Episode #9: Pete Mladina, Northern Trust, “What You Thought Was Skill Was Just Risk Premia”

Episode #9: Pete Mladina, Northern Trust, “What You Thought Was Skill Was Just Risk Premia”








Guest: Pete Mladina is the Director of Portfolio Research for Wealth Management at Northern Trust. He is responsible for the application of leading research to the Wealth Management investment process. In addition, Pete has written many academic papers, and is a guest lecturer on financial markets and investments at the University of California, Los Angeles.

Date: 7/18/16     |     Run-Time: 1:07:26

Summary: “Do I have enough to fund my retirement?” “What’s the optimal lifetime asset allocation?” Those two questions, stemming from a recent academic paper written by Pete, help launch Episode 9. The answers point toward Pete’s solution for retirement challenges, something called “goals-based” asset allocations (as opposed a singular, static “all-in,” asset allocation applied to your entire capital base). In other words, your specific goal – say, college tuition, a second home, maybe a trust – dictates the asset allocation of the associated, earmarked funds. From there, Meb and Pete transition to a discussion on factor-based investing, starting with “term” and “market” factors. According to Pete, “Ninety-five, ninety-six percent of the return variation of all managers and funds in the Morningstar database are explained by…basic factors.” Meb then asks, “What are the best diversifiers to a traditional portfolio?” Hint: Pete’s response includes Meb’s “desert island” strategy. They then discuss whether individual smart beta factors such as “value” should be evaluated relative to their own historical valuation. Your own answer will likely reflect whether or not you believe markets are mean-reverting, a topic often debated. They then touch upon risk factors as applied to REITs before diving into a discussion of the Yale Endowment allocation. Pete tells us that Yale’s outperformance over the decades really boils down to just one thing: exposure to venture capital. The rest could be replicated in a factor-tilted portfolio. They wrap up with a reader question: “How do you know when your strategy no longer works?” Find out Meb’s and Pete’s answers in Episode #9.

Comments or suggestions? Email us Feedback@TheMebFaberShow.com

Links from the Episode:

Running Segment: “Things I find beautiful, useful or downright magical”:

  • Philosophy: Return to the things we enjoyed as children – Pete
  • Fiverr – Meb

Transcript of Episode 9:

Welcome Message: Welcome to The Meb Faber show, where the focus is on helping you grow and preserve your wealth. Join us as we discuss the craft of investing and uncover new and profitable ideas, all that help you grow wealthier and wiser. Better investing starts here.

Disclosure: Meb Faber is the co-Founder and Chief Investment Officer at Cambria Investment Management. Due to industry regulations, he will not discuss any of Cambria’s funds on this podcast. All opinions expressed by podcast participants are solely their own opinions and do not reflect the opinion of Cambria Investment Management or its affiliates. For more information, visit cambriainvestments.com.

Sponsor Message: Today’s podcast is sponsored by The Idea Farm. Do you want the same investing edge as the pros? The Idea Farm gives small investors the same market research usually reserved for the world’s largest institutions, funds, and money managers. These are reports from some of the most respected research shops in investing. Many of them cost thousands and are only available to institutions or investment professionals, but now they’re yours with The Idea Farm subscription. Are you ready for an investing edge? Visit theideafarm.com to learn more.

Meb: Welcome to the podcast, friends. Today, we have a very special local guest, Pete Mladina. Welcome to the show, Pete.

Pete: Thanks, Meb. It’s great to be here.

Meb: So, I’ve known Pete for going on a decade now, and Pete is currently Director of Research at Northern Trust under Wealth Management. And his background, did a local undergrad at UCLA, did grad school in Europe. Were both those business? Study something else?

Pete: Economics and business.

Meb: Okay. And then stints at myCFO, Waterline, which then Northern Trust acquired, right?

Pete: That’s right. It acquired about five years ago.

Meb: So, I first met Pete…we both exist with one leg in this academic writing world, and Pete has written an enormous amount of papers. And so I remember way back, when I was doing my first book, conversing with Pete over some ideas. And so we started a yearly habit of grabbing beers at a great local gastropub called Simmzy’s, talking about financial markets. I imagine boring to tears anyone sitting around us, but having a good time. So today, we actually grabbed a few beers left over from Patrick’s podcast a few weeks ago, and they’re kind of the remnants. So we have a Lagunitas Pale Oat Ale, and what you have there?

Pete: I have a Celebrator Bavarian Double Bock.

Meb: Yeah, and then the other three were totally undrinkable, sort of weird flavors, so no one in the office even wanted them. So today, we’re cleaning it out regardless. So Pete and I have pretty esoteric interest in markets, and a lot of this could have the ability to go down a dark, deep, wonky finance hole. So we’ll try to pull back every once in a while, if we’re getting a little too deep on this. But Pete’s written a bunch of papers. How many papers have you written at this point?

Pete: Numerous articles. I’d say six academic peer review journal published papers, and one about to go to peer review at this point.

Meb: We talked about this the other day. I actually have much less patience for the time required for peer review process. I need much more immediate feedback. But there’s a funny story. When I put out my first and only academic paper, it was in my late 20s. And Journal of Wealth Management, which you’ve had a bunch come out in. And I remember checking it every day, trying online to see when it came out. Because I was really excited, really proud of it. And it comes out at night, and it says Quantitative Approach to Tactical Asset Allocation, by Melanie Faber. And I was so sad, and just panicking, and just emailing everyone at the staff of Journal of Wealth Management. If you’re listening, I apologize now. But saying, “Please, dear God, tell me this is not coming out in the print version.” And thankfully, they fixed it in time. But not only did they get the name wrong, they got the gender wrong as well.

So anyway, what we are going to do today, we are going to talk about a lot of different topics. I’m going to try to use as guideposts some of Pete’s papers. And so, there’s a lot of really interesting ones. If you’ve been a blog reader for years, you’ve definitely seen me mention a number of Pete’s ideas before. So let’s talk about a more recent one. And this was a paper called Optimal Lifetime Asset Allocation with Goal-Based Lifecycle Glide Paths. That’s a mouthful. It came out in Journal of Wealth Management this summer. That’s actually pretty recent.

Pete: Right. Recent issue, I think.

Meb: And for all the papers that we can link to, we will. Some of these are behind some of the journal gates. But there are shorter versions, and articles, and we’ll put them all in the show notes online at the podcast link. Why don’t you tell us a little bit about what this paper is about, why you wrote it, and what’s some of the main takeaways? And I’ll interject and questions along the way.

Pete: Thanks Melanie, I mean Meb. And I, just really quickly, I agree with your sentiment. Every time I finish one of these papers, I promise myself I’ll never do another one. And unfortunately, ultimately, my curiosity gets the best of me. I’ve never been able to write a book like you have though, because I don’t really have the patience to go more than 30 or 40 pages.

Meb: That’s the thing is that many of my books are like 30 or 40 pages, so I just call them books and get away with them being books, whereas many of your papers are probably longer than mine.

Pete: You’ll have to teach me the art of expanding then. I’m very excited about this particular paper. I’ve done a lot of different research that I think is really interesting and helpful to investors. I’m all about, “Hey, what can we do to create the research that’s empirical, and academically rigorous, and the economic insights that are relevant and practical to real-world investors?” All in the effort to create better investment outcomes for both our clients at Northern Trust, and frankly, just broadly in the universe. And I’m really excited about this particular paper, because it really does a lot in solving, I think, a lot of problems that have been out there for a long time. Whether you’re talking about…more important, perhaps, to most people is the retirement space.

