Are index funds killing competition? By Matthew Yglesias

Read time: 10 minutes

Are index funds killing competition?

An intriguing question without a clear answer

There are four drug stores within a short walk of my house, which gives me a lot of flexibility if I want to run multiple errands. But it doesn’t offer much in the way of actual competition because three of the four drug stores are owned by CVS, which is far and away the dominant player in the DC drug store marketplace. As a consumer, to the extent that I enjoy meaningful competition, it’s from the fact that most of the items sold at the CVS stores (or the one Walgreens) can also be purchased at one of the nearby supermarket chain locations. Some are available at 7-11, some at Target. And, of course, all brick and mortar retailers face competition from Amazon.

So it is a competitive retail market, it’s just not as competitive as “wow, there are four different drugstores nearby!” makes it sound, because from a competition standpoint, it matters who owns the stores.

But who owns CVS?

Well, nobody in particular. It’s a broadly owned public company, so its main shareholders are generic investment funds — Vanguard, BlackRock, and State Street are the top three owners. Walgreens is different. Through a baffling series of business deals, an Italian guy named Stefano Pessina managed to merge his family’s pharmaceutical wholesale company with a French one, which then merged with another company to become Alliance Unichem, which took over a British pharmacy chain called Boots, which in turn took over Walgreens. So now Walgreens is predominantly owned by this Italian guy who lives in Monaco for tax purposes. But after him, the top three shareholders are Vanguard, BlackRock, and State Street.

And this poses an interesting conundrum.

Just as you might live in a neighborhood with four pharmacies, three of which are CVS, a large share of companies are owned by the same broad investment funds that own all the other companies. That doesn’t reflect anything nefarious. For most people, the right thing to do with their money is to put it in an index fund that owns stock very broadly. But if most people put their money in funds like that, then most companies will be owned by the same people who own all the other companies. Which doesn’t seem like how a competitive market is supposed to work.

More interesting than oligarchy

Late last year, while most people were offline and spending time with their families, Bernie Sanders alluded to this cross-ownership issue on Twitter, saying that the tendency of BlackRock, Vanguard, and State Street to be major shareholders of basically every company is “what oligarchy is about.”

I don’t think that’s a very good description of what’s going on. If those three companies were all closely held private firms, then of course their widespread ownership of American companies would be a form of oligarchy.

But that’s not the actual situation.

Like a lot of Americans, the biggest single asset my family owns is our house. Then after that, we own shares of stock via some Vanguard target-date retirement funds and a broad Vanguard ETF. We are richer than the average American household, but we are not billionaires or oligarchs. And the reason Vanguard owns such a large share of so many companies isn’t because of the activities of a small number of oligarchs, it’s because of the activities of a large group of affluent-ish Americans with modest stock holdings that, accumulated across the entire population, add up to a huge amount of money.

Oligarch-types are normally the reverse of this. I’m sure that Elon Musk and Jeff Bezos have some diverse stock holdings. But the main reason they are so rich is that they own very large stakes in Tesla and Amazon, respectively, and those are valuable companies. Tesla is an example of a company over which the BlackRock / Vanguard / State Street troika exerts unusually little influence, since Musk personally owns 13 percent of the company and exerts disproportionate control over corporate affairs. Similarly, Amazon’s largest shareholder is Bezos. If you redistribute the wealth of super-rich founder types to the great American middle class, that could have various benefits. But also, an even larger share of corporate control would end up in the hands of BlackRock, Vanguard, and State Street.

The alternatives to cross-ownership are rich entrepreneurs and weird oligarchical family holding companies.

You see this clearly in the auto industry. The biggest owner of GM is Vanguard. The biggest owner of Ford is also Vanguard. But Chrysler has been incorporated into a big multinational company, Stellantis, whose largest shareholders are the Agnelli family (the founders of Fiat) and the Peugeot family. The biggest owner of Tesla is a weird guy who likes to do reactionary tweets. For Musk, it makes sense to have a lot of his eggs in the Tesla basket because he’s betting on himself. For the Agnellis and Peugeots, most of whom are not actively involved in Stellantis management, it makes less sense, but the tax consequences of trying to diversify would probably be bad. But for a normal middle-class professional trying to save for retirement, it would be (on average) bad investment practice to develop strong opinions about the valuation of Ford vs GM vs Tesla. It’s very hard to beat the market, and if you try, you’ll probably fail. Sound investing advice says it’s better to own everything. And that’s how we end up with tons of cross-ownership.

Is this bad?

It’s kind of funny that a handful of big fund management companies show up on the top five shareholder list of every company. But it’s not obvious that this is bad.

Ford and GM are both heavily owned by the big three, but large blocks of Ford shares are owned by the government of Norway and by Newport Trust. Both Newport and the Norwegian sovereign wealth fund own shares in lots of different companies, but they are not index funds. They deliberately take large positions in select companies based on their internal analysis. And those companies do not invest in GM. Conversely, Warren Buffett’s company, Berkshire Hathaway, has a large stake in GM, but not in Ford.

One possible interpretation of the situation is those kind of major investors are the real owners/controllers of various companies and the big three operate as basically irrelevant silent partners.

They are, after all, mostly just cut-outs for millions of small investors. Letting small investors own diversified portfolios gets us better returns on our investment. Letting us get better returns on our investment increases the savings rate. That higher savings rate means it’s easier for companies and entrepreneurs to access capital and make investments. And that ease of investment means higher productivity, more economic growth, and ultimately higher wages. The cross-ownership is a funny fact but not actually relevant to anything.

Except, maybe it is relevant.

