Owen Lamont, Senior Vice President, Acadian Asset Management

[00:00:01] Hello, this is Dean Curnutt and welcome to the Alpha Exchange, where we explore topics in financial markets associated with managing risk, generating return, and the deployment of capital in the alternative investment industry. Now a portfolio manager at Acadian Asset Management, Owen Lamont has had a long career in both the markets and in academic research on them.

[00:00:28] Earning a PhD in economics from MIT in the 90s and then teaching at the University of Chicago shortly thereafter, Owen makes the point that these two storied institutions approach empirical finance from vastly different perspectives, with the MIT approach to explaining market anomalies, utilizing behavioral finance, and the University of Chicago embracing market efficiency.

[00:00:51] Our conversation is about some of Owen's current work, starting with the observation that equity correlation has been exceptionally low, owing to the manner in which large cap growth stocks are disconnected from the rest of the market. As part of this, we explore the original tech bubble of the late 90s, contrasting it to present market leadership. Here, Owen makes the point that the original internet stock craze had dramatically more equity issuance than we see today.

[00:01:18] Owen puts equity issuance and short interest in a category of factors that have particular significance from an information content perspective, calling both firms and short sellers smart money. We talk further about the AI trend in markets and Owen's concerns that the massive corporate spend may be overdone. He points to research in the academic literature that shows that high capex firms have some history of underperformance and offers competing theories on why.

[00:01:47] He gravitates to explaining excess investment in AI from the lens of over-optimism among both investors and companies. Among the other topics we cover is Owen's take on the min-vol factor. That is, the empirical finding that low volatility stocks outperform the market on a risk-adjusted basis.

[00:02:06] In a manner similar to the tech stock craze of the late 90s, the underperformance of the low vol factor over the last five years owes to the incredibly strong performance of the riskiest stocks during this timeframe. On a going-forward basis, Owen is optimistic that low vol stocks can deliver better risk-adjusted returns. I hope you enjoy this episode of the Alpha Exchange, my conversation with Owen Lamont.

[00:02:34] My guest today on the Alpha Exchange is Owen Lamont. He is Senior Vice President and Portfolio Manager at Acadian Asset Management. Owen, it's a pleasure to welcome you to the Alpha Exchange. Happy to be here. I'm looking forward to the conversation. You and I have some similar roots. We crossed at the University of Chicago about three decades ago. I as a student, you as a professor. So it's great to reconnect.

[00:02:57] And I'm looking forward to a conversation about market risk, the intersection of academic finance research and the practicalities of markets. So let's get into it. Let's learn a little bit more of your background. You got a PhD from MIT about 30-odd years ago. So lots has changed in markets since then. Why don't you just give us a little bit of chronology of your background, both in academic research and in markets? Sure.

[00:03:24] This might be a little inside baseball, but I come from MIT. There's a different tradition of MIT and Chicago. So MIT is the saltwater school of economics where they're more willing to believe that markets have mistakes, that the market is inefficient, that there are bubbles. Bubble is not a dirty word. And at Chicago, where I was a professor, totally different view, of course, at the time. I'm sure you remember Gene Fama and Merton Miller was still alive at the time.

[00:03:54] I'm kind of the opposite end of finance from you. I come from the corporate finance, which has not as much to do with your fancy derivatives that you talk about. I'm a longtime listener of your podcast, but you often use words, Greek letters that I'm less familiar with than I should be. So I come from a macro slash corporate point of view, and I gradually became interested in the stock market and asset pricing.

[00:04:18] So like you, I was privileged to be at the University of Chicago and learn from the masters about the deep ideas underlying modern portfolio theory and modern finance. But I'm a bit more behavioral. Thaler was there at the time, and that's more my kind of thing. So my academic background is, I would say I ended up being interested in behavioral finance and stock market pricing. And I happened to be at Chicago right during the period of tech stock mania.

[00:04:46] And that was certainly a period that changed my mind and made me feel more behavioral. And I think the right way to describe tech stock mania, and you lived through it too, is there were a lot of people who believed in efficient markets, and their belief was diminished by tech stock mania. And it's not true that everybody capable of changing their mind did change their mind. But it's certainly true that only those capable of changing their mind did change their mind.

[00:05:16] I felt like that was a real blow to stock market efficiency. So that's my academic background. I ended up being a tenured professor at Yale, and I took the amazingly poorly timed decision to resign tenure at a 300-year-old institution. I was in 2007, right before the huge financial crisis, to go work for a hedge fund. And since then, I've been in and out of academia. I'm mostly out now, but I taught at Harvard in the Department of Economics for a while.

[00:05:46] So you've given so much interesting stuff for us to run with there. And as you sort of think through the long and storied history of the University of Chicago and its belief in this notion of market efficiency, I know we both know Cliff Asnes, he would say that his belief in efficiency is probably less now than even a decade ago, right?

[00:06:08] You live through episodes not just like the tech bubble, but things like the meme craze, which was more protracted, more pronounced, shorter lived. And so there's so many ways to interpret this, right? I think it's telling that both Fama and Thaler are Nobel Prize winners with a kind of antithetical approach. One person's bubble or behavioral phenomenon is another person's, there's a risk premium here, right?

