Episode 330 - Davidson Heath: How Are Index Funds Affecting Financial Markets and Corporate Governance?

November 7, 2024 Kelsey Robichaud

bio: university of Utah

Davidson Heath is Assistant Professor of Finance at the University of Utah’s David Eccles School of Business.

Davidson is a former derivatives quant and trader. He holds a Ph.D. (Finance) from USC Marshall and an MBA (Finance) from Chicago Booth.

See this content in the original post

What happens when index investing dominates the market? In this episode, we’re joined by Davidson Heath, Assistant Professor of Finance at the University of Utah David Eccles School of Business, to explore this question and its surprising answers. Davidson’s research dives into the unintended impacts of passive investing, examining how it influences price stability, corporate governance, and even the way we define shareholder responsibility. He unpacks how index funds, while supporting price efficiency, may be weakening governance structures by reducing shareholder oversight, a shift that could have lasting effects on corporate accountability. We also discuss the promise and limitations of socially responsible investing (SRI), as Davidson introduces the term “impact washing” to describe how some SRI funds fail to achieve real change despite their green branding. In a forward-looking segment, Davidson shares insights on the collaboration between AI and human intelligence in finance, giving a reassuring perspective on the future of machine and human co-existence in complex decision-making. This episode is a must-listen for anyone curious about the hidden dynamics of passive investing and the evolving role of technology in finance!

Key Points From This Episode:

(0:02:22) Davidson’s paper On Index Investing; Why active managers are important to indexing.

(0:08:42) Conclusions on how index investing is affecting price efficiency.

(0:11:10) The role of shareholders in corporate governance.

(0:13:06) How the incentives of index funds to monitor portfolio firms differ from active funds.

(0:15:10) Measuring how well index or active funds are monitoring the companies they own.

(0:16:54) How the expense ratios of index funds affect their quality of monitoring.

(0:18:08) What shareholders can do to monitor and make themselves heard.

(0:20:31) How index fund ownership affects other firm-level governance issues.

(0:21:30) Recap and takeaways on index funds and the market.

(0:25:39) The impact of socially responsible investing (SRI) and how successful they are at selecting firms with better environmental, social, and governance (ESG) characteristics.

(0:28:08) Unpacking “impact washing” in SRI funds and its consequences.

(0:33:04) Insights on how ETFs are replicating index funds.

(0:37:03) The implications of Davidson’s findings for index ETF investors and the markets.

(0:38:57) Details on Davidson’s Cyborg Trading project and how it’s using AI to complement human intelligence.

(0:42:42) How Davidson defines success: being a part of a worthwhile endeavour.

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're hosted by me Benjamin Felix, Portfolio Manager at PWL Capital, and Mark McGrath, Associate Portfolio Manager at PWL Capital.

Mark McGrath: Episode 330.

Ben Felix: That is correct. Episode 330, which is crazy in itself. Today we're joined by a fellow Canadian who is currently living in Utah as a professor there. He's an Assistant Professor of Finance at the University of Utah's David Eccles School of Business. And it is Davidson Heath. Super interesting guy. He was a derivatives quant and trader prior to academia. He went to UBC out in British Columbia. And also, to Queens. And then he did an MBA at Chicago Booth. PhD finance from USC. And now he's an academia. And he's got some incredible research on a bunch of different topics. But a large theme of what we talked about is index funds and how they are impacting price efficiency of stocks. How they're impacting corporate governance? How they're impacting the liquidity of stocks? And then we also talked a little bit about socially responsible investing. And that segment was super interesting.

Mark McGrath: Yeah. It's hard not to spoil it in the intro, but there was a couple mind-blowing comments and stats in there. Yeah, the whole episode was really, really good. Obviously, he's brilliant. Grew up in my neck – I mean, you were born on the island too. But he mentioned that he was from Quadra Island. There's a population of 2,000 people on Quadra. My best friend happens to live there. What a small world. It's really cool. But clearly brilliant. Very well-spoken. Great episode. Loved it.

Ben Felix: It is crazy though that this episode is three people – well, you're not from the island, I guess. Three people from BC. Close enough. Two of us from Vancouver. Or no. He was on Quadra. Not on Vancouver Island. But in a very close area of British Columbia. Small world.

Anyway, I think this a super interesting conversation. We talked about Davidson's paper on index investing. The first paper that we talked about on a podcast episode, I don't know, years ago at this point. But when that paper came out just talking about how index funds are not affecting price efficiency, I thought that was really interesting at the time. It was pretty cool to get to talk to Davidson.

We do mention during the conversation that he's actually the person that introduced me to Scott Cederburg's work, which obviously ended up being a couple of really interesting episodes with Scott. Thanks again to Davidson for that introduction.

All right. I think that's enough of an intro for this. We can go ahead to our conversation with Davidson Heath.

***

Ben Felix: Davidson Heath, welcome to the Rational Reminder Podcast.

