Once Upon a Time There Was a Big Bubble

One of the most beloved episodes of Star Trek: The Next Generation is called “Darmok.” On stardate 45047.2, the crew of the USS Enterprise, led by Captain Jean-Luc Picard, encounters an advanced alien species called the Tamarians with whom it is unable to communicate. Though the Tamarians’ words and syntax can be grasped, what they’re saying seems to make no sense. “But are they truly incomprehensible?” Picard wonders. “In my experience, communication is a matter of patience, imagination.” And yet a barrier remains. The frustrated Tamarian leader, Dathon, keeps repeating the same peculiar phrase: “Darmok and Jalad at Tanagra.”

The Tamarians, it turns out, speak only in narratives. They lack abstractions. They rely on stories to express what they mean. “Darmok and Jalad at Tanagra” is shorthand for the tale of an improbable friendship between mutually suspicious, potentially hostile leaders of different cultures who ended up as allies, fighting side by side against a common enemy. By invoking that narrative, Dathon meant to signal to Picard what he hoped would happen between them. What makes this episode memorable is its depiction of a culture in which communication occurs only through stories, taken as analogies or allegories. It is as if people conveyed their every thought to one another by saying, “Hitler at Munich,” “when Nixon resigned,” “the collapse of the Soviet Union,” “Hillary Clinton’s e-mails,” “the election of Barack Obama,” “the killing of George Floyd,” “your sister’s first marriage,” or “coronavirus.”

For narratives to communicate as well as they did among the Tamarians, a society needs a clear understanding of the stories that it shares; it needs stories that are universally known and a consensus about what they mean. In the absence of this sort of understanding, a story might divide people sharply; it might mean two or three things (“Donald Trump impeached”).

The “Darmok” episode resonates because humans have a lot in common with the Tamarians. Narratives provide a kind of social glue. The Bible, for instance, offers a collection of stories, usually taken to impart specific lessons; I might ask how your day was in pandemic-era America, and you might reply by mentioning the Book of Job. In the United States, our national bywords include people, policies, and place names: Abraham Lincoln, Robert E. Lee, the New Deal, Stonewall, Anita Hill, Harvey Weinstein. The meanings of these stories are constantly disputed and shift over time. That is testimony to their power.

Economists are interested in data, not stories. If you want to know whether inclusion of calorie labels on condiment wrappers or soda cans helps to decrease obesity, you should not rely on an anecdote about an overweight neighbor who paid no attention to those labels. If you are curious about whether increases in the minimum wage will decrease employment, you will not learn much if someone tells you that a tomato farmer her cousin knew declined to hire more tomato-pickers because of a rising minimum wage. To be sure, behavioral economists draw on psychology to emphasize our all-too-human departures from perfect rationality. We know, for example, that people often suffer from “present bias,” which means that they emphasize the short term over the future. We know that people tend to be overconfident about their plans, expecting that projects will take less time than they actually do. We know that people tend to be “loss-averse,” meaning that they dislike losses more than they enjoy corresponding gains. These and other findings are based on data, not on stories.

Robert Shiller is one of the founders of behavioral economics. He has spent much of his career compiling and studying large data sets on subjects including real estate prices and movements in the stock market. He has used those data sets to explain “irrational exuberance,” for example unjustified optimism about increases in stock prices. But in recent years, Shiller has placed special emphasis on what he sees as the power of stories. His particular interest is in how specific narratives affect economic phenomena—recessions, depressions, consumer choices, bubbles, the success of new products, and rapid growth. Shiller’s claim in his latest book, Narrative Economics, is that economists and others have overlooked the importance of narratives in economic life. Consider, for example, how memories of the Great Depression marked so many people who came of age in the 1940s, giving them a pervasive fear of the possibility of large-scale economic collapse. For some, that fear provided a long-term cloud over daily life; for others, it led to a lot of saving and an unwillingness to take economic risks. By emphasizing narratives, Shiller aims to mount a fundamental challenge to standard economic thinking—and to open up new territory for analysis.

Narrative Economics was published before the novel coronavirus struck, but in a sense the pandemic is an important point in his argument’s favor. It has already become a narrative—about the volatility of the stock market, about the negligence and failures of certain public officials, about China, about the devastating effect of pandemics in general. Of course, the content of the coronavirus narrative is not fixed and firm; we are in the midst of it, and people sharply disagree about its meaning. But there’s a story there all the same, and it’s affecting investors, employers, employees, and voters. Because the virus is disproportionately harming the poor and people of color, it is contributing to narratives about class and race as well—and to the sense of injustice over police violence that arose this summer after the killing of George Floyd.

