Episode 326 - Dr. Sunil Wahal: Applying Financial Science

28:58You did nothing. So you are inert. At the end of 2023, how much of the market portfolio would you still have? Well, it's not 100%. That number is way south, way south, like 30%. So you have to trade. Even the market portfolio trades. Passive funds are not passive. They're just less active. Vanguard that I love, I have a ton of respect for Vanguard, their total market index, that's not a passive portfolio. It's just a less active portfolio. It's terrible. It's terrible. And then you get into these silly side discussions about, well, is passive taking over the world? Is it going to cause problems?

29:37Price discovery, yada, yada, yada. I mean, that's sort of almost not the issue. There is no passive portfolio. I like that label of less active portfolios as opposed to passive. That's literally all it is. So that sounds like good news. So in your view, we don't have to give up our factor investing. No, I think implementation matters tremendously. How you do it, you've got to minimize costs. They're not zero. They can't be zero in equilibrium, but you have to minimize costs. You pay attention to turnover. You do what you can in kind. As much as you can do in kind makes a ton of difference. You trade sensibly, but you're still demanding liquidity.

30:21The idea that you can sit on the bid side of the market when you're buying and not demand liquidity, that's nonsense. That's not true. You're still demanding liquidity. We know how markets really work. You can still move prices and you do move prices when you sit on the bid side of the market. I want to move on to some questions about how institutions make investment decisions. And this is of personal interest to me because of some investment committee work that I do. So I'm really looking forward to this. how do institutions decide which investment managers to hire? Well, they have, quote unquote, a process.

30:57And the process involves an investment committee. And the process involves consultants often. Doesn't it always? There's always a consultant in this. So committees and consultants are delegated. I've sat in investment committees. You've sat in investment committees. anytime you sit on any committee. The outcome of the committee is very much a function of the quality of that committee. And if I were to sit on a committee for deciding medical issues, I'd be a disaster. Like, you don't want me on that committee. So what kind of people do we want on an investment committee? Well, we want people with expertise in financial markets.

31:44We want people with practical knowledge as well, because at the end of the day, these things have to work. And you have to worry about conflicts of interest. That's also important. Many investment committees do not look like that at all. And the agency problems in those sorts of investment committees are pretty strong. They're often quite severe. So what you see is decision-making that from your podcast perspective is not rational. And that's sort of sad. We bear the cost of it. And how do managers tend to perform after they've been hired? Well, they do exactly what you'd expect. They perform average.

32:30So what does the process look like? Well, you hire someone who's done really well in the past. and then there's some mean reversion in performance and they deliver average performance. That's not necessarily bad, but it depends on what your expectation was. If your expectation was this is going to be rent tech, good luck with that. You're not going to get rent tech. What you're going to get is the average outcome. So that's kind of what they look like. Which is exactly what you should expect. But I think people look at past three-year, five-year performance and say, yep, that manager looks good.

33:08But as you say, there's no reason to expect them to do better in the future. They should deliver average performance, which is exactly what they do. Yeah. That's actually another pet peeve of mine that a lot of professional investment setups have this very artificial focal point of a three or a five-year performance. I don't know what the difference is between three years versus two years and 11 months. I don't know how it makes a lick of difference. We don't see that in private equity, by the way, but certainly in public equities and fixed income and things like that, you can see in the flow data, if you look carefully, the moment firms or managers cross that three or five-year benchmark, you see an uptick.

33:53And that's because people have generated a focal point in their mind that says, I got to wait for five years. Why? because I don't know why, but that's why. And everybody else is doing it, so it must be right. Unnecessary. Yeah. So fund managers often get selected based on recent past performance and then deliver average performance. How do institutions decide when to fire managers? This is a little more nuanced. So you do see institutions firing managers for performance. You kind of expect that, and you want some degree of that to happen. So discipline in financial markets for investors is quite important.

