Social Norms as Nudges

Social Norms as Nudges

The general lesson from this line of research is clear. If choice architects want to shift behavior and to do so with a nudge, they might be able to achieve this by simply informing people about what others are thinking and doing. Sometimes the thoughts and practices of others are surprising, and hence people are much affected by learning what they are. A voluminous body of research finds that informing people about the social norm can be extremely effective. Here as elsewhere, there is no substitute for actually testing hypotheses. Populations are not all alike. (Some nudge units have a good slogan: “Test, test, test.”) But consider just a few examples.

In the context of tax compliance, an experiment conducted by officials in Minnesota produced big changes in behavior. Groups of taxpayers were given four kinds of information. Some were told that their taxes went to various good works, including education, police protection, and fire protection. Others were threatened with information about the risks of punishment for noncompliance. Others were given information about how they might get help if they were confused or uncertain about how to fill out their tax forms. Still others were just told that more than 90 percent of Minnesotans already complied in full with their obligations under the tax law.

Only one of these interventions had a significant effect on tax compliance, and it was the last. Apparently, some taxpayers are more likely to violate the law because of a misperception—plausibly based on the availability of media or other accounts of cheaters—that the level of compliance is pretty low. When informed that the actual compliance level is high, they become less likely to cheat. It follows that either desirable or undesirable behavior can often be increased, at least to some extent, by drawing public attention to what others are doing. (Note to political parties: If you would like to increase turnout, please do not lament the large numbers of people who fail to vote. But do tell people that many of their neighbors vote!)

Using this strategy can save the government a lot of money, as illustrated by one of the first experiments conducted by the UK’s Behavioural Insights Team. The goal of the experiment was to see whether taxpayers who owed money could be nudged to pay off their debts more quickly. The results were analyzed by team member Michael Hallsworth in collaboration with three academic economists. The subjects (who did not know they were part of an experiment) were taxpayers, such as business owners, who had income that was not subject to the withholding tax and had not paid in full. Several different letters were tried and compared to a control letter just reminding people of how much money they owed.

The winning formulation was to say, “Nine out of ten people in the UK pay their tax on time. You are currently in the very small minority of people who have not paid us yet.” Notice this short message conveys (truthfully) both that most people pay on time and that you are in the minority of those who don’t. A follow-up experiment found that the message could be further strengthened by making it local, as in “Nine out of ten taxpayers in Manchester pay on time.” The impact of these letters was substantial, increasing the number of people paying within the first twenty-three days by as much as five percentage points. That may not sound like a large effect, but as with many such interventions, the costs are negligible. The government was already sending reminder letters, so why not nudge people at the same time?

The choice of reference group can also affect the response. Celebrities and so-called influencers might believe that they are best positioned to inspire regular people like us to change our ways. The hotel study described here found that guests responded strongly to a norm set by people in similar circumstances. A 2008 study determined how best to nudge hotel-goers to reuse their towels. You’ve probably already guessed that asking guests to save the environment by reusing their towels wasn’t as effective as a message communicating a social norm: “Join your fellow guests in helping to save the environment. Almost 75 percent of guests . . . help by using their towels more than once.”

The “guests” part turned out to be key. The researchers also tested messages highlighting reuse rates of people matching different aspects of the guest’s identity (for example, gender), as well as the behavior of people who had stayed in the same room. Although participants reported that other identities were more important to them, their behavior was ultimately influenced most by people who had stayed in the same room! The researchers called such specific in-grouping “provincial norms.” As any adolescent will tell you, peer pressure is real.

With respect to both public opinion and law, much of the world witnessed a cascade on an issue that motivated a whole chapter of the original edition of this book: same-sex marriage. When that edition was published in 2008, many nations, including ours, were badly torn on the issue. Lots of people felt strongly that same-sex marriage should be allowed and recognized; indeed, they thought this was quite obvious. In the United States, interracial marriage had once been illegal in many states, but in 1967 such laws were declared unconstitutional. Supporters of same-sex marriage believed that the same arguments should apply in this case. Yet others felt equally strongly that same-sex marriage was abhorrent. It is noteworthy that in 2008, the official position of Barack Obama, a left-of-center mixed-race presidential candidate committed to civil rights and equality for all, was that marriage is between a man and a woman. We thought that we had discovered a solution, very much in the spirit of libertarian paternalism: privatize marriage.

