Invest Like the Best with Patrick O'Shaughnessy - Cliff Asness – The Past, The Present & Future of Quant [Invest Like the Best, EP.111]

Episode Date: November 13, 2018

My guest this week is Cliff Asness, the managing and founding principal at AQR Capital Management. 20 years after its founding in 1998, AQR manages $226 Billion dollars across a number of quantitatively based investing strategies. Cliff was an original quant researcher and he has long been one of the financial writers and thinkers that I look to for education and for inspiration. I distinctly remember reading one paper in particular—value and momentum everywhere—somewhat early in my career and thinking: this is the kind of research I want to do forever. You can always tell when talking to Cliff or hearing him speak that he just loves researching markets. There is a deep intellectual honesty in his work, and a respect for thinkers at different ends of the market spectrum, from Gene Fama and Ken French, to Jack Bogle, to Dick Thaler and Robert Shiller. Our conversation is about all things quant—past, present, and future. Cliff touches on many of the big issues facing quant investing and tells some great strong along the way. I hope you enjoy our discussion. Let’s dive in. For more episodes go to InvestorFieldGuide.com/podcast. Sign up for the book club, where you’ll get a full investor curriculum and then 3-4 suggestions every month at InvestorFieldGuide.com/bookclub. Follow Patrick on Twitter at @patrick_oshag   Show Notes 1:47 - (First Question) – Favorite superhero 2:43 – Why ‘Ka nama kaa lajerama’ is part of his twitter profile. 3:38 – How portfolios have shifted the way they use factors in a portfolio 10:15 – What are good questions clients are asking right now             13:24 – Contrarian Factor Timing Is Deceptively Difficult 15:40 – Does technology impact investing strategy 22:14 – When to share information vs keep it proprietary for clients sake 26:40 – How their research process is governed 31:14 – How they will incorporate machine learning into their process 34:21 – What they will do when red flags show up 37:01 – Wackiest question from a client 41:47 – The Three Sharpe Ratio Strategy             41:53 – Liquid Alt Ragnarök 48:10 – Does his thinking change when it comes to asset allocation vs portfolio building             50:17 – Parallels Between the Cross-Sectional Predictability of Stock and Country Returns             53:01 – Sin a Little 57:14 – Trends in fees and pricing 1:02:43 – Thoughts on private equity markets 1:11:03 – Common attributes of really good researchers 1:13:21 – What is he most curious about right now 1:15:43 – What excites him outside of finance 1:17:00 – How much he discusses his work with his kids             1:18:35 – The Devil in HML’s details 1:19:36 – Kindest thing anyone has done for him   Learn More For more episodes go to InvestorFieldGuide.com/podcast.  Sign up for the book club, where you’ll get a full investor curriculum and then 3-4 suggestions every month at InvestorFieldGuide.com/bookclub Follow Patrick on twitter at @patrick_oshag

Hello and welcome, everyone. I'm Patrick O'Shaughnessy, and this is Invest like the Best. This show is an open-ended exploration of markets, ideas, methods, stories, and of strategies that will help you better invest both your time and your money. You can learn more and stay up to date at investorfieldguide.com. Patrick O'Shaughnessy is the CEO of O'Shaughnessy asset management. All opinions expressed by Patrick and podcast guests are solely their own opinions and do not reflect the opinion of O'Shaunacy Ashton asset management. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of O'Shaughnessy Asset Management may maintain positions and the securities discussed in this podcast. My guest this week is Cliff Asniss, the managing

and founding principle at AQR Capital Management. 20 years after its founding in 1998, AQR manages $226 billion across a number of quantitatively based investing strategies. Cliff was an original quant researcher, and he has long been one of the financial writers and thinkers that I look to for education and for inspiration. I distinctly remember reading one paper in particular, value and momentum everywhere, somewhat early in my career and thinking, this is the kind of research I want to do forever. You can always tell when talking to Cliff or hearing him speak that he just loves researching markets. There is a deep intellectual honesty in his work and a respect for thinkers at different ends of the market spectrum, from Gene Fama to Ken French to Jack Bogle to Dick Thaler and Robert Schiller.

Our conversation is about all things quant, past, present, and future. Cliff touches on many of the big issues facing quant investing today and tells some great stories along the way. I hope you enjoy our conversation. Let's dive in. I always like to start these things very seriously, so we'll start with your favorite superhero and the reason why. I would say my favorite superhero, and I guess I have to be consistent because I've said this before, I've always been a Captain America fan. As to why is harder, because I've been reading comic books since I was six, coming up with a

an origin story from my own liking of this character. I don't remember particularly when I was six, but I don't think it was necessarily the America aspect or a lot of patriotic, whatever. I'm pro-America, let me go on the record there, but I don't think that was it. I think it was, if I were a DC fan, I probably would have been a Batman person.

It was the lack of superpowers and kind of competing with the superpowered people and keeping up and, in fact, being their leader often when he was just a dude who was in shape. It's like eking the most out of your potential or something like that. Well, that's literally what the super soldier formula is supposed to. It does not give you superpowers. It maximizes human potential.

I love it. Second reference, that's funny. Why do you have Ka Namma Kalajerama on your Twitter profile? That's more sci-fi geek stuff. Oh, man, is this getting obscure? This is Robert E. Howard, who was the original creator of Conan. Also created a character named King Kull, who many people have pointed out is in many ways. staggeringly similar to Conan, kind of a different version. But his perpetual enemies were the

so-called serpent men. I say so-called. They were serpent men, but they could take human form. And I can't even pronounce it as well as you did, but that was a shibboleth, that in some versions of the story, if you said they would be revealed, or in some versions, they just couldn't say it. And you'd figure out who they were. Like their facial structure or something, wouldn't allow them to say it. So that's just random gibberish to see if anyone knows. to see. I love it. I'm trolling for fellow people who had, who had no friends when they were nine. Well, well, let's get into our mutual world of kind of factor investing and it's been certainly an interesting year. I'd love to spend, but by interesting, do you mean crappy? Yes, I do. Very little

sleep during the entire year. So I'd love to kind of start at the high level and maybe we'll get into some of the specifics, especially around liquid oils. I thought that paper was just, you know, really thought provoking. Maybe beginning by asking the major changes that you've observed in the ways that factors, broadly speaking, are used by institutions in a portfolio. And underneath this question is sort of wondering how you would address a group new to factor investing and how you recommend that institutions in today's world use factors. By far the most major change is, of course, them just being much more widespread. I'm old enough to remember when there was no such thing as factor investing. I'm not even sure we called quant investing in the beginning. It was more, let's see if this academic stuff

can make real clients money. It was about as formal as it got. We don't take an approach of saying this is the absolute best way. We do, and I'll get to it, have a version where we say, if you're willing to do X, Y, and Z and live with A, B, and C, this is the highest risk-adjusted return we think you can get. But we actually have a presentation on this where we talk about things on a spectrum from traditional long only to being allowed to short, to no derivatives allowed, to being allowed to use derivatives, to no leverage allowed to be allowed to lever. In theory, and I believe in practice, the first best use, if you are an absolute Vulcan who is going to stick with it and is unsusceptible to unconventionality problems and whatnot,

is to go through every place you think quantitative factors apply. do a multi-factor long-short portfolio, lever the ones that are very low-risk. It's not just about leverage. De-lever the ones that are very high-risk. I always use the example of if you built a factor, a long-short factor approach, it was equally as good in a risk-adjusted sense

for bonds and commodities. And you put a dollar into each long-short program. You're a commodities manager. You have bonds are an interesting hobby for you. So if you put many fewer dollars into commodities and more dollars into bonds, you always have to be careful with leverage. There are limits, but it's really not about more risk. It's about equilibrating them. So first best use in many ways is to do every factor you believe in, every place you believe in it, scaled roughly to be equal contribution because you may or may not believe they work equally everywhere, but to be roughly comparable contributions.

And even then, the sad factors are a life of the. life of a quant is we're probably shooting for risk-adjustive returns. Sharp ratio's imperfect, but I'll use that. One would be exceptionally nice. If we're doing everything, you know we can get into it more, but you know I think the world has great inflated expectations of that. And managers are guilty of that. The clients are guilty of that in some ways. In saying, well, we can't tolerate down this. Managers are all too happy to tell them, no, we have this product that's delivered a two and a half. I mean, you know the math. If you could deliver a one sharp ratio in a long, short, uncorrelated sense, you know, that's more than double the risk adjuster return of the

S&P 500, uncorrelated to it. Theory would say you'd put more money into that than the S&P 500. Nobody actually would do that, which brings us to my next point. We also run portfolios like that. We also run portfolios that are long only, no leverage, beat the benchmark, tilted, in a sense, some multi-factor, some even single-factor. and the momentum side where we think maybe that's a complement to people who already have a value. A lot of value exposure, yeah. There, we can't claim the return against the benchmark is the same risk you just return as when you're allowed to do every asset class using leverage or de-leveraging to balance them.