34:34So I don't want that. And the equilibrium can't be that that number goes to zero. So some poor performance termination is important. But there are other reasons. People make shifts in asset allocation, and they might make them for market timing reasons, or they might make them for very legitimate strategic asset allocation reasons. You have to add an asset class, so you've got to shrink something else. That's pretty reasonable as well. There's also risk issues. I mean, you can see this if you looked at the press in the last few weeks or months, you saw WAMCO, right? The issues that arose at WAMCO, and what do you see?

35:12A flood of outflows. So that's not unexpected that there's risk associated either with personnel or compliance issues, all these kinds of things. So that answer is a bit more nuanced, I think. There's more going on. You mentioned poor performance. How much tolerance do institutions have for poor performance? There's this narrative that's been going around for decades now that they are quick to the trigger, so to speak. Very, very quick to the trigger. My co-author, Amit Goyal, he and I have worked together on a number of things for a long time now. We teamed up with a practitioner in Europe, Ramon Toll, and we surveyed.

35:56This was in collaboration with institutional investors. I should give them credit as well. We surveyed some very, very large institutions around the world to ask them about that sort of tolerance. It was bigger than I expected. It wasn't three years and you're out of here kind of thing. It was a lot longer. So there maybe reflects the sophistication of the very largest institutions, but they were tolerant. So it's clear that they had some understanding that returns are noisy and that you have to do more than that. So this goes back to your investment committee part that you're asking. Then the investment committee has to do its due diligence, which means not just looking at performance, but diving deeper into the organization and doing the due diligence and asking, do you know what you're doing?

36:47If there is underperformance, what's the cause? And can I tell whether it's random or something systematic in your process that I ought to be worried about? So they tend to be more nuanced. I like that outcome. You briefly mentioned sophistication. What effect does that have or how does that interact with underperformance tolerance? It does quite a bit. I mean, the tricky part is being able to separate signal from noise, right? That's what sophistication really gives you. People who go into a very quick, just like a little attribution system and say, oh, you underperformed because you had money in this sector or this security or this market and it underperformed, attribution is backward looking.

37:30And that's not a bad thing. Doesn't tell you anything about the future, but it does help you diagnose where the issues arose. And so I think the sophistication part is being able to separate signal from noise, right? being able to say, here's why this happened. And if you can tell that you make better decisions going forward. I think that's the key. Okay. So let's talk performance. How do manager hiring and firing decisions affect the institutional performance on average? If they're poorly done, they're going to be a drag because what's going to happen is that there are very significant transactions costs associated with turnover of managers in the institutional realm.

38:16So you see this, you go from a legacy portfolio to our target portfolio, you've got a transition over the securities. And there are firms out there that will do this for you, some of it in kind, and they'll trade the rest. But anybody who's ever sat on a trading desk will tell you that every once in a while, you might get a phone call that says, hey, so-and-so large organization is transitioning from large value to large growth, which means we got some stocks that need to be sold and stocks that need to be bought. I don't need that. That information should not enter financial markets ahead of them because it's going to cost them something.

38:57So transitions, portfolio transitions are costly. And if you ignore the marketing hype that you can do this for zero, nothing is done for zero. It costs them something. Ultimately, that's a drag on the portfolio for that institution. If it's an endowment or a foundation, you're going to be able to pay your students or your beneficiaries less. If it's a retirement plan, that's going to cost you something. Frictions are a part of life. That's so interesting. So the winning managers don't keep winning and firing losing managers, I think from your paper kind of shows that it doesn't really matter.

39:35So if they just held the managers that they had to begin with, they would have been before transition costs just as well off. That's definitely not what happens in real life though. Yeah. It's not what happens in real life. Real life is full of frictions, all kinds frictions and all you want to do is minimize those frictions and you're never going to make them go away. So I think it's a, maybe a hopeful, but a hopeless exercise at the same time to make frictions go away. Just minimize, do the best you can. How do relationships between managers, investment managers and hiring institutions, how do those relationships affect hiring decisions?

40:15You know, I have a cousin who's in the, is a doctor, he's in the dating market. And he's 30 something. And guess what he does? He gets on his dating app. And that's how he's looking to find a partner, right? But he will also take recommendations from us and say, hey, maybe you should reach out to this person. So relationships matter. They're a part of the equilibrium that we live in, and in which trades happen and finding a partner is a trade. Is it not? That's exactly what it is. So it's part of that equilibrium and relationships do matter. So they can be very helpful because knowing people allows you some information advantage.

