Properly Diversifying a $1.6Trillion Portfolio ft. Sebastien Page

[00:00:06] Sebastien 

 

Historically, again, as long as the shock is driven by growth, news, or growth-related volatility, treasuries have been a really good hedge to portfolio risk when everything else starts moving together unless you’re explicitly buying protection or doing momentum as we’ve been talking about. The wonderful thing about treasuries is historically has been that you actually made a return. So you would get paid something better than free insurance.

 

You were getting paid for the insurance. You had a yield. Plus, it provided insurance in the portfolio. Wonderful. But when rates are at zero and shocks are driven by uncertainty around rates and inflation more than growth per se, then they don’t play the same role in the portfolio.

 

[00:01:00] Intro 

 

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.

 

Before we begin today’s conversation, remember to keep two things in mind. All the discussion we will have about investment performance is about the past, and past performance does not guarantee or even infer anything about future performance. Also understand that there’s a significant risk of financial loss with all investment strategies, and you need to request and understand the specific risks from the investment manager about their product before you make investment decisions.

 

Here’s your host, veteran hedge fund manager, Niels Kaastrup-Larsen.

 

[00:01:55] Niels 

 

For me, the best part of my podcasting journey has been the opportunity to speak to a huge range of extraordinary investors from all around the world. In this series, I have invited one of them, namely, Alan Dunne to host a series of in-depth conversations on the topic of what it takes to be a world-class allocator. In today’s world, portfolio construction is fast moving to the top of the agenda of many investors as they tried to analyze and understand the riskiness of their portfolios.

 

With ever-increasing uncertainty around the globe, being well-diversified across many different strategies and themes in your portfolio can mean the difference between ruin and survival when the next crises emerge. The aim of these conversations is to try and understand the experiences that have influenced these highly specialized allocators and the processes they follow to harness the best returns for their clients so that we can all become better-informed investors. With that, please welcome Alan Dunne.

 

[00:02:56] Alan 

 

Thanks very much for that introduction, Niels. Today, I’m joined by Sebastien Page. Sebastien is Head of Global Multi-Asset and Chief Investment Officer at T. Rowe Price in Baltimore. T. Rowe Price manage $1.6 trillion in assets under management and they have about 450 billion in multi-asset portfolios that Sebastien and his team are responsible for. Sebastian is also the author of Beyond Diversification: What Every Investor Needs to Know About Asset Allocation, which I’ve been reading over the last while and I found to be an absolutely excellent insight into the world of asset allocation and multi-asset investing. So Sebastien, it’s a great pleasure to have you on today. Thanks very much for joining me.

 

[00:03:39] Sebastien 

 

Thanks, Alan. Thanks for inviting me. I got to say I’m excited about doing a longer form podcast. I do too many two-minute spots on CNBC and there’s so much pressure. Now we can have a real conversation.

 

[00:03:52] Alan 

 

That’s right. I’m looking forward to delving into a lot of the stuff that you talk about. I do follow your posts on social media. I know you’re quite active on LinkedIn posting as well. So I’d encourage our listeners to check that out as well. But maybe to kick off Sebastien, you obviously have a big role overseeing a lot of portfolios in T. Rowe Price, $450 billion in assets. Can you give us a sense of your journey in the investment management industry, how you got started, how you have progressed to that current position, and also maybe a little bit of context on the type of portfolios you’re running.

 

[00:04:28] Sebastien 

 

You could say finance is part of my DNA. My father was a finance professor for almost 40 years. He’s now retired. So I grew up with conversations around the dinner table of the capital asset pricing model of Modigliani and Miller. Alan, I don’t know if you have a glass of wine with your friends and it’s getting late after dinner if you talk about the capital asset pricing model.

 

[00:04:56] Alan 

 

Generally, that’s not something that gets the conversation going really well, but I’m assuming in a household of finance geeks, it really could get things going.

 

[00:05:05] Sebastien 

 

I grew up around this. My father never pushed me in that direction, but I naturally, as I said, because it’s part of my DNA, just gravitated towards it and now fast forward a few decades later, I’m here at T. Rowe Price, I just published this book. I would say about my job, Alan, I absolutely love it. Running a large global organization, you mentioned $450 billion in assets, we have over 220 different portfolios, we have one of the leading retirement target date franchise in the world and the job involves, of course, investment oversight, which means I have to stay on top of capital markets, I have consume vast amounts of research, but also, I’m running a business.

 

So you need to set a strategic vision, make sure the vision is executed well, get a recruit, develop talent, you get to manage large product development projects. Also, Alan, I’m a member of the firm’s management committee, which is interesting for me because in that role, I help manage our entire firm. You mentioned $1.6 trillion, we have over 7,000 employees, 16 different countries. I would say we have a good brand. But we also have an ethos of keeping our head down and delivering investment performance. We’re an investment-led firm entirely dedicated to active asset management. I don’t know, maybe some people in our audience today don’t know that much about T. Rowe Price.

 

Sometimes, we’re the biggest asset manager they’ve never heard of. So the bottom line, Alan, is my journey has been one that started very early in life and I studied finance, and I’ve worked at State Street and PIMCO before T. Rowe Price. Where I am here is a very broad job. I learn something every day, investment challenges and leadership challenges. So you mentioned, we’re doing a little bit of- we’re trying to grow on social media, so doing some LinkedIn. If you follow me there, I’m actually also posting about psychology and leadership as much as capital markets and investment research.

 

So sorry, Alan, we’re plugging the social media. But at this point, I have 11,000 followers, it’s not that impressive, so we got to get your audience to come in.

 

[00:07:45] Alan 

 

Let’s hope so. Let’s hope we’ll see a bit of a jump up in followers after the podcast gets released. The other aspect of your role which you didn’t really talk about is that you’re a prolific writer and publisher of research papers and that’s something that’s obviously been a big part of your career going back a couple of decades. It looks to me having read your book that the book was a culmination of all of those efforts. Was that really the motivation in bringing the book together?

