Guest: Cliff Asness is a Founder, Managing Principal and Chief Investment Officer at AQR Capital Management.
Recorded: 3/20/2024 | Run-Time: 01:05:24 
Summary: In today’s episode, Cliff & I start by talking about some quotes he may or may not have said in the past. Then we kick around a bunch of topics. We talk about diversifying by both asset class and geography, the challenge of managing short-term expectations while keeping a long-term perspective, and how AQR is implementing AI and machine learning in their investment process.
Comments or suggestions? Interested in sponsoring an episode? Email us Feedback@TheMebFaberShow.com
Links from the Episode:
- (1:20) – Welcome to our guest, Cliff Asness
- (1:45) – Cliff’s article in the Financial Times
- (2:30) – Parsing Cliff’s real quotes from fake ones
- (8:56) – Thoughts on bonds as a diversifier
- (13:37) – Moving away from market-cap weighting
- (19:27) – Managing client expectations
- (29:30) – AQR Data Sets
- (29:56) – Alternative trend-following
- (43:48) – Position resizing for alternative strategies
- (48:06) – AQR implementing AI in their investment process
- (51:46) – Cliff’s most controversial opinion
- (57:39) – Cliff’s most memorable investment
- Learn more about Cliff: AQR; Twitter
Transcript:
Meb:
Cliff, welcome to the show.
Cliff:
Thank you, Meb. Thanks for having me.
Meb:
I’ve listened to a few of your podcasts lately. They’ve been great and there’s been some themes that you’ve been talking about. I’m going to try to avoid those today. So we’re going to try to do a couple of things a little bit differently. We both love value, but you’ve published a lot over the years and you guys have a really great summary piece called 20 for 20. You guys are 20 years old as a business plus now?
Cliff:
No, we’re 25 years old now.
Meb:
Listeners, we’ll put this in the show note links. It’s like 500 pages. It’s well worth a read. And if you’re not going to read the whole thing, just read John Bogle’s intro, which is great. But as a way of jumping off points in this conversation, we’re going to use a few of Cliff’s quotes. Now because Cliff recently had a quote in the Financial Times that said, “We don’t think AI, at least in our field, is as revolutionary as others do,” and that feels like fighting words. But the fun part we do is we said, all right, we’re also going to ask ChatGPT for some quotes that sound like Cliffs but aren’t. And we’ll see if Cliff can tell the difference if he said it or not over the years, and we’ll call those CliffGPT. So to begin, we’ll start with one and we’ll start broad and get wonkier. The first one is, and these are also potentially fighting words. “If Warren Buffett had 90% inequities and added 60% bonds, I believe he would’ve done better.” Is that something you might’ve said? And if so, you want to explain?
Cliff:
It’s something I did say.
Meb:
What do you mean by that?
Cliff:
I was talking about the general notion that levering the best portfolio is a better idea than putting all your money in the best asset. Something I wrote about starting in 1996, still comes up all the time and really smart interviewers, but asked me an interesting question. They just said, “Hey, Warren Buffet’s just been equities, in fact, probably somewhat levered equities. Are you saying he’s not optimal?” And I said, “Yeah, he’s phenomenal. And my colleagues have written a paper on exactly how phenomenal he is called Buffett’s Alpha. But the principal doesn’t say that he’s doing something wrong. It says he’s not doing everything he could do.” If you added a very modest amount of risk in some other asset that’s not very correlated like bonds, it doesn’t have to be bonds. You’d only have to do a drop less of what he was doing. And I do think if he was applying that principle over time, I don’t think he’s hurting. I don’t think it’s a big criticism. But yeah, I think the principle would’ve still applied.
Meb:
As we think about diversification. So Buffett famously has said, and a lot of this is probably due to simplicity, but “90% on the S&P and 10% just keeping cash,” or whatnot. And to me, that feels horribly undiversified. And so there was an entrepreneur on Twitter a while back and he had just sold his business, let’s call it 30 million, some amazing outcome, and he’s like, “You know what I’m going to do with this? I want to be boring. I want to be boring and basic and I’m just going to index and put it all on the S&P.”
And part of me felt this little something where I was like, you can call in my mind investing in just the S&P a lot of things, but boring in an asset class that can, and equities in general, any country loses half or more. I don’t know if boring to me, feels like the right thing. And so question to you as you think about diversification. So let’s say you’re talking to this entrepreneur and he’s like, “I’m just going to put all my money in S&P, SPY. What might you say to him? And here’s the second quote for you, Cliff or CliffGPT, “Investing without diversification is like going to a buffet and only eating dessert. Satisfying in the short term, but disastrous in the long term.”
Cliff:
Sounds less like me.
Meb:
That’s CliffGPT.
Cliff:
I agree with the sentiment, but I try not to be quite that trite.
Meb:
Yeah. All right. Well, you’re correct. You’re two for two.
Cliff:
Has no hint of obnoxiousness to it. And usually, something I say has at least a little hint of that. This is a long conversation. This is the whole reason why you shouldn’t have all your money in one country’s equity market, why you shouldn’t be over influenced by the fact that it’s the market that has won. Those of us who believe in diversification, and I know you’re in this camp, but we have a tough road to hoe and the S&P has crushed most of everything else for a long period, but you always have to play the game going forward. I wrote a piece, I don’t know, somewhere between one and 20 years ago. I no longer remember times. I remember two periods well, the last two weeks in high school. Everything else is something that happened in between. But I wrote a piece called The Long Run is Lying to You saying, “Yeah, for 30 years the S&P has beaten the world. This is just equities now.”
Most of that, call it 80% of that, has come from richening, from people paying more for US cash flows, dividends, buybacks, whatever you would like and has come from outgrowth. Some of it has come from outgrowing the world, but most of it has been richening. You don’t assume you’re going to richen again by that massive amount every 30 years. That’s I think, on the face of it, silly. So playing the game going forward, diversification across anything non-US equities, bonds, other strategies from trend following to truly hedge long short strategies to merger arbitrage. I’m talking about my book now. Those are the things I believe in, but it’s really hard to make an ex-ante argument why you’d want to have all your money in an asset that has gotten much richer than anything else.
