Peer effects in the adoption of solar energy technologies in the United States: An urban case study

Building upon recent literature, we combine a novel spatiotemporal variable with spatial methods to investigate and quantify the influence of the built environment and jurisdictional boundaries on spatial peer-effects (SPEs) in inner-city areas. We focus on the Hartford Capital region, using detailed data at block-group and PV system levels for the years 2005-2013. This region is part of a state, Connecticut, actively engaged in supporting PV system at residential level. Adoption of PV systems varies substantially, and state policies are mediated by town-level regulations. We initially employ typology analysis to investigate the heterogeneity of the block groups with higher adoption rates. We then use panel FE and spatial estimations to determine the existence of spill-overs of SPEs beyond town boundaries. Our estimations suggest that new PV systems have a more limited spatiotemporal influence in inner-cities. We identify spatial spill-overs from neighboring block groups even between towns, suggesting that SPEs transcend municipal barriers. We do not find significant results for built-environment, although we identify several data limitations. Our results suggest that centralized, non-voluntary support policies may have larger effects if implemented beyond town-level, and that SPEs change their determination power depending on the underlying built environment.