“It’s always important to know that individual data points, may actually lead you astray and quite significantly so as an economy does move and change in a relatively slow fashion.” – Anders Lindell (Tweet)
Our next guest turned a challenging market into an opportunity to transform his strategy and build something substantial.
Welcome back to another episode of Top Traders Unplugged. Today we have the former CEO, now Chairman of Informed Portfolio Management on to discuss the road to formation of his firm, their Systematic Global Macro strategy, their unique approach to investor interaction and much much more.
Thank you for listening to the show, please welcome our next guest, Anders Lindell.
In This Episode, You’ll Learn:
- The relevance of controlling production at paper and pulp mills while Anders was a developing man
- His history during his early days at JP bank as an analyst
- The impact of the shifts in the fixed income markets of the early 1990’s which inspired a new strategy
“How do we deal with tactical deviations from these long term strategic targets? You can end up in a situation where you deviate intentionally or you end up there because you’ve allowed markets to push you in that direction.” – Anders Lindell (Tweet)
- The strategic focus that was the starting point for the formation of IPM
- Who Anders learned from early on and who helped him cultivate a career in the hedge fund industry
- The programs IPM run today, when they started and the assets run inside of each:
- Systematic Macro Program – with about $3 billion AUM trading in account format since 2002
- Systematic Equity Program – with about $4 billion AUM trading since 2006
“Once we have identified weakness in areas for further improvement, that’s when we try and define how to fix it. That goes into research rather than scouting the market for general forms of alpha.” – Anders Lindell (Tweet)
- How to find and retain talent and the entrepreneurial spirit drives in house management decisions
- How IPM as an organizations manages the in house operations
- How Anders hopes to see their potential realized in terms of double or tripling their AUM for future markets
“We want the markets, even if they start deviating, that they will mean revert at some point in time and we don’t want that point to be 5 years out.” – Anders Lindell (Tweet)
- What Anders looks for when spotting talent for new team members
- Ideas for incentivizing team members to appreciate their careers
- The long term view of IPM’s track record and how to evaluate it
- More on the objective of the strategy of IPM and the environment in which it’s been designed to work well
“If we find ourselves in a market environment where people simply don’t care about fundamentals, then this model could not be expected to perform optimally. It could still deliver but then it becomes more of a random game.” – Anders Lindell (Tweet)
- Why the opportunities and volatility in emerging market equity trades is decompressing globally
- The meaning of global macro and how they structured all the information into a systematic trading approach
- How the IPM model differentiates it’s self from discretionary trading models and how they run them
- The role of fundamental information in identifying market strategies
“What differentiates us from a discretionary trader is that we have identified a large number of opportunities and they are coded in the form of risk factors and we are always in the market for all of those to a varying degree.” – Anders Lindell (Tweet)
Sponsored by Swiss Financial Services and Saxo Bank:
Connect with IPM:
Visit the Website: www.IPM.se
Call IPM: +46 8 20 19 29
E-Mail IPM: firstname.lastname@example.org
Follow Anders Lindell on Linkedin
“When we model, we model everything in relation to global composites. This is to get rid of the need to model absolute development.” – Anders Lindell (Tweet)
Niels: You're listening to top traders unplugged, episode number 023 with Anders Lindell, co-founder and Chairman of Informed Portfolio Management. This episode is sponsored by Swiss Financial Services.
Imagine spending an hour with the world’s greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world, so you can take your manager due diligence, or investment career to the next level. Here is your host veteran hedge fund manager, Niels Kaastrup-Larsen.
Niels: Welcome to another episode of Top Traders Unplugged. Thanks so much for tuning in today. I know how valuable your time is so I do appreciate you spending some of it here with me, and also thank you so much for sharing the pod cast with your friends and colleagues. It really does help me expand the reach of the pod cast so that more people can learn from my amazing guests. On today's show I'm talking to Anders Lindell, co-founder and Chairman of IPM. Anders was influenced early on in his career by some extreme market events such as the equity crash of 1987 and the attack on the Swedish central bank in 1992, where overnight rates reached 500%, to essentially go down the path of finding value in going against price drift and dislocation, away from the fair value, or intrinsic value of markets. This path has taken him to the top of the hedge fund universe with more than 7 billion dollars under management in a relatively short period of time. IPM's global macro strategy is fascinating in its structure, and so is the story of how Anders realized his success.
For those of you who are new to the show, I just want to let you know that you can find all of the show notes including a full transcript of today's episode on the TOPTRADERSUNPLUGGED.COM web site. Now let's get started with part 1 of my conversation. I hope you will really enjoy this.
Anders, thank you so much for being with us today. I really appreciate your time.
Anders: Thank you Niels. It's a privilege to be here.
Niels: Fantastic, now as I was preparing for our conversation today, I couldn't help think about the many times you're here on the financial media. How a particular analyst or economist suddenly revise the estimates that they have, or the forecast that they have for a particular market. Very rarely are these analysts held to their initial forecast. I also seem to remember an old saying that if you give 10 economists the same set of data they will come up with 10 different opinions of what it means, and what's going to happen. I think most people would agree that this is not a good foundation of managing money. You shouldn't be swayed in your opinion every time there's a new set of economic data being released as this happens nearly every day. So I'm interested in finding out through our conversation if you, at any point in your beginning, belonged to this type of analyst or economist that would revise forecasts and opinions based on new data, or whether you've always had this ruled based, or mechanical approach, where you organize fundamental data so that it can be implemented in a non-emotional and disciplined way. With your investment universe truly being global, I'm really excited about all the possible topics that we can talk about today. But, before we go into all of these details about your company, where it is today, I'd really like if you could take us back to the beginning, and tell us your story and what let you to take this path. Feel free to go back as far as you want...what were you like as a teenager, or whatever you feel like sharing with us today.
