The transcript from this week’s, MiB: Jim O’Shaughnessy, O’Shaughnessy Ventures, is below.
You can stream and download our full conversation, including any podcast extras, on Apple Podcasts, Spotify, YouTube, and Bloomberg. All of our earlier podcasts on your favorite pod hosts can be found here.
~~~
This is Masters in business with Barry Ritholtz on Bloomberg Radio
Barry Ritholtz: This week on the podcast, boy do I have an extra special guest. I know Jim O’Shaughnessy for, I don’t know, maybe 20 plus years, something like that. We actually first met in the Green Room at CNBC, like early two thousands and found, we shared some similar likes and and philosophies, and I’ve been a fan of his book, what Works on Wall Street, pretty much from when it came out. This is a fascinating conversation about a person who has worked through multiple locales and seats in finance, not just running systematic investing at Bear Stearns, but creating O’Shaughnessy Asset Management, creating a unique custom index product that ended up attracting the attention of Franklin Templeton, who, who paid some undisclosed and ungodly amount of money for the whole firm. And now in a later phase of his career doing os Shaughnessy Ventures and, and the Os Shaughnessy Fellowship.
I first know him from really the first quant book, what Works on Wall Street. That was a half a century of data analysis, really was never accessible to the public before. I found the conversation to be fascinating. And I think you will also, and at this point I am obligated to do a disclosure. My firm, RITHOLTZ’s Wealth Management, has been working with O’Shaughnessy on their direct index platform. Really, we were one of the first beta testers. We now have over a billion dollars on that platform, maybe coming even closer to another big round number. With no further ado, my discussion with O’Shaughnessy Ventures.
Jim O’Shaughnessy, welcome back to Bloomberg
Jim O’Shaughnessy: It’s great to see you Barry, and congratulations. Wow, that’s amazing.
Barry Ritholtz: Congratulations to you. I I, I’m still my firm just had its 10th anniversary. You guys, anytime I see the phrase for an undisclosed amount, my brain automatically says, wow, that has to be a lot of money. If it’s, if they’re not disclosing it, it’s material but undisclosed. That’s a lot of cash.
00:02:26 [Jim O’Shaughnessy] It could be like trading places and the normal bet of a dollar.
00:02:30 [Barry Ritholtz] That’s right. The usual, the usual bet Mortimer one, $1. So, so we
know each other from way back when you first came into my orbit from the book, what Works
on Wall Street, I read it from cover to cover. I was on a trading desk when that came out and I’m
like, huh. So there’s some science and math behind this. It’s not just rumors and whatever
happens to cross TV that day. I’m intrigued. Before we get there, let’s talk a little bit about what
you were doing prior. Tell us about the early Jim O’Shaughnessy.
00:03:05 [Jim O’Shaughnessy] Well, I was always fascinated about markets in general, which
stemmed from a very angry conversation between my uncle and father about IBM and I. I had
just been allowed to go to the adult table, right? And I was sitting next to my dad and he and myUncle John were going hammer and tong about whether IBM was a good company or not. And I
was listening and it was all about the chairman. It was all about, you know, things that I looked
at as kind of soft intelligence. Squishy, squishy. And so I just thought, I asked at the dinner, I
said, well, would it make more sense to like, look at how much money they’re making and what
their earnings are and how much you have to pay for that? And they both just literally glared at
me.
00:04:00 [Barry Ritholtz] That’s hilarious. Kids, they don’t know anything do you?
00:04:03 [Jim O’Shaughnessy] Exactly, exactly.
00:04:04 [Speaker Changed] It’s the chairman. How tall is he? I like the cut of his jib.
00:04:08 [Speaker Changed] It’s almost as if you were there. That bug got implanted. That mind
worm got implanted in my brain.
00:04:15 [Speaker Changed] How, how old were you when that this
00:04:17 [Speaker Changed] Happened? I was 17.
00:04:18 [Speaker Changed] Oh, so you are just going into college.
00:04:20 [Speaker Changed] Yeah,
00:04:21 [Speaker Changed] Absolutely. And and you were a Minnesota kid, is
00:04:24 [Speaker Changed] That right? Yep. I grew up in St. Paul, Minnesota and beautiful
00:04:28 [Speaker Changed] Country, certainly in the summer. Anyway, gorgeous.
00:04:30 [Speaker Changed] The winters are tough. Yeah, yeah. Well if this were the old USSR,
right? That is where all the political prisoners would be.
00:04:39 [Speaker Changed] Send them to Minnesota.
00:04:41 [Speaker Changed] That’s hilarious. But, but so I started doing research on essentially
the Dow 30 because it was manageable. 30 stocks I could list by hand showing how old I am
because you literally, there were no computers that we could use at the time. Simple things like
what’s the price, what’s the dividend, what’s the price to earnings, book value, et cetera. And I
found a definite trend, right? I found that buying the 10 stocks in the DAO with the lowest PEs
from 19, like 35, I think I started through when I was doing it, and this would’ve been about
1980, absolutely decimated the 10 highest PE stocks. Wow. So, wow, I love this. In the
meantime, I had computers and the only reason I actually got to write what works on Wall Street
was because Ben Graham didn’t have computers. If he had had them, I would’ve had no chance
’cause he would’ve done it.
00:05:44 Basically what I wanted to see was, is there any rhyme or reason to all of these reasons
people say they like or hate a stock, right? Where is the proof, where is the empirical evidence
that say buying the low PE stocks from the Dow works very well over many market cycles? So I
wrote a first book called Invest Like the Best, in which I basically showed you how you couldclone your favorite portfolio manager by taking his or her stocks, putting them on a big database
like Compus stat, seeing how they differed from the overall market and then using those as factor
screens to get down to a portfolio that looked acted and most importantly performed like your
favorite manager.
00:06:32 [Speaker Changed] Now, the average investor typically didn’t have access to CompStat,
to big data, to big computers. And so they relied on you who did, and if I recall what works on
Wall Street, you back tested like half a century worth of data, something like that. And it was the
full market, not just the 30 Dow stocks.
00:06:52 [Speaker Changed] Yeah, absolutely. And and also not just the full market, it was also
any company that had been around but went bankrupt or got taken over the very, very needed
research database on Compu stat.
00:07:08 [Speaker Changed] So no survivorship bias, none you back that out. That’s great.
00:07:12 [Speaker Changed] Yeah. Yeah. Because some of the early academic studies were, they
had a lot of survivorship bias. They didn’t properly lag for when you actually knew a number. So
they just assumed, right, well there’s the number on March 31st, I’m gonna use that number.
Well, you didn’t really know that for most of history until maybe May or June.
00:07:39 [Speaker Changed] Really interesting. So you run these numbers, what sort of
strategies do you find perform best?
00:07:46 [Speaker Changed] Well, we found that on the value side, smaller value stocks that had
some catalyst and had turned a corner and their prices had started to go up was a beautiful
strategy.
00:08:02 [Speaker Changed] Small cap value with a touch of
00:08:04 [Speaker Changed] Momentum. Momentum, yes. Okay. On the growth side, we found
momentum works really, really well. As we continued the research, we found, okay, there’s all
sorts of caveats. So for example, we learned after a severe bear market IE one in which the
market had to declined by 40 or more percent. Wow. Not a lot of those. Not a lot, thank God. But
momentum inverted and the stocks with the worst six or 12 month momentum actually did vastly
better than the ones with the best. And if you think about it, even for a minute, it makes sense,
right?
00:08:43 [Speaker Changed] The deepest value.
00:08:45 [Speaker Changed] But what happened was a lot of really great stocks during the bear
market got pushed way low in price. And so people, when the market was recovering, jumped on
those stocks, they were like, I can’t believe I’m getting, you know, these earnings at six times
earnings for an IBM or a, you know Qualcomm, right?
00:09:06 [Speaker Changed] That’s the baby with the bath water
00:09:08 [Speaker Changed] Strategy. Exactly. And so, but we found, you know, that value
actually works. Now it hasn’t for a long time, but we also found that large stocks with highshareholder yield, IE dividend yield plus buyback yield was an excellent way to identify big
stocks that are obviously much more conservative than the smaller fry in the small cap world.
00:09:40 [Speaker Changed] Hmm, interesting. So, so let’s talk a little bit about your work at
Bear Stearns. Really, where I first met you in the two thousands, you were head of systematic
equity at Bear Stearns Asset Management. I’m assuming you are applying a lot of the lessons you
learned in what works on Wall Street to the bear institutional and retail investing strategies.
00:10:01 [Speaker Changed] Absolutely. And you know, let me just say Bear was really a great
company, very unfortunate what happened to it during the financial crisis, but the reason I love
Bear is, you know, a lot of big banks talk about being entrepreneurial. Bear Stearns really was.
