Shownotes
As equity markets grind higher and trend strategies navigate sharp reversals, Moritz Siebert welcomes Nick Baltas of Goldman Sachs for a conversation that moves beyond performance to examine structure. Together they unpack the machinery of the $1.3 trillion QIS industry - from index design and client behavior to the subtle forces shaping capacity and crowding. They discuss how trading speed has become a key axis of dispersion, why volatility remains the hidden cost in systematic portfolios, and what resilience in markets might really be masking. This is not just about strategy. It’s about how products scale, and how ideas hold.
-----
50 YEARS OF TREND FOLLOWING BOOK AND BEHIND-THE-SCENES VIDEO FOR ACCREDITED INVESTORS - CLICK HERE
-----
Follow Niels on Twitter, LinkedIn, YouTube or via the TTU website.
IT’s TRUE ? – most CIO’s read 50+ books each year – get your FREE copy of the Ultimate Guide to the Best Investment Books ever written here.
And you can get a free copy of my latest book “Ten Reasons to Add Trend Following to Your Portfolio” here.
Learn more about the Trend Barometer here.
Send your questions to info@toptradersunplugged.com
And please share this episode with a like-minded friend and leave an honest Rating & Review on iTunes or Spotify so more people can discover the podcast.
Follow Nick on Twitter.
Follow Moritz on LinkedIn.
Episode TimeStamps:
00:23 - Moritz opens the show and introduces Nick
01:16 - Nick’s quick life update and setting the tone
02:45 - Market resilience vs. fragility in 2025
04:18 - Performance rundown: CTAs, trend, equities, bonds
06:10 - October reversals: metals and livestock giveback
07:32 - What’s working: equities, gold, copper; sugar shorts
08:58 - Trend speed, April V-shape, and dispersion
10:40 - Position exits, re-entries, and neutral zones
11:55 - How QIS differs and why it’s opaque from the outside
14:40 - How big is QIS? Asset class split and caveats
18:05 - Who uses QIS: from asset owners to hedge funds
21:30 - What QIS really provides: rules, access, and efficiency
28:05 - The four pillars of product design (research, clients, tech, markets)
30:20 - Why thousands of indices exist (customization, wrappers)
33:22 - Delta-1 vs. options on indices: when each makes sense
36:58 - Vol targeting trade-offs for optionality and recovery
38:55 - Built for scale, but with capacity discipline
42:00 - Innovating without data-mining traps
47:12 - Crowding: congestion, vol carry, and trend mechanics
56:35 - Wrap-up and what to expect next week
Copyright © 2025 – CMC AG – All Rights Reserved
----
PLUS: Whenever you're ready... here are 3 ways I can help you in your investment Journey:
1. eBooks that cover key topics that you need to know about
In my eBooks, I put together some key discoveries and things I have learnt during the more than 3 decades I have worked in the Trend Following industry, which I hope you will find useful. Click Here
2. Daily Trend Barometer and Market Score
One of the things I’m really proud of, is the fact that I have managed to published the Trend Barometer and Market Score each day for more than a decade...as these tools are really good at describing the environment for trend following managers as well as giving insights into the general positioning of a trend following strategy! Click Here
3. Other Resources that can help you
And if you are hungry for more useful resources from the trend following world...check out some precious resources that I have found over the years to be really valuable. Click Here
Transcripts
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.
Moritz:Hello everyone and welcome or welcome back to the Systematic Investor series on Top Traders Unplugged, where each week we have a look at the global markets through the lens of a rules-based in investor.
My name is Moritz Seibert and today I'm standing in for my friend Niels Kaastrup-Larsen, who would normally run this show, but at this very moment he's in route to Canada and therefore unable to get in front of a mic. Joining me for this episode is Nick Baltas from Goldman Sachs, a regular guest on this show, as many of you will surely know.
Today, Nick and I will focus on the QIS space at large, discussing questions such as how large is the global QIS business? Is it growing? Who are the main clients? How many indices are out there? What's the outlook for the space, and so on?
So, with that as an intro, let me say hello to Nick and welcome him to the show.
Hi, Nick.
Nick:Hi, Moritz. Very good to be here.
Moritz:Good to have you here. How have you been doing recently?
Nick:I've been doing well. It's been a busy period, I should say, and I spent a good amount of time with family over the last couple of days. I had my sons christening, so I had family and friends flying over from Greece. It was a nice occasion. So, that was like a small kind of break. But now I’m back in action, but I'm doing very well.
Moritz:Congratulations on that christening.
Nick:Thank you so much.
Moritz:Anything else that's been top of your mind, on your radar? It's the question that Niels likes to bring up, market wise. Yesterday we've had a Fed decision. Maybe that has created some turbulence, I don't know.
Nick:Yeah, I think what is on my mind – on the top of my mind, I would say this year has been pretty much like a bit of a debate between fragility and resilience. You kind of have events that you feel will turn against the overall market and the positive sentiment and the growth and then you have like no results, and so on, and so forth, kind of going against it. And the market seems to be quite resilient in terms of kind of absorbing this information and kind of consuming it.
So, I would say if there's one thing that, you know, keeps me still awake and maybe kind of on my feet, as the days and months go past and we're kind of approaching November, it's this kind of pendulum between a fragile market versus a resilient one.
I mean, I cannot call the shots, but you know, if this is one thing that, you know, probably entering the last phase of the year will be a memory of the year, that would be the one.
