Odd Lots - Cliff Asness on How Markets Got Dumber in the Last 10 Years

Episode Date: November 13, 2025

The Odd Lots podcast has been around for 10 years. Unfortunately, markets have gotten less rational over the same time frame. At least this is the contention of Cliff Asness, the co-founder and CEO of AQR Capital Management, a quantitative investing firm that's been around for nearly three decades. Asness' approach to investing is rooted in academic theory, having studied under the legendary Eugene Fama at the University of Chicago. In the world of social media and meme stocks, it's tough out there for the academically minded. And that's forced Cliff to adjust his approach over time. On this episode, we talk about the history of quantitative investing, market efficiency, and the emergence of AI/ML in his process. We also talk about the reality of investing other people's money, and the challenge of sticking with one's convictions at a time when temporary forces are working against you.See omnystudio.com/listener for privacy information.

The news doesn't stop on the weekends. Context changes constantly. And now Bloomberg is the place to stay on top of it all. Hi, I'm David Gurra. Join us every Saturday and Sunday for the new Bloomberg this weekend. I'm Christina Rafini. We'll bring you the latest headlines, in-depth analysis, and big interviews. All the stories that hit home on your days off.

And I'm Lisa Mateo. Watch and listen to Bloomberg this weekend for thoughtful, enlightening conversations about business, lifestyle, people, and culture. On Saturday mornings, we put the past week's events into context, examining what happened in the markets and the world. That on Sundays, we speak with journalists, columnists, and key political figures to prepare you for the week ahead. Join us as soon as you wake up and bring us with you wherever your weekend plans take you. Watch us on Bloomberg Television. Listen on Bloomberg Radio, stream the show live on the Bloomberg business app, or listen to the podcast. That's Bloomberg this weekend.

Saturdays and Sundays starting at 7 a.m. Eastern. Make us part of your weekend routine on Bloomberg Television, Radio, and wherever you get your podcasts. Bloomberg Audio Studios, Podcasts, Radio News. Hello and welcome to another episode of the Odd Lots Podcast. I'm Tracy Allaway. And I'm Joe Wisenthall. Joe, it's a big month for us. Big month for us.

We've been doing this for 10 years. I know. I can't believe it. Do you remember the first episode? Yeah, of course, with Tom King. Yeah. And then our second episode, I think, was about bananas. For some reason.

Yeah, I think it was. You're right. It took us a long time to figure out what we were doing. It took us a long time to figure out what we were doing. And I don't think at that point I would have expected that we'd be doing it 10 years later. I don't know what I was expecting. We were just turning on a microphone in a radio studio and talking for a while. We started doing it because we wanted to have a podcast and talk to interesting people.

I think we were hashtag blessed. No one was listening for a very, very long time, which gave us a long runway to figure things out. So we got lucky. That said, you know, 10 years, it is in fact a long time to be doing this. And a lot has changed in that period. A lot has changed in that period, sometimes mind-blowing. And we've talked about this before, for sure. But that things that we were covering is capital and news at the time are now capital age history.

And it's like these things that are we sort of take for granted. Everyone was there. It's like, no, children, let us tell you what it was like in the old days when people were worried the world was going to come to an end because, you know, Greece. is sovereign debt and all this stuff that we just sort of part of the landscape is people don't remember it. No, one of those things has to be the idea of value or fundamental investing, right? Like, let us tell you about the days when price actually mattered and had a limit to what investors would pile into. Yeah, that's exactly right. Let us tell you about the days when people used to talk about PE ratios. And this stock, oh, it's at a 25 PE, we better sell it and buy the stock

at a 15P or whatever. Yes, that feels quaint. Maybe it'll be back there one day. But for now, know, given how many things in the market seem to be, the Graham and Dodd kind of stuff, feels a little old. A little old fashioned, right? Yeah, okay. So one of the big themes that has emerged in the 10 years that we've been doing this podcast is everyone seems to have grown more stupid, I would say. Hopefully that's not related.

That's not like a correlation thing. It might be. It could be. All right. But, you know, we have all this gamification of investing, people betting on lines going up or down, people betting on random meme coins, things like that. And I think, you know, we talk about a lot on the podcast, but this is actually a fundamental shift in the market. If you think about the market as something that's supposed to be about capital allocation, alignment of incentives,

people are investing in something because they think it's going to be profitable in the future at the right price. And now people are just sort of piling into stuff because other people are doing it. And again, line go up. Deep down, I still believe that the value of a stock should reflect the net present value of all future cash flows. I know you're an EMH guy. But I've, it's been a little bit hard with some of these things. And, you know, the other thing, too, that is sort of change is that like half of our episodes these days are kind of AI related in some way. And so I think there's a lot of interesting stuff going on, particularly at the intersection of tech and applying tech to both investing in tech, but then the application of tech to investing and so forth.

So yes, much has changed. A lot has changed. And we have the perfect guess. to talk about how everyone has grown more stupid over time. We're going to be speaking with someone we've wanted on the show for a really, really long time. I'm very excited about this. It's Cliff Asnes, the co-founder and CIO of AQR. Thank you so much for coming on all thoughts. Thank you for having me. What's it like to be a rational person living in an irrational world, Cliff?

Yeah, you've been doing it for a long time. Well, a rational person is not always how I'm described. But there's rational investing and there's rational conduct in your personal life. So let's just distinguish. those. You guys in your intro said like, it's a little, it's a little scary for the next hour because you said like a third of the things I want to say. Oh, shoot. Okay. I wrote a rather gigantic piece in the Journal of Portfolio Management. They were having a 50th anniversary. None of us were quite old enough to have been there in the first issue, but they were looking for

the old guys to write kind of retrospectives. And the piece I wrote was called the less efficient market hypothesis. Now, I think you guys probably know this, but my disqualification. I think you guys probably know this, but my dissertation advisor was a little-known guy named Eugene Fama. We've heard of them. I was his TA for two years. I grew up in the EMH. I love that you guys just say EMH and your audience knows where you're talking about. That is not the norm for me when I talk about these things.

