April 9, 2020

Episode 93: Cliff Asness from AQR: The Impact of Stories, Behaviour and Risk

April 9, 2020

No one credible ever said that investing was a simple endeavour. It might have some simple guidelines, that if followed are more likely to yield positive results, but the ins and outs of the markets, decisions and their impacts, movements and crashes are never straightforward one-dimensional cases. Our guest today, Cliff Asness, really brings this point to bear, showing the nuance and multiplicity of all the topics we discuss. As the experienced owner of AQR and a wealth of knowledge and insight, Cliff shares a host of ideas and thoughts on as many topics as we have time for. We start off the chat talking about market efficiency before moving into the murky waters of value. We hold value investing to be sound, as does Cliff, yet the last few years have stretched even our commitment to this philosophy a little. The perspective that Cliff is able to share, drawing from his formative years in the investing world in the '90s is invaluable and a lot of what we talk about gets contrasted to the tech bubble of that period. The conversation also covers the size of stocks and portfolio allocation. Although Cliff has strong opinions on most of these issues he does a great job of showing the lack of definitive answers to any one of them, allowing space for new knowledge and outlying evidence to make its mark. We also get into finding the right kind of investor for your own style and goals, the role of good communication in finance and the influential article that Cliff wrote about 'pulling the goalie'. In it, Cliff lays out what the data tells us about certain late-stage situations in which it is statistically wise to make more risky choices. For all of this and a fabulously entertaining conversation, listen in with us today!

Key Points From This Episode:

• Cliff's perspective on market efficiency and the impact on his portfolios. [0:03:48.5]

• Value investing in today's climate where value has taken such a knock. [0:08:30.8]

• Stories and behavioural effects on value; how we understand ups and downs. [0:13:36.2]

• Conversations Cliff has had with clients in the tougher times. [0:21:04.5]

• Comparing the companies driving growth now with those in the '90s. [0:23:46.2]

• The size effect and why Cliff does not subscribe to this philosophy. [0:25:17.1]

• 60/40 portfolios; are they still alive? Why Cliff thinks you can do better! [0:33:07.7]

• Cliff's experiences with institutions and advisors and contrasting the two. [0:36:31.5]

• Informed decisions on who to invest with; thoughts on finding the right advisor. [0:38:28.7]

• Pulling the goalie and why risky behaviour can work in certain circumstances. [0:40:42.5]

• The value of communication skills in the game of financial advising. [0:47:29.7]

• How Cliff defines success for his own life! [0:50:07.9]

Read the Transcript:

So off the top, Cliff, can you talk about market efficiency a bit and how your perspective might be different than someone like Professor Eugene Fama who was your PhD advisor at University of Chicago?

Sure. For one thing, let's back up with a definition. Fama beats this into your head at the University of Chicago, market efficiency is simply the idea that prices reflect all information. Whenever this comes up, I jump in, and first, not that Fama needs defending at all, the dude's righteously enjoying a Nobel prize, but sometimes people think he is more dogmatic than he is. Fama has this moment, and I took the class and then TA'd it twice, so I sat through this moment three times, where in his class, he looks out and tells the class that markets are almost assuredly not perfectly efficient. And only at the University of Chicago would this have like 60 people gasping. Anywhere else in the world, people would be like, "Yeah, we either don't care or know that."

He always presents it as a point on a spectrum. He probably thinks they're more efficient than most people. And getting to your question, he probably thinks they're more efficient than I do. I was always at least a minor heretic, not a major one, it's not like I think markets are wildly inefficient, but I did write a dissertation for him on a few different things. But one of the main topics was this successful simple price momentum strategy, still one of the most difficult findings to reconcile with efficient markets. To his credit, I was nervous about writing that for him, and he said, "If it's in the data, write the paper."

But even with that, I think I moved a little bit away. And then doing this live for now 25 years has probably moved me at least towards the middle of the spectrum. If one direction is markets are very, very efficient and one is they're wildly inefficient. They're too hard to beat for them to be wildly inefficient, I'm pretty sure of that. But having lived through the technology bubble of '99, 2000, the GFC, the current real difficult times for value investing has certainly pushed me not way further, I'm not Dick Thaler yet, but I'm not pure Gene Fama

So how does this difference affect how you actually construct your portfolios?

That's funny. I don't a ton, to be honest. Truth and interpretation and what you believe matters, but so many of these things in financial economics, these so-called anomalies or risk factors, can work for a variety of reasons, for instance, value. There are three possible reasons value has worked for the last 100 years. One is it represents a pure rational risk premium. Cheap stocks are riskier in some sense than expensive stocks, and you get paid for bearing that risk. The next is irrationality, cheap stocks are cheap, at least partially by era and vice versa.

The third is it was all dumb luck and it's never going to work again. I always have to throw that one in to be intellectually honest, I don't actually believe that one. So if Gene and Gene works very closely with Dimensional Fund, a firm I respect the hell out of. If they believe value is mostly, and I doubt many people there only believe in one explanation, and I certainly don't, but if they think it's mostly a risk premium, and I think it's mostly, which would probably be going too far for me, but if I think it's more than they do a behavioral effect, we don't do anything radically different with that information. We do have some differences momentum.