[00:06:37] There's a rational explanation. And even the tech bubble, which of course, God, that's got to strain credibility in a lot of ways. But in some sense, the market's just pricing the optionality of things that it just doesn't understand. Again, hard not to call that an absolute bubble. Hard not to look at the meme craze as well and just say, boy, that's difficult to look at GameStop,

[00:07:02] a stock that goes from 30 to 300 in a week or so on nothing and call that a rational or efficient market. So what would you say if you were to sort of step back and look at the academic literature or the thought process broadly, knowing that an MIT is different than a University of Chicago, how have these events in markets colored the way broadly folks are approaching academic research?

[00:07:31] Hmm. Well, let me first comment on what you said about the tech stock bubble. There are many finance academics, some of them at Chicago, for example, Veronese and Pastor, who would say the tech stock bubble was perfectly consistent with rationality. And I think what Fama would say is out of the tech stock bubble, there came Microsoft and Apple. And if there had just come like two or three more stocks as good as Microsoft and Apple, that would have justified the prices.

[00:08:00] So there's plenty of people who would say the tech stock bubble was not necessarily irrational. I think one thing that has happened and that I would say the economics profession is not really have a good model for is crypto and Bitcoin. And that's where the response of Fama is quite amazing to me, because maybe I shouldn't say amazing, but he has the same response that I do to Bitcoin, which is it doesn't seem like it should be worth anything.

[00:08:29] And that is a not something that Fama would usually say. Fama would usually say, here's a price. Our null hypothesis is that the price is correct. So an emerging theory, OK, a rational theory for Bitcoin and for GameStop as of 2021. And the rational theory or a rational theory, and this is articulated by John Cochran,

[00:08:52] also formerly of Chicago, is there is a convenience yield that you get from holding Bitcoin. And you might even think of it as a liquidity service, or you could think of it as some sort of unmeasured benefit that you get. And we have the same puzzle explaining money or gold. Why is gold worth anything?

[00:09:16] So what they would say or what John Cochran would say is that Bitcoin is just part of the grand tapestry of difficult to explain ex post, but probably rational events in human history, such as gold. And he would not agree that there's anything irrational about Bitcoin. So I think the thing about crypto and Bitcoin is there is such a universal belief among economists,

[00:09:42] or at least older economists like my age, as opposed to 20-year-old economists, that Bitcoin and crypto are some kind of inexplicable event. I'm not in academia anymore, but that's my impression outside. There's an excellent series of surveys run by the University of Chicago. It's called the Clark Center surveys. And economists are way on one side of the crypto debate, traditional side. Right. And it's become one of these almost toxic conversations sometimes.

[00:10:08] I think the community, unfortunately, with all the cheerleading that happens and statements like have fun staying poor, it becomes a difficult conversation to have. I like what you had said about gold because that resonates a lot with me as well, which is some form of it has value because we say it has value. And that's another podcast guest is a gentleman by the name of Jordy Visser.

[00:10:32] And he's done a lot of thinking on AI and just the speed with which AI is a force for advancement, but also destruction. You have no defense in some ways against something that can read every human word ever printed in a matter of an hour or so, right? But what he says about Bitcoin and gold is that they have brand. And I like that term. It's some version of goodwill, I suppose. And brand is hard to measure, but it is real.

[00:11:02] Why do we value gold the way we do? It's not something we can really use. It's just something we've decided. Obviously, it was a medium of exchange in the past. But to me, Bitcoin kind of got there. It doesn't mean it'll stay there. It's still very, very speculative. I'm sure there's lots of fraud still in the system. But it has become something. It represents something. And obviously, you can't really apply any sort of fundamental framework to something that

[00:11:30] has no yield or earnings or management team or anything like that. I think you hit it on the nose. It's brand. Gold has a really great brand. They mentioned gold and every fairy tale has gold. The Bible has gold. So I think in one of my pieces I mentioned, I don't even know if I'm going to pronounce this right, renium. It's a mineral that's equally as scarce as gold, but nobody cares about it. You know why? It doesn't have any brand. So I think the bull case specifically for Bitcoin is that this is a brand that's endured.

[00:12:00] It's endured since 2008 or whenever Bitcoin wasn't then. So that's really the bull case that there's limited space in the human brain for an enduring asset. And maybe Bitcoin is up there with diamonds and gold. So all that glitters is gold. All that glitters is renium. That just doesn't sound as catchy. It was only discovered recently. Yeah. It doesn't glitter at all, as far as I know. Well, I alluded to AI.

[00:12:27] And so you've done some interesting writing and a lot of thinking on AI and specifically with application to kind of an area of market risk that's got me super interested. I've done a lot of writing about the, you know, I just call it anti-correlation among stocks. There's low levels of correlation and then there's this. So you've written a piece, Growth is from Mars. And what was the second part? Value is from Venus. Value is from Venus. Okay.

[00:12:53] So why don't you maybe frame out some of the history of this? Again, to reference Cliff Asnes and sort of his thinking on the value spread, which just got to outrageous, never before explored levels during COVID and maybe the year after. So walk us through some of your thinking, maybe starting with growth versus value, and then we'll talk about MAG7 and this very low level of correlation across stocks.

[00:13:19] So the value spread, which as you say, has been well discussed over the years by Cliff Asnes, I guess I would say it has been discussed with increasing anger and dismay by Cliff Asnes and other intelligent people. The value spread is a description of whether value looks historically cheap relative to growth. And it is true that the value spread is currently very wide and was especially wide in 2021. But it was also very wide in 1999-2000.