For a long time to come, people will be influenced, both politically and economically, by their understanding of the pandemic, whatever narrative about it crystalizes for them. Some will stay away from the stock market out of a newfound fear of volatility and risk. Others will make expensive contingency plans, or hoard masks and cleaning products. Many will live with a kind of ambient anxiety, depression, or despair, which could make them turn to drugs or self-harm. Some will engage more in protests, or study epidemiology, or employ private tutors, or bolster their business-interruption insurance policies. Some will become EMTs; some who dreamed of being EMTs will no longer pursue that path. The same event has already given rise to myriad narratives, which compete with one another and which can lead people to live in what seem to be different social universes.

Shiller hopes to develop “a new theory of economic change,” one that emphasizes the importance of “popular stories that spread through word of mouth, the news media, and social media.” He seeks to focus on “(1) the word-of-mouth contagion of ideas in the form of stories and (2) the efforts that people make to generate new contagious stories or to make stories more contagious.” The latter point is important; influential narratives are often generated by people with explicit economic or political goals (“the failure of Obamacare” or “the hapless neoliberalism of Barack Obama” or “the safety of gold”). The narratives associated with Me Too have provided the foundation for challenges not only to sexual violence and sexual harassment, but also to gender inequality of multiple kinds. In Sexual Harassment of Working Women (1979) the legal scholar Catharine MacKinnon employed searing narratives of male abuse of power.

Narrative economics, as Shiller calls it, studies “the viral spread of popular narratives that affect economic behavior,” including decisions about whether to spend or save, where and how much to invest, whether to go to college, and whether to take one or another job. When narratives spread, not only individuals but also entire economies can move in surprising ways. Shiller wants “to improve people’s ability to anticipate and deal with major economic events,” and he thinks that if we incorporate an understanding of economic narratives, we will do exactly that.

To explain what he has in mind, Shiller draws attention to a recent narrative involving the cryptocurrency Bitcoin, whose value increased from $0 to $300 billion in just a few years. Without evaluating it or making predictions about its future, Shiller emphasizes that “Bitcoin has no value unless people think it has value.” He suggests that the occasionally spectacular spikes in Bitcoin’s price have been made possible by a mysterious narrative featuring a secretive creator named Satoshi Nakamoto, who, in Shiller’s words, “has never been seen by anyone who will testify to having seen him.” And this narrative draws attention to an imagined utopian future in which most people buy and sell all sorts of things with Bitcoin and thus become “part of a broader cosmopolitan culture,” liberated from the authority of nation-states and their provincial treasuries.

Shiller uses the example of Bitcoin to illustrate the importance of social influences—our beliefs about what other people believe—on people’s economic choices, and to show the centrality of emotionally evocative narratives to those influences. A new product—the Tesla, for example—might connect itself to an inspiring story of public-spirited technological innovation to combat climate change, and attract investors and consumers for that reason.

These influences extend to public policy. Shiller points to the Laffer Curve, developed by the economist Arthur Laffer (to whom Donald Trump awarded a Presidential Medal of Freedom), which purports to show that cutting taxes (including on the rich) will actually increase government revenues. Unfortunately, the Laffer Curve is wrong; tax cuts do not increase government revenues.

As Shiller sees it, the Laffer Curve is not a theory but a story, a pleasing tale of what might happen, and that’s what accounts for its power. It started to spread in 1978, when the Wall Street Journal reporter Jude Wanniski wrote a popular book, The Way the World Works, reporting that Laffer originally drew his famous curve on a napkin over dinner with two White House officials, Dick Cheney and Donald Rumsfeld. Laffer himself denied the report, but, Shiller writes, the vivid tale of “a drawing on a napkin helped make the story go viral,” which may well have made the theory more popular, even persuasive. Drawing on psychological research, he contends that “when authors want their audience to remember a story, they should suggest striking visual images.”

Shiller argues that such images take hold of people’s imaginations and have power for that reason. With respect to the economy, there are “perennial narratives,” which

affect economic behavior by changing the popular understanding of the economy, by altering public perceptions of economic reality, by creating new ideas about what is meaningful and important and moral, or by suggesting new scripts for individual behavior.

These include the view that real estate has some special value, that a return to the gold standard would stabilize the economy, and that either large corporations or labor unions are inherently evil. Shiller explores the nature and persistence of these narratives and also their various mutations.