 

[00:08:15] Sebastien 

 

Yeah, exactly, Alan. I wanted to put it all together. All the research I’ve done over the years, what I’ve learned from my father, what I’ve learned from my colleagues, some of the top investors in the world, and just try to bring the asset allocation as done by the pros, if you’ll allow me, by professionals to financial advisors, and other individuals who might be interested in just stepping up their game in asset allocation. I’m trying to reconcile. I would say as an industry, we still haven’t reconciled, and I think you’re doing this with this podcast series as well.

 

One of the most important questions in investment management, how do you combine numbers-driven quantitative approaches with investor judgment? So Alan, you’ve read the book, I’ve reviewed over 200 academic papers. I try to make them accessible. I don’t put in all the math there. In fact, there’s no math in the book explicitly and I also rely on my experience and the experience of my colleagues. So I’m really excited that the book is out and that you’ve read it. Thank you.

 

[00:09:39] Alan 

 

I have indeed, and I’ve learned a lot reading the book and some of the material that we’re going to get into, it’s just very interesting. It actually was results that I felt I knew intuitively, but you had the academic and the, I suppose, empirical evidence to support some of the things that you might have a sense of in terms of the performance and behavior of markets, particularly in crisis periods and we’ll touch on that.

 

Speaking of crisis periods, I might be a stretch to call the current markets a crisis, but with the NASDAQ down in excess of 12% a month-to-date, we’re recording on the 28th of January, it’s been a pretty rough start to the year in markets and we’ve had the Fed this week. I think coming into the Fed, maybe there was an expectation that we would get some soothing words from Jay Powell and a bit of a rebound. It hasn’t been the case. What are you thinking on markets at this point?

 

[00:10:36] Sebastien 

 

I thought you might ask. I know this podcast series is more about evergreen content and how you improve portfolio construction, and in the case of this particular series, asset allocation, but I guessed you might ask about capital markets, and I was thinking about it. Alan, I don’t know about you, but I don’t like acronyms and our industry, we like to use acronyms. So I thought, okay, how do you describe what’s going on right now and how you think about investing in 2022? Let’s just go wild on the acronyms because they’re actually meaningful even though we love to hate them.

 

So we’ve had ZIRP, which has led to QE, which is then led to FAIT. Ultimately, TINA and now FOMO, and that’s where we ended last year. So should we translate for the listeners? ZIRP is Zero Interest Rate Policy. What happens when you have rates at zero? Well, the central bank needs to do more and more, and it gets hard. You can’t bring rates, nominal rates. It’s been done, I suppose. But you can’t push them too far into negative territory. So you do quantitative easing, QE. Now, when that runs out of ammunition where you feel you’re already doing enough, you have FAIT. You got that one, Alan?

 

[00:12:04] Alan 

 

Flexible average inflation targeting.

 

[00:12:06] Sebastien

 

Right. So the Fed says, well, we’re not going to really target 2% per se, we’re going to target- well, we’re going to target 2% but on average, so flexible average. So we can let the inflation run above 2%, and boy, is it running above. We just printed 7%. Then if we average out to 2% over time, then that’s okay. So all of that central bank stimulus leads to very, very low rates, and essentially, the least attractive expected returns on bonds that we’ve had pretty much in history. So TINA, There Is No Alternative, and investors are pushed into risk assets to get not only a yield, but just the rate of return and especially if you look for rate of return after inflation, it gets difficult.

 

That creates asset bubbles, risk assets go up and up, and they go up and there’s Fear Of Missing Out, FOMO. So you end up in an environment like we are now where investors start to worry about bubbles, there’s real inflation that we haven’t had in a long time, and then we got to deal with rising rates. In 2022, the Fed actually has a dual mandate now. The Fed just can’t press the accelerator and let inflation take care of itself. That’s where we are now.

 

[00:13:37] Alan 

 

It’s interesting. You’ve gone through the various phases that we’ve had over the last number of years really I guess since the financial crisis and that’s really nicely framed where we’re at. The big question I think that is lurking there is are we possibly in the midst of a regime shift? This is something that really will impact not just the economic outlook, but it’s also a key question from an asset allocation perspective and it touches on the big question that every asset allocator is talking about at the moment, the outlook for the 60/40 portfolio.

 

Maybe before we get into that, you touched on two themes in markets at the moment: rising inflation concerns, which have been massively into the fore and the corollary of that, which is the expectation of rising rates. So as you’re looking at ahead this year and you’re positioning your portfolios, are they really heavily impacting how you’re positioning or are you taking more of a longer-term strategic view in terms of the outlook at the moment?

 

[00:14:46] Sebastien 

 

Alan, we do both. We spend a lot of time on strategic asset allocation and also a lot of time and resources on tactical asset allocation, which for us is think about it as a 12-month horizon set of views on capital markets. Rates and inflation are key risks for 2022. We’re also keeping an eye on China. We’re currently underweight stocks relative to bonds, but not by much, so about 2%. If you take a 60/40 portfolio, for example, we’d be at 58% stocks. Again, we have 220 different portfolios that do different things and I think we’ll get into that today.

 

But if you look at our total risk, and you can calculate things like a total equity beta which basically just looks at the cyclicality in the entire portfolios, we’re close to neutral because while we’re underweight stocks, we’re actually long value versus growth. We’re long small versus large. We have a slight overweight to credit and emerging markets as well. So when you put it all together, we bring the portfolio’s risk given that cyclicality back to neutral. If you look at those asset classes, so-called reopening asset classes, they’re quite cheap relative to say their counterparts and it basically tells you that the recovery trade is cheap, value is a historical cheap relative to growth, small, historical cheap relative to large.

 

Credit is expensive. But you have to approach credit with an eye towards default rates which are historically low. So you look at this and you say, well, stocks look expensive, bonds look expensive, and I know a lot of your audience is in the hedge fund business, they’re following the hedge fund business so they’ll understand this concept of relative valuation trading or positioning. From that perspective, you can get exposure to the reopening in a way that is quite cheap. What’s interesting year to date is that value stocks are actually becoming more safety.

 

[00:17:15] Alan 

 

It’s interesting.

 

[00:17:17] Sebastien 

 

Talk about regime shift. This could be a regime shift. We talked about this in our asset allocation committee, redefining safety because the narrative has been that mega-cap growth stocks are where the safety is. But now, we’re sitting here recording this in January 2022, markets are down, value stocks are outperforming growth stocks. So there’s not only a cheap exposure to cyclicality in the reopening, but also redefining safety and that’s why we’re positioned towards value.

