0:02
Hi, I'm Jim O'Shaughnessy and welcome to Infinite Loops.
Search this video
0:02
Sometimes we get caught up in what feel like infinite loops when trying to figure things out.
0:07
Markets go up and down, research is presented and then refuted, and we find ourselves right back where we started.
0:13
The goal of this podcast is to learn how we can reset our thinking on issues that hopefully leaves us with a better understanding as to why we think the way we think and how we might be able to change that to avoid going in infinite loops of thought.
0:36
We hope to offer our listeners a fresh perspective on a variety of issues and look at them through a multifaceted lens — including history, philosophy, art, science, linguistics, and yes, also through quantitative analysis.
0:49
And through these discussions help you not only become a better investor, but also become a more nuanced thinker.
0:56
With each episode we hope to bring you along with us as we learn together.
1:02
Thanks for joining us, now please enjoy this episode of Infinite Loops.
1:11
Disclaimer: Jim O'Shaughnessy is chairman and Co-Chief Investment Officer of O'Shaughnessy Asset Management, where Jamie Catherwood is an associate.
1:17
All opinions expressed by Jim, Jamie and podcast guests are solely their own opinions and do not reflect the opinions of O'Shaughnessy Asset Management.
1:26
This podcast is for informational purposes only, and should not be relied upon as a basis for investment decisions.
1:30
Clients of O'Shaughnessy Asset Management may maintain positions in the securities discussed in this podcast.
1:39
Jim O'Shaughnessy: Well, hello, everyone.
1:39
It's Jim O'Shaughnessy with my colleague, Jamie Catherwood for another episode of Infinite Loops.
1:44
Today, I have a very interesting guest, who I first met at a soirée that Patrick, my son, and Brent Beshore put on, where I wish you could see Dan.
1:58
Dan Rasmussen was the founder and CIO of Verdad.
2:04
I think I got that right. Verdad?
2:04
Dan Rasmussen: It sounds good to me.
2:07
Jim O'Shaughnessy: Verdad?
2:08
I like it, Verdad Advisers.
2:08
Now, Dan, I got to say, you, honestly, you are the Manchurian candidate, and the reason I say that is because you are so pleasant-looking and you're a good-looking guy.
2:16
You just look like you fit in.
2:16
You could probably walk in to any club in Manhattan and they'd say, "Right this way, sir."
2:21
Yet, what I love about you and what I want to talk about today is I was making some notes for this, and right in my mind pops the Dr.
2:27
Dre song with the Eminem lyrics that, "Me and Dre stood in front of the burnt down house.
2:40
There we were with matches and a full can of gasoline and still no one found out."
2:44
Dan Rasmussen: It's the first time for the record I've ever been compared to Dr. Dre.
2:47
So, this would be one to ride home about.
2:50
Jim O'Shaughnessy: So, what I love about it, though, is you are a classic debunker.
2:55
I love that because what you debunk, we have a lot of things in common and we're going to talk about them, but one of the things that is great is, do you have to have security when you're going to a VC conference or- Dan Rasmussen: I did a conference, a big endowment conference in Toronto, and that was actually my opening line was that I said, "Well, you guys, thanks for paying for my flight up here and my hotel and also for the bodyguards."
3:20
They had to escort me after, and then I just spent half an hour.
3:25
Just to be really obnoxious, I did a survey and we said, "What are your expected returns expectations for private equity?"
3:35
There was a real endowment professionals.
3:35
So, I think the majority said 4% a year outperformance public equity.
3:39
Dan Rasmussen: Then I said, "What do you think the average credit rating on your private equity companies is?"
3:44
Dan Rasmussen: I think, again, the majority of some said investment grade, and then I just went through it and said, "Well, here are the last 10 years, which are below four and the actual credit rating is triple C or single B for 90% of it.
3:52
So, you guys are way off on returns and way off on risk, and probably your job is risk versus return.
3:57
So, we might have an issue." Gee!
4:02
I thought I was going to heckled or thrown eggs, but I managed to get out of there alive. Jim O'Shaughnessy: Yeah. That's really cool.
4:06
Like me, I bet your motto is nullius in verba, right?
4:09
Dan Rasmussen: You're going to have to do the translation for me, Jim, because it's been years since my Latin.
4:16
Jim O'Shaughnessy: Take no man's word for it.
4:17
Dan Rasmussen: I like that.
4:19
Jim O'Shaughnessy: It is the motto of the British Royal Science Society that Isaac Newton founded.
4:24
Dan Rasmussen: I love that. Yeah.
4:24
I take one thing on faith, and that's the resurrection of our Lord and everything else I'd like to see the evidence, please.
4:31
Jim O'Shaughnessy: There you go. Okay. Well, let's jump in.
4:32
I'd like to start with your newest piece of work, which I found fascinating, about emerging markets and crisis.
4:37
I'm going to let you go, but emerging markets have always been a challenge, especially for quants, because the data is iffy.
4:51
There's a lot of problems, but if you would, would you walk us through the thesis and then some of the information you have supporting it?
5:00
Dan Rasmussen: Yeah, sure. It's so interesting.
5:00
I mean, I think there are these investment fads that everybody believes for a period of time.
5:02
I mean, I think one of the 2010s is private equity is the best thing since sliced bread, but before private equity was the best thing since sliced bread, emerging markets were the best things since sliced bread.
5:14
You had this amazing World Bank report, I think in '93 or '94, saying, "The Soviet Union has fallen and we're going to see this massive economic resurgence across all of these emerging markets, and you should put your money there. Go invest."
5:29
Dan Rasmussen: The returns since then have just been disastrous despite everybody from the World Bank to some quant funds saying, EM is cheap, but it's the best thing."
5:35
I think my question was, "Well, why? What happened?
5:41
Why didn't emerging markets stocks work?"
5:41
I mean, it seems like they're illiquid, they're obscure. That should be helpful.
5:46
Certainly, at some point they've got to be cheap and that should be helpful.
5:50
So, where's the problem and why the classic?
5:56
Dan Rasmussen: The biggest headline error was Harvard management, who if you rewind back to 2007, it wasn't the Yale model, it was the endowment model because Harvard was crushing it, and they had a totally different approach that they
6:09
were doing all in-house management and whatever, and they hired Mohamed El-Erian, who's the World Bank guy that wrote that report in '94, and he said, "Well, gee, not only are emerging markets a great thing to invest in, it helps close the gap between rich countries and poor countries." Dan Rasmussen:
6:24
Dan Rasmussen: So, it's good for the world, but look at the returns from 2001 to 2007. They've been so good.
6:30
Harvard needs to really amp up.
6:30
It's bad in emerging markets.
6:30
What's related to emerging markets is commodities.
6:34
I think Harvard put 20% of its endowment in emerging markets and commodities in 2007 on El=Erian's guidance.
6:39
He published a big book that year about global convergence between poor countries and rich countries.
6:47
Dan Rasmussen: Then you don't hear about the Harvard model anymore for a reason, right?
6:50
They finally finished undoing.
6:50
They finally finished undoing all of those things, right? It took 10 years. It was so illiquid.
6:53
Now, actually, probably, it's a relatively decent time to be owning commodities in emerging markets. So, it's an irony.
7:02
By the time you got out of it, there's actually some interest.
7:05
Dan Rasmussen: So, I started looking into it, right?
7:05
I think my first question was to look at crisis in emerging markets.
7:07
I've gotten really interested in crisis, and one of the reasons I like crisis is because you know when you're in one, right?
7:18
I mean, I think the challenge of, say, if you're analyzing bubbles, well, whose definition of a bubble are we in a bubble now?
7:22
We're in a bubble before?
7:22
We're in a bubble in '98, right?
7:26
You never really know if you're in a bubble or not, but you absolutely know if you're in a crisis. It's just so clear.
7:29
Dan Rasmussen: So, as a quant, it appeals to me because nobody is really going to argue with my definition of a crisis, right?
7:34
They're not going to say, "Oh, no, no, no.
7:34
Argentina wasn't in a crisis then.
7:38
Yeah, the equity market was down 55%, but it was fine.
7:38
The market was fairly valued," right? You get it, right? We're in a crisis.
7:43
So, it provides this unique, I think, time for quants to say, "Hey, if there was a time when we know something special is going on in the markets or it's a unique economic environment that we all agree is unique.
7:52
We all know we're in a crisis.
7:57
Dan Rasmussen: So, I got started studying those in the US, and then I said, "Well, the natural next place to study them is in emerging markets," right, which have been so crisis-prone.
8:01
I think maybe 3x or 4x as many crisis as developed markets.
8:07
It seems a constant series.
8:07
There's a mix of, first of all, whenever there's a global economic crisis, there's emerging market crisis on steroids, right?
8:16
I mean, everybody just flees to safety.
8:16
No one wants to be in there, right?
8:21
You think of the logic of that, right?
8:21
Dan Rasmussen: I mean, maybe you're not as sophisticated as Harvard, but you've decided you've got a 90% US portfolio, but you bought that Indian bank that your friend recommended.
8:26
Well, when the market is down 30%, what's the first thing you liquidate?
8:31
That's stupid Indian bank. Why do I own that?
8:35
I don't know anything about India.
8:35
Maybe the government is going to steal it.
8:38
It's probably down 50% for a reason. Let's sell it.
8:40
Dan Rasmussen: So, you see this very exaggerate.