And essentially if you were doing your thing and playing by the rules and doing well, they let
you alone. Which was pretty important for me because when I got there, it was right after the dot
bomb. And a lot of the brokers had done pretty poorly because they were in a lot of those names.
And so I convinced Steve Dantes, who was then head of private client services that wouldn’t it be
better if we did a packaged portfolio, a a separately managed account. And we offered at one
time, I think we were all the way up at 10 to the brokers so that they could use a more systematic
time tested way of investing for their clients,
00:11:11 [Speaker Changed] Bringing a little discipline into what had been, at least in the
nineties, very much a cowboy type of environment. And I’m not just referring to Bear, the entire
retail stock brokerage was wild.
00:11:24 [Speaker Changed] Totally. He was very open to it. We ended up putting together a
separately managed account platform that the brokers embraced. They loved it because literally
they did what they did well, which was calm the client during bad times, try to keep ’em from
getting too excited during great times, but they also loved the idea that it had a very explicit
explanation for why they were putting that client in that portfolio. So that was a lot of fun. By the
time I left Bearer, my group controlled about 70% of Bear Stearns asset management long
00:12:04 [Speaker Changed] Only. And that was a lot of money, wasn’t it?
00:12:06 [Speaker Changed] It was, it was about $14 billion.
00:12:09 [Speaker Changed] Okay. So you mentioned you left Bear, let’s put a little flesh on on
those bones. Your timing was perfect. You exit Bear in 2007, is that right? To, to set up Nessy
Asset Management was the thinking, Hey, I want to do this out on my own shop, or were you
sniffing something out in oh seven that’s like, Hey, maybe I don’t wanna be attached to a giant
Ocean liner taking on water.
00:12:38 [Speaker Changed] You know, that’s funny. I spent the next two years after that trying
to convince reporters that I really didn’t know anything. Why I left Bear was because I felt that I
really wanted to be on my own. Again, I really wanted to be able to just talk about quantitative
investing. Bear was a boutique, so there were a lot of different managers, right. Liked them all,
all thought they all were great, but I really, really wanted to focus just exclusively on Quant. And
secondly, we had upgraded a lot of our systems to the idea that would become Canvas. Right.
Because remember Net Folio was our first try at that.00:13:23 [Speaker Changed] That was nineties, right?
00:13:25 [Speaker Changed] 99. Yeah. Really. Well of course, you know, you know the really
funny story here is in April of 1999, I wrote a piece called the Internet Contrarian. And in that
piece I said 85% of the companies currently ex in the internet space are gonna be carried outta
the market feet first. The, I’ve never seen a bubble like this in my history of investing and what
did I do next, Barry? I started an internet company.
00:13:57 [Speaker Changed] Well, just because the stocks are a bubble doesn’t mean this internet
thingy isn’t gonna catch on. That’s true. Right? It’s quite true. It’s, it’s, there are, you know, it’s
funny, we forget in the thirties, forties, fifties, there was only Ma Bell. Every company used
telephones. Yep. The way we describe internet companies, if you use the internet as a core part
of your platform is difference between the dot coms and the nineties and people who have just
really integrated the technology into their business. Right? Absolutely. So I think Net Folio is not
a.com, but a com that used the net as a way to reach more people and give them access to data.
Well,
00:14:39 [Speaker Changed] It’s really funny because I made a couple, well I made more than a
couple of mistakes, but one of the big ones I made was we designed Net Folio as a B2C
company, right? So we called, we were taking on at the time mutual funds, which were
dominant. We didn’t have ETFs while we had them, but they were in their, they
00:15:00 [Speaker Changed] Were very early days.
00:15:00 [Speaker Changed] Very very early days. Right. And so we had
00:15:04 [Speaker Changed] What, what did the spiders just turn 25 recently? Yeah, I think of
something like that. Yeah. So, so 99 is like, it was really the beginning.
00:15:12 [Speaker Changed] Oh, totally. And and basically the idea was it was the first online
investment advisor. And the reason that we thought it would work so well was personalization,
tax management, right? All of those things. So for example, we would, they were all run by
quant models that we had developed, right? And, but it gave the user the ability to say, let’s say
they’re anti-smoking, right? And Philip Morris is one of the selections they could just check,
Nope, don’t want it. Up comes the next stock that meets the criteria. And so it had a lot of really
great features, but the tech was not quite there
00:15:53 [Speaker Changed] Yet. You were 20 years ahead of where you would end up in the
late 2010s, right?
00:16:01 [Speaker Changed] I I, I was, I, I really do have to give my son Patrick the credit for
resurrecting the idea because when we were at OS A MI said, listen, we left Bear right into the
great financial crisis. And I put the team together and I’m like, I don’t think that we’re gonna be
able to sell many long only portfolios after the market has collapsed by nearly 50%. So let’s
spend our time developing internal technology that works the way we work. The off the shelf
stuff really wasn’t cutting it. And so the project to get there was multi-year and Patrick oversaw
that and then he walked into my office one day and he goes, you know, dad, we’ve been usingthe death star to kill a mouse. And I’m like, okay, I like the metaphor, but what do you mean?
And he started talking about AWS talking about Net Folio and he’s like, we have the perfect tech
now that our clients, OLS being one of them could use. And I’m like, brilliant, let’s go with it.
00:17:10 [Speaker Changed] So we’re gonna talk a little more about Canvas, but I wanna stay
with the launch of OEM in oh seven. So a, you don’t need to disclose this, but I’m gonna assume
you had a lot of bear stern stock options that you had a vest on your exit. So you probably had a
pretty good sale, pretty good print on on those when you first set up Nessy, you running your
traditional models, things like cornerstone value and cornerstone growth. And I’m a big fan of
your micro cap sleeve, which really operates parallel to venture capital returns only using public
stocks. Am am I getting that more or less right?
00:17:54 [Speaker Changed] Yeah, actually we wrote,
00:17:55 [Speaker Changed] We use that
00:17:56 [Speaker Changed] Also. Yeah. We wrote a paper saying that it was the poor man’s
way to get exposure to private equity.
00:18:02 [Speaker Changed] Private equity or venture capital or both?
00:18:05 [Speaker Changed] Both really private equity closer because the, the micro cap, I love
micro cap investing. The only real reason that we offered that was because I loved it so much.
Really
00:18:16 [Speaker Changed] Well, and the data backs it up, right? Oh,
00:18:18 [Speaker Changed] Totally, totally. It is. Micro cap is an amazing place if you’ve got
the right tools to sort through the thousands of names in the micro cap universe, because you
would not want to buy an index of micro cap stocks. For the most part they’re micro caps
because they kind of suck. However, there are so many diamonds in the rough, in micro cap that
if you have a strategy like a quant strategy that can sort through these thousands of names, you
can do extraordinarily well. I love the strategy and,
00:18:59 [Speaker Changed] And I know the os a micro cap sleeve is what I call it, has just
really shot the lights out. Especially last year when the market was having a pretty good year.
They They did pretty well, didn’t
00:19:11 [Speaker Changed] They? They did. They did. Now remember you introduced me as
chairman of om. I’m no longer. No longer. Yeah, I, they let me retire. And actually Patrick is
now chairman emeritus over at OS a. Let’s
00:19:26 [Speaker Changed] Talk a little bit about Canvas. And again, full disclosure, we’re a
client, we were a beta tester. We love the product, which is kind of ironic because I used to hate
direct indexing every time I would demo or see a product. It was clunky, it was klugy. You
would get these statements that were like hundreds of pages long. You guys kind of figured out
the secret sauce for how do we make this clean, usable, and easier to understand. Tell us a little
bit about the genesis of Canvas.00:20:02 [Speaker Changed] Well first of all, we call it custom indexing as opposed to direct.
And the reason I make that distinction is because as you point out, the direct indexing products
of that time were clunky. They were difficult. You got reams and reams of paper reports and they
were really only focusing on tax benefits. Right? What we wanted to do with Canvas, which is
custom indexing is as the name implies, give you as the advisor full control over what your client
portfolio wanted to look like. You got the advantages of tax harvesting, you got the advantages
of being able to mix indexes in with active strategies. But you could also do a social investing
fund if you want it. But the way we did it was we didn’t presume what your client was going to
think of as good social investing. So often when you see some of the ESG portfolios, they’ve
been predetermined as to what is going to be included.
00:21:13 We give you the tools to turn a dial up or down on whatever you want. I think last I
looked, there were over 58 separate things that you could fine tune around on the idea of ESG.
We wanted to give the tools to you because you knew your client vastly better than we did. And
we thought, let’s try, as you mentioned, you were one of the beta testers. That was actually one of
the smartest things we did. I think because we had really good advice from a lot of people that
we knew in both venture and other places. The first thing that many of them said to us was, do
not try to go big with this originally. Find advisors who you trust who will give you real
feedback. In other words, they won’t shine you on if they don’t like you. You guys were very
good at telling us what they did.