 

I work with some really, really good investors at T. Rowe Price, some of the top stock pickers and what they often tell me is position yourself 12 months from now, just think of the world 12 months from now. What will the world look like? Our view is that it will look like a world that has almost entirely reopened, that has materialized, pent up spending in the services space and where cyclicality has done well, this whole repricing towards reopening has occurred within 12 months. So that’s how we’re positioned.

 

[00:18:41] Alan 

 

It sounds like you’re in aggregate a little bit cautious on the equity market, but not massively so, and the big thing you’re looking for is what a lot of people have been talking about this rotation trade as I guess some of the factors which have struggled in the last decade or so such as value now showing signs of coming back to life essentially.

 

[00:19:04] Sebastien 

 

Yeah. We’re going through a COVID repricing cycle. 2022 is going to be a year of normalization from a lot of distortions. So it’s not a normal credit cycle and you have to adjust your investment process to navigate this repricing cycle. Going back to your question on regime shifts, one of the important questions is, are we now in a rising interest rate regime and what does it mean for investing?

 

[00:19:43] Alan 

 

From that perspective, obviously, you are more underweight in stocks and you’re slightly overweight bonds. So clearly, you’d think that that fixed income can hold in reasonably well even in the face of rising rates. Is that because you don’t see an aggressive tightening cycle or are there particular segments of the fixed income markets offering opportunity?

 

[00:20:06] Sebastien 

 

That’s a really good question because the way we’ve structured the fixed income side is also for rising rates.

 

[00:20:12] Alan 

 

Okay.

 

[00:20:12] Sebastien 

 

So we’re short treasuries in our tactical asset allocation in our asset allocation committee. We’re long credit, as I mentioned. We like bank loans as an asset class. If you press me to say, what’s your top idea for 2022, here’s an asset class that generates 4 or 5% yield in a yield starved world that tends to do well when rates rise, that has minus 30% correlation with treasuries, that has made money 22 out of the last 23 years, that has less exposure to the volatile energy sector, and ultimately, the idea is that those bank loan rates reset every three months. So as rates rise, you can actually keep making money.

 

So on the fixed income side, we’re underweight duration, if you want to think about it from a risk factor perspective. But one of the key questions is, are rising rates really bad for stocks? We know they’re bad for bonds to the extent they’re not priced in. The reality is that- I talk about this in the book. The analogy I use is a puzzle. Here’s a puzzle for you, Alan. What do coffee, egg yolks, red wine, and rising rates have in common?

 

[00:21:46] Alan 

 

You’ve got me with that one. You put me on the spot.

 

[00:21:50] Sebastien 

 

You don’t really know if these things are good or bad for you.

 

[00:21:53] Alan 

 

Okay.

 

[00:21:54] Sebastien 

 

I don’t know, I try to watch what I eat. But you try to figure out if egg yolks are good or bad and it changes every year. Same for red wine, same for coffee. I don’t think I’ll ever give up coffee. So rising rates are they really bad for risk assets? That’s the narrative like, oh, let’s all be worried about rising rates, stocks will go down and they can go down for sure. But last 12 months, for example, the two-year yield was at around 10 basis points 12 months ago. It’s gone up by over 100. It’s over 110 today and stocks are down year to date, but they’re still up 10, 15% over that time period. In fact, Alan, we did this study, we went back 30 years, and we found 72 different 12-month periods during which the 10-year rose more than 50 basis points. So 30 years, you got 72 12-month periods, we just used rolling 12 month data during which the 10-year rose more than 50 basis points.

 

The average return for stocks during these rising rates regime, if you will, is 17% and the hit rate is 100%. Stocks make money 72 times out of 72. So why is that? Why are rising rates so bad for you? By the way, they’re bad for bonds and we’re seeing this right now. But why are they so bad for stocks, at least historically, this year could definitely be different, but at least historically. Well, it’s pretty simple. The Fed doesn’t raise rates in a recession and the Fed will raise rates when there’s enough growth to absorb those rising rates.

 

We do have a lot of growth, 6% last year, 4% this year. But the difference this time and again, going back to the theme of potential regime shifts is that for the first time in several decades, the Fed really has to worry about inflation. So we’ll see what happens.

 

[00:24:06] Alan 

 

Interesting. It’s something that there’s been a lot of research about I think from the sell side as well, people pointing to the fact that exactly what you’re saying, hey, what’s the problem? Often, stocks do go up particularly in the early part of a tightening cycle. I guess they’ll always be a suggestion, is this time different or are we seeing a break with the past in terms of either Fed policy or the fact that we haven’t seen the Fed starting to tighten policy with inflation already at 7% in the last number of decades?

 

So there is always that I guess tail risk, maybe not even a tail risk, but certainly a possibility that what we’re witnessing is a different scenario. What’s linked to that is the concern that you could see stocks and bonds going down at the same time and that’s the regime shift that we haven’t seen. Is that something that keeps you awake at night?

 

[00:25:05] Sebastien 

 

Yeah, I like that question because as we sit here in the beginning of 2022, this is what’s happening, stocks and bonds are down together because of inflation and rates shocks, and shocks mean unexpected developments on things that weren’t priced. I wrote a paper with some colleagues back when I was at PIMCO on the stock-bond correlation. I’m a little bit of a correlation nerd. I like to study correlations. We found that as long as the shocks were driven by growth, GDP, the correlation remains negative. But when the shocks are driven by interest rates, or inflation, the correlation can switch to positive territory.

 

By the way, the stock-bond correlation is very hard to model or predict. If you go back 80 years, it shifts sign all the time and then it goes to regimes too, going back to your earlier question where it’s positive for a long period of time or it’s negative like it’s been mostly last decade for a fairly long periods of time. So yes, 2022 could be a year where the stock-bond correlation is positive and then you got to start looking at things like what’s the role of cash tactically in the portfolio. I’m sure we’re going to get this given that your audience here wants the role of alternatives. I’m sure we’re going to get into this and I’m looking forward to talk about it, but also an asset class like bank loans, going back to my earlier comment.