I was not a perfect efficient marketer even back then. Neither is Gene, by the way. Gene's not a zealot. About third week of class. I know this because I took the class three times. I didn't fail, but as the TA, I didn't. I sat through it, three full years. And like the third week, he always tells the class, markets are almost certainly not perfectly

efficient because Gene's a brilliant guy and recognizes that perfection is a really stupid idea. He's been on the show, by the way. Oh, I didn't know that. Yeah, yeah. And I don't know if that came up, but he's always very honest about that. He probably thinks they're considerably more efficient than I do these days. And I think I probably think they're more efficient than maybe the active average retail trader. But I wrote a dissertation for him on the success.

of price momentum. That is not a very Gene Pharma dissertation. A Gene Fama dissertation is, I've studied price momentum and it loses gobs of money and these idiots on Wall Street do it anyway. That's kind of a fishing market. Look how silly they are. And he was great about it. I remember I kind of mumbled. I'm like, I want to write a dissertation on price momentum. And by the way, I find it works very well. What was that cliff? It works very well. And he said if it's in the data, write the paper. So I've drifted at least a little bit further. from efficient markets even way back then. Now, price momentum, I'll do a lot of segues.

You're going to have to stop. No, that's good. I do parentheticals within parentheticals. Price momentum is often thought of as this index of irrationality. It can work for two different reasons. It can work because of, yes, feedback loops, chasing. Line goes up, just like you said before, and people pour in, which isn't very connected to reality. It's just chasing returns.

But it can also work because of what the behavioral finance. people would call underreaction. News comes out that should move the price by so-and-so. On average, we have found, I say we, it's the royal we of academia and private researchers, that the price moves the right direction, but it doesn't move all the way. So if you trade on that, there's still a little bit to go. That's a quaint thing. You wouldn't want to bet on one stock that way, but if on average that happens,

and you could do that through observing the price or the fundamentals, there's usually a little bit more to go. So it's not always irrational momentum. But here's what I observed. In the very beginning of AQR, AQR launched in 1998 after we had a fabulous run at Goldman Sachs. There's survivorship bias in this. You don't get to start your own billion dollar hedge fund

unless you have a fabulous run at Goldman Sachs or something equivalent. We had a good first month. You know how the story's going to go when you say you had a good first month, right? And then? That first month, by the way, it was August of 1998 when the S&P was down 20% on the Russian debt crisis. and we're doing high fives. Like we say we're market neutral and we're up a little bit in a crash. And it was my first of many lessons never to high five in this business.

When you fully retire and divest, you get one high five. Well, no, it's the end when you do the high five. I'm a quant who is also superstitious, which, if that's a contradiction, what the heck. But the next 18 months was the crescendo of the famous dot-com or tech bubble. And that was not kind to us. It was particularly not kind because we decided to start with a extremely aggressive market neutral fund. So momentum helped, as you can imagine in a bubble, but value was just destroyed. And that was most of the model back then. Things have really broadened out. We hit spreads between cheap and expensive. This is something we invented at the time,

and now a lot of people do. People said sort stocks on valuations go long, the cheap, short, the expensive. But the very obvious question of how cheap and how expensive, are they sometimes pretty tightly clustered, are they sometimes wider? Does that mean they're better or worse going forward? Was not asked before. So we invented this measure and the spread between cheap and expensive for 50 years had looked like a well-behaved series. It moved around a fair amount. And then I'm drawing on my hand, if I only see the video. Oh, yeah. I should mention this is an audio medium. And then in late 99, 2000, went to just way wider than anything ever seen for 50 plus years. We also showed that historically, those were better times.

You never saw that before, but when it was wider, were better times for value. We stuck with it. We made money round trip. Life was good. If you had asked me after the round trip, which was harrowing, you know, even if we love our process, you never want to start a business with poor returns. If you asked me at the end of that, which was probably a couple years later, 2003. Do you think you're ever going to see that in your career again? No one asked, thank God, because I think I would have gotten it wrong, but I think I would have said, oh, probably not.

Hopefully I wouldn't say definitely. No one who does what any of us do for a living should say definitely. That's a bad word in markets. 40% is the favorite term, right? It's always a 40% shab. But A, it was the craziest thing numerically in 50 plus years. B, the question presupposes. It's built in that I and people of my cohort will still be a right. round, right? And we'll probably be closer to in charge. So how's it going to happen again? And then it happened again. Even before COVID by late 2019, that spread between cheap and expensive was approaching.com extremes. And then it blew past it in COVID. It went to what I, in a geeky math joke that no one ever gets called a 125th percentile. There is no 125th percentile. It's just a new

hundredth percentile. I'm trying to convey that it went further. And we survived that one. too. We suffered somewhat, and then we made more than all of it back, round-trip, good. Most of the time, we really don't look like value investors, by the way. Only in extreme bubbles do we seem to have that property, and I think it's smaller now than it used to be. The last five years have been quite strong for us, and it's not been a very good value market. God, there's so many questions that I have that I want to ask that are sort of embedded or related to your answer. I'm going to ask, actually, just like a very narrow question, sort of skip ahead and something, though. You Make the spread between the most expensive and the cheapest stocks.

Yes. How much easier it is to simply compile that spread is it today versus the technology that you're working with when you're starting your career? For that, I got to disappoint you and say not that much easier. Okay. We had the databases. Okay. A lot of technological advancement is about speed. Okay. And new data sets, what quant will call alternative data is something we're very into.

but the classic data, if you're looking for price to sales ratios, I'm old, but we had telephones. We had telephones in Bloomberg's. But that does bring us to me observing these two episodes. And I stepped back and I asked a question, what the hell happened? Why did we see something crazier than 50 years? And then why did it happen again? And that led to this paper, the less efficient market hypothesis. I do believe that markets have shown, and I think I have some good guesses.

as to why that they are more susceptible to bouts of crazy than they used to be. Just before going further, one quick definitional question. Efficient markets, because one way you could define whether a market is efficient is, are these securities disconnected from what you would say, the net present value of the all future free cash flows? Another way that I often think about it, but I'm no quant, is are there obvious opportunities for the active manager to make money? Sure. Just for our purposes here, how do you define market efficiency? Much closer to the first one.

