Odd Lots - ARK's Head of Research on How They Find the Next Huge Winner

Episode Date: February 11, 2021

In a world dominated by passive investing on one end and retail YOLO traders on the other, there aren't many star fund managers these days. There's one big exception though. Cathie Wood, the head of the ARK family of funds, has become a celebrity due to the incredible performance of her stock picks. So how do they do it? On this episode, we speak with Brett Winton, ARK's Head of Research, who explains the process they use to find disruptive technologies, and the companies that will win from them.See omnystudio.com/listener for privacy information.

Thanks for listening to Oddlots. Follow the show on Amazon Music for more future episodes or just ask Alexa, play the podcast, Oddlots on Amazon music. Hello and welcome to another episode of the Odd Lots podcast. I'm Joe Wisenthal. And I'm Tracy Allaway. So Tracy, here's something that I never thought I would see again. So I first started following markets in the late 90s, you know, dot com era. And something that I never thought I would see again. So I first started following markets in the late 90s in the late 90s, you know, dot com era. and something that I never thought I would see again in my career after that ended was the superstar fund manager. Okay. Why is that? Well, actually, that's not totally true. What I mean is more the superstar stock picker because, of course, back in the old days, there were a lot of like star stock pickers, fund managers, you know, Peter Lynch comes to mind some of the other tech investors back then. but these days with ETFs, with online brokerages that make it really easy for individuals to buy stocks on their own, it really sort of seemed to me like that era was just gone.

Right. So I suppose there was this idea that the time of stock picking has come and gone and that if you want to make good returns in the market, you should just pour all your money into something like an S&P 500 ETF, like a Vs, SACs or something. something like that and just stick with it. And don't bother trying to outperform the market because over a longer period of time, even the best stock pickers had eventually underperformed. Right. I think this mantra of don't try to pick stocks, A, if you try to pick stocks, you're probably going to underperform the index. And B, if you come across, say, a mutual fund or a fund manager

who's good at picking stocks, oh, it's probably just luck. It's not going to last to, you know, even if there is someone who can beat the market, how are you going to know whether it's actually worth putting your money with them? And so like this idea that everyone should just index that trying to beat the market is kind of a loser's proposition. It's really been drilled into people's heads. And I think like, you know, for years, they're really, we just haven't had a sort of another,

a new Peter Lynch or Buffett. You know, there's like Stark Quants maybe, some bond fund managers who are known. But the idea of like someone who is just really a. associated with a great track record of picking individual stocks, hasn't been a thing for a while. And yet, and yet a star stock picker emerges over the horizon. Yeah, exactly right.

So obviously, that really, for the first time in a long time, there is currently a fund manager, a stock picker, who has amassed an incredible track record, an incredible following. And of course, we're talking about Kathy Wood. She is the CEO and chief investment officer of Arc Invest. And there is this total fascination with Arc and this family of actively traded ETFs that have just done a phenomenally well in terms of returns, but also attracted an extraordinary amount of investor cash in the last couple of years.

Right. So the ARC ETFs, I mean, I'm looking at their performance. They have, you know, five different thematic portfolios alone that have basically doubled over the past year, which is pretty amazing. If you think about it, it's amazing enough for just one stock to double in price like that and in just the space of 12 months. But to do it across multiple ETFs is really remarkable. And I think within their actual portfolio, there's a tiny, tiny number of stocks that haven't risen recently. And I'm not even sure there are any, actually. It's a really amazing performance.

It's really extraordinary. Actually, I'm looking at at the end of 2020, for 2020, their performance of ARKKK, which is the sort of flagship innovation ETF that ARC has, was up 152% for the year. Extraordinary returns. And if you look at the holdings, they're just all of the companies that have absolutely killed it in the recent environment. Tesla is the biggest. one, but other names, Square, the payments company, phenomenal, Roku, huge winner, Zillow, Spotify, Teledoc, which, of course, had an incredible year, thanks to the rise of Romo

medicine and so forth. So it is a just extraordinary number of winners that this is a fund and the related funds, there's a related fund for finance and medicine that have, that they've brought Just the track record is incredible. If anyone follows Eric Belcunis, who's sort of Bloomberg Intelligence's ETF analyst, I feel like three quarters of his tweets these days are just about how extraordinary this family of funds and the performance of Ark Invest has been lately. Yeah, absolutely. And you mentioned Kathy Wood already, but it's sort of, it's given rise to a cult around her, I guess. I don't want to say cult because that has. negative connotations, but certainly there's been a lot of admiration and fascination with what

she's been doing over at Arc. Yeah. And we're recording this January 20th. I saw Erica Belcuna's tweet. Just today that Arc has taken in is the third most popular fun family right now in terms of new money coming in so far year to date. That exceeds the money coming into Black Rock's I shares family, which is much bigger. I mean, that's like that's like the name brand. that's like basically the Coca-Cola of ETFs. So to have a sort of small boutique fund firm with a few actively traded fund pulling in more than I shares, it's just, it's staggering stuff. Yeah, absolutely. So we are going to be talking to someone from Arc today, right?

We are. So the question is, how do they do it? How do they find, how do they pick stocks? I mean, Tesla is obviously this huge winner, but it wouldn't have been a huge winner for them unless they had. been in it for a lot longer than most people. So the question is, how do they find and pick great stocks that trounce the market? Everyone would like to know. Well, I'm also, I'm also interested in how they deal with inflows as they get bigger and whether or not that makes it harder to have a sort of active ETF that is focused on stock picking. So this is going to be a really interesting

conversation. I can tell. Yeah, I'm super excited about this one. So we're going to be speaking with Brett Winton, he is the director of research at ARC. He's been with the company since its founding in 2014. Previously to that, he worked with Kathy Wood at Alliance Bernstein. They've worked together since 2007. So with any luck, we are going to learn at least some of the secrets of ARC from Brett and how they do it. Although I should say, you know, to some extent, maybe it's not a secret because part of what they do is their research is very open. and it's very transparent.