That's an interesting one. I would say I believe in it more than Gene does, but again, to use the example, AQR and DFA, we both use momentum in our process. They use it to screen stocks before trading, we use it as a formal factor in the stocks we like and dislike. So I do think it's fair to say we are bigger believers in it. But then we're talking about degree, not actually a difference in what we believe, just in how much we believe it. So for some of the things, there are differences, but for many of them, we may actually have significant differences in what we believe, but it doesn't actually matter that much for something like a value factor for when you go to implement it.

Yeah. We had Marlena Lee, Dimensional's co-head of research on the podcast a while back, and she agreed that it's probably some mix of a risk-based and behavioral explanation for value. On the topic of value investing, you wrote a letter at the end of-

Do we have to talk about value investing? Can't we wait three years until we're proven dramatically right and then talk about it?

Won't that be nice when that happens?

Yes. Yes, it will. But go on.

We use Dimensional's products, so we're valued tilted as well. At the end of last year, you said it was time to send a little in terms of increasing exposure to value, and then value stocks, they continued to not do well even to a higher degree than previously. How do we think about, we and you, as value tilt investors, how do we think about sticking with this strategy?

Well, a few things. First of all, that's just true. And I think you know I wrote a piece on this that I called Quantitative Whining. I don't run from these kinds of things. To be fair, when you stake out a position, it's a multi-year view. It would be really nice if the day after you both put a tilt on and wrote about it, the world collectively went, "Oh God, they're right," and you made money for the next 14 months in large gobs starting that moment. Again, it help for that. You don't put it on thinking, "Oh, I know it's going to lose a little more, but I have to do this."

No, you think the odds are on your side, but calling a top or bottom, you quoted us to saying sin a little. We think market timing and style timing, moving in and out of things like value momentum, low beta, size is very hard to do. We call it an investing sin, not because of any religious belief, but because we think it's very hard to do and it often causes more harm than good. But we have never said don't sin, we've said sin a little bit. At extremes, particularly extremes of valuation, do a little bit more at a level you think you can tolerate if you do get unlucky and have to quantitatively whine for awhile.

We're still perfectly comfortable with it. We like it. I'd feel nervous if I didn't have enough value in my portfolio for the next few years. That's actually fairly standard. You wish you didn't have it in the last few years, and you're thrilled you have it going forward. Now as to how to stick with it, I have no magic bullet. I do believe that these episodes teach us something very important. They teach us that it's way harder to stick with these things than people realize. Take again, you can use Gene Fama for everything, he's done so much. Take his and Ken French's HML.

Again, we might differ a little bit with them on this, we tend to use more factors. I don't even know, they might... They're actually fairly agnostic on this. They will tell you different value factors, all kind of work and they're just versions of the same things. So we like a basket, not just price to book, but it doesn't matter. Take their HML and graph it over time back to the '20s. You will look at that and you will look at some horrible periods. You will look at the last eight to 10 years, you will look at the tech bubble, you will look at some other times in history and you'll see it all ends pretty good, and you'll go, "Yeah, I could stick with that." Meanwhile, lots and lots of people didn't.

I remember during the tech bubble, we got great credit from our clients for sticking with value after the fact. And one client was trying to be nice, but they said, "Julian Robertson, an investor I respect very much, closed his fund near the top of the bubble or the bottom for value, and you got stuck with it," and it was meant to be like a complement. And I said, "No, you don't understand. Julian had many billions of dollars left to spend time in New Zealand because he didn't want to deal with the rest of us idiots, I didn't have that option, so I don't think it was necessarily bravery or whatever, but it was reflective of a strong belief we had."

And to stick with it, education, I think we're finding it much easier to communicate with people, frankly, just having gone through it at least one and a half times before. The tech bubble was disastrous for value, there was a period of the GFC that was quite bad for value. We, the collective we, people who do this and have done it for a long time, the world doesn't fully understand. And I have my days where I don't understand it, I'm not saying I don't. I like to sound stone-cold rational these podcasts and I like to throw things around my office when value goes down three days after I said it would go up. So I don't want to fake anybody out, out there. The rational side is the right way to invest.

But, these things, especially single factors, have fairly low risk adjusted returns. I will use the term sharp ratio, it's not a perfect measure. Whenever I use sharp ratio without a caveat, someone in the audience raises their hand goes, "What if returns are fat tailed or something?" And I'm like, "Yeah, I know, I know. I know sharp ratio is not perfect, but we need a language to talk about something like a value factor." Even a very good value factor is a modest, positive sharp ratio. If it's as good as the stock market, the stock market has about a 0.4 long-term sharp ratio.

That means it nearly has occasional bad decades, so does the value factor. The evidence for the value factor, I think is at least on a... I don't think it's as good as just being net long stocks, particularly given theory that there should be a positive versus premium to being net long stocks, but it's certainly in the ballpark. So you have evidence and you know you will see periods like this, but living through them is very hard. So we try to write, we try to speak, we try to point to the historical record of what we've done other times this has happened, and then we suffer because none of that makes it easy.

I think one of the biggest challenges in times like this is the stories that people create to explain what's happening. And one of the ones that we keep hearing, obviously this is the thing people always say when things are going the way they hoped, but it's different this time because of the types of companies that are dominant in the market and the types of companies that are driving the growth of growth. Is there any merit to that argument or that story?

