My new paper with Nathaniel Horner and Inês Azevedo just got accepted in Environmental Research Letters. 

In this paper, we use a database assembled by the Lawerence Berkley National Laboratory to assess the benefits and costs of the vast majority of rooftop solar photovoltaic systems currently installed in the United States. We used data from recently installed systems to project what the benefits and of a system installed today might be.

We find that a combination of subsidies and net metering policies make the systems attractive to their owners in many places. However, if we narrowly define the public benefit of these systems as the pollution (both CO2 and other pollutants) their installation avoids, public subsidies far exceed public benefit. Moreover, the analysis suggests that these subsidies have accrued disproportionately to systems installed in counties with high incomes.

The research lends supports to efforts to update policies for integrating distributed generation into the grid. These policies should seek to design and target incentives so that the costs to the public and to individuals more closely match the benefits.