Expected Returns

Episode 220: Jonathan Berk and Jules van Binsbergen: The Arithmetic of Active Management, Revisited

Jules Van Binsbergen is the Nippon Life Professor in Finance at the Wharton School of the University of Pennsylvania and conducts both theoretical and empirical research in finance. His current work focuses on the relationship between financial markets and the macro economy, as well as the organization, skill and performance of financial intermediaries.

He received his PhD from the Fuqua School of Business at Duke University. He joined the Wharton School in 2014 and previously taught at the Graduate School of Business of Stanford University and at the Kellogg School of Management of Northwestern University. He is a research associate at the NBER and a research fellow at the CEPR and is a co-host of the podcast series All Else Equal: Making Better Business Decisions.

Jonathan Berk is the A.P. Giannini Professor of Finance at Stanford Graduate School of Business (GSB). His research is primarily theoretical in nature and covers a broad range of topics in finance, including delegated money management; the pricing of financial assets; valuing a firm’s growth potential; the capital structure decision; and the interaction between labor markets and financial markets. He has also explored individual rationality in an experimental setting.

Professor Berk has coauthored two finance textbooks: Corporate Finance and Fundamentals in Finance. The first edition of Corporate Finance is the most successful first edition textbook ever published in financial economics, and is a standard text in almost all top MBA programs around the world. At Stanford GSB, he teaches courses in Institutional Money Management and Critical Analytical Thinking.

Professor Berk’s research is internationally recognized and has won numerous awards, including the Stephen A. Ross Prize in Financial Economics, the TIAA-CREF Paul A. Samuelson Award, the Smith Breeden Prize, Best Paper of the Year in the Review of Financial Studies, and the FAME Research Prize. His article, “A Critique of Size-Related Anomalies,” was selected as one of the two best papers ever published in the Review of Financial Studies, and was also honored as one of the 100 seminal papers published by Oxford University Press. In recognition of his influence on the practice of finance, he has received the Graham and Dodd Award of Excellence, the Roger F. Murray Prize, and the Bernstein Fabozzi/Jacobs Levy Award. Professor also cohosts the "All Else Equal" podcast that focuses on making better decisions.


Do you feel like you have a good grasp of financial markets? Think again! In this episode, we take a plunge into the world of financial markets with experts Jules van Binsbergen and Jonathan Berk. Jules is a Professor of Finance at the Wharton School of the University of Pennsylvania and Jonathan is a Professor of Finance at Stanford Graduate School of Business. They also host a popular podcast called Else Equal, which explores the science and strategy of making better financial decisions, and have written several academic papers that challenge the status quo. In our conversation, we discuss their research on the relationship between manager skill and fund performance, the best ways to measure performance, and reasons why benefits are in favour of the managers. We also explore the dogma surrounding mutual funds, the differences between active and passive management, and how to measure efficient capital markets. Listeners will also hear perspectives that challenge their understanding of capital markets and viewpoints that completely disagree with previous guests. Although we have covered this topic before in previous episodes, this conversation will fundamentally change the way you view financial markets and how to think about them.


Key Points From This Episode:

  • What information fund performance contains about manager skill. (0:04:04)

  • Reasons why manager skill and performance are unrelated. (0:04:59)

  • We learn how manager skills should be measured. (0:06:57)

  • How to choose the appropriate benchmark to measure value added. (0:09:26)

  • Find out if you can use factor-mimicking portfolios to measure risk-adjusted returns. (0:12:05)

  • Whether funds that directly target risk factors can be used as an investable benchmark. (0:16:35)

  • What the skill of active managers are when skill is measured as value-added. (0:20:52)

  • The proportion of value-added between security selection and market timing. (0:23:20)

  • Discussion about how persistence manifests when it is measured by value-added. (0:25:43)

  • Find out if investors should analyze mutual fund companies as opposed to managers. (0:32:36)

  • Discover why research has focused on individual security pricing and not on evaluating manager skill. (0:34:25)

  • We unpack the reasons why it's a zero net alpha as opposed to a negative net alpha in equilibrium. (0:38:19)

  • We delve into why the research took so long to apply rational expectations to fund investors as with the stock market. (0:42:46)

  • An explanation of how equilibrium zero net alpha fits into Bill Sharpe's arithmetic of active management. (0:48:16)

  • Who benefits from the high amount of skill available within the sector. (0:51:11)

  • Whether the increase in millionaires around the world drives inequality. (0:56:12)

  • Hear if it is possible to identify skilled fund managers before the benefits of their skills are absorbed by fund size. (01:01:41)

  • The implications on efficient market hypothesis for the stock market. (01:05:36)

  • Advice for investors, considering that the benefits of skill are in favour of managers. (01:08:37)

  • Details about their research on how multi-factor asset pricing models are not representative of risk. (01:12:45)

  • We end the show by learning how our guests define success in their lives. (01:19:08)


Read the Transcript:

Ben Felix: This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision making from two Canadians. We are hosted by me, Benjamin Felix, and Cameron Passmore, Portfolio Managers at PWL Capital.

Cameron Passmore: Welcome to episode 220. This is a pretty fun ride of an episode with two guests. We have Jonathan Berk and Jules Van Binsbergen join us. And this was one, I think it's fair to say, Ben, spicy conversation with two academics that have done a ton of research and are not afraid to speak their minds.

Ben Felix: This is one of the episodes where you realize how little you actually know about financial markets. We, during this conversation, overturned so many things that most listeners will have taken for granted in the way that they think about how markets work. And that's amazing. That's the best kind of episode. We were just talking, Cameron, about how our favorite episodes are the ones that fundamentally change the way that you see the world. And I'm fairly certain that this episode will do that for most of our listeners.

Cameron Passmore: And I'm pretty sure listeners are going to say, "Oh, I wish you'd asked them this or ask them that as a follow-up." We were time constrained, we had a list we wanted to get through, so we did our best. But boy, there's a lot of things that will, what do we say, threw a grenade down your belief system. It's quite something.

Ben Felix: We spent most of our conversation talking about their ... I guess the whole conversation really talking about their work on the relationship between manager skill and fund performance. Now I said the relationship, the reality is there is no relationship, which is the basis of their research. People may remember that Fama told us that the best evidence of efficient markets ... When he was on our podcast. The best evidence for efficient capital markets is that active management is a negative sum game. This conversation completely turns that on its head and you'll hear Jonathan directly disagree with Fama, which in terms of the way that people think about financial markets, that's about as fundamental as it gets.

Cameron Passmore: Jonathan's a Professor of Finance at Stanford Graduate School of Business, co-authored two very popular textbooks on Corporate Finance and Fundamentals in Finance. He was the associate editor of the Journal of Finance from 2000-2008, has a PhD from Yale, and is originally from Johannesburg, South Africa. And then Jules is a Professor of Finance at the Wharton School of the University of Pennsylvania, and has a PhD from Duke. And if you're interested, you can listen to their podcast called All Else Equal: Making Better Decisions. It's shorter episodes, about 30 minute episodes. Really good. Really well done.

Ben Felix: They have episodes with people like John Cochrane and Cliff Asness, other people that we've had on our podcast. It's very different from our podcast, but if you're a finance nerd, it's definitely a good listen. They have a very good dynamic between the two of them and that comes out in this conversation as well. We were a bit worried, this is our first episode ever, I think with two guests. Full episode. I guess we've had short conversations with two guests. But at the beginning we all chatted about how we were going to alternate, but it was seamless. They work extremely well together.

Cameron Passmore: Very well. Anything else, Ben, add to this?

Ben Felix: No. Get ready to have your beliefs changed.

Cameron Passmore: Exactly. All right. So with that, here's our conversation with Jonathan Berk and Jules Van Binsbergen.

Ben Felix: Jonathan Berk and Jules Van Binsbergen, welcome to the Rational Reminder Podcast.

Jonathan Berk: Well, we're very happy to be here.

Jules Van Binsbergen: It's an honor. Thank you for inviting us.

Ben Felix: Excellent. All right. We're going to start with your work on manager skill, which we are chatting before this, it's mind blowing work. Anyway.

Jonathan Berk: Just keep saying that, nobody else does.

Ben Felix: It really is. And it's important. What information does fund a performance measured by net or gross alpha contain about manager skill?

Jonathan Berk: Well, I think ... I don't want to say what Jules and I's contribution is. I don't think that's up to us to say. But I will say this, that understanding that the alpha does not measure manager skill, and then instead, net alpha is a comment on investors, not a comment on managers. And understanding that gross alpha doesn't measure anything. And that the value added measures the manager's skill. I think that it's an important point. And Jules and I ... You know about our podcast. It's an application of what economists called Equilibrium Thinking and what we call on the podcast, All Else Equal thinking. Which is once you think in equilibrium, these things are obvious. If you don't think in equilibrium, you make the mistake.

Ben Felix: Now the relationship between manager scale and performance is a pretty commonly held belief. Why is it that they're actually unrelated?