41:00So presumably if my wife or I recommend someone to him, then there's some degree of vetting that's already happened. But they can also go the other way. You can imagine situations where so-and-so makes a recommendation and there's a side payment or that it doesn't really help you. So does it work this way in asset management? Of course it does. The relationships are key and they should be. It's not like they should be non-existent. If you just look at conferences that take place, that independent organizations like Institutional Investor run, those are essentially networking conferences, right?

41:39There's some content, but they're essentially networking conferences. So relationships matter and they matter for whether an asset manager gets hired or not. And that's fine. I think expectations are a different issue and investment committees still need to do their due diligence, but yeah, they matter. So you have this paper that looks at this and establishes that this relationship does exist, that if there are relationships between the two entities, they're more likely to get hired. How do relationship-based hires affect performance? Does the trust help them find better managers? Not really.

42:16I mean, financial markets are all about trust, right? I give you my money to manage or you give me your money to manage. That's all trust. Everything in financial markets is about trust. I pull out a dollar bill, a loony or a dollar. This is still trust, right? It's trust in the government. It's trust in institutions. So that's all it really is. So if you hire based on relationships, you don't do better, you don't do worse. You get what you get. You get about on average. And that's not necessarily a bad outcome. It depends on what your expectations were. If your expectation was, I'm connected to you and we make really good investment decisions and I expect to earn really high performance, that's great.

43:10It all depends on what your expectation is and what the counterfactual is. Because the alternative could be, oh, I handed my money to Bernie Madoff. Yeah, that's an interesting point. I think either in the abstract or the conclusion of that paper, there's an interesting comment about how those relationships seem to be a little bit one-sided. Because when these two people know each other, two entities know each other, the investment manager gains surely from fees. And the customer, the institution, kind of just gets nothing. Yeah. Not all relationships are – it's not like a marriage, which one would hope that marriages are two-sided and matching equilibria.

43:49That's not always the case. sometimes these things can be one-sided you know the data in that paper were fascinating to begin with i mean they came from this organization called i think they've been rebranded but they used to be called relationship science so very good organization a lot of intellectual capital behind it and it's amazing what they track i mean if you want to know ben how do i connect to Bill Gates, they'll find a way for you to connect to Bill Gates. It might be through two or three intermediaries, but they'll find a way to do it. If it's there, it's there. It's clever. I like it.

44:28Amazing. Let's move on to private markets a bit in this realm. How do institutions choose which private market firms to invest in? Private markets are a tough asset class. I really, really do feel for institutions and actually retail investors as well who want to get into private asset classes. I used to sit on an investment committee that had to get into private asset classes because their clients really, really wanted it. High net worth, accredited clients that really wanted to get into private markets. It's tough. I think the key aspect of what makes it really tough is what we call access. so I would like access into Andreessen Horowitz or Sequoia but guess what they're not going to give me the time of day I'm a little piddly guy and that means if I'm excluded from those because I think that there is persistence in performance and if there's any persistence and performance in private markets, it's in venture.

45:35So I'd like that, but I can't get in there. So what am I going to do? I'm going to go to other places. So this is a bit of a lemons problem. So I'm excluding the right tail, which means I'm going to be working with a truncated distribution of managers. And that's tough. So if you want exposure, that's not so easy to do. So it's all about selection from that truncated distribution. Unless, of course, if you're the Yale Endowment, you can pretty much go wherever you want. Then you have no access problem. It's really about managing that. It's tough, not easy. One of the things that I took away from that paper is that institutions are pretty willing to invest with first-time general partners or people with maybe not a lot of experience managing money.

46:25What's the relationship there between first-time or young GPs and future performance of their funds? Yeah. So this has to do with how competitive markets are. So your investment committee needs to allocate, let's say, 10 % to private markets. You don't get access to Andreessen Horowitz. So you're going to find whatever you can get access to. And maybe you have a consultant that Cambridge or whoever comes along and says, We can get you access to some of what we think are good ones. It's still not going to be Andreessen Horowitz. And we also think you should take a look at these new folks on the market.