And I've heard you talk about this, too. The better strategy you can't stick with is not the better strategy. So we try to spend a lot of time talking to people about where on the spectrum, really with a few different dials, one being how aggressive you can be. One thing about being a quant, I think it's way easier than for a traditional manager to dial the risk to different levels, essentially. If someone's more aggressive, you just do everything at a larger size. In principle, that should be possible for more traditional manager. But when you're running kind of a concentrated book of a few stocks, what that even means gets a little murkier. So one is how much risk you can take.

Two, a very correlated concept, because it's about being able to stick with it, how on conventional you can be. They're correlated, but they're not the exact same thing. There are people who could take a fair amount of fluctuation as long as it's not considered weird or ex post imprudent. Why did you go do that strange thing? And there are people and organizations more wired to be okay with less conventional approaches. So it's art, not science, and we do a lot of scenarios. We try to guess it worse cases. Those are two of the more dangerous words in finance, worst cases. I'm very fond of pointing out that the most common. common model for a worst case is a real life or a back test, what's the worst thing that's

ever happened? And I'm fond of pointing out the kind of trite but true, I think, observation that you know that's a bad model because it missed the one that happened. Right. Right. It wasn't the worst case until it happened. But it's not just quants. There's nobody doing any kind of investing who doesn't have to have some form of what's the worst case and can I tolerate this. So looking at history, taking a haircut, being more conservative than history, something. Something you want to be considerably more conservative than history. If you've never seen any weird events, they are out there. If you're trying to come up for a worst case, that's much worse than the Great Depression

for the economy and for stocks, well, that's not too productive to me. I'm willing to say the Great Depression is kind of a, if you think the worst case is much worse than that, than the proper assets were firearms and can goods. Right. It's not a political statement. It's just a joke. Just a fact. So, you know, figuring out, telling people, we think these factors can,

add value long term in a traditional sense or in a highly non-traditional sense, in a geek, risk-adjusted, how much we can move the dial on the portfolio, the more, the better we can do. But we don't push for that. You never want someone to be pushed to their comfort point. It means they're beyond. I think we all overestimate our comfort point a little bit. When the tough times hit, you discover not that everyone runs immediately, not that everyone's a total wuss about it, but you're a bigger wuss than you thought you were.

This is a good year to be having this conversation. You mentioned this idea of like Vulcan before. And I think everyone assumes they see these beautiful log, growth of a dollar quant, back test charts, that this is this cure-all for portfolios. But this year, it's been a tough year, things like Value Not Working as well. What are the most common concerns or questions you get from people that are living through this as clients that you think are good, you know, good questions to be raising? Well, you added that last part.

I get some wacky questions, too. Well, we can listen to those, too. I'm the wacky ones. We mostly get great questions. Remind me to share some of the wacky ones. I got at least one good one. The most valid question I ever get is, is there just too much money in this? Don't get me wrong.

We're building up to. I think I have a really good answer for it. I've worked. Putting your reps on this one. But the people who ask that, particularly prior to the people who ask that before a bad year. Yes. Because it's much easier to worry and to ask things after a tough year. But either way, that's always.

a legitimate question. There is no strategy that could, this is one of the more obvious statements, I hate when people use infinity as an argument, but there is no strategy that can run any amount of money it chooses without affecting its risk-adjusted return, either the individual manager or in the aggregate of similar strategies. So the possibility that strategies like we pursue, that you pursue could be arbitraged away by too much money, it's a very plausible question. Now, people do tend to ask it much more in a bad year again. If anything, I don't think that's quite right, because I think if a strategy was arbitraged away, this is a subtlety, but I think it's often missed. It would just look more like a random number going forward. Sometimes people think arbitraged

away means you automatically lose money going forward. So it's really not a very good explanation for a bad year for a factor or a set of factors. But it's a fair question. So it's the number one you got to deal with. We try, and I know you've read our stuff, so I'm speaking for your listeners. And you just you just have to nod as if this is new, interesting information to you. But we always have to try to be more specific. Sometimes at its most general, you know, has the world changed? The answer is, yes, I don't know how. It's changed for the last 150 years. The stuff has worked over that period. You've got to get more specific. So too much money, even there you have to get more specific. People knowing about a strategy doesn't ruin a strategy.

It has to be their actions. We have come up with two major things we think we'd see if there was too much money in the strategy. One is, for us, this goes back to 1999 during the tech bubble when we tried to come up with a way to measure how cheap or expensive a factor was. I still call it this occasionally, a terrible term. The value of value, if you're applying to value, you can apply it to any factor, though. You take your favorite valuation measures and you look at how cheaper expensive, the long, Zarr versus the shorts. All SQL, we always like cheaper.

One thing you might expect if way too many people were doing this would be great compression in these numbers. We've written on this, we wrote a paper with the terrible title, Contrarian Factor Timing is deceptively difficult. I'm still embarrassed by that. Barn burner. I think we're pretty good at titles. And they're usually like declarative, simple titles or they have some humor in them.

I was part of this paper. So I'm not, I'm not casting aspersions. I signed off on it. But afterwards, I'm like, that is the IBM of titles. It's the most boring name for a, interestingly, we named the firm AQR, which was the IBM of names, applied quantitative research. Not necessarily because of the name. But the short answer is it varies across factors. And it's a tough one.

The world doesn't have a common language for this yet. If you have slightly different valuation factors than I do, you might have a slightly different answer. But we all know once you're doing price divided by something, they're fairly correlated concepts. And all the factors we look at are within their kind of normal historical range. If you look at the value factor itself and you choose the most well-known version out there, the Fama French price-to-book factor, HML. It's long the one-third of stock's cheapest on price-to-book, short the one-third that are most expensive on on price to book. We don't use only that by any means price to book get some heat these days. I wouldn't call it perfect, but it is the lingua frank of systematic value investing. If you look at that

historically, the expensive stocks have always, with one exception, traded in a ban between three and six times the price to book of the cheap stocks. That exception, of course, was the technology bubble in late 99, 2000. It went up to, depending on precisely whether you use their measure or our way of doing it, you know, 12 to 14 times more. And it was something that we developed back then as a way to try to gauge, all right, we think something crazy is going on, but can we put a number on it? It came back. It has been within that band ever since. I think it's probably on the cheap side of that band again. I think the price to book factor looks cheaper than some of the other value factors. On net value is probably a little cheaper than average, but not much. The other factors,

some are cheaper, some are more expensive. The only one really at all expensive versus history is the low beta factor. Even that is within historical bounds when we don't find a lot of timing of the factors too useful. Is there anything that could break this idea? So I would agree, obviously, that narrowing spreads would be one source of excess return for a factor into the future. We get asked sometimes about industry or stock obsolescence as the result of technology, like this sort of winner take all or winner take most idea that, yeah, like in history spreads were this. but we didn't have this like incredible dominance. I think I know your answer because, you know, we've looked at similar data. It's so hard because these very general questions, could it be different now that there is a dominance of a few stocks? I am of the view that when you have, you know, 100 some odd years of evidence across acid classes, across geographies,

across other times of modest technological change like rural electrification and, you know, going from steamboats to railroads, that the stuff is reasonably robust. Now, any one measure, you can always tell a story where it is broken. And Price the Book comes under the most heat for this. The idea that there used to be a lot of physical plant required, and now it's more of an idea. There's some accounting gimmickery. A gimmickory or reality.

Like, you know, Netflix and whatnot has a huge networking effect. That's very protected moats. That's not very capital intensive. This won't shock you. We don't think that's what's going on. For one thing, we've always given some weight to price to book, but never literally since we started in late 94 at Goldman Sachs. It's never been a dominant factor for us. We diversify across a lot of measures that I think most people would agree are more robust to this kind of issue.

Second, this gets very more specific to us, though it might do exactly the same thing. I don't know. We've always taken most of our value risk, not across industries within industries. We wrote a paper on that. Asnes Porter and Stevens. It's never seen the light of day. But it was 1995, so it was early. I'm still a little bitter we didn't get that one published.

One of my co-authors left to start his own firm, and we're friends again, but we didn't talk for a few years. I think we missed our window to publish that by just not getting back to it. But I think it was an early paper trying to break up industries versus non-industries. There's a lot of hypotheses for why value might work better within an industry. One of them fits what they're talking about today, just. might be very hard to compare. You might have, I'll never say permanent, but long-term growth differences that justify value differences. It may be accounting differences across industries. So the fact that we diversify across a lot of value measures, the fact that we don't take

big industry bets makes this idea that there's a one or a handful of stocks dominating has changed the world. And then we just look at the empirics. Empirics are the famous stocks everyone talks about something sounding like a fang with M's added and other things subtracted. The ever-evolving acronym. Very, very small part of the systematic and value investing story this year. I mean, yes, systematic value investors probably been more on the wrong than the right side of those stocks. But it's not the story.