Jonathan Berk: Well, I think that the insight is that backward looking over time, you are learning from the performance of a manager about what the skill of the manager is. But given the fact that the size of the fund adjusts to what we think today is the level of the skill of the manager, going forward, it doesn't have that predictive power. And I think that the best way, the analogy, to think about is ... Think about how a stock is priced. If I asked you ... If your grandmother came to you and said, "You should really buy Apple stock because I love Apple products, I have every Apple product that's ever been produced." And so for that reason, she should put all your money in Apple stock. What you would say is, "Well, listen, the price of the stock Apple is already high, so that going forward you should not be expecting to get much alpha out of this."

In fact, going forward, the return is just a measure of the riskiness of the stock. And so I think our paper is just an application of exactly the same insight, but then to another asset, which is the mutual fund. So even though backward looking performance of a stock gets you to the high price, and therefore a high price today tells you that a firm is a good company. Going forward, there is not that same relationship. Same thing when managers, backward looking, over time you learn about their skill, but you should not expect a net alpha going forward to be an indication of the skill of the manager, because investors have competed away that rent by all wanting to be with that manager. They've already flocked and flushed that manager with money. The fund is already big. Just as much as with the stock to stock price is already high. And so going forward, that's going to just erode to the performance.

Jonathan Berk: It's a classic All Else Equal mistake, you ignore the fact that if you have found a positive alpha opportunity, other people would've found it too. Nobody ... Everybody wants a positive opportunity. And so obviously supply and demand says that can't survive.

Cameron Passmore: That's pretty clear. So how should manager skill be measured?

Jonathan Berk: Well, as we say in the paper, there's a pretty clear measure. It's called ... What we defined to be value added, which is the gross alpha multiplied by the size of the fund. So if you think of gross alpha as the return the manager makes in excess of the benchmark on the fund he's managing. So it is the extra return he's adding multiply by how much money he manages, that gives you the total amount of money he has added, or as we like to say, extracted from markets. That's the measurement of skill. When you look at that way, it's pretty obvious that that is in fact the measurement of the skill because that's what the manager makes.

Jules Van Binsbergen: And so what I think our paper does ... And I think there was quite a bit of confusion about it before, is there are two separate questions. The first question is, is the manager, by trading, generating a pie? Yes or no? The second question is, if we divide that pie, who do you think gets most of it? And so if we are in a situation where there are lots of investors that are all competing for the skill of very few managers, then of course the investors don't have any negotiation power in that relationship. The few managers that have the skill do. And so in a situation where there's this amount of asymmetry in negotiation power, who do you think goes home with the results from the skill? It's going to be the manager themselves. And so for that reason, the fees that they collect is going to be exactly equal to that value addage in equilibrium, which is the same thing as saying that the net alpha to the investors is zero.

Jonathan Berk: I always joke ... It's not totally a joke when I say Mark was right, because what did Mark say? Mark said, "All the rents accrue to labor." Well that's exactly modern financial markets. All the benefits of your labor accrued to you, capital is not in short supply. There is no reason capital should earn rents. There's lots of competition in the capital markets, that competition competes away any rents and all the rents go to labor. And the modern economy is exactly that. That's why Steve Jobs was so wealthy. That's why Mark Zuckerberg is so wealthy. It's because they get the rents, the capitalists don't get the rents.

Ben Felix: You mentioned a benchmark being needed to measure value added. How do you choose which benchmark to use?

Jules Van Binsbergen: Well, so I think that there are a whole bunch of different candidates that the literature had out there before. And so what we really wanted to do was ... We noticed that when we teach the MBA students or any other investment management class, we always use this lose phrase, that says something like, "Oh, you should just put it in passive. Or you should just index your money." But we wanted to give some empirical content to what that statement actually means. And we came to the conclusion that it means something very different depending on what time sample we're looking at. For example, before 1977, it's actually pretty difficult to index your money at all. It's very difficult to find index funds. Then in 1977, Vanguard introduces the S&P 500 index fund, and so then you can least index that. But since then, I think a whole bunch of different funds have been introduced as passive funds.

By the way, I also think there's a deeper point here where we really need to question the clean separation between active and passive. There's some very interesting recent work that shows that ETFs, which everybody loves to call passive, are really very active investment options. And just because it's called passive and it sounds popular, doesn't mean it's not ... It is passive. It's actually very active. But what we did in terms of choosing the benchmark is that, well, at least everybody can always invest their money in some combination of Vanguard index funds at very low fees. If we take that linear combination as at least always the counterfactual you can pick, as an active fund, you need to beat that. That is going to be the bar that we hold you accountable against. And we found a set of very nice ... I think it was 13, 11 or 13 funds, I think, that Vanguard has.

And then there's still a little bit of a question that you can still ask. For example, Vanguard charges fees too, on their index funds, so just indexing doesn't actually mean that you get the S&P 500 index itself. There's still transaction costs, there's still management costs, and then there's still the distinction between ... Vanguard has three different groups that they offer this to. They offer to retail investors that get a different feed, and admiral investors that get a different feed to institutional investors. Which one is a reasonable benchmark? And I think also there we have quite some nice insights because the price of diversification, which is what the fee is that Vanguard charges, is therefore different depending on the amounts of money that you invest. And so you can tailor the benchmark a little bit there, even depending on what sort of investor you are thinking about.

Cameron Passmore: Interesting. Why not use factor mimicking portfolios, like the Fama-French or Fama-French Carhart to measure risk adjusted returns?

Jonathan Berk: There's a two part answer to your question. If you believe the factor portfolios are really a risk measure, then it's perfectly fine to use it. But there's not a shred of evidence that these things are risk. It's just an invented term. You notice it has a positive excess return and you call it risk. I think that that is academics getting papers published. The second reason to use the factor mimicking portfolios is to say, "Forget about where they measure risk. They are the next best alternative investment opportunity." If you think about what an alpha is, it's how much am I adding over my next best investment opportunity?

And you can interpret the factor portfolios as the next best investment opportunity. And I think there Jules and I take major issue with it, because the factor portfolios don't take into account transaction costs. They're not investible portfolios. So, if you could find a factor portfolio that's an investible portfolio ... If Vanguard said, "Okay, we are going to offer this factor portfolio as an investment portfolio," then that is fine to use. But you can't use a factor portfolio that doesn't charge transaction costs and compare that to an active manager who has transaction costs, that's just not a fair comparison.

Jules Van Binsbergen: Let me add two more things to that. The first thing that we do in our papers that we say, let's take these Fama-French factors as strategies and let's see how well they do against the Vanguard benchmarks. And what you find is that you have this enormous outperformance of the Fama-French factors compared to these real time available Vanguard index funds. And so that just shows you that holding managers accountable against such a high bar may not be the right thing to do. And so what I always joke about to the students in class, and it very much adds to what Jonathan said was, that this was the procedure. There's a manager who invents a strategy that works well. There's an academic that sorts the same thing, and says, "Hey, there's a puzzle, there's an anomaly, and publishes the paper."

The next academic says, "Hey, these puzzles cannot exist in rational markets. Let's just put it on the right hand side of the equation of the risk model without there being any justification for putting it there." And then the next academic says, "Actually, I think if we evaluate that first manager against this new risk model, there's absolutely no evidence that they can outperform anything." And we go through that cycle over and over and over again until, I think for hedge funds at this point, we're at 13 factor models to try to evaluate managers against. And I think that has gotten so far away for what a reasonable counterfactual investment opportunity for investors is, that I think we need to try to bridge that gap and better understand what that gap means. And so we think that the counterfactual investment opportunity set is a perfectly reasonable bar in many ways. And we discussed it today in class actually, even that combination of Vanguard funds may be too sophisticated for a lot of investors. The ideal thing we would want to measure is this.

Suppose that there is an individual investor that puts all their money in one stock and it's also the one stock of the company they work for. It's the most undiversified investment that you can possibly imagine because when the firm fails they lose and their job and their money. Now suppose that that person can be convinced to put their money into an active fund, then you can already see that the whole discussion of active versus passive becomes entirely irrelevant because so much value for this investor is added just by getting that person to move from that counterfactual to this new counterfactual. You could even say, "Is the linear combination of Vanguard index funds a reasonable counterfactual for all investors everywhere in the world?" I think we, in that sense, actually already picked quite a high bar for the active managers to beat. But it's at least it is an investible opportunity that a rational person could do, which I think is already a huge improvement over the Fama-French factors, which is not an actual investment opportunity that a rational investor could do.

Ben Felix: I have a follow-up question, but I just want to make a comment for our listeners because you guys were very hard on factor models just now. Those were not baseless comments. And we're going to ask you about your empirical work on asset pricing models later. But our listeners ... Because we often talk about risk factors, I know our listeners are going to be like, "What are they talking about?" My follow-up question, what about using funds that do directly target risk factors, like products from Avantas or Dimensional, as the investible benchmark for this purpose?