Now, one thing I often find useful when this is just a verbal trick, but when talking about this, someone goes, “I’m going to have all my money in the S&P, it’s done better.” And then you go, “Well, why don’t you have all your money in the best stock in the S&P?” And they’re like, “Wait, maybe we can do better than that. Maybe that’s very risky.” Or if they’re very sophisticated, “Maybe that’s not the most return I can get for the risk I’m taking.” Okay, but the power of diversification stops at exactly 500, huh? It doesn’t.
So there are a lot of ways to make this argument. It’s one that comes up again and again and again. You can go to other markets when they have terrible periods, 10 to 20 years in Japan, and I’m sure someone in Japan from about 1990 for a long time afterwards was saying, “Why invest anything in Japanese stocks?” And that was wrong… Even if it worked out, it was wrong ex-ante for precisely the same reason as investing everything in the S&P because it has been so good for so long, is wrong. And then as usual, I’d probably lose the argument.
Meb:
Most people, they say, “Okay, I hear you Cliff. I’m moving away from SPY, so I’m going to diversify with bonds.” And this quote, Cliff or CliffGPT, the option to hold a bond maturity and “Get your money back if you assume no default risk like we used to assume for US government bonds, is apparently greatly valued by many, but in reality, valueless. We’re at a point right now,” and that’s the end of the quote where bonds, many of which are in pretty big drawdowns, but it feels, and maybe you run in different circles, like most investors are ignoring that, complacent to that. I feel like if stocks were down 30, 40, 50% in some places, people would be losing their mind. There would be riots on Reddit. Talk to us a little bit about bonds. First of all, did you say this? Second of all, bonds are the next diversifier and why aren’t people panicking?
Cliff:
I said something like it. I don’t think I said exactly that, but I said something like it in a different context. Talking about owning bonds directly versus owning bonds through a mutual fund, and this is this whole, if you own bonds directly, you always get… Again, assuming no default risk, you always get your money back. So they’re not risky, but if you own a mutual fund, the NAV moves, I’m going to go with me.
Meb:
It is you and it is from the piece you’re talking about. And the reason I bring it up is because so many people when I bring up why aren’t people freaking out about bonds being down so much? They say, “It’s because you’ll get your money back.” And I say, “That’s not really how this works, but if you think that’s how this works, it’s okay.”
Cliff:
They shouldn’t be freaking out because again, a diversified portfolio across all the things we’re doing has been doing fine. What’s been doing well and what’s been doing poorly switches through time. If they wanted to freak out, which I still don’t recommend, 2022 is probably the year to freak out when stocks and bonds aren’t doing well. A lot of the true alternatives, hedge alternatives actually did very well in 2022. So even there, it’s a bit of a commercial, I apologize, but if equities plunged a ton next year and bonds went up because of that plunge perhaps, I wouldn’t suddenly say it makes a lot of sense to patent making bad equities. First, the thing you should always think about is the whole portfolio. I think maybe the number one error, the number one error people make has to be not being long-term enough, but they’re related.
I think the number 1B era is looking at line items overlooking line items in a portfolio and not their whole portfolio. You have regret. You say, “Why did I do this?” Bonds had gotten very expensive and I’m not saying that was easy to time. They stayed expensive for a long time. Low yields, real and nominal just kept getting lower. Bonds have returned to looking somewhat more historically normal and they’ve returned relatively quickly in the last two years to that. So that entails some ugly returns. Whenever markets get to such extremes, like the Shiller CAPE in early 2000, guys like you and me always say things like, “Well, you could lose slowly on this by just making much less than you normally do for a really long period. Or you can lose quickly.” And if we’re being honest, I’m assuming you say the same thing, we say, “Yeah, we don’t know what’s going to happen over the next year or two, but we know it looks worse than normal.”So you should never panic and you should always look at the whole portfolio.
But if you were going to panic, the time was when we had negative real yields on bonds, not back to where you’re getting 2% on a long-term tip. Looking in the rear view mirror again, yeah, sell your bonds a year and a half, two years ago. Looking through the actual front window, they look pretty reasonable now and a lot more reasonable than they’ve looked in a while and your whole portfolio has survived extremely well if you were diversified. So, I see no reason to panic at any reason here, but the one reason to not panic is not that you’ll eventually get all your money back. I hope I had all the knots correct there. I had about a double or a triple negative. Returning your money in nominal terms is just not that interesting.
Meb:
Yeah. I think the challenge for so many investors thinking in terms of nominal and real is a big bridge. Hardly anyone that I speak to on a consistent basis thinks in terms of reality, at least on the day-to-day. It tends to be more nominal, even though most I think at their core know that the real is really all that matters.
Cliff:
To your point, day to day, the inflation adjustment is pretty irrelevant. Outside of Weimar Germany, we don’t get 100% inflation in a day. So in a day, and I’m going to guess, you don’t look at our P&L, which unfortunately, I look at way too much during the day for someone who preaches the long-term constantly. I’m a major hypocrite on that front, but during the day I don’t think, well, what’s this adjusted for today’s inflation? But over multiple years even and certainly decades, it’s the only thing that’s relevant.
Meb:
So we’re speaking to this young entrepreneur again. He’s like, “All right, I hear you guys, US stocks, bonds,” and you guys had a little piece at AQR where you’re talking about three great ideas for this year and I’d like to eventually spend more time on the third, but the first two, I feel like we have to touch real quick. This conversation around moving away from market cap weighted global, which is all US, which if you’re American, is like you put like 80, 90% in the US. And so here’s another quote, which is like, “Thinking about moving away from market cap weighting [inaudible 00:12:09] contributions to smart beta investing are akin to adding rocket fuel to a traditional engine, it propels performance to new heights.”
Cliff:
Certainly doesn’t sound like me.
Meb:
Sorry I had to include one of those, but all right.
Cliff:
That’s okay.
Meb:
As we talk about, CliffGPT has a sense of humor.
Cliff:
For anyone watching here, Med’s trying to stir up trouble.
Meb:
No. CliffGPT had a good sense of humor. This is using OpenAI. You’re probably tired of talking about this, as I am, but talk to the person who says, “All right, I hear you. I’ll diversify US stocks and bonds. But this whole foreign thing, this whole value thing, this whole emerging thing, which there’s some overlap on the Venn diagram with those discussions today, it hasn’t worked. So why should I be thinking about this, man? Come on, help me out.”
