"Simplicity and robustness, some of the basic things that were fundamental to our philosophy at QCM because I believed in them, still prevail today." - Aref Karim (Tweet)
How do you transition from working alongside the capital management industry to starting your own hedge fund? Our next guest grew up in Bangladesh but fled to London during the Bangladesh Liberation War in 1971. He worked as an accountant and then went on to join the Abu Dhabi Investment Authority (ADIA) where he pioneered the organization's futures investment department. Learn about his personal setbacks and successes, his innovative investment strategies, and how he founded Quality Capital Management in the UK.
Thanks for listening and please welcome our guest, Aref Karim.
In This Episode, You'll Learn:
- About Aref's childhood in Bangladesh.
- How his father wanted all 10 of his children to attend university and instilled in them the belief system that made them successful.
- Why Aref decided to study English and Literature even though his background was in the sciences.
"My father quite fancied me to be joining the foreign service - he thought I could make a good foreign minister or ambassador for the country." - Aref Karim (Tweet)
- The story of Aref's escape to the United Kingdom because of war and social upheaval in Bangladesh.
- How he went from being an accountant in England to working for the largest sovereign wealth fund in the world, ADIA.
- Aref's perspective on the history of the hedge fund industry and the alternative investment industry.
"The degree of sophistication increased many fold because of technology, the power of computers, the power of being able to go into tighter and shorter time frames." - Aref Karim (Tweet)
- About the beginnings of ADIA's futures department that Aref helped to start.
- The early days of the futures industry and Aref's perspective on trend-following.
"I don't believe in too many moving parts, because I feel like it's a lot of smoke without the fire." - Aref Karim (Tweet)
- How he overcame personal setbacks when his wife unexpectedly passed away, leaving him with three young children.
- About Aref's return to the UK and his decision to start his own CTA and start trading in futures.
"The more variables you introduce - the more that things can go wrong or go out of sync." - Aref Karim (Tweet)
- About the genesis of Quality Capital Management.
- How Aref's investment strategy evolved and the specifics of his current trading strategy.
"Our propensity to buy is far greater than our propensity to sell, and this always keeps that element of buoyancy on the long side." - Aref Karim (Tweet)
- How he measures the "Flow" of the market.
- Why his strategy looks at changes in volatility and doesn't care whether it is a bond or an equity.
- How QCM went from using a few indicators to no indicators at all.
Resources & Links Mentioned in this Episode:
- Learn about Aref Karim on his Wikipedia page.
- Learn about the Bangladesh Liberation War that displaced Aref from his country of birth to the UK.
- Learn more about the Abu Dhabi Investment Authority (ADIA).
This episode was sponsored by Swiss Financial Services:
Connect with Quality Capital Management (QCM):
Visit the Website: www.QualityCapital.com
Call QCM: +44 (0) 1932 33 44 00
E-Mail QCM: email@example.com
Follow QCM on Linkedin
"Our basic premise is that all these assets are one and the same so we use the same approach whether it is commodities, fixed income, or equities." - Aref Karim (Tweet)
The following is a full detailed transcript of this conversion. Click here to subscribe to our mailing list, and get full access to our library of downloadable eBook transcripts!
Welcome to another episode of Top Traders Unplugged. Thanks so much for tuning in today. I know how valuable your time is, so I appreciate you spending some of it here with me. Also, thank you to those of you who are sharing the podcast with your friends and your colleagues. It really does help me to expand the reach of the show so that more people can learn from my amazing guests. On today's show I'm talking to Aref Karim, Founder and CEO of Quality Capital Management. Aref brings a unique perspective to our conversation as he started off his career on the buy side of the hedge fund industry, and not only that, he did it within the largest allocated to hedge funds at the time and perhaps even today, namely the sovereign well fund of Abu Dhabi, also known as ADIA. I think most people will learn a great deal from the insights to this unique organization that Aref shares, but I also want to mention that at the very end of our conversation, which you will hear in the next episode #030, Aref surprised me when I asked him what I had not covered in the interview. I think his answer says a lot about his attitude towards transparency and the challenges that you face as a hedge fund manager. Now for those of you who are new to the show, let me just mention that you can find all of the show notes including a full transcript of today's episode on the TOPTRADERSUNPLUGGED.COM website. Now let's get started with part 1 of my conversation. I hope you will enjoy it.
Aref, thank you so much for being with us today. It's great to have you on the podcast.
Thank you Niels for the invitation.
You're very welcome. Now, you started QCM about 20 years ago, but it wasn't your first career choice, so to speak, because you started out as an accountant, if I'm not mistaken. I think in fairness, it's the next move in your career that many people are fascinated about, namely how you go from working as an accountant in the UK to joining one of, if not the largest sovereign well fund in the world. It wasn't just any sovereign well fund; it was ADIA (Abu Dhabi Investment Authority), and this I believe, has great relevance to our industry as a whole and to our listeners in particular. ADIA has been known to be a very early adopter and a great supporter of the alternative investment strategies. Perhaps, in particular, the systematic type trading strategies that would also include commodities in the investment universe. So, Naturally I'm excited to be able to dive into the topic with you today because I think we can all learn from these experiences. The transition from being on the buy side of the industry to then starting your own hedge fund is also quite unique. I thought because you are on, and you have such a long career in this industry, that I wanted to start in a slightly different place today, namely by asking you, perhaps to give an historic perspective of the hedge fund industry or the alternative investment industry from your point of view looking at it over a 40 year plus period. Maybe you can share some of your insight as to how it was back in the day, and how you've seen it evolved over time?