8:40
Then like Turkey today, you see what I call the idiosyncratic crisis, where the government makes some terrible policy decision or they're exposed to some industry and the country just goes through this collapse, and those as we call the idiosyncratic crisis and the global crisis but I segmented them out because I think they have different drivers, right?
8:58
Turkey is selling off not because there's a liquidity flight from emerging market, but, literally, the prime minister of Turkey did something really dumb.
9:05
Dan Rasmussen: So, if you segment those out, and you define a crisis, how do you define a crisis?
9:07
I just define crisis as a 50% drop in the equity market.
9:11
I got that from this Yale professor, William Goetzmann, who's a professor, an expert on bubbles and- Jamie Catherwood: He's the man. Dan Rasmussen: ... negative bubbles. Really fascinating guy. He says 50%.
9:23
I think he looked back 1600 and said 50% is his definition of a negative bubble.
9:29
A negative bubble and a crisis are similar or at least as an investor you're looking for crises that are negative bubbles.
9:33
So, there's too much fear, too much panic.
9:33
So, I use that definition, the negative 50% and then started to look at what comes next.
9:41
Dan Rasmussen: Broadly, what I found is that if you were to buy emerging market equities only during global crisis when emerging markets were at a negative bubble, so you just said, "Okay. The S&P 500 is down 20.
9:50
I'm going to buy every EM country index that's down 50% or more and hold it for two years and then sell out and go back to cash," right?
10:00
You'd have dramatically outperformed buying and holding emerging markets.
10:05
Dan Rasmussen: Then I think the next part is if you look at those idiosyncratic crises, actually, buying the country equity was not a good risk return because, oftentimes, like in Turkey, yeah, go buy Turkish equities right now, great idea.
10:16
Let's see what happens next.
10:16
What does the prime minister do next?
10:16
He's just hacked the central bank.
10:19
What's he going to do next? Right?
10:21
Dan Rasmussen: You see that play out in a lot of these markets where the equity market never recovers, but what's interesting is that the dollar-denominated government data actually does pay off relatively, reliably.
10:26
So, we developed this simple rule in global crisis by EM equities and in idiosyncratic crisis, buy the debt but don't buy the equities, and that pairing, I think, worked really well.
10:37
So, that was the outline of the paper was just saying, "Hey, buy and hold an emerging market has been a disaster.
10:42
In part it's been a disaster because of these constant crisis and liquidity flight."
10:48
Dan Rasmussen: We can talk more about liquidity and the role that plays in emerging markets. It's really interesting.
10:50
So, if you just segment it out and only bought emerging markets during this crisis opportunities or thoughtful about whether to do the equity or the debt, you just did so much better than listening to all the experts who were telling you to ...
11:02
I mean, you could almost think of EM as the original ESG, right?
11:07
Before ESG was buying green energy stuff.
11:07
It was helping out third world countries.
11:12
So, emerging markets had this big overlap with ESG, which also was part of the big story.
11:17
So, I'll pause there, but that's where we got to with the research.
11:20
Jim O'Shaughnessy: So, we think in a similar fashion when we published the generational buying opportunity in March of 2009.
11:23
It wasn't because I had any great insight.
11:31
It was because we looked at the data. I said, "Okay.
11:31
So, these 10 years ending February '09 are the second worst 10-year real rate of return for US markets back to 1870.
11:43
Let's look at what happened in the other 49," right?
11:48
Jim O'Shaughnessy: So, obviously, after three years, the other 49 showed incredibly positive results, and then coupled with that was also the fact that the long bond had outperformed the equity market over the previous 30, 20, 15, 10, five-year period.
12:07
The last time that happened was in 1944, I think.
12:07
I haven't reread it recently.
12:12
Jim O'Shaughnessy: So, I love the counterintuitive aspect of this, and I agree about emerging markets.
12:15
The buy and hold there, I would not advocate, right?
12:22
So, I really do like this theory a lot.
12:22
I also remember, though, when I was younger and diving into this stuff, I subscribed to the journal of portfolio management and all the geeky stuff.
12:34
That's when you had to actually get the thing in the mail.
12:34
I was thinking as I was looking at the notes I was writing up, there was a series of articles.
12:39
I can't remember who wrote them, but they were looking at international markets.
12:48
Jim O'Shaughnessy: The author, and I think that this was actually in 1981, so I would have been 21.
12:51
The author took apart international markets first by value.
12:58
So, PE, price to book, et cetera, and then he had a follow on I think in 1980, whatever, '89 maybe, where he took them apart by dividend yield.
13:04
His strategy was buy the five countries from the MSCI international index that have the lowest valuations or the highest dividend yield.
13:19
Jim O'Shaughnessy: The challenge was it didn't go on to work.
13:20
So, how do you account for that within the confines of what I find a very intuitively seductive strategy? Dan Rasmussen: Yeah.
13:32
Well, Jim, I think it's something that as a value manager over the past few years is something that I wrestled with a lot.
13:34
Why does sometimes cheapness not work?
13:41
How can you be a value investor through 2018, 2019, and 2020 and not have asked that question?
13:46
So, maybe that's in the ether, right?
13:46
Clearly, value doesn't work sometimes.
13:51
Dan Rasmussen: I was talking to some really smart investor the other day and he said, "Well, we came to this realization," this firm, they're a fundamental management, "We came to this realization you shouldn't just buy stocks just because they're cheap.
13:59
That's not enough for a reason."
14:03
Dan Rasmussen: I said to myself, "Oh, shoot!
14:03
What have I done with my career? What an idiot I am."
14:05
Dan Rasmussen: By golly, that's been the right answer for 2018, 2019, and 2020.
14:09
So, whatever, but I think I look back and say emerging markets are one case study for that or a broader area where I think they had to be really cheap for it to work.
14:19
So, I think what I said is if you think maybe cheapness doesn't work if it's marginal, right?
14:24
Well, maybe it should work more at the extremes, when things are really cheap.
14:32
Dan Rasmussen: Actually, if you ran our crisis analysis piece and instead of a 50% drawdown in the market, you started tilting it back like, "What if the market is down 20% or 30%?"
14:39
Actually, alpha just comes way, way down, right?
14:39
So, there's something about extremes, I think, that matter in markets, that I think matter in quant, right?
14:48
Dan Rasmussen: There's another study I saw a few years back that said that returns are eight times as predictable in crisis as they in non-crisis.
14:54
I think there's some truth to that, right?
14:58
Think about when should value work.
14:58
Well, it should work at the extremes because that's when the behavioral bias is clear.
15:03
That's when the liquidity is real.
15:03
I'm actually maybe drawn to some of these liquidity-driven arguments or behavioral arguments, right?
15:11
Dan Rasmussen: There's got to be ...
15:11
If you're thinking that's not even cheap, you want it to be cheap and to be reading newspaper headlines about why it's terrible, right?
15:17
You don't want it to be cheap and no one is talking about it.
15:17
The real extreme stuff is what I'm really quite interested in because if you look at this last decade, which has been quite terrible for value, when has valued worked?
15:26
Well, it actually really worked a few years over the last few years, right?
15:32
It worked really well in 2016.
15:35
It worked really well from March 2020 on.
15:39
Dan Rasmussen: So, what were the difference between those periods and the periods when it didn't work.
15:41
One of the reasons is we're coming out of economic crisis, right?
15:44
There's a quasi recession in 2015.
15:44
There was an obvious one in Q1 of 2020.
15:51
You see this consistent pattern, where during these crises and, actually, there's the factors from scratch paper, which you guys put out, which I made as a sign greeting to my interns.
16:01
It's so darn good, but if you think about that framework from the crisis perspective, the growth stocks do better coming in to the crisis. They really do.
16:11
Dan Rasmussen: You paid a higher multiple and by March of 2020, you are happy that paid that higher multiple for Amazon.
16:14
It actually did better, and that actually really mattered in that early stage of the crisis, to own a better company with better margins and whatever.
16:23
You are actually doing better.
16:23
So, paying higher prices was fine coming in to that, but then you had this weird reversal, where by March of 2020 you're thinking, "Well, wait a second.
16:33
I got the benefit of owning an acyclical secular growth story and everything that's cyclical was cheap before because people didn't want to own cyclical, and then the cycle hit, and their revenues are down 20% this quarter, and their profit is down 60%, and then you say, "Well, let's fast forward a year," and that factors from scratch, right?
16:56
You move forward from time, rerun your model, "What's my growth expectation now and, therefore, what's going to happen to my multiple in the future?"
17:03
Dan Rasmussen: You have this conclusion and say, "Oh, my gosh!"
17:04
Actually, value stocks are probably going to grow faster than growth stocks and they trade at a bigger spread valuation wise than they did pre-crisis.
17:12
So, you're buying them even cheaper even though their expectations are much better and you're probably expecting significant multiple expansion in a year or for now once people recognize that.
17:21
Dan Rasmussen: I think that those dynamics play out in these crises and they play out in emerging markets as well, right?
17:28
Those are the ultimate cyclical markets because they're so commodity-exposed.
17:32
Jim O'Shaughnessy: So, mechanically, how do you go in and buy?
17:36
Do you buy in the market or do you buy ADR or you said debts or you can't be doing ADRs, I don't think? What is your process? Dan Rasmussen: Yeah.
17:44
So, we're still working on it.
17:44
This was more of a thought piece than something we implemented.
17:47
It's something we like to implement in the future.
17:47
We just missed one of the great buying opportunities in emerging market value, so we'll wait for the next one, but I think the idea is basically what you want to do. You can do it two ways.