There can be. Just because the world is declared to be different this time 100 times and it's not 99 times out of 100, doesn't mean it isn't one time out of 100, so you should not be dismissive of these things. One of the chief value factors that's come under fire is the price-to-book factor, this idea that there are intangibles that are impossible to measure with paid-in and retained capital, so maybe book isn't quite what it was. And I have some sympathy for that. We've always used book as one of, with no special standing, one of an array of factors.

I do point out with some irony that I think book to price has been one of the better factors for the last couple of years, better being less bad in the value world. If you care about price in any way, you've not had an enjoyable few years, but I do find a little ironic that the one that comes under the most fire, my ex professor's original factor is currently underperforming by less. So I do think that you have to keep an open mind, but you do have to remember that your mind can not be so open that your brains fall out. Most of the time, the world does not change. Value does not work historically because it nailed the right way to price a company.

Value works historically, and I'm going to take the behavioral explanation here only because it's rhetorically difficult to give the behavioral and the risk-based explanation, both of them every time. But please, I do think both have meaning. I don't want to be staking out a big side here. But if value works for behavioral reasons, it means people look at the fundamentals. It could be book, it could be sales, it could be cashflow, it could be all kinds of forms of earnings, both trailing and forecasted. It can be many other things. And on average, pay a little too much for the darlings and a little too little for the laggards.

Typically, those are the stocks, the darlings are the ones that have been doing better for the last few years and the laggards are the ones who have been doing worse, not just in price, but in real life performance. So if that's true, value adapts, it doesn't really matter. If people have this bias and there are a whole slew of fundamentals, they will always pay a little too much for those fundamentals. Now, a few other things. When people criticize the factor as well broken, they often think they've proven why it's gone straight down for eight years. That's not the case. A broken factor would look like someone added random noise to your measurement, it wouldn't systematically get it wrong, because it's impossible.

There's no systematic arbitrage that I know of in the market to just make money. Would you have confidently shorted the value factor 10 years ago, because, hey, book to price is going to be less important going forward? So people often make explanations that could cause, in my view, perhaps a modest deterioration in a factor as if they've explained the three standard deviation draw down. So there's that. Finally, value is far from the only thing in most people's process. Fama and French have expanded what they do. We certainly look at profitability, low-risk, fundamental, and price momentum.

We saw about a seven, eight-year period, call it from 2010 through two years ago where value was having a lousy period, not every day because that's not how these things work, but on average the whole time, and life was actually pretty darn good for us because all those things many of which are about picking up fundamentals. Just because a pure value manager would be crazy to say every stock should trade for the exact same price to book, I don't think anyone believes that. People think the marketing that goes too far or I'll break my own rule, gives you a risk premium for the cheap stocks, but it doesn't say they should all be the same. So I think of a lot of the other factors as excellent compliments to value.

Now, for us, I can tell you, that stopped working about two years ago. At the exact same time over those two years, we've seen these spreads between cheap and expensive, not just on a price-to-book basis. Again, I think most people have that in their head when they make these world has changed change criticisms. Price to cashflow, price to forecasted earnings have been disastrous value factors. And there's very little story where the world is different and we don't care about forecasted earnings anymore. In the last two years. The other things that are meant to pick up fundamentals have ceased to work while the spreads between cheap and expensive stocks in terms of valuations, something that I kept pointing out, weren't that wide for a long time.

I'm late to the party, thank God, in saying do more value. Because for multiple years, I took the other side saying don't sin. And by the way, cheaper is expensive is not that extreme. If value loses on the fundamentals, which sounds like a disaster, you didn't lose on price, you didn't lose on sentiment, you lost because the stocks you were, the value liked performed worse, earned worse, had less fewer cash flows, lower sales, shrinking companies. That's actually not the worst thing in the world for us because many of the other things in a process like ours or a process like Dimensional's that considers profitability, that uses momentum as a screen, can pick up on that.

I will tell you the one environment I have found impossible to do well in is the one I think we've been in the last couple of years, where value loses for irrational reasons. And I know it's a loaded word, irrational, but at least more irrational reasons. Price momentum is about the only thing that has a hope of doing well in that kind of environment. Even earnings momentum or fundamental momentum tends to start to fall by the wayside because it's not fundamentals being rewarded. There's a limit to how much price momentum anyone, including me, will put it in their process. It has a pretty vicious left tail.

It's wonderful long-term strategy. And at the right size, you can live through that left tail, but too much is too dangerous. So I do think if anything, the evidence is value has lost for the last couple of years, at least, not because it's broken, not because it's different this time, but because it's the same thing as before. People are going to look crazy. And I do think many of the other factors, not picking up value in these last couple of years and the value spread widening at the same time is pretty strong evidence for that theory.

Let's talk about another one of the early documented factors.

You guys are allowed to disagree if you think I'm full of it, you're allowed to push me.

I want you to be right. Cameron and I always talk about how, wouldn't it be nice if... It's like what you said, when we're talking about this three years from now and we've had a great run for value, I don't have the answers, there's nothing to disagree with with what you said, it's all an unknown. I read your note about the value spread, it makes it look like now's the best time ever to invest in value. Maybe not as good as 2000, but-

Yeah, you can measured it a lot of different ways. I don't think it's as good as 2000. I do think it's pretty much the best outside of 2000, best being usually a pretty horrible experience to get to the best, but best going forward. The GFC, there are some measures of value that got cheaper, some did not get as cheap. Again, once you introduce more than one way to measure value, you won't have exactly the same results, but they're broadly similar. That want to be here three years from now to say, "Ooh, thank God we've stuck with that." I think it's probably a little easier for me having had that experience several times.

