Introduction

Patrick: [00:00:46] My guest today is Michael Mauboussin, whose research on all aspects of investing is a treasure trove. Michael is a Managing Director and Head of Global Financial Strategies at Credit Suisse. In this wide-ranging interview, we discuss the current state of the asset management business, explore all stages of the investment process, and consider what edges may still exist for those trying to beat the market. Please enjoy. Thank you Michael so much for being here with me today. I have to start by saying when I started in this business as an active manager, my first role was to learn about investing and I was a philosophy major, so I didn't know much at all. And so I went to the bookstore and bought a bunch of books. And the first two that I read, one was by David Dreman called Contrarian Investment Strategies, which I still love to this day. And the second book was a book that you wrote with your mentor called Expectations Investing. And so we're going to get into that book today for sure, but I thought an interesting connection, so it's a pleasure now to be here talking about the industry with you.

Michael: [00:01:46] Thanks Patrick, and thanks for buying the book.

Michael's Personal Habits

Patrick: [00:01:48] So before we get too deep into some of the interesting ground we're going to cover today, and we'll talk about the investment industry, about active investing very broadly, and what the future may look like, I'd like to start by establishing your personal system. So obviously you're prolific, I guess I would call you researcher above all else, but writer, teacher, reader, so I'd love to understand what you think the key things are that you do, maybe not every day, but most days or at least weekly, what are the kind of habits that you have that has led to the various accomplishments through your career?

Michael: [00:02:23] I don't know about the accomplishments, but I think a lot of it is the foundational stuff that you probably hear a lot of people talk about. So for me, a few things are really essential, and the first big one is sleep. By the way that sounds silly, but it wasn't until I was well into my career that I figured out that sleeping an appropriate amount for me was an extraordinarily important thing for my productivity. So I always used to think cheating on sleep would let you pick up some productivity, but I found actually that was not the case at all, so that's a big one. Exercise is another huge one for me personally, and things like diet. And the other thing is always been dedicated from fairly early on to allocating time to reading. And people often say to me like, "Jeez, seems like you read a lot", and what's not always obvious is it's a trade-off. I'm not doing a lot of other things. So if you asked me about popular TV programs, I know nothing about them. So those are choices, and I'm not sure they're all the choices that everybody would want to make. But for me, those have been some of the really key things. And then of course just quality time with family and other things that keep a balance. So those were, I think, the big structures. And again, it's hard to say that everybody should be doing those things, but I think the sleep, diet, exercise, trio is so central to success, probably no matter what you're doing.

And people often say to me like, "Jeez, seems like you read a lot", and what's not always obvious is it's a trade-off. I'm not doing a lot of other things. So if you asked me about popular TV programs, I know nothing about them. So those are choices, and I'm not sure they're all the choices that everybody would want to make.

Patrick: [00:03:40] So a couple of questions about reading, and this is a popular topic these days, we seem to be in some reading renaissance. Everyone really recognizing the value of reading very broadly and often. Are there ways that you look for the next thing to read?

Michael: [00:03:53] For me it's a mess actually. So if you go into any of my offices, there are piles of books everywhere, and I'm not sure that I've got a very good method to the madness. Usually, there are... And I suspect Patrick the same as you, is in reading multiple books simultaneously. Usually, there are a couple of things that are more business-related, so whether it's finance, or business, or what have you, and then a couple of things that are far fields, whether they're science, or some element of psychology, or so and so forth. And I have rules to things like if 3 people, like if Patrick you call me up and say, "Hey, you ought to read this book", and I hear that same thing from 3 different people, usually I stop and I pick that one up. But yeah, that's basically the method. So I'm all over the place. I was on break for the last couple of weeks and I read 6 books in 2 weeks, and they're all over the place.

Patrick: [00:04:40] You mentioned the opportunity cost, do you ever worry about reading too much in a given field? Especially if it's an area that you're trying to figure something out. My angle is that it might color your priors, that it might make you less able to have original insights in the space.

Michael: [00:04:57] Totally. Listen, I think it's hard for all of us. We were the product of our cumulative experience, so it's hard for us not to have biases for sure. And interestingly, and I wrote a piece a few weeks ago about celebrating my 30th anniversary of joining Wall Street, and I gave a little background, which I hoped didn't bore people to tears, but the reason I did that was to... And I think I wrote it, like, "This is where I'm coming from, and these are my biases." And so if you want to know where my shortcomings are, or where my thinking is colored a particular way, I'm going to reveal my cards on that. That said, I think that the idea of reading widely and exposing yourself to lots of different points of view, being actively open-minded, I do think reading helps cultivate that quality in a person, and I do try to adhere to that as well. So I'm willing to read things. Even there are Wall Street folks that publish things that I don't necessarily agree with, but I think they're well-written and they're thoughtful, and I think they're sincere, and I read those things to try to understand all different points of view.

Patrick: [00:05:58] Are there one or two categories outside of, let's say, business and investing, that you find yourself returning to often as a reader?

Michael: [00:06:05] Yeah I think that... And I don't know if it's outside of business and investing, but certainly, the literature and psychology have just had a profound influence. And I think everyone's starting to get that. I do think there was the phase where we are going beyond understanding these cognitive biases that have been so well-documented, to asking the questions about what we do, and I've seen that's a big part of what you do and what your organization does. The second area that I always found I've always been fascinated by is, I might call the intersection of biology and economics. And so economics as a field, has grown up has been fairly mathematical, I think, deeply inspired by physics. And it turns out that most economic systems have much more of a biological field, which was much harder to model. And I think we're migrating in that direction. So in particular things like, are markets as complex adapt to systems? What does that mean? What are the implications for how we think about efficiency? Those kinds of questions, I think, have always been fascinating to me.

Patrick: [00:07:03] Reminds me of a book called The Nature of Value by Nick Gogerty, have you read that book?

Michael: [00:07:07] I have, yeah.

Patrick: [00:07:08] Really interesting intersection of biology and markets for those that are interested. So as a learner, obviously you're a big reader, but also a writer and a teacher at Columbia, I think you've been doing that probably almost 25 years now based on the note you put out a few weeks ago. One of those three things, your preferred way of learning, do you think, obviously reading, you learn a lot by reading, but is there more merit, or more learning to be done through writing and teaching maybe?

Michael: [00:07:34] It's an Interesting question Patrick, and I'd love to get your take on this as well. I feel that for me, writing and teaching, in particular, become very valuable mechanisms to consolidate understanding. So the truth is, I read a lot of stuff and I don't retain as much as I would like, to be honest. And I think to myself to really consolidate those thoughts, writing them down in a cogent and intelligible fashion, or teaching them even more important to a live audience of people who are engaged and questioning, really allows me to solidify that. Going back to your initial question, you said, "What motivates me every single day?" It's probably this combination of input, which is learning new things, or discovering, or delving into particular topics. But then consolidating by outputting, which might be teaching, or writing, or sharing thoughts with people. And until you can write it clearly, or until you can teach it in a way that's effective, I'm not sure you really understand it. And the last thing I'll say is you probably think back to the great professors you had, whether at Notre Dame, or you can go in high school teachers and so forth, who the people that really inspired you? And almost always it wasn't the people that were the most complex. They seem to make challenging ideas, difficult ideas, exhilarating ideas, accessible. And that is what it's all about for me, and that's what teaching is about, and that's what good writing is about. Is somehow you feel that moment as a reader, "Man, I get this, and this is exciting, and it's accessible to me. I may have to work a little bit, but it's accessible", and that's the goal.

The Active-Passive Debate

Patrick: [00:09:15] We used to call the good philosophy professors, or the best ones Hemingways. Because in philosophy, obviously things can get very complicated and jargon-like, and then you just see eyes gloss over, and I totally agree. I think that for me, it's writing, I don't teach as you do, I'd love to someday. But for me being able to write about something coherently helps me understand probably more than reading. Reading for me is as much recreational, because like you said, I forget. I keep notes, I've got a great repository of notes, but I forget most of what I've read. Okay, great. So that's really the foundation I expected to hear you say, but it's good to hear that some basic things like sleep, and health, and reading, and just... Sounds like exploration above all else of have interesting ideas drive your output. So now what I'd like to do is jump into, first the investment industry. We'll start broad and then get more specific, and talk first about this very popular dynamic or question of active versus passive management. I would say it's fair that this is the biggest question on the entire industry's mind these days. We're probably at about maybe a third, maybe 40%. I've seen that as estimates that high of equity assets, at least that are passively invested these days. And one question people ask is, "Where is this going?" But before we get there, I'd love to hear from your perspective, what you think the key functions are? The benefits, if you will. In aggregate, not individual managers, but in aggregate of active management.

Michael: [00:10:42] Well, I think the benefit is quite clear, which is by the way counters your own objectives, but the objective overall for society is that active managers gather information and impound that information accurately in prices. So in a sense, they create or contribute to efficiency. And efficiency from a societal point of view is actually very important because it allows, ideally at least, that our assets go to the best and highest use. So that the whole function is incredibly useful. Now the key, of course, is like I said, the better active managers are at this, the more feudal they're actual appears to be. So in other words, if markets are truly efficient, no one cannot perform, or very few people cannot perform beyond what chance would dictate. So to me, that is the essential role.

Source: BIS

Patrick: [00:11:29] So obviously we need some. It seems pretty clear that we can't be 100% passive. If we're at 30 or 40% today, where do you see this going? Do you think if you had to guess, and we know how futile forecasting like this is, but if you had to guess a rough percentage of, if there is some equilibrium, if there even is that equilibrium, where do you think that settles out in say 20 years?