Davidson Heath: Thanks very much, Ben and Mark. It's a pleasure to be here. I am a big Rational Reminder fan. And I rarely miss a pod. So this is a treat for me.

Ben Felix: That is very cool to hear. It's maybe worth mentioning while we're chatting about that that you introduced me to or at least told me about Scott Cederburg a while ago. And those turned out to be some pretty cool episodes. I'm sure the listeners thank you for that as well.

Davidson Heath: Very good.

Ben Felix: All right. Let's kick this off. We're going to start with your paper on index investing. The paper's literally titled On Index Investing. Why are active managers important to indexing?

Davidson Heath: Active investors create the liquidity in the market. And by definition, they're the ones who end up taking the other side of passive investors trading. And active investors also collect information and try to buy up stocks that are underpriced and sell or short stocks that are overpriced. And in so doing, they move the prices to being more correct. Active managers also set the prices at which index funds buy and sell.

Mark McGrath: And what are your models predictions on how index investing affects information production about individual stocks?

Davidson Heath: In the model, as passive investing goes up, it just pulls out a larger and larger chunk of the stock's market cap, which previously was available to active investors to hold and to trade. Since active investing is a zero-sum game, the stakes are now smaller for the active investors to compete over. As a result, active investors do less trading. And they gather less information about individual stocks.

Ben Felix: When you look at this, how does index inclusion affect the composition of active and passive investors in a given stock?

Davidson Heath: That's an important question because the key research question that we're trying to address, the effects of the rise of passive on financial markets, you need to separate the growth of index investing from all the other events that have happened over the same time period. That is you want to avoid spurious correlations.

In the paper, we look at stocks that switch between the Russell 1000 and 2,000 indices on a predetermined schedule. The Russell 1000 is the 1,000 largest. That's the large-cap index. And then the next stocks numbered 1001 through 3000 is the Russell 2000. Small-cap index. Now both those indexes are market-cap-weighted. If you go from the bottom of the 1,000 to the top of the 2,000, that produces a jump of about 2% of the stock's market cap in passive fund holdings.

Mark McGrath: And how does index investing affect investor behavior and stock price dynamics?

Davidson Heath: We find that stocks that get a bump in this way to their passive holdings, they have higher trading activity, higher turnover relative to their float. They have higher short interest. They have higher volatility. Their daily return volatility goes up. And they have higher correlation with the market as a whole.

Ben Felix: Stocks with more index ownership see those changes.

Davidson Heath: That's right. And in some cases, it seems pretty clear and straightforward as to why the short interest component. Many index funds make back a sizeable amount of their revenue on loaning out the stocks that they hold for short sellers to short. It makes sense that index funds relative to active funds would create a bump in the supply of shares to be shorted, for example. The other outcomes, things like correlation with the market, return volatility, it's a little harder to see exactly what's causing what.

Ben Felix: How do you measure information production, which is one of the things you guys are trying to figure out?

Davidson Heath: We look at three measures. We look at Google searches for the stock's ticker. Views of the firm's SEC filings. Think about their annual and quarterly reports. Mostly on the SEC's EDGAR website. And buy-side analyst reports about that stock.

Ben Felix: That's really interesting. It's literally how much are people looking at this stock and how much are they writing about it. Yeah, that's cool.

Mark McGrath: What are your empirical findings on how index investing affects information production?

Davidson Heath: For all three measures, when the stock gets a bump to their index fund holdings, we find that information production goes down.

Ben Felix: And how does it affect price informativeness?

Davidson Heath: The key question of course is really about information production about a stock is not itself an important thing on a society-wide level. What matters is how accurate, or informative, or efficient the stock prices are. We measure price efficiency three different ways as well. We look at variance ratios. The variance ratio is this measure of random walkiness of the stock price. It goes back to if the ratio of your volatility. Your one-day volatility to your 10-day volatility is about square root 10, then that says that your stock is relatively a random walk.

And going back to Paul Samuelson way back in the 1950s, a properly anticipated price should be a random walk is statement about price efficiency.

We look at variance ratios. We look at this aggregate mispricing measure that Rob Stambaugh and co-authors constructed. And we look at post-earnings announcement drift. How quickly new information is incorporated into the stock price? And on all three measures, we actually find zero change across the board.

Mark McGrath: Random walking is a great term, by the way. I'm going to see if I can use that somehow. How is it that information production falls but price informativeness is unaffected?

Davidson Heath: That's the key puzzling or interesting fact in the paper. Now, our model predicts this. But it is actually a very simplified and stylized prediction. The intuition is that as the active PI shrinks, as a larger and larger proportion of the stock's float is taken out of the active market by the index fund investors, the active PI is shrinking. As a result, as we said, the returns to gathering information shrink. But the level of noise trading by uninformed or less informed active traders also shrinks as well. And as a result, the ratio in that active segment of noise trading to informed trading stays the same. And as a result, price informativeness, price efficiency is unchanged.

Ben Felix: I know you guys use a Grossman-Stiglitz framework. But this is the Grossman-Stiglitz paradox happening in real life.

