The Derivative - Surviving Adversity, Building Resilience, and Evolving a Quantitative Investment Firm with Joe Kelly of Campbell

Starting point is 00:00:00 Welcome to The Derivative by RCM Alternatives, where we dive into what makes alternative investments go, analyze the strategies of unique hedge fund managers, and chat with interesting guests from across the investment world. Hello there. Happy spring, which unfortunately comes with no more ski trips for me this season. But hey, what are you going to do? That doesn't mean we're not going to get into some scary ski stories on this pod, though. Yep, we're back this week with Joe Kelly, a partner at quantitative investment firm Campbell & Company.

Starting point is 00:00:35 And as well known as Campbell is, some of the stories and background of fintech startup during the tech bubble, his role in shaping the firm's collaborative culture were new to me. So come for the stories, but stay for the dive into the quantitative, multi-strat, systematic macro world of Campbell and how they've grown from their trend roots into a multi-billion dollar manager who is approaching things in quite a unique way. Send it. This pod is brought to you by RCM's Outsource Trading Desk, which everyone from startup hedge funds to big firms like Campbell use 24-6 as their outsource trade operations.

Starting point is 00:01:08 Check it out at rcmalts.com slash 24. rcmalts.com slash 24. And now, back to the show. All right, everyone. We're here with Joe Kelly. Joe, how are you? Hi, Jeff. Good. Thanks for inviting me. Good to be here. Good to see you. And we were just talking, you're up just north of the city here in Chicago. North of the city, as everybody likely knows, Campbell's based in Baltimore, but I'm lucky enough to raise my kids around our family and work for the firm from Chicago. But you were in New York for a while?

Starting point is 00:01:53 Yeah. So Campbell, we had an office in New York pre-COVID. So we ended up in Connecticut for a couple of years, did the commute down. Really good situation. But then as is the case with a lot of groups, COVID hit. We actually had a realtor take us out of the lease pretty much for even money. And nobody knew. I mean, this was early, late 2020. So nobody knew if that was going to be a good deal or a bad deal or how long this was going to last. And my wife's from Chicago. So we took the opportunity to kind of use the flexibility going forward.

Starting point is 00:02:33 Yeah, that's perfect. Welcome. A little bit of a cost of living increase from Connecticut to North Shore, probably. You know, we thought it would be the opposite, but it was it was about a flat trade. Taxes, believe it or not, are higher in Illinois or maybe, you know, maybe you knew that. But we had been outside of Illinois for about 18 years at that point, always in the industry, but on kind of both coasts. And so, you know, it wasn't it was it was pretty much an even trade overall. I love it. Um, and as my listeners know,

Starting point is 00:03:07 I'm always going to talk skiing if I got a skier on here. So yeah, you're a skier, but had it go a little sideways two years ago, wasn't it? Yeah, it's actually, my last operation was less than a year ago. It was, uh, 11 months ago months ago. And so, you know, if you've heard this, I apologize, because probably sick of it by now. But December of 22, I was out in Utah, had a fall, we initially thought I got hit. But, you know, to be fair, at this point point my memory's a little fuzzy um woke up uh was knocked out and woke up uh completely paralyzed i think you know this story jeff we just talked briefly so yeah i want to hear a little bit yeah so laying you know laying face up

Starting point is 00:03:59 um paralyzed on a slope and you never know what's going to go through your mind. I think that's the most fascinating part is you don't know until you're in that situation. So I had first was kind of fascination and this, you know, it happened long enough that I could actually think through this. Second was how am I going to remodel my house so that we could we could manage this. And then, you know then you know to give you the very short version i went from you know paralyzed with the wind knocked out of me where i thought you know maybe that was kind of the end to um wind knocked out of me which was the remodeling

Starting point is 00:04:39 the house portion of the memory and then you, you know, very slowly akin to like a, you know, something you'd see in the movies, toes started coming back, fingers started coming back. And, and then the adrenaline really takes over and adrenaline, like you've never ever witnessed before. So I sat up from, you know, sort of of that position this is all over the course of about four minutes but i did realize i was in trouble um what hill were you on again i don't even know if i'm allowed to name the resort but it was in utah i've actually never named the resort on a recording just out of in utah that's good enough prudence and so um waited for help help never came big snow day had to pop my skis on ski down um go to the clinic uh flew home to Chicago with

Starting point is 00:05:36 blood pressure of 190 over 120 and at Evanston Hospital discovered that I had broken C2, which is right at your brainstem. And it usually is fatal, Christopher Reeve's type of fatal. And I'd also broken T2 and T3 in my thoracic spine from the fall. And the guy at Evanston came in almost white as a ghost, great young doctor, put me in a collar. And then the process started. And did he say, how the hell did you ski down and fly from Utah to Chicago with all this going on? Yeah, that's that's kind of. The subject of what I've played out in my mind a hundred times is like, almost like a Ted talk. Um, you know, I, I, from there,

Starting point is 00:06:29 I'll give you the lessons learned in about 10 seconds, but from there, found a world-class doctor at rush, just lucky, um, two hedge fund prominent hedge fund people, um, that we had never known got us into this gentleman, which we're extremely thankful for. One's here in Chicago, one's out in New York. Might have a spine center at Cornell named after him. And from there, two operations, January 23rd, May 23rd, fusing me, and I would show you the picture if I could. Bone healing and then unfusing, of which there's, I think, about four of us in the United States.

