The Exchanges

Q And now over the years, the trading side evolved into what's become well-known in the industry as TOPS. And why don't you touch a little bit on that evolution and what TOPS has become?

A That was really Ian's genius and vision, and it's actually started With a discussion about how we measured our brokers. And like most people those days, it was pretty random. It was kind of a quarterly finger in the air. Who's done a good job. They get X share of the wallet. And Ian said, well, there must be a better way of measuring this and doing it systematically. And we asked Anthony Clay, at that stage was a 21 year old summer trainee, still at Oxford, to think about how to do that. And he came up with the idea of giving the 40 brokers who covered us at that time a virtual portfolio, and said, you run this virtual portfolio, we'll measure whether you actually create any alpha, any value. And much to my surprise, because I was pretty skeptical about it, I had the typical arrogance of a buy-side guy, They actually generated a lot of alpha. And so we, having measured it for about a year, we then said, or Ian said, let's put some capital behind this. And we put, by that stage, Eureka was a kind of two, two and a half billion dollar fund. We put two hundred million, so 10% of the fund, into a program which basically replicated the ideas of all of the brokers who were sending them in, the ideas into us in real time. And it got off to a huge flying start.

AI assessment note: “giving the 40 brokers who covered us at that time a virtual portfolio”

Q of scrutiny around the world in active management in general, particularly in say the long only equity markets. Because of this persistence of called broad-based underperformance relative to index funds. So how do you weigh out what you see in these inefficiencies that can move and adapt and you have to adapt? And on the other hand, the ability to capture them with so many smart participants in the marketplace.

A I think the main concession I would make to the Chicago school to efficient market theory, the only concession is that markets Are gradually getting less inefficient. And that's because they are professionalizing. So I think the US has gone from in the 19 seventies, 50% retail to today, 15. And I think in 19 oh seven, it was something like 85% retail. And China today, Is 85% retail and 15% professional. Professionals inherently have much more information, especially today, than the retail investor. And therefore, as the people around the poker table get smarter, it gets more difficult. And so we are seeing a process of winnowing out of the people around the table. That leads you therefore to look at China as a better source of alpha, theoretically at least, than the United States. Which is the most professionalized market. But at the same time as that is happening, you're getting all kinds of new inefficiencies emerging. So, you know, whether it be index funds, which are themselves just a form of lagged momentum, or whether it be the fact that a significant number of part of the hedge fund industry kind of totally focuses on quarterly earnings, and you get this phenomenon now of massive crowding around quarterly earnings, which actually is becoming less and less efficient as a way of extracting Value. Or the recent David Portnoy phenomenon, retail buying. Tesla is up 400% in th…

AI assessment note: “as the people around the poker table get smarter, it gets more difficult.”

Q So one of the things you talk about in the book is the importance of size, which is a more nuanced and complex issue than size is the enemy of performance. But on the other hand, you said as the industry evolves, you get a little bit more concentration in the, in the alpha generators. So how have you thought about size of asset management firms?

A There's a chapter which called size matters. And the real argument there is that Although you need a certain amount of size to have critical mass and pay the bills, beyond a certain level, size is most of the time a disadvantage. And so it's a sadness to me in a way that there are so many barriers to entry now and that it is getting more concentrated. But the point that follows from that is we've built our business all the way through by recognizing that size matters at a, beyond a certain point. Your returns are handicapped by the friction costs of trading or by your footprint in the market, your liquidity footprint. And that's why we closed Eureka when it was two billion in 2001, gave back capital. And we've frequently closed our funds all the way through the life of the firm. And the paradox about Marshall Waste is we've grown to be the largest equity hedge fund in Europe, Because, in my opinion, we constantly closed. Because other people grew to be big and blew up. Because they were too big. Their size fell for the wrong reason. In our case, we said, right, the maximum we can do in this strategy is X. 1,000,000,002 1,000,000,005 billion, it's now closed. And that put the onus on us then to say, well, how can we innovate to find other ways of generating alpha? And so that's how our growth has been slow. And it's been based on that hindrance. Constraints are also very creativ…

AI assessment note: “beyond a certain level, size is most of the time a disadvantage”

Q And so what was the original Marshall Way strategy?