Davidson Heath: Exactly. Exactly. It's a very stylized model. It abstracts from a lot of things. But I think Grossman-Stiglitz is a classic for a reason. It captures some very deep intuitions that we like.

Ben Felix: Super cool. What's your main conclusion on how index investing is affecting price efficiency?

Davidson Heath: I could say a lot of things here. But I would say first of all that the model is very simple and stylized. But it captures some very reasonable intuitions on two sides. The first side is index funds have grown enormously. Passive investing has just grown by leaps and bounds. But index funds have a very strong incentive to minimize their effects on the stocks that they hold. In other words, to minimize market impact.

Vanguard, State Street, and BlackRock have a ton of really, really smart hardworking people who work there. And that's their job is to how can we minimize market impact. On the other hand, at the same time, active investors always have a strong incentive to search out mispricing or information and trade on it. That's what they do.

And furthermore, I could say a former PhD student of mine, Da Huang, who now is a professor at North Eastern, his PhD thesis actually explored that as passive investing rises, which active managers drop out? And it turns out that it's disproportionately the worst performers. Those with the worst performance and the worst returns among active managers who see the most outflows are most likely to close up shop. On both sides of the rise of passive, it makes sense that market quality and efficiency of financial markets should be pretty robust to the rise of passive investing. There are very strong forces on both sides that are pushing us back to equilibrium.

Ben Felix: Do you have a thought on at what point it becomes a problem? At what point passive becomes a problem in terms of price efficiency?

Davidson Heath: Again, in our simplified stylized model, you go all the way up to 9999 of the market is passive. And there's literally one guy with $1. And that guy sets the prices and things work out fine. I don't think we will ever get to that point. But I will also say I don't think we'll ever get to the point of it's going to tail off at some point.

The strongest comment on this I could make is that people love to trade actively. And in fact, the main use of ETFs, for example, which are these giant passive vehicles, is for people to trade baskets of stocks in an active way. People are now day trading ETFs, which is such a bizarre counterintuitive thing to be doing in some sense. I think the market can evolve a lot more. And you will still see plenty of active trading, plenty of information gathering. And I think prices will still be pretty efficient for the reasons I said earlier.

Ben Felix: Makes sense.

Mark McGrath: What role do shareholders have in managing agency conflicts between themselves and the managers of the firms they invest in?

Davidson Heath: There's a second paper which we executed on the same topic of the rise of passive and the effects on financial markets. And then this paper really on sort of corporate governance. And the agency conflicts between shareholders and managers is pretty much an evergreen topic in finance. And, arguably, it's the original principal-agent conflict. There are probably thousands of academic studies going back to Mike Jensen who recently passed away. Rest in peace. Back to Berle and Means in the early 20th century. And as usual, with most classic questions in economics, even Adam Smith wrote about it. I actually have – I realized we quoted this in our paper. So I wanted to point out, "Smith 1776. The directors of such joint stock companies, being the managers of other people's money rather than of their own, cannot be well expected that they should watch over it with the same anxious vigilance with which the partners in a private co-partner watch over their own."

Ben Felix: Yeah. It's been around for a while. Okay. That matters a lot. Shareholder's role in corporate governance. How significant is index fund ownership of individual stocks?

Davidson Heath: It is significant. The big three passive fund providers, namely BlackRock, Vanguard, and State Street, between them hold 20 or more percent of all the available shares of every single US public corporation. And for other countries, other stock markets, the fraction is lower. But they are catching up as money flows in and they continue to try to diversify.

And the fraction is continuing to rise everywhere because passive investing continues to rise. The train has left the station. And I should say that passive investor's influence on corporate governance is even larger than that shocking number would suggest. Because owing to their fiduciary duty, these funds show up at every shareholder meeting and they always vote on every item. Other investors generally don't.

Ben Felix: Interesting.

Mark McGrath: That is interesting. I hadn't thought about that. How are index funds incentives to monitor are their portfolio holdings different from those of traditional actively managed funds?

Davidson Heath: It's very difficult to say. It's complicated. And it is still an open debate how the incentives of index funds to monitor differ from active funds. If you ask BlackRock, or Vanguard, or certain of my academic colleagues, they'll say, "Look, these are gigantic long-term shareholders. And as a result, they have a very strong incentive to monitor their portfolio firms."

On the other hand, if you ask others in the industry or if you ask other of my academic colleagues, I would point at a beautiful classic paper now, 2017, by Lucian Bebchuk, Alma Cohen, and Scott Hirst. In that paper they point out that index funds, because of their fundamental business model, they have very little resources or incentives to monitor their portfolio of firms.

