By Jon M Jachimowicz, Columbia University
More than two dozen governments, including the U.S., now have a team of behavioral scientists tasked with trying to improve bureaucratic efficiency to “nudge” their citizens toward what they deem to be higher levels of well-being.
A few recent examples include a push by the socialist French government to increase the numbers of organ donors, a conservative UK government plan to prevent (costly) missed doctor appointments, and efforts by the Obama White House to boost voter turnout on Election Day.
While the government’s use of our psychological quirks to affect behavior rubs some people the wrong way, most of us can agree that the above examples achieve positive ends. More organ donors mean more lives saved, fewer missed doctor appointments mean the government or health industry is more efficient, and increased voting means stronger citizen engagement in democracy.
But “nudges” themselves are value neutral. That is, they can be used to both achieve altruistic ends or more malicious ones. Just as behavioral science can be used to increase voter turnout, it can also be used to suppress the votes of specific individuals likely to favor the opposing side, as reportedly happened in the recent U.S. presidential election.
The nudge, in other words, has a dark side.
My research explores how behavioral science can help people follow through on their intentions where they make better or longer-term choices that increase their well-being. Because choices are influenced by the environment in which they are made, changing the environment can change decision outcomes.
This can be positive to the extent that those designing interventions have good intentions. But what happens when someone uses these insights to systematically influence others’ behavior to favor his or her own interests – even at the expense of everyone else’s?
That’s my concern with President Donald Trump, whose campaign appears to have exploited behavioral science to suppress the vote of Hillary Clinton supporters.
What’s in a nudge?
Behavioral science is a relatively young field, and governments have only recently begun using its insights to inform public policy.
The UK was the first in 2010 when it created its Behavioral Insights Team. In subsequent years, dozens of governments around the world followed, including Canada with its Behavioral Insights Unit and the U.S., which in 2015 officially launched the White House Social and Behavioral Sciences Team.
The teams’ missions are all relatively similar: to leverage insights from behavioral science to make public services more cost-effective and easier to use, to help people make better choices for themselves, and to improve well-being.
In the UK, for example, the Behavioral Insights Team was able to persuade about 100,000 more people a year to donate their organs by tweaking a message people received when renewing their car tax. Here in the U.S., the Social and Behavioral Science Team helped the Department of Defense increase the amount of retirement savings accounts for service members by 8.3 percent.
However, as I have pointed out elsewhere, our environment (and the government) is always exerting some influence on our behavior, so we’re always being nudged. The question is therefore not whether we will be nudged, but how and in what direction.
For example, when you sit down to dinner, the size of your plate can make a big difference in how much you eat. Studies show you’re more likely to consume less food if you use a smaller plate. So if the government is handing out the dinnerware, and if most us wanted to avoid overeating, why not set the default plate to a small one?
But now let’s consider the dark side: a restaurant might hand out a small plate if it means it can charge more for less food and thus make more money. The owner likely doesn’t care about your waist size.
Any intervention based on behavioral science is therefore neither good nor bad. What matters is the intention behind it, the aim which the nudge is ultimately supposed to help achieve.
Potential for abuse
Take the case of what Cambridge Analytica – a company founded in 2013 and reportedly funded by the family of billionaire conservative donor Robert Mercer – did during the election. This team of data scientists and behavioral researchers claims to have collected thousands of data points on 220 million Americans in order to “model target audience groups and predict the behavior of like-minded people.”
Essentially, all that data can be used to deduce individual’s personality traits and then send them messages that match their personality, which are more likely to be persuasive. For example, highly neurotic Jane will be more receptive to a political message that promises safety, as opposed to financial gains, which may be more compelling to conscientious Joe.
So what’s the problem? In and of itself, this analysis can be a neutral tool. A government might want to use this approach to provide helpful information to at-risk populations, for example by providing suicide prevention hotlines to severely depressed individuals, as Facebook is currently doing. One might even argue that Cambridge Analytica, first hired by the Cruz campaign and later by Trump, was not acting unethically when it sent such personalized messaging to convince undecided voters to support the eventual Republican nominee. After all, this is what all marketing campaigns set out to do.
But there is a fine ethical line here that behavioral science can make easier to cross. In the same way that people can be influenced to engage in a behavior, they may also be discouraged from doing so. Bloomberg reported that Cambridge Analytica identified likely Clinton voters such as African-Americans and tried to dissuade them from going to the ballot box. The company denies discouraging any Americans from casting their vote.
Alexander Nix, CEO of Cambridge Analytica, talks about what his company does.
How might Trump nudge?
So far, it’s unclear whether or how the Trump administration might use behavioral science in the White House.
Trump, like most Republicans, has emphasized his desire to make government more efficient. Since behavioral science is generally a low-cost intervention strategy that provides tangible, measurable gains that should appeal to a business-minded president, Trump may very well turn to its insights to accomplish this goal. After all, the UK’s Behavioral Insights Team was kicked off under conservative leadership.
The White House Social and Behavioral Science Team’s impressive interventions have led to hundreds of millions of dollars in savings across a variety of departments and at the same time increased the well-being of millions of citizens. The future of the team is now unclear. Some members are worried that Trump will use their skills in less benevolent ways.
Trump’s apparent use of Cambridge Analytica to suppress Clinton turnout, however, is not a good sign. More broadly, the president does not seem to value ethics. Despite repeated warnings from government ethics watchdogs, he refuses to seriously deal with his innumerable conflicts of interest. Without the release of his tax returns, the true extent of his conflicts remain unknown.
And as we know from behavioral science, people frequently underestimate the influence conflicts of interests have on their own behavior.
In addition, studies show that people can easily set aside moral concerns in the pursuit of efficiency or other specific goals. People are also creative in rationalizing unethical behavior. It doesn’t seem to be a stretch to imagine that Trump, given his poor track record where ethics is concerned, could cross the fine ethical line and abuse behavioral science for self-serving ends.
A virus and a cure
Behavioral science has been heralded as part of the solution to many societal ills.
Behavioral economists Richard Thaler and Cass Sunstein, co-authors of the book “Nudge” coining the term, have been strong advocates of using the field’s tools to improve government policy – when the intentions are transparent and in the public interest.
But might the current administration use them in ways that go against our own interests? The problem is that we may not even be aware when it happens. People are often unable to tell whether they are being nudged and, even if they are, may be unable to tell how it’s influencing their behavior.
Governments around the world have found success using the burgeoning field of behavioral science to improve the efficiency of their policies and increase citizens’ well-being. While we should continue to find new ways to do this, we also need clear guidelines from Congress on when and how to use behavioral science in policy. That would help ensure the current or a future occupant of the White House doesn’t cross the line into the dark side of nudges.