03 Top Traders Round Table with Marty Bergin, Grant Jaffarian, and Mike Boss - 1of2

Niels

Welcome back to Top Traders Round Table, a podcast series on managed futures. My name is Niels Kaastrup-Larsen, and I’m delighted to welcome you to today’s conversation with industry leaders and managed futures which is brought to you by the CME group. I’m joined by Marty Bergin, the owner, and president of Dunn Capital Management, Grant Jaffarian, portfolio manager of the Advanced Trend program at Crabel Capital Management, as well as Mike Boss who is a director within the Capital Introductions Team, at Société Générale here in the U.S.

First of all welcome, and thank you for taking time out to join me for our conversation about managed futures. Before we jump into the specific topics, share with me a short version of your own investment journey and how you got to where you are today. And Marty, Dunn has one of the longest journeys within managed futures, so why don’t we start with you. Tell us about your own and Dunn’s history when it comes to the CTA industry.

Marty

Thanks Niels. Dunn started in 1974, and I’m sad to say that I wasn’t along at the beginning. But Bill was originally working in the government contracting business in DC, and thought there was a better way to make a living, so he started investigating trend following. Originally applying it to equities, he figured out the population was too large, so he moved on to futures, which there was only about twelve at the time.

Since then it’s been a success. He’s not part of the Turtles, which most people mistakenly believe he is. But, he was doing the same thing the Turtles were doing, just neither party knew of one another.

On my involvement, I’m the CPA. I was originally working for an accounting firm that did business with Bill and audited all his funds. Bill and I worked together for eight or ten years. In 97’ he offered me a job, and it took me about two seconds to say “yes,” and I’ve been at Dunn ever since.

I started out at the bottom and worked my way through the process. Then seven years ago we entered into a transitional agreement where I would take over the firm from him. It was supposed to be a ten-year plan. Five years into it he accelerated it and told me he thought that things were going pretty well. I said, “Well, how am I going to know if you’re not happy Bill?” And since Bill is our largest client, he said I would know if he wasn’t happy. And he’s still our largest client, and we still break bread occasionally. He’s still in the Palm Beach area, so he stays pretty close to the firm, but he’s not active on a day to day basis.

Niels

Sure, sure. And how about you Grant? I know that you’ve played different rolls in the managed futures industry over the years. Of course, Crabel is another well-known CTA. Share with us some of yours and Crabel’s background story, please.

Grant

Of course! Crabel has been around really… Toby started the firm, almost three decades ago. We’ve been running institutional assets, or at least high net worth assets really for over two decades.

We’re known as pioneers in the short term trading space. Which is in many ways markedly different than the trend following space. We tend to have very low correlation to the average trend follower, purposefully.

Right now, today, we focus on strategies in our flag ship which we call the Multi-Product that target about a twenty-four hour hold time duration. Which is quite fast relative to I’d say the larger sample of the managed futures space.

What’s more, we’ve expanded our offerings now. I’m the portfolio manager for a product we call Advanced Trend, which is directly accessing trend following returns. So, whereas our flag ship is uncorrelated and short term, we do now have a trend follower that really leverages all of our experience in both what we learned – the momentum on the short end of the scale; and all of our thought processes around really long term trend following that we’ve deployed inside the Multi-Product for decades now, coalescing into the advanced trend product.

I came to be a part of Crabel after spending nearly a decade at a fund the funds, Efficient Capitol. I was the chief investment officer. While at Efficient we ran somewhere in the vicinity of two, two and a half billion dollars in assets that we allocated directly to managed futures traders.

We were a little bit different in that we focused (certainly during my tenure there) not only on some of the best-known asset managers in the managed futures space, but we typically had a number of allocations to early stage managers as well. It was a wonderful opportunity to watch the development of, now, some really household names like Winton and other big asset managers grow from a very small initial size (when we first started allocating with them). It was fascinating.

I left Efficient in 2012, initially desiring to work with startup managers. I saw a real struggle that they had in establishing themselves. Certainly, it’s become a lot more challenging as the biggest managers have grown. I thought there was a need, and potentially I could help them develop their businesses. I started a small group called Alpha Terra, which was acquired by Crabel Capital Management in March of 2014. We sort of dissolved Alpha Terra, and now we just use the brand identifier called Alpha Terra for Crabel offerings. They’re fully Crabel vehicles, but we use the Alpha Terra designation for products that are not short-term systematic strategies. Of course foremost of which is our advanced trend product.

Niels:  Great! And Mike, as someone who interacts with many players in the managed futures industry each day, I would like to learn more about your background and how you engage with them on a daily basis.