Michael: [00:11:52] It's a great question. And let me... A couple of thoughts on how I might approach this, and I'm actually trying to put a pen to paper a little bit, to be more formal about this. The first thing which is implicit in my prior comment about active management is that it creates a positive externality that passive investors can take advantage of, and that is market efficiency. In other words, you can't have an index fund and charge 5 or 10 basis points without having some sort of a price-setting mechanism. And that those costs are largely being borne by active managers, and ultimately their customers. So active management creates a positive externality that passive can take advantage of. In terms of the framework, the one I keep coming back to is Grossman-Stiglitz, 1980, a very famous paper. And by the way, 1980 is actually important to note the date because the late 1970s, really through the '70s were probably the peak of the enthusiasm for the efficient market hypothesis. So that was when the Chicago school was really the rage and people really felt that efficient markets were the answer. So this paper came out in a sense it was a counterbalance to that. And the argument was basically, "Look..." The paper is called on the Impossibility of Informationally Efficient Markets, and their argument was, "Hey guys look, there is a cost to gathering information and in impounding prices. And as a consequence, there needs to be a requisite benefit and we're going to call those anomalies, or inefficiencies, or whatever it is." And you could argue those are unbalanced or NPV zero, but they have to exist.

Otherwise, there's no incentive for these active managers to do this. So this becomes the fundamental question, and you can think about this maybe dynamically on both sides, which is, "How big must the inefficiencies be, in order to keep some aspect of the active management business spinning to keep prices effectively efficient?" We're probably at the juncture where the active side had gotten a little bit too big, maybe a little bit too fee-heavy, so the opportunities weren't sufficient to justify the economics of the active business. So you're seeing that shift over, that's probably happening, but to me, obviously, markets can't be 100% passive to state the obvious. And the other interesting question is, "Might there be... How many people do we need versus, for example, technology or quantitative methods to allow that information to be impounded prices?" So it's a super interesting question, but that is the way I'm going to think I'm going to come at this to say, "How much is being spent in active across the board?" By the way, not just mutual funds, but hedge funds and otherwise. And then, "How big must the opportunities be in markets?" Global equity markets are very large, but, "How big must these opportunities be for this to be in rough equilibrium?", and then, "How do we get there?" So, I don't know how that does settle?

Patrick: [00:14:43] That's the framework.

Michael: [00:14:44] To me that's the framework to go at it. It's cool.

Patrick: [00:14:46] Yeah, it's a really interesting question, obviously, such a complex market and hard to know what the answer is. I think the idea of active managers as providers of liquidity is something that they can at least earn some of their fees through that method, especially if there are fewer and fewer of them. And maybe if you're writing a paper, I'd love to hear your thoughts about how we might be able to tell when there's too much passive. Is it index funds running higher tracking than we're used to? That might be another interesting question to explore in the paper. And I guess maybe that's a long-term question in the more immediate term. I'm an active manager, you talked to a lot of them and worked with a lot of them throughout your career. So a big question for the active management community is, "Well, in this period of transition, assuming that this kind of linear market share gaining from passive continues over the next decade or two, is more and more passive good or bad for active managers?" Now you've heard very famous people on both sides of this. So Buffet decades ago was famous for saying he liked passive because there was less competition, but now he advocates that people invest in index funds. I'm going to read one more opinion, which I think is a really interesting one here, "I believe that indexing will turn out to be just another Wall Street fad. When it passes, the prices of securities included in popular indexes will almost certainly decline relative to those that have been excluded. More significantly, as Barron has pointed out, “A self-reinforcing feedback loop has been created, where the success of indexing has bolstered the performance of the index itself, which, in turn, promotes more indexing.” When the market trend reverses, matching the market will not seem so attractive, the selling will then adversely affect the performance of the indexers and then further exacerbate the rush for the exits." So that was actually Seth Klarman writing in Margin of Safety in the late '80s, early '90s, in what is an otherwise extremely valuable book. Got that one a little bit wrong, certainly not a fad. So I'm curious where you fall on this continuum, is more indexing make it easier or harder for active managers?

Michael: [00:16:46] You going to give Seth Klarman that one? No. So I think a couple of things that are, well, certainly I don't think it's a fad and a couple of observations that would make. The first is to state the obvious, that alpha or excess returns in markets by definition have to be zero before fees for a given period. It's a particular year or whatever it is by definition. Because it's a Y-intercept of zero, so above it's got be zero. So for you to have positive alpha, there's got to be someone who's got to have negative alpha. And that's a very important thing to bear in mind. So to me, one of the interesting... And I'm making this more as an assertion than necessarily a statement that I can back up fully. But my sense is, a lot of the people who were the weaker players in the game, that is other institutions that were less sophisticated, or less qualified, or mom-and-pop, have been leading the move toward indexing. And as a consequence, the people that remain in the active game tend to be the smarter guys, not the dumber guys. And if that's the case, and I think that's true, it sparks what I've called the paradox of skills, not my idea. But the paradox of skill basically says, "In activities where skill and luck contribute to outcomes", which are certainly true for investing, "more skill leads to more luck." And the key there is not the absolute level of skill, which I think uniformly we've seen gone up, it's the relative level of skill. So the relative level of skill is going down over time.

The decline in Standard Deviation of Excess Returns for U.S. Large Capitalization Funds (Declining variance in skill)

Source: Credit Suisse

So the very best in the average guys are less different. And as a consequence, there's a parody in the fact, and so luck takes over. So that's a whole first big thought on this is that "If it's the case that the weaker players are leaving, it actually makes it not easier, but more difficult for the remaining active managers because they're competing with the other smart guys." The second thing about indexing is that if you own, say, the S&P 500 Index, you're going to earn the index return. You could argue perhaps that some prices within the index, some stocks are mispriced, but you're going to earn the index return. So if that's your benchmark, and I don't think it's an unreasonable benchmark for active managers, that's a reality. And the third thing is if there are these mispricings that would actually feedback into opportunities for active guys, or in other words, arbitrage will kick in and so... Now there have been a couple of things, there's a paper by Russ Wermers, University of Maryland, where he's talked about, "Stocks that are highly passively held, tend to be less accurately priced or efficiently priced than stocks that are more actively held." Did I say that properly? So that's a difficult case to make.

Patrick: [00:19:33] Yeah, for sure.

Michael: [00:19:33] But there might be something there as well. So, "Do I get a sense there might be a liquidity argument pointed out with liquidity and active guys?" There might be a liquidity argument, so it's not that we can go tumbling into passive and have no active managers, that's not going to work either. But we just have to be measured about this, that for most people, I think, being passive or indexing is not an unreasonable strategy. But that doesn't mean that there's going to be no room for people to do things that are active.

Patrick: [00:20:01] So you say it's being led by the exodus, let's say, or the shift in market shares being led by less skilled, let's call them investors. But I wonder about the remaining composition of active management. So you think about, "Who's losing the assets? Which active managers are losing their assets?" So a couple of thoughts here, the first being that obviously, people tend to fire managers that have underperformed. There seems to be, maybe not super strong but, some mean reversion in performance. So maybe these guys are getting fired before they were going to go on to do quite well. And then the second idea is, if you look at the rise of what we call closet indexing in the active space, especially here in the US, I wonder if, and that's been a very successful thing, by the way, managers that have, you could call it smart beta, there's a lot of different versions of it. Fundamental guys do this too, where massive funds effectively, you're getting a beta exposure, but it's in the active bucket. You could argue that that's not a very skilled player. Even if the manager themself is a genius, if you manage 2, $300 billion, or a $100 billion dollars, or something like that, it's pretty hard to differentiate yourself at that scale. So I wonder whether or not, certainly as like sharp ratios have fallen things like that, there seems to be some evidence that relative skill is falling. But I also wonder whether or not some of the industry dynamics, people managing career risks, this idea of closet indexing, et cetera, affects kind of this alpha pool that's leftover?

Michael: [00:21:26] I agree with all that. I think you're nailing it. The first comment I would just make on regression, and you make this point accurately. And I think that if you talk to clients, you talk to consultants, they'll all go on about this topic yet they still do it, which is this idea of firing. At the end of the day, they go, "Hey Patrick. Well, look at what you're doing, and your process, and your people, and your philosophy, and that's what we care about." But at the end of the day, look, the truth is, the past performance is probably the most powerful indicator, whether you're going to get hired or fired. The thing is that, so you fire underperforming managers and you hire outperforming managers, in the subsequent periods it doesn't mean that the worst guys, the bad guys...it basically means there is going to be regression, so you expect bad performance to be... It's all within randomness. So the expected value of the next outcome is some alpha close to zero, and for both the good guys and the bad guys. So that's the first thing.

The second thing on the closet indexing is absolutely fascinating. Which is, and there have been other things, so now as a mutual fund, you have to disclose the benchmark you're competing with, and that's a fairly new development in the history of mutual funds, that could be a precipitating event. If markets are being more efficient, it actually may be completely rational as an asset manager and someone who wants to retain assets, to start to index your benchmark to a greater degree, because the chance of you shooting the lights out, maybe not be very high, but the chance of you getting crushed, also not very high. So as long as you stay in the game a little bit, you can retain your assets, you've got a pretty good business. And you're absolutely right about career risk, which I think is a really really big deal in our business, which is, Keynes got that famous quote, “Worldly wisdom teaches better to fail conventionally than to succeed unconventionally”, and I think that's a big factor. So if you start to do something that's really off the beaten path, even if it's economically justified, if it doesn't work out in a relatively short period of time, you're going to be under a lot of pressure. And that's also something that Seth Klarman, notwithstanding his indexing comments, has made a lot of really interesting, and I think very thoughtful observations about clients and the nature of clients. And I think if you ask him about the success of Bell Post over the decades, I think he would say that having the right clients was an essential ingredient, not just they're great investors, but the central ingredient in their success. Because it allowed them to execute the process in a way that they may not have been able to do with a different set of clients with different objectives.

If markets are being more efficient, it actually may be completely rational as an asset manager and someone who wants to retain assets, to start to index your benchmark to a greater degree, because the chance of you shooting the lights out, maybe not be very high, but the chance of you getting crushed, also not very high. So as long as you stay in the game a little bit, you can retain your assets, you've got a pretty good business.