[00:13:49] So the value spread kind of looks like a bubble indicator or a craziness indicator for the U.S. stock market. And maybe that statement is becoming increasingly untenable as it remains wide for so long. But that's a statement about the level of prices. So the way I would say it is, historically, it has been sometimes true that value looks cheap and sometimes true value looks less cheap. OK, so let's look on to changes in prices.

[00:14:15] So if we look at the return on large cap growth stocks and we compare that return of that portfolio of large cap growth stocks to the rest of the markets, the rest of the markets like small caps and large cap value, that number has always been around 80%. So it's just like a stock is a stock. When the market goes up, all stocks go up. Maybe some go up more, some go up less. So what's changed in the past four years, I'd say, has been that that number, the correlation

[00:14:45] of large cap growth stocks with the rest of the market has gone down. And it has gone down way down, especially earlier this year, to low levels. So that's a statement about bifurcation, that the market is bifurcating into two parts. So it's not negative correlation. It's not like gold is sometimes negatively correlated. Sometimes risk on, risk off. You have treasuries negatively correlated with the stock market. They're not negatively correlated, but there's a surprising low correlation. So why is that happening?

[00:15:12] I think one explanation for that low correlation is just that big large cap tech stocks are just different from traditional consumer durables or whatever, that there has been a fundamental those stocks are just like responding to different shocks. And so there's nothing necessarily irrational about two sets of securities becoming less positively correlated than they usually are. So it's just a fact that we're living in a market, at least in the past couple of years,

[00:15:40] where it's almost as if we have two different markets, like the Chinese market or the American market are two different markets. We have the large cap growth market and the rest of the American market. They're less correlated with each other than they are with the rest of the MSCI world, for example. So that's a bifurcation. And is it good? Is it bad? I don't know. You could say that it's bad. The world is falling apart. And this is part of a world where all our institutions are collapsing. That's the bad case scenario.

[00:16:10] I don't think that scenario makes much sense. Is it a symbol of speculative excess? I'm not sure about that because this correlation wasn't particularly low in 1999. So I would almost say if something didn't go crazy in 1999, it's not a good bubble indicator. So that's the discussion of whether it's bad. Is it good? Well, when you have stocks and when those stocks become less correlated with each other, the total aggregate market portfolio as a whole becomes less correlated and becomes less volatile.

[00:16:39] So you could say that it's good in the sense that it's less risk for everyone. So the good case scenario is we happen to live in an economy where two types of companies are increasingly less similar to each other. And that makes our stock market safer because it's more diversified. Yeah. No. And first, just to sort of capture it in one of your statistics, you say big cap growth, the current disconnect with respect to this low level of correlation is in the 99th percentile all the way back to 1926.

[00:17:07] So that goes back a ways that obviously goes through war and various economic cycles through various market dislocations. So that goes back quite some time. Just exploring this bifurcation as bad versus good again, and maybe overlaying it again back to the tech bubble. One of the little event tests that I did, really simple. I just looked at the three or four bellwethers today in big cap tech versus then.

[00:17:36] So today, maybe it's Apple, Microsoft, NVIDIA. Maybe it's Amazon, something like that, or Google. And back then, it was probably Cisco, Intel, Microsoft. I don't know, Broadcom. I can't remember the last day. I had four of them there. Sun Micro is a great example. And the correlation of those stocks during that period was so remarkably higher. So it's not just that big cap growth is vastly less correlated than normally would be to the rest of the market.

[00:18:04] It's vastly less correlated to each other. That, to me, is super interesting. And I don't know enough about AI to really dive into it. But one thing these companies certainly do share is a massive CapEx spend and some valuation premium that I think is assigned to this working. They do seem to be in the business of efficiency and productivity and the markets paying up now for those expected benefits.

[00:18:34] Now, they are still printing money. And they weren't doing that back in 2000. I'd love to just hear a little bit more about how you juxtapose the super caps now, the tech super caps now versus then. It's got sort of similarities and differences. I'm going to disagree with you. I think that the super caps back in 1999 were very profitable. Profitable, but extremely high PEs. I guess maybe that's what I meant to say. They were more excessively valued. Okay.

[00:19:00] So you're saying now that the Magnificent Seven of 1999, whatever those guys were, they had much higher PEs. Much, much higher. Yeah. I'll agree with that. So in both cases, we had high E and they got to higher PE. And one response to that is, okay, well, just give us a year and maybe we'll get to PEs of 100 on the Magnificent Seven. So I would say the similarity is in both cases, you had E going up.

[00:19:27] You had very profitable, larger cap firms, not so much for the startups, you know, your pest.com and stuff, but very profitable existing firms. Second area of agreement is, as you said, there was high CapEx. And in 1999, it was WorldCom laying down tons of fiber optic cable. And today it's companies building data centers. What's the big difference between today and 1999? Big difference is issuance. In 1999, you had companies like Cisco, their number of shares were going up because they

[00:19:57] were issuing shares one way or the other, whether it was outright season equity offerings or by paying their executives and shares in some way. So 1999, we had a huge wave of IPOs and a huge wave of issuance by existing companies. Today, we have none of that, no issuance. We don't have a wave of IPOs. And I think most of the Magnificent Seven are repurchasing equity. They're not issuing equity. So that's the big difference to me. And that would be the major reason I would say that I don't think we're in a bubble today.

