors coming in where you have some secret trade that you’re doing that they’re going to tell them about and you’re not going to tell about the last guy who was in. Everybody’s going to have exactly the same information. So, nobody has an information advantage when they come in. And often, by the time these institutions are looking at people, they’ve already been successful.

So, they may have been one out of 20 funds that was successful. So, there’s a massive kind of selection bias built into it. In the managed future space, in particular, on the allocator side, one of the things I’m writing about is actually that this idea that managed futures is trapped in the hedge fund bucket.

And I’ve got these charts that basically show that if you look at the past 10 years of good data, managed futures have hit the trifecta of what a hedge fund is supposed to be. Low correlation to equities, does well when you need it the most, generates high alpha. It should be the cornerstone of every hedge fund allocation. And it’s 6% of the space.

But that gets into a broader question of the people that you’re talking to, the other 37 conversations they’re having during the course of the week, they’re talking to people who are telling them about like, I found this stock and I’m capturing an illiquidity premium. And the market’s wrong about this.

This affects me and going out and talking to advisors as well, is that there’s a tremendous amount of anxiety in these people about trend following and whether it should work and why it works. They’re presented with 24 years of data from the SocGen CTA index, that it’s an absolute no brainer that this should be maybe not 40%, but 5% of every single portfolio and they don’t do it. And so, in part of my travels in allocator land, and the summer where I kind of sit and spend a lot of time thinking and digesting, and I realized it’s not that they believe it hasn’t worked, it’s that they believe it shouldn’t work.

And it’s very similar to what we just talked about with replication. The evidence is there, guys. You can take people who are totally evidence-based in everything they do, and you tell them about replication, and it’s more that they want to believe it doesn’t work. They hope it doesn’t work because they somehow have economic skin in the game.

But, like our friend in Canada who ambushed me on your podcast because it’s disruptive, it’s threatening, it’s change, it’s changes of fashion. So, part of what you’re seeing is that the allocators that you’re talking to, I think in general, they were trained that it shouldn’t work.

I gave a talk to, to a group of people, the CFA society in the US about this. It’s that these guys are trained that when we’re going to invest in this guy and we’re going to find the best guy in the space. And he’s got some special piece of information or way of looking at this or some judgment or something that he’s going to bring to the space. And he’s buying things, and the market’s wrong. You know, the market’s not seeing it and he’s going to see it earlier than everybody else.

And eventually the market’s going to figure out and be right about it. And the idea of trend following is that actually there’s no special information. You guys are all looking at the same return streams for these asset classes. In a sense, you get out of these things when the market slaps you in the face, usually.

The signal to get out is when you’ve already walked into a propeller on this thing or it just stopped working. And it’s just kind of been dead money for some period of time and the trades are in plain sight.

I talked to people about the positioning of our 10 simple futures contracts and they want to hear some insight like, wow, why are they long 5%, you know, crude oil or something? Because it’s been going up guys. There’s no story about it. And if it goes back, if it turns around and goes back, they’re going to be out in two weeks.

So, that’s the problem. Because what that does is it makes these guys incredibly anxious when they look at the space. Because then, also, you’ll start talking about futures contracts or, like Mount Lucas, who I’ve talked to a lot about this. They’ll talk about economic transfer of wealth from commercial players to speculators. And because we know futures contracts, and we talk about futures contracts. A futures role is not a scary thing for us. Futures notional amounts are not scary amounts long and short businesses aren’t scary amounts.

But for most allocators, used to talking about a guy buying a stock, as their anxiety goes up, they feel compelled to look, and look, and look, and look, and look, and look, to make sure there’s nothing that’s going to blow up on them that’s going to hurt them. And since there’s really nothing you can look at, on the investment side, to figure this out, they then focus on how many people do you have?

Look, it’s a challenge for us, but part of our strategy is to partner with larger firms to mitigate that. So yeah, we don’t have an army, but they do.

[00:43:09] Niels

Yeah, and of course also the… I guess so, I feel so, about the UCITS space. Maybe it would be the same with ETFs. I don’t know ETFs as well, but there are a few extra, independent checks and balances each day.