47:02So this is a new organization. They used to be at XYZ Farm, and now they've created their own organization. You have demand for investment services. You want exposure to private markets. This is your best shot at getting that exposure. So what we see in the data are exactly that. You see large institutions allocate large amounts of capital in aggregate, individually small, but in aggregate large, to what we call first-time GPs or young GPs that don't have a performance history. because you can't see their performance, right? The fund lasts for 10 years. You don't really get to see multiples or IRs or anything like that until 10 years later.

47:50So three years into the process, you have no idea what things look like. So it's a little bit shooting in the dark. And in my view, it's basically demand and supply. Allocations to private markets are up. You need to get it invested. Where are you going to go? Subject to the access problem, you're going to go to some new folks and out of the woodwork are going to come some new folks. GPs are going to get created to meet that demand. Yeah, that's wild. So there's so much demand for private assets that a GP can kind of just spin up a story and launch a fund without a track record and still attract capital.

48:26I mean, for some GPs, there'll be, I mean, this is why, again, it comes back to the devil is in the details, right? Selection is incredibly important. And some of those, I open up a shop, call myself a GP. Am I going to attract capital? Maybe, maybe not. But if I've spent 10 years or 20 years at a private fund, I have a better shot at attracting capital. And some fraction of those will do well and some fraction won't. So what are the main lessons that institutional investors should take from your research? You mean overall in terms of selection and stuff? Yeah. I mean, honestly, if you ask me what the single most important thing is that an investment committee at an institution does is it's selection and monitoring.

49:14That is really the key to what they do because it's not like they manage the individual portfolios. Everything is farmed out to professional organizations that do that. It's selection and monitoring. And selection and monitoring is very much a function of the quality of the investment committee and the people that you have, if they know what they're doing, if they're well-versed in the science and in the practice, I emphasize, I think both are incredibly important, the science and the practice, then they have a better shot. I mean, if you take something like the Yale Endowment and compare it to, there's a story in today's Bloomberg about the Harvard Endowment and its relative underperformance over the last two decades or so and how poor it looks and how expensive it's been for Harvard University.

50:07If you contrast the two, the Harvard management companies rife with or was rife with agency problems. And if you look at Yale setup or Stanford setup, it's not the case. So people matter and getting the right people in the right places really matters. Now, your research shows this, and you mentioned you've been on investment committees. You've probably seen it firsthand. Investment committees feel this need to hire and terminate managers. I don't know if it's to look like they're doing the right thing, their stakeholders, or I don't know what it is. Maybe you have an idea on that too, but how would you communicate to an investment committee?

50:46Or maybe how have you communicated to committees that you've been on that doing that, that turnover of managers may not be the right thing to do. Yeah, this is a behavioral lesson, isn't it? Not doing something is doing something. It really is. That's essentially what it is. So you pay attention. I mean, I've gone in front of investment committees who've asked me similar questions and I keep telling them the same thing. Not doing something is doing something. You do your job. Your job is hire, monitor, monitor, monitor, monitor, monitor, and then really if necessary, fire and then hire again. But you don't have to do this.

51:28It's not like a haircut. You might need a haircut every month, but you don't need to do this process every month. It's understanding that the signal to noise ratio in financial markets is very low. That's really what it boils down to. It takes a long time for us to learn from prices. But that means you have to find other ways to learn. So you have to understand your investment managers, understand what they're doing, how they're doing it, whether you believe their investment thesis or not, what it relies on, where the pain points are. Do they reside in turnover? Do they reside in personnel? Like you have the WAMCO example from last month, if a person leaves, is that, well, Bill Gross, if you think about that.

52:18Person leaves, everything collapses. You don't want that kind of risk. So a committee's job, think about risk. That's their one job. And what lessons from this research do you think apply to retail investors? It's kind of interesting. When I think about institutional investors, so I think about the head of an IC, right? The head of an investment committee is a retail investor. They're an individual. So they have the same biases and preferences that individual investors have as well. So I think every lesson applies to individuals as well. The institutional and retail arena are not that different.