They post models. So we're going to really learn, hopefully, how it all works out. Brett, thank you very much for joining us. Thank you for having me. Happy to be here. So you've worked with Kathy previously at Alliance Bernstein since 2007 with ARC since 2014. Why do you sort of compare and contrast big picture, and that we'll get into details, what the research process looks like at a sort of traditional,

asset management firm versus the sort of open transparent research approach you take it art. I think it's interesting in that I was overhearing your intro and you were talking about stock picking. And I was hearing that and I don't think of what we do as stock picking, at least at its inception level. So we really look at the technology level first. And so we specifically seek to identify disruptive technologies that basically technology platforms that future historians will look back upon and say, oh my gosh, that was a signpost technology. That was as big as the computer. That was as big as electrification. And there's an established criteria for identifying these technologies for it's called general purpose technology theory, but they all follow steep cost

declines. They all cut across sectors and they all themselves are platforms of innovation. And so that actually matches that we're investing in those kinds of technologies matches with three critical weaknesses that I see in traditional fund management that create inefficiencies, pricing inefficiencies that we seek to exploit. So the technologies follow steep cost declines. The drama of those cost declines don't manifest over the next three or six months. So it actually can look very linear over a short time horizon. And so it doesn't really impact an understanding of that.

cost decline doesn't impact the way analysts model the company on the cell side over the next quarter or two. But if you take a step back and you have an intentionally longer term point of view, you can actually come to radically different conclusions about what the future state of the world is likely to look like relative to others, just by having an understanding of the mechanics of how a cost decline occurs and then what the demand elasticity of that price difference is going to be. So that's like from the beginning, we set ourselves up to say, hey, we're not going to, we're not going to try to trade stocks or identify securities that are mispriced on the basis of price to earnings or price to sales or any kind of shorthand for valuation. And we're not also, we're also not going to try to do a full DCF because a full discounted cash flow model because then you get to cheat with how you use the discount rate and your terminal rate of growth. Instead, we're going to say if we own one of these companies, five, years from now, if we're then forced to sell it to a technological pessimist, what will that person be forced to pay, given the cash flow generation of the business at that time? And so just by underwriting

the positions over a five-year perspective, we've been able to, and still are able to identify really radically underpriced securities. So I think of it is we're value investors in intangible assets. Intangible assets are very difficult to understand how much. cash flow they can generate. But we really do the work of trying to figure that out over the time horizon that matters in the part of the capital structure that we're in. Equities are infinite in duration. It's really kind of dumb to underwrite them over a year or two because the market volatility is, you know, really high. Like, I can't tell you what the next 12 months of equities is going to look like. I can actually say with reasonable assurance, over five years, this position is underpriced.

And so that's like the first major inefficiency we exploit. And at the technology level, what that means is, so take Tesla, which is a position that everybody's well aware of, because we have a perspective on what the cost declines of batteries is going to do, it allows us to demonstrate to our satisfaction that we think by 2023, there will be electric vehicles that are sticker price comparable to internal combustion engine vehicles, and the average internal combustion engine vehicles. So you'll walk into a dealership and say, do I want a Toyota Camry that cost me more over time? I have to take it to the dealership more often because it breaks down more often.

And it costs me basically the same amount of money out of pocket. Or do I want the downmarket equivalent of a Model 3, which is faster off the line, cost me less money over time. And it cost me less money today. It would be a real surprise if people didn't shift over to buying electric vehicles. So with that as your initial perspective, you can say, hey, So if we are at, we think there are going to be 40 million electric vehicles sold by 2025. If we're at 40 million and the rest of the market consultants, et cetera, or something like 7 million, well, there's probably some inefficiently priced assets exposed to that technology.

So by identifying, by underwriting over a time horizon that's reasonable, we believe, and by identifying basically fertile terrain where technologies are misunderstood, then we can basically, basically, basically concentrate our exposure into equities that are more likely to be mispriced. So that's the, I just talked a lot, but that's the first of three inefficiencies that we exploit. So I wanted to back up for a second because one thing, one thing I wonder a lot about in tech investing is what's the firm's collective background here? Does it come primarily from finance or is there a lot of technological expertise? And what I mean by that is you mentioned that you're looking at technologies across a long-term time horizon.

Some of your ETFs are very, very technical. I know you have a genomic revolution ETF, for instance, and I think you're looking at space exploration as well. So how do you build up the tech expertise in order to be confident in your calls on what's going to work out in the long term? Well, that actually leads right into the second inefficiency that I see. S exploiting, because all of these technologies are cross-sector technologies, they actually kind of cut across the skis of traditional sector-based analysts.

You know, the auto analyst had been told for years and years and years by the Tier 1 suppliers and by the automotive companies that electric vehicles were actually not a meaningful technology. Not only that, they were a niche technology. Not only that, the people trying to do it were crazy and were probably going to bankrupt and we'll be able to get the assets when we need them, and we can invest in it later on if required. And so, and that auto analyst who's sitting in basically that echo chamber of

information with the quarterly calls with the CFO, the annual calls with the CEO, who has spent his, and in this case it's always a he in autos, his entire career basically being like, well, now Ford's better than Nissan and now GM is better than Ford, has developed a pattern of thinking. that basically doesn't allow him to really understand whether or not an electric vehicle will be competitive with the existing legacy technology or given him a good tool set by which to assess whether or not it will. And so when I was at Alliance Bernstein, Aluminow, which was a genomics company,

is 90 plus percent of genome sequenced in the world go through aluminum boxes. At Alliance Bernstein, it was covered by our industrials analysts. And, you know, he didn't know what to do with it. He almost would have had to take on a whole new, like, set of learnings in order to successfully cover this company that to him looked extremely high priced. It was not like Dana Hur or Honeywell to him. And so he was perpetually a neutral. So the way that we approach kind of technology and markets is our analysts are assigned by technology rather than by sector. And, So it helps that our batteries analyst is able to Sam, who's great. He does the cost decline work on batteries. He understands how it's going to feed into certainly the electric vehicle industry, but he also can look at kind of the energy storage within the utility space. Think about how that impacts kind of the propensity for utility spend. He can understand how it's going to impact aerial drones and their ability to both deliver parcels from place to place or deliver people from place to place. By selecting for people that are expert in the technology, we believe we get

an edge against basically sector experts who are used to a kind of competitive landscape that doesn't get dramatically upturned. Another example is in banks. I'm sure many bank analysts kind of poo-poohed and overlooked square. Well, our view is that, you know, you have a digital bank branch in your pocket, and most of certainly the U.S. is going to begin banking in that way over the next five years. And Square is acquiring customers through its peer-to-peer transfer app, the cash app, at something like $20 per customer. Traditional banks pay upwards of $1,000 per customer account. The product set that Square is going to offer is going to expand,

and that retail bank branch infrastructure is going to depreciate much more rapidly than any of the executives are willing to acknowledge, or, maybe even understand. And so a traditional bank's analyst is not necessarily even empowered to turn around and suggest square to his portfolio managers. It might be off limits to him. And so kind of by focusing on the technologies that matter, with the analyst focused directly on those technologies, actually unlocks a lot of other sort of like misunderstood opportunities, companies that fall through the cracks, and sectors that are really ripe for getting disrupted. that have people who have built their whole careers on understanding that the structure of the sector as it currently exists, rather than as it likely is to exist.

