#87 - Eduardo Repetto: Financial Science Meets Portfolio Innovation

0:05Welcome to the Insightful Investor Podcast, a weekly series that seeks to share industry, investment, investment, and market insights. We define insights as concepts that are counterintuitive, widely misunderstood, or underappreciated. In other words, unique ideas that you probably won't hear elsewhere. I'm Alex Shahidi, the host of the podcast and co-CIO of Evoke Advisors, a leading investment advisory firm. Learn more about our show at insightfulinvestor.org.

0:38Today's guest is Eduardo Rapeto. Eduardo is CIO and co-founder of Avantis Investors, which manages about$75 billion as of the end of June. He leads the research, design, and implementation of the firm's investment strategies and is widely recognized for applying financial science and innovative thinking to portfolio management. Before Avantis, Eduardo was the co-CEO and co-CIO at Dimensional Fund Advisors. Welcome, Eduardo. Thank you. Thank you for coming to me here. Breaking news. We are 85 billion now, but that's okay. You're growing so fast, we can't keep track. Good clients. That's why we have to be thankful.

1:22Good clients. That's right. Let's go back a few years. You started with a PhD in aeronautical engineering at Caltech, and then obviously you pivoted to finance. Was there a defining insight that drove your transition? And are there any similarities or differences you found between solving scientific problems in engineering versus challenges in investing? No, that's a great question. So I'm a geek. So I have a civil engineering degree from Argentina. Then I got a master's in mechanical engineering from Brown University. Now the PhD in aeronautical engineering from Caltech. But my thesis was more about theoretical models that you solve with computers trying to explain physical things like properties of materials and whatnot.

2:16And so whenever you do a PhD, you face a problem that has not been solved, and you're trying to read everything that is run, learn from every work that everyone has done, and try to push the frontier of the current knowledge. And if you think about that in an abstract way, that's very useful for investing in finance. There is a lot of work being done. A lot of people have written for forever, more than 100 years about finance. Everyone is pushing the frontier of knowledge a little bit more. And what you try to do is bring something to investors that is a little bit better than whatever was in the market before.

3:03Try to provide some enhancements. And you try to do it very efficiently so you can deliver that at a very good expense ratio. So people have not only good product, but also have a good set of fees. And so the concept, the after way of thinking about science and evolving science and then bringing it to business, it's good. You say, well, you should have studied a PhD in finance. Well, you know, I love what I was doing at the time. And now I have been doing investments for more than 25 years. I love what I'm doing now. It's interesting. You mentioned maybe you should have had a PhD in finance. But what you miss there is you miss the coming into a relatively new industry with a fresh perspective, with a toolkit that you developed in a different discipline that you can apply to this space.

3:58And that could potentially lead you down a different path than if you kind of grew up on the finance side. That's a good point. Look, when you do the kind of work that I did, you need to understand math, you need to understand physical sciences because you have to apply this math on the physical models into something that you're trying to solve. And then you need to try computers and optimization in order to be able to solve these complex problems. And all that is very useful here. You have a very good point. Well, DFA's academic culture, which is unique in that sense, made for an excellent transition from academia and offered a unique learning experience.

4:38How did that context help you bridge your academic training with practical portfolio management? Look, I was a Caltech and I decided that I didn't want to be a professor. So it was a point that I had to start applying to become a professor somewhere around the United States or anywhere, or do something else. One of my former bosses, a mathematician, he has a doctorate in mathematics, very weird stuff, went to Wall Street. And he told me, hey, you will have fun, do that. so i went to interview in new york i live in los angeles so my wife is from los angeles i met her here so i went to interview in in new york in in january oh man i even forgot to have any winter coats it was cold and once you live here the weather is so nice i said better if i find something here and dimensional dfa was hiring at the time and dfa was based in santa monica and it was a small company if you think about that i think that i was employee 107 if you remember right and the fa was around 25 billion under management and they hired me they hired me to work with people that are amazing people like your friends jim fama then i was interacting with myron shawls bob merton with humongous you know amazing brains in finance i have written you know the knowledge that we have today, they have big influence in all the knowledge that we have in finance today.

