If you're Jeff Bezos, you're not going to have some random dude manage your money and hope for the best. You're not gonna open up a Robinhood account and risk it all on meme stocks like GameStop. You're going to hire the type of investor who has a Ph.D. in mathematics and drives a Bugatti, a go-getter who wakes up with a turmeric latte and pores over satellite images of factories in Asia to predict the earnings of some 3D-printing company most of us have never heard of. We're talking about the best of the best in finance.

Billionaires and gigantic institutional investors turn to these financiers because they want their investments to make the most money possible, which requires making the right calls on both buying and selling stocks. New research confirms they're whizzes when picking which stocks to buy. But when it comes to the other important part of their job — picking which stocks to sell and when — even these titans of finance are no better than a drunken monkey throwing darts.

The story of this study begins several years ago, when two of its co-authors were feuding in graduate school. Lawrence Schmidt, now at the Massachusetts Institute of Technology, was trained in the old school of economics, which asserts that investors behave rationally when they buy and sell stocks, carefully sifting through information on companies and making the best trades they can. Alex Imas, now at the University of Chicago Booth School of Business, was trained in behavioral economics, which asserts that people's trading decisions can be — and often are — steered off course by the pesky flaws of our brains. "And we started this paper with the idea of settling the fight between the two of us," Schmidt says.

As luck would have it, they met a professional investor named Rick Di Mascio. Di Mascio runs a company called Inalytics, which tracks the investing activity of major financial companies' investors to assess and improve their performance. He had spent years collecting rich data on the trades and portfolios of the titans of finance. Like Imas and Schmidt, he wondered whether these elite investors make systematic mistakes when trading. And after seeing a presentation given by Imas at a conference, Di Mascio offered the economists his glittering data set. Now equipped with the data, Imas and Schmidt could finally settle their feud.

Fancy-schmancy financier vs. drunken monkey

Schmidt, Imas and Di Mascio joined forces with Klakow Akepanidtaworn, a financial economist now at the International Monetary Fund. The economists analyzed data on 783 big-time investors from January 2000 to March 2016. To give you a sense of how elite these investors are, they manage portfolios averaging almost $600 million. They typically focus on trying to maximize investment returns for one or two important clients, like a multibillionaire, a massive pension fund or a sovereign wealth fund.

The first part of these economists' study evaluated the investors' performance. To do that, the economists compared the investors' trading decisions with what they could have done instead. And the economists decided to compare those decisions with the simplest alternative investment strategy they could think of, "which is almost literally throwing a dart at a list of the names that exist in their portfolio and buying or selling that instead of the company that the investor actually chose to buy or sell," Schmidt says. In other words, it's fancy-schmancy financier vs. monkey randomly throwing darts.

[Editor's note: This is an excerpt of Planet Money's newsletter. You can sign up here.]

The economists' first big finding is that these financiers are real whizzes when it comes to buying stocks. They've got skills that could justify charging clients high fees. The average stock they chose to buy outperformed the random dart-throwing monkey by 1.2 percentage points. That might not seem like a lot, but with the power of compound interest, it really adds up over time. It makes these investors rock stars in the world of finance. They're earning those Bugattis.

But then the economists looked at these investors' performance when selling stocks. It turns out they're bad, much worse than the monkey. The stocks the investors sold ended up going up in value faster than the stocks they decided to keep. If their clients had instead hired the monkey with darts to randomly choose which stocks to sell, the clients' portfolios would have earned 0.8 percentage points more per year. Again, that is a huge amount in the world of finance. Goodbye Bugatti, hello Ford Focus.

So ... what the heck?

The economists next tried to figure out why the elite investors are good at buying stocks yet bad at selling stocks. And the basic theory they landed on is that these investors spend much more brain energy on buying stocks than selling them.

The title of their study is "Selling Fast and Buying Slow," a reference to Nobel Prize-winning psychologist Daniel Kahneman's book Thinking, Fast and Slow. The book summarizes decades of Kahneman and his late colleague Amos Tversky's research about human decision-making. This research sowed the seeds for what has blossomed into behavioral economics.

Kahneman offers a framework for thinking about how humans think. He says people have two systems in their brain. System 2 is the more deliberative, rational way we make decisions, the one we use when we can sit down and slowly ruminate about the world. System 1 is the automatic, instinctual way we make decisions. It's the part of the brain that our ancestors developed when they were getting chased by lions on the Serengeti. When you're getting chased by lions, you've got to think fast. And so we've evolved to use these heuristics, or simple rules of thumb, to navigate a complex world on the fly. These mental shortcuts work a lot of the time, but they can also lead us to systematically make decisions that aren't in our best interest.

"The title of our paper is basically saying that people use the deliberative System 2 when making buying decisions and use the more intuitive, automatic System 1 when making selling decisions," Imas says.

To test this theory, the economists looked closer at which stocks the investors tended to sell and which ones they tended to hold. And it turns out that the investors aren't horrible at selling all stocks. If a company releases an earnings statement and all of a sudden the investor has an impetus to think more deliberately about that company's stock, their decisions over whether to sell it dramatically improve. Their selling decisions are also much better when it comes to the best- and worst-performing stocks in their portfolio. It's like when a stock becomes shinier, they pay more attention to it and start acting like a savvy financier again.

It's pretty darn surprising that elite investors are falling asleep at the wheel when it comes to a big part of their job. Previous research has found that small "retail" investors, like ones buying stocks in GameStop, are steered astray by their mental flaws. But we're now talking about the crème de la crème of the finance world. There are literally millions, even billions, of dollars on the line.

Imas believes that at least part of the reason is a general behavioral economics insight that people make worse decisions when they lack feedback. When investors buy stocks, they have something to look at to see how they're doing. They can learn from past mistakes in buying stinkers and adjust accordingly. But when they sell a stock, poof it's gone. They're not looking at the alternate universe where they held onto the stock and made gobs more money. They're not learning from their past selling mistakes.

Schmidt says asset managers may be more focused on buying stocks because buying stocks is sexier than selling them. When you buy some new, obscure stock that you expect to rocket for some smart reason, it makes you look good at your job. You can take the head of some sovereign wealth fund out to dinner and explain to her why you're a genius for investing in some neat company.

But the bottom line is that these investment managers are failing to maximize returns for their clients. Without some changes to improve their selling decisions, they'd be better off strictly focusing on buying stocks and letting "a robot manage their selling decisions," Schmidt says.

And so Schmidt, who has long ascribed to the traditional economic view that investors trade stocks rationally, appears to have lost the fight that originally inspired this study. "My big takeaway from this paper is even when we look at a sample of extremely talented, highly incentivized expert investors, they are still people," Schmidt says. As simple as that sounds, it contradicts a school of thought that dominated economics for decades — a school that Schmidt once wholeheartedly embraced. Now, he says, "I think I'm becoming converted. I think I'm becoming a behavioral economist."

Did you enjoy this newsletter segment? Well, it looks even better in your inbox! You can sign up here.

Copyright 2021 NPR. To see more, visit https://www.npr.org.