On April 4, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. So this is really interesting, this sort of structural problem in stock identification as a result of people being bucketed into traditional sectors rather than starting at the technology level. I think that's two inefficiencies. What is the third inefficiency that you seek to exploit that you see within the sort of traditional investment selection approach? Well, you alluded to it, Joe, but all of these technologies are themselves platforms atop which other innovations are going to be built.

And so it's really easy to suffer from a failure of the imagination. You really can't like so some people try to model these technologies. they say, these are the three areas where it's selling today, and we're just going to drag out the growth rate, and that's going to be the size of the market. You can't pre-imagined all the things that are going to happen on top of it. And so the only hope to begin to understand the potential scope and breadth, the areas where it will apply and where it won't is to expand your information footprint, to be totally transparent about what you believe is going to happen. And so we publish blogs. we publish white papers, we talk on podcasts, we have our own podcast, it's FYI for your innovation, you should listen to it. And we do that because when we produce this information, when we're

transparent with our forecasts, that information attracts other information. People come back at us and say, I don't understand how you got to that conclusion. You're wrong about solid state batteries. They really seek to combat us to argue from first principles whether or not we're right and why. And so this helps us to understand both the limitations of our ability to tell what's going to happen in the future and to get a sense for some things that we may not have imagined that we should be underwriting in to our fundamental models. And that's very different. You know, just being able to access Twitter is not something that is allowed in a lot of fund management shops. That seems crazy to me. Our analysts are on Twitter. They're interacting with the community. This is the way you understand how the world is going to work. We believe that it gives us a competitive edge that from the beginning, we've designed ourselves to be able to compliantly do that and to kind of operate in the technology circles where these technologies are actually being grappled with and built and deployed into the market.

So I wonder if you could bring a lot of the discussion that we've been having and sort of try to. to solidify it with a single stock example. I wanted to talk about Tesla because, of course, Kathy made a pretty famous call a couple years ago. I think it was for Tesla's stock to go to $4,000. And it has since hit that on a split-adjusted basis and you have a new price target on it. What is it that your methodology and your organizational structure was seeing about Tesla in particular that other aren't seeing. There was a lot of criticism and incredulity about that call when you made it a couple years ago. So what was it that you saw? For one thing, so we have a cost decline on lithium ion batteries. We think we have a better understanding of what goes into an electric vehicle than probably

any other shop on the street. I don't know. Certainly anybody than anybody publishing. That informs both our top line forecasts. I alluded to it. We believe 40 million units will be sold by 2025. of electric vehicles and because they'll be cheaper than traditional cars. When we first began that top line forecasting, the EIA and OPEC, these policy agencies thought that electric vehicles were going to sell in the 200 to 300,000 unit annually through the end of their forecast through 2040. So first of all, you do that and you say, well, we must be doing something wrong. Like, what are we misunderstanding here?

And it turns out I don't think we were misunderstanding anything. It's just that there's not actually a great set of incentive structures for people to make reasonable first principles long-term forecasts. I've been surprised. I thought that we would basically, from inception, I thought we would be kind of duplicating the work of consultants like McKinsey Global Institute, but we would understand the mechanics of how those forecasts were built. And so that would give us an edge. But we got totally different results. Sometimes. Sometimes we got similar results, in which case it's very easy to say, well, this is kind of priced in. You know, there's probably not much that's of interest to us here. But sometimes we got very different results. And I think it's for consultants because they actually are paid to cater to the biases of the executives that hire them. But that aside. So first of all, we think Tesla is, you know, right now 25% share of the electric vehicle industry, 40 million units in 2025. If they maintain their share, which we think they can scale production at that rate, then that would put them at 10 million units in

2025. We think that the electric vehicle industry is going to consolidate. It's actually the products are differentiated on a software basis much more so than internal combustion. So you could get to a naturally higher margin in that industry relative to traditional automotive. And so you can do pretty simple math to actually say Tesla as an electric vehicle manufacturer, just leaving aside Robotaxy and vertical integration into Ride Hale and the insurance product is still a compelling position even at these valuation levels. So it's given kind of the state of their technology stack versus others, you could end up in a situation where just like Apple, they're extracting most of the profitability out of the industry, even though they have something like a 20% share of

the industry. And so that's possible. From the beginning, we've always thought that Tesla was interesting not just for the initial vehicle sale, but because of their ability to monetize the fleet of assets that they have in the field. You know, you can say that Tesla has the largest deployed fleet of robots in the world and that those robots improve over time. And conditional on them delivering the ability for one of those cars to drive itself around, which is, you know, actually quite a difficult technical challenge. You know, you can imagine that those vehicles that Wall Street still believes is I sell a car for $50,000, maybe even optimistic Wall Street thinks I get like $5,000 of operating earnings off of that.

If instead I, the owner of that vehicle, am able to turn it into a taxi, it could do 100,000 miles a year and might generate to me, you know, $20,000 in operating earnings per year to Tesla, to me the owner, you know, additional cash flow on top of that. And so you go from a single sale operating earnings event to every asset they've ever sold generates cash flow for them year after year after year. And we don't think that's 100% probability. In fact, we think it's a relatively low probability. We think there's a 30% chance within our model that they're able to deliver robotaxy capability to the deployed vehicles in fleet. But that call option is worth quite a bit because you have, depending on how aggressive they are

at building electric vehicle factories, they basically get an Uber-like model in a natural monopoly type position in all of the vehicles that they have deployed. So, you know, speaking of that business, I mean, I was looking, you guys. you put your models, your financial models, on GitHub, or at least some of them. And so anyone can go to your site and then go to your GitHub and find the assumptions that you use to figure out the economics, basically, of robotaxies and all this. And how much is worth and how much that's going to cost and how compelling that model of car usage would be versus traditional. is that something like putting it out there like that, having a open source model that anyone can play with,

is that something that previously at your old shop, is that something that you wanted to do and there's just sort of like no way, you know, that sort of openness, what you're describing, being able to tweet, social media, argue about this stuff, get feedback. I mean, I recall, I think a couple years ago, seeing you or maybe your firm get into like going back

with like someone at FT Elphaville, like arguing about your Tesla model. Might have been you. Like, is that something that prior to Arc that you had wanted to do and felt like was missing? And talk to us a little bit more about that aspect of the approach. It's always been clear to me that you get more out of the information ecosystem by providing information into it.