53:00The exception might be taxes. From a retail investor's perspective, you've got to pay a lot more attention to taxes. Or as I tell my students, hey, you get to buy that car with after cost and after tax dollars. So pay attention. Is it a QDI or not? Are you going to get a tax bill when you didn't even trade? Pay attention to those kinds of things. All of that matters. But other than that, it's the same set of agency problems, the same set of behavioral ideas that exist at the institutional level as exist at the retail level. I think it's exactly the same thing. I want to ask one more on institutions, just based on what you're talking about there.

53:47You mentioned costs and taxes in the case of retail investors. We also talked about the Harvard story where they've paid some pretty substantial fees. How important are costs for an institution managing a portfolio? Oh, they're incredibly important. I mean, it's a friction. All of these sorts of things are frictions. So if you think about an asset manager that delivers a portfolio, an asset manager that delivers a portfolio basically gives you a set of portfolio weights, and those portfolio weights change over time. You as a latent owner, that's what you have. Okay, so what are the frictions and costs associated with delivering you that set of portfolio weights?

54:28Can I do it myself? The answer is no, because I need research, I need data, I need to be able to trade this stuff, etc., etc. So all of those costs essentially get summed up in two or three different places. One, they get summed up in an expense ratio, and that's explicit. You get to see that. The second place they get summed up is in the after-cost returns, because the costs associated with trading as an investor, you don't get to see those because you just get returns that are net of those costs. So those are invisible. And then the last part is the tax part. So all of those get summed up. It's incredibly important to pay attention to those.

55:14Things that look really good on a pre-tax, pre-cost basis, Momentum is the classic example. There are many organizations that run momentum portfolios. I don't own a single one. I would not touch it with a barge pole, having worked in this area, built portfolios, and knowing what I know about trading over the last 28 years or so. That's a recipe for writing checks to Jane Street, which they need checks written to, but I don't need to write them more checks than they get. I got to dig into that a little bit more. I know you've got a paper on momentum and we didn't have it in our list of stuff to discuss, but since you mentioned it, it's a question that I get all the time about, Ben, you talk about factor investing, but what about momentum?

56:02Should I be investing in momentum? You said you wouldn't touch it. Can you just expand a little bit on why? Let's talk about the other strategy. Take value, take profitability, take all of those. Those are anchored strategies. They're anchored to a fundamental. Momentum is inherently an unanchored strategy. The only thing it cares about is the prior price, and the prior price itself is unanchored. So any strategy like that is unanchored to fundamentals, which means it can have very volatile outcomes. And because it's not anchored, it goes up and down, right? You can see the volatility in the outcomes, and the ship goes all over the place as the tide rises and as it falls.

56:45And that means it can have crashes. And we know that there are, if you Google momentum crashes, you'll find, I don't know, probably at least a dozen papers by well-known authors that say, we study momentum crashes. Well, they crash because they're unanchored. So momentum as a strategy, because it's unanchored, generates very volatile outcomes. And for an investor, okay, just take Piddley on me. I have some objectives. My objectives are I got to pay my kids college tuition. And I don't want to be in a situation where I can't write your college tuition bill because my momentum portfolio really crashed.

57:27I just don't want that. I can't plan for it. So uncertain outcomes, or as Ken French likes to describe the unexpected, right? The unexpected is evil. I agree with him 100%. I don't want that. And that's what momentum gives you. And on top of that, it's incredibly expensive to trade. No matter how well you adjust your trading bands and portfolio breaks and yada, yada, yada, there are a dozen ways in which you could try to minimize turnover. But if you minimize turnover, you're giving up expected returns. That's kind of the way it works. The first part of your answer almost sounded like a risk-based explanation for why the momentum premium exists.

58:13If you think of risk in that context. Right. Yeah. I mean, anything that's unanchored makes me nervous. You can think about other strategies, right? So think about like low vol or betting against beta, those kinds of strategies. They're kind of unanchored, but they're based on prescriptive models. So they say cap M betas, low beta stocks earn higher returns than they should. Well, it's true in the data, but you tell me when's the last time you believed in the cap M to begin with. So am I going to use that as an investment strategy? From my perspective, the answer is no, because it's telling me to reject a prescriptive model, but I already agreed to reject the prescriptive model.

