Climate change opinion and recent presidential elections

We review patterns of climate change opinion and polarization, and we estimate the effect of climate change opinion on recent U.S. presidential elections, bringing together three lines of evidence. First, we review recent polling studies that document climate change opinion, including: how it varies across regions and demographic groups, how it compares to other issues in importance, how it fits into broader political polarization, and the degree to which voters trust the two major political parties on the issue. Second, we use individual-level data from the Voter Study Group to estimate the effect of climate change opinion on voting behavior in the 2016 and 2020 presidential elections, controlling for demographics and other issue opinions. We do this using logistic regressions and probability forests. From these results, we project how the popular vote might have changed in the 2020 presidential election if voters' preferences for the parties' climate change stances had been different. Third, we build a model that can project Electoral College outcomes probabilistically in hypothetical popular vote scenarios. We find that climate change opinion has had a significant and growing effect on voting that favors the Democrats and is large enough to be pivotal to the outcomes of close elections. We project that climate change opinion probably cost Republicans the 2020 presidential election, all else being equal. We are aware of many possible implications of our findings for policymakers and advocates from a variety of political persuasions. However, we do not editorialize about or discuss these implications. Instead, we lay out our methods and results as matter-of-factly and objectively as possible, and we let our findings speak for themselves.