Okay. The practical question of can you make money from these is if there are big deviations from fair value, and this relates to some stuff I did in a less efficient market hypothesis, if there are big deviations, the classic, and you'll notice a giant caveat, if you can stick with your position, not a small thing, you will make money. There are opportunities. And in fact, I think a less efficient market, in that sense, in your present value, sense almost has to deliver bigger opportunities for people who can stick with it. It also makes

it considerably harder to stick with because the extremes you have to live through and the length of time those extremes can go on for. One thing that I'm dying for an academic to take me up on this, I'm too old to do the math on this. But I've never seen a model that looks at pain, disutility, the negative of losing money in terms of how long you've lost money for, not just magnitude, right? And in real life, I can tell you a drawdown that is one and a half times bigger, but with six months instead of three years, is ridiculously easier to live through. So the bigger disconnect from reality, again, I haven't even told you why I think it's going on. But if I'm right, that we have these bouts of it, it's not necessarily everyday things are crazy, but these bouts of it

is a two-ed sword. It's a bigger opportunity for people who can stick with it. And, and and it's harder to do. And I find that, not that what I think is fair is particularly relevant, but I find that really fair, harder to do, but more lucrative if you can do it. I don't think the efficient market idea of buying what is fundamentally mismatched against its future cash flows adjusted for its risk, forecasting those cash flows, what risk really means. We still have some open issues there.

Yeah. But I don't think that will ever go away unless markets are perfectly efficient. But how easy it is to identify and how easy it is. is to stick with. I think crazy markets make it easier to identify what to do and harder to do it. I like that. I do want to ask you why you think this is happening. But before we do, this is a very basic question, but sometimes the basic questions are the most interesting. But when you say it's painful to try to stay rational during these bouts of irrationality, why is that exactly? Because I get that there are funding costs. I get that there are carrying costs. But on the other

hand, you're a big hedge fund with deep pockets. This is presumably what investors, are paying you to do? Is it just the emotional trauma or stress of having to explain to everyone why you're taking the position that you have when it's not paying off yet? That is a big part of it. I have a running fight with one of my co-founders who never gets upset. I'm always upset. But I'm more upset when we're losing money. This is sort of our dynamic. I get more upset than Joe does. Where he'll come in my office during a bad period and I'll be upset. It'll be a bad day in a bad period. Why are you upset? And I'm like, well, because people are yelling at us and I, we're losing money. And he's like, but you're pretty sure we're going to win, right? I'm like, yeah. He's like, and you have all your own money and your kids money in this, right? So you're not doing anything different. Do you wouldn't do that if you weren't? I'm like, yeah. He's like, so we're going to win. It's just a question of when. Why do you care? And I look at him like he's from Mars and go, why do you not care? And we finally figured out, it's kind of obvious that I talk to clients a lot more than he does. It's a lot easier to have that attitude when you're just sitting in your office.

Also, we are not immune from this. Even if people love you and think you've done well for 25 years, you have a bad two years. You get redemptions. And sometimes they're of serious size. We fell by like half over about three years. And that's not fun. You have to shrink your firm. You have to let some people you love go.

So there is some real pain that goes with it. That period did not shake my confidence in the actual investment process. I feel bad. I think I did better than our investors because I kept adding and saying take the ball up on what I do. Not everyone can do that. And I know more than they know, not in a weird insider sense and just I should be more confident in my own process than anyone else's. But it is excruciating. The amount, I remember someone I admire tremendously when Stan Drucken Miller retired to run his own money. He's still very active in markets, I think. He wrote a note that resonated with me because I forget the exact details, but the essence was,

It's too painful and too upsetting to run client money. The man never had a down year. And I'm like, if Stan can't take it, those of us, and we've had a lot more up years than down years and life's been good or I won't be sitting here. But we've had never three, but two plus years of pain. And I'm like, if Stan can't take it, man, this is harder to do than it looks like. The whole world thinks you're stupid when you're losing money. No matter what you can point to, no matter what evidence you can point to, I should say the whole world. Your mom still likes you.

But a lot of your world. And no, a lot of investors stuck with us, and I love them for it. But you do lose a lot of people. So it's not fun for a business to go through that. Intuitively, though, is to your point measuring the disutility of long drawdowns versus deep drawdowns. This must be a very acute thing for anyone who's not just managing their own personal portfolio for the reason that you've just described. Yeah. I've talked about this with other money managers, this idea of length versus severity.

And I've never had one who's not like, yeah, length is much worse. Well, one of the things is when something goes on, it's often a similar story for two and a quarter years. Right. So you go back after six months where rationality is getting absolutely punished. You get a lot of sympathy from people. You show them, look, bigger bargains, we're right, we're going to be right. You go back six months later. They're like, okay.

You go back six months and six months later. later, they're like, you're just saying the same thing. And eventually they must just think, maybe you're a dinosaur. Like, maybe you just don't get this new wave that's going on. Is it possible that we really have a regime and change? Again, it's not everyone. We have the tremendous amount of investors stick with us. We have ones who double up, who get it, that those are often opportunities.

But you shrink when you lose money for a while and you grow when you make money for a while. It's the ironclad rule of this business. And sometimes it's just backwards. I'm Francie Lacquan, an award-winning journalist. and I've got a new podcast, Leaders with Francine Lacqua from Bloomberg Podcasts. I've interviewed everyone from heads of state to fashion icons about the news of the moment. But I've always been curious who are these people as leaders. I don't think there's one right way to be a leader.

Make decisions. A poor decision is always better than no decision. Listen to new episodes every other Monday. Follow leaders with Francine Lacroix wherever you get your podcasts. So one of the things that has been happening recently that has changed from 2015 when we started doing this is retail participation in the market. And it just feels like it is such a big thing for retail investors now to use things like options, even, you know, one or zero day options, the kind of stuff that you might more traditionally associate with someone running a hedge fund. Now they're trading. No, we're not stupid enough to trade zero day options. All right.

Some hedge funds then. Does the increased presence of retail in the market change the way that you do business at all? I think it contributes to what we're talking about of this dislocation. I hate this because I sound very elitist when I diss on retail. But there's a lot of academic work that shows retail on net loses. Net is important. They go through periods where they win. They go through periods where they win a lot.

You know, if you buy a meme stock and a triples tomorrow, you don't always lose. but on average retail transfers money to Wall Street and to institutions. So if there are bigger force in markets, you're going to have more of that going on. And it jives perfectly because that would raise the opportunity. If they're generally on the wrong side of things, but there are more of them, many more of them than they used to be. They can be right, even if they're wrong on the facts, they can be right on the numbers for longer than they used to be.