And it's not any one industry it's calculated in. So, you know, historically, ridiculously robust to technological change. We've already tried to design it to be as robust as we can. to future change, and we have a healthy amount of momentum and other type quality type factors in there because we know it's not always a value world. See, one thing, and we're guilty of this too, it's very subtle to talk about performance attribution for a multi-factor approach. The single tagline on this year, and if you're going to go with one tagline, this is the right one, is it's a very bad year for systematic value. But it's also not nearly enough. And on my website,

I've detailed this at times. Sometimes I accurately say that's the worst factor on the year. And if that factor, if you flip the sign on that factor, we'd be having a wonderful year. That's really not, it's very deep for me to say. When you're losing, if you flip the sign, you would be winning. You know I get whole blogs out of concepts like that. But for the prior, the value, the tough period for systematic value has been going on mostly since the GFC.

That is not just this year. And post-GFC has been a very good time. on net for our factor models. And value has been in there and losing most of the time. So it's actually more accurate. It's more accurate, but also a totology that I'm not sure really helps to go. This year, the value is doing poorly and the other factors, which had been making up for it for a long time, low beta, profitability, momentum are not. That is literally true. I don't really care which factor goes up. I care desperately that the whole goes up. So the prior six, seven years of value, not every year, but generally having a very bad period, and us doing well did not bother me

in the slightest. I did not bemoan that we didn't time it at short value because I don't think anyone can do that very well. I think that's way too ex post. And I said, hey, great, we've done well. This year, it really is that combination. What ends up being just remarkably unsatisfying, And I admit, quants can do performance attribution to 11 decimal places and still be unsatisfying. Because why most years, even when one style like value suffers, the other styles that we think are quite good more than make up for it, and some years they don't. Well, we have, I think, pretty strong stories for why these things work on average. That we think we understand. But why in one particular year momentum didn't catch it right and more than make up for value.

as it does in many other years. I freely admit we're betting and very happy to bet long term on that more often than not it will. But when it doesn't, or when some of the other factors don't step up, we are left saying that thing we do

that's worked for us for 20 years at AQR and 25 years since our Goldman Sachs start and 100 years in back tests did not work this year. And people seem to find that unsatisfying. And not everyone, some people are okay with it, but I do wish ultimately we could put some more whys on our stories, but it's not the nature of quant.

You know, like we spend our lives, and I know you do the same. You do some things that are more concentrated, but some that are more diversified. We diversify. We have many, many hundreds of stocks. There's never going to be a simple story. It's going to be the sum of these things usually wins. We think we understand why. We are more than happy to bet on it winning in the future.

This is the one out of four or five years. It doesn't win, and it's the one out of ten years. loses badly. I know you're as committed as we are to just open education. Like you share a ton of research and that's been a key part of your DNA, but also the firm's DNA. How do you think about that from your client's perspective, if it's your client's right to sort of own your best work, if you will, thinking about sharing versus keeping something proprietary? Like is there a set of rules by which you might actually not share something? How do you think about that? For one thing, I would step back and describe just for a minute, we've set up our business and we try to be ridiculously

open and honest about this, that we run some things that we call style portfolios, that they can be quite involved. We use the arrogant term craftsmanship on, you know, we've learned, we think we just learned a lot from 25 years about putting these together, but there's nothing in those that only AQR knows about, that we try to produce as well as we can, an excellent version of what's out there. And then, by definition, I will not be talking about it on this podcast, but we do think along the way, and we're always trying to improve upon that. We have some proprietary aspects. So we have versions of what we do that pursue both styles

that we never get away from that. And for want of a better word, alpha. Here I'm using the word alpha to mean after the style tilts. As to what gets into what portfolio and what we share with clients, there are things that we still consider proprietary that we would even share those with a client that we absolutely was certain with a pure investor

who is, if a client has internal quantitative proprietary trading, I have to admit, there may be a little bit of a judgment. If we don't share something with you, it's probably a compliment because we're doing something. You could do some damage with this if we shared this with you. But we do have to make a judgment on some things. And in general, the arc of history runs towards proprietary becoming a style. And we wrote a paper a number of years ago on exactly this simple point,

but just that alpha, something may have been alpha in the past. and it may go away completely if it's a very arbitrageable. I'm not sure that's actually a word, but if it's very arbitrageable thing to take out of the market. But I think it's very common. If you go back literally, you keep saying 25, it's probably 24 years, 1994. I'm going to go back to saying 25, but let me just be honest about that. I patted it by a year. When we started doing systematic value and momentum, I think it was effectively alpha to our clients.

Very few people, your dad being one of the examiner. but very few people even knew about this in the world, let alone we're offering both long only and long short versions. So only known to a few people, positive, expected return uncorrelated to the market. I don't know a better definition of alpha. Over time, and back then, even then, we didn't do it super simple,

but even if you did just straight price to book, I think you had a positive expected return, and it was alpha to the world. I don't think I could look at anyone today and say, I'm going to go long, low price to book, short, high price to book. and earn you some alpha. And earn you some alpha, or obviously more relevant, charge you alpha fees for it.

A lot of this comes down to fees. As things become moved from alpha to a more known but perhaps still good. And again, that argument that we haven't seen the prices collapse. I won't go into it in depth, but we haven't seen the transactions costs go through the roof on these things. That idea that the premium is still there, but it is known, that's a supply and demand. You simply can't charge for something. a secret. If you have something that is a secret and it is good, the challenge, of course, is getting someone to believe it. You tend not to have back tests in 72 places and published papers

by definition. You don't have published papers by all kinds of luminaries. But ultimately, getting back to your original question, we make a call on what we think is well known. And I think we try to decide it to maybe reveal a little too much in the favor of styles, meaning if it's on the edge, we'll say it's known out there. But there is still some things we think we've developed that are not well-known enough to be called a style, but there's no magic formula for it. It's partly surveying the world and the literature

and what's written by all of academia and our worthy competitors and saying, is this out there? You never know for sure, because this would be an odd coincidence, but it could be every one of us have independently discovered the same thing and simply chose not to write about it. So it's not a perfect model, but I think it gets most of the way there.

Can you talk about how the research process then is governed, whether or not you think in terms of this informational versus analytical type of edge, if that's how you parse things, if it's a sort of top-down type system or more of an organic bottom-up system? I'd be fascinated to know how that process works internally. Oh, so would I.

You should really work on that. Sounds like bottom-up. Well, I think there are both aspects to it. There's a lot of bottom up in that we have a lot of people. I'll brag about the Vite. We have some pretty high power finance people here who are all on their own. I have a sneaking suspicion. Some of these people read this stuff, you know, Saturday morning instead of the paper.

You know, they get their coffee and they sit on their deck and their robes and they're reading the Journal of Finance. I used to do that. I admit to anyone I don't do that anymore. I get to it later in the day, but not first thing in the morning. So there is just a fair amount of someone reading the latest paper, the latest working paper, from God knows where saying, hey, you know, it doesn't sound like we're doing this, and it's at least a plausible story. We then bring, I hope, a ridiculous amount of firepower to it.

Let's really tear apart the theoretical story. And often this is just a function of scale, and we have pretty decent scale at this point. Let's test it everywhere the author's story should apply, just methodically. because all you're creating now is a higher hurdle to data mining. It can still be data mining. But when you find some similar spirit story works in different asset classes and different geographies and different time periods than the original author. And ultimately, that can be one source.

We do a fair amount of top down in that. And this is increased over the years. I think there are advantages to being bigger and disadvantages to being bigger. And this is some of both. You have to get a little more formal in your process. So we have seminars where, Sometimes it's someone from AQR, often it's someone from the outside, where we're just bringing in and saying, present your favorite latest paper. Very often, it's not directly applicable to making a model 0.02 sharp ratio better.

