Why Indexing Is Not Making Markets Inefficient

The Concern Sounds Plausible…

With billions of dollars flowing into market-cap weighted index funds, it seems like a valid concern that larger stocks, which make up a larger proportion of a given market-cap weighted index, are being irrationally bid up in price whilst smaller stocks are being forgotten. If this is happening, large popular stocks such as those in the S&P 500 could become systemically overpriced and smaller, lesser-known stocks could become systemically underpriced.

That concern sounds intuitive. If new money is constantly being directed into market-cap weighted funds, and those funds mechanically buy more of the largest names because those names already have the highest weights, it is natural to wonder whether indexing is creating a self-reinforcing loop. Indeed from this perspective, the big would get bigger not because their fundamentals justify it, but because the structure of ‘passive’ investing keeps sending their prices upwards via ‘irrational’ demand.

…Until You Realise That Prices Are Set by Trading

To understand how index funds might affect stock prices, we need to understand how prices are set. Prices are set by trading. Each trade is effectively a vote for the price to go up or down. The aggregation of all of these votes is the current price, which is the market’s best guess at the value of a company.

If index funds were the only entities placing trades, buying up more of the biggest stocks simply because they have the largest weights in a market-cap weighted index could indeed cause serious issues. In that scenario, prices could become detached from fundamentals because there would be no meaningful force pushing back against mechanical demand from passive investors.

But this is misguided thinking. The relevant question is not how much of the market is indexed, but how much of the trading index funds are doing. If index funds are not doing the majority of trading, then active managers are still dominating price discovery and thus this ensures that prices remain a fair reflection of fundamental value. BlackRock makes this exact argument in its 2017 paper, estimating that for every $1 of US equity trades driven by index strategies, active stock selectors trade roughly $22. The same paper also argues that trades arising from ETF creations and redemptions account for only around 5% of all US stock market trading (BlackRock 2017).

If Indexing Did Create Distortions, Active Managers Would Have an Incentive to Correct Them

We also have to think about market dynamics. Even if index funds did get to the scale where they could create meaningful price distortions, each distortion would represent an opportunity for an active manager. If there are more distortions, there is more profit available from correcting them. That should attract more active capital back into the market.

This is closely related to the Grossman-Stiglitz paradox. If the market were perfectly efficient, everyone would index, because there would be no reward for trying to beat the market. But if everyone indexed, there would be nobody left doing price discovery and prices would stop reflecting underlying information properly. At that point, active management would once again become profitable, because there would now be mispricings to exploit. Grossman and Stiglitz argue exactly this in their classic 1980 paper: perfectly informationally efficient markets cannot exist in equilibrium because, if prices fully reveal all information, nobody has an incentive to collect that information in the first place (Grossman and Stiglitz 1980).

So markets cannot be perfectly efficient because, by the nature of perfect efficiency, they would become inefficient. There must always be at least some active management in the system to keep prices broadly in line. The exact tipping point is not something the literature pins down with precision, but the self-correcting logic is the key point. Once passive gets large enough to create serious distortions, active has a greater incentive to step in and exploit them (Grossman and Stiglitz 1980).

Indexing May Be Pushing Out Bad Active Managers, Not Breaking Markets

If the costs of uncovering and evaluating relevant information are low, then it does not take much active investing to keep markets efficient. From this perspective, the rise of indexing may actually be pushing bad active managers out of the market, leaving only the more skilled ones behind. If that is true, then the remaining active capital could actually be making the market more efficient, not less.

That, in essence, is the core argument against the claim that indexing is breaking markets. Passive does not need to do price discovery itself. It only needs there to be enough active capital left in the system to keep prices broadly sensible. Once that condition is met, the mere growth of passive ownership does not imply a collapse in market efficiency.

Most ETF Trading Does Not Touch the Underlying Securities

With ETFs becoming an ever more popular fund structure, understanding the basic mechanics of ETFs matters. The majority of ETF trading happens on the secondary market. That is simply ETF unit holders trading with one another. It is only when there are deviations between the price of the ETF and the value of the underlying securities, which we can think of as excess supply or demand for ETF units, that authorised participants step in to create or redeem units. Unlike secondary-market transactions, the creation and redemption process does require trading in the underlying stocks.

This distinction is crucial. In ‘Setting the Record Straight: Truths About Indexing’, Vanguard shows that the vast majority of equity ETF trading, 94% on average, is done on the secondary market. That means ETF unit holders are generally buying and selling from each other without touching the underlying securities (Rowley, Hirt, and Wang 2018).