[00:43:23] Andrew

Yeah. I mean, you’re working with a more tightly controlled regulatory constraint in USITS land or ETF land, in the US. There, you generally don’t have long RSBs. People can get out tomorrow if they want to.

[00:43:41] Niels

Yeah, true. Yeah. Good stuff.

All right. So, we kind of jumped from topic number one to topic number three, which is great. They were kind of tied together, but so we’ve actually kept people in suspense for even longer than I thought because I promised a world premiere, something that they will not have found on any other podcast up until this day.

So, tell us, what’s this new, exciting thing.

[00: 44:09] Andrew

Well, it’s actually not even our announcement to make. It’s that SocGen has launched another index. So, to take a step back, anybody who’s less familiar with the space, SocGen has what I would call the two industry standard benchmarks for hedge funds.

There’s the SocGen CTA index. That is an index of the equally weighted net-of-fee returns of the 20 largest managed futures hedge funds. The index goes back to January 1, 2000.

And every January SocGen tells you who those 20 managers are going to be. And, generally, it’s the 20 largest managers in the space that are open to new investment. And when we went down the replication paths, we viewed this as the industry standard.

If you ask an institution, how did manage futures do last year? They’ll tell you, generally, whether this index went up or down. And the data is great. By hedge fund standards the data is great. It’s daily data. There is very, very little selection bias. You don’t have smoothing of returns that you get in certain categories.

Then there’s the second, which is a SocGen CTA Trend Index, which is 10 of those 20 that are heavy on trend, but it’s the same thing. Now, I don’t know exactly when this one was launched. It wasn’t launched in 2000. It was launched at some period of time afterwards.

The way you can think of the indices, the second one, the Trend Index, is like 1.3 times the first version with a 98% correlation. So, it’s a little bit more leverage, a little bit better returns, a little bit bigger drawdowns.

The issue that we’ve been wrestling with is when we talk how do we get more people in this space? One of the issues that I run into is that people want simple and straightforward. So, why would you have an index in the first place?

There were no hedge fund indices in the 2000s. And the funny thing is when I first took, started looking at replication, it was about replicating hedge fund indices. And I went to talk to some hedge fund luminaries.

I was sitting across from this legendary stock picker named Steve Mandel talking about replicating hedge funds and hedge fund indices. And he’s like, I’m sorry, what the f*@# is a hedge fund index? Why wouldn’t he know about it?

So, the hedge fund indices were part of the institutionalization of the space. And, and you want them for two reasons. The first is you have to decide how much do I want in this strategy versus everything else? And the way people decide that is that you look at a long-term track record. And this is called a capital markets assumption. And it is the foundation upon which you run your model to decide how much you want in all of your 10, or 20, or 40 or 50 different categories.

So, the index of a category provides the raw material, the data, over a long period of time. Because within that category, like the S&P 500, you wouldn’t say how much do I want in U.S. equities? Well, I’m going to use Nvidia or I’m going to use, I don’t know, some company that’s done terribly. You’re looking for a broad representation of the space. So very, very useful in that.

The second way that they’re useful is in manager evaluation. I want the best guys in the space. How do I get the best guys in the space? How do I know if he’s better than everybody else? Well, I can use it for benchmarking purposes and manager evaluation. And then often guys at pension plans, et cetera, will be judged on the basis of how well they do relative to the category. Are you actually picking guys from above average?

So, the two problems with both of them as an allocator is, first, you can’t actually invest in them. So, generally when you’re doing this, if you decide you like that category, your default option is to invest in the S&P 500.

So, the whole growth of this index, multi, multi, multi trillion dollar index business, people talk about it like, fee reduction and other things, and growth of ETFs, and other things like that. But a lot of what was really driving it was, if you’re the asset allocator and you decide that you want X exposure to emerging markets or X exposure to large cap stocks, when you pick the index product, you’re never wrong because there is no manager selection risk.

If large cap equities go down, and you’re in the S& P 500, you’re fine because you’re always right. The second issue with the indices is their after fees, which is outside of hedge fund land and private equity land that’s a weird thing. We don’t say, how did the MSCI world do over the past 20 years? But we throw in, you know, 80 basis points of fees for a long only manager.

