Episode 364 – Martijn Cremers: Is the Conventional Wisdom on Active Management Wrong?

27:28It has seen a slow, continued decline. So in other words, the aggregate or average active share of equity funds in the United States has continued to trend downwards, roughly speaking, for the last three decades. And more funds now are somewhere between 60 and 70 percent active share. How do you describe the concept of a fund's active fee? The starting point for thinking about active fee is to relate the fees that you're paying for how much active management are you receiving. The exposure to the markets, again, for you as equities, is, of course, very inexpensive. So beta needs to be cheap. Investing in broad-based market indices should be low cost.

28:10So higher expenses, therefore, should reflect that you're receiving more active management. So how can you interpret whether the level of fees of any product is appropriate without understanding how much active management you are actually receiving in the product? And that's where active share comes in as a basic measure of how much security selection the fund is doing. So that brings up the active fee, which we define as the ratio of the level of expenses of the actively managed funds divided by the active share, the ratio of that. In the paper, we define it more precisely as how much more expensive the active fund is relative to the passive fund, so the difference in fees over the active share.

28:51So it's, in my mind, a proxy for hurdle rates for how much the active manager has to outperform by to earn back the higher fees that the fund charges. So maybe I can go through a very simple example. So let's say an actively managed fund charges fees that are 100 basis points per year higher. It's just a simple example. Then it costs to invest in the fund's benchmark. So then the hurdle rate for that manager is that that fund needs to outperform by at least 100 basis points per year on average to earn back the higher level of fees. And then, as just an example, say that the fund has an active share of 50%.

29:33That means that half of the holdings of the fund are one-for-one overlapping with the holdings of the benchmark. Of course, it matters which holdings are overlapping. But the overlapping holdings themselves won't help to earn you back the higher fees. As a simple example, I'm completely ignoring risk now. Let's say both the fund and the benchmark have 5 % invested in Apple stock, the same weight. Then whatever the performance of Apple stock is in the market, ignoring diversification and risk for the moment, it will affect the fund and the benchmark identically because they have the same portfolio weight in Apple stock.

30:10So whatever Apple stock does is not going to help the manager earn back the higher fees. So the funds can only earn back the higher fees to the extent that the fund is different than the benchmark. And so with an active share 50%, that's only half the portfolio. So you can then think about the hurdle rates of the active part of the portfolio, the active share of the portfolio. And that's why you divide the difference in fees by the active share to get the active fee, which in my very simple example would be an active fee of 200 basis points per year. Exactly. Yeah, that's really interesting. It's the fee that you're paying on the active portion of the portfolio.

30:48Is that a way to think about it? Exactly. So if you're paying an active management fee on a mostly closet index fund, you're paying a really high fee for the portion that's active. Yeah, that's interesting because you're paying active fees on beta for a portion of it, which means if you separate that out, the amount that you're paying just for active could be really, really high. Yeah, that's it. Exactly. It's also a way to compare fees across very different products. You always have to be careful with interpretation. At the first level, exactly as you said, I think of it as how much are you paying for stock selection, if that is the main thing that the fund is doing.

31:22It makes a lot of sense how that can be very different across funds. Is there a relationship between fund size and closet indexing? Yes, but only for really large funds. And so many of the really, really large funds tend to have lower active shares or even low active shares. Their fees also tend to be lower. So again, you have to adjust for that. But we did find that the very largest funds are more likely to be closet indexers. But if you take the very largest funds out, then we didn't find a strong relationship between fund size and closet indexing. And how do closet index funds perform compared to the more active peers?

32:00I think that's probably the main takeaway from our first active share paper. I think that's the strongest and most robust prediction I have is that closet index funds tend to strongly underperform with strong statistical significance. They typically underperform, if you form portfolios of closet index funds, by about the fees that they charge. That's interesting. Do I infer correctly from that? That research was less about active share predicts positive performance from high active share funds, but more so that it predicts negative performance from closet index funds? Yes. So if you go back to our 2009 paper, we had a very strong statistical result on the low equity share funds.

