AgPa #77: Too Much Passive Investing?

The Rise of Passive Investing and Active Mutual Fund Skill (2023)
Da Huang
SSRN Working Paper, URL

This week’s AGNOSTIC Paper is a quite recent working paper that examines the impact of passive investing (URL) on the US stock market. The debate about a potential tipping point when too many assets go passive is ongoing and often quite emotional. Depending on who you ask, you hear everything from “fundamentally broken” markets to the idea that we only need very few skilled active managers who compete for all the alpha.1See for example this article from ACADIAN. This week’s paper provides some interesting theoretical and empirical results on that matter.

Everything that follows is only my summary of the original paper. So unless indicated otherwise, all tables and charts belong to the authors of the paper and I am just quoting them. The authors deserve full credit for creating this material, so please always cite the original source.

Setup and Idea

The paper has been on my list for quite a while, but really came back to me when I saw the following interview with David Einhorn at the 2024 IRA Sohn Conference.

First of all, I have huge respect for David Einhorn and what he did with Greenlight Capital. In fact, I used a quote from his famous Allied Capital speech (URL) in the introduction of my master thesis (URL). I even sent the thesis to Greenlight and David himself replied via email. To me, this was absolutely amazing and I think it says a lot about a person when the multi-billion dollar hedge fund manager personally replies to a random graduate student.

That was the bias-disclaimer, now to the video. He basically argues that the market is “fundamentally broken” because many fundamental investors who analyze companies either went out of business or changed their strategies during the rise of passive investing. According to him, there are thus not enough capital flows to reward good companies and punish bad ones. In fact, he says that because index funds simply buy everything “[…] market value has actually diverged from fair value.” In other words, passive investing makes the market less efficient and seems to have negative side effects.2He also makes some statements about IPOs, shareholder activism, and corporate governance. As those are not covered by the paper, I skip them at this point. While there is certainly truth in those statements, I believe some important details are missing and this week’s paper is a good starting point to understand why.

Data and Methodology

The paper combines various theoretical and empirical elements. In the first part, the author develops a model to derive potential rational responses of active managers to the rise of passive investing. I will not go into too much detail, but I think the general mechanics are quite intuitive.

The model consists of investors, active managers with different levels of skill, and one passive benchmark. Depending on how they evaluate their own skill, active managers decide about the risk they take and the fee they charge. Investors of course don’t know the skill of active managers before investing. They only know the overall distribution of skill in the industry. However, and this is a key idea of the model, investors use their realized out or underperformance to gradually learn if they picked a skilled manager or not.

How does passive investing relate to this? The idea is that the amount of passive assets also helps investors to learn about skill. Take the extreme example of only one active manager in an otherwise passive world. This manager should outperform because she has no competition. When she underperforms, this is a very strong sign that she has no skill whatsoever. If you don’t make profits as monopolist, you shouldn’t be an entrepreneur. If you do not generate alpha as the only active investor, you shouldn’t be a fund manager. As we will see below, this mechanism is a key result of the paper.

The author also brings the model to the data and collects a sample of US mutual funds from the Center for Research in Security Prices (CRSP) database. CRSP is a standard data vendor for the US market and ensures very high data quality. Most importantly, the mutual fund database is free from survivorship-bias and includes all terminated funds. The final sample consists of more than 20,000 funds between 2004 and 2021. The start of the sample comes from data availability of an index-fund indicator variable and is thus not arbitrary.

Table 1 of Huang (2023).

To estimate the impact of passive investing on market efficiency, the author further collects data on single stocks and uses the methodology from AgPa #2. This vector autoregressive (VAR) approach is full of statistical wizardry, but the overall idea and outcome is quite simple. It decomposes stock price movements into four underlying drivers: market-wide information, public firm-specific information, private firm-specific information, and a noise component that captures all the rest.