The bottom line is that the fundamental business model of an index fund is polar opposite from the business model of an active fund. Index funds do not try to generate alpha. Their only job is replicate the index as closely as possible, as cheaply as possible. Moreover, the index fund industry is brutally competitive. Index fund investors are shopping now on the basis of individual basis points per year in fees. It's viciously competitive. And you've seen this race to the bottom where the fees are staggeringly low now. You can park up money in VTI and Vanguard will replicate this giant total market index for you for five basis points a year. It's incredible. The Bebchuck et al paper argues on that basis that index funds therefore have neither the resources nor the incentives to monitor their portfolio firms. And that controversy is what we wanted to look into and try to get some answers.

Ben Felix: It's super interesting to think about, because I have often thought that what's the difference between large active managers holding most of the assets versus index funds from the perspective of this type of thing? But there is actually a difference because of their fundamental business model. How do you measure monitoring? How do you measure how well index funds or active funds are doing monitoring the companies they own?

Davidson Heath: Our main measure of monitoring, we look at how funds vote at shareholder meetings on contentious items. Those are items where the firm management and ISS, the third-party recommender, ISS, Institutional Shareholder Services, made different recommendations on how shareholders should vote. And we see those items as being where the funds really have to show you where their allegiance lies.

Mark McGrath: I looked up VTI just to be sure. I think the expense ratio is actually .03. It's even lower. Market beta is basically free at this point. Empirically, how well do index funds monitor relative to active funds?

Davidson Heath: Index funds are much more management-friendly. For example, we find that across the board on these contentious items. For example, when management recommends a yes vote but ISS, the third party, recommends a no vote, active funds vote yes 44% of the time while index funds voted yes 53 % of the time. That's a 9-percentage point gap, which in this voting literature in this area is huge.

Ben Felix: We have this agency conflict between the shareholders and the managers. And index funds are more often voting with management compared to active funds.

Davidson Heath: That's right. And you'll notice that the active funds come in at 44%. You're talking about a world 20 years ago where, on average on contentious votes, management did not get what they wanted. As we move more and more and more toward passive recall, passive-voted managements weigh 53% of the time. You're potentially tipping into a world where, on average on these contentious items, management gets what they want.

Ben Felix: Okay. We talked about the low cost of index funds being potentially part of the overall incentive issue here. How do the expense ratios of index funds affect their quality of monitoring?

Davidson Heath: We find that index funds with higher expense ratios, the voting gap actually shrinks. That is to say they are relatively less management-friendly. And they vote a little bit more like active funds would. And I should say this finding is consistent with another classic fairly recent paper by Jonathan and Katharina Lewellen. And that paper argues that index funds do in fact have some incentive to monitor based on their feeds. You don't get bonused if you run an index fund on the basis of generating alpha. That's not your job.

But if, Mark, you run VTI and you have some intervention you could do, that would be accretive to firm value for. First of all, they point out your fiduciary duty would require you to do that intervention. Second of all, they point out, if you can raise one of your portfolio firm's values, your AUM goes up and you collect extra fees on that. And, also, investors may notice that your returns are looking a little better. And as a result, you may attract more AUM through that indirect channel. And the fees do provide you with an incentive. And our finding is consistent with that story.

Mark McGrath: Intuitively, that seems to make sense. But, well – how do we know that index funds aren't engaging with management beforehand and then voting in support of the proposals that they've pre-negotiated?

Davidson Heath: Yeah. This is a really good question. There are three main pillars as to how shareholders can monitor and make themselves heard. This is corporate finance 101. We divide it into voting, exit, and engagement. We said on voting there's this really, really sizeable gap that we observe. On exit, you almost don't even have to check. But we did go and check just to be sure. And, yeah, it turns out that active funds turn over their positions enormously more than index funds do. Therefore, the threat or the possibility of exit is enormously weaker for index funds.

That leaves just the possibility that they're making it all up on the side of engagement. That they're engaging way more behind the scenes. Now engagement is very hard to measure. I do have a couple of projects in the hopper that I won't tip my hand yet. But we're working on that. But on this project, we do have one window on engagement. Because if a mutual fund owns more than 5% of a particular firm, it's a block-holder. And it has to file one of two forms with the SEC in order to disclose that fact.

Form 13D is the form that the fund will file if it has the intent to – there's some legal language. But it's something like the intent to engage with or alter the conduct of the firm. And then there's also form 13G, which the reporting requirements are less onerous. But it requires that the fund has no such intent. We went and we looked. We scraped the SEC website and we did a bunch of data analysis. And it turns out, interestingly, it's pretty simple to state, index funds always file the second form, 13G. In fact, we could not find a single event in our sample ever in which an index fund filed the activist form 13D where they stated that they intended to engage or alter the firm's conduct.

Ben Felix: Man, that's crazy. The voting index funds are doing worse. But there's this possibility where they're doing engagement behind the scenes. And they're always voting with management because they've convinced management to do what they want beforehand through engagement. But they're never following the form that would suggest that they're actually doing any engagement.

Davidson Heath: Correct. You want to be a little skeptical, of course. So you say, "Well, those 5% block holder, that's maybe a minority of observations. You're not seeing everything. Well, yes. But those 5% block-holder holdings are precisely where you would expect them to engage the most because they have the most ownership and the most at stake. And they don't.