So I do think this is part of it. And I apologize to all the really smart retail investors. Anytime you talk about averages, you're dissing a whole lot of people who don't deserve to be. But on average, retail loses. Zero day options, my God. They're making exactly the people they claim to despise on Wall Street, very rich by trading these things.

And it's just fan duels. I love that. All right, we got to get to the question of why, right? But one last question sort of leading up to it. How do you establish that? What do you look at in the market that right now you say this is a less efficient market than once it was? What is the measure or is it just feel or you just feel sort of crazy like all of us observed? Most of it is quantitative.

I do start with this thing I've talked about already. This spreads between cheap and expensive. I don't just look at spreads. I wrote a piece back in 2000 during the tech bubble. It's called bubble logic. It never got published because I tried to make a book out of it. And the bubble came down too fast for me. That was good for my business, but bad for my author career, where I didn't just look at

price multiples. I looked at it in a more holistic sense. What growth do we need to justify these multiples trying to come up? The only time I will use the word bubble is when I've tried very hard, and it doesn't mean we'll come up with the same answer, but this is the framework I use. I've tried very hard to come up with assumptions, even if I don't like them and think they're at the outer edge of possible, that could justify these. prices. And if they don't come close, twice in my career, I've been willing to go, no, I'm willing

to use the B word. I should tell you right now, when it comes to within stocks, the whole market is a different issue. The market is quite expensive right now. But when it comes to this spread between cheap and expensive, I'm not using the bubble word. That same measure, it's not the 125th percentile anymore. It's the 77th percentile. It's wider than on average, maybe making it a little more attractive if you can stick with it, that big if. But I don't use the word bubble for 77th percentile. I used it for blowing through. So five years ago, I was saying it's a bubble. Now I'm just saying this is a little bit of an odd market. So I want to ask you more about the bubble, but we keep touting that we're going to ask you the why question. So let's ask the why question. Okay, why are markets

becoming less efficient? Okay. Well, first of all, and I say this in the piece, a lot of conjecture going on. This is an op-ed. As statisticians, we'd like a 50-page? Yeah. A 50-page op-cha. A 50-page academic op-ed. Look, this is our first time doing this. You will be convinced I could do a 50-page single-spaced op-ed.

I believe you. I know you can. It's just as a quant, as a statistician, you'd like to see 100 bubbles, have stats on each one. They're all, they'd rhyme, but they wouldn't be exactly the same. You'd tease out what's going on. If you see two in a 35-year career, you're not going to be able to. prove this statistically. But I believe in my conjectures. I'm not soft selling them. I just want clear that nobody is going to walk away saying he proved it. I list a few reasons in the piece

why markets might be prone to bouts of bigger disconnects from reality. My second favorite is one I'm sure you've talked about, I know I've listened to you guys talk about it, the rise of passive investing. I am not a passive hater. They're really smart people. Mike Greens out there, the amount that guy hates passive. I don't know. The Middle East has never seen hate like the Mount Mike Green hates passive. But again, he's a smart guy. He makes interesting arguments. I'm not that guy. I think passive has been a huge positive for an investor welfare. I actually was lucky enough to be fairly good friends with Jack Bogle. And he was a hero of mine. We had a podcast briefly. And he came on. And he came on. I'll find out. In fact, I'll tell you part of that story in a second.

So I'm not a passive hater. But here's what we know. We know the whole world. cannot be passive. And when I say passive, I mean in a Jack Bogle market cap weighted sense. Sometimes people use passive for people like us and they really mean rules-based. And that's not how I use the word passive. You mean someone who owns the entire market? Yeah, we're long, short and lever. How you get to passive on that? I don't know, but some people do. I own the entire market. We had Jack on the podcast and he absolutely agreed everyone can't be passive. Now, of course, being Jack Bogle, he thinks at that point the marginal investor should still move to passive, But he recognizes the obvious fact that if 100% of the people are not looking at prices,

nobody's looking at prices. Who's figuring out if Nvidia is worth more or less than the corner drugstore, right? So the market gets very weird there. We don't even understand what happens. It's a singularity. I use a physics analogy. We don't know what happens there. PhD students in finance, and I used to be one of these, we'll stay up late at night

in their cups talking about what happens if everyone was passive. what would it even look like? We know it's very weird, and I doubt all the weirdness happens between 99.999% passive and 100. So we're on a curve. We're a lot more passive than we used to be. Even that, you're probably aware it's hard to measure exactly how it's just passive. Direct passive, true market cap weight you can measure. But what about people who take, you know, 80 basis points of tracking error?

They're kind of mostly passive. Or people pegged to like custom indices now. It's kind of funny. You guys remember the princesses? bride. Yeah, yeah. Remember mostly dead? Not fully dead. You're mostly passive. So I think fewer people thinking about prices, fewer people willing to take the other side when things get a little crazy. If one side really starts to get crazy, there are fewer people out there. That has to exacerbate these swings. My actual number one reason, though, you talked about the gamification.

for me it's the overall, I'm going to sound like a very old man yelling at the sky on my lawn right now. You know, get off my lawn or yelling at the sky. Some Simpsons thing. Yeah. I'm Abe in this case. Old man yells at cloud. Yeah. Exactly.

Thank you. Social media and the broader environment that you guys were talking about. I don't want to do politics except to say, I don't think you find many people. There'll be some, but I don't think you find many people who don't agree with the idea that this environment has made. our politics worse and more dangerous. Confirmation bias. We live in our own bubbles. We have algorithms that push us further and further towards, you start out as a moderate belief, but it keeps pushing you towards extremes. And pretty soon you're saying stupid things like, hey, that Tucker Carlson, he's a good guy. All right, I might have revealed a little politics there. Yeah, you did a bit of politics

there. So all of that adds up to making our politics worse, more prone, in particular, to swings and extremes. Markets are not arbitrage mechanisms that's sometimes misunderstood. They're voting mechanisms. The price is a weighted average vote where the weight is by dollars. I'm lucky enough to get more vote than the average person, and Warren Buffett gets a lot more votes than I get, if we all disagree on opinions. The reason it's not an arbitrage mechanism, and here I'll get a little geeky. Imagine you're reasonably sure this thing is mispriced in that Graham and Dodd sense.

and you think it's massively mispriced. Trading enough to make it a third less mispriced is not very risky to you because it's not that big a trade and it's very high expected return because it's that mispriced. Now you've moved it back to a third or maybe half. The next part of the trade is much riskier to you because you already have the trade on. So you're just adding more. Oh, I see. And it has half the gain because you've already closed it by half.