59:04Yeah, interesting. Basically, like the five-factor model explains the low volatility kind of argument, right? Yeah. Man, I wasn't expecting to go into momentum or low vol, but here we are. I don't think anyone's going to complain about it. On institutions, I had one other question in my head. Why don't more institutions just index? You got to be seen to be doing something. That's the hard part. That's the agency friction. It's really hard to make it away. Do we see some institutions do that? The answer is yes. So when you say index, you mean, in my sense, like less active, right? Not a passive portfolio, just less active, hold market cap and leave it alone and update.

59:47There are institutions that do that, but there's not a ton of them. The reason is because you got to be seen to be doing something. Yeah, it really is incredible. That seems like the right answer, but there's been this indexing revolution for retail investors. Meanwhile, you look around at institutional portfolios and not a whole lot of them are at least not pure index. Maybe they're using more indexing or less less active strategies, but I don't know, a lot of pretty serious active stuff going on out there. You look at the Nakumo data, for example, I mean, you look at those portfolio allocations, you know, I don't know what the heck you people are doing, but you're all over the place.

1:00:27It's just what it is. It just costs you something. Yeah. Super interesting. And also it's a shame. Like you mentioned earlier, what's the objective of these institutions? If it's an endowment or retirement plan or whatever, They're affecting real people. And a lot of those dollars are going toward investment managers and alternative investment funds and all that kind of stuff. We talked earlier about competition in private markets. How competitive is the market for mutual funds? I mean, asset managers in general, I think whether it's mutual funds or ETFs or whatever the wrapper or the delivery vehicle is, the asset management market in general, I think is quite competitive.

1:01:09I have a paper with a co-author that looks at this in mutual funds, but you can see this in other markets as well. And think about the way that this business works is you create a product, you deliver that product to investors. The cost of entry and exit for an investor into a product is relatively low with the exception of taxes. And if you think about the marginal cost of the producer, there's some fixed costs for me to set up a fund complex and create a fund. But then the marginal costs, especially in days now, like where the technology is really good, the marginal costs are not that high. So what does that mean?

1:01:51You get new entrants, new entrants, generate competition. And the more competition there is, I mean, I don't know anybody who doesn't like competition. You get better products at better prices, quality improvement, price improvement? What's not to like? I love it. I wish there are even more electric vehicle producers because what happens? Prices fall, quality goes up. That's great. Yeah, it makes a lot of sense. You've got a pretty new paper. It's mind-blowing. This is a cool paper. How does diversification in mutual fund investors affect the performance of the funds that they invest in? It's actually really intuitive, really, really intuitive.

1:02:32it. So suppose you're setting up a business, okay? You're setting up a business, ignore mutual funds. You're just setting up a business and you have what we call a corner client. Your corner client represents 50 % of your business from your perspective. That's a pretty important client from your perspective. That's a big client. Okay. Now suppose you have a customer who's a small client of yours. So 5%. But from their perspective, their investment in your business is 80%. So they're a small client to you. You are a big provider to them. Okay. So now let's go back to your business. Your business, you are worried that that client that you have that represents 50 % of your revenue stream says, I don't like you anymore.

1:03:28I'm going elsewhere. You're going to be scared of that, right? You don't want that. So what do you want? Well, you want a diversified client base. That makes sense, right? I don't know any business that says, oh, I only want one client and it's the government. Okay. I'm just kidding about the government pop, but you know what I mean? You don't want a concentrated client base. You want a diversified client base. So that's pretty easy. So if you have a diversified client base, if you have one client leave, no big deal, not as big a deal. The other ones give you more business. That's a good thing.

