Why Index Funds Are Like Subprime CDOs

CDOs are collections of mortgages with rules about how they pay out. During the last bubble they were sold (and rated by supposedly respectable third-party arbiters) as among the safer investments available. This was because the CDO had a safety feature where a certain percentage of the mortgages were expected to default, and so you didn't need all top-quality mortgages, lousy ones were okay, too. Thus the sudden availability of "NINJA" (No Income, No Job, no Assets) loans, which were completely inexplicable to basically everyone. No one would have lent their own money to poor credit risks, but the CDOs would buy that loan and stick in in their security in a heartbeat. The problem came that once the expected maximum level of default was breached, the CDO started paying out very little or nothing, and the value fell to near zero.

So his analogy is that the same thing will happen to index funds, which invest on the principle that you don't need to pick top-quality stocks, that you just buy all of them. As far as it goes, there is a significant similarity with CDOs. As an index investor, I don't try to find my own good stocks, I don't even need someone skilled in stock picking. I'm relying to an extent on financial engineering and the rest of the market to make sure that I'm not grossly overpaying for the lousy companies that are in the index. He notes that there is a huge multiple on many of the stocks, such as the 266 that have less than $150 million in daily trades, but represent trillions held by index funds. Trading is the "price discovery" mechanism of the market, and lightly traded stocks are subject to all kinds of manipulations and volatility, to be sure.

But there are several places where I think he's making a major stretch with this argument. First, the CDOs were sliced into "tranches", so when you bought a CDO you didn't get the underlying assets, just the right to a payment stream. The index fund sticks very close to the current market value of its assets; you get what you would have gotten if you just bought all 500 stocks individually, without any financial magic. For instance, on a $10K investment the Vanguard S&P fund trails its index by about $100 over 10 years. There's no daylight for shenanigans there, so I think the financial engineering argument is categorically false.

It also doesn't bother me that the stocks are thinly traded relative to the assets, because one of the big advantages of passive investing is that you're not trading all the time. Again looking at Vanguard, they turn over less than 4% of the stocks in a given year. But that's what you should do if you don't want to get killed by trading fees. Buy and hold and all that. For there to be a problem you'd have to see that those stocks were getting volatile, or that the index funds were constantly buying at a disadvantage. It is true that there is a certain amount of trying to beat the index funds to the punch; kind of hard to keep from telegraphing your investment choices when they're literally written into the name of the fund and you need to buy for a half-billion dollar fund. But these are tiny in magnitude. If anything, things are much more efficient and rapid then they were before computers took on most of the trading.

Finally you get to the thing I worry he has a point. If 100% of the money was passive, there would be huge opportunities to exploit. There's no law that says passive investing is going to be better than active. It's been true so long that maybe it's taken as gospel when it shouldn't be. Nothing is forever, and anyone who argues "it's different this time" is probably wrong, eventually. But there's still a ton of money out there in active funds, hedge funds, pensions, etc. If they saw a good opportunity, they would take it. There's too much money to be made by sharp-eyed investors to let the market as a whole get to the point where bad stocks and good stocks are treated the same.

https://investor.vanguard.com/mutual-funds/profile/portfolio...

Warren Buffett lucidly pointed out that the average performance of active investors will be....the market average. You cannot, by definition, have a majority of investors beating the market.

And once you add in fees, index funds produce above average performance, as they have low fees.

So far so good. But, the index funds are free riding on the decisions taken by active investors. Active investors do a useful service to the world by moving capital away from inefficient companies and towards efficient companies.

If the market gets worse at this capital allocation, we can expect overall lower returns.

In other words, there's average market performance but also the factor of efficiency in capital allocation.

This is related to but separate from what Burry talked about, I think. His central point seemed to be that valuations were based on very thin trading, and that when money is taken out of index funds, it will be very hard to find sufficient buyers without a large drop in prices. Add in to that complex derivatives etc used in making all the etfs work.

That bit is somewhat beyond me though. Anyone got a good analysis of how that may play out?

As more of the market moved to indec funds, a smaller amount will be controlled by active investors, which will make the entite market dumber (I would say less efficient, but that would include the cost of managing the fund). As the market gets dumber, an active investor can make more profit without needing to be any better then he currently is.

At some point (in theory), the marginal profit one can make with an active fund will equal the added cost.