Jules Van Binsbergen: Yeah, you could. But first of all, those are active funds and they charge active fees. Let's start with that. And so those are part of the active space, which in some sense is already an interesting fact, given the brains behind DFAR. That's something to at least consider. But if they become tradeable factors and tradeable portfolios, then the after fee returns of those things, you could choose to take those as the counterfactual investment opportunity for investors. Again, taking into account the different levels of fees that different share classes need to pay, if that's relevant.

Jonathan Berk: I think the important thing to understand is we have to be very clear on what passive investment is. A passive ... When we teach it, one of the things we say about the market portfolio is if you truly hold the market portfolio, you are ensuring yourself against insider trading or asymmetric information to the full extent that you can. And so that defies the passive. If you move away from that, for whatever reason, it could be a totally legitimate reason, like I am working in tech and I don't want the tech part of the passive portfolio. It's a totally legitimate reason to move away from the passive portfolio. But if you move away from that, it's no longer a passive portfolio.

Ben Felix: Can you reiterate that or say it a different way? I think what you just said was important.

Jonathan Berk: Are you talking about why the market portfolio is an optimal portfolio to hold if you don't have information?

Ben Felix: Yes.

Jonathan Berk: This is, I think, a very important point. We obviously emphasize when we teach finance. But I think that in general this is not emphasized enough. Which is if you are trading, anytime you trade and you don't have good information, if the other side of the person has information, you lose. That's the nature of the game. If somebody has more information than you and you trade with them, you are on the losing side. The extent to which ... That if you know don't have information and there is information out there, what you want to do is don't trade. And the way you don't trade is you buy the market portfolio, you buy it once, you hold it and you sell it. And actually it's more subtle in that.

It's not just don't trade. But why the market? Why not just any portfolio? The reason why the market is particularly good is you're buying it in the economy wide weights. You're not emphasizing any stock. Somebody with information is going to emphasize some stock over another stock. If you are on the other side of that trade, that means you have the opposite. You're doing exactly opposite to them and you will lose. So what you want to make sure is you're never exactly opposite of somebody with information. How do you do that? You just buy the market. Because the market, you are not opposite to anything. You're in the weights of the whole economy.

Jules Van Binsbergen: And I think another thing that brings up, that I think isn't appreciated enough is, the way that we define the market, and the way that Jonathan just did, there's only one way to do that. It's the market capitalization weighted total of everything. And so in that sense, if we had the index fund space and the index fund space was about holding the market, there should only be one product that is offered that offers to market. But I think at this point we can agree that there are tons of different passive funds. And I will think that one of the funniest one is, that one of the largest index fund providers in the country, if you read what their perspectives of their value fund says, it says, "Value stocks ..."

It's a value index fund because it's rules-based and they say it's passive. But then it says, "Value stocks are generally associated with underpriced stocks that give you an outsized return." But is that really ... As soon as you sell an outsized return, can we still call this a passive strategy? Does this still qualifies it? It sounds really like an active fund to me. That's at least how active funds generally were being sold. And so holding the market and passive, that actually ... There's now a wedge between those two statements. It's not the same thing anymore. It used to be the same thing. But in today's investment world it isn't anymore.

Ben Felix: We did do an episode with Adriana Robertson talking about her paper, Passive In Name Only, which is, I think, a lot of what you're referring to. Back to skill. When we measure skill as value added, what does the evidence say about the skill of active managers?

Jonathan Berk: Oh, it's overwhelming. Managers are highly skilled. Something like they add something like 3,000,000 ... An average manager adds $3,000,000 a year. There's important caveat to that. On average, they add 3,000,000 a year, but the distribution is highly skewed. Most managers destroy value, but a few managers add an enormous amount of value. And the reason of course is that the skilled managers have all the money. There are lots and lots of managers that are destroying value, but they're not managing much money, they're not doing a lot of damage. Most of the money is concentrated in highly skilled managers.

Jules Van Binsbergen: And so I think the point Jonathan just brought up is also very important actually in the passive space. Sometimes you see people say statements like, "The fact that there are passive funds that charge really high fees is evidence that markets are really ... The mutual fund markets really are irrational and that there's a problem with them." But we cannot just look at funds on a fund by fund basis and then see how it's allocated.

If we find out that the vast majority of money is allocated to the very large index funds that do it correctly, and yes, we can find a few where this allocation didn't optimally happen, that still means that there's overwhelming evidence in the direction that the capital is going to the places where it needs to go. Even if sometimes ... Yes. And I think Jonathan has some work on that. For the very worst performing funds, there is a little bit of evidence that people stick with those a little bit too long. That may also be true, but that doesn't mean that we should throw out the baby with the bathwater. Generally, the best managers get the most funds and therefore add the most value.

Jonathan Berk: Look, the world is full of charlatans, the idea that we're going to get rid of all the charlatans in the world is a little naive. So the question is not, "Can I find a fund where there's basically a manager ripping off investors?" Of course, you're going to be able to find a fund. You can find charlatan doctors, and you can find charlatans in all areas of the economy. That's not the interesting question. The question is are those funds, are those charlatans, large? Are they ripping off many investors? And the answer is definitely not. The vast majority of investors are in funds that are adding a lot of value.

Cameron Passmore: Interesting. Of those highly skilled managers, you're talking about Jonathan, how much of the value that the ad is coming from a security selection? How much of market timing?

Jonathan Berk: That we can't answer. Because we didn't look at ... And this is a very subtle point, I'm not sure we want to get into it over here, I could briefly talk about it. But stocks selection in aggregate looks like market timing. Imagine you ... Let's go back to the argument that the market portfolio ... If you are holding the market portfolio, it means you've insured yourself against asymmetric information. Think about that in aggregates. Say, I have a bunch of managers that are selecting stocks. And let's say everybody's rational in this world. If that's the case in aggregate, you could just hold the market. And so then you say to me, "Well that doesn't make any sense." How does it possible for managers to select stocks, and in aggregate make money? And yet at the same time, investors can ensure themselves against this asymmetric, by holding the market?

And the answer of course is market timing. That in aggregate, all the managers by themselves are selecting stocks, investors are holding the markets. So in aggregate, on the other side of that, all these stock selectors aggregate to the market because they're trading the same. But you realize sometimes those managers have information and they're in the market, in aggregate, and sometimes the same information and they're out of the market in aggregate. Add investors on the other side, and that's how they make the extra money. That's when I say to you, investors can ensure themselves against ... By holding the market, and by not trading, they do have to trade twice. They have to trade when they get in and they have to when they get out. When they trade, they will lose money against informal investors. But the point is you want to minimize that, so you don't have to trade more than twice. This is a subtle point. It's hard to do this on a one hour podcast to actually go through these subtle points.

Ben Felix: I think you made the point well though. Passive investors are still trading against active managers. And if we know active managers are extracting value from the market, passive measures must be losing on those trades. One of the ways that ... From the research, like the Fama-French and Carhart research, that a lot of people are familiar with, they're looking at persistence in skill. How does persistence look when it's measured by value added?

Jules Van Binsbergen: Very interestingly, it's very persistent. I think we need to clarify two things. There are two measures that we look at. One, as Jonathan mentioned, is investor rationality measured by the net alpha. And that is the persistent that most of the literature is looked at. And then there's the persistence and the value added that you asked about. And the interesting thing is that in the competitive equilibrium that we were talking about, they're very clear predictions about both. The net alpha is not supposed to be persistent. And the reason is, as soon as there is a net alpha, investors flock to that manager, give the manager more money, and therefore drive the net alpha back down.

The lack of persistence there is in support of the mechanism. But for the value added, we're supposed to be seeing large persistence. And so what we did was we sorted managers into deciles, we looked at what their value added was up until that point, and then evaluate what the value added is for the next one, three, five, seven, and 10 years. And what we find is that even up to 10 years into the future, we see that the managers that had value added in the past still keep on having large value added in the future. So indeed, in that sense, value added is also very persistent.

Jonathan Berk: We actually show something else as well, which I found astonishing, once we saw it, because it's such a comment on the rationale of investors. Instead of sorting by past value added, we sorted by managerial compensation. What's managerial compensation? It's the fee they charge times the size of the fund. Now, the thing about managerial compensation is investors determine that. Investors determine managerial compensation. They decide how big the fund is. And if you sort by that ... Now we're sorting by who investors want to put their money in, who investors think are the bests managers. If we sort by that, you still get incredible persistence, but it's better. Investors are better able to sort managers than past performance. The investors are able to figure out who the good managers are and then put their money with the good managers.

Ben Felix: That changes the view of mutual fund investors as irrational performance chasers.

Jonathan Berk: That's ... I think one of the biggest misconceptions in the entire academic literature. The idea that mutual fund investors are naive. Every single thing we have done in this space has pointed to the rationality of mutual fund investors to the point that Jules and I now think that mutual fund investors are more rational than stock investors. If you look in the stock market, you see more evidence of irrationality than you do in the mutual fund space.

Ben Felix: Wow.