 

[00:27:04] Alan 

 

Very good. We’ve focused I guess a lot on tactical questions so far. It’s all been about the possibility of rising rates, etc., and how to position in this environment. But you did touch on the fact that you have two different processes or two components to your overall process, strategic asset allocation and tactical asset allocation. a lot of the research that you’ve done, obviously, is from the perspective of constructing highly diversified and sensible portfolios from a strategic asset allocation perspective. So maybe to give us a sense on the approach to strategic asset allocation, within that, is there an emphasis on portfolio optimization? How do you think about that process.

 

[00:27:52] Sebastien 

 

So for strategic asset allocation, we take a much longer time horizon. A lot of our clients, investors are saving for retirement. So here we are talking about the life cycle problem, which is very hard to optimize, very hard to solve. I think Bill Sharpe described it as the thorniest hardest problem in all of finance because you have multiple time periods, you have an accumulation phase, you have a decumulation phase. Ultimately, you have taxes. It’s a complicated problem.

 

We take a long time horizon, we build capital markets assumptions, we use Monte Carlo simulations and then we look at different asset mixes and we try to maximize utility, which is basically a way to rank the portfolios in terms of how well do the portfolios do in reaching your ultimate goal, which is, for example, spend through retirement in a way that’s risk conscious, however you define the so called utility function. But this is an optimization problem. It’s a multi-period optimization problem. There’s a lot of academic research around it.

 

I even have, Alan, in the book, you’ll recall, an example where I actually show people how to maximize utility with a very simple toy example that you can do in Excel in about five minutes so you can get the idea. But we run these optimizations and we solve not only for the asset mix in the components, but also for the glide path, how your risk profile should change as you approach retirement. Now, you asked a question about the use of portfolio optimization. It’s really useful for strategic asset allocation. We use it. We’re not dogmatic about which model we use.

 

We like to use multiple models, multiple assumptions and look for “robustness”, answers that seem to work across different types of models and assumptions. But it’s surprisingly controversial in academic and quant circles whether you should use optimization, whether one optimization model is better than the other, whether you should do risk parity, whether you should do Black Litterman, whether you should take into account tail risk exposures. There’s just a lot of controversy. If you go to a quant conference, it’s almost as bad as bringing up religion or politics. I have a story about this in the book. I was sitting at a quant conference and I was pretty half asleep as you tend to be at quant conferences. I hope I’m not insulting anyone.

 

[00:31:10] Alan 

 

Our quant listeners won’t be offended, I think.

 

[00:31:12] Sebastien 

 

Sometimes they get exciting though. The presenter was Bernd Scherer. He’s a well-known professional. He’s actually straddled both academia and practice. He’s worked in investment management and he’s worked in academia. He was presenting some portfolio optimization models, and there was someone who asked a question who was clearly a fundamental investor, who thought this was all nonsense. I don’t know what they were doing at the quant conference, but it made for an interesting interaction.

 

The person raised their hand and said, all of this optimization is useless because you can’t really estimate the inputs. You don’t know what the expected returns are, you don’t know what the volatilities, the correlations are. He brought up, and this is a good way to insult a quant, the GIGO critique, garbage in, garbage out. So we should just do away with optimization. It’s just fraught with error and it doesn’t tell you anything was the argument. I’ll always remember what Bernd Scherer said because it was the best rebuttal I’d heard and it stayed with me through the years.

 

He looked at the person asking the question, and he just said, “If you don’t think he can estimate expected returns, you shouldn’t be in the investment business.” The point was investing is making a forecast on the future and what the optimization tools do is they help you build scale in how you process all the relevant data. But ultimately, there’s judgment in the process as well. So I’m being a little bit philosophical here, but that’s how we approach the use of these tools.

 

It’s not like you come to me and you say, how do you do strategic asset allocation, and I give you a formula with some data. Sometimes, our clients would love that. We could just empower them to do it. It doesn’t really work that way in practice.

 

[00:33:25] Alan 

 

Well, I think the optimization that people would probably be most familiar with would be mean variance optimization. I think in your book, you also have an interesting anecdote about an exchange between Paul Samuelson and Harry Markowitz, where Samuelson points out that Markowitz was actually wrong with his portfolio optimization and mean variance approach, which I guess highlights how this is a very debated point in academic circles.

 

You might touch briefly on the challenge with variance optimization and then secondly, you say there’s this preference for utility maximization and you do give a good example of how to plug that into Excel, and I’ve done that myself following your advice in the book. But it’s a very difficult topic or concept to define mathematically, utility. If you were to speak to an investor, they’d be able to describe what types of scenarios.

 

Obviously, consistent returns are favorable, deep drawdowns are unfavorable, long drawdowns are unfavorable. Getting all of that into a formula that you can maximize seems like a stretch. Utility maximization, while it may be intellectually interesting, is it something that can be of practical use to multi-asset investors?

 

[00:34:54] Sebastien 

 

Yeah, you’re bringing up two really important questions around strategic asset allocation. The first one has to do with fat tails and I think we’ll talk about this- I hope we’ll talk about this a little bit more. The second one has to do with just in practice, how do you define preferences? If you ask how do you define utility, basically, what you’re asking is, what are you maximizing? What are you trying to achieve? What’s the figure of merit to use the term that Markowitz used? By the way, this interaction between Paul Samuelson and Harry Markowitz was quite interesting because it was about fat tails.

 

Samuelson was arguing that you should optimize directly on the probabilities of fat tails. Markowitz was saying that, no, because mean variance is a good enough approximation, and we ended up getting involved. Interestingly enough for your audience, we have a paper out there about portfolio optimization for hedge funds because that’s where you have the most non-normal return distributions, and it is about utility. Now, on your question of defining the utility function, it has to be iterative. You have to take a shot at it or run it, and then wash, rinse, repeat and see what you get.

 

If you’re a financial advisor or if you’re an asset allocator who deals with a board of trustees, for example, then you have to run iterations to show people what they get in terms of risk, return, exposure to loss, long-term wealth accumulation, and then try to back out their preferences for these different trade-offs by tweaking your utilities function in the sausage factory in the background. For financial advisors, just ask someone, how much do you care- people use a lot of surveys. So how much do you care? Are you prepared to lose 10% in a year in order to increase your expected return from 6 to 7.5%?