Do I have enough to fund my retirement? What’s the optimal lifetime asset allocation? Not just that asset allocation today. But over the course of my life, when I consider everything I’m trying to do, all the goals I may have, the risk preferences I may have. It may have more than one goal. These things can be heterogeneous, rather than a simple liability profile like a pension plan.

Meb: What’s an example of some of these goals?

Pete: Well, the clearest and most important one is a lifetime consumption goal, and you might think of that as your consumption retirement. If you’re a high net worth investor, you may be consuming from that portfolio today. So that’s the simplest one, and that’s the liability of a pension plan. It’s analogous to that type of liability. But the reality is, in private wealth, goals are more multifaceted, and more complex, and more heterogeneous than just that simple lifetime liability. You may have, in addition to ongoing consumption, there could be ad hoc consumption goals, a second home, air transportation, a yacht, whatever may be the case. Or wealth transfer objectives to the next generation, or…

Meb: Maybe funding a trust, or…

Pete: Absolutely. Or philanthropy, which is a really critical goal for very many wealthy investors. And all of these can have different risk preferences. They occur at different points of time. They have different priorities. And so therefore, they ought to have different asset allocations attached to them that all roll up into an optimal lifetime asset allocation.

Meb: So as you’re thinking about this…you also talk about things in this paper like human capital. So one, goals-based already is a little bit different than most people already think about it. Most people, as investors, they think about a all-in asset allocation. They think about whether they have ten accounts or whatever. They think, “This is how I allocate.” And the goals-based is a little more about, “Hey, you’re saving for buying a house in five years, or 20 years, or whatever it may be. You’re saving for retirement. You’re saving for this endowment.” All those may have different asset allocation profiles, and so you would end up investing in different ways for all those. But one of the other interesting parts about this paper is, also, you talk a little bit about human capital, which a lot of advisors and investors don’t think about.

Pete: Well, you hit on a really important aspect of this. So first of all, let me step back a second and say as a first principle, we argue that assets should serve a purpose. And if they serve a purpose, that purpose is really to fund a lifetime of financial goals. So if you start from that perspective, “Assets serve a purpose. That purpose is to fund a lifetime of financial goals,” you step back. And you would simply note that for the most part, the idea of maximizing return per unit of risk is not really a goal. It’s a means to achieving a goal. And so when we realize that assets serve a purpose, it’s to fund a profile of goals, we have to work backwards in solving that puzzle from the perspective of those goals. And those goals, at the highest level, really are either consumption or gifts. You can either consume them yourself, or you’re going to give them away. Either to your kids, or to philanthropy, or to taxes.

Meb: Or government, yeah.

Pete: Or to a state tax. So you want to be really thoughtful in about how you think about your goals. Because if you’re not, you’ll have a disproportionate share of it go away to a state tax. So, that’s the first principle. From there, if we recognize that assets serve a purpose to fund a lifetime of goals, we realize that optimal lifetime asset allocation has to include a number of concepts. It has to consider your unique profile of goals as a starting point. It has to consider your unique profile of assets, some of which are portfolio assets, but a lot of them may not be portfolio assets. They may be things like human capital, pensions, future inheritances. There’s a number of things that are assets that ultimately will fund these goals and affect the optimal lifetime asset allocation.

Meb: So, I assume you guys use some in-house software for this, to aid in some of the approach?

Pete: Right. This is a novel approach to asset allocation. It’s theoretically sound and built on what’s called an intertemporal capital asset pricing model.

Meb: That’s a mouthful.

Pete: Yeah, it really is. But just so you know, leading financial economists have pretty much argued more recently, 2014, 2015, that recent, that really an ICAPM, or an intertemporal capital asset pricing model, ought to be the benchmark for multi-period, or what we call intertemporal portfolio theory. So we’re going down a rabbit’s hole a little bit here. But it’s important to realize that what we built here is a very practical goals-based approach anchor to what the leading financial economists today are arguing is really the benchmark for portfolio theory.

Meb: I’m trying to think, as far as the retail facing crowd, the advisor, the average investor out there, I know Morningstar does some here. It’s not particularly developed. I know Betterment is a goals-based advisor. I’m just not that familiar with this software, and who else is actually doing this in this space. If I do a little research, though, I’ll add some to the show notes. I can’t think of many off the top of my head. I know Morningstar talks a lot about this. I don’t know if they have a good software suite that aids in it. What are some of the takeaways for this? So, if you had to take a step back and say, “Okay, how does this really impact people’s allocation?” Is there a certain kind of broad takeaway you can mention such as, “Oh, well, it ends up in people having more in bonds.” Are there any sort of hard and fast takeaways, or no?

Pete: No. You hit on a number of things here. First of all, what you get essentially is a customized glide path over your own lifecycle, unique to your own assets, goals, and risk preferences.

Meb: Because most glide paths just say, “More in stocks and risk assets. The older you get, the less.”

Pete: And there is no theory behind it other than, implicitly, you have reduction in human capital. And that’s really the only academic justification for a glide path or a changing asset allocation through time, is you have more human capital when you are younger, and less when you’re older. So your bonds replace your human capital as you get older. There’s a couple more justifications, like mean reversion and things like that, that can justify dynamic asset allocations. But fundamentally, the outcome here is not a static portfolio that might be an output of traditional portfolio theory. But actually a dynamic and adaptive asset allocation, a unique customized glide path, unique to you, that’s adaptive through time, based on your unique assets, goals, and risk preferences, and how those change over the course of your life. And I think that’s the novelty.

Yeah, there are other goals-based methods. In my opinion, I argue it in the paper, some of them are built off of, for example, using shortfall probability as a definition of risk. That’s problematic for a number of technical reasons. The most important of which is that you might have a low shortfall probability, but that also may mean that that fundamentally with…the trade-off there is that you have a higher potential magnitude of potential shortfalls. So that’s the trade-off.

Meb: I also appreciate in the paper that you talk about people living to the age 100.

Pete: That’s a 98th percentile outcome. That’s adaptive framework, where you can pick and choose your life expectancy, and you can adjust and adapt over time. And also, I point out, I think, an important application here, more as you go into the defined contribution 401K space, is going to be longevity insurance. This idea of deferred income annuities. Not as investment vehicles, but as any good insurance product should do, it’s really about transferring risk. And in this case, it’s about transferring longevity risk, the risk that you’ll outlive your portfolio.

Meb: And this paper was mainly considering stocks and bonds. You include some other assets. What’s the universe in this?

Pete: So, the universe is all assets. And it’s interesting. You point to stocks and bonds. Maybe we framed it that way, as risky assets and safe assets, really. And largely, that’s an interesting transition point to, really, risk factors. Because really what we are talking about there are the super risk premium risk factors that are available in capital markets. The first of which we’ll call the market factor, which is essentially equity, or equity-like risk, or default risk. And that’s what mostly driving the returns of private equity, public equities, hedge funds, high-yield bonds. And then there’s the term factor, which is really the risk you take on for bearing interest rate or duration risk, and that’s more common to investment-grade bonds. And so, really, it’s not so much stocks and bonds. It’s risky assets and maybe we’ll call them safer assets. But fundamentally what we’re talking about there are two super factors that dominate the returns and capital markets, and that’s the market factor and the term factor.

Meb: So do you guys use this asset allocation approach? I assume it works with institutions as well. It’s a little bit different. Many of them have super long-term horizons, a fairly set spending schedule, but they tend to have different goals and often unrealistic return assumptions. But I assume it would apply similarly to institutions as it would to individuals as well.