00:22:06 [Speaker Changed] Like, and Michael, Michael Batnick in my office, one of my
partners who was over the moon when he first saw this, every time another product came in, it
would take me 30 seconds to poke holes in it. And he, he came breathless into my office, dude,
you gotta see this. And I’m like, yeah, yeah, okay, another garbage direct let show tee it up. And
it took about 30 seconds to go, oh my God, how? How do we get a piece of this? This is
fantastic. The interface, the design, all of the bullet points that all the boxes are checked were
great. Let’s stick with what we no longer call ESG and Meyer Statman famously called values-
based investing. Some people have called it woke investing, but that’s really the wrong phrase.
I’m fascinated for example, by the Catholic bishops whose endowment says, look, we don’t want
any abort efficients there any drugs that do that. We can’t invest in those, those companies. We
can’t invest in hospital chains that perform these sort of surgeries or insurers. You have the
ability to say whatever your personal preferences are, you could just tune those out of pick an
index, the s and p 500, the Vanguard Total Market. You could say, I don’t want X or Y or Z and
out it comes. Tell us a little bit about that.
00:23:27 [Speaker Changed] I felt that that was really, really important because everybody has
different ideas. As you point out, the Catholic bishops wanted to exclude certain things, others
might want to include certain things actually felt, it would be very arrogant of us to determine
what good social investing was because we had managed money for a variety of religious
institutions. And guess what, they all have different takes on what they want to see. We did one
where, for example, you couldn’t buy any company that did anything with animals with eyes.
That was an interesting one. Huh. But then on the other hand, we had a client who wanted to see
more female board members and females in the C-suite.00:24:15 [Speaker Changed] And you could, you could screen for that. We can screen for that.
And there’s a bunch of research that shows those companies. Now you don’t know if it’s
causative or just merely correlated, but those companies tend to outperform the, the request we
probably hear the most is no gun stocks, no tobacco stocks. Yeah. Kind of interesting.
00:24:33 [Speaker Changed] Yeah. The tobacco guns, those are pretty large groups where
majority of investors want nothing to do with them. But the other thing that’s cool about our dials
on canvas, you, let’s say that Ritholtz has a wild-eyed libertarian walk in who happens to have a
billion dollars. And he says, you know what? I want the gun manufacturers I want, I’m a big
Second Amendment guy. Right? Right. Or I want the pharmaceuticals. Or I want the tobacco.
Gimme
00:25:04 [Speaker Changed] The sin stocks, gimme gambling and alcohol.
00:25:06 [Speaker Changed] Well, and you know, the joke there was that my first company,
O’Shaughnessy Capital Management, we used to keep a joke portfolio, which was called the Eat
drink and Be Merry for tomorrow. You die Barry. It killed it. Right? Killed it.
00:25:20 [Speaker Changed] Sure. So what ends up happening very often is when there’s a non-
financial reason for kicking a stock out out of a lot of portfolios. Eventually a company with still
having decent financial prospects, it becomes cheap.
00:25:37 [Speaker Changed] Yep, absolutely. But the thing with the social style investing, we
wanted you to be able to reflect your client’s unique needs and there really wasn’t anything like
that. I don’t know if there is now, but I I haven’t seen anything like that.
00:25:55 [Speaker Changed] Well, certainly not to this degree of granularity. By the way, when
we first were beta testing canvas internally, my view was, hey, people are gonna want to use this
for value-based investing, then they’re gonna want Deconcentrate. If I work for Google, do I
really need all this tech exposure? My income is coming from there. Let me diversify that way.
And then tax loss harvesting was gonna bring up the rear. I had it exactly backwards in large part
because, I don’t know, maybe a year into it, we had the Covid crash Market falls 34%. And
coincidentally bottoms just near the end of the quarter, that rebalance, you know, typical tax lost
harvesting your own a dozen mutual funds, eh, you pick up 10, 20 basis points against the
portfolio of losses to offset gains. The hope with this was, it would be 50 60. We had clients
getting 200, 300, 400 basis points. And I’ve talked to some of your staff or former staff and
they’ve told us some unique use cases where the numbers are are bonkers. First off, explain to the
audience who may not be familiar with this, what is tax loss harvesting?
00:27:13 [Speaker Changed] So essentially what it does is we had to build a non-trivial
algorithm that could monitor every portfolio we were managing on behalf of clients. And as you
know, they can go all the way up but get maximized tax losses or all the way down, you don’t
worry about them. So for example, you wouldn’t care about it in an I rra, right?
00:27:37 [Speaker Changed] Any qualified account. Right, right. No one00:27:39 [Speaker Changed] Cares. But, but the purpose was that we found through our research
that a tremendous amount of alpha was being left on the table. And that was the alpha from tax
lost harvesting. When you’re in a market like the market we had when we went into C, the bear
market ensued in under other circumstances. Well kinda you’re outta luck. But in this particular
case, that creates the kick in for harvesting the losses, reducing the overall tax needs for the
portfolio. And you could really look at that as that’s money in your pocket. By the way, we had
the benefits completely backward too. Tax loss harvesting was at the bottom of our list as well.
It’s,
00:28:27 [Speaker Changed] It’s arcane and technical and you don’t really think about it, but we
have clients who were either, you know, startup founders that cashed out or they inherited or, or
just own stock with a very low cost basis. You know, it’s always funny when you see a $5
million portfolio and some stock has blown up where it’s 80% of the holdings, Hey if, if you
have $5 million and 4 million of it is Apple or Amazon or some combination of big stocks, that’s
a lot of single stock risk. And to a man, every person says, Hey, you should diversify. The
answer is always, I’m gonna get killed in capital gains taxes. This worked out to be a really good
way to say, we are gonna work out of your concentrated position over 3, 4, 5 years. And then
2020 comes along and what should have been a five year process took half as long.
00:29:24 ’cause you had so many losses. So, so for those people who may not be familiar with
this, let’s say you own 10 mutual funds, right? And some are up, one or two are down, you sell
the ones that are down, you replace it with something very similar. Hey, now I got a little bit of
loss even and my portfolio looks the same, but I have an actual realized loss that I could use to
offset my real gains. But those losses are three, five, 10%. They’re nothing. On the other hand, if
you have a direct index or a custom index that has a couple of hundred stocks, well the worst
stocks in those portfolios, they’re not down three, four, 5%, they’re down 40, 60, 70 5%. You sell
the ones that are down, you replace them. And this is one of the things I like about Canvas. You
identify the replacement stocks that are, is it fair to say mathematically similar? They look well.
00:30:21 [Speaker Changed] So they come from, they come from the same strategy. So yeah,
you could say they were mathematically similar.
00:30:27 [Speaker Changed] So the overall portfolio, more or less retains the same
characteristics. You’re just realizing losses, deep losses on some stocks and replacing them with
something relatively similar.
00:30:40 [Speaker Changed] Exactly. And you know, we’re just basically making math work for
us. And because the entire thing is operated within the Canvas architecture after getting the
algorithm, which was non-trivial,
00:30:55 [Speaker Changed] What do you mean by non-trivial algorithm? It
00:30:57 [Speaker Changed] Took a hell of a lot of work. Okay. To be able to make that function
properly. And as we worked with firms like yours, it became very, very clear to us that that was
gonna be a big deal in Canvas. So we wanted that algorithm to work perfectly. But as you also
note, we wanted the nearest neighbor, if you will, that would replace that stock to not affect theoverall metrics of your portfolio. So it’s gonna look, act, and perform very much like the earlier
portfolio, but you’ve already taken that wonderful tax loss so that you can offset the gains from
elsewhere. The other use case that we thought would be number one was, you know, you have a
concentrated position, let’s say Google, right? Don’t give me any tech exposure. Right. Or give
me tech exposure only in this tech, which is like hardware for example, right. That I can do. And
that type of use case would work hand in hand with the tax loss, making it a much, much more
efficient, more money in the investor’s pocket. In terms of final outcomes with the portfolios,
00:32:15 [Speaker Changed] What, what was the uptake on that approach? Were people
enthusiastic about
00:32:19 [Speaker Changed] It or? They were, but they were not nearly as enthusiastic as we
anticipated they would be. Right? There were a few advisors that we were working with who
worked specifically with founders and early employees who had a lot of options in that particular
and usually tech. But we also did work and do work with a lot of people who just amassed
through employment, a huge position in their particular company. And they wanted to have the
rest of the portfolio be built to compliment and offset, if you will, any further investments over
there. So it’s worked actually quite nicely.