Absolutely. To put things into perspective, let me talk a little bit about my younger days and childhood, etc. but very quickly. That has some bearing on the evolution, if you like, or the path essentially that I have taken that has lead up to the industry that I am in today.
I come from a place which today is called Bangladesh - the country of Bangladesh which, for many years has been one of the poorest countries in the world. It was formerly known as East Pakistan. Up until 1947 the British had ruled united India, and Pakistan was a part of it. Then in 1947, they got their independence and my father - I'm the youngest of a family of 10, much like a Catholic family, if you like, in the west. We had very spiritual parents, but, on the other hand, fairly liberal and he had his own struggles coming from a village. He made it into the civil service, educated himself and went all the way through University and then joined the British government. He worked up the ranks and eventually became a senior officer in the government of Bengal, under the British rule until 1947.
I wasn't born then; I was born in 1953, a few years later, but the childhood was a very modest one. It was a lot of fun - mainly outdoors growing up with very basic kind of sporting kit if you like from playing football. Because it was under the rule of the British, cricket was very popular too, and, on the other hand, flying kites, and playing with tops and marbles and things like that. It was very much outdoorsy kind of life. My father placed a high degree of important on education, and he wanted to make sure that all 10 of the children eventually went to University, because he had gone through his experiences, and he found how beneficial it was in terms of his own career. His income was very modest, but never the less he managed to put me into a private school, which was an American school, ironically enough - one of these American missionary schools - St. Josephs if I remember. I went through the schooling there and then eventually went from there to college for another 2 years. This is a pre-university kind of degree in secondary education, and that too was with a convent American College - Notre Dame.
Now my focus at the time was very much into the sciences: physics, chemistry, math, and so on, and so forth. What had happened was I had 4 brothers - 3 other brothers including myself, and my eldest brother eventually became quite successful after joining the army, and he went to Sandhurst, in the UK, he went to West Point, in the US, he joined the armored core, he became eventually a general. My second brother went on to do advertising, and he built up the largest advertising company in Bangladesh, and then my third brother went into civil service, and he began with the government. Each of my 3 brothers had been very successful in their own right. My father quite fancied me to be joining the Foreign Service. He thought I could make a good foreign minister or ambassador for the country. The focus of my education kind of changed a bit because I was in 2 minds - when you're young you think, oh, my brother's joined the army, so I'm going to go into the air force, so I was fascinated with aircraft, and jet aircraft in particular. I used to build model planes and so I was quite into that. Having said that, when my father ruled that out and said look, I think you should go somewhere else. You do have a lot of paternal influence, and in those days, when you have large families, the father is the father figure. He is the one who tells you what to do, giving you certain allowances of course.
So he suggested that I join the Foreign Service, but then the wars broke out, and in-between I was thinking, perhaps I'll do architecture. So, in some ways, English as a language fascinated me so that I thought, OK if I'm going to go into the Foreign Service, then perhaps I would rather prefer to do it through English as a language because you didn't need to have the sciences to join the Foreign Service. So I basically suggested that to my father, and he said OK, if that's what takes your fancy. Essentially from the science curriculum that I was following in college as I went to University, I switched to English and English literature, and I did English and economics at University. That was the time when the war of independence affected me, and I'm now talking about that the independence of Bangladesh started. The University closed down because there was total disruption there, and I essentially decided that to continue my education I would move to West Pakistan, which is the other wing, into Karachi and finish my University degree there. So I finished that in Karachi and I went on to be in the first year doing my Masters. That was the time when I got a break to leave the country, because war was going on, and it was very difficult to, as a Bangladeshi as we were seen to the secessionists. So we weren't allowed to leave the country. I managed to get special permission through some people I knew who helped me, and I came to England not knowing exactly what I was going to do.
I could not go back to Bangladesh because there was a lot of trouble there. That is when I fell into accountancy because I found that to be a good solid professional career that is very reputable, respectable and my father and my family essentially thought that was a good thing to do so I started doing my chartered accountancy. I hated it initially, because of coming from the art world, and having done literature. Never the less, I persevered because it was a long 4 years training period that you have to do, and you have to do professional educations, it's a bit like these days if you do CFA, though ours was a longer course, but something similar. While you are training, you are actually...while you are studying basically you are also training with a firm, so you are attached to a firm. I qualified from there and then I joined the profession with a firm of accountants. By the time I finished I quite actually enjoyed doing accountancy because what it brought into perspective, for me was this ability to see the bigger picture, when you are looking at a corporate balance sheet and group balance sheets, and you consolidate accounts and such like. It was fascinating that you were able to go macro and then zoom in and then go into micro aspects of a company's operations, analyzing it to the T and being able to make recommendations, etc. At the same time the whole operations is element as well in terms of internal controls and such like, those are also quite fascinating.