Anders: Fantastic Niels. Well, we shouldn't go so far back as my teens; that would be far too much detail for this particular conversation. I think it's material for my future development and also the ordinance of this firm that I actually started out as an Engineer. I went to Technical University in Sweden and focused on control theory and mathematics. Indeed, my first work or job immediately after University was actually selling and designing, building, and implementing control systems for paper and pulp mills. Obviously, this is a very structured process. One has to account for various inputs and data given by a large number of various sensors. One has to respond or program the system to respond in a repeatable and controllable fashion. So I think this is actually quite relevant.
Following that, I spent a couple of years doing that, I further educated myself in the field of finance, and then I started up a career at a fixed income trading house called JP Bank, a domestic Swedish bank at the time, and now we're in 1993. Here I started out as an analyst and basically spent my first year there analyzing commercial paper programs for various corporate (interests) that the bank represented on the market. Moving on, I moved to the position of economist. This is basically very much focused on macroeconomics, and indeed as this was a domestic house, 99% of the focus was actually Swedish government finances and political development. As I'm sure you'll recall, those were quite interesting days following first in the early 1990s, sort of famous attack by a certain George Soros on the bank of England, and probably, globally an 1/2 coordinated attack on Swedish currency into bank rates, moving as high as 500% in the late fall of 1992. Then we had our own crisis. Basically, what we did then was to inform out clients, to the best of our abilities, of any underlying macro-economic and political trends that would lead to a significant improvement in government financing and the debt situation. This is indeed something that a large number of global macro funds focused a lot on as they took significant positions in Swedish government debt.
So, in touching on your initial question - have I been an analyst? That sort of revises and changes. Well, to some extent, obviously, one has to account for new information as and when it gets released. On the other hand, I think it's always important to know that individual data points may actually lead you astray, and quite significantly so, as an economy actually does move and change in a relatively slow and moderate fashion. It would be rare, unless we're talking about potentially countries like Argentina or whatever, to see developed economies turn 180 degrees overnight. This has always been, in my mind, a slow moving process. One has to account for obviously new data, but only as an average, and trying to see the longer term development rather than focusing too much on individual data numbers.
So these were very interesting times, obviously trading fixed income and Swedish gov debt in the early 1990s. As the Swedish economy started improving in the mid-1990s, and basically throughout the 1990s, spreads to international markets - most notably the German market, started compressing quite a bit, which is obviously what our clientele had focused on, and hoped for. But at the same time as things started stabilizing, actually the fixed income market domestically in Sweden got a lot less interesting than it had been. So by 1997 or so, I actually started thinking, together with my co-founding partner who also worked at JP Bank at the time, we actually started thinking about a next move - what could we do together that would make sense going forward. Neither of us felt that this particular market was the place to stay around for another few years.
So early in 1998, and with a background of many of our clients being not only international hedge funds, but also a large number of more traditional asset managers - pension funds, life insurance companies, etc. most of them in Europe. We started thinking a little bit about our experiences in dealing with them, and one topic that came up over and over again was, from a pension fund perspective and a life insurance company perspective, what's the most important decision that you have to make, as a long term investor, to gain relevant returns on your funds? The asset allocation decision seemed to us to be central.
Obviously there are two parts to this. Most of these funds then and now have a strategic long term asset allocation, where they set targets for any number of years - 3 to 5 years typically - how are we going to be invested and what asset classes are we going to be invested in? But there's also a shorter term quite interesting decision that is being made frequently by these funds, and that is how do we deal with tactical deviations from these long term strategic targets? You can end up in a situation where you deviate intentionally, or you can end up there because you have let markets basically push you in that direction. So irrespective of what has actually taken you to that point, how do you deal with the situation? Do you want to continue to deviate from your benchmark and allow that tactical debt, whether intentional or unintentional, to continue to play out; or do you want to reduce or indeed even increase that positioning relative to the benchmark? So this tactical asset allocation decision is immensely important.
Now, at the time most pensions and lifers, I dare say at least in Europe, made these decisions in a very traditional way that obviously had made a lot of sense up until then. Basically they had their in-house economies, they invited the investment bank and other more independent economies, then they read a lot of research, and then they had an investment committee that sat down on monthly or quarterly frequency, compiling and analyzing all of this data from various sources, and then making a decision to the best of their ability. Obviously they were also exposed to the analytical community and economist community in changing their views and revising based on data points, so there's a lot of noise going into this process. Another thing one has to bear in mind is that most of the decisions made by these committees, or other organs making those decisions were typically relatively modest in size. People, based on this information, knowing full well that it's noisy, and it's hard to analyze, rarely dared to make the bigger bolder bets that would actually make the needle turn at the bottom line. So perhaps I'm exaggerating a bit, but the typical outcome of this would be to change your 60/40 stock bond exposure to 58/42. This may or may not have been right. You may not have had the solid background to do it, but irrespective of which, it's a long a cumbersome exercise that rarely contributed a whole lot to the bottom line.