So arbitrage will not take things. If on net more people believe something stupid, stupid's going to win, and we're going to be at least somewhat off of real prices. And I find it remarkably easy to believe that this same environment that makes our politics go a little crazy, for markets to be efficient in any degree, not even perfectly efficient, this famous idea of the wisdom of crowds has to be helping us a lot.

The hypothesis that we're all geniuses is never going to fly, right? So the wisdom of crowds, and you know it well, says even if on average most people don't know the answer, the stupid answers cancel and the right answers don't because they're the same. My favorite example of that, and this might be dating myself, is Regis Philbin and who wants to be a millionaire, right? Remember the show, multiple choice show? It's frightening that you're describing that as old, but I guess it is. Yeah. It's been off the air. Sadly, Regis has passed away.

It's old. Sorry. Sorry. But you had to answer multiple choice questions. If you miss one, you're out. They start off ridiculously easy and they get harder and harder. You had multiple cheats, like three cheats. Phone a friend.

Phone a friend was almost useless. A, your friend usually wasn't much smarter than you. And B, people, at least to my eye, I didn't watch every episode, but seemed to choose friends who knew the same stuff they knew, right? You really want to choose a friend who's in like a totally different field. I think in some countries, the friends also had a tendency to deliberately give the wrong answer because they just didn't want to see their friend actually win money. The most cynical phrase ever is nothing succeeds like a friend's failure.

There we go. I don't believe. I'm not condoning that, but it is out there. The other one was eliminate two of the wrong answers. That's great. Obviously, even if you have no idea, you go from one out of four to one out of two. The other one was poll the audience. And at least to my non-exhaustive examination, it seemed to work pretty much every time.

even if the question was hard. Because imagine you have 100 people in a room, 10 of them know the answer. The other ones are guessing. The 90 distribute evenly over the four. Maybe not perfectly evenly, but roughly evenly. The 10 all land on B. So you pick B because it's bigger. Work pretty much every time.

There's a crucial assumption in that. The audience has to be relatively independent of each other. And they did that. They weren't talking. It was silent voting. If the audience all gets to talk, maybe the 10 convinced the 90, but maybe they don't. Maybe a demagogue with a better Twitter feed convinces everyone. And if you ruin the independence, and I think have we ever come up with a better vehicle

for turning a wisdom of crowds into craziness of mobs than social media? I'd be hard pressed to describe it. I find this to be very compelling. James Sir Wickey at the New York Times. He came out with that book, Wisdom of Crowds in 2005. But that was right before social media blew up. And then, you know, there's that famous book in 1841, extraordinary popular delusion and the madness of crowds. So we've always sort of understood that crowds can be both mad and wise.

And I find this very interesting, the idea that perhaps the linkness of the crowd is what sort of flips it from wisdom to madness. I think that's exactly it. You maybe can come up with counter examples. But I think most of the time, if the crowd is making independent decisions, and this can go for political voting, you can go for markets. You're going to get some degree of wisdom. When the crowd is all making a unified decision, it could still work out. But you are much more susceptible to what the quant technical term is cray craye.

You can get the news whenever you want it with Bloomberg News Now. I'm Amy Morris. And I'm Karen Moscow here to tell you about our new on-demand news report delivered right to your podcast feed. Bloomberg News Now is a short five. minute audio report on the day's top stories. Episodes are published throughout the day with the latest information and data to keep you informed. Yes, there are other products like this from a variety of news organizations, but they usually rerun their radio newscasts throughout the day. That's not what we do. We create customized episodes that can only be heard on Bloomberg News Now. And we don't wait an hour

to publish breaking news. When news breaks, we'll have an episode up in your podcast feed within minutes. So you're always getting the latest stories and developments. Get the reporting and the context from Bloomberg's 3,000 journalists and analysts we're all over the world. Listen to the latest from Bloomberg News Now on Apple, Spotify, or anywhere you listen. I want to pivot a little bit and talk about AI. You've written or you've talked recently, I forget exactly the word that was used in some of the headlines, succumbing to the machine. Surrendering ourselves. I regret saying that.

We try to talk a lot about AI. I still don't know exactly what it means in any context. What does it mean to surrender to the machines in the AQR context? Well, first, not every journalist has Bloomberg's high standards. That's right. I am reasonably certain I said partially in that. And the word partially got dropped. Even skipping all the details.

If you're going to use AI in your process at all, almost by definition, you were going to lose a little intuition. And it bothered me for a couple years. I think I slowed us down on AI by a year or two just by saying, you know, we've always prided ourselves on the balance of we intuitively understand why we think this makes money and the evidence that it makes money. And when you go to AI, you're normally giving up some. Not all.

We actually do try very hard to figure out why we think this works or doesn't work. But if you weren't giving up some intuition, what the heck is the AI doing? If it's simple and you could have just seen it with the naked eye, it's hard to imagine. It's helping. But let me give you a concrete example. We like good momentum.

We like it in price. We like it in fundamentals. One way people on the quant side have tried to measure this for years is something like earnings calls, trying to decide if earnings calls are good news or bad news. And if people underreact to good news, you want to buy when it's good news. Is this just like how many times people say great quarter guys or are you looking at something else? It's going to sound about as silly as that. Okay.

You build up tables of words and phrases with numerical values and then you say what's the numerical score of this? And they can be much more subtle than this. I'm going to use a real simple example. The word increasing plus one. Right. And I'm sure you see the flaw. If the actual sentence was massive embezzlement is increasing, you know, our best. on that one. The amount of fraud we're seeing in our private credit deals is increasing.

