To get at the mechanisms that might be driving this persistence, the study looks at persistence with a sample of movers. It turns out that as soon as a household moves, Opower would stop sending their mailer but could still observe energy consumption in the house. This sample seemed pertinent because it could help determine how much of the intervention’s persistence was driven by adoption of energy- efficient technology. What else could explain a mailer’s effect staying in the home after the original household moves? Interestingly, we find that the effect of the social nudge does stay in the home after the original household moves, and, depending on how you analyze it, this can explain the majority of the persistence found in past studies.
One implication of this finding is that informational campaigns might be able to spur longer-lasting changes in behavior by focusing on settings where there is readily available tech that can be adopted. If there isn’t, though, the researcher or policymaker could still leverage our findings by also designing their own tech to spur persistence. For example, letting households know they should save for their retirement might lead to longer-lasting changes if it also gives households the option to enroll in a default savings program.
Q; Could you briefly describe any other research projects you’re currently working on?
A: Last year I partnered with the rideshare company Lyft to try and answer this provocative hypothesis in behavioral economics. The hypothesis is that workers are income targeters. That is, they work until they hit a target income on a given day. For a company like Lyft, evaluating this hypothesis is important, because it means they should be targeting their incentives to drivers who are facing a low hourly wage on a given shift. To assess this hypothesis, Lyft drivers were randomly assigned to receive $2 or $40 after completing a ride late in their shift one day. In response, we see that receiving the $40 causes drivers to end their shift at the same rate as drivers getting the $2. These data reject the income targeting hypothesis.
Q; Do you have plans to make use of Johns Hopkins resources, such as departments and agencies within the university and medical system, in your work as a Carey researcher? If so, what would they be?
A: Definitely! There are so many applications of behavioral economics to health. Reaching out to partner with researchers in the School of Medicine is high on my list of plans. Nothing super concrete yet, though!
Q: Concerning your work, what do you see as the main challenges, as well as any opportunities, to emerge (near term and/or long term) from the COVID-19 pandemic and resulting lockdown?
A: For an experimenter, these are strange times! We can’t conduct in-person research, which is how the vast majority of experiments are conducted. At the moment, I’m just keeping my head down and trying to wrap up old projects.
Q: Are you teaching any classes during the 2020-2021 academic year? If so, what are they? And how have you been preparing for the pivot from traditional classroom instruction to remote learning?
A: I’m currently slated to teach in the spring. My class is called Economics for Decision Making. As a new faculty member, I’ve never taught this class in-person, so I'm not sure I’d consider this a pivot, but one thing I’m excited to try is to illustrate the concepts to students by having them participate in online experiments.