10 Top Traders Round Table with Ryan Abrams, Ela Karahasanoglu, and Carrie Lo - 2of2

Niels

…on the scene about manager selection, I think we’ve all learned about the benefit of diversification and how this is the only free lunch in finance. But, at what point does diversification become de-versification – meaning at what point do you need to stop adding more managers? I don’t know, who would like to go first on that one?

Ela

I’m happy to go first. We do have a little bit of that. So, just to go back to the previous point, the difference between the managers and how that has changed over time. I was working at a quant macro CTA between the periods of 2005 and 2011. So it was an interesting period because I was really trading. I was the head of trading as well as the quant research in 2008 which was the time where everything really happened. Markets went haywire.

Around then the CTAs (it’s really a misnomer – I should call them managed futures) and the space weren’t as different or as broad, if you will, at the time. The trend following was trend following but you didn’t really have the newer, big data, systematic type of strategies – Artificial Intelligence and the ones that have come into place that are not really exactly trend following or are not even connected to trend following at all, but they are still in the same bucket of strategies.

So I think that has changed the spectrum and that it has changed the profile of what we call managed futures. It’s not one style, it’s difficult to bucket them. It’s really a whole host of strategies. It’s been difficult to pick and choose. There are a lot of strategies that look like each other, having covered that space for some time and having worked in it. It still is not easy to pick and, as Ryan alluded to earlier, the correlation really converges under certain environments. You don’t really get a whole lot of diversification, which brings me to the question that you just asked.

We do have fund of funds, predominantly in our portfolio. That was necessary, given our size and the resources at the time. We’re still a relatively small team, so it does still make sense. When you fund the funds, what that means is that you actually have a portfolio that ranges from 10 to 30 managers. If you have, say, two to three managers now you’re roughly at 40, 50 to 70, 80 managers line-up. Then everything really looks pretty similar and you start seeing allocations that are not material.

So, the way I would think about it is there’s no specific number that would be applicable to everyone, but having worked with institutional investors over the years, the magic range, if you will, depending on what you’re trying to cover – how many types of strategies or asset classes you’re trying to cover, if you’re looking at one asset class I would say it would be running between 10 and 20. That’s where you would be getting decent diversification into an asset class. if you’re looking into a subset of an asset class you could even get that with maybe 3 or 4.

Niels

Sure, that makes perfect sense. Now, let me get back to you, Carrie. I wasn’t entirely sure from your first answer whether you do use beta replication or whether you only use the real thing, if I can put it like that, in your portfolio. My follow up question would be, I’m just curious here when you look at these types of investments and certainly also during the implementation phase, do you try to time these strategies a little bit? Or, is that just something that you say, “Well, we can’t do it?”

Carrie

Sure, as far as getting access to time series momentum beta, that is what we are trying to do in our trend following program with the medium/long-term exposure.  We did want pure trend exposure. We selected managers that had very little or no other types of diversifying models in their programs. The Systematic Risk Premia program, it’s a little bit different in that we define that typically, as long/short various style factors like value carry momentum across multiple asset classes in, generally, a market neutral fashion. So that definition is a little bit different.

As far as timing trend following, we ask all of the vendors that we meet with if they can predict when trends will happen and none of them can. So, I think it would even be harder for me to do that. So we have a systematic allocation exposure to this area. We don’t know when the next crisis will happen, what form it will take.

Niels

Sure. It’s not only the crisis areas where these strategies do well so that makes perfect sense. Now, of course, we’ve mentioned the word a couple of times now, so I think it’s time to talk about it and that is the word fees. I think, and this could be a little bit controversial, but I think a lot of institutions have been sold by firms that offer the low flat fee products that, in certainly the trend following space, it’s so easy to capture these returns, and that therefore it should be very cheap to deliver.

So, I’m interested in your opinion about what is the right kinds of fees nowadays and what are the right levels of fees, and if fees are more important than net performance. From a psychological point of view, when you must present these things.

The reason I ask this is because I haven’t really seen any of these low-cost products, at least in the trend following space, that consistently and over a longer period have outperformed the established trend followers. I even noticed, the other day, that there’s a new index, from Bridge Alternatives, that tracks the five biggest managers that provide these low-cost trend products. That index even outperformed the SG CTA Index. So, personally, I’m not a big believer in this approach, but I know that may not be the opinion of you.