This point is important for two reasons. Firstly, it helps explain why ETFs make up such a small proportion of overall trading in the underlying market. Secondly, it helps explain why concerns about the liquidity of the underlying holdings are often overblown. Most ETF trading does not touch the underlying securities at all. Vanguard and BlackRock both make this point clearly, even if they express it using slightly different datasets and framing (BlackRock 2017; Rowley, Hirt, and Wang 2018).

So, broadly, active managers are still doing the overwhelming majority of price discovery in the underlying securities, and prices are still being set primarily by active trading at the margin rather than by passive ownership itself (BlackRock 2017; Rowley, Hirt, and Wang 2018).

Michael Green’s Concern

Michael Green, portfolio manager and chief strategist at Simplify Asset Management, argues something much more troubling. He essentially says that passive index funds run by the likes of Vanguard, BlackRock, and State Street have become too large and are now causing price distortions that could have huge negative implications down the road. A useful public summary of this debate appears in the Rational Reminder discussion featuring Michael Green and Randolph Cohen, which frames the disagreement around whether passive flows are making markets increasingly ‘inelastic’ and whether that could produce meaningful long-run distortions (Rational Reminder 2024).

His argument is that so many pension and investment contributions are now being funnelled into market-cap weighted index funds that the largest stocks receive a constant mechanical bid. In his telling, the biggest companies absorb a disproportionate share of new contributions simply because they already have the largest weights in the benchmark. This, he argues, leads to these companies trading at valuations that may differ materially from intrinsic value. Prices keep rising not because fundamentals justify them, but because the flow of money into index funds keeps pushing up the largest holdings (Rational Reminder 2024).

The real problem, in Green’s framework, occurs when the direction of flows changes. During accumulation years, workers continue contributing into pensions and investment accounts, which supports demand for the underlying securities. But when people retire, they stop contributing and begin withdrawing. If asset managers hold very little cash, then they must ultimately meet redemptions by turning to the market and selling underlying securities to raise cash. That, in his view, could place downwards pressure on prices at exactly the moment when valuations are already too high (Rational Reminder 2024).

Green’s concern is therefore not just that prices may be somewhat distorted today. It is that the structure of modern pension saving, especially through automatic contributions into defined contribution plans, SIPPs, 401(k)s, and target-date funds, could create a system in which inflows mechanically push prices up for years and then set the stage for a painful reversal later. He also argues that older defined benefit schemes did not inflate stock prices in the same way, because the institutional structure of retirement saving was different.

Interestingly, Green does not think this should necessarily cause an individual investor to abandon a buy-and-hold indexed strategy. His argument is more game theoretic. If prices continue to rise irrationally, then it may still make sense for the individual to keep participating in the system whilst that remains true, even if the broader social outcome is ultimately fragile. In other words, what may be individually rational need not be socially optimal (Rational Reminder 2024).

Randy Cohen’s Response

Randolph Cohen, senior Lecturer of Entrepreneurial Management in the Finance Unit at Harvard Business School, disagrees completely with these views. His response to the first point is that target-date funds do not simply push money blindly into markets. They also rebalance. If market prices move in such a way that the portfolio drifts away from its target asset allocation, the fund automatically trades against that drift. So, for example, if rising interest rates cause bond prices to fall sharply and equities do not fall enough, a 60/40 portfolio might drift to 70/30. Rebalancing then requires buying the now cheaper bonds and selling the relatively more expensive equities. In that sense, the mechanism can work against mispricing rather than reinforce it. This is one of the central counterarguments in the Green-Cohen debate (Rational Reminder 2024).

Cohen also rejects the claim that the move from defined benefit to defined contribution pensions can explain a huge inflation in equity prices. His view, as summarised in that discussion, is that older pension structures are also materially invested in markets, and that the flow effects Green is worried about should decay too quickly to justify the scale of valuation distortion Green suggests. Thus, whatever price effects flows do create are not remotely large enough, or persistent enough, to justify the idea that markets should be 50% lower than they are today (Rational Reminder 2024).

From this perspective, target-date funds, or indexing more generally, may actually make markets more efficient rather than less. They can introduce a disciplined rebalancing process that pushes capital towards relatively cheaper assets and away from relatively more expensive ones.

If the Distortions Were That Large, Active Management Should Be Able to Exploit Them

This point is important. If indexing really were causing substantial and persistent price distortions, then active management should be able to exploit those distortions and generate excess returns, or ‘alpha’, over and above a given index.