So, I spent a couple of years trying to think about how to solve that. What we found is what SocGen has basically done. So, I think replication is the best way to do that. Because I mentioned that we’ve been doing this for eight years. We haven’t changed anything that we do. So, you can have an index like the S&P 500, but even the S&P 500 gets rebalanced. And you can have smart beta indices that get rebalanced more frequently, et cetera.

So, the idea of the replication strategies was to create the index solution for people who are interested in the managed future space, broadly. And that’s why we said, I’d call it index like, because it’s not exactly the SocGen CTA index, but it’s close.

And the way we built it was to aim for pre-fee and what we think are pre-trading cost returns. So, the whole idea of replication was to solve those issues, those two issues. What SocGen has done is we basically gave our replication methodology to SocGen and SocGen has launched something called the SocGen CTA investable hedge fund index.

So, it’s not the hedge fund index, but rather it’s a way that investors can now look at the space. And the index is again, it’s their calculations, et cetera, et cetera, but it goes back to 2002 and it’s now available on Bloomberg, and then will gradually be available in other places.

So again, this is SocGen’s index. It’s not a DBI index, but the underlying engine of it is replication. And it’s the replication model that we basically have contributed to them for this.

[00:50:59] Niels

So when they go back to 2002, of course, they’re doing a sophisticated “back test” of the replicator model all the way back.

[00:51:07] Andrew

Right. Okay. So, the great issue with any index, I wrote a paper called The Danger of Indices because in the whole alternative risk premia space people can be abusive in how they use indices. So, the whole alternative risk premia space was, was driven by investment banks. Look, you put a quant in the room and you say, look at value versus growth over time. They don’t come out with something that looks bad if it’s going to become a product.

So, what happens is people sit in the room and they turn dials. And maybe we should rebalance quarterly. Well, that looks better. Okay, maybe we should do this. Maybe we should redefine value. Maybe we should redefine growth. However they end up doing it, it’s a natural human thing to build models that have fantastically unrealistic returns.

And there is somebody who replicated the replicators, who after we had a discussion with them about this launched their own version. I don’t know exactly what they do, version of what we do, in an index form. And they launched in the beginning of this year, if you look back at their long-term track record, if you continue the meeting after looking at their long-term track record, caveat emptor, let me say.

You are dealing with guys who know nothing about the space, who are in the index business. So, what’s different about this is that it’s built off something that’s been working for eight years.

So, before that, yes, you’re right. It’s basically saying if we had figured this out in 2002, not in 2015, how would we have done over that period of time? Which, in itself, there’s always an inherent unrealism, but by index and backtesting standards it’s as clean as you get.

[00:53:05] Niels

Yeah, I agree with that. But of course, as I think you also pointed out in previous conversations, when new firms come out and they show you this shiny new thing, and it’s all a backtest, all I’m saying is that when they present this to the world, and I can’t wait to see the return since 2002, just people need to know what are the limitations?

[00: 53:33] Andrew

Again, this is SocGen. I mean, look, SocGen, they’re pros. And they do indices. So, they’ll have all these disclaimers. So, I talked to an RAA, in the US, who’s never invested in managed futures. And we manage an ETF, let’s leave it nameless because I, you know, I don’t think I’m supposed to talk too specifically about it, but they’re trying to figure out whether they want to invest in the space. So, the SocGen CTA Index hedge fund data, you used to be able to just pull it off Bloomberg. It’s difficult to get ahold of now.

Let’s say they get it. So, they can do the analysis that Man Group and Winton group and everybody else does. And they can look at the back over some long period of time and they can compare it to what they owned and everything else. And they’re going to want some. Now they have to implement it.

For us, I’m having a conversation with them and I’m saying, I’d like you to consider us as your choice. And they’re like, okay, but if you’re replicating this, why are you doing better again?

Why do you have a bigger drawdown for the peak than they do? How come you underperformed last year, but you’re underperforming so much? It feels very single manager like, which it’s not really. But (back to your point about replicators) they’re talking to one firm who does this for a living. And so, there is this feeling of idiosyncratic risk.