Suppose the price of jet fuel rises, forcing airlines to raise prices to cover the higher cost of flying. But then thanks to a healthy labor market, it turns out that people only cut back on flying a very small amount — traffic has fallen a little in response to higher prices, but overall revenue is up. Now the price of jet fuel starts to fall. We’d normally expect airfares to fall, too. Not because the airlines are nice guys who want to pass the savings on to us, but because of competition. Someone will get the idea of cutting prices to expand market share and thus become more profitable than ever. But once Delta cuts fares, Southwest and United and American will need to cut fares to match. As a result, Delta’s efforts to grow market share will be self-defeating and consumers end up winning. The cool thing about capitalism, though, is that even if Delta management knows that trying to grow market share by cutting prices won’t work, it makes sense for them to do it anyway. They are facing a prisoner’s dilemma with regard to the falling price of jet fuel. If they don’t go for growth, a competitor will beat them to it, so they may as well go first.

And here is where we worry about cross-ownership. The key to the prisoner’s dilemma is that the parties can’t communicate and coordinate. In business, if executives gather around a table and work out a price-setting deal, that’s an illegal cartel. But if the companies have the same major shareholders, you can’t stop the shareholders from knowing about this dynamic and perhaps subtly influencing companies’ behavior.

The empirical record is murky

So what is actually true? Is widespread cross-ownership muting competition throughout the American economy and causing problems?

There was some buzz around this topic years ago, but it seemed a bit ambiguous and I largely forgot about it until Sanders’ tweet. That discussion surfaced some relatively fresh research from Florian Ederer and Bruce Pellegrino that claims this is a big deal. The authors try to take a nuanced look at both the extent to which different companies ownership networks overlap and also the extent to which the product markets they compete in overlap. The idea is that CVS and Target overlap more than CVS and Tesla, but less than CVS and Walgreens, and we should quantify degrees of overlap along with degrees of common ownership.

Their quantitative analysis suggests a meaningful diminution of competition generating both deadweight loss and reallocation of surplus to producers:

We estimate our model for the universe of U.S. public corporations using a combination of firm financials, investor holdings, and text-based product similarity data. We perform counterfactual calculations to evaluate how the efficiency and the distributional impact of common ownership have evolved over time. According to our estimates the welfare cost of common ownership, measured as the ratio of deadweight loss to total surplus, has increased about nine-fold between 1995 and 2021. Under various corporate governance models the deadweight loss of common ownership ranges between 3.5% and 13.2% of total surplus in 2021. The rise of common ownership has also resulted in a significant reallocation of surplus from consumers to producers.

I was kind of hoping to turn this into a flaming hot take like yesterday’s anti-dentite jeremiad, but when Maya looked at the research more comprehensively, it turns out the landscape is still kind of murky. But I like to think that one virtue of the subscription-based model is that it’s possible to sally forth every so often with a non-viral, somewhat equivocal take just because I think it’s interesting and informative.

A 2019 overview of what they call “the common ownership hypothesis” by Matthew Backus, Christopher Conlon, and Michael Sinkinson congratulates early pioneers in this field for raising an interesting question, but says their empirical methods aren’t correct and that there’s no solid evidence of problems. The Ederer and Pellegrino paper is, I think, in part a response to that — an effort to construct estimates based on better identification. On the other hand, a survey by Kristopher Gerardi, Michelle Lowry, and Carola Schenone published in 2023 by the Atlanta Fed says that “across the newest papers employing the most credible identification techniques, there is relatively little evidence that common ownership causes lower competition.” Because Ederer & Pellegrino came out in working paper form in 2022, I was hoping the Gerardi, Lowry, and Schenone paper would discuss it. They do not. Instead, they talk about a different paper from Ederer co-authored with Miguel Antón, Mireia Giné, and Martin Schmalz that is dedicated to exploring a specific causal channel — finding “that managerial incentives are less performance-sensitive in firms with more common ownership.” Gerardi et al raise various objections to their identification strategy that seem somewhat plausible but not definitive.

They also have this big ol’ table of studies, showing that there has been a lot of research in this area and the studies disagree.

I wish I had a more satisfying conclusion to offer, but it mostly seems like an intriguing hypothesis with a lot of research and somewhat mixed conclusions.

What have we learned here?

One thing to say about this is that compared to some of the recent plagiarism scandals or the Jamaican metallurgy fiasco that I covered late last year, there is something pleasantly earnest and rigorous about it all.

We have a politically relevant topic with some policy implications, but no obvious ideological upshot. If it’s true that cross-ownership is a big problem for competition, that probably calls for some new regulations on the mutual fund industry, but it’s not the basis for a social revolution. Everyone seems to agree that finding appropriate identification strategies to get the direction of causation correct is difficult, and so they are trying hard to find plausible strategies. The results aren’t all identical, but that’s at least in part because people are looking at different variables of interest. Over time, the papers seem to be getting better: The initial wave of literature was interesting, but critics found flaws in it and new studies have emerged that try to address those flaws. This is how the knowledge-generation process is supposed to work.

Part of that, unfortunately, is you don’t get a super-clear headline for your popular press writeup. But that’s okay. The system still works, at least to an extent, and I’m glad more people are looking into this issue.

What I’d like to read next are more thoughts as to the potential policy options for addressing it. If the only way to counteract the potentially negative impact of cross-ownership is to do something very costly, then we’d want to be incredibly sure that the negative impact is real. But if we have low-cost options, then we should probably be willing to act even in the face of uncertainty. For now, I’m just not sure. But it’s nice to see a careful, responsibly managed research debate.