Starting point is 00:07:14 Unfusing doesn't sound pleasant. Unfusing was 11 months ago. That was for the benefit of range of motion and at the curse of cutting back into some pretty significant muscle that had started healing. So kind of, you know, and then Jeff, I probably saw you, you know, in June, which was super aggressive and not very smart. Yeah. You were still a little, uh, yeah. Yeah. On stage. And yeah, exactly. So I ended up spending about three months at an organization. I think they're just in Chicago, but Shirley Ryan, which is known as kind of a stroke center, but it's also

Starting point is 00:07:52 a really prominent spinal cord injury center. And I was lucky because I, I was just, it was just bone. So it wasn't, you know, motor function other than, you know, bone infusion. But kind of the two things that came out of that first is sitting in for three months with these folks that do have spinal cord injuries and are coming back from strokes. I have not paid, you know, an extraordinary amount of attention to people that were disabled on the street, in stores, et cetera. Now I have my kids, they realize how incredibly strong you have to be to go through that experience. The attitudes were amazing, truly, truly humbling. And, you know, most of the time, a lot of levity and being around that was incredible because, again, I was one of the ones that was

Starting point is 00:08:45 going to heal the second. And this is more of what the Ted talk is. Um, and Jeff, you have kids. So this will resonate with you. Like resilience isn't like a point in time thing. And I'm not saying like, this is like some big heroic thing, but like a lot of us, like I played youth sports, played a college sport, went through a, you trading career around just fighting through to make a career out of it, climb mountains, et cetera. And all of those little pieces, all of those small things get you off a ski hill with a broken cervical spine. I mean, it is going back to that adrenaline moment. I would just love to impart that resilience piece to my kids, to my family, to anybody that realizes or that doesn't want to go through the hard stuff now because they don't realize that it makes you so much stronger later. So, uh, your body was like going through the catalog and pulling like, Hey,

Starting point is 00:09:49 I need a little bit of this resilient. I need a little bit of this one, put them all together and, and get off this hill by. Yeah. And your brain and, you know, you could sit there or you could go. And, and at some point you can't sit there anymore um and not having a choice helps too so and so was it if it had been an inch this way or that way it might have severed the spinal cord and all that nasty stuff yeah so so the reason why i was paralyzed is normally that shears off and severs your spinal cord um mine sheared off and bruised my spinal cord for no reason. Um, and so that temporary four ish minutes or whatever it ended up being, um, that's the way they described it to me. And, uh, and thankfully again, it wasn't displaced enough where they had to permanently

Starting point is 00:10:38 fuse me. Um, but the, the, this world-class individual at rush hadn't, hadn't done a lot of these surgeries. Wow. And so, million-dollar question, will you get back out on the ski hill? Yeah, I mean, it's a great question. We had a short vacation booked this year, and it affects your kids more than you think, and it affected my wife more than any of us knew.

Starting point is 00:11:05 So we punted this year, but I'm 100% going. I mean, if you're a skier or any of your listeners, it sounds like you've had a couple people on here that are passionate. I was a young kid in Michigan, youngest of seven. We skied. So that was my babysitter for a few years growing up. So I got to get back. Yeah. And it, it had the weird effect on me after you told me that story. And I was like, and as you told me then, I won't say too much, but it was on like a

Starting point is 00:11:36 blue, right. You weren't doing anything crazy. You were going down a groom blue from what I remember. Yeah. That's why we, that's why we think I got hit is because it was a little more of a congested area yeah so someone might have come out of the woods and so it had the weird effect on me i was like well i'm gonna keep going full full bore right if i could get hurt just going down a blue i might keep going off cliffs and doing this doing this crazy stuff because if it's your time it's your time and who knows how it's going to happen yeah it might not be the best uh for all the people who are listening here for risk management techniques, but my twisted brain, that's what I got out of it. I'm like, well, heck, if that can happen, I'm just going to keep going hard.

Starting point is 00:12:16 I appreciate that. And I don't disagree. If you would ask me what I want to do when I'm 85, it's I want to have enough, you know, mobility to ski. So we'll see. We'll see what the next couple of years old. Oh, well, we're glad you are up among the living and get to move your neck around. Yeah. Yeah. Thank you. So you mentioned there, I'm going to dig into that. You mentioned some travails back when you were in your trading days. Want to dig into that and give us a little bit of the personal background? Yeah, I don't. Maybe that was a little bit of a misnomer. I think

Starting point is 00:12:59 it's the same sort of challenge every young trader goes through breaking into the industry. A friend of mine worked for some independents in the single name pit over at IBM and others. And he described it as kind of an English major learning three-dimensional chess in Chinese and there's money on the table. And this is when we were all in our 20s. So so like a lot of people, I came out, I went to Penn out in Philadelphia, came out, was a Midwest kid, ended up interviewing with O'Connor and Hull. Um, and they were kind of looking for, you know, two things, mathematicians and athletes. You were lucky if you had some of both. Um, and so I ended up at Hull. Hull at the time in 1993 was the upstart if you remember um

Starting point is 00:13:49 40 people and blair hole uh famously kind of bought the bottom in the s&p uh during black friday black monday and um we have a blog post about that it was actually the mmi index it was the mmi okay right and there was no volume in there. So when he put in, he bought like 10 lots and it basically created this arb where everything kind of bottomed out. We'll put a link to that. It's a cool story. Yeah. Amazing.

Starting point is 00:14:15 And so I got to spend some time around him. The firm was just kind of getting bigger. Art Margulis, a bunch of other people that are kind of big names in the Chicago financial engineering community, et cetera. And so came up through that organization for a couple of years, spent some time in Germany, back here in Chicago. And then, as you recall, Jeff Hull got bought by Goldman in 98. I moved on to a smaller prop shop and then founded at that point. And we were kind of working on this

Starting point is 00:14:46 in the background at that stage where two other people in the trading community and I came up with this concept. This is tech bubble now, 99, 2000, 2001. And our friends and the industry of tech had created about a trillion dollars of wealth that was paper wealth in employee stock options. And using our background, we created what you would call today FinTech, kind of a decision engine sitting on top of a broker dealer to provide liquidity within the construct of the corporate agreements to employ stock option holders. So this could go through benefits. It has people retaining instead of paying tax, achieving their goal, et cetera, all these amazing things. It was kind of option overlays or when you should add the option overlay?

