The pandemic, which could conceivably have brought the country together, has instead contributed to our growing political divides. Partisan affiliation is often the strongest single predictor of behavior and attitudes about COVID-19, even more powerful than local infection rates or demographic characteristics, such as age and health status, as we show in our new paper, The Real Cost of Political Polarization: Evidence from the COVID-19 Pandemic. Accordingly, a state’s partisan orientation also explains its public health policies, including the timing and duration of stay-at-home orders, bans on social gathering, and mask mandates.
In the paper, we analyze survey responses from just under 50,000 U.S. adults surveyed repeatedly by Gallup from March through August, as well as publicly available policy and political data from a variety of sources.
The implications are unfortunate. Ideally, public health policy would be driven by theory and evidence, not the relative power of partisans. State leaders have missed opportunities to adopt mask ordinances, limit gatherings in the most dangerous indoor spaces, and allow businesses to operate safely. Bad COVID policy, distorted by partisanship, has cost lives and jobs, as our work shows.
Polarization of individual attitudes and behaviors
When the pandemic started, two of the most popular news programs—both on Fox—covered it very differently. Tucker Carlson emphasized the disease’s severity while Sean Hannity downplayed it, according to striking research from Leonardo Bursztyn and his co-authors. In survey data, they find that Hannity’s viewers waited longer before significantly changing their behavior compared to Carlson viewers, who were otherwise demographically similar. More Hannity viewers predicted more infections at the county level. The “Hannity Effect” illustrates a much larger pattern: Access to information is heavily distorted by our media diet, and that has real consequences for attitudes and behavior.
Using Gallup data, we document large and persistent partisan gaps in levels of fear over COVID-19, social distancing, mask wearing, visiting work, and the scope of expected economic and social distribution (Figure 1). We also show that political party support is usually the most important variable in explaining these attitudes and behaviors, dominating county-level infections and other demographic variables.
Politics shapes COVID views
Notably, these gaps in attitudes and behaviors persisted even after the disease burden shifted from being disproportionately high in counties won by Hillary Clinton in 2016 in the spring to counties won by President Trump over the summer. For example, New York City and the surrounding metropolitan area were especially hard-hit in April, with roughly 20% of the population eventually becoming infected, according to CDC estimates of antibody…