# At what point do index funds become unreliable?

Passively investing in exchange-traded funds that track some broad index is often recommended for private investors that don't have the time or expertise for active investment. In particular, index funds are recommended over actively managed funds, because the former, on average and in the long term, have better returns than the latter.

In some sense, index funds piggy-back on active investment decisions by other investors. The price of the stocks that are bundled (or tracked) by index ETF form because active investors assess and analyze the risk and profitability of these stocks (or speculate how those will be perceived by other investors in the future) and sell or buy based on these decisions. In other words, index ETF don't try to beat the market, but they rely on other investors who do.

Now that index ETF are becoming so fashionable, could there be a tipping point at which the market signals that active investors send become so diluted that this "index ETF parasitism" collapses? How would this look like and would it affect only those who invest in index ETF or would it affect the stock market more generally?

To make this question perhaps more on-topic: Is the fact (or presumption) that index ETF rely indirectly on active investment decisions by other market participants, as explained above, a known source of concern for personal investment?

• Great question- I have been wondering the same - when is it time to go against the current? I doubt anyone will have a definitive answer, but I'm hoping for some insights. – Aganju Oct 2 '17 at 15:09
• This might be better for economics.stackexchange.com. It's really not a personal finance question. – Ben Crowell Oct 2 '17 at 19:20
• The fact that so many people try to beat the market today, despite evidence that shows how unlikely success will be, shows to me that we will never have a shortage of active management. – BlackThorn Oct 2 '17 at 21:30
• Each year loads of the active funds DO perform better, but the same fund won't perform better over a number of years. It's not that success is unlikely, it's that success is fleeting and costs too much. You call it semantics, but this nuance is important. – quid Oct 2 '17 at 22:15
• @JoeTaxpayer, absolutely, I also think it's strange that people think they can figure out roulette, but people do. Most of my money is in low-fee S&P Small/Mid/Large cap index funds for the same reason, mathematically it makes sense to take the average return directly especially given that the alternative comes with a ~15x-20x higher fee. It's not that it's impossible or horribly unlikely to beat the average, loads of funds do beat the average, that's how averages work. But, over 30 years the law of averages catches up and the additional fee ate your lunch. It's a simple math issue. – quid Oct 3 '17 at 17:32

A great deal of analysis on this question relies on misunderstandings of the market or noticing trends that happened at the same time but were not caused by each other. Without knowing your view, I'll just give the basic idea.

The amount of active management is self-correcting. The reason people have moved out of actively managed funds is that the funds have not been performing well. Their objective is to beat their benchmarks by profiting as they correct mispricing. They are performing poorly because there is too much money chasing too few mispricings. That is why the actively managed industry is shrinking. If it gets small enough, presumably those opportunities will become more abundant and mispricing correction will become more profitable. Then money will flow back into active funds.

Relevant active management may not be what a lay person is thinking of. At the retail level, we are observing a shift to passive funds, but there is still plenty of money in other places. For example, pension and endowment funds normally have an objective of beating a market benchmark like the Russell 3000. As a result they are constantly trying to find opportunities to invest in active management that really can outperform. They represent a great deal of money and are nothing like the "buy and forget" stereotype we sometimes imagine. Moreover, hedge funds and propreitary trading shops explicitly and solely try to correct mispricings. They represent a very, very large bucket of money that is not shrinking. Active retail mutual funds and individual investors are not as relevant for pricing as we might think.

More trading volume is not necessarily a good thing, nor is it the measure of market quality. One argument against passive funds is that passive funds don't trade much. Yet the volume of trading in the markets has risen dramatically over time as a result of technological improvements (algorithmic traders, mostly). They have out-competed certain market makers who used to make money on inefficiencies of the market. Is this a good thing or a bad thing? Well, prices are more efficient now and it appears that these computers are more responsive to price-relevant information than people used to be. So even if trading volume does decrease, I see no reason to worry that prices will become less efficient. That's not the direction things have gone, even as passive investing has boomed.

Overall, worries about passive investing rely on an assumption that there is not enough interest in and resources for making arbitrage profits to keep prices efficient. This is highly counterfactual and always will be. As long as people and institutions want money and have access to the markets, there will be plenty of resources allocated to price correction.