It doesn't make it easy. I actually sometimes tell friends, "It's astounding how hard this still is, even though I've seen this movie multiple times before." But when you see your first time, that's harrowing.

Yeah. When you've gone through this in the past, like you said you have, what have been the conversations like with clients? Like when clients come and say whatever they say, what do you say back to them?

Well, it's actually a fairly similar conversation here, "Has something changed?" You certainly tend to grow assets in good times and lose assets in bad times. The world has that bias, that if value works for behavioral reasons again and if momentum works for behavioral reasons, somebody has to be doing something a little irrational to create that. It's hard to try to make your living from that and then root for it not to apply to you at any time. It's easy to root for that, it's just not going to happen. But by and large, the conversations are pretty good. No, one's happy in a terrible period, but our clients are pretty rational. People they've chosen to do this, they more often than not, and I won't say we don't have tough conversations, but more often than not, they've seen the long-term evidence, they've looked at those periods.

It is always hard to live through life. One thing always fascinating is much like it's easier for me having seen this before, a client who's been a client for a long, long time finds this much easier than a client who happened to buy in 23 months ago at the peak. And it's funny because you know that makes no sense. Two clients looking at the same historical data, two anybodies looking at the same historical data shouldn't come to a radically different conclusion based on their own personal experience, but they often do. It happens to employees. An employee we hired 23 months ago, and I shouldn't say this, there's some things that have done well over this time.

Credit strategies for us have done well, low-risk investing has done well, but by and large, it's been an excruciating near two years. Employees, they've not seen us ever do really well across the board. Someone who's been here 10 years has seen that for a long time. Someone who's been with us 20 years or going back to Goldman Sachs, we're down just a few from the core team at Goldman Sachs, that's 25 years. They take it a lot differently. So I just can't stress that enough. We should use the historical experience of others and the data to learn more than I think we do. But I do think we're lucky that we have clients that by and large get it. And it doesn't mean they'll stick with us for 100 years.

If a couple of years from now, we're still saying, "Well, the value spread is now three times the tech bubble," we will be a much smaller firm. And the expected returns going forward will be way, way higher than they even are today, it's a fact of life.

But one of the big differences, Cliff, comparing now to the late '90s is so many of the big name stocks that are driving the growth part of the marketplace are names we use every day like our iPhones, like Netflix, everyone's getting stuff shipped by Amazon. Whereas the names back then, it was more a revolution, but not with day-to-day proven products. And I think that does play with people's psyche.

Yeah. While I do think value is too cheap, in some sense, at least some irrationality has creeped in making it so, I don't think it's nearly as bad as the tech bubble, for two reasons. One I've already mentioned, I think just price spreads were wider back then, but the other is what you just said, the companies in the tech bubble... How do I put this technically? I'm going to use stochastic calculus. The companies that were very expensive back then were crap companies. The companies that are very expensive now are not, I think they're just too expensive for even what they're going to deliver, and I think history has a lot of evidence on that.

With the tech bubble, it was two-sided, there was a valuation difference that was gigantic. And to your point, a lot of the companies that actually were expensive were on the face of it's silly. We're not talking in this round about drkoop.com. I'm dating this for a lot of people, but drkoop.com, eToys. I have a little collage I bought. I went on eBay, which is ironic to make fun of the tech bubble by going on eBay, but I bought a bunch of stock certificates of some of the more egregiously silly companies after we were proven right. So I do agree with you, it is not the tech bubble, but I do think it is at the very least as good going forward as it's been historically and probably somewhat better.

But the tech bubble hopefully will remain a unique thing in our financial history. That won't be true. Somebody someday will experience something similar. I hope I'm not doing this during the next tech bubble.

You haven't talked about the size factor yet. What are your thoughts on that?

I think you know my thoughts on that. I think it's a leading question.

That is true.

We are simply not big believers in the size effect. We have not been, since inception, since our days at Goldman Sachs, we've never explicitly traded the size effect, meaning we've never preferred small just for being small. We think the original evidence Roth bonds, '81. '82, I forget just a few years before I was at Chicago, is generally acknowledged to have been somewhat overstated. No knock on professor bonds, he's awesome, but they've changed the databases. You always have to guess at the price of delisting stocks. You know it's probably not too good, but the databases have changed, and I think the results on delisting is a little more extreme. Small stocks delists more than large stocks. The very original findings are still there, but weaker. Second, small stocks are less liquid, meaning a lot of standards, statistical techniques, looking at their volatilities in particular, looking at their betas.

It's easy to understate them because they don't trade every day. We've done simple adjustments for that, Scholes and Williams did these kinds of things a long time ago, we've done it to hedge funds, we've done it to small stocks, their betas are a little higher than people think. So they do. Let me be clear, two good things about small stocks, they do tend to outperform large stocks. We simply think that's almost if not entirely due to their having a higher market betas. Second, and this is a real one, there is actually a good to still be overweight small stocks, even if you fully agree with what I just said.

