When push comes to shove, it looks like it's going to take more than a "nudge" for people to change their bad habits. So says the United Kingdom's House of Lords Science and Technology Sub-Committee. Today it issued a report  on behavioral change policy that finds "nudges" and similar behavior interventions are ineffective in influencing behavioral changes when used in isolation.
Everyone has been nudged before—usually without them even knowing it. Take the chocolates conveniently placed in the checkout line, for example. A nudge is any action that seeks to change people's behavior by altering the environment or context of their decisions. A more healthy nudge might be to make fruit the default side order—rather than French fries—or to make stairs more prominent and install fewer elevators.
Lawmakers have taken an interest in nudges because of their nonregulatory nature and supposed cost-effectiveness. The U.K. government is keen to address societal issues such as obesity and carbon emissions by finding ways to change behavior without using regulation. But the Science and Technology Sub-Committee's report finds that a mixture of interventions is required, including regulation and taxation.
"We hope we will persuade the U.K. government that these interventions are valuable as part of an armory to persuade people to change behavior," says committee chair Julia Neuberger.
That view was echoed by Theresa Marteau, director of the Behaviour and Health Research Unit at the University of Cambridge Institute of Public Health, who submitted evidence for the report. "The concern we raised, and was raised much more forcibly by the House of Lords report, is that government and policymakers were in danger of ignoring some of the very potent means or levers for changing behavior if they focused only on nudging," she says.
The report also finds that much of government policy in this area is not based on evidence. This may be, the peers conclude, because there is little existing evidence about behavior interventions to draw on and little is being funded. In the United Kingdom, only 2.5% of all health research funding is for prevention. And only 0.5% of that is spent on behavior interventions, according to Marteau. She adds that in the United States the figure is around 4% spent on prevention. The Food and Drug Administration's new labels  on cigarette packaging that feature a cadaver or other graphic images are the United States's own attempt to implement nonregulatory interventions.
What little research exists on behavior interventions suggests that though nudging may work, the effect is slight and more often negative. "The general impression is that nudging without using regulation is going to have at best only a small effect in the context of the very large changes that are required," Marteau says.
In a review of existing literature, Marteau did find descriptions of nudging being used to improve health behaviors—one study reports a 70% increase in the amount of fruit bought by school children at lunchtime after fruit was placed by the cash register —but these interventions have not been evaluated on a large scale or assessed for cost-effectiveness. "We found a number of interesting ideas, but very few had been evaluated in general populations and very few, I don't think any, had looked at the long term and the sustainability of change," she says.
Nudge studies in the United States have also had mixed results. In 2008, the Sacramento Municipal Utility District tried to reduce household energy consumption  by issuing bills that compared customer's energy use to their neighbors. But people weren't shamed into cutting back: usage fell by only 2%.
The report recommends that the government finds a way to make the available evidence more accessible to policymakers and appoints a chief social scientist to communicate research-based advice to policymakers.
But nudging has potential. "One has to be very careful not to throw babies out with bath waters," Merteau says. "I think what is needed is more research to find out the characteristics of nudges that are effective and the context in which they will be most effective with and without regulation."