Jonathan Berk: Look, I've just told, these investors know who the good managers are they direct their money. Now we have other work that explains why you might expect that. One of the things you have to understand is there is an intermedium, meaning the mutual fund ... The executors of the mutual fund company, and they're making decisions about how they deploy their talent in the company. And those decisions are very highly informed. You know your own employees and you correctly put your employees in the right place. When I give my money to the mutual fund company, to the mutual fund, I'm taking advantage of the fact that they are assessing this talent. And that intermediary, I think, makes the market much more rational than stocks where you don't have that intermediary.

Jules Van Binsbergen: Another way of saying that is, we know there's a flow performance relationship. That means you can build large AUM for yourself as a manager by outperforming. But that flow performance channel through investor money is a relatively slow process. It takes you quite a few years of outperformance before you've grown your fund. The quickest way to make a promotion as a manager is if your mutual fund company decides that when Peter Lynch retires, you are going to be assigned to that particular mutual fund. And that then means you get tons of money allocated to you.

But there's a downside too, there's a catch too, which is investors don't have to stay with that fund if they don't buy it. To the extent that the mutual fund company says, "This is the successor of Peter Lynch." And then after we see that the fund size drops by double digit percentage numbers, that does mean that people are saying, "Yes, we believe this manager is very good. But not as good as Peter Lynch, and therefore we are withdrawing some of the capital from that particular fund." And so we indeed did see that fund shrink after Peter Lynch retired and a new manager was assigned.

Jonathan Berk: Well, Jules, let's just be clear, the mutual fund company's doing their best to replace Peter Lynch with somebody as good as Peter Lynch and they had the most information to do that. They know their own employees. And actually if you look at it, in the end of course majority shrunk a lot, but initially it did look like the initial manager took off after Peter Lynch were able to continue doing what Peter Lynch was doing. Because it takes ... You don't see the big outflow until two or three manager replacements after Peter Lynch. And I don't know why they were replacing all those managers, but there is some extent. But of course that's a very specific case. And of course I think it's very unfair to use that in the sense that ... I think everybody agrees that Peter Lynch is the greatest money manager that's ever lived. And so obviously it's going to be impossible to replace somebody of that quality. But generally what we find in our research is that mutual fund companies are exceptionally good at replacing managers with managers that are just as good.

Jules Van Binsbergen: Well, so to be precise, what we find is that when a mutual fund company turns over the managers and reallocates the funds between them, we see that the value added after goes up. Meaning that the relocation actually works. I do think that it is not hard to believe that the mutual fund management companies have more information about the managers than the investors do, because one of the things that investors have as a downside is that one of the most important information sources that they have is realized returns. And so that's what they can learn from.

They can do a little bit more research on the manager, but the mutual fund management company can observe every action and every choice, and can also observe the rationale behind it and the whole decision making process. And most importantly, what the mutual fund management company can observe is all the stocks that were not picked, and which were considered, but were turned down for particular reasons. They can really see the decision making process. And that I think gives them an informational advantage. And indeed the data suggests that that informational advantage pays off. More value added after management turnover, on average.

Cameron Passmore: Should investors be spending more effort on analyzing mutual fund companies as opposed to managers?

Jonathan Berk: Yes. And they do.

Cameron Passmore: They do?

Jonathan Berk: In other words ...

Cameron Passmore: Interesting.

Jonathan Berk: Look at the space. I think the space is dominated by five companies. So, it is. So yes, they invested at the beginning. Most investors give their money to one of five companies.

Cameron Passmore: Huh? That's true. I'm just curious if they think about the fund company level as opposed to the individual manager level.

Jonathan Berk: We actually have evidence on that. When we wrote the paper, here was the issue, we showed definitively that when the firm ... When the mutual fund company makes managerial changes, the value added goes up. But if you think about that, how does that benefit the company? I mean, right, the value added goes up, but their fees on the sides of the funds. So unless investors see the managerial change and react by investing more money, they're not going to get any benefit out of this. But that's exactly what you see. When the mutual fund company makes managerial changes, more money is invested, and exactly the equilibrium, because of course if more money was not invested, there'd be a positive alpha. I told you value added went up. That's exactly what you see. You see an inflow of funds driving the alpha down to zero and the mutual fund company grabs the rents, then so yes, investors are aware of this.

Jules Van Binsbergen: The flow relationship even holds at the firm level in some sense. That's one way to think about it. The whole mutual fund firm grows in its total AUM across all funds in response to the reallocation. Even if the percentages in the fees stay the same because the total AUM has grown, the total revenue of the mutual fund company has just gone up.

Cameron Passmore: Why has so much of the literature and study on this then focus on individual security pricing and not at evaluating manager skill?

Jonathan Berk: Well, let's just start with the fact the literature was very confused. Until 20 years ago, we wouldn't even question the idea that alpha measure skill. Nobody would even question. Today, this day, Jules and I still referee papers where they use alpha as a measure of skill. And so if you are going to make that fundamental mistake, it's going to lead to a lot of inferences. Of course, it led to the main inference that managers weren't skilled. And that was dogma in the industry, in my entire academic career. Until I wrote the paper that was dogma, which is that managers weren't skilled and investors were irrational.

And so if you come to it from that perspective, you don't even think about asking questions like why do mutual funds companies exist? Why are there only five of them? What are mutual fund executives doing? How is talent allocated? One of the reasons Jules and I wrote that paper about talent was ... Obviously we were interested in mutual funds, but as economists we were much more interested in a bigger question, which is how do firms operate? How does talent get allocated within the firm? This is a big economic question. It's very difficult to get data on. Most firms are not opaque. The great thing about a mutual fund company is its transparency. You can use data to actually answer a question like that.

Jules Van Binsbergen: And to add to your question, I think that one argument that was made was because no mutual fund manager can do anything, or beat the market if you will, that would then be supportive evidence for the fact that the stock market was perfectly, what has been called, efficient. But the problem with that argument is that arguing that the stock market for that reason is perfectly efficient, comes at a humongous cost. Which says you must argue then at the same time that the entire multi-trillion dollar mutual fund market is completely irrational and everything that happens there is therefore not efficient. And so given the fact that there's so many investors that operate in both markets at the same time, it's weird to argue that when they operate in one market, they do everything perfectly and everything works. And then when they move to this other markets, suddenly they lose their head and everything about it is irrational.

And so for that reason, I think that the insight ... The impossibility of efficient markets that we need to have active managers that get prices to where they need to be, that also has large real economic value, by the way, I think is an important one. But it also implies that there is a place for these active managers and therefore the truth is much more in the middle. And we can even argue, as Jonathan said earlier, once we move to more rationality for mutual funds and less rationality and more frictions, if you will, in regular stock markets, you can even ask the question, are there more frictions in mutual fund markets or are there more frictions or irrationality in stock markets? And I think that the literature should spend much more time trying to evaluate the relative rationality of the two as opposed to picking the two corner solutions where we just say stock markets are perfectly rational and perfectly efficient. And mutual fund markets are just, in the entire enterprises, irrational.

Jonathan Berk: Jules spoke about it earlier, but I just want to reemphasize, that the argument that is used for what people call the efficient market hypothesis, if you use that argument in mutual funds, then a logical result is the alpha is zero. So making the argument that if I see a zero or negative alpha in the mutual fund space, that means stock marks are efficient, is a nonsense, right? Because you expect to get zero alpha or negative ...

Jules Van Binsbergen: Zero net alpha.

Jonathan Berk: Zero net alpha. That's the implication of efficient markets. You expect to get it. You do not expect to get a positive alpha.

Ben Felix: Can you talk more about why it's a zero net alpha as opposed to a negative net alpha, in equilibrium?

Jonathan Berk: Well, remember what net alpha measures, net alpha measures the rationality of investors. When I teach this, I do the following. I say to students, "If, let's say, we saw a positive net alpha in mutual funds, what would that mean? That would mean markets weren't competitive. That would mean they were positive net present value opportunities on the table. Investors had an ability to make money and they weren't taking those opportunities. And that only could happen if markets weren't competitive. And what does it mean if the net alpha is negative? That means investors are making mistakes, they're investing in negative net present value opportunities. They are committing too much money to mutual buyers. That's what a negative net alpha means. And a zero net alpha means the market is both competitive and the investors are fully rational. Now the question is, do we in fact see a negative net alpha? And one of the things Jules and I noticed is that much of the negative net alpha derives from two mistakes.

The first mistake is not taking into account transaction costs. If you compare a manager that is paying transaction costs against a factor that isn't, of course it's going to be negative. That's the first reason why you see a negative net alpha. The other reason ... And this I think is a pretty important observation. The other reason is the dataset that the academic studies used. For reasons that Jules and I cannot fathom, the first study only looked at domestic US funds. I don't mean funds sold in the United States, I mean funds that only invest in US stocks. If you do that, you cut the data set by two thirds. You throw out one third of the data set, for no apparent reason at all. It's like international investment.

Jules Van Binsbergen: You throw out two thirds of the dataset. You're left with one third of the data.