6:09And I was working for them. And when you come with a PhD from Caltech, this idea that you work five days a week for eight hours a day, that doesn't work. It's just, this work is continuous. And so the ability to interact with these people, that was amazing. And it was like redoing a PhD, but now using the toolkit that was coming from engineering and learning from finance and trying to use all that together. And they were very generous with their time and their knowledge. And I think that I was providing something useful also, because if not, they were not going to give me all the titles that they gave me all the time.

6:50So it was a perfect match. And it was also best in LA, so I know how to ask my wife to move. That would have been difficult. And so it was amazing. It was really a blessing to be able to work in a company like that, small enough where you were able not only to be involved in research, but also portfolio management, in trading, in marketing, in legal, in sales, in all the different disciplines because the company was small and you were developing something new. So it was great. Well, you rose all the way to co-CEO and co-CIO of the firm. And after 17 years at DFA, you decided to leave and eventually start something new with Avantis.

7:31Would you talk to us about the brief time away from the industry and what prompted that decision? Yeah, look, at some point, I thought that it was time for me to leave DFA. So I left. I just went to the chairman and say, hey, I'm resigning. Like a month later, they make it public and whatnot. And I did nothing. I did nothing like for two years and a half. You speak with a lot of people. When you say you do nothing, you don't have a job, but you're speaking with a lot of people and whatnot. But after more than two years, like two years and a half, something like that, we started Avantis. And Avantis is basically an investment company, even though we are not a company.

8:14We are a subsidiary of a larger company that was created to, we will speak later, to systematize active management. You have a strategy to have all the benefits that you have from diversification, from indexing at low cost and everything, but trying to put more financial science and more technology in order to deliver better investment solution for foreign investors. We did it with the backing of another company. That's why we're a subsidiary of a larger company called American Century that has been around for like 68 years. and i know these people for like for 25 years and so when they reach out and say do you want to help us i say yeah i would be happy if i don't have to move and if you're going if you're willing to have low fees and that's exactly what they wanted they wanted to organize and start to have low fees and so it was a perfect match so we came to market nearly six years ago five years and ten months ago with ETF, the mutual fund, and separate accounts.

9:22And here we are, 85 million later. The evolution of investing has moved from the traditional divide between pure active management and passive strategies like those pioneered by Vanguard into a new era. Firms such as DFA introduced what many call smart beta and factor tilts, which further closed the gap between active and passive. And my sense is that Avantis seems to represent potentially the next step in that evolution, which is a smarter form of passive investing that systematizes aspects of active management beyond simple factor tilts. Is this your vision for the future of the industry? And how do you think about how it benefits clients?

10:06It's a very good question. And I know that it's philosophical, but it has a lot of insights. because we always think about index funds where the portfolio manager replicates the index and start picking on the other side. But technology has allowed a lot of people that without having to do fundamental analysis, visiting companies, starting to run the world, speaking with the CFO, technology has allowed a lot of people to close that gap. So there is a lot that you can do today that years ago we would have thought that as a stock picking, then now you can do it in a very efficient way, at a very low cost to analyze securities.

10:50One of the issues that you have when you're stock picking, that you want to speak with the CEO of a company, you have to travel around the world to go and speak with that. That's a very expensive proposition. So you have a high cost to analyze securities. So you finish with portfolios that are not as well diversified and with high expense ratio. If you are able to systematize that process, bring technology, bring a computer, bring data, bring model to systematize the process, then you are analyzing those securities at a much lower cost. So what allows you to have way more diversified portfolios with low expense ratios and still have value value.

11:30And you can rebalance them on a daily basis if you need to, because the cost of analyzing the Securities is low. And so what you have seen is that a lot of people, including us, have moved this technology, creating closer to what was stock picking before, and in my opinion, making it better because we can deliver portfolios that are more diversified and low expense ratios. So yes, the gap is getting closer and closer. And so what puts a lot of pressure on the original stock picking guys, because the value add gets smaller. You have to reduce your fees, so you have to find a way to increase that value add, what is very difficult.