The author also uses the data to test several hypotheses of the theoretical model. Depending on the context, he therefore uses a bunch of other statistical models. Reviewing all of them in detail is beyond the scope of this post. Since the paper is not yet published, we can also not free-ride on the peer-review. However, the author cites a lot of high-quality research to justify his model selection at each point and the paper has been discussed with reputable scholars. While still recommending some caution, I do believe the methodology is quite robust.

The key explanatory variable in each analysis is the size of the passive industry. The author simply defines that as the passive fraction of all mutual funds’ assets under management (AuMs).3Some people argue that this approach underestimates the true amount of passive investing. Check for example this episode of the Rational Reminder podcast for an in-depth discussion about that issue.

Important Results and Takeaways

Passive investing in the US grew tremendously

To set the stage, the author provides some descriptive statistics about the rise of passive investing over the last years. We have all seen charts like this and it doesn’t require much comment. However, I still find it useful to have it once from rigor scientific analysis and the high-quality CRSP data.

Figure 1 of Huang (2023).

The chart mostly speaks for itself and according to some sources, passive investing has finally overtaken active management since the end of this sample period in December 2021. In addition to the overall market, the author also provides statistics on 8 different styles of mutual funds.4In my opinion, this style classification is somewhat strange and certainly not comprehensive. I therefore recommend to take this chart only as indicative evidence. While there are considerable differences, the overall trend towards passive is always the same.

Figure 4 of Huang (2023).

Passive investing forces unskilled managers to quit

The standard argument against passive investing goes like this: index funds don’t care what they buy, so the shift from active to passive is bad for markets and makes them less efficient. That seems intuitive at first glance and is certainly true in extremes, but it ignores a very important aspect. It does not only matter how many people compete for excess returns, it also depends on how smart they are.

Think about a sports tournament. Is it more difficult to win against 32 teams or against 64? It depends. I would rather compete with 64 amateurs than 32 professionals. The same is true for markets. We cannot say that markets become less efficient just because a large number of active managers exit. For the critique to hold, the smart ones must exit and this is far from obvious. In fact, markets should become even more efficient when more dumb investors go passive. Again, the sports analogy. What is more competitive? The group stage of the world cup with 32 teams or the semi-final with just the four best ones? Probably the semi-final…

The author therefore derives and tests the following hypothesis: “Greater passive investment makes underperforming funds more likely to exit”. This follows exactly the logic of the model I explained earlier. When more assets go passive, investors should punish underperforming managers more quickly. The reasons are obvious. First, they simply have the option to go passive. Second, they should learn that managers who still underperform after some of their (skilled or unskilled) competition left the field were probably not skilled in the first place.

This outcome may be somewhat counter-intuitive at first glance. While passive investing does decrease the number of active managers, it simultaneously increases the overall level of skill in the mutual fund industry. At the second thought it should be obvious though. If you are a skilled active manager, you have nothing to fear. Why should investors go passive when you deliver them outperformance? Passive alternatives go after the bad active managers who cannot deliver on their promises.5Those are in my experience also the active managers who criticize index funds the most… Some of them even label ETFs as “dumb” which I believe is akin to a physicist who questions gravity. But I am digressing to my usual footnote cynicism… Sorry for that! The mechanism is again the same as the knockout-stage of the world cup. You have fewer teams, but those who survive are most likely better than those that left.

The author also finds evidence for this hypothesis in the data. Using a statistical survival model, he shows that more passive investment indeed accelerates exits of underperforming funds. Importantly, those results are also robust to a whole bunch of other potential effects. Overall, he concludes both theoretically and empirically that more passive investment drives primarily unskilled managers out of the market. Needless to say, having fewer unskilled managers should be good for markets and investors.

Surviving active managers have more skill, but take less risk

The author next examines the reaction of the (skilled) active managers who remain in the industry. In this respect, he analyzes two hypotheses: “Greater passive investment makes surviving managers take less risk because higher risk-taking slows down the rate at which investors learn managers’ skill” and “Greater passive investment makes the return dispersion of the active management industry decrease”. I really like the following chart to illustrate the underlying idea.