[00:20:26] We have many, many bubble symptoms, but the major symptom we do not have is issuance. And one can imagine in the coming year, that'll change. But I think in 1999, the general consensus was, first of all, the growth stocks were too expensive. And second of all, the value stocks were cheap. They had low PEs. And I'm not sure that's really the consensus today, just because many of the Magnificent Seven, they have such high E that their PEs are pretty low.

[00:20:55] So when you think of value stocks being neglected and despised and overlooked, whatever Europe being despised and overlooked, I think the best way to measure that's always going to be relative. And I know that people like to say, oh, there's this company that only has a PE of five. It's going for a steal. I'm not 100% convinced that's the right way to think about it. What is your overall take on the value factor? As we're discussing, it's clearly got some time variation. Sometimes value is cheaper than other times.

[00:21:24] Your view of just portfolio construction over the long term, where does value sit in the factor zoo, so to speak? Do you find that to be something to lean into? Yes. Okay. So I would say of all the possible factors that could possibly work, value has an exalted special place because the way you get a return is you buy something for a low price and the price goes up.

[00:21:52] Now, that's probably an inelegant way of saying it. But the one thing we know about stocks is the capital gains you get, what you sell it for today minus what you bought it for a year ago. So if anything is going to be predicting returns, it's going to have something having price or valuation. Okay. That's an overstatement. But you generally expect as a law of math that valuation would work. It's possible you could live in a world where momentum worked and value didn't work, but I'm skeptical.

[00:22:19] So having said that, that's like a theoretical statement that if I had price divided by fundamental value, that would be a really good thing. But the problem is we don't observe price divided by fundamental. We do a price divided by some stupid proxy like earnings. So I would be willing to stipulate that due to changes in the economy, it's impossible to construct a good value measure. Don't think that's true, but it's a logical possibility.

[00:22:47] So I would say I think that if I were to predict 100 years from now which factors worked over the next 100 years, I think some version of value is still going to be that. That's the animal in the zoo. That's the elephant. That's not going to die. It's going to live a long time. I'll give you what I think are the other two of my favorite indicators that I would put in. And they're both correlated with value.

[00:23:15] And what they are is the first one is issuance, which we've already discussed. So like firms that issue equity, those are overpriced. Firms that repurchase equity, those are underpriced. Those are good firms. And the second one, this one might become obsolete due to changes in the legal structure. My second one would be short sellers. Whatever short sellers are shorting, I don't want that thing. And whatever short sellers don't short, I like that thing. So what I'm trying to say is firms are smart money and short sellers are smart money. I want to do what they're doing.

[00:23:44] So I'm saying 100 years ago, I think issuance will continue to be a good factor. The short selling, that one might not work because short selling might get banned as it has been banned in other countries. So there's this concept of shareholder yield, which is some version of not issuing, doing buybacks and so forth. And that makes its way into certain factor strategies. I think your short selling one is a great one, too, because you're right.

[00:24:06] There is pretty darn good information content into someone who's done the work and is willing to bet against. Short sale research is one of the kind of boutique products out there, which has probably seen better days. I think muddy waters shut down.

[00:24:23] But people would pay for that because I think there's a view that if you get in there forensically and shine a bright light on accounting uncertainties, the Jim Chanos type stuff with Enron, there can be real value there. And of course, there could be real asymmetry. Short selling could be very convex if you get it right. Things can go down very quickly to the extent that there is maybe malfeasance that's uncovered or something like that.

[00:24:49] Well, on the value front, it's probably not the case that deep value stocks are engaging in a lot of the massive AI spend that we see from some of the mega caps. And so I just want to read a statement that you made here as part of your piece on AI spending. You said, I don't know if it'll be wasted, but I do know that historically high planned investment has been followed by low stock returns in subsequent years. So I'm just going to leave that out there for you to explore for us.

[00:25:18] I found it to be a really interesting statement and I'd love to learn more about it. Sure. So in 1999, we saw tons of CapEx capital expenditures, not only firms laying fiber optics, but firms building factories, firms spending in ways that some of which might not have been classified as CapEx, but were investments in AI. So in good times, when profits are high and the stock market is high, you see firms really ramping up investment in physical capital.

[00:25:48] That's what CapEx is. So they're buying trucks, they're building factories, they're buying land, they're building buildings, whatever. And historically, that time when firms are really optimistic and spending a lot, that has been a bad time to invest in the stock market. It's kind of the opposite of blood in the streets. It's the time when everyone is super optimistic and capital is cheap and they're spending more.

[00:26:12] So I think that fact, and by the way, that's also somewhat true for individual firms as well. If you see a firm that is spending a ton on CapEx, obviously different industries have different CapEx needs. But if you see a firm that has unusually high CapEx expenditures, that is generally a firm that's a growth firm that's going to underperform. But let's go back to the aggregate fact. That fact is, we have like five different theories to explain that fact.

[00:26:39] The high CapEx, especially high planned CapEx is associated with low returns. And I think we can think about this moment in history. And I think some of these explanations fit this moment in history and some of them don't. So the first explanation I'm going to give, and this is kind of a everybody who has ever heard of NPV should understand this explanation is, if the discount rate is low, firms are going to do a lot of projects. And if the discount rate is high, firms are not going to do a lot of projects.