17:55
You can buy the country level indices, essentially, and just sort by which country got hit the worst and buy the country level indices. That's a fine approach. We tested it that way.
18:06
The results of doing it that way and the other way I'm going to describe or equivalent.
18:14
Dan Rasmussen: The other way to do it is to do a simple factor model and within the liquid emerging market stocks, which you got you there is a lot of illiquid stuff, but within the liquid stock, just do a classic Fama-French value sort, sort by price to book or sort by price to book and profitability and buy the top rank things, and that works really well, too.
18:30
The results are virtually identical because you're ending up with the same stocks, right?
18:34
You're going to end up waiting those countries, anyway, but those are the two ways to do it.
18:38
Dan Rasmussen: With debt, you want to buy the USD or Euro-denominated, the foreign currency debt, not the local currency debt of the government.
18:44
Although I think there's some interesting potential for the quasi governmental infrastructure debt or other things like that.
18:50
I've talked to some people since the report came out that suggests that there's an interesting opportunity across the credit markets in those countries during those periods.
18:58
So, that's something we're looking into as well.
18:58
So, I'd say this was our high level theory point to the opportunity and then now we're digging in at the underlying name by name implementation questions.
19:09
Jim O'Shaughnessy: So, you mentioned before we started recording that, and I was teasing you about needing security.
19:12
Are you finding that institutional or endowment type investors are open to this kind of thing or is the narrative just too difficult for them to overcome?
19:25
One of the things that we found is the career risk part is much more intense than a lot of people think, both for the manager and for the person giving that manager money, right?
19:38
So, it's a double whammy.
19:38
Do you think that the efficacy of the theoretical approach that you, which is great, by the way, and I urge everyone to read it in full. It's 80+ pages. How long?
19:48
I read most of it, but how many pages?
19:53
Dan Rasmussen: Something like that.
19:54
Jim O'Shaughnessy: Yeah, yeah.
19:54
So, it's really well done, but what's your reaction that you're getting from traditional endowment investors, traditional institutional investors? Dan Rasmussen: Yeah.
20:05
It's quite interesting, actually.
20:05
I mean, I think that if you think about quant equity or what works in quant equity, I mean, you have to go into illiquid corners of the market to make it work really well or at least its best returns are in micro and small cap, right?
20:20
There's still good returns in the rest of the market cap spectrum, but the biggest hit of it is in micro and small.
20:24
Dan Rasmussen: The cost of that is illiquidity, capacity constraints, and volatility.
20:26
I actually find that's a tougher pill for a lot of institutions to swallow because, eventually, we'd like to size up to 200 million. Can you do that? Not micro cap. I'm at 25. That would be big.
20:43
So, it's just not enough to move the needle.
20:43
The alpha opportunity there for a lot of the big institutions isn't enough to move the needle.
20:47
So, they're interested in it as a smaller allocation if so, but I think a lot of the institutions they want to deploy larger amounts.
20:56
They want to find strategies they can really scale in to.
20:59
Dan Rasmussen: I think at least the portions of quant equity we focus on in terms of the micro cap deep value international, that sort of stuff, it's a bit harder.
21:08
It's actually better for retail and family offices who have a smaller amount of money makes a bigger difference, too.
21:12
I think that the crisis stuff, actually, is quite more interesting to institutions in part because it's quite scalable or at least more scalable because when liquidity flight, when people are panicking, you can dump a lot of money into these crises before you start to move markets because there's just so much selling.
21:28
Dan Rasmussen: So, and then I think the second thing is that if you think about all the institutions are really moved to this endowment model where they have big private equity and venture allocations, and one of the really obvious big things you see in the data is that the capital calls of private equity and venture capital, especially private equity, are really cyclical.
21:48
Dan Rasmussen: So, when valuations are high, they call more capital.
21:51
When valuations are low, they call less capital.
21:51
This isn't a conspiracy theory that the private equity people are total idiots and don't get how markets work. They get it.
21:59
They know that the opportunities are better when things are cheap and not as good when things are expensive, but there are two other dynamics going, which are that private equity, you have to have a company to buy and you have to have debt to buy it.
22:11
Dan Rasmussen: So, if you have some really great family business you've owned for 100 years and you've decided that this is the year to sell it and then the market crashes, you'd think, "Probably we should wait a year.
22:17
I don't want to sell my family business this year when the prices are so low," right?
22:21
So, the sales volume dries up, no one is selling.
22:26
The next thing is the debt market is closed.
22:26
So, all the banks and the lenders say, "Heavens know we're not giving you more money to go out and business.
22:30
Do you know uncertain the market is right now?
22:34
We don't even know if we're going to collect our interest on the existing companies you guys own. Holy smokes! Absolutely not."
22:42
Dan Rasmussen: So, private equities are just dead in the water.
22:44
There's nothing they can do.
22:44
Then all their deal guys, they're redeploying the portfolio companies anyway because the shit is hitting the fan.
22:47
They don't know if they're going to be an interest payment.
22:49
So, you have this big problem, actually, for a lot of the institutions where their capital flows are super pro-cyclical.
22:54
where their capital flows are super pro-cyclical. So, actually, I think that this crisis strategy or the idea of maybe you should put some money in public markets during times of equity market
23:05
dislocation from your private portfolio because you're not going to be deploying any work on your money in your privates, anyway, and by the time the market recovers, that's when private equity is going to be capital again, anyway, so you'll be getting money back from publics and you can go deploy it in privates again. Dan Rasmussen: I think that's actually really resonated with
23:19
Dan Rasmussen: I think that's actually really resonated with a lot of people, which has been nice.
23:21
I think the other dimension of it is that there's a certain type of investor that when the markets go down panics, and then there's another type of investors that says, "Okay.
23:29
Find me the craziest, most aggressive thing I can possibly do to take advantage of the chaos," and I think that I like to be that guy.
23:34
When the markets are absolutely panicking and you want to go long, call me and we'll find a way to deploy some capital.
23:43
I like that because I think it's much more likely to work than a lot of other alpha strategies.
23:46
Jamie Catherwood: What do you think of private equity during all this, the last year with COVID?
23:48
Because I remember during March, April, Minsky moment was the hot buzzword because people were saying that it's going to be the Minsky moment for private equity, where all these companies that they had levered up with debt are going to come crashing down and that's going to hurt private equity, but it seems that there haven't been any high profile issues so far.
24:13
So, I'm sure you have thoughts on this.
24:13
I don't know why I'm asking if you have thoughts.
24:18
Dan Rasmussen: There are a few things going on.
24:18
I'd expand the discussion from private equity to private credit, too, which is now all of these firms have their own private credit arm and they're lending to each other's deals, and blah, blah, blah, right?
24:28
So, there's a few interesting dynamics associated with that.
24:30
Dan Rasmussen: One is you did see, and those tend to be publicly traded, a lot of those private credit firms are publicly traded as BDCs and those sort of things, right?
24:36
So, you can actually get some insight on what's going on.
24:36
They had a massive drawdown in March, right?
24:39
So, they were down, the BDCs were down 40%-50%, right?
24:44
So, people were treating those BDCs as triple C credit, which is what they are.
24:48
Dan Rasmussen: So, that was an interesting moment, but what happened is that everything snapped back so fast and, actually, private credit has this extend number 10 ability with ...
24:55
They're creatures of the private equity firms, they don't want to market, they don't want to admit a bankruptcy, they don't want to call a covenant violation.
25:03
They just do an amendment and they just keep rolling over.
25:07
Dan Rasmussen: So, what was interesting about what happened this year is you saw that private credit play out in a really interesting way, right?
25:10
All the private equity deals that were financed by private credit, they were fine, right?
25:18
The private credit guys say, "Hey, whatever. Take a quarter down. Send us interest.
25:18
Let's talk again next quarter," right?
25:21
So, that system worked really well to extend number 10.
25:25
Just, "We'll give you a little bit more debt, little worse terms, and we'll deal with it later."
25:29
That was fine for both parties.
25:29
Dan Rasmussen: I think the other dynamic, which I probably underappreciated, which is quite interesting, is the extent to which private equity over the past few years is really tech private equity.
25:40
I think it might be upwards of 50% of the deals are software or software-related.
25:46
I mean, it is a big, big macro beta on behalf of these shops.
25:46
If they're looking at industrials, they would rather look for an industrial vertical software company than buy an actual industrial manufacturing plant.
25:56
I think that also played out to their benefit in 2020.
26:00
Dan Rasmussen: So, I think that what ended up happening is that private equity, and this was my thesis, originally, was that private equity between call it 2006 and really post-crisis, private equity made a switch from being leveraged small cap value, which, again, I think is a great strategy, to being leveraged small cap growth, which I think is a terrible strategy.
26:20
Dan Rasmussen: So, my thesis to certain private equity is you probably don't want to put 40% of your endowment into leveraged small cap growth.
26:28
If you've ever looked at a long-term back test of small cap growth, it looks bad, and if you add leverage to that, especially a leverage that is at triple C bankruptcy level, there are two things that don't get compensated, bankruptcy risk and overvaluation risk.
26:41
So, why take both in one illiquid package?
26:41
Yet, growth was what worked last decade.
26:48
So, they didn't get punished.
26:48
In fact, they got rewarded for doing that.
26:52
Dan Rasmussen: So, I think what's going to happen and the key acid test is what happens next.
26:53
You started to see cracks.
26:58
Greensill was one of the first big private credit lenders to go KO and it had all the hallmarks. It was illiquid, right?
27:05
You didn't actually know what the loans were.
