Trust Usually Helps Communities Thrive. During a Pandemic, Not So Much.

“It’s very intuitive,” says Georgy Egorov, a professor of managerial economics and decision sciences at Kellogg. “If you like other people and you trust other people, then of course everybody does the right thing.”

But a recent study by Egorov and colleagues suggests that the truth is more complicated: trust may not always serve communities well in terms of promoting healthy behaviors. In fact, it could encourage people to act more recklessly. And as the delta variant fuels a new surge of COVID-19 cases, reckless behavior has become more dangerous for everyone.

Using data from Russia and the U.S., the researchers found that in more ethnically diverse areas—where people have less trust that their neighbors will follow social-distancing guidelines—residents are more likely to stay home to avoid becoming infected. And, of course, when people stay home, they don’t spread COVID-19.

The result is an unusual example of a situation where fear—not trust—leads to a positive outcome, at least in the short term. “Most of the time, lack of trust is bad,” Egorov says. But at this particular moment, from a public-health standpoint, “it does help.”

Stay In or Go Out?

The study arose when Egorov and his colleagues wondered which social factors might affect people’s willingness to isolate themselves during the coronavirus pandemic. The team included Ruben Enikolopov and Maria Petrova at the Barcelona Graduate School of Economics and Alexey Makarin at the Einaudi Institute for Economics and Finance.

People’s trust in each other often depends a great deal on how similar they are, Egorov says.

So the researchers decided to start by focusing on ethnic fractionalization, which is the likelihood that two random people in a community are from different ethnic groups.

At first glance, it seems reasonable to assume that citizens of a homogeneous society would stay home more. After all, people care about their community, so they’ll isolate themselves to avoid infecting others.

“But if you try to think through the steps of this argument, then it becomes a little bit more ambiguous,” Egorov says.

Consider the following scenario. In a homogeneous city, people trust others to do the right thing—that is, stay home when they’re sick or know that they were exposed to the virus. But seemingly healthy, unexposed people then feel emboldened to go out because they figure they’re unlikely to encounter someone who’s infected.

The thinking is that “I shouldn’t be afraid of anything,” Egorov says. “I can freely continue my life as before.” Yet, of course, if some of those “healthy” people are in fact infected and asymptomatic, they could fuel the outbreak.

In a diverse society, the opposite pattern might play out. With little trust that sick people will stay home, people who feel healthy therefore isolate themselves to avoid being infected. In other words, citizens might stay home not out of altruism but fear. And that fear helps to keep the community healthier.

When the Coronavirus Arrives

To test this hypothesis, the researchers focused first on Russia, where the outbreak started in Moscow and spread outward. The researchers examined cell-phone data that captured how much people were moving around in 302 cities from late February to late April 2020. The team also determined the level of ethnic fractionalization in each city.

Egorov and his colleagues compared changes in movement before and after the first case was reported in each city, controlling for factors such as population density, average income, and education level.

Before the first case report, the amount of movement was fairly similar in cities with high and low ethnic fractionalization. Afterward, the two groups diverged: People’s movements were more restricted in more diverse cities.

Egorov and his colleagues were somewhat concerned that the date of the first reported case might not be fully objective. For example, if individuals in more diverse cities trust doctors—rather than each other—less, they might be less inclined to get tested or even call a doctor when they feel sick. This would artificially delay the first reported case in more diverse cities. And if people are suspicious that there might be a delay in reporting, it might cause an overreaction when that first case is finally reported, because people might suspect there was already a major outbreak underway. To deal with this problem, the researchers used previous patterns of travel to and from Moscow to predict the arrival of the virus in each city.

Again, more diverse cities showed more social isolation after that date. A one-standard-deviation rise in ethnic fractionalization was linked, on average, to 3.1 percent less movement.

“We pretty much unambiguously see the direction of the effect,” Egorov says.

A similar pattern emerged when the team looked specifically at xenophobia levels in each Russian city. They calculated xenophobia levels based on the number of ethnic hate crimes and people’s online searches for ethnic slurs. As expected, cities with higher xenophobia tended to reduce movement more.

A Look at Patterns in the U.S.

Finally, the researchers investigated whether this same general phenomenon also occurred outside Russia. This time, they used aggregated data from Google on people’s mobility across the U.S. to calculate the percentage of people staying home after the first case report in each county. That figure rose by 0.46 percentage points with each one-standard-deviation rise in ethnic fractionalization.

Egorov cautions that the U.S. evidence is weaker because its cities tend to be more segregated than Russian ones—meaning that people living in a highly diverse area might still live in homogenous neighborhoods. Still, “we have reason to believe that our effect is not Russia-specific,” he says.

Appealing to Selfishness

Egorov notes that the study focused on a city’s initial response. Longer-term behaviors depend on more complex factors such as the amount of testing, contact-tracing efforts, and how the government chooses to trade off public health measures with economic recovery. Many of these efforts may indeed be aided by higher levels of trust.

But the results suggest that when an outbreak first arrives, it might be prudent for public-health officials to tailor their messages based on the community’s level of diversity, Egorov says. In a homogeneous city, public health officials could encourage social isolation by appealing to altruism: even if you feel healthy, you might be asymptomatic, so you should stay home to protect others.

In a diverse area, officials could emphasize more selfish reasons. Other people could be contagious even if they look healthy, so you should stay home to avoid becoming infected. “Altruistic messages might fall on deaf ears,” Egorov says.