Trend Following during Trump Tariff Turmoil ft. Katy Kaminski

[00:00:01] 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 Katy Kaminski and I, Niels Kaastrup-Larsen, where each week we take the pulse of the global markets through the lens of a rules-based investor. Katy, it is great to be back with you this week. How are you doing? What’s going on where you are?

[00:00:39] Katy

You know, things are good. You know, we’re moving forward. It’s mid-April. Here we go.

[00:00:45] Niels

That’s something. You know, speaking of mid-April, I think it’s fair to say that spring has arrived over here in Europe. And usually I associate spring with something positive, something optimistic after a long dark winter. But, as I’m sure we’re going to be talking about today, this spring feels a little bit different, but we’ll get to that, I guess.

Anyways, we’ve got a great lineup of talking points today focusing of course on current events and also your latest paper on crisis and correction and we might call it the TFTTT, which I sort of came up with sort of as the Trend Following in a Trump Tariff Turmoil, but we’ll see if that hangs on social media or not. Anyways, we’ll be tackling some of those issues today.

But before we do that, it’s always fun for me to learn what’s been on your radar outside all of these financial matters that we so often talk about.

[00:01:48] Katy

Well, I mean, I’d say it’s just been, it’s been very wild.

I think everybody I’m talking to, it’s just hearing different perspectives from talking to friends of mine in the regional banking industry to talking to friends of mine in academia because I know here in Boston, we have Harvard, we have MIT. So, you know, people are definitely sort of trying to navigate like what is going on here.

And it’s also April, which, you know, as I would usually say, my tolerance for cold is starting to run out. So, I’m very excited about that pivot into summer and spring. And I think here we’re not quite there yet. We had some ice this weekend and I was like, no, I’m ready for some warmer weather. So maybe May will be our month.

[00:02:38] Niels

Yes, absolutely.

Well, I’ve got a few things on my radar that I wanted to share with you. Some of it is a little bit related to what we do professionally, some of it is not. But I couldn’t help… So, my day started out by me reading a headline about J. Powell and how he had been out, I think at the Chicago Economics Club or something like that, Economics Club of Chicago, I think it’s called, where he’d been out saying that, you know, these terrorists were much bigger than they expected and that this was really making their job pretty difficult.

And to kind of distill it down, he sounded like, you know, right now there’s not much they can do. It’s almost like saying, you know, we’ve been put on pause because of what’s going on here. And, and I thought it was pretty blunt.

Some of the quotes that I saw, where he basically said that they were finding themselves in a very challenging scenario in trying to keep their dual mandate goals and the tension between those goals at the moment. So, I thought that was kind of interesting.

And then just before we press record here, a few minutes ago, I see this headline coming in where apparently the President had been out saying, maybe it was last night or maybe it’s this morning, saying, well, actually, Jerome Powell’s resignation can’t come too soon. I mean, he was kind of fed up with having some critique.

What I will say is that we all know it’s been some pretty tumultuous weeks we’ve had, and I certainly don’t think it would help the situation if someone like Jerome Powell was somehow ousted from the Fed. I think just for my own, I guess, personal opinion. I’m not expecting you to comment on this, and so that’s fine.

But one thing I do think might be interesting to you as well is that I don’t know if you follow the Bank of America Fund Manager Survey. Apparently, I think it comes out every month and it came out on Tuesday this week. And, you know, guess what? For the last, I think, two years, the most crowded trade, MAX 7. In this report, it’s no longer the MAX 7 that’s the most crowded trade. So, I wanted to put you on the spot here, Katy, and ask, do you have any guess what the most crowded trade is right now?

[00:05:11] Katy

Short the Russell?

[00:05:11] Niels

And you’re going to be in it. I’ll guarantee you’re in it. And so are we.

[00:05:19] Katy

Energy? I don’t know.

[00:05:21] Niels

Long gold.

[00:05:22] Katy

Long gold. Well, that. Yeah.

[00:05:23] Niels

Yes. Right.

[00:05:24] Katy

Yeah, I forgot about gold. It’s just always going up. So, you know.

[00:05:28] Niels

Yeah. So apparently, we are now part of the… And I speak for CTAs in general as I’m sure most people will have a nice long signal in gold by now. So, long gold apparently is the most crowded trade according to their survey. So, I thought that was interesting.

My final little gem for you this morning, Katy, is, I think, and you have to correct me if I’m wrong here, I think that it’s just been what you call Tax Day in the U.S. is that correct?

[00:05:56] Katy

Yes. Tax Tuesday, this last Tuesday.

[00:06:00] Niels

Okay, so I read, and this is a little bit of a stretch when you say that there’s a relevance here. There’s actually a sequel to a movie coming out. There was a movie in 2016 called The Accountant where Ben Affleck played an accountant with autism and who was both using math and martial arts in his job auditing the books of a corrupt businessman. So, he’s like the John Wick of accountants.

And I thought that’s kind of good timing when it’s just been Tax Day that you’re sending out a new movie with the accountant. Are you going to watch the first one?

[00:06:37] Katy

And the challenges that people might have – post tax stress, you know, something. So, I may not want to watch. Who knows? Because it’s like, oh, finally done. Let’s wait till next year.

[00:06:46] Niels

Did you watch the first one?

[00:06:48] Katy

No, I didn’t.

[00:06:49] Niels

Well, then you have to watch it.

[00:06:49] Katy

Now I’m going to have something to watch.

[00:06:50] Niels

You have to watch it before the next one comes out.

Okay, fair enough.

All right, enough of that silliness. Let’s move on to the hardcore stuff, the trend following update that we normally get into. And of course, no doubt, April has been eventful and challenging for many strategies, but trend following certainly as well.

If I was to generalize, I would say that it’s been mostly the financial sectors that have been causing the problems, like equities, like currencies. But of course, we’ve also had… Well, let me put it this way. We haven’t had much offset in say, fixed income that, you know, often happens. We’ll talk about that. However, within the fixed income sectors, I do think CTAs will have found one market, in particular, that has really been causing a lot of trouble, and that’s the JGBs, because that really got a short squeeze in the last couple of weeks. So, a pretty difficult environment to navigate, we know that.