Moritz:Yes, I agree we cannot call the shots, and you know, personally I'm just, you know… Well, we're trading in a rules-based way, so we don't really care about equity valuations or anything like that. But to see equities performing so strongly again this year is really surprising. It's just nobody's able to beat the S&P 500. The Sharpe ratio of the S&P 500 has now decoupled from everything else. It's the best investment in the world.
Until it isn't.
Nick:That is true. Until it isn't.
Moritz:Until it isn't.
Nick:Fair, exactly.
Moritz:Yeah, so, maybe with that as a backdrop, it's a good opportunity to get through a bit of a performance update. These are numbers as of Wednesday evening, October 29th, so, month-to-date and year-to-date. The BTOP50 index 1.89% up month-to-date, and now positive 2.35% year-to-date. SocGen CTA 1.7% for the month up, and just minus 1% for the year. So, that is a good recovery, now in October. The Trend index plus 2.15% for this month and around flat for the year. The Short-Term Traders index from SocGen is up about 0.50% this month, and still down quite a bit - minus 4.6% for the year.
And as I've just mentioned with the equities, the MSCI World Index plus 2.8% this month, plus 22% this year. Aggregate Bond index plus 1.2% percent this month, plus 7.2% this year. And the S&P 500, the darling of everyone it seems, plus 3% yet again in October, and plus 18% and a bit this year. Isn't this amazing?
Who beats the equities? It's just tough, huh?
Nick: n that. I think, what was it,:So, to my earlier point I think we've been seeing this debate between, is the market going to continue going up in the way that it's doing it? And I think maybe the fixed income market is a bit more debated these days and obviously the place that bonds would hold in asset allocation portfolios. But I think when it comes to equities, they keep on surprising us in a way.
Moritz:Yeah, I mean it's kind of like you say, the equity risk premium where you expect to make 6%, 7%, 8%, something like that percent per year, that's like the long, long, long term historical average. But for the past five years that's not true. The past five years it's kind of like every year is between plus 15% and plus 20%.
Nick: Barring: Moritz:Yeah, I don't remember what… So, the COVID period you mean?
Nick: No, no,: Moritz: e S&P have a negative year in: Nick:I would think so, actually, quite negative, yeah.
Moritz:Okay. Okay. So maybe that's the exception.
But if they're up, if the equities are up, in my recent memory, it's always kind of like more than 15% in a year.
Well, that is what it is. In terms of trend following, because that is a topic that's near and dear to our hearts, looking at October, it's been a great start to October. I mean I don't know the books of other CTAs but, as far as we're concerned, we've had a great time until around the mid-October, or the 20th of October, something like that, with long positions in precious metals, and long positions in livestock. It's just been a great month. And then a lot of give-back of open trade profits on exactly these markets; Palladium, gold, silver, platinum, you know, feeder cattle, live cattle, livestock, lean hogs. Those markets have been reversing and it's kind of like back to around flat to be quite honest.
So, really just you turning their inconvenient give back of these open trade profits. But you know that is, it is what it is.
Nick:I'm seeing similar things. On our side, it's kind of maybe still on the positive side for the month. Um, to your point, most of the return was driven initially by commodities and now they gave pretty much everything back. Equities are still kind of holding up there, and rates keep on being the same story for the year; keep on whipsawing, keep on losing on the trend side. And main drivers, yes, are pretty much equities, and then, I would say, some of the kind of long exposures in, in gold, aluminum, copper and good short on the sugar, kind of delivering some good performance so far.
Moritz:Yeah, sugar is a recent short entry that just continues to go down. Still, some of the grain markets still look relatively weak even though they've recovered a bit from their weakness in the past, say, 10 days or so. But I agree with you, the big whipsaw this year is the fixed income markets. It's on and off all the time.
Nick:We were doing like an attribution analysis, somewhat like a straight line down.
Moritz:I’m not surprised. How's it looking for currencies for you guys when you do the attribution?
Nick:So, I would say for the month it’s just flat pretty much, with the biggest kind of positive contributors being JPY and maybe on the other side is Aussie dollar and GBP. Year-to-date it's kind of the second worst past rates. You know, if there's one asset class that over the year has actually contributed positive, it's equities. No surprise again, right? No surprise.
And obviously the question then is how quick or slow the speed is to be able to sustain an exposure through April. So, get the recovery and then obviously work from then on or substantially degross at that point in time, and then missing some of the recovery and therefore building up again the exposure would take some time.
So, it's a question of speed and volatility response are on the V shape. And I think that's the primary driver I would want to believe. Not that I have too much data to support it, but I think that's the main driver of dispersion that we've seen this year in the trend following space.
Moritz:I think that's true. It's the liberation day or the early April period sell-off. In our case it's not just been the equities. We got kicked out of some of the equities, but we also got kicked out of some of the currency positions that we had. And there were just a lot of positions, that we had on the books, that went to nothing or even reversed at that point in time.
And, like you say, it's a function most likely of trading speed. Do you get back into these markets? Did you lose your position or not? And that very easily can make a difference of 10% for this year.
Nick:I guess, quite interestingly, if you were to be very quick, you'd kind of capture some of the V shape at the time, but then the subsequent part of the year wouldn't work out in your favor. So, I think at the time we're just talking about the short-term traders actually doing well in the V shape. And that's I think part of the reason that you capture some of the first drawdown in deliberation day and then some of the recovery around the tariff pause.