27:09
They're marked in strange ways, marketed to different people, right?
27:09
I mean, it was just you played that out and, yeah, that was trade finance, it wasn't private equity debt financing, and it's a little bit like saying just because LTCM blew up that all hedge funds are bad, and I don't want to be that guy, right?
27:20
Obviously, Greensill is unique, but don't you look at that and say they're by the grace of God go 15 in other private credit first?
27:28
Dan Rasmussen: So, I think that it's a little bit of a scary ecosystem, but the tech exposure and the ability of private credit to extend number 10 has largely gotten them through this crisis at least.
27:38
Jim O'Shaughnessy: So, that leads me into the other aspect of your work that I actually love because that's the Molotov cocktail throwing part of it, and I mean that with the greatest respect, which is this idea of the persistence of bad models, which you have a paper on, and which I find very interesting.
27:52
John Meriwether, who you just brought up long-term capital management, absolutely blew it up.
27:58
I knew those guys, and was saying, "Leverage guys, illiquid, not a good combination," but then he went on and raised another fund.
28:11
Jim O'Shaughnessy: Then you have things like, as you note in your paper, the two guys, Fama and French, who came up with the capital asset pricing model, after decades come out and say, "Oopsie! It doesn't work, guys."
28:25
Then you've got Markowitz, who is the founder of the efficient market line, but does he invest his money that way? No. No.
28:32
He does a naïve distribution heuristic where he does a third, a third, and a third, and when asked why he did that, he cites he didn't want to be Heber Grant, right? Dan Rasmussen: Yup.
28:46
Jim O'Shaughnessy: So, I mean, have you ever really take a 30,000-foot look and just is there some kind of pattern that involves human behavior, that involves the halo effect, that involves theory taking precedence over actuality, like capital asset pricing model, all of this stuff?
29:01
Interesting theoretical framework doesn't work.
29:09
I've always cited it as one of the primary differences between practitioners and academics, right?
29:18
So, when I did What Works on Wall Street, I did it from the point of view of a practitioner. I was interested in ... Okay.
29:24
I wanted the theory to make sense as well, obviously, and I tried to stay as close to the scientific method as I could.
29:36
In other words, you should be able to replicate my research if you have the same tools and the same dataset and it should have something that makes a little bit of sense, but what do you think about this?
29:47
Why do we see time and again the persistence of, "Okay.
29:53
Well, yeah, that's what the data says about private equity and venture capital, but goddamn it, that's where I'm going to put my money"? Dan Rasmussen: Yeah.
30:00
I mean, well, I think let's start the models first, right?
30:02
I mean, I think it's about making sense of the world and how we make sense of the world.
30:07
I think a lot of people we want to have a plan and we want to have a prediction of what's going to happen.
30:11
Often, the prediction of what we want to happen is motivated by our desires and biases and goals, right?
30:16
If I am a Russia hawk, I am convinced that every year Russia is going to do some aggressive foreign policy thing and the thing that we should do is attack.
30:29
So, I'm going to put out a string of papers every year about imminent threat from Russia and that's just my gig, right?
30:36
Dan Rasmussen: I think people have those heuristics or biases or attitudes, and as they think about the future, they think that it makes a lot more sense in part because we're sense-making about the past, right?
30:43
We're trying to say everybody is trying to have a linear story of how they got to where they are and everything makes sense in retrospect and there's a cause and effect relationship.
30:52
Even that, when you actually start to talk to people and you realize that other people might have totally different cause and effects for the same event that you think happened, obviously, for X reason and someone thinks, "Well, you come up with that reason."
31:05
Y reason is the reason it happened, but we're so convinced our explanation of the past is right and that our cause and effect relationships are right, we then extrapolate into the future and think it's a lot more predictable.
31:15
Dan Rasmussen: So, if you look at a lot of the problems of these early models, the dividend discount model, the initial problematic one, it says forecast all future cash flows and discount based on a discount rate, which is based on the riskiness of the security. You're like, "Ah!
31:29
So, if we knew what all the future cash flow were and we knew how risky the company was, we'd perfectly be able to evaluate," right? What's the problem?
31:34
Well, we don't know what the future cash flows are and we don't know how risky it is.
31:37
So, given two imaginary inputs, yes, perfect foresight, yes.
31:43
Dan Rasmussen: Then you move through and like Markowitz's diversification, it was just his first step.
31:46
I mean, it's almost comical, right?
31:46
You read the paper and be like, "Well, if the dividend discount model is right, which it is, wouldn't you just put all of your money in the highest expected return stock, just put it all in that because that's the highest return, but people don't seem to do that, right?
31:58
So, why don't they do that?
32:02
Dan Rasmussen: Then he's like, "Ah, because of volatility," right?
32:03
So, yes, the high expected return, it's going to reach 50% a year.
32:09
So, you would put all your money in it except that it goes up and down a lot, so you decide not to.
32:12
Therefore, we have to introduce beta, right?
32:12
Therefore, now we want to have a diversified portfolio, so we have efficient frontier, right?
32:18
You just keep playing these things through, and they're all built on this very unrealistic assumption that the future is predictable either in future cash flows or the discount rate.
32:30
Dan Rasmussen: I think then you say, "Well, what's the alternative to that?"
32:31
I'm a big fan as I think you guys both are of Phil Tetlock, right?
32:36
I mean, you can start with nihilism and say, "Nothing is predictable. It's all an illusion."
32:40
We know absolutely nothing and we're standing behind the veil of ignorance, which, by the way, is a pretty good default assumption, right?
32:49
That leads you to things like buying value stocks, right?
32:49
I mean, there's some logical conclusions.
32:53
If you're behind the veil of ignorance, why would I pay for a company that everybody else in the market thinks it's going to grow 50% a year and so it trades at 100 times BE.
33:01
I probably shouldn't buy that because that seems really overconfident if we know nothing.
33:05
I should just sort by price because that's going to get me the stuff that people are most pessimistic about," right?
33:10
Dan Rasmussen: So, starting with nihilism and saying nothing is predictable is step one.
33:12
Then step two is saying, "Well, what things are more predictable or less predictable?"
33:16
That's base rates, right? That's your book.
33:20
That's why I love your book.
33:20
What Works On Wall Street is a base rate gospel, right? We've got it right here. I look it up.
33:24
Every time I think of, "Would ROE work in this back test?"
33:28
I just look at the page and scare down the base rate tables, right?
33:33
It's so useful because it's telling you, "Hey, maybe we don't know what's going to happen in the future and we're mostly behind the veil of ignorance, but in similar situations, what have been my probability distributions of outcomes?"
33:44
Dan Rasmussen: I think when you start moving in that direction, then you say, "Well, when do the expected return and the probability is the highest?"
33:51
You're going to quickly get to crisis because crisis, expected returns are really high and your hit rate probability is really high.
33:55
So, it's like the most predictable thing among a set of world of unpredictable things.
33:59
It's like when your grandmother calls you and says, "Should I sell all my stock?"
34:03
That's the time to buy, right?
34:03
There are these simple mental heuristics that actually are pretty predictive and powerful.
34:10
Dan Rasmussen: Then when it comes to an alternative way of sense-making, which is someone says, "Gee!
34:13
I've noticed that my predictions don't come true that often," or "I'm not that good at predicting the future," and there's a step of awareness there.
34:23
Then they say, "Well, how should I predict the future then because I need to predict it."
34:26
I'm not going to say we don't predict it.
34:26
So, "I'll find an expert.
34:30
Let me find an expert who can predict the future."
34:33
Dan Rasmussen: This creates this huge market for expertise and people that claim expertise.
34:35
I think that that also drives these behavioral things, the herding.
34:41
I think people say, "Well, David Swensen put a huge percentage of his portfolio into private equity and he thinks it's a superior form of capitalism, and I'm not as smart as David Swensen. He's thought it through.
34:50
So, why don't I just do what he does? He's the expert. Let me just copy him."
34:54
Dan Rasmussen: There's this mimetic behavior that is the result of, I think, the people falsely going from either, A, make your own predictions and just be like a maverick like, "I'm going to make my own plans and my own predictions," B, start with extreme nihilism, assume unpredictability and rely on base rates, but those are outliers.
35:11
The vast majority of people in the middle of the bell curve say, "I'm not that good. I'm not that smart.
35:16
Let me find somebody who is and just follow them and copy them."
35:20
Dan Rasmussen: I think that's what drives a lot of these behaviors, and that's what drives, honestly, I think a lot of the things we see like momentum and trend in markets is just the human mimetic behavior, where somebody puts out a forecast and other people start to believe it, and then they all start betting on it.
35:35
Jim O'Shaughnessy: Very [crosstalk] Jim O'Shaughnessy: So, one of the things that I always struggle with as I'm approaching this, and I think you obviously know the way I look at the market very closely because you- Dan Rasmussen: You published some things that I've read. Jim O'Shaughnessy: Yeah.
35:47
I did a couple of things.
35:47
So, one of the things that I always run in to that is embedded, I think, in people's thinking, which I think is incorrect, which is one of the primary, using another Latin phrase here, things that, academics especially, stick in to everything, and that's ceteris paribus, other things being equal.
36:09
I have a degree in economics, right?
36:09
I remember I was an even more ... See, you're so polite.
36:15
I wasn't nearly as polite.
36:15
I was the one always with my hand up.
36:20
Jim O'Shaughnessy: So, ceteris paribus, but other things are never unchanged, right?
