The Power of Nudges, for Good and Bad
By Richard Thaler -- New York Times
Nudges, small design changes that can markedly affect individual behavior, have been catching on. These techniques rely on insights from behavioral science, and when used ethically, they can be very helpful. But we need to be sure that they aren’t being employed to sway people to make bad decisions that they will later regret.
Whenever I’m asked to autograph a copy of “Nudge,” the book I wrote with Cass Sunstein, the Harvard law professor, I sign it, “Nudge for good.” Unfortunately, that is meant as a plea, not an expectation.
Three principles should guide the use of nudges:
■ All nudging should be transparent and never misleading.
■ It should be as easy as possible to opt out of the nudge, preferably with as little as one mouse click.
■ There should be good reason to believe that the behavior being encouraged will improve the welfare of those being nudged.
As far as I know, the government teams in Britain and the United States that have focused on nudging have followed these guidelines scrupulously. But the private sector is another matter. In this domain, I see much more troubling behavior.
For example, last spring I received an email telling me that the first prominent review of a new book of mine had appeared: It was in The Times of London. Eager to read the review, I clicked on a hyperlink, only to run into a pay wall. Still, I was tempted by an offer to take out a one-month trial subscription for the price of just £1.
As both a consumer and producer of newspaper articles, I have no beef with pay walls. But before signing up, I read the fine print. As expected, I would have to provide credit card information and would be automatically enrolled as a subscriber when the trial period expired. The subscription rate would then be £26 (about $40) a month. That wasn’t a concern because I did not intend to become a paying subscriber. I just wanted to read that one article.
But the details turned me off. To cancel, I had to give 15 days’ notice, so the one-month trial offer actually was good for just two weeks. What’s more, I would have to call London, during British business hours, and not on a toll-free number. That was both annoying and worrying. As an absent-minded American professor, I figured there was a good chance I would end up subscribing for several months, and that reading the article would end up costing me at least £100.
I spoke to Chris Duncan, a spokesman for The Times of London. He said his company wanted readers to call before canceling to make sure that they appreciated the scope of the paper’s coverage, but when I pointed out the inconvenience this posed to readers outside Britain, he said that the company might rethink that aspect of the policy.
In the meantime, that deal qualifies as a nudge that violates all three of my guiding principles: The offer was misleading, not transparent; opting out was cumbersome; and the entire package did not seem to be in the best interest of a potential subscriber, as opposed to the publisher.
I ran into another type of bad nudging when I tried to book a ticket on United Airlines to travel from Chicago to New York. Before the airline would allow me to buy online, it required that I indicate whether I wanted trip insurance. This sort of required choice is justifiable when there is a genuine concern that the consumer might make a big blunder. I would endorse posing such a question to anyone opting out of all of an employer’s sponsored health insurance options, for example. We wouldn’t want someone to go without health insurance just because of a careless oversight.
United went a step further, though. It asked me to select “yes” or “no” before buying a ticket and highlighted the “yes” option as “recommended.” Required choice plus a recommendation qualifies as a strong nudge. But is it a nudge for good?
Consider the details: The trip insurance would cost $20.13 for a flight that itself would cost only $300. Keep in mind that the insurance covers just the nonrefundable portion of the ticket. (Normally the charge for changing a flight on United is $200, and the insurance would cover this.) I also learned, deep in the fine-print weeds, that to collect on the policy, unless the flight were canceled or delayed, I would need a doctor’s note saying that I was too sick to fly safely, and would have to file such documentation within 72 hours of canceling the trip.
If my flight connected up with a cruise, the insurance might make sense, because with the insurance I could receive compensation when a delayed or canceled flight made me miss my rendezvous with a cruise ship. But I wasn’t going on a cruise.
As an alternative, I examined a fully refundable ticket, which would cost $856. United still recommended trip insurance and charged more for it ($49.23) even though this ticket was refundable. Rahsaan Johnson, a United spokesman, said the company recommended “insurance to protect against unintended expenses that may result from circumstances beyond the travelers’ and the airline’s control,” but said the pricing is set by Allianz, which provides the insurance. Daniel Durazo, an Allianz spokesman, said it priced the insurance as a percentage of the ticket cost, but would not say whether that formula made sense.
On the positive side, opting out of these offers was easy, but the nudges flunked my other tests: They were neither transparent nor in the best interest of most customers.
These examples are not unusual. Many companies are nudging purely for their own profit and not in customers’ best interests. In a recent column in The New York Times, Robert Shiller called such behavior “phishing.” Mr. Shiller and George Akerlof, both Nobel-winning economists, have written a book on the subject, “Phishing for Phools.”
Some argue that phishing — or evil nudging — is more dangerous in government than in the private sector. The argument is that government is a monopoly with coercive power, while we have more choice in the private sector over which newspapers we read and which airlines we fly.
I think this distinction is overstated. In a democracy, if a government creates bad policies, it can be voted out of office. Competition in the private sector, however, can easily work to encourage phishing rather than stifle it.
One example is the mortgage industry in the early 2000s. Borrowers were encouraged to take out loans that they could not repay when real estate prices fell. Competition did not eliminate this practice, because it was hard for anyone to make money selling the advice “Don’t take that loan.”
As customers, we can help one another by resisting these come-ons. The more we turn down questionable offers like trip insurance and scrutinize “one month” trials, the less incentive companies will have to use such schemes. Conversely, if customers reward firms that act in our best interests, more such outfits will survive and flourish, and the options available to us will improve.