If you're slower, you'll probably just stomach the whole V shape, and then recover with it, and then performing to this day, to your earlier numbers, basically capturing the equity trends. And I think there's some bitter spot, in between being too fast and too slow, whereby you just deliver, at the wrong time, that the market's eventually recovering. And it actually gave you a week to also crystallize the deleveraging of those exposures.
Moritz:There we go. Yes. If we could only know these types of things before, but we don't. So, we have to stick with our own systems and some of them are longer-term and some of them are medium or short-term, and that's what we do. In our case there are more longer-term but still we did get kicked out of some of these positions and then it really takes a while to get back on.
I think there were just two or three equity markets, if I remember that correctly, where we didn't lose our position. I think one was the Hang Seng, I think the share price index, the Aussie index, and the South African equity index. Those three we kind of like capped, but all the other ones; TOPIX, S&P 500, NASDAQ, go down the list, all the European indices, we just lost all our own positions in April. And then it probably took until, I don't remember, July, August, something like that to get back in. Yeah, quite some time.
Nick:And that's basically I guess some sort of threshold based criterion to keep or not the position. Because we have zero, flat zero.
Moritz:Yeah. We have our exits, you know, different types of exits. Some are trailing stops, some have other rules. But we're not forcing ourselves to reverse the position into shorts. There is a kind of like neutral segment or area, and that's what we've hit.
Nick:No trade zone.
Moritz:Exactly, exactly. So QIS, look, this is an interesting business. It's also, in a way, I would say, a little bit opaque to the outside observer - like to most people, really. They know QIS means Quantitative Investment Strategies, it's a business and activity that pretty much all of the larger investment banks engage in these days. It's you know, focused on risk premium indices, advanced beta type of strategies, that type of stuff, designed and then sold to clients.
But it's kind of difficult for people to see how large it is, who the participants are, who the clients are, how these deals are transacted, how the exposure is transferred from a bank to a client, and these types of things. So, I thought with you on the show today, that would be a good topic to just speak about the QIS business, more from a macro perspective, at large and chat about how large it is.
rm, in the first half year of:So, asset managers also have quantitative investment strategies. Some of you folks may know some risk premia UCITS funds or, you know, there are some asset management firms specializing in risk premia strategies that would kind of count towards that that 50%. But it's a relatively, I would say, large number, US$1,300 billion. And the largest part of that is equities.
Equities is more than half of all the exposure, of all the strategies that are linked to equity type of strategies, equity long/short factor type of exposures, followed by multi asset class strategies, then commodities and quite surprisingly I was thinking that fixed income would have a much bigger footprint. But fixed income is only US$50 billion. Even though fixed income markets are gigantically large markets. But in QIS space, on a relative basis, it kind of seems to me that they're underrepresented, as is FX with only US$27 billion, and credit with US$13 billion. So that is a backdrop. Do you think these numbers make sense to you, Nick?
Nick:Yeah. So, look, do the numbers make sense?
There's a reporting convention here that I believe different banks are utilizing, and maybe asset managers. I think on the asset manager side it’s a bit easier because that's the actual funded exposure. When you look into broker dealers, the way that QIS products are distributed is either via kind of unfunded swaps or via sometimes call options.
So, the way that we can think of assets here can be debated. Do you do a look through leverage adjustment? Do you do delta adjustment, if that is an option? There are different conventions. It is unclear to me whether those conventions are properly utilized and uniformly applied across the reporting banks.
But I guess, in the grand scheme of things, if you just take at the minimum the broker dealer sum, which is close to the US$600 billion mark, I would say with some confidence of plus or minus 100 (I'm just making it up really. I don't really have hard data to support it), I think it's in the right ballpark.
Now, if you take asset managers into the mix, maybe there is some double counting because obviously asset managers are very active in the QIS space and some of them do utilize QIS products. So, in the context of reporting, you might end up having both broker dealers reporting numbers for specific strategies which are eventually deployed by asset managers that also report numbers.
we can identify, maybe around:To your question on the asset class split, it's interesting to see that the equity long-only space is by far dominating the asset manager world. And that's more like the smart beta of the world. So, anything that is quant enhanced and delivers an equity exposure that is supposed to outperform a typical benchmark would come into that pocket.
It's less so on the bank side. I think the bank, or the broker dealer asset class contribution is more evenly without that being even, but more evenly split between the asset classes. I would say part of the reason why you see less on the fixed income and FX side is that historically liquidity has improved.
If you move away from the Delta 1 space on the option side, which is typically the one that banks are much more focused on, is something that grew over the years. I think the equity space is more natural. That's the equity long/short and that's basically the genesis of the QIS businesses. So, it has grown over the years. I think commodities is the other one.
You go back to those rolling futures and how you can indexify commodity exposure, both on the beta side as well as on the long/short side multi asset is again kind of a CTA type of a place. There is also a good amount of application in the retail space when it comes to multi asset.
I think FX, fixed income are in their growth phase, if we were to say. And even if we look into kind of academic activity, it has been massively more on the equity side and less so on the other asset class.
So, I think it's a combination, I guess, of features of the market. But I would certainly flag that all asset classes are now becoming part of the toolkit of broker dealers.
Moritz:I also now see a lot of multi QIS offerings in the way that banks would combine say their commodity carry strategy with their FX carry strategy and an equity carry strategy in order to create a cross asset carry basket that's also multi asset. But back to the numbers. I think just to put this in perspective, the numbers could be conservative, they could also be aggressive.