Now, there has been a few bright spots. We already talked about one of them, namely gold and maybe a few other of the precious metals. We’ve had a little bit of relief also from energies, which has been kind of a crisis offsetting market, as we’ve seen before. And then also, I guess, another part of the portfolio which is usually something we can rely on if we are positioned correctly, and that is the short-term interest rates. But again, not universal. Not all short interest rates are as strong as they are in, say, Europe at the moment.

So, I was going to ask you a slightly different question before we get into all of the good stuff that you brought along. And we’re going to go through all the performance numbers before we do that. But I was curious. I know you and I, as trend followers, we’re not really allowed to be watching the markets too closely, getting emotionally involved, but is there one or two markets that you find more interesting at the moment to follow?

[00:08:46] Katy

I mean, I think the most interesting trend, for me, and you kind of hit on it a little bit, is, you know, if you look at what has happened. So, we had this big shock where sort of the market tried to price in sort of higher than expected tariffs and the potential disruption of sort of trade relations. And I think the biggest shift that people have to really focus, on and what has been interesting to me, is that you have this scenario where equities sell off aggressively. And in that same scenario, yields went up and the dollar sold off.

And to be honest with you, that is a very, very different reaction to a crisis that we don’t typically see. So, for example, usually when the equities sell off, the dollar rallies as you see sort of reinvestment and capital flow.

And then the same thing is true for fixed income. It’s a safe haven asset which people run to. Which, you know, kind of, to me, is something to note. And the reason I say this is that, you know, it to me shows that there’s some vulnerability in US assets that could change the way that bonds and the dollar behave in this new regime.

And so, I think my general view about any sort of trend is when, as the trends shift quickly, depending on where you’re standing before they fall, that’s going to impact how you perform until you can adjust to where the new trends and the new macro developments are.

And this particular environment was precisely the same story. What’s interesting is going to be where do we go from now? And so, what are the new themes that emerge as a function of a new world? Because in some sense we’ve moved into a new regime: a new regime for global politics, a new regime for business negotiation, a new regime for thinking about
inflation as well and does that become a real threat.

So, when I look at the two asset classes that are typically responding as our safe haven, so bonds, suddenly you have this threat of people potentially wanting to dump US assets. So, that makes fixed income vulnerable. And you have the issue that you could have inflation.

So, even if you cut rates on the short end, and I know you’re just talking about Powell, and sort of the rock and the hard place that he feels that he’s in, but even if you cut rates on the short end, if inflation comes in because of this, then you have the potential for a huge steepener. And we saw a little bit of that peeking out as well. So that makes fixed income very hard for anybody right now.

And then you turn to the dollar. We’ve been talking for months about dollar dominance. In theory, trade imbalances and improving that should be pro dollar. But what happened, actually, had the opposite effect, at least in the short term, where you saw that the dollar actually has lost some competitive advantage. And so, people are starting to talk about, you know, as a key currency, what happens to the dollar.

So, this is showing me themes that people aren’t used to. Trend, longer term, can pick up a real seismic shift in policy, not sort of a free fall. And that’s kind of what we went through.

We’re going to shift through sort of the wreckage of this and figure out what trends are going to work and what things are going to happen going forward. And that’s where sort of the rest of this year could be quite interesting, I think, now that we’ve already made that shift into a new regime.

[00:12:30] Niels

Yeah, I completely agree. A couple of thoughts came in while you were talking. One is just on the latter part of your point, I don’t think it’s just about this year, really. And it seems like we are in, as you say, some kind of seismic change or shift. And I think this could be setting up, and also frankly, and maybe we’ll talk about this a little bit later, when we do look at performance and we do look at drawdowns of all these CTA indices the reality is that this is it. This is where it doesn’t feel comfortable getting in, but this is historically exactly where you want to get in.

Anyways, that’s a different story.

But I also think it’s important, when you talk about these reactions, and some of the reactions we’ve seen, say, in bonds, some of the reactions we’ve seen in dollars are not typical. Of course, I’m sure it’s happened before. If we go back and look at all the crises, there will have been a crisis similar to this.

But of course, at the end of the day, it’s not necessarily the market move that made our industry and others lose money because of the shift. It’s obviously how we were positioned going into this particular period. So, all of that will sort itself out, as you say, once the new trends emerge and all of that stuff. We need to navigate through this and we’ll come to that.

But I mean, as eventful as it is and as painful as it is to go through this periods, this is also, I think, one of these really important moments, not just in global finance, but also in the little world of CTAs, for sure.

Let me turn a little bit further into our trend following update before we leave that. So, my own trend barometer, just as we normally do, finished yesterday at 45. That’s completely neutral. Again, it uses slightly shorter timeframe so not completely unexpected.

When we look at the numbers, these are as of Tuesday because we’re recording on Thursday morning your time. So, BTOP 50 down 5.2% for the month, down 5.26% for the year. SocGen CTA index down about 6% for the month, down 8.36% for the year. SocGen Trend down 6.48%, down 10.84% for the year. And then comes the Short-Term Traders index down only 1% roughly this month, and down about 1% this year. So, definitely doing better in the short-term timeframes.

And then we look at the traditional indices. MSCI World down 4.41% as of last night, 6.5% roughly for the year. In terms of bonds, the S&P US Aggregate Bond index down 52 basis points in April, but still up 2% for the year. And the S&P 500 Total Return down 5.93% after yesterday’s sessions, and down 9.95% so far.

Anyways, I think this conversation today, Katy, there are so many directions we could go. I’m going to try and start out with your paper and then we’ll see. And, of course, in many ways you should be the ones going through this. But it is interesting. I mean it’s very timely. It’s always great that you manage to publish these papers right, when they’re needed.

But it’s, to me, it’s not just kind of any other research note, and people should go and download it of course. It’s also a kind of a diagnostic tool for people. It’s kind of a framework for understanding whether we are looking at a short-term market dislocation or are we standing at the edge of a structural regime change? You talk about catalyst, you zoom out and you take a big picture and look at these things. So, let’s do it step by step and try. And even though people can’t see the paper in front of them, maybe we can visualize it and then they can go in and grab it on your website later on.

[00:16:33] Katy

So, Niels, I’m really… This is the type of thing, you can imagine this stuff is happening, and what does Katie do? I’m like, I need to write a paper.