Figure 3 of Huang (2023).

Panel (a) shows the stylized expected return distribution of an unskilled manager. Why unskilled? Because the expected outperformance (Fund Excess Return) at the peak of the density function is negative. Now, remember that the active managers in the model can only decide about their level of risk. They cannot just increase their skill out of nowhere. I believe that this is a quite realistic assumption as it certainly takes multiple years to gain investment skills.

What is the idea now? Managers want to signal skill to investors and gather assets. How can they do that? Well, they should try to outperform as consistently as possible. Consider the following example. The first manager outperforms the benchmark every year by 1%. The second one underperforms for 9 years, but posts giant outperformance of 10% in the final year.6I know that geometric and arithmetic returns differ here… It is just an example. Who do you think has more skill? Probably the first one…

This is exactly the idea of this chart. The unskilled managers in Panel (a) cannot afford to take little risk because that would most likely result in underperformance. Instead, the model predicts that they increase the level of risk to get some chance of outperformance at the expense of more risk (shaded green area).

While it may sound theoretical until here, I think this is quite realistic. When you know that you cannot do serious stock picking, you just buy a “lottery ticket” and put a large position of your fund into a stock like Tesla. If it works out, you outperform massively. If not, you underperform, but this isn’t so much of a problem as you would underperform without the “lottery ticket” anyway.

Naturally, the picture is exactly the opposite for the skilled manager in panel (b). Truly skilled managers have no reason to take too much risk. That just introduces an unwanted probability of underperforming the benchmark (shaded red area).

Those ideas are in my opinion quite intuitive and practical. Many investors (though admittedly not all) nowadays evaluate their active managers on risk-adjusted measures. So they tend to either implicitly or explicitly examine the consistency of outperformance. You can always search for genius and it don’t say the Warren Buffetts (URL) of this world do not exist. But if you play the game in terms of probability, a manager who consistently outperforms with little active risk probably signals more skill than the apparent genius who invests in something like Tesla once a decade.

The author again takes those hypotheses to the data. He uses tracking error and turnover as measures for risk-taking. The evidence supports the ideas. The surviving active managers indeed tend to take less risk when more assets go passive. He therefore concludes that US mutual funds seem to compete with passive benchmarks by offering lower-risk funds with more consistent outperformance.

Importantly, he also mentions that those results differ from studies of mutual funds in other countries. For example, he cites research that finds more risk-taking of mutual funds in international non-US markets. This of course raises concerns. The author argues that skill-differences among international and US fund managers could explain the differences, but more research is certainly necessary to validate the results out-of-sample.

In another analysis, the author also shows that the mutual fund industry becomes more homogenous in response to passive investing. What does this mean? Well, we have seen that the surviving active managers take less risk. As a consequence, the potential return outcomes for investors narrow. In addition to that, remember that alpha or outperformance are scarce resources and mostly zero-sum games. So when you wash out the unskilled managers via passive investing, you end up with a more skilled active industry that tends to focus on the few pockets of alpha that actually exist.

We are probably not yet at the point of too much passive

Finally, what does this all mean for market efficiency and David Einhorn’s statement about the “fundamentally broken” market? Looking at the previous results, there are two opposing mechanisms.

First, killing the unskilled managers should be beneficial. You can think about market efficiency as how difficult it is to outperform. As we have seen, passive investing primarily drives the unskilled managers out of business, so the surviving ones face tougher competition. All else equal, it therefore becomes more difficult (or at least as difficult as before) to outperform.

Second, passive investing incentivizes active managers to take less risk and thereby makes the industry more homogenous. All else equal, this could be indeed bad for market efficiency as it creates pockets of the market that no one dares to attempt. In my interpretation, this second effect is much closer to Einhorn’s view about the “fundamentally broken” market than the first one. In another interview (URL), he mentions this quite explicitly and hypothesizes that investors seem to ignore very healthy companies in the mid or small cap segment.