32:41We did have a result that the high equity share funds as a group outperformed, but with much less statistical confidence. Wow. Yeah. Okay. And then we had a third result, which I think is also a very robust result, which was that we found strong positive performance persistence among the high equity share funds. So high-actorshelf funds that had strong outperformance in the past, that they were more likely to continue their outperformance in the future. So I think two of those three results, I think, are fairly robust. The one result that has not been very robust is that the average for the typical high-actorshelf fund outperformed.

33:22That was true for a particular time period that is not proven true since then or over the full period now measured. Okay, interesting. So the strong performance of high active share funds in the sample in that paper looked pretty good, but out of sample, it has not been so good? Exactly. The other result is that the combination of high active share and strong past performance, that that leads to continued strong future performance that has been robust as far as I know. Okay, interesting. So it's the unconditional high active share. If you don't combine active share with anything else, then you don't find in recent decades It's the average high active share fund has outperformance.

34:03But the positive performance persistence, as far as I'm aware of, has been robust. Probably related to persistence, how persistent is active share as opposed to performance, but active share itself? For a typical manager, active share is quite persistent. We look at five-year periods of time where the typical fund stays in its active share decile. So it's quite persistent. There's some mean reversion, but very slow. tracking error is much more strongly mean reverting. And so active share is quite persistent. You see significant changes in active share when a fund massively changes its size. Typically when size goes up a lot, you see active share often, not always, but often come down a bit.

34:47And when the manager changes or the whole management team changes, you may see some significant changes in active share as well. Yeah, okay, interesting. I was going to ask about that. With the Birkin Green 2004 idea that money flows to skilled managers to the point where they can no longer express their skill effectively. Is that that idea? When a fund gets really, really big, its active share goes down? That is what we find, yes. That is not a strong result unless the fund gets really big though. Okay. Empirically, how does active share relate to the three pillars of active management? Yeah.

35:18So in our paper I wrote in the Financial Analyst Journal, I tried to summarize what I I have learned from doing research on active management. I put that together in what I call the three pillars framework that I argue summarize what I think active managers need for long-term success. These are three things that they all need all at once. Not just two out of the three, you really all need all three at the same time. So they need skill, conviction, and opportunity. Skill to identify investment opportunities. than the conviction to pursue an investment strategy that requires a certain courage of one's convictions.

35:56It's not easy to replicate. And then you need opportunity internally in the fund complex from your investors that you don't have too many constraints. But you also need to operate in an external market where you actually invest your cash where there are actually investment opportunities. There are stock selection opportunities. And then I relate those to active share. So for example, if you have a high active share, it doesn't mean you are a skilled manager. You don't need skill to have a high active share. Active share is not a measure of skill. But if you have strong convictions, you would naturally have a higher active share.

36:34If you have a lot of opportunities in the market for security selection, it is natural that you would have higher active share. And so a high active share manager, in my view, will naturally be a manager with strong convictions and more opportunity, but not necessarily skill. But if the manager doesn't have skill and the markets are very competitive, then how long can high active share managers with strong conviction and opportunity, how long can they survive in very competitive markets? And that's why I say that in the end, well, if the market is really competitive and you have a manager with a high active share strategy that has good long-term performance, then that manager surely must have skill.

37:12Can you talk more about patience specifically? and how important that is? Yeah, patience. I use it as my example for investment strategy that requires conviction. So if you go back to conviction, if a manager's active strategy doesn't require conviction, then why would it persist? Is a strategy successful? And what prevents too much money from flowing in? What prevents too many other managers from doing exactly what this manager is doing? And so without the strategy requiring a certain courage upon convictions where there are some economic frictions that you need to overcome, why would it persist?

37:49Key example is Warren Buffett. It's actually very hard to do what he does. And so I think it's currently the case that, especially in mutual funds, most investors evaluate their managers over fairly short periods of time. And so it's a fairly impatient world. And so it's hard to be patient in an impatient world. There's a certain mismatch between the expected investment horizon of a patient manager and the time over which that manager is evaluated by the investors. And even just hypothetically, let's say the patient long-term manager is right about a certain bet. That manager may be wrong about that bet in the short term, and he's evaluated by the investors in the short term.

