Electoral Backlash against Climate Policy: A Natural Experiment on Retrospective Voting and Local Resistance to Public Policy

Retrospective voting studies typically examine policies where the public has common interests. By contrast, climate policy has broad public support but concentrated opposition in communities where costs are imposed. This spatial distribution of weak supporters and strong, local opponents mirrors opposition to other policies with diffuse public benefits and concentrated local costs. I use a natural experiment to investigate whether citizens living in proximity to wind energy projects retrospectively punished an incumbent government because of its climate policy. Using both fixed effects and instrumental variable estimators, I identify electoral losses for the incumbent party ranging from 4-10%, with the effect persisting 3 km from wind turbines. Voters also discriminate by correctly punishing the level of government responsible for the policy, providing evidence that voters are informed. I conclude that the spatial distribution of citizens’ policy preferences can affect democratic accountability and exacerbate political barriers to addressing climate change. ∗Thank you to Liqun Chen, Matto Mildenberger, Chad Hazlett, Daniel de Kadt, Rebecca Silverman, Kathryn Buggs and Nicole Goodman for assistance with this project and to Chris Warshaw, Danny Hidalgo, Jens Hainmueller, Laura Chirot, Alex Hertel-Fernandez, Gina Bateson and the participants at ONSEP 2013, the UT Energy Forum 2014, APSA 2013, MPSA 2014, CPSA 2014, DUSP PhD workshop and GSWIP MIT for feedback on earlier drafts. I also acknowledge financial support from Social Sciences and Humanities Research Council and the MIT Martin Family Fellowship. Replication data are available at the AJPS Dataverse at http://dx.doi.org/10.7910/DVN/SDUGCC