Most systematic, and I would imagine more judgmental strategies if you believe in those, but I know it for systematic strategies, do seem to have higher alphas within small. So I think there is a saving grace. If you are a small value investor, you should not be particularly upset about what I'm saying. There is some evidence that value is stronger there, but as a pure will small beat large by more than its beta? No, we just don't see the evidence, ever. We got a lot of evidence on this.

Yeah, it's interesting. And it's interesting because somebody going out and buying a small cap, ETF, hoping for a good outcome, not necessarily get it, maybe if you go small cap value.

Yeah. I'll even give you a reason to go pure small that's a little bit different than what we're talking about. I'm always talking about the intercept against your beta. What do you add? Alpha is a loaded word too. When I say alpha, that's not necessarily inefficient market alpha, I just mean an intercept against your beta. And when you say small stocks don't have one, that's bad. The original work adjusted for that and the exciting part was they had an alpha, but they do deliver a higher beta with no financial leverage.

To get that same thing out of the stock market, you'd have to lever it up somewhat. So you can either do an unlevered over the smaller dollar allocation to small stocks. And that might be a small positive to them, to the ability to get a little bit higher beta without any, as our British friends would say gearing. I don't think by any means from the fact that factors work better within small to that advantage, I think a lot of people are using small very rationally, but I do think the story for small is much weaker now than we thought 24 years ago, standalone.

And by the way, I said story, let's use the word story in a different way. I've never, and this is part of the reason we never traded it, never fully got the story for why small should beat large. Very often it falls back to a liquidity story, but then let's use liquidity measures. There are more direct measures of liquidity than size, analysts maybe are less interested in them, but that's a mispricing, not a risk story. I've never really fully embraced the story for small. I think the story for say, value both either in a risk or behavioral sense is way stronger than the story for pure small.

You mentioned alpha and I agree it's a tricky term, but I'm going to use it in the inefficient markets context. So alpha, it's getting harder to generate, I think as asset pricing models are getting better. We can explain more returns through risk or behavioral factors, systematic factors that used to show up as alpha, even the example of the paper that some of your AQR colleagues wrote explaining Buffett's Alpha, which is a fantastic paper. So based on knowing all of that, do you think that investors should still be looking for alpha in a traditional sense?

This is an incredibly hard question. When you say in the traditional sense, do you mean like the non-quantitative go pick over individual stocks or the very factors we're talking or both?

Maybe both.

Okay. It's an odd question because I certainly get it that alpha could be harder to find, on the other hand, my position, particularly again, with respect to the value factor, is that things are more mispriced than normal today. So the notion that things are more mispriced because alpha's hard to find is a little bit of a contradictory one. Whenever you're clubbed over the head and you do lousy, but things are more attractive going forward, it has been harder to find because you've lost, but I find it hard to argue maybe going forward in a very long-term way, that will be true. But if we look today, I think things are, again, if you accept that markets aren't perfect and some rationality, have more than normal.

So in that sense, medium to long-term alpha's easier to find right now, it's just harder to live with until it starts to work. People argue about the rise of passive. There already is a lot less active management than there used to be, a lot more passive, call it somewhere in the middle. I hate when people contrast us with active management because we're differing from an index and in some cases, we're long and short. I think that's pretty active. I certainly don't think of us as passive, but some people do use that characterization, I would say, we're somewhere in the middle.

A lot of people sit around saying this move to passive should raise alpha opportunities going forward. Now, that's actually more complicated than it seems, people throw that out and then just go with it. You still can't get away from the zero-sum problem for everyone outperforming, someone's underperforming, but it is certainly possible, if there's a big move to passive, prices are more irrational, the swings are further away. You still have to be convinced that you're one of the people who'll be on the right side of that. The average still can't beat the average, but I've heard plausible arguments that alpha will be easier going forward because there are fewer people pursuing it.

You asked a great question, but I reject the premise. The other part that you didn't say, but people often do and maybe it was part of it is information is easy to get now. Doesn't that make alpha harder to get? Well, this one, I know we heard, I like to beat to death the analogy with the tech bubble, I do apologize for that. It was very formative in my life, my career, not my life, I don't want to... In the tech bubble, we heard repeatedly that this can't be irrational because information is available to everyone much faster and quicker and instantaneously in many cases than it used to be, information had been democratized.

And I had hair to rip out back then, I used to rip my hair when people would say that because for one thing, it was using an internet bubble to justify the internet bubble. Second, having information and pricing it well in a world that perhaps does occasionally have irrational mob psychology is not the same thing. You can walk up to a crazy mob with pitchforks and torches and give it all the information in the world and it will still look at you, call you a witch and burn you. So I reject the premise of your original question. It could one day be the case, I don't think it's the case now

Interesting answer, makes sense. Many advisors, Cliff in our industry suggest that the 60/40 portfolio is dead. Is it dead? For a dead thing, it keeps kicking the crap out of everything else, doesn't it? But value proposition is basically the 60/40 is dead, therefore you need us to go to the supermarket of alternative strategies, hedge funds, whatever it is.