Starting point is 00:15:35 Just straight up hedges that were tax aware. And at the same time, because there were hedges, you wouldn't be taxed on the original exercise because most of the time the extrinsic value was just being given away. And at the time, if you remember, and this goes a long way back, but post-tech bubble, there was a famous article about JP Morgan, sorry to name names, but buying long-term options from Microsoft employees who wanted to sell. And they were selling them for a dollar because they were underwater and there was eight years and all of the traders were saying, wait a minute, you're not underwater, there's eight years. And JP Morgan did an amazing job of providing liquidity,

Starting point is 00:16:16 but also they were hedging their own long-term options in the market and that's what we were doing. What were some of the names some of the like pet.com and things crazy like that or it was um yeah more mainstream ibm you know we found out lou gerstner had a huge straddle on or strangle on at one point and that kind of drove some of our thinking around the c-suite having access to these things and the enormous pool of non C-suite that had emerged tech bubble. So like there was a group called Sapient that was one of our vendors that, you know, that's probably as edgy as it got. That's.com and the others, you know, we, we were lucky.

Starting point is 00:16:59 We didn't do too much there, but the SEC needed to approve a no action letter on this. We spent about seven years with the SEC getting that process up to speed and eventually they approved our concept. That was three years after we ran out of money. That's my old line. No one doubts you're a pioneer. It's whether you'll make it over the Rockies. Yeah. Well, we were we were young and frankly, addressing a need that people didn't know they had, which is the death knell for startups. Right. Don't don't don't try to, you know, create something they need to demand. Yeah. Well, and since Spider Rock has now done this, right? And done it well. Yeah. Yeah. So what that did though, we all know plenty of traders that have made a lot of money

Starting point is 00:17:53 and they have great careers, but I needed a little bit of a broader horizon. I don't know why. So it took me from being a trader and it was one of the only things I knew that could take that skill set but also allow me. I had an argument with Myron Scholes on the phone, and in person, he slammed his book and said, we got to make this a private enterprise. And our whole theory was cutting out the middleman and making it a democratized enterprise. And so we had some disagreements with some very big people. Super fun. a democratized enterprise and so we had some disagreements with some very big people super fun um but it it made me realize i want to run businesses in alternatives rather than be the you know on the on the end of the uh trading spectrum and so along along came rotella at that point so sorry for the very long story no worries it's We've had Rotella on the pod before, Bob.

Starting point is 00:18:45 Yeah. Jack Deesh or who? We had Bob himself on way back when, but then Jack Deesh afterwards, yeah. Okay, good. So Bob Rotella, Rose Rotella, I mean, salt of the earth, great people. Robert had moved from Chicago, Jeff, as you remember, out to Seattle. And there were some questions about whether that would be successful. It actually made him work harder because now he was in a place he loved. And Joe Canapari and I kind of built the business back up after that move because they lost some European business. And it was great. We launched Molinero. We launched

Starting point is 00:19:21 a few other people in the industry that you know, Jeff. And in 2011, Robert decided that he had run the course and didn't want to manage outside capital anymore. And that's where Jagdish kind of moved into his seat. And from there, I kind of bounced around. One of your questions earlier was why Campbell. During that period, we were all presumably doing the same thing. Campbell was not. Campbell was heavy into macro in the early 2000s, heavy into Steadarb in single name equities. And when you went out and you thought you were competing, you weren't. Bruce Cleland was doing a really good job. Rotellas were very friendly with the Campbell folks. So I got to know the business, but I competed a lot with them. And so I got to know it from a very sort of early stage,

Starting point is 00:20:12 bounced around for a couple of years post Rotella, Russell, a couple of private enterprises trying to learn the private side of the business. And then Campbell, 2016, had a seat open. And Mary Foreman at Morgan Stanley called me up and said, would you guys move? Yeah, would you guys move at that point from San Francisco to Connecticut for Campbell? And I said, immediately, I said yes. How old are your kids at that point? At that point, they are three and one. All right. So that made it a little easier, right? If they've been on the high school team or whatnot

Starting point is 00:20:49 and friends. Yeah, big time. And so if you know much about San Francisco, obviously it's a tech community. My wife is also in the business. The hedge fund industry out there is not large and very sort of long-short equity-centric, impact-centric. So for me, getting back to Quant, the only option in San Francisco at that point was BGI. Getting back to Quant was my DNA. And so I immediately jumped at the opportunity, moved to Connecticut, obviously started running the business. And from there, you know, I would describe sort of the modern version of the firm, not attributable to me, but it was, it was being created, like, you know, what I call 3.0, right at that point. And it took a couple of years, but we're kind of, what we describe as not, certainly not a finished project or product, but like the modern view on who we want to be is, is now kind of in place after several years of working through that.

Starting point is 00:22:00 I've always admired Campbell from afar because it's very non-founder-ish, for lack of a more scientific name. Most of these firms, Winton, founder here, blah, blah, blah, founder, right? It all ties exactly to the founder where, at least from my view and a lot of your investors, I don't know, do they even know who the founder was? Do they care? Yeah. You guys have done a good job of basically lessening that impact and like, no, this is a firm. Here's what we're doing as a firm. So talk a bit about that. That's intentional, obviously. Yeah, it is. I mean, there's an ownership team. I'm super lucky to be on it. So in some rooms,

Starting point is 00:22:38 I'm the marketer. In some rooms, I'm a partner. But we spend all of our time. I mean, when we want to talk about where we want this to go, it always ends up back in culture. So the nuts and bolts end up being, there's five people on an IC that run the portfolios. That's intentional. Very low ego across the firm. We're sort of lucky to have kind of that as a derivative of the way we've been running the firm over time. But from the culture perspective, having this concept where, so if we essentially kind of bucket our strategies across broadly kind of trend, as you'd expect, while macro, short-term, market-neutral quant equities. In some firms, you would expect the evolution of this to be siloed. And we're not