• I'm a bit in doubt on the idea that high-speed algorithmic trading is really making the market more efficient. New information about companies is not flowing at sub-second rates. Most of these algorithms are preying on price lags between exchanges or the taking advantage of patterns in other competing algorithms etc. Holding a stock for a few milliseconds adds nothing to market efficiency. Most of these efficiency arguments are made by people who don't want anyone really looking into what they are doing or why they pay traders to use their exchanges. – JimmyJames Oct 2 '17 at 19:48
• (1) Many algorithmic traders are not high frequency traders. A large proportion of hedge funds are, at the end of the day, computer programs that aggregate market and economic information and trade on it. (2) The high frequency traders you have in mind know nothing of the economy but essentially replace human market makers who also know nothing. Bid/Ask spreads have fallen, ergo they are more efficient. – farnsy Oct 2 '17 at 19:58
• "Preying on price lags between exchanges"... in other words, coordinating prices across exchanges? :) – hobbs Oct 3 '17 at 2:06
• @JimmyJames How about when you want to buy and the price in your market is too high relative to other markets? HFT will submit sell orders in your market (and buy orders in another), driving the price down and your fund will get an unexpectedly good price. There are two sides of each trade and there is no guarantee that wrong prices will benefit you. HFT benefit by shrinking the bid and ask and disseminating correct prices--both good things. Mutual fund managers are not the ones complaining, it is the human market makers who used to benefit from those inefficiencies who are upset. – farnsy Oct 3 '17 at 19:47
• Not terribly interested in anecdotes nor mass-market books. There is a healthy peer-reviewed academic literature studying HFTs and a strong consensus among experts that the overall effect has been positive. I'm sure bad things happen in the space, just as they did in markets run by real people, but that's not the same thing as the overall effect of all algo traders. Spreads and slippages are at historical lows now. My answer is not about how great every HFT firm is; it's about how market quality is plenty good and not threatened by passive investors. – farnsy Oct 3 '17 at 23:27

The argument you are making here is similar to the problem I have with the stronger forms of the efficient market hypothesis. That is if the market already has incorporated all of the information about the correct prices, then there's no reason to question any prices and then the prices never change. However, the mechanism through which the market incorporates this information is via the actors buying an selling based on what they see as the market being incorrect.

The most basic concept of this problem (I think) starts with the idea that every investor is passive and they simply buy the market as one basket. So every paycheck, the index fund buys some more stock in the market in a completely static way. This means the demand for each stock is the same. No one is paying attention to the actual companies' performance so a poor performer's stock price never moves. The same for the high performer. The only thing moving prices is demand but that's always up at a more or less constant rate.

This is a topic that has a lot of discussion lately in financial circles. Here are two articles about this topic but I'm not convinced the author is completely serious hence the "worst-case scenario" title.

These are interesting reads but again, take this with a grain of salt. You should follow the links in the articles because they give a more nuanced understanding of each potential issue. One thing that's important is that the reality is nothing like what I outline above.

One of the links in these articles that is interesting is the one that talks about how we now have more indexes than stocks on the US markets. The writer points to this as a problem in the first article, but think for a moment why that is. There are many different types of strategies that active managers follow in how they determine what goes in a fund based on different stock metrics. If a stocks P/E ratio drops below a critical level, for example, a number of indexes are going to sell it. Some might buy it. It's up to the investors (you and me) to pick which of these strategies we believe in.

Another thing to consider is that active managers are losing their clients to the passive funds. They have a vested interest in attacking passive management.

As more actively managed funds are driven out of the market, the pricing of individual stocks should become less rational. I.e. more stocks will become underpriced relative to their peers. As stock prices become less rational, the reward for active investing will increase, since it will become easier to "pick a winner". Eventually, the market will reach a new equilibrium where only active investors who are good enough to turn a profit will remain. Even then, passive investment will still do roughly as well as "the market" since it has low overhead and minimal investment lag. There is no reason to expect the system to collapse, since it is characterized primarily by negative feedback loops rather than positive feedback.

