Some of the difficulties of driving behavior change

Ariella Kristal and Ashley Whillans discuss why "nudging" may not be enough for shifting commuter behavior.
Some of the difficulties of driving behavior change

Several years ago, Ariella ran the five studies described in our paper with a sample of over 50,000 people in order to try to reduce the number of people driving their own cars to work. None of the interventions she tried led to economic or statistical significance, so she figured it would be difficult to publish these null findings (this was before the launch of Nature Human Behaviour). Then, Ariella met Ashley, who was working on similar commuting projects, and Ariella felt compelled to warn Ashley about the types of interventions to steer clear of (i.e. “pure nudges,”) when trying to reduce the number of people driving alone to work. Over the course of her research, Ashley also encountered other researchers and practitioners trying, and often failing, to significantly change commuting behavior without more dramatic interventions such as heavy (dis)incentives, parking bans, or flexible working arrangements. 

Given that Ashley has published null-results before, and given that we both find it incredibly important to share both “what works” and “what doesn’t” when it comes to behavior change, we decided to team up to share these findings to prevent policy-makers and organizations from investing their time, energy, and money into avenues that are unlikely to be effective. We think it is important that large-scale field experiments that fail to achieve significance (both in the statistical, but also in the economic sense) should be shared widely--to avoid wasting taxpayer or employer money on related initiatives. 

Nudges, or small changes to the decision-making environment that are non-coercive and do not significantly change economic incentives, have gained popularity with policymakers over the past ten years, as it has been successful in a wide range of domains. As we discuss in our paper, there are several important factors about commuting that make it different from other domains in which nudging has succeeded. As opposed to paying your taxes (which is something you have to do anyway), getting a flu shot (which, though unpleasant, will improve your health) or saving for retirement (for which you will reap direct benefits in the future), switching to sustainable commuting modes is not something people feel the need to do, or even feel like it will benefit them. While there are some monetary benefits, many of the advantages of having fewer cars on the road (such as less congestion, faster commuting times, and higher air quality) will only come to fruition if most people change their behavior. 

In addition to being a collective action problem, commuting is also a habitual behavior. Most successful nudges change infrequent behaviors. Saving for retirement, which sounds habitual, is most effective when a nudge defaults people into being automatically enrolled in contributing to their retirement accounts. Even in household energy use, a seemingly analogous context to sustainable travel, where nudges led to significant energy reduction, it was recently shown that up to 55% of the energy reduction in the house persisted even after the original occupants (who received a social comparison nudge) moved out (Brandon et al., 2017). This led the researchers to infer that the majority of the gains resulted from a one-off capital investment, as opposed to sustaining a habitual behavior change. In the words of Richard Thaler, nudges work well for behaviors where you can “set it and forget it.” Therefore, nudging to buy more environmentally friendly vehicles or defaulting individuals into working from home, may be more fruitful avenues for nudging commuter behavior. But if organizations or policymakers are serious about changing commuter behavior, it may be worth focusing on these more heavy-handed approaches.

Our paper in Nature Human Behaviour is available here.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in