[00:27:09] So one explanation that should be true, as long as you think firms are thinking about NPV is, when you see a high CapEx, that means firms are using a low discount rate. And what is a low discount rate? What's the translation of low discount rate into normal language? Low discount rate means overpriced, means the stock market is overpriced. At least that's the behavioral view. And then you could also say, well, it's a rationally low discount rate because risk is really low. That would be another theory.

[00:27:36] But no matter which theory you use, both of those are saying the stock market is priced to deliver low future returns. So that's sort of the vanilla mainstream, the thing that everybody would agree on that statement. Then you get more convoluted. And I would call these behavioral corporate finance explanations. So one is the thing I already mentioned, issuance. Back in 1999, Cisco was doing a lot of CapEx and it was also issuing a lot of equity in order to fund that CapEx.

[00:28:04] So why were firms issuing equity in 1999? It's because they thought the equity was overpriced. At least they didn't think it was underpriced. So issuance, again, is one of my favorite indicators and issuance at CapEx often go hand in hand. Not today, but historically. Another explanation, and this doesn't really resonate with me, but it's called catering. So catering is the statement that firms will just do whatever they think investors want them to do to make the stock price go up.

[00:28:31] So the story would be firms don't really care about AI, but the investors care about AI. So the firms are going to invest in AI in order to please outside investors and get its stock price to go up. And that's the story. Doesn't especially work for me at this moment. And I think we see catering in other parts of the stock market, for example, related to crypto and firms pivoting to crypto in order to attract attention. And that's catering, okay?

[00:28:59] Doing some stupid crypto project in order to make your stock price go up. But firms investing in AI, that doesn't seem especially like catering to me. What matters is whether the stock market is going to reward companies or not for investing in AI. And I think the stock market may change its mind in recent weeks about that. Okay. Another explanation, and this does resonate with me, is just everybody is overoptimistic. So investors are overoptimistic. The AI companies are overoptimistic. Everybody is bullish on AI.

[00:29:29] And that explanation would also explain why we don't see issuance. We don't see issuance because the Magnificent Seven don't perceive themselves to be overpriced. They perceive themselves to be fairly priced. And as you said, there's a good reason maybe to believe that. But historically, times when earnings have gone way up and everyone's investing, and even if there's not issuance, you might think those are times where people are just too optimistic. They've just overreacted. And that certainly is something that seems to have happened in the 1990s as well. Okay.

[00:29:59] Last explanation is companies are just wasting the AI capex. And that comes in two flavors. First is, and this is part of a tradition in corporate finance that goes back to the 1970s and 80s when there were conglomerates. Companies just like to spend money. That's just what they do. They like to build big projects. And they'll just do it whenever they can. And now they can because they have a lot of free cash flow and no one's stopping them.

[00:30:26] So that's kind of the empire building wasteful spending version. And that's just a version that said they just like to spend like drunken sailors. Usually they're constrained because they don't have enough much capex, but right now they have tons of cash flow. So I don't know. That argument doesn't seem particularly compelling to me. Another argument under which they're wasting is competition neglect. And this one maybe is relevant. So competition neglect is there's a company building a huge data center.

[00:30:55] It's spending $100 billion on a data center. And it is realizing it's doing that because today there's a shortage of data centers. But what it is neglecting to take into account is every other company is building a data center. And three years from now when the data center is finished, there's going to be too many data centers. So that's a story that is often told in bubbles and too many railroads. Too much fiber optic cable in 1999. Too many houses built in 2006.

[00:31:24] And that does maybe have some resonance for me. So those are all the many explanations of why it could be bad to see high capex today. I don't think high capex is like a smoking gun. It's not ringing the bell. But it's one more piece of news that makes me think the market's overvalued. And is this a portion of the academic empirical literature, this idea? And so take us back a little bit.

[00:31:49] Obviously, the granddaddy of the Molotek bubble in 99-2000 is your most prominent example, I assume. But are there other economic cycles that sort of tend to become part of that? No, 1999 is not the granddaddy. 1999 is recent. Economists of the 1920s often use the word overbuilding, overbuilding. And so there are these railroads built in Britain in 1850. Too many railroads, you know, like three railroad lines going to the same medium-sized city.

[00:32:19] And one explanation of that is competition neglect. They didn't realize somebody else was building a railroad. And that's probably not the total story, because there's also an argument that it was a rational competing to be the first guy to build a railroad to the city. But competition neglect is something that has been documented experimentally in labs, where it's not just about firms. It's also about people. Like, I'm going to get 10 students, and they'll have to do this task.

[00:32:45] And I'm not going to be able to describe accurately what the experiments are, but it ends up like the people in the experiment are acting as if no one else is reacting to them. And they're acting myopically, as if they're not really playing out the whole sequence of events. It's as if they're the only actor and no one's going to react to them. So that's where the term competition neglect comes from. And it's one of many theories that explain outcomes.

[00:33:12] I think a good example of competition neglect might be electric vehicles in 2016, 2017, going into 2020. Tons of different startups that were starting electric vehicles, electric trucks, whatever. And the argument was made, this is not my argument, but people have made the argument, including Brad Cornell. He calls it the big market delusion. Brad Cornell, writing with various different co-authors.