Take the idea that machines are going to replace workers. Shiller goes back to classical Greece to show how potent that idea can be. Around 350 BC, Aristotle raised the possibility that people might be replaced by mechanical instruments, describing a situation in which “every instrument could accomplish its own work, obeying or anticipating the will of others.” That fear has arisen periodically in the United States, with prophecies in the 1950s about the baleful effects of an automated future. In fact, the word “automation” was coined around 1955, when electronic data processing became pervasive and sparked fear of job losses. The fear arose again in the 1980s, when the culture was obsessed with robots (think Star Wars) and when the “new automation,” including electronic word processing and computerized storage equipment, was held responsible for unemployment. Today it is easy to find newspaper or magazine stories on the topic, with intense focus on the potentially harmful effects of artificial intelligence and machine learning.

In Shiller’s view, there are many unanswered questions here; we do not know nearly enough to be able to “predict the effects of labor-saving and intelligent machines on livelihoods and work in the future.” But he emphasizes that the narratives about their adverse consequences can have effects on both politics and investment, ultimately affecting stock markets and the success or failure of companies large and small. Many people are enthusiastic about investing in companies that specialize in artificial intelligence, for example, even if there is no good reason to think that the investment will pay off.

For many years Shiller has devoted attention to booms and busts in real estate, stressing his own finding (based on surveys) that in some periods, people believe that real estate prices are bound to go up, essentially forever. “Exuberant real estate narratives,” as he calls them, can lead to a kind of boom, with sharply rising housing prices. The reason is that such narratives tend to be contagious. If people share a tale to the effect that real estate is a fool-proof investment, they can make it so, at least for a while. Belief in the inevitable growth in real estate prices creates a self-fulfilling prophecy, in the form of a bubble. But bubbles pop, and this narrative can lead to disaster, such as the Great Recession of 2007.

Stock market bubbles are best understood in similar terms, produced as they are by narratives of inevitable growth and by people’s belief that other people are optimistic (and hence likely to invest). Shiller has claimed that the lengthy bull market since 2009 in the United States is partly the result of a “story” that drove up markets during that period. Once people stop believing in that story, a sudden and steep downturn will occur. The market’s immediate collapse in response to the coronavirus pandemic can be seen to have vindicated Shiller’s argument, though its recovery since raises numerous puzzles.

Shiller encourages economists to learn much more about how narratives work. He wants them to assemble focus groups with people from different socioeconomic levels and to elicit conversations about economic narratives. They can learn a lot, he believes, from listening to what people have to say. Economists might work with historical databases of personal letters and diaries. They might interview people, asking them to tell stories in response to questions about economic decisions.

Intriguingly, he suggests that economists should investigate databases of religious sermons, because “they touch on moral values” and because “value judgments about what is right and wrong are undoubtedly relevant to changing economic decisions.” If, for example, priests loudly denounced the immorality and harmful effects of speculation, stock market investments, gambling, and greed, economic historians might see a lot of caution in the economic sphere. Above all, Shiller is interested “in getting closer to the human reality behind major economic events,” urging that the patterns of our thinking help explain why economies grow or stagnate, and “experience periods of rapid progress and periods of regression.” To be sure, many historians do something like this already; Shiller emphasizes the importance of orienting economists toward these narratives.

Here Shiller is onto something important and often neglected. Many of our most significant decisions are made not by careful cost-benefit analysis but by stories that come into our minds—including keen recollections of personal decisions that went well (moving to a new city, for example) or badly (dating a charismatic narcissist). Something like this is true of public officials as well as ordinary people. The stories of Vietnam, Watergate, and the Iraq War have loomed large in the minds of high-level politicians.

In business, prominent narratives can have a large influence—for instance, a recent decision by an assortment of companies to act in accordance with moral values might be a response to a narrative about such a decision, by one company, that also turned out to be highly profitable. Narratives of success or failure often make all the difference, including to those who influence large-scale movements in the economy. When people decide whether to buy or rent a home, to enroll in a savings plan, to stay at home during a pandemic, or to invest in equities or bonds, they often rely on salient stories in their own lives, or those of others. Shiller is convincing on these points, and in this he makes a major theoretical contribution.

Yet Shiller conflates the power of social influences in general and the power of stories in particular. All of his examples point to social influences on people’s beliefs, but some of them do not involve stories at all, at least not in any obvious sense. At times, he seems to stretch the very idea of “narrative.”