00:33:03 [Speaker Changed] Hmm. And then in 2021, Franklin Templeton comes knocking at
the door. They’re an investment giant with a trillion plus dollars on their books and they’ve been
pretty acquisitive over the past few years. Tell us a little bit about how that transaction began. If I
recall correctly, you guys weren’t out shopping the firm to be sold, were
00:33:26 [Speaker Changed] You? Not at all. We were, it’s a funny story. We almost got kind of
a cold call from a gentleman at Franklin Templeton. I was sort of like, give it to Chris Loveless
or you know, who’s the president of the firm. And ultimately Patrick spoke with him and came
into my office and he is like, Hey, Franklin Templeton is really interested in Canvas. I’m like,
okay, did they want to use it? What do no, no, they, they wanna buy it. And I’m like, okay, well
let’s do a due diligence on Franklin Templeton. They’re massive as you know, right? I think
trillion and a half in assets under management. And we were really having great results as you
know, with Canvas on our own. We thought about it for a long time and you know, we really
wanted custom indexing to be a new category of asset management.
00:34:21 And we felt really proud about that because it isn’t too often that you’re able to invent
kind of a new category, right? Of investing. And as we chatted about it and talked it out, we’re
like, you know, we’re at an inflection point here. We are a relatively small boutique, even though
this is working really, really well. If we want custom indexing, custom portfolio creation to
really make the big time, it probably makes sense for a much larger asset manager with all sorts
of advantages that we did not have to, to take it and run with it. So we let that be our guide. And
after doing quite a bit of due diligence on the people at Franklin, we were like, okay, let’s
negotiate about selling the firm to them.
00:35:13 [Speaker Changed] Talk about good timing. Morgan Stanley bought one of your
competitors in that space. Vanguard rolled out their own product, which quickly amassed, you
know, billions and billions of dollars on it. So this has worked its way into the mainstream, eventhough it’s still relatively, I, I don’t wanna call it a niche product ’cause it’s bigger than that, but
it’s not ETFs, it’s not giant yet, but it’s still growing at a pretty rapid clip, isn’t it? Totally.
00:35:46 [Speaker Changed] And and I think that ultimately we might look back 10 years from
now and, and have the thought, can you imagine that people just bought packaged products,
right? I mean, like, my God, no tax advantage, none of the customization, none of the
immunization for concentrated positions that I have. And so we definitely think that this is a way
of investing that, well, you know, once a client sees their portfolio under Canvas and with the
customization, it’s really, really hard to go back to thinking, ah, you know what? I think I’ll just
go with five mutual funds or five ETFs. I don’t really care about much of the other. I think that,
you know, these things take time, but I mean, again, your, your firm is a classic example here.
You were able to use custom in a way that was good for your firm, good for your clients. Right.
And you know, the clients that we speak with, love it.
00:36:54 [Speaker Changed] Yeah, no, they all love it. Our, our, that’s been our experience. It’s
really Mark Andreessen’s software is eating the world. Yep. Writ large. Because there, there are
two aspects to this, and I’m gonna circle back to the database part of it in a bit. But the front end,
the user interface and the software that allows a very simple set of choices and that you could,
you know, go increasingly down the rabbit hole and find more and more and more issues
certainly is a big factor. A lot of what is done, the technology just wasn’t quite mature enough 15,
20 years beforehand. And when you look at it, it’s just, well this is just software. It’s just a user
interface and a way of organizing it. But now let’s circle back to the database, which I recall you
saying was the secret sauce. Tell us a little bit about the database that you’ve been working on for
a quarter century that drives Canvas.
00:37:57 [Speaker Changed] So we use the Comstat universe, they cover virtually every
company that trades both here on American exchanges and elsewhere. And it, it is kind of the
gold standard really in terms of databases.
00:38:14 [Speaker Changed] How does it compare to something like CRISPR or some of the
other?
00:38:18 [Speaker Changed] Well, so CRISP comes to us from the University of Chicago Center
for Research and Security pricing. The downside of CRISP is, it’s a first off I love Crisp. We
used it in the most recent edition of what works, but it doesn’t have enough of the fundamental
factors attached to it. In other words, it’s mostly price history, price history. And it also tries and
generally succeeds to include all of the names that might have been around trading on the Amex
or the New York Stock Exchange or nasdaq. But the challenge is, a guy by the name of
Macquarie wrote a really compelling paper talking about how a lot of the historical data, not
Compus stat, but further back, right in the twenties and thirties, used to come from the papers
Wall Street Journal. Yeah. And, and also wasn’t nearly as thorough as say the Comstat is. In fact,
one of the things that we were doing before Franklin Templeton approached us is we were
literally digitizing old Moody’s manuals.
00:39:26 Huh? They go back to 1900. And what we wanted to do was marry into the crisp data,
all of the fundamental factors that would’ve given us the ability to run a 1900 through 1955.When CompStat begins test, we, we ran some test runs, we did price to book and we did a couple
others. And what we were finding and won’t surprise you, generally speaking, same kind of
results, right? With the exception price to book. We actually took price to book out of our
composites, you know, how we have the composites for value and momentum and all of those
things. And we took price to book out because of the research that we did that covered the Great
Depression in the thirties. You know, and I know if you’ve taken any finance courses, price to
book previously had been used as a proxy for likelihood of bankruptcy.
00:40:21 Right. Well guess what? During the thirties, a lot of those low price to book companies
went bankrupt. Well, when your book value collapses. Exactly. It’s the book isn’t much value.
Right. Exactly. Exactly. So we did find some learnings where we jiggered with the composites
that we use. That’s another thing we do. We don’t use a single factor. And my first version of
what works on Wall Street, we would sort down for the final portfolio on a single factor. And we
found that that wasn’t nearly as effective as a composite of factors. Again, a lot of people, the old
joke about quants, right? What do you guys do golf all day? You know, you’re just running your
models. Well, we don’t golf all day, but what we do do all day is research the underlying models.
What we’re always trying to do is improve them, but it’s evolutionary not revolutionary.
00:41:19 Listen, the foundations are very, very similar by the way. They make a lot of sense too.
I used to say if we changed it and, and walked out onto Lexington Avenue here and we found a
food truck, right? And we went up and long line, everything looks good. And we talked to the
owner and we said, how much you, how much are you clearing a year? And he says, well, I’m
clearing a hundred thousand. And we’re like, well, would you take a buy offer from us? And he
goes, yeah, you can buy it for 10 million. You and I are gonna go get outta here. There’s no way
we’re gonna buy this. Right? Well change it to a stock ticker. There’s a lot of stocks trading right
at that kind of multiple. And so when you look at the underlying strategies, they make intuitive
economic sense.
00:42:06 And so the data set that you’re using becomes of paramount importance. The other
thing I found was that, and this one disturbed me a little, I I haven’t looked at this recently, but
when I was doing it several years ago, you could get really different numbers if you went to
Bloomberg or if you went to Reuters or if you went to Dow Jones or any other innumerable
providers of data. And so that was another huge project for us. And also part of the data set that
we’re talking about. One of the other things that I was widely hated for by my research team was
we went on a multi-year data cleansing exercise because we found that a lot of it had a lot of hair
on it. And so I made no friends on the research desk when I said, listen, we’ve got to get this
pristine. And so our data cleansing of the universe also is another real important distinction
between just generally available data and that which we are using. Huh.
00:43:14 [Speaker Changed] Really, really interesting. Let’s stay with price to book. ’cause I
wanna ask your opinion on something and you’re the perfect quant to bring this up to, which is,
all right, so we’re, we’re talking about price to book back in the day when manufacturing required
a lot of men and material and, and capital and you had big factories and railroads were laying
thousands of miles of steel and you know, you were building these forges and foundries to make
cars. The modern era, especially with technology, there are a lot of intangibles that don’t seem tofind their way to book value. Things like patents and copyrights and algorithms and processes
that are proprietary that really are the whole value of the company, but somehow never show up
in, in metrics like price to book, which has led to some people, and, and, and I’m not positive
who to name, I don’t wanna mischaracterize anybody, but some folks have said, we’re mispricing
companies that operate in the tech space ’cause we’re not giving them the appropriate credit for
all of this intellectual property. Is that an overstatement or, or is there some truth there?
00:44:32 [Speaker Changed] I think there’s more than some truth to that. We published a paper,
it’s called the Veiled Value, and it looked at the idea that brand value, that all of the items that
you just delineated were not being captured in
00:44:48 [Speaker Changed] Trademarks, logos, all
00:44:49 [Speaker Changed] Of those straight across the board, research and development
straight across the board. When we took a look at that, we found that you could figure out a way
to price that into the model. So you are absolutely right. This is one of my bugaboos things like
GDP, all of the metrics that we continue to report and get obsessed about, basically they’ve lost a
lot of their meaning because they were designed for the world you just articulated, right? They
were designed for manufacturing, they were designed for physical things. And we moved off that
for many, many decades. Now,
00:45:27 [Speaker Changed] From Adams to Bits was a big transition,
00:45:29 [Speaker Changed] Huge transition. And so we think that we, another aspect of
research, right? When when we got the idea, you know, we think we’re missing something here.