Quant can survive looking stupid a lot if that's 47% of the time. If 53% of the time it's getting it right, fine. And those things had some efficacy, but they weren't great. What we do today is we train ML. It's called natural language processing. It's the subfield of ML to analyze corporate statements. But what it does, and this is going to be the geekiest thing I'll say, It represents every corporate statement. It looks across them and represents them as a set of numbers, what the geeks would call a vector of numbers. Then what we do is empirics to say, all right, we have 50 years of this across many firms. We have different vectors or numbers for every earnings call. What combination go long when the first number is good, short when the second number is

high, blah, blah, blah, what best combination forecasts? That seems to be. be correlated to what we were doing before, these word count things, just considerably better. It does a better job than word counts. Language is very nonlinear, whether something is good, whether that word increasing is good. AI is not perfect, but its chance of figuring out if increasing was good or bad, especially when it's trained on tons of these is better than us. Here's where you lose the intuition, though. I skipped a step.

I like to do that. It's fun. It's sneaky. if you ask myself or even some of the younger people who are much more tooled up on machine learning than I am these days, what does that vector of numbers actually mean? You often get a, we really can't tell you that. We can say it's summing up in a mathematical sense, the information content of that. We still get intuition because this indicator acts like a short-term momentum indicator. So it's picking up what we wanted to pick up. And it's also done

fabulously well for us for multiple years in real life. But we are giving up intuition at one stage that we used to not do. So my answer was meant to be much more subtle that there are give-ups in intuition when you move to something like machine learning. They almost have to be or else again, what are you doing? I was uncomfortable for that for a while, so I was starting to do Mia Culpah saying that I slowed us down. And it came out as, well, I used to hate this, but now I have no job, the machine runs it, which is probably going to be true in eight years, but not yet. This reminds me actually of something I wanted to ask, because another big multi-year decade trend is the rise of the multistrats. And at AQR, as I understand it, you have a multi-strat model

in there, but it's more centralized than some other places. Can you go into a little bit more detail? Sure. They get used interchangeably, but I think there's the difference between multistrat and multi-manager. Okay. Multi-Strat just means. means I wrote a dissertation on choosing U.S. stocks. We have applied similar things to stocks around the world to currencies, to commodities, to directional bets through trend following. They are correlated, but low. So we think of these as different strategies. And we think a set of our strategies is better than a single one. Or, you know, it's just the power of diversification. So we're big believers in a particularly if you have a common philosophy, so it's not just

just fitting the data, it's fitting into an overall theme of what you believe in. We're big believers in multistrats. What we share with Multi-Manager is that belief in diversification. Multi-manager is what it sounds like. We are one team building these. We might, of course, have little separate teams at AQR, but we're one firm building these. They are farming it out to different people. To be frank, if you had told me their business model 10, 20 years ago, I would have been very cynical that it worked. If you told me, A, what the total fees are going to be when you add up everything that's passed through and a fair amount of them, and they vary in how quick they do this, if something's not working for

what I would consider a very short while, they stop doing it. And I know there are a lot of low to medium sharp ratio risk-adjustive return strategies that are really good long-term, but have bad periods. So if someone told me that model, I would go, you're going to charge a ton and you're going to throw out people who have a bad two quarters? No. And there have been people who've obviously proven me wrong. And I'm humble about this. I don't fully understand why. They must be very, very good at actual selecting the alpha. Right. And this is something that I, we don't do. We're internal quants who build our own models. We've never, maybe it'd be interesting one day.

We've never tried to apply that to choosing outside people. But plenty of people, by the way, have tried to start multi-manager and failed. So it's not like you just apply this, charge a ton, hire a bunch of active managers and fire room if they have a bad two hours, just automatically works. It does seem, though, that, like, talent is the thing that's, like, capping multi-strat expansion, right? That's the limiting factor. They vary, but some of the major ones are giving back money.

Even if you believe in this model, and I, even from afar, I have to be a believer. I've seen the results are too good for too long, in my view, to have a decent chance of randomness. I don't exactly know what they're doing. I'm still waiting for Ken and Stevie and Izzy to send me their exact process. That would be wonderful. Izzy doesn't share it with you? No, even better if Medallion would send me their exact process. But I respect it.

But the amount of individual alpha from teams that can be out there has to, all alpha is finite. but has to have a decently tight cap. And the ones I've seen are fairly disciplined about this. And again, it varies. I can't speak for everyone. I know we compete for talent with them much more than we used to. They have non-quant parts of what they do, but they have quant parts of what they do. And for young quants, they're often considering an offer from AQR and Citadel to pick just one example.

We win our fair share. We lose our fair share. I think we probably pay a little less in the short term. we don't fire you if you have a bad week. So some people are... And I'm not picking on Ken. He's not doing that. I'm just saying...

No, but we know that... It's a more cutthroat environment. We've done a lot of episodes on that environment. We know, like, they cut pretty quickly if you lose money. I want to go back a little bit to the connection between sort of AI, machine learning, and interpretability of the factor. You know, the quintessential... You know, the risk is, of course, that you find something that works strictly from data mining,

right? Tickers that's... Start with B tend to rise on Tuesdays. I usually use the CEO's middle initial. Yeah, stuff like that. We know like, but, you know, even when you were talking about momentum and you said, well, there's two reasons theories for why momentum can work. Let's just go back to the canonical like quant factors.

Is there complete consensus about why these factors work? Not at all. Okay. You start out with great. Because like intuitively like cheap stocks go, but my impression is that there's even still disagreement about why. Yeah. You start out with The Great Divide.

And when they split the Nobel Prize between Gene Fama and Robert Schiller, my co-founder and I wrote a piece on the cover of institutional investor with that title, The Great Divide. Lars Hanson also won a share of it. I shouldn't leave him out. But Schiller and Fama were juxtaposed as the efficient market guy and the inefficient market guy, which is fairly close to – I don't only either of them are zealots, but it's fairly close to true.

So you start out if something has worked historically and made a lot of money, over a long period. You start out with what you led with. Is it data mining? Let's say you convince yourself it's not. Okay. Let's just put that way. And that's really important that I'm not, you know, poo-pooing that step.