Pete: Yeah, absolutely. This framework applies…well, first of all, there’s a couple of ways you can think about this. The framework that we talked about a moment ago, the goals-based framework, really applies to anybody or anything that has a liability profile. Now again, it’s most relevant to private wealth where goals are heterogeneous, a number of different goals with different risk preferences. But it’s equally applicable and easy to apply to an institution with a homogeneous liability profile, a population of plan participants. Now, what you’re really getting to is what I’ll call factor-based asset allocation.

And really, with the introduction of smart beta products, you see increasing interest not just in capturing smart beta. Which is a term I hate. We refer to it as engineered beta-type solutions. But fundamentally what you’re referring to is those underlying fundamental sources of return that drive capital market portfolio returns, asset class returns, and in fact, manager returns. And at the highest levels, I noted there are two. For each asset class, you can go deeper. And I’m sure we will in a moment, but what I want to point out is that at Northern Trust, we were pioneers in factor-based asset allocation. And even if you don’t have a profile of goals to apply this method that I outlined in this paper, really at the same time, even in the absence of goals, you can still apply this for more robust diversification.

Because fundamentally, what we’re doing here is we’re optimizing to find the optimal mix of risky assets, inclusive of hedge funds, private equity, etc. And then the optimal combination of safe assets, which are dominated by, basically, that term factor. And then we’re able to…again, relying on this CAPM framework, or ICAPM I should say, intertemporal CAPM framework, we’re able to combine these things in different mixes to create what we call robust efficient frontier. And we’re using robust here in the classic sense of the word, which means it can withstand any kind of risk environment. Because fundamentally, risk assets, and risk control assets, term risk, and market risk are uncorrelated.

Meb: So, I’ve done a little bit of a two step forward and backwards in between these two topics. But as we transition into thinking about factors…and basically, thinking about smart beta, or any of these terms, just moving away from a lot of these market cap weighted indices into tilts or other asset classes. I’ve got about eight different papers I’m thinking of. Is there one in particular you think we could start to highlight and then move down the list? What’s a good segue, do you think? Do you think it’s…?

Pete: You mean, of the ones I’ve written?

Meb: Yeah, the ones that you’ve written. I’ve got a bunch.

Pete: Let me tee it up first, and then maybe I’ll let you figure out which way you want to go. It’s really interesting, because when you…and again, this research started with work by Gene Fama and Ken French in 1992 and 1993. Where, prior to their work in this space, there was recognition that, “Gosh, it seems like small-cap stocks outperform large cap stocks. It seems like value stocks outperform growth stocks,” things like that.

Meb: And those were the big three. It was market, value, small-cap. And then the biggest, the fourth, came along later.

Pete: Yeah, momentum, which I know you’re…

Meb: My favorite.

Pete: You’re hugely interested in momentum, one of the most robust premia, came actually just a year later, a year or two later. And what’s interesting here isn’t just that these things offer premia when you do them, when you build them in a long-short framework, or even in a long-only portfolio. It’s not just that they offer premia, a source of return that is statistically significant. That’s how I’ll define a premium. It’s not just a random result, but a robust or statistically significant premium. It’s also that they do a very good job of explaining, or spreading out risk across different portfolios. And so these things have been termed risk factors or compensated risk factors. Not simply because they come with a return premium, that’s extremely important, but also because they explain the return and risk across a broad set of different portfolios. And so that becomes really interesting, because not only could you think about tilting away from the broad market to capture these risk premia in a long-only or long-short framework. But at the same time, you can think about evaluating total portfolios from this framework of risk factors, asset classes from this framework of risk factors, or even strategies and individual managers, funds of individual managers. Whether they’re hedge funds, fixed-income managers, or public equity managers.

Meb: And so as you are starting to look at distilling the returns of some of these managers, or strategies, or asset classes, is there a list of factors? Do you say, “Hey, look, here is the four main ones”? Or maybe, “Here’s the ten main ones I use in descending order of importance.” Or, “Here’s the 70.” How many are there, typically, you’re using for this analysis?

Pete: That’s a great question. You’re seeking the most parsimonious model, really. And a lot of so-called factors, especially now with smart beta products, people are inventing their own so-called factors. And a lot of these don’t meet the same definitions that I noted earlier. Robust risk premium, they’re uncorrelated with each other, so they’re unique and different from each other, And that they explain the spread, return, and risk across a broad set of portfolios. I call that the three primary criteria. Not all of them hit it as well. But yeah, we’re looking for the most parsimonious model. At the portfolio level, we think in terms of those two big ones, the term factor and the market factor.

As we dig into the individual asset classes, I think you can broaden it out. So in equities, especially nowadays…it started off as you note, Meb. The first three were market, the market factor, the return premium of stocks over cash, the size factor, or the return premium of small over large stocks, value factor, momentum factor. And now we know there’s a newer factor that was discovered in 2013 by an academic called Novy-Marx that looks at profitability as a factor. And Fama-French more recently actually added profitability and low investment, what some people call the combination of two with quote-unquote ‘quality’ as that additional lens to view equity returns. As we go into fixed income, there’s term risk and there’s credit risk, credit or default risk. It turns out, really interestingly, that credit/default risk in the fixed-income space is what we call linear combination with the market factor. So fundamentally, it’s arguable that that’s not a unique and different source of return. That’s just a manifestation, maybe a nonlinear manifestation, of market risk.

Meb: It’s meaning you’re getting the same thing in a different form, so there’s some overlap.

Pete: Yeah. Well, I think you’re getting the same thing. You’re getting market risk. But as Merton [SP] has shown a long time ago, and is increasingly becoming part of fixed income models, he showed it’s basically a call option on the assets of the firm.

Meb: This happens in a lot of asset classes. So a lot of people will buy eight different U.S. equity ETFs or mutual funds, and think they’re diversified. They’ll own a large-cap growth, a small-cap value, mid-cap go anywhere, and they just basically get the S&P. And another way of saying it. We say this now. It’s not always true. But buying, say, emerging markets, commodities, and short dollar is all the same trade, or at least it has been for the last few years. They’re totally different asset classes. Totally different, but there’s probably a fair amount of overlap.

Pete: So, again, getting back to what a true factor is, versus things and risks folks consider in their portfolio that don’t rise to the level of the true factor. So you may consider currency risk. You may consider the price of oil. There are a lot of things you can consider in your asset allocation. But the reality is, none of those really rise to this level of pricing the risk and return of portfolios. And I think that’s the really important nuance of factors. And when you step back and you build the analytical tools from a factor-based perspective, you have a very powerful perspective that you can apply to evaluating the skill of managers, evaluating the risk of managers, and at your broader portfolio.

So, for example, we found overwhelmingly in the equity side, we can use market size, value, and momentum. And frankly, market size and value do the heavy lifting there. But overwhelmingly, 95%, 96% of the return variation of all managers and funds in the Morningstar database are explained by those basic factors. And really, you can add momentum and profitability to that mix. As we noted earlier, additional factors in that space. And the lion’s share, almost all the alpha, goes away. And so really, that gets back to…once you realize that, you really realize that these things are driving not just the return, but the compensated return, and therefore the risk, the compensated risk, of these strategies.

Now, I wrote an article, or my team and I wrote an article I should say. I’ve got a number of great folks on my team that help me out. But we wrote an article on the equity side that applies to these basic principles to the mutual fund database over the last five years, and we found precisely what they found in the academic space. I wrote a paper last year on hedge funds. Now there’s certainly a broader set of risk factors involved in hedge funds. So, hedge funds are multi-asset-class portfolios.

Meb: We said before that saying ‘hedge fund’ is like just saying like, “Dog.” It’s a hugely different universe. Now, when you put them all together, it looks like one thing. But you’ve got to parse them by style and what they do. So, long-short equity guy looks nothing like a dedicated short seller, which looks nothing like a man who chooses a course.