My take-away from the accountancy world is something which I feel is very, very kind of important, and what has also been extremely beneficial for me, is I believe in my mind to have done the arts degrees. So today when somebody asks what are my hobbies and interests, I'm into photography, we can talk about Ansel Adams, the pictures of black and white, large format of the Yosemite Valley or Henri Cartier Bresson and his photo journalism and that kind of stuff that he did. Yousuf Karsh is an Armenian Turkish guy who did all the portraits in black and white of Churchill and so on, and so forth, all the way through to current day - Annie Leibovitz with Vanity Fair and such like. At the same time (I was interested in) literature from poetry to fiction, whether you're talking magical realism, Salman Rushdie or Gabriel García Márquez, these are all very interesting for me. Even today, as we run our strategies in the systematic world, I think that the ability to be able to think left side, right side, I keep telling most of my team members here as well that this is absolutely crucial. Because to be able to analyze and to be able to use your scientific skill sets and knowledge that you picked up. To be able to be creative at the same time is extremely important.
While I was in the accountancy profession here, I had an opportunity, funny enough, to go to Abu Dhabi on my way to Dhaka, Bangladesh on a holiday, only because a sister of mine happened to be living in Abu Dhabi. She was married to an engineer. Her husband sadly passed years ago. He was working for an oil company, so they basically said look, you've never been to Abu Dhabi, come visit us. So I said OK, why not. So now I was passing through Abu Dhabi, and I spent a few days and I thought, you know what, this is actually quite a nice place. Money was tax-free, etc. and not just that but the standard was quite livable. Of course, you didn't have all of the facilities, etc. but you could be able to go away when you needed and get your batteries recharged if you like. At that time, Abu Dhabi was not as developed as it is today.
Anyhow, this opportunity came up with ADIA, and so I joined this institution. They made me an offer, and it was initially in the financial management side, but what I found was that, and I'm now talking about 1982. My first daughter (I've got three children) was about a year old in fact. She works with me here in the business development side. I started at ADIA for the financial management side and at the time the institution did not have an alternative investment portfolio. What it did have was just a commodity portfolio. By that they were holding passively quite a substantial reserve in terms of gold, like most central banks tend to do. They wanted to develop another sort of non-correlated investment vehicle within the institution because a large part of their assets were in classical, traditional equities and fixed income kind of things, plus some real estate. My boss at the time, the director, to whom I reported directly, he was a very interesting man, and he was very encouraging. I have to say that in Abu Dhabi I met some very nice people within ADIA as well, wonderful people, with whom I still maintain a good relationship. They were very encouraging. My boss at the time; he was very encouraging for me to...I guess he must have seen the potential of me being more useful if you like, on the investment side. Without even realizing I got more and more and more involved in the investment side, and that led me on to this project, if you like, of starting up an alternative investment portfolio.
The difficulty was that there was already, of course, a substantial amount of investment in traditional assets and equities and fixed income as I mentioned earlier and so we could not touch that, and hence, the alternative was to go through futures. At the time, and in fact quite a few years prior to that, if you remember John Lintner, the late professor at Harvard, he wrote one of those pioneering books on managed futures and CTAs and so on and so forth. Managed futures as an asset class, as he described it at the time. He did the typical efficient frontier study to show what was an optimal allocation, if I recall I think it was something in the region of 30% or so that it made sense. When you have substantial deep pockets as ADIA does with an industry as a managed futures industry being so tiny at the time, it was almost impossible to put any substantial investment with any one manager. Probably the largest CTA at the time was Mint with Peter Matthews and Larry Hite and Michael Delman. So they were managing something like 500 to 600 million at the time. There were some other managers who were in the region of maybe 100 to 200 million mark. Never the less, our mandate, therefore, was to see if we could invest with some of these managers that were non-correlated to the traditional asset class, so they wanted basically some active strategies.
With a universe that was relatively small at the time, and having seen the 1987 stock market crash, it became, in some respects, even more important for the institution to diversify if they want some form of a hedge I guess. Not that you could hedge with such a small portfolio in relation to such a substantive mainstream portfolio that we were running. Obviously I'm in, without going into too much detail because it is an institution that is very discreet. One of the most enjoyable things about ADIA was the sheer power and the understated way that it operated, and it had a lot of clout, but at the same time it never flashed the clout, so that was really nice. So anyhow, we went around very discreetly, and I was kind of driving that process to a certain extent, into finding or identifying managers and CTAs. We were tough in our negotiations in terms of fees simply because it is a sovereign fund. They were obviously not used to seeing these large management fees and those fees as you recall were like 6%. Mint used to charge 6% and 15% I think it was, and even subsequently the fees were typically at 4% or 5% and 15% to 20%, this is just simply unheard of you know. We had to do some very innovative negotiated fee structures that we came up with to be able to symbiotically benefit both sides. From their perspective the big selling point for us and it was basically that this is high-quality money, we're very discreet, we cannot disclose names, etc. but this is a very long-term relationships that we're looking to build and so on, and so forth.
