The recent volatility in the stock market undoubtedly drove many casual investors to search popular financial news outlets for clues to the market’s next leap or plunge. But listening too closely to stock commentary can actually be detrimental to investors’ portfolios because commentators unwittingly describe the stock market using metaphorical language that can cloud an investor’s judgment, shows research by Professors Michael Morris and Daniel Ames.
Morris and Ames, along with coresearchers Oliver Sheldon of Cornell University and Maia Young of the University of California, Los Angeles, examined two common types of metaphors used in stock market commentary: agent metaphors, which describe price trajectories as volitional actions (such as “the Dow fought its way upward”) and object metaphors, which describe price changes as movements of inanimate objects (“the NASDAQ dropped off a cliff,” for example).
While these financial pundits may merely be trying to spice up their reports, the researchers found conclusive evidence that agent metaphors are more likely after days when the market has trended up rather than down. In addition, they found that exposure to agent-metaphor descriptions of trends leads investors to expect trend continuance rather than correction on the following day.
The danger in this situation, says Morris, is that the stock market does not actually follow any particular patterns. “But saying ‘The market moved randomly today’ isn’t much of a news story,” he notes. “The job of a commentator is to provide the story behind the numbers. A story requires events, characters and motives, and hence the metaphorical description.” What the commentators don’t realize, he says, is that the metaphors they draw upon for up days and down days follow a pattern, and that pattern perpetuates a bias of investors to believe that uptrends offer meaningful signals about the future but that downtrends do not.
Morris and his research partners began testing their theories on the role of metaphorical language by asking subjects to study one day’s price activity and, after listening to commentary from a TV market analyst, predict the next day’s trend. The researchers manipulated the type of language subjects were exposed to in the market commentary, varying among agent, object and nonmetaphorical language. As predicted, participants exposed to agent metaphors were most likely to expect that day’s price trend to continue the next day.
What is it about agentic words that make people hearing them susceptible to the trend-continuance bias? Humans have a natural inclination, says Morris, to turn away from randomness and attribute events to intention — a tendency that is heightened by the animation and action that agentic descriptions convey.
The same forces behind the judgments humans make about one another are at work in the judgments we make about the stock market. “There is a lot of evidence showing that although behavior is largely situational, we misread other people’s behavior as reflecting some kind of intention or disposition. For example, if I meet you at a party and you’re laughing and telling jokes, I’m going to expect the same level of extraversion the next time I see you,” Morris explains. “Our findings show the same is true when you’re trying to understand the stock market: if you think a change in the stock market reflects the desire of the market to go up or down, you’ll expect the same thing to be present tomorrow.”
To determine when and why commentators gravitate toward specific types of metaphors, the researchers combed the transcripts of a daily CNBC show during a six-month period, searching for use of agent and object metaphors. Again, they found a close correlation between metaphor use and market performance. “Our results showed that the rates of agent and object metaphors in market commentary depend on the overall direction of the daily trend,” says Morris.
The CNBC reporters used agent metaphors more often to describe a market index on days when its trend was upward, and they used object metaphors more frequently when an index trended downward. On one up day, for example, a CNBC commentator said, “The NASDAQ index jumped 122 1/3 points,” while two days later the same commentator described a poor-performing market by saying, “The S&P got caught in the downdraft.” This difference was especially marked when the stock chart showed smoothly upward or downward movement as opposed to a trend interrupted by high volatility. To check that this pattern was not a by-product of a particular historical period, the analysis was replicated during both bull and bear market periods.
This tendency to describe an ascending trend agentically has its roots in our brain’s basic instinct to distinguish animate from inanimate entities in the environment based on trajectories: we impute animacy to things that move upward, against the force of gravity. “If you’re hiking and spot something moving down the hill, it could be a falling rock. But if you see something moving up the hill, you immediately assume it is animate and you interpret the event agentically — you impute a motive or purpose,” Morris explains.
The researchers also conducted additional studies to rule out alternative explanations for the pattern. For example, to make sure that the pattern did not merely reflect that stock commentators have a career interest in the market going up, the researchers sought to replicate the experience in a laboratory setting, using undergraduate research participants playing the role of stock market commentator. The students were shown a series of charts representing daily market index performance and were asked to describe the day’s price movement. After coding the subject commentary for metaphors, the researchers found that these students’ metaphors, while not as sophisticated as those of professional market reporters, also showed a bias toward using agent metaphors on up days and object metaphors on down days. This suggests the bias arises from basic aspects of human psychology that are evoked by the task of describing trends.
So, taking this research into consideration, what can average investors do to keep metaphor-laden stock market commentary from influencing their view of the markets? Morris offers a few recommendations: Don’t rely on intuitions about trends that come from looking at charts. “Charts seem to trigger metaphorical thinking that clouds decision making. And,” he adds, “the financial advisers who use the most colorful language may not be the ones who steer you the best. While their reports may be more interesting, their metaphors may trigger intuitions that interfere with clear analysis.”
A lesson for teachers of investing is that one must either work against or work with students’ inclination to understand the market anthropomorphically. Morris notes the clever approach in Benjamin Graham’s classic The Intelligent Investor, which encourages investors to think of the stock market as a manic-depressive person whose erratic behavior changes daily. This specialized agent metaphor helps readers intuitively grasp market volatility and the opportunities it presents, says Morris.
As for stock market commentators, Morris believes they should strive to be evenhanded with their metaphors. “It’s fine for commentators to use metaphors, but they should try to balance them,” he explains. “If a commentator is always going to anthropomorphize the market, they should do it on down days as well as up days so they don’t send an unwitting signal to their audience that uptrends are more real than downtrends.”
Michael Morris is the Chavkin-Chang Professor of Leadership and Daniel Ames is the Sanford C. Bernstein & Co. Associate Professor of Leadership and Ethics in the Management Division at Columbia Business School.
Read the research
About the researcher
Professor Morris is highly regarded for his research on social judgment, the study of how people make sense of events observed in their environment...Read more.
About the researcher
Professor Ames's research focuses on social judgment and behavior. He examines how people judge themselves as well as the individuals and groups around...Read more.