 

Those types of survey questions can help you tease out so-called utility functions. Advisors use them, and now you have something we’re working on as well, robo advisors. You can go and you can plug in the information and you will get a portfolio out of it. But just the most basic one is, what is your risk tolerance? Because based on that, that’ll change your strategic asset allocation, that’ll change how you optimize the portfolio for return against risk. I don’t know, Alan, what’s your risk tolerance in a mean variance? Is it 3, 3.4? What does it mean?

 

[00:37:48] Alan 

 

Yeah, it’s a great question. I guess your point is that the investment industry and the advisory world is effectively incorporating utility into their process by asking these questions like, how would you react to a 20% drop in equities or would you be more inclined to double up or sell out? So that’s effectively trying to get some greater insight into the preferences. If it was the case that somebody was risk-neutral in the academic sense or wasn’t subjected to things like loss aversion, what then? Is traditional portfolio optimization sufficient? Is there enough information in just looking at volatility and return? I guess what we’re getting to is what we’ve been dancing around is the point that the downside of the mean variance doesn’t take account of the higher moments of the distribution. So can you explain all of that to us?

 

[00:38:46] Sebastien 

 

I don’t think it’s sufficient to just look at mean and variance. Now, if someone says I want the highest Sharpe ratio, which is basically the return divided by volatility, you can adjust for the cash return, that’s a utility statement. That’s your figure of merit. Find me the portfolio that will give me the highest Sharpe ratio over time. It’s a simple one and it relates to mean variance optimization, but I think it’s a misleading one because it leads you to load up on negatively skewed assets and risk premia. I can manufacture low volatility by selling insurance, but my exposure to loss is quite high, either implicit or explicit in the data, sometimes you can even see it in the data.

 

So my short answer to your question is it’s not enough and you do need to model exposure to loss directly because volatility just only goes part of the way at estimating this exposure to loss. But again, you also have to go back to what’s risk tolerance. My longtime mentor Mark Kritzman. He teaches financial engineering at the MIT. He’s the CEO of Windham Capital Management. He’s been a great friend and mentor for a few decades. He had this answer once that I always remember when- I think I talk about this in the book. A client would ask, “How do you determine risk tolerance?”

 

Mark had a bit of a dry sense of humor, but this this is actually true. He said, well, there’s research at MIT where we looked at- he wasn’t involved directly, but looked at hormones in your bloodstream, how much testosterone do you have, how much of, I don’t know, the markers are, and that would tell you how tolerant you are to risk and then he would say- look at the client and say, generally, our clients tend to think a blood test is a bit too invasive, so we look at other ways. So it’s an abstract concept, but it really matters.

 

So what it boils down to when you think about portfolio construction, strategic asset allocation, is that yes, volatility in the traditional sense matters. Yes, you can and should estimate expected returns if you call yourself an investor, and in fact, if you invest, you have no choice but to make a call on expected returns. But yes, you do need to look specifically at exposure to loss. You can do that by scenario analysis or other techniques. Then you need to make a call as far as how much risk you’re willing to take. I’m sure in your audience, we have hedge fund traders, they make that call almost every day, right?

 

[00:41:49] Alan 

 

Absolutely. I think in your book, you had a comment there that stood out. It says there is a need in our industry to move scenario analysis from the back office to the front office and it’s this idea that you need to be cognizant of the different scenarios and I guess, of tail risk. I guess that’s part of the downside of mean variance optimization is that it doesn’t account for the possibility of fatter tails. Isn’t that right?

 

[00:42:24] Sebastien 

 

Yeah. Clients will ask, do you use scenario analysis, and asset allocators and investment managers will produce the numbers and show the scenario analysis. But if you dig into it and you say, how is the scenario analysis influencing your asset allocation decision? Then that’s a more difficult question for a lot of asset allocators. They might actually just revert back to Sharpe ratios or other simpler measures of merit. So that’s what I meant when I said bring scenario analysis to the front office. Fat tails, they really matter.

 

Those large losses that aren’t predicted by volatility models, they really matter and if you just look at Sharpe ratios, it can be pretty misleading. Someone once told me that the only people capable of generating a Sharpe ratio of 3.0 were Bernie Madoff, and we know how that was done, or quants running backtests.

 

[00:43:36] Alan 

 

I feel like we’re giving quants a hard time, Sebastien. I think we might have to delve into that one a little more. More obviously, that’s the downside of overfitting your data to or your to the data.

 

[00:43:51] Sebastien

 

I have to say, I’m just saying this tongue in cheek because that’s my background actually. I think most quants nowadays realize that the more backtests you run, the more you risk overfitting the data and I think the industry has made a lot of progress that way, but it really matters. There’s an example from Andrew Lowe. I saw that you had him as a guest in your podcast series. He published papers back in 2001 about hedge fund risk management. It’s a really interesting case study of how Sharpe ratios can be misleading. I think most people know this, by the way. I don’t think we’re saying anything new here. But here’s an interesting way of thinking about it.

 

He looked at monthly data from ‘92 to December ‘99 and he created this investment strategy this is all in the paper that required no investment skill, zero investment skill, no analysis, no foresight, no judgment. A monkey could do it type of investment strategy. But despite the simplicity, the strategy doubled the Sharpe ratio of the S&P 500 over that time period. It went from 0.98, almost doubled to 1.94 and it only had six negative months compared to 36 for the S&P 500.

 

So Andrew Lowe in the paper, again, tongue in cheek says, by all accounts, this is an enormously successful hedge fund with a track record that would be the envy of most managers. Then he reveals the strategy that requires no investment skill and it relates, of course, directly to fat tails. You know what it was, Alan?

 

[00:45:43] Alan 

 

I know what it was because I’ve read the paper. Obviously, it’s selling out of the money options, so picking a premium. The period that we’re talking about is the 1990s when we didn’t have a significant drawdown in the S&P 500 and this looked like an absolutely fantastic strategy.