That kind of worked. The industry, in those days, to go back to your original question, was dominated, in the futures area, by trend followers. There were some variations, like for example Paul Tudor Jones, who had a cracking year in 1987 and had the lime lights there with his 120% returns, if I recall the number to be right. He was doing a little bit of technical training, swing trading, and so on, and so forth. Predominately trends dominated things. Markets were so inefficient in those days that you could literally ride these trends for extended periods of time. Most of the CTAs adopted a very linear kind of approach, and that is to say market by market, they had technical signals of some sort, and then you just essentially ride out those larger trends. In particular even in currencies, I remember that you had some major, major trends - obviously this is pre-EC.
The evolution of the industry over the years has been incredible. For a start, it started out initially quite a bit in the US. US CTAs were a lot more, in terms of numbers, compared to Europe and what you had in Europe, it was just primarily Mint that was the only one that was well-known at the time. What you saw was a gradual diffusion from the US, a migration if you like in terms of the talent side of it into Europe, and particularly the UK as well. From an investor perspective, from ADIA's perspective, it was great not just geographically but there was an evolution in terms of the approach so that the typical trend following linear approach of market by market type of thing with its traditional stops and so on and so forth and just diversifying across multiple timeframes, that approach started to change. At some point the risk allocation aspect of fund management started to progress and this then took us into various portfolio methods, whereby it wasn't just the signals, it was how you allocated your risk budget to different markets and so some variation of the traditional market with this mean-variance approach and variations of those.
So in some ways, the degree of sophistication increased manifold because of technology. The power of computers, the power of being able to go into tighter timeframes and shorter timeframes, not that I'm a great believer in that, but that's just a philosophical thing. Never the less that allowed a lot more diversity. With that came the ability to build portfolios or approaches that were much quicker, in terms of frequency. So higher frequency and with all the low latency stuff that's out today it's even more so, and also variations in the strategies. So what used to be almost a one model or one type of approach, basically through trends started to morph into multi-strat type of things. You had essentially mean reversionary type of approaches built in along with trend models. Basically, this allowed the ADIA portfolio for us to build on it.
I certainly was never a great believer in a huge amount of diversification, so we kept the numbers to a manageable level - not too many. I also was a great believer in taking on some degree of volatility because anybody who was just smoothing away the vol was essentially, for me, taking away the skew potential of the strategy and ADIA is a sovereign fund so obviously we could stomach some volatility so long as it's at the portfolio level, you can just cancel them out. That was our job, right? So our preference was not, at the time, to necessarily to go with traders who were so particular about controlling vol and so on and so forth that their returns were kind of meager. So we essentially wanted more and a little bit of spicier kind of managers. But there was quite a diversity of these. Of course a lot of these managers over the years, the successful ones, who built on the business and built them up to sizable businesses today, have changed their methods, and this is something that everybody knows about, that there has been a degree of style drift, call it evolution or whatever, so that research led them into trying to focus a lot more on the risk side, and in the process, the returns have diluted somewhat from the punchier returns we used to get.
So that's kind of the evolution of ADIA. Then I had a personal setback. My wife unexpectedly passed back in 1992, and so I had 3 young children at the time and so two of the others were actually born there, and so my original plan of wanting to come back to the UK which was where I was based got delayed to a certain extent simply because this was such a major upheaval in life, so I needed time. I didn't want to introduce other variables in terms of complicating my life. So I stayed on for another couple of years. My boss at the time, he really wanted me to stay on. So anyhow it worked out OK. Then I left in 1995, and I knew pretty much what I was going to do. At one time I thought OK should I go into the fund the funds side of things because that's where a lot of my experience was, but then I had already started to do bits and pieces of modeling and stuff. In 1995 when I came back, I decided that I was going to found my own business which was going to be a CTA business, primarily starting off with futures. Hence, QCM was born in December of 1995 we started trading with some friends and family money. Probably one of my first investors was Commodities Corp., which at the time was seeding some of the smaller CTAs starting off, so they also seeded us. That was basically the start. I'd already developed a model. It was a relatively straightforward simple trend following model, and initially, the number of markets we were covering were just the US markets. There were 25 of those, and there we go, and thus unfolds the QCM story.
Thank you so much for sharing that. I wanted to ask you something before we leave the ADIA story, which I think, is quite...to me at least I'm curious about it. Back in the day, how did you get comfortable with a manager? My perspective here is now-a-days, for an institution to make an investment with a manager you have to look like an institution and you have to have quite a big infrastructure and all of these things that basically prevents allocators today from investing with small managers. But back in the day when, I'm thinking that things were not as organized, and some of these managers who had obviously built billion dollar businesses today, I'm sure they weren't quite there at that time. What were you looking for? What was important to you in evaluating a manager back then?