36:22
When action A happens, all sorts of these other actions happen often in response to it.
36:28
It was actually one of the hallmarks and cornerstones of the economic theory that was dominant as I was in university, which is rational expectations and really took the wind out of the sails of the Keynesian econometric people because it was like, "Wait.
36:46
So, you're saying that if we double tax rates, people's behavior is going to change?"
36:52
Jim O'Shaughnessy: It's funny because I read all the academic papers because I think that there's something going to there, whether for me to disagree or agree or think about why they're thinking the way they do."
37:07
The desirability bias is another one, right?
37:07
So, people find what they're looking for.
37:13
So, it's very difficult to go in to a situation as you say as a nihilist.
37:21
It's easy for me because that's the way I think.
37:21
I just think, "Well, I don't know."
37:26
So, I'm just going to see what directionally has worked, and then I think, "Well, that makes a lot of sense."
37:32
Jim O'Shaughnessy: Back to momentum, we've tested momentum using the CRSP database, which is from the University of Chicago, Center for Research and Security Pricing.
37:39
Momentum works in every decade from the '30s on, really reliably with one exception, following a crisis.
37:46
When a crisis occurs, you would invert and you would buy the worst decile by momentum, not the best.
37:57
So, I find it's still challenging for people, and I think that ...
38:08
I studied mimetic behavior because I have to, right?
38:19
Jim O'Shaughnessy: Another thing that I think is, for the most part, which is seen as heresy, which is narrative, generally, follows price. It does not lead price.
38:29
So, what are your thoughts about what's going on right now, which I find, by the way, highly entertaining with the meme stocks and with David Portnoy, who is a marketing genius and he uses weapons-grade persuasion.
38:42
He's really good at what he does, but what he does is not in the field of portfolio management. It's in marketing. How does this end? What do you think? Dan Rasmussen: Yeah. It's so interesting.
39:05
I mean, coming back to the rational expectations argument, I'm a fan of an alternative theory called rational beliefs, which Mordecai Kurz at Stanford came up with, where he basically said everybody has their own belief system that's rational.
39:22
His definition of rational is you can line up all the available facts about the past to justify that belief system, right?
39:32
Dan Rasmussen: It just turns out that all the facts about the past don't only justify one belief system, which is a big shock to the academics, but it turns out that people that look at the same data often come to different conclusions for very legitimate reasons.
39:43
We do not know who is right because we can't see the future.
39:49
So, until the future happens, we can't adjudicate who is right and who is wrong.
39:49
In Mordecai Kurz's series, actually, that explains the volatility of markets.
39:55
It's divergent beliefs among market participants than then have to adapt to a large percentage of those beliefs being wrong when new data comes out, right?
40:06
Dan Rasmussen: That just makes intuitive sense to me, right?
40:09
How many times have I been wrong? A heck of a lot, right?
40:12
So, probably other people are wrong, too, and so a lot of these beliefs must be rational, and that probably does drive volatility because, yeah, if I'm wrong and I see I'm wrong, I do sell something, and that does drive the price.
40:21
So, I think that makes total sense.
40:25
Dan Rasmussen: Then I think that you then try to think about how do these beliefs get constructed, right?
40:31
I think one of the things that has happened, if someone says something and then it happens and it comes true, then people believe it or they're more likely to believe it.
40:36
If it happens two or three years in a row, then they really believe it.
40:41
This is, by the way, a funny analog to the fund management world because I think this is actually how a lot of very smart institutional allocators make their decisions about managers.
40:50
They bring them in year one and they say, "Hey, what's your thesis about the next 12 months?"
40:58
Dan Rasmussen: They say, "Well, here's my thesis."
40:59
Dan Rasmussen: "What are you betting on?"
41:00
Dan Rasmussen: They bring him in a year later and said, "Was your thesis right? Did you execute on it?
41:02
Then what's your thesis for the next year?"
41:06
Then they wait the next year.
41:06
If they've gotten it right two years in a right, great, but, actually, it's funny.
41:08
I mean, as a side note, a lot of managers fail the first time.
41:08
There are thesis [inaudible] a year later.
41:12
So, it's a funny world, but I think what you've seen the last decade is a lot of these growth stocks.
41:24
They did better than expectations. The price went up. More people bought them.
41:29
It was this big building on itself of story.
41:34
Dan Rasmussen: I think what's fascinating about Portnoy, he's just manufacturing at warp speed.
41:36
It's like, "I go out and tell everybody to buy John Deere's stock.
41:40
I saw a deer outside my bedroom this morning, and then Deere stock is up 5%, and I tell them I was right.
41:44
Then people pile in because they think I'm a genius, and then it goes up another 5%, and it's very short order, and then I'm going to switch to another meme."
41:53
It's the evolution from newspapers to Twitter.
41:53
It's like the evolution is from JPM's market strategists putting out something once a quarter to Dave Portnoy switching his momentum stock based on animal he sees outside his bedroom, right? It's fascinating.
42:09
Dan Rasmussen: Now, I think there's some benefits, right?
42:11
I mean, I think that it's nice to see.
42:11
I think there's two interesting things about what I think is generally a horrible story, which is that there are probably going to be, A, a lot of new micro and small cap stocks as a result of the SPAC craziness, which is great.
42:24
Now, a lot of disasters for the people that invested in them, but give it two years and those value investors are going to have a field day and we'll finally be able to ...
42:29
I feel like public equity investors for years have dealt with, "I've decided to move my allocation to private equity because the number of listed stocks is going down since 1998."
42:41
Dan Rasmussen: You're like, "Well, if you look at what stocks comprised all the stocks in 1998 because it was a lot of SPAC-like things, but now we'll finally all of a sudden, well, now, shouldn't you be shifting your money back into public equities because the number of public listing is going up again?"
42:51
So, I'm excited about that.
42:54
Dan Rasmussen: Then I think the other thing is that it's interesting to see the cycles of behavior.
42:58
I mean, you think of how dominant the index funds, buy index funds.
42:58
I'm only buying index funds.
43:02
People stopped owning individual stocks.
43:02
I mean, they really did, right?
43:02
People were shifting out of their portfolios.
43:07
Now, maybe there'll be a research and some people actually buying individual stocks again, which as a stock picker, I like and maybe the data for index funds is much better than the results from people buying individual stocks, but I still love the idea of people buying individual stocks.
43:18
So, I think that's a good thing. Jim O'Shaughnessy: Yeah. I agree.
43:21
I think that nails the thesis that I have about the whole thing, which is the thing that bothers me and I get concerned about is I love younger people, especially, becoming interested in investing.
43:32
I think that's a good thing.
43:39
The challenge is that the ratio for everyone who's going to be quasi rational about it and actually do some study and homework, there is 100 that aren't going to do that.
43:54
They're going to get burned, and then they're going to swear off stocks and say, "That game is rigged."
43:59
Jim O'Shaughnessy: So, I agree with you in terms of the creation of new supply in SPACs.
44:01
I agree with you on the liquidity problem.
44:07
My worry is that given our base code, I've taken to calling it human operating system because when you talk about beliefs, people's shields go up really quickly.
44:21
I just wonder what will be the reaction. What do you think?
44:21
When the inevitable crash and burn happens, how many people just will walk away from the market forever?
44:27
That's what I worry about, young people especially.
44:33
Dan Rasmussen: Right, right, especially given what a lot of them are choosing to invest in, which are the worst of the most overvalued or look at what's going on with electric vehicles stocks, right?
44:45
I mean, I know I said you can really tell when you're in a crisis.
44:45
It's obviously, and you can never tell when you're in a bubble, but how can you not look at electric vehicle stocks and that sort of stuff and say, "That's obviously a bubble," and yet, that's what most of the investor enthusiasm is driven towards.
44:59
Dan Rasmussen: I think there are probably two components.
45:00
There's two components of it.
45:00
I think one is coming back to the emerging market, if you have enough individual investors putting money into something, that money actually funds stuff.
45:08
I mean, the company could actually use it to do something just like the emerging markets could use to build bridges or stuff.
45:12
Now, the investors never made a dollar off that stuff, but the people that were running the companies actually might, and maybe another way to think about the EV bubble is to think, "Well, actually, maybe there's so many people committed doing electric vehicle future that they want to just give these companies money to just go build cool stuff," and they don't really care if they make money or not, right?
45:31
Dan Rasmussen: I don't know what percent of people that own EV stocks believe or think that way but I think it's possible.
45:33
It's probably the same thing driving enthusiasm for rockets, right?
45:37
Well, if Jeff Bezos can dump a few billion of his fortune into building rockets, why can't I take some of my 401(k) and build rockets, too? I get that, right?
45:47
Maybe something good can come off it, but I think one of my new favorite quotes is from a mentor of mine. He says.
45:52
"It's easier to know what will happen than when will it happen."
45:56
I think it's obvious that there's going to be a train wreck in these growth shares and when it will happen, I don't know, but I agree the consequences of it for a lot of people are going to be really bad and it's sad.
46:05
I think rather than following Wall Street bets, people should go buy What Works on Wall Street and they'd be a lot better off.
46:11
Jim O'Shaughnessy: The unpaid plug.
46:13
I'm going to lead with that.
46:13
So, I agree and I've been thinking about a thesis of bubbles that I heard from several people, which is this idea that you just expressed.
46:25
In other words, bubbles leave a lot of very cool stuff after they popped, right?
46:25
So, you think about the dot com bubble.
46:33
Well, there was a lot of infrastructure that got built.