12:16And obviously the benefit to clients is they could theoretically get whatever they were getting before at a lower cost and potentially even better because it can be more diversified. Perfect. I agree. So Avantis aims to systematize active management, something you mentioned earlier. What does this mean beyond simply picking factors and implementing them? Yeah, look, if you have enough factors, you can always think about some of the factors. But basically, what we have been trying to do is trying to link everything to evaluation of a company. So there are 400 factors that have been documented in the literature.

13:02There is a paper called the Factor Soup that's telling you how many out there. And the paper is five years old, so probably another hundred right now. But if you are having all these factors in isolation, they really don't tell you much. For me, all this factor analysis makes sense if then I can use that knowledge and put it back into evaluation of a company. Instead of saying, I'm going to buy companies because they have low level of accruals, one factor, one variable that has been considered reform, that doesn't, low level of accruals doesn't tell me much because you can have a company that has no accruals and supposedly high accruals is bad, but you can have a company that has low accruals but makes no money, the price is very high, and the balance sheet is full of liabilities.

13:53So yes, accruals is an important variable to take into account, a factor, let's say. But I need to put that in the context of the valuation of the company. And that's what we're always trying to do, is trying to see all this research that is out there, by others and by us. I'm not shy of saying by others. There are many, many academics out there that are doing a lot of research. And trying to learn from all these variables, all these ratios that they find that have information, and try to see how can I use those ratios or these variables into a valuation framework that consider the company holistically, not just the balance sheet alone, not just the income statement, the company holistically, like if you were going to buy a company.

14:35And that's what we do day in and day out. And I'm not telling you what we do is perfect. Perfection is something impossible to reach. So tomorrow, hopefully, we do a better job. and a year after, a better job because there is knowledge all the time coming to market. And markets are always changing and as information becomes more widespread, you have to keep adapting as well to stay ahead. You are absolutely right. And regulation changes. Accounting standards 20 years ago, they're different than today. So companies have to disclose different things. Okay, how can I use all that to have a better holistic view of the company?

15:14Aventus has achieved remarkable growth. I mentioned$75 billion, and you're already up to$85 billion. What were the key factors behind that rapid scaling? Okay, that's a great question. So I always think, make life simple, I always think about the restaurant. If you want to open a restaurant and you want it to be full, you're going to have good food, good service, and low price. yeah you have very good food very good service and very low price the restaurant will be full and that's how we came to market to aguita bandages the idea was we want to have the state of the art very good investment strategies provide very good service but try to do that at very low expense ratios so we're going to provide them value added supporting the investor with good client service but doing it in a way that is attractive to them and so that's basically the recipes.

16:07There is not much more than that. Yes, we have to have the good strategy. Yes, we have to have the good service, but the fee also has to be attractive. And it works. And it works very well when you don't have as much in terms of capacity constraints. If you're running a restaurant and you only have so many seats, you can actually set your price too low, right? And it's too full. But in this space where you have not unlimited capacity, but a lot of capacity, then it's a little bit different of a formula. Yes. And you said something that is very, very important. It's when we develop a strategy, we really develop strategies thinking about having a lot of investors.

16:50We're not thinking about trying to have only two tables in the restaurant. We're trying to develop strategies that are broadly used and they can take a lot of customers. And so that's a very important point. Do you see investing as predominantly a scientific endeavor that can be systematized, or is there room for art? And where does intuition fit into all of this? I think that there is always room for art because no matter how much you systematize, how good your models are, the models are not perfect. It's getting more and more and more difficult. Imagine, if you say, Eduardo, can you reduce the time that it takes you to run one mile by one minute?

17:34Probably, yes, I can reduce one minute because I haven't been training running miles. But if I train a little bit, probably I can reduce a minute from 10 minutes to 9 minutes or whatever it is in number. But the more and the more you train, the more and more good everything is, reducing it is just getting more and more difficult. At some point, reducing one second is a big challenge because you are so good in what you do. So what is happening here is this model, the system designs, is evolving to make the art more systematic. I don't need so much art because the models are better and better, but they are never perfect.

18:17And so there is always room out there for someone that discovers something, that has some insight. Now, it's difficult. That business is difficult, very difficult, because science has evolved dramatically. Data has evolved dramatically. Computers have evolved dramatically. Not only do they have to do it very, very well, they have to do it at a low fee. Because if not, there is no business. And so it's a tough business at one, but that is true. And something you said, I think is really important, which is you can add value on both sides. You can add value in terms of systematizing what was active management into a system that can scale it much higher.

