On the grand scene of financial innovations, the exchange-traded fund was fairly innocuous at first. It took a good 15 years of slow realization for people to figure out how disruptive they would ultimately be. “Exchange-traded fund” isn’t even a very good name, since closed-end funds were also exchange-traded.

And the ETF wrapper still isn’t the ideal investment vehicle, at least for the purposes of traditional active management. For active strategies, the open-end fund structure is still superior.

I’d argue that the ETF revolution is less about ETFs and more about indexing; about how people have come to view stocks less as stocks and more as blobs of stocks.

After spending a career in indexes in some capacity, I am no longer much of a stock-picker. All my ideas are top-down. I like oil. I like France. Instead of finding an oil stock or a French stock I buy all oil stocks or all French stocks. Seems easier—what if I bought the wrong oil stock or the wrong French stock? I’m happy to take the average.

Finance is funny. There are a lot of things in the capital markets that make sense when a few people do it, but not when everybody does it. The thing about buying all oil companies is that you are buying the bad ones as well as the good ones—and driving all of their valuations higher.

Back in 2004 at Lehman, we observed that correlations among stocks in the same sector were increasing, at least on an intraday basis. Think about it: if oil goes up, companies like Chevron, ExxonMobil, and ConocoPhillips should all go up—on a micro, minute-by-minute basis. So, you could still make money in the long term by picking the best integrated oil major, but day trading them against each other became useless.

Fast forward to 2017, and the stock market is a sea of baskets trading against baskets. A lot of people have learned their lesson—the only thing that works is buy and hold. Which I think is a good thing!

But a lot of people still find it hard to stick to buy and hold.

Anyway, there are now more indices than stocks. Some people say this is a bit ridiculous. Well, back in the mutual fund boom, there were more mutual funds than stocks, and everything turned out fine. Is indexing causing distortions? Is it wrecking the markets?

Maybe a little. But even then, the argument is very nuanced.

Commodity Indexing

A few months ago, I wrote a rather controversial piece at Bloomberg View talking about the last time people got really excited about indexing.

It was the mid-2000s. People stopped looking at gold, oil, and corn idiosyncratically, and started looking at them as a blob—as part of an asset class. In some respects, commodities do respond to the same macro factors. If the dollar goes down, most commodities will go up. And so, commodity indexing was born.

Commodity indexes had been around for some time (like the GSCI, or Goldman Sachs Commodity Index), but few people allocated money to them. What changed?

The major investment banks began to offer swaps and structured products that mimicked the price of the GSCI, and other commodity indexes. Money plowed into these strategies, and the prices of commodities rose in unison. Not too dissimilar to what has been going on in stocks—with the S&P 500.

But people failed to consider that hot money can flow out as well as in. One of the less-talked about aspects of the financial crisis is that the commodity bubble unwind was especially vicious. Almost nobody talks about commodity index swaps anymore—the whole strategy has been discredited.

Will stock indexing someday be discredited?

The Philosophy of Indexing

The whole point of an ETF or index is that you can buy a bunch of stocks or bonds which are more or less alike, or share similar characteristics. You don’t have to go to the trouble of picking individual stocks or bonds, which can be time-consuming.

So let me ask you a question: do all US stocks share similar characteristics?

Well, they are all in the US (duh), and are subject to the same political and economic forces. But that’s where it ends. You can’t get more different companies than Amazon and Newmont Mining. Or Goldman Sachs and US Steel.

So why put them all in the same basket?

I’m driving at something here—My theory is that indexing is most useful on narrowly-defined groups of stocks and less useful on the really big blobs.

The worst offenders are total market indices, EAFE, and of course, the world index. Why would anyone get long the planet? I happen to be very bearish on the planet. Cybersecurity stocks, on the other hand, make more sense.

Plowing a bunch of money into “the market” simply because it is “going up” is not smart.

Once more, with feeling—putting all of your money in SPY is not going to cut it. Because someday, everyone will want all of their money out of SPY… at the same time.

Get Thought-Provoking Contrarian Insights from Jared Dillian

Meet Jared Dillian, former Wall Street trader, fearless contrarian, and maybe the most original investment analyst and writer today. His weekly newsletter, The 10th Man, will not just make you a better investor—it’s also truly addictive. Get it free in your inbox every Thursday.