There could be funds out there, asset managers, maybe hedge funds, just people trading risk premium strategies that are not reporting to that survey. Think about a hedge fund, maybe that hedge fund is not covered by that survey and they may run whatever, like a few hundred million in notion of exposure in risk premium type of strategies, but they're not factored into this survey.
And then it could also be, on the other side, like you say, there could be double counting which could happen if, for instance, you swap exposure of a strategy to an asset management firm and the asset management firm then on sells it to their client. You report it and the asset management firm reports it.
And then what you've mentioned, the delta adjustment or the leverage adjustment is, you know, if you're selling a call option linked exposure to one of these strategies, it may have an initial delta of 0.5. But the question is, do you report that delta adjusted exposure or do you report the full notional of the option to the survey?
So, let's take these numbers with a pinch of salt. But the, I guess, overarching statement here is that that space is, I would say, pretty large now and it's growing. And there's a chart included in the survey where essentially these assets, with the exception of what you've called like one or two, three winter quartiles, like three months or so, but they're essentially going up.
Nick:Yeah, I mean that's a fair statement. It is a space that has been evolving in a variety of ways. Maybe it's asset class allocations, maybe it's new markets that are traded, but also users on the client side have kind of increased over the years. You know, we can also discuss that part.
Maybe the one thing that I would flag, the one last thing that I would flag and then happy to go to the client side, is that there is also an element of substitution, to your point. So, if today there is a, I don't know, US$100 million exposure in a CTA that is now transferred into a QIS product, the market still consumes the same level of liquidity demand. It's just now marked under the QIS kind of badge and therefore would be seen as an increase in the multi asset pocket of a survey as such, when in reality just a substitution effect from what otherwise was like a funded exposure on a CTA. I'm just making that example whereby a substitution effect could also be present here, more so than just genuine growth - another asterisk for you.
Moritz:Yes, yes, there are many asterisks. You've mentioned clients, Nick, and I think you've mentioned that there's new clients coming in. So maybe let's talk about that. Who's interested in this? Who is your usual client and how has the spectrum of clients changed over the year?
Nick:Yes, so the QIS space started maybe 15 years ago. We can point in a timeline as to where was the first, maybe, QIS trade. I would think of it either coming from the commodity space and those rolling futures indices that were built to deliver kind of economic exposure into commodities without having to hold physical commodities. So, that was one reason why those QIS teams initially were formed. And then the other space or the other place, if you like, that we saw origination of ideas and indices was post GFC, and some of the work done by some of the Nordic pensions to replace or maybe enhance their risk profile by risk factors, and kind of transitioning or repurposing how they think about diversification away from asset class mix into kind of a factor mix. And that was the time that some of the banks started working on equity market neutral strategies, some volatility carry strategies, you know, version 1 at the time.
And, in this regard, the primary users in the early days were either asset owners… So, we can talk about large pension funds, sovereign wealth funds across the globe. You know, there are some regional kind of shifts, that happened over the years, and happy to go through that. And then asset managers, these are the primary users.
If you do kind of a fast forward to today, you can widen the spectrum quite substantially and you can go anywhere from private banks to hedge funds. So, the utilization of the technology and the utilization of this IP has massively expanded across the segment.
And one could think of a QIS as offering nothing more than, I guess in a good way, kind of a convenience store in the sense that there are various type of systematic exposures that can deliver performance, or various ways of delivering a defensive profile via, for example, rolling put options and being very thoughtful on which tenors and which strikes to utilize, whether the delta is hedged or not. All I'm just discussing is nothing more than a set of rules.
It is, to a certain extent, not too dissimilar to saying, look, if you really want to have the equity market, you need to find the 500 larger names and you need on a quarterly basis to reconstitute what those 500 names would be and allocate market cap weights to them. This is nothing more than a set of rules. And this is called S&P 500.
And once upon a time there came a contract, that is called the futures contract, that delivers an unfunded exposure to the equator’s premium. So, that in itself facilitated whoever wanted to get equity exposure, to avoid going to the market and physically replicate 500 names into a portfolio of what the S&P will represent, and basically get on a contract nothing more than the performance of that construct that, in itself, can be put into a document and can be reflected into an index profile.
So, if today we claim that volatility is rewarded and whoever is willing to sell insurance should be compensated, that in itself can be described with a set of rules by selling some options and trying to reduce the spot participation, the so called delta hedging, we can go back to the academic roots of the volatility risk premium. If somebody were to deliver ‘a futures contract’, in this regard that would be an OTC swap, then this allows the end investor to be exposed to the volatility risk premium not too dissimilar to being exposed to the equator's premium. And what sits underlying that investment is nothing more than a rules-based index.
So, the way that I think about the QIS space is that it delivers different risk profiles in an accessible format and maybe reduces the operational burden and delivers this operational ease to the end investor. And there are other benefits, you know, cash efficiency, and so on, and so forth, that we can also discuss. But the primary reason why we now see this proliferation of this type of investing, nobody suddenly thought that systematic is the way.
But systematic is the way to deliver into an index format what otherwise could be a cumbersome and tedious process. And the end allocator, being a private bank or being an asset manager or a hedge fund, would simply have that exposure and utilize their internal models on timing, utilize their internal models on sizing, focus more on kind of bringing their alpha, which is more of a timing element rather than of an implementation element.
