Peer Effects in the Diffusion of Solar Photovoltaic Panels

Abstract

Social interaction (peer) effects are recognized as a potentially important factor in the diffusion of new products. In the case of environmentally friendly goods or technologies, both marketers and policy makers are interested in the presence of causal peer effects as social spillovers can be used to expedite adoption. We provide a methodology for the simple, straightforward identification of peer effects with sufficiently rich data, avoiding the biases that occur with traditional fixed effects estimation when using the past installed base of consumers in the reference group. We study the diffusion of solar photovoltaic panels in California and find that at the average number of owner-occupied homes in a zip code, an additional installation increases the probability of an adoption in the zip code by 0.78 percentage points. Our results provide valuable guidance to marketers designing strategies to increase referrals and reduce customer acquisition costs. They also provide insights into the diffusion process of environmentally friendly technologies.

Keywords: social contagion, diffusion models, empirical IO methods, probability models, word of mouth, endogeneity