Actually, what happens is that in these moments, as trend followers, it’s always very interesting because in the heat of the moment, on those type of days… And I think to give people some perspective, when the S&P 500 moves 1%, that’s sort of like a 1 Sigma move. So, if you move 10%, that’s a 10 Sigma move, like approximately. So, I always think about that as sort of a barometer of like how intense is the world?

And so, when you’re in those days where things are moving at multiple sigmas of your typical statistical distribution, it always feels much more intense and it feels like things are in slow motion sometimes. And that’s why I love trend following, because you can’t make decisions. You need to like, think and you need to watch and kind of follow your process. And it’s what tends to actually work in many very extreme environments.

But it is at a moment where, you know, you can get stuck in that one day. And so, what happened with me is that whenever we’re in some sort of drawdown like this, of course there’s an emotional reaction. So, my plan is like, all right, we’ve been here before, let’s look at the data. Let’s do some analysis, let’s ask ourselves a question. When equity markets are in a drawdown, how does trend typically respond? And what is a typical trajectory of opportunity set, in terms of like when the market is in a drawdown or correction, how do we typically react?

And second of all, you know, what is sort of the importance of how deep the drawdown is, how long it is, and what are sort of the type of environments we’re in? Because if you think about it, you know, it seemed like it was pretty huge, what’s happened. But just like a Covid type environment, it was very intense for a couple of days, not a couple of months. It may have felt like that because it was a big move, but you know, you have to kind of step back and look at the data.

So, we went back and we’ve written a couple papers about crisis or correction throughout the years and luckily for me, it’s sort of easy to rerun this analysis because you know, you just dust off the code and you know, there you go. So we wrote a paper, Yingshan, one of my colleagues, and I, updated some analysis and wrote a paper called Trade War, Crisis or Correction: Managed Futures and Crisis Alpha in 2025.

And what this paper is really doing, similar to some other work that we’ve done in the past, is really looking at how does managed futures do across different drawdowns for the S&P 500, peak to trough, across different dimensions? So depths of drawdown, in terms of how much pain you endured – so how far did a drop, and then of course length. so duration. If you think about sort of pain, did you have it quickly or was it something that drew out over long periods of time?

And so, what’s really helpful with this analysis is that if you look at drawdown, similar to what we’ve experienced anywhere around the 15%, there’s not a lot of data points above 15%, so around 15%, anything less than 15% actually tends to be negative. And the reason for this is that it’s what we talked about at the beginning of this discussion. When markets get a shock, it’s often in the opposite direction of current prevailing trends, in various orders of magnitude depending on what’s going on. And thus, trend following often has to adjust and react.

So on average, although sometimes we get lucky and happen to be in the right trend before they explode, oftentimes we end up on the wrong side and have to adjust our positions to get into whatever that new theme is going to be. And so, what you see here is that the first part of the drawdown is often a negative environment for trend following.

And as you move farther down, you know, deeper drawdowns and longer drawdowns, that’s where you see, you know, that type of crisis alpha, the second responder type of response that consultants like Maketa have kind of coined in the sense that you really sort of can’t be prepared for an initial shock. But trend is able to adjust and adapt to defined times like the tech bubble, the GFC in 2022. And so, I think for us, what this means is that we’re then through the first part of the pain.

What’s interesting is what happens next, and sort of what opportunities could be available in a world where we saw a massive macro shift or massive policy shift.

[00:21:35] Niels

Yeah, absolutely. And it’s important, maybe in kind of a way to distill it, that we as trend followers, we don’t really chase the sell-off, but we ride that unraveling that happens in the slipstream of the sell-off, however long that may take. Ideally long enough for us to actually find some opportunities. So yeah, that’s really super useful.

Feel free if you want to take in the next step of the paper or I can throw you some questions that spring to mind as we go along. I’m completely open here, Katy. Whatever you prefer.

[00:22:16] Katy

So, the first part of the paper, we show you all the drawdowns and we have this beautiful bubble chart. Because I love bubble charts.

[00:22:22] Niels

I see some people like that chart a lot because I’ve seen it in other papers now.

[00:22:25] Katy

Yeah, there’s nothing like three dimensional. Like, you know, having three ways of explaining something. But you have red, and you have green, and then you have these big bubbles for good stuff and red bubbles for bad stuff. And the size of the bubble tells you how big it is. And so, it’s a really, really great visualization of crisis alpha.

[00:22:46] Niels

Can I just interrupt you? I’ve never created one of those charts myself. Is that called a bubble chart or what is it actually called?

[00:22:54] Katy

I call it a bubble chart. don’t know, like, they’re awesome. So, basically, you know, it’s a way to visualize data in one more dimension. Because a bar chart just wouldn’t do it for you. So, this bubble chart is really, really good.

And then the second part of the paper was really sort of, okay, you look at that chart and then you say to yourself, okay, for events with 15% or less drawdowns, or 17%, or not extreme. So, kind of in that beginning of a difficult period, how did trend do?

And we kind of show that trend struggles in the onset of a crisis, especially when it’s quick and deep. Then the second part was really looking at the few data points that we have that are longer or deeper. And I divided those into two sections. One was short, sharp drawdowns and the other was extended crisis periods.

And this was very comforting for me to see this graph because, you know, like I said, when you’re in that day when the S&P has multi sigma moves, it feels so big. Because it is big. It’s a statistical outlier.

But when you actually look at history and you kind of put in perspective, how long has this drawdown been going on and how does it compare to say a Covid response, and how does it compare to say a tech bubble response? You kind of see where you are. And so those two graphs are very interesting.

The first one is about where we are versus extended crisis periods. And when you look at that graph, it actually looks a lot like the start of the tech bubble. And that really, I thought, you know, there’s some rhyming there. You have AI, you have some disruption. It’s not the same but, you know, it turns out trend did very similarly at the beginning of the tech bubble, but ended up being something that worked very well throughout a very extended, difficult period for equity markets.

And if you look at the graph, and I think this is the other part, and you’re probably looking at it right now, it’s actually hard to see this recent event on the graph because it is so short. It’s like 20, 30 days.

[00:25:12] Niels

Oh yeah.

[00:25:13] Katy

And the tech bubble is 500 something days.

[00:25:15] Niels

Yeah.

[00:25:16] Katy

So, you kind of say like, oh my gosh, if we’re going to go into a much more extended crisis period, you know, we have the whole graph to follow here. Like we’re not even there yet. And so that kind of made me… It kind of put things in perspective.