So that's a quick kind of… or maybe not as quick, but that's, I think, a quick overview as to how I think the space is evolving and the utilization across different client segments.
Moritz:Very good. And the observation that I personally made and maybe also many others, is that tying into what you said? You said, it's a technology. I think that's a good term to use and that comes along with operational efficiency.
So, when you think about the early days of QIS, as you've mentioned, there's, you know, asset owners or insurance companies or large asset managers contracting with you. And maybe they wanted to have a new variation of equity value. Maybe it's something that they're doing in-house, and maybe it's just they want to have another source to diversify, and do something like that.
But it's not like high frequency, you know, it's kind of like, oh, you're running this daily or weekly or monthly or something like that. But then that has changed when most banks came up with, hey, we now offer intra-day vol or we offer intraday momentum trading, which is technology, which is an operational thing that not everybody can work with.
And I guess that's where the hedge fund clients come in, where they go like, well, look, if the bank can, every five minutes or whatever is, do a vol carry intra-day strategy on the S&P 500, then I might be better off just buying it from them as opposed to implementing that internally, myself, because it's difficult to do that.
Nick:And you're right in this one. There's an element of IP that comes with the idea of designing a strategy that delivers some return profile that can be alpha enhancing, that can be defensive. But to your point, there is a massive, now, integration of technology.
The way I think about the QIS space, over the years, is as a combination of four main pillars in terms of product development. There's certainly research, industry, academic research that is evolving, you know, how we think of asset prices. And that is definitely a source of inspiration.
Number two is the entire client segment in terms of appetite. You know, it used to be either a hedge fund replacement or complement and it eventually became more of a specific economic outcome focus. I want something that is a bit more defensive and I'm happy to take on the negative cost of care. Or maybe I want something that is enhancing my yield, and I should be very focused on what is a spot contribution. Maybe now somebody's focusing on inflation, maybe the economic regime. So, there are different ways of us delivering product for specific economic outcomes. So that's the second pillar, how the client utilization of the product has shifted.
I think the third one is technology. (There is no order by the way. These are just equally sharing the load of inspiration.) So, technology, to your point, being able to access markets, being able to trade markets, being able to trade faster markets, this is a massive enhancement that we have in our product offering. And then I would say the fourth is how markets evolve. Once upon a time you could not trade some of the frontier commodities. Now you can. Once upon a time there were no short aid options. Now there are.
So, there's a variety of things that eventually happen around us allowing us to almost do out of sample testing on universes that otherwise would not be able to do so, at least back in the day, and not net of costs.
So, I think if you were to ask me, these are the four pillars that over the years have been contributing to product: research, technology, client utilization, and market liquidity evolution.
Moritz:Got it. What do you think Nick, guesstimate, how many different QIS indices are out there, or not different, total number of QIS indices?
NIck:Across all banks?
Moritz:Or maybe start with you.
NIck:I mean certainly thousands.
Moritz:Yeah, it's thousands. It's thousands.
Nick:Well, I mean, we can discuss the reason why that can be the case. And there is certainly an element of customization. So, for example, you mentioned equity value. You know, somebody can say look, I don't know what's the best equity value definition? Maybe it's, you know, book to price, maybe it's dividend yield, maybe it's a combination.
So, there could be variations of the same product purely to reflect that different design choices could suit different investor needs. We can speak about CTAs, you know, if somebody would be willing to avoid having equity participation on the upside purely because they want to have more of a defensive profile. That's a different index. So, you know, the fact that we don't, and that a bank cannot act as an asset manager, and there is no fiduciary element, that in itself leads to the need of having a pallet of possibilities.
And we can talk about an equity investment that is seen from a non-US investor that is exposed to withholding tax. So, that has to be a net total return implementation versus a gross total return. So that in itself is two indices.
So I think, because there's a lot of discussion as to how many indices, and data mining buys, and so on, and so forth, I think there's an aspect here that should be definitely mentioned, and that is purely the fact that any different implementation bid, for a specific client-ask or a specific need, has to be a different index. A client doesn't want to have ags in their commodities portfolio; that's a different index.
So, there's a lot of customization that is purely driven by a specific need and less so of, “Oh, we found a better way of doing it. Let me just replace it now because I've just done my overfitting exercise.” This in itself has to be addressed, but this is, I guess, one way to address the magnitude.
Moritz:Why there are so many, and they all have their tickers, and they all run on the Bloomberg terminal, I presume. And you put them into Python code or whatever it is to run them. I guess, now, these days, in Europe you need an independent calculation agent - benchmark regulation.
Moritz:You do you have that internally, like Chinese Vault, or do you use external parties?
Nick:There's a variety of models, as you say. You know, some banks utilize internal calculation agents that, obviously, are fenced. So, there's a lot of governance around this business.
Some others would be using external calculation agents. This is purely a business decision as long as, obviously, the governance is in place and the business overall is well run. But it's certainly the case.
Moritz:Looking at the types of exposures or products that your clients typically request, is most of the notional transacted in Delta 1 space, linked to a swap or node, or is most of it kind of optionalized and therefore nonlinear index exposure, like somebody buying a call option on the index?
Nick:Yeah, so, historically the vast majority has always been in kind of the Delta 1 space in the sense that they would do some form of a Delta 1 wrapper that can be typically a swap. That's what you see in the space.