Patrick: [00:23:51] Yeah, I think I was once asked if I could have dinner with any investor, who would it be? And Klarman was my answer largely because of the business that he set up, as much as his investing skill. Obviously, he's a deep, fundamental guy. I'm a quant guy, so not a whole lot of overlap there, but I think that his, this idea of client alpha finding the right partners, I would even call them partners more than investors may be the most, going forward as passive continues to gain, share maybe the most important aspect of what active managers do. Maybe that wasn't the case in the past. But I think going forward-

Michael: [00:24:26] I agree with that. You know, and he's got a lot, man. I remember him saying this at one point where he said, "a great client is one who cashes a check when we hand it to him and writes us a check when we ask for one". I thought that was really neat. Right? So in other words, when we run out of investment opportunities, at least episodically run out of what we perceive to be good investment opportunities, we're going to give you your money back and you don't complain cause we've typically had a good performance. And when we see opportunities, typically when clients are scared, they will stroke the check and allow you to put money to work. And it's a judgment call on both sides, but exactly, I think there's, there's the huge alpha and getting the right people behind you. And by the way, things like you think about Berkshire Hathaway, the fact that you've got an insurance business that generates float, so you have a guaranteed sort of source of capital. That's just a huge advantage over long periods of time, right? Because you always have money to deploy. So when somebody has some deal to strike with you for an 8% preferred or whatever it is you can write the check and-

The Expectations Investing Framework

Patrick: [00:25:25] Yeah. Float definitely is a source of power. I think you look at the second half of Buffet's career it's as much about recognizing the power of float as it was kind of value. So great. So, I broadly agree with the kind of where we're going in terms of active and passive, that there will be a role for active and understanding that I'd love now to get into active management. So you've written a ton about this. I would categorize you as a very meta thinker writer, trying to convey useful systems to active investors for thinking about things like value, portfolio construction, capital allocation. So I'd like to get into a few of those. I mentioned at the outset of the conversation, your book "Expectations Investing", maybe you could start there. That was a really useful book for me. I know a lot of fundamental investors who hand that book out. So maybe, if you could, describe what that framework is.

Michael: [00:26:18] In a very brief history of this is that you're a philosophy major. I was a government major. So I came to Wall Street, very uninitiated with basic ideas of business and finance. In some ways that were a liability, in other ways, it was an asset because it did prompt me to go back to first principles on a lot of things. And in the late 1980s, a colleague of mine handed me a copy of Al Rappaport's book, "Creating Shareholder Value" which had a lot of really cool ideas, but chapter seven in particular in the original book was called stock market signals to managers. And the argument was that, hey, you're a CEO of a company if you're investing so as to earn above the cost of capital, that's actually not enough, right? For your stock to go up, because the market may expect you to do that already. The key is for you to do better than what the market expects. So, clearly, his target audience was corporates, but as an investor, the bells are going off. So I started writing research as an analyst, as a junior analyst using a lot of these Rappaport techniques. And that allowed me to meet him in the early 1990s and sort of towards the late 1990s, he came to me and said, "Hey, listen, maybe we should write this book, "Creating Shareholder Value" book for investors. So that's where the idea for "Expectations Investing" hatched. By the way, it couldn't have been launched at a more dismal time, or actually, our website was launched on September 10th, 2001 the day before 9/11. And of course, as you recall is the middle of a three-year bear market. So when we were preparing the book it was the roaring 1990s, by the time it came out was a less fortuitous time. But the idea behind Expectations Investing is a very fundamental, simple one. It doesn't mean people shouldn't buy the book. But here it is in 30 seconds.

Patrick: [00:28:00] Which is now on Kindle.

Michael: [00:28:00] But here's the idea very simple, right? Which is stock prices reflect a set of expectations for future financial performance, by the way, all asset prices do. Second, as an investor, you should be able to hopefully have some strategic and financial analysis to allow you to determine or get a sense of that fundamental performance. And number three is you're looking for mismatches. So, at the end of the day, and by the way, growth versus value investing, what is the distinction? There really isn't that big a distinction. All investing collapses to buying things for low expectations, right? So it's all about that basic concept. So we developed this framework to try to be accessible specifically for investors and the main point I like to make, and there's a dash of hyperbole when I say this, but I always say that the biggest mistake I see among active managers is a failure to distinguish between fundamentals and expectations. In other words, when things are going well, we want to buy stuff, whether it's a company or the market. And when things are going badly, we want to sell. And what the expectations approach forces you to do every single time is to step back and say, let me compare fundamentals versus expectations. And it's really that gap that allows you to generate excess returns. So I know, for example, you guys probably think about this. You can use quantitative methods to identify low expectations. And I think that's a completely valid and thoughtful approach to this, but you never want to lose sight of that ultimate objective, which is that gap between fundamentals and expectations. So, that's what that book is about. And I think it's still very valuable as a way of thinking today. And our industry continues to be replete with rules of thumb and old wives, tales and heuristics that work a lot, but often don't work. So we're trying to peel the onion back a little bit and get more to first principles.

In other words, when things are going well, we want to buy stuff, whether it's a company or the market. And when things are going badly, we want to sell. And what the expectations approach forces you to do every single time is to step back and say, let me compare fundamentals versus expectations. And it's really that gap that allows you to generate excess returns.

Patrick: [00:29:51] So I'm curious, kind of getting into growth and value a little bit more. You say in the book that most stocks need 10, maybe 15 years of value-generating cash flows to justify their current price. Obviously, we know what we know about forecasting, kind of like the psychological biases. This is another category that we know that forecasting in markets is exceptionally hard. It's hard for one year, let alone 10 to 15 years. And now you've got companies, the average lifespan in indexes at 10 or 15 years. So things seem to be speeding up. You say that growth and values, really not all that different, but I wonder if assuming that another way of saying growth and value are low and high expectations if this framework is more appropriate or will produce better outcomes for active investors amongst value stocks that let's call them statistically cheap, low PE, low price-book value stocks.

Michael: [00:30:42] Yeah, no, Patrick, I think you're right on. First of all, that time horizon, that 10 to 15 years was written, I think it was at the time a contemporary, I think it was true at the time. I think it's less true today. So I actually think that believe it or not, expectations are less today in many ways than they were back then. So, that's the first comment, but you're exactly right. So I think Buffet and other people have talked about this, I mean, I always struggle with this distinction between growth and value because, at the end of the day, the common denominator is low expectations. So saying it differently, I mean maybe another metaphor if I could bring one in, would be that of a high jumper, you know? So the expectations are where the bar is set. The fundamentals are how high the company will jump. And you're looking for companies that can clear the bar, right? So there may be cases where the bar is at eight feet and the company can jump 10 feet, the classic growth company where it's going to be great, and that's actually a very attractive investment. There may be places where the bar's two feet in the company can only jump afoot. In which case it's actually, although cheap, not attractive because expectations are actually too high. But as you point out, absolutely that all things being equal, lower expectations are the common denominator, and we can often get toward them with statistical measures. So, when people say about value investing, and low price to book, or high yield, or low multiples of earnings, that in and of itself is not what I'm after, but those may be the proxy for low expectations.

Patrick: [00:32:04] Right.

Michael: [00:32:04] So if I can clean that up a little bit, or maybe that's the pond at which I want to start fishing, that's great, but I also don't want to do that of the exclusion of more growth-oriented businesses, because there may be cases where they can do well. Now, the other thing you talked about forecasting and I think that's really an important thing. You know, we've done a ton of work recently on thinking about the rate of the fate of excess returns. So if you have a high return on capital business, how fast does that fade back down to the cost of capital? And you can make some distinctions across, for example, sectors or industries to start to get a feel for that. They're probabilistic statements, but it's not like you're flying completely blind on some of those things. So those are all really interesting questions. And I think the unifying theme is low expectations, but you make a couple of distinctions that are really important for people to bear in mind.

Patrick: [00:32:52] Yeah, just one other observation here, which is that I growth is so interesting because obviously it underperforms all sorts of evidence that very expensive stocks underperformed longer term, but the category of expensive stocks also produces some of the best individual outcomes. So you can call these lottery-type outcomes, massive success stories where you buy a Facebook that's expensive, but the subsequent returns are still incredible because there was a gap, right? Expectations were high. They should have been higher. So, that happens sometimes. But for us as quantitative investors, we just like to say, okay, we just want probabilities in our favor. We're never going to own a Facebook or an Amazon that's too bad because obviously, those have been two of the best stocks for investors in the entire world. But categorically growth just has expectations that are too high. So that's what I really took away from your book was let's find systematic proxies for low expectations, make big bets on that category, not necessarily individual names. And I'm sure there are great investors that have the skill to do that on the fundamental side, but from a quantitative perspective, it seems like cheapness is one good proxy for low expectations. Now in the book, you're critical of, I think, justifiably, so of simple things like a PE ratio or earnings as the appropriate proxy for the health or success of a business. Could you explain why that is and maybe what you would prefer over earnings as an alternative?

Michael: [00:34:18] Yeah. So earnings, in particular, I guess the argument we- It's interesting that Al Rappaport again, my mentor and my collaborator, his Ph.D. is actually in accounting. And so I think he's now associated more with finance and creating shareholder value and those kinds of concepts. But at the end of the day, what he studied was the world of accounting. And I think most people probably are of the firm belief that the lifeblood of a business and not just public companies, but private companies or anything, is the cash that it generates over its life. And when I mean cash, I mean money, you can put in your pocket, take out capital, not capital cash money you can put in your pocket. And so the question becomes the degree to which earnings represent were a proxy for cash flows. And over the very, very, very long haul, those things do get reconciled, but for periods of multiple years, decades, and so forth, there can be a fairly large dichotomy between earnings and cash flow. So, that was really our point of emphasis is to say, earnings can be very misleading in that regard. The second thing about earnings is that at the end of the day, capital allocation, distills to investing, and by the way, there are people issues as well, but for, let's say for the money, you want to invest in opportunities to generate returns in excess of the opportunity, cost of capital, right? So you want to create value and it is possible, it's trivial to demonstrate this mathematically, it's possible to grow earnings and destroy value simultaneously. So in other words, you're investing below the appropriate rate of return. So even though you're growing, you're actually digging yourself into a bigger economic hole. And for those couple of reasons, I think earnings have a lot of limitations. Now, the PE, this is just a spinoff of that, right? So what is a PE ultimately about, and what I love to do with PEs, the way we wrote about this is going back to talk about first principles, very famous, 1961 paper by Merton Miller and Franco Modigliani on valuation. And you know, it's funny because today you read a finance textbook, it feels like a lot of these issues have been settled, but you know, 1961, there were a lot of loose strands in the air. And these guys try to pin them down. And the question was, Hey, what does the market value? Does it care about earnings? Does it care about cashflow? Does it care about dividends? And there were people that argued every one of those things and the way they came out of it, and they said, listen, actually, all these things don't matter. It's all these things together because they're all tethered, right? But there's a formulation in that paper, which I love. And they actually have some very elegant language around it, which is accessible today and relevant today. And they said, look, you could think about the value of the business in two pieces. The first is sort of the steady-state value. So, they earn a dollar today. You know, they're going to earn that dollar into perpetuity. And the second piece is what they called, well, it's now called the present value, growth opportunities, basically future value creation. So one of the ways you might think about this as say, a restaurant chain. The stores in place today would be the steady-state value, and then the future stores they build that create value will be what you're paying for the future.