That's a good question. Essentially the instruments we were trading gave us huge comfort - futures. Because as you know with futures everybody talks about these days after the 2008 crisis and the transparency and the liquidity, and so on, and so forth. We recognized that years ago at the time, and not just that but it was fundamental to our hedge fund portfolio build because we had to convince our investment committee that these were highly liquid. If we didn't like it, we could get out anytime, not that that was our intention because it's a sovereign fund it was very much a long term strategic kind of portfolio. The first comfort was purely from the fact that these are regulated markets - the futures, and one price for all transactions costs, even though those days brokers did charge quite a high level of commission, but still it was good. The fact that we did not have to fund the entire investment, we used it, at one time, almost like an overlay strategy. You could put in the margin amounts. We had completely separate segregated accounts with top notch banks, so we wanted to make sure essentially from an operational perspective and from a credit perspective we were fully covered, and that meant that the next thing to really be more concerned about was the strategy element.
On the strategy side what we wanted was consistency. This was very crucial, at least for me this was very crucial. We did not like managers that were constantly changing and swapping their strategies and connecting more models and so on and so forth. So much so that we used to have clauses in our agreements that if any changes in the models are to be made than we should be consulted first, or we should be informed.
I remember that, actually.
This was quite pivotal because what we didn't want was the business of style drift. Particularly because, remember not all of them had very long history either. The second thing is what we did not want was this risk of style drift. Then the bonus came, obviously, in terms of identifying managers who did have some operational infrastructure on top. So those were getting bigger ticks on the books. But having said that, it was really more important that the team as a whole, their backgrounds and their strategies all made sense. We also wanted quite a bit of transparency in terms of strategies and we have always said that look, it is not our intention to try and copy this in any way what so ever, but we need to understand because if we don't understand we're not going to be with you when you do through these difficult times, through drawdowns. If fact, they did offer a huge amount of transparency. Not that there was a great deal of variations or differences between managers, because the bulk of them really were following trends really, so there was such an overlap to some extent. It mainly came about through timeframes and perhaps the portfolios were a little bit different.
Time moved on, and we started to get a lot more sophistication and at the same time, towards my last few years, we started to move...not away I should say as the bulk of it was still trend following and CTAs, but we started to venture into other approaches or strategies so we took on some guys who were doing global or tactical asset allocation with multi-factor models. We never took on any shorter term high-frequency guys, because mind you even the level of sophistication was not there that much because technology was not that far advanced. It was advancing, but it hadn't reached quite the stage that we felt comfortable with. Then we went into currencies, specialists... so the job became a lot more challenging to identify strategies and approaches that were quite a bit different from just what we were already doing that was our bread and butter with the CTA side. In so doing, with some of the global TA guys, these were some of the larger fund managers who were venturing into, and they had to really fine people working and great brains and so on and so forth and they were in a way testing out their kind of approaches in the models using futures to essentially trade equity index futures and fixed income, and currencies, primarily sticking to financial markets and not so much commodities. We did identify a number of those and so we invested in them. So the portfolio overall then kind of morphed into a combination of CTAs and other strategies. So I don't know if I answered your question.
That's very interesting. So we're now in 1995. You're heading back to the UK, and you're starting QCM but what would you say those 13 years at ADIA, what is the one really important thing you brought back with you that was very important to you in starting QCM?
I think in line with my accountancy background that I referred to earlier in terms of being able to see the macro picture, what really helped and complimented that was essentially my tenure at ADIA because they were also the big picture kind of guys, looking beyond the noise, looking at the bigger things, relationships in terms of markets. That almost takes you towards a philosophy that abhors and sort of moves away from anything where there are just too many things going on. In other words, simplicity, robustness, some of the basic things that are fundamental to our philosophy at QCM, because I believed in those, and this still prevails today. It hasn't changed.
I just did not believe in too many moving parts coming into the equation because I feel like there's a lot of smoke without the fire, essentially, and too many competing forces. While I was at ADIA, keeping the portfolio relatively straightforward, simple, etc., and which did very well during that period, and I'm sure it continues to do so now. I used that philosophy to be developing our own strategies here, and I referred earlier to the left side, right side, and the hobbies and interests. We want to encourage creativity but at the same time it's not creativity for the sake of creativity, in the sense that you don't want to just tinker with things all the time. What we've been able to try and focus on is what are the main sources of returns? Where's the real alpha coming from? I'm a great believer that it all comes from more the portfolio approach. I know there's a number of managers out there who essentially take a very linear approach of doing market by market, and they have the best model in a certain market or certain sector asset class, that that is where their alpha comes from. I kind of take a view that the less indicators that we use and the more portfolio relationships that we track to move positions around is a lot more beneficial. So in other words asset allocation in our kind of relatively shorter term approach as opposed to long-term asset allocation from an institutional perspective at a higher portfolio level is probably more what would be risk allocation - tactical dynamic risk allocation. I do believe that a lot of alpha comes from there rather than trying to get these signals right to be able to call the directions right whether we should be long or short.
Was that a focus from the beginning? Because you mentioned earlier that you kind of started out as a CTA and my impression is that you maybe morphed yourself actually into more of that...
