TY - JOUR

T1 - Can pecuniary and environmental incentives via SMS messaging make households adjust their electricity demand to a fluctuating production?

AU - Møller, Niels Framroze

AU - Andersen, Laura Mørch

AU - Hansen, Lars Gårn

AU - Jensen, Carsten Lynge

PY - 2019

Y1 - 2019

N2 - The increasing deployment of renewables introduces substantial variability into the production of electricity, requiring demand to be more movable across time. We analyze data from a large Danish field experiment (2015-2016) to investigate whether households can be prompted, via SMS messages, to move electricity consumption, and if so, whether these are motivated by pecuniary or environmental motives. The analysis has two steps: As a novel approach, we first use automatic model selection, which allows a different time-series regression for each of the 1488 households studied. From this, we obtain a cross-section of estimated SMS effects, which we then regress on the motive type. Since households can opt out there is a risk of self-selection. We therefore control for the size, income and average consumption of the household, and the age, educational- and labor market status of the SMS recipient. The results suggest that SMS messages can to some extent motivate households to move consumption. Although a stronger financial motive seems more effective, mixing financial and environmental motives seems the most effective. Finally, women and elderly people are more inclined to move consumption.

AB - The increasing deployment of renewables introduces substantial variability into the production of electricity, requiring demand to be more movable across time. We analyze data from a large Danish field experiment (2015-2016) to investigate whether households can be prompted, via SMS messages, to move electricity consumption, and if so, whether these are motivated by pecuniary or environmental motives. The analysis has two steps: As a novel approach, we first use automatic model selection, which allows a different time-series regression for each of the 1488 households studied. From this, we obtain a cross-section of estimated SMS effects, which we then regress on the motive type. Since households can opt out there is a risk of self-selection. We therefore control for the size, income and average consumption of the household, and the age, educational- and labor market status of the SMS recipient. The results suggest that SMS messages can to some extent motivate households to move consumption. Although a stronger financial motive seems more effective, mixing financial and environmental motives seems the most effective. Finally, women and elderly people are more inclined to move consumption.

U2 - 10.1016/j.eneco.2019.01.023

DO - 10.1016/j.eneco.2019.01.023

M3 - Journal article

VL - 80

SP - 1050

EP - 1058

JO - Energy Economics

JF - Energy Economics

SN - 0140-9883

ER -