19:02And you can lower fees. And by lowering fees, you're adding value as well. If you're providing, even if it's not 100 % of what you're replacing, but 90 % of it, and you're lowering fees substantially, cutting them in half, for example, then you're already ahead. So there's room on both sides. Absolutely right. There's a concept that I've heard that I thought was interesting. It's called that the markets are being efficiently inefficient. In other words, they're not perfectly efficient, but just inefficient enough to incentivize participants to try to seek out unique insights and strive for outperformance.

19:38Is that how you think about markets? Yes, yes. You cannot have a perfectly efficient market where no one, because every time you have news and information has to get into the market. So if there is no incentive to get the information into the market, so the market cannot be efficient. There is something called Grossman-Sticklitz paradox. That basically, this is what it says. And I didn't invent it. I cannot take credit for this by any means. And so what it says is there has to be some incentive for people to incorporate the information. Now, Alex makes an announcement. How that announcement gets into the price so someone will act fast, someone is trying to do something.

20:21Now, we're speaking of marginal benefits because the market is so, so good in incorporating that information that you're speaking of very, very marginal benefits. but there has to be always at least some benefit. Why should investors expect higher returns from certain types of companies? And how do you use diversification to reduce single company risk? Let's start with the single company risk. Look, this goes back forever. So you're speaking about diversification. No, it's something that is absolutely necessary. So there is a concept in investing that is don't take unnecessary risks. Yeah. If I give you two securities, I have the same expected returns.

21:13Will you pick one or will you pick both? You probably pick both because something can happen to one. So it can be horrible news tomorrow related to one. So if you have two, you are reducing your risk. You're having the same return because you have the same expected return. but you are reducing your risk. So the concept of diversification is extremely important and has been extremely well-documented forever in finance. And so that's something that we embrace. We have tons, many, many thousands of securities in some of the portfolios, and we do that. The second concept that you ask is, do every security have some security have higher returns than others yeah and it's better than the question the other way around do we think that every security has exactly the same spectra returns no matter what's the valuation no matter how high the price is and how bad the earnings or how bad the balance sheet is, or the IWR run, do we think that every security in the market has the same expected returns?

22:29And think about some security has very high price, very bad balance sheet, so a lot of liabilities, and very bad cash flows. Do we think that security has the same return as security that is the opposite? Very good balance sheet, very good cash flows, and low price. And immediately say, no, this idea that every security in the market has the same expected returns cannot be true. And there is nothing that makes it true. There is no axioma, no law of nature that says every security will have the same expector returns. Even though the information that we view as public and available should be reflected in today's price?

23:09Yes, it should be reflected in today's price, but that doesn't mean that the expector return of Every security is the same. So look, there are tastes and preferences, for example. If all of us hate oil companies, for example, what happens with the price of oil companies? It goes down because we don't want to buy it. No one wants to buy it. And how down will it go? It will go down until someone says, well, at the price, I really like it. because it's such a low price, such a big opportunity that they like it. And the same may happen with financials or technology or one particular company. So there is always some, at least there is some preference.

23:57There are other things, by the way, that says, look, there is no reason why every security will have the same expectant. And some securities may have higher than others. The question is how to identify that, how to say which company has higher returns than which other company. And that's where all the financial science and all the data, all the analysis and all the research by many academics, by practitioners, by us, you know, everyone has different ways to do that. And I think that our way is the best we can do, the best that the market can do, because if not, we will evolve, we will get a better way.

24:38And simplistically, it's not as straightforward as the companies that have performed the best in the past have the highest expected return in the future because the price has changed relative to what I said earlier. That probably you can be sure on that because a company cannot be the best performer forever. And why is that? Because if a company is the best performer forever, at some point, the price becomes infinity. And we know that prices cannot be infinity. And so there is a concept, some concept of some mean reversion, because if something is going so well, competition will come. We'll push, you know, on the moment that you have competition, we'll push that company, we'll lower their charge for their services or by losing market share.