Pete: Totally, but see, here’s the interesting thing. So you’ve got to think about, “What’s the set of factors that could be available to a broad multi-asset class portfolio?” And so let’s combine those factors we talked about on the equity side. Primarily, we’ll say the market factor, the size, the size factor, value, and momentum. Let’s combine the factors on the fixed income side, term and credit. And it turns out there’s a couple of other risk premium related to value and momentum that are available through futures-based strategies, and long-short commodity momentum is an example. Trend following, which I know you’re a big proponent of, Meb, is another example. And in the currency space, it turns out there’s a return premium to basically a carry tray, which is owning high interest yielding currencies and shorting the low interest yielding currencies.

Meb: So, and why is this important is that as the markets get more competitive and efficient, you have what formerly was seen as alpha, what people could buy that was hard to capture, that people didn’t understand, increasingly get commoditized as what we call maybe alternative beta, or beta. Meaning you can distill it to a simple rules-based strategy, and people shouldn’t charge much for it. So you can go buy a Vanguard value fund, maybe a bad example, or a momentum fund, and pay less than half a percent. Whereas most active managers charge 1%, 1.5%, or 2% and 20%. And so you say in one of your papers, you said, essentially, “Look, a lot of these guys you can distill down to these low-cost factor-based strategies.” However, you do admit that you said, “Focus on your asset allocation, but consider expanding your universe to either capture unique and different risk premia available from alternative assets.” Maybe talk a little bit about that. Or do you guys spend the time in search of ways that there is unique factors, or managers, or active managers? What’s your approach to say, “All right, this whole area has been commoditized. But maybe there’s still value over here”?

Pete: So, you hit on a number of things. And I’ll try to take them one by one at least, at least the ones I want to take. So first, yeah, you’re absolutely right. Thirty years ago, you would have seen a manager who beat the S&P 500 over the a 10 or 15-year track record. You would have said he’s a great manager. Nowadays, we have value factors. We can run regressions, and we can test, “Was that really skill, or was that just generic exposure in a tilt towards value stocks, that nowadays we can implement in a very efficient, low-cost way?” And 99 out of 100 times, we find that’s exactly the case. So nowadays you know that what used to be skill and a credit to that manager back in the day before we knew all this stuff…that was skill back then, because they had inside information that the rest of us didn’t have.

Now that we know that, it’s a different story. We’re going to pay for what’s worth paying for. So, that’s that. On the issue of…just to get back on the question of hedge funds. So we would apply a total portfolio, we call it a portfolio factor model, to hedge funds. Because we don’t really know, with any individual hedge fund, the mix of risk premia or risk factors they may be trying to capture. And when we do that, what we find is that different hedge fund strategies, whether you’re talking about equity long-short, event driven, relative value, macro, whatever it is, they’re nothing more than just different bundles of these risk premia. That’s all they are. Now, getting to your last question, again, that was the key insight of that paper. Not just recognizing that, but then, “What are the portfolio implications?”

Well, most of our clients, most of your clients, most people own generic portfolios of stocks, bonds, and cash. But it turns out that there are other risk premia out there that are different than just long-only stocks, bonds, and cash, that you can own in a long-short framework. Whether we’re talking about value in equities that you own long-short and you lever up to some expected return, or you’re talking about a currency premium as I noted a moment ago, you go long highly yielding currencies and short low yielding currencies, and then you lever that up to a suitable expected return. Whether you’re talking about trend following, as another source that’s different than just long-only stocks, and bonds, and cash. So fundamentally, once you know these things and you can find a manager who can be really efficient and cost-effective in the implementation without all the bells, and whistles, and nonsense, you have a very good approach to expanding beyond a very traditional portfolio of stocks, bonds, and cash.

Meb: And what do you think some of these, maybe the ones you just mentioned are the examples, but what do you think are some of the best diversifiers to a traditional portfolio? So, stocks, bonds, cash. If you had to add something, go talk to a policy portfolio committee and say, “All right, you guys have got stocks, bonds, and cash. You can only add one or two things. What would you add?”

Pete: That’s a great question. And I think to answer that from the perspective of how I think about things, I need actually step back one more step. And so I think about…so there’s backward looking statistical significance, then my own forward looking confidence. And so, a lot of these risk factors have what we’ll call statistical significance far higher than the significance of bonds, returning a premium over cash, or equities returning a premium over cash. So you think about momentum or trend following, for example. What we call T statistics, or the statistical significance of those strategies, are huge going back in history. Some strategies are harder to implement than just owning bonds or just owning cash. There’s transaction costs involved, taxes. These are very practical drags that have to be considered. So I think about it from a confidence perspective. I know that we live in basically a capitalist society, that fundamentally funds our economy with debt and equity. And debt and equity is essentially term risk and market risk.

So on a forward-looking perspective, when I look at risk premiums that I have the highest confidence level in, I look at the market factor, the returns of equities over cash, and I look at the term factor, the returns of bonds over cash. So that’s my starting point, and that’s why most people own stock, bond, cash portfolios, I’m quite sure. Now there may be, looking back, a lot of statistical significance with value trades in particular, momentum trades in particular, trend following in particular. And the strength of this significance is stronger in some asset classes than other asset classes, and we focus in on those where the significance is greater. It tends to be a less significant phenomenon in bonds than it may be, say, in currencies or equities in particular. So that’s how I look at it. If you’re looking to diversify your portfolio more, or capture more return, your next confident place to go after finding that optimal asset allocation between stocks, bonds and cash, I would say is the different variations of value in momentum across certain asset classes. I think that’s the next confident place to go.

Meb: We always talk about…do you guys, do you like managed futures, hate it?

Pete: Yeah. To us, managed futures is simply trend. And so then the question becomes, again, “Is it a so-called skilled manager who has some sort of a managed future strategy, or is this fundamentally nothing more than a very basic generic trend strategy?” And that becomes your risk factor. And similar to that value manager I gave you an example with a long time ago, I’ve looked at…a number of managed futures strategies have come across my desk. And I can say it’s a very similar story to what we’ve seen with value managers 20 years ago. What you thought was skill is just a risk premium. And that’s true with…

Meb: Yeah. Look, managed futures, [inaudible 00:31:53] guys, I love you. It’s my desert island strategy, if I had to pick just one. But I think the vast majority of them, you could write down rules that are not too different than what Charles Dow wrote down 100 years ago and quantified, and then be pretty similar. The devil is in the details, of course. But I think in general, the basics of it, I agree it’s not, in my mind, a two in twenty strategy, but something that could be implemented for pretty cheap. And it seems like a lot of them do a pretty basic, a pretty similar style strategy.

Pete: Well Meb, let me give you a good example. And I’m not going to name names, but there are…

Meb: No, go ahead. Go ahead. Fire away.

Pete: No, I can’t, for many reasons. But the really interesting thing in my mind…and again, it gets back to utilizing these tools, utilizing these methods and sophisticated tools. And again, you can’t buy any of these tools off the shelf. You’ve got to building them yourself, because there aren’t vendors selling the same type of academic risk factor exposures as we’re using. And so I’ll give you an example on trend. And again, I put trend as a distant cousin of momentum when I mentione value and momentum being the two premia across asset classes I’d be interested in outside of stocks, bonds, and cash.

Meb: And just a quick note for readers. Momentum is comparing assets. So hypothetically, is gold doing better than stocks or bonds over a 12-month period? That’s a traditional. Think about a car as going around a race track. Which one’s in the lead? Which one’s accelerating? Whereas trend is simply, “Is a market going up or down?” And there’s many ways to look at it, but that’s the basic differentiation between the two. Keep going.