Ben Felix: That's wild. How does index fund ownership affect other firm-level governance issues?

Davidson Heath: We find no effect at all on measures of intrenchment such as – think poison pills, or staggered boards, things of that nature. On the other hand, those entrenchment measures – dual-class share status. Another classic measure of entrenchment. On the other hand, if you think about dual-class share status or poison pills, those are very slow-moving. You generally need like a super-majority vote just to change that. You might well expect that they'd be very slow to react.

On the other hand, we do find pretty clear effects on managerial compensation. A bump to your index fund holdings and a reduction in your active holdings is followed by lower pay performance sensitivity and higher total compensation for the firm's top managers. And that seems pretty clearly worse for firm governance.

Mark McGrath: So then what's your main conclusion on how index funds affect monitoring and corporate governance?

Davidson Heath: Let's recap. Index funds, we find vote in a way that is significantly more management-friendly. They're less prone to exit or engage with their portfolio firms. And increased index fund holdings are followed by lower pay performance sensitivity and higher compensation for top managers. And, overall, it paints a picture that the rise of index funds may be leading to worse oversight and less monitoring.

Ben Felix: Which is bad. At some point, that's destroying value for shareholders. That's the problem.

Davidson Heath: You're absolutely right, Ben, that it's going to eventually be value-destroying. But even more than that, you can point to Broccardo, Zingales and Hart, for example, and other folks who have been pointing out the rise of the desire for socially responsible conduct, socially responsible investing, and socially responsible conduct by the firm. So-called stakeholder capitalism. And worse governance and less monitoring and oversight by your shareholders is worse for all stakeholders.

Ben Felix: You've got a paper related to that topic too that we're going to get to in a second. But we've talked about how index funds are not really affecting price efficiency because there's this equilibrium. And that's all fine. That's great. But they're also deteriorating corporate governance, which is not so great. What are your just general thoughts on are index funds good? Or is there a problem here?

Davidson Heath: For the investors who are able now to park their money in VTI and get maximum diversification for, as Mark pointed out, three basis points a year, this is unquestionably just a gigantic boon. And hats off to Jack Bogle, Vanguard, and all the folks working in the passive fund industry. They've done a great thing.

What are the potential down the road unintended consequences of this? That's the real question of this whole line of my research. We think of financial capital more broadly in the sense of everything on the right hand of the firm's balance sheet is funding for the firm one way or the other. And that's financial capital. And we generally taxonomize it according to two dimensions, cash flows and control rights.

Bonds have very, very defined cash flows and relatively weak control rights, except if you're in default. Stocks have much hazier cash flows but stronger control rights. This kind of thing. Now, my dear, dear friend and brilliant co-author, Matt Ringgenberg, once put it really nicely. I think we were out drinking bourbon at some place downtown in Salt Lake City, which is where we get all our best ideas, and Matt said, "The market for cash flows is really, really competitive and efficient. And the market for control rights really is not." And I think that's a great little aperçu that stuck with me.

On the financial side, there are lots and lots of really strong incentives, and money at stake, and mechanisms that push us back towards efficient liquid functioning. On the governance side, if you think about the market for corporate votes or the market for corporate control, much less so. That's the reason why I suspect that, with the continued rise of passive, we will probably still continue to see financial markets functioning well.

In fact, if you just look in the very broad patterns over time, financial market's liquidity and most measures of efficiency have just been getting better over time over the last two decades or so. If there are unintended consequences to this rise of passive, I think we're going to see it on the side of corporate governance, the side of control rights.

Ben Felix: Do those things at some point collide though?

Davidson Heath: That's a difficult question. I'm skeptic on this. I'm more we should just do batch auctions at the end of every day and 90% of trading and talking about trading. We should shut down MSNBC and Fox Business. They're pointless. We should just close the financial pages and just do one big batch auction at the end of every day. My view is that most of that is just people buying and selling in a casino as it were. I agree, Ben, that in principle, worsening governance could eventually lead to consequences for returns and value. But I don't think it's as bad as you might think.

Ben Felix: That suggest that the agency conflicts, while they may exist, managers aren't evil value-destroying beings.

Davidson Heath: No more than anybody else. No more than shareholders.

Ben Felix: That's interesting. How well do SRI funds do on selecting firms with better ESG characteristics?

Davidson Heath: Across the board, SRI funds invest in firms that have better environmental and social conduct. That pattern is very, very clear and very, very robust.

Mark McGrath: And how do you measure whether SRI funds improve the behavior of the companies they invest in?

Davidson Heath: This is the second part of the question. The first part of the question was do they do what they say by investing in firms that are scoring better currently on environmental and social grounds? The second question is do they have impact? And, once again, to test this second question, you really want to watch out for spurious correlations in the data. Firms might be improving their DNS conduct for other reasons.