So, Ela, why don’t you start on this one.

Ela

So, there’s actually two questions in there… Let me see which one I would like to tackle. I can start with the fee and maybe and the combination of management and performance fee and really, it’s the flat fee products versus the flagship products that are the home-grown versions of those – the non-beta products if you will. We’ve done some work on fees since I joined and it has been a big deal recently, as we’ve all been reading in the media. So, my perspective on the fees is really that there needs to be a good combination of management and performance fee. We’re talking in the alpha land, not in the alternative risk premia land.

The way I think about it is management fee should be per unit of volatility, rather than just a set number. So, just the 2/20, the way it has come about has emanated from 10% vol, 10 % return – this sort of thinking of 1 IR or 1 sharp, depending on what you use for benchmark or risk rebate. That’s the pedigree and that’s how this number has come about. The 2% of 10% of return portfolio is not the same thing as a 2% of a 5% of return portfolio. So, it does really eat into performance faster, which really drops the information ratio faster. This has not really happened.

The volatility of a lot of the strategies has come down. The fees have not been adjusted, the management fee if you will. The performance fee is really an ad hoc number. Our perspective has been that we want to align ourselves with the managers, so we want to pay the management fee for two reasons: one, the management fee really pays the bills and keeps the lights on and that is what managers live on, are based on. As the assets grow, is that really as important? Also, without a management fee – a performance fee only structure could create an environment that could lead to excessive risk-taking to meet performance.

So, management fee really creates a basis for the management to take a longer term approach and not be rushed to get to the return so that they can stay afloat. If there’s no performance there’s no money. They need to get to performance. It needs to be a combination of the two.

We’ve been working on it and we’ve made good headway except in managed futures. Macro and managed futures have been difficult ones, partially because a lot of the strategies are really targeting volatility, so the vol levels are not lower, but the returns levels are challenging, but I think the evolution has not seeped into that part of the market. Saying that, there are exceptions to the rule.

There are managers who have gone down the path of doing alternative risk premia or changing their fee structure. I would agree with you that the performance may not be as good as the flagship product, but depending on what you’re looking at, the marginal benefit may not justify the performance. I’m not saying that it does not, but we do consider whether it does make sense to go for the full-blown flagship product versus the flatter fee, the set product that just gives you what you need or the medium to long-term product.

There is no yes or no answer. We actually combined the two, so it’s the smaller portion of that macro portfolio: roughly 10% to 15% that’s beta – alternative beta. The rest is really alpha strategy, but we do have also hedge funds beta type of strategies in there that doesn’t replicate, but it does look to exploit hedge fund risk premia. Those are also cheaper products that do deliver a similar return profile with actually a lower volatility.

Niels

Sure. Now, Carrie and Ryan, feel free to add anything if you want, otherwise, I wanted to (in being considerate of our time) jump to the next question unless you have something to add to this about fees.

Ryan

Really quickly I was just going to say that I echo a lot Ela’s those point. I think in looking at return per unit of fee or unit of risk per unit of fee makes complete sense, and distilling that into what percentage of the gross return is allocated among the managers and the investors is the easy way to distill that. Then, I think from there, looking at this through the lens of alpha and beta that really helps determine what’s appropriate and where the market settles in terms of pricing for return streams are really high quality and novel.

Investors would be willing to pay more for returns that are not novel and very easy to get: stuff like equity index exposure. They’re basically free and trend following is somewhere in the middle. As a result, that’s why the fees are or tend to be somewhere in the middle as well. Depending on how simple the strategy is or how much additional thought or work has gone into things like risk management and portfolio construction, to the extent that those things add value managers should be able to justify higher fees and investors should be willing to pay them.

Niels

I’m going to be jumping back to you, Carrie. So, if you do have anything you want to add to this, feel free to do so. My question is more with regards to, this has been a very long-term project for you, as you mentioned, and it took substantial resources and time. What did you learn along the way that you, perhaps, could pass on to other pension funds and investors who are considering adding risk mitigation allocation?

Carrie

That’s a great question. It certainly does require a lot of resources and time. It’s a very complex area and we didn’t approach hedge funds as an asset class. We looked at each strategy individually and really hedge funds are just a legal structure. It’s really the packaging around what type of return stream we’re trying to access.