Pete: So, the interesting thing is when you evaluate these managers, we have available to us a very generic trend following strategy as a risk factor. And it’s very simple to implement. And fundamentally, what’s interesting is, we can run regressions against their performance. And what you find is they’ll have a beta, or an exposure…which, by the way, you can dial in your own trend data by using leverage or deleveraging with cash, right? So it’s not so much how much return you want. You can choose that yourself. And we find is it’s all the beta, and then there’s a lot of idiosyncratic or uncompensated risk. So their strategy as they communicate it to you around the table may be very complicated in nuance, with all these different issues with regard to time horizons and trend signals. But at the end of the day, what does it aggregate to? It aggregates to nothing more than some exposure to this generic trend strategy, and a bunch of additional risk in all those active bets.

Meb: This reminds me of a quote, and I may have used this on the podcast. I’m sorry if I have, but I love quoting Charlie Munger. And he has a quote where he tells a joke. He says, “There’s this guy and he went into a fishing shop, and he was looking at all these lures. And he asked the store manager, he finds one in particular. He’s like, ‘And its purple with streamers that are sparkling, and everything else.’ And he says, ‘Wow.’ He’s like, ‘This is amazing.’ It’s a $20 lure. He says, ‘The fish actually bite this?’ And the store manager looks at the fisherman and he says, with a smile says, ‘Sir, I’m not selling to the fish.'” So a lot of the investment managers that the dress up these highly intricate, complex strategies, in many ways allocating to a simple blend of these various factors, could accomplish the same thing. It’s actually ironic, because I was fishing last week in western Colorado outside of Gypsum, and Dotsero, and Eagle, if you know where that is, listeners. But I was with my brother and a fishing guide, and the fishing guide pulls out, I think it was a pheasant tail. We were fly-fishing. There was literally purple with a little streamer on it, which then caught a bunch of fish. And I was dying laughing, because I’d never seen a purple fly before. And I was asking him about it. He says, “Yeah, yeah. The fish love it here.”

So, who knows, maybe there’s something to that joke. But so you see this as a commoditization of strategies over time. If you look back 50 years ago…and there’s a good book, and I’m blanking on the name of a bit, but it was about the Harvard endowment, and how they were very early into a lot of asset classes, such as timber. And then as more and more people get into it, it gets a little commoditized, and it makes me think of even simple value strategies. So when Fama and others were talking about it, price-to-book was one of the first ones. And then price-to-book since publication, and it used to be, may not be anymore, DFA’s favorite value factor, it’s been one of the worst of the value factors. Maybe because a lot of money chases it. I don’t know. But in general, it becomes something where when money…and here’s a good question for you real quick, because there’s been a lot of debate about this recently in the investment community.

We have people lining up on both sides, whether it’s Rob Arnott, whether it’s Cliff Asness, is talking about these smart beta factors. And some people say, “Yes, you should just allocate to the ones you like, and wash your hands, and be done with it.” And others say, “Well, actually, you need to incorporate evaluation of these actual factors.” So, for example, are low vol stocks cheap or expensive relative to their history? Are value stocks cheap or expensive, etc.? And Research Affiliates has written a lot about this. And he says, “Actually, yes, you should take that into account, because you may have a factor strategy. That, because of its expensive nature, actually negates the entire premium you may get.” Do you guys think about that at all? Do you look at it? Do you get excited about smart beta, timing, and all these things? Or what’s your opinion on that?

Pete: Yeah, so you hit a bunch of things there. First thing I’m going to say really quickly is you’d mentioned low vol. And it’s really interesting, because when you add those two quality factors, I noted profitability and low investment, to all those low vol strategies, that goes away. So fundamentally, that gets back to, “We have be careful in defining factors, because sometimes they’re redundant with existing factor definitions.” Similarly, the value premium or the value phenomenon, regardless of how you define it, it’s all a manifestation of the same underlying phenomenon.

So you can use price-to-book, P/E ratio, price to cash flow ratio. Whatever you use, you’ll find that as you build those returns, they’re all linear combinations of each other. Which basically means they’re different manifestations of each other. But over a five-year period, they may look and feel a bit different. But in terms of the entire return premium, they’re fundamentally the same. And so again, it requires this set of tools and understanding of…what I wouldn’t say is it necessarily a commoditization of products, but a recognition of the fundamental underlying risk factors that seem to have driven portfolio and strategy returns, really since the beginning of empirical research.

Now, will they be arbitraged away? And then I’ll get to your question of factor timing. Will they be arbitraged away? I don’t see any evidence of that right now. And that said, if they did, that is really a testament to adaptive and efficient markets. Because fundamentally, that would be moving things to Sharpe’s original CAPM. And so that may happen, but I don’t see evidence of that. Simply because looking at value stocks versus growth stocks. Obviously, value stocks have under-performed the last few years. But that happened by way of not so much value stocks getting more expensive. It’s happened by getting cheaper, but growth stocks getting more expensive. So it wouldn’t surprise me if that moderates at some point.

Now, despite what I just said, the question about factor timing is really a question about mean reversion and the evidence of whether there’s mean reversion in market returns. Now, there is some very modest evidence in equity markets over an intermediate period, let’s say five to seven years, that equity market returns mean revert. It’s statistically significant, so it’s there, but it’s barely statistically significant. And half of academics argue that equity returns mean revert and the other half don’t, so there’s certainly a debate over that.

Meb: So, you’re not to be going to the investment committee at any point saying, “You know what? We got to get out of Japan, or value stocks, or whatever, because it’s gone too crazy.”

Pete: Well again, that goes back to whether there’s mean reversion, if there’s evidence of mean reversion or not. Because if returns don’t mean revert, then they’re independent, completely independent, which means what just happened has no relevance on what’s about to happen. And so, let me extend it past the market factor. The credit factor, again, default risk which I said was related to the market factor, again, there’s some modest evidence of mean reversion there. But again, it’s easily debatable. Now, as we move beyond those to the value factor, the size factor, things like that, the evidence is much weaker for mean reversion.

So, I think my strong practical advice is if you, again, want to move away from a generic portfolio of stocks, bonds, and cash, your next best place is to look to these risk factors. But unless you think you’re even better at timing the equity markets, you shouldn’t dare time these. The other thing I’ll say where I agree with Cliff Asness in his debate with Rob Arnott is not only do I think for most people, what I mean by that is everybody, they should stay the course. If they decide to take on a factor tilt and just bear with it, much like you’ll feel the pain of equity sometimes too, when you decide to own equity in a portfolio, it’s no different. Market factor is a factor, no different than a value factor. Hold on, Meb. Let me finish this idea really quick.

You can’t say that small stocks are cheap, and so therefore the size factor is attractive. Because now you are, this is where I agree with Cliff, now you are fundamentally intermixing the value factor with the size factor. I think it’s intuitive to define a value factor by evaluation ratio. At least that’s intuitive. Whether or not there’s mean reversion is a different question. But you know what? You are blending factors, and they’re supposed to be independent. As soon as you take that same paradigm and apply to size, apply it to momentum, apply it to profitability, you have problems in the independence of factors when you do that.

Meb: I was interviewing at a quant hedge fund in the early 2000s, will remain nameless, in San Francisco. And I remember going in, and they admitted they use something like 70 or 90 factors. I can’t even remember what it was. Just every possible factor you could possibly program into a computer. And I remember thinking, I was like, “Huh. I feel like most people, at this point, they all have the same PhD.” So if you type a ticker of a multifactor fund, so LSV, we wrote about in our last book, really famous quants, I think out of Chicago. And managed a bunch of quant funds, have done great. But you type in a ticker of one of their holdings, and you’re going to see a laundry list of other quant shops that all own the same stocks. It’ll be D.E. Shaw, maybe a AQR, Bridgeway, a bunch of other shops. And so you do see a lot of the factors that, when you have maybe properties of cheap, high-quality, momentum, whatever it may be, you know the data sets are such that everyone got facts that…everyone’s got all the same data.