38:31And that mismatch creates an economic friction that makes it hard to be a patient manager and active. If you look empirically, we find that most active managers are not very patient. They hold stocks for one to one and a half years on average. Most patient capital is not very active. It's the combination of patience and high active share that is very rare. And so then if you think about markets where I posit, there are some long-term investment opportunities that only patient managers can exploit. And for reasons I just tried to explain, there's relatively few patient managers that pursue those because this is really hard to do unless you have really, really patient investors.

39:18There may be particularly good opportunities for patient high-acted share managers. That gives me a follow-up question. You talked about the lack of persistence of the unconditional active share out of sample. If you look at patient, high active share, active managers, has that been persistent? Yes. The exercise that we did is we look at active funds in two dimensions independently, active share versus patients. What we found was that on average, the high active share managers did not underperform, but there was also no strong outperformance in our sample, except for the high-actose ear managers that were also patients.

39:58So all of the outperformance, also in my earlier study, really came down to that group, that is both high-actose ear and patient. The less patient high-actose ear managers also did not underperform, so arguably it may still be a decent starting spot to start with. But it was really the combination of patients high-actose ear that on the one hand is rare. On the other hand, it drove the outperformance among the high active share funds. Interesting. I have to ask you, how was the active share concept initially received by both academia and the industry? So by now, it's a fairly standard measure, like tracking error.

40:35Initially, I think a lot of academics were very skeptical about our results because it goes against the grain of the conventional wisdom of active management. when we originally did our paper, we were so astonished of how many low active share or closet index funds there were in the market. I think that in practice or in the industry, it has become a standard measure that is useful to, on the one hand, find closet indexes, but also on the other hand, find funds that actually do a lot of security selection. And there's lots of funds that do that. It's a very simple measure to find those funds. It definitely does go against the conventional wisdom.

41:10We've been beating the index funds drum for a long time at our firm. And yeah, this is definitely something that somebody who promotes active management can say, well, hey, look, there's this research that says that there are cases where it can make sense. There's a critical paper in the Financial Analyst Journal, 2016 paper that I'm sure you're familiar with, deactivating active share. I'd love to know what you think about their critique. Yeah, it's good to have a debate. And I also was glad that they were able to replicate our original results without any problems. That's always a relief. There can be issues with replications.

41:46If you go back to our paper, I talked about three big results. One, low equity share funds with strong statistical confidence continue to underperform. Then unconditionally, that equity share in our particular sample outperformed. That result did not prove robust over time. And then the third result is that if you combine high equity share with past performance, we found strong positive performance persistence that in a recent financial analyst journal we documented was persistent and that was robust in the last couple of decades after our first paper was published. And so this 2016 paper that you're referring to focused on the second of our two results, the one that has proven to be the least robust.

42:30They made a very important point that I talked about earlier, is that there's a very big difference in the typical active share of a large cap fund and a small cap fund. Large cap funds tend to have significantly lower active shares on average. And most small cap funds have very, very high active shares. There's simply not much variation in active shares among small cap funds. Almost all are very high. The real variation in active share comes from large cap funds. What they do in this paper is they then sort active share within the style. They sort on active share only within large caps and only within small caps.

43:11And then they look at basically within style and then the aggregate, if that makes sense. So they do it sequentially. So they decide whether or not a fundus high or low active share within a certain style. In my mind, that kind of confuses a bit what active share is trying to do. And then what they find is that, if I remember correctly, is that the main results that high active share funds outperform in that original sample, that is not robust once you adjust what active share means. And so effectively what it means is they are adding large cap funds with active shares that are actually not that high to the high active share group because they do it just within large caps.

43:56And so in my mind, they're adding noise to active share and that weakens the results. But at the same time, it's a very important point that style matters for active share a lot. And that if you're looking at a small cap fund, almost all of them have a very high active share. But I don't think that you can reduce active share to just style. And that's also why I point to all of the other results about low active share funds significantly underperforming, which they also report, but don't talk about in their paper. and then the positive performance persistence, as well as a large range of other results in my too many papers I've now written in this field as well.