So, if we do really believe that we could have a recession, we could have inflation issues and these could take time to persist. This is only the beginning of the story. The story is going to take time to unfold.

So that was actually a very good picture for me. I like graphs, they give me some emotional support in those wild days. And then I also graphed the trend following drawdown versus short, sharp drawdowns that actually recover. And you have a few of those. You have 2018 and you also have 2020 that are some of the recent ones.

[00:26:06] Niels

So, Volmageddon and Covid.

[00:26:08] Katy

Exactly. And if you look at those events, it actually looks very similar to Covid and it’s not that much different than 2018. So, it kind of showed me like, you know, Katy, we have been here before. You know, markets do these things from time to time. And this is sort of a natural cycle of what it is to be a trend follower and dealing with different crisis environments.

So, I think that’s why data, in pictures, is always a way to kind of zoom away from those days and ask yourself the question, okay, we’re only 30 days in, 20 days into the drawdown, what does this mean? And if you look at the longer term data, that really helps.

And I think perhaps my friend Alex Greyserman helped me with that because some of our 800-year analysis, I try to think about that sometimes as well. I’m thinking, like, 800 years, have we seen this in 800 years? And the answer is probably yes. But, you know, anyways, data is helpful.

[00:27:11] Niels

Data is helpful. You know what? Someone who helped me with some data, of course he didn’t intend necessarily to help me, but he shared it on, I think on LinkedIn. And I think last week I also cited his work, but maybe I didn’t get his full name, but he’s called Tyler Lovingood. And he did something which is always very useful and that is just to remind us of the worst drawdowns, and so on, and so forth.

So, he divided up the drawdown of how does CTA typically react when you have the first 10% of an S&P drawdown and then what happens once you get from 10% to 20%? So, I did a little update myself because his chart was a few days old.

And just for those people who may not have this data, the actual drawdown from, I think it was February 19th high, or something like that, where the S&P had its peak, and then until the low, which was April 8, before we had a little bit of maybe manipulation of the stock market, who knows? The drawdown of the S&P, close to close, was 18.74%.

So, we’re actually not at the 20% level right now, which is probably not a bad thing because actually both the SocGen CTA index and the Trend index, in that period, were actually down about, you know, 8.6% for the CG index and about 10.76%.

So, even as we get closer to the 20% yet, we haven’t quite turned the corner in terms of producing positive performance, but who knows where it’ll end up. But it’s kind of similar to some of the work you’ve done, and I think it is super helpful in periods of a bit of emotional stress, I’m sure, to just keep this in mind.

So, I want to do something a little bit different, Katy, because I want to throw something at you because you did coin the word ‘crisis alpha’, and you and I have talked in the past about how I felt that sometimes crisis alpha could be challenging because everybody loved the term.

But then every time there’s like a “drawdown” of the S&P, people would say, well, hang on, that’s a crisis and where’s the alpha, and all that. Anyways, I want to try something different with you today. And that is, do you think that there are different types of crisis, so to speak? And I know this is a little bit out there, right?

But when I look at what’s happening, and I kind of date myself by saying that I’ve been involved in the markets for 40 years, by now, I would almost classify this as chaos, not necessarily crises. And I was thinking, could there be a difference between what would be a good Crisis Alpha solution to a good Chaos Alpha solution?

Not saying that I could back this up academically in any way, shape, or form in terms of what’s different, but this is about how I feel. I feel this is different. I feel this is chaotic. I felt Covid was a crisis. It was out of our control. And you know, all of that stuff. This just feels chaotic, and we don’t know why. So, do you think there could be something, shades of gray, if we call it that, between one type of crisis and maybe chaos?

[00:30:32] Katy

I love that you asked this question because I want to reiterate the first initial paper on Crisis Alpha you probably remember, that I wrote a bazillion years ago. It was literally called In Search of Crisis Alpha. The point is that Crisis Alpha is not a guarantee. It’s just this idea of how do you try to find strategies that might be able to do well in different crises.

And so that means that different tools and tricks are going to be important for a different type of crisis. And so that was the original idea. And that explaining that, you know, this is not a guarantee, this is just a search for crisis Alpha in that, you know, there are things that might be able to capture it. And as you have different types of crises, different types of strengths are going to be important.

And that’s why I am such a fan of some of the work by the consultant Maketa, because they actually explain this in a really great way where they talk about the first responders, the second responders, and third responders. And you have to think about building a portfolio with attributes that might be able to catch it.

So that’s sort of more the way I think about it. Instead of thinking about, you have one tool that’s going to get you that response. I know that one tool, it’s called insurance and it’s expensive. So, if you want to do it more strategically, chaos crisis alpha, there’s few strategies that can do that. It’s often volatility strategies or other types of approaches and sometimes it’s even cash, you know.

[00:32:09] Niels

Well, it’s funny you mentioned that because I’m pretty sure I met with one of the authors, Ryan, at the latest conference in Miami, and I don’t remember the specifics of the paper except I agree with you, it’s a good way of explaining it. And I know CTAs and trend followers, we got into this, you know, the second responder, and that makes perfect sense.

I do think though that, and this is just for illustrational purposes, I do think though that a lot of people would have classified long US Treasuries in the first responder group. And of course this time that didn’t work either.

So, I think this is a good illustration that we can do all these classifications, we can come up with all these fancy financial terms, but at the end of the day there is, as you say, there is no guarantee and people need to be open minded. And I think it kind of also leads me to another topic I wanted to talk to you about before we pivot to a related area of this.

But I think the other question people will have when they listen to us is, well, why are managers’ performance so different this time around? And you know, some managers have been around for 30, 40 years, are having their worst drawdown ever. And you would think… And I’d love to hear you, as a researcher’s viewpoint, because in my simple mind, not being anywhere near any of the research we do, you think, well, hang on, we’ve been doing this for 50 years. We’ve been researching all of this. So, we should have been proved. Okay, that’s one thing.

We have many more fancy tools. We can react more quickly, slowly, whatever. We do all this analysis and we have all this experience, we’ve gone through many different crises. Yet for some, at least, this seems to be an environment that their current configurations are just not handling as well as they have handled previous difficult periods. So, take us into the engine room, take us into the brain of a researcher and how do you think about this in terms making robust improvements? Because we’d like to think that we’re improving with all the resources we put into research. How do you think about this?