1:04:05That's the benefit of diversification from your perspective. And now what about that client that you had, that small fraction of you? So 5 % of you, but 80 % of their investment is going in there. into your business, they're sitting there worried, hey, I'm giving 80 % of my capital to Cameron Ben, but they've got this other client who's 50 % of their business. If they run away, uh-oh, that's going to be bad for me. This is the mutual fund business in a nutshell. It simply says, if you diversify across clients, different client types with different liquidity needs, then you're less likely to have what we call bank runs, where one person runs for the exit and everybody else is, oh, crap, they might run for the exit.

1:05:00We got to go, too. And that can be very harmful to investors. So there was an investment committee I used to sit on. I sat on this committee for, I don't know, many years, at least half a dozen years. And one of the things I really liked about the CIO of this IC was that every quarterly meeting, what she insisted on was that we start with a look at all the asset managers that we use for this organization. and all the individual products, the funds or ETFs or whatever, and what their flows had looked like over the last quarter, the last year, the last three years, the last five years, whatever the numbers were.

1:05:43And she insisted on it. I thought that was an incredibly intelligent organizational thing to do. And it made a lot of sense because you sort of sit there and go, okay, if I'm seeing a fund that's been bleeding assets, I know it's going to cost me in terms of performance. And it might cost me in terms of tax bills, at least in a mutual fund setup. Part of fiduciary due diligence, you want to be able to track those things. And so the flip side of that from the fiduciary perspective is the fund would like to have a diversified client base. And the investor would like to have the fund to have a diversified client base as well.

1:06:27And so you don't want to put all your money in one basket. It's pretty simple, but it shows up in financial markets. It is simple, but I had not spent a lot of time thinking about, I mean, it's like the bank run concept sort of for mutual funds. I think you mentioned that as an example. It's fascinating. Is there a relationship though, between performance and concentration of investor base in the mutual fund space? For sure. So the nice thing about mutual funds is that we can detect things with so much precision. So we measure flows on a daily basis, like every day. So we know what each fund's flow looks like on that day.

1:07:05And we know, so there are measures of how diversified their client base is. So when they have outflows, so imagine two funds, one with a very diversified client base and one with a not diversified client base. They both have outflows. So they're both going to incur costs, right? Because securities have to be sold. And that cost is going to be borne by shareholders who remain in the fund. So if I stay in the fund, I'm going to pay one way or the other. And so that's why in my example earlier, my illustration in the IC, we keep track of flows because we don't want to pay those costs. But all else equal, if you give me two funds, one that has a more diversified client base and one that has a less diversified client base, I want the one with a more diversified client base because I pay less.

1:07:54I still pay, but I pay less. So how do you find that fund with a more diversified investor base? We use a bunch of different proxies for it. So one proxy is you can look at, this is getting nerdy and technical, but the covariance of flows across share classes, Right. Funds will have half a dozen share classes, retirement class A, retirement class B, retail, advisor, blah, blah, blah. So you can look at the covariance of flows across those. That's sort of the simplest way to do it. That tells you. Think about 2008. If you're in 2008 and all your clients head for the exit or COVID, all your clients head for the exit.

1:08:40that's not what you want. You want some patient clients to stay there to offset. And you also want some clients to say, you know what, all of you are heading for the exit. I think expected returns are high. I'm going to allocate more capital. That's what you want. That paper, like I mentioned, when we started talking about it, I found it to be just mind blowing because it's not a concept that I'd thought about in the context of mutual funds. But like you said earlier, it makes so much intuitive sense once you explain it. But not easy to have been the first one to really think about it. So that's why it's a cool paper.

1:09:13How well have actively managed global equity mutual funds performed for retail investors? A lot of it depends on what your benchmark is. So there are two ways to think about a benchmark. One way to think about a benchmark is sort of the way that the investment community thinks about a benchmark, which might be like ACQUI or whatever. Pick a benchmark that way. I think a better way to think about a benchmark from the downstream investors perspective is what financial objectives were you trying to meet and were you able to meet those financial objectives? is. So what do active global equity funds do?

1:09:53They invest around the world. Will they give you the global equity premium? Depending on how they are managed, how concentrated they are, how they handle costs, et cetera, et cetera. Some of them will, some of them won't. In general, the more active they are, the more costly it is for them to implement their approaches in general, not for every single one. And those costs add up. So I think of a continuum of less active to most active, as opposed to totally passive, which doesn't really even exist. And as you move along that scale, you hope that expected returns rise and that costs don't rise commensurate with expected returns.