The last few decades have seen a shift from active to passive investment because increased market transparency and efficiency have reduced the labor required to keep pricing rational. Basically, as people have gotten better at predicting stock performance, less active investment has been required to keep prices rational.

• Boom and bust cycles do not necessarily imply poor pricing. Certainly they don't implicate passive investing. In fact, artificially low volatility and markets that do not respond to real news would be a better signal that pricing is poor. – farnsy Oct 2 '17 at 19:00
• I'd disagree that a wide scale market bust does not imply poor pricing, but I'm not "blaming" index investing. I don't think "people have gotten better at predicting stock performance" (your answer does a great job of addressing this point) or that "less active investment has been required to keep rices rational" because I don't think prices have stayed rational consistently over the last few decades. – quid Oct 2 '17 at 19:12
• @quid stock prices relative to each other are kept rational by active investment. stock prices relative to the us dollar are outside the scope of my answer. – james turner Oct 2 '17 at 19:21
• I guess we do agree about the facts, but not about what James Turner is saying. I think he's just playing fast and loose with "rational." Remember, even participating in a bubble that you know is a bubble can be, strictly speaking, rational. – farnsy Oct 2 '17 at 20:04
• @farnsy i'm using the term in a technical sense en.wikipedia.org/wiki/Rational_pricing it's important to remember that even if nearly all market participants are bad at predicting future returns (as reflected by volatility), prices can still be rational if they closely reflect the expectations of the most prescient market participants because they do not expose arbitrage opportunities. – james turner Oct 2 '17 at 21:17

private investors that don't have the time or expertise for active investment.

This may be known as every private investor. An index fund ensures average returns. The bulk of active trading is done by private institutions with bucketloads of experts studying the markets and AI scraping every bit of data it can get (from the news, stock market, the weather reports, etc...). Because of that, to get above average returns an average percent of the time, singular private investors have to drastically beat the average large team of individuals/software.

Now that index ETF are becoming so fashionable, could there be a tipping point at which the market signals that active investors send become so diluted that this "index ETF parasitism" collapses? How would this look like and would it affect only those who invest in index ETF or would it affect the stock market more generally?

To make this question perhaps more on-topic: Is the fact (or presumption) that index ETF rely indirectly on active investment decisions by other market participants, as explained above, a known source of concern for personal investment?

This is a well-covered topic. Some people think this will be an issue. Others point out that it is a hard issue to bootstrap. I gravitate to this view. A small active market can support a large number of passive investors. If the number of active investors ever got too low, the gains & likelihood of gains that could be made from being an active investor would rise and generate more active investors.

Private investing makes sense in a few cases. One example is ethics. Some people may not want to be invested, even indirectly, in certain companies.

• "to get above average returns an average percent of the time, singular private investors have to drastically beat the average large team of individuals/software." Yep, that's exactly right and that's exactly what happens. The average large team of individuals/software comes with drastic expenses, expenses which are a drain on returns, so goes the passive-investment conventional wisdom. – Beanluc Oct 2 '17 at 19:43
• 'An index fund guarantees average returns.' Recommend you change guarantee to ensure or provide. I say this because guarantee has a special meaning in the investment world. Since indices are the benchmarks for returns, by definition index funds provide average returns. Also I think where you said scrapping you meant scraping. – Xalorous Oct 3 '17 at 20:38

Private investors as mutual funds are a minority of the market. Institutional investors make up a substantial portion of the long term holdings. These include pension funds, insurance companies, and even corporations managing their money, as well as individuals rich enough to actively manage their own investments.

From Business Insider, with some aggregation:

24% Mutual funds and ETFs
16% Pension and government retirement
10% Insurance and hedge funds
27% Household
22% International and other


Numbers don't add to 100% because of rounding. Also, I pulled insurance out of household because it's not household managed. Another source is the Tax Policy Center, which shows that about 50% of corporate stock is owned by individuals (25%) and individually managed retirement accounts (25%).

Another issue is that household can be a bit confusing. While some of these may be people choosing stocks and investing their money, this also includes Employee Stock Ownership Plans (ESOP) and company founders. For example, Jeff Bezos owns about 17% of Amazon.com according to Wikipedia. That would show up under household even though that is not an investment account. Jeff Bezos is not going to sell his company and buy equity in an index fund.