What we had also observed, was that over in the US a few firms had already started helping pension fund and long term institutional investors with these decisions by way of providing basically an investment service to help them make more informed decisions based on a solid set of information, solid methods, and to really do this in a repeatable fashion. So this is really what triggered us to start up this firm. In early 1998 we sat down and said, let's try and do this to help, mostly European pensions and lifers with their tax classification decisions. So this is really the starting point of the firm.
Niels: Sure. Before you bring us up to speed on what happened from 1998 to where you are now, I just want to go back to the early 1990s, because you mentioned, of course, in 1992 the Swedish central bank was offering 500% overnight rates, and I also seem to remember that. In 1994, I think it was, we had a bit of a surprise from the Fed that caught some people off guard in terms of the rate environment, and I just wonder, with these extremes happening so early on, and I say early on in your career, but I don't mean it like that because I'm not entirely sure exactly when you would say that your career started, but these extremes, do you think that they really were the ones that impacted you in focusing on relative value - saying I want to be a little bit of a contrarian when these things happen because it doesn't make sense. We're not going to have 500% overnight rates forever. We need to do something going against that, or ...
Anders: Yeah, it's a good observation, and you can actually throw in another big event in 1987, when I had actually, on my personal account, still in school, started trading a bit, and I actually had rather big positions in the options market without really knowing too much of what I was doing. Obviously I noted that trading options for the better part of 1986 and early 1987 I'd made decent profits for a student, and then I pretty much lost it all in the ensuing equity market crash. That was sort of formative. It's a brutal way of learning. Before you start doing stuff and trading stuff, you should really know all of the ins and outs rather than just going with the flow. I think both of the events that you mentioned, the 1992 500% rates, here in Sweden, the 1994 Fed hike, but I think perhaps even more I learned these longer term outlook and the value of patients and solid analysis from many other counterparts that we traded with - basically the global macro hedge funds. That sort of coming into the market with the view clearly stating that hey, with currency at 523, 524, 525 to the Deutsche mark and spread (so several 100 basis points) this is going to play out in our direction over time, but you have to be patient. These are positions that you build to take a longer term view and really profit over the course of a couple of years, rather than trading in and out and individual months, or trading too much based on market volatility following the Fed hike, etc. So I think this way of analyzing markets, with a longer term view, trying to get rid of the noise and really performing the analysis based on sound theory probably came from there.
Niels: Did you have any mentors when you first started out that you looked up to, or someone who looked after you in a sense?
Anders: Not really mentors. Obviously some of the folks and analysts at these global macro hedge funds were quite interesting people, and I tried to learn from them. We also had management at JP Bank and head of the fixed income department at the time was a certain Kent Janer, who subsequently started up the hedge fund Nectar. He was a demanding boss. He required nothing less than perfectionism, really in our analysis even down to the way we produced reports and how we phrased ourselves, and what conclusions that could be drawn and not be drawn, etc., etc. He was also a person who was among the first to bring in, to Sweden, a more quantitative approach to trading fixed income based on his experiences from working in London a couple of years earlier. So, perhaps he wasn't necessarily a mentor, but he was certainly a person that set the bar and the standards.
Niels: Before we leave your story completely. Today, of course, I know you have stepped down as the CEO, but being part of IPM and being Chairman of that today is obviously a big part of your life, but what do you spend your time doing when you're not focusing on the business.
Anders: Oh, there are a lot of things that can be done. I have three kids. That's a time consuming business. Obviously, approaching 50 years of age, one enters into a period where one has to start looking after health a lot more. So I've taken to mountain biking, long distance running, but these are more of sort of a house keeping thing, but it's also fun. The greatest fun off work would probably be downhill skiing. I spend a lot of time doing that.
Niels: Fantastic. Great stuff. Now I think for today's talk we're going to be focusing on the systematic macro program, but perhaps you could just mention the programs that you run today, and when they started, and what kind of assets you run in each of them.
Anders: Sure. Basically we only have two programs here at IPM; one is the systematic macro program, which is a systematic global macro hedge fund, and the other one is an equally systematic but long only equity program. In the macro box, we are currently managing about 3 billion US dollars, and in equities about 4. However, the macro program can also be run in slightly different forms, so you can carve out, for example, the currency program and trade that alone; you can carve out, and people do carve out currencies and fixed income and trade that as a separate strategy. They're all part of the broader program. All we do when we carve something out is we simply turn off the other processes. The macro program has been trading in account format basically since 2002. Our pool vehicle started in 2005; that was a currency vehicle, then in 2006 the full macro program. Both are domiciled on the Caymans. The equity program has been running since early 2006.
Niels: Fantastic. Great stuff. Now before we jump into the first topic more specifically to IPM, I want to ask you... you mentioned the traditional 60/40 bond stock, or stock bond, depending on whether you were European or whether you were a US institution - they seem to have a little bit of a different asset allocation, as far as I remember. The world has changed in the last 10, 20, 30 years of course, and I just wonder, from a really big picture point of view, how do you see them dealing with this asset allocation, not from what you do, but from what they do and how they may interact with firms like you. Are they becoming more open, so it's not just 48/52 that they changed to or whatever it might be, are they starting to take bolder decisions in their own asset allocation?
