A Stated Preference Study Exploring the Zero Price Effect in Public Electric Vehicle Charging 

Abstract
Although there is economical and marketing evidence of consumers assigning additional benefits to free products and bundles, there is limited research into the behavioral consequences of offering public electric vehicle charging for free. Previous exploratory research by Maness and Lin analyzed possible economic and environmental benefits from offering free public charging infrastructure and policy. Their work found that providing free public charging infrastructure could increase plug-in electric vehicle (PEV) sales and thus cause decreased reliance on fossil fuels in the personal transportation sector. However, their study assumed the increased value which consumers would place on free charge events. This project proposes to establish an early estimate of the value of free charging in the United States. This is due to the lack of a control treatment and lack of variability in charging prices. To solve these problems, this project proposes to study consumers' responses to a free charging program through a stated preference approach. Under this approach, valuation behaviors were explored by generating 24 PEV charging scenarios generated using an orthogonal design. Respondents were presented with 12 such scenarios of charging location choice; three charging stations are presented with varying attributes – charging cost, charging time, detour travel time, and nearby amenities. The choice experiment data was analyzed using a discrete choice approach. Determining the zero price effect entailed adding a dummy variable to the model to indicate when an alternative’s fueling cost was zero. In the linear case, the ratio of this free price dummy variable and the charging price coefficients produces an estimate of the zero price effect. The analysis was performed across two model specifications – one in which the price parameter enters utility linearly, and another where price enters non-linearly. The model with a non-linear price specification was found to not provide a statistically better fit according to a likelihood ratio test. Results from the linear price model suggest that the zero price effect is valued at 1.50 $/charge. Additionally, the value of time (VOT) in regards to charging was valued at + 11.65 $/hr. Similarly, the value of time associated with the detour is found to be valued at + 12.20 $/hr. Although a limitation of stated preference studies is in determining the correct scale for the coefficient estimates, the value of time estimated is close to the USDOT personal travel VOT (50% of hourly median household income). Additionally, the model results suggest that the zero price effect may be about one-eighth of one’s value of travel time.