Jonathan Berk: Sorry. Right. Rule number one, when you're teaching PhD students, it's never throw out data. Or if you're going to throw out data, you have to have a very good reason. If you put that data back in there and you measure the fund against a transactable benchmark, you get a net alpha of zero. Exactly what you expect. Which is consistent again with the fact that mutual fund investors are actually highly rational.

Jules Van Binsbergen: To add to that, I think one helpful way to think about it is stock selection and mutual fund selection are exactly the same game. When I say to you, "Buy Apple, because Apple is a good company," you say, "Wait a second. First tell me how much I'm paying for Apple because maybe Apple is already overpriced, so it's a good company, I admit, but that doesn't mean that it's a good deal. Maybe I'm paying too much for it and therefore the alpha could be negative." The alpha is negative if the price is too high. The alpha is positive if the price is too low. Now let's translate exactly that insight to the mutual funds. The question you need to ask is, given the quality of the manager, just like the quality of the company, how big is the fund? If the fund is too big, expect a negative net alpha.

If you think the fund is too small, and therefore not enough capital has been allocated to this fund, then you can just as much with the low stock price for Apple, you buy the stock, in this case, you get into the mutual fund, and then you can still expect to get some alpha going forward. And there are examples in literature where you can do exercises like this. Although I think a lot of the papers that do this, they draw to wrong conclusion. For example, let me give you a very simple example that I think is a very nice example. Managers with foreign sounding names have positive alphas. Now, some people say, "Well, from that I can learn that foreign sounding managers are skilled." And the answer is no, remember, the net alpha teaches you something about the rationality of investors, not about the skill level. What have I learned from this? Investors are not allocating enough money to managers that have foreign sounding names.

Ben Felix: That's wild. That's crazy.

Jules Van Binsbergen: Whether or not they're better or worse than managers with English sounding names is an entirely different question. The only question that we've established is have these managers been allocated enough money given how good they are? That's all we've established.

Cameron Passmore: Wow.

Ben Felix: It's crazy stuff. When I read your paper, your 2004 paper, Jonathan, for the first time, this stuff blew my mind then. And it's blowing my mind again now. Why do you think that it took so long for the literature, for your papers I guess, to apply rational expectations both to fund investors just as we do at the stock market?

Jonathan Berk: That's a deep question. That's a deep question about how research works. You could ask another question ... Rick and I wrote the paper, the first draft I think was 2002 maybe, that's 20 years ago. And it's still not fully appreciated. 20 years have passed. I think the answer is that we have an idealized view of how research occurs. And I don't think this is just economics, I think this is everything. But that in fact the reality is very different. That human behavior is much more important determinants of how research progresses. And so if a famous professor gets up and has a theory that might be ... At the time people are not ... Don't want to question ... Correct me or becomes dogma, but turns out to be wrong, it takes an extremely long time for that theory to be overturned. And so there's the famous Kuhn book where he talks about ... The Structure of Scientific Revolutions.

But I think that the Kuhn idea that you have to wait for the young people to arrive for the idea to die is actually naive. I think it's much worse than that. I think it's the ideas might outlive the people because of fads, and ... It's hard for me not to comment on this in today's society. And Jules and I are going to do a podcast on the question of freedom of speech and things like that. Jules and I did not think of freedom of speech as some kind of religion. It's in the constitution, it's part of our culture. We think of it as a critical part of making a business or research work. Coming back to your point, I think that within academics, within research, we don't have enough people who actually are willing to step out with new ideas.

There's a lot of pressure in society to not say new things, even it comes down to small things like, "Oh, I'm going to sound like an idiot, so I don't want to say it." And I think that largely explains why research progresses in this way. There's this book on cancer, called The Emperor of All Maladies. And one of the things when you read that book, you realize how you could have a medical procedure, a medical treatment for which there is no basis whatsoever, not one ounce of evidence that it works last a hundred years. The way research works is this idea of just, "Oh, we'll figure it out. We'll always do the good thing." It doesn't work that way."

Jules Van Binsbergen: And I think it's even worse. I think sometimes we even mean revert. Sometimes we take steps backwards. I had PhD students in a project in my class evaluate some of these things and come up with the best examples they could find. At some point there was this theory that because out of a rotting piece of meat you just get flies coming out of it. There was this theory that life would just start to arise automatically out of other previously living things. And so for a while that was the theory. And then at some point somebody did an experiment, then one of them was covered so that the air could get through, but the flies couldn't get in, and the other one wasn't. And then you would see the life come through in one and not in the other. And people said, "You see, it's really not the theory that we had before." That was then rejected, but decades later just returned again.

We just went back to that theory even though it had already been disproven before. And so I think research isn't even a linear path towards progress. Sometimes we make several steps backwards. And I think the other thing that Jonathan has said, which I think is incredibly important, is we need to be very careful that scientific consensus that arrives from moral pressure is no consensus at all. When 97% of scientists say that the sun evolves around the Earth due to moral pressure, we can say there's scientific consensus on this issue, but that scientific consensus is meaningless. And so we want to listen to the voices of Galileo Galilei, who are the outliers, the ones that everybody thinks are morally reprehensible. Everything was wrong with them. He was accused of everything under the sun and even was forced to recant what he said earlier. And so I hope that, particularly in the US and these days, we can sustain an environment where people are allowed to bring forward non-conventional theories so that we remain open to them and keep testing them and use the scientific method to make progress with them.

Jonathan Berk: One of the things ... You could take a step back and think about, and we should think about, is how physics was able to transform to theories that was so seemingly outrageous. I can think of no other feel where somebody stepped up to the plate, and said, "Time is a function of how fast you move. Or a particle is not a particle, it's a wave function." I think it's astonishing how physics was able to transform in a way that I don't think many fields could. And the question we can ask ourselves is, what is it about physics that they were able to do that?

Ben Felix: It's fascinating hearing you guys talk about this because it probably took me eight years of having gone through the CFA program and done an MBA with finance concentration and all that stuff to learn about your work on manager skill. Whereas before that it was all, "Well look, the net alphas are negative." That commentary with that context is very interesting.

Cameron Passmore: Many listeners will be familiar with Bill Sharpe's Arithmetic of Active Management. How does the equilibrium zero net alpha fit in that framework?

Jonathan Berk: Bill Sharpe is a good friend of mine. But unfortunately, on this particular one, he is wrong. And let me be very explicit. Let me just explain Bill Sharpe's arithmetic so we all know what it means. Bill Sharpe's arithmetic is luck. Sum of everybody's portfolio is the market portfolio. The sum of active managers ... The active managers have a portfolio. Then we have everybody else, that sums to the market portfolio. That means that sum of everybody's portfolio can't outperform the market portfolio. The only way active managers could make money is if they take it away from everybody else. It has to be a zero sum game.

Where I think Bill made a mistake was, he forgot about the fact that people pay for liquidity. Another way of saying that is, I know ... Let's say I am a fully rational passive investor, and I say, "I would like to invest in the market." I know that there's going to be two places where I'm going to expose myself to information, once when I get in and once when I get out. And I take that as given. And I say that's the cost of transacting. And that breaks the arithmetic because of course when I go in and I get out, they're going to make money off me. Even though, I'm always holding the markets. When I get up, I buy the markets. When I get out, I buy the markets.

Jules Van Binsbergen: By the way, there are lots of investors that are getting in every month and getting out every month. Everybody does ... In their passive investing are saving for retirement. There are tons of investors that are constantly getting in and out of passive vehicles. This is not just a theoretical thing, this is also pragmatically very important. But secondly, I think the second argument that needs to be made here is you cannot just split up the world and say the passive investors hold to market. And I call everybody else active. And then I just equate those active managers ... Or these active investors with active managers. There's still a very large group of people in the ones that don't hold a market that actually are not active mutual fund managers. So even there, the arithmetic already doesn't add up. You need to think about all the other people that are holding individual stocks or are doing things on their own or think that they're smarter than everybody else and are in the market.

Those investors obviously can get exploited by that behavior as well. And so an active manager that trades against those people will also win because they have an informational advantage. But then on top of that, indeed even all the passive ones need to get in and out all the time. And that leads to further distortion of the summing up constrained. And so it's a good benchmark to think about. But as with many benchmarks that we have in finance, say for example Modigliani-Miller and other things, as soon as we've established the benchmark, we do have to immediately start thinking about the frictions so that the frictions make the equilibrium interesting, not the frictionless benchmark that was proposed here.

Ben Felix: We've established from your research that managers are in fact skilled and that investors are able to identify them ex-ante, who actually benefits from that skill though?

Jules Van Binsbergen: The managers. The example I do in class is a very simple one. And we actually did it in one of the podcasts, in the All Else Equal podcast, when we talked about capitalism. I first say to the students, "We've all spent the whole day making shoes. We're a hundred students into class, 99 people make a left shoe, one person makes a right shoe. One of the students in the class made a right shoe." And then I say, "A pair of shoes goes for $100." I also made a left shoe. And so I approached the only person with the right shoe and they said, "Karl Marx said that how much time you spent making something determines its value. We both spent the whole day making it. I propose that we both get 50 bucks and then we sell the one pair of shoes."