Being very specific about what your objective is for that investment, and communicating that well is, I think, very key because if your object is a different type of return stream as opposed to a risk reduction type objective than the way you approach it could be very different. Just having a consistent story and a lot of data to back that up is very important.

Niels

Sure, absolutely. Ela, do you approach it differently? How do you approach this from your experience in what you’ve learned and passing that on to other people who might be looking in this direction?

Ela

The experience has been pretty much along the lines of what Carrie mentioned. I’ve done a lot of analysis; I’ve written long, long technical notes in my previous role and talked to institutional investors. But, we have one client and that’s WSIB Investment Committee and talking to them we have created extensive research and we worked on numbers and it has been a lot of second-guessing ourselves.

This was not any easy path and even though I do understand the managed futures space, I’ve been in it for over 10 years now and I do understand the details of it, but there are so many little nuances that we needed to get the story right. Every portfolio is slightly different.

So, we had to do our homework and we had to go out and understand what else was out there. Before embarking on this I think it’s important to understand what is really the end game and to understand the timeframe that you’re looking at. If it is a shorter timeframe and people sometimes get deluded into thinking that it’s a quarterly return profile. Quarterly returns don’t really mean much. It’s really longer timeframes.

What I’ve learned over the last year that I’ve been here is that it’s about repeating the same message again, and again, and again, and ensuring that you have all the information to back you up and all the research because the questions that sometimes come up are something that you may not have thought of, or you wouldn’t have thought of in a long time, but that needs to be addressed. You need to fill out all the questions in your mind before you embark on it.

It’s not just about the first step, you need to think of the entire endgame and that could be years. I’ve learned that it’s not just about the first part, it’s the vision. You need to share the vision and have a vision of where the portfolio is going. From an institutional perspective, I think that’s key because that’s not where you’re here now, but where you might be standing in five years from now and what the needs might be and how you need to be flexible.

Niels

Sure. That’s a great answer.

Ryan, I’m going to ask you a slightly different question, if you wouldn’t mind. A lot of investors who would want these strategies often what them right after a crisis where they realized that they should have had them but they didn’t. So, money chases returns and I guess that is also the case for institutions. Therefore, there was a big flood of money into the CTA and trend following space, managed futures, after the crisis of 2008.

Of course, as history often does, it proves that it hasn’t been the greatest time necessarily to be allocated in, in recent years. I just wanted to ask you, with all the research that you have done, if we should pass on some kind of advice or perspective – similar to what Ela just alluded to – how should investors look at this space when it comes to returns? Is there any reason for concern by the fact that we’ve had lower returns in the last few years?

Ryan

I think this goes back to trying to just understand the very fundamentals of what the strategy is, how it works, and where the returns come from. I think that, hopefully, helps with this question.

So, trend following generates its returns from trends in the market and a lot of trends tend to happen during dislocation periods. This recent period, since the crisis has been… I don’t know if it’s been unique, or all that different, or unusual but there have not been any big dislocations to speak of since the crisis. I think a lot of that has to do with the extraordinary actions of policy makers in really suppressing dislocation or suppressing volatility as a matter of policy.

I don’t think that will go on perpetually, or will go on forever. There have always been dislocations in markets and there always will be, but you can have these periods where returns are flat, or slightly down, or not that exciting for many years. This can be the case with any investment.

Investments go in and out of favour. You can have periods – certainly for equities, or bonds, or any other investment – that work performance is flat or down for a long time. If your portfolio is diversified, hopefully, some other things are working for you. There are other pistons in the engine that are firing and are allowing you to keep going and propel yourself on.

Just because there have not been any dislocations in the last few years isn’t, in my opinion, a reason not to be allocated to something like trend following. When facing a particularly disappointing multi-year growth period this wouldn’t be a good reason to no longer be allocating to equities, or not adding equities to the portfolio if the shoe is on the other foot and starting were different.

Niels

Sure. Absolutely.

You tend to wish that something or someone could make trend following great again, but I’m not sure what that might be right now.

Now, Carrie, throughout this long process of building your teams and the mandates, what are the issues and risks that you are concerned about that may still be a bit unclear how to handle? Is that something that you continue to research, and maybe you have some ideas that you can share?

