Most governments aren’t subtle when they want citizens to do something.
The U.S. spends close to $1 billion annually on advertising--trying to convince citizens to do everything from taking flu prevention shots to reporting unattended suitcases at the airport.
But now agencies are finding that subtle “nudges” can motivate behavior much better than ads, fines, or deadlines.
Nudges, or small changes to the context in which decisions are made, are the subject of a new analysis by Harvard Business School Professor John Beshears and colleagues, recently published in the journal Psychological Science.
The paper, Should Governments Invest More in Nudges? answers its own question with a resounding “Yes.”
“We suspected that nudges on an impact-per-cost basis would be superior to traditional approaches such as a financial incentive or an educational campaign,” says Beshears. “But we were surprised to see the extent to which it is true.”
According to behavioral scientists, nudges are dollar for dollar a hugely cost-effective way of causing people to change behavior and do the kinds of things that government wants them to do, like save for retirement—which are both for the good of society and for their own good.
For example:
On the first day of a new job, the paperwork is coming at you fast and furious, including a packet of information about how to sign up for the retirement savings plan.
What's more likely to make you fill out the sign-up form, information about the importance of saving for your golden years or a deadline that requests you fill it out within 30 days?
If you are like most of us, it’s the latter.
“In the flurry of activity in starting a new job, it’s easy to say, ‘I’m going to fill it out later,’” says Beshears. “But then there is never a convenient time.”
Providing a deadline, even one that isn’t strictly mandatory, cuts through the procrastination cycle, spurring employees to action.
Nudges are less expensive and more productive
According to their analysis, money spent on nudges can in some cases be more than 40 times more effective than the next most effective method, a dramatic result for governments dealing with scarce resources.
“If you take a particular policy objective as a given, nudges give you a lot of bang for your buck,” Beshears says.
A nudge, as defined by behavioral economists Richard Thaler and Cass Sunstein, is a way of changing the environment in which decisions are made without meaningfully changing financial incentives.
“It could be changing the way information is presented or the process by which you select from the available options in such a way that it tends to lead people toward a certain path,” Beshears says. “But it doesn’t coerce them in any way or remove any options from the menu. T
hey still have freedom of choice.”
In the past several years, governments across the world have been creating “nudge units” to identify policy domains where such prods can be effective — starting with the United Kingdom and spreading to include Germany, Australia, the Netherlands, and the U.S.
Until now, however, no one had ever systematically analyzed their effectiveness in terms of impact per dollar spent.
Beshears is part of an informal group of behavioral scientists who, a few years ago, discussed doing just that.
Spearheaded by UCLA’s Shlomo Benartzi, and including Beshears, Thaler, Sunstein, and the Wharton School’s Katherine Milkman, among others, the group settled on four areas of particular interest to nudge units in the U.S. and U.K.—retirement savings, college enrollment, public health interventions, and energy consumption.
They then identified a single metric of success in each of the four areas and reviewed every paper that was focused on that success metric and that was published in a top academic journal in the last 15 years.
Analyzing the results, they didn’t always find that nudges were the most effective ways of moving the needle — when money was no object, financial incentives were often more effective in making people take action. “It’s no surprise that if you pay someone thousands of dollars to do something, they will do it,” says Beshears.
On the other hand, nudges are often cheap to implement—sometimes consisting only of changing the way a particular form is written — making them much easier on the budget.
The most dramatic impact the researchers found was in the area of college enrollment.
In an effort to increase the number of students applying for higher education, federal and state governments have put large amounts of money into grants and tax breaks to decrease the cost of college.
But Beshears and his colleagues found the most cost-effective way to increase college enrollment was just to make applying for financial aid easier.
A 2012 study worked with tax preparer H&R Block to pre-fill the FAFSA, the federal form required for financial aid, at the same time individuals filed their taxes.
That simple intervention led to an increase of 1.53 students enrolled in college for every $1,000 spent on the program.
The next most effective method, providing monthly stipends to students, led to only a .0351 increase per $1,000 — a 43-fold difference — while tax credits were almost negligible in tipping the scale.
“When we are making the decision whether to go to college, it’s not staring us in the face that we will get a tax credit,” Beshears says. “So, it’s not entering into our calculations at all.”
The researchers found similar, if less dramatic, results in other categories.
A nudge on energy conservation that compared households to their neighbors led to almost twice as many kilowatt-hours saved per dollar spent as traditional education and incentives.
A nudge to schedule flu shots led to almost a 50 percent increase in vaccination take-up per $100 spent over an education campaign.
Nudges are less effective in motivating polluters
Beshears attributes the effectiveness of nudges to the way they target our thought process.
He takes a page from Princeton psychologist Daniel Kahneman, who identifies two types of thinking: System 1 decision-making is fast and intuitive, while System 2 decision-making is slow and deliberative.
As much as we might want to believe we are thoughtful, rational creatures, the truth is that humans spend most of their time in System 1 thinking.
“Most of what we do day in and day out is act on autopilot as a way to conserve our cognitive resources,” Beshears says. “If we used System 2 the entire day, we’d get too tired.” Nudges tend to work with System 1 thinking, he says, because they make certain choices the more natural and convenient option.
That isn’t to say that nudges are always the most effective way for governments to make their citizens do the right thing.
In cases like enforcing pollution regulations, for example, fines are likely more effective, in part because the decisions that go into building a factory are slow and deliberative, taking into account all of the incentives using System 2 thinking.
For everyday decisions made by citizens, however, governments should at least consider nudges as a way to supplement — if not replace — traditional incentives.
“When governments are thinking about how to deploy scarce resources, it’s important to do this kind of cost-impact analysis to think about the merits of different policies,” Beshears says.
“Doing so makes the case that nudges are often a particularly fruitful avenue to invest in.”
Harvard Business School Working Knowledge offers an accessible look at the latest research and ideas from the faculty of Harvard Business School.