To see how social influences work, it is useful to start with the idea of “informational cascades.” When people don’t know about a given subject, they are often highly attentive to the information provided by the statements or actions of others. Suppose that Rick is not sure whether it makes sense to invest in a new real estate property. He may become enthusiastic about investing in an airy three-bedroom apartment in his neighborhood if his trusted friend Andrea says that real estate is a terrific investment. If both Andrea and Rick believe that real estate is a terrific investment, their friend Cynthia may end up thinking so too, at least if she lacks independent information to the contrary. If Andrea, Rick, and Cynthia believe that real estate is a terrific investment, their friend David will need a good deal of confidence to reject their shared conclusion, and he may invest in real estate as well.

This is a highly simplified example, of course, but variations can be found every day, online and off, as people react to the signals given by the beliefs and actions of others. This conforming or assimilating process can produce cascade effects, as large groups of people end up thinking and doing something different or new, often shockingly quickly, simply because other people are seen to think and do it too. To take a contemporary example, all over the world the social response to the coronavirus has been a product not only of law but also of an informational cascade, often driven by two words: Stay home. And yet in many places, there has been no such cascade. When social groups are divided, it is often because different social cascades develop, leading people to radically different judgments about what they ought to do.

In many cases, cascade effects will occur, or not, depending on seemingly small factors, such as who speaks first in a meeting, the initial distribution of beliefs, the level of trust, and people’s thresholds for abandoning their original beliefs in deference to the views announced by others. In hindsight, people are often tempted to think that a large-scale shift—the civil rights movement, the Arab Spring, Brexit, the election of Donald Trump, the Me Too movement—is inevitable or the product of deep cultural forces. But it might well be the result of something more serendipitous, such as a prominent news story at just the right time, or agreement between a movement leader and someone important who was formerly undecided.

Many of Shiller’s examples do not involve stories as such. For instance, he refers to a “narrative” that points out “the folly of get-rich-quick schemes.” Is that really a story? It sounds more like a cautionary note. Shiller also emphasizes a memorable quip by Ronald Reagan: “Government’s view of the economy could be summed up in a few short phrases: If it moves, tax it. If it keeps moving, regulate it. And if it stops moving, subsidize it.” In its time, that quip did catch on widely, but it does not sound much like a narrative either.

Or consider Shiller’s “perennial narratives.” Suppose that you think real estate prices will keep going up. If so, you need not have any particular story in mind; you might instead be making a prediction about what will happen. Such exuberance is often driven not by stories but by propositions about the future. The same is true about pessimism. The claim that machine learning and artificial intelligence will produce unemployment is not a story at all; it is a prediction about a likely causal chain. To be sure, it might go viral. (Actually, it has.) But taken by itself, how much does it have to do with narrative?

Nonetheless, Shiller is right to suggest that narratives can be uniquely memorable and influential, because they focus people’s attention and move their emotions in ways that abstractions usually do not. Humans are not Tamarians, but with respect to human behavior, a brief but riveting tale can be worth a thousand pages of careful analysis. We know this from daily life, and there is suggestive research to that effect as well. If doctors want to encourage a patient to have a risky operation, they might do best not to refer to data but to tell a story about a similar patient who chose that operation and is doing very well today. If you want to encourage a friend to get her son vaccinated, it might be best to tell a story about a child who wasn’t vaccinated and who got really sick with the flu, instead of quoting statistics. If politicians want to encourage people to welcome immigrants, it’s smart to describe the heroic feats and extraordinary inventions of particular immigrants who made spectacular contributions to society, rather than to give a summary of the latest academic research.

At the same time, stories can be manipulative, which means that using them to influence people’s economic, political, or medical decisions can raise ethical questions. A good storyteller can be a trickster. Suppose, for example, that a company tries to sell a new weight-loss product, regaling potential purchasers with a vivid (and true) story about a formerly obese teenager who used its product, lost fifty pounds, and is now able to run marathons in under three hours. This might make for a persuasive advertisement, but a single story, appealing to people’s hopes and emotions, doesn’t tell us a whole lot about the actual effectiveness of the product, or how likely it is to benefit most people.

Societies need data, conveyed in a clear and comprehensible way. But if our goal is to understand large-scale movements in the economy and otherwise, we could make a great deal of progress by exploring the effects of stories on people’s choices, and also the extent to which sharp differences in the perceived meaning of well-known stories end up creating parallel universes, by splitting people along political, economic, religious, and cultural lines.