So, the challenge of finding unique and different factors in a world of…I rememngber back in the days of talking to portfolio managers, and how they used to get value-added information. There was a lot of borderline grey areas of legal, illegal, taking a CEO out, getting him drunk, and letting him just spill his guts type of ways to trade. That’s a lot harder these days, I think. And so on the lookout for new asset classes and factors, I think, is pretty tough. And actually, in one of your papers, which you may or may not be able to talk about, cut me off if you can’t, that’s talking about real estate. Can you talk about that?

Pete: Yeah, I can talk about that.

Meb: So I think this is a good example. So a lot of people, and this is timely where they’re splitting out…it was S&P or MSCI. I can’t remember. Splitting out REITs as a separate sector of the universe, and looking at real estate investing. And a lot of people talk about the main asset classes, stocks, bonds, currencies, commodities. A lot of people talk about REITs or real estate as an asset class. But really, it’s kind of an amalgamation of some of the other factors or asset classes you’re talking about.

Pete: No, absolutely. It’s private real estate in particular, because you have smooth returns because of appraisal-based pricing. And so a lot people viewed…a lot of private investors owned private real estate to the almost exclusion of all other asset classes. It’s the darling asset class for a lot of investors, a lot of private investors anyway. And what we found there…to step back, you can pretend you have 70 factors. But again, we prefer to lean on the work and the definitions of the academic space, because I can run some tests and show you that 70 factors are reduced to six or seven truly independent factors. That have the same…

Meb: But it sounds so much more rigorous if can have a hedge fund that has 3,000 factors.

Pete: It’s less rigorous, because you’ve weakened and loosened your definitions.

Meb: How do you sell a fund that only has three factors? It’s so much harder. All right, keep going.

Pete: So, here’s the interesting thing. In a nutshell, again, real estate is the darling asset class. From an institutional perspective, it has these powerful diversification benefits. From a private wealth perspective, if you talk to all sorts of people we encounter who…they don’t want to diversify multi-asset class portfolios. They’re fine with their seven or eight apartment buildings and two commercial real estate properties. And so, the really interesting work here is…well, let’s step back in what we did in this paper. And it continues to be a working paper. But we evaluated REITs knowing that, okay, at the end of the day, REITs are essentially traded income-producing properties that have some leverage built in them, equity REITs anyway. And we know that, hey, well that is effectively real estate, very similar to a core real estate portfolio that may not be traded on an exchange, and may be levered or may not be levered.

So we took a look at that REITs and we said, “Okay, based on this factor-based perspective, what are REITs? Do these risk factors do a good job of describing them?” And in a nutshell, what I can say is that REITs are interesting and not interesting at the same time. What they are is, really, they offer surprisingly a rich mix of factors that are really analogous to a portfolio that may be something like 60% small value stocks. So again, size factor, value factor, market factor, 60% and 40% high-yield bonds. So again, factors, term factor, credit factor, and high-yield bonds. And so that’s really interesting. And I think that 60%-40%, that risk tilted 60%-40% portfolio, explains why among so many private investors they’re comfortable with just a real estate allocation. Now here’s the interesting thing.

Meb: Because it ends up looking like a diversified stock, bond…

Pete: Right, “Hey, I have a diversified 60%-40% portfolio that’s risk tilted with small value and term in credit betas.” Now here’s the interesting thing, though. We can replicate that with a stock and bond only portfolio that utilizes these engineered or smart beta concepts with far less risk than that REITs strategy. So you wouldn’t just own those REITs, because fundamentally that’s suboptimal. You can own the same compensated return premium that comes with those REITs for far less risk, and an engineered beta solution that includes intentional exposures or betas to small value stocks and to high-yield bonds. Now, okay, that’s REITs. And I know what you’re saying. I don’t own publicly traded REITs. I have my seven or eight apartment buildings and my two or three commercial office buildings in my area, and that’s what I own.

And the interesting thing is we can run additional unsmoothing tests, and I think that’s the other novelty of the papers. We introduced a new approach to unsmoothing private equity, or private real estate returns, that are consistent with risk factor betas. And what we find then is when we do that, the entire return premium of private real estate is completely explained by basically their current and lagged exposures to REIT returns. And in fact, there’s additional risk that’s uncompensated in private real estate, unexplained but uncompensated. So again, even in the private real estate space, again, it boils down to these risk factors you can own in stock bond portfolios.

Meb: So why even bother with private real estate?

Pete: Well, I think the same is true with private equity, except we observe a robust or statistically significant alpha in private equity. And so that may be due to a higher prevalence of manager skill in private equity than public equity. It may be because of asymmetric information, sourcing ability, whatever. Or in a liquidity premium that comes along with private equity. But in private real estate, we actually don’t observe that same alpha like we do in private equity. And so that’s a great question. And in the paper, we leave it to future research. There could be a higher prevalence of alpha among a small subset of managers. And so we make that argument for hedge funds, for example. Although on average, hedge funds may not be attractive. There could be a small minority of them, if you employ these risk factor tools and methods, that contribute meaningful value to your portfolio, whether quant or traditional skill-based. And the same may be true of private real estate. We just, again, on average, it’s not the case. But it could be that a small subset offer robust alphas that could significantly improve a portfolio of generic stocks, bonds, and cash.

Meb: This is a similar topic in another one of your papers that probably may have originally brought us together. I can’t remember. But maybe ten years ago, wow, where you were talking about the Yale endowment. And I think the title was something like The Yale Endowment Returns. Is it skill, or is it just risk exposure? And maybe you could walk people through just the thought experiment you did with the Yale endowment, which has been one of the best performing institutions over the past 30 years. And the experiment you did, and conclusions you came to from that portfolio. And we’ll talk a little bit about private equity in that context too.

Pete: Obviously, a very interesting topic. And it really gets back to, again, risk factors applied to portfolios, applied to asset classes. We talked about real estate hedge funds and apply it to managers. And in this case, we were applying to portfolios. And the issue’s been these endowments really only report once a year. Which means even if they’ve been reporting for 20 years, we only have 20 return observations, which make it a little difficult to run traditional return risk attribution on them. And so in that particular paper, we took more of a holding space perspective based on how they were describing their allocations. And we tried to replicate those allocations from a holding space perspective. And in reading through their annual reports, you would realize their equity allocations tended to be small cap oriented, value-oriented, things like that, that we then were able to rebuild their portfolio and run a number of tests.

And ultimately, we concluded that the unique source of their excess return was fundamentally explained by really just their allocation to venture capital. And really, everything else was replicable, again, by a pretty straightforward factor tilted portfolio that could be owned in a portfolio of only stocks, bonds, and cash, traditional assets. And really, we were able to disentangle or decompose those returns so that we were able to identify that it was uniquely venture capital. And since then…I can’t remember the authors. There have been one or two additional papers that they did. Unfortunately, they didn’t site my work, because they probably wouldn’t have written their papers if they had read it. But they actually confirmed independently. It came to a similar conclusion. Now as a follow-on to that, about a year or two ago at Northern Trust, we wrote an investment commentary that expanded it beyond Yale, and looked at a, let’s call it an overall Ivy League endowment portfolio. And I know you’ve done a lot of work on that, Meb.

