An Aggregate Perspective on the Geo-spatial Distribution of Residential Solar Panels

Residential solar panels in the United States (U.S.) are inefficiently distributed in terms of optimizing solar-electrical production. Controlling for local solar electricity generation potential (insolation), the residential solar share of electrical consumption is relatively higher in cloudier locales like the Pacific Northwest and Northeast than it is in sunnier areas like the Western U.S. and Florida. Rebates designed to increase residential solar adoption in places like Florida and Texas with relatively low solar-electrical shares are ineffective and may lead to net decreases in the residential solar share if housing and electrical consumption are complementary. This is because electrical consumption increases faster in response to a decline in effective residential solar prices than actual demand for panels themselves, thus driving down the solar share despite additional installations. Through the lens of a county-level structural model of demand for housing, electricity, and solar panels, we find that this phenomenon is especially prevalent in locales with high demand for cooling services (e.g., air conditioning, refrigeration, etc.) due to high numbers of cooling degree days. Inability to effectively store solar-produced electricity may be to blame. Our results thus suggest that future policies should subsidize nascent battery technologies in place of direct solar-panel installation rebates if the goal is to increase the residential solar share of electrical consumption.