And then regardless of who the student is or what their political beliefs are, by the way, they always turn down that offer. They immediately say, "No way, you're not getting $50 for this." And I say, "Well, explain to me why you feel that that's the case." And then they say, "Well, I have all negotiation power because they're 98 other people in the class that I could go to for a left shoe. I have the thing in short supply, you have the thing in large supply. So who gets all the rents?" Then I say, "Well, how much do you think the person with the right shoe gets?" And invariably the class comes to the conclusion that it's somewhere between 99 and a hundred dollars, so that almost everything goes to the person who has the one right shoe.

And so after saying that, I say to them, "Well, to do investment management, I need two inputs. Instead of the left and the right shoe. I need investible money and I need mutual fund managers skill. Now tell me which one of the two is in short supply." And so very quickly they come to the conclusion that having investible money makes you in no way special, because everybody has investible money. And so it's very difficult to call that a skill in short supply. The skill of the manager is something that is in short supply. Who is going to go home with the rents from this process? Just like the right shoe owner, they go home with everything, the manager goes home with everything. And so the value added is entirely absorbed by the manager in terms of fees. Which by the way is exactly the same reason for why the net alpha is zero. Because in this left and right shoe example, the net alpha is the amount of money the left shoe owner goes home with. And that's going to be close to nothing. So there you have it.

Jonathan Berk: Listening to Jules talk, it's just fascinating because there's all this angst in the world right now about how unfair the world is and everything else. And if I think 100 years ago you turned around to somebody and you described the world today, they would've said we were living in paradise. Think about it, 100 years ago, there was a situation where capital earned rents. Rich people, they happened tojust be rich because their parents were rich, were able to earn rents on their capital. Completely gone away. There is no way capitalists earn rents. If you want to earn a positive alpha on your money, you have to work damn hard at it. Go and try find an investment that you can do. It's exactly the utopia that people wanted, which is the only people that make money are the people who provide a skill in short supply. People who provide something nobody else can provide that everybody wants. And the way the utopia is, those people get the value of what they do.

Jules Van Binsbergen: No. And so the other thing to add to that is if you think about how many people directly or indirectly through their pension plans are capital owners today. Almost everybody in the economy today that participates in the pension plan is actually capital owner. And I think that one ... It's a bit of a tongue in cheek, but if you look at how underfunded all the pension plans are, I think that being a capital owner doesn't exactly give you the fantastic rents that everybody thought it was going to give.

Because otherwise it should be relatively simple to fund all these pension plans, if to where all these excessively large rents on capital owners. And so all of the pension plans are all struggling to get an extra percent alpha too on their investments. The rates of return on capital are so astronomically low. Interest rates are finally starting to come back up, but they've been at such low levels for such a long time. And even stock markets in terms of their future returns have ultra low levels. So I think that's all consistent with the idea that the returns on capital are just not very high anymore.

Jonathan Berk: Think about the world we live in where some guy with no capital whatsoever, but a good idea, can easily raise capital to open a business and make himself wealth. That's what most people would think of utopia 100 years ago. We have access to capital like that. Anybody who has this access to capital.

Cameron Passmore: It's interesting, Jonathan, you mentioned that, because I listened to that episode on capitalism this morning and then I turned on the news and there's a new report out from one of the big banks citing the large increase in the number of millionaires around the world. And then that bridged into a discussion about how this is increasing the inequity in society, which is directly in contrast of your conversation you have with Professor Cochrane on that episode.

Jules Van Binsbergen: Well, can we first start to establish that if inflation rates are 10, 20, 30%, that the number of millionaires will go up? Just because ... Let's start with that because the value of that million is not the same as it used to be, so let's start with that. But I do think there's quite a bit of research on wealth inequality and I think there are very important measurement issues that need to be taken into account. For example, when we establish wealth inequality, some researchers here at the Wharton School, it matters a lot where the social security is measured as part of the wealth of American households. Yes or no? If you count it, then the increase in wealth inequality is really not that big at all.

And particularly if you go to this wealth inequality database by Zucman and Saez and so forth, if you plot the wealth and equality numbers, not just for the last decade or two, but since say 1850, you see this enormous decrease in it. Exactly consistent with the statements Jonathan just made. And then we can debate a bit whether at the end it comes back up a little bit depending on how you take social security into account. But even though I think we should have a serious debate about wealth inequality and whether that's a problem, what we should do about, it's also fine to acknowledge how much progress in that sense has been made and how much it's been decreased since say 1850.

Jonathan Berk: I can't resist. Obviously these newspapers need to sell newspapers, so nothing better than to a headline that says more millionaires and all this other stuff. But they're just make following observation. One of the big differentiators in our society in terms of wealth inequality is whether you have a bachelor's degree or whether you don't. And those same people constantly complain about wealth inequality have just proposed to forgive student loans. Now what is that? There is a strict transfer of wealth from people at bachelor's degrees, two bachelors degrees. And what do you think that's going to do to the wealth inequality? It's going to increase wealth inequality.

It's very hard for me to listen to this, these self-serving debates and newspaper selling things. When in fact the data is overwhelming, especially on a worldwide basis. What are we ... The same people that complain about immigration and worry about the environment and the whole world are suddenly, "Oh no, it's wealth inequality in the United States." Well, if you really care about the world and you care about everything, what are you focusing on the United States for? Let's talk about wealth inequality in the world, which is at an all time low. Again, 100 years ago, if you told people the fraction of the world that is living out of poverty, they would've told you we lived in utopia. It would've been inconceivable to them the fraction of the world today that living out of poverty. Everybody eats today. What has been achieved is absolutely astonishing, on a worldwide basis now.

Jules Van Binsbergen: If you see the whole world as one country, indeed, and the wealth inequality inside that country has gone down tremendously, because a lot of developing markets didn't really ... Not have any wealth a while ago and now they have lots of wealth. And so all of that wealth inequality has decreased. And arguably that has come at the cost of somewhat higher wealth inequality in developed countries. But again, if you care about the world and you care about everybody in the world, then maybe that is worth it because there are many people that have been lifted out of poverty in the last couple decades alone. And indeed, if you look at world hunger, we're going to reach a lot of targets for world hunger much quicker than people thought. Now we'll see how the recent energy crisis and inflation crisis and the war in Ukraine and all of those things are going to contribute potentially to reversing that trend. But we were on a pretty good path before all of these things hit. That is for sure true.

Jonathan Berk: I don't know if you've read this book by Steve Koonin, called Unsettled. But in there, there's a graph that's just fascinating. And again, people don't appreciate this. He graphs the number ... I think it's the ... Either the number of deaths or the costs of natural disasters over the last 100 years. And that graph is precipitously down. It's exponentially declining over the last 100 years. And just take a step back, think about how much welfare has increased by the fact that the effect of natural disasters has been so minimized over the last 100 years. At the same time, we're hearing people telling us that the natural disasters ... It's the end of the world with all these natural disasters. It's an appreciation of the fact that the welfare that we've created has been enormous and that we should be celebrating the world we live in precisely because of the equality and the lives saved and the quality of life.

Jules Van Binsbergen: Because it's not a given that we can keep it. If we're not careful with it, we could lose it again. There have been many events in human history where people have turned on each other and where ... There are many recent examples of countries where changes in regimes have really led to large decrease. Think about Venezuela, what happened there? It's not a given that we can keep it. We have to appreciate what we have. Doesn't mean we shouldn't improve it if we can. That's also important. But let's not throw away the baby with the bathwater. That seems important too. Let's keep what was working before and let's try to improve the things that weren't working.

Ben Felix: I want to bring our conversation back to a topic that seems relatively much less important than what we just veered off into. Is it possible to identify skilled fund managers before the benefits of their skill are absorbed by the size of their fund?

Jules Van Binsbergen: Well, I gave you an example of that. If it's the case that you think people are making a mistake. And so if a foreign sounding name means that people are not on the ball and therefore they're too slow in at capital allocation, just as much as if you think that the information about Apple that just came out is not properly incorporated by the market in the price because they're making a mistake and there's a good reason to believe they make a mistake, you can bet on that. If you think that there's a mistake made in the capital allocation process, and there are reasons for why people are irrational in the decision making process, then you could try to exploit that. But it does have to be something where you know that the market ... You know something better than the market, either because the market is making a mistake or because something before they do. Just as this with stocks.

Now that said, we do see that the flow performance relationship, which is this equilibrating mechanism, it is not as fast as with stock markets. With stock markets or with bond markets, I think within an hour or 90 minutes, generally we see prices quickly converge to the new equilibrium level that they have. With the flows of the mutual funds, there are some frictions there that can sometimes take a little longer for the flows to happen. They still happen. But I think that the flow performance relationship peaks after six months or something, three to six months. The little bit of evidence that we do have on alpha predictability is therefore at the very short horizon, but it very quickly mean reverts and disappears.

