Life Is Full of Big Games

From optimizing our daily commute to choosing whether or not to purchase Apple’s latest gadget, our lives are filled with instances of strategic decision making. And while most of us are vaguely aware of the larger systems at work (the city’s transportation network; the marketplace for technology products), our main concern is to navigate the chaos of modern life.

Ehud Kalai, a professor of managerial economics and decision sciences at the Kellogg School who recently delivered the prestigious 2015 Nancy L. Schwartz Memorial Lecture, has spent his career as a game theorist trying to model this type of chaos. He views commuters and smart-watch buyers as participants in “big games”—scenarios that involve the repeated interaction of many “players,” other examples of which include financial markets, healthcare management, and the education system. Kalai is especially interested in what game theorists call “rational learning”—in other words, how people learn to play. “Rational learning,” he says, “can ultimately lead to more stability.”

Of course, with any game, there will always be periods of instability. When a city builds a new highway or expands its subway network, the average daily commute is disrupted, and it takes time for people to reestablish their preferred route. Likewise, whenever a company introduces revolutionary technology—the P.C., laptop, iPhone, etc.—a period of minor chaos ensues while the public decides whether to get on board. In such cases, Kalai says, “the fundamentals of the game have changed, and players need time to adjust.”

It is never clear how or when the rules of the game will change. Take, for example, the demand for butter. “When margarine was first introduced, there was a period of high instability,” Kalai says. Nobody quite knew what to make of this new butter-like product. But when the FDA released a report outlining the health risks of traditional butter, many consumers switched to margarine. Then, the FDA determined that margarine was unhealthy, which set off another period of instability in the market (this also gave rise to “light butter”).

For the most part, we like our games to be stable. Not only does stability mean that we get to work on time, it also makes us feel that our purchases and investments are safe bets. Not that stability alone is the only desired goal: while high unemployment may be stable, it is certainly not desirable. And yet, as Kalai points out, there are cases in which stability contributes greatly to overall welfare. “It means that plans are followed through, and that outcomes are predictable. You can think about the predictability of the flu vaccine as opposed to the unpredictability of Ebola treatment.”

So what can we do to reduce the instability in our systems? Kalai—together with his Kellogg research partner, Eran Shmaya—has developed theorems to show that rational learning can help reduce chaos in big games. “The question of learning had not been clear,” he says. “But now we can say that whenever there is a change to the fundamentals of a game, people will, within a certain finite period of time, learn how to predict outcomes, and that will lead to a new period of stability. That’s what the math tells us.”

Not that such learning is easy. Whether it is traffic patterns, smart-watch sales, or the spread of a deadly virus, “the behavior that people are trying to predict is constantly changing—and there are still all kinds of misunderstandings,” Kalai says. Rational players may learn how to predict outcomes, but that does not mean they fully grasp the fundamentals of the system. “Their knowledge is not sufficient to predict what would happen in hypothetical situations,” Kalai says. Knowing that the iPhone would outsell the Galaxy is not the same as knowing how the smart-phone market works; a correct diagnosis does not mean that we understand the disease.

Still, if game theory can prove how people learn to play specific games, it might be able to explain a wider range of social phenomena—anything from online dating to financial markets. Kalai says he was watching the Money Channel one day when he heard the following sentence: “More educated traders and transparent daily information lead to more stable market prices.”

“I thought: ‘Well, that’s my theorem.’”