A So the original Marshall Way strategy was, it was the Eureka fund, which is still the, the flagship, and it was European equity long short. We were about the third hedge fund in Europe. So we're well behind the curve vis-a-vis the US, but we were relatively early in Europe. We divided the fund into two components, really, what we call core and trading, and the core investments were Ben Graham or Philip Fisher, long-term quality companies. Philip Fisher, you like companies which have great quality business and a quality management, so they've got to be lucky and smart. And then the trading side, which was Much more high turnover, very catalyst based, looking for every opportunity in the market. And one was, one was lumpy, concentrated, and the other was less concentrated and much more active. And that played to both of our backgrounds. The background was in trading, my background was in longer term investing. And that was, that was the original proposition. We raised fifty million dollars of rich half was from Soros. The rest was from family and friends, which in those days was a lot of money, and it was enough to get started.

AI assessment note: “it was European equity long short. We divided the fund into two components, really”

Q wealth management firms moving to Ridgeline gain a decided advantage. That's why customers call it miraculous, game-changing, and an awakening. If that's not how you would describe your investment management tech, request a demo at ridgeline.ai. And now back to the show. Let's talk a little bit about the short side specifically. Very different game than the longs. What are the key lessons you've learned from shorting all these years?

A I say in the book, the kind of tautology, the statement of the obvious, that shorts and longs are very different. And if you look at the long-term record of the Eureka Fund as a proxy, our long-term annualized alpha on the long side is about nine percent, and our short side about three percent. And I think that short side is actually probably quite a creditable result in the context of the industry. And provided your short side is positive, short average positive, then it allows you to really employ your long book very aggressively. The first reason for a difference is that the information bias of the market is set up completely for the long sides. The brokers essentially seek to please companies. 70 to 80% of all recommendations are buy or hold. That's 50 plus percent of them are buys. So your short seller is competing against in a world where information bias is against him or her. It's also competing in a world which is inherently, it's more expensive because you have a borrowing cost, so there's a bigger hurdle before you make money. It's much more competitive because when you're short a stock, obviously you're competing against long and short sellers, but you're effectively when you're borrowing a stock, you're competing only against short sellers. And they are, because they're essentially hedge fund managers, they're typically amongst the smarter people in the market. And…

AI assessment note: “It is a more difficult game, and it requires more trading.”

Q You know, for a long time between TOPS and the fundamental investing You've had some semblance of optimization and machine work, sort of data crunching and investing that way, and then of course on the fundamental side. What's been your assessment of this question of man versus machine?

A Well, I started that chapter, it's not a quote from anybody, but it's a kind of paraphrasing of Kasparov. A machine beats a man, but a man plus a machine beats a machine. And that was certainly the conclusion he came to after he lost the Deep Blue. And so we think that the best Way forward is to blend the two. Now we, at the moment we keep our business in a way separate. So we've got one side, which is fundamental, which is using more and more data and more and more processing power and new, all kinds of new data sets to help give our managers an edge. And on the other side, we have top systems, which is always different from any other systematic system because it uses human beings and it's prime in human Cognition is its primary driver of the origin of the idea. So we try and keep those two things apart for the good of the business, actually, because you want to maintain lots of different alpha streams and alpha sources and keep them distinct. But wherever the industry goes, and You can have all kinds of debates about where the industry will be in 10 years. We think that we are in an incredibly good position because we have these two building blocks, and certainly you can't call the end of fundamental investing at all. I think fundamental investing evolves, but it evolves to incorporate more and more data, but still with guys pulling the trigger at the top.

AI assessment note: “A machine beats a man, but a man plus a machine beats a machine.”

Q the one hand, you could say, oh, these are all equally weighted by positions, adjust for liquidity, whatever it is. And on the other hand, you could then optimize on the particular skills across stock selection, a sector, a market regime. One of the 10 and a half lessons in your book is the notion that the best portfolio construction combines concentration and diversification. What do you mean by that?