Cliff:
Sure. I always have various ways I tackle this as I imagine you do, but one I do is try to invert the problem. Again, should a Japanese investor be all in the US? Because when you do that, you’re making a statement. Unless you have some particular weird consumption, you can make more economic arguments why you should invest more in your home. I’m going to just think about it as a portfolio. If you’re just thinking about it in terms of real returns in a portfolio, investors in different countries faced with the same facts should not come up with different conclusions. So should all the countries assume the US is just going to do better going forward? That is what has happened. It is sobering that something can go on for 20, 30 years. It is comforting that it didn’t end your life to be global instead of the US.
Another way to put it is that some country had to win. If you started 30 years ago, you know some country’s going to win over the next 30 years. 30 years ago, would you tell your 30 year older self, should you wait over the next 30 years and then make sure you have all your money in what has done the best over the 30 years that passed? Again, I’m trying to surpass my prior sentence for complication and parentheses in one sentence, but all of this, you’re never going to commit to everyone. The people who say, “I’m just doing what has worked,” I think they lose long term. They win for an annoyingly long period to those of us who don’t think they’re right, but that’s why there are opportunities to do better. So I like arguments like that.
If France had won over the last 30 years and not the US, would you say put all your money in France? And I think to a person, even the French would mostly go, “No, that’s silly.” But the US is such a big market and because we’re all colored by lifetime experience, I’m going to make up the numbers that were in my paper. But if the US started at two thirds of the price of non-US stocks and it’s now one and a half times the price of US stocks, for 30 years, you got a huge benefit of this repricing. It’s now one and a half times. Do you really want to assume that going forward? And what’s your loss for diversification at this point? It’s a hard argument.
The world pointing and saying this just hasn’t worked and the notion that something could not work over such long periods, it defies credulity to people. Yet when you do the math, which I know you and I have both done, doing the right thing going forward cannot work for a really long time, particularly small edges. One thing we should be honest about is nobody says a globally diversified portfolio has triple the sharpe ratio of a US portfolio. It’s a modest advantage if people like you and I are right about that. And when you look at that modest advantage, the chance it doesn’t work over even very long periods is real. It is still a modest advantage. I don’t have a better one than that. I would’ve convinced the world a long time ago if I did.
Meb:
One of the things you do a great job of is, I think, being transparent, honest in both the good times and the bad times. I think that lends an air of legitimacy to what we do in our world. We all know it’s easy just to go out and talk about what’s working and say the things that investors want to hear because marketing what’s working is way easier than marketing what’s not working.
One of you all’s papers is about running money for the long run, I think it hits home in a lot of areas. First of all, I thought of you this morning because I got an email from somebody who said, “Hey, Meb, I just sat down with 10 heads of desks.” They all agreed to a T, cash is moving off the sidelines into fixed income and a little part of me died thinking of you, but not the topic of what I’m going to talk about. The other topic I got, and I don’t want to shame this person, but is a multi-billion dollar advisor who was like, “Hey, we’re going to allocate to one of your funds. It’s going to be a starter position and we’re going to see how it does.”
Cliff:
Over the next what period?
Meb:
Exactly. So I said, “I wrote to our salesperson,” because I didn’t have the conversation. I said, “I almost want to draft a guide to, hey, we’re going to mail this to people once they buy one of our funds or strategies and say, ‘Here’s some things to think about or what you’re going to do,’ because I’m not sure what you’re going to learn in the next…” For them it’s probably one, three, six months. You talk and you get a quote. If you have a three-year period where something doesn’t work, it ages you a decade. You face immense pressure to change your models. You have bosses and clients who lose faith and I cannot really convey the amount of discipline you need.
So many people, I feel like this guy when he’s buying this fund, he’s going to say, “All right, if it outperforms over the next one, maybe two years, we’ll buy a bunch more. If it does poorly, we’ll probably sell it.” Which in many ways is probably the opposite of what people should be doing. As someone who’s been doing this for a while, talk to this never ending struggle with not just retail but professional client expectations and how to work around it, how to set up guardrails, what do you do? What do you tell them?
Cliff:
First of all, there’s no magic bullet. I run a firm that used to be double the size after a value drawdown from 18 through 20, which we were, if I may brag on your thing, right about, and have been proven right about. So obviously, I don’t have perfectly persuasive skills to get people to get past this short term, though I have tried many times, so I’ll try a few of them here. Let me start out with one difficulty that both you and I probably face. I wrote a dissertation on momentum investing with data ending in 1990. You want to feel old? When the outer sample data in your dissertation is longer than your in sample data, you’ve been doing this a while. You wrote a canonical paper on trend following, which is in the CTA world, we call it trend following. In the individual equity world, we call it momentum. I have no idea why.
So one thing when I start explaining this, someone who knows a little bit, who’s smart, often looks at me and goes, “Hey, you’re the guy who says to buy what’s going up. So [inaudible 00:19:07] a little bit of hoisted by your own petard there.” I’ve never quite figured out what a petard is, but let’s not go into that. My response to that is a few fold. First, I didn’t say to only do that. Maybe the reason I graduated writing for Fama and French was I said, this is a wonderful compliment to value investing and not instead of value investing, and here I am quoting myself, this is not ChatGPT Cliff, though I’m sure it would do a very nice version. I have often said that a lot of investors act as momentum investors but with a time horizon more appropriate for value.
So if someone said the following, “Over the next three months, if you do well, I’ll add a little bit more because things tend to trend.” I have no beef with that. I’ve written about that. You’ve written on that, but that’s generally not what they mean. This is a case when someone’s making a bad argument with us , maybe thinking of a good one that would lead to their behavior. The problem of course, is I think investors think they learn a lot over three months about the true long-term, unconditional mean, and there’s a reason these things persist. Eventually going to get to your entire question, but one interesting subpart is you and I live in a world of a paradox where if we ever convince enough people we’re right, the advantage to a lot of what we do goes away. So if suddenly nobody did this, opportunities go away, markets become more efficient.
So this notion that math is very hard to convey of how little you learn in three months. You know who learned a lot in three months? LTCM learned a lot in months. Barings learned a lot in three months, and thankfully, neither you nor I have ever had to face something like that and I don’t think we ever will. But barring that, even multiple standard deviation events, you learn very little and people think you learn a lot. So you make your argument. You try to explain this, I’m not being nihilistic or defeatist. There are a lot of people who will get it and will understand, but there’ll always be some who don’t and without sounding too riley, cynical, we need those people. If someone’s going to consume alpha, someone has to supply alpha. That’s a very anodyne way of saying they’re getting it wrong when they do that.