[00:34:31] Katy

This is a Great question. And it always reminds me of an experiment that my advisor, Andrew Lowe, used to do with his MBA students. He would give them a penny, the whole class, and he would have the whole entire class toss that penny a certain number of times. And then he would turn to people and say, did anyone get seven heads? Did anyone get eight?

And there always was some person that got those heads or tails, right? More heads or tails. And when you have enough volatility and enough movement, you get return dispersion. So, you get a wide range of outcomes.

And if you think about a very extreme market environment, like what we experience, or like Covid, so much matters about where are you standing before you fall. So, you can think about all these trend followers on a cliff, right? Some of them are here, some of them are here. When the earth shakes, depending on where you’re standing, your fall is going to be different, right?

And so when you have a shock, it is very hard to predict, you know, how your system is going to react because it depends exactly your magnitude of your positions and how that is relative to the direction of that shock. And so, over time, it’s very common, especially in extreme environments, that there is a wide dispersion because there’s wide movement of those returns.

And so, you know, if you held more European equities than someone else, if you had short signals instead of long signals, even marginal differences there, you had big differences in return over those days. And so, for me, it’s really about kind of understanding that on average there’s going to be a wide spread. And just like those pennies, if you do the experiment again, you’re going to see, again, another mismatch of people, where do they end up?

And it’s because it’s a shock and it’s sort of a probabilistically unpredictable shock that you know you’re going to end up with that return dispersion depending on where you’re standing before the event occurs. And so, I’ve seen this throughout my career multiple times.

And I think that’s where it’s very useful to talk to the fund of fund managers, and those that really know all of us very well. They know that sort of return dispersion is part of our strategy. And that’s why they know that when this happens, it happens to the best of managers and the worst of managers.

It just happens. It’s a shock. And you end up finding people in different places along the distribution.

And I’ve written a couple papers about return dispersion, the Things that we can do. I mean, there are things that we try to do. We try to aggregate multiple different approaches for how we measure trends. We try to incorporate risk management to adjust for changing volatility. We’re doing all those things. But the world is idiosyncratic and there can be extremely time series dependent moves that are like a shock.

And in those shocks it’s very hard to predict exactly how that’s going to be. The only thing we can do is try our best to try and risk manage that. And it doesn’t always… It’s very normal, is what I would say.

[00:37:52] Niels

Yeah, and I want to dig a little bit deeper into that in just a second because I do think it’s important just to repeat, during a time like this, kind of the main drivers. And also it relates a little bit to a new wave of products that are being launched that are very relevant for what’s happening right now.

But before we go there, unfortunately, I don’t remember who wrote this paper or who made the comment. I just simply don’t remember. But I remember that during one of the recent years, this is like in the 2020s, I think, that’s how I recall it. There was someone who had done an analysis that showed that, let’s just take moving averages as an example. I think they mentioned that the difference between having 100 day moving average and 105 day moving average, in that one particular year, was like a 20% performance difference. It was like crazy. Where you think, how can that be? But it was.

I don’t know if you remember that paper, or whatever, but it is exactly to your point. Some of the choices we make, sometimes we’re on the right side.

And I think, you know, even when I speak to some of the people who are in the replicator business, they will admit that, yeah, they were lucky this time around that they actually, you know, were too slow, or slow enough, or fast enough, or whatever you call it, just to navigate the last couple of weeks a bit better maybe.

[00:39:18] Katy

It goes in cycles and I think that’s the industry, and that’s why you have to look at longer term because it’s longer term. Things change and then the next crisis or the next part of the crisis could be very different.

So, I think, you know, you saw that in 2022 as well. In that bond trade, some managers got that better than others and, you know, it just had to do with how they design their systems. And you see return dispersion on both sides.

And that’s something that I’ve written some papers on that as well, and we even have a chapter in our book, and ironically it’s called, it’s chapter 11, which is a funny one on return dispersion. And it’s fascinating to me because we talk about sort of the value of parameter diversification and what you can do about it. What you can do, because there are things you can do. And it does lead to, you know, having very simple parameters can create more return dispersions. That’s why managers have learned to kind of build robust signal processes that combine different methodologies and approaches.

And I worry about very simple implementations because, you know, if you have that 105 day moving average versus the 100 day, you can have big differences just implicit in your sort of lack of parameter diversification.

[00:40:46] Niels

Yeah, I mean, for sure. And it seems to me, at least, that even though I normally would say that, well, whether you use one kind of trend following model or another kind of trend following model, it’s not going to be too big of a difference. And that might be true in the long run, but certainly when we go through these specific periods, shorter periods, “crisis periods”, even down to what the type of trend following approach you choose can make a big difference.

And of course, without starting a debate with some of the people listening here, even something like whether you have static position sizing or dynamic position sizing will make a very big difference. It certainly worked well in the last two or three years just riding the trend, but it’s also pretty painful when things turn like they’ve done this time around.

Anyways, I want to pivot now because we still have a little bit of time left and you know, having products that can help people, this has been one of the main arguments, I think, for our industry for a few decades that, you know, adding trend following to your portfolio, to your full portfolio is just such an “easy decision to make” because, you know, you can improve your upside, you can reduce your downside. So, of course people start producing these products.

And a few years back, certainly the people like Standpoint, we have Corey Hoffstein, we have quite a few great, great friends of ours in the industry have provided people with solutions, now, where they can just buy a ticker and it’s all great. And of course, right now, this is exactly when these products will show their relevance, so to speak.

Now, of course, at the moment, people might feel that the returns are stacked against them because it’s not, as you say, we are hurting, the beta is hurting. It’s not really that easy to navigate right now.

But when you have to choose, this is my angle a little bit. When you have to choose your alpha engine when designing your trend model, there are a few things we have to focus on that makes us one engine different from another engine, besides whether you’re using say moving averages, or breakout, or whatever it might be.

But it’s trading speed, it’s the universe of markets and how you weight them, and it’s the risk management. So, I wanted to maybe talk a little bit about each of them just to help people understand some of the choices, some of the things, and how you feel that they impact. There is no doubt that, historically, if we go back to maybe the ‘70s, the’ 80s and the ‘90s, trading speed was probably different to what it is today.