 

[00:46:02] Sebastien 

 

Exactly. The strategy was to load up on tail risks. Pick your analogy, pick up pennies in front of a steamroller and so on, and you can really generate a nice Sharpe ratio that way even if there’s some tail exposure in the data. In this case, there was very little of it, because to your point, it was the ‘90s but anyone looking at the investment process.  If you ask me as an allocator how I look at alternatives in hedge funds, I’m going to look for embedded short optionality strategies. Is this strategy selling insurance and whether or not it’s in the data. I need to get my head around this implicit or explicit tail risk exposure.

 

There are ways to get at that. Risk premia, for example. Many of them are short optionality. Carry credit is short optionality. The Merton model says if you hold a corporate bond, you’re selling an equity put. Volatility-type risk premia, liquidity premium, they’re all short an option. It’s not necessarily a bad thing because you get paid for it. But it’s a very different probability distribution. As an allocator, it matters, especially if you’re an allocator who does the strategic asset allocation in a way that’s aware of exposure to loss beyond just looking and comparing Sharpe ratios of the different options.

 

[00:47:41] Alan 

 

Yeah, I’m glad you brought this topic up because so many of our listeners and participants are in the managed futures industry and heavily rely on trend following. When we’re speaking to clients, this is an important part of our pitch about the benefit of trend following strategies vis-à-vis many other hedge fund strategies, and it’s exactly what you’re talking about. Lots of hedge fund strategies might have a better Sharpe ratio in certain periods but they effectively have a short volatility characteristic, either directly or have implied the same characteristic. So the point that you’re making is that you can’t just look at Sharpe ratios, you have to look at the higher moments of the distribution and the tails.

 

Also, within managed futures and trend following what you tend to see is a convex profile, which is that in those more extreme moves, you actually get even stronger performance on occasions because you get stronger trends and in markets. You touched on something interesting there and in the book as well, which I found really interesting. There was a reference to back in the financial crisis, I remember the quote in some shape or form of the executive from Goldman. I think it was in overseeing 25 standard deviation moves a number of days in a row, and it was somewhat tongue in cheek, but the whole point was that we were seeing moves in markets that were way outside expectations if you use some normal distribution.

 

What you show very interestingly in the book is these moves might actually not be that unusual after all, and that certain assets, or possibly all assets are better described by thinking in terms of distributions in two different regimes. So you have a distribution for a normal, maybe a risk-on regime, and you have a distribution for a risk-off regime and they can both be normal. It’s when you combine those two different regimes that you get this non-normal characteristic, which I thought was explained really well. Is that something that you then embed in your portfolio construction process?

 

[00:49:59] Sebastien 

 

First of all, Alan, you’re the ideal reader because you really get it. It’s kind of a puzzle if you think about it, right? Ask someone a normal distribution plus a normal distribution, can it equal a non-normal distribution? If you have two regimes, risk-on and risk-off, then when you blend those regimes together over time and you try to model it as one regime, you end up with really substantial fat tails. So this modeling is really important. There’s so many ways to incorporate it into portfolio optimization, into risk modeling. You need to account for the probability that you switch between one regime or the other and you can model your risk that way, you can assign subjective probabilities between regimes.

 

By the way, we’ve had to extend this as asset allocators over the last two years because it’s not only around risk-on versus risk-off regimes, which have different correlation properties, different average return properties. It’s also been about COVID-on versus COVID-off regimes. So now you have almost a two-by-two matrix of regimes. The COVID-on regime has interesting characteristics that are different from the COVID-off regime. To boil it down to one example, large-cap growth stocks do much better in stay-at-home COVID-on. Value stocks, reopening companies, do much better in COVID-off, reopening.

 

So different ways to think about regimes, but they do matter and they’re a persistent feature of financial markets. Not only are they a neat way of explaining and modeling tail risk, but they also lead to correlation asymmetries. Everybody talks about this. to circle back to your point about managed futures and that they tend to be more defensive when everything goes haywire and when you get a market crash, generally, across asset classes, we all know you get much higher correlations when markets sell off. What people don’t necessarily pay as much attention to is that you get much, much- diversification works really, really well when you don’t want it, when markets rally.

 

Just to give you an example, Alan, we published a paper titled When Diversification Fails in the Financial Analysts Journal, and the correlation between US stocks and non-US stocks, we had almost 40 years of monthly data here, was 87% when US stock returns were selling off in the bottom 1-5% of the monthly returns. So 87%, no international diversification. But what was it when markets were rallying top 1-5% of monthly returns going back 40 years? Oh, great diversification, minus 17. So you have this asymmetry and if you just use the average correlation, it’s pretty meaningless. In fact, it’s quite misleading.

 

I like to use the analogy of you have your head in the freezer, your feet in the oven, and you can claim that your average body temperature is perfectly fine, that you’re comfortable. That’s what you do when you use average correlations in risk modeling. Just one more thing on this, to your point about defensive strategies, strategies that sell on the downside, momentum-based strategies, managed volatility type strategies, they’re interesting in that context because you get better correlation asymmetries. By the way, all other asset classes, almost all of them except for treasuries have this undesirable asymmetry. But let me just look here, I have this right here.

 

Cross-asset momentum is fairly correlated with US equities on the upside, big rallies in US equities. You get almost 80%. So you participate in the upside with cross asset momentum. On the downside, it’s actually lower and it’s pretty uncorrelated, below 20%. So you reverse that asymmetry. Here’s the sales pitch for hedge fund managers that do momentum.

 

[00:55:06] Alan 

 

You’ll be in demand from CTAs to join their marketing teams. This is exactly part of the positioning of CTA is this possibility for a positive correlation in an equity bull market and negative correlation in an equity bear market. I think it’s interesting because the story is even more dramatic when you look at some other asset classes. Take the example of credit, and I guess credit-focused hedge funds will be the same. So, one, on the upside, the correlation isn’t that positive, then on the negative side, it becomes more positive, which is not what you want. The second thing is in those periods when equities are going down, volatility will tend to go up. So their equity beta is actually becoming bigger and it’s having a more detrimental effect on your portfolio if you’re allocating to a strategy as a diversifier.