Well, the funny part is I am certainly one of the people that thinks you can do better than a passive cap weight at 60/40, yet, I feel an incredible desire to take the other side of any argument that anyone ever makes. If you tilt towards value, small, momentum, low beta, you're differing from a pure 60/40 portfolio. If you're a believer in what's called risk parody, you are not 60/40. You have more bonds and fewer stocks because 60/40 is very dominated by stocks. I do believe those are long-term edges, I believe in international diversification. So is 60/40 domestic or is it global?

Every single one of the things I mentioned, some have done okay, some have done poorly. Nothing has been quite as good as long and strong US stocks for quite a long time. So I was joking in the beginning when I said it's not dead because it's kicking the hell out of everyone, kicking the hell out of everyone for a long time also makes you expensive. We are in the camp that both stocks and bonds, not a disaster, we're not trying to time a crash, but have lower long-term returns from here than they normally do. And that's a fairly simple argument.

Price to any fundamental for stocks for the market wide is quite high, and you might've noticed, bond yields are quite low. People almost always agree on bonds, it's very hard to imagine bonds making you three times their yield for the next 10 years. If you're forecasting 10-year returns over inflation, the empirical results for equities, kind of use your favorite measures, Shiller CAPE has become the ubiquitous one. I'm old enough to remember that when that first came around in '96, but not only do I think on average, people can do better than 60/40, because I believe in these other things we're talking about.

But I think 60/40 is probably, I say probably because we don't get enough 10, 20 year periods to ever say for sure, but it's probably poised to deliver decently lower returns going forward. And you'll always look silly making that argument when it keeps winning, but that is when the argument becomes the strongest going forward. So I don't think if you talk about cap weighting everything, and you said 60/40, that's actually sneakily. And I know you know this, a more active portfolio than people realize, because even if it's close to the average weight of stocks and bonds, when stocks are expensive and bonds are cheap and vice versa, the market's not 60/40.

So if you're 60/40, it's a rebalancing strategy. But if you extend that to a cap weighted index strategy that's long everything, that's never going to go out of style because that's the average of what we all hold. That is always going to be a decent default strategy, but I'm one of the people who believes you can do somewhat better from all the same things we keep talking about.

So you guys, AQR, have like liquid alternative products, for example, products that won't necessarily move with the market and are not designed to do so. In terms of your clients, you started out dealing with institutions and then later on started dealing with the advisors that pushed the funds out to individual investors, retail investors. Did you notice a difference in the clientele?

Can we use a nicer term than push the funds out?

Yes.

That sounds so salesy. That introduce the funds in a highly intellectual and deep way to their clients.

Yeah. That's better.

You actually mentioned in that question one of the important things we did. When we moved to do mutual funds, which I think we started in late '08, or early '09, we early on restricted it to financial advisors and made a special effort to meet with them, to gather them together, to meet with them individually and in group offsite kind of things. This is not the case for everyone, but there are a tremendous amount of financial advisors who are institutional quality. I don't think an AQR wants to take an ad out in a magazine and try to get someone sending in a check. Not that I think that person would do poorly, of course, I think they would do well long term.

The question is whether they will be with it long term. The only strategy I know of aside from overtrading transactions costs, that's the sure loser is one that you're guaranteed to abandon when it's really having a tough time. You're not going to enjoy your full long-term benefits from that. And I think if you went straight to the individual, you'd have a huge problem that way. I think the institutions who are with us are very solid, but I think the financial advisors, many of them are institutional quality. And frankly, we treat them like institutions.

So I see less of a difference, but there certainly may be some selection bias in who we try to work with and then maybe even bigger bias in who wants to work with us. I might not be seeing the mean, is what I'm saying.

No, no, but it makes sense.

I'm trying to put that nicely.

Cliff, I believe your firm has over 80 PhDs on staff and dimensional is also very heavy bench strength with academic backgrounds. Vanguard has them, Blackrock has brilliant people, Alpha Architect, RAFI. There's a bunch of firms that have lots of brain power. So it seems like there's lots of brainy-type stories. How do you think investors should sift through this to decide where to invest their money?

I don't know, there are a lot of financial podcasts out there, how do podcast listeners decide to listen to you guys? I didn't mean that to be snarky I meant that to be analogous. I think you talk to a whole bunch of people you think are smart, I would never advise someone to choose just one. Some people do, but there are differences among all of us, and I think the average of smart people is going to generally be better than even individual smart people. So figuring out who you think is smart, not investing in every one of them, that becomes unwieldy, but investing in at least a modest portfolio of a few people pursuing a philosophy you believe in perhaps with some differences and tweaks is going to long term yield a good result.

The world where you discover something that's true and can add long term and you're the only one who gets to do it forever, it's just not a realistic world if you're not named Jim Simons. There is one guy out there who throws a monkey wrench into my argument, but I'm going to skip him. I'm going to skip him for good reason, we can get back to that if you want, but I think you named, only firms that I respect. Don't get me wrong, some of them I'll fight like cats and dogs with on occasion, but we tend to be like two different sex and the Protestant reformation who hate each other and are 98% overlapping.

Oh, I'm going to get in trouble, we don't hate each other, but we fight like cats and dogs occasionally about that small things we differ on. Is there a size premium? How much you should time a factor, but there's not a firm you mentioned I wouldn't be comfortable with making part of my allocation to. While I won't go there, there are plenty of firms and plenty of styles that I would not be comfortable saying that about. So come up with your own belief and philosophy, look for the firms that you think are excellent, that are consistent with that, and pick your favorite few and then stick with them.