Mike

Well, I have to say when I graduated from college and I got a job as a runner for one hundred and twenty-five dollars a week in 1984, I called my mom, and I was so excited, she cried, “how could that degree lead you to that?” But the very first thing when I asked one of my mentors what to do, he sent me up and got graph paper. This is what we used to do back in the day, we charted.

So I learned charting, and we called it technical analysis at the time, and that was my indoctrination into momentum. After working for a couple of Wall Street banks, I joined Société Générale’s Legacy Organizations in 2002. We have a long, long history in our legacy companies of working in the CTA industry, and pretty much starting at the very beginning and growing with them.

In regards to day to day, I love it, because there’s just a lot of incoming from our client base because most of these large CTA managers are our clients and a lot of them really need our guidance, especially from a regional perspective. In particular, I would say I’m involved with our European and Asian clients helping them navigate the North American investor landscape which can be very, very confusing to a non-American.

Niels

Sure, excellent, thanks. Now today we’ll cover a number of different topics that I think many investors and traders have to deal with in their day to day work. But, since we have two legendary CTA firms sitting around this table, I want to kick off with a question relating to trend following, or the part of the industry that relates to trend following. The question is about trend following as a strategy, which has clearly been the backbone of the managed futures industry for many decades, but there seems to be a tendency, at least in recent years, where classical trend followers are re-shaping themselves to become more multi-strategy type firms. Grant, how do you view this evolution? Why do you think this apparent change has taken place?

Grant

Trend followers tend to have, as we know, a reasonably high correlation to each other. That doesn’t necessarily mean that their returns look identical. There tends to be dispersion in terms of absolute returns. But, when investors look at the trend following landscape, I think it’s become difficult (particularly as institutions have used managed futures more readily, and trend following more aggressively) to differentiate one trend follower from another. This is especially true when contextualized with the growing number of flat fee and low-cost trend following options. This puts tremendous pressure on an historically full fee trend follower – I’d say one in twenty, or two and twenty to justify the fee load.

This is particularly true when, in any given year, the correlations are very high. The dispersion can be either positive or negative. How do you do that? Well, one way to do that is to theoretically offer your investors something more than simple trend following, in a sense. So if you do not believe you can offer a compelling reason as to why your core trend following offering is worth fees relative to flat fee competitors, offer more. So we’re seeing an abundance of growing strategies. Of course, that is a positive in a sense that theoretically it increases, (hopefully) the sharp of a manager’s core vehicle. But it’s also a negative in the sense that, for many institutions, it becomes very difficult to understand exactly what they’re paying for and whether it’s worth it.

Niels

Marty I know that trend following is very much at the heart of what Dunn Capital is doing but do you see any dangers that investors should be aware of from this shift in focus that some trend followers have been undergoing? Or, are they just adapting their strategies to cope with a less friendly environment that we’ve seen for trend following since the financial crisis?

Marty

Well, I agree with a lot of what Grant said because what you’ve seen over the years is that there are a large number of trend followers who haven’t adapted or have done the research required to not differentiate themselves, but to take advantage of the technology and the data and the things that are available today to do trend following, or develop risk management tools and portfolio design to make their systems better and still be a trend follower.

So, trend following isn’t complicated, it’s time and noise, and anyone can design a trend following system. The problem is that you have a lot of trend followers who have been around a long time and haven’t adapted their program. So, in my mind, you have kind of a midline tier of trend followers who’s performance has been poor. I mean that’s the bottom line.

So if you’re in that boat and you haven’t done well, there’s two things you can do. One is you can… because trend following definitely has value. I mean it’s the one program that gives you the insurance during bad markets – so credit crisis markets, and any kind of stress that gets introduced into the system when all other financial strategies become correlated, trend following tends to make money, and it tends to make a lot of money. So people want to have an allocation in trend following. So how do you get that allocation?

Well if you go into just your average ordinary simple trend following program, you want to get low-cost. There’s a lot of people out there providing very low-cost trend following. What we’re trying to do… I mean in one respect we’re low-cost because we never charge a management fee, but on the other side, we do charge a 25% incentive fee. So I tell people to look at the net returns if we can provide returns that are better than the average, and we can get 25% incentive fee, so be it. I’ve never had a complaint from a client that paid me an incentive fee. The fact that we don’t charge money when they’re losing money, and as a trend follower there’s going to be periods where you lose money. But we continue to do research, we continue to do things to try to make our system better over the three to five year periods, and I think we’ve been fairly successful.

Niels

Sure, sure. Mike, I want to bring you in on this topic, since you see a lot of the flows related to these strategies. What have you seen when it comes to investor appetite within the managed futures space? And perhaps specifically, what kind of strategies do investors go for at the moment when they decide to invest in this space?