Sometimes it is, but very often it's not. But it keeps that kind of academic spirit going. Not that we always achieve this, but our goal is to be a fairly high-level academic finance faculty who happens to be much more focused than them on actually. making money. Right. Post frictions. That sounds easy to say. Yes, post frictions is funny. I look back at my dissertation, and I'm mildly embarrassed by it in that I was looking at some of the standard things, value, size, momentum. I was not the unfortunately for me, but one of the very early papers on momentum. But the trading I did ignored transactions costs and worse was trading equal-weighted portfolios. In some sense, there's a lot of dumb luck that the stuff was good enough to survive

when we got more real world about it. But I occasionally, and I'm not really embarrassed because most of the rest of the literature was doing it exactly the same way at the time. The top down, making sure there are scheduled times to do this, particularly when we have changes to our approach, changes to a model somewhere. We take a very applied academic approach

where often someone will present, just like they're in an academic seminar. The model we have for that is a few of the founders met in University of Chicago's PhD program. And the famous, I think it's still going on, Tuesday afternoon finance seminar there. It's where I had to present my dissertation. They sat in a U around you,

and this is classic for many years. Eugene Fama sat all the way to your right, and Merton Miller, who sadly has passed, but he sat all the way to your left. And so you kind of got a little whiplash. But I actually found it not that being of a place, but it had kind of a mean reputation. I think when other people would come visit,

it was known as a seminar where people would instead of going, well, that's real interesting. They'd go, I don't buy it, that kind of thing. But it's kind of what we were raised on. So people here have to kind of convince other people of their idea. So I really, in some sense,

I wish you could formalize this more and have a research factory, but it just doesn't work that way. You try to do a fair amount of structure to make sure things are always going going on. There are always thoughts. We're always looking at the current papers. We're always trying to produce in the literature also. And a fair amount of bottom up where people, everyone wants to be the hero. I love that. Everyone wants to come up with a great new thing that we're not doing yet.

So that's, that's wonderful too. You know, in this business, it's definitely been the scientific method, Lee Square's regression, big tools being used to try to improve the portfolio process. Now we're coming up against this period when all this kind of fancy machine learning stuff has become extremely popular to at least pay lip service to in an investment process. I'm curious your take on how you and AQR will manage the transition to, let's say, a more pragmatic type approach, which is how I kind of think about machine learning, sort of finding nonlinearities in this data and how you'll get comfortable with putting that sort of thing into practice if you do. I know you know this, but we made a big hire in the machine learning area, Marcos Lopez de Prado, who's one of, if not the big name.

Like you said, a howitzer. Yeah, I did say at one point that machine learning is an arms race and he's a howitzer. I hope he takes that as a compliment when he hears this podcast. But he also knows and is a big proponent and when he joined us of two things that we share in, two beliefs we share in common with him on machine learning. One is to sound trite, evolutionary, not revolutionary. I don't think you ever saw it. I sent him one of my papers from forever ago on the interaction of value and momentum of looking

for non-linearities in aversion. very low-tech way, just with bivariate sorts, and it was one of the early papers that did that, and he was kind of excited about that, and there were ways to take machine learning and push it, but machine learning is basically, and you called it yourself, a tool to look for non-linearities, interaction effects, or often the same kind of linear interaction effects, but more complex relationships than ordinary least squares. What we also agree very much, or else, you know, wouldn't have been a great idea to have hired Marcos. is it is unlikely at AQR, you will see a machine learning process that has no economic explanation.

I admit, and I think Marcus would agree, how well we can do it creating things where machine learning helps, and we really feel we understand why. It's an open question. It's what I'm optimistic about, or else, again, we wouldn't have done this. But we do require that economic story to go. It's got to be interpretable, in other words. And you just don't have enough data in our world. If we got to observe a few million years of stable U.S. and global economy. stationary.

Yeah, and nobody ever came along with a technological change. Oh, I'm kidding. That's the one we don't worry about. But you could really use the scientific method where you can do all kinds of A-B experiments. And we just can't. We get to observe the world once. Often it's Pandora's box. Somebody observed something and you can't unring that bell.

Someone opened the box. They rang the bell. We now know that fact. So plausible stories, economic logic, add a sample test, of course. That's always my favorite. I admit I love to talk about. economic stories, but finding a place we haven't tested something yet. It's still the most exciting and nerve-wracking thing because you don't actually know if it's going to work. But we're

trying very hard and we're very serious by taking the same approach to machine learning and check back in two years. Yeah. Back just to round out the idea of kind of this year, I want to hear the story about the wacky question. But before we get to that, before we get to that one, I'm curious how you think about what you would do if some of these things were flashing red. if it was transaction costs, let's say, or let's say the narrowness of the spread. You know, cheap's always going to be cheaper, but the narrowness is, it breaches its all-timer. I have to be honest and say it's what I hope I would do because I think there is an ethical action to be done.

If we ever saw spreads smash down, if three to six, to use that Fama French example, was the norm, and suddenly it's always going to be greater than one because, you know, you sorted it on price to book. I always joke, if it's not greater than one, your spreadsheet is broken. And then some of the team here has asked me to stop saying spreadsheet because it makes me sound old. They're like, Cliff, you know we don't do that in spreadsheets anymore,

like you did in the 90s. And I say, yeah, of course I know that. And I'm going back to my office. I assume there was still spreadsheets floating around. But if that ever pick your favorite, smash down to one and a half, and then kind of flat line there for a while. You know, anything that happens for three days that comes back

was a crazy, markets have crazy events, but kind of flatline there. I think, and again, I made up one and a half, and we'd have to do some work to try to figure out what the threshold is. But I hope and believe we'd be a firm who'd say, we're going to stop doing this. And if we have any products that only do that, we're going to stop doing those, maybe hopefully, hopefully knock on wood for AQR. We have something else that is recommendable.

But even in the extreme, I do believe, if I woke up tomorrow convinced, the sky opened up and told me we have a zero sharp ratio. I believe I'd close her down, but I've never been faced with that, with the sky opening up. So I can only brag about the virtue I believe I would have. Bravery is always easy from a safe distance, right? Yes, it is.

But across everything you do, you saw the transactions cost triple and the expensiveness go through the roof or cheapness, depending on how you want to define it, of what you do. I don't know how you can make the argument that you should just keep doing it. We make the opposite argument.

We've not seen that. Therefore, we're very comfortable continuing to do what we believe in and has worked live and over the super long term. But if we saw the opposite and we don't do it, that would be a ethical failing. And right now there's an AQR lawyer somewhere. My general counsel, Billy, is now saying he really shouldn't preadmit to that beforehand. And I'm telling you, Billy, it's okay. I will do it. If we ever have nothing, I'll tell people we have nothing.

So what was the wacky question? Over time, I'm sure you get them too. I'm not just sucking up. Our clients are good. They ask good questions. They're in this for the right reasons. You still get wacky things if you do this long enough. The wackiest question I ever got, and we've really had three tough periods in 20 years

at AQR and about 24 years going back to Goldman. They're about 10 years apart, which a number, this is not the wacky question. This is semi-wacky. A few people have asked me, do you think there's anything too? that. And I answer that one, honestly, you know, this is the third data point in three very different environments with the ups being much bigger than the downs. These are just, you know, I want to sound like they're always down. These are three episodes that have been more than eclipsed by the good times. No, we don't really think there's anything to that. But then I admit to those people that after

three harrowing 10-year events. Ten years from now. Going into 2028, I'm going to pay a little extra attention, at least. I will get a cold chill going into 2028. But that's not quite the wackiest. The wackiest we ever got was, I believe this was during the technology bubble. These actually blend for me, some of these events. A client came to us. We made an error when we started our firm. We talked about how you could do this on a spectrum, how aggressive to be and how unconventional to be. We, of course, led with our chin. When we were at Goldman, we did the same thing. Since 94, we've told people we could do this at different points on a spectrum and tried to work with them on where to be. You can't launch a new firm doing everything.

So we, of course, launched the 20 to 25% vol, fully unconventional. Strap in. Fully long, short, much as it's long, market neutral version, and ran teeth first into the tech bubble. That is when we created that research saying, you know, we do think it's crazy. I'm a Gene Fama student who still worships at the altar of Gene Fama. but I don't think markets are quite as efficient as he does. And I got jolted a little bit towards more willing to use the word bubble by living through that. We created all this analytics saying that we ended up being right.

So I'm quite proud of it. We think the expected returns are extremely attractive now. And we don't say that all the time. Today, I don't tell people, get in now because the returns are quadruple normal. I say get in or stay in now because we have our normal level of positive expected returns. So I'm not the guy who always jumps to, oh, we've had a bad person. period. That time, I really thought this is conditionally super attractive. It could kill anyone for the next six months, timing it to the day as a fool's, Aaron. So I had a client who was

redeeming. And most of our clients stick with us through that, which was actually amazing, given it was at the start. This is a bad year after 20 net good years. Not all good years, but 20 net good years. That's much more tolerable than bad year one. I was very thankful to our clients, but here was one who was leaving. And I said, you know, we've talked a lot. We had a bunch of meetings on the attractiveness. Do you not agree? And they said, no, we think you are the most attractive thing in our portfolio right now.

But we think, and it ended up being faster than this, thankfully, but we think it will be four years until you make the money back. And we can't wait that long. And I admit, and I will issue no names, and I'll embarrass no one. I had hair back then to pull on. I'm saying, so we're the most attractive thing in your portfolio, not by our estimate, but by your estimate. And you're selling us because we're down since inception, and it's going to be a while until we make the money back.