Anders: There's a whole range of answers to that. I think, generally, the answer will be yes, on sort of a global basis. I think if you look at US institutional investors; they probably come further than their European counterparts in allocating very significant parts of the risk budgets to folks like ourselves, and generally hedge funds. Whereas, in Europe, finding a long term institution having a hedge fund allocation exceeding 3%, 4%, 5% would be unusual. Finding the same in the US would be the norm, being north about even up to 10%, 15% that would be the norm. But then you have, obviously, various examples. If you look at your own native country and the big pension fund there, ATP they instituted, I believe about 10 years ago, a radically different asset allocation structure from the traditional, where they basically, and you can correct me if I'm wrong here, but basically what they do is they sort of equal weight their risk budget across 10, 15 different asset classes ranging from infrastructure to traditional markets.
So people are doing a large number of different things, but I think the general observation holds true that North Americans, generally speaking, are much more into seeking alternative sources and allocating significant parts of the risk budget to those alternative sources of returns, than Europeans are. Obviously this also has something to do with the state, generally, of the pension systems. If you look at US pensions, corporate and state, they are generally underfunded to a very large degree - 60%, 70% funding ratio, whereas most European countries that have funded pension systems, would have significantly higher funding ratios. Many of them actually 100% or better. So this obviously changes your long term strategic allocation. Yes, you have to match liabilities, but if you are underfunded, and if you are running something at 70% funding ratio, than you better seek methods to make up for the shortfall pretty quickly.
Niels: Sure, sure. Now it's interesting that you mentioned ATP clearly that that's a big, big pension fund in Denmark, and they have been, as you say doing things a little bit different in the last years. What I seem to remember also from the press, because they did set up an alpha team, and so on and so forth, to actually take some of these strategies in-house, which I think, from what I know, you actually are not completely unfamiliar with, that some of your clients might be wanting to learn from you, but we can talk about that. But what I also seem to remember about them is actually that they ran into a little bit of problems from, whether they were internal or sort of image point of view about having people sitting inside a pension fund who could in theory earn a lot more than an average salaried person because they were doing strategies and maybe earning some kind of performance fee, so I think actually part of that, unfortunately, has been shot down, but that might be isolated to them specifically.
Anders: It's a general problem. Obviously if you are going to bring in people that can actually do this, you have to pay them market average. Otherwise, you're not going to be able to. Additionally, I think most of the people that have tried insourcing have reversed and stopped doing that. I think one of the reasons is political and recently regulatory, but it's also the fact that inside an organization like that you're not really going to have the entrepreneurial spirit that sort of constantly improves on this process. Whereas if you outsourced to other parties, that live each day by what they make, that is going to be a different story. So inherently I think insourcing these rather specialized forms of management is very difficult.
Niels: Yeah, interesting. Now you mentioned roughly 7 billion dollars under management today. I want to talk just a little bit about the organizational structure. Running a big company like that, how have you decided to organize it, and what things, if anything, are you able to outsource today in terms of taking advantage of technology, or specialized firms helping out, and just tell me a little bit about your organization as it stands today.
Anders: I think the first point to note here, by way of background is to note that we are 100% systematic, and by that we mean that everything that we do inside these two programs is actually coded into software code. That's sort of a starting point, but it's also very important, because with that structure we can do things, and we can scale the organization, or the output from the organization completely differently from people that act more exclusively disgressionarily.
So what we have at IPM, at the heart of IPM, is a research team comprising about 10 people, supported by another 4 systems developers. Next to them, out here on the trading floor we have three traders and a couple of guys on risk, and this is at the heart. So their daily task is really running and maintaining the research team I'm talking about they really run their maintaining and further developing researching the models that we employ. Obviously they are also the people hitting the button on a daily basis to generate the trades and generally supervising that, but that's in relative terms, a small part of the daily work, typically. So that's the main part of the machine.
Supporting them we have our own back, or rather more middle office, that does the traditional things - they check for risk and they check positions and they reconcile with counterparties and clients, but that's actually only 4 people. The remainder of the firm, then is traditional functions, legal, compliance, IT (as opposed to systems development IT would be infrastructure), business development, and key account management, or as we refer to them as investment strategists. But it's really... the organization is really designed with us being 100% systematic in mind. Really, to compare with some, we are probably organized pretty much like your average CTA.
Niels: Sounds like it.
Anders: There wouldn't be a big difference there, although the inputs that we use and the models that we use are completely different, but operationally we function pretty much the same.
Niels: OK. In terms of growth of the business, I can imagine that maybe legal and compliance is an area that is always going to grow the way the world goes, but other than that, where would you like to see the AUM go to, based on what you have now? It sounds like you don't really need a lot of changes even if you had a significant increase in AUM.
Anders: That's a question that we always get, and the typical answer would probably be that I'd like to see our potential realized, and by that I mean that we'd like to get our AUM proportional to the size of the asset pools in various markets globally. That's a cryptic way of saying that we could see a lot more coming in from the US. We just started marketing over in the US a few years ago. It's obviously a hard market to break into, and today we have a very small amount coming from there, but if we're going to move to a position in the coming three years where our AUM split reflects the global assets, then we would probably double or even triple our AUM which is, of course, very ambitious over something like three years, but longer term we'd like to get to that point. And yes, as you observed, that does not mean that we are going to have to employ a large number of people. Obviously some business developers and a few account managers, but in terms of research and other operations, not a whole lot.