The Steady-State Price-Earnings Multiple (1961-2013)

Source: Credit Suisse

So you could think about value broken down into those two components. Now the steady-state component, if you say dollar earnings, the multiple that you assign to that would be basically the inverse of the cost of equity capital. So I don't hurt myself, we'll pretend that's 8%. So that's a 12 and a half times multiple. So that's basically saying if I pay 12 and a half times, or less, for current earnings, and those earnings are sustainable, I'm paying for nothing for future value. And then the second piece. So what are you paying above me on? So if you think about the market, I'm just making this really simple, say the market's kind of multiple say in the high teens call 12 or 13 times as the base, the rest of that is for future value creation, right? So if you look at that series and we did some simple ways to do this, go back to 1960s, let's say 55 years about two-thirds of the value of the S&P is steady-state, and about a third is future value creation, right? Which doesn't seem crazy, right?

Percentage of S&P 500 Attributable to Anticipated Value Creation (1961-2013)

Source: Credit Suisse

So to me, this sort of dismantling the PE, and by the way, the future value creation, I'll just extend on for one moment, has to do with three things. One is how much you can invest, at what rate of return, for how long, right? So now it's a really nice way to take this sort of abstract number and break it into these pieces in a way that I can think about analytically. That's a lot more intelligent than tying back to value investing or low expectations investing. As you said, if you can find companies where you're paying very little for future value creation, where there's a demonstrable record of value creation, you think there's some reason to believe it will persist. That's pretty darn good, right? Because that means you're paying basically nothing for future value creation and a lot of upside opportunity. Likewise, you might have businesses that current earnings are not sustainable. They're descending really rapidly. It may look really cheap, but walking right into a value trap, right?

So if you look at that series and we did some simple ways to do this, go back to 1960s, let's say 55 years about two-thirds of the value of the S&P is steady-state, and about a third is future value creation

Patrick: [00:39:08] Yeah. It's an interesting way of breaking it out and really thought about that way, but the idea of a value investing strategy is not relying on the future, which is so hard to know. And if we know anything about the future of fundamentals, maybe the most ironclad rule is that they're mean-reverting. So, you take an industry like the retail industry today, which has been obviously, totally hammered, beaten up really badly. Amazon's capturing tons of market share. They trade at, in many cases, extremely cheap valuation multiples that I'm sure if he did the work you just described, what you'd find is that there's basically nothing in there for future growth. If anything, people probably expect a decline in sales or revenue for the industry in the next five years or something like that. So maybe that's another way of understanding why value has worked, that you're basically not paying for the future at all. You're, you're paying for the current stream and maybe less.

Michael: [00:39:58] Absolutely. And by the way, my colleague at Columbia business school, Bruce Greenwald, who's brilliant and an amazing teacher. He often says valuation, the way he would teach it would say start with a balance sheet. So literally walking through the balance sheet and appropriately marking down the value of assets and liabilities to understand some sort of modified book value. Then look at earnings power, which is a steady-state value we've been talking about. And then third and only in quite reluctantly pay for future value creation. So that's sort of the classic value investor, Ben Graham mindset. But you're right. Now that said, and I know you said you'll give a pass on things like Amazon, and these other high valuation companies, that said there may be ways to think about where growth will be value-creating, which kind of companies will do well. So I know as a discipline you're saying we'll let that go. But I'm also saying that there may be ways for people to think about those without violating any of the principles we're talking about.

Patrick: [00:40:55] So do you think, using Amazon as the example, because obviously you really couldn't capture this, maybe using like language processing, you could do it in the future, but you couldn't capture this quantitatively. Do you think there are qualitative ways of finding visionary companies/leaders like Bezos who say in his famous '97 letter, listen to focuses on the customer on long-term free cash generation, and then actually walk that walk because obviously it's been a tremendous success story? There are other stories like that, and those companies tend to be exciting and expensive. So a quantitative system will never own them. Do you think there really are repeatable ways of finding those companies ahead of time? Or is it just catching the lucky star?

Michael: [00:41:34] Yeah, I think it's probably more the latter, the lucky star phenomenon. That said, I've been interested in Amazon for a long time personally in part because a good friend of mine was the analyst at the time of the IPO Bill Gurley, who's now at Benchmark Capital.

Patrick: [00:41:51] My favorite writers.

Michael: [00:41:52] Yeah, really smart guy, really great guy. And Bill was at Deutsche at the time, and he basically said, hey, these guys get it. And the second thing is he introduced me, and I got to know quite well at the time their CFO was Joy Covey. And Joy, she tragically died a few years ago, but she was a remarkable person and thinker. And I think that her influence is why that '97 letter looks the way that it does.

Patrick: [00:42:20] Really?

Michael: [00:42:20] And so I think she was the one that placed great emphasis on ROIC, great emphasis on free cash flow first and foremost. Jeff certainly got the customer-centricity, but they understood that scale could be really important. So they always kept their prices. They were always a little bit of ahead of where they needed to be. She understood the cash conversion cycle. So I think that you look at any time we talk about success, anytime you see great success, there's been a lot of skill and a lot of luck together in big, big doses, both. And I certainly think that would be relevant and it would explain Amazon. But I think that they sort of developed this ethos early on with this focus on cash. That's been really important. The second thing I'll say, which is interesting, and I don't know if you guys are looking at this factor, you probably have in some way, shape, or form, but I've been very interested in this work by Robert Novy-Marx on gross profitability and quality. And the simplest measure for that is typically gross profits divided by assets. And you look at Amazon on almost every traditional measure, multiples of earnings and EBITDA and so forth. And it really leaves you swimming a little bit, but on gross profitability, on quality, they look really good. So their revenue growth has been quite robust, but also their gross margins have been going up. So their gross profitability, Novy-Marx gives some sort of guidelines for things that look attractive and they would be in the top quartile of that. So I guess the counter indicator and probably had maybe has been a better predictor of the stock price performance. But yeah, without commenting on that company, I don't know anything about it in any greater detail than that, but I think there were certainly some elements of what they were doing even 20 years ago that would give you an indication that they were thinking about the world the right way, at least.

Patrick: [00:44:09] Yeah. Novy-Marx, and I guess the whole idea of quality has become a really popular one in the quantitative world. Probably the most recent big factor, one of the big four or five factors. Our opinion is that quality is more useful. There's more utility as a negative screen. Meaning companies and quality now mean so many things it ceases to have any meaning, but when we say quality, we mean balance sheets, leverage, how companies report earnings, how companies invest their capital. Conservative choices basically are how we would define quality and good returns on invested capital. What we find is that the bigger negative or positive excess return is in the bad tales. So companies with really bad quality do consistently badly, versus the market, those were the highest quality of the highest, Novy-Marx's gross return on assets do outperform, but not to the same extent that the battle is underperformed.

Michael: [00:44:58] Right. Yeah.

Patrick: [00:45:00] So a step for listeners, if you're stacking that factor up to something like value. Value, the excess returns are much more monotonic.

Michael: [00:45:07] Yeah.

Patrick: [00:45:07] So the cheap stocks outperformed by a little less, but roughly the same amount as the expensive ones underperform. So, if you're thinking about using quality in your process, think about it as a negative screen, like Howard Marks would say that bond investing is a negative art.

Michael: [00:45:22] Exactly.

Patrick: [00:45:22] So, I think that that's an interesting factor.

Michael: [00:45:24] How do you guys do capital allocation stuff, too? So there's now also another it's related to quality, I guess, but a really strong threat, but it's in the new five-factor model Fama French on investments, right? So there's a lot of evidence that companies that grow have rapid asset growth, which may be, I don't know if it's associated with poor quality, tend to underperform in those with more modest, in some cases, even declining asset growth tend to have better returns. What do you think about that one?

Patrick: [00:45:49] Yeah, I think it's a great segue to get into capital allocation. And for people that are interested, one of the best papers is I don't ... I think you actually updated it last year.

Michael: [00:45:59] Yeah, we're going to have another. Hopefully, we'll have another one in the next couple of months.

The State of Capital Allocation

Patrick: [00:46:02] Yeah. It is an all-inclusive look at kind of the state of capital allocation, major changes and trends, things like CapEx, buybacks, acquisitions, et cetera. It's really useful as a background. But to answer your question and then to get your opinion, we think about capital allocation as probably the most neglected tool for analyzing a business, one that can be really valuable. So we find the latest Fama French five-factor model that high increases' percentage increases in capital spending tend to lead to abnormally negative future excess returns and vice versa. So consolidation of the asset base has actually been a good thing, even though it sounds like a bad thing. We also find those particular kinds of buyback programs. And this is where I really want to get your opinion to see if you agree or disagree, have been a pretty phenomenal signal. And by a particular kind, I mean, high conviction done at cheap prices, which is probably how you should always do them. That's certainly not the case. And in aggregate, I don't think there's much to say about buybacks in aggregate because it's kind of like M&A, it follows the market cycle and it sort of peaks, or in dollar terms when markets peak, because there's more excess cash. So the short answer to your question is, yes, we prefer shrinking assets basis over rapidly expanding ones. We like buybacks over a lot of reliance on external financing. All of those things kind of sync with value to some extent, but have been, we think, neglected and very powerful tools for selection, both for selection and for avoiding companies. So how do you think about buybacks? You know, it's become a whipping boy in the press, even for some famous politicians. I think Hillary Clinton's mentioned it. Larry Fink at BlackRock has come out saying that companies should be spending less on buybacks. Where do you fall?