That’s what resulted in the paper about brand value and goodwill and all those things not being
taken into account by investors at all. And so we found ways we could do that with factors and
improved the efficacy of the underlying models significantly. I
00:46:00 [Speaker Changed] Think one of the greatest quotes ever issued by a statistics professor
is George Box. All models are wrong, but some are
00:46:09 [Speaker Changed] Useful. Exactly. I quote him all the time because he’s absolutely
right. The idea that you, you’re gonna get anything to perfection is a fool’s errand. Right? I I have
a writer that we’re working with under O’Shaughnessy Ventures, one of our new verticals, which
is Infinite Books, and he’s got a great quote, which is, perfection is a 100% tax.
00:46:34 [Speaker Changed] Really interesting. Let’s talk a little about O’Shaughnessy Ventures,
starting with your mission statement. OSVs mission is to fuel creators in the worlds of art,
science and technology with the advice, data and resources they need to stay focused and get
great ideas out of their heads, off of their whiteboards and out into the world. Discuss.
00:47:01 [Speaker Changed] I had a thesis that started to develop around 2017, 2018 as I
watched old playbooks that used to work beautifully stop working. And so I came up with this
idea that we were in a great reshuffle where all of the old models were collapsing and people
were kind of freaked out. They were like, this has worked for decades, why doesn’t it work
anymore? And I think that one of the reasons it didn’t work anymore was because the tools, thetech tools and the platforms and the internet and all of that put together allowed for much more
innovative business models in a variety of industries, right? So if you look at the verticals of
O’Shaughnessy Ventures, you’ll see what we think, right? So we have what we call infinite
adventures, that’s venture capital. But I love, in the old days they used to call venture capital.
Adventure capital, right? And the one I really loved, liberation Capital,
00:48:11 [Speaker Changed] Which I thought to find that what is, what is liberation? And I’ve
heard the phrase
00:48:15 [Speaker Changed] Yeah. In the old days, the so-called Hateful eight that wanted to
leave Shockley. Right, right.
00:48:20 [Speaker Changed] The early days of semiconductors. Yeah. And the the pre Fairchild
semiconductors.
00:48:25 [Speaker Changed] Exactly. Exactly right. Good call. And, and back then, the idea that
a group of engineers, or even, you know, regular business people would leave a big company
that was well funded by a bank or a series of other investors was almost unthinkable. And so
what came to be known as the Hateful Eight who created Fairchild got pitched by a variety of
investors, external investors saying, why don’t you guys just start your own company? He finally
talked them into it. And that’s when he used the term, this is your liberation capital where you
can focus on just what you wanna focus on making better semiconductors. You don’t have to
play any of the politics of the big company. You don’t have to answer to people who don’t really
understand what you’re doing. Right. The people in New York that might have owned it or
financed it, had very little understanding of what semiconductors were all about in the fifties and
sixties. And so I like that part very, very much.
00:49:32 [Speaker Changed] That’s the genesis of Intel, right? Yeah. Of as well as a, a whole run
of other semiconductors can trace its roots back to Fairchild, right?
00:49:42 [Speaker Changed] E Exactly. And so there we’re looking for companies that we think
will expand the opportunity set for very clever entrepreneurs and creators. Another vertical is
infinite Films. Why that? Well, we think we’re approaching a period where you can make films,
documentaries. You can use AI to augment your filmmaking in such a way that the people who
couldn’t make movies in the past are gonna be able to make them in the future. You
00:50:16 [Speaker Changed] Could legitimately make a film with an iPhone now. Yes, you can.
That couldn’t, you couldn’t do even five years ago is kind of on the border.
00:50:24 [Speaker Changed] Barry, some of the things that I’ve seen as submissions to infinite
films, oh my God, really? Like, literally I’m 63. If, if I had seen that as a trailer for a movie at a
movie theater like 10 years ago, I would’ve thought, wow, this is amazing. This is cool. And then
the guy at the bottom says, by the way, I made this on my iPhone. That’s
00:50:49 [Speaker Changed] Crazy. That really is
00:50:50 [Speaker Changed] Crazy. And and so that unlocks tremendous talent that never had
access to the Hollywood infrastructure. So our thesis is there are tons of really creative peopleout there who now have the tools to make great movies. Another thing I wanted to do was, where
are the Rudy’s of movies today? Now Rudy’s of course, is about the kid who goes to Notre Dame
and he’s five foot nothing and weighs a buck, nothing. And he gets on the team, the Notre Dame
team. Why was that such a great movie? Because it’s incredibly inspirational. It gives the viewer
like, you know what? I can take a shot at it, I can do it. Hollywood seems to have completely
forgotten about making these types of movies. And,
00:51:38 [Speaker Changed] And just for people who might not remember the movie, Rudy, it’s
the story that drives the whole thing. And, and the characters. There’s not a whole lot of
expensive special effects or, you know, they, they’re not flying out to Nepal. It’s all done pretty
much on the cheap. And, and that’s the area of film you’re looking to explore. Narrative driven,
accessible stories,
00:52:03 [Speaker Changed] Narrative driven, accessible stories that we’re also changing the
underlying economics on. So here’s how we’re gonna do that. Everyone who comes and works on
one of our films is gonna own a piece of that film.
00:52:20 [Speaker Changed] Backend points.
00:52:22 [Speaker Changed] Backend points. But for everybody, we’re not gonna use Hollywood
accounting. Our accounting is very, very straightforward. Here’s what it costs us to make it.
What happens after we recover those costs? You own X percent. If we manage to sell it or
generate revenue from it through the multiple platforms you can put it out on, you are going to
benefit from that. The other thing that we’re gonna do is we’re gonna give young people a shot.
Right now, if you wanna try to beat, let’s say you graduate from NYU film school and you
decide you’re gonna go out to Hollywood and you’re gonna pitch all of these studios. Good luck
that you wanna luck. Yeah, good luck. Because it ain’t gonna happen, right? There is almost a
guild like system out in Hollywood where, you know, you, it’s, it’s kind of the idea that, yeah, I
wanna get in the Screen Actor’s Guild, how do I do that? Well, to get in the Screen Actor’s Guild,
you have to be in three movies. Well, wait a minute, how do I get in the movie if I’m not in the
Screen Actor’s Guild? So there are a lot of really old fashioned rules. And it’s not just Hollywood
by the way, it’s much of media. It’s much of all of the things that we consume every day. And so
basically what I did was say, what industries that I find fascinating that I’m interested in have the
greatest arbitrage ability. Huh.
00:53:48 [Speaker Changed] I I love that concept. And you know, it’s funny you mentioned films
because that dynamic tension of indie films. Look, look at how great a 24 has been doing
amazing, a, a as a, as an independent studio. The timing is really good. And the technology tools,
the ability to film on a phone edit on your laptop, and then distribute it by uploading to YouTube
or wherever,
00:54:16 [Speaker Changed] Barry, that’s the key. There’s always cultural lag, right? You’ve, you
know, the s-curve for tech adoption, right? It’s real. And let’s change industries and let’s look at
publishing, right? So we are launching Infinite Books. Why? Well, because the current
publishing industry is still playing under 1920 rules. Not 2020 rules. We no longer have to have
minuscule amounts going to the author. We can, because of the tech, because of our ability toproduce that book, give the author much more of the upside. So for example, we’re gonna give
anywhere between, depending on what the author wants us to do for them, it’s gonna always be
above 50%. Mostly it’s gonna be 70%. But that’s just the start. Imagine Barry, you write a book,
you bring it to Infinite Books, and I say, Hey Barry, what other languages do you want this
published in? And you’re like, I don’t know, maybe Spanish, maybe French maybe done because
of ai we can translate the entire book and have it available for the French or Spanish speaking
markets. Even better, let’s say you wanna do an audio book and you wanna read it ’cause you’ve
got a great voice. I say, Barry, do a minute on this for me, say express, surprise or anger or
whatever. It will model your voice and you can read your book on all the audio books. But
what’s really cool is we can translate your voice into French, into Spanish, into Russian, into
anything. Wow. And so all of these tech advantages are being left just lying around on the floor,
right? And we think that’s crazy. We’re
00:56:11 [Speaker Changed] Still early days of the transition. Oh,
00:56:14 [Speaker Changed] Very early
00:56:15 [Speaker Changed] To technology, to ai, to all these changes in platforms. It’s amazing
how slowly it takes place. I, I think our, our mutual friend Morgan Housel described how long it
took from the Wright Brothers doing the first test flight in Kitty Hawk before it even made its
way into newspapers.