Obviously, there are other wrappers like certificates, notes, depending on the contractual relationship that a bank would have with a specific institution. More recently, and certainly for the insurance space, but also more recently we've seen, in the institutional space, some interest for option profiles.
And just to be clear for the audience, we're not talking about the constituents of an index. The constituents can be Delta 1 or vol. You know, it can be a volatility carry, it can be a trend following strategy. We're just talking about their wrapper. So, your question is really how can I get access to a volatility carry trade, or how can I get access to a CTA profile?
This is typically done, as I mentioned on a Delta 1 type of profile. But recently… And when I say recently, maybe the last two, three years, that we've seen a bit more interest in fixing a price and basically capping the loss; so, getting a call option on some of those indices. I would say it's a fraction, it's a small fraction, at least, of the executed trades.
There is a philosophical question here, I guess, to be put in place. Let's say you're selling volatility to get a volatility carry profile and then you want to put an option on that profile. It's almost going in circles, right?
So, the mere reason why you're supposed to be benefiting from a premium, when you sell options, is because you're willing to take on the risk. This thing is going to, at times, occasionally experience a vol spike. So, almost writing a call option to it, in itself, nullifies the existence of the premium in the first place.
So, why would you have to get optionality on something that is dropping and still benefit from an upside, while the price for it should be the exact premium that you're benefiting from selling it? So, almost this viscous circle nullifies the value of pricing a call option on some of those negatively skewed profiles.
There are still ways that we can think around that. But I guess the high-level answer to your question is that it's primarily on the Delta 1 space. And when it comes to an option, there are types of transactions, maybe more on the underlying Delta 1 space and not on the option space. But rarely do we see that as frequently as we do the Delta 1 transaction.
Moritz:When you do the Delta 1 transaction, it removes the requirement or the kind of motivation for you guys to have a vol target or vol cap. I mean, it may still have this as an intrinsic part of the strategy element, but I reckon, when you sell a call option, the underlying index will definitely be vol controlled or vol targeted to a certain type of level.
So, there's a lot of that stuff that kind of slacks through the market as well. Because, you know, when somebody buys that call option, really the only party that will quote a price is you. They can't go to a… Well, they could potentially go to a different bank, but they're not going to get a good bid ask, if anything.
So, you know, if you’re the source of liquidity and therefore you're protecting, I guess, your book by having this vol control in place and selling the option at a premium, that's amenable.
Nick:That is fair. That is fair. It's certainly easier if you think about linear structures underlying it. So, when we talk about equity momentum, equity momentum, in itself, is vol targeted, and that in itself is helping equity momentum because negative skew and equity momentum can be moderated by vol targeting. So, that in itself, to your point, almost comes not even for free, but for a benefit of the performance while facilitating some option pricing to operate on top.
So, I guess another way of kind of answering your question is that there are pockets of a QIS offering that are more accommodative for option pricing and some others that can become a bit more nuanced purely by the, I guess, the economic principle around the premium, but also mechanically. If you were to volatility target a strategy that is supposed to be reacting and recovering post the volatility spike, you're kind of acting against its ability to recover.
So, then the whole question is, okay, fine, we can do the vol targeting, but then what you're ending up with is a suboptimal profile for the respective strategy. So, all these are important considerations.
Moritz:What do you think? By and large, would you build your business or the QIS business for massive scale, like an ETF provider would, where practically, theoretically, say, the AUM of an ETF are unlimited. It can grow to whatever size it wants to grow. Do you have that same thinking in the QIS space?
Because one thing that I recognize is that, for instance, the CTA or trend offerings or some of the more popular offerings, they're not exposed to some of the smaller markets. They're really lacking from these type of indices that would more like be the super deep and liquid markets. Is that a fair statement, that you're building it for scale?
Nick:Generally we build product for scale because that, in itself, is associated with how we can see the profitability of this business. At the same time, we are always exploring what are the pockets of the markets that can deliver some good performance, even if the scale cannot be at sizes of a much more liquid profile.
So, we can spend time talking about frontier commodities or some of the non kind of benchmark commodities. Obviously, they don't have the liquidity that you'd expect to see in some of the larger markets. But certainly there could be variations, to your earlier point, there could be another index that simply has a broader set of assets as underlying. With the, I guess, consequence being that the overall capacity that this profile can support is now a fraction of what a subset of assets could.
So that is, I guess, to answer your question in a different way, we try to solve for different scales. And if there is an argument to be had with regards to premia being more pronounced with lower liquidity or pockets of lower liquidity, if there is a liquidity premium to be harvested or illiquidity or whatever, it's worth it.
We're always keen to see what the enhancement can be, with the asterisk being that we cannot scale it as much as we could with a very liquid universe. So, we don't shy away from exploring those pockets of liquidity. But it's probably the most important thing when we design strategies, being very, very thoughtful on capacity. So, to put it into an extreme kind of statement, nothing is built for infinite scale, to be clear. Everything has to be done in a very prudent manner, very conscious of market liquidity, how much we transact, how much we consume, how much we roll, all that lot feeds into the product design.
Moritz:And that's also because you have the option to, at some point say, stop. If you wanted to stop selling an index, you could close it, whereas an ETF cannot. Right?
Nick:And we do. I mean, at the end of the day, you need to protect the investors and you need to protect the market and you need to act in a prudent fashion when you transact and interact with the market. So, there are multiple cases whereby a specific design cannot sustain anymore without being market impactful. So, we're like, that's it for now.