Whilst I do personally believe that some managers are skilled enough to generate consistent excess returns, even after costs, those managers are exceptionally rare. It is even more difficult to identify them in advance. So, even if one accepts that some degree of price distortion may exist at times, it does not follow that this creates an easy or reliable opportunity for investors to improve outcomes by abandoning indexing for active management.

Jonathan Berk and Jules van Binsbergen are useful here. They argue that skill can exist in active management, but that investors do not necessarily capture it easily because competition drives capital towards skilled managers and erodes the abnormal returns available to clients. In other words, even if skill exists, exploiting it in practice is not straightforward for end investors (Berk and van Binsbergen 2015).

That matters because one of the strongest practical objections to the ‘passive is breaking markets’ story is simply this: if the distortions were that large, active managers should be able to profit from them more consistently than they appear to do in reality. The fact that identifying outperformance remains so difficult is not proof that no distortions exist, but it does make the most extreme version of the story harder to sustain.

The Big Three and Corporate Governance

There is also a separate, though related, concern about the growing influence of the ‘Big Three’ themselves. This is more of a corporate-governance issue than a pure market-efficiency issue, but it often gets mixed into the broader passive-investing debate. Lucian Bebchuk and Scott Hirst argue that index funds own an increasingly large proportion of American public companies and that the stewardship decisions of index fund managers can therefore have a profound effect on corporate governance (Bebchuk and Hirst 2019).

That is a legitimate issue to think about, but it is still a different issue from whether indexing is making markets inefficient. The concentration of voting power may be a governance concern. It is not, by itself, proof that stock prices are no longer being set efficiently.

The Academic Caveat

To be clear, none of this means passive investing has no effect on markets at all. Marco Sammon finds that higher passive ownership reduces the degree to which stock prices anticipate earnings announcements. He estimates that the rise in passive ownership over the last 30 years has reduced the amount of information incorporated into prices ahead of earnings announcements by roughly one-fourth of its whole-sample mean. That is an important caveat, because it suggests that higher passive ownership may weaken one dimension of price informativeness even if it does not broadly break price discovery altogether (Sammon 2025).

So the sensible conclusion is not that passive has no effects. It is that the evidence does not support the much stronger claim that indexing has made markets broadly inefficient or incapable of pricing securities sensibly.

Concluding Remarks

Stepping back, I think the key point is this. The fact that more people are indexing does not mean prices are no longer being set by informed trading. Prices are still set at the margin by those who trade, and active managers still account for the overwhelming majority of that trading in underlying securities. Most ETF trading does not touch the underlying stocks at all. And if passive investing ever did become so dominant that prices stopped reflecting fundamentals properly, that very fact would create stronger incentives for active investors to step in and exploit any distortions (BlackRock 2017; Grossman and Stiglitz 1980; Rowley, Hirt, and Wang 2018).

So the argument that indexing is making markets inefficient is, in my view, wrong. There may well be edge cases, frictions, and pockets of distortion. There may also be important questions around market structure and corporate governance. But that is very different from saying that market-cap weighted index funds are breaking price discovery altogether.

If anything, the more plausible conclusion is that indexing has put pressure on bad active management, not on market efficiency itself. As long as enough active capital remains to do the work of price discovery, and there is strong reason to think that it does, indexing can continue to grow without causing the kind of systemic pricing failure that some critics fear (BlackRock 2017; Rowley, Hirt, and Wang 2018).

References

Bebchuk, Lucian A., and Scott Hirst. 2019. ‘Index Funds and the Future of Corporate Governance: Theory, Evidence, and Policy’. Columbia Law Review 119 (8): 2029–2146.

Berk, Jonathan B., and Jules H. van Binsbergen. 2015. ‘Measuring Skill in the Mutual Fund Industry’. Journal of Financial Economics 118 (1): 1–20.

BlackRock. 2017. Index Investing Supports Vibrant Capital Markets. New York: BlackRock.

Grossman, Sanford J., and Joseph E. Stiglitz. 1980. ‘On the Impossibility of Informationally Efficient Markets’. American Economic Review 70 (3): 393–408.

Rational Reminder. 2024. ‘Episode 332: Randolph Cohen & Michael Green’. 21 November 2024.

Rowley, James J., Jr., Joshua M. Hirt, and Haifeng Wang. 2018. Setting the Record Straight: Truths About Indexing. Valley Forge, PA: Vanguard.

Sammon, Marco. 2025. ‘Passive Ownership and Price Informativeness’. Management Science 71 (6): 4582–4598.