Now in business, you try to convince people not to do that. I’ve had people say that. I once had an investor, I think they said two weeks, and this is actually a funny story. I was doing a presentation in Switzerland, this was the late ’90s, so it was much more of a fund of funds world, Swiss Fund of Funds were big, and I made a joke. I didn’t want to name the investor, it was not this investor, and I said, “So I was talking to Baron Rothschild about his investment and he said, ‘If you have a good two weeks, we’ll take it up.'” Got a good laugh in the crowd because they did get how silly two weeks are.
In the car on the way out, my partner who’s more on the business side where I’m more on the research and trading side, got a call from a representative of the Rothschild family to go, “Why is Cliff lying about them on stage?” And I was like, “I don’t want to be rude, but you guys are still around?” I was kidding. It was a joke. I know I didn’t say that to them, but then I made one of my favorite jokes ever that got zero laughs at the end of the whole thing. I did apologize because I used their name not knowing that they’d be offended or even that they were there, to be honest, and I said, “Can I still make fun of the Hapsburgs?” And they said, “No, they’re still around too.”
Meb:
Yeah, man, you can’t offend the Illuminati, Cliff. They’ll come after you, man.
Cliff:
I’m careful about that. But you learn rounding to zero. You learn rounding to zero in a year about true unconditional long-term means. Again, barring true disaster, you need an uncomfortably long amount of time to learn much. You know what you can also learn? I don’t want to have to go to Barings collapse extremes. If you think you’ve developed a three sharpe ratio strategy, first, you might ask yourself why I think I’m better than Jim Simons’ medallion fund? But if you think you’ve developed that, you can also learn reasonably quickly you don’t have that because yes, three sharpe ratios don’t really have a lot of down years. But barring a true disaster or a very unrealistic assumption, if you think the sharpe ratio of your strategy or the advantage of your strategy over another pretty good strategy is a 0.5 information ratio, sharpe ratio, whatever you want to think about it. That’s not a two standard deviation event to lose for a decade and that’s a wonderful advantage to have. That means your advantage is a better risk adjusted return than the stock market’s advantage over cash.
Stock market’s been more like a 0.35, 0.4 and one lesson, which is a good one that people seem to have learned is that we should be really long-term about something like the stock market. People understand there could be draw-downs. It’s the same for any real life advantage with a reasonable risk adjusted return. So at the end of the day, I try real hard, I give the mathematical arguments, I give the folks here arguments. I give them, if you want to be a trend follower, fine, but only do it with a little bit of your money and only do it to time trades in or out of a good strategy. Again, I’d be a hypocrite if I said, you can’t do that and then I let the chips fall where they may. You’re going to convince some and you’re not going to convince some.
Meb:
One of my all-time favorite quotes on this topic that it might be a top 10 favorite quote from me in general, Ken French, who we have coming on the podcast soon, the French and the French Fama, your old thesis advisor had a quote where he said, “The inferences that people try to draw from asset classes or strategies for one, three, five or even 10 years is crazy,” or something like that, he said. And how many people when we sit down in a meeting that are going to allocate or contemplate it, saying, you may not learn anything about our strategy in the next one, three, five years. That might not even be enough information to really determine if this is particularly good or not. How many people are going to nod their head and say, “Yes, I agree with you.”
Cliff:
Not shockingly, of course Ken is right and we’re saying the same thing in different ways. Another way I put it is, and I’m no cure for this again, but the time horizon that matters for statistical inference is dramatically longer than the time horizon that matters in the real world. Unfortunately, for the real world, it’s the real world that’s wrong. Sometimes, people will throw quotes out, right? The market can stay irrational longer than you could stay solvent. The strategy you can’t stick with no matter how good it is long term isn’t good in a practical sense. There are two things about those quotes. They’re both true and horrific. They’re true. The strategy you can’t stick with is not a good strategy. The horrific part, and I should only say part, is that they’re often thrown out as an excuse not to try.
If somebody says the time horizon is dramatically different, we’re going to work really hard to extend our time horizon. Maybe we won’t get it up to 10 years, but maybe we’ll get it to five years from two years. We’ll talk it through. We’ll do scenario analysis. When we invest in something, we’ll go, “Look, what if we get the three standard deviation negative years? The two standard deviations of negative five years, would that change our belief in this?” If that happened with a valuation change, and particularly, it shouldn’t change your belief in it because you know exactly why it happened and you either think it’ll reverse or you don’t think it’ll happen again. So there are a lot of things you could do, a lot of arguments you can make, but quite simply, and I’m comfortable with this statement, 99 out of 100 times, Ken French is right. Don’t ask me the one out of 100. That’s between me and Kent.
Meb:
One of the things that I wanted to spend a little time on is a topic, one of your top three ideas of 2024. You guys value emerging markets. Then alt trend strategies and listeners of this show have probably heard a decent amount about your traditional price-based trend, whether that’s through long flats or whether it’s through CTA strategies. I was tweeting the other day because some of these, the last couple months have been particularly interesting. Here’s another quote of yours, maybe, give away the answer. “In trend following, which has been doing a lot more things, more esoteric securities. The best example is alternative commodities such as milk, coal, or my personal favorite Malaysian palm oil.”
Cliff:
I do get a kick walking around the firm and hearing people talk about Malaysian palm oil. I always do a little bit of a double take. I’m like, “We trade that?” And they’re like, “Yeah, we trade that.” I’m like, “Cool.”
Meb:
Yeah. Some of the very esoteric trends, you see cocoa makes NVIDIA look like a flat line on some of these charts that have been going crazy up and down, but the really interesting I wanted to spend a minute on is you guys have extended this concept of pure price-based trend, which listeners, AQR does such a great job and not just publishing the research. There’s data sets online, we’ll put some links in the show notes to go play around with for the nerds out there. But you guys have started to move into some more over the past decade I think, all trend ideas where it’s based on fundamental momentum. Could you maybe explain what some of these concepts are for listeners, because it’s, as you mentioned before, pretty interesting in the sense that it’s similar but different and diverse enough to where these things could be complimentary.