 

You make the point in your book that actually, you can have what is a seemingly diversified portfolio, but you may have greater exposure to loss in that portfolio because of this characteristic of correlations going to 1, volatility picking up. So it really raises the point about it’s not just about pulling a bunch of assets together and saying, hey, we have a diversified portfolio. Look, we’ve got US equities, foreign equities, credit, emerging market debt, we’ve got some credit strategy hedge funds. You’re actually leveraging up the same risk factor with that kind of diversification. So as an allocator, you really have to understand the nature of diversification.

 

[00:56:52] Sebastien 

 

Yeah, I have to give you an A plus on following the book, Alan. You hit the nail on the head. An example I mentioned is from Marty Leibowitz where around the opioid crisis, a 2-asset class portfolio had a much smaller loss than a 1O-asset class portfolio because of that effect. The non-linearity in credit is key. I don’t hate credit, I like credit. But it is a short optionality position especially when you get into the turbulent market regime to circle back to the prior discussion on risk regimes. So that’s a feature of capital markets. If you’re an asset allocator, it’s really important. If you’re evaluating hedge fund strategies, it’s probably the most important risk question you need to answer. So how do you do it?

 

Well, maybe you just like- here’s an easy way. Ask the hedge fund manager to give you a bunch of historical monthly returns and just run a regression on factors, run a regression on carry credit, volatility, you can even model the liquidity premia. If the hedge fund loads up on all these short optionality strategies, then you can model it as having a similar exposure to loss. So that’s a relatively easy way to assess this exposure in a hedge fund. Liquidity is its own animal in terms of risk premium.

 

[00:58:53] Alan 

 

I think we’re making a pretty good case indirectly here for time series momentum, and I think you do touch on that a little bit in your book as a diversifying strategy vis-à-vis other strategies. You do say I think duration is the only real source of diversification. I think you have a quote like that, but even that is open to debate, I guess. We started off talking about rising rates and you talked about the COVID crisis, and there was a moment in March 2020 when the Treasury market pretty much froze over. Thinking about tail risk, thinking about fat tails, how robust is treasuries? I know maybe the answer is the fact that you’re short treasuries at the moment or maybe not. But is duration still a durable source of diversification do you think going forward?

 

[00:59:55] Sebastien 

 

That’s why people fundamentally are questioning the 60/40 asset mix as the basic model that if you have no view is a good default. It’s really about the 40, the bonds, and ultimately, the role of treasuries in portfolio construction. Historically, again, as long as the shock is driven by growth, news, or growth-related volatility, treasuries have been a really good hedge to portfolio risk when everything else starts moving together, unless you’re explicitly buying protection or doing momentum as we’ve been talking about. The wonderful thing about treasuries historically has been that you actually made a return. So you will get paid something better than free insurance.

 

You were getting paid for the insurance. You had to yield plus it provided insurance in the portfolio, wonderful. But when rates are at zero and shocks are driven by uncertainty around rates and inflation more than growth per se, then they don’t play the same role in the portfolio. I have a paper that we just finished. It’s going to be published in the Journal of Derivatives with my colleagues Stefan Hubrich and Bob Harlow titled Tail Risk Hedging in A Low Rate Environment. The question is, if treasuries don’t give you the same free or even better than free protection in your portfolio, if we are indeed in somewhat of a regime shift towards rising rates and rates uncertainty, is it time to revisit the cost benefit of directly buying protection for the portfolio? We do this evaluation.

 

If you just think about it that way, you might think, well, we should all just buy equity puts and keep a high equity exposure and that should be better than de-risking towards treasuries. It’s actually not that simple and it’s not a free lunch. Protection, if you buy it explicitly and you do nothing to manage the cost, it remains quite expensive. You can always go to cash too to de-risk your portfolio and reduce your exposure. It’s not as convex an exposure. So we show all the trade-offs. The trade-offs change. They tilt in favor of buying protection.

 

But what we do in what we call our risk-managed equities strategy is yes, we buy some protection but we also add a managed volatility overlay as well as other ways to mitigate tail risk. So it’s a diversified approach to managing downside risk that’s more efficient and lower costs than just buying puts.

 

[01:03:01] Alan 

 

Interesting.

 

[01:03:01] Sebastien 

 

Although buying puts can be part of it if you’re really concerned about extreme gap risk.

 

[01:03:06] Alan

 

It’s interesting how this is becoming more of a common approach, I think, to managing downside risk in equities. The idea of managed volatility is something that CTAs have always embedded in their portfolio management and effectively reducing position sizes as volatility picks up. So that combination of managed volatility and the potential for using time series momentum are two interesting and possibly more cost-effective ways of managing downside risk. Is that fair to say?

 

[01:03:43] Sebastien 

 

Yeah, but not a free lunch. Sometimes, these things just sound too good to be true. But when markets bounce back quickly and you’re all de-risked, you don’t participate. So it has to be actively managed ultimately and it has to be part of the portfolio in a way that makes sense in terms of exposure relative to what else you have in the portfolio as an asset allocator and what the trade-offs are. That’s the point of our Journal of Derivatives paper, which is going to come out in a few months, is have the trade-offs changed and what are they exactly between just going to cash or buying treasuries?

 

I think liquidity risk is the ultimate risk here that you need to manage and we’ve seen this during COVID. Alan, you mentioned that treasuries during COVID and those weird few days were- it’s been underreported I think how strange it was that the most liquid, safe refuge asset class in the world faced a bit of a crunch and markets were falling like a rock and rates were going up. The Fed had to step in because the plumbing was bursting and there were issues in the guts of the financial system. It’s a bit nerve-wracking in hindsight.

 

[01:05:18] Alan 

 

We will keep an eye out for that upcoming paper. Before we finish up, there was a few topics I wanted to touch on from more of a non-investment perspective. You did say at the outset that you’ve been writing more and thinking more about leadership and I guess the softer side of the skills required of working in the investment industry. I saw you had a LinkedIn post touching on lessons from the stoics recently, which was very interesting too. Obviously, you’re an investor, but you mentioned that you’re a part of the leadership team. So you are managing high performance investment teams in T. Rowe Price. What have you learned from that experience and any insights to pass on from your reading in this regard?