Starting point is 00:23:33 the only one to come up with this concept, but we think there's more alpha and kind of more of a force multiplier in having those teams collaborate in a peer review process when ideas get documented. So by the time, you know, that comes into the portfolio level, there is no individual ownership of anything. It's shared IP across research, that very low IP leakage, believe it or not, which is always a question, but what ends up happening and sorry, Jeff, believe it or not, which is always a question. But what ends up happening,

Starting point is 00:24:07 and sorry, Jeff, I talk about this all day long, because I think culture drives returns. But what ends up happening is when, let's say you have two models that have been developed by two different, three different teams, four months later, when it gets in the portfolio, because there's sort of no individual fighting someone else for risk or trumpeting their efforts over somebody else's efforts because it's collaborative. Their promotion is based on it or whatnot. Yeah, exactly. Those teams are actually incented not necessarily to choose their model or fight for their model, but look at in the portfolio context, what's the most additive model? Maybe it's both, but if we have to choose one or the other, everybody on the research team is aligned with choosing the optimal model within a given

Starting point is 00:24:56 portfolio. And that's optimal can be defined a number of different ways. But we come back to the management level and the way that works is because we compensate everybody based on total P&L. And so now, and that's kind of the kicker, right? Because now, and we have superstars, we have amazing minds in research that want to be part of that culture as much as they want to be part of something that pays them millions and millions and millions of dollars. Because oftentimes when you get into that second culture, the pressure's on, you don't get to talk to anybody.

Starting point is 00:25:35 And your PhD is your PhD and kind of ends there. Not the case here. So that draws people in. And then investors think about it. Investors know the optimal models are going in and they know our research team's getting paid on their portfolio. So again, this was not in place 15 years ago. I was lucky because the wheels were turning. That collaborative piece was already in place 10 years ago, but it's really come of age. And so contrast it with that 15 years ago, it was more of a set of models. I think, and I wasn't there, so this is second hand information,

Starting point is 00:26:15 but I think they paid more people on individual efforts when somebody wanted to trade FX options or, you know, et cetera. And so it was a little more piecemeal. Bruce, who was a really great leader, liked that framework, right? He liked to have people compensated for some great idea, but, you know, GFC and other areas, you know, teach you that, you know, egos can be a good thing. They can be be a toxic thing and so we're trying to have enough of that but also incent people to have sort of that learning mindset and so i'm not sure yeah yeah but a lot of these firms that people get upset right because they're like hey my right i didn't get bonus on my pnl because i got netted out against this idiot who was trading uh new

Starting point is 00:27:04 zealand swaps or whatnot. But so, right, they get upset and they go and start their own firm. So your goal is, hey, it's not your P&L. You're also not necessarily getting netted out against the really bad trade, but the whole portfolio. Now you get to participate regardless of how you perform. Yeah, you got it. you know i mean then management wise it's pretty easy to to find the people that you know maybe could hide under that we have none of those um we only have 65 people yeah um so we have we have we're lucky we have more outperformers than underperformers and um and people like i people, like I said, not for everybody,

Starting point is 00:27:45 it really isn't, but people have really sort of bought into that, that are with us right now. And it seems like an anti New York model, right? Like it seems like a, not to upset all my New York friends, sorry, but right. That'd be like, Hey, I want alpha dogs in here and you fight for what's yours and you prove your worth. Yeah. It's a you know it's it means a lot of things but it it might have a harder road in a big discretionary shop whether it's macro or long short equity where people are sourcing ideas um you know and they want to get paid on doing the heavy lifting on that individual idea. But even in the quant world, I feel like we're starting to see these pod shops that are mostly all quant. Sure, maybe they have one guy doing some discretionary stuff, but they're putting together multiple quant models and each pod is running one of those quant models.

Starting point is 00:28:43 So that has grown significantly, right? Do you guys consider yourself sort of a pod shop? Luke Gromen, So it's not necessarily pod, but we do, I mean, we use systematic multi-strat internally, externally, the nuances with investors come back to that collaborative culture versus competing for sort of risk. And then honestly, sometimes those work quite well together in a portfolio. And so we're trying to hold off on the individual. And again, we think there's alpha. I mean, I'll give you an example. Our macro team thought there might be some opportunities to trade intraday data in some pretty historically long folds in macro. And they couldn't do that unless our short-term team did some heavy lifting with them. weak hold W-E-E-K holding period macro ideas using intraday data that got to market faster

Starting point is 00:29:49 because of that collaborative effort. You weren't interested in holding the W-E-A-K positions? Exactly. I like that. And so you mentioned that they're coming in and the investment committee has to identify them as optimal. You said there's a few ways to define that, right? So as a quant shop, is it hard to just be like, oh, there's the best sharp or whatever risk adjusted ratio you use. Let's plug it in. Like, what are some of the other metrics that you use to say, how does it have to help the portfolio? Yeah. Yeah.

Starting point is 00:30:24 I mean, it's um the i think the days of having an arms race around model count are over you remember those um they used to drive me crazy at you know at rotella and other places where you know sometimes you'd add a zero sharp model just to say you had more models and then, you know, inevitably correlations weren't what you thought they were during periods of stress and you weren't well diversified, et cetera, et cetera. So for us, you know, we identify a model as a unique idea, you know, timeframe independent, right? Because you can apply the same idea to five different time frames.