Cliff:
Sure. Just for everyone listening, Meb is throwing me a hanging curveball to talk about wonderful investments that have worked out, so thank you, Meb. We’re overloading the word alternative, but it’s neat… I haven’t thought of it the way you just said it. You can think of the two things that [inaudible 00:28:53] over the last 10 years we’ve done to improve our trend following strategies as two forms of alternative. One, you mentioned already, just getting into more esoteric securities. It turns out each… Of course, they have lower capacity than the [inaudible 00:29:07] dollar or the S&P, but there are a lot of them, and we find the trends are very strong in these things, and it’s not just esoteric commodities. It’s structured trade, shape of a yield curve, very equity factors that we believe in themselves, the long short factors. We call those alternatives in that they’re not what CTAs had mostly done in the past.
Most of the CTA world, including ours, was price trend on the major markets in stocks, bonds, currencies, and commodities. That’s one form of alternative. It’s not quite as extreme as micro caps, but think of that as, there are small caps and micro caps that have more alpha than we’re not extracting that all from there. Another way to use the word alternative, which I might steal from you, and I in fact, I’m stealing it right now, is an alternative way to measure a trend. The historic way to do it, and I’m taking Kohl’s to Newcastle here by explaining this to you I know, is price trend. Of course, every trend follows every CTA, every managed futures account has their own favorite way to do this, and we all will pound the table and claim ours is the best when obviously, mine is the best and Meb’s is second, but that’s… Or whoever’s interviewing me is second and it happens to be Meb today.
No, seriously, that can be simple price momentum over different horizons. That could be a moving average type process, that could be a more complicated mathematical filter, but they’re all based on price movements and trying to capture the tendency of what’s been going up tends to keep going up. In the individual equity world and sometimes insight comes from just luck and here, I think we got lucky. The fact that we are quantitative equity traders as well as trend followers means we have spent our life in the individual equity world and in the individual equity world since pretty much the late ’80s. These are the 1980s, I’m not that old. I wrote my dissertation on price momentum, but forms of what we might call fundamental momentum are earnings being revised up on a company, our earnings surprises coming in positive, our margin’s improving, our inventory’s growing or shrinking. Those have been part of the quantitative world that we would broadly call a fundamental momentum, forever.
So at some point, and I’m sure this has happened to you, Meb, and anyone listening, when you come up with something that’s really you think cool, but also obvious, you get a, wow, this is great. We can make things better. And you get what, a moron I am for not thinking of this 25 years ago because it was there. I console myself and I don’t think a lot of people are doing this yet, so we can’t be that stupid, but I do think we could have come up with this earlier. At the market level, analogies to fundamental momentum hold.
To make the simplest one is equities because I just described it. Instead of just looking at the price momentum of the US and the price momentum of Japan and then every other country, look at the aggregated earnings revisions with some waiting scheme that approximate the index year trading, our earnings being revised up faster or slower. Our earnings surprise is coming in faster or slower. When you get out of equities in the FX world, what does the balance of payments look like? Both in terms of realized economic announcements and economists’ forecasts of the future, revisions to their forecast?
There are many more, I’m not going to do it justice, but call these broadly speaking economic trends. Well, it turns out a few things. One, economic trend works at least in the very similar ballpark to price trend. We didn’t throw out the price trend. Second, call it 0.5, 0.6, correlated. Meaning, if over the same markets, Meb built price momentum and obviously the specific form of price momentum will change this, but broadly speaking, if you built price momentum and I built pure fundamental momentum by where the fundamentals are getting better and sell where they’re getting worse, we’d be about 0.5, 0.6 correlated. This is also luck. If they were 0.9 correlated, we’d say pretty cool results, but not that helpful. 0.9 Doesn’t move the dial very much. It’s just more of the same. Maybe you do some for robustness if one breaks, you have the other, but it’s not going to change your average sharp, your long-term sharp very much.
If they were zero correlated, you’d say, “Wow, that was really great, but it’s totally implausible,” and you wouldn’t believe me because that would mean prices measured over the same period as fundamental changes had nothing to do with each other. They were never leading to the same positions. So again, dumb luck, maybe some real economics going on, but from a lucky perspective, it landed in that sweet spot where, as a rather nontrivial improvement to the risk adjusted return, but not an implausibly uncorrelated one that we just wouldn’t believe. Finally, you can always improve a pure trend following strategy by saying, and some have done this over the years, add an FX carry strategy to trend. If that were your whole portfolio, that’s a beautiful addition because FX carry on average works. Sometimes it blows up and those have tended to be best periods for trend following strategies.
I say have tended to be, it’s not a guarantee, but we live in a world of risk. They’ve tended to work well together, but that’s not how most investors I think look at trend following. I think most, if they allocate to it, as of course I think they should, and I think you probably agree with me on that. They look at it as, I want to make money long term, but I want to do particularly well when the blank hits the fan, particularly in equity markets, which are most of our wealth. FX carry does not have that property. It has the opposite property.
Turns out, again, I don’t think this is luck. I think this is embedded in economics, but from our perspective of designing it, it was very lucky. Fundamental trend absolutely preserves and maybe even improves that property of doing well when things fall apart in the traditional world. Some disasters the market sees first and price momentum will get better than fundamental momentum, the market saw a deterioration to come. In some disasters, the market is whistling past the graveyard. The fundamentals have started to crumble and price didn’t pick it up as much. You don’t know which, we’re not claiming we can time which one, but the two together both improve the risk adjusted return and seem to be a more robust hedge on those terrible periods for global stocks and bonds.
Meb:
The way I think about it, the narrative is a price-based trend. You’re looking for that meat of the move and the fundamentals on some of the cases that these reversions where the trend ends or trend begins said differently, can be pretty big moves at times and having some of these additive signals. And listeners, like Cliff is saying, it doesn’t have to be a magical three sharp like these little dial turns that if they don’t correlate. It’s funny because Ray Dalio actually has a wonderful quote where he says, “The holy grail of investing is putting together 10, 15 non-correlated investing strategies.” I had a little fun poking at a recent book. I’m not going to name the author, but the name of the book is called The Holy Grail of Investing, which has a little hubris naming, anything in our world, the holy grail of investing, but then managed to put together about eight return streams that were basically all equity beta along in some form or another. They’re all the same trade. I said, “Well, you’ve perturbed the intent of this quote.”
Cliff:
You took the Dalio concept and utterly destroyed it. You and I would write something with that title, but it would be a sarcastic title. We would not write something with that title being serious because we’d laugh ourselves out of the room.
