In 2000, Sweden privatized their social security system. By 2006, over 90% of Swedes were investing in the government-designed portfolio. They were nudged … well. (More)

Nudge, Part II – Choices, Money, and Health

This week Morning Feature will discuss Nudge, the 2008 award-winning bestseller by Richard Thaler and Cass Sunstein. Yesterday we considered the difference between Econs, the perfectly informed and perfectly rational economic actors of theory, and actual Humans. Today we see how Humans can be nudged toward better decisions regarding money and health. Tomorrow we’ll conclude with issues of personal freedom and objections to nudging.

Note: Richard Thaler is a Professor of Behavioral Science and Economics at the University of Chicago. Cass Sunstein was a Professor of Law at the University of Chicago when the book was written; in 2009, President Obama appointed Sunstein to be the Administrator of the Office of Information and Regulatory Affairs.

In the cafeteria

Many of us remember the morning announcement at school: “Today’s lunch menu is Salisbury steak, mashed potatoes, gravy, and green beans.” Or at least that’s what they called it. We called it “mystery meat.” But we ate it, unless we brought our own lunch: a “luncheon loaf” (mystery meat from home) sandwich with mustard, or maybe the classic PB&J. Regardless, those were your choices. Eat what they cooked, or bring your own.

Nowadays, most school cafeterias offer a buffet line. That means more choices, and that’s good, right? Well, it depends on what your kids choose.

Thaler and Sunstein introduce Nudge with a hypothetical example of Carolyn, the director of food services for a city school system. She has dinner with her friend Adam, a consultant who works with supermarket chains. Adam knows that the arrangement of items in a supermarket affects which items sell most, and he says Carolyn could nudge the students’ lunch choices by how she arranged the offerings in the buffet line.

Carolyn decides to experiment with different arrangements in different schools, and track what foods the students choose. Desserts are at the front of the line in some schools, at the end of the line in some others, and in a separate line in the rest. In some schools the french fries are at eye level. In others, the carrot sticks are at eye level. Carolyn discovers that the arrangement of items in the buffet line will increase or decrease consumption of some foods by as much as 25%.

Choice architecture

Carolyn decides to put this information to use. How should she arrange the schools’ buffet lines?

  1. Arrange the lines so students will choose more nutritious meals.
  2. Arrange the lines at random.
  3. Arrange the lines so students will pick the same foods they would choose on their own.
  4. Arrange the lines to favor items from suppliers who offer the largest bribes.
  5. Arrange the lines to maximize cafeteria profits.

Option #1 is a bit intrusive, as Thaler and Sunstein admit. But as they also note, the alternatives are even worse. Carolyn will be practicing what they call choice architecture. The students will still choose their own lunches, but the arrangement of the buffet line will ‘nudge’ them toward more nutritious meals.

Importantly, there is no such thing as a “not arranged” buffet line … and that’s true of almost every choice menu in our lives. The key to Nudge is that if we understand more about how context shapes our decisions, we can both practice better choice architecture in our own lives and demand better choice architecture from the businesses we patronize and the governments we elect. We will still make our own choices, and indeed the authors’ concept of libertarian paternalism may give us more choice opportunities than we have now. But better choice architecture will ‘nudge’ we ordinary Humans toward the better choices we would make if we were idealized Econs.

The Swedish experience

Take Social Security, for example. Republicans and other libertarians say we should privatize it because “you can invest your money better than the government.” A government-run system is One Size Fits All, while a privatized system would Just Maximize Choices. That’s better for everyone, the theory goes, because you get to choose for yourself. What’s more, competition would spur innovation, drive bad ideas off the market, and result in better retirement plans for everyone. Yay!

And that might work well, if Humans were Econs. Alas, Humans tend to save too little, and most of us don’t make good investment choices. We worry too much about short-term outcomes, yet don’t update our plans when we should. (Ask any Human Resources manager how many people change options during the annual 401(K) free enrollment period.) We tend to be too risk-averse, yet invest too heavily in our employers’ stock. (Ask the folks who used to work at Enron.) And we tend to follow the herd, so we buy in just as something peaks or sell just as it’s hitting bottom.

The Swedish government understood that Swedes were Humans, and kept that in mind when they privatized their social security system back in 2000. They asked several experts to design a basic retirement plan, and they made that the default choice. When the privatized system debuted, the government told Swedes the basic plan was there, but encouraged people to check out other options and “make the choice that is best for people like you.” There were hundreds of offerings, many of them advertising on Swedish TV. And for the first year or so, many Swedes chose their own plans.

But the government’s “make the choice that is best for people like you” ads stopped after a few months, once most Swedes had already made their choices. As did most of the other plans’ ads. Not surprisingly, most young Swedes entering the workforce went with the default plan. So did many other Swedes, when they realized the government experts had set up a better plan. In fact, it wasn’t even close. The average private plan had an annual fee of 0.77% – $77 per $1000 invested – compared to the default plan’s fee of 0.17%. And the default plan grew by 21.5% from 2000 through July 2007, while the average private plan grew by only 5.1%.

Of course the Bush administration studied that success and applied the same thoughtful choice architecture when they passed the 2003 Medicare Modernization Act….

The Medicare maze

Or not. Thaler and Sunstein quote President Bush’s comments to a group of seniors in 2006:

The more choices you have, the more likely it is you’ll be able to find a program that suits your specific needs. In other words, one-size-fits-all is not a consumer-friendly program. And I believe in consumers. I believe in trusting people.

If only Humans were Econs. Under the MMA, the default option for most seniors was nonenrollment. For those who were automatically enrolled – the poorest and sickest who were on Medicaid – the default option was random assignment. Most states offered 50-60 different plans, covering different drugs at different rates with different deductibles. Seniors were told to consult their doctors and pharmacists for help, but neither doctors nor pharmacists could make heads or tails of the dizzying options.

One study found that seniors could save an average $700 per year by choosing the right plan, if only they could figure out what the right plan was. Fortunately, the government set up a website.

Or unfortunately. Thaler asked a friend Katie – an economist who researches health care coverage – to use the website to choose the best plan for Katie’s mother, based on her prescription history. Katie finally got a list of three plans that looked good, after several hours’ of searching. Thaler then tried, with the same prescription data. Several frustrated hours later, Thaler got an different list of three plans. Thaler thought a younger, more tech-savvy person would have an easier time, so he asked a graduate student to try. Several hours later, the grad student got yet another list of three plans. Finally Thaler gave the task to their student intern, a Teen Jeopardy genius …

… who struggled through the online maze and got a fourth list of plans, all using the same data. It turns out the plans update their drug prices constantly, and there’s no guarantee the best plan for your mom today will be the best plan for your mom tomorrow. In a 2007 study by Nobel laureate Daniel McFadden and two other researchers, two-thirds of seniors did not choose the plan that minimized their out-of-pocket costs … even in a simplified simulation with only four options.

The dogma of Just Maximize Choices does not work in a world inhabited by Humans. We need nudges, and tomorrow we’ll discuss why nudges give us more freedom … not less.


Happy Friday!