I asked if there was some weird accounting reason where, you know, you can imagine some crazy system where they get paid that way. That would not be a good system. But he's like, no, we just, we're uncomfortable with something that takes a while to make back. like, okay, well, I've already convinced you were the most attractive thing in your portfolio, and that was not enough. So I have done all I can do. To be honest, looking back, I was younger and even stupider back then. I think I was missing something. I think I was probably talking to someone whose superior told them sell this dog, who believed in it personally and had to come up with something to tell me. And that didn't dawn on

me maybe for a decade. For about a decade afterwards, I would tell this story as if I was talking to someone who's a little crazy. I now think I was talking to someone I actually feel bad for, because I think they were put in a pretty bad situation of having to explain to me. And the person was quite honorable. They could have just sold their boss out. They could have just looked at me and said, hey, I want to stick. And that guy, Phil in the corner, he's easy. But that was probably the wackiest conversation I've ever had. We think you're the most attractive thing in our portfolio. We need out. So I always think one of the most valuable contributions that Quant Research can make is helping

people understand what not to do. You wrote in the Liquid Alt-Ragnarok piece about this kind of pejorative three-sharp ratio strategies. I want to talk a little bit about that and how maybe from an allocator's perspective, those could be almost like the lottery tickets of our world. Maybe describe what you mean by a three-sharp ratio strategy and why pursuing them or chasing them that may not be the best idea. First, your listeners don't know this, but you properly made air quotes when you said three-sharp ratio strategy. And every firm gets their own isms that are, things you bat around. Three sharp ratio strategy at AQR is generally, as you said, used as a pejorative. That doesn't mean we wouldn't love to have strategies that are really three sharp ratios.

That doesn't mean that some don't exist. Our friends out at Renaissance technologies in Long Island is always the one that comes up. By no means do I think their returns are not true. I think they're incredible and they're the exception that proves the rule. I love when people ask me about them because everyone's from all the client will go, are they better than you? and I'll go, hell yes, but they will not take your money, and we're pretty damn good, and we will. So I'm not sure why that's relevant. We do point out that even the quest for that so-called three-sharp ratio, they're a good example. They will take your money, but in products that actually look a lot more like what you and I do, they won't take your money in their never lose money medallion.

If you actually find that holy grail, somebody who's produced something that is for real, that is inefficiency no one else knows about or an efficiency everyone else knows that somehow they're just much better at executing. It does seem to be a pattern. And in the few instances where I think those things are real, they kick out the clients anyway. So yeah, you get a few years, but that's not even a long-term thing.

I believe that real-life strategies, we tend to use a range of about 0.5 to 1 as a sharp ratio that we're targeting. If you look at our real-life records, that's a pretty good, pretty reasonable for what we've achieved in different portfolios. There's no magic to it.

We're all guessing at these numbers ultimately. We're looking at history, applying costs, trying to think of whatever haircut for them being more well known, even if the spreads are still attractive. But importantly, strategies with risk-adjusted returns like that that can be done in true institutional capacity can move the dial a ton more than a tiny allocation to a three-sharp ratio strategy.

Sometimes people, we live in the risk-adjusted world too. And in our world, and again, I'm using sharp ratio as a shorthand for hopefully a more sophisticated, subtler version. You wouldn't use it for a very big left-tailed strategy or whatnot. I always worry people will think I'm just too much of a true believer in that. But we too compete and try to get the highest risk-adjusted return we can. But our whole industry occasionally forgets that risk-adjustive return is there in the service of total return. Dollars are what counts. a 0.5 sharp ratio strategy, which is actually better than the S&P 500,

diversifying to the S&P 500, that you can apply to a $100 billion versus a three-sharp ratio strategy that you can apply to $100 million. If your objective function is making clients better off, I would argue the first one has made a lot more people a lot better off. It's harder to live with. One thing particularly nice about the business of a three-sharp ratio strategy, even if it for most people is rather limited in scale. I do think some of them exist,

but I think they tend to be very hard to scale strategies, is if you've truly achieved that, you don't have a lot of nervous years. You have some nervous weeks where you go, oh, two down days in a row, maybe it's all over. But ultimately, we're in this to create economic value. And I don't think that generally creates as much economic value as something that's real, still there.

I know you're a believer in this too, probably exists because it is hard to stick with. You know, I do admit this. I get thrilled that I don't think these things have been arbitraged away when we look at the prices, the cost, but they could be. And if they had started, if value ever, it never was a three sharp ratio, a simple value strategy,

but if it was, it wouldn't be now. It would have been arbitraged away. I do think there is, for strategies that are real, and I keep saying that, I think there are strategies out there their data mine that won't repeat, that are simply don't pass that test of enough places where it's worked for enough time and enough different out-of-sample tests with enough of a good story. But if you pass all those hurdles, and in my word, in my vernacular, you're a real strategy. I think there is almost an economically natural sharp ratio it gets down to where it is

tolerable by some, should be tolerable. Again, I used the word Vulcan. If we were all, Mr. Spock, should be tolerable by many, but is not. It's hard to live with. And so you get into this range of, I think it has to, in general, I use numbers a little bit bigger than the S&P's sharp ratio, not for bragging rights and not because that's about been what we've achieved. But I do think people require a little bit more on doing something unconventional. And some of that's fair. The story for why there should be a positive equity risk premium is better than any story you or I have. Now, you and I might believe in value investing. We might believe in risk, explanations, behavioral-based explanations. I love these stories. I think they're absolutely true.

I have more confidence that theoretically, using common sense, you should get paid for putting your money at risk in the stock market. So in that sense, it's not crazy that it can survive maybe the lowest sharp ratio because the story is so good. And even there, I call it low sharp ratio. It's the same thing, the stock market, that people talk about stocks for the long run and love to write sometimes great papers saying if you have a long enough time horizon never goes to zero, but your chance of losing goes down and down and down. And all they're doing is lauding what a 0.4 to 0.5 sharp ratio strategy looks like if you stick with it. It's a little bit of negative order correlation in stock returns makes the long term a little bit better.

That's not most of it. Most of it is if you really have a 0.4 to 0.5, and you stick with it for 20 years, it's a very small chance you're going to lose money. Most of what we talked about is very cross-sectional. So factors within industry or in an asset class, we haven't really talked about the same ideas applied at the asset allocation level. So do you think it's true that it's really just kind of value, momentum, volatility all the way down and up that these are the three sort of basic ideas that govern. Maybe it's trend instead of momentum.

It's, you know, risk parity instead of defensive or something like this. How do you think about the same ideas applied to acid allocation, whether or not they're more, less attractive. Some thoughts there would be great. First of all, let me answer the question before I answer it the long way. I think it is mostly the same ideas repeating. I think the ideas of value that cheap tends to be expensive if you have a reasonable metric to measure it on.

Momentum or trend. Really the same idea. People just in the asset allocation world call it trend and in the cross-sectional world call it momentum. But the idea that what's been going on probably due to underreaction of some kind, but the jury's still out on that. It tends to have power. Quality factors, which include low beta for me, and carry factors, which in general,

we don't talk about a lot when we're talking about individual stocks, which for some reason we talk about most of the time, largely because not that it doesn't work, just because they're very, very correlated to value. But as you get into other asset classes, that link is broken. You go look at currencies, something like value, you can get much more subtle than this, but a simple measure is just purchasing power parity. Where can you buy more stuff? Carrie is very little to do with purchasing power parity.

Whether you're in the real or nominal world, it's really about short-term rates. So there you almost get a fourth dimension to it. I'm now sounding a little Rod Serling-ish. I didn't mean to do that. When you say asset allocation, there are different levels. You could be cross-sectional in what some people call the asset allocation world. For instance, the first paper we wrote as a group in,

in 1995, I think. I'm going to get the title close to right. I'm not going to get our own title, even our own title exactly right. We were absolutely parroting Fahman French's paper on the cross-section of stock returns. We wrote the paper title was very close to the parallels between the cross-section of expected stock and expected country returns. That's a bad title. You've gotten better. Yeah, our titles got better for a while and then we've recently tailed off again.

But early on, this was dumb luck. You know, if you're onto something good and you work hard, I do believe the world's a meritocracy, but how far you go with it is a lot of luck. And we were in the right place at the right time in a lot of ways. At Goldman Sachs, we were asked to help choose countries, which countries in the world a long-only equity manager should allocate to. We were asked to do this because the current long-only product, which was more of an active, traditional bottom-up stock-picking product, had done well country by country, but we're in the wrong countries for about three years. I still feel guilty about this because it may or may not have been statistically significant. But one way at a big organization you get asked to do something is someone else is either doing it poorly or getting unlucky. So they said to us, can you guys do this?