Meb:
Well, I think anyone who’s been in this world long enough, I think I said on Twitter the other day, “Man, I’m going to a crypto conference today because I have some friends, local at USC at this Bitcoin conference. I love just to go get this sentiment and check out what’s going on and get a vibe for how it feels.” But I would say to most of my crypto friends as their markets hit new all time highs, the number one trade that I usually, if I look for and allocating to someone as a professional money manager is humility or at least this knowledge that you’re going to get taken out to the woodshed eventually at some point just surviving that is probably the biggest key to being an investor.
Cliff:
Markets are extremely humbling things. No one’s ever accused me of being personally over humble, but when it comes to an investment process, appropriately humble, meaning you realize your edge, if you’re as good as you think you are. Your edge will lead to some really painful times and some really extended painful times. Humble is just accurate there. So yeah, I’m someone who doesn’t acknowledge that and I’m not saying that about the crypto people. I could defend them by saying they’ve actually stuck with it through some crazy ups and downs, so I don’t even know how it applies to them, but a manager who’s had a really good long-term run who doesn’t appreciate this, I’d run from.
Meb:
You just gave me a great idea. We both joked on Twitter a fair amount about this volatility laundering in the world of private equity, what we should really be doing is a crypto fund that only marks once a year because then you take the volatility not from 20 down to seven, which the private equity people try to do, but you’re taking it down from 90 down to maybe only 20. Talk about a magical transformation. That might be the trillion-dollar idea.
Cliff:
It might be. It faces the same problem we’ve always had. Every 10 years, somebody I know either our firm or even some similar firm says, “What if we did something like private equity? And I go, “The problem is we do know the prices and it’s somewhere between unrealistic to illegal not to tell people what the prices are.” Now, in private equity, I’ve been a gadfly to that world, though it’s not put on the numerator side. People do debate how much money they make IRRs versus returns or whether you have to be top quartile or whether the average, I’m not getting into that. I’ll take them at face value on the numerator. It’s been on the reported vol and they absolutely could market to market every day.
Trust me, if the market crashed 50% tomorrow, the public markets, these are brilliant investors who know companies at a deep level that I can only dream about. They could tell you roughly what they think they could sell it for. For some weird institutional quirk, nobody makes them do it. So yeah, I would love to run that fund. You would love to run that fund. Bitcoin in that would be… Let me take the other side for a second. The vol and bitcoin may be a feature, not a bug. If people didn’t get to see it every day, they might actually lose interest in it. But in private equity, I’m pretty sure the laundered low vol is a feature to people.
Meb:
On the trend following piece, sticking there for a second, we’re an ETF shop, so we allocate because we have to, to direct securities in ETFs, but I’ll give you guys a compliment. There’s no better compliment to give than saying, “My mom has owned you all’s managed futures mutual fund for many years.” So Mom Faber, the fact she’s been allocated to this fund is a compliment.
Cliff:
Meb, I’m glad you brought that up because I was going to raise it. She’s a really annoying client. She emails me every day. She’s upset if we have a down day. Everyone should know. I’m kidding. Thank you, Meb.
Meb:
She has problems with palm oil. She’s probably the best investor I know because she’s from, we’ve spoken to this many times on the podcast, where each investor has their own journey, which is highly colored by the markets they’ve experienced. So we’ve had investors that grew up in Peru and were subject to hyperinflation on the podcast and people in all sorts of countries around the world that have had very different experiences. Even within the US you started investing like she did, in the ’80s and ’90s, like buy and hold equities. You just buy stocks, you hold them. People like me that graduated college in 2000 had a totally different experience on and off.
Cliff:
You had a fairly terrible decade.
Meb:
Yeah. But I didn’t have any money. So that’s the beauty of starting in a bad time. So let’s say you’re sitting down with an institution or even a high net worth, the Rothschilds, maybe an institution, we’ll call them, [inaudible 00:41:35] I don’t know.
Cliff:
I’m not going to insult them again.
Meb:
And let’s say they got whatever it is, 10 million, a hundred million, a billion, 10 billion, and they say, “Look, all right, Cliff, we hear you. We’ve moved from just US stocks and bonds, market cap weight. We’re going to add some of these, a little bit of global diversification and other asset classes, so we’re now to a long portfolio of global assets. How should we think about position sizing the results?” And they say, “We are totally open to having a portfolio where we want it to be optimal, sharp. We want it to grow but have manageable volatility. How much should we put in some of these all types of strategies?” How do you begin that discussion and what do you think is a reasonable amount for a person who has that question for you guys?
Cliff:
We’ve all probably been down the road with optimizations. If you sit down and write down what you think you can make on uncorrelated alts, it may or may not be realistic, but you sit down at Excel and you run a solver and it says, put 83% of your money in there and then you go, well, nobody’s going to do that. And then you back into what most people will do, which is highly non-mathematical. It’s a little bit of a guess. To be brutally honest, I don’t think we have much better than that. If you believe in these things, most ex-ante, it could be a formal optimization or just an informal one, will want a lot of those. For one thing, if you’re willing to do any gearing at all, you don’t have to do that much less of what you’re doing already.
These are not very correlated assets, and I don’t mean the average hedge fund out there. We’ve written that the average hedge fund’s about 0.8 correlated to just the S&P 500, but if you truly construct long short portfolios that are as short as they are long, similar betas, a decent version of any of the existing factors in the literature and some that are more proprietary. I’m sure you have them, I have them. Those are the things I’m not allowed to talk about on YouTube, but you end up with that exercise that I mentioned earlier. Remember I said people use an excuse if you don’t invest in it, if you can’t stick with it won’t be good. That doesn’t mean I’m not a practical person who says, you got to run that exercise. You got a method act. You got to say, “All right, this is what it’s going to feel like when everybody else is making money and I’m not making as much because I believe in these things.”
What I said before was, you can’t use that as an excuse. Your job is to build the best portfolio. So you fight and push and argue with yourself and others to get to the point where you could do as much as you can, but more than that is to no one’s advantage. So we can run all the optimizations we want, but spreading your bets across a bunch of alternatives because you’re not going to be able to stick with one, running the scenario, if you’re an individual you got to run it yourself, but if you’re an organization, what’s it going to look like when you get the down two standard deviation five years for the alternative part of my portfolio? Am I going to close it down? Am I going to stick with it?