Classic advice, pick a strategy and stick with it.

It's so easy to say.

We're in Canada, I don't know if we mentioned that, and because we're up in Canada-

You did seem excessively nice.

So being here, we thought it was important to ask you.

Except if we put you people on ice skates, then you become vicious baskets.

That's what we want to ask about. You wrote a paper in 2018 about when the coach had pulled the goalie in hockey, when should they pull the goalie?

Well, a few things. First of all, considering you guys are Canadian, I have to acknowledge that myself and my coauthor did this originally because we were interested, we thought we had a neat model for it. And as we acknowledged in the paper, there have been about six papers we didn't know about all, of course by Canadians trying to come up with analytic solutions. Our solution was somewhat different and a lot of them are different. There are different ways to skin a cat to get at the same idea, ours and most of the others, maybe even all of the others, basically say considerably earlier.

For those, I know the Canadians listening are all familiar, but for those who are not, if you're losing with 10 seconds left in the game by a goal, in general, except for some highly specific situations, which people love to throw at us like gold differential to make the playoffs or something, in almost every situation, losing by two is no worse than losing by one, you still have a loss in the standings. Therefore, you do something that would normally be very, very stupid, you remove your goaltender. Why is it stupid? Because normally you have a goaltender because it's smart.

When you remove your goaltender, your chance of scoring goes up, the other team's chance of scoring goes way, way up. So in a neutral situation, early in a game, tied, this would be suicidal. But near the end, when you're losing, their chance of scoring going up, doesn't matter. It's a direct analogy to an option, you're an out of the money option, how much you lose by doesn't matter. And out of the money option likes volatility, even at the cost of expected return. So you will probably lose because you are losing and you've just raised their chance of scoring by more.

But you've also raised your chance of scoring by more, and that is more important in the situation. So now the question is, when do you do this? And the method came up with, I really I'm tooting my own horn, but I still love it, we did it so simply, we didn't try to be too clever. We said, we know we would do a 10 seconds left, so let's assume that well, what do you do at 20 seconds left? Well, you could do something called dynamic programming. It's a very simple form of dynamic programming where you assume you will act rationally, but 10 seconds left, to makes it very easy to figure out what to do with 20 seconds you left and then you just stepped backwards.

When we did that, we got all kinds of stats. Of course, how often people score with the goalie pulled, how often they're scored against, those are the key ones. We came up with a solution that was about five, five and a half minutes, you should pull. And then it gets even freakier because if you're down by two goals, you should pull with about 11 minutes left, which sounds mad. To any hockey fan who would see that they would freak out, but after staring at it and thinking it was crazy for a while, I eventually decided that... Again, don't hold me to 11 minutes, it could be seven minutes or 14 minutes. These models are not precise, you change the numbers.

The key is earlier than you would intuitively think by a healthy margin, but with 11 minutes left in the game down by two, you are highly likely to lose this game, therefore, you score once by the way, with 11 minutes left, you pull the goalie, you score one, you put your goalie right back in, because now you're back to the situation of losing by one. And if you're still losing by one with five minutes left, you pull them again. So we like some of the Canadian papers, got some fairly radical results that we thought were cool. Now, I will tell you one thing, I'm both pleased by this and even mildly annoyed by it.

I have written a ton of stuff in my life, some of it even good, and I've gotten more attention for the hockey paper than every finance thing I've ever written combined, partly it's because Malcolm Gladwell did a podcast where he talked about it, which I was convinced he would not use because I was convinced it was most boring geeky thing on earth, but when he did it, there were serial killers involved and decisions where mothers had to leave their children with serial killers, because that was actually, it was like pulling the goalie, that was the right call to run away and try to get help, but instinctively horrible.

And I'm like, we wrote a paper about dynamic programming and Malcolm Gladwell made it interesting with serial killers. Hopefully you guys can do the same if you throw a few serial killers in the editing process. But the fun part of the paper, there's the hockey part, I'll also brag that I got to talk about the paperwork Louis Lamoriello, who generally agreed that it should be pulled earlier. I don't think he's ready to go for five and a half minutes, but that was an exciting moment for me to talk to the current GM of the Islanders, former GM of the Devil's and the Canadiens.

Again, the hockey papers were more fun and more attention than anything we've ever did in finance, but as you guys I'm sure know, we spent half the paper talking about investing. After we were done with the hockey, we said, "All right, let's look at our day jobs." And the analogies in investing we thought were obvious and gigantic. The key one being, doing something different than the crowd that is often likely to hurt you even if it is stone cold, rational, even if it's provably rational, is very, very difficult. Imagine the hockey coach who looks at our results completely buys them and does it, it will not work most of the time.

How do you think the press and the fans will treat that hockey coach who's pulling the goalie with five and a half minutes left and having it fail the first four times, and he's waving around some essence in brown paper, doing something different than the crowd is hard, it's also rewarding if you can do it. And another investing analogy, you don't have to do all of it, meaning you don't have to pull five and a half minutes early, if Aaron and I are right, and everyone else is doing it at one minute, if you do it at two minutes, you have an advantage. And maybe that's a popularity risk you can live with.