Niels: Just one sort of final area before we jump on to some more the program specific issue, how do you spot talent that you want to hire for your business? What are the skills that you are looking for today, and are these skills different than maybe 5 years or 10 years ago?
Anders: It's an interesting question that we can spin different ways because it actually applies to many of the things that we do also in research. We're not out there just looking for general talent. We get a few applications per week from very talented people globally and domestically, but we're not too interested in that. Rather, we have taken the opposite approach. We're saying here's our team, here's what we want to do, what do we need to get there? Do we have the talent we need, or is there a specific area that is not completely covered? If we find such an area, then we actually go looking for that specific talent, rather than the reverse. The same actually holds true for our general research, how we further develop the model, the same approach, we are looking at the model; we're trying to identify obvious strengths, but most of the focus is quite obviously on weaknesses, so once we have identified weaknesses in areas for further improvement that's when we try and define how do we fix that, and that goes into research, rather than scouting the market for general sources of alpha.
Niels: Final question on this, Anders, and that is, how do you compensate and retain talented people, because the world has changed, and I'm not sure that it's just about financial reward anymore? What, in order to make people stay with a firm long term...and we know in our business that if you have someone good, especially in research and so on and so forth, that are a little bit sensitive, ideally you want to keep people for the long run, how do you see that?
Anders: I fully agree with what you say. Perhaps outside the financial centers of London and New York, but increasingly even there, I think people are motivated by job satisfaction - having interesting jobs, being recognized for what they do properly rather than what you get in your paycheck. Obviously the paycheck is going to remain important in this business, and obviously there's going to be competitors trying to poach our employees, much as anyone else's, but we reward people here based on what they contribute to long term business development, and, importantly, at this firm, we don't have remuneration systems focusing on individual returns, like portfolio managers or prop desks or whatever. Really, everyone gets compensated based on company EBIT, and then we distribute based on contribution to long term development, idea sharing, things like that entirely. Obviously it does help, from a retainment perspective, that we are based in Stockholm. Relatively few competitors would actually target Stockholm as a market to go looking for talent they can poach.
Niels: Interesting. Thank you so much for sharing that. That's a really interesting insight. Now I want to jump to the next subject which I tend to call track record. What I mean by that is that looking at your track record as a whole, knowing that the market environment has changed over time, is there any particular way that one should look at your track record - certain stages, certain points in time where you may have done some upgrades. How do you look at it when you look back on your track record right now?
Anders: I think a very important point is that what we do, as we invest client money based on fundamental input and economic analysis of that input, is that our program is going to be relatively long term in nature. Holding periods are going to be significant, and it's really not meaningful to discuss three month performance, six month performance, even annual might be misleading, so really you should look at our track record probably in the form of a three year moving average or something like that. That's point number one.
Point number two is you are referring to a changing environment, and yes obviously as we all know the macro environment has changed quite a bit over the past 10, 15 years, not least over the past 6 years with the central bank interventions and ongoing interventions, etc. etc., but what we do is really playing markets in a very relative sense against each other. So we're trading US equities against German equities. We're trading Canadian “govys” against Japanese “govys”, and everything that a model does is done in a very relative sense. We don't really care about the absolute levels, for the most part - that's 90% of the story. This means that, in terms of regimes and changing environment, we're probably less sensitive than someone that models each instrument in its own right, and based on each instruments absolute path.
So really the way to evaluate this program is to recognize you should be looking at 3 year moving average, bigger dislocation in various markets may go against us for some time, but typically they do come back, and they almost always do. To my mind they have always done so, but it may take a little bit of time - so this longer term strategy and really relative. I'm not saying that we're 100% immune against changes in regimes, but I'm trying to say that, looking at evaluation factor, all currencies, relative inflation between two countries, in the end relative inflation is always going to have a role in the pricing of a currency pair, irrespective of everything else that goes on. Over time that will play a role. It may not play a role for 3 months, 6 months, even 18 or 24 months, but eventually it's going to play out. So if you trust your fundamentals and you have the patience to sit it out, knowing that it will bear some real drawdown risk, and it's a good thing to do.
Niels: You mentioned, of course that the last 6 years have been different, abnormal, maybe even in terms of the role that central banks are playing, and if I understand you correctly you are saying that actually over time these things will play out and it will go back to normal. What extremes, if I can use that word, do you think that we are seeing right now in this environment that you think will be sort of the ones, when they do go back to normal that we are going to feel the most?
Anders: It's difficult to rank these extremes. Generally, my personal view on this is that, obviously way outside of our investment model view, but my personal view would be that we are in a very stretched territory, both when it comes to fixed income valuation and levels, but that's really controlled by central banks, but I also think, if you look at equity markets generally that are being driven and have been driven for a very long time by cheap money. You can extend this and look at credit and high yield or what have you. I think the bigger portion of asset markets that they are actually in stretch territory. That they are heavily dependent on the continuation of cheap money. So when that changes, and indeed it's going to change at some point, we're probably in for a rough ride. Having said that, obviously central banks also know that if they suddenly reverse this, they would have a whole community of depressed pension funds and other people to take care of, so I don't think they are going to make any drastic moves.