Absolutely, absolutely, that's 100% right because I started off as a CTA then I realized that keeping static views on markets in terms of weights and allocations was not the most efficient way of going about doing the job. So we said, OK, we have to make this movement more dynamic and have some limits within each market, and also within each sector. Allow it to float a little bit in terms of the weights. That approach, to me, tied or linked our philosophy and blended nicely with the basic economic rationale that we were after, and that has not changed in any way. The basic interests and changes of economic paradigms - the inflation/deflation themes that paired out. You might say if you just do trend following you can do that too. You do except that you are doing it singularly, market by market and that to me is less efficient than to be able to say, OK, how do stocks and bonds relate to each other during different changes of economic regimes. That opens up another dimension completely. I believe that there is greater alpha in that rather than just through trying to linearly get the direction right.
Absolutely 100% yes, our strategies evolved and I would say that probably in 2005 was our first major change in direction. Although I say 2005, because you can't really put an exact timeframe on it because these things evolve. I would say by 2005 we had almost switched completely into a dynamic allocation approach, but still using some indicators for directional calls. That we phased out slowly over a period from multiple - 6, 8 indicators, down to 3, then down to 1, then now, today, we don't use any indicators what-so-ever. I'm very excited. I love what we do and how we've evolved to what we have evolved to. I do believe this is a potent strategy. Obviously I know we've all gone through some hard times in the last 3 years or so, but I think that what we have done to our models in terms of the slight enhancements that we made. We've made them significantly more powerful so that the strategies and the models are able to handle unencountered kind of environments such as what we have seen in the last few years a lot more efficiently and a lot better, of course time will tell. The good thing is there's no fitting this. This is something I adopted right from the beginning is that even when we were trend followers it was the same input, the same parameter that was used across all markets. What I was really after philosophically was more the strategy and not the market itself, which is quite an important thing from my perspective. If you believe that you've got a good, sound strategy, then all these things go up and down. Of course, there are certain nuances that you can take advantage of, but that can come later. You can add certain things to take into account those things. At the base level I think it's quite important that markets should be seen in a very generic way, almost in a similar kind of way.
I wanted to try and see if we can visualize this because even for me, these things can sometimes be quite difficult to follow. Let me start out with how I think, in a sense, you started off, and like many trend followers: so essentially you could have a model that you test across all types of markets and, let's just say that you buy the 50 day high and you sell at the 50 day low, and you use some kind of risk management overlay to do that. You could have other models using moving average crossover systems; it doesn't really matter. So that's one way of capturing trends. Now, how, if you should describe visually what your model evolved into and what are the things that you are looking for and how do you systematize that? How would you describe that?
OK, so if we think about returns, alpha essentially can come from a couple of sources, with the 3rd source really more market return or market moves if you like, and that you don't have any control over. What you do have control over is essentially the direction - whether you are long or short, that's up to you, and the level of conviction that you put on a trade, so the weight that you put into it. You can vary both of those. When we started off we focused a lot more on trying to find the best way to cull the directions, and we essentially diversified that recognizing that we could be out of synch in terms of the cycles of the trends. So we had a shorter term, a medium term, a longer term, which is quite typical for most CTAs.
The second variable of changing the weights - we kept that on constant. So we would vol adjust them in terms of the position size, which is just an explicit kind of an adjustment from pure changes in volatility, but we would keep the risk weight almost static in the market, and market base. A lot of CTAs still continue to do that, and it works, by the way. It's the heuristic kind of approach, rule of thumb type of thing - OK so just equally weight all markets. That still works, there's nothing wrong with that, but we found that the errors that you engage into when you get the signal wrong, and then you try to take care of those errors - you try and correct those errors - whether you do it through explicit stops, or whether you have another model to shorter term to cancel the other one to take off the position size fairly quickly, those errors are a lot more difficult to handle. Now it's all hunky-dory when trends are really going well, it's not a problem. When you get into troubled periods, that becomes a bit of a challenge because what you end up with is a lot of errors in terms of the noise and these tend to compound, and you get stuck in it for a long time.
Then we said, OK, in which case how about changing the weights so that when you are going into a problem area let's have the diversified timeframe approach to help you out so that you're not engaging unnecessarily in a high degree of activity but at the same time, what if you were to reduce the weight there and shift that weight somewhere else? So that's playing with that second variable. Most managers, most strategists today tend to focus, I guess (I'm assuming just by feel and by looking at numbers, etc.) that they play with both. In other words, they're playing with the diversity of signal generators and at the same time they're playing with allocation weight. The problem is you are having two moving parts, and this is the issue. The more variables you introduce, the more that things can go wrong and go out of sync. You have to control that in a certain way. The signal generation part is a little more controllable because it tends to stick for a bit, unless you are a real high-frequency trader.