Or if it’s actually cheapened in some meaningful way, which doesn’t always happen with losses. If you’re losing a momentum strategy, it doesn’t get cheaper going forward, but more often with a value strategy, it does. Would I even add to it? To say the remarkably obvious, doing half the optimal amount and being able to stick with it is ridiculously better than doing the “optimal amount,” which is not optimal because you can’t. Obviously, there’s tremendous art. Obviously, I don’t want people to use that as an excuse to do nothing. I don’t want it intellectually and as a selfish businessman, I don’t want it and I think it’s wrong, but you still have to do that exercise. The one that drives me crazy still must be done.
Meb:
Where do you think most people’s sweet spots are settling here? Is it 10, 25%? Because I know most… I guess it depends on what you call…
Cliff:
16%.
Meb:
What you call alts. We’re an outlier here because we’re a trend-falling shop through and through, so we’re what we would call our flagship is darn near half, but we’re crazy. I feel like most institutions, if they’re on the avant-garde and I’m not including private equity as an alt, yeah, a quarter seems like it would be the upper bound.
Cliff:
That is the upper bound. Yeah. Of course, there may be outliers. The 99th percentile is going to be in the 20s. You don’t see a lot of major institutions with half in what you and I would call alts and you said it right, private equity might be great if it outperforms its beta after fees, but it is not an alt in the sense that you and I are using it and your 50% might be crazy in a, make Meb’s life easy sense, but it is probably not crazy in an actual mathematical sense. That’s the nicest thing I’ll say about you, Meb, not crazy.
Meb:
I’ll take it. You guys have written a lot, you alluded to earlier, talking about the fundamental trend following where there’s some ideas and kernels and little bits you pick up and say, “Well, I didn’t think about this earlier. Just look forward to the horizon.” What are you guys thinking about now?
Cliff:
Couple things. One, it is a good opportunity for me to clean up a quote of mine you started the whole podcast with, my comments on AI that it’s a little oversold and it’s more evolutionary, not revolutionary. I don’t think I’m actually a hypocrite, but it might sound that way, that we have a major effort in AI. Brian Kelly of Yale is an AQR partner. I think he’s the best, but I’m biased. He’s right up there with the top machine learning investment researchers in the world. We think he can make what we do better. Maybe the only example I’m allowed by my team to use is natural language processing, forms of fundamental momentum where you could do a lot better than the old keyword searches. You probably remember quants would take a press release or an earnings call and look for the words increasing and do plus one, decreasing minus one with the obvious flaw always been if the line was massive, losses are increasing, you probably shouldn’t have given that a plus one.
As good quants, you could be wrong a lot as long as you’re right a little bit more often than you’re wrong, so that wasn’t a disaster. It might’ve even been good, but modern techniques make that a lot better. We think machine learning is going to be pretty important when it comes to allocating among factors. We don’t mean short-term timing and you have to constrain it. If you let machine learning try to do every factor in the world, it’ll grossly over fit. But if you started with a stock pond of factors you believe in, we think there’s some improvement there. Where I say it’s oversold or it’s not as revolutionary is a simple statement. It’s still just statistics. It’s data going in, with an answer coming out. It’s higher powered statistics with more computing power than we had before.
As we’ve known forever more statistical power can be used for good or evil. If it’s used to purely over fit, it’s going to make things worse. I am in the simultaneous, seemingly contradictory, but I hope not, position of saying we have a huge effort underway. We think it’s going to make our world a little better or maybe even just prevent our world from getting worse. The world might always be trying to arbitrage your strategy away. You never know if you’re swimming to go forward or to tread water, but I don’t think it’s going to change everything tomorrow. I give you an example, what is the equity risk premium? What is the premium for high quality over low quality stocks or cheap over expensive stocks? I don’t think ML is going to tell us squat about that. We have small data, not a large data problem with that. You know what’ll tell us a lot about that? Another million years. That’ll tell us a lot about that, but it’s not some esoteric nonlinear relationship that you need to tease out of the data with ML.
So I think as long as people don’t start thinking their world’s about to get three times better because of this, yeah, of course you use the new techniques and you try to apply them wherever you can and you try to make things better as much as you can and I’m very optimistic about that. But I do think this might set me apart a little bit, that many in our industry tend to oversell the new thing as a panacea, as a holy grail, if we can go back to earlier. I don’t think it is. I think it’s making our lives better, but not utopia.
Meb:
This is somewhat of a softball question for you because for everyone else, it often tends to be a little harder, but for you, it’s flipped on its head where we say, if Cliff goes and sits down at the local diner, happy hour, poker table, wherever, with a bunch of buddies in our world, so contemporaries and what’s a belief you hold that 75% of your peers if you stated, would shake their heads at?
Cliff:
I’ll give you two. One is, we’ve written a bunch on, I know you’ve seen it. We are not believers in the small firm effect and that is part of the firmament of quant investing. It’s one of the Fama French, three or five factor models and maybe the one out of a hundred things I mentioned before that Ken and I don’t see eye to eye on. We think it’s 100% coming from small and microcaps having higher betas. I’m very biased, but I think we’ve put some convincing stuff on that and certain times, things become part of the firmament and then they’re just very hard to just plot. They’re just quoted as conventional wisdom. So I think we probably could convince a lot of quants we’re right, but I think many will just automatically include that. We do think at least gross after trading costs, it’s closer, but most factors work better in small, so if you’re going to have a factor tilt, doing more small might make sense. The pure small firm effect is not that, it’s just small against large.
Meb:
It might be easier to convince the people at the table now as small caps have been sucking it up relative to the large cap market cap weighted index than it was maybe a number of years ago.
Cliff:
As you know, I go somewhat the other way on that if I have to go either direction, I would never dain to use that. Actually, I’d use it in a second if it would help me, but that’s not the right argument. The other one, and I think a lot of people would disagree with this, but I’m not actually certain. I think over my career, the market, call it the relative pricing of stocks, has gotten less, not more efficient. And that sounds really weird. I think we all think that we live in this whiggish world where things get better, technology gets better, access to information, ubiquity of information, it’s at all of our fingertips. We can trade instantly and at lower costs. How could that lead to a market getting less efficient? And I plan on writing about this soon. I’ve been saying that for about two years, so no one should hold their breath because it’s a hard one to write about because there’s not going to be a proof. There’s nothing to point to on this.