You got to do that. Then you've got to ask yourself why. The two main contenders are a rational gene pharma kind of market where some stocks are riskier than others. And whatever you're using to say go long these and short these is loading on that risk. And if something is risky, you should make more for investing in it. One of the problems is identifying, A, there are subfights within these fights. What do we mean by risk? Is risk beta like the capital asset pricing model? This is where finance academics are like every other

academic because some fights within the fight. Define your terms. It's Talmudic at this point. But is risk beta like the capital asset pricing model? Well, the capital asset pricing model is a beautiful, elegant model that has failed everywhere it's ever been attempted. Is risk more multidimensional? Is risk something like what happens in a great depression, which is very hard to measure? The other argument is markets are not perfect. People make errors and it's behavioral finance. So if a cheap stock beats an expensive stock because it's inherently adding some risk that you can't diversify away, that's a gene pharma explanation. If it beats inexpensive stock because people went too far. They went past the Graham and Dodd point. Yeah. And if you can hold it,

you make money when reality sets in. That's more the Bob Schiller point. I love Gene. He's my hero. I have probably drifted from 7525 Gene to 7525 Bob over my career. World's complicated. One of the hard parts is everyone wants to win, but both explanations can be true and they can be true at different times. Life isn't so so simple. But those are the two biggies. But once you get into behavioral finance, Now you can fight about why. What behavioral bias? Notice the, you could sum up the two reasons I gave you for momentum as underreaction and overreaction. Overreaction is chasing, right?

Undereaction is the information came out, didn't move far enough and I hopped on that bandwagon. You know you're in a little bit of a dodgy area when your two best explanations sound a little like antonymns. Now, I'm being intentionally. It's so funny. We know it works. We're just not sure whether it's literally opposite things or intuitively opposite things. We know it's one of them and there's some debate. I'm being a little too negative. You can actually, they're good papers teasing out. The underreaction is easier to show. You can actually show the fundamentals do catch up.

I think the overreaction probably kicks in more in bubbles where there is just more feedback trading. Again, both explanations can be true and they can vary in their intensity over time. The overreaction might have a very small part of it, except plus or minus two years around a major bubble, when it may be the driving force. So we have to get really comfortable. We don't have to have the be all, end all, single explanation. But if there are a few good explanations and no real counter ones and we have really strong 50-year empirical results, yeah, we're okay with taking some bet on this. You never want to put all your money on one thing. That's not a quant thing.

I'm going to try to connect a bunch of the sort of decade-long megatrends that we've been discussing. But I'm thinking behavioral finance, gamification of markets, big data, AI, all those fun things combining into sports betting, which seems to be getting bigger and bigger than ever. And the reason I ask is because I was talking to someone in a bar the other day. And they said that AQR was doing more sports betting stuff. Someone in a bar was talking about AQR doing sports betting. Yeah, I know. Well, it probably says more. about the bars that I'm in and the people I'm talking to than anything else. But yeah,

they mention that. I'm picturing the Star Wars canteener right now. I got to tell you. So I'm just curious. Is that a thing or are you maybe interested in the sort of prediction market aspect of things nowadays? Susquehanna is getting into it. I think no one, again, tells you exactly what they're doing. My sense is they are into it and are good at it. This is insipion for us. We are considered. and looking at it. We have a long history, Toby Moskowitz, who's a Yale professor and an AQR partner. Great deal, by the way. You get paid for two full-time jobs that are 80% overlap. I'll keep that in mind for the future. If Toby's listening, just, you know, your lucky man.

He wrote a book called scorecasting that was all about sports. But the sports analytics, the money ball type stuff looks a lot like what we do. It looks a lot like people are not pricing this right. people. I wrote my most downloaded paper ever, which is really a little annoying because it's like the only one I've written outside of my field was on when to pull the goaltender in a hockey game. Oh, yeah. And I wrote it with a colleague and friend, Aaron Brown. We built a model and we came up with you should be pulling the goalie if you're losing by one with five or six minutes left, not with two minutes left. And I'm pretty sure we're right. We spent a lot of time in that paper saying, why aren't they doing it and making analogies to investing?

You know, when people know the right thing to do but can't do it because the public opprobrium, if they get it wrong, will be much worse. Right, like, aren't they saying, like, football teams should go for it on fourth down a lot more than they do? That's like one of these things that have- And they have started to. Yeah. There is a feedback where some of these things are slow, some of them are fast.

Baseball has largely absorbed a lot of these things. You know, on-base percentage was one of the major insights of Moneyball, and nobody's missing that one anymore. But we still think particularly in the betting markets are probably not as rational as Toby's scorecasting book. So we do think there might be opportunities there, but I don't want to overstate it. Just because I am cynical about online sports betting, I got to tell you two things. I was always a very libertarian guy. I still think I am.

But a couple of things have made me less libertarian. Children, the existence of my children will make you go, yeah, maybe people shouldn't be able to do anything they want. want when they turn 21. There should be some rules. And sports betting and maybe walking down the New York City streets and getting a contact high. Sports betting, we're already, you know, some of these scandals were starting to see. Yeah.

How do you not have them when, you know, some of these people make a ton of money, but they don't all make a ton of money. And there's a lot of money in sports betting. Anything you do for entertainment is fine if you do it at entertainment size. If you go to Las Vegas and you go, the odds are on the house's side, I'm going to lose $500 over two days, probably if I win great, but it's going to be fun. It's factored into the cost of a Vegas vacation. But like I have a whole bunch of 20-something year old, mostly guys in my extended family, a lot of them are sports betters. And they're not even, it even pisses me off that

they're not even rooting for their teams anymore. They're kind of rooting for this player to score on their prop. It is a strange dynamic. So do I think that gamification is highly related? I think if you go look at the Robin Hood app and Fandul, I think you will find they are far more similar on their feedback and how they treat things and in how their investors do on average. Particularly sports betting, you know investors lose on average because the sports betting companies make money. It's like saying on average insurance is a bad deal. You know how I know? Because insurance companies are profitable. Doesn't mean it's stupid to do because if it's in a risk you can't tolerate. That could be a fair trade.

price you pay for peace of mind, which is something I've come to appreciate over the course of 10 years. There is zero chance that the average online sports better is making money. Yeah. I just have one last question. You know, I think if you looked at like some particularly crazy times in markets, here's something that's changed very much since we started the podcast first 10 years ago. I think if you look at some crazy times in markets, whether it's, you know, 2019 when some

of these measures were getting extreme, or just, you know, the SPAC mania of 2020. one theory that one might have offered is ZERP. And he's like, oh, this is, now we have, you know, we haven't been a ZERP for a long time. And yet some of this sort of speculative mania crazes has not gone away. How surprised are you that the move from zero percent to say five percent or whatever didn't have more of a sapping effect on some of the behavior and speculative fraud or just sort of animal spirits in these markets? I am mildly surprised.