00:56:38 [Speaker Changed] Exactly. Takes forever. And it does. And this leg, even in our 24 7
always online environment remains, right? It like, if you think about it, it makes tons of sense.
People are habitual, right? They, they get into habits, they do all of these things. Now, I think
that the pandemic really sped up a lot of these trends. Things like work from anywhere.
O’Shaughnessy Ventures is a work from anywhere enterprise. We have people in Singapore,
India, uk, all over the world because we can, and the idea that we have to have a traditional
office, the idea that we have to do any of those traditional things goes right out the window. It
becomes a much less costly enterprise when you can do it this way. But we back to infinite
books, like we also are going to at the author’s decision, right? We’re not gonna force anything
on our authors.
00:57:44 But if the author wants an AI agent to, let’s say for example, your new book, let’s say if
it were an Infinite Books publication and you note noted that it quadrupled sales in Omaha,
Nebraska, how about having an AI agent find out what podcasts in Omaha are interested in The
subject Barry’s written about, how about sending them a query letter? How about sending them a
clip from the book and saying, you really ought to have him on your show or podcast, or write
about him in your substack. All of the tools that are available to us work today and people aren’t
using them. And so we suspect that this is going to really, I hate the word revolutionize because
that’s, you know, come on. But it’s,
00:58:34 [Speaker Changed] It’s certainly gonna accelerate, accelerate
00:58:37 [Speaker Changed] Train. That’s a, that’s a better00:58:38 [Speaker Changed] Word for it. Right? So, so I wanna talk about another aspect of
Osuna Sea Ventures, which is the fellowship program, which I find to be absolutely fascinating.
How does this work? Tell us a little bit about the Nessy Fellowship
00:58:53 [Speaker Changed] For, for most of history, a genius could be born, live, and died
without even knowing they were a genius. Right? Far less other people knowing it. Right? We
were really bound by our geography and by our networks. And those networks were pretty small.
Like, who’d you grow up with? Who’d you go to school with? Who’d you marry? Where are your
kids going to school? What church do you go to? That kind of stuff.
00:59:17 [Speaker Changed] Pretty random. Pretty random. Where you were born was just dumb
00:59:20 [Speaker Changed] Luck was kind of dumb luck. You could move of course, but
changing your digital zip code is a hell of a lot easier than changing your physical zip code. But
more importantly, we now are interconnected. I can find somebody who’s a genius who happens
to live in Bangladesh. I would’ve never under the old system ever known about that person. Now
I have the ability to know about that person and find and fund them. The whole idea behind the
fellowships was we wanted to come up with something that highlighted the fact that there are
tons, millions of brilliant people who in the past just didn’t have the right connections, didn’t
have the right credentials, you name it, to get into a place where they could get funding, they
could make their idea come to life. And so the idea is quite simple. We’re gonna find and fund
them and see what comes from that. I think that it allows for so many things. Like it allows, we
have a guy who got one of our grants, which is the smaller amount. It’s 10,000, the fellowships
are a hundred thousand over a year. No strings, no
01:00:35 [Speaker Changed] Strings attached. Here’s a check for a hundred k, go do something
interesting. We don’t care
01:00:39 [Speaker Changed] What it’s exactly. And we wanted to do no strings because like, we
don’t want gotchas, we don’t want, but you’ve gotta do, you gotta give us right of first refusal.
The, the way I look at it is if, if we got somebody so wrong that they’re gonna take a hundred
thousand fellowship from us, develop something really cool, decide to start a company around it
and then take it to a different person for funding. Well, we made the mistake. Right? Right.
Because generally speaking, what we’re finding is they love being part of the community.
Because I’m also a huge believer in cognitive diversity, right? There’s a great quote that is like,
no matter how smart somebody is, no matter how insightful, no matter how brilliant, you can’t
ask them to make a list of things that would never occur to them. Right? And so essentially what
happens when you get all of these really bright people in our fellowship and grant community
communicating with each other, wow. The ideas that come out of those cross pollinization of
ideas are really extraordinary. So, but this, this sounds
01:01:51 [Speaker Changed] Like this is really an incubator of sorts.
01:01:53 [Speaker Changed] It can be, but it needn’t be, here’s a great example. One of the guys
that we gave a grant to, his name’s just, that’s his staged name, was an accountant in India who
decided he really had music in him. And he really wanted to do a musical video using traditionalIndian songs and singing in Hindi and other Indian dialects. He went super viral, tens of millions
of downloads of his song. He’s being put on all of their Good Morning India. You know, we have
Good Morning America being written about in all of their newspapers. And essentially that was
because we thought, wow, this guy’s got talent. Let’s see what happens. We’re not incubating him
for anything, right? If he goes off and signs a deal with a music company, we don’t do music. So
God bless.
01:02:50 [Speaker Changed] This sounds a little bit like the MacArthur Genius Awards, where
01:02:54 [Speaker Changed] Here’s a chunk of money, go be a genius. There’s just so much
potential around the world, Barry, that I feel compelled to amplify. Everybody loves to bag on
the generation before or after them, right? Listen, the kids today, young people today are digital
natives. They know how to use these tools in ways that we boomers probably are never gonna
get to. And I say, let’s empower them. Let’s demonstrate to the world that this makes real
practical sense right now. Let’s take somebody else who is turning his grant into a company. It’s a
guy in Africa who faced a problem I knew nothing about, which was the cost of sanitary napkins
For women who are menstruating is out of reach. They are all imported from the west and they
can’t buy them because they don’t have enough money. Well, he came up with an idea where his
mostly female staff and researchers use banana leaves and other biodegradable products that they
can make on the ground in Africa sell for a fraction of the cost that the imported ones work just
as well.
01:04:19 Now, I believe he is turning that into an enterprise. He’s founding a company. We’ll
take a look at investing in it because of course he’s asked us to. It can be on the business side,
definitely an incubator. But on the social side, on the music side, on the art side. So for example,
this year I really wanna have a fine artist get one of these grants because again, I want really
people to be able to see there is so much talent in the world and I always try to look for things to
root for as opposed to against. There’s So it’s so easy to root against something, right? You don’t
have to be terribly bright to say, that sucks. That sucks. Here’s why. How about doing things the
other way around? How about finding things you can root for? And then the results have been
kind of like the coolest things we’ve ever seen. Like the guy going viral in India, like we have,
we funded a guy trying to advance open source quantum computing. He now is a big deal in
quantum computing. Wow. And it’s a great thing to do in general. Tell us
01:05:31 [Speaker Changed] About some of the first few you tried. Who, who were the people
that were the first couple of recipients of
01:05:38 [Speaker Changed] The guy, the fellowship guy I just mentioned, right. WA with the
quantum computing. He had me at Hello. ’cause I love that stuff. What,
01:05:45 [Speaker Changed] What about people who are looking at markets and the economy? I
know that that’s a, a peeve of yours.
01:05:50 [Speaker Changed] Oh, absolutely. The thing there is, we wanted it to be significantly
different than our traditional quant. One of the reasons I became so interested in machine
learning and AI was I viewed that as the next frontier for quant. The dirty little secret of a of wequants is if, if you really press us and ask us to really explain your model like you would to a 5-
year-old, we’re using pretty much the same stuff, right? Yeah. So what we wanted to do there
was push the needle as far as we possibly could. But then one of the first people to get one of the
fellowships was a married couple, Nat and Martha Sharp. And what they wanted to do was make
a documentary about non-traditional schools for their kids. They have a bunch of young kids
below, you know, the age of seven. And they put out a great documentary about a particular
school, which was really novel.
01:06:55 And so we really are all over the map in the type of person or groups that were willing
to consider yet another was a refugee in Ireland who found that she couldn’t figure out a way in
her native language to work her way through the halls of the bureaucracy to figure out how do I
get a place to live? How do I do all of these things? So we funded her to make an app. And then
finally another one that I just love is we have a doctor who came to us and said what he wanted
to do was make an app for an iPhone or an Android where you could completely non-invasively.
I could point the phone at you, get your vitals on the phone just by the camera on the phone.
Really? Yeah. Wow. And what was cool for us was we really pushed him.
01:07:50 We’re like, why, why, why, why? And finally at the end of our interview with him, he
was near tears. And he went, the real reason for this is my dad died of a stroke and I was in
medical school and I didn’t save him. I didn’t even know that he had a problem. And so this is
why I am so passionate about this, to get a lifesaving thing in the hands of and on something that
we all carry with us, right. These smartphones is what motivated him. And on top of that looks
like it could also be a great business.