So maybe you can touch a little bit on that since you did do the 800 year analysis. Maybe you remember some of that. Just kidding here. But anyways, there has been an evolution in that sense. So, let’s just talk about speed because I also imagine that when you design a trend following product, perhaps specifically to be a portable alpha engine. Meaning it will have an equity beta, or bond beta, or whatever beta you have combined with it. Maybe there are some of these things that are more important than others, actually, frankly, compared to just designing a standalone trend following product.

I don’t know, I’m just throwing it out there for you to think about for three seconds before we all expect a clever answer.

[00:44:28] Katy

Yeah, I mean, so thinking about trading speed, this is interesting to me because some of the analysis that we have done has definitely showed, looking at the index and sort of decomposing it by speed, we have seen a consistent shift to slower trend systems. And you know, that makes sense, in trading costs and also capturing sort of long term risk premia, which actually has worked really well until more recently, I would say. There are some caveats to that.

If you are only in longer-term trading speeds, your ability to pivot, especially if it’s a longer dislocation, is going to be less possible. So, in general, over history, having a little bit more reactivity can help you capture some of the crisis in history. That doesn’t mean over something like recently where it was very fast.

And so, I think, in general, it’s about balancing how much reactivity you want in your system versus how much you’re kind of willing to lean in on long-term risk premia. And if you’re leaning into long-term risk premia, you’re looking a lot more like equities too. So, there’s an issue with that.

And so, I think that’s where investors who are looking for risk mitigation. They know that shorter-term trend systems have a lower Sharpe, especially more recently. But they have that diversification characteristics that can help, especially if you’re thinking about adding it to equities.

Okay, so you’re opening up a whole can of worms here, Niels, because you’re not talking about just trend, you’re talking about a new problem, this idea of portable alpha solutions. And essentially portable alpha solutions are combining another investment with trend following.

And the exciting part about this that a lot of people don’t know, or at least why I’ve always been a fan of these type of approaches in some degree, is the cost effectiveness in terms of margining and the collateral efficiency that the futures markets provide so that you can actually get that 200% exposure, or let’s say you want 150% exposure, without having to put in more money. And I think that’s something very unique to the capital efficiency of futures.

And so, I think that’s why this is an area that’s definitely booming and that people are starting to get used to sort of this concept of, you know, combining different investments together. So, you get your equity investment without having to take it out of your equity bucket, but you’re adding trend on top of it to kind of give you the return of trend to smooth out the ride of your equity.

So, the reason I say, you know, we can’t just talk about the speeds in isolation is, when you start to do a combined portfolio, you change the objective function a little bit, right? Because if you’re looking at a trend program, there’s a couple objectives. One is the best possible return, another is the highest Sharpe ratio. Another is the most diversification, sort of the most like crazy self alike, the most risk mitigating.

When you combine those two, your new objective can become a little different because now you’re doing a portfolio construction problem where you’re combining equities with trend. And there’s a couple of approaches you could have. One is slap them together, you know, and just like stack them up and you know, just let them rip. Right?

Another is, wait a minute, these two things are not independent. We need to think about how they relate to each other. So, for example, if trend is fully in equities and you have 100% equities, ooh, you know, good luck. That’s intense. Right. So, you’ve now opened up a rebalancing problem, a question about what goes best together? So how did those two ingredients combine? So, we were now trying to solve our investors problem for them, in some sense.

And so, that means that when you start talking about speed, when you start talking about which markets, when you start asking those questions, there’s a new problem to solve when they’re combined.

And so, I personally think, and this is something I’ve done a lot of research on, and there is a great paper years ago called Taming of the Skew, that I really love. It’s the idea that when you combine equities and trend, they’re naturally very cohesive, but that problem of how you balance them over time is something that investors do struggle with. So, when equities are up, they say, what have you done for me lately? And then when equities are down, they say what have you done for me lately?

And if we can at least solve that problem a little bit better, there’s some opportunity there. So, it makes, potentially, the investment a little more holdable, but it also allows us to utilize the value of some rebalancing between those different assets together.

So, I think it’s an exciting area. But again, like I said, it’s not just a trend following question. It’s opening up a new can of worms and a new optimization problem, like, how do you solve that portfolio problem now that you have control of the equity? As opposed to that usually we talk about, hey, we do really well with your equity. Put us in your alts bucket and you figure that problem out. Now it’s, we need to figure that problem out.

And we can either throw our hands up and just put them together, which is a classic portable alpha, or we can kind of start to say, you know, there are smart ways to think about risk management of those two things. So, let’s talk about recently. Like, this week, this last month, would be a good case study for that.

[00:50:09] Niels

Sure. I’m sure there will be lots of people going back to the drawing board and just checking up on ideas. One thing more, so, just before we go maybe to more the thing about, you know, the markets you choose and how that may impact, just staying on speed, just whether it’s part of a portfolio alpha product or whether it’s a standalone trend following product. Do you think there’s any case for treating markets or sectors differently in terms of speed or are you more a subscriber to the fact, let’s not be too clever, let’s just run the models the same across the whole portfolio. How do you think about that?

[00:50:55] Katy

Oooo, that’s a really tough one because it’s a yes and it’s a no. And I hate when people answer yes and no at the same time.

But you know, if you look statistically, there are certain techniques where you can actually sort of lean a little bit into certain speeds for different asset classes based on, you know, using nonlinear approaches and things like that. I would say those types of choices need to be on the margin because the truth is there’s just not that much statistical evidence for such.

So you could go through periods where, you know, you should trade FX really short-term and then you go through another period where you should trade FX really long-term. And so, I would say that I always think about whenever you want to specialize your view, instead of diversification you need to look for statistical evidence, and you need to adjust your view as a function of the level of statistical evidence. And the statistical evidence is slim. But there are things, you know, some things that you can do, it’s on the margin. It’s not that there’s a magic speed. You know, like 105 days.

[00:52:07] Niels

And I just want to say to people, do not go home and start trading 105 days or 100 days. We’re just throwing these out for…

[00:52:13] Katy

That was Niels’ number.

[00:52:14] Niels

Exactly.

[00:52:15] Katy

Not mine.

[00:52:18] Niels

Okay, so let’s adopt this question framework for markets traded, because, again, when people do their analysis they’re going to find markets that don’t look great, and they’re going to find markets that look fantastic. Cocoa is a good example of a market that didn’t look great for 15 years and suddenly it looked great. So how do we think about this?