Carrie

With macro and trend we looked at a lot of historical data. But, because these strategies can behave very differently going forward, of course, there’s no guarantee as far as how they will mitigate some risk going forward. So, there’s always that uncertainty, but we find confidence in the investment process with these managers and their experience. It just takes a lot of due diligence to get to that point. With systematic risk premia, there’s been a ton of academic research to support these types of strategies and their value, but they haven’t been tested through a downturn and that remains to be seen.

Niels

What about you, Ryan? Just maybe briefly, is there anything that keeps you up at night when it comes to this part of your portfolio?

Ryan

No, not really, that’s one of the really nice things about being diversified along a lot of different dimensions. It’s OK if a few line items in our portfolio don’t do much or don’t do anything because some other things should be working.

There’s a nice “sleep well at night” fact with trend following just because of how the strategy works to the extent that if the position and the managers, at a point in time, aren’t working, the strategy is self correcting from the perspective that those positions will turn over and change, and losses will be stemmed that way. It’s, of course, possible that you could have a really long period where a manager is chopped in and out of positions. I think that’s kind of been the case for a lot of strategies over this past year or past 18 months.

Look, we look to the very long empirical histories that strategies have. There’s a lot of comfort in the economic intuition behind trend following, that suggests human nature will lead investors and market participants to over and under react to new information. It’s evolutionary in our DNA. I don’t think that’s going to change anytime soon.

So, keeping the perspective that you’re a long-term investor and sometimes that means your horizon is maybe longer than you’d like it to be, in terms of years, but it makes sense to you to hold something like this in a diversified portfolio because it does well in a certain kind of market environment and balances the risks of other things that don’t do well in those kinds of environments.

It’s important to try and mitigate and to try and minimize your exposure to blind spots, as it were. I think that trend following covers a blind spot, or blind spots, that are not easy to cover for a lot of other investments that investors hold. Yeah, from that perspective it’s a really powerful tool in the toolbox.

Niels

Absolutely.

Now, as we start to wrap up our conversation I wanted to ask all of you a slightly related question. Working with managed futures and trend following within your portfolios and doing all your research and implementation of them, what has been the most surprising finding that you’ve come across during this process? This could be a good surprise, or it could be a bad surprise.

Ela, what have you come across in your world?

Ela

Since I took over the portfolio I wanted to understand exactly how it was lined up. One of the things that I found out was that… Now, we’ve sort of evolved it and realigned the portfolio, if you will, but I realized that there was a lot of conditional correlation. So, in certain market environments just the strategy started behaving exactly the same, which did not really show up in the regular correlation.

So if you looked at certain timeframes, if you bulked the type of strategies (this is not just for managed futures but for other types of strategies as well) you lost that whole diversification that you think you’d be getting from it. I think it goes back to the strategic asset allocation, which should be taking over in periods of distress, and the manager selection becomes a minor point.

So really, under normal circumstances, manager selection is as important as your strategic mix. But, when there’s financial distress the strategic mix actually drives your portfolio performance. So it depends on how much you put into equities, how much you have in bonds, or how much you have in privates, and hedge funds, and specifically (let’s say) in trend following, is going to dictate, rather than which managers you selected.

I always knew that, but looking at the portfolio, looking at how some of the strategies did behave similarly was quite interesting and realized that the space had become… There were too many managers and everybody was looking very like each other, especially under similar circumstances. So, we did take some steps and we reduced the number of managers and the line-up. We put higher conviction exposures into the portfolio. It was interesting to see it in action.

Niels

Yeah. Absolutely.

What about you, Carrie, what surprises have you had in your research and in your work?

Carrie

Along those lines of correlation, and as mentioned before, when looking at trend following it seems like there’s a very standard definition of what it is, but when you look under the hood it is very complex and a lot of these small decisions can result in very drastically different outcomes. In addition to the complexity of the investment process something that hasn’t been mentioned yet, was how important the execution is as well on the back of all of that.

I’m also just very excited about the industry and what’s being done on the machine learning front, or what different types of data are being used. It continues to evolve and that’s what’s very interesting about it.

Niels

Yeah, yeah. Absolutely.

Ryan, I’m going to jump to something different as we’re wrapping up this. Maybe you want to chime in on this. I wanted to ask you also, the three of you if there’s anything that you want to bring up with each other. I wanted to certainly give you the opportunity to ask each other a question since we’re here. I don’t know if you have a question to one-another, or I’ll run to my final question, but feel free if there’s anything you want to bring up.