Now, the argument that I can make to them empirically, and most will believe it over time, is the idiosyncratic risk of what we’re doing, relative to this option and SocGen CTA Index, is far less than putting all of your eggs in the basket of say one mutual fund or one single manager ETF. It’s not black and white. It’s, shades of gray. And I think we’re lighter.

So, the value of the index for that guy is that he can pull the data on that index and decide if he wants it. If he wants it, his easy allocation is to buy a product that’s going to have very, very similar returns over time. And now he can use something that’s actionable to evaluate whether he likes another manager more.

That’s a big step that is yanking this business into the 2020s, where, from an allocation perspective, it’s really been stuck in the 2000s.

[00: 56:02] Niels

And will this investable index be made public on the website where you could download the data, do you think? You said it was hard to get hold of the data?

[00:56:11] Andrew

It’s going to be up to SocGen. The intention is for it to be publicly available for this reason. And again, from our perspective, one of my mentors is ‘grow the pie’.

A lot of allocators that I talked to, on the institutional side, don’t pay full fees. So, they beat the SocGen CTA index. They diversify. They beat the SocGen CTA index, in a sense like what we’re trying to do synthetically through fee arbitrage.

And so, if you look at the long-term returns of something that’s hundreds of basis points better because the fee disintermediation has already been baked into it, you start to see higher allocations. You don’t go through that period of late 2010s. You don’t go through that dead money period of late 2010s. You’re still making some money above cash.

So again, my goal in all of this is the $300 billion that’s in this industry, to me, is woefully low relative to the diversification benefits. Part of that is just again, you go outside this.

Remember the mutual fund space, which has been around for 15 years, it’s $20 billion. It’s not even a footnote in a footnote of the asset management industry, the ETF world, I mean, you know, Corey Hoffstein is having tremendous success and he’s going to close in on a billion dollars soon in his return stacked products.

But a billion dollars out of $10 trillion means he is the shining light of bringing a valuable product to market and has one basis point of the space that’s growing.

So, these are all just different steps, and I deal in a different world because I’m dealing with how do I convince people who’ve never invested in the space to come into it? And this is something that I think will be very helpful for them.

[00:58:13] Niels

Yeah. I remember when I started in this industry, I was very excited about the growth potential because back then the industry as a whole was less than 10 billion. At least now we’re up to 400 billion.

But there’s still, as you say, a very long way to go. Even though let’s hope not everybody goes into trend following and CTAs, that might make it more difficult for us in some ways.

[00: 58:35] Andrew

For the first time, the media will be right. That we’ll finally be moving markets for the first time.

[00: 58:40] Niels

Yeah, exactly. That’s true. This was great.

I think we’re going to keep the last topic because we’re kind of on the hour now. So, we’ll keep the last topic. It’s a topic we’ve discussed before about sizing the CTA space, based on a paper that Winton wrote.

So, people should go and get the paper from, from Winton and maybe we’ll pick it up on the podcast at a later date. But this was great. This was wonderful, Andrew.

I really, really do appreciate the fact that we can have these candid conversations to where we don’t always see eye to eye on things, but we see eye to eye on the big picture, of course. So, this is wonderful. And thanks for bringing some world breaking news to the podcast.

Maybe I should do the thumbnail in yellow, like breaking. I don’t know, we’ll see.

[00: 59:26] Andrew

Scrolling across the screen.

[00: 59:28] Niels

Yeah, exactly. Absolutely. So, if anyone out there listens to this conversation and think this is indeed great, and I appreciate everything that Andrew does for this, why don’t you go and leave a rating and review on your favorite podcast player so that more people can discover and listen to conversations with Andrew like this. On this note, we’re going to wrap up our conversation.

Next week I’m going to be joined by Rob. He’s back from his extended summer holiday, as he always has. I’m sure it’ll be fun, and therefore you should send in some questions for the upcoming episode. You can do that as always to info@ toptradersunplugged.com.

That’s it for now. From Andrew and me. Thanks ever so much for listening. We look forward to being back with you next week. And in the meantime, take care of yourself and take care of each other.

[01:00:43] Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released. We have some amazing guests lined up for you.

And to ensure our show continues to grow, please leave us an honest rating and review in iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.