Jonathan Berk: If you're willing to move your money every three months, there is evidence that you could exploit the fact that invest ... The flow performance relationship is slower. But that said, if you have inside information, then of course you can make ... I don't mean inside information is illegal inside information. That privileged information. Then of course you can make money because there lots of managers that have too little money to manage relative to their true ability. But the point is it's very hard for individual investors to have information. It's not hard for the mutual fund companies. I mean, that's our other paper.

Cameron Passmore: Oh wow.

Jonathan Berk: The mutual fund companies know who the good managers are, and they move the capital.

Ben Felix: I've got a question. We had Gus Sauter, who was the former CIO at Vanguard, for many years. He made this comment to us. He said that because of his position at the company, he felt that he did know that there were some active managers that he could allocate to and he believes in active management. And that sounds like it perfectly aligns with what you guys are talking about.

Jonathan Berk: Exactly what we found.

Ben Felix: Wow.

Jonathan Berk: That's exactly what we found.

Ben Felix: Very interesting.

Jonathan Berk: It is interesting. Jules and I are academics, and it's very interesting as an academic to then talk to practitioners about what they themselves are doing. One of the things you may say to me is, "Well, don't the practitioners already understand what they're doing?" And they do, in their own world. But generally practitioners have never taken a step back, and thought about how they fit into the big picture. And that's where, when you have discussions with them, they suddenly begin to see where you are coming from. And the big picture. It's an interesting process.

Jules Van Binsbergen: Well, to give you one example, one place where I think that the way the practitioners talk about the problem is exactly consistent with what we've been talking about is this, two mutual fund managers meet and they have to determine the higher key between them. Do you think that the first information they exchange is what the alpha over the last quarter, the last year, the last five years was? What do you think is the piece of information that they first say to each other?

Ben Felix: Their income or the size of the fund.

Jules Van Binsbergen: How big is your fund? And so that obviously is exactly consistent what we've been saying, that is that the skill, how competent they are is very quickly communicated through how much money you manage, not through what alpha you've been making.

Ben Felix: We had Gene Fama on our podcast a while ago, for episode 200, so 20 episodes ago, and his comment was that the strongest evidence for efficient capital markets is that active management is a negative sum game. What are the implications from your research on efficient market hypothesis for the stock market?

Jules Van Binsbergen: Well, as you said earlier, I think they're much closer to each other than was previously suggested. To the fact that stock markets are completely rational and mutual fund markets are completely irrational. That's a corner solution that doesn't seem reasonable. I think that we need to move both towards each other. Mutual fund markets are more rational than we thought before. And I think that more and more evidence in the stock market literature seems to suggest that. For example, all of these factors that we studied before, there's more and more evidence that they are not proxying for risk, which suggests that maybe the stock markets aren't as efficient as we thought before. And so now I think the real research question, at least to me, and I'm sure to Jonathan, is also interesting, is where exactly do these two meet? Or do they even cross over to the other side?

And so I think that one way to think about it is this, all markets in the economy have a certain level of competitiveness. You have stock markets, you have bond markets, you have labor markets, you have product markets, you have housing markets, you have mutual fund markets. And so ranking them by how competitive they are is much more interesting exercise than to say, is the stock market efficient or not. Efficient just means perfectly competitive. And perfectly competitive is a theoretical construct that we know in reality can never be true anyway. In fact, the famous paper on the Impossibility of Efficient Market says that that corner solution can never be true. It's not a matter of whether they're efficient or not, it's just how competitive are they and also relative to each other. Now, I think that we can all agree that financial markets are much more competitive than labor markets because if you want to fill a position in your firm, you are searching for I don't know how long under tons of costs that are associated with having to do that.

I think that finding a house is a much less competitive market in the sense that the search friction of finding the right house or the buyer and the seller bringing them together. There are tons of fees with intermediaries that charge for that. Those are much less efficient. Even product markets, I would say, are much less competitive than financial markets. I shouldn't say efficient, I should say competitive. But the key question that we were just debating is which market is more competitive mutual fund markets or stock markets? And how does that relate to each other? And I think that a lot of research should go in that direction and trying to establish those two things.

Jonathan Berk: Jules, I think, is being a little bit too polite to Gene Fama. Gene Fama is just wrong about this. If you observe that the alpha in mutual funds is negative, tells you absolutely nothing about efficient markets as he defines it. Its tells you that too much money has been allocated to active management. His own theory, which he apparently does not appreciate, says he expects the alpha to be zero. If the alpha's negative, that is evidence against his theory. It's evidence against the idea that people are rational. His theory requires competitive rational markets. He's just wrong about that. What can I say?

Cameron Passmore: Given that the benefits of skill go to the managers, what should individual investors do?

Jonathan Berk: Well, what I tell my students is, I say, "Look, some people love to follow the stock market." And I'm sure many of your listeners fit in that category, that's why they listen to this podcast. I don't. I couldn't care ... I find it boring. I couldn't care less happens stock market. For me, if I were to spend my time looking at mutual fund managers and trying to find good ones, I would find that a disutility, right? But there are many people that think that's different for. They enjoy the process of finding managers and finding stocks. If you're a person like me, what you should do is stick your money in passive management and forget about it, which is exactly what I do. But if you're one of those people that enjoys the process of investing and searching for managers, then what you should do is search for active managers. What Jules and I have already said is the flow of funds relation is slow.

If you are ahead of the game and you're prepared to move your money every three to six months between managers, you can make extra money. It's just like the stock market. If you really enjoy researching stocks, you can try to compete with the T-row prices of the world, and most likely you won't, but you could try, and you're getting enjoyment out of it. And so that's my advice. My advice is you don't get enjoyment out of it, put your money in the market and forget about your investment, and start withdrawing when you retire. And if you do get enjoyment out of it, you're welcome to try to find good managers.

Jules Van Binsbergen: But it is this catch-22, in the sense that it is the competition between the investors that are trying it that makes it ineffective. As soon as this Apple stock is underpriced, the fact that everybody immediately goes towards it and tries to buy it at this lower price and the competition between them makes the market maker update on what the right price needs to be. And therefore instantaneously the price jumps up. And then in the end, very few people, if any, are going to be making money out of that process because the competition is so intense. But it is the catch-22, because in the sense that if nobody would engage in that process, then of course it would never happen.

Therefore, there needs to be a little bit of friction for this. I think it's like the oil that makes the wheels go. You need a little bit of friction there, as Jonathan describes, can make it work. But in the end, you do need to realize that you're playing a super competitive game. And so in super competitive games, there are not a lot of winners. It's much easier to play a game where there's not a lot of competition and then everybody can make lots of rents. The financial markets, you can say many things about them. There're some frictions, but that they're very uncompetitive. I wouldn't describe them that way.

Ben Felix: With allocating the funds, you're not trading against the big institutions, but you are allocating against big institutions. You guys talked earlier about how fund companies play an important role in allocating to managers.

Jules Van Binsbergen: They are already making sure that the capital allocation can happen quickly. But we also said that the company as a whole gets more money allocated because the value added off all of the managers goes up because this allocation step helps. And so there's still ... If no extra money would be allocated to the company, the mutual fund company as a whole, then the fee revenue of the company would stay the same. And all of the benefit from this relocation step would accrue to the investors. There again, there's this competitive step where the investors compete with each other for getting those rents, therefore make the whole company's AUM bigger and therefore in the end, again, go home with little to nothing. It's a sad ... It's the same thing with the left and the right shoe example, the 99 people that are competing with each other. If they could collude, they could get something out of that situation. But if they can't collude, unfortunately the outcome is they're not going to get very much. That's just the reality of it.

Ben Felix: I'm conscious of time. I do want to ask more about asset pricing models though, just to finish our conversation. You guys have both talked about how multifactor asset pricing models, there's a lot of evidence suggesting that those are not representative of risk. Can you talk a little bit about your research on that? Maybe just quickly, how did you test that? And what are the implications for multifactor asset pricing?

Jules Van Binsbergen: I think that ... The way we did it is simply this. We said the flow performance relationship is an indication of people chasing a positive NPV opportunity. If the stock is underpriced, then you go for it. If the mutual fund is too small, you go for it. If a manager outperforms, that means you update positively on how good they are, which means that the current size is not as big as it should be because the manager is bigger and can handle more money than you thought before, and therefore you get to see this flow. But if we see people voting with their feet, why don't we use the voting with their feet to figure out what benchmark they use to cast their vote? And so what we said was, here are 10 different models, benchmark models, let's compute outperformance relative to each of these 10 models.

And again, some models a manager may have outperformed, and against another they haven't. And then we see whether the voting with their feet coincides with that measure of outperformance. And then we're going to see which model best predicts how people vote. And so when we did that, we found out that the CAPM ... Outperformance relative to the CAPM actually was the best model. In the sense that that best predicted how people vote with their feet. And that these additional factors had no, if even negative, explanatory power in terms of predicting how people vote with their feet. And so therefore, at least it seemed that people are not using those models when they make their capital allocation decisions. Now you can still say, "Well, these are not the people that set what we call in finance the stochastic discount factor." They're not the ones who determine what the benchmark models should be.