Starting point is 00:31:05 That's not five models. We try to be just conservative about that. We don't think that's the way you need to sort of compete as an arms race. But having, you know, the expected correlation of a given strategy, you know, being zero or negative can be a lot more interesting, even if, you know, the sharp is not two. So there's a lot of moving parts around it, but it's, you know, it takes a really robust risk framework to feel like you have gained a level of comfort around your assumptions. That's obviously a huge, huge concept within these development teams. And then obviously walking into risk rather than running into risk, implementing it in the portfolio with lower levels, and then post-launch reviews, et cetera. And so it's a little bit of a moving target, but healthy, sharp, not just diversifiers. Diversifiers,

Starting point is 00:32:14 that really help. And now we're finding negatively correlated models to the other pieces of the portfolio, which really add not only diversification, but we're not looking for capacity, but it could be in some of those areas as well. And do you ever do anything? I think I used New Zealand as the example before, so I'll stick with New Zealand. Do you ever be like, hey, we need this New Zealand exposure? We don't have anything that's exposed to that geography or that level of GDP or that currency type? So you and I and your team have talked about China and we can walk through that. That's more of a market efficiency rather than a regional exposure, alternative market type of idea. But

Starting point is 00:33:01 for us, just like a lot of other managers, if you have a momentum signal conceptually, you should be able to apply that pretty universally. But then you get into short term and macro, leaving off market neutral equities for a minute. And there are some very market specific ideas that use different sets of data that are not applied universally, but we don't, and we don't tend to say, you know, we need South Africa exposure. Most of the derivatives we trade, you know, are not regionally exposed, but they're, let's say commodities or whatever else. But it's more around, you know around what is the fundamental underpinning of why this should work,

Starting point is 00:33:47 why it shouldn't decay overnight, and why it would be additive to the portfolio. And then we had the guys from Floor & Court on a couple weeks ago, and they're going into all sorts of unique stuff. So if you started to lift the lid on that and be like, hey, if we could get access to this via otc or swaps or whatever this would be a great uh additional model on this product i give those

Starting point is 00:34:13 guys a lot of credit um i really do and as well as others that were early to that game as you know you know the operational lift around that is significant. You have to invest in that. However, I would love to have a beer with them and talk about counterparty risk and how sort of inherent to our industry is the lack of counterparty risk, but there's some alpha there. So how much, what's that trade-off? Over 50 years, are you getting paid enough for that? Yeah. And we all know that there's firms that should not have gone out of business that do.

Starting point is 00:34:51 So at that inflection point, and this is circa 2018, we could have gone down that road or we could have gone down the road of instead sort of looking at our skill set and figuring out what markets we do have an advantage in rather than applying some universal concept to as many markets as we could. And we took that route. We took the route of, look, we have a very specific skill set that we're trying to build inside the firm. A lot of it's sort of non-priced skill sets that are specific to sometimes one market, sometimes five markets. And we went down that road of trying to generate alpha that we think could be more sustainable than being the fifth or sixth or seventh player in XYZ energy market that is seeing the alpha erode quite a bit. If it's like momentum based alpha. And I'm not saying it is, but conceptually, the more players, the more that could versus kind of our approach.

Starting point is 00:35:59 You took that path to the right, to the left, whichever one it was. So that was the and you bumped on this a few times, but let's just tighten it up. So there's trend, there's macro, and there's long short equity, or the three main buckets, you would say? And short term, that is really, really very unique. So those are the four buckets. And short term, just short term, just in and out, like we're talking high frequency trading

Starting point is 00:36:31 or no, within WDK? Right up against it. We're not looking to provide liquidity. We do, you know, we are changing signs a few times a day, sampling a lot of data, not using just a shortened momentum signal in short term or ball breakouts, all the things we kind of grew up on. More recently, it's been around things like where is there more information

Starting point is 00:37:01 and data that might not be, you know be the last hour? It might have been a day ago, it might have been a week ago. So that's been a hugely interesting area. And as you know, the degrees of freedom in short term, when Robert Rotella went into short term, and we would say, from one day to two days, you got twice the opportunity set and um and so that hasn't changed at all there's so many opportunities the two sort of sort of most diverse groups for us are which can can be a lot of different ideas or short-term and then this just popped in my head sorry but do you kind of view these right these four pillars you kind of view trend and macro as more like long ball for lack of a better term and long short equity and

Starting point is 00:37:53 the short-term stuff as kind of short ball i don't want to put it as short ball but you're kind of those are more generating this consistent return and you have those other two pieces that are positive skewed and looking for the outliers yeah i, I would put trend in that camp. I'd put macro neutral to that camp. I would put short-term loving volatility, like as you said when we were just coming on around your example around March 23. And I would say, you know, quant equities doesn't care about that camp. So we have two that don't care. And, you know, one that loves volatility, but it doesn't have to be, you know, in any, it could be, you know, a reduction in stress. It doesn't have to be running into stress. Whereas momentum, as you

Starting point is 00:38:42 know, you know, plays that momentum role in the portfolio. And so not surprisingly, and this is like our multi-strategy, right? We do subsets of this if investors wanted pure trend, for instance, as a risk mitigation, et cetera. But our sort of biggest capabilities portfolio, the real goal is psychoagnostic. Those four tend to rotate

Starting point is 00:39:12 into opportunities very differently and that's why those are there. Imagine a simplistic multi-strat that's got carry and then trend on the other side and when carry gets hit trend is likely to do well and they kind of you know here carry's there to pay the bills and then trends there to cover it when it gets hit you're saying that not necessarily that level you're just

Starting point is 00:39:36 saying hey these are all non-correlated to negative correlated and they work yeah when i mean that's a perfect example when we all got to know camp in 2002, it was trend carry instead of. And now we have over 100 variations on different techniques underlying all of this. And we've cycled through hundreds over the years, right? So that's just the state of it today. But obviously, when you look back at that portfolio versus the portfolios today, the only one that is very similar is kind of trend. Even carry has evolved quite a bit. But everything else has evolved so much that it's fun to have data going back that far, but it's a totally different process now. But it's a great, I mean, you nailed the example. Yeah. And how do you, how do you weigh and consider like, okay, if these, we have all this cool stuff,