25:24So no, this idea that the company of the past that did very well will continue doing well forever and will become huge. No, no, no, no. That doesn't work. And it's more, the higher the price, remember, the price is higher and higher and higher, probably the more difficult time that company has to continue delivering higher returns. Because unless you increase your earnings as much as you're increasing your price, you're becoming more expensive. So as you identify these companies and that process, do you view that as alpha or is it smarter beta or is it something else? That's a great question. So you're asking is what is alpha?

26:12And for your listeners, alpha is a lingo in our industry that means that you are adding value. You're having higher returns. but you can have higher returns for example by buying instead of let's suppose that you have a balanced strategy 60 % equities 40 % fixed income yeah so you can add more value by buying more equities and less fixed income because equities have higher returns is that alpha no that's not alpha that's just changing your allocation to buy something that all of us will agree has higher returns. And so the question from Alex is, what you are doing is really identifying something that cannot be achieved with changes in asset allocation, or it's just a change in asset allocation, what you are doing?

27:04And the answer probably is a little bit of both. and why I'm saying a little bit of both. Because I cannot tell you what will be the best model of tomorrow. Whatever we think that is alpha today, identifier something today, tomorrow will maybe be in a new model. And that model may explain everything that we do today. But what we're always looking at is what companies are in the market that given their balance sheet, given their cash flows, they're trading at a relatively low price. So the price is pushed down. The price is less than what it should be if it's not because it's being pushed down by a higher discount rate.

27:54That's what we're looking in the market. And so if we look that alpha, we have alpha. If we look that it's some kind of beta, well, it's some kind of beta. But we're really looking good balance sheet, good cash flows, and low price. Those are ideal companies for us. And you touched on this a little bit, but is that edge, that investment edge, is it likely to persist or fade over time? Okay, that's a great question. So if you ask if there are companies that will continue to have higher returns than others, if there are stocks in the market that will continue having higher returns than others, they will never be the same.

28:37But if you look for companies that have good balance sheet, good cash flows, and low price, it will change because the price of the company may go up or down or the balance sheet may become bad or good. It will change. But this should be a good recipe to have higher returns. So there is always going to be some premium performance, some excess return from investing in this kind of company related to the rest of the market. If you're asking, is someone coming to the market in the future doing something better than us? So you're thinking about Avantis itself? Yeah, that can happen. Someone can come and do something better than us.

29:13And so we're not going to be, let's say, the best, but then we become the second best. What they're saying, I'm coming second in every game, and that's not good for anyone. And so our job at Avantis is trying to be on top of all the designs, on top of all the developments, and trying to improve our processes, our models, our computers, everything, so we can deliver the best we can. So people look at us and say, I want to be investing with those guys that are number one, not the number two. And so that's a goal. But the premiums in the market will always be there. There is no reason to think every security will have the same spectrum.

29:55That's no logic for that. And obviously the challenge is identifying those before. So some factors like value and profitability may seem simple or even obvious. Why do you believe these can persist as sources of excess return, even when widely known and importantly, relatively easily implementable? if you think about factors the question is you don't know if they will persist and go forward because a factor is basically a ration like for example a value factor a traditional value factor is buying company have low price to book yeah is there a reason why companies that have low price to continue doing well unless if you link that to evaluation of a company you really don't know so if you have low price to book i can give you a lot of companies have low price to balance information because they make no money and so if you make no money your price is low does it make you attractive as a as an investment yeah i think about a imagine buying cheap sushi you get a good deal by buying cheap sushi unless if you tell me that it's good quality and also cheap you cannot tell me that it's a good deal and so when you're speaking about factors in isolation you have to be a little bit worried about what kind of biases that is introducing the portfolio.

31:18Now, if you think about a lot of these factors together, you say profitability and value together, so now it gets more interesting because basically what you are doing is thinking about the valuation of the company. So if you have something that is valuations, so imagine that I give you, have a business that has very good balance sheet, very good cash flows, and the price is low. You're going to tell me that's a good investment. I never told you that investment is selling donkeys in Babylonia five, six, seven years ago, selling water in Mars 200 years from now. I never told you if I was in emerging markets, developed markets, small cap, large cap.