Anecdotally, I can say we’ve been in love with measuring the spread between cheap and expensive stocks literally since 1999. I’m on my 25th year of measuring this thing, and that was the biggest one in 50 years. And by late 2020, we exceeded it. So in some simple, maybe circular because maybe this started me down this path, but just looking at things, yeah, we think things have gotten crazier twice than we’d ever seen before, but I think the world, and maybe not this table of quants, but maybe if I just sat down with investors in general, who would assume that more information super cheap, instantly delivered with cheap trading costs leads to more efficiency. I find it very plausible to think that leads to less efficiency, to more overconfidence. I’ve joked that the people who believe this leads to more efficiency, are the same people who believe social media must make us all like each other more.
I don’t think that’s quite happened. If you want to go to an extreme example, and I’m not saying this is a fair example, I’m going to extremes, but the US meme stock example is a great example of people thinking because we have all the information and everyone has the internet and we can all talk to each other. The internet is also a great vehicle for creating manias. The wisdom of crowds has always been fascinating. If the crowd is independent, and I know you’ve talked about this too, the crowd is independent. Crowds are incredibly wonderful. You always want to poll the crowd if you’re playing with Regis. He’s since passed. I probably should not have used him as an example, but you can ask the crowd the most esoteric question, very few people know the answer to. It’ll be 35%, get it right, and 65 divided by three get it wrong. And you pick the 35 and you’re going to be right because the signal comes through the noise.
But if the crowd gets to talk to each other, the crowd becomes a mob and the power of polling the crowd completely will go away. And the internet does a lot of wonderful things and some terrible things, but it’s also made it much easier to turn a wonderfully independent crowd into a highly dependent mob. So both from real life experience and some scars from battles I’ve won but have still been battles and from observing a lot of these things, I have the probably iconoclastic view that markets are less, not more efficient than when I started my career in the Stone Age. And frankly, to investors, that’s bad and good news. The bad news is that if things can say some degree of crazy, I don’t think things are always crazy and I object to people who always have a bubble to point to.
But if you think that could happen occasionally, it could happen to a larger magnitude and last longer, making it hard to do the thing we keep referring to. Stick with what you’re doing. It should also be almost as a tautology, making it more lucrative if you can do it. If you make your money from markets not being perfect, if they’re going to be more imperfect, you’d hope long-term, if you can stick with it, it would make you more money. And frankly, that strikes me as fair. A world where it’s harder both because it can be more severe and last longer, but you make more if you can do it, strikes me as the fairest thing in the world, but I do think that’s probably a 25%, not a 75% view of mine.
Meb:
You’ve done thousands of trades throughout your career, like a true quant. When you go on TV, half the time they want to know positions or whatever. You’re like, “Look dude, I don’t even know what we own. I’m a quant.” That having been said, over the years, do you have a most memorable trade or investment? For you could be a day. Anything come to mind? Most memorable?
Cliff:
Yeah. I’m going to tell you a story from my Goldman Sachs days. This wasn’t really a trade, it was more of a day. So we were doing what we’re still doing, just its version 17.0, market neutral, fairly aggressive, early version of what we do. So a standalone font. Goldman partners had seeded it. It was in October. Somehow, it’s always September or October. It was in October of ’97. It was something that has been called the Asian Debt Crisis. Everyone forgets about this one because it didn’t lead to the end of the world, and we’ve had so many bigger crises since then. But the S&P was down about 7% in a day, which is a huge number on any day, but it was also a very calm period. So this was absolutely shocking to people. The then head of Goldman John Corzine is making the tour around Goldman to see how bad it is because a place like Goldman’s is still going to be net long, a lot more than it’s that short.
And he was probably having a pretty bad day. He got to us. I had shown him as part of a VIP tour a month ago. We had an Excel spreadsheet. We added F9 to update the P&L. It was the Stone Age of technology, but it would say how we’re doing that day in the fund that Goldman’s partners had invested in. We didn’t man it all day. Why should we? We checked it every once in a while. We were voyeuristic about it, but not at every second level.
So John, the freaking head of Goldman Sachs walked up to it alone and it says, we’re up 7% and the S&P’s down 7%. I run out there to say, “John, we’re not up 7%.” He’s like, “What are you talking about? Is it broken?” I’m like, “No. One of the major trades we have on the relative value trade is short the US and long Europe.” And I forget why. Some value quality moment, it had to be some combination and one, I don’t think I said it exactly like this, but fairly close. One great way to look like a genius is to short a crashing market against being long or closed one.
Meb:
Technically, you were doing volatility laundering before it was cool. It’s just not even-
Cliff:
I was, and no one knows how to not volatility launder that one because they’re not open at the same time but I was. The funny part is I told them, “Look, if Europe opens up down as, or maybe a little more than the US tomorrow we’re going to be flat,” which should have been a home run. We had made a lot of money in a prior two or three year bull market. And if you say you’re market neutral, it’s a nice acid test. It’s still one day, but it’s a day you really would like to be flat. You cannot take a person from up seven to even what should be a home run of flat and have them still be remotely excited. If I had gotten in there and told them flat beforehand, I think he would’ve said, “That’s awesome. I’m getting bad news everywhere else.”
Instead, I was a major disappointment to him that day. But later that day, it’s funny, this actually did happen the same day, the head of Goldman’s private equity, a great guy comes by and just goes, “How are you guys doing today?” And I was smart enough, I had learned my lesson. I said, “Flat.” He said, “That’s awesome.” He was smart. He thought that that was awesome. And then he said, “Me too.” I think I had my first or maybe my second or third argument about private equity vol that same day. So I just tied a lot of our talk all together right here.
Meb:
I love it. Cliff, you’ve been very gracious over the years publishing a lot of content. I put you on the Mount Rushmore of investors that also share their content up there with a lot of people we know and respect. If people want to find what you’re writing, your missives, what you’re up to, where do they go? Best place to find it.
Cliff:
Aqr.com, we have a tab called Education. I have a section of that is probably the least helpful, which is my… I don’t like to call it a blog, but it really is. And then data sets and current papers we have written. For any of the researchers you see, you can go out to the SSRN and look up their page. Most of that will be on our website too, but there might be another way to access it. Or you can listen to Meb and I talk on Twitter.
Meb:
My goal for you one day is to eventually write a paper that has a title and actually no main body content, just footnotes. If you read Cliff’s stuff, listeners, you got to read the footnotes. That’s where all the good stuff is. Cliff, it’s been a blast. Thanks so much for joining us today.
Cliff:
Absolutely. Thank you-