Anyway, the most generous description puts individuals as controlling about half of all stocks. Even if they switched all of that to index funds, the other half of stocks are still owned by others. In particular, about 26% is owned by institutional investors that actively manage their portfolios.

In addition, day traders buy and sell stocks on a daily basis, not appearing in these numbers. Both active institutional investors and day traders would hop on misvalued stocks, either shorting the overvalued or buying the undervalued.

It doesn't take that much of the market to control prices, so long as it is the active trading market. The passive market doesn't make frequent trades. They usually only need to buy or sell as money is invested or withdrawn. So while they dominate the ownership stake numbers, they are much lower on the trading volume numbers.

TL;DR: there is more than enough active investment by organizations or individuals who would not switch to index funds to offset those that do. Unless that changes, this is not a big issue.

• Could you name 'Household' as 'Individually Held' instead? – Xalorous Oct 3 '17 at 20:41
• Household is what the source calls it. I'm reluctant to break that connection. Particularly as I already adjusted to remove insurance from household. – Brythan Oct 3 '17 at 20:45
• Oh, well, they wanted to do their own thing I guess. I'd always heard such holdings called Individual holdings. But they're the ones who make millions a year talking about business, not me. – Xalorous Oct 3 '17 at 20:47

All humans are inherently active investors. You make an active decision when:

• Picking an ETF based on its objectives (note - no ETF is truly passive: All remove insolvent companies.)
• Deciding to invest passively instead of actively
• Deciding to invest at all

All the people currently sitting on index ETFs aren't inherently passive, they have just decided that at this time, they are happy with the value they're getting from them. They refrain from active investing because they don't feel like the effort needed is justified by how much they would get.

So it is unrealistic to think of people suddenly becoming passive. When the volume of active trading decreases, competition also decreases, so it becomes easier to make money. This will attract more people to active investment, and existing actives will become wealthier and create more active volume.

Also, it is not true that ETFs are parasitic, because they do not take any value away from the market. They simply allocate capital indiscriminately, so that the only profit/loss is from whatever wealth is created/destroyed by the sum of our economic activity. So far, we live in prosperous times where humanity creates wealth, and indexing is profitable. If the economy ceased to productive, that would surely change. Even the people who are absolutely clueless about markets would at least sell their ETFs and do something else with the money (which would probably still benefit some company).

Therefore the real risk is not the extinction of active investors, but the destabilization of the economy. This is on several levels, such as your country's economy, your bloc's economy and the planet's economy, depending on what exactly your indexing strategy is. But the point is, when the economy stops producing more than it destroys, indexing is dead. But this has little to do with insufficient active investment.

There is another arguably less real risk, which is market panic and a crash. I call this less real because while traditionally it is a risk, it does not persist and indexes often quickly recover. This likewise does not arise from a dearth of active trading, but panic from market participants. This is another case where you can see ETFers are not passive at all: When their index goes red, they all want to sell, and many do.

So to answer your topic, there are obvious points where index funds become unprofitable:

• When the index sucks (eg. poor objective)
• When the fund management sucks (eg. high fees)
• When the economy goes in the toilet (eg. war)
• When people panic (eg. bubble burst)

But I don't think there will ever be a tipping point that causes a collapse. A tipping point would require that either active or passive investment was self promoting: The more people do it, the more people are induced to do it. But it's the opposite: The more people do it, the less attractive it becomes vs. the other choice, because competition lowers your income potential.

As for ETFs being better than active funds, I regard that as academic at best, propaganda at worst. These aren't really things that can be meaningfully compared, there are all sorts of active funds and some do much better than the index. The only thing that is safe to say is that active investment is more work.

But is there any consequence at all to capital inflow to ETFs? I think the most pertinent one is to consider that a lot of people invest in ETFs are very low information. Their investment decision rests on "it went up, therefore I buy". When it stops going up, these people are liable to panic and sell, because their whole thesis has been refuted. The most obvious consequence of this is that the magnitude of both bubbles and crashes will be amplified.

Another minor effect is that because indexes favor large companies, they promote an environment where the rich get richer, but this is effectively countered by arbitrage from even a few active investors.