Are Simple Behavioral Interventions Cost Effective?
Nudging behavior can be inexpensive and useful.
Posted Aug 23, 2017
Governments, nonprofits, and companies all try to influence people’s behavior. They do that in lots of ways. There are many education programs that aim to affect behavior. For example, the US Environmental Protection Agency routinely produces public service announcements to educate people about things they can do to protect the environment. Organizations also use incentives to affect behavior. The U.S. Government gives tax breaks to people who donate money to charities in order to spur charitable giving.
Over the last 10 years, organizations have also been drawing on behavioral science to try to use simple interventions from psychology to affect what people do. Many governments have created “nudge units” aimed at creating these simple interventions.
Now that a number of these behavioral interventions have been implemented, it is worth knowing whether they have had an effect on people’s behavior. It would also be nice to know whether these behavioral interventions were cost effective. That is, it might be better to focus resources on education or incentives.
This question was addressed in a paper in the August, 2017 issue of Psychological Science by Shlomo Benartzi, John Beshears, Katherine Milkman, Cass Sunstein, Richard Thaler, Maya Shankar, Will Tucker-Ray, William Congdon, and Steven Galing.
They compared interventions in four different domains: Retirement savings, college enrollment, energy conservation, and flu vaccines. In each domain, they compared the effectiveness of one or more simple behavioral interventions with education and incentives. They were interested both in the overall effect of these interventions as well as their cost effectiveness.
In many cases, the behavioral interventions were highly cost effective compared to other approaches. For example, a company that required employees to state the amount of the contribution they wanted to make toward their retirement in the first month of employment saw an increase of at least $200 in savings per employee with an intervention that cost only about $2 per employee. That is an increase of $100 in savings for each dollar spent. The next most cost effective intervention was an educational intervention at a university that cost about $4 per employee and increased savings by about $59 per employee (about a $15 increase in savings per dollar spent).
Of course, if the goal of an organization was just to increase savings regardless of cost, there are other strategies. The Danish government provided tax incentives leading to an average increase in retirement savings of $540, but this intervention cost the government about $200 per person, and so the program increased savings less than $3 per dollar saved.
Similar results were obtained in each domain. For example, a simple message sent to consumers telling them how their energy consumption compared to that of their neighbors cost about $1 per household to provide and led to a 2 percent reduction in energy consumption (saving about 27 kWh of electricity per dollar spent). Financial incentives led to larger overall reductions. A study in California gave people a 20 percent reduction on their energy bill if they reduced their electricity consumption 20 percent compared to the previous year. The program led to a 4.2 percent decrease in consumption overall, but was also more expensive than the report, and so only about 3 kWh of electricity were saved for each dollar spent.
It is important to point out that not all behavioral interventions work that well. Giving people information about the health and environmental impact of their energy consumption was inexpensive, but had a tiny impact on energy consumption.
Finally, while behavioral interventions are often great, they can also have unintended consequences. For example, in the domain of flu shots, automatically signing up faculty and staff at a university for flu shots did increase the number of people who got flu shots, but it also led to a large number of people calling the health center to change or cancel appointments that decreased its cost effectiveness. In this domain, a combination of education and incentives at a health care facility led to a large increase in the number of people getting flu shots and was more cost effective than the behavioral intervention.
These analyses suggest that simple behavioral interventions should remain a part of the way governments and organizations try to affect people’s behavior. They often involve simple interventions that give a lot of bang for the buck. That said, it is also worth doing some testing of the effectiveness of these interventions. At times they can lead to unintended consequences. At other times, more traditional methods (like incentives and education) are actually better.
Benartzi, S., Beshears, J., Milkman, K.L., Sunstein, C.R., Thaler, R.H., Shankar, M., Tucker-Ray, W., Congdon, W.J., & Galing, S. (2017). Should governments invest more in nudging? Psychological Science, 28(8), 1041-1055.