Mike

Well as I think everyone here is aware, you know hedge funds, and not just CTA’s, but hedge funds in general, are having a difficult time justifying their fee structures because of basically performance related issues. So it’s not just CTA’s. I would say definitely the good news; the good news is some very… you know some pension funds have done deep, deep dives into the strategies and convinced their boards that they can bring in returns when others are not and when they’ve kind of fitted CTA’s into twenty years of their own investment portfolios, they see a smoothing of the returns, but they see less volatility, higher sharps.

So there are flows coming in, in particular from North American entities. That’s the good news, the bad news is the flag ship, or maybe not the bad news, but the flag ship allocations are to the lower fee, lower cost trend. There is some good news around that, in that we see investors kind of building… doing a kind of flag ship allocation to low fee trend and then building allocations of non-correlating managers around them. Then kind of back to the bad news again, is those allocations we see are really going to the very, very large names who we’ve always known. The larger ones get most of the lion’s share, and we’re continuing to see that. So kind of good news/bad news, I would say, with North America leading the pack for allocations now globally.

Niels

Sure, but if you look at some of the other parts of the world and the other regions that you interact with, is the appetite very different from North America there? Or is it just sort of different ways of implementing?

Mike

Well I think Marty can definitely comment on this because I know Dunn was earlier in the UCITS… with the UCITS offering. Yeah, so anywhere from mid-continent Europe to Southern Europe. I wouldn’t say all of a sudden, and I think we had kind of an early start and interest in UCITS with it kind of getting quiet, and now if you want an allocation from that area, you do need a UCITS vehicle. That was Marty’s comment. They have had a nice organic growth of theirs. I would add, in Northern Europe, I think the institutional investors have embraced these strategies. They do tend to go to higher fee, larger managers in Europe. Fairly quiet in Asia, with probably just a notable… we’ve seen quite a bit of growth of some of the risk premia products in Asian markets.

Marty

To just add on to that because you’re absolutely correct, I mean the money goes to the bigger managers, and the bigger get richer. But I think what we’re starting to see because we’re in the next tier down; we’re a smaller manager. We’re starting to see people who have allocations and they kind of build their main allocation to a big manager and then they reach out and try to build around that allocation. We’re starting to see those flows now, and I think you’ll see more of that as we move forward as they start looking for the specialized manager or the manager that’s getting a little bit better returns to enhance their portfolio. Once they get comfortable with the big guys.

So you allocate to Winton, or somebody in that framework. You get comfortable with managed futures; you understand how the program works, how it evolves during the market environments. Once that comfort comes about then people start allocating more in different things.

I think, Grant, you have to be seeing it from the short term space. I mean Toby has to be accumulating assets because it seems to me that’s the next natural diversifier, is going from the trend following to the short-term space and knowing that this is one manager that has actually done it for a period of time and made money over time.

There’s a lot of short term space where they do well for three or four years, and then it just goes off the cliff for whatever reason. Crabel has shown that they can do it and stand the test of time, so I have to think that that is a huge advantage. Is there anybody who’s been doing it longer?

Grant

Not, certainly, at the institutional quality level, I think that we have at Crabel. I think that that is reasonable to say. You’re right, there have been flows, we certainly talk to a number of investors these days that start the conversation by indicating they’re not interested in trend following. They’re talking to us because we offer no correlation to trend.

One of the difficulties you run into in short term trading is the reality that you have no choice organizationally but to be essentially a technology research shop first. For instance, at Crabel, we’re over ninety people now. We essentially have less than five in business development and client relations, because everyone else is deployed in technology research, etc.

When you’re trading short term you have to think about pursuing technological advantages that really are not so much competing against institutional asset managers, but the high-frequency trading community; alpha degradation shows up a lot more readily; and all of those things basically mean that capacity is a little bit more limited of course. So for instance, we’re very strict on our capacity. All of us at Crabel have a tremendous amount of our net worth invested in our own vehicles. The last thing we want to see is alpha decay and sharp decline because we’ve allowed institutional investors in beyond the appetite that we can handle from a capacity consideration. These are issues that we look at, really, every day on an intra-second level. So it’s wonderful to see in-flows, we welcome it, but we’re also not an unlimited supply of these things. We have to be cautious on the capacity side.

Mike

If I could pipe in: from a clearer executer perspective I’m just curious (and this is open to both of you), I think I’ve seen an evolution of… well, we’ve all seen evolution from voice to DMA, electronic execution, to now, I think, the third level of some pretty high-tech algo-execution. I think back in the day when returns were headier, there wasn’t as much attention paid to slippage, and just curious and to open to both of you, have you done a lot of research? Have you implemented new execution algos? Are you monitoring your slippage more than ever before? Again, open to both of you.