I can't say that at Alliance Bernstein, I wanted to do that. It wasn't even within the, like, realm of possibility. of a thing that I could have done, right? Like, think about the way in which traditional fund management, the analysts themselves are buried. They're, like, not allowed to speak on behalf of the firm in any way. Like, whereas we think that the analysts need to be able to conversantly discuss what they believe, both in written form and orally,

so that they uncover their own weaknesses. Like having to publish to the world makes you a much better and more diligent modeler than if you're not going to do so. You know, if you dig into the like underpinnings of models that aren't published, they're always a mess. They're always poorly documented. There are always things where people have been like, I just picked that because I wanted to, right? Whereas both the kind of auditing process we have to go through in order to get something really. ready to externalize and the research process along the way in which you're always seeking to distill it to its most elemental and understandable form actually forces us to be better at our jobs.

And then once you publish, you get better again because you get this great feedback loop of people telling you what you got wrong, which at the time is kind of like having your eyes gouged out. But at the end of which, you know, you are stronger and more. more certain about the things that you were right about, and you've uncovered the critical weaknesses in how you underwrote the business or the technology or whatever you're talking about. And so it really helps us internally and externally in the iteration cycle of trying to get closer to what's going to be the ultimate truth of how things play out. We have a much more open format than other organizations that I've been involved with.

And I think it has helped both in strategic decision-making internally and certainly in portfolio management and research. I think that the hard part has not been getting people to share their work, but it's really to do the hard first principles work in the first place. I think there's a lot of people, particularly within the financial industry, who are not used to having to be intellectually ambitious and accept that you're going to forecast something with over a time horizon where it's unfair to try to forecast it over that time horizon, but it's actually fair so long as you understand the error bands around what you're forecasting.

And so people prefer because it's more comfortable to pay $5,000 for the, you know, the market's report that tells you what the market is going to be five years from now and be like, okay, well, I'm just going to use that. Well, if you think about within equities and particularly high priced equities, which is the terrain that we operate in, you know, most of the value of that business is in whatever the terminal rate of growth you put on the DCF. And so effectively what they're doing is they are outsourcing the most critically sensitive part of the valuation work to an entity that they haven't necessarily due diligence at all. And that is not really well incentivized to create a forecast that's correct. The hard part is not getting people to share.

It's getting people to make a reasonable first principles forecast that is different. So it's really like the, I think forecasting within the financial industry for the most part is people taking other forecasts and going plus or minus 20% from that other forecast. Right. Right. And that's not, that's just not how we approach it. And having having a process and a discipline of not approaching it that way, not saying, well, I want to see what everybody else has done. And then I feel better about this. So I'm going to go slightly higher or worse about this. So I'm going to go slightly lower. Instead, it's like, well, this is what we think it's going to be. And sometimes you do all that work and you're like, okay, well, that's not that

interesting because it's similar to what everybody else thinks. And sometimes it's wildly interesting because you're like, what is everybody else thinking? And so then the process of discovery of either what did you overlook or what are they overlooking is that's, that's why. where all of the inefficiency lies. Now, early on, we tried to do, I think we, like, the analyst didn't run their own Twitter accounts. They were like, they shared them and they were more corporate. And that didn't work very well. Like, you do have to have somebody publishing under their own name, speaking in their own voice to a certain extent, so that they can both own the mistakes and own the triumphs of, oh, I understand this thing. Right. And so kind of the, the, the,

creating an environment where analysts are owning kind of that intellectual accomplishment and the learning process and how it filters into the learning process is a big part of the secret sauce. And that's not available. I mean, at Alliance Bernstein, I'd do a lot of work and then somebody else's name would go on it. You know, that doesn't feel very good. Like, what's your incentive structure for doing all that work if, like, it's, it's just published under somebody else's name. So I think that having kind of a real, like having the analyst as close to the metal as possible of what's going on in the world and having their perspective on what's going to happen in the world, like come right up against that interface is really critical to just thinking about

things better. Do you have a different approach to finding and hiring analysts at ARC than at a place, like Alliance Bernstein, and can a different type of person get an interview or get their foot in the door there than might say the traditional filter at a traditional asset management company? Well, I mean, we're likely to hire someone soon who doesn't have a college degree, so that probably gives you a sense. The usual route is you hire MBAs or you hire cell side analysts, you know, and those are often very smart people. but there is, I think, a selection bias that happens early on within kind of the financial ecosystem that, you know, cuts out some of the creativity that you need in order to end up with the different result. Another, like, nuance that I think is interesting about our process is if you imagine how do you make money in or do well by your clients when managing equities? Well, if you are wrong, that's fine as long as you're uniquely wrong.

Right. Right? Like, if you're uniquely wrong, everybody thought you were crazy anyway. And so it's not priced in. It's when you're wrong with everybody else that you get into trouble. And it's when you're uniquely right that you actually, you know, compound your holdings. And so if your forecast were on average worse, but unique, that is better than having forecasts that are closer to the actual truth, but the same as

everybody else. And so, you know, you need to have a diversity of kind of cognitive perspective in some way relative to everybody that you're competing against, even if it yields kind of like more volatile results. You, on average, have more differentiated results, which then gives you both downside protection. Everybody thought you were crazy anyway. upside potential, well, nobody expected this. So I think having, and that requires a degree of kind of hardness against some of the social pressures that I think operate in a lot of Wall Street. And so we don't, yeah, we don't select from the same pool of candidates, or at least we haven't, and we've often end up screening out kind of more traditional financial candidates just because