In our case what we have done, the approach that we have taken, has almost eliminated the signals side of it: the binary and long/short element of it. There are a couple of reasons for that; one is markets in terms of return generation process, are overwhelmingly more profitable on the long side. This is a no-brainer, anybody can see that, the markets are floored, there's a logic behind it, they're floored at 0, theoretically up-side infinite, there's risk premier built into some of these assets, so our propensity to buy is far greater than our propensity to sell. This always keeps that element of buoyancy on the long side. That being the case, therefore the need to change signals - long/short, long/short, etc. is perhaps not as great in our minds as is the need to be able to have some control on the dynamics of the weight shifts, because it's through the weight shifts you can actually look at relationships: risk assets and non-risk assets. So where does this money flow essentially when risk is coming off, where is this money flowing to? It is clearly going into some form of defensive asset, non-risk - whether it's at the shorter end of the maturity spectrum into cash or whether we're going into the middle or whether we're going into longer term fixed income. That's another issue, but nevertheless, it will be somewhere there in the defensive area.
What I'm trying to suggest here is that the ability to play correlations, and I'm not talking about explicit statistical correlations, because those don't mean a great deal in my mind in terms of managing portfolios because your timeframe is too short, short term correlations are actually unstable, notoriously they get skewed by outliers, etc. I'm talking about correlations in term of the relative value nature of the correlation. One is going up; another is going down. If equity markets are going up where is this money coming from? The chance is it's coming from cash or bonds, etc. so you are divesting a little bit there, so there is a cash flow, or capital flow that is taking place which is, in my mind a lot more ascertainable, a lot more predictable if you like. To be able to jump onto that is a lot more efficient than to be able to call it through looking at price signals.
How do you measure flow?
In our case, we have a proprietary methodology and we essentially we are, without going into the details of our strategy, we look at volatility and we look at more of the vol of vol itself. So we're going into a completely different landscape altogether. It is no longer prices at all. We look at this acceleration of volatility in the markets and we co-relate it to other positions that we are holding and how they are behaving and we focus a lot more on the up side of volatility as in profitable volatility, or beneficial volatility for us. Our positions get changed. Essentially the portfolio is getting rebalanced on a daily basis. It's not going up and down, up and down because we're not high-frequency guys. What it is doing is essentially saying, OK, if there's a move from the defensive assets into risk seeking assets in the quest for risk premium then those risk assets are going to likely go up, and you're going to have dullness, if you like, in the fixed income area. In other words, it's going to be less.
We measure it by what we call, in a proprietary way, convexity. So we say the convexity is greater in the risk assets and less so in the non-risk. So the system is automatically looking for the greatest convexity in our portfolio will attend to essentially shift allocation of weight to the ones that are looking more attractive from a vol pickup perspective. Interestingly that methodology through use of only one model, which our approach is a simple one model approach. We don't have a value model, momentum model, and so on, as some do, but what happens is through tracking these changes of convexity in a way we find value opportunities. Just to give you an example, let's say you've got some position, S&P 500, it comes down, it gets sold, so we essentially take risk away from there, and the risk automatically gets moved and shunted somewhere else. It could be another risk asset that is doing well; we don't know. Typically, let's say, if it was just a fixed income and equity portfolio, it would go into the fixed income side, and there I mean short rates too, not longs.
Let's assume that the S&P, following the selloff, now the market just suddenly turns, so what's happening is that the convexity suddenly turns back positive for us in the S&P, so now we're not waiting for a priced base indicator - a level at which to buy in the S&P, we're constantly selling and buying. It's kind of a relative value game, almost like equities and fixed income, so in this case we assume that it's just a two market portfolio. So as risk comes back on, risk on trade comes, on that starts to pick up in terms of vol acceleration, so naturally again the money flows back, or the risk flows back again from the fixed income side onto this. So it's this dynamic that we are after, and that is a non-linear alpha. This is for us the biggest source, and so much so that when we describe a strategy we ask what are the two sources of returns for us predominantly?
One is what we call the portfolio alignment alpha, and that's what this is - the dynamic between markets within the portfolio looking at convexity in different markets and the process is indigenous, it's happening within. We're not imposing; we're not saying we have a level here to buy, or a level here...which is subjective. This is just automatically, dynamically adjusting. That sets the basis for the portfolio alignment alpha, and that also has the advantage, we believe that it can handle more difficult environments better. We had a bit of a problem earlier on because we had that one indicator still that I mentioned to you that we got rid of these indicators, and we had one indicator that was still a very long term indicator and that was preventing you, so you had a hurdle to cross, where now it's very smooth and it flows without any hindrance. All we're doing is we're actively buying and selling essentially. So wherever convexity is positive it's picking up, and we're going to be buying, and wherever convexity is negative or flattening out, we're looking to sell, so long as there is another opportunity there. Opportunity cost is quite important for us. When we are doing relative value we always relating one proposition with another, so we're saying does this look more attractive compared to that? I'm forgoing something here in order to increase my bet on something else. That's how we're keeping the whole portfolio on its toes and constantly seeking this convexity.
Do you have to, when you construct a portfolio, do you have to pre-determine and say these are the risk-on assets, and these are the risk-off assets? Let me tell you why I ask this, we have been brought up in a world where we always look at fixed income and bonds to be the safe place, the safe haven - that's where we go to, but on the other hand I would not personally at least be surprised that in the next few years we might see a complete reverse of that situation where actually equities could be the safe haven because of a major upheaval in the credit markets and the sovereign debt issues suddenly come back with a much higher speed than we have ever seen before, so I'm just thinking out loud here, does a portfolio or a methodology like that does it care whether it's one or the other as a label? Will it just pick up the changes in volatility?