Starting point is 00:40:34 do I get to a point of like T-bill rate of return, right? If I have all this cool stuff, that's non-correlated. Am I really, if I'm taking no huge position, am I taking any position? Yeah, no, it's, it ends up being this concept of embedded leverage, which again, isn't a Campbell thing. But if we're going to deliver you net 10 ball, let's say, which in today's environment, even in my travels in Japan two weeks ago, single digit vol used to be the bogey. Now, whether it's the hedge ratio or other things, you need to be able to deliver a healthy amount, but a healthy amount is also evolved to a stable amount, right? High sharp depends on stable vol, normal distribution, et cetera, et cetera. So if you can deliver stable vol and net all of that complexity

Starting point is 00:41:25 down to a 10, you're doing something special. And that comes back, we have an individual named Grace Lowe on our team who heads the risk piece. And that's really the part that does rotate risk across these several parameters, models, markets, et cetera. And to do that and nail that stability piece, I mean, that is where you get really happy investors is minimizing those outliers. You might not always make as much money as

Starting point is 00:41:56 you want to, but not having outliers over time is great. Both sides. Both sides. So you're not seeking to make 58 in a month in cocoa or whatnot yeah no and and that's where and i think it's really it's really interesting because we serve two different masters specifically to camp so one master is you know give me an absolute return,

Starting point is 00:42:26 you know, psychoagnostic, one sharper, better on 10 ball, you know, et cetera, et cetera, with all this diversification. That's sort of what I would describe again. And you've heard me say this a couple of times on panels, et cetera. If you have to use the term CTA, that's like CTA 3.0. And don't forget, it includes quant equities. There's a material part of the portfolio, market neutral. But we also serve the risk mitigation, you know, Makita driven and others. But I think they did a great job leading the industry into this risk mitigation class, long bonds, trends, et cetera.

Starting point is 00:43:11 And we've got quite a few clients there as well. And that goal, as you describe it, is put up the 58%, maybe not in one market because we haven't, but the goal is convexity. Sometimes convexity comes at a price. And that's where certain firms have done a really good job talking about, especially with boards, portfolio level. Don't look at these line items. Portfolio level impact of this is truly sort of incredible. And going really well for people right now, obviously. And do you feel some of the stuff you've added, right? So what we'll call loosely crisis period performance, right? So that risk mitigation, when there's a 2008, when there's a problem, this should be there.

Starting point is 00:43:57 But adding some long-short equity, adding some of the stuff that's not necessarily convex and positive skew, how does that right versus what you used to be of the big trend like how do you weigh out of like okay we've added all these pieces and we're very cautious to not really ramp down the convexity yeah no it's um sort of two things come to mind first is transparency right so you have to you have to walk into meetings and frankly, distill what that goal is, which leads to much better conversations around what you might be able

Starting point is 00:44:30 to deliver because we don't, nobody wants, you know, we all work too hard to have these very short-term relationships if you're not fully transparent. I mean, that is a enormous tenant of the way we run the business. And then the second is the way we think about it, and there's no magic number, but let's say because we can net some very diverse approaches down to a 10 annualized risk, we have to run each of those at higher leverage just due to the portfolio effects. And think about what you can possibly run trend at inside of a portfolio that maybe has zero correlation to everything else. But, you know, again, you're going to net to a 10. 10 is your number.

Starting point is 00:45:19 Sorry if I'm giving away trade secrets. I don't think I am. But because of that, when it's working, you get, and I know I can't name numbers, but you get quite a significant bit of upcapture in that portfolio as risk is consumed into that piece. And then how good is your risk system at now locking that in, rotating risk, and not having that give back that pure trend tends to have. And you try to capture the middle, as everybody knows, get into it, capture the middle, and have something else rotate in maybe to capitalize on that reversal.

Starting point is 00:46:02 And so we do think about it as X percent up capture, 90% up capture, and let's say 10% down capture. And thankfully, it's kind of gone that way in the past couple of years. And when you get with an investor and they say, hey, you made X and this firm made 1.3X as a pure trend follower, they should have been in that strategy in the first place if that was what they were looking for. Yeah, right. You're going to be unhappy. Are you going to be as happy on the way down when they're 1.3X the loss or maybe 3.3x on the loss? Yeah. I do think, kind of as a sidebar, I think some of the bigger pensions that have gotten into the space, and let's just call it pure trend, and have done well, I give them credit. I give everybody involved because some of them

Starting point is 00:47:00 really needed to have diversifiers work. And as far as we've seen, it seems to be working. So it's to credit, frankly, to everybody involved in the process, the teams there, the consultants, the managers, it's so cool to see it work. Well, 23 made me so nervous because I'm like, here we go again, right? Gains in 08, losses in 09, gains in 22, losses in 23. Everyone's going to bail on it again. But I think to your point, I think they'd be a little more educated and be like,

Starting point is 00:47:31 hey, this is part of it. See it through. Yeah, no doubt. And so, and I mean... Sorry, go ahead. I was just going to say, I mean, you do have to step back every once in a while and think about the effect of all of this. And we try to on people's retirement, et cetera, et cetera.

Starting point is 00:47:50 It might kind of be cheesy, but that's rewarding. So, again, I think it's just kind of a cool couple of years that should lead to better results. I mean, we're talking about 20, 30-year timelines, but if they stick with this, it's going to have a real impact. I always get mad when some hedge fund blows up and there's people cheering like, oh, those fat cats lost their money. I'm like, it's actually some fireman's pension that probably lost their money, right? It's not these billionaires that lost their money. It's some pensioner. So be careful cheering it lower. So I want to get into, if we can,

Starting point is 00:48:35 what you guys call, now you just said quant macro, but before you've said systematic macro. So quant or systematic macro versus trend. Some of the differences there you've talked about in the past. So let's just dive into that before we wrap up. I think you were the one, we had that panel in Chicago again, shortly after my second surgery. I think you were the one in the audience who kind of nailed sort of the non-price comment about what's the difference is because we're all kind of, kind of, you know, riffing on what anybody else was saying, but for us, the way we describe it unfairly or fairly, because I know you, you know, you work with a lot of our peers and this is going to be a broad statement.