And then maybe you could touch a little bit on, I know you did it already but, this thing about if you are building the alpha engine for say a portable alpha product linked to equities, because that’s probably the norm, should you then actually have less exposure to equities, or should you say no actually my trend engine needs to be the best ideas product. This will include this amount of equities regardless, and so on, and so forth. I’m just curious about how you think about the portfolio construction or the markets construction.

[00:53:12] Katy

Let’s start with the markets. I would have to agree with this idea that if you’re a pure trend follower, you believe that trends can come from anywhere and you’re going to position based on the strength of those trends. And I always remember multiple periods throughout history in the space where one asset class or one particular asset was dead.

So, my favorite is metals. I remember for decades people saying, we should never trade precious metals, we should never trade industrial metals. They don’t do anything they can’t help. What have you done for me lately?

And you know, metals have been very exciting in a world where inflation matters, and where you have Covid, and supply chains. And so, you know, I think the key is if you’re really a pure trend follower, you let the signals tell you when to trade those metals.

It means you may not trade cocoa for a long time, very effectively, or very much. But you know, when cocoa actually moves, you’ll be there. And so, that’s my view with trend following, really, across multiple different assets, is really just following that process of following the strength of that trend. So, when that actually occurs, you’re there. That’s pretty consistent with this philosophical idea is that you never know where the next trend may be.

And that’s where the diversification comes from a fundamental approach where you have to decide, well, metals work or they don’t. Instead we say, we don’t know if metals work, but they may work one day and when they work, we’re going to be there. And that’s how I see cocoa, for example – cocoa and coffee and whatever… Oh, cocoa and coffee. Oh, they’re just Ags. They never trend.

All you need is a major supply chain crisis and it’s like, ooh, better stock up for Easter, you know. So, I’d say that my view on markets is really about being agnostic and having an approach that follows the true nature of trend so that you can capture those unpredicted and sort of often unexpected trends that are quite interesting for diversification.

The challenge of course, with that is that when you look across trends in our space, there’s the idiosyncratic trends and then there’s the sort of global macro trends, right? And in certain environments, you see where those global macro trends are really the big driver of a lot of trend performance and movement.

So, 2022 is a great example of this. Fixed income, it may be everybody knew yields were going up, but it was a huge trend. And it’s bonds, it’s not cocoa, it’s not something esoteric. It’s something everybody owns.

So sometimes it’s those trends and then other times it’s also adding some diversification to add things like EU emissions, or when you think about what’s happened in coffee, or cocoa, or soybeans. And a great example I’ve been looking at this month is soybeans, for example, have gone up really a lot while you’ve seen energy and a lot of the commodities really going down. So, you’re kind of seeing that value of having not just the macro trends but having other things as well.

[00:56:21] Niels

Yeah, absolutely. So, before I get to the last point that I wanted to just touch on, which is relating to how these models are different, and why one manager might perform differently to others, and so on, and so forth. The one thing I was questioning a little bit in my own mind, thinking about these portable alpha products, and also obviously we had Razvan, from Aspect, on a couple of weeks ago talking about portable alpha. It’s this thing about whether I think in many ways there’s lots of advantages of just showing people one return stream, it’s one product. You don’t need to worry about the two line items because clearly people don’t like the trend line item when they have it separately.

However, it also makes it harder to work out, for investors, how good is the alpha engine? They just see a return stream unless they really spend a lot of time analyzing how much performance came from one or the other.

So, enough for now, but I think it is interesting to find ways of actually keeping a little bit of a track on how good are all these products coming out in terms of delivering that alpha as part of the portable alpha. Anyways, a separate point.

Anyways, the last thing I wanted to talk to you about today was the third thing that I think is becoming more and more important, actually, in determining the difference in performance between managers. And that’s the risk management. And I think it’s actually also an area where we will be able to continue to learn and improve our models.

And there are two things that I just want maybe to give you as a starting point and then, you can kind of improvise from there. But there are two things that I find very useful to think about when it comes to the risk management. It’s how we deal with correlations.

Now I’m fully aware that some of our long standing friends in the industry will say, well, correlation doesn’t matter, you don’t need to worry about that. In your position sizing, you keep it all separate. Fine. There are many others who will say, no, correlations are important. We need to build that into our risk framework. So, that’s also an approach.

Also why I think it’s relevant, this time around, to talk about is that as we’ve already touched on that this time correlations “behaved differently to what we would expect”, again, depending on the positions we had on.

So, the last thing I wanted to just talk about today is just the role of volatility in risk management, because, as Yoav Git and I talked about, I think last week, when it comes to a lot of the changes in exposure that happens around trend following, around turning points and so on, and so forth, you could say that having models that are different in design can lead you to get signals at slightly different times compared to your peer group. And that obviously helps with say, liquidity in the markets and trading at different times, maybe as some other models.

However, if there’s one thing that probably unites us in terms of when we want to trade, it’s actually adjustment for volatility in the position sizing, because that’s something we all register more or less at the same time. So, talk to me about correlations and volatility, and how you think about that or how people should think about it.

[01:00:05] Katy

Well, I think, you know, correlation is very important to me, and the reason I would say, not just to me but like to many CTAs, is because as you’ve seen… Let me just take the example of stocks and bonds.

[01:00:18] Niels

Okay.

[01:00:19] Katy

You know, when you look at stocks and bonds and how they trade and how they behave, if they’re trading based on inflation or rising rates, they have positive correlation. Which means that if you’re long both of them at the same time, you have more risk on. So, you know, intrinsically you need to think about correlation because correlation tells you where your risk is.

I mean, the challenge with correlation is that it’s volatile and we saw that this month. So, in the first part of this sell-off, you were registering positive correlation between US assets and equities. And then, later on, that actually flipped. So now you’ve actually seen that shift.

So, it does open up a can of worms of how do you balance between reacting, and measuring, and adjusting to those correlations that do define the risk in the current environment versus how much do you overreact to sort of something that is kind of volatile? And I think this has become a very, very interesting question post 2020 because, as soon as we started… And we have a great paper, last summer, that we talked about – that Bonds Behaving Badly paper. So, bonds are still behaving badly by the way.