Niels: No, no. In terms of the strategy of the years, before we dive into the strategy itself, what would you say has been the main upgrades, or the evolution of the strategy over the years? I wonder also, in that particular point, about..I understand the point about research and finding new ways of improving existing models, but is there actually something that you would say model decayed, to a point where you would say a model doesn't work anymore? Of course, I can assume that there're certain things that could completely change fundamentally that would make it redundant, but other than that, where you just say this model might not work. It's not just about the new research I'm interested in; it's actually also about whether there has been anything that you had to take out along the way?
Anders: I think, generally speaking, this is again a long term evolutionary process, rather than sort of big changes, short term. Typically we would introduce two, maybe three changes to the model on an annual basis. Those could be additions of new risk factors or sub models in a more general lingo. They could be new risk analytics, methods being introduced into the system. They could indeed be entire new models or asset classes. We just started trading last year emerging market currencies. We haven't done that previously. Historically though, I would say probably one of the biggest changes was when, four year ago, almost to the day, we relaxed the traditional tactical asset allocation restriction, namely for every dollar you're long something, you have to be a similar amount short. Which really comes from the classic CTA world where if you overweight something, you have to underweight something else. We still retain this in all of our relative models where we trade equity markets against child or bull markets against each other and currency markets, but what was previously referred to as the global asset class decision, where you went long $100 dollars of global equities based on some composite, then you had to go short $100 worth of bonds - that one we relaxed four years ago. So today, we can actually be, in that dimension, we can be long both global stocks and bonds, or short both, or whatever accommodations.
Niels: Why did you relax it? What prompted you to actually make that change? That's a big philosophical change perhaps...
Anders: Yeah, it's a big one, but it's also a very natural one because again, as we moved away from the classic DTAA type setting and most clients did by saying this is another alpha source - this is a global macro systematic, then there's no particular reason to have it, for starters. Another reason is, from a pure trading a risk perspective, it is not a very balanced way of doing decisions because what was targeted at the DTAA level was really dollars, so buying $100 worth of equities and shorting $100 worth of bonds, obviously that's not going to make you risk neutral in any sense of the word. So from that perspective it is not really a meaningful restriction. Additionally, if you look at asset class correlation structure, it has changed quite a bit over the past 15, 20 years. So as that was built into the model - hot coded into the model, it was probably, even with the way we've designed the model today, prior to 1995, 1996, 19997 we would probably have found ourselves in a situation where we were long stocks and short bonds most of the time or vice versa, so the correlation structure. Whereas over the past few years we have been long both or short both more often than not.
Niels: Yeah, yeah, absolutely. Now I wanted to jump to the actual strategy and talk about that. A couple of questions initially is how you would describe the objective of the strategy, and also, each strategy has a certain environment that it performs well in, or it's been designed to perform well in, and vice versa, and I think maybe investors, to some degree, may not recognize this. They buy things because they think they are just going to continue to make money, but they don't necessarily appreciate that there are certain times where the environment is simply not favorable for a particular strategy, but it doesn't mean the strategy is wrong, it just means that you have to accept that there is nothing that makes money every single day. So how would you describe the strategy, the objective and the environment it has been designed to work well in?
Anders: This is a strategy that we can run at pretty much any risk level, and we can always discuss what we mean by risk here, but let's say that we can run this at any level of expected volatility in the program. Most investors, and as we cater only to institutions they would start we would start feeling uncomfortable north of 20%, 25% expected. Your typical institution, really what they need to get out of something like this is 10% net or there about. So what we've done is that we've said let's find the sweet spot for the strategy, at sort of 10% net per annum to the client. We need to run this at a particular risk level measured as acceptable volatility to make sure that we deliver in the end, and indeed, that's pretty much where we are over the course of all of the published track record which is based on one of our investment funds, we're at 9.7%, 9.8%.
With a strategy like this, this is not accomplished by running a sharp of 1 1/2 or 2, we don't think this is realistic for this particular type of strategy, given that we're always in the market in all of the instruments that we trade on a large number of themes, there is a significant risk of drawdowns on any different dimension, anything north of a 1 over the longer period doesn't really sound very plausible. A reasonable target for a strategy like this and this goes for most systematic managers, would be just south of 1 in terms of sharp.
What type of environment is it defined to trade best in? Ideally, obviously it should trade well in any type of environment, but generally, as we build a model based on fundamental input, we are concerned with GDP level, we are concerned with inflation; we are concerned with changes in money supply rates, etc., etc. We are relying on markets that, over a reasonable time period, responds rationally to these fundamental variables. That's a fancy way of saying we want the markets, even if they start deviating, or they go against a particular underlying fundamental development that they will mean revert at some point in time and we don't want that point to be 5 years out; we want that point to be 1 year, 1 1/2 years out. That means that if we find ourselves in a market environment where people simply don't care at all about fundamentals - let's say they care more and only about what the Fed does, and then in such periods this model could not be expected to perform optimally. It could still deliver, but then it becomes more of a random game. The type of positions we may enter such a period in holding may be the right ones, and they may also be the wrong ones, and there's really no way of telling. So in that period and that shift from period A to period B, the model can be a little bit vulnerable.
Niels: So if I try to summarize, what you're trying to do from an objective point of view, is to take risk in relation to the opportunities you see.
Anders: Yes, very much so.