The author uses the VAR model to bring this question to the data. Overall, he finds the same results as Brogaard et al. (2022) in AgPa #2. The amount of noise in stock prices decreased over time and also in response to more passive investing. Those are generally good news and it suggests that replacing unskilled active managers by passive investing indeed improves market efficiency.

With respect to the types of information, however, the evidence is less clear. The amount of market-wide information increased at the expense of public firm-specific information. This supports the idea of the “fundamentally broken” market as investor indeed seem to shifted their focus away from individual stocks. On the contrary, however, passive investing doesn’t really change the share of private firm-specific information. The author regards this as evidence that truly skilled active managers who uncover private firm-specific information simply continue to do their job as before.

So is the market “fundamentally broken”? For the market as a whole, I (and the author) don’t think so. Most active managers still underperform their benchmarks after costs (URL), so I think there is still some way to go before we reach the problem of too much passive. Does this mean Einhorn is wrong? I don’t think so either. Like for most innovations, the rise of passive investing certainly comes with negative side-effects. Under-covered pockets of the market, challenges for corporate governance, a more homogenous mutual fund industry, and concerns about market inelasticity are definitely challenges that need to be addressed. However, I do believe that truly skilled active managers – and Einhorn is certainly one of them – will always have their place in the world and don’t need to complain.

Conclusions and Further Ideas

I really like the paper, but there are a few things to consider. First, as of today it is just a working paper. So we should be especially careful with the empirical part. Having said that, I think the theoretical mechanisms alone are interesting and though-provoking. Second, the methodology is not straight-forward. Much of the paper is built on rather complicated models. The model how investors and managers react to passive investing, the model to decompose the information in stock returns, and so on. Models are the best we have to examine problems, but they never capture the full reality and assumptions can be wrong. We shouldn’t forget this and there are certainly many things to criticize.

Shortly after finishing this post, for example, I listened to an episode of the Rational Reminder Podcast (URL) in which they examine research that comes to more cautious conclusions about the impact of passive investing. As I am a partly-passive investor myself, I will certainly cover this research in the future to find out if I am missing something.

For the moment, however, I do believe the paper is right with respect to market efficiency and tougher competition among surviving active managers. On the other hand, I have trouble to resonate some of the remaining results with my perceived reality from the investment industry. In my and many other investors’ view, you don’t want to be in the middle of the active-passive spectrum. You either go passive and take no active risk, or you put in the effort to search a truly skilled manager and give him the tracking error to capitalize that skill. On the other hand, I also observe an increasing supply of “enhanced index” funds which is exactly what the author describes. So maybe I and the fellow investors I am talking to are just special in that sense…

This content is for educational and informational purposes only and no substitute for professional or financial advice. The use of any information on this website is solely on your own risk and I do not take responsibility or liability for any damages that may occur. The views expressed on this website are solely my own and do not necessarily reflect the views of any organisation I am associated with. Income- or benefit-generating links are marked with a star (*). All content that is not my intellectual property is marked as such. If you own the intellectual property displayed on this website and do not agree with my use of it, please send me an e-mail and I will remedy the situation immediately. Please also read the Disclaimer.

Endnotes

Endnotes
1 See for example this article from ACADIAN.
2 He also makes some statements about IPOs, shareholder activism, and corporate governance. As those are not covered by the paper, I skip them at this point.
3 Some people argue that this approach underestimates the true amount of passive investing. Check for example this episode of the Rational Reminder podcast for an in-depth discussion about that issue.
4 In my opinion, this style classification is somewhat strange and certainly not comprehensive. I therefore recommend to take this chart only as indicative evidence.
5 Those are in my experience also the active managers who criticize index funds the most… Some of them even label ETFs as “dumb” which I believe is akin to a physicist who questions gravity. But I am digressing to my usual footnote cynicism… Sorry for that!
6 I know that geometric and arithmetic returns differ here… It is just an example.