1:10:44Unfortunately, they do. So it's a tough gig. It's not going to be easy. But that's, again, where the implementation comes in. And what does this suggest about the efficiency of markets outside the U.S.? So this is something we hear from investors often, right? That the argument, I can't tell you how many times I've heard this, where somebody says, well, we believe that U.S. markets are really, really efficient, but emerging markets are not, or markets outside the U.S. are not. And so we're going to use active management there and not active management or passive management here. And my response is, well, you're really using active management everywhere.

1:11:23It's just a matter of degree. And now you tell me you have a prior that let's say emerging markets are less efficient than U.S. markets, but you show me the evidence that that's the case. And when you ask for evidence, you don't get it. And that's the issue. So it's really hard to tell whether emerging markets, for example, are less efficient than US markets. I can't tell. And I think that if I asked many of my esteemed colleagues in academia, they would probably tell you, at least those who think about these sorts of things, they would tell you the same thing. It's really hard to tell. Or in a Bayesian sense, it's a very diffuse prior.

1:12:14Like, I can't update this. That's a very nerdy answer. I love it. Okay, we got two more questions for you. You've worked with firms like Dimensional, you mentioned, for many years and now with Aventis. How important do you think it is for financial economists to spend time with practitioners? So back when I was young and I didn't have this much gray hair, I remember reading a little piece of writing by Fisher Black, who I think was in some sort of practitioner journal, maybe a financial analyst journal or journal portfolio management, something like that, in which he basically made the case that every self-respecting economist, financial economist, should spend some time in industry.

1:13:00And the case was the following. It says, if you're an academic, you're interested in prices. Where do prices come from? Prices come from agents trading. They come from trading. They don't magically appear. There's no such thing as a Walrasian auctioneer. So they come from real trading. And so if you want to understand prices, you should spend some time and figure out where prices really come from with industry. And then he flipped the argument around the other way and said, OK, if you're a practitioner, you're watching markets clear every single day. You're watching trading, behavior, et cetera, et cetera.

1:13:37But sometimes you are so into the details that you miss the structural aspect of what should be rather than what is. And because you miss that structural aspect, it's worth your while to occasionally walk over to your local university or wherever and attend a seminar or two and learn to think how financial economists think about this because they think about it in a structural way. So from my perspective, I think it's incredibly important. And I'm fortunate. I thank many people for the fact that I've been able to do this over the last three decades or so. My first engagement came with, I don't know if you guys know, but Ted Aronson.

1:14:27Ted Aronson's original firm, Aronson Johnson & Ortiz. Ted Aronson, I think, is still ahead of the CFA, maybe? one of the original value investors. I think it's incredibly important for academics to spend time in industry and vice versa. You learn a lot. And this plethora of research that we have that says, oh, here's another signal and here are some other expected returns that you can, or an alpha that you can generate. A lot of that would go away when you realize, It doesn't work that way. It's not how prices were formed. Our final question for you, Sunil. How do you define success in your life?

1:15:09Isn't that an interesting question? I'll tell you what a good day is. So a good day is when I can exercise my brain and do some research. A good day is when I can get into the canyons in Arizona and get into some wilderness, hopefully with a friend or two, get some hikes in. And a good day is when I come home and have a fun dinner with my wife and talk to my kids who are neither of them are here. They're both in colleges and various places. So that's a good day. So what's a good life is an accumulation of good days. So that's the way I think about it. So you asked success, and I described happiness.

1:15:48And success and happiness are not necessarily the same thing. You can be very successful and unhappy. I kind of prefer to connect them. Like I have both of them at the same time. It's kind of like value and profitability, if you will. So what's success? Success is an accumulation of good days. Leave the world a better place than when you found it and be happy while doing it. I think for me, it's pretty simple. Very cool. Well, having you join us has made today a good day. So, Sunil, great to see you again. And thanks for joining us. Thank you so much. It was fun.