And similarly, looking at all of these portfolios, again, we were able to dial it down into risk factors on average. And even among let’s call them elite investors, those that run Ivy portfolios, all you had to do was expand it to private equity. And then premium involved in private equity, as we noted, venture capital in particular. And you were able to really explain away all of the Ivy League endowment excess return. So it’s not a manager skill issue. It’s really just thoughtful, long-term allocations to risk premium public markets. And in private markets, areas where there is either a liquidity premium or things like venture capital, where skill and sourcing matter.

Meb: We’ll give them credit in the sense that the one thing they did do well is that, decades ago, they did have a global allocation, and did start to tilt. Whether through active managers, or through quant strategies, too many of the things we were talking about. So they were early, in many cases, to a lot of this.

Pete: I agree, Meb. And I think that’s a skill. And so, again, I don’t want to discount that, because I think that’s precisely right. It’s like Warren Buffett’s returns, and a couple of academics from AQR have been able to replicated it with a combination of value, profitability, which are now factors as we know, and leveraging service float. And so that’s a classic example. Yeah, it’s great to do it ex post. And now that we know that, that is informative on a forward-looking basis. But it’s another thing to actually realize that 20 or 30 years ago, and implemented it. And back then, nobody knew that. So that was skill. Totally agree.

Meb: So hypothetically speaking, and we did an article called Should CalPERS Be Managed By A Robot? And the thinking was, “Hey, look, they have this large global allocation. You could just go by a bunch of ETFs for them, probably. Just low-cost separate accounts, replicate their portfolio, not have to worry about all the headache of managing this with 300 employees,” etc., etc. But theoretically, for an institution, you say, “Take a step back, a very honest step back,” at this point. So, “You know what? All right. For these main asset classes and strategies, we can allocate to these alternative beta, whatever you want to call, these simple rules-based or tilted funds. And then, we’re going to dedicate our entire resources to the place where we really can add value, like private equity.” So should Yale just index everything and then go say, “Hey, we’re just going to focus on finding the best-of-breed private equity,” and then that’s that. Would that be reasonable? It seems like that would be a reasonable idea.

Pete: Yeah, I think it’s reasonable. But of course there’s agency costs that are involved with regard to professionals who sit on the investment committee. And then there’s also a sizing issue. You get so large. Can you even get a meaningful enough allocation to hedge funds, for example, knowing that when you are aware of these tools and methods, really it’s only a small subset that are interesting? Can you even get to the allocation necessary to move the needle and improve your portfolio enough? And I think there’s probably more you can do in private equity than hedge in that regard. But the committed capital in private equity is not too much different than what’s invested capital in hedge funds today. I think the interesting example is Norway sovereign wealth fund. Where they went down this pathway, and pretty much decided what they’re going to do is have broad, global exposure that’s passive or semi-passive, and they’re going to try to harvest risk premium that we’ve talking about.

And I think it’s a recognition that once you get so large, it probably makes sense to do some form of that, if not indexing. Some form of passive or semi-passive investing that you can implement and stay the course. The other thing I will say, the value that a lot of these pension plans, these large plans can offer, isn’t so much on…a lot of it is really thoughtful fulfillment. But I think they’re way off on their asset allocation. They’re not realistic about the liability profile they have, the return assumptions in this environment that are necessary to achieve those, and the implications for funding ratios, etc.

Meb: Yeah, we tweeted out a study that showed what a lot of these plans still expect is for their returns, and particularly for their hedge fund returns. And it was so crazy that I can’t even begin to go back to it.

Male1: Let me just step back a second. And what amazes me is, among a lot of those folks, there’s not even an understanding of the difference between an arithmetic expected return and a geometric one. A geometric one is what you actually get to earn and consume. But people model portfolio optimization and Monte Carlo’s with arithmetic returns. And so they’re embedding in expected returns that are typically arithmetic, not understand the geometric flow. Or not everybody, but I encountered this so much it scares me.

Meb: We call those volatility gremlins. It’s what we talked about. I just looked it up. The institutional investors expected their hedge funds to return 13% net, 13%. And let’s forget about the actual hedge funds that returned, what, half of that. Historically, even the best performing indexes. And this is also something you showed in one of your papers. So 13% net means, at two in twenty, that’s about what? 20% gross?

Pete: Well, there’s a bigger issue there. When you realize the risk factors that drive hedge fund returns, and then you have expected returns for each of those risk factors, that goes way down. Far below equity, somewhere between bonds and equity. And as an example, we can show you that the average hedge fund is roughly approximated by a simple factor mix that you could say is about 35% equity and the rest cash. And so that’s the risk return profile of the average hedge fund out there. So you can associate…with no statistically significant alpha, by the way. So you can associate an expected return of whatever your forecast is for cash, which is close to zero. And then equities, and then take a third of it. Now again, that’s the average of what you should be interested in. Hedge funds are unique and different source of return, not necessarily high returns, because really, the value of a select hedge fund is a diversification value.

Meb: All the hedge funds we know are better than average, right?

Pete: We’re all are better than average, on average.

Meb: One more question. Here’s a twitter question for you. And we’ll probably start winding down, because we’re hitting the hour mark. We did a twitter question, and it was aimed at me, but I will let you tackle it first. They said, “Meb, how do you know when a strategy,” and we can apply this to factors too, or even active managers, and there may be different answer for each. But, “How do you know when it basically no longer works? Or how do you know it’s been commoditized? Or how do you know that it’s time to sell something that previously you had a reason for buying?” Take that broadly. Answer any of those variants that you think would, “How do you know when your strategy no longer works?”

Pete: Yeah, I think that’s a very, very, very tough one, and I don’t think you’ll know for 10 or 15 years. That’s part of the problem. You don’t know anything in this business. You can only attach confidence levels to them. And I think to be a good investor, you’ve got to start thinking that way. It’s not black or white. Everything is grey. There’s an expected outcome. There’s a variance around that expected outcome. And with that comes essentially confidence levels. And so I’ll give you an example. There were many times in history that a value premium or value tilt would have under-performed a growth tilt or the market for five or 10 years, or more.

So it didn’t mean that it stopped working, it turns out. And so a lot of these things, empirical research, where you have evidence that supports a claim, really takes time to play out. So I think as time goes by, you have to evaluate all of the past empirical research, and look at it, and you have to make a judgment call.

Meb: And we think the same way, where we look at a lot of the strategies and we say, “We only want to allocate the strategies that are pretty basic, and you can explain, and understand.” Basically explain to a 15-year-old, and say, “This is why this works in value,” and a lot of these people can get on board with pretty easily. The challenge I think, of course, comes with particularly active strategies and active managers. So where you have a lot of other factors in play. How big are they? Do they have too much money to manage that they can’t allocate? Do they just have a divorce that you may not know about, and they go on tilt? Many other things going on. And active because you’re betting on the active manager is a lot harder, I think, than actually a quant style strategy. Because you can look at reasons why it works or doesn’t work. It’s a hard question, I think.

Pete: Well, absolutely. I’ll take it a step further and say the challenge there is that what you think is outperformance may just be factor risk. And that even if you even if you observe outperformance after solving for factor risk, then you have to ask yourself, “Is it likely random, or is it likely statistically significant and therefore true?” And almost nobody does it. We do that all the time in Northern Trust, is we evaluate the activity managers we use. But it’s a critical question, because a lot of managers will just perform well, even risk-adjusted, randomly. And you have to solve for whether what you were observing is likely true skill or randomness. And I can tell you that the prevalence of true alpha that’s statistically significant is very, very, very thin across all asset classes. And without these tools and methods, I’d say, “Good luck.” But with them, you have a chance. And I think that the chances maybe less than 1% or 2% of managers per asset class. It exists gross of fees better than net, so it’s very much a function of finding those managers. And if you can negotiate their fees down, that’s what’s really required to get there.