I said, of course we can do this. We'll whip that up for you in a jiffy. We went in a room and we were like, how the hell are we going to choose countries? Pretty quickly, we said your reference. Is it the same thing? I think you're referencing the old joke about the person who think it's turtles all the way down. Embarrassingly, we all remember it the same way. It took us a couple days to say maybe it's just turtles all the way down.

We were like, how are we going to do this? We said, let's treat countries as if they're individual stocks. Which they are, right? Their aggregations of them. What's the DAX? What's the Futsi? If priced a book with your favorite value measure, you take the whole cap of the Futsi and divided by the whole book of the Futsi.

If that's two and the DAX is a four, either the DAX in the FOMA French world is lower risk or in the Dick Thaler world is they're over-optimistic. I have now hypothesized an optimistic, an over-optimistic German, which seems inconsistent, but let me get away with that. And then we tested, you know, if you do these simple strategies, does it hold up? Of course, everything I talk about is kind of a cooking show. The cake's already baked in the back. I'm pretending I'm leading you up to this dramatic reveal, but you know I'm not. And we found that didn't work at all, and we threw darts instead. No, we found it did hold up. But there, as long as it's like to,

like, even among, say, 20 equity markets, you don't have nearly the cross-section that you have of individual stocks, but they're much less volatile, much less of a left tail than individual stocks, too. So there you can still stay cross-sectional. If you're truly talking about timing the stock market, yes, it is still value and momentum, and the others, if you can come up with them, it is much, much weaker, in my view. Auntie Ilman and I have written one more academic, one more practitioner. The more practitioner version was entitled Sin a Little. Oh, Tom Maloney was a co-author on this too. I didn't leave you out, Tom. Oh, I did leave you out, but then I saved myself. We take the famous Schiller Cape. AQR has done this a million times. I first did this

in the tech bubble. I think I was one of the early people doing these bucket, the past by the Schiller Cape. Look at the next 10 years. Equity returns. You see this beautiful pattern. You know it is we all, every quantum is very well. Everyone's like a right of passage to run this test. And I still think it's real in terms of setting rational expectations. I'm still a believer. We have some stuff on our website. Some of our team has written papers, even questioning that statistical power. It comes down a little bit to priors. It fits my prior very well. And the one data point we have of history has worked very well. But even the long-term tests are dodgy to tell statistically. We have less data than it even looks like. But I am a

believer, if the Schiller cape is 35 versus an average of 16, I would use a lower expected return for the equity market going forward. That doesn't mean it's timable. In principle, it should be. If value and momentum work, why not? So Auntie and I, other people have done this too, but we tried to build a really good rolling out of sample test. The bucketing it by Schiller P and seeing what happened, and this is us doing it, not Bob Schiller, so I'm not saying he cheats, but it's cheating because it's looking at the whole period at one. In reality, you had to look only backwards. So simple thing, just look, you can look at any rolling window you want.

We show it's fairly robust to this. Do a value strategy that goes long when the Schiller Cape is low versus the history you've seen to date. You do get a positive return over 140 some 120 years of testing it. It is a very narrow edge. It is very hard. You do not want to make a career. trying to beat the stock market, trading the stock market based on whether it is cheap or expensive. I could have this wrong.

I refer you to the paper if I'm not precisely right. But I think value alone for timing the stock market had a flat 40-year period. This has been a bad eight months for factor investing. I am extremely comfortable going in front of our clients again and saying, it's been 20 good years. We've seen this before. I don't want to scare anyone. I'm not predicting this by any means, but good strategy can be down for a number of years.

I don't think I would be too successful, nor should I be, going in front of people at year 39, and saying, all right, we've not made you money for 39 years, but we've never failed for 40 years in a row. It'll be a moot point, because no one has a nickel left who would have done that. Trend or momentum, as it applies to direction,

for pure market timing, is a little more powerful. I think they are both, powerful, but we just can't measure value as well. You have this roll, right away, you have this rolling window looking backwards that moves over time. So there's no perfect measure for what it should be. The cape isn't a perfect measure. Value is simply a harder thing to compare through time. Trend or momentum, you can get much fancier, but one year price momentum usually is pretty darn good. So it's a very robust measure. So I think, I don't think it's necessarily

that momentum works and value doesn't, but it practically is because when it comes to true timing, and the same is of the bond market. Bond market, it might be real bond yield and 12-month momentum. I think valuation should matter. It is negatively correlated to trend again. It's not as strongly negatively correlated as the cross-section, but it is. So you want to do a little bolt of these. The reason Auntie and I called our paper sin a little is we basically are telling people market

timing is a sin. It will probably hurt you if you do too much. The very basics do hold up at a very weak level. Therefore, if you must, and we're actually, if you read the paper, we're not telling people a sin a little, we're saying, if you must, do a tiny bit. Do a little bit. And do it in the direction that you would imagine when the market looks cheaper but is improving, be a little bit longer and vice versa. But they should be very small moves because it's not that good. I'd love to get your perspective on a couple very big trends in our industry and in the function of markets in general. So first in our industry, you mentioned earlier fees. So the big story is downward pressure on fees. That has to continue almost. Obviously, free indexes are about maybe as low as you can go.

Maybe we'll get paid soon enough. But if you look forward and I always think about like farming where, you know, there's huge amount of people who are farmers and now it's nothing. Is there a scenario in which our industry is like that? That 20 years from now, there's a fraction of the number of people trying to price, actively price public market securities. Is there a scenario? Yeah. Is it the probable scenario? The probable scenario is fewer. I'm not smart enough to know if it'll look like the middle ages today and farming. One giant question is how many people, how many organizations, how much of the world's resources have to be devoted to accurately pricing securities. Now, to someone who hates markets,

Their default answer is almost anyone doing this is just betting, and it's not a useful function in society. And I will take the strong other side of that. A huge advantage of a market economy, and I'm getting political, but versus a planned economy, is that smart people in a competitive world are trying to set prices. And there's a tremendous positive externality from people doing it. It doesn't mean everyone has to be an active manager, right? If the whole world we're actively trading stocks, the information's already there. And like all math problems, the last piece of information, taking it from one basis point mispriced to zero basis points mispriced, is worth approximately nothing. So, you know, to give you another example, Jack Bogle will tell you flat out something very obvious, but still fun to hear from Jack. Not everyone can market cap index.

Somebody has to think about individual stocks. Quants can help, but I actually think you still need traditional managers. Quants can lower the spreads on a lot of the normal things, a lot of the systematic things, and keep them in line. But I don't literally know that a candy store and Apple computer are, if it's in their price to books, I'll know. But, you know, somebody has to actually go, no, Cliff, that's a candy store. So we do need active managers. We have no idea how many we need. I've defended them, so maybe I get some leeway to say this.

I do think historically we've probably had too much for the last 50 years. We've probably had more people trying to beat the market than is necessary to create an efficient market. And therefore, there was a lot of room to have fewer and have prices come down. At the end of the day, I'm mainly filibustering the question because I can't get to the final answer. I still think, you know, people sometimes talk about this like we're very late in the game. The last numbers I saw for mutual funds were somewhere in the mid 30% were indexed. And I always have to dig. Sometimes when people say indexed, they just mean quant.

and some of those things are not close to a cap. When I say index, I generally mean cap-weighted. I call them Jack Bogle. Because what you do, what we do is not an index fund. We vary a little or a ton from the index. You know, in the extreme, we run things that are long, short, and levered. Some people actually will call those index funds because there is a form. Rules-based.

Rules-based is a good way to describe it. So, you know, in the mid-30s, that's up dramatically from a much smaller number 10 years ago. So the move is huge, but it's not like we have nobody picking stocks. And I'm going to guess, and this is a pure guess, that the number broader than just mutual funds is fewer index funds. When you get into individuals and company holdings, I don't think they're mimicking index funds very often. So I think there's probably room to go further on indexing, but all I can tell you is further from here and not to zero. And the zero world, we don't even know what that looks like. We all have physics envy in our field. I call it the singularity.

Nobody doing this and us all trying to copy the index that nobody's actually looking at. No one knows. It will never get to that world, so it's fine. But no one knows what that world looks like. Fees, I'm proud of the fact that we were pretty early in this. We've never done real index funds, Jack Bogle index funds. But in separating our business, like we talked about already into what's true alpha and what's a style and pricing it different, I, I, I, I, Will, this is a little fox guarding the hen house, I will say, I think the obsession between two and 20 and 20 bits was worthwhile. And the obsession between five and three bips, it's not that people are wrong. If there are two things you're absolutely convinced are equal,

absolutely convinced, and one is three bips and one is five bips, take the three. You know, If someone offered me two free basis points, I take them. But that means if there's any reason you think one is better than the other, it is highly implausible. It's not worth two bips. So I do think this has just become almost a weird macho soundbite thing. We hit zero. And it certainly gets attention.