01:08:28 [Speaker Changed] Wow. That’s, that’s really interesting. Let, let’s stay with AI and talk
about medicine in particular. I’m fascinated by the concept of AI running through billions or
even trillions of molecular combinations to identify promising drugs, some of which are already
out there, some of which haven’t been created. But it really gives us the ability to take millennia
worth of experimentation and do it in a really very short period of
01:09:00 [Speaker Changed] Time. It’s a world changer. The ability to, as you mentioned, take
different molecules where there isn’t a drug addressing a certain problem. And or taking existing
research from drugs and repurposing it. AI can go into all of those spaces that we humans simply
can’t do and find the connections on an existing drug. You know what this drug was originally
done for malaria. Well, it doesn’t work for malaria, but it works really well for this disease over
here. And then new drugs that the discovery is going to be amazing. And you gotta remember, a
lot of this stuff can be done what they call in silico. You don’t have to test it on humans or
animals. You can test it on the clone of we humans that you set up in the computer. Hmm. And,
and so these types of things, like, I honestly don’t think it’s an overstatement to say like this, this
AI and its many use cases belong up there with the wheel and fire and the printing press because
it is a multi-use technology that’s going to affect everything from drug discovery to financial
analysis.
01:10:25 What about, we had train an AI to generate nothing but null sets, right? Like if you’re a
medical researcher and you’re trying to get funding, what do you wanna do? You wanna provesomething new, right? You don’t, you’re not gonna get funded to prove, you know that aspirin
works, but you wanna find something new and you also want it to be a positive finding. So what
happens is the incentives preclude a lot of brilliant scientists from looking for things that don’t
work and yet, like the dog that didn’t bark in Sherlock homes, right? There’s a lot of really cool
information. Useful information via negativity. And so one of the things that we wanna do is just
have a large language model, churn out hypothesis after hypothesis that is gonna generate an null
set, publish them to a database that all scientists can have access to because there’s a wealth of
information in the stuff that doesn’t
01:11:30 [Speaker Changed] Work. Here are things you don’t wanna waste your time on.
01:11:32 [Speaker Changed] Exactly.
01:11:33 [Speaker Changed] Let, let’s talk a bit about stability. ai. You’re on the board of
directors, you’re the executive chair, and you started back in September, 2022. Pretty, pretty
good timing. Tell us a little bit about what stability AI does and how does this relate to the rest of
Nessy Ventures?
01:11:50 [Speaker Changed] So stability, AI builds foundational open source models. I had a
very pointed point of view that with a technology this powerful, I did not want it controlled by a
panopticon controlled by a few. And I saw that with that kind of power could come some pretty
negative externalities. And so stability AI was the one that really caught my eye because they
really were the ones who shot the gun back in the summer of August of 22. They released a
stable diffusion model, which generates images, right? But they did something that no one had
done before. They released that model with all of its weights. Now, not to get too geeky here, but
the only way people can build on that type of model is to know what the weights are. And so
what they did was show it all. They released the whole thing, full
01:12:59 [Speaker Changed] Open source, fully transparent,
01:13:00 [Speaker Changed] Open source, fully transparent, and bury the Cambrian like
explosion of creativity. That happened almost immediately, really proved to me. Yeah. Back to
cognitive diversity, right? When you allow all of these clever people, the ability to play with it,
to tinker it with it, you get a much better model. For example, that’s why Linux runs the web.
Linux is open source, right? And it does so because a bunch of different people work on different
problems. And so my point of view was I’m all for the open, I use open ai, I use all of the
commercial
01:13:43 [Speaker Changed] Large. What, what are some of the commercial apps you
01:13:46 [Speaker Changed] Work with? So, so perplexity,
01:13:48 [Speaker Changed] I love perplexity. It’s on my phone. It’s really, really useful.
01:13:51 [Speaker Changed] Open ai. I’m looking at Claude, the new Claude
01:13:55 [Speaker Changed] That you knows can be driven by either Claude or, or there’s like
four different engines that drive it. Exactly. Which is, it’s01:14:03 [Speaker Changed] Really
01:14:03 [Speaker Changed] Interesting. Which
01:14:03 [Speaker Changed] Is one, one of the things I love about Yeah. Perplexity.
01:14:05 [Speaker Changed] It, it’s just a great, and it’s cheap and it’s so useful. Exactly. Every
interview I do, I, I don’t start with perplexity. I finish with perplexity. Yep. And what did I miss?
What did I get wrong? Although you still have to be careful ’cause every now and then, like
O’Shaughnessy is not the rarest of names. You know, I had Bill Dudley, former New York Fed
chair and I learned that he was a running back in the NFL in the forties, which is kind of
interesting ’cause he wasn’t born till the fifties. But every now and then something will pop up.
That is a little off. I, I love the phrase hallucination for that. What else do you use besides
perplexity and chat? GBT
01:14:50 [Speaker Changed] Assume, well obviously stability, ais various models
01:14:54 [Speaker Changed] And are they available, are they accessible to the lay person? Like
that’s the beauty of perplexity?
01:15:00 [Speaker Changed] They they are, but through different APIs we really wanted to focus
on being the builder, right? So we did not want to try to compete in the direct to consumer space.
And so what we’re focusing on is multimodals, including generative models, including specific
models for medical research. Obviously generative art models, movie models, et cetera. The
thing I wanted to mention when you were talking about perplexity and it coming up with, I also
passionately believe that the models that are gonna wi win or not the models, the approach that’s
gonna win is human plus machine. The so-called Sansar model. I think that you’re gonna see,
you know, we’re gonna see a deluge of AI only generated stuff, content, movies, et cetera. And
to be honest, most of it’s gonna suck. Right? Right, right. The magic comes when you add a
human in the loop. The magic comes by being able to partner with that and co-create and
sometimes iterate on your own stuff.
01:16:15 Right? And like you said, the ideas that you can generate through putting your own
stuff into the various models is really cool. We invest in a startup called Wand, and what they do
is it’s for graphic artists and it’s an ai, but it has an actual tool, thus the name Wand. And what the
artist is able to do is feed their own work into the model and then ask it, Hey, spit out variations
on it. And then the artist will look at it and say, wow, I never thought about it that way. That’s
really cool. And then he or she will iterate, iterate, send it back. And this is an iterative process,
but what’s really cool is they end up in places. We had one artist say to me, I would never have
thought to do it this way, but I absolutely love it. It’s his work. He’s iterating on his own work,
but he’s using a tool, the wand that makes it infinitely easier for him to get these great ideas.
01:17:21 [Speaker Changed] Huh, really interesting. Last question before we jump to our
favorites. We ask all our guests, which is, I wanna bring this back to stocks. I know thanks to
Perplexity as an example, but there are lots of other tools. I find myself going to Google a whole
lot less than I used to. And in fact, the Google search results, like suddenly you realize these are
crude, they’re much less useful than they used to be. They’re feto with a lot of advertising and alot of Google internal products dominate that first page. What else is ai? What other companies,
what other sectors might AI affect either positively or negatively?
01:18:12 [Speaker Changed] Well, honestly, how much time do you have? It’s, I I think that AI
is going to transform virtually every industry. And one of the things that people, they get afraid
when they hear that. And, and my view is quite different. It’s, it’s going to transform for a lot of
industries. The pure drudge work, the pure copy and paste stuff. Why do you want, do you like
copying and pasting? I hate it. And so it also is going to be able to create jobs that we can’t even
conceive of right now. Right? Like two years ago, would you have known what a prompt
engineer was? No, I certainly wouldn’t have. Right. And yet there’s a lot of people doing really
well pursuing that as a career. And so I think that entertainment is going to be materially affected
media, materially affected search as you well point out. Like you can do a customized search just
for Barry and it, you know, depending on how much information you wanna give that AI about
yourself, you’re gonna be at a place where you’re gonna be able to say, Hey, what was that place
that I had lunch with Jim last time? We both really, really liked it. I would like to go there again,
and guess what? It’s gonna give you the name and address of that restaurant because it has access
to your calendar, it has access to all of that type of stuff. It,
01:19:39 [Speaker Changed] It feels like, I’ll never forget, I, I tweeted out this really interesting
Roman pizza place and Roman Pizza is a different type of, and, and I just, you know, I I you Siri
to speaking to the iPhone, Hey, we had a fend. This is really different than your usual pizza. And
somehow it showed up on Twitter as woman pizza and like, wait, I’m standing right in front of,
of the place. Any correlation between my, my geotag and business I’m in front of, it just felt like
technology should have figured that out. Yeah. What you’re saying is that sort of access to your
contacts, access to your, where you are, access to your calendar once there’s an intelligent agent
running all of that, a lot of these sort of silly, why can’t Siri talk with this person? Why can’t
Alexa? It just seems like the pre AI era was filled with a lot of pretty dumb ai. It’s starting to get
smarter.
