People overreact to news about stocks (and other things) they own

Michael Maiello | Mar 26, 2021

The car in your garage is precious to you—dents, rust, quirks, and all. That same auto on a used-car lot would probably hold less appeal. This is the essence of the endowment effect, a foundational concept in behavioral finance that highlights the power of ownership. There’s a gap between what people will pay to buy an object and what they’ll accept to sell the same thing that they own. 

Research by Chicago Booth’s Samuel Hartzmark and Alex Imas and University of Colorado postdoctoral scholar Samuel Hirshman (a recent graduate of Booth’s PhD Program) suggests that ownership also has an important effect on how people process information. Positive information tends to inspire exuberant optimism, while negative information tends to spark dire pessimism, both out of proportion with the signals themselves. This finding has implications in the stock market, as it may explain how investors react to news and why people trade as much as they do. 

In their main experiment, the researchers gave each study participant an asset—think of it like an imaginary stock—then shared information that was an indication of its quality. Participants were shown six equally priced stocks, given three of them, and told that each had either gained or lost value. Then participants had to forecast what each stock would be worth in the future. The experiment involved 15 rounds of price changes and was designed to eliminate any biases about the stocks or the participants as traders. 

“We intentionally made this as stylized as possible,” says Hartzmark. “This is light-touch ownership: ‘Here’s a stock. It’s yours.’ We’re not tying in biases that arise from selection.” And yet participants overreacted significantly to signals about stocks they owned. After seeing positive signals, the participants thought stocks they owned were better than the others. After seeing negative signals, they thought their own stocks were worse. In contrast, participants appeared relatively rational when interpreting information about stocks they didn’t own.

The pattern also turned up in consumer data. The University of Michigan Surveys of Consumers asks investors to forecast how they think their stocks will perform, and the researchers analyzed survey data collected between 2002 and 2019. Ownership affected expectations, just as in the experiments. When it came to expectations about future market performance, stock investors responded more positively to recent positive returns and more negatively to recent negative returns than did people who didn’t own stocks.

When people own something, they pay more attention to signals about it, and their increased attention leads them to overextrapolate and overreact, the researchers suggest. While the experiment shielded participants from real-world financial, employment, and status pressures, it’s easy to imagine that the magnitude of these impulses would only grow under such strains. 

The researchers also studied how information processing affects the original endowment effect. In the experiment, they gave participants one of two power banks (portable cell-phone chargers) to own. After this, the participants saw a series of Amazon customers’ star ratings of both power banks and were asked how much they would be willing to pay for each of the chargers, and how much they’d be willing to sell theirs for. 

In line with ownership-driven overreaction, good reviews of a charger effectively doubled the endowment effect—that gap between what people pay for an object and what they’re willing to sell it for—and bad reviews wiped it out entirely.

The findings may be helpful for investors hoping to moderate the urge to sell when prices drop or invest too enthusiastically when they rise, and it may help explain why people trade more than models indicate they should, which is a long-standing puzzle in finance. 

“Ownership is so obviously part of everything that we tend to forget about it,” Hartzmark says. But as he, Hirshman, and Imas write, “Our findings suggest that ignoring the influence of ownership on the learning process can lead to erroneous conclusions.”