Lower is always better for the same product, though most of the heavy lifting in terms of consumer welfare, I think has been done in terms of those products are available now. But just because I'm going to whine about that five versus two doesn't matter, doesn't mean anyone's going to listen to me. What's your own take, maybe even from your own investing perspective, on the rise of the prominence of private markets? And obviously, there's still huge opacity that's there. The firm's still getting two and 20 or higher. And obviously, transactions are much harder. There's a lot of differences. But this is a common theme is that companies are staying public longer, you know, fewer IPOs, fewer publicly listed securities. How do you think?

think about this as someone that wants to be, you know, you preach diversification and a lot of the value of the world is, is private now. So how do you think about that, maybe even personally? Well, first, I'll cut to the chase. In DFA wrote something on this recently. I thought it was very good. I still think there are more than enough stocks that are public. I have no worry that the thousands of stocks out there and the giant amount of market cap. So this is not a self-serving. I don't have a, I don't have a horse in this race necessarily. And I would if I thought it was threatening, if the public mark is dwindled to a tenth of their prior size, it would certainly be constraining to those of us who work in the public world. I've had a lot of schizophrenia over time

and changing views on privates. I've got to be even more careful. I think I believe rather consistent things, but I've interpreted them differently over time. I have a 21-year-old story. This was the Asian debt crisis of 1997. We thought this was a crisis. We had not to see it. seen the tech bubble or the GFC. So, and it had been a calm five or six years. So we, you know, we thought this was pretty harrowing. I think the S&P had a down 7% day. I mean, down 7% is pretty frigging bad.

You don't need to cheat and quoted in points to make that bad. That's a bad day. But we had been, we had started at Goldman Sachs for a few years at that point. We had our market neutral, aggressive equity, not equity, aggressive everything product. And that day, our P&L system, said we were up. In reality, we guessed we were flat. Now, this was bad for me because the then head of Goldman Sachs John Corzine literally came by and looked at our screen, which I'd already toward him. I had showed it to him once. And he was kind of going everywhere Goldman Sachs had risked that day

and saying, how we doing? Goldman Sachs had a fair amount of money with us as partners capital. And the screen said like up 3%. And this is also a different era. Nowadays, we have a thousand people and 70 of them are looking at the exact value. There was actually no one at the P&L screen at that point, which kind of tells you something about what the 11-person group back then was versus now. But I literally come there to find the head of Goldman Sachs kind of getting the only good news he's gotten all day. And I had to say, no, John, I actually think, you know, looking back, would have been a white lie just to let him believe it.

Couldn't do it. Couldn't do it. John, our best guesses were flat today. And he'd go, why? It says up 3%. Well, we're in general short the U.S. against Europe. The U.S. is crashing in the afternoon. A great way to look like a genius is to be short a market that's crashing and long a market that's closed.

He's smart guy, ex-trader. He totally got it. But we thought it was freaking awesome that we were flat that day because we had made a lot of money in a bull market for several years. And we weren't rooting against the world, but you know, you want the acid test. You want to say, how are we going to do? So we were prior to John coming by doing cartwheels about our estimate that we were flat.

It's one day. I don't want to overstate it, but it was excited. You can't take someone from up 300 basis points to flat and still have them be excited, even if it's rational that they be so. This is all true story, but it actually has a point that is to deal with private equity. First, we got a little bit, we got a one-day version of what it feels like to be private equity, where you don't have to mark everything to market. It was not the same thing.

It wasn't dampenval. It was directional. It made us look too good. But for once, that was on our side. Second, in the same day, the then head of G. Sam's private equity effort, I think it's long enough that no one can figure out who this is. So I will reveal no names, but the then head comes by and says to me, he doesn't look at the screen, but just says to me, how are you guys doing today?

What's kind of a concern like everyone was? There was a lot of ashen faces walking around Wall Street. day. And I had smartened up. I was at the P&L screen. I said, I think we're about flat. He goes, that's great. He was smart enough to go, that's great. And he pauses and he goes, us too. And then the young cliff, I used to be volatile with a temper. Patrick, you guys, hopefully you can hear him laughing. I am still, I was worse. I have actually mellowed to somewhat, but, but, but nowhere near, no one, no one near to actually mellow. I'm like, what are you talking about?

You guys are way down today. I didn't mean that meanly. I just meant that factually. He goes, no or not. And I'm like, aren't you levered long, small, and midcap stocks essentially? He said, yeah. And I said, if you went to sell them today, wouldn't they be down from yesterday? And to his credit, he said, way down.

And then he added the line you know is coming. But we don't have to say. sell them. And here's where I've changed a bit on this. Roll back 21 years, I just thought nobody got it. I thought these people were all foolish and this was really stupid. This was an arrogant kid thinking the world is dumb and just doesn't get the obvious. I look at it now and I think, first of all, I'm not saying this is the only reason by any means. Private equity can have an alpha reason, too. It should be the main reason, not diversification. But I think a lot of people pursue it with very open eyes, knowing that that inability to mark to market allows them to be better investors.

And I've heard people consciously articulated since then, and I came to the conclusion myself. It doesn't stop me from occasionally being bitter that people like you and I have to report every day simply because we can. I have tried to tell people, how about we tell you every 10 years? Turns out when you do have a place to look up the value of your portfolio, people throw around words like fiduciary and whatnot, they find it very odd if you refuse to tell them. So in some sense, in a perfect, again, that Vulcan world, that should not, that should be a negative, not a positive, right? Illiquidity is generally thought of as a negative in the economic world. In a behavioral sense, in a tie yourself to the mass sense, in your own work, if your investors generally gave you

10-year lockups and you knew they didn't look at it for 10 years, I'm sure you're pretty confident you're going to do well over 10 years. You'd sleep better at night and you'd probably end up with a net better result. You could take more risk, for instance, if you, if you, if you knew that. It's not the world you and I get to live in. I used to arrogantly kind of think they didn't get it. I now think they're looking for great companies. They're trying to produce private alpha, but I think they get it a lot more. And I think that's, that is a real purpose of it is that it allows people to move out the risk spectrum in a way that they can tolerate. I use the, the kind of phrase, they're fooling themselves, but with open eyes. They know what's

going on. Now, as to the rise of privates, partially it's that, that ability to monetize positive risk premium with some vol-smoving. Partly, it is the lack of desire for companies to go public. There's the demand than a supply side, right? The demand side is the demand for return-smooth assets. The supply side is how many there are. And the private world, you know, people talk about Sarbanes-Oxley. I'm supposed to be some libertarian capitalists. I'm supposed to hate Sarbanes-Oxley. I think that's generally overdone as an explanation. I'm not. I'm not. I don't have an alternative one. I'm supposed to like that explanation.

The government came along and did this horrible thing and now no one will go public. That might be some part of it. I have great difficulty determining how much it is. But as a point of fact, yeah. Applying it to our own firm, I can't really articulate for you all the reasons, and I'll never say never,

because if you say never, you end up have to apologize for lying to people 30 years from now. But I wouldn't find going public particularly attractive. Now having to do quarterly earnings calls and, you know, let's try to take our business that fights to be long term and has a tendency to be too short term and make it much more short term. It does not seem like a good strategy to me. And I don't think that's unique to our business. You've referred to a lot of your research colleagues and a lot of well-known researchers here at AQR that you've worked with over the years. I'm curious if you had to think back

or- Who I like the most? I won't reveal that. We won't go there. But what? I have four children. Would you like to know who I like? We should go on stack rank them. So if you had to identify, I guess, common attributes that you think really good researchers specifically share, what would those be? A throwaway is I have managed to hire many people considerably smarter than me. There is a certain raw horsepower to some of these people that kind of blows you away. Questions like this, Patrick, I always find awkward because I always sound like I'm full of it because they're flowery answers right there. but a genuine curiosity. The top researchers here, and I'm going to guess everywhere,

are excited about finding out new stuff first. Secondary to that, they are greedy, economically maximizing individuals who say, you know, now that we found out something new, we can make clients and therefore ourselves money from this. So I'm not saying they're devoid of pecuniary desires, but nobody comes and goes, Look, look at this. We can make more money.

They come and they go, check this out. You're marginal to this factor. Totally makes sense. And there's aversion and fixed income that holds up, which makes us even more sure it's real. They're excited and curious. So I guess you need to start with really smart. I don't mean to be obnoxious, but excited and curious and just bad at it. Doesn't get you.

And I might have been good at it at one point in my life, but I wouldn't consider myself among our top researchers at this point. So I'm bragging for the firm, but not for myself. You do need some of the raw talent. But raw talent that's only trying to say, how do I make more money tomorrow? It's not going to get there. It's not even the right present.