We argued that governments should get out of the business of defining marriage as a legal category. Instead, we suggested, they should simply regulate legal domestic partnerships just as they do business partnerships. In the United States, such partnerships are called civil unions. At the time, many governments (states and countries) did allow same-sex couples to get civil unions, but such couples justifiably felt that they were being discriminated against, since only opposite-sex couples were eligible for marriage, a term that bestowed both legal and social status. (In the United States, civil partners also did not qualify for many legal rights, including tax benefits bestowed upon spouses.) In our proposed regime, all such legal provisions would be defined by the status granted by governments, but marriages would be purely private affairs, conducted by religious groups or others according to whatever rules they wished to use. Marriage, as such, would not be an official category defined by the government. We hoped this approach would soften an intense disagreement: if the government got out of the marriage business, the whole area would become less contested.

Writing in the updated edition of Nudge, we still liked that idea, but twelve years of history had rendered it largely moot. Much to our (pleasant) surprise, over that period, many nations took a different and much simpler approach. They would just allow and recognize same-sex marriage! In the United States, which was our particular focus, this happened exceptionally quickly. In 2012, President Barack Obama said that he had changed his mind, announcing that “same-sex couples should be able to get married.” In 2015, the Supreme Court of the United States ruled that under the U.S. Constitution, same-sex couples have a right to get married. To the surprise of many people (including us), the court’s decision did not produce much of a backlash. That decision made same-sex marriage legal in every state of the union.

Around the world, there has been an uncommonly rapid movement toward the same conclusion. As of 2021, same-sex marriages were recognized in about thirty countries. The following list gives legal milestones in selected jurisdictions; the dates do not all mark nationwide recognition. These include Argentina (2010), Australia (2017), Austria (2019), Brazil (2013), Canada (2005), Denmark (2012), England and Wales (2013), Finland (2015), France (2013), Germany (2017), Iceland (2010), Ireland (2015), Mexico City, Mexico (2009), the Netherlands (2000), New Zealand (2013), Norway (2008), Portugal (2010), South Africa (2006), Spain (2005), and Sweden (2009).

Linger over those dates, if you would. Every nation on the list had long refused to recognize same-sex marriages; for generations, most of their citizens would have ridiculed or abhorred the idea. Yet in a short period, all of them came to treat same-sex couples the same as opposite-sex couples, and generally did so without a lot of controversy. How did that happen? Why did so many people (including us) fail to anticipate it?

A full answer would take a book of its own. The account offers two explanations. First, many gays and lesbians who had not disclosed their sexual orientation, and who had never sought even to ask for same-sex marriage, came out of the closet. Every time someone shared, “I am gay,” or “I am a lesbian,” or “I am bisexual,” a small nudge was put in place. Once people began revealing their sexual orientation to friends and family, the floodgates started to open, perhaps especially when their family members were politicians who were members of a party opposed to this change.

Openness about sexual orientation in the workplace also had an effect, particularly when that workplace was the United States Supreme Court. Justice Lewis Powell Jr.—who cast the decisive vote in a 1986 case that set back the gay rights movement—famously told his fellow justices that he had never met someone who was gay. In fact, one of his law clerks that term was gay. But in 2013, two years before the decision that legalized gay marriage nationwide, thirty members of the National LGBT Bar Association were admitted to practice in the Supreme Court—the first time openly gay lawyers were admitted to the Supreme Court bar as a group.

Second, the authors argue that social influences helped the change spread. With respect to same-sex marriage, cities, states, and nations witnessed both informational and reputational cascades. Increasing numbers of individuals joined a swelling chorus, amplifying the volume of the message to which they were themselves responding. A statement that was long punished by old norms (“I am in favor of same-sex marriage”) was suddenly rewarded by new ones. As we have seen, a norm or practice that is understood to be emerging, or to be increasingly supported, can operate as a powerful nudge, even if it is not yet supported by a majority. In the relevant period, the emerging norm was unambiguously in favor of same-sex marriage. That created a kind of self-fulfilling prophecy. And we must say that as much as we liked our idea, we are very pleased with this outcome.

The social influences that we have emphasized here can easily be enlisted by private and public choice architects. Both businesses and governments can use the power of social influences to promote many good (and bad) causes. In fact, they are doing that every day. With respect to the good, much more remains to be done.