[00:33:37] He's made the argument that each of these electric vehicle companies was priced to perfection as if it would be the winner. And you have the existing automotive makers that were also priced. So it's perfectly fine to have one huge electric vehicle manufacturer that has a high market cap. But when its market cap goes up, the other one's market cap should go down. And that's not what we see. It's as if each stock is being priced by the biggest optimist, but no one's really doing

[00:34:06] the adding up. And so the whole sector gets overpriced. I don't know if this is a real corollary to that. But if you go back a couple of years to the Archegos blow up, where the dealers were providing leveraged financing on equity swap as if they were the only one doing it and not aggregating the exposure, and the same thing happened in LTCM. Every derivative salesperson, of which I was one, thought they were a hero because they were doing a lot of business with LTCM.

[00:34:35] Meanwhile, look to your left and look to your right, and Goldman and Morgan Stanley, and they're doing at least as much. And in sum total, it's just mind-bogglingly large amount of risk. And the same thing happened with Archegos. In that case, that was fraud. He was lying to the dealers about his other positions. But yeah, I do think that that does capture the flavor of it. The competition neglect in the electric vehicle, though, it's not like there was a big secret there were all these electric vehicles. But yeah, it is the exact same thing, yeah.

[00:35:02] So let's finish up with a little discussion on some of what you've written on this idea of low vol. You know, a strategy that kind of came into focus, I'm going to guess, like a decade ago. I started to see some ETFs pop up before we jumped on this call. I looked at there's USMV, there's SPLV. I think the first one's got $24 billion of assets under management. And so the big picture, and I'm going to let you clean this up, but the big picture

[00:35:32] is that CAPM sort of governs or guides us along to think of the expected return to be a function of a risk-free rate plus some beta that the asset has for a higher or lower beta. And what low vol says is you could actually buy these lower vol, lower beta stocks and lever them up a little bit and do better. And I just remember the ETF starting to pop up and got pretty big.

[00:35:59] And I was thinking, OK, is this a risk factor that is going to go away very quickly or an edge that's going to go away very quickly? Because we're just going to bid up Procter & Gamble and Colgate or take your pick of a low vol stock. So I don't know, frame it out for us, what you've looked at here and the state of the state in terms of low vol strategies. First, let me react to your statement that the rise of low vol is the thing that made low vol not work.

[00:36:28] I'm not suggesting, I'm saying it, I'm putting out that. That was your fear. That was your suspicion. A fear, yeah. And I think people say the same about value, that there are so many people going into value in 2014 or something that it made it not work. And that is not what happened. That is the opposite of what happened. The opposite of what happened is the risky growth stocks went way up and the allegedly crowded value safe stocks did not.

[00:36:56] So, and you mentioned the value spread, the value spread widened. If everyone crowded into value, the value spread should narrow, but we saw the opposite happening. So here's my take. And I'm going to go back into my personal history mode. So I arrived at Chicago as an assistant professor of finance in 1995. And I didn't teach the investments course, but I saw what the other people were teaching. I taught corporate finance. And they were teaching the CAPM and they were teaching a specific version of the CAPM, which

[00:37:24] is the black version of the CAPM. They were talking about how the security market line is too flat. And I remember in 1995 thinking, these guys, these guys teaching, they're teaching this like outmoded historical relic that only a dinosaur would be interested in the fact that the CAPM, that the securities market line is too flat. And what I did not realize is that that fact that the securities market line is too flat or that the risky stocks have returns that are not high enough, that is the basis of a

[00:37:54] profitable strategy, which is just what you say. You buy the low vol things and you lever them up. So, as you say, the low vol, you could call it defensive or minimum vol. There are many words for this, many phrases for this strategy. I think it did get started about 10 years ago, about 2014. And it included ETFs and it also included, of course, many separate accounts at institutional money managers, institutional asset managers.

[00:38:20] And it has not done well in the past five years. So why is it? Why is it that the strategy that has done pretty well and when you test it over 50 years, it works well? It works as expected. Why is it not done well in the past five years? And I think we have a good reason. We have a good answer for that. And the answer is the following. This is a strategy that does well in a typical year.

[00:38:49] It does well on average. The time that it does not do well is a time of a huge bull market where risky stocks go up. And you could even say that it doesn't do well in a bubble when the risky stocks go up. So what we know from history, and this is before the boom in low vol after 2010 or so, what we know from history is this strategy of buying low vol stocks, buying low beta stocks,

[00:39:18] it did terribly in 1999. It did terribly in the late 90s. And that's because in the late 90s, the risky stocks were going way up and the boring low vol old economy stocks were not. So the one time in history when this strategy is not going to work is a highly speculative market where you get the market prices go way up and the risky stock stuff goes even more. And that is a description of us in the past five years.

[00:39:46] Risky stocks doing really well. So that is, I think, a good explanation for why low vol strategies have done poorly in the past five years. What does it predict about the future? To me, it suggests that low vol is going to do well in the future because you can't have the stock market doubling every five years. That can't continue forever. That's got to slow down.

[00:40:10] So I think if you look at the past 50 years and you say, OK, there's two five-year periods when low vol did poorly. Going forward, are we going to be in that same situation? I don't think so. So to me, we have this historical regularity, which is low vol stocks tend to have returns that are just the same on average as high vol stocks. You have short periods where the historical pattern doesn't work, and then you have it reverted. So when was a great time to do low vol?

[00:40:37] The great time to do low vol was after the tech stock bubble deflated 2000 to 2005. So if you think we're in a bubble now or something like a bubble and it's going to get worse, that would be a great time to do low vol. So that's not like a very compelling reason to do low vol, but it's just a reason based on history. It usually works. We know the kinds of markets it doesn't work, and I don't think those markets can persist forever.