46:39
People lost fortunes on it, but now there's the infrastructure.
46:43
Jim O'Shaughnessy: In a similar fashion, I like your metaphor for emerging markets.
46:47
Well, people might lose everything they put in, but that bridge gets built or that road gets paved.
46:52
I'm a huge traveler and I think travel is such a fantastic education, but I was in Baton, which is a wonderful place, by the way, but literally car ride is fear of death because the roads aren't paved and there's rocks stumbling down, and I think it would be great if Baton got a huge infusion of capital and put that towards infrastructure because I'm bullish on the prospects of all of these various emerging and/or frontier type companies.
47:27
Jim O'Shaughnessy: I don't mean to sound indifferent to people's losses.
47:30
I'm certainly not, but my theory has always been, well, as part of our grand experiment here, we're going to have lots of failures.
47:44
Failures are opportunities to correct errors.
47:44
Without failures, no progress, basically, because then you're living in spaces and the precautionary principle abides, and if the precautionary principle abides, you're going to try to get everyone to stop doing everything.
48:04
Jim O'Shaughnessy: I think of the old forecast from the '70s.
48:07
Julian Simon has a great book called The Ultimate Resource.
48:07
Basically, the theme of his book is the ultimate resource is human ingenuity.
48:13
There was a guy, Ehrlich, I think he's still around, but he was a superstar in the '70s because he outdid Malthus by 10 times.
48:18
He gave lectures about how we're running out of everything, people are going to be starving by the billions in 1980 at the very latest, and it's too late to do anything about it.
48:38
Jim O'Shaughnessy: Well, obviously, in fact, there's a famous bet between Simon and him in which Simon let him pick what elements were the rarest, right?
48:50
He picked 10 and Simon said, "They will all be cheaper 10 years from now as opposed to gone."
48:56
Jim O'Shaughnessy: Ehrlich was like, "Shot the easiest money ever made."
49:00
Jim O'Shaughnessy: Well, you know what happened, and Ehrlich skunked away, but I think that this is something I'm working on right now because I definitely think it's part of our human source code, this idea of both being pessimistic about
49:17
the future and the present and I think that's because of the evolution of our genes didn't keep up with the evolution of our memes, I mean memes in the memeplex transfer of knowledge, ever get cumulative societal evolution, has gone parabolic, and we humans, we're still optimized for the stone age in a way. Jim O'Shaughnessy:
49:40
Jim O'Shaughnessy: Do you think about those kinds of things or are you just more of a knots and bolts like, "Well, I'm going to be just I know nothing," which, by the way, I think is the best place to start? Any thought process?
49:52
I always say, "I know nothing. Let me learn. Let me learn." What do you think? Dan Rasmussen: Yeah. Well, it's funny. I'm a big fan of E. O.
49:59
Wilson and Sociobiology and some of these evolutionary concepts. E. O.
50:04
Wilson, I think, said, "What are the two most successful species on planet Earth?"
50:09
The two most successful by body mass just the aggregate mass of living creatures are ants and humans.
50:15
Wilson says, "Well, what do ants and humans have in common?"
50:20
They have in common that they're both extremely social animals.
50:20
They also have in common that they're the only two species that wage war, which is very interesting, right?
50:25
There are lots of these fascinating like ants will wage very sophisticated wars with other ant tribes just like humans do.
50:33
It's fascinating, but they're wildly successful as well.
50:36
Dan Rasmussen: You think about evolution on a broader scope.
50:40
Evolution is a deeply cruel process, right?
50:40
I mean, if you think about the actual thesis behind it, it's like, well, 99 of these mutations will result in death and one of the mutation will result and it's not good the species.
50:49
You can analogize that in some sense to some of these futurology aspects of the stock market, right?
50:53
99 out of 100 EV companies will fail, but maybe the one that one were producing cool for people, and that's neat, right?
50:59
So, what could be good for society might be bad for investors.
51:06
Dan Rasmussen: I think a lot about or I've been trying to think a lot about not just how to belief structures relate to markets, how can we get a sense of what people's belief structures are, how can we look at them over time, what are they saying in surveys, and how do those surveys result in different behaviors in markets, which I think is a really interesting question and fun, but also different utilities, right?
51:26
What do people want out of the market?
51:26
How do the young people want something different out of the stock market than older people?
51:30
Dan Rasmussen: For example, how do people's actual preferences relate to what they own?
51:32
I think starting to think about your start with your SPAC-type perspective, everybody's goal must be to maximize absolute returns, and then you just walk that back and you say, "Well, how many people who I met that actually very coldly calculate expected returns and hold the highest expected return thing?" None.
51:51
So, there must be something else that work. So, what do people want?
51:51
What is the purpose of their investments, and how does that overlap with beliefs, which I think is so important with the development of meme stocks and the power of these growth stocks at hyperly huge, huge market most and the desire of people to invest in innovation ETFs.
52:11
You just can't separate what's going on in markets from the reality that markets are made up of people.
52:16
Dan Rasmussen: I think people do have some behavioral attributes.
52:17
That's one of the things I love about crisis and crisis investing, right?
52:22
It's obviously hardwired to what people do in a crisis. They're scared. They listen to experts.
52:28
They extrapolate from the recent past.
52:28
scared. They listen to experts. They extrapolate from the recent past. So and so who predicted the crisis and short at stocks three months ago thinks it's going to get even worse and so you should probably listen to their advice because they were right versus, I think, this alternative approach, which is to be driven by base rates and to say, "Rather than thinking
52:50
what worked over the last decade in the market, why don't I think right now, given where valuations are, where the yield curve is, where the high yield spread is, and what would similar situations have looked like in the past in terms of expected return across different assets and, therefore,
53:06
maybe my behavior should be dramatically different because the stuff that worked best last decade might be the most overvalued this decade and, conversely, the stuff that's most out of favor might actually work." Dan Rasmussen:
53:14
Dan Rasmussen: Unless you start from a base rate-driven approach, you're going to be so driven by narrative and trend extrapolation and confirmation bias and the herding that everybody else is doing and so you should probably do it, too, that I think it can lead to a lot of bad decisions. Jim O'Shaughnessy: Yeah.
53:31
I've been a quant it's so long that it's in my DNA.
53:34
I will get physically uncomfortable if I'm in a crowded trade because I just had this intuition.
53:41
If I'm in a crowded trade, I'm wrong.
53:44
Dan Rasmussen: Yup, yup, yup, yup.
53:45
Jim O'Shaughnessy: Even though I'm following the base rates, even though I'm doing all of that and I'm following the results of the model, man, I do not like seeing company.
53:51
I like people calling me that, "You're insane. Why are you doing that?"
53:56
because that gives me a much higher degree of confidence.
54:00
Jim O'Shaughnessy: One of the last things that I want to ask you about is something that has been a pet project of mine, which we hadn't been able to put all the brainiacs over the OSAM Research on because we're spending all our time with canvass.
54:13
I'm sure you're pretty much familiar about that. Dan Rasmussen: Yeah. Congratulations.
54:17
Jim O'Shaughnessy: Thank you. Thank you.
54:17
Team, entirely team and Patrick. Fantastic. I love it.
54:19
So, I have a thesis that basically is the stock market is a complex adaptive system with feedback.
54:27
The reason it clears most of the time is because as you noted a moment ago, people have very different expectations and desires from participating in the stock market, right?
54:43
So, we could say that their viewpoints are heterogeneous. They're very different.
54:48
So, I could sell Apple to you and we could both be right by our own thinking, right?
54:54
You're buying it for a child's future education.
54:54
I'm selling it because I want to give the money to my grandson, something very simple.
55:02
Jim O'Shaughnessy: This is why markets are generally efficient in terms of, and I don't mean that in terms of reflecting all, I mean, they clear.
55:07
In crisis as we know, they don't clear.
55:12
So, that changes when feedback turns into what I call an information cascade.
55:21
That information cascade is mimetically so powerful that people's opinions become homogenous.
55:31
In other words, they're all thinking the same thing, right?
55:31
I think that when that happens, you're talking about the end of a bubble or the beginning of bad things, talking about a black swan, which by definition can't be predicted, but my theory is you might be able to confirm that a black swan has occurred.
55:54
Jim O'Shaughnessy: The reason I'm going into this is because your thesis on emerging markets and crisis is a similar with a different narrative, but it's a similar point of view as this idea of mine.
56:02
I think that the ability to confirm a black swan, because by definition you can't forecast a black swan, but the ability to confirm that one has actually occurred would give on an enormous advantage in markets if one had the flexibility of mind to say, "Okay.
56:24
The data is now confirming this for me.
56:24
The black swan did, in fact, happen."
56:32
Jim O'Shaughnessy: I think, though, that, again, my hypothesis is this is going to need machine learning.
56:40
This is going to need huge dataset, which, by the way, my entire Twitter feed is designed in caps.
56:47
Dan Rasmussen: It's all one big experiment.
56:48
Jim O'Shaughnessy: Of course.
56:50
Dan Rasmussen: Thanks for playing.
56:51
Jim O'Shaughnessy: What do you think about that?
56:53
Do you think that it's possible?
56:53
Because going into this and it's been my hypothesis for 20 years, really, and machine learning came along and I'm like, "Ooh, I might be able to actually run some real test on this."
57:04
Do you think it's possible to confirm, not forecast, but to confirm a black swan has occurred and, therefore, I might want to be doing the opposite?
57:18
Dan Rasmussen: What's a tangible example, Jim?
57:21
Can you make it less abstract?