44:33There's a chart in a white paper from Vanguard showing increasing dispersion of benchmark relative performance with increasing active share. How concerned do you think investors who are seeking high active share funds, like they hear this podcast and say, cool, I'm going to go find a high active share fund, maybe a patient high active share fund, how worried should they be about that increasing dispersion? The increased dispersion among high active share funds is inherent. If you're dealing with a fund with a lot of security selection, then that fund is more different from the benchmark. And so you would expect that fund to be more likely to have returns that are different from the benchmark.

45:13And that creates the dispersion. With all respect to Vanguard, I think of that issue is not really an issue about dispersion per se. I think of it as an issue of idiosyncratic risk first. And when you think about what investors should be doing, I think investors should think about their overall portfolio, not the dispersion of idiosyncratic risk of a particular fund, but rather how, if you take your overall portfolio, what happens to your overall portfolio if you add or do not add that particular fund to your overall portfolio? And what you worry about is not the idiosyncratic risk of a particular fund, but how adding that idiosyncratic risk to the overall portfolio actually changes the dispersion in your overall portfolio.

45:57So where you don't like dispersion is in your overall portfolio. But adding a fund with more idiosyncratic risk, as you may very well do if you add a high access share fund, may actually be good for diversification if the fund that you're adding is completely different than anything else you have in your portfolio. I think that's what's missing from the Vanguard discussion. They do not think about the overall portfolio. They don't think about diversification and how diversification can actually be improved by adding a product that has less overlap with what is already in the overall portfolio, or more generally, how idiosyncratic risk actually helps diversification.

46:36And further to Ben's question, Martin. So we have listeners that range from a curious DIY type to a lot of experienced seasoned professionals who listen. So in your opinion, how should they decide whether to invest in index funds or to actually hunt for patient high active share fund that might beat the index? I don't give investment advice, but I appreciate the question. I knew the disclaimer was coming. Yeah, I have my usual disclosure as well that I serve as independent director of an active manager. If I would give advice, it depends on what you're trying to achieve and what you're trying to avoid.

47:12if an investor is very focused on avoiding on the performance and they don't have time to study active managers. And it makes sense to invest in low-active share funds that are very inexpensive. I think one aspect that we haven't talked about is downside risk protection. And there's actually considerable evidence that we also go through in our challenging conventional wisdom paper is that active managers have real skills in risk management and particularly downside risk protection. And if investors are very interested in that, but also are interested in potential outperformance, but they're also really worried about protecting the downside, especially in times that the market is not performing well, then they may consider high active share funds that have strong past performance, strong evidence track record of protecting the downside in bad times over a longer period of time at a reasonable fee.

48:04But that takes research, that takes time, that is really not easy to do. But unlike the conventional wisdom, it's not a foregone conclusion. It sounds like, based on your research, that you're going to underperform if you try and choose an active fund. Yeah, that's my conclusion. As long as, again, it's a fund that is worth its fee, that's actually very different and not a close index fund. All right, let's move on to your relatively recent paper on active share in fixed income, which is a really interesting concept. Can you talk about how the arguments in favor of indexing in equities hold up when we move over to fixed income?

48:40It's a completely different world than we found. Yeah, we're excited. It's a new paper. It's taken a bunch of years to write, but I'm excited about it. It starts with the observation is that passive funds, the actual passive funds in fixed income are actually not very passive. They don't actually follow the benchmarks very closely. The actual passive funds in fixed income or index funds, ETFs in fixed income, they have wealthily high access years and they trade a lot. And the reason for this is that other than in equities where very close replication of the actual benchmarks is fairly easy and low cost.

49:19It's not at all the case for fixed income. Benchmark replication is far from straightforward. Lots and lots of securities in fixed income that are in the benchmarks are very illiquid. So it is practically impossible for a passive fund manager to closely follow the benchmark. The second broad point is fixed income markets are still much, much less efficient than equity markets. And the mutual fund industry, so to speak, is much less competitive, relatively speaking, if you compare U.S. fixed income to U.S. equities, especially large cap equities, most of the money is. What are the main challenges for index investing in fixed income?

