High-frequency analytics and residential water consumption: Estimating heterogeneous effects

This paper estimates how high-frequency online Home Water Use Reports (HWURs) affect household-level water consumption. The HWURs under the study share social comparisons, consumption analytics, leak alerts, and conservation information to residential accounts, primarily through digital communications. The data utilized in this paper is a daily panel dataset that tracks single-family residential households from January 2013 to September 2019. We found a 6.2 % reduction in average daily household water consumption for a typical household enrolled in the program. We estimate heterogeneous treatment effects by the day of the week, the content of push notifications, and baseline consumption quintile. For the latter, we provide an illustrative test to emphasize how mean reversion can severely bias a naïve panel data estimator for heterogeneous treatment effects when the source of heterogeneity is the outcome variable. We also find evidence that leak alerts effectively reduce water consumption immediately following the alert. • High-frequency online Home Water Use Reports (HWURs) lead to a 6.2 % reduction in water consumption for a typical household enrolled in the program. • We estimate significant heterogeneity in the treatment effects depending on the day of the week, the content of push notifications, and the baseline consumption levels. • Our analysis indicates that leak alerts are particularly effective at reducing water consumption immediately after they are issued • We illustrate how mean reversion can bias a naïve panel estimator for heterogeneous treatment effects when heterogeneity arises from the outcome variable.