they don't have as idiosyncratic a point of view. I wanted to ask you about another specific, well, I don't want to say specific thing, but another specific technology, since you group yourselves by technology, you have a crypto analyst who looks at blockchain and Bitcoin. And I think Kathy has been quite bullish on the technological potential of blockchain as a whole. Could you maybe walk us through the thinking behind that and again, connect it to your overall methodology and structure. Because I think there were quite a few cell-side analysts talking about, you know, how Bitcoin's a bubble, but blockchain is the technological future. But you at Arc took

a different approach and basically said, buy Bitcoin and by blockchain-related technology. So what was the thinking there? Yeah, I think you can from a very high level think about how all contracts that we sign actually have this fail mode where the political entity that enforces them sometimes just decides not to enforce them or changes the rules. Like imagine you've signed a contract and then suddenly, you know, somebody, the other counterparty in the contract reneges, he doesn't pay up and you go to the government and say, well, you have to force this guy to pay because he didn't pay. And the guy is actually, you know, has an end with the government. The government's like, no, thank you. And so the promise of crypto assets generally is basically that, you, that

that final layer of kind of contract settlement happens regardless of the underlying political circumstances. So you can broaden that across, think of like all of the various contracts in the economy. Think of a structuring desk in an investment bank. It's basically set up to create complex contracts with counterparties where they are assured that those contracts will actually be made good because the counterparties are really well respected established institutions. Well, kind of smart contracting platforms like Ethereum and others allows for an experimentation layer where you can, you know, instead of having to work in Morgan Stanley in order to structure those products, you can be Joe, you know, coder and create kind of those contracts with the protocol itself serving as the ultimate counterparty that we'll see that those contracts get executed upon. Well, currencies are also contracts. In fact, you could argue they're the most valuable contracts.

And there's a social contract between me and the U.S. government that somehow the purchasing price of the dollar is not going to diminish more than, I guess, 2% a year or whatever their target is. Right. So Bitcoin basically supplants that social contract with kind of its protocol for security of the asset. So I think it's a profoundly interesting kind of set of ideas across the entire crypto asset space that over, of all the tech. we look at over probably the longest adoption time will have actually the most dramatic financial and technological impact. So, you know, from the beginning, we thought it was interesting. We had a crypto, a blockchain analyst in, I believe, 2015 is when we hired him. And it was because even though there were not necessarily many investable assets, we understood that the technology at that time, we understood that the technology was within our product suite. We understood that the technology was interesting enough and going to create sufficient disruption

that it was worth beginning to invest in understanding it, understanding where it was going to go, understanding what the best mechanism by which to create client exposure to it. And so that's what we did. We have vehicles in which we can get client exposure to crypto assets. We, you know, it was a smart move, both at the time and going forward. And I think that if you're within the financial services industry and you're not thinking very deeply about how digital wallets, crypto assets, and neural nets and artificial intelligence are going to change what you're doing over the medium term,

then you're not operating intelligently within the firm that you're operating. So I have a follow-up question. I'm going to try to phrase this, as diplomatically as possible. Your outperformance speaks for itself. Your returns have been absolutely excellent in recent years, but there are people out there who would point to that performance and say that you've been riding a tech bubble, or you're buying into the stocks that have very compelling narratives that seem to capture the wider imagination, but that haven't actually been proven yet in terms of earnings. They're just training at, you know, massive valuations and

getting more expensive, but the earnings haven't actually kept up. So what do you say to those people, to people who say that you're basically momentum training on enthusiasm for unproved technology? Wait, I didn't phrase that very diplomatically, did I? No, that's fine. I mean, listen, our job is to understand what the value of something is going to be, as we currently phrase it, five years from now. And sometimes, so fuel cells is a great example where, you know, we did our first work on fuel cells in 2014, 2015. We did a cost decline on it. We determined that within passenger vehicles, we don't think they are going to be cost competitive with electric vehicles until the early 2030s. And that was contingent on something like the

Toyota Marai selling in Toyota Prius-like volumes over the course of a decade. And so, So, you know, having gone through that exercise, it was very easy for us to kind of, you know, look at that entire stack of assets. And every time one comes up and says, oh, well, this is what's different. We've already done the work to understand, you know, the key cost assumptions you need to make and how those costs are declining and what that means for the future unit economics of that technology. And then you can say, okay, that's not something we're going to invest in.

Now, we could be wrong, right? And we're wrong all the time, right? But I think that actually doing the work to underwrite the asset is really difficult. It's not easy. And so there's a difference between I am going to invest in the blockchain ICT company because they say the word blockchain and invest in a company that I think is underpriced over a time horizon that is meaningful to my client. I'm really glad you brought up fuel cells because I was actually going to go there next. So maybe probably people are aware. Some of the hottest stocks right now are fuel cell stocks.

Plug power is one. Fuel cell energy is another one major winners in 2020. And I have a personal interest in this area because, and I'm not saying this to brag or anything like that, but I actually like traded these exact stocks when I was in college. In the late 90s, 99 and early 2000, these same exact stocks, plug-powered fuel cell, they've been around forever. And I, you know, they were crazy overvalued then, but I got kind of lucky and that helped pay for college. Anyway, the point is, I'm not trying to brag. It's just anyway, my point is at the time, it was like, okay, this is right around the corner and fuel cell, fuel cells are going to be on the road by like 2010.

Obviously, that didn't happen. So you're saying it's not different this time, that this is just yet a. another series in an extremely long history of people getting over-excited and over-optimistic about this technology and that once again, it's further off than people think. I mean, there are niche applications where you can underwrite it. I'm not going to disparage or endorse plug power, but there are clearly buyers of fuel cell-driven forklifts because you can have the hydrogen right there on site and it makes sense. To, you know, all Our understanding of how you would have to underwrite that asset to justify its price is you would have to think it's going to get into the truck or passenger vehicle business.