57:24
Jim O'Shaughnessy: Of course.
57:24
The last great financial crisis, the collapse of real estate, the collapse of CDOs, all of that.
57:28
This is what got me really wanting to test it, honestly, because for the year prior to that, I was still at Bear Stearns at the time, but I had been walking around telling any other senior managing director, "If you can short your house, short your house because this is insane.
57:45
This will not last," but I'm such a quant I didn't act on it because I didn't have any quantitative information or base rates or all of the things that I've relied on for my entire investment career saying, "Yes, Jim. You're right." That's why I got it.
58:07
Jim O'Shaughnessy: We think now, by the way, that everyone knew that it was a black swan. No, they did not.
58:11
Go back to the source material.
58:11
If you go back to the source material, there were months and months after the decline had begun, where everyone was saying, "Oh, no, no. This is fine. This is fine. This is fine." It wasn't fine. Dan Rasmussen: Yeah.
58:29
I think it's fascinating.
58:29
I think it brings to mind.
58:29
I mean, I think when you start to think about belief structures, there's two related concepts come to mind, which are consensus and salience.
58:35
Consensus is how much does everybody agree on it, and then salience is how impactful on a mimetic level is the event that changes the narrative?
58:40
What are the events upon which that consensus is built?
58:48
What are the fact patterns in which that consensus is built?
58:53
When the black swan comes, how disconfirmatory is it and how quickly are people to adapt and change their beliefs?
58:59
Dan Rasmussen: That is a fascinating question.
59:02
I don't know how you'd test it or prove it out, but I think the idea that the only way to break a consensus bubble like that is with a salient event that changes the underlying fact pattern. It got to be right.
59:12
There's got to be something that destroys the narrative.
59:17
By the way, this is purely speculative, but- Jim O'Shaughnessy: Yeah, but we should make that clear.
59:20
This is purely speculative on my part as well.
59:22
You're a very smart guy, and that's why I'm seeking your thoughts on this because like usually, I'm going in to testing this hypothesis with the null set being expected.
59:33
So, I'm expecting I'm going to find nothing, but I'm still going to look.
59:38
Dan Rasmussen: So, one random one about current market events, and I've got a few different ones, but I think what a salient event that destroys a narrative.
59:44
I think one really popular narrative right now, and there are a number, but it's we're not in a bubble like 1999 because the FANMAG stocks are actually good businesses with high margins and very high growth.
59:55
If you look at their contribution to the stock market's earnings growth, it's been massive.
1:00:00
Therefore, their valuations are worth a very big premium and it's not a bubble, right?
1:00:08
Dan Rasmussen: Other things other than the FANMAG stocks might be a bubble, but not the FANMAG.
1:00:11
They're just great companies at the right price.
1:00:11
That species of narrative, I think you could talk to everybody from great value investors to great growth investors and they'd be like, "Yeah, I agree, of course."
1:00:20
What could possibly be wrong?
1:00:23
Dan Rasmussen: One of my very speculative thesis, and I'm almost certain to be wrong in this, but if you just have one of these big ones, have a negative revenue growth year or even ... What happens? Take Amazon.
1:00:35
Let me play out a narrative where COVID turned out to be the best year ever for Amazon's retail operation. Why?
1:00:40
Because people were sitting at home, they had nothing to do, and they're terrified to go to the store, so they just bought stuff online and had it shipped to them and shifted all their purchasing online, and then the minute the pandemic ended, people thought, "Wouldn't it be fun to go shopping again?
1:00:54
Instead of sitting on my computer buying stuff online, I should go out and play soccer with my friends."
1:01:03
Dan Rasmussen: Then what if Amazon's retail sales didn't just grow less than expected, but went down?
1:01:06
What would happen to this entire narrative?
1:01:12
That salient event would say, "Oh, my God!
1:01:12
Maybe they're not growing into perpetuity.
1:01:16
Maybe they're not just great business.
1:01:16
Maybe they're cyclical, too.
1:01:16
I start to think that's just one who knows how low a probability case that story is, but certainly a very low probability one, but you start to think about other potential salient events that could puncture a very consensus narrative.
1:01:30
Dan Rasmussen: The other one, I think, on the private equity side is a top tier private equity firm doing what Greensill just did, where all of a sudden, holy smokes, it just blows up and people say, "Oh, I get it.
1:01:42
An illiquid 10-year lockup can lead to this type of behavior when I trust them to give me the marks."
1:01:49
Dan Rasmussen: I think you just need a few of these really salient events to have such a dramatic shift.
1:01:54
I think there's a great study that looks at value and growth stocks and how they react to news.
1:01:58
Basically, what they found is that value stocks tend to have a positive reaction to news and growth stocks have a negative reaction to news because what's sustaining the growth stock valuation is a narrative about really great past data, and the future just unfolds relatively unpredictable and it's inevitably going to puncture it for whatever reason.
1:02:14
Whereas value stocks is built on a very negative narrative that strings together the past facts and then a new fact occurs, and it's rarely ever just on a straight line linearly down like all the past facts.
1:02:21
So, I think that there's got to be some way to test that or run machine learning or broaden that thesis out, but I would be fascinated to see it, Jim.
1:02:31
Jim O'Shaughnessy: So, you've actually just described some of the format of some of the questions we're going to be asking through machine learning.
1:02:37
So, we have this thing at OSAM called OSAM Research Partners.
1:02:41
Jesse Livermore, one of the most brilliant guys I've ever met in my life, he doesn't have anything to do with the stock market, but when you give him the database, oh, my God!
1:02:50
We have a fellow who's also an expert at machine learning, and he came up a couple of years ago and gave the firm basically a day on machine learning, and it was great because it took away all of the magic that people infer or imply is present in AI.
1:03:10
Jim O'Shaughnessy: Kevin just was like, "No, no, no, no, no.
1:03:12
Don't believe any of that.
1:03:12
That is what marketers are telling you, but I'm going to tell you why this is still very powerful."
1:03:18
Really, it's like just think of the old decision tree, but think of it at hyper speed, right?
1:03:23
So, it becomes much clearer why it works because it does in a fraction of the time what we as human beings literally either couldn't do or if we try to do it, I often think about NASA with their human calculators, right?
1:03:45
Back then, they didn't have computers and they called their staff calculators because that's what they would do.
1:03:53
Jim O'Shaughnessy: So, this is a very temporal phenomenon, I think, because what machine learning does is it compresses what would take humans if they even dare to try it, decades, decades, well, like the vaccine, right?
1:04:04
How long did humans try to [inaudible] How long did it take AI?
1:04:10
So, that part makes a lot of sense to me.
1:04:17
Jim O'Shaughnessy: So, because it demystifies it, it's not like some magic.
1:04:18
It's a very profoundly pretty simple at its core.
1:04:26
What it does, however, is it gives you the ability to compress into a day what would have taken a decade for we humans to do.
1:04:33
So, there is this theory, for whatever reason I'm very popular with young people like Zoomers and millennials.
1:04:43
I think one of the reasons for it is the gifs, honestly, but one of the things that I love is there's these three schools of thought.
1:04:54
I'm not going to tell you about the other two.
1:04:54
I'm going to tell you about the one that uses meme most.
1:04:58
Jim O'Shaughnessy: Inevitably, when I dig in to it, I find that the holders of this view are almost universally techies.
1:05:08
This view is a 60-year-old White male, it is inconceivable that he could come up with these gifs that match perfectly with the response.
1:05:16
Therefore, it is AI that they're testing surreptitiously and not telling us about.
1:05:27
Dan Rasmussen: That's amazing. I love it.
1:05:29
Jim O'Shaughnessy: Well, of course, you know what I do. I feed it.
1:05:32
Dan Rasmussen: Yeah, yeah.
1:05:33
Jim, I won't tell anyone the truth, I promise.
1:05:36
Jim O'Shaughnessy: I really appreciate that because before we started recording, I admitted to you that.
1:05:38
Anyway- Dan Rasmussen: The hologram that I'm talking to admitted it.
1:05:44
Jim O'Shaughnessy: Exactly.
1:05:45
The virtual Jim is I keep saying in beta, "He's not beta."
1:05:45
This has been great, Dan.
1:05:55
So, one of the things we always ask at the end is it's fun because it really allows us to see where your head is at in terms of if you could do something and, actually, people would do what you said.
1:06:08
So, our question that we're going to pose to you is this.
1:06:08
We're going to wave a wand and you become emperor of the world for a day. You can't kill anybody.
1:06:20
You can't send anyone to a camp for reeducation, but you get to hack human OS, and you get to implant two ideas or types of behaviors that humans would engage in after your day as the impresario manifestor of the empire.
1:06:36
What two things are you going to put in people's minds and change their behavior about?
1:06:43
Dan Rasmussen: Well, that's such a terrifying question to be honest, Jim.
1:06:46
I mean, I'm a believer in the idea of subsidiarity, that the decision should be made at the level closest to which they're made or even high, right?
1:06:54
Someone say the price mechanism is the best mechanism.
1:06:59
I can see no circumstance in which implanting pretty much any idea in anybody would result in anything good.
1:07:04
I'm so opposed to consensus, Jim.
1:07:04
I just feel like maybe the one would be just telling people just don't ...
1:07:09
Dan Rasmussen: Then you think if you tell them don't trust authority, so where does that lead, right?
1:07:13
So, I just don't think it's good, right?
1:07:13
We need human cognitive and intellectual diversity, and diversity of opinion.