[00:41:01] So through the lens of MIT more behavioral or University of Chicago a bit more rationality or risk premium argument, how do you arrive at the why of low vol? Why is that line flat, the SML? Okay. So there's a bunch of different theories, and this is not my super area of super expertise, but I'll do it my best.

[00:41:25] One theory is that, this is a behavioral theory, is that people just like to buy lottery tickets. The boring stocks that make toothpaste and deodorant, those are boring stocks. They don't go up and down. Nobody wants to hold them. They want to hold the exciting technology stocks. So that's like the exciting technology stocks are overvalued. They're just always overvalued. And because they're risky, they have low returns. So things that are risky, people bid up the price and the subsequent return is lower. We call that lottery-like preferences.

[00:41:55] A second argument, and this is a little bit more rational. I'm not sure it's 100% rational. It's the Sini and Pedersen argument in their paper, Betting Against Beta. It has to do with leverage constraints. And the idea is that people want to lever up, and they want to hold more than 100% of their wealth in the stock market, say, but the bank won't lend them money.

[00:42:19] So instead of just holding the market, they hold the risky stocks in the market in an effort to get around these leverage constraints. And so, again, they're driving up the price of risky stocks, and they're driving down the subsequent returns of risky stocks. And they have some evidence that they think is consistent with their story. So that's not about individual stocks being risky. That's about market exposure.

[00:42:45] So I'm not sure which of these stories is the best one, the flatness of the security market line. I feel like that's not a very behavioral thing. I feel like the Fred Zin and Pedersen thing is probably closer to the type of argument or the type of explanation I'd want to hear. But neither of those explanations. I think those two explanations might have different predictions going forward. Because the Fred Zin and Pedersen thing is based on people having limited leverage.

[00:43:11] And if you think nowadays people have more access to leverage, maybe from levered ETFs or something, then maybe that would make it go away. I don't know. Well, let me ask you this. As you just mentioned, the levered ETFs. We have these brand new products. They don't resemble anything that Graham and Dodd would have taught us to look at. Two times inverse on MSTR, which buys Bitcoin and issues converts to do so. You've got ETFs with derivatives in them.

[00:43:40] We've got things like Polymarket, right? Which is you could bet on will the Ukrainian war end in 90 days? Will Chris Rock speak at the Oscars? And folks are all over these sites. Our obsession with sports, especially in the US, is now overlaid with FanDuel and DraftKings, of which there's 100 derivative plays that you could do intraday. And then, of course, we have the meme stock episode of 2021, which for me was a doozy.

[00:44:09] I just thought, wow, that is something else to see the force of folks trying to corner different stocks and it working for a period of time. And you mentioned at the outset, I look at things like implied volatility. I mean, these things were just literally off the charts. So my question is, when you think about the behavioral elements of markets,

[00:44:32] has society changed in a way that might force us to think differently about how asset markets work? Our penchant for gambling and risk, it seems like we like to take risk these days. Okay, well, there's a lot in your question. Let me first say there is nothing new under the sun. And we talked about Graham and Dodd, which I think was written in the 1930s. Okay, in the 1920s, they basically had levered ETFs. They were called levered closed-end funds.

[00:45:00] And what was the dynamic there? The dynamic there is you had an institution called Goldman Sachs, still exists today, that wanted to make money. And you had a bunch of retail investors. They called them the little fellow. The little fellow wanted to invest in stocks. And Goldman Sachs was happy to provide these levered instruments to them. So I think there's always going to be financial intermediaries who will,

[00:45:25] as long as the regulations allow them, will provide levered instruments for unsophisticated investors to invest in. So that's my description of levered ETFs. And it's not that human nature has changed. It's that levered ETFs didn't used to be legal in the United States. It didn't used to exist. And in some countries, they're still illegal. For example, in Korea, levered single-stock ETFs are not legal. So what do the Koreans do? They buy the levered single-stock ETFs in America.

[00:45:54] It's not that Koreans have different human nature. It's just that they're able to do it. So one aspect of it is, I think, that there's part of this narrative of bubbles, that bubbles are a time when people are horribly reckless and they borrow. And bubbles are pumped up by leverage. You get that story in Japan in the 1990s, for example, it involved leverage. And people certainly say that for the 1990, 1929 bubble.

[00:46:20] So for the 1999 bubble, I don't think that had much to do with leverage. I think that was mostly a purely unlevered phenomena. And sure, there was WorldCom and stuff that had debt that they defaulted on, but that wasn't the main thing. I think we also had a little bubble in 2021. Didn't seem very levered to me either. So I think that the existence of the levered ETFs is more just what the regulators will allow as opposed to human nature has changed. So a couple of other things you said. First of all, about the prediction markets.

[00:46:49] I love prediction markets. I think most economists are in favor of prediction markets. Markets are good. We want more markets to solve human problems. And it's hard to forecast stuff and markets embody information. So Bob Schiller, the great behavioral economist, is all in favor of using markets to solve human problems, like GDP markets or whatever. I don't think you'd be that interested in markets about what Adam Sandler will say or whatever. But okay, going on to the general idea of gambling.

