Behavioral Insights

The behavioral insights approach is perhaps best understood as a lens through which we see policies, programs, and services—and that enables us to introduce new options, enhance existing ones, and reassess current activity.
The new options offered by behavioral insights have tended to attract the most attention. As the case of food consumption shows, the evidence may have surprising implications that cut against our assumptions and open up new ideas. As discussed later, many of these ideas may deal with the way that choices are structured—what has been called “choice architecture.” For example, many people display a “compromise effect,” whereby they use a mental shortcut of “go for the middle option.” Awareness of this compromise effect may reveal new approaches. Studies have shown that soft-drink consumption can be reduced by removing the largest cup size on offer and adding a smaller one at the bottom of the scale, since people often choose the middle option, regardless of its size. And thinking more carefully about the order in which options are presented offers another new opportunity. For example, sales of a sugared soft drink declined when it was moved from first to third in the list of options in electronic touchscreen kiosks in 622 McDonald’s restaurants.
However, the behavioral insights approach does not just offer new tools. Using it as a lens can also, for example, reveal how existing actions may be creating unintended and undesirable behaviors; it could highlight the biases that affect policymakers themselves; it could show the flawed assumptions about behavior in a proposed rule and show how it can be changed; or it could show that the best solution may not be attempting to change behavior at all, but rather redesigning services around what is already occurring.
This point is worth stressing because it counters some misperceptions about the true scope and value of behavioral insights. One is that the behavioral insights approach is just an alternative to more traditional instruments like information, taxation, or legislation—and has nothing to say about these options. As just shown, behavioral insights can provide recommendations about how these approaches should be put into practice that can make the difference between success and failure. Another misperception is that the approach just deals with tweaks or incremental changes—or that it is focused solely on individuals’ decisions. As shown earlier, this approach can be used to completely rethink and redesign systems or policies like sugared drink taxation. A final misperception is to see the behavioral insights approach just as an optional extra “tool” that policymakers can use (or not) if they feel like it. But since most government policies are concerned with influencing behavior (from murder laws to sex education), behavioral insights will have something to say about most policies.