The Effect of Messaging on Electric Vehicle Purchases

How to cite

Gillingham, Kenneth; Weber, Stephanie. 2017. "The Effect of Messaging on Electric Vehicle Purchases." AEA RCT Registry. AEARCTR-0002248. https://doi.org/10.1257/rct.2248-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Kenneth Gillingham* , Stephanie Weber

* Primary investigator

Registry ID
AEARCTR-0002248
DOI
https://doi.org/10.1257/rct.2248-1.0
First registered
Last updated
Status
In development
Trial dates
2017-05-31 – 2017-09-30
Countries
Private
JEL codes
Q4, R4
Registry record
http://www.socialscienceregistry.org/trials/2248

Abstract

Electric vehicles made up 0.7 percent of the U.S. vehicle stock in 2015, despite state and federal subsidies and incentives. This study analyzes the effect of email messages with different emphases on the decision to purchase a subsidized electric vehicle. Households with residential solar photovoltaic installations are randomly assigned to receive an email alerting them to a substantial ($10,000) electric vehicle rebate and emphasizing either economic or environmental incentives for electric vehicle ownership. A third, control group is not informed for the rebate until after the experiment. Households interested in the incentive must fill out a survey to access a promotional code, after which they can purchase a subsidized Nissan LEAF. We will compare the effect of the alternate messaging on survey completion and electric vehicle purchasing behavior.

Experimental design

Households participating in the residential solar investment program through the CT Green Bank are randomly assigned to one of two treatment groups or the control. All groups are informed of their eligibility for a $10,000 discount on a 2017 Nissan LEAF, but the economic treatment emails emphasize the financial incentives for EV ownership, while the environmental treatment emails focus on environmental motivations. The treatment emails are sent several weeks before the control group is informed of the incentive.

Design details

Randomization method
Randomization was performed using a uniform random number generator in Stata.
Randomization unit
household
Clusters
21,751 households
Observations
21,751 households
Arms
7,164 households control, 7,178 households financial treatment, 7,409 households environmental treatment
Primary outcomes
The main outcomes are actual purchases of the Nissan LEAF and the responses to the survey about electric vehicle preferences.

Cited by

No citation data available for this registration yet.

Keywords