A It's a paradox, because the two things should be antithetical. And so diversification, Markowitz, risk managers love diversification. It brings clear benefits in terms of return per unit of risk, and all good portfolio construction should be aiming to achieve a minimum level of diversification. But for stock pickers, my view is that most Managers, few managers have more than 10 or 20 high convictions at any one time. And for that reason, one of the constraints in the top system is that we are the contributor portfolios. The contributors are only expect to run about 10 names in their book. We say, we only want your highest convictions. Don't give us, you, don't you worry about diversification. Just give us your highest convictions and we'll worry about the diversification. And because we want that concentration within our individual managers, we also encourage them to put a high amount of their risk in their top convictions. And so those two things, how do they combine? Well, they don't combine at the level of one portfolio. They combine at the level of the product that you deliver to the client. So that's why Eureka evolved from being one strategy when Ian and I ran it. To being now combining 15 strategies, which we think is the kind of minimal level of diversification you need, and that then delivers to the client a set of very interesting alpha streams. Each one is relatively…

AI assessment note: “They combine at the level of the product that you deliver to the client.”

Q And what did you come to believe about how to add value in the equity markets?

A I went through quite a long process of looking at different types of ways of analyzing companies, EVA and CF, ROI, and kind of different ways of putting all the numbers together, but the bottom line was, I suppose it's Ultimately, you're looking for companies where the long-term value is not appreciated by the market, and the long-term value is defined as the sum of all the cash flows. So you've got to understand why the cash flows of the companies will generate in the future will be significantly more than the market is appreciated. So it's very much weighing, Ben Graham weighing machine philosophy, but also combining that with catalysts. So it's never enough For me to just be looking at understanding a long-term undervaluation, you've got to have a, a catalyst which will draw people's attention to the valuation. So that was the, that was the kind of approach that I grew to like, and it used pretty early on.

AI assessment note: “Ultimately, you're looking for companies where the long-term value is not appreciated by the market”

Q So if we look at Marshall Waist today, How did the investments get made across these strategies?

A So we've always worked on the basis, first of all, that we were size constrained in everything we did, and that led us to allow different strategies to develop their life and to distribute the capital as much as possible, really. And because of all the benefits of diversification, which I talk about in the book, and, and the benefits of blending You can blend lots and lots of different alpha signals. So the way Marshall Waste has evolved is we have a huge number of different signals which were blending together in the systematic side, which includes both the TOPS business we still call TOPS and a systematic business in its own right. Then we have the fundamental side which has evolved from having essentially one strategy when we started to now there's around 15 strategies run by fundamental managers and Covering different sectors and different geographies, but they also then benefit from all of the infrastructure of the firm and use a lot of information that they can get from the systematic side as inputs into their decision making. And we've also now introduced quantum mental investing, which is effectively use that word to refer to non-traditional types of information sources. So mobility data, emails, social media data, Retail, et cetera, et cetera. So that's another source of information and alpha signals, which we can blend into the results. But what you have in the end is…

AI assessment note: “we have a huge number of different signals which were blending together”

Q So when you first started that interest really in the equity market, how did you start to learn what you came to believe about how markets work?

A It's been a very, very long apprenticeship and I'm still learning. So I didn't learn massive amount actually in Zurich. I then went to business school in Seattle and then I went to Wahlberg's and went into the European investment department and had some very good mentors. And really learned on the spot, and it was learning by doing, and learning by my mistakes from the beginning, which is how I started to learn. And the, the right or wrong thing about Prochip, Mercury, or Warburgs, Mercury was the financial business, was that they gave you money to manage almost immediately. So actually, in my case, far too early, but I was given money to manage within six or nine months, which is ridiculous. And I paid the consequences with some of the things I got wrong, but I learned by doing, really.

AI assessment note: “really learned on the spot, and it was learning by doing, and learning by my mistakes”

Q We could dive in for hours on this. I know we're going to turn to the book, but I do want to ask, when you have, say, the top structure, which are sort of external resources, Broker recommendations, and then you have these 15 internal strategies. What do you see as the strengths and weaknesses as an organization of sourcing the ideas internally versus externally?