Michael: [00:47:48] Well, we've written a ton. And I'm going to say that I think a lot of the articles I read certainly by journalists tend to be drivel, I guess, would be the word I would pick. I just don't think that they're informed or thoughtful. That said when I think about buybacks and you gave one bucket of it that, obviously doing it with conviction when the stocks are cheap. We like to think about buybacks by categorizing them into three basic buckets. First, I'm going to call the Market Efficiency Bucket, which is, "Hey, we're a large company, we pay a dividend. Typically, we're going to buy back stock methodically, every quarter, every year, some rough amount of our earnings. And sometimes we'll overpay, sometimes we'll underpay, but basically, markets are efficient it'll come out in the wash." Right? And there are some companies that are certainly in that category. The second category is the Intrinsic Value Camp, right? Exactly what you said, which is these are companies that really do have a feel for the value of their own company and tend to make high conviction purchases when it's cheap. They tend to lay off when it's dear and they actually know how to make that distinction. When I was an analyst many years ago, the company that was the poster child for me in that camp was Ralston Purina. And Ralston Purina is, by the way, Bill Stiritz the CEO of the time was one of the subjects of Will Thorndike's book, The Outsiders about Capital Allocation. Stiritz was just a very thoughtful shrewd guy. Basically when Stiritz announced the buyback program, and they tend to be very small by the way, pretended like you didn't have any money. You could almost buy right along with him. Then when he would stop buying it, you could basically sit on your hands. It was just an amazing signal because they really did have a sense. The Intrinsic Value Camp, and as you described, if you can find those companies, I think you want to be right inside or behind it. The third camp I call the Impure Motives Camp, right? Which is buying back stock for reasons that don't have to do with one or two. The big ones are things like offsetting dilution from either option programs, or employee stock option programs, or accretion and earnings, because we're now in a world where you can borrow, especially short-term borrowing at close to zero, and almost no matter what your multiple is, you can buy back stock. It's creative to earnings as creative to ROE. By the way, if your bonuses are tied to earnings growth and ROE, the treasurer becomes the hero because he or she can devise a system to get you there. Those are the people who are not paying attention to economics and that's where I have a problem with it. Right? So as you said, M&A and buybacks tend to be very procyclical. Why? Because companies have... And by the way, the other thing we should say about buybacks, which I think is really a fascinating thing is, I can argue to you with some assumptions that dividends and buybacks are mathematically equivalent, right? In the real world, they're not exactly but they're pretty darn close.

Buybacks and Dividends for the S&P 500 (1999-2015)

Source: Credit Suisse

But it turns out psychologically they're totally different from management. Management thinks of dividends as a quasi-contract. It's sacred once it's been established, you want to raise it if you can. You never want to cut it unless you're really under the gun. But buybacks are considered sort of as residual. We paid all our bills. We made all our investments. We got money sitting around, what do we do with it? Let's buy back stock. They're notwithstanding, they really serve the same purpose. They're in completely different mental categories. So why the buybacks, why the procyclical because that's when companies are doing well, is when they have the money.

I can argue to you with some assumptions that dividends and buybacks are mathematically equivalent, right? In the real world, they're not exactly but they're pretty darn close. But it turns out psychologically they're totally different from management. Management thinks of dividends as a quasi-contract. It's sacred once it's been established, you want to raise it if you can. You never want to cut it unless you're really under the gun. But buybacks are considered sort of as residual. We paid all our bills. We made all our investments. We got money sitting around, what do we do with it? Let's buy back stock.

Patrick: [00:51:12] Right.

Michael: [00:51:13] Same with M&A, right? So it's the same basic idea.

Patrick: [00:51:16] You've perfectly described the difference psychologically between the two. It strikes me as very odd, right? That in many cases the dividends are paid at all because investors can... It's so easy now, right? You can create your own dividend and why not prefer if you don't need the cash to not pay taxes on it now, every time it gets paid out, let it stay in the business earn a little bit more of the business. Can you envision that changing? It seems to make a lot of sense rationally that it would change. What do you think?

Michael: [00:51:44] Yeah, no. I mean this topic has been around for a really long time and there are a lot of people chasing high-yield stocks, whether they're Real Estate Investment Trusts or high-yielding stocks in general, for the yield. As you pointed out, you can make a homemade dividend, which is going to give you the exact same consequence and in fact will be much more tax efficient. Whether people are just not sophisticated enough to do that but if you think about it, that's something a financial advisor or even a robo-advisor, if you said, "Hey, I'm going to own X, Y, Z company. I want to own..." They should be able to set that up for you. Sell a slight amount, becomes a synthetic dividend or homemade dividend, and your tax consequence is much better. I mean, I hope we get there, but it's interesting because I think most people just... And by the way, it's another reaching question. This is another topic for another day, but I read about this and I think this drives me crazy too because of people... When you sit down with your financial advisor, they're almost always talking about Total Shareholder Returns for whether stocks or for the market. The challenge is almost no one owns the Total Shareholder Return. The reason is a TSR assumes 100% reinvestment of dividends with zero tax consequence and almost nobody reinvests 100% of their dividends with zero tax consequence. If you have an automatic reinvestment program in your 401k you got it, you're nailing it. Most of us get a dividend check, people just go spend the money or if you're even a Mutual Fund doesn't own TSRs because Mutual Funds get dividend checks and they just go to their cash balance. It's this really interesting idea that we'd run around with these ideas called Total Shared Return but in fact, the people that actually earn that return are tiny. Taxes alone exclude most people from doing it but even in tax-free accounts, most people don't.

Patrick: [00:53:31] What I find amazing is that given all that income yield there is such a powerful motivator. Now you've got a scenario where we like dividend yield as a factor, mostly because it was a value factor, but in many cases, it ceased to be because now high-yielding stocks, the correlation between the stocks' dividend yield and other measures of cheapness has basically gone to zero because they'd been bid up to many cases 20 times earnings. This is low growth, stodgy businesses, trading at dear multiples because people like that income. I think for investors out there, income if you need it if you really do want those dividends can still be a good thing, but you need to be mindful of the price you're paying for that. The market as a whole hasn't seemed to be mindful it seems to be this kind of one-directional trade where, and another example, yielding stocks are doing phenomenally well this year.

Michael: [00:54:19] One of the things I want to mention, Patrick, on the buybacks which are... I always say to people, active investors, "If you own the shares of a company buying back stock, doing nothing is doing something."

Patrick: [00:54:30] Right.

Michael: [00:54:30] That something is increasing your percentage ownership in the company. That's the other thing to your argument about, can we create these homemade dividends? In fact, if you just sold the prorated amount, you would maintain the exact same percentage ownership in the company, and have cash in your pocket, and be more tax efficient. All three things simultaneously. It's doable, but you have to be alert obviously to be able to do that.

What's your Edge and is it Sustainable?

Patrick: [00:54:55] That makes me think of a chapter, I think the first chapter in 'Outsiders' on Henry Singleton, where I think from peak to trough he bought back 90% of his shares so your "Do nothing," meant he went from being one owner to one that owns a lot more of that business. Now for them, it worked out fantastically well, Teledyne did really well. But I think a fascinating topic, so I appreciate the opinions on it. As we run down this line of inquiry on what active managers should care about these perception/reality gaps, looking for low expectations, versus the real fundamental story. I've heard it said that the edge that remains, and this is the question certainly as a portfolio manager myself, when dealing with prospective investors or clients, this is the question of the day. "What is your edge and how is it sustainable?" You hear that there are a couple of different kinds of edges. There's an analytical one, an informational one, and a behavioral one. Those are the kind of the big three camps, I might add organizational and client alpha that we talked a lot about earlier. That having a business set up the right way and the right clients can be a huge advantage as well. But taking those first three categories, if you were an active investor and you had to choose, or I get to bestow you with a keen advantage in one of those three categories, which do you think is the most important of those categories?

Michael: [00:56:08] I think, part of it is, I have my own personal inclinations on this, but that may not be relevant for all people. By the way, I hear stories and I meet with, for example, certain Hedge Funds that are doing extraordinary things in the information-gathering business. Huge budgets on data. In fact, my oldest son studied computer science and he's done a lot of machine learning. I asked him recently, "What are the biggest barriers to really improving these models?" And he said, "It's less about the analytics and much more about the quality of the data." I believe that there is some opportunity with information but for me, I do think it's analytical/behavioral. In the book on Skill and Luck, the way I rephrased it is, skill is an analytical set of skills and you point out the edge and its portfolio construction, it is managing and mitigating the behavioral issues, and hopefully taking advantage of those when they work for you. But I also mentioned the organizational one, I think you're absolutely right. And you think about quality decision-making is probably parked dispositions a little bit it's who you are. Part of your training, what you studied, and part of its organization. I don't know exactly what the allocation is, but those are probably the three big buckets. So to me, when you think about where the pockets of inefficiency, usually two or three things I like to think about. One is as an institution, are there places where you can compete against mom-and-pop individuals? In public markets in the US that tends to be a come and go. Obviously, the last time we had a huge wave of individuals was the late 1990s and that created huge opportunities for institutions. The second is, are there forced sellers? And it's really interesting that there are people that are doing things for non-fundamental reasons. Now, it's intriguing to bring in the indexing idea here, but usually, Fall Selling is things like margin calls so people have to do things they don't want to do could be banks having to reduce their risk-rated assets but the classic one is spin-offs.

Patrick: [00:58:20] Sure.