And I think investors do appreciate that there is this scrutiny and there is this transparency. And existing investors also see that as a good thing, in a way, that we are basically the guardians of their exposures.
Moritz:Maybe two more things, one, this is just an observation. Every couple of weeks I receive emails from, it's not just banks, it's all sorts of like people offering product in that space and they come up with something new, something that here, therefore, has not, in their view, existed. It's kind of like, here's a new index, here's a new strategy or a new variation of a strategy.
And I sometimes go like… It's kind of like the same that you've had before, just a little bit of a new twist to it. But other than that, it's pretty much the same thing. So how do you balance this kind of like being forced to innovate and coming up with new indices, maybe in an approach or in an effort to stay relevant versus just over optimizing things and churning out new product for the sake of producing new products?
Nick:Yeah, that's an amazing question actually. Why it's an amazing question is because it's part of human nature to observe, and then think, and then try to do better, specifically in that space. So, if I observe my trend following strategy in April, then probably there is a temptation to say, oh, if only we're a bit slower or if only we're a bit, I don't know, overweight this asset class or the other asset class, or if only we're not doing dynamic sizing or we're doing like not a static size, whatever. So, this temptation exists and, I think in itself, it's quite inherent to have those biases of overfitting.
So maybe another way of asking the same question is to say, how do you control data mining in a way? Because frankly, if there's a new idea, this idea is probably driven by the fact that something must probably have worked better and, therefore, let me just launch this tweak because that tweak will just make the recent performance look better.
So how do you control for it? I mean, my personal view is that this is genuinely product culture. There's no better way to be conscious of data mining unless you just call it out. So, it is part of our, at least as far as I see it and given the responsibility I have in the product teams, that I would flag, look, now we're data mining.
Fine, let's look at it, let's get a better sense of what the results suggest. But we should be very conscious of these data mining biases. And as soon as you acknowledge them, then it makes, I think, your brain iterate in a very different fashion because then you see all those new kind of very skilled individuals, that get hired from university, being exposed to those biases. But as soon as you call them out, then they start ingraining a culture whereby any new tweak they do is not about making the Sharpe ratio look better anymore.
So, I don't think there's a right answer to your question in the sense, okay, that's the blueprint. I think it's a combination of acknowledgment, a combination of calling it out, a combination of being very thoughtful and studying some of the work that has happened in the academic space to control for determining biases. I think there is no way you can avoid data mining.
If I were to show you 10 back tests, you would basically drop the bottom one just purely because you wouldn't think about it. So, when it comes to changing a strategy or changing a design, what we try to strive for is (I tend to say that quite often) that explaining the underperformance is more important than, you know, being lucky on the upside. We would love to outperform, but if we cannot explain underperformance, that's the worst thing that can happen.
So, at the end of the day, we are only successful because our clients are successful. And, you know, we live and die by those performances. I know everyone would say the same thing, asset managers and banks alike. But there is no other way for us to prove to our clients, over the years, that sustained (not necessarily outperformance), explainable underperformance, in light of what we initially suggested that we will build, would bring to us in terms of a kind of confidence and reward.
So, I'm not sure whether I answered your question. I don't think there's an easy answer to the question. That's the process that we kind of go through.
Moritz:Yeah, I think you did answer it in the sense that you're following a rigorous process as far as research and science is concerned, and you're not just producing products for the product's provision's sake. Like, you know, there's no schedule. Every week you have to come up with new index, which would be ridiculous. So, you want to come up with something that's meaningful and different compared to what you had before.
Nick:And also, no reaction to emotion. I mean, we could have reacted in SVB, we could have reacted last year in the dollar/yen unwind, we could have reacted in April when it comes to trend following which just basically sat behind, saw what the markets were here to basically tell us in terms of maybe a new regime, maybe not, I don't know. But reacting on emotion and just making the last drawdown outperform is purely the recipe for underperformance.
Moritz:One thing, and maybe as a final question, we've mentioned the US$1,300 billion, and we're not exactly sure whether that number is right or wrong. We've asterisked this and conditioned it on a couple of parameters. But by and large, where do you think, if anywhere, could the QIS business have become too large? I mean, do you see effects of crowding in some of the markets or in some of the strategies?
Let me give you just an example and I'm absolutely not sure whether I'm right or wrong. But for instance, when you look at commodity indices and liquidity provision around index roles, business day 5, 6, 7, 8 or 9, which, you know, that the GSCI type of roll period. It’s a very successful strategy, probably 10 to 15 years ago, providing liquidity during that role, and then reversing the position, you know, five days later. I think that strategy has largely decayed, then roll days have changed. So maybe that's just one example.
I'd like to hear your view on this of, yeah, enough money has crowded into that space so that the effect has essentially gone away and the market is now balanced. Would you say that's a fair statement? And if so, are there any other pockets of the business where you see similar effects?
Nick:Yeah, so both are very good questions, and that’s a question that comes very often, I should say. So, how should I answer?
Let me just first answer, maybe, picking up your example. You're referring to commodity congestion, I guess, and the fact that there is passive exposure to the benchmark roles, specifically for institutions that do not want to hold physical commodities and want to be just financially exposed to commodity prices and therefore they need to follow an index of all things because it's easy in this regard. But that index is following a predetermined schedule.
So, there are days that you need to sell the futures you're holding and buy the new one. So, over the course of a few days, that is supply, net supply of the contract you want to move away from and net demand for the contract you want to go into.