ESG and socially responsible investing became a very hot topic for everybody in the last decade or so. In this case, in order to try to duck those spurious correlations, we made use of a quirk in the Morningstar star rating methodology. As you know, Morningstar rates all extent mutual funds and assigns them one through five stars. And it turns out, there's a quirk in how they do that that produces unrelated flows of investor capital in and out of funds. You can use that and say, "Okay, that's free from those spurious correlations." And we used that approach as well as some others. And we looked at the changes in those 18 real environmental and social outcomes that I mentioned.

Ben Felix: I don't know if I understood the Morningstar star rating thing. Can you explain it again?

Davidson Heath: Sure. What we needed was some change in how much SRI fund investment there was in a given firm. And you needed to be unrelated to anything else, like broad trends and hotness of the topic. Even hotness of the SRI funds as a whole. It turns out that when you wrap around into January of each year, there's a rounding that happens where an SRI, where any given fund, can come in with, let's say, the unadjusted Morningstar number is like a 3.49. In that case, it gets rounded down and that fund gets three stars.

A similar fund that had a slightly better return, even a slightly better five-year lag return, might come in at a 3.51. It gets rounded up and it gets displayed and bumped up in the rankings with four stars. That fund will get higher inflows. But you can see that that pattern is not related to anything about the fund itself. It's purely just a quirk of the Morningstar star ratings.

Ben Felix: How well do firms do on improving the behavior of the firms they invest in?

Davidson Heath: They don't. The bottom line is that, across those 18 different real firm outcomes that we measure and across a variety of approaches from simple associations, to time series analysis, to the carefully identified shocks due to the Morningstar star ratings, across the board we find zero effects of socially responsible investing on firm conduct.

Ben Felix: Zero.

Davidson Heath: Zero everywhere and across the board.

Ben Felix: And it was 71% of firms that said they are doing that?

Davidson Heath: 81% of SRI funds claim that they impact or that they work toward impact.

Ben Felix: 81. SRI funds are saying that they are going to pick stocks that have SRI characteristics, which they are doing. And then a lot of them are saying they're going to change firms and they are not doing that.

Davidson Heath: Correct.

Ben Felix: Very interesting.

Mark McGrath: How much effort does it look like SRI funds are actually putting into changing the firm behavior then?

Davidson Heath: It was documented in a previous paper, not ours, that SRI funds are significantly more likely to vote for environmental and social-oriented shareholder proposals. And the gap there is quite large. It's maybe even 10% or 15%, which is very, very large. They are voting in the way that you would expect.

However, we then looked at whether or not SRI investment makes at the firm-level these ENS-focused shareholder proposals more likely to pass. And they don't. It's unclear whether the funds are putting in all the effort that they can and it's just not effective. Or whether they're putting in a token effort. Either of those could be correct. But either way, we again see no effects on firm conduct.

Ben Felix: Interesting. That's not really greenwashing. Because they're holding green firms. What would you call this?

Davidson Heath: Very good. We say it's right. These SRI funds are not exactly greenwashing because they are delivering at least part of their claim that they will park your money in green firms. At a wonderful conference in Oxford, one of the audience came up with the word impact washing. I don't know if that's going to catch on. But we said they are not greenwashing. But it looks like they are impact washing.

Ben Felix: Like you mentioned earlier, ESG and SRI investing, I think it's died down a little bit now. But even a few years ago, it was this big thing that everybody was asking about. What do you think investors need to understand based on your research before they commit their capital to a socially responsible fund?

Davidson Heath: SRI or ESG funds will invest your money in a green subset of firms. However, they don't appear to have any impact on actually any real-world outcomes on actually changing any firm's behavior. And if you want to think about the reasoning, I would say, as econ 101 would suggest, look again at their resources and their incentives.

On the resource side, these funds are actually still very, very small. In dollar terms, you commonly see – again, a couple of years ago, this was very hot. But you would see these stories where would say ESG investing or SRI investing is growing really, really fast. And you see this chart that goes from bottom left to top right. But those would always be in relative terms.

In dollar terms, the entire SRI investing industry is still on the order of tens of billions of dollars. Maybe it's gotten up to over 100 billion now. I don't know. That's a lot of money. Don't get me wrong. But when you consider that the aggregate value of the US Stock Market is currently around 50 trillion, if you look at the big passive managers, I've claimed that the rise of passive is potentially having widespread effects, BlackRock just passed – was it 11 trillion in total assets under management? That's pretty plausible that they could be having these big effects.

But most SRI funds, the total SRI fund holdings of a typical firm in our analysis that we found was about 0.1% of the firm's market cap. Even if they are showing up and engaging and voting like crazy, it's difficult to see how they're going to have a real effect. That was on resources. The second thing on incentives I would say is that the ESG ratings that these funds compete on and get rated on are not based on some measure usually of impact. They're based on the ESG ratings of the firms that they hold currently.

The clear incentive is to park your money in already green firms and then pretty low effort from there on out.