Michael: [00:58:20] Right? It's amazing how persistent spinoffs have been, notwithstanding this literature has been around for a really long time. And then the last one, I call them diversity breakdowns. I think it's more of a behavioral thing, which is what we know ever since markets have existed. And by the way, it's not just marketing it's in society in general, that we tend to correlate our behaviors in meaningful ways from time to time. So for markets to be efficient, one of the key underlying conditions is, we call it investor heterogeneity differences. Different points of view, different strategies, different time horizons, and so forth. And that tends to contribute to market efficiency. If we correlate our behaviors, whether we all just adopt the same decision rules, or if you a contrarian investor, sit on your hands, you don't participate. So in effect, you pull yourself out of the game, we get these diversity breakdowns and we'll go back to Seth Klarman. He's got this line which I absolutely love. He says, "Value investing is at its core, the marriage of a contrarian streak and a calculator." And as I try to unpack that the contrarian streak says, "Hey, if everybody's bullish on this, I'm going to consider the other side of it." Or "If everybody's bearish, I'm going to consider the other side of it." Right. So at least weighing the other side of the story. But note that it's not just the contrarian streak, because being a contrarian for the sake of contrarian, not a good idea because if the consensus is right. It's these calculators the second piece where, because everybody believes the same thing that has led to this mismatch between fundamentals and expectations. And I don't know that it's the case, but you mentioned this thing on here are these high dividend stocks trading at 20 times earnings because everybody wants to yield. That would be the case where I'd say, let me examine this carefully. Everybody wants to own this kind of stock. We know why they want to own this kind of stock. So their decision rules are collapsing to one. Now let me pull up the calculator and see do those valuations now make sense in an absence of just a pure yield. Right. I love that Klarman framework. So it would be a behavioral/analytical combination perhaps, but at the core investing is a social activity. It always has been, and it always will be. And so as a consequence as long as there are humans around, even if they're programming machines-

Patrick: [01:00:39] The breakdowns will happen.

Michael: [01:00:39] We're going to have a little bit of those things.

Patrick: [01:00:41] Yeah. You see it too in the dollar versus the time-weighted rate of return of even index investors, right. Where typically the investor return is worse sometimes drastically so. In ETFs, it's way worse. Maybe because they're easier to trade and people are treating ETFs like they used to treat stocks. Maybe that's the new edge to exploit if you think in terms of mom-and-pop. But a pretty fascinating topic that maybe what you want is consensus good or bad because if everyone agrees, there's probably an opportunity there. Whether everyone agrees the outlook is horrible or everyone agrees the outlook is fantastic. Maybe that's when you want to be a contrarian.

Michael: [01:01:19] I do think you need to take the next step though because as I always like to say if the movie house is on fire by all means run out the door. Right. So in other words, there are cases where positive feedback, which I mean is in a technical sense, moves the system away from a certain outcome, which is good. Right. But like you said, more often than not certainly that's the pond you want to fish in, right?

Patrick: [01:01:39] Yeah.

Michael: [01:01:39] And there will be opportunities.

Patrick: [01:01:41] We've talked a lot about buying. What about selling? Is it a similar skillset? Do you think that it's just, a relative assessment of your opportunities where you own one gap perception, reality gap, and it's narrowed and now there's a wider one? Is it that simple? Do you think it's a different skillset?

Michael: [01:01:59] Yeah. I mean it's hard. I think it probably is a different skillset, and we read about this in Expectations Investing. And I think there's not much I could add to what we wrote, that there usually three reasons that you would sell a security. The first is you screwed up, you made a mistake. And so you had some sense of a fundamental outlook, which has not proven to be accurate and as a consequence, the thesis is no longer true. And by the way, that's a really hard one because we all, especially, an active, non-quantitative manager is you fall for thesis creep, right? In other words, typically the stock has not done well, so it looks cheaper. And so you're really hesitant to make a decision. So one is just, you got it wrong. The second is, you got a right. So you have some sense of what the expectations should be and how the financials were unfolded and those things become aligned and so that's the good reason to sell something. And it's an interesting one even if you go back to Walter Schloss, who was one of the great value investors. He would often say, "Don't be so quick to sell things even when they get to what you think is fair." Because he's like, "Just take your time." Right. Which is really interesting. So things can probably exceed what you think they're worth. And so you'd probably don't need to be in a big rush, but that's sort of the victory lap. That's a good thing. And the third is exactly what you said, which is every day you have an opportunity set and you should be trying to figure out where the best opportunities are given what's going on. So there may be a case where you think stock A is attractive, but now you identify stock B, which looks even more attractive. So you sell A for B just to improve your relative position. So those to me would be the three big categories or probably subcategories of those things. That's how I would probably think about it. The hardest one is probably the first one because you have to admit that you're wrong, which most of us don't want to do.

Portfolio Construction

Patrick: [01:03:46] So the last piece of the puzzle is portfolio construction. I know you've written a lot about these things like the Kelly criterion, mean-variance. What do you think is the state of portfolio construction skills amongst active managers? And if you had to choose a method for how you weighed a portfolio, once you've made your selections, what would it be?

Michael: [01:04:06] I'd like to do much more work on this and I'm a little bit frustrated in part because of my lack of technical abilities in this, but let me just mention one thing I found fascinating. We wrote a piece not too long ago called Form Files Function. It was about how your organization should be structured in order to serve your edge. And one of the academic papers in there that I found intriguing was a study of about 14 Mutual Fund Families that they looked about 65, 70 funds run by analysts within the Fund Family. So I'm making this up but it might be to the T. Rowe Price analysts fund versus an appropriate T. Rowe PM run fund. Okay. That's the basic idea. And what they found was that the analyst-run funds did better than the PM run funds.

Patrick: [01:04:51] Wow.

Michael: [01:04:51] In the same family. Yeah, that's kind of interesting. So what's going on there? So a couple of things you might point to, one is most analysts-run funds are sector constrained, right? So in other words, they may not be exactly sector-neutral, but they're going to be close to sector-neutral. So it's pure stock picking. And going back to our discussion, we had talked a little bit about offline, about active share. Active shares, either you're weighting stock differently, or you're doing something outside the index, but there's a good example. These guys are all stock-picking they're not messing with sectors. By the way, PM's tend to not listen to their analysts as they get older, but that's separate. The one that was not explicit, which I thought was interesting is to your point was what about portfolio construction in general. So that may be a related sector thing. So that's intriguing. Okay. So going back to your specific question, I think the way to approach this is to encourage portfolio managers, to answer a series of questions, because what you should do for portfolio construction is really a function of what kind of edge you think you have. Do you think you're going to have lots of lots of little, little opportunities? Do you think you're going to have a few ginormous opportunities? The second is going to be, what are you trying to do? Are you trying to maximize your wealth in any period? So you want to have as much money as you can possibly have by December 31st, 2016, or do you anticipate parlaying your wealth? So your bankroll on December 31st becomes your bankroll on January 1st and so on and so forth. That leads to a separate set of questions. And third is, what are your constraints? Do you have drawdown constraints, sector constraints, leverage constraints, trading constraints, so forth, right? And I think if you answer those series of questions that will lead to a portfolio construction approach, that makes sense for what you're trying to do. Now in our industry, and tell me if you have a view or sense that it's different than this, I think most portfolio managers and Mutual Funds and even to some degree, Hedge Funds probably have a feel for what they're doing with portfolio construction, and then have a risk manager look over their shoulder and say, "Hey, Patrick, I don't know about this or that." Or "You might want to think of that construction."

Patrick: [01:06:57] Borrow might have more influence-

Michael: [01:06:58] Yeah borrow might have -

Patrick: [01:07:00] - on people's portfolio construction than the...

Michael: [01:07:03] Yeah. Right. Exactly. And there's something for that, but you really haven't answered the questions as to whether that's the approach. So I have to say, and you've sort of alluded to it that one of the things I believe or have a sense is underutilized is a Kelly criterion. Right. So Kelly...

Patrick: [01:07:18] Can you describe it a little bit?

Michael: [01:07:19] Yeah. So Kelly was a physical is at Bell Labs back in the 1950s. And it was interesting and basic problem gambling. And he was talking to Claude Shannon, who is a father of information theory. And Shannon said, "Well, there's an interesting application of information theory in gambling." And so the Kelly criteria basically says that the more information you have... So think about it this way, that information is in a sense, a move away from entropy so it's something different than randomness, right? So if you say markets are perfectly efficient, there is no information in markets, right. But if there is some edge there's information, right? And so the idea of a Kelly system is that your bet size should be proportionate to some degree to the amount of information that you have. So Kelly formalized this in the 1950s, he died tragically, he was quite young. So there's this idea, but the basic idea is called Geometric Mean Maximization. So you're looking at geometric means, not arithmetic means, but geometric means over time. Okay. So probably the most famous user of this is Ed Thorp. And Ed is famous for writing the book, Beat the Dealer about card counting back in the 1960s. And what's interesting about that is that Thorp not only wrote about card counting techniques, which are quite novel at the time but also describe the Kelly system for betting. So the point was if you know your bankroll and you know your edge, you know precisely how much you should bet on a particular hand, which is super cool, right?

Patrick: [01:08:38] Yeah.

Michael: [01:08:38] So you're maximizing your return, giving your information, your edge. And so Thorp went on to found Princeton Newport Partners, eventually his own money management firm, and delivered unbelievable numbers. Now it turns out that a pure Kelly formula leads to pretty wicked drawdowns.

Patrick: [01:08:53] Crazy debts.

Michael: [01:08:54] Right? So in other words, in drawdown. So it gives you a higher probability of higher terminal wealth at some point, but you don't want to be on that roller coaster. Right? So you can do things like partial Kelly's, you can tone it down a little bit. But that to me is something I'd like to do more work on, and there some guys doing good winning out there, but to create a series of important questions to ask a portfolio manager about what he or she is trying to do and what his or her constraints are. And then once you've answered those questions honestly, and correctly, it becomes an algorithm. I mean, becomes math right at the end. And the other thing I was talking to one of our quant guys the other day about this it turns out that because of covariances sometimes PM loves stock A more than stock B, but the model says you should own more of B than A, and it doesn't make sense. But that's the other thing is you get yourself out of an individual stock mindset into a portfolio structure, and that's another different dimension. So the way I would envision doing this is ultimately having an algorithm say, "Patrick, given what you said, your inputs here's what your portfolio should look like." And I'm not sure you should necessarily adhere to it perfectly, but it's really nice to know what that portfolio would look like and why you're different than what that portfolio is.