Okay. I left this out, but I actually had three possible reasons in my paper. And the third one was super low interest rates for a long time for why things can get crazy. Now, the counter argument is once you break people's brains, they don't necessarily repair themselves instantly. So I think it was probably a contributory factor. Again, very hard to prove. A lot of people in 19 and 20, when the spreads being cheap and expensive were here, were saying it was a super low interest rate environment. And gross stocks have more cash flows in the future.

future. Low interest rates means they're worth more. We did the math on that. It explained like 2% of the extra value spread. And in 99, 2000, interest rates were quite hot. That's right. There was no zero. 99. So it's not a unified field theory explanation. Do I think it helped kickstart some? And again, we're in the soft guesswork. But I think these are educated guesses. And I listed it as one of my three. I think it certainly kicked us off on some of these things, certainly loosen the bounds of rationality, absolutely free money, we'll do that. I don't think it's not human nature that they take away ZERP and everything comes back. Would I have thought it was a bigger effect in going back to, you know, at one point we hit about 5% on the 10 year, almost 5%. Would I have thought that would

have mattered more? Yeah. Only in 2022 did we see one ugly year over that, but it has mattered less than I thought it doesn't mean it will never matter. All right. Cliff Asmus. Thank you so much for coming on odd lots and kicking off our 10-year anniversary celebration. Yeah. Thank you so much, Cliff. That was fantastic. That was really a pleasure. I had a lot of fun.

Thank you. Joe, that was so much fun. That was great. Literally the perfect guest. Literally the perfect guest. One, well, there's so many things that stuck out from that conversation, but one of the things that stuck out from the conversation is this idea about the thing that flips the wisdom of crowds into the madness of crowds, right?

And the idea that maybe the wisdom of crowds theory works, as long as everyone is sort of isolated and independent and making their own choice off of the information available to them. But it starts to fall apart when everyone is tied to everyone else and sort of in the same social network. And, you know, if you think about one of the dominant themes in markets in recent years, it has been people hurting into the same positions, right? I found that to be really fascinating. I think that's a, it makes it a lot of intuitive sense. It probably can explain a lot of things about the world of politics, about the world of markets, etc. This idea that we're all just sort of one connected global village, as Marshall McLuhan put it.

We're all just gossiping with each other. All the time, but just constant talk, talk, talk, talk. No, that's a very interesting idea. I also thought it was interesting the idea of length of drawdown versus depth of drawdown and the former being more painful, which strikes me as something I hadn't really heard anyone talk about that before, but especially from the perspective of a manager of other people's money. It's a very highly intuitive. That makes a lot of sense to me that, okay, like, yeah, you had a bad quarter, whatever. Eventually, you're like, oh, your ideas are just out of date. It's been three years since you made money, maybe time to rethink some of your fundamental

assumptions and how powerful that must be. That's one thing I've learned over the course of 10 years. Yeah. The other thing was the idea of markets not necessarily being an arbitrage mechanism, which I think is very counterintuitive to the way a lot of people will think about markets and this idea that like, well, sometimes you can't compress the price all the way to where it should be rationally or logically or according to EMH or whatever because the reward just isn't necessarily there to get to that like final 10%. It's interesting, too, to think about the sort of link between patterns and interpretability, why something works. And in our conversation a couple of weeks ago with Ian Dunning of Hudson River

trading, it's like, they do not put a lot of emphasis on interpretability. There's a pattern and they have some reason to establish that the pattern works. It makes money. The idea that they then have to also come up with an economic story about why it works is not so important to them. Maybe that's because it has to do with timeframes. Obviously, Cliffs trading time frame is going to be very different than a high-requency trading firm like HRT.

But it is interesting. And then it's interesting to think that even in the most established quant patterns, like, why do cheap stocks outperform more expensive stocks over the long? long term, right? Even there, there's dispute about why this pattern holds, even though it feels a little bit more intuitive. So much interesting stuff here. It's also, I guess, it sort of warms my old cynical heart that maybe the edge for humans will be spotting the regime change, right? Which, you know, at least there's something left for us to do if it's not just pure pattern recognition. There's that one very difficult thing. Good luck to us. Yeah. All right. Shall we leave it there?

Let's leave it there. This has been another episode of the Odd Thoughts podcast. I'm Tracy Alloway. You can follow me at Tracy Allaway. And I'm Jill Wisenthal. You can follow me at The Stallwart. Follow our guest, Cliff Asness. He's at Clifford Asnes.

Follow our producers, Carmen Rodriguez, at Carmen Armin. Dashel Bennett at Dashpot and Kale Brooks at Kail Brooks. For more Odd Lots content, go to Bloomberg.com slash Oddlots with the daily newsletter and all of our episodes. And you can chat about all of these topics 24-7 in our Discord. Discord.g. slash odd lots. And if you enjoy Odlots, if you like it when we speak to guys like Cliff Asnus, then please

leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening. This is Tom Keene,

inviting you to join us for the Bloomberg Surveillance Podcast. It's about making you smarter every business day. I'm Paul Sweeney. We bring you complete coverage of the U.S. market open. We cover stocks, bonds, commodities, even crypto, all the information you need to excel. And I'm Alexis Christophers. Bloomberg Surveillance also brings you the analysis behind the headlines. We do that through conversations with the smartest names in economics, finance, investment, and international

relations. We do all this live each and every weekday that bring you the best analysis in our daily podcast. Search for Bloomberg Surveillance on Apple, Spotify, YouTube, or anywhere else you listen. On the East Coast, listen at lunch. And on the West Coast, listen as soon as you wake up. That's the Bloomberg Surveillance Podcast with Tom Keene, Paul Sweeney, and me, Alexis Christophores. Subscribe today, wherever you get your podcasts.

Bloomberg Surveillance, Essential Listening, each and every business day.