You would have to think that the cost decline on the technology would carry it into a competitive position with alternative mode of technologies in those domains. And just on the electric vehicle side, it's really hard. you have to make a lot of assumptions about somehow the assets or the technology being bought up to drive the cost sufficiently low to make it unit economic compelling. And so you have to generate the hydrogen somewhere, first of all. So that costs money. Your operating costs are much, much higher. Then you have to fund the build out of the hydrogen fueling infrastructure, which is

it's not easy. It's like a hard coordination problem. Tesla, even from the beginning, we thought I was skeptical of Tesla's supercharger network build-out. I thought that that was not a layer that they needed to compete in. As it turned out, I was dead wrong. That definitely differentiates their product because range doesn't become as much of an issue in making the sale. Because people can imagine doing the road trip they want to do with their new car, which is, if you can't do the road trip, what's the point of getting the new car?

right? But a supercharger costs a tenth what a hydrogen station does to like a full supercharger station versus a hydrogen station. So you know, you're talking on the order of $100 to $200,000 versus a million to $2 million. You have to do a ton of execution, have like a ton of sell-in of the technology that it's hard to see how it happens because you can't chicken and egg. The infrastructure in place is sufficient to drive demand for the underlying technology, sufficient to get it low enough in price, that it's cost competitive with existing modes of transport. So is it possible? Yes.

Can we reasonably underwrite it? Not at this time. That was a good answer. But that's what, I mean, you know, when Facebook bought Oculus, right, suddenly all these analysts are coming out with VR is going to be, you know, 18 million units by 2018 or whatever. And we looked at it. We did a model on it. And we couldn't get enough people to buy the headsets for a AAA game developer to justify underwriting a game developed specific for the headsets.

And so you just couldn't, it didn't make sense, right? Like you go through and you try to say, what is this market going to be? And if it doesn't, like, if the modeling that you do doesn't make sense, then you're not going to take aggressive positions on the basis of it making sense. So we never built kind of VR heavily into our NVIDIA model because that, you know, it was a dry hole. And a lot of, I think our role is to figure out actually the things that you can dismiss a whole category of by doing a single piece of work. Like that's really important because you have, you know, now there's a gazillion SPACs coming at us, right? And you need to have some lens by which you approach these assets and say what they are fundamentally worth that allows you to easily establish whether or not something is of potential interest to portfolio management and to your clients.

A lot of short daily news podcast focus on just one story. But right now, you probably need more. On Up First from NPR, we bring you three of the world's top headlines. every day in under 15 minutes because no one's story can capture all that's happening in this big, crazy world of ours on any given morning. Listen now to the Up First podcast from NPR. What separates good leaders from transformational ones? I'm Jessica Chen, and in season two of Leading By Example, we'll sit down with executives like Grace Chen of Bertie Gray to find out. It's important to understand where you spike, but also really acknowledge where you don't and find people who can fill those gaps.

Listen to leading by example, executives making an impact on the IHeart radio app, Apple Podcast, or wherever you get your podcasts. I wanted to go back to something we mentioned in the intro, which is the extraordinary inflows that we've seen into ARC alongside the extraordinary performance that we've been discussing. have those inflows change the way you invest at all or your research process. Does it perhaps become harder to identify new opportunities, the more money you have to put into a certain company or a technology? Every investment decision that you make, you would want to make frictionlessly at the exact size that you want at the instant that you want to do it. And that's not possible no matter how much you're managing.

The research process has always remained the same. We always start at the technology level. We understand the direction that the technology is going. Because all of the technologies that we're investing in are exponential, you're actually creating a lot more opportunity. I mean, if you look at our open source Tesla model, you can drag out, you can see what our next year needs to be given our expectation for EV sales. and it actually meaningfully increases your expectation for value of the company, just dragging out to the right by one year.

You know, and kind of the SPAC phenomenon is creating more publicly traded equities that we could potentially invest in within the technology areas that we're interested in. And, you know, taking in a lot of flows into our assets, luckily within the ETF construct, that's relatively easy. and we are always selective about, you know, how we deploy and what's the most efficient way to get exposure to the inefficiencies that we see. You know, you mentioned the big picture technology families that you start with. So you have this like top-down approach to figuring out the big areas. Genomics is one, robotics. How did you come up with those? I mean, what is there, is there a methodological?

process for figuring out and planning a flag on the ground and saying, okay, this is going to be really big? Yeah. So they all have to, and so I alluded to it, but there's this theory called general purpose technology theory where it's like these academics have agreed upon what the criteria are for really meaningful technologies. They all have steep cost declines. They all cut across sectors and they're all themselves platforms of innovation. And so we try to apply that framework to the technologies that we're interested in. We think there are five fundamental technology platforms that are all entering the economic marketplace today, gene sequencing and editing, AI and particularly neuronets, robots, particularly collaborative robots, energy storage, and the advances

in battery technology, and then blockchain cryptocurrency. And so we believe that yet future historians will look back and identify all of those as big technological. buckets, but, you know, within taxonomies, there are always weaknesses, right? And you could draw the lines in a slightly different area. So we tried to look back and see, like, what technologies did historians agree upon were these general purpose technology platforms over time? And there's not consensus at even looking backwards. So, of course, there's not consensus today. What are the major technology platforms. But I think it's a, so the other, from those five technology platforms, there are also, we have 14 underlying technologies that are discreetly modelable, where we have a

good understanding of the cost decline of kind of how it cuts across sectors and the equity market capital cruel that we expect those technologies to achieve over time. I'm a big believer in getting really dirt simple with your assumptions of things so that you can tell if they make sense. So it's kind of like, you know, what are robots going to be worth? Well, what if we start out and say, what about every manual laborer employee in the world and we say, well, we're going to supplement this person with a $10,000 tool that's a robot? Like, what would that market be worth? What would be the cash flow accrual to the robot manufacturers in that instance? And then so how much would you assume is occupied in terms of enterprise value by the companies that are catering to that

economic opportunity. And so if you do kind of that, very high-level assumptions about the technologies that we track, you would assume that there's going to be $50 trillion in market cap accrual to our technologies over the next decade. And so this gets back to the capacity question, there's going to be a lot of economic value created. And there's going to be major, major businesses that accrue out of it. You know, like with it, if you look at how we've modeled autonomous robotaxies globally, we think that autonomous robotaxies, the platforms that enable that are going to be worth more than the global energy sector as a whole within five years. Just, and you actually don't have to make radical assumptions to get there. You say,