Grant

Well I’ll start. The answer is an emphatic yes, to a degree that’s sort of hard to really fathom until you sit amongst our team for a great deal of time. The investment that we’ve made on the execution side has been extremely expensive, time intensive. It’s been a very, very steep learning curve. And as mentioned earlier, the reality is we find ourselves competing against Jump and Virtu and Tower more often than we do our former managers.

I think if you think about why that is, I think I would suggest two different paradigms of viewing the world of execution from an institutional asset management perspective, for the managed futures community. The first is, of course, managed futures has developed over really four decades, Turtle trading, etc. Around the idea that you know, we’re going to take trades; we’re going to hold them for a duration of time. They’re going to produce several points of return. When we actually get into those trades, or out of those trades we may give up a tick here, a tick there, but really in the context of the number of points we look to capture on a trading opportunity, it’s negligible. Don’t worry about it. There’s that paradigm, and that paradigm makes sense.

You’re essentially looking at this big numerator of the profit per trade and this very small denominator of the cost structure inherent in sort of capturing it, in terms of execution, and that paradigm makes a lot of sense. In fact, that paradigm is very prevalent even now. I run across a number of investors who, when we talk about our trend following product, say, “Yeah we don’t really”… all the legacy trend followers we talk to, they’re pretty skeptical that execution matters for trend following because of that ratio – huge numerator, in terms of profit per trade but a very small denominator in terms of cost.

However, there’s another way to look at that, which is the mathematical sort of impact of cost on an investor’s… essentially their NAV – their rate of return for an investment in a product. So, the other way would essentially be saying, well how many round turns per million dollars of invested capital will I be making to run my trend follower? As  an example. And a common number might be, say, a thousand round turns per million.

Which is to say that every dollar of cost equates to ten basis points of you NAV a year as a trend follower. It might not seem like a lot, a dollar, that’s… you know in the E-mini S&P we’re talking about a tick width that’s twelve and a half dollars, you’re talking about less than one tenth of a tick theoretically. Exchange costs run anywhere from two to four dollars per contract. So twenty to forty basis points there.

The reality is, that it’s ten to fifty basis points, maybe two to four percent if you’re paying a tick of slippage entry or exit. Even for a trend follower doing one thousand round turns a year, well that’s a real number! So if you shift your paradigm from instead of considering the size of the opportunity relative to the cost, to saying but the cost exists, and what exactly is that impact? You start to, I suppose, sort of wake up to the fact that there goes four to five percent of your NAV a year if theoretically execution doesn’t matter, and we’re talking about a trend follower.

So that cultural shift happened for us about five years ago where we said, “You know what, alpha is important, we’ve always been an alpha shop, but my goodness do we pay a lot in execution!” And we always thought it was relatively small, and inconsequential. But guess what? It immediately hits the bottom line, it’s essentially a fixed cost, and you can fix it. The problem with fixing it is you have to learn, and you have to learn in a hurry, and you have to compete against folks who are spending literally hundreds of millions of dollars a year to maintain their co-locations, advance their algos, and execute in let’s say double digit micros. We’re there now, but my goodness what a road!

Marty

So one thing people have to understand is when we talk about the cost of trading, it’s not the commission cost that we’re worried about, it’s the cost of slippage in making that trade. So the larger you get, the more difficult it is to move your book. I assume Grant, you are still using actual traders to do the trades? Or is most of it automated at this point?

Grant

One hundred percent automated, co-located.

Marty

Okay, so our approach is exactly the opposite, in we have that ability because we’re a trend follower. Grant’s exactly right, the cost of slippage to us as a relative basis of profitability is very small. Now it doesn’t mean it isn’t important because it is, and we monitor it constantly because all our research has to make an assumption about slippage, and so we’re always trying to reduce that slippage and become better at it.

Our approach is we have actual traders that do the trading for us; then I provide them with a sweep of tools. Part of those tools are the ability to trade algos, part of it is the platforms and the algos that are on platforms, and then part of it is leaving it to their understanding of the markets or what they think they can do with a market and leave it in their hands. And there are times when they’ll hold a trade back you know, for several hours with the idea that the market’s coming to them, and execute the trade when they think it’ll be the most profitable.