Ben Felix: I guess if someone wants to hedge climate risk, they can still hold ESG firms. But if they want to change the way companies are behaving, SRI fund is probably not going to help so much.

Davidson Heath: I'll shout out another classic paper by Sam Hartzmark and Kelly Shue pointing out that if you put capital towards these green firms and starve the brown firms of capital, raise the cost of capital for the brown firms, that might make them even more short-termist and extractive. It could even have the opposite effect that you desire.

Ben Felix: Yeah. That's a cool paper. We had Sam on and he talked about that paper. Moving on to some of the other stuff on how ETFs are actually replicating indexes, can you talk about how ETFs approach replicating their benchmark index?

Davidson Heath: This is another topic, another paper in the similar direction of how do these changes in institutional investing and even institutional investing technology, how do they affect underlying financial markets and perhaps the real economy? This paper starts with really a thought experiment. If I hand you, Ben, a billion dollars and I say, "Okay. Go replicate the total US market index." You are probably not going to go out and lift every single offer on every single stock and just hit the buy button repeatedly. Because you're going to bang the market in the wrong direction. You're going to get terrible execution.

Many of the members of your target index are tiny stocks with very low index weights and they're highly illiquid. And after one second of thought you realize, "Okay, I could safely omit those." As a matter of fact, nowadays you could get about 25% of the index by buying the magnificent seven giant tech stocks because they're so huge.

In general, most index funds and ETF system especially, because liquidity and transaction costs are so important for ETFs, use this so-called sampling approach to replicating their underlying index.

Mark McGrath: And how do ETFs that sample decide which securities to include in their sampling basket?

Davidson Heath: This is a case of where my former student, Da Huang, who I mentioned. Great guy. He and I were literally like, "Let's write down a simple model of if we were trying to solve that problem and just see what comes out. Because out of sheer intellectual curiosity." And it turns out that this is pretty close to what fund managers actually do at least to a first approximation. They're probably much more sophisticated. But they're laser-focused on the trade-off between tracking error and transaction costs. They want to minimize both. But you can't set one to zero without blowing the other one out as we just said.

And it turns out you want to pick stocks that have a high weight in the index. Obvious. You want to pick stocks that have a high correlation with the index. A high beta to the index. Because you're getting a two for one there. And third, you want to pick stocks that have very good liquidity, very tight bit ask spread so you can minimize transaction costs.

Ben Felix: What effect does the sampling choice have on authorized participant arbitrage?

Davidson Heath: The way this works is that the ETF sponsor, who's the custodian, daily, they post their basket of stocks and proportions and they say, "That's my basket that I'll exchange for shares of the ETF." And then they leave it up to the authorized participants who are generally big, high-frequency trading firms to arbitrage the basket against the ETF shares. And that's how ETFs achieve this astonishingly low-tracking error millisecond-to-millisecond and at the same time have very low transaction costs.

If the basket consists only of the very large liquid stocks, that's where the APs are going to go to trade. And if the basket omits that left tail of the many small illiquid stocks, then trading in the ETF is not going to prompt any trading in those small illiquid stocks as a result.

Mark McGrath: And how does increased AP or authorized participant arbitrage in the most liquid stocks affect market correlation and market quality?

Davidson Heath: We find that it consumes liquidity and makes liquidity worse in those stunts. And it leads to higher intraday volatility of the prices and a higher correlation with the market as a whole.

Ben Felix: And then how does that compare to illiquid stocks?

Davidson Heath: In illiquid stocks, we find either no effects at all, which is intuitive. Or in some cases, small opposite effects to what you find in the large liquid stocks. And that could be because trading activity as a whole is reallocated and tilted out of those stocks and into the ETFs and the large liquid stocks. As we said, ETFs are a wonderful passive investing creation which are chiefly used by people to day trade on an active basis all day long.

Mark McGrath: What are the implications of your findings for index ETF investors?

Davidson Heath: For ETF investors, this whole mechanism that we described and this whole phenomenon is again just unquestionably a boot. Because sampling replication is what gives us this universe of ETFs. I think, in fact, I don't know when this happened. This might have been a few years ago now. We passed the point where there's more ETFs traded than there are actual individual stocks. You have this universe of ETFs with perfect liquidity and perfect millisecond tracking error on-demand. And sampling replication is a big part of that.

Ben Felix: What about the implications for financial markets more broadly speaking?

Davidson Heath: For financial markets, it's much more of a question and much more nuanced. A huge part of trading activity is now in ETFs. Most trading and, indeed, most price discovery nowadays is happening in ETFs. Not in individual stocks, for example. In individual stocks, especially in times of market disruption or crisis, you increasingly see mile-wide bid ask spreads, severe illiquidity, and stale prices. And the illiquidity and the stale pricing in individual assets, especially those that were already the smallest and least liquid, is potentially a concern for the long-term function of the markets, I would say.

Mark McGrath: Switching gears a little bit. What is cyborg trading?