Meb: There’s a lot of papers in the academic literature that talk about asset classes and active strategies, and say, “Actually, these guys do generate some alpha. They just take all of it in their fees, and maybe then some.”

Pete: I think that’s what we’ve found. And I think sometimes the trick is, if you have the ability…again, we’re large organization, so we do. But you have the ability to negotiate down fees as low as [possible]. When you find a manager who has raw, pure skill, regardless of expenses they charge, that interests me. And then the question is, “Can we get it so that the skill survives the fees?” And usually the answer is no. But hopefully, you’re able to find a couple than can. And then you’ve got to ask yourself about confidence levels, right? So I believe in the market factor going forward more than I believe in the value factor, the momentum factor. But I believe in the value factor, the momentum factor going forward, more than I believe in the persistence of an individual manager’s alpha.

Meb: So I imagine we can go for another hour or two, but we’ll have to just have you back on in another 6-12 months and see what other papers you’ve written. I’m going to ask you one more question. We ask all the guests if there’s anything you thought of that you find particularly beautiful, useful, or somewhat magical? Any ideas here?

Pete: Yeah. It’s funny. You asked me to think about this a little bit. And I’m going to hit something I’ve been thinking about more recently. I think it’s an insight, but maybe not. For folks like us, Meb, sometimes I think we work more than we should. And I find myself, maybe it’s the curiosity in me, I find myself up at 11:00 at night writing some paper that I don’t need to write for my career. And I wonder why I’m doing it, and I realize I shouldn’t be doing it, actually. And I’m trying to find that balance between these things that I do find interesting, but the rest of my life that has absolutely nothing to do with this.

And I’ve recently realized, I’ve tried a number of things, whether it’s…I play tennis. I love to go fishing. This is first I’ve realized you’re a fisherman. We’ve got to connect on that later. I’ve realized really lately that although I have tried a number of things since I’ve been 20, for example, I find myself reverting back to and enjoying the things I enjoyed when I was in 6th, 7th, and 8th, and 9th grade. I think there’s a formative time then, around middle school, where I find myself just loving those things. And fishing is an example of that, or playing tennis, or whatever may be the case. And as a part of that, the things I’ve learned later, I’ve been less good at and less persistent with. And as part of that, I have two young girls now, and they’re coming of that age now. I just want to make sure they have enough experiences between 6th grade and 9th grade so that when they are adults, they’ll have these hobbies and these interests that they’ll love to go back to, because of the things they enjoyed when they were that age.

Meb: Are you going to be able to…are any of them…you’re heading to Europe soon, right?

Pete: I am, yeah. We’re going on vacation, actually, to Croatia, where I have some of my family’s background. And the beautiful Dalmatian Coast, I just had some number of weeks there.

Meb: I just had some friends that were sailing that coast, and I was jealously looking at all their photos. It looks pretty awesome.

Pete: Yeah, there’s 1000+ islands, great history. It’s like Greece, but a little further north. And huge Roman and Croatian history there. It’s a fabulous, beautiful place. I don’t want to market it too much, because it’s already starting to get too crowded.

Meb: Yeah. I’m trying to remember back, all things that I loved doing in 6th, 7th grade. So obviously, playing sports. It would have been laser tag back then. There’s a good investor we’ll have on this podcast later who talks about, he says, “As the younger generation starts to age and hit middle-age and older, and starts having the assets…if you think about investing, but in the collectible space, or trading space of art and collectibles, it’s, ‘What did that generation covet when they were younger?'” And so this generation it may be classic muscle cars of the ’60s, and all the things that people were interested in. And I’m trying to think about…I’ll think about it later, but things that I would be interested in investing in that I really coveted. Maybe it was a Ken Griffey, Jr. baseball card. I don’t know. Upper Deck, ’89.

Pete: All I know is I wish I did more things when I was in seventh, eighth, and ninth grade that I could do as an adult. Because you’re right when you hit sports. I played a lot of basketball and a lot of soccer. I just can’t do that anymore. But I can play tennis, right? I can go fishing.

Meb: The one that I just got back…so after I was on this crazy road trip, Canada, Colorado to give a speech, and then to Oregon where I have a bunch of guy friends that are…I was by far the worst golfer. So Bandon Dunes we went and played, and I had bought some clubs, and rushed to get a handicap. But what a beautiful location. The biggest challenge I have with golf is it’s such a massive time chunk. It’s a straight up 4 to 6 hours. That’s pretty tough. If I could go out, and no one else out there, and go play, and put on a podcast…

Pete: It’s a great example. Whether it’s golf or sailing, another thing I used to do a lot of, they both take a lot of time. And so the trick is finding these things that don’t take as much time but you still enjoy and love. And again, if you get kids, I’m trying to think about those things so they learn how to do them.

Meb: Yeah, just drag them along. Just drag them along. All right, well mine is totally different, and is a website called Fiverr. Have you heard of this?

Pete: No.

Male: It’s F-I-V-E-R-R. It’s a website where you can go on and basically pay anyone five bucks to do anything. And that’s a stretch, but say you need a logo, or you want someone to call your friend and sing him happy birthday in Al Pacino’s voice. Or, there couple I did where, for Valentine’s Day one time, I had a bunch of drawings done. First of all, it’s $5, so none of these are going to be Leonardo-type drawings. But some of them actually became very good, and some of them were really funny. And interesting story, and we’ll play this in a following podcast because I don’t have it on me, but we…if anyone’s ever heard Tosh.0, the comedian on Comedy Central, he has a disclosure in the beginning he reads where it’s like, “Don’t try this at home. These are things that will kill you if you do this at home.” But he had someone dub it where it was Barack Obama’s voice from all of his speeches. And so it’s very clear that he’s not reading it, that’s it’s cut up, but it’s pretty funny.

And so we put it on Fiverr to try to get people…we have disclosure we read at the beginning about Cambria, and, “This isn’t investment advice,” yadda yadda. And so we had to try to get a job for both Hillary and Trump ahead of this election time, just to see. And they’re so hilariously bad. I wish I had them to play for you. I’ll play them in the next podcast. We’re certainly not going to include them, because they’re so horrifically bad, but it’s pretty funny. But anyway, this website’s wonderful. Even to the point where I was like, “This is such a good idea. But why don’t they start one where it’s $20 instead of $5.” I was trying to buy the domain 20spot, but it turns out it’s a café in San Francisco. They wouldn’t sell it to me. I don’t know if they’re still around. I’ll have to look it up. Anyway, that’s a good one to check out. Pete, great having you here today. Where can people find more information, if they want to find you somewhere? Is there a good place to…home page, anywhere?

Pete: Yeah. Right now, yeah. Right now, it’s, just go to NT.com [northerntrust.com] and you can stick my name in there under “Our Research”, or “Our Experts”, “Research Insight”. You can view a lot of these papers that Meb and I have been talking about today. And we put out a piece at least every quarter, often more, and I try to do an academic paper once every year or two, so it’ll keep coming.

Meb: Good. Thanks for being here. Look, as a reminder, listeners, you can always find the show notes. We’ll link to all of Pete’s papers that the internet will let us. But you can always find the other episodes at mebfaber.com/podcast. Thanks for taking the time to listen today. We always welcome feedback and questions for the mailbag. We’ll read them in a future podcast on air at feedback@themebfabershow.com. You can always subscribe to the show on iTunes. And if you’re enjoying the podcast, please leave a review. Thanks for listening, friends, and good investing.