 

[01:06:07] Sebastien 

 

I think it’s really important for our industry, this question of how you manage high-performing teams. For years in our industry, we’ve had this issue where top individual contributors because they’re really, really good are put in people management positions. I mentioned at T. Rowe Price, we are an investment led organization. My boss is CEO of T. Rowe Price, Rob Sharps, was an investor for a long time and that’s part of our DNA. But the leadership side sometimes across our industry is a second thought and it shouldn’t be in my mind and that’s why I’m reading more about it.

 

It helps me in my job of managing a high performing highly opinionated, strong-minded team. I’m learning a lot just digging into psychology. I think leaders have a lot more to learn about psychology than you might think. We think of psychology in terms of pathologies and issues like depression and anxiety, but there’s this whole world of academic psychology that goes into the other direction, positive psychology, and how do you define fulfillment and this concept of engagement and flow.

 

There’s psychology around personality psychology, all sorts of interesting research that I think if you put it all together, and that’s what I’m trying to do in those posts on my LinkedIn, leaders in investment management can really benefit from.

 

[01:07:58] Alan 

 

Interesting. You touched on the idea of flow which is an interesting one. I think it’s the guy who we can never pronounce his name. I think it’s [inaudible]. The example in his book I think he gives is the guy who could be the like- I think it’s possibly even the options trader, who’s so engrossed in what he’s doing that time just passes. I guess that sounds like a typical day for yourself. You’re saying you’re consuming a lot of research, you’re consumed with the markets. Do you think you have to actively try and put an environment in place that people can experience that and have you thoughts on that? What is a good working environment then that people can feel that they’re in the flow?

 

[01:08:51] Sebastien 

 

It’s a wonderful state. It’s not just positive emotion. It’s not just feeling good. It’s just feeling engaged and using all your skills on tasks that are hard enough but not too hard. Think of a tennis player in the middle of a match, think of a musician. They’re all in a state of flow. So what does it mean for you as a leader? Well, I think you want to set goals, you want to develop metrics where people get rapid feedback that are all conducive to flow. It’s not that certain jobs are- well, it is to a certain extent that certain jobs are- in certain jobs, it’s easier to get into a state of flows. But I think in all jobs, you can get into a state of flow. I can mow the lawn and get into a state of flow.

 

So how do you- conversations. I think the conversation we just had, Alan. I haven’t thought about anything else for the last hour besides our conversation. So we had a wonderful, as far as I’m concerned, flow experience. So we learned that from psychology and if you’re a leader and you create an environment that encourages people to be engaged, and there’s just many ways to do it But goal setting is really important. How do you calibrate the goal? To be hard enough? To recruit people skills? But if it’s too hard, people get disengaged. If it’s too easy, they get bored out of their mind. Yeah, how you set goals and then how you pick the metrics and keep track and give feedback.

 

[01:10:40] Alan 

 

Good stuff. Last topic before we let you go. You touched at the outset on the fact that your dad is a professor in finance. Obviously, that must have been an inspiration for you and you must have learned a lot and you seem intent on passing on the knowledge. I noticed on LinkedIn, you’re doing some posts where you’re explaining very technical topics to your daughter, Olivia, I think. There’s this idea of the Feynman technique, that if you want to know that you understand something, you write it down on a piece of paper and explain it to a 12-year-old and this is exactly what you’re doing with the benefit of cameras and LinkedIn. How have you found that process? What have you learned?

 

[01:11:19] Sebastien 

 

It’s been a lot of fun. It’s surprisingly gone a bit viral by LinkedIn standards. I had this idea because I wrote my book to be accessible but I failed to make it fully accessible to someone without a finance background. So if you go on Amazon, I have two or three reviews, like one star, this is too complicated, and this is from probably readers who just don’t have the investment background to begin with. So it hurt me a little bit to read this and I thought, how can I just get better at explaining things? In the middle of COVID I got nowhere to go. So my daughter was there in the room and I thought, okay, I had this light bulb.

 

I said I’m going to explain to her things like time value of money, the magic of compound interests. So I sat her down, we set up a couple of cameras, recorded, and it wasn’t rehearsed. What was really interesting in one of the clips is her genuine reaction to what happens if while you work, you save $140 a week, which is not like- for a lot of people, it’s a fairly substantial sum.  I think if you earn $60,000-65,000 a year, it’s probably doable. Well, you can retire with a 5% return. You can quibble over the 5%. You can retire with a million dollars. Just showing her the numbers in her face, realizing why compound interest is- actually, stay invested, stay diversified, as I often say.

 

So it’s been a lot of fun. I think there’s a senator from Colorado who reposted it and said every kid needs to learn from it. T. Rowe Price has an initiative called Money Confident Kids where we promote financial education to teens and preteens and we’re going to record a few more of those and see how it goes.

 

[01:13:31] Alan 

 

Well, great, I think it’s a really interesting, a lot of fun by the looks of it as well, and a really good thing to do to get young kids interested and up to speed on these topics. So well done. But thank you very much for coming on the show. It’s been absolutely fascinating, fantastic, lots of great insights. I think we’ve referenced a huge number of papers as well. So hopefully, we can get those in the notes and people can check those out. As I say, you are reasonably active on LinkedIn so people can find you there as well and I guess check out T. Rowe Price for more information. I’d just like to say thank you very much, Sebastian. It’s been a pleasure having you on. With that, I’ll pass back to Niels.

 

[01:14:18] Niels 

 

Thank you so much, Alan and Sebastien, for a great and fascinating conversation. I really enjoyed learning about Sebastien’s thoughts on what it takes to manage $1.6 trillion, not only from a pure investment point of view, but also the psychology and the leadership skills that it involves. It was also fascinating to hear about their thoughts on how to redefine safety when it comes to assets, how shocks can impact the stock-bond correlation and the seductive nature of the Sharpe ratio and how to double it in a whim. Of course, one of my favorite parts as a father myself was the work Sebastian is doing with his 10-year-old daughter Olivia.

 

Make sure you go and follow Sebastien’s and Alan’s work because as you can tell from today’s conversation, there are many exciting facets to building a well-diversified portfolio, and we really look forward to exploring many more of them as our series continue. From Alan and me, thanks so much for listening, and we look forward to being back with you on the next episode. In the meantime, take care of yourself and take care of each other.

 

[01:15:32] Ending

 

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released.

 

We have some amazing guests lined up for you and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.