Skip to main content

To many young people, the idea of efficient financial markets -- the idea that, in the words of economist Eugene Fama, “at any point in time, the actual price of a security will be a good estimate of its intrinsic value” -- probably seems like a joke. The financial crisis of 2008, the bursting of the housing bubble and gyrations in markets from gold to Bitcoin to Chinese stocks have put paid, at least for now, to the idea that prices are guided by the steady hand of rationality. The theory won Fama an economics Nobel prize in 2013, but he shared it with Robert Shiller, whose research poked significant holes in the idea decades ago.

But believe it or not, there was a time when efficient markets theory occupied a place of honour in the worldview of economists and financial professionals alike. This is chronicled in my colleague Justin Fox’s excellent book, The Myth of the Rational Market: A History of Risk, Reward, and Delusion on Wall Street. Though Fama did empirical research that seemed to support market efficiency, at its core the idea is based on fairly simple logic -- if people consistently buy assets for more than its fundamental value, they’ll lose money. If people lose money, they’ll be pushed out of the market, and only investors who tend to pay the right price -- often referred to as “rational arbitrageurs” in economics papers -- will remain. Economist Milton Friedman legendarily stated this idea in 1953 when he wrote:

People who argue that speculation is destabilizing seldom realize that this is largely equivalent to saying that speculators lose money, since speculation can be destabilizing in general only if speculators on the average sell when [an asset] is low in price and buy when it is high.

This logic is simple and compelling, but it doesn’t have to be true. In the real world, people who trade based on fundamentals don’t have infinite resources -- short-selling on the expectation that a stock will fall in price is expensive and risky, borrowing shares to place those bets costs money and financial backers can withdraw their money before a trade pays off. Because of these limits, trading against a mispricing isn’t true arbitrage, because, as economist and investor John Maynard Keynes is said to have quipped, “the market can stay irrational longer than you can stay solvent.”

In the wake of the 1987 stock market crash, economists pushed back strongly against the idea of efficient markets. Keynes’s remark was formalized into economic theory by J. Bradford De Long, Andrei Shleifer, Lawrence Summers and Robert Waldmann in 1989. The idea was that so-called noise traders -- a coordinated herd of investors who either overpay or underpay for a financial asset -- can act in concert to push prices out of line with fundamentals. With their limited resources, rational arbitrageurs can’t always risk pushing back against the irrational tide -- sometimes, like a movie character who throws down his shotgun and runs when faced with an onrushing gang of zombies, rational traders can end up moving in the same direction as the speculators. When this happens, the noise traders can actually make money and remain in the market, defying Friedman’s formulation.

It’s a compelling idea. But it’s very hard to test because it’s hard to observe rational arbitrageurs in action. Typically, no one knows which trades are placed by the rational folks, and which are by misinformed speculators.

A recent paper, however, does something along those lines. In ETF Arbitrage and Return Predictability, economists David Brown, Shaun Davies and Matthew Ringgenberg take advantage of the way exchange-traded funds are structured. An ETF typically has a designated set of traders called “authorized participants” (APs) who are able to carry out arbitrage between the fund and its underlying assets, whether stocks, bonds or commodities. When the price changes, APs respond by buying and selling the underlying assets, and by either creating or redeeming shares of the ETF, until the two values come back into line. They are, by design, rational arbitrageurs.

Generally, an ETF’s APs do a good job of keeping the fund’s value close to the value of the assets it owns. Many studies confirm this. But Brown et al. find that APs’ arbitrage coincides with a deviation of asset values from their fundamentals.

When traders other than the APs push around the price -- the changes in the prices of the assets tend to reverse themselves over the subsequent months. Anyone watching the APs’ arbitrage trades -- which are public record, since they involve the creation and destruction of ETF shares -- can then bet that the recent rise or fall in the price of the assets underlying the ETF will be reversed. And make a lot of money. Under efficient markets theory, that’s not supposed to happen.

Now, the APs’ job is to equalize the price of the ETF and its assets, not to make sure the asset prices themselves reflect fundamentals. But the fact that they exist implies that there are at least some rational arbitrageurs in the market. And the fact that ETF asset prices have predictable reversals implies that rational arbitrageurs aren’t trying very hard to correct prices that are out of line with long-term fundamental values.

In other words, even markets with some rational participants can behave irrationally. Speculation can move prices around for irrational reasons, and rational traders often either can’t or won’t bother to correct them. However, it’s worth noting that the effect is less pronounced in 2012-2016 than in 2007-2011, suggesting the possibility that this particular market inefficiency may have been a temporary phenomenon.

Efficient markets theory never really fits the facts, but it never quite dies, either.


Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.

Interact with The Globe