Now in tracking slippage, I’ve actually had positive slippage for several years. Now this year, not so much. So there’s going to be times when they’re going to pay more than the average, and there’s going to be times when they’re going to do better than the average. I just think with as much of the industry that’s going to the automated route, because it’s not just short-term traders like Grant, people in my long term trend following space are also going to automated trading. I mean they don’t want to have the cost of a trading desk. It’s much easier to have the technology, have a co-located server, use the algos that are off the shelf, and trade the position. I would assume, Grant, you guys don’t use algos off the shelf, you’re probably doing things that are designed in-house?

Grant

Yes.

Marty

I mean the bottom line is if you’re using algo off the shelf… there’s other algos designed to take advantage of that algo. That’s just the way it is, so you get what you pay for.

Niels

I just want to go back to invest and demand a little bit, but also continue a little bit on the cost issue. Clearly, the low fee products that we talked about earlier on, and you know, where they claim that they can capture the return of certain strategies, have received a lot of attention from investors in recent years. So I want to hear your sort of opinion about these products Grant. Why don’t we start with you? What do you make of this development in general?

Grant

I think it’s fantastic! It’s absolutely fantastic! I mean the reality here is we’re all in the business of rewarding our investors at the end of the day that’s our priority. And for the most part, a growing percentage of our investors are the pensions and endowment communities of the world. And it’s exciting to actually theoretically add value to the bottom line for folks that really need it. The idea of a competitive landscape where investors pay for what you’re actually doing is really exciting for us.

In terms of low-cost trend following, specifically, the world has sort of evolved so far to be a world of full fee trend following that claims to add a lot of advantages that justify higher rates of fees and low-cost trend followers. And very often the very same manager may offer one of both. That… you know, and there are some wonderful shops out there, in no way do I intend to disparage them, however I do want to suggest that is an interesting potential conflict of interest in the sense that the last thing that you can afford to have happen (if you’re offering two products, and justifying full fees of one by being superior to one that is lower fees), is have the one that’s lower outperform. That really jeopardized your core flag ship vehicle.

So theoretically, intuitively, and I don’t work for these organizations. I don’t know how they think about this issue, but it seems to me that it would be wise to somehow handcuff what you’re doing on the low fee side. Investors expect low fee to be more vanilla anyway. And hopefully in that handcuffing (either in the number of markets you trade, perhaps in how you execute, perhaps in the number of strategies you deploy, etc.), you can guarantee that you won’t have a sustained period of time where your low-cost alternative outperforms your flag ship. It’s a big risk.

Now, what’s happened more recently, and there are a number of shops that were certainly… I guess one of them who have not historically offered their flagship as a long term trend follower; they’ve been uncorrelated. So in our particular instance, we have a trend follower that’s been developed over decades of time that’s, again sort of coalescing our research over all sorts of different time frames. And we’ve manifested it into sort of the heart of our thinking an optimal everything that we can think of that adds value in a trend following way, in what we call Advanced Trend.

When we came to market with that particular product in 2014 we found ourselves debating internally, what do we do here? Because on the one hand it’s our peer group, and we look to outperform every full fee trend follower out there. We love the idea of having that competitive landscape to compare ourselves on a daily basis. On the other, there’s a meaningful appetite for low-cost fee, and frankly, there should be, because much of what trend following returns are relatively replicable.

So we ended up trying to cut both ways and said, “You know what? We’re going to pursue everything we can think of in terms of high quality trend following, but we’re just going to offer it at a low-cost fee level.” We did this very much with the intention of creating that fee pressure. And we’re somewhat unapologetic about it.

We’ll see what happens from here, but that’s sort of the landscape. Historically it’s been full fees doing everything, low fees theoretically doing more vanilla, and now there are a number of entrants that among them say, “Wait a second, we think that you should pay low fees for what use to be full fee.” We’ll see where it goes.

Niels

So can I just to clarify? So what do you charge for your trend following? And also what do you think the real cost of running a trend following program is?

Grant

We charge one in zero. That’s our one percent management fee, no incentive, or zero percent management, ten percent incentive at our investor’s choice. Or really any combination thereof. We estimate that just in terms of our execution infrastructure we’re talking about having to support it with several ten million dollars, multiple ten million dollars worth of investment in executional loan. That doesn’t really speak to all the other costs around data, client relations, office space, etc.

So, to break even with our trend following product we’d have to run in the vicinity of four billion dollars at our fee structure to break even at the fee level we’re charging for this product. The reality for us is, of course, we’re a short term manager by legacy. All of that infrastructure we maintain because we have to for our short term program and it all is leveraged. So, we’re in something of an advantageous position, I suppose, in the sense that we do not need to plow all of our trend following management fees right back into execution. Much of that is supported, it can enhance instead ongoing research and development, but really four billion is about the break even if that’s all we were, was a trend follower.