Starting point is 00:49:16 Don't take it personally, but you know, we want to separate kind of momentum from macro. I mean, there's always going to be that impact right i mean i'm not going to stand here and say it's not but there are certainly some quant macro systematic macro managers and concepts in the marketplace that rhyme a lot with momentum. We try really hard to take a lot of relationships that I would describe as the way economies interact based on either the dominant driver of that economy, a given commodity is the easy example, or different trade agreements, different impacts, obviously like economic release data, differentials around whether it's something simple like a given country's yield, which we don't do a lot of sort of curve trades, but also kind of the direction of that without

Starting point is 00:50:27 sounding like it's momentum, the delta of that on a very, very short-term basis. And so I guess my point is take concepts that are not price-related, but price can certainly help or be a conditioning factor on it. Take concepts that distill dominant pieces of an economy and how that's going to have a knock-on effect across assets within a given economy. And what is that going to do to its partners, trade partners, et cetera. And so that's kind of what we mean by macro. And I touched on this earlier. That means that we trade certain data sets in single markets because it just is very specific to the behavior of XYZ market. And there are some we trade more broadly. you know, some of the reward and risk that used to be a big part of carry and now, you know, make it a little more dynamic around how that changes through time. And, you know, time can be

Starting point is 00:51:31 a day, um, which is very different than that, you know, year long, huge left tail, uh, previously in the macro constructs construct. So the one, the one give by me is I will include carry in macro, but we try to do that a little bit differently so that it's not just a premia. In fact, we try to actually constrain from more known methods and just even in momentum, believe it or not, Jeff, we try to constrain against maybe like an indexed factor and try to find where we provide alpha and allocate risk to that. Even as we described it, at the expense of just raw convexity for the benefit of alpha where we can get it. Does that make sense? Sorry for the long answer. No, can get it. Does that make sense? Sorry for the long answer. No, no, no.

Starting point is 00:52:28 I like it. And the example we used back on that panel was, and well, I'll start with the confusion perhaps for investors and people looking under the hood of, you could be in very similar positions to momentum, but for different reasons. So I think the example we used was the trend followers long copper because it the 30 days above the 100 day moving average or whatever and it hasn't come back

Starting point is 00:52:52 down yada yada some classic momentum signal and you guys are long copper because the chilean peso did x y and z right and but you're both long it and have been for weeks. Who knows? So that's sort of the confusion of like, but I guess that same signal is used as exit or do you get in on that macro factor and out on some other factor? Same. In a continuous signal. So that may actually be a position that's 20 models.

Starting point is 00:53:20 And, you know, because there's going to be scaling in and out daily, we'd like to think that just, you know, that benched against a, you know, a very simple trend signal should over time provide a higher sharp, even if it's the same opportunity set. So that sort of portfolio effect should generate a higher sharp, and that's not rocket science. Then just that singular signal, even if we're both long. And inevitably, what we've come across in the market is there are more people in macro that are directional, and there are not a lot of relative value systematic relative value players and if you just look at that increasing your number of bets if you look at the diversity of that etc again all theory but like sharp should increase by adding more relative value approaches because we have we have momentum over here yeah um and so macro is place. I think I said this on stage, not only is it data

Starting point is 00:54:26 and the opportunity set, but relative value is where we push very hard on investors. Because it's hard to tell five managers that have been in business for more than 20 years apart when we're all using the same terms. Yeah, exactly. And so relative value, we'll use that copper trade still. Do you seek out, does it have to have an opposite pair? Do you have to be short silver or something on the other side? It doesn't have to, but it might use, it doesn't always have to be that country's FX, right? It might be a data set that looks like relative value data set. And then there's also this conception of there's relative value portfolio construction. So

Starting point is 00:55:09 at the model level, it might not have an exact pair, but at the sector level, subsector sector portfolio level, it looks like it does because there's different effects in the portfolio. And so that gets a little bit in the weeds. Yeah. We're saying that same fundamental input could be, could trigger a short in some roughly related market. Yeah. It's not, it's not like you would consider pairs trading in, in equities where,

Starting point is 00:55:40 where there is a population or a single name. But it's more of the portfolio effect at times. At times, it's, yeah, it is an exact pair, but at times it's the portfolio effect. And then it also becomes like a bit of a voting machine, right? So if you have the same position in trend and quant macro in short term for a few hours, like that's just the strength of that signal, right? Multiple, multiple models have all said belong X. Yeah, those are all independent. Exactly. With the exception of risk management where we see,

Starting point is 00:56:12 you know, we've got, as I said, we've got a process that looks at our position, looks at the market's position, you know, tries to come up with kind of a context for how much money we've made or can make versus the market in a given trade. And then it'll constrain across rather than any individual model as well. I'm jumping a little over the place here, but what are your thoughts on replication? So everything you guys are doing, how complex it is. And if I write, people are like, oh, we can replicate this by being long euro dollars and short gold and whatever. Yeah. Impossible.

Starting point is 00:56:53 Yeah. Short term, you might get lucky. Long term, you know, we're moving every day too. So we're not reconstructing the portfolio every day, week, every year even, but it does look very different today than it did two or three years ago, five years ago. And again, that comes back to transparency with what you're trying to do with investors. But yeah, you could come out and replicate our 2002 portfolio today. No problem. Maybe not even the stat RPs. Way too complex. And I used to, even back to Robert Rotella, we used to say there's just no way. He was very anti uh sma for that reason and and we we would say time and time again you just you couldn't do it at that point with him but you're you're part of an index

Starting point is 00:57:53 that is being replicated so again and the interesting thing andrew beer was on the podcast to go back and listen to that one but right he's saying hey yeah i know they're doing tons of complex stuff but if you regress the last quarter's performance these three markets can explain the performance can air quotes explain the performance right and there's a huge lag effect and all this stuff so i i'm with you that i'm dubious but it's yeah a really kind of you know sarcastic answer would be you know replicate the short-term traders index by doing nothing. Yeah, right. But there are some world-class managers making up that index that you'd be lucky to invest in at any point.