Well, look at global miles driven and these are going to price at something like 25 cents a mile. So they're going to be cheaper than actually buying and owning and operating a vehicle in the U.S. That costs you 75 cents a mile if you buy a new one. Right. And so it's going to be the default way by which people get around. These autonomous taxi platforms are going to scrape a platform fee just like an Uber or Lyft. If you back into the aggregate cash flow that you expect, given your adoption curves and everything else, it's, you know, measured in hundreds of billions of dollars within five years. So it's natural that the market would

pay at least a reasonable cash flow multiple on that cash. So there's a combination of like, what's the cost decline look like? How is it cross sector and when has it gone cross sector? What other things are going to be built on top of it? And thinking about how meaningful is this going to be economically over the medium to long term? And if you can kind of dimension that it's meaningful and cross-sector and steep cost decline and itself a platform of innovation, it's very likely that this thing is going, there's going to be a lot of value created here. So it's worth devoting the intellectual capital to understanding it and understanding the puts and takes of how it's going to get to market and which part of the value chain is going to be the

most cash accretive and how that part of the value chain is underwritten today relative to how you think it should be. Would you ever consider starting a SPAC or is it too much of a departure from the current model? Well, I mean, I think that there are pluses and minuses of SPACs. I think that there is a degree of nervousness that at least I have right now, that when people raise SPACs, like they are heavily incentivized to figure out something to buy with them. Nobody returns the money, right? And so it's almost like you create a time bomb of IPO, right?

Like you IPO, but the real IPO is when you merge with the other entity. And the people who are controlling whether or not you merge, yes, you have to get the shareholders to vote. But the people who are controlling it, like it's basically like, you know, buy something or you lose it. It seems in some ways backwards to how companies should come to the capital markets and that they should come to the capital markets when they're ready, not because there's a pool of money that's going around trying to find everything that could possibly go to the capital markets.

I'm nervous about that. On the other hand, if you think about what has happened with late stage venture, which is where a lot of these companies would otherwise have been funded, is that there you are only allowing accredited investors to invest in kind of these technology companies. and, you know, the late-stage venture capitalists get money through their carry and their management fee that's also quite punitive to the end shareholder, and you're cutting out the entire, you know, Joe investor who's not accredited. And so you could argue that this is a way to democratize access to these late-stage venture-type assets. I think that to me, there seems like there's a lot of misbehavior going on in the space. And usually our bias is to, when there's a lot of

capital going after something, to be wary of it. But I can't, you know, comment, you know, directly. But the lack of disclosure for the underlying companies, I think, could lead misbehavior on top of misbehavior. I talked about how consultants, like McKinsey and stuff, they, their forecasts weren't as good as I thought they're going to be. Well, we also look at the forecasts of the management teams within these SPACs. And that is a difference from an IPO. And, you know, in S-1, you're not going to get a management team telling you what they think they're going to print in revenue five years from now. Within the SPACs, the management teams are. And there, as a general rule so far, looking at what management teams have forecast, we have a hard time

hitting that. So let me sort of, and, you know, I think we can wrap up soon. But let me just sort of, this sort of gets to a bigger question. And Tracy sort of hinted at it, you know, this sort of the claim among our detractors that you've done a really good job basically riding this big bubble. What happens if at some point we are in a bubble? And some people would say we're in one now, but, you know, there are times in which in retrospect you're like, oh, there's definitely, we're in a bubble. There was no good tech to buy in December of 1999. Anything that you bought then, pretty much in anything related to tech, was probably going to be underwater for years if you had purchased then.

Maybe some of them have obviously done well since then. Could that, what do you do if you come across that environment where all of your models are saying in these areas of innovation, in these areas of tech that we're into, we just can't make the numbers work for anything. that's of like quality. Is that a concern? Is that a situation that you've thought about? Like, how do you think about that question? I can say that right now we can still find a lot of inefficiently priced assets. So at least as we model or as we expect the world to unveil itself, I don't see it, you know, within the context of the positions that we put client money into. I think that, you know, financial markets are full of, in some ways, saying a bubble,

I think is, you know, it's lots of burbling. And sometimes you get a, you know, a bigger degree of burbling. But there are always, you know, there's the ICO boom in 2017. There's if today or, you know, three months from now, the equity markets are down 40%. Right. You know, then we would all look back and say, oh, well, the SPACs were the sign. It was obvious. didn't you see, you know, and if you are investing money in the equity markets, equities are infinite in duration, you should not be doing that on the basis that the one-year result is going to be meaningfully indicative of whether or not it was a good decision. That it's the wrong time horizon.

Yeah. Right? And so, like, I, like, my comfort level is that we look out five years. And I say, this looks very reasonable over five years because we're not making, we're not going out five years and saying, then I'm going to pay an elevated multiple. I'm going out five years and saying, I'm going to be a forced seller to someone who only pays the market multiple for the cash flow coming off a business with this kind of margin profile and capital intensity.

And, you know, our return hurdle for the positions we underwrite is 15%. So we think it's going to roughly double over five years. Well, you know, so I have a lot of ways. in which to get exposure where that's at least the way we forecast the world. Now, could we look really dumb 12 months from now? Yes. In fact, I think it's likely that at some point, people will think that ARC was a scam and that we were, you know, we don't know our left from our right and we're doing things wrong. And our discipline and our mission is to continue to say what we think is going to happen and to try to, you know, buy basically intangible assets at deep value, regardless of

the market environment. The 10-year rates and that they came down as much as they did during the pandemic, it provides the highest leverage to the longer duration assets, right? And so naturally, if, you know, over 10 years, you can only get 1% compounded on your money, well, then something that's not going to produce cash flow for you until 10 years from now, but that cash flow could be monumental looks a lot more attractive because you're, you know, the competitive rate of, of, cash flow generation is just much lower. That had an effect on the overall market multiple. And we don't try to take a stance against the overall market multiple, as in we don't try to position ourselves that, you know, I'm not going to, I'm not trying to allocate between equities and fixed income, right?

And so I just try to underwrite the equities, you know, qua and other equity exposure. You know, financial markets, I was being accused of having committed career suicide because of our Tesla position. And that was, you know, that was Memorial Day of like 2018. Yeah. You know, that was that was not that long ago. The, the markets mania is much more volatile than our fundamental valuing of the company. So the way in which we manage the portfolios is we're typically short-term contrarian. If something's rallying, it's often rallying if it goes up because it beat on earnings,

that doesn't change what we think the company is going to look like five years from now. So we'll often sell off that gain to buy into something that, you know, suddenly was investing too much in R&D, so they missed on earnings. And it's like, yes, give me more of that. And so that actually doing the work over five years, provides us a lot of kind of anchoring that allows us to manage positions within the portfolio as they respond to news that we don't think is actually fundamentally meaningful. In the event that the markets start to sell off for whatever reason, because rates are going

up because, I don't know, geopolitical risk diminishes. But, you all are the ones that get to explain daily why markets do what they do. Then we'll respond to that. That's why we actively manage the portfolios. But I certainly, I would hate to have to. to say what's going to happen in three months. I think that's much harder than saying what's going to happen over five years, actually. I think it's a really challenging game because it requires you anticipating what other people are going to then think rather than trying to forecast what's going to happen kind of objectively in the world.