49:58The benchmarks in fixed income have too many securities. So many of these securities are fairly illiquid. For example, the main market-wide benchmark in fixed income in the US is the Bloomberg US aggregate. It contains more than 12 ,000 different securities. I think about half of those don't trade on any given day. So in general, many of these securities are not very liquid at all. And that means if you are a manager of a passive fixed income fund, you face a real trade-off that is not really faced by your equity counterparts. So a passive fixed income fund faces a trade-off between, on the one hand, tracking the benchmark, but on the other hand, you want to also control or really limit the trading costs and provide liquidity to their investors.

50:46And it's really hard to do both. It's really hard to closely track, but do so at a low cost that provides liquidity to your investors. We had Hank Bessembinder on this podcast a while ago, pretty recently actually, and he talked about the performance of equity mutual funds at long horizons and how it gets worse and worse the longer you go out. How do active bond funds perform relative to index funds at increasing horizons? Bessembinder makes a very powerful argument for equity funds. I find his work very compelling. So again, it's very different for bonds. The average active bond fund has outperformed their passive counterparts.

51:21Not that much, and statistically significant is marginal. In our sample, look at a fairly large, long sample. It's about 28 basis points per year, with some but not strong statistical confidence. And what explains that? Well, first, as I elaborated, passive investing in fixed income is very difficult. Trading is very costly. On the other hand, fixed income markets are much less efficient, and the industry there is less competitive than for US equities. You talked about how bond index funds don't really look like the index. How does the active share of a typical bond index fund look? The active share of bond funds is one of the things, of course, we do in our paper.

52:03We adjust it for the fixed income markets. So we compute active shares at two different levels. The security level, which is the bond issue level, which is we can directly compare to the active share of an equity fund. Very few equities have more than one share class. But you can also compute the excess here of a bond fund at the firm level or the issuer level. For example, IBM Corporation issues many, many different bonds. So you can look at the IBM bonds at the issue level, or you can aggregate up across all the different issues that an issuer like IBM has issued and then calculate the portfolio weights at the level of the issuer, the firm.

52:45and then you first aggregate up the weights, you end up with two different active shares. The median passive bond fund has a bot-level active share, so security-level active share, of 55%, which is incredibly high. And you can compare that to the median active share of a passive equity fund, which is about 2%. In other words, passive equity funds are actually really, really low active share, 2%, the median, while for passive bond funds, those are substantially different than their benchmarks. And how does bond liquidity affect bond index fund active share? If you look at where the active share of these passive index funds come from, then it comes primarily from them avoiding the more illiquid issues in their benchmark.

53:35and those are costly to trade. You need to provide liquidity to your investors. There's a lot of turnover and so they minimize costs by avoiding trading in some of the less liquid parts of the benchmark, which also, of course, that means they also have more tracking error that comes with that. But doing that increases their active share and can explain why the median active share of the passive fixed income funds in our sample, including ETFs, is 55%. You talked about a couple of the reasons is that it's different in fixed income than it is in equities. What about the skewness argument? That's one of the things that we talked to Hank Bessenbender about.

54:11Does that apply in fixed income markets? It doesn't. So I find Hendrik Bessenbender's skewness argument very compelling. I think it's a very strong argument pertaining to equity funds. In equity funds, distribution cross-sectionally of equity returns is very strongly positively skewed. that a few names, few stocks have extremely high positive stock returns. And those few names drive the performance of the market, drive outperformance. That's where the idea of buying the haystack comes from. That outperformance requires that you to be in those names. And one way to do that is be very well diversified and buy the haystack.

54:52That argument of very strong positive skewness strengthens the diversification benefits of indexing of passive management. We find that that argument simply doesn't apply to bonds. Bond returns are not very positively skewed. They have hardly any skew at all, especially compared to stocks. So the positive skewness argument is simply moot. We repeat several of the exercises that Professor Bessenbinder has done, and we don't find that it applies to bond markets. And for example, one of the things that he does in his papers is he chooses random stocks and see what is the percentage that a random stock will underperform where you perform portfolios that are not super well diversified.