Niels: And also we know that we have been in a world where a lot of countries and certainly developed countries have tried to coordinate, at least that's how I see it, to try and coordinate their economies as more and more integration and if I look at different cycles of countries, they... at least for a certain period of time, they were starting to come together much more coordinated. But now, of course as we know, and I completely agree with you, the world seems very stretched, and there seems to be a lot of tension building up in trying to make everything the same, for lack of a better word. So your strategy, in a sense, may find it more difficult to find opportunities as they're trying to coordinate, but now where the dispersions are building up and maybe more opportunities for this particular strategy are emerging.
Anders: Yeah, it's an obvious sort of observation. If we started out in the early 2000s or even in the 1990s, it would have been a large number of opportunities: trading, Dutch equities against German equities, or German equities against French equities, today that's not really meaningful. We still trade many European markets, but it would be rare for the model to find, or to recommend a position where we would explicitly go long Germany against a short position in France, simply because of the integration. This is driven not only by authorities, central banks, etc., but it's also driven by a very natural development. The world is getting much more integrated, certainly in Europe. Countries and economies are getting a whole lot more integrated, and companies are getting a whole lot more global today, a very relevant question is to say if you are trading the Dutch equity index, what are you actually trading? Are you trading the Dutch economy, or is it a global energy play? So that's one level of observations.
Yes, from a model perspective, as it gets more integrated and/or as authorities - central banks try to control the market, we are going to have to look elsewhere for opportunity, which is one of the reasons we started trading in emerging market currencies. There are other things in the research pipeline, other equity markets, potentially other asset classes, that we are going to start trading over the coming few years, pending a successful research process. As the world becomes a more integrated place, yes, you are going to have to start looking at other instruments, other asset classes, other segments of asset classes to pick up on what you need to pick up on. That said, as we have been now in a long period where central banks have been very aggressive - whether we call that 6 years, or 3 years, or 4 years, we've been in a long period, and people have starting getting used to this. So, on a relative basis (given that this is a global phenomenon), on a relative basis markets actually do trade around, and it's meaningful to trade markets against each other on a relative basis, although the opportunities, per se, may be a little bit smaller, and volatility is certainly a lot more compressed.
Niels: I think we'll probably touch upon this when we come to risk management, because obviously maybe the fact that as you say opportunities become smaller. We've seen certain players compensating by leveraging up and maybe that's come back to haunt them. I want to touch upon that a little bit later. I want to dive into the strategy, and I want to ask you what global macro really means to you, and you have you taken that concept in building your model? What kind of themes are you focusing on, and how you've really structured taking so much information that you can obviously get when you are talking about the globe as your playing field, but structuring that into a systematic trading program?
Anders: One way of looking at this is to say that any global macro trader - and let's forget about systematic or discretionary for the time being - any global macro trader would be interested, based on sound analysis, sound principles, identifying price discrepancies, based on either something that is supposed to mean revert over time but is now moving south and you are taking a position based on it's going to be moving north at some point, or you are taking a position based on you shorting a risk premier that seems to be excessively compressed at the current time, or you are going long in an opportunity. So you try to use sound analysis to identify these situations, and what really differentiates us from a discretionary trader would be to say that we have identified a large number of such opportunities and they're coded in the form of risk factors. We're always in the market for all of those, but to varying degrees. So, as you said earlier, we're actually going to take position directly proportional to identified opportunity.
Whereas your discretionary manager, he would probably distill this and try and find 1 to 3, maybe 4 of the strongest themes and then bet everything there, and then he'd capture his profits, and then moves to the next area. Whereas our model is for each of the markets that we trade, for each of the themes that we trade, pretty much in the market at all times. This is at the heart of the model. We believe in mean reversion for certain phenomena, we believe in trading obvious risk premier. You can go long or short depending on where you find yourself, etc.
So the model is built... on the one axis you have five dimensions, and those would be the sub models that we trade, so again we trade relative models of emerging market currencies, developed currencies, equities, and fixed income. Then we have this, previously known as asset class, but directional components, so that's five dimensions on one axis. The other axis you would find the types of phenomena that we are trying to pick up on, and this is common for all of the five sub models. We are trying to identify valuation drivers. So, for example, you could probably think that something like purchasing power parity is a sound valuation methodology for currencies, so that would go in the value box and similar type factors for other asset classes. The next dimension would be risk premier. Well, obviously risk premier long term or bonds, in currencies you probably be talking, you may say that the carry is a risk premier, and indeed I think it is. So things like that will go into risk premiers. We're trading both of those. Thirdly we find macro-economic factors. As aim to apply monies expands or contracts, as government debt issuance expands or contracts, as trade balances shift between countries, you could take position on that. Finally, we have a 4th box that we refer to as market dynamics, which is really our way of saying here is the box where we put stuff that is really idiosyncratic to each of the asset classes that we trade. So, we would look at the government bond curve, the deal curve or term structure if you like, that would be something that we focus on in our efforts in bond markets. It doesn't really apply the same way when you trade currencies. We are trying to forecast things like investment flows, cross board investment flows for currencies, which is probably much less relevant for our equity trading. So, 5 sub models, sharing the common 4 themes and all of them are equipped with factors identifying value, risk premier, macro developments, and then the 4th box the market dynamics.
Niels: Sure, sure. So you have this structure. So in practice, and I hope you can explain it much more elegantly than I can, but I'm trying to visualize it for our listeners just as how do you do these things? So is it something along the lines where you would say, OK let's look at inflation, and say inflation is now going up in the US, so if that happens, we have a model that looks at this, and it gives a certain score into a bigger pot, and then you have other things that gives other models that look at different things, and they all give a little bit of a score, if I can use that word, and that then determines, overall, what your position should be in that particular currency, in the stock market...how do you put all these things together?