And I think a lot of people play that game, but it's not that interesting to me. I think it's really hard. Brett, that was fantastic. Really, I'm really glad we got a chance to talk to you. I learned a ton in that conversation, and I appreciate you taking the time. My pleasure, Joe, Tracy. Thank you. Thank you so much.

Thanks, Brett. Tracy, that was really cool. I mean, obviously, I've been aware of Arc and their amazing stock picks and particularly their sort of vindication on the Tesla pick. but I'm hearing overall, like, their sort of like general approach that was a very useful and interesting. Yeah, I agree. There were two things that stuck out from that conversation for me. And again, I don't mean to navel gaze in the media too much, but one of them was the way they organized themselves around technologies rather than traditional sort of analyst or industry sectors.

Yeah. And I have to say, I think that's something that a lot of media companies have struggled with over the years, you know, particularly when Bitcoin came out, for instance, there was a lot of discussion about it. Should it be done by market reporters? Should it be done by commodities reporters? Does it fit into an investment team or the tech team? And everyone kind of struggled to fit it into a traditional category. But had you just looked at it as a sort of crossbeat technology like blockchain, maybe it would have been easier to conceptualize, I suppose. And the same thing for, you know, electric batteries and things like that. Yeah. So that was really interesting. And the second thing about refining your work through public interaction and discourse also strikes

a chord. Both you and I are very active on Twitter and social media. I think journalism in itself is a very public activity since every time you publish something, you're probably going to get some sort of reaction or feedback to it. And in the end, you can use that to refine your thought process. You think more strategically about your model or your subject matter or whatever. And I don't know, I just see a lot of parallels between what his analysts at ARC are doing and what some journalists are doing or could be doing. Yeah. That definitely stood out to me. And it's one of these things where, like, I get as a journalist so much value from interacting on social media, arguing with people, having people, like, try to, like, pick apart my point.

And it's very intuitive after he describes it. It's not something I had, like, really thought about. Like, I do think analysts or by-siders or sell-siders, like, I do think it's, like, good to publicly interact. But after hearing him, like, describe it, that benefits, what they do with, like, posting all of their, uh, their, their models making them public that you could sort of instantly see. how an asset management firm could really use that to the advantage. And it's also good marketing. I mean, you know, it's it stands out. I mean, it's good for refining your arguments. I, you know, I remember those sort of like fights about Tesla, especially, as you mentioned back in 2018, when there were serious questions about whether the company was going to make it. But it's also good

marketing. And it makes it stand out. And I can't think of any other firm right now that's doing anything similar, but I could see a lot more sort of embracing that model. Absolutely. The other thing that I was thinking about was our conversation around value investing and this idea that actually if you kind of redefine how you're looking at a company's value, then maybe your universe of value stock starts to look very different. So this idea that, you know, the way ARC is looking at it, Tesla probably was a value stock back in 2018. And I suppose also to Brett's point, it depends on your time horizon, right? Yeah.

I mean, you know, like, you got to be pretty confident. And I thought that was really interesting about the sort of like the value of genuinely original ideas. Because there is a lot of, you know, within the space of like the million people who say cover Apple or cover Facebook. As he put it, you know, it's like maybe the bullish ones, take the consensus and add 10% or 20%, the bearish ones subtract 10 or 20%. But the idea that's like coming at a problem where you genuinely seek to uncover ideas that aren't just some deviation from consensus is a sort of a very interesting challenge. But again, you can see if you're like really confident about it and you feel like you understand it, you can come up with interesting ideas that you can have some conviction for, make a meaningfully size. bet, so to speak.

I'm trying to think if there was one other thing that stood it out to me. But yeah, let's leave it there. This has been another episode of the All Thoughts podcast. I'm Tracy Allaway. You can follow me on Twitter at Tracy Allaway. And I'm Joe Wisenthal. You can follow me on Twitter. Oh, I remember what I was going to say.

Can I just say it real quickly? I'm just going to say it right here in the outro. I thought that was, you know, I'm personally biased because I've always, I've been interested in fuel cell companies for a long time. But I did think that was a pretty interesting example of the case that they're not just bubble riders, that there are like these sort of sexy areas of the stock market that they're not participating in. And that if they were just sort of a firm that was like riding bubbles or riding hot trends, that they would be participating in that area.

So I thought that was interesting. It's like, here's a thing. A bunch of investors are super excited about it. They are not. It's sort of like I thought a useful counter example of this. idea that they're just in all the sexy areas. Anyway, I just wanted to say that. I'm Jill Wiesenthal.

You can follow me on Twitter at the stalwart. Follow our guest, Brett Winton. He's at Winton, A-R-K on Twitter. And of course, check out all of their white papers and models at their website. Follow our producer, Laura Carlson. She's at Laura M. Carlson. Follow the Bloomberg head of podcast, Francesca Levy, at Francesca Today. And check out all of our podcasts at Bloomberg, under the handle, at Podcast.

Thanks for listening. Hi, I'm PJ Vote. My podcast Search Engine has a new two-part series for you. Of all the new technologies coming out of AI, the most transformative one might be driverless cars. They're already on the road in 10 American cities, and they're quickly coming to more. We tell the story of how we got here.

The secret team at Google that spent 15 years building what might be the safest vehicle on the road, and we cover the fights brewing in blue cities, where unions and politicians are working to keep those cars off the streets. Listen to search engine wherever you get your podcasts. What separates good leaders from transformational ones? I'm Jessica Chen, and in season two of Leading By Example, we'll sit down with executives like Grace Chen of Bertie Gray to find out.

It's important to understand where you spike, but also really acknowledge where you don't and find people who can fill those gaps. Listen to Leading By Example, executives making an impact on the IHeart Radio app, Apple Podcast, or wherever you get your podcast. podcast.