55:31How many stocks does it take before this skewness effect wears out? So for equities, it requires, they do this for memory, but I think hundreds of stocks before this skewness effect wears out. And for bonds, it doesn't. If you have 30 bonds, then the skewness effect is moot because the skewness in bonds is much less positive. And so the diversification argument, the schoonish argument doesn't really apply to bonds. Yeah, it makes sense, I guess, because you wouldn't have big upside single bond issue for some reason. Oh, exactly. I've long taught fixed income, I think for 10 years at Yale, and I taught it at Notre Dame when I started here.

56:12And I always like to tell my students, bonds are very strange instruments. The best you can ever hope to get from a bond is exactly what it promises to pay you. Exactly. Of course, I'm not looking at any embedded options or convertibles. If you just look at a straight bond, then the best you can ever do is that bond gives you exactly the cash that it promises. In other words, the upside is strictly limited. In just ways go way down, then you may have a decent upside. But there's a strict limit to the upside. You still have the whole downside. And so that explains exactly why bond returns are much less positively skewed than equities because equities don't really have an upside limit.

56:52Yeah, right. So you implicitly answered this in answering the previous questions. Why do you think active bond funds have remained relatively popular despite active management declining in popularity within equities? Yeah, that was the starting point of our paper where we just observed that empirically speaking, these massive outflows that we've observed in the active management industry for equities hasn't happened in bonds. And so passive investing in bond funds has also become more popular, but we've not seen the massive outflows out of actively managed bond funds. I think a lot of that is just about performance.

57:30Active bond funds have performed fine. They have generally outperformed passive bond funds. Passive bond funds, as I explained, are actually not that passive. And in our paper, we have a lot of results related to active share. We find that the high active share bond funds have done substantial outperformance. We found significant positive performance resistance among high-actual share bond funds. And we also find that the high-actual share bond funds have significant protection on the downside. If you look at the maximum drawdown over left-eight quarters, the high-actual share bond funds perform pretty well there.

58:05In other words, the case for active management and bond funds seems relatively strong. It's interesting. Anecdotally, Cameron and I have talked to peer firms like us. And it's not uncommon to find a firm that's really passionate about index investing within equities, but they are still using active management in fixed income, which I think speaks to the points you're making here. Our final question for you, Martin, how do you define success in your life? That's a great question and a hard one. Well, what is success in life? I think for me, that's defined by a life that is full of faith, full of love and full of hope.

58:39And those are great gifts. And that includes, I would say, faithfulness to your particular vocation. For me, my vocation as a husband, to my wife, as a father, to our children, but also my vocation as a professor of scholarship, teaching, and administration. At the end of the day, for me, my faith, love, and hope are the great three things that are hard to measure per se, but that do define success for me. What a wonderful answer. Great to meet you. Great to have you on. No, thanks for having me. It was great. Thank you. Awesome. Thanks, Martin.

59:38Thank you. constitutes an offer or solicitation to buy or sell any security. This communication is distributed for informational purposes only. The information contained herein has been derived from sources believed to be accurate, but no guarantee as to its accuracy or completeness can be made. Furthermore, nothing herein should be construed as investment, tax, or legal advice and or used to make any investment decisions. Different types of investments and investment strategies have varying degrees of risk and are not suitable for all investors. You should consult with a professional advisor to see how the information contained herein may apply to your individual circumstances.

1:00:16All market indices discussed are unmanaged, do not incur management fees, and cannot be invested indirectly. All investing involves risk of loss and nothing herein should be construed as a guarantee of any specific outcome or profit. Past performance is not indicative of or a guarantee of future results. All statements and opinions presented herein are those of the individual hosts and or guests, are current only as of this communication's original publication date and are subject to change without notice. Neither OneDigital nor PWL Capital has any obligation to provide revised statements and or opinions in the event of changed circumstances.