Patrick: [01:10:12] Yeah. We mentioned borrow, which is one way of measuring risk exposures to different kinds of things like say value investing. There are ones that could measure exposure to something like the price of oil and what you find I think amongst traditional fundamental investors is that they like stock X. They like it this much more than stock Y and building a portfolio that way may maximize return over some long period, but has unintentional bets baked into it. So you may say, you're exposed to the price of oil. And they say, "Well, I don't have an opinion on the price of oil. So I don't want that exposure." That seems to be an increasingly key part of how sophisticated managers are building their portfolios, even if they're a pure stock picker that using some of these risk-based tools for portfolio construction is an interesting application. Now, whether or not that's a good thing, or whether it's deadening some of their edges is an open question, but that's how we think about it is what do we want and then to get that, can we remove, let's call them unintentional bets or risk exposures. Can I get the same thing I want in our case, a portfolio with great valuations, with really good capital allocations behind them, things like that? Can I get that without taking on a massive bet on this or that?

Michael: [01:11:26] And by the way, I think a lot of the multi-strat guys do a lot of that, and as you know they're pretty sophisticated at that, so I agree with all that. And I think the other big challenge in all of this stuff is the notion of non-stationarity which is a fancy way of saying that these relationships will change.

Patrick: [01:11:41] Covariance has changed.

Michael: [01:11:42] Yeah. So that's the offset is you also want to have some degree of humility or measuredness about how much you can actually really specify, but you're definitely right. I mean, being aware of things that you're not aware of in your portfolio or even exposure of a particular stock that's just an incredibly important thing and useful thing.

Patrick: [01:12:00] So last question on the role of active management...

Michael: [01:12:00] ... important and useful thing. Right?

Free Chess

Patrick: [01:12:01] So, the last question on the kind of world of active management before we get to a couple of fun closing questions, I'm curious what your general opinion is on let's say factor or quantitative investing. Obviously, I'm biased. This is what I do, but in terms of its effectiveness, its potential for the future, but also whether or not you think that these factors which are cheaper and cheaper in a broad sense are taking away some of what used to be called skill from other active managers.

Michael: [01:12:32] Yeah. I mean, it's an incredibly interesting question and you go back to the work by Paul Meehl in the 1950s was demonstrating that it's often the case that algorithms can help human decisions and decision-making, and there've been meta-studies demonstrating that we'll call them more quantitative methods or more algorithmic methods, very rarely do worse than humans. In the worst cases, they tie and they often do better than humans, right? So, there's not in our field, but in other fields, there's a lot of evidence that this way of thinking about the world works. The trick, a couple of tricks about the factors and you're much more intimate with this than I am is, I mean, sort of these open questions about, are we capturing a risk component or is there sort of mispricing? And that's an interesting angle to consider. The second is some of the episodic nature of some of these things. Bands wrote his famous paper about small-cap outperforming large-cap in 1981. And he said, "I'm going all-in on that in 1981." For 20 years you were wandering around the desert, right? So there's some of that as well. So to me, I mean the question would be, if you're going to call it a qualitative or a fundamental investor, can you blend some of the quantitative techniques with these other qualitative techniques? And so this is almost like this hybrid. And so to me, as you can tell, I'm very sympathetic to quantitative methods. There are a lot of things that quants can do much more effectively. They can look at many more situations, they can test things much more rigorously as you point out. Sometimes they'll generate counterintuitive answers that at least are worth contemplating carefully. That said, also these factors or even quantitative techniques also reflect the choices of the researchers, the biases of the researchers. And there's going to be components of, there are going to be episodes where they work or don't work. So yeah, I think, I mean, I would think that they can be combined. And by the way, the last thing I wrote a little piece about this and tell me if you're a sense of this is different, but there was a really interesting work done by the CFA Institute. And they talked to both the quant community and the sort of we'll call it fundamental community and these were like totally different camps. And in fact, they could barely talk to each other, right? So the quant guys were like, "Gee, I have a $10,000 computer. It could be better than all these analysts over here." And then the analysts were like, "These quants don't understand the subtleties of my industry." And they're just, they're sort of bulldozing with these factors that don't understand the nuances. Look, their truth is somewhere in between, and if that's the case, it'd be nice to see some sort of reconciliation.

There are a lot of things that quants can do much more effectively. They can look at many more situations, they can test things much more rigorously as you point out. Sometimes they'll generate counterintuitive answers that at least are worth contemplating carefully. That said, also these factors or even quantitative techniques also reflect the choices of the researchers, the biases of the researchers. And there's going to be episodes where they work or don't work.

Patrick: [01:15:20] What I find amazing is that, as you say, it's in the middle somewhere and I see a lot. We use the PE ratio, right? It's one measure of cheapness in a suite that works. What's fascinating to me is I would think that PE ratios lose a ton of nuance and they do. They have to. We talked about it earlier with earnings. You talked about it in your book, but when you do a sort of a cumulative or annualized excess return of just a simple basket of cheap stocks buy PE versus a simple basket of cheap stocks by something that a lot of fundamental guys like more, and I personally like more, like a free cash flow to enterprise value, something more nuanced that's capturing more of what's true about the business. The annualized excess return is very close. It's basis points difference over a long period. Now there can be big swings from time to time, but I think that's pretty amazing that maybe it's just that you want low expectations and it doesn't really matter how you measure it, even though I think intuitively free cashflow makes a lot more sense.

Michael: [01:16:14] Yeah, I agree with that.

Patrick: [01:16:15] I'm curious if, you wrote about maybe two years ago, a paper about free chess and you mentioned hybrid, and that's basically what free chess is, which is yeah, now we've got computers that can beat the best chess players or go players or whatever, but if you combine humans with sophisticated whatever tools they want to use technology tools, that's actually the best. So two questions. One is, what do you think that the humans are bringing to the table in that equation that allows them to win? And second, is, do you think that this free chess analogy is fair or applies to investing as something people should pursue?

Michael: [01:16:51] Yeah, it's great. They're both important questions. The first one is what do humans bring to that equation? To the best of my understanding. And I'm not a chess player, is that if a human deeply understands what's going on in the game and they understand how the program works, there can be junctures in the game where the human sees a move that's better than what the computer is proposing. And one of the interesting things about that is there was a freestyle tournament 10 years ago. And the two guys that won were two guys in their twenties from New Hampshire and sort of this like the middle of nowhere guys. And they looked at their chess ratings or their ELO ratings, basically how good they were. And they were very, they were good, but they're not great chess players. So the skills are being great at chess, the skill was knowing the programs intimately and knowing what was going on in the game, and knowing when to override. Right? So that's, I think is the answer to that. And by the way, I had a really very special opportunity to visit a bit with Gary Kasparov in February this year-

Patrick: [01:17:52] Wow. Neat.

Michael: [01:17:52] -and I asked him this question and he felt that he suggested, he thought these freestyle teams would continue to be better than the programs. Now I think the margin is not great, but it still appears to be the case. So, that's interesting, okay. So does it apply to investing? And I think that look, chess is a complete information game. It's a closed system, predictable movement. Computers can do a lot of computation, right? So markets are different in all those dimensions, right? So that's true. That said, and it goes back to my comment a moment ago is that we do know that in almost any field, whether it's interviewing techniques or making clinical decisions as a physician, that introducing algorithms or quantitative methods improves people's choices in decision-making processes. So I do, I mean, I wouldn't go crazy with this freestyle analogy, but I do think that the melding of some quantitative methods with some fundamental methods if it can be achieved would be a really fertile area for exploration. So we talked about, is it behavioral is analytical? And if you think about it, your quantitative methods help you across all those things, right? They help you behaviorally because you're going to want to do what you shouldn't do. You mentioned time-weighted versus dollar-weighted returns, and it's going to help you analytically, if nothing else to be more efficient, to look at more companies, more factors, more things that may be opportunities.

...we do know that in almost any field, whether it's interviewing techniques or making clinical decisions as a physician, that introducing algorithms or quantitative methods improves people's choices in decision-making processes. I wouldn't go crazy with this freestyle analogy, but I do think that the melding of some quantitative methods with some fundamental methods if it can be achieved would be a really fertile area for exploration.

Patrick: [01:19:21] It seems like another way of asking, what is the role or value of intuition? And intuition is in a bad spot these days. People make fun of it as silly and thinking with your gut. People think that that's a bad way of making decisions. We kind of live in the age of the algorithm, but I do wonder if someone with significant experience at dealing with market situations, we're just information processing machines ourselves, probably more powerful than we give ourselves credit for whether or not the free chess thing works and could continue to work in market because Soros gets the back pain or whatever it is, there is something that intuition is nothing more than a recognition I think of a pattern, right, that you've seen before and that you see again. So maybe there is a role and for me, this is way out of bounds because there's no gut or intuition in what I do.

Michael: [01:20:11] I have a thought on this because I mean, I wrote about this in my book called Think Twice and I have a fairly strong view on this, which is that intuition does exist by the way. And I think pattern recognition is a good way to describe it. I would use the Danny Kahneman formulation of system one, system two writes, "This one being your experiential system, which is fast and automatic and very difficult to train and system two being your analytical system, which is slow, purposeful, deliberate, and so forth." And I think that intuition applies when you've trained your system one in a certain way, to be able to see patterns and chess, going back to chess is a great example of that. So we know you can show a chess master a board, and he or she can pick out very quickly who has the advantage and what the best moves are, and so forth. They're very good at that because they've internalized it through all this practice. But if you put yourself into an unstable and non-linear environment, those tools all go away. And so I know that investors talk about pattern recognition and there may be certain patterns like corporate performance and so forth that may be a priori identifiable, but Greg Northcraft, psychologist. Got this line I love. He says, "There's a really important distinction between experience and expertise and experience means you've been doing something for a long time. Expertise means you have a predictive model that works." And we confuse those two things all the time. So people go, "Oh, how'd you get that stock right?" "Oh, I saw the pattern." Right? Well, the key is, are you rigorously measuring all of your intuitions? And I suspect if we did that in a sort of rigorous way, we would find that intuition really in investing there may be subsets of investing where it works, but in the entirety, I'd be very skeptical about its power personally.