Great point and that's exactly the case. It does not distinguish; it does not look at the labels it just looks at streams of returns essentially and the volatility in these and the way that we have designed for the model to identify the pattern that we're looking for. It adjusts for that. Therefore, it doesn't matter if it’s the other way around, if one becomes a risk asset and the other becomes a defensive asset. It's purely the vol itself. That's the beauty of this that this is, in a way, totally agnostic to that. It just so happens that when you look at it from the other perspective the bigger expectation - macro expectation in terms of the relationship, (the way it pans out between risk and non-risk and inflation/deflation sort of periods) is at the moment that you will tend to see that risk premier starts to enrich as inflation picks up, growth picks up, and so there is a natural tendency for the system to say, hello, I want to be long this and guess what, this is S&P or guess what this is a commodity, and so on, but it doesn't care. It just identifies those where the action is happening a lot more.
The nice thing, if you take a year like 2008, it was interesting because this is the crisis period, and I know that most CTAs did very well in that selloff building up to the Lehman collapse in September or October. In our case, we made quite a bit of money earlier on in 2008 through some of the commodities and to a certain extent equities, but more commodities really. In particular, if you recall oil went all the way up to $150. Then you had this major reversal in the early part of July or the second week in July. During that period we, again using our dynamics that we discussed through the convexity method, so the portfolio was aligning itself. The alignment process came about through, yes eventually going...because we were at the time a lot more longer term in terms of the indicator keeping us there. Now it's a lot more dynamic, so predominantly we're long term, but at the same time we can be quiet quick in changing our position if need be and that's just simply a function of convexity. We eventually did go short, obviously these equity markets, and we made quite a bit of money there. The bulk of the returns came earlier on from the shift into liquidity premium essentially into fixed income, into short-term interest rates and bonds and the dollar. The system doesn't know, it just sees a huge pick up in convexity and moved in those areas, and that's how we're able to generate very attractive returns in 2008.
Yes, does the model look at pairs in any way shape or form? The reason I ask is commodities, I can imagine that it's not that easy, necessarily to combine, obviously depending on which commodities you have in the portfolio but commodities are a little bit difficult to pair up with something in a sense the financials are a little bit easier at least to my knowledge. How do you do that? How do you mix commodities into all of this?
Here in our approach we don't like to fit anything so, therefore, our basic premise is that all these assets are one and the same, so we use the same approach whether it's commodities, fixed income or it's equities. Having said that, one of the enhancements that we made last September, and when I mentioned that we eliminated the last indicator, was to do with exactly addressing this aspect of it that commodities are obviously notoriously volatile and unpredictable, particularly weather-related commodities. Industrial commodities such as metals, energies, are a little bit more fundamentally, and I say in a broader looser term, tradable whereas the others are very difficult, because things just change so quickly.
So one of the things that we identified is that we needed to do a bit of an adjustment in the vol measure for our commodities. You can think of it in a risk parity type of an adjustment, it's that kind of adjustment recognizing that this asset class, because commodities don't earn any income, fixed income does, equities do, so consequently it's pure price fluctuations - capital appreciation/depreciation and so as a result it does tend to be a lot more volatile, so we made an adjustment to the vol to almost vol normalize, if you like, commodities with the other: with equity and fixed income. The other part is when you ask do we do pairs? No, we don't. It's the one whole portfolio of instruments or time series if you like where we are long and the same time we've got another bucket of all the shorts basically, and they're playing each other off in a way, but not in a paired fashion. It's almost like a portfolio fashion where we're pairing them off. Within each, it's doing a relative value type of approach.
Speaking about the portfolio and the universe of markets, and you mentioned commodities. I noticed that you offer, as far as I'm aware, the same strategy but with or without commodities, and I think that's a big debate actually because it's clear that people in the last few years who have not had a big allocation to commodities probably did better from a performance point of view in the last couple of years. What do you think, or what is right or wrong in terms of should you have the commodities, or should you not have the commodities? How do you view that?
I think it all depends on your perspective. Our AFP, the Alpha Financials Portfolio, that came about through a specific request from an existing client who was already invested in the JDP, it was actually a fund the funds and they had in turn a pension fund investor who specifically asked if we had a pure financials portfolio, so the request we got from our investor was that, obviously they were quite happy with our strategy and so asked us if we could run some simulations based exactly on the same approach, which we would not change anyhow for another portfolio. We always tend to use the same approach and we just took commodities out. The interesting thing is that people say, so is this a carve out, and I say no it's not a carve out. In portfolio terms with the labeling, yes it looks like a carve out, but in terms of its behavior, because of the way we trade, the relative value type...
Ready to learn more about the world's top traders? Go to TOPTRADERSUNPLUGGED.COM and sign up to receive the full transcripts of the first 10 episodes of the show and visit the show notes where you can find useful links to other amazing resources. Thanks for listening and we'll see you on the next episode of Top Traders Unplugged.