Ryan

Nothing in particular. It’s nice, it sounds like we’re all simpatico on a lot of things philosophically so I guess it’s nice to learn and I think it will be fun to potentially catch up with each of you offline and just learn a little bit more about what you’re doing in the managed futures and risk premia and hedge fund parts of your portfolios.

Niels

Good, good.

Ok, I have one final one, which is related to what you want there, Ryan. I’m sure you get together with other institutions from time to time at industry conferences and I wanted to ask you what is the general openness by pension funds and similar institutions to these strategies? In the press, you hear about some, and even some big ones pulling out of alternatives and others, like yourselves, increasing commitment to this area, what is your impression when you meet and the microphones are turned off, of the overall instructional appetite for this area? Carrie, maybe you want to share some of your…?

Carrie

Sure. The most obvious example of that is CalPERS withdrawing its hedge fund investments. We actually share a number of the same board members and I think we went about approaching hedge funds in a very different way and we emphasized the education of why specific strategies, and building it out slowly, we were able to build comfort with our program. Certainly, when other investors reconsider their hedge fund allocation we’re all looking at each other as peers and want to understand their decision process and whether it applies to our situation or not. I think as we can gather from this conversation, we’re all facing a number of similar challenges on fees and justifying the role of these strategies and the complexity of them.

Niels

Sure. Absolutely.

What about you, Ela? Up north is the attitude towards this space friendly?

Ela

We have a very interesting mix of institutional investors here, as you know. We have Ontario Teachers here, CPPIB, OMERS, they’re the sophisticated, the larger – and all institutional clients I have meant so far are sophisticated, but when I say sophisticated I mean with internal resources, large assets that they can devote to doing more. There is more acceptance on that end versus as you go down the size spectrum.

It’s not to say that smaller institutional clients do not have hedge funds or they’re not along the lines of looking into managed futures, but it’s been a little bit more conservative when you get into the smaller sized institutional investors. That’s been gradually changing as I go to events and I’m speaking at events, and I tend to speak on managed futures more often than on the hedge funds per se. I see that there is more acceptance and there’s more understanding. I think the education is now being more widespread and we’re moving in that direction.

It’s under the guidance and the light of the larger plans who are at the forefront of this effort in Canada. Of course, we’re very connected, not just in Canada, but across the border in North America in general as well as in Europe. Everybody is aware. It’s slowly shifting.

It’s not an overnight process, so there’s a lot more talk on adding alternatives, but right now the interest is more on infrastructure or predominantly on private markets rather than hedge funds. So, I wouldn’t call them the outflow, I think it’s moved towards it but it’s more gradual.

Niels

Sure, Sure. Ryan, just finishing up with you, what’s your impression on your side? Are there enough people who have read your study and become convinced?

Ryan

I think that it has come a long way but maybe still has a long way to go. I think a lot of investors are making small allocations to these strategies or are at least open to learning about them, but certainly much more than was the case 5 or 10 years ago. The experience in 2008 led a lot of investors to look at what the potential of strategies like this is. So, it’s come a long way. I don’t know that there’s much else to add beyond that.

Niels

No, that’s fine.

We talked about a few different key topics in our conversation today, but of course, is there a question that you feel that is really important that I didn’t ask you today that you want to share as we wrap up this conversation?

The silence…

Ryan

I don’t have anything.

Ela

You did a great job.

Carrie

I think so too.

Niels

Good, good, good.

On that positive note let’s wrap up this awesome conversation about managed futures and about how it helped some of the biggest investors in the world mitigate risk in their portfolios.

Carrie, Ela, and Ryan thank you so much for sharing your thoughts and opinions on today’s topic. I really appreciate your openness during our conversation. To our listeners around the world, let me finish by saying I hope you were able to take something from today’s conversation onto your own investment journey. If you did, please share these episodes with your friends and colleagues, and send us a comment to let us know what topics you want us to bring up in the upcoming conversations with industry leaders in managed futures. From me, Niels Kaastrup-Larsen and our exclusive sponsors, CME group, thanks for listening and I look forward to being back with you on the next episode of Top Traders Round Table.

In the meantime go check out the amazing free resources you can find on CMEgroup.com as well as TopTradersRoundTable.com.