Patrick: [01:21:58] So two final investing questions, and then just one or two more about some funds takeaways for listeners out there. So the first, and these are kind of different flavors of the same idea. Let's imagine that you have one goal and that's it in life, which is to earn the best return you can over, let's say the next 20 years. Investing on the long side only, only in public equity markets, so you're just a vanilla stock investor. You have two options. The first is you get to be the PM and you get to pick three analysts. And one of them could be Seth Klarman, whoever you want in the world and you're going to be making active decisions. The second option is you buy Vanguard, the total market index. Which of those do you go with?

Michael: [01:22:41] Yeah, so if I were really committed to it and I really want to do that, I would go for the first one, believe it or not. And I think that, and that may be a false belief that I could pull this off. But I do think that with the proper, and I mentioned before, I mean, I think with people with a proper disposition and proper training, and if we created an environment that we could probably do reasonably well. Maybe not crush the market, but do something that was certainly comparable, and especially if we could go anywhere, that would be helpful. So now that said, and I just want to be clear that most of my personal invested money is indexed because I don't have the time and there are conflict possibilities, but I don't have a lot of time to do it. And I do, so I do think for most people, it's a very reasonable thing to do.

Patrick: [01:23:27] So you may have already answered it by saying most of your money, your personal money is indexed, but so the first question was really aimed more at prospective or current practitioners out there, managers themselves. Another way of framing it would be in a way that might be more useful for allocators, which would forget you doing it yourself with three analysts. Let's say that you got to interview 50 managers, which allocators get to interview whoever they want because they have the money to invest to fuel businesses. If you had, let's say, 50 managers that you could choose or be randomly selected, you don't know ahead of time which ones you're going to like best, but in the end, you have to invest in one or more of them. Again, the second option being Vanguard. Is it the same answer, or would you go with Vanguard then?

Michael: [01:24:07] I might go with the same answer, which would be, try to pick those. But I got to say that a couple of things. You relegated me to equities and long-only and so forth, but to me and I got this from David Swensen in the Yale Endowment, I mean, he talks a lot about looking at differential performance across asset classes, and basically argues that when there is a substantial difference across, there's a lot of divergences, a lot of variants that's where there's an opportunity to express skill. And so certain things like U.S. Treasury bond portfolios. It's just really hard. It's just the nature of what you're doing to distinguish yourself. And then, so those kinds of things would lend themselves to passive, but there may be certain asset classes that would lend themselves to a better performance where skill can really express itself, so that would be the first thing I would look for. The second thing, and we talked briefly about this offline, but this idea that I would probably want people to have to try to do something a little bit different than the benchmark, right? So if I'm going to own, for example, an S&P 500 Vanguard index fund, if I want someone to try to beat that, I want their active share and I want them to be somewhat different than that portfolio in a thoughtful fashion to give me some shot. Right. So, and the other thing is, of course, the fees come into play and all this stuff, but yeah, I think that there are some characteristics of fund managers that do well that we can identify, doesn't assure you're going to do it, but probably skews the probabilities in your favor.

Most Satisfying Aspects of Michael's Career

Patrick: [01:25:32] Yeah. Well, great. So now just a couple of closing questions, one looking back one looking forward. What aspects of your career or life have been the most satisfying in hindsight? You just wrote the paper about your 30 year anniversary in the business. Are there one or two things that really stand out as great memories?

Michael: [01:25:52] Professionally you mean. Yeah, because like I say, the first thing is, first and foremost and probably when life draws to an end is what you think about is just family stuff. Right?

Patrick: [01:26:02] Yeah. You mentioned five kids.

Michael: [01:26:02] Yeah, five kids and having a great wife, great partner, and very satisfying so that's really, at the end of the day, that's the most important thing, but professionally I've been incredibly blessed to be around. I've had just a few people that have been really pivotal and constructive that allowed me to go off and do a lot of things that I've loved to do. So, Al Rappaport, I mentioned a number of times, and Al and I still collaborate on work and he's great and has been an incredible source of inspiration and learning and teaching. So I'm still his student, so I ask him questions all the time, which has been great. Bill Miller, who I worked with for nine years at Legg Mason Capital Management, Bill is another guy who is an extraordinary reader, incredible intellect, a very thoughtful guy, and learned a lot from him in many different ways. So that's another guy. And then the guy who actually, the reason I rejoined Credit Suisse was the CEO at the time Brady Dougan, who was a guy, another very supportive guy over the years. And so if you said to me though at the end day, "What is the most gratifying?" It is this process I mentioned before that having the opportunity every day to input and output, right? So every day to learn something, to try to be a little bit smarter, a little bit more thoughtful about the world, and then to be able to share those things in a way that hopefully is accessible for others, right? So hopefully making the world just a little bit better place to synthesizing some of this work. That I would say. And I have to mention, I've had now a 20-year affiliation with the Santa Fe Institute in Santa Fe, New Mexico. It's a non-degree conferring institute.

Patrick: [01:27:34] Multi-disciplinary.

Michael: [01:27:35] Multi. Across beyond disciplinary boundaries and the unifying themes of study of complex systems and understanding the world and that from that lens and just an incredible group of people, researchers, open-minded rigorous been a great source of different thoughts and great people. So that's been another, just incredibly valuable component to my intellectual life.

Patrick: [01:27:59] Very neat. How about the future? What areas are you most excited about? What has you getting up early to go keep exploring?

Michael: [01:28:07] Yeah, I mean there's, I just think, I tell my students at the beginning and at the end of the semester, in some ways life has never been more exciting in the world of investing because we really don't understand a lot of stuff, notwithstanding all the stuff we've done. Right? I don't know that we really understand what risk is that well. I don't think we really understand market efficiency all that well. How do we think about a company's competitive position? There are frameworks, but all these things are work in progress. And so to the degree, to which we can help chip away at some of those thoughts, that to me is very exciting. What's immediately on the plate is we've been doing a lot of work integrating base rates, so this idea that forecasting benefits from the not only sort of unique information about what's going on, but also for integrating past experiences. So we have a huge piece called the base rate book, where we look at corporate performance back to 1950, which we really think can help guide people's thinking. We're updating the capital allocation piece. You mentioned that. That's also really cool work.

Patrick: [01:29:04] Yeah, it's great.

Michael: [01:29:04] A look at how the top thousand companies in the U.S. have spent all their dollars in the last 35 years, look at the trends, drawn a lot of academic research to show what's better, what's worse and so forth. That's a lot of fun. And you mentioned this, we mentioned this in some detail, but I keep coming back to is portfolio construction. If you said, "If outcomes are a function of your edge and how you bet on that edge," I think that we spend as an industry a lot of time thinking about the edge, much less time on portfolio construction. So that's another area that we're really excited about. So it never ends, I always think to myself, I'm going to run out of stuff to talk about, but my list continues to be long and more ideas than time. That's the problem.

Patrick: [01:29:46] So we started with your personal system and wound there too. I'm curious and I'll frame, everyone asks the favorite book question, so I'm going to frame it a little bit differently, which is again, bestow you with a power, which is that you can force everybody out there to read and absorb, really absorb just one book. What would that book be?

Michael: [01:30:03] Yeah, that's a tricky one. I think I would go with Danny Kahneman's book Thinking, Fast and Slow, and the reason I am, and maybe it's just a personal taste, but I am very fascinated by our decision-making processes as people. And I think Kahneman and certainly his collaborator Tversky have done as much or as is more than anybody in the last 50 years to contribute to our understanding of where we do things well and where we don't do things well. And I also think it tends to be a gap in our curriculum. You think about young people, whether in high school or in college and certainly post-college, many of them have not taken a formal class in decision-making. Now as a philosophy major, you probably studied logic and so you understand some of these principles and sort of resonates with you more, but it is remarkable that we throw people into the world without thtenem understanding or having learned about some of these principles of decision-making. And so it's obviously professionally super helpful, right? But it's also, I think extraordinary just for life, all the decisions you face, whether they're personal decisions or personal finance decisions and so forth. So I think that would be the book I would go for. There's, I mean, there are many, many more-

...it is remarkable that we throw people into the world without them understanding or having learned about some of these principles of decision-making.

Patrick: [01:31:13] Yeah, I know. A whole catalog.

Michael: [01:31:15] -but I'll leave it at that.

Patrick: [01:31:16] And then the last two you mentioned, sleep, obviously, I think it's pretty straightforward there, but also diet and exercise. I think nutrition and exercise are probably areas that we know very little about that are interesting areas of research for those in other fields. So I'm curious, kind of your take. What's your routine, what are your kind of prescriptions for how you eat, and how you think about moving?

Michael: [01:31:40] Yeah, I'm reading Daniel Lieberman's book right now called The History of the Human Body. I don't know if you know that book-

Patrick: [01:31:44] I don't know that one.

Michael: [01:31:44] -but it's sort of evolutionary history of humans and it's absolutely, it's a fascinating read because it's tracing the human body over let's call it a couple of million years, so where we've come from and so forth. Yeah. And look, I don't think there are any prescriptions that I have that are not for my common sense. I was an athlete growing up. I always enjoyed playing sports. I still do play sports. So for me, getting out and exercising is something that I've always enjoyed doing. It's always been part of my routine, so it's not a big onus for me. I think for some people it's more difficult. And then, I think diet is there's just, for me, it's mostly common sense stuff. I mean obviously, lay off all the bad stuff, and that's one of the points that Lieberman makes in this book is that we've obviously evolved to desire certain things like salty foods, sugary foods, fatty foods, and so forth. And there is a huge evolutionary advantage to finding those things and consuming them for our forebearers. But the food industry has figured out a way to deliver those things in bulk with no nutrition and cheap-

Patrick: [01:32:54] Supernormal stimuli I call them, yeah.

Michael: [01:32:57] Supernormal stimuli. So you just got to figure out how to do everything in a good measure, I guess.

Patrick: [01:33:01] Well this has been an absolute blast. I really appreciate all the time. And hopefully, everyone out there finds it useful for making some decisions about what they're going to do, whether it's with their career in the business or how they're going to allocate their money. So I really appreciate the pretty long interview. I appreciate your time.

Michael: [01:33:16] Thank you, Patrick. It was awesome.

Created by @ekmokaya