Now, if the market is elastic, you would not expect any price impact from this activity. But lo and behold, for years, because there was this passive requirement for commodity exposure, it had kind of an impact, a negative impact from selling at the same time that everyone is selling and buying at the same time that everyone's buying.
So, there was this kind of strategy called congestion that said, look, I really know when that benchmark growth will operate, so let me just do it a few days before. So, I'm going to buy whatever else is going to buy a few days before they do so and I'm going to sell whatever else we're going to be selling a few days before they do so.
ghtly so, as you said, around: time with the team, that's in:So, if this were to be massively crowded, as in QIS investors that do congestion, at the margin you would expect to see very negative performance. Because now you're the one congesting the market and whoever is doing the benchmark rolls is actually benefiting from it.
But oddly enough, we saw the strategy flattening out and the flattening out of the strategy, it's more that the initial demand for this liquidity is no longer there. In other words, whether you pre-roll or post-roll or roll together with a benchmark roll, you have no impact whatsoever.
The market started becoming elastic again. It started absorbing this net demand and net supply. And I guess, if you like, the outcome of this analysis suggested that investors became smarter in the way that they rolled the exposures and they now, or at the time, were doing more enhanced data exposure in commodities rather than too many investors crowding out the congestion trade. And this is very subtle, but very important. The strategy became flat, not underperforming.
Lo and behold, in:So that's, I guess, statement number one. But where I want to focus on, as a consequence of this discussion, is that this demand for commodity rolls is not driven by some sort of risk sharing mechanism. It's a pure asset allocation decision. So, the premium, in itself, shouldn't exist in the first place. And this is the space that, you know, if there is too much of crowding, it can go away. Or obviously if there is no demand for those benchmark rolls, as it happened to be the case, there is no premium to be harvested anyway.
It's very different when we start looking into maybe volatility selling. So, if we sell volatility, there is an underlying risk factor that there is a segment of the market that is not willing to take, and we're happy to basically take the other side by selling those options and be paid the premium that whoever is willing to hedge is willing to pay and pay more.
So, can we now observe crowding implications? Maybe the premium falls. I tend to use this basic example. If we all become car insurance sellers or car insurance providers, the premium to insure your car is going to drop purely by the economics of competition. But that doesn't, in itself, suggest that the premium is going to go to zero, maybe negative, unless somehow we will all become better drivers.
So, there's a fundamental link here between what is the price to pay for specific insurance, and what is the underlying risk sharing mechanism, and where the market will clear in terms of pricing that risk. That's the price of the premium, the risk premium, if you like.
But you know, suggesting that this can become negative, I think it's a very bold statement. And the reason why I'm saying so is that there is a barrier to entry. Who is going to be selling this insurance if it drops below a threshold?
So, the more kind of crowding you end up observing, the more subdued the returns would be. It’s still positive in the longer term, but then the barrier of entry will just increase. So, there's an equilibrium here that is achieved purely by the fact that it's an underlying risk sharing mechanism, and that's different to congestion.
And then obviously, the last segment of alpha seeking strategies would be more behavioral. Trend following, I think, it's a prime example here. We can debate what crowding will bring to trend following. There's an argument to be had whereby trend following is actually benefiting from early crowding.
The problem here is more risk management. The problem here is how those negative skew events can be moderated. So, the implications are very different.
So where am I going with all this stuff? I think crowding implications have been overstated because they haven't necessarily been looked through the economics and the mechanics of the strategies. But that in itself should not basically put us behind the curtain and say, look, there's no problem at all.
ere in the, I guess, years of:Now I think the biggest (be careful with my wording), the biggest consideration that banks and QIS desks should have is on the externalities that come with similar products being designed by banks and similar investors holding similar products for specific objectives, which are also very similar. So, if you hold a 60/40 portfolio and an overlay, and I hold a 60/40 and an overlay, both of us will react on a macro signal that would make us deliver the overall portfolio, and as a consequence of that both of us will probably de-lever our alternatives. So, those components now can expose the realized correlations that we haven't seen in a backtest. And I think those externalities and contagion impact of orchestrating activity to macro shocks is more the concerning factor than how do we design a strategy to make it be less crowding impacted.
So yes, there's a lot of focus on that. There's a lot of focus together with our trading desk as to how they see and experience the markets. But if there is one pillar of focus, when we design those products, is to build them for scale, to your point, but for very conscious kind of scaling rather than let's just consume the liquidity of the market however much it can deliver to us and move on with our lives.
Again, we are only assessed by our performance. So, all these are very important considerations. Very long winded answer, but I know it's not an easy question.
Moritz:Not an easy topic. At the end of the day this is what counts, right? It's the performance. If your indices do not perform, if they do not provide the premium that we suggest or expect there to be, then the volume will dry up, people will leave and you'll see reversion of your flows. Such as life.
Look on that note, I think that's a good wrap. Let's close this week's conversation.
We hope that everybody has enjoyed it as much as we did making the episode for you. I hear from Niels that next week Alan will be joined by Yoav. So that should be a fun conversation. It's also your chance to have them tackle some of your questions, if you like. You can send them, as usual, by email to info@toptradersunplugged.com. Niels will pick them up and do his best to bring them up.
So, from Nick and me, thanks so much for listening and we look forward to being back with you next week.
Ending:Thanks for listening to the Systematic Investor podcast series. If you enjoy 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.
And 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.