Anders: That's not a bad way to phrase it, actually, let me try to rephrase it back and paraphrase it a little bit. So the first thing to know is that when we model, we model everything in relation to global composites. This is really to get rid of otherwise the need to model absolute developments. So when we are looking at government bonds, or government bond futures to be precise, we are actually looking at how the prices and the factors driving those bonds behave in relation to the global basket of such government bond futures - the same for equity index futures - the same for currencies. We're not looking, and this is probably enlightening, we're not looking at the traded currency pairs, which is what most people would be looking at the Dollar/Yen rate or the Euro/Dollar or whatever, but we are actually looking at the Yen in isolation against a global basket of currencies. We are forming synthetic instruments when we model. Additionally everything is modeled in relation to its own history, and risk adjusted. The fancier way of saying it would be we're looking at normalized risk adjusted and really what comes out of this is ZED scores - number of standard deviations away from equity rim - long term equity rim. So when we are looking at, for example something as simple as 10 year government bond rates as an indication for value in bond markets, we're taking, for example, the JGB rate and we're comparing that to this global composite, but we're not taking it outright, we're taking it's risk adjusted deviation from where JGBs normally trade, and that's what we are comparing. So it wouldn't be meaningful to take the JGBs and compare them to T-Notes because they would always be expensive, right?
Niels: Sure. Sorry to interrupt on that point, because I think that's an interesting example because what is the norm in JGBs anymore? 5 or 10 years ago we thought maybe that was 3%; I don't know, I'm just picking a number here, but, of course, now it's been locked at almost nothing for so many years, when do you change the norm? Meaning, over how long a period to you actually need to look in order to say this must be the norm, and we are now away from that?
Anders: For most markets and for most models we're actually using as much information as we can - the longest reliable information that we have. The reason for doing that is really that markets will change, and we will face different scenarios, but it is very, very difficult for us to say that the past 5 years are going to be representative for the coming 5 years, or the past 10 years. Bank of Japan just started their program a couple of years ago, well; a year and a half ago, and that changed the equation. Who's to say that they are not going to stop next year. So if we say that the norm today should be the past couple of years JGB rates, and then apply that looking forward, and then all of a sudden the Bank of Japan stops doing what it is currently doing, or they run into problems and cannot do it and JGB starts to explode, we can be caught entirely off guard. So try to be unbiased by using as much information as we can. Obviously if we were to find a situation...one extreme would be if a central bank actually starts pegging a currency, then they want it changed to the fact that we can't trade that instrument anymore because that instrument clearly will not respond to fundamental information.
Niels: Is it important for you to understand why a market might move back to fair value, or could you simply accept it as being everything moves in cycles and therefore at some point we're going to go back to the mean?
Anders: It is important. Generally underselling is usually important. We think it's important that our clients understand what we do and how we do, but in our modeling, we want to be able to understand what are the forces that would push prices away from or back to fair value? What are the forces that would cause a particular risk premier to become over or under priced and then subsequently return onto some longer term equilibria. So it is important, and indeed we are trying to model those forces. If you look at our model today, there's going to be factors that are stronger in one direction than the other.
Niels: So, in a sense, even though it's systematic, it's not a matter of just quantitatively assessing that there is a relationship that is stretched, it's actually also fundamentally understanding why it's stretched.
Anders: Yes, yes, very much so.
Niels: How many markets do you trade today?
Anders: Currently about 40 markets.
Niels: Am I right in saying that you actually don't trade anything over the counter?
Anders: Well, currencies would be over the counter for the most part, but otherwise exchange rate, government bond futures, exchange rate equity index futures, and then currency forwards.
Niels: And it's all financials, so no commodities at this stage?
Anders: That's correct, which is not to say that we're not going to start trading commodities at some point in the future. At the current time, we don't.
Niels: Are there any of the types of strategies that you use within each theme that you could try and visualize for us. I know you've talked a little bit about it. For example, I'd like to talk a little bit about carry. I'm no expert, but obviously carry has been quite a big source of return for many people for a while, and actually my guest last week was quite concerned about some of these trades that are being put on. In his opinion, by large, asset managers to compensate for not making so much money in the directional arena, because if we look at currencies, at least developed currencies, volatility has gone down dramatically in recent time, and so maybe talk a little bit about how you see, if you are going to drill down in your carry models, how do they work, what do they look for in your world?
Anders: Well I think that as a starting point, and this is, I don't know who said that first, but carry trading is really about picking pennies in front of a steam roller. The greedier you get the closer to the steam roller you are going to be. Typically...
Ending: That's all for this episode of Top Traders Unplugged. We'd love for you to be a part of our community, so head over to TOPTRADERSUNPLUGGED.COM and let us know what you thought of this episode in the comments section of the show notes. Take action, get involved, and suggest who you would like to see as a future guest on the show, or how you think we can improve. Constructive comments will be rewarded with 30 days of free access to our premium member area. So head over there now, and we'll see you next time on Top Traders Unplugged.
Become An Insider
Subscribe for free and be the first to receive new and exclusive interviews with the world's top traders. As an insider we'll also send insightful bonus content direct to your inbox.Free Instant Access »
You might also like:
Date posted: 18 Aug 2014no comments