1:07:18
Dan Rasmussen: I honestly think that one of the sad things that I worry about, I don't like consensus. I really don't.
1:07:24
I hate it when it's forced consensus or when everybody forces you to think some way.
1:07:30
Actually, my wife and I, we came up with a phrase, which we like to use sometimes as a tool where we say, "Well, is it possible that a smart and well-intentioned person could disagree with you on that?"
1:07:39
You'd be shocked how often the answer these days is no, right? "No, it's not possible."
1:07:44
Dan Rasmussen: You just say, "Well, I guess I'm just going to stay really quiet because if I say what I think, you're either going to think I'm evil or stupid and I don't want you to think either of those things.
1:07:55
So, you're just going to think I'm quiet."
1:07:58
Jim O'Shaughnessy: Well, I love that response, and that's why I came up with the question, by the way, because I think just like you do. I don't like consensus.
1:08:07
I don't like being told what to thought, but it's part of the scientific method.
1:08:07
The ethos that the enlightenment gave us is that we have to allow criticism.
1:08:12
We have to allow people, in fact, not allow, we have to seek out people disagreeing with us because if they don't, our theory will die in the water because it hasn't been questioned.
1:08:30
Jim O'Shaughnessy: So, one of the things that I was expecting you to say was that you would make people more open to discussion and less fearful about stating their actual preferences because another thing that we could do an entire hour on is stated preferences and revealed preferences.
1:08:49
I don't think the gap has ever been wider, and I've been around for a while, and I pay attention to these things.
1:08:57
Jim O'Shaughnessy: It's an epidemic among young people.
1:08:57
I talk to a lot of young people and that's the first thing out of their mouth is, "I'm afraid to say what I think," and that is not an environment that leads to progress, innovation, change.
1:09:15
That is an environment that leads to witch trials and I have a list like Senator McCarthy.
1:09:23
I don't have a political party because none of them reflect the way I think about things.
1:09:27
I guess I'm anti-authoritarian of any kind, of the left or the right.
1:09:27
So, if I were doing it, it wouldn't be anti-authoritarian.
1:09:34
It would be think for yourself, I guess would be one of the ones that I would put.
1:09:39
Dan Rasmussen: Maybe I'll give one, which is- Jim O'Shaughnessy: I knew you're going to get- Dan Rasmussen: This Solzhenitsyn quote where he says, "Live not by lies."
1:09:47
Jim O'Shaughnessy: Oh, I love it. That's so great.
1:09:50
Dan Rasmussen: You see a lot of people live by lies. It's very sad, right?
1:09:53
"Why are you doing this?"
1:09:53
Dan Rasmussen: "Because I just want people to think X, Y or Z about me, so I'm not going to reveal the truth of what I think."
1:09:59
Dan Rasmussen: I mean, it is horrible, right?
1:10:00
We live in a wonderful free country.
1:10:00
Why should we be afraid to express our views or, in many cases, afraid to listen to somebody express views we disagree with or get so angry about it? It really is sad.
1:10:08
I think part of it is a failure, this rational expectation verus rational beliefs debate that happens in markets, right?
1:10:18
People need to read out their Isaiah Berlin or whatever and realize, "Gee!
1:10:18
It's not that this person is stupid or evil.
1:10:22
They just see the world differently.
1:10:22
They interpret the world differently.
1:10:27
They see the same facts and come to different conclusions.
1:10:27
They have different values or different hopes or different expectations that are reasonable.
1:10:31
If I can just ask them more about it, I can figure out what their reasons are and it's probably going to be interesting," but people just assume, "Wow!
1:10:39
Well, if you disagree, you must be evil or stupid," right?
1:10:44
That's why people don't want to express their views, right?
1:10:48
Jim O'Shaughnessy: I love it.
1:10:48
Live not by lies, what a great one.
1:10:52
I'm stealing that from you immediately.
1:10:52
Dan Rasmussen: Well, I stole it from Solzhenitsyn.
1:10:54
Jim O'Shaughnessy: Great. That's great.
1:10:57
That's wonderful because that's a subpart of my other theory that we were discussing that if you have a mechanism that you have greater insight into revealed preferences versus stated preferences, well, that's an interesting data.
1:11:13
Dan Rasmussen: Right, right, right, right, right.
1:11:14
Jim O'Shaughnessy: Listen.
1:11:14
So, Dan, this has been as fun as I thought it was going to be. You are a wonderful guy.
1:11:16
I wish you all of the greatest success.
1:11:21
Verdad, what's next for the company?
1:11:21
Are you adding people, subtracting people?
1:11:26
Dan Rasmussen: We hope not to subtract anybody, but, no.
1:11:30
We're doing a lot of- Jim O'Shaughnessy: Did I slip that in there? Dan Rasmussen: Yeah. Exactly.
1:11:35
We're deleting four people tomorrow.
1:11:35
We hope it's not you, but we're going to pull the names out of a hat. No. We're ideally not.
1:11:39
We're doing a ton of research.
1:11:44
I think that what I've gotten really interested in, stemming from this crisis work and we've been doing a ton of work on crisis, is saying, "Okay.
1:11:49
Well, what to do when you're not in a crisis?"
1:11:53
I think we've been trying to look at a lot of the research into macro variables and how those predict different asset classes do and can you structure approach to asset allocation that's countercyclical, that takes into account expected returns, that takes into account where the macro is.
1:12:11
Dan Rasmussen: So, I'll give you just one of the random case studies of something that we've been looking into, which is totally random and anecdotal but fun is oil futures, okay?
1:12:18
So, if you look at the long-term of oil futures, it's got pretty much on a real basis a 0% return, a max drawdown of 90%, and a crazy- Jim O'Shaughnessy: What's not to like? Dan Rasmussen: Right.
1:12:35
So, you look at oil futures and you're like, "Holy shit!
1:12:35
This is just the most ... Why would anyone ..."
1:12:38
Certainly, we don't want to have a permanent allocation.
1:12:38
It's nothing that looks like that, but it just terrifies me.
1:12:41
It's just seen as max drawdown expense.
1:12:45
So, then what I try to pierce together was to say, "Okay.
1:12:45
Well, certainly, sometimes oil futures have to work, otherwise, people wouldn't be so drawn to them."
1:12:49
What are the conditions under which they work?
1:12:53
Ran a huge ton of different test and tested all different variables and read the literature on it and came up with two that are really interesting.
1:13:03
Dan Rasmussen: One is that when the real price of oil is below its long-term average, that's a positive signal.
1:13:06
So, okay, right, that makes sense.
1:13:11
When oil prices drop, people drill or, sorry, when oil prices drop, they stop drilling.
1:13:14
Jim O'Shaughnessy: Stop drilling.
1:13:15
Dan Rasmussen: Mean reverts.
1:13:15
When prices get really expensive, people drill, the price goes down. So, you're like, "Okay. Got it."
1:13:17
Oil should be mean reverting.
1:13:21
It turns out oil prices are mean reverting.
1:13:21
So, buying oil futures when oil prices are below long-term mean, okay, good.
1:13:26
That only gets you part of the way, though.
1:13:26
The other interesting one is my favorite macro variable is the high yield spread, which is a great contemporaneous measure of economic conditions.
1:13:34
When the high yield spread is really wide, borrowing is shut down, you're in the middle of a recession, and then when the spreads are really tight, it tends to be late cycle inflation, excessive credit creation, et cetera.
1:13:47
Dan Rasmussen: Actually, it turns out that if you look at oil futures and you segment it out by quartile, oil futures do really well when high yield spreads are really wide, right?
1:13:55
So, think of March of 2020.
1:13:55
If you then bought oil futures, you did well because the economy then recovered.
1:13:59
Then they also work really well when high yield spreads are really, really tight, which is because there's inflation.
1:14:03
If you pair those two things and said and call it 25% of market environments where those two conditions are met, where prices are below average, and the high yield spread is extreme, that's basically 100% of the returns in oil over the entire history have come in periods that met those two pre-conditions.
1:14:25
Dan Rasmussen: You've been better off not even touching oil futures the other 75% or 80% of the time because you're owning an asset with 0% return and 90% max drawdown.
1:14:30
So, we've been trying to think, "Are there other asset classes or places, not just emerging markets or small value stocks, where you actually only want to be in them on occasion because the conviction level is just so much higher under certain conditions than it is under others.
1:14:46
So, there are these conditional probabilities or base rates.
1:14:49
Dan Rasmussen: I know this is pushing far out into the heretical because I think a lot of people are so averse to any market timing and, of course, there's a lot of data mining risk to any type of investigation along these lines, but I think it's worth pursuing.
1:15:01
So, that's where a lot of our research efforts are focused.
1:15:05
Jim O'Shaughnessy: Fantastic.
1:15:07
For everyone listening, I always think that Dan and team's research is great. I always read it. I think you should, too. Dan, thank you so much.
1:15:12
How do we find you on Twitter and how do we find your website?
1:15:18
Dan Rasmussen: My Twitter is @VerdadCap, and my website is verdadcap. com.
1:15:21
We write a weekly emerging markets.
1:15:27
It comes out at Monday mornings at 9:00 AM.
1:15:27
I've been doing it for a few years now.
1:15:31
If you like this podcast and would like to read more of our research, I encourage you to sign up for it.
1:15:36
Jim O'Shaughnessy: Terrific. This has been great. Thanks a lot.
1:15:37
Dan Rasmussen: Thank you, Jim.
1:15:38
Jim O'Shaughnessy: Cheers.