A The advantage of sourcing them externally is you can do what you want with their ideas. Their feelings are never going to be hurt. You can virtually hire and fire. So you can effectively, if you choose, People have six hundred million of capital, which follows the ideas of a person. They won't know that, but you could, using that, almost in an undiluted, you could, if you choose, on an undiluted way, use one person's ideas. You have to put them back on your book. You have all of the headache of having to do a contract, et cetera, et cetera, which is a much more complicated thing. The managers that we do have on our books are all people we're very committed to, and we have a very, very Low turnover compared with any other of the comparable funds who have lots of managers. People stay with us a very long time, most of them indefinitely, because we believe in backing their skill and we believe they're skillful. There is a stickiness to that which we don't have with the external contributors. And then the other thing is with the external contributors, you can blend their signals with lots and lots of other signals to create effectively almost a completely new product. And again, nobody's going to get hurt. Whereas if I took alpha signals from internal managers, And, you know, so this guy's good at healthcare, so I'm going to use that, the healthcare signal. He's going to, or she is…

AI assessment note: “The advantage of sourcing them externally is you can do what you want with their ideas.”

Q How have you thought about this real question of the evolution of the industry?

A We think about it all the time. There are some trends which are, they seem to be so structural that nothing's going to shift. So the, the index, the trend of indexation, which is about two or three percent a year, whatever, that's almost like a straight line. So there's going to be more and more money going to indexation, which I would call that hollowing out of slightly sleepy active managers in the middle of the spectrum. And you will end up with a large amount of equity assets, which are run passively. And then you'll have a small group which is run very actively. I run only through hedge funds and alternative investments, and you'll have the best of the best active managers will still have a role to play, I think. I don't think you'll ever get to a point of being completely passive. There'll always be huge new pockets of inefficiency opened up by the way the industry evolves. I'll be very cautious about making big predictions about the industry anyway. I think what we're more interested in is Anticipating the opportunities of the next two to five years. So things we're looking at at the moment, we launched an ESG fund, and that's a very exciting area, which we, having been pretty cynical about it as an alpha source, we now think actually it's going to be pretty alpha rich. And we're also looking at the crossover space, which is a space between public and private. And that's…

AI assessment note: “hollowing out of slightly sleepy active managers in the middle of the spectrum”

Q What have you found are the optimal ways to then transfer, let's say, the presence of stock selection alpha Into a portfolio that delivers?

A In the TOPS part of our business, your target, you measure alpha specifically is the thing that you're trying to capture. And in fact, the other word for TOPS is alpha capture. So the thing that you're putting into the portfolio is not A return. The thing you're optimizing is not a return objective. It's an alpha objective. So it's the performance of those ideas against whatever the benchmark is. And you create information ratios and risk manage around the alpha. And for our managers, it's actually the same thing. When we optimize between our managers, we optimize to their alpha. We're not optimizing to return. We're optimizing to alpha. And so they get capital if they do well with their alpha. And again, and so for our, for our fundamental managers, We end up with the same level of complexity in how we evaluate their alpha as we do on TOPS.

AI assessment note: “The thing you're optimizing is not a return objective. It's an alpha objective.”

Q So given that risk, why write a book?

A Well, I guess partly I wanted to read, write it before I failed. So I get it out. And I just thought that was Actually, this started as a talk at a conference, and I thought, actually, this, I wouldn't mind just getting this all down, because I think it's worth writing it down. It's just enough to make a book. It's quite a short book. And the other part, I suppose, is philosophically, there's something I wanted to get across, which I feel strongly about, because the underlying theme is you, all the way through, is really fallibility, view and fallibility, and inefficiency of markets, and uncertainty, and All of these things, and that message needs to be out there, especially today, because there is a, on a much broader scale, there is a kind of a crisis of epistemology in the world. People are arguing now, not only about what they believe, but about how they know anything. Empiricism versus rationalism, but then in the new woke, white fragility environment, there is a complete rejection of reason, as a Simply something that comes from the patriarchy, and so on, and you must actually reject reason, because there is no such thing as truth. So truth is simply what you perceive, and what a person who's more powerful than you tells you is the truth. So there's a crisis, in my opinion, of philosophy, and that applies to our industry, curiously, because there is a, the way I wrote abo…

AI assessment note: “philosophically, there's something I wanted to get across, which I feel strongly about”