Very similar if AQR, DFA or Alpha Architect or any of the great firms you mentioned has an optimal portfolio that they believe in, you don't have to do the whole thing, maybe that's too hard to stick with. You can move towards not to an optimal solution. So we had fun, we had a whole bunch of others, but we thought they were tremendous analogies to all the difficulties in doing what you know is right in investing. So we ended up being an investing paper anyway, but I don't kid myself, people read it for the hockey.

Those are really good analogies. And they tie into a lot of things that we've talked about today. Cliff, your ability and AQR in general, I think, like the papers you guys produce are just unreal, your ability to communicate is excellent. I don't know how else to describe it, and speak the way that you're able to communicate the ideas we're talking with this podcast are just unbelievable.

Feel comfortable leaving it there as a statement not a question. No, go on, I'm kidding. I'm kidding. Go on.

How important do you think that ability to communicate is to the success of your clients, AQR's clients?

This won't shock you, you said it, now I sound like I'm bragging, but I think it is hugely important to success. The number one thing for success is to have an investment process that will deliver over the long term. I don't think anyone communicates so well that if they don't have an investment process with a lot of evidence for it, a logic to it, some combination of the two. I always get scared if I talk about something like this, people will think that's number one. No, number one is having a process you believe in, but we keep this as almost a theme of our talk. Even if you believe in a process, if you're not Jim Simons, who I'd mentioned before, the famous Medallion fund, occasionally they have a bad Thursday and that's very upsetting to them.

On the other hand, I've also pointed out that they take out a ridiculous amount of money for a few people every year, but can't do that at institutional scale. So I like to say they kicked all the clients out and kept it for themselves. So every once in a while, a client will ask me, and I know I've changed the subject, is the Medallion fund better than you guys? And I'd go, which surprises them, "Oh, hell yes, but we're pretty darn good, and we will take your money." So why are they relevant? And actually, they do run some public products that are very good, but I don't think better than the rest of ours. They become human quants when they try to run it at institutional scale.

So if you have a process with a good, but not Jim Simons sharp ratio, we hope to do somewhat better than this, but if you can have your added value, not your total sharp ratio, have a similar risk adjusted return to the stock market, 0.4, 0.5, kind of sharp. That's phenomenal long term. It's very hard to live with. Again, we've never had a bad decade, the longest we've had is about two years, but 0.4 sharp ratio processes of bad decades. So if you believe that's what you've discovered, communication, education, empathy, these things are all what gets you to the goal, which is sticking with a good process for the long term.

So I am proud of our ability to communicate, it is not at all ever a substitute for a good investment process, but a good investment process without it, it's not actually that useful. A good BD, not Jim Simons, but modest, risk-adjusted return that's real and can be done at billions and billions and billions of dollars for institutions and individuals, if you can't communicate it well, it can actually not even be that helpful because no one's going to stick with it when it hits its inevitable tough time. So it's very nice that you phrase that as a compliment, we've had our failures too, we're not always great communicators.

Our last question for you, we've talked about success, how do you define success in your own life?

Let's start with the basics. You can't ask someone with four wonderful kids how to define success. And if they don't start with that and raising good kids and happy kids, then there's something wrong with that person. And I will start with that. How we had kids is always an interesting quantitative story. We had two sets of twins born 18 months apart. My wife does not like it when I refer to our family planning as, and I'm quoting here, I'm quoting myself in the past, a gross failure of risk control. She finds that somewhat unromantic, though, I'm pretty happy with the kids, so we're not giving those back.

So happy, adjusted, well adjusted, successful kids is number one to any parent. After that, I think you may probably meant the question in more of a business sense, but in a business sense, building something, at this point, it's about building something that outlasts me, building something that is part of the firmament in finance, that I could be retired guy, hopefully 50 years from now, I'll be 103, so that's optimistic, looking at other people running it and there will still be great research, and I'll be able to say I was one of the four people who started that, that will be phenomenal success to me.

The spirit of the question, it is more about you personally than it is about business. The business answer was great, but other than the kids, how do you define success personally?

I'd like to see the Rangers win another Stanley Cup. It was 54 years from 1940 to 1994. It's already frigging 26 years, it's astounding, it's like half the original drought. I have a feeling I might be a 70-year-old chanting. In the garden, I used to chant in 1994 all the time. I actually, opponents in 1994, I have a feeling I might be around to hear people tramping 1994, but I am a big sports fan. I'm a Yankee fan, there, I cannot complain, as a ranger fan, it's quite the opposite. It's been 11 years since we've won. So for a Yankee fan, that's depressing, but Yankee fans have an embarrassment of riches, but I would like to see the Rangers win another Stanley Cup.

Happy kids, a business that's part of the investing firmament and a New York Ranger Stanley Cup, it will be hard for me to imagine me not being happy with those three. Though, I can be quite creative about things to be miserable about, so check back with me if they happen.

All right. Well, cliff, you and I are both 53 years old, so I think in 50 years when we're both 103, hopefully we can have you back, hopefully before then, but for sure, when we're both 103.

Yes. And we can both gloat about being right. It'll be fun.

Awesome. Well, thanks very much for joining us. This has been incredible interview.

Oh, I've enjoyed it. Thanks guys.