Starting point is 00:58:40 Yeah. And so there's a little bit of a nuance between selecting the individuals and uh and replicating the total yeah yeah if you will and again like i'm not you know there are way smarter people than i am doing that um but that's just my one of the funny little piece there is he needs you guys to continue to be super successful so that he can replicate if you go away there's nothing to replicate. Good point. So what's next for Campbell, for you?

Starting point is 00:59:16 What's on the horizon you guys are going to move? You're saying this is always changing, so I'm sure the research process is out of this world. Talk to us about what's next. What are you guys looking at yeah it's good it's um so we sort of come up with themes i mean at the end of every year you sort of look back and say what do we do what's our thing going forward what's currently in the pipeline that's always changing i mean some things just don't work some things work and they're pushed through and then what's your next thing. So sort of, you know, thematically right now,

Starting point is 00:59:49 and this isn't going to be surprising, but we, we have upgraded quite a bit of our short-term capabilities to have, you know, a lot more of what we call data density that at a minimum validates what we're doing across a lot of different parts of the portfolio. And that sort of the most optimistic leads to some really cool ideas, as I've described it around information and data, et cetera. So short-term is going to be, if we came on next year, it would be short-term. The year after, it's a big part of where we see opportunities even now. And those have played out, again, in certain pieces of the market where there was high stress over the past year and a half.

Starting point is 01:00:35 So it's self-fulfilling. I mean, it's sort of reinforcing that effort. The second, and I laugh at the guys I work with, you can't get through a meeting. If you don't say AI at a meeting, I'm going to take something away from you. But finally, kind of true applications for it, like code review, like engineering, sort of supplementing engineering, we're using it right now as a supplement to risk management to identify market themes that maybe as humans, like we think we know what's going on, but there's five more ideas out there. Some of them are completely irrelevant. You know, maybe there's a few that are worth paying attention to.

Starting point is 01:01:21 This is going to be the year of event risk, not surprisingly, because of the election, but then how does that impact certain markets, et cetera. So that's something we've had in place for a little while now. The event risk piece we've had in for about almost 10 years now. And so AI across risk, across engineering and code, not for us, and I'm happy to say this right now, not for client, sort of not putting risk capital behind it right now. We see efficiencies rather than pushing it into ideas that we don't know the fundamental underpinnings of, if you will. So that's something that's going to be really interesting to watch because everybody wants to know if it's a risk. But at the same time, we could develop some really cool techniques that right now are being done by 10, 20 people at certain firms. And I called it, and I did this an hour before we got on, Jeff, with our chief technology officer.

Starting point is 01:02:29 I, at times, kind of call it the great equalizer. And it's been a lot of technologies over the years that we've called the great equalizer. But it's really interesting for that reason. And you know better than I do, the firms that might be saying they're putting risk capital behind it i i we're not there yet um kind of hand on heart yeah and the few that i've read about right man ahl that was like three years ago they were in an article and they're it basically told them to buy the s&p after trump got elected i was like that's it that doesn't sound very, right?

Starting point is 01:03:06 Because they were like, oh, all these dips have been bought. And so the AI said, buy the dip. Like, well, that doesn't sound very sophisticated. I'm out. Yeah. Yeah. We do see, you know, again, in research for models that have capital now, but are not using this, you know, is there a more predictive technique which is kind

Starting point is 01:03:25 of a holy grail especially around look back etc um to be determined so that's that's sort of another and then as you said it just gives it could be like you have a hundred extra employees or something right just like manpower yeah i mean everybody's worried about taking their job but what about the jobs that you know the sort of jobs that adds to current teams in a, you know, in a, you know, sort of a subtle way, not real jobs, but obviously capabilities, efficiencies, et cetera. So that's significant. And then firm level, we want to keep this culture going, you know, we're limited capacity. So we've, we've got a couple billion left to capacity outside of those pure trend strategies that tend to scale quite a bit. And so eventually we'll start exploring other asset classes, the accident to trading to running a fintech.

Starting point is 01:04:34 It's fun running this firm. And so that was kind of my goal. And it doesn't always work out, but it's going in that direction really well right now. Well, you're a pro. Br brought it back full circle to the accent. Well, we're glad again that you guys are doing what you're doing. Keep it up. It's been fun to watch from afar and let's get out on the slopes, me and you one day. I'd love it. And we appreciate everything you do, Jeff. Thank you very much.

Starting point is 01:05:05 Thank you. We'll talk to you soon. Okay. Thank you. All right. Thanks. Okay. That's it for the pod. Thanks to Joe and the folks at Campbell. Thanks to RCM for sponsoring. Thanks to Jeff Berger for producing. We'll be back possibly next week, but more likely the week following that. Peace. You've been listening to The Derivative. Links from this episode will be in the episode description of this channel. Follow us on Twitter at rcmalts and visit our website to read our blog or subscribe to our newsletter at rcmalts.com. If you liked our show, introduce a friend and show them how to subscribe and be sure to leave comments. We'd love to hear from you.

Starting point is 01:05:51 This podcast is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. All opinions expressed by podcast participants are solely their own opinions and do not necessarily reflect the opinions of RCM Alternatives, their affiliates, or companies featured. Due to industry regulations, participants on this podcast are instructed not to make specific trade recommendations nor reference past or potential profits. And listeners are reminded that managed futures, commodity trading and other alternative investments are complex and carry a risk of substantial losses. As such, they are not suitable for all investors.

The Derivative - Surviving Adversity, Building Resilience, and Evolving a Quantitative Investment Firm with Joe Kelly of Campbell

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.