If you're trying to maximize present value of your own holdings, it is not the right route. Somehow getting yourself excited about what you're doing is the right route to maximizing present value for yourself. It's a little ironic. Yeah, it's so interesting. What has you most excited? I'm talking to someone. You use the word curiosity all the time. I'm conscious of quoting you back to you.

All good. What has you most excited or most curious in the research setting right now about the future? Like what research project or idea or asset class or anything is most interesting? Well, we already talked about machine learning. But I have to raise that. It is our major new effort. We have lots of other things that are more, even more evolutionary that are trying to push the envelope a little bit further. I'm optimistic. But even the guys. working in that area know that partly in the next, I don't know, two years, five years, we're going to learn a lot about whether we can apply these things in a way that we're economically comfortable with. And if so, we're going to make this decently better. So it's both exciting,

but also new information to be learned. As a business person, this is a little less intellectual. I'm very excited about fixed income, which sounds like an oxymoron to say. I'm very excited about fixed income. We've traded fixed incomes literally since 1994 at Goldman. We, We traded sovereigns, on value momentum and carry. We've traded currencies. We've traded where to be in a yield curve. It's more recent, not super recent, but more recent that we trade credit instruments. And it's extremely recent that we're very, very comfortable trading against traditional

benchmarks. It's one of the areas where managing a market-neutral hedge fund was logistically far easier. If you have a view that these 20 credits are better than these 20 credits in a market-neutral hedge fund, you go along these 20 credits and short these other 20 credits. when you run against a complex fixed income benchmark that's quite opaque, that only has historical data back to X, and you're asked to take a fairly low amount of tracking error to it. So if you get the benchmark wrong, it could swamp what's left. That is a business logistical engineering challenge, but we spent a lot of time on it in the last few years. So as a business person, I'm intellectually excited about it too, but it is more of the same.

I'm intellectually excited at the same ideas. You look at credit. Spread against a standard default model of forecast is a value indicator. Spread on its own is a carry indicator. Quality indicators are profitability, cash flow-based interest coverage. Even the BAB factor, low beta, low spread duration holds up. So I am intellectually excited to see the same ideas hold up. But it is exciting for us to be able to finally enter the most boring part of the business.

If you had to have, last couple questions, if you had to have a conversation like this one of similar length and detail, but it couldn't be about finance at all. What would be the next topic in for you? I probably should say raising four children born a year and a half apart. The honest answer is probably something more to do with comic books or hockey. But I'm going to go with the first one. Actually, I already ruined it with the second one. But yeah, my wife and I had two sets of twins born a year and a half apart. I don't think, don't take this wrong way, Patrick.

I don't think she listens to your podcast. So I think I'm safe saying this. For a number of years, I made the same joke in industries. When there's someone from our industry around, I would call our family planning a, quote, gross failure risk control. And it turns out she was not happy with that. So I'm using it now with the assumption that she doesn't already have you on automatic download. But you learn a lot about parenting and you wish you get to do it all again because now you finally get to think you're good at it near. shouldn't say near the end, but they're four teenagers. The die is cast now. We have four teenagers

at home. As crazy as it was to have four toddlers at home, having two 14-year-olds and two 15-year-olds is ridiculous. I'll just tell everyone that. You probably didn't need to be told that. I'm curious how you approach with them, like what you do, like how heavy-handed or light-handed have you been with teaching them about business, about investing? And do you have a strong opinion one way or the other? Pretty light. Look, I'd be thrilled. if at least one of them wanted to. At this point, I don't forecast any of them become financial quant geeks.

I'm talking to someone who followed an amazing footsteps of your fathers. I don't see it. Their interests are, they like words more than numbers. And I'm pretty sure they're mine. But they take after my wife a fair amount on that. So when they were very young, I described

it extremely generally. I help people save for retirement. It's obviously that could be interpreted in a million ways you could, but I thought it was a And over time, I've been telling them more and more. My kids know that it's a tough year for Quant. And that's interesting. Last time it was a tough year was August of 07, the famous Quant Quake,

was the last really bad period. And they were three and four. So I didn't have to go home. Now, I mean, they're only nice about it. They're like, everything, okay, Dad? And I tell them the truth. Had three bad periods. You were non-existent and you were babies for the other two.

It just happens occasionally. and it's part of what we do. And they go, okay, but they look, and they go, you don't seem to be in a great mood. And I go, that's because I'm a hypocrite because I love to talk about the long term, but in a bad day, pisses me off. Let me tell you about HML.

Yeah. Oh, well, it's funny you say HML, because this blends perfectly. This is a true story. I have a paper co-author with Andrea Frizzini in the Journal of Portfolio Management called The Devil in HML's Details. Frank Fibbozzi, wonderful guy.

I've known him forever, the editor of the JPM, calls me to tell me the papers except the amount of email. Calls me to tell me that I'm going to make up some of the story as I go, but it's essentially correct.

Calls me to tell me the paper is accepted but tells me I have to change the title. I'm like, why? He's like, too many people these days think HML is the internet acronym for hate my life. And I convinced him.

Frank, you know, you are running the Journal of Portfolio Management. This is, if I was writing for Teen Beat, I understand that they would assume it was hate my life, not the high minus low book to price. And he acquiesced, but I got to check this because I don't know if I made the final cut, but I believe we still have a footnote, literally saying it's not the internet acronym. It's the Fama French version. So there got to be a lot of value managers who are using that acronym in both ways this year. Yes, no kidding. Well, family is a great excuse for my last question for everybody.

which is for the kindest thing that anyone's ever done for you? The kindest thing anyone's ever done for me, it's related to my career. I had a few professors at Chicago who weren't even my direct kind of, the ones I was most close to, take some first drafts I wrote

and rewrite them, just rip them apart. It was a cruel kind thing. I'll give a shout out to Cam Harvey, a wonderful editor of the Journal of Finance for many years. he took one of my papers early on, and it was very nice because he thought I was on to an interesting idea. You know, I guess it was a compliment, because I don't think you would rip apart, and he literally rewrote like a three-page introduction anew for me, which I thought was just incredibly

generous. And I always, in the same vein, have to add my two dissertation advisors, Fama and French. Probably the luckiest thing in my career was being their student at the right time. and I don't know if kind is the right word, but they didn't have to take me on as a student. They didn't take many students. So luck, you know, blessed to be there at the right time and to have them, Gene Fama, being okay with me writing a dissertation on price momentum.

That might be, that one might be closer to a pure act of kindness. You might have heard me tell the story before, but I was genuinely terrified to tell them. I had found this, Jagedish and Tipman were definitely ahead of me, but I had found it independently and was all excited about. It was very disappointed to found out they had found it too. but I went to tell Gene I did a lot of the work

because you don't go there with just an idea you go there with empirics and I'm like Professor, I didn't call him Gene then I still don't call him Gene he tells me to call him Gene and I call him Professor

I want to write a dissertation on price momentum and then I mumbled the second part and I find it works very well he's like what was that I'm like and it works very well because you know it's Gene Phama-esque

Chicago-esque dissertation to say these crazy people on Wall Street chase these price trends and they don't add value and after transactions cost subtract value and Gene would never do this

but there's an implied ha ha they're foolish they're foolish again a strategy and again maybe one day someone will come up with an efficient market's risk-based explanation people have tried

I don't think they've particularly succeeded at that but it is not a wonderful result for the efficient market hypothesis that that 12-month price momentum is an effective strategy now I wasn't stupid enough to say this do this instead of value

that would have been both bad portfolio management because they work well together and also bad graduation strategy bad dissertation strategy but jean said to me and it's funny it took me a little while but this was probably one of the nicest things i've ever heard and it was almost a religious statement from him he kind of paused for a second and he said if it's in the data write the paper which for for him was kind of like the ultimate ethical statement and he was and i did and he wanted me to go on the academic job market for that i ended up staying at Goldman. But he was very supportive of the paper, even though to this day, I, you know, I don't think he's the biggest fan of momentum investing out there. The firm he works with DFA does incorporate

it into their process. So I'm sure that means Gene must have signed off at some point because they, I don't think they, I don't think anyone's allowed to do something if Gene absolutely slipped it in. No, if you, when you're Gene Fama, you get a say. So I'm assuming he, he at least allows a little, but it's not his favorite result to this date. But the intellectual honesty, I will take as an act of kindness. Wonderful. Well, this has been so much fun. I can do this for hours with you, so I appreciate your time. Thanks, Patrick. This was great. Hey, everyone. Patrick here again. To find more episodes of InvestorFieldguide.com forward slash podcast. If you're a book lover, you can also sign up for my book club at investorfieldguide.com

forward slash book club. After you sign up, you'll receive a full investor curriculum right away, and then three to four suggestions of new books every month. You can also follow me on Twitter at Patrick underscore Oshag, OSHAG. If you enjoy the show, please leave a quick review for us on iTunes, which will help more people discover Invest Like the Best. Thanks so much for listening.