But you know, once we moved into a world where fixed income was more vulnerable, let it be whether the US assets are in danger or whether or not it’s an inflation threat, suddenly you’re in a world where correlations do matter and change a lot. And so, I think there is a lot of importance to thinking about these questions and incorporating them in some manner into your process to actually understand the amount of risk that you’re taking.

I’m not saying it’s an easy thing because obviously you need to measure that correlation, you need to manage that correlation, and you need to be skeptical volatility of that correlation. But I think if you just kind of say, I don’t care, I think that is could have adverse consequences.

So, if you just assume that stocks and bonds have negative correlation and you pile into both of them, and you have a risk model that has a long-term estimate or correlation, guess what? You may be taking a lot more risks than you say.

[01:02:39] Niels

Although, just in defense of those who don’t use correlation, I just want to say actually I don’t think they care whether they are positive or negative. Meaning they don’t have any assumption about it. They just say here’s my entry and here’s my stop, and that’s it, and I don’t need to worry about it.

[01:02:54] Katy

But they still will be exposed to that correlation.

[01:02:56] Niels

Of course, yes, yes. But they will have their stop to kind of protect them on the way up. The funny thing is, I agree with what you just said but the funny thing is, again, going back to my idea about chaos, that maybe during a period of chaos, actually trying to be too precise about correlations, might not actually be a great thing because it’s so damn difficult. Sorry for my French here.

And correlation, by the way, is not something you can measure over short-term periods with success. I think you need to give it a bit of time. So, it is a challenge. But volatility is another interesting and maybe increasingly important part. Not just how you measure it maybe, but actually how you use it.

And as you know, lots of people have kind of fear mongered about the fact that we all use VAR models and we’re all exposed to the same systemic whatever, shock in that world. And luckily, we haven’t seen any fallout from that? But just talk to me a little bit about var… not var, sorry, about volatility.

[01:04:07] Katy

Oh, good. Yeah, I’d rather talk about volatility. Volatility… You know what’s hard with volatility is we use volatility in so many ways. And so, we could talk about sort of the more static and less aggressive way that we use volatility. And that has to do with position sizing.

And so, for example, currencies have volatility, let’s say the yen around 10% and then commodities have a volatility around 40%. So, guess what? You need like a ratio of 4 to 1. So, you need less exposure to those commodities than you do. So just sort of in balancing risk, you know, there is that natural, like how risky is that asset? So, you know, our exposures over time are sort of balanced on those risks, similar to risk parity or something like that. So, we see investments as a risk opportunity and that’s one way that we sort of intrinsically use risk.

And so, you can imagine when the world changes, like for bonds and in the year of 2022, when bond vol was 2x, that’s where you have to start thinking about, well, how much risk do I have in each of these assets? And then another way that we use volatility is in managing the overall portfolio and also in signal generation.

So, you know, intrinsically in any sort of signal generation, you have to look at how much signal do I have versus a noise. And so intrinsically, in how strong the trend is defined is always the volatility measure as well in some way or another. Not always standard deviations, sometimes different things as well, depending on what you’re doing. And so, I think intrinsically in that, volatility is an issue because it’s something that’s inputted in how strong the trends are. It’s also something that comes into play and how much exposure we have.

And so, when you need to start thinking about volatility, and it’s true, everybody knows, like forecasting volatility is very hard. But let me tell you an analysis that I did – oh, nerd out for a second. I’m sorry. One of my favorite analysis I did a couple of years ago is I had a researcher looking at this topic and I said, let’s imagine that I could perfectly predict the volatility of an asset in my portfolio so that I actually knew the volatility. And it turns out that even if you know the volatility, it doesn’t improve your performance that much.

What it is, really, is about balancing risk management and thinking about that. So, like actually predicting volatility, on aggregate, over longer horizons, there’s less juice there. But people have this idea like, oh, if you could better predict volatility. But the truth is volatility tells you the range. It doesn’t tell you if you went up or down.

So, it turns out that is not what’s going to give you that out, like better performance. And so, I think the challenge with thinking about volatility for us is these moments, like the recent one.How much do you incorporate shocks, jumps into your estimation and sort of how do you balance, like, different views on volatility when you manage it? And I think that’s the eternal question, sort of.

And I think, recently, there was also that event with nickel that also brought this up, you know, where you had this big shock. And the question is, how risky is risky? Because that’s what volatility is. It’s really trying to understand the range of outcomes.

[01:07:43] Niels

Yeah, absolutely. Well, Katy, it has been great searching for crisis alpha with you today. Thank you so much.

[01:07:51] Katy

We’ll keep searching, Niels. We’ll keep searching.

[01:07:53] Niels

We’ll keep searching. Absolutely.

[01:07:54] Katy

The search is on now.

[01:07:56] Niels

The search is on.

But the other thing that people should be fully aware of, especially if they’re relatively new to the podcast, and that is, you know, we may have been doing this for decades, but we’re still students of trend following. I mean, there is always something new to learn. I learn something every week when I speak to all of you clever people from the research side.

So, it is great, and it’s so appreciated by me and I’m sure by all our listeners that you take the time to not only write the papers, but that you come here, and you share all your insights, Katy, so, thank you for doing that.

We have another great friend of ours coming on next week, Nick Baltas from Goldman Sachs. And if you have any questions for him, he’s obviously doing it from a slightly different perspective in his role.

[01:08:45] Katy

Ask him about his co-movement. I want to know about that.

[01:08:48] Niels

Okay. All right. Well, he may. He’s now starting to prepare, I’m sure, co-movement.

[01:08:55] Katy

He has some great work I love his paper on co-movement and crowding. So, ask him about that.

[01:08:58] Niels

Okay, cool. We learned today that gold now is the most crowded trade by fund managers. I thought that was a fun fact, And I’m sure CTA’s at some point will be blamed for something relating to why gold is going up or at some point when it’s selling off. But time will tell.

Anyways, if you want to show your appreciation for Katy and all the other co-hosts, go to your favorite podcast platform, leave a rating and review that you feel that they deserve. We are so appreciative for that. And then, of course, come back next week when Nick joins and we’ll continue our search for crisis alpha.

In the meantime, from Katy and me, thank you so much for listening. We look forward to being back with you next week. And of course, as always, especially around these times, take care of yourself and take care of each other.

[01:09:54] 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, our 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.