01:20:46 [Speaker Changed] Yeah. And and that’s the thing, going back to your Wright brothers
example, you know, when the Wright brothers did that very brief flight, it was only a matter of
eight seconds, something like that. Yeah. I think it was 12 seconds. Right? And I think they went
like a hundred and odd feet. Like you could see why a lot of people would go, eh, hey, they
didn’t accomplish much, but I like the person who was watching and said, this changes
everything. Right? And so that’s kind of how I see ai. Of course we’re in the early innings of this,
and of course it’s going to, this is the worst you’re ever gonna see it, right? It’s going to improve,
improve, improve. But the other thing I wanna really underline here is it’s the quality of the data
that you train your AI on that determines its value to you.
01:21:36 And one of the big reasons I’m a huge believer in private ais is that you will feel if you
know that no one else can have access to that, right? You’re gonna give it a lot more access to
things than you might otherwise. That’s happening right now. Wow. And so one of the things,
you know, a lot of people see this as, you know, like the, the, the great model that will figure
everything out. I don’t see it that way at all. I see it as a lot of smaller but incredibly useful AI
agents doing specific things for each of us. Again, canvas fits in beautifully here we are now inan era of mass customization. We are in an era where it’s going to be able to design it just for you
and your likes and dislikes. That that’s really profound when you think about it. Really
01:22:34 [Speaker Changed] Fascinating. So let’s jump to our speed round. Our favorite
questions we ask all of our guests, starting with what has been keeping you entertained these
days? What are you either watching or listening to?
01:22:46 [Speaker Changed] So we rewatched true detective, my wife and I, I would highly
recommend rewatching the first season of that. It was brilliant. It led us into a rewatch of the
entire series. And, and now we’re on number three. The second one, here’s one of the funny
things like in memory, I kind of, my wife and I were both kinda like, yeah, that second one
wasn’t very good. It was good. And so we’re doing that Masters of the Air that’s on Apple tv just
01:23:18 [Speaker Changed] Started on Apple. Yeah, it looks great.
01:23:19 [Speaker Changed] Really loving that. I loved Band of Brothers. So we’re, we’re both
really, really liking that. And then we are also watching a series, or I guess I should say
rewatching a series which kind of kicked off the idea of the golden age of television. It was one
of the earlier ones. I’m not the Sopranos, but The Wire.
01:23:45 [Speaker Changed] Now I recall The Wire being very brutal and difficult to
01:23:49 [Speaker Changed] Watch. It’s, it is, but what’s so cool if you choose to watch it again,
you see that the reason it kicked off that kind of TV was because it was brutally honest about
things. It wasn’t trying to lie to you about anything. And the characters are incredibly complex,
even though even the evil guys are incredibly complex. And, and so watching it now from the
vantage point of like 20 years or more, it’s really amazing.
01:24:26 [Speaker Changed] Huh. Really interesting. Tell us about your mentors who helped to
shape your career.
01:24:32 [Speaker Changed] Primarily. I, I would list my grandfather. I was lucky enough, he
was very successful in the oil industry. And I am the youngest of the third generation, at least the
males. I have one younger female cousin and she’s just a few months behind me. But I lived in
the same town my grandfather did. And after my grandmother died, he would come to our house
twice a week for dinner. And literally, I would literally sit at his knee and he was a wonderful
storyteller. He was a wonderful teacher. And he taught me this idea of predating that I have
written a lot about and use all the time. Another was a wonderful man, not related to me at all by
the name of Jim Myers, any entrepreneur. You hit some rough spots. Sure. And I had hit a really
rough spot and was basically broke and trying to pay for a house because we’d moved to
Greenwich and keep my business afloat and all of that.
01:25:40 And the banks are like, dude, like you, you’re an entrepreneur. This is back in the
nineties. Yeah, sorry, we’re not gonna give you a a mortgage. He stepped in and he’s like, Jim, I
believe you’re gonna be tremendously successful. And gave me one on a handshake. Wow.
Which I was able to repay rapidly. But more than that, just being a super high quality man. He
taught me more about real business than any textbook. And ’cause I was young. Right. And Istarted with him when I was in my early twenties. Wow. And just a, just an amazing man. And
then finally the, the other mentors that I would say are like the greatest minds of history. I love to
read. I particularly like to read biographies about people I admire. And you know what, Barry
life was not easy. We remember them now, right? Like, oh, they were this huge success. When
you read their biographies, you see they went through a lot of muck to get where they got. And
so kind of universal lessons
01:26:49 [Speaker Changed] There. So perfect segue. Let, let’s talk about some of your favorite
books and what are you reading right now?
01:26:55 [Speaker Changed] So right now I am reading about four different books. And I, I,
which
01:27:03 [Speaker Changed] By the way is an occupational hazard for folks like us. Yeah.
Because there’s always a book I’m prepping for a podcast. There’s a book I’m reading for work
and then there’s a book. I’m just like, I’m gonna relax and read this. Yeah.
01:27:15 [Speaker Changed] So for fun, right now I’m reading Burn Book by Kara, what’s her
last name? Swisher. Swisher. Which I find very interesting.
01:27:25 [Speaker Changed] She’s always
01:27:26 [Speaker Changed] Fascinating. Yeah. Kind of an inside look. My only comment there
was she, she might be a little guilty of the things that she accuses, the people she doesn’t like.
Sure. But other than that, it’s a fun and kind of a rollicking read. I am reading or rereading
several of the books from Wild Durant’s Story of Civilization, which I read as a kid, a young
man loved and thought, you know what, we moved recently. And so I was going through all my
books and I found that and I’m like, I should reread some of these just to see if it still stands up.
Barry, it’s still great
01:28:05 [Speaker Changed] Stuff. Right.
01:28:06 [Speaker Changed] Really, really stands up. And then just finished a, an additional
biography about Teddy Roosevelt, Teddy Rex. And then finally I am reading a lot about AI and
scientific development. The book I’d recommend there is written by a pair of authors. One, an AI
expert, the other, a great storyteller. And it’s called AI 2041. 10 Visions of our AI future. Huh.
Highly recommend.
01:28:38 [Speaker Changed] I’m gonna check that out. We, we’ve been talking about the Wright
Brothers, did you ever read the David McCullough biography of the Wright Brothers? I did.
Fascinating. Right. Really, really, really fascinating. And our final two questions. What sort of
advice would you give to a recent college graduate interested in a career in either quantitative
analysis, finance, asset management? What’s your advice for them?
01:29:02 [Speaker Changed] My advice is to focus on the parts of learning that might not be
included in a business or finance degree. My line is that markets change second by second, but
human nature barely budges. Millennia by millennia arbitraging, human nature is the last
sustainable edge in investing. And so if you read about evolutionary psychology and biology,regular psychology and biology and history, what you’re gonna see is no history doesn’t repeat,
but it rhymes. And you can see in, you know, all you gotta do is go read a book about the South
Sea scandal where Isaac Newton, one of the most brilliant guys of his era, lost a fortune causing
him to lament that he could measure the motion of heavenly bodies, but not the madness of men.
And guess what? We’re not changing. So you can read it in a market related way or just
understand human nature better. You’re gonna be miles ahead of the people who are just
studying math or finance or economics.
01:30:14 [Speaker Changed] Hmm. Really interesting. And our final question, what do you
know about the world of investing today? You wish you knew 40 or so years ago when you were
first getting started?
01:30:26 [Speaker Changed] I think maybe just the advice that I just gave. I wish that I would’ve
known 40 years ago that markets are, market prices are determined by human beings. And if you
are ignorant of all of the ways that we let things affect us from whether we’re hungry or not, or
whether we’re angry or whether we’re calm, I would’ve understood that it was not just numbers
on a page that markets are full-blooded, almost human-like things because they’re driven and
created by humans. If, if I could have told Jim of age 23 that it would’ve hastened, but also
improved the pretty circuitous path that I took to becoming a quant.
01:31:20 [Speaker Changed] Really interesting. Thank you, Jim, for being so generous with your
time. We have been speaking with Jim O’Shaughnessy, founder of OS A M Asset Management,
and currently CEO and founder of O’Shaughnessy Ventures and host of the Infinite Loops
podcast. If you enjoy this conversation, well be sure and check out any of the 500 previous
discussions we’ve had over the past 10 years. You can find those at iTunes, Spotify, YouTube,
wherever you find your favorite podcast. Be sure and sign up for my new podcast at the Money
where we speak with an expert and give you information on a topic relative to your money in
short, eight to 12 minute batches. You can find those in the Masters in Business podcast feed, or
wherever you get your favorite podcasts. I would be remiss if I did not thank the crack team that
helps us put these conversations together each week. My audio engineer is Sebastian Escobar.
My producer is Anna Luke. Sean Russo is my head of research. Atika Val Bru is my project
manager. Sage Bauman is the head of podcasts. I’m Barry Ltz. You’ve been listening to Masters
in Business on Bloomberg Radio.
~~~