[00:47:19] There's nothing wrong with gambling. People like gambling. They should do it. And Milton Friedman talked a lot about gambling. And one of the things Milton Friedman said is, if you're in a country where gambling is illegal, and you want to gamble on the commodities market, that that's fine. That's a purpose that the commodities market serves. Here's my view of gambling in the United States. I think if people want to gamble, it's fine. I would prefer that they don't do it in the stock market. I would like them to gamble on sports or something like that,

[00:47:47] that does not influence capital allocation. And like you, I was horrified by the GameStop episode. And the thing that horrified me about it was not that there were irrational prices, but that it was an adversarial anti... We were like hunting down the short sellers. The short sellers are evil people, and we're going to hurt them. And that is not the way I understand the world to be. I don't know if you saw the movie Dumb Money. Oh, yeah.

[00:48:18] 99% of the people who see the movie Dumb Money don't realize that the short sellers are the good guys. They think the short sellers are the bad guys. Or that's my view anyway. So I do think that it is true that in America, attitudes towards gambling have changed. And that's just part of social... You know, we see all kinds of changes in attitudes towards sexuality, changes in attitudes towards all kinds of social things. That's one of the things that we've changed our view on is whether you should be allowed to gamble on sports.

[00:48:46] So I'm not terribly upset by that. I think what I would be concerned with is speculative gambling affecting stock prices and somehow a degradation of market quality. And I think our experience in COVID was when you don't allow people to bet on basketball games, they're going to bet on the stock market. That's part of what the meme stock thing was. So I say, let's have all the betting on basketball games we want so that they can get that out of their system.

[00:49:16] Yeah, no, just super interesting to see. So let's close out. One last question for you. Just a big picture ask you to share what's on your mind. You're doing a lot of writing. That's on the Acadian website. Some of your recent pieces I really enjoyed. We're all in the business of trying to find alpha. It's a hard thing to come by, but we're trying to find those, whether they're strategies or ways of constructing a portfolio,

[00:49:44] ways of finding diversifying assets that allow us to get more return for a given amount of risk or the same return and less risk. What's on your mind there? Are there areas of investigation that you're actively looking at with members of your team? Can you share a little bit on that front with us? Absolutely. Okay. So we've talked a lot about AI. And what I have not said yet is AI is an amazing benefit to humanity.

[00:50:10] And there's no doubt that AI is going to change our economy in fundamental ways. And if you thought the internet bubble and the internet was an important development in 1999, AI is like 10 times more important. So we talked about the possibility of corporations overspending on AI. I mean, maybe that's bad for their shareholders, but it's good for society to have cheap AI all over the place, helping AI doctors, whatever it is we need.

[00:50:38] So I think we're in the business, you and I, of analyzing data and understanding complicated quantitative settings. And AI is perfect for us to help us with that. So in Acadian, we are very interested in AI. And when you say the word AI, that could mean many different tools. But we are certainly interested, and I am certainly interested in ways to automate the process. We talked about the factors.

[00:51:07] So the factor zoo is the outcome of a bunch of fallible human beings with questionable incentives, including me, publishing claims, statistical claims over the course of 30 years. And that is not the right way to arrive at an understanding of what determines alpha. That is like saying the right way to build a car is to have 20 different people build a part of it over 20 years.

[00:51:33] No, the right way is to industrialize it with a rational, statistical, scientific framework from top to bottom and not some kind of handcrafted, hand curated thing. So AI is going to impact many professions, and it's going to impact especially the profession, or it has already impacted in a good way, the profession of asset management. And that's just alpha, okay? And there are many other parts of our job, such as writing or researching.

[00:52:03] We now have a magical research assistant that will summarize research, and they will sometimes lie and hallucinate. But that's true of regular human research assistants as well. So I think it is a wonderful time to be alive and to be in the business of science or science-adjacent business that we're in.

[00:52:23] And I don't know if you remember what it was like to be a college student, but part of the joy of being a college student and part of the joy of our job is a week ago, you didn't know something. And now you know something. It's like you went from being this ignorant person in this one area to suddenly having a sense of mastery that you have changed. And AI is really part of turbocharging that. I know I sound like some kind of bizarre crypto enthusiast who is saying crypto, but this is AI.

[00:52:52] This is a real thing that is doing science fiction. We're living in a science fictional time of history, and it's amazing to be alive. And I don't know. How do you feel about AI? Am I over the top here? No, I think it's going to be incredibly interesting. I think that to some extent, there is a very small cohort of folks, and I certainly wouldn't put myself in that cohort, that really, really understand the potential implications of where we're going.

[00:53:19] And 99.9% of the folks do not. And I'm in the 99.9%. I know enough to think, okay, this is going to be pretty profound, and it's really hard to know where we're going. It's going to be very disruptive, but also it's very difficult to think that this is not humanity in a lot of ways at its best with respect to advancement. I'm super interested in what happens to the investment business. I hope it doesn't take the kneecaps out of a podcaster.

[00:53:48] We already know that there are robot podcasters. And I'll tell you, those robot podcasters, they are way more flattering than you are. AI has really learned how to navigate human ego by saying how great you are. So, yeah, I write these little pieces, and AI is great at writing pieces. So, yeah. Right. Well, Owen, it's been a pleasure. I'm very glad we got a chance to do this. I've appreciated your insights and looking forward to keeping in touch. Great to talk to you.

[00:54:18] All right. Thanks a lot. Thanks again, and catch you next time.