But as we already established, there's a very large group of people that participates in mutual fund markets. And if for that whole group, these models don't describe how they're voting, then maybe we need to start questioning whether these models are really the right models and whether they're measuring the risk that we thought. To summarize it very simply, we just ask the question, do investors view Fama-French factors as outperformance or as risk? And the answer is they view it as outperformance, not as risk. And so it's an alpha. They view it ... They count it as alpha, not as a risk premium that you can ... Because ... Another way of saying it is a risk ... You shouldn't reward a manager with more money if all they did was take more risk according to that risk premium. You should only reward them if it was outperformance. And so clearly they count it as outperformance.

Now, in the beginning there was some criticism, and people said, "Oh, but this is just because it's unsophisticated investors." But then there were some people that did it for hedge funds. Where I think people generally agreed that the investors are quite more sophisticated and they found the same result there. And then there were even some people that tried to do it for real investment decisions inside firms to figure out what firms themselves uses the risk model. And again, the same ... No, I say that quite incorrectly. How firms decide to repurchase their stock or issue more stock. And so that's also an investment decision that's made inside the firm by managers. And again, the same result shows up. It's again going to be the CAPM. CAPM seems to be quite a dominant model that seems to be showing up all over the place.

And if we were really fair with ourselves, I think if you asked the number of finance professors that in their core finance class taught anything but the CAPM, I don't think any of them really went to any of these multifactor models. And that is also revealed preference, because if you really thought that this was the risk model that everybody should be using, why isn't it taught everywhere as the risk model that people should have? And then on top of that, the finance literature also votes with their feet because it wasn't the case that we all converged to the multifactor models. What we started doing was we just report the results of the paper for five different risk models because nobody really wanted to take a stance on what it really was. You do it. And for the CAPM and for the Fama-French three and four and five factor models plus the ... Clearly, I think we were just building this confusion there.

And I've always found this disappointing because as an asset pricer, I always thought that we were just given one job, and that is tell people how to adjust for risk. And if there's one job that we have not delivered on it is to tell people how we need to adjust for risk. In that sense, as a field, I think we've largely failed so far. And so I think there's work to be done. Now the CAPM works the best in terms of explaining investor behavior, but it doesn't work perfectly either. I think the best summary, Jonathan, correct me if I'm wrong, is that the CAPM is the best we have so far, but I'm not sure it's the ultimate answer. But it's certainly ... The other things that we've tried may not be improvements.

Jonathan Berk: Let me just add the following anecdote, which is I think when you write a good paper, when you do good research, you're surprised in some fundamental way. And if you ask me about that work, what fundamentally surprised me, it's the following. Before I started, if you'd said to me, "Okay, Jonathan, what's going to happen?" I think I would've predicted, what we got to find is that investors use the market portfolio. That they benchmark against the market. We didn't find that. We found that the CAPM worked. In other words, they're not benchmarking against the market, they're adjusting for beta. Because one of the other models we have in there is the market model. And it didn't perform as well as the Capital Asset Pricing Model. That in fact, the essence of the CAPM, which is that you measure risk with the beta. You understand fundamentally that idiosyncratic risk can be diversified away. And all you care about systematic risk, absolutely, was verified in this research.

And that came as a big surprise to me, especially given the inability of the model to explain the cross section of stock returns. And so that ... Where Jules said ... There is something missing, no question about it. We as a field have failed to fully explain risk models. Having said that, Bill Sharpe's Capital Asset Pricing Model is the best we have, and there is a lot of substance there. And when I teach it, that's what I tell students to do. Look, we don't have the answer, but it's not that we know nothing. The best we have is the Capital Asset Pricing Model, it imposes a lot of discipline, and it's what you should use when you are trying to make a corporate financing decision.

Cameron Passmore: Wow. This has been an incredible discussion. And we have one final question for each of you. How do you each define success in your lives?

Jules Van Binsbergen: Jonathan, you want to go first? The two of us has talked about this a lot and so ...

Jonathan Berk: You shouldn't ask a 60 year old man such a question. I think one of the sad issues of all us ... And I tell my students this all the time, "Is look down, not up." We're all very competitive people, very successful people, and we are always looking up. Whenever we achieve success, we go, "Oh my God, that's not enough." And I tell my students, "No. If you want to be happy, look down, not up. Look at what you've achieved, not what you could have achieved." I think the best definition of success is to really appreciate what you have done in life. And I don't mean just ... Great. I admit that I'm immensely proud of some of the research that I've done. Astonishing to me that I could do stuff that could actually inform people. But there are other things. I don't want to sound soppy about this. I can't stand it when people sound soppy. But realistically, there are other parts of your life that are important.

How successful are you as a parent? How successful are you in other areas of your life? And I think success should be what have I achieved and what could I have possibly thought about when I was young? As I like to say to Jules, Jules knows I'm always complaining, but if you'd said to me when I was growing up in my house in South Africa that one day I would be on the faculty at one of the world's greatest universities, at the time when one of my colleagues won the Nobel Prize, I would've said you were dreaming. You were absolutely dreaming. And I could never in a world achieve that level of success. And yet, here I am in exactly that position complaining about, "I'm not successful enough." One piece of advice, look down, not up.

Jules Van Binsbergen: The more I see what's happening in society today, therefore I think I've chosen the right profession in the sense that I think our biggest job is to pass on to the next generations, the collective body of knowledge that we have accumulated. And I think that if you look at certain cultures and certain events, world events and historical events, losing that passing on is not so hard. It only requires a generation or a generation or two to not pass on the culture, the education, the drive, the value system that we have in terms of how we treat other people, that we treat other people like individuals, for example, and not as members of groups. And we were really struggling with that right now. And so I think that ... And there, I think being a parent comes in a lot too. But also I think being a teacher and being able to justify, in some of the recent classes that I've developed and started teaching, I wanted to communicate much better why as a field of economics and finance, we arrived at what we do today to begin with.

In other words, Wharton students come here, we can teach them the formula of the CAPM, and then they can go home, not with that formula, and apply it or not apply it. But I haven't taught them really how to critically think. I haven't really taught them what the underlying philosophy is of why markets are competitive and what the benefits of competitive markets are and why the expected return on investment is only a function of its risk and not of other things, which is very much a consequence of these competitive markets. And so the underlying philosophy and a lot of the enlightenment values that we've been taught, I think bringing those back into the classroom, I think is that. And so if you ask me how do you define success, the more of that I can pass on to next generations in whatever form, I think that's how I would define it. In that sense, I think our teaching mission is just as important at this point as our research mission.

Cameron Passmore: Great answers. And thanks guys for joining us. This has been so much fun, so enlightening. And hopefully you'll come back one day. This has been great.

Jules Van Binsbergen: Thank you so much for inviting us.

Jonathan Berk: Thank you guys for inviting us. I really enjoyed this. I've got to go. I actually have to go and teach. All right, guys. Bye.

Cameron Passmore: Thank you for your time. Bye.

Ben Felix: All right. See ya. Thanks a lot guys. Thanks so much.

Is there an error in the transcript? Let us know! Email us at info@rationalreminder.ca.

Be sure to add the episode number for reference.


Participate in our Community Discussion about this Episode:

https://community.rationalreminder.ca/t/episode-220-jonathan-berk-and-jules-van-binsbergen-the-arithmetic-of-active-management-revisited-discussion-thread/19429

Links From Today’s Episode:

Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582.
Rational Reminder Website — https://rationalreminder.ca/ 

Shop Merch — https://shop.rationalreminder.ca/

Join the Community — https://community.rationalreminder.ca/

Follow us on Twitter — https://twitter.com/RationalRemind

Follow us on Instagram — @rationalreminder

Benjamin on Twitter — https://twitter.com/benjaminwfelix

Cameron on Twitter — https://twitter.com/CameronPassmore

Jonathan Berk — https://www.gsb.stanford.edu/faculty-research/faculty/jonathan-b-berk

Jules van Binsbergen — https://sites.google.com/view/jules-van-binsbergen/

Else Equal: Making Better Decisions podcast— https://www.gsb.stanford.edu/business-podcasts/all-else-equal-making-better-decisions

'Mutual Fund Flows and Performance in Rational Markets' — https://www.jstor.org/stable/10.1086/424739

'Measuring skill in the mutual fund industry' — https://www.sciencedirect.com/science/article/abs/pii/S0304405X15000628

'Mutual Funds in Equilibrium' — https://www.annualreviews.org/doi/abs/10.1146/annurev-financial-110716-032454?journalCode=financial#article-denial

'Matching Capital and Labor' — https://onlinelibrary.wiley.com/doi/abs/10.1111/jofi.12542

'Assessing asset pricing models using revealed preference' — https://www.sciencedirect.com/science/article/abs/pii/S0304405X1500149X

'How Do Investors Compute the Discount Rate? They Use the CAPM' — https://www.cfainstitute.org/en/research/financial-analysts-journal/2017/how-do-investors-compute-the-discount-rate-they-use-the-capm-corrected-june-2017