Davidson Heath: Cyborg trading is a project that I co-direct with Aspire Commodities, which is one of the top, top macro strategy/hedge funds/family office founded by Adam Sinn. And the first thing to know is that cyborg trading is named after the classic 1980s cheesy action film starring Jean-Claude Van Damme. If you haven't seen it, Ben, you're too young. But it's worth watching. Just to give you some idea, the villain's name is Fender Tremolo.

Mark McGrath: What year is that?

Ben Felix: Sounds ridiculous.

Davidson Heath: 1980s sometime. I saw it in the theatre.

Ben Felix: What's the fundamental hypothesis of cyborg trading?

Davidson Heath: The fundamental hypothesis of the project is that machine intelligence, AI, and human intelligence are not substitutes but they are complements on a fundamental level. All of the recent stunning advances, and they are stunning in AI, have come from an approach of fitting more and more larger and more flexible models with billions and now potentially approaching a trillion free parameters. And essentially no built-in a priori decision rules.

Human intelligence does not work that way. We use a very small sample of data generally. And we combine it with these built-in decision rules and heuristics that have been shaped by a billion years of evolution. As a result, the machines are getting better and better and they're superhumanly good at sifting the data for subtle statistical patterns that a human can never match because we have attention and cognitive limitations.

But the machines will, in general, be very bad at novel situations and at antagonistic situations because, fundamentally, there's no world model. It's only these robust statistical patterns. They're very bad at dealing with something they haven't seen before. You may have seen the news about the most powerful AI Go playing robots which, when they play a conventional strategy, are super humanly good and can't be beat. But it turns out a simple but very bizarre counterintuitive strategy enables even a relative amateur at Go to beat the most powerful go AIs.

So far, the project has, knock wood, been going well. But the broader point is that this is enormously reassuring for me personally as a human being and as a dad. Because the implication is that, for the foreseeable future, AI is not going to outcompete and replace humans and put us all out of a job. Rather, the biggest gains are going to come from figuring out on an individual level and also on a corporate level. And in terms of organizing our workflows and our hierarchies in such a way that the AIs and the humans can work together optimally.

And we've been seeing that in more and more fields recently. Radiology is an example. Coding. The coding copilots. The profusion of those as an example. Even Ethan Malik and co-authors did an experiment where they said exactly it was some sort of cyborg model that did the best when they gave AI assistance to BCG, Boston Consulting Group, consultants. To me this is an enormously reassuring message.

Ben Felix: You're telling us that we're going to be able to keep our jobs basically.

Davidson Heath: Fingers crossed, baby.

Ben Felix: Man, I've been playing with AI a bit in simple stuff, like interpreting research papers. And one of the things that I find is that it'll say stuff that sounds really smart. But if it's a paper that I already know well, there will often be something where I'm like I know that's wrong because I know this paper. But I would not have known it was wrong if I didn't already know the paper. And the AI sounded really smart. And I've called it out and been like, "Hey, that's wrong. And I know it's wrong." And the AI is like, "Oh, sorry." What? But it sounded so confident. It's almost like you need a certain level of expertise to know whether what the AI is saying is right.

Davidson Heath: I think that's exactly right. And I think humans will be in that role for a good long time to come.

Mark McGrath: I noticed that, too, Ben. I'll catch up making mistakes and you call it out immediately. It's like you're right to question that. Thank you. And then it gives you the correct answer. But to your point, if you didn't know that there was a mistake there. It's hugely helpful and useful in a lot of cases. But to blindly trust it I think is dangerous at best. I don't know a lot about AI. But there's other tools out there that you can just train only on your own resources that you've fed at that don't hallucinate as much, I'm told. And so, that's a really interesting area of AI I think. But this has been incredible. Thanks, Davidson. Our final question, how do you define success in your life?

Davidson Heath: Although my CV says finance professor, somewhat oddly perhaps, I do not consider money to have any relationship or to be really any part of success. Money itself is not a good in both literal and figurative terms. It's just a point system. I think instead that if you have something in your life that you consider to be worthwhile and a worthy task for you, and that contributes something into the world, could be building a trillion-dollar company, could be pushing a broom at that trillion-dollar company, could be painting a painting or writing a song. If you can pour yourself into some task or endeavour that you feel is worthwhile and contribute something into the world, then, in my opinion, whether your account says $10 or 10 billion dollars, you are a success.

Ben Felix: That's a great answer. Do you get that from the research that you do?

Davidson Heath: Yes. I would say one realization I had fairly recently was that I was much happier as a quant than I was as a trader even though the money and prestige was so much better on the desk. But working as a quant, I could sit there and I could build like a little software tool or I could do an analysis and write a paper. And I could take as long as I needed to get it just right. And it felt like working in the wood shop with my dad back when I was a kid doing carpentry stuff. And I realized that's what I really like. And the money – what is money? It's just some points on a screen? It's like a video game.

Ben Felix: All right. Well, this has been great, Davidson. Thanks a lot for joining us.

Davidson Heath: Thank you, guys, very much. It's my pleasure.

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.