Why partisanship affected Americans’ pandemic perceptions

Aug 05, 2020

As the COVID-19 pandemic began its spread across the United States, politicians and pundits were divided on how great the public-health threat was. Washington University’s John Barrios says that divide was reflected in the social-distancing behavior of individuals, whose risk perceptions appeared to be informed by media commentators and the leaders of one party or the other.

Video Transcript

One would hope that when your life is on the line that partisanship drops out, right? We’re both being affected by this notion that something comes up, and it’s going to kill you. It shouldn’t matter if you’re Republican or Democrat, your reaction should be quite similar.

The way we think about partisanship affecting individuals’ responses to these public-health events stemmed [in part] from just reading the newspaper, thinking about how people have been debating. We see these partisan divides, these very black-and-white views on certain topics.

However, a lot of these public-policy debates with respect to health are very uncertain. So even though COVID-19 is coming, there’s uncertainty as to, “Am I going to be affected? Is it affecting larger cities more than rural areas?”

So perception really matters in terms of how you behave. And it’s even more important in these epidemics because of the externalities that our behavior has. I could be very conservative because I perceive COVID-19 as really bad and I could get it, right? I could be very conservative; yet, if my neighbor doesn’t really care, it doesn’t matter how much I stay in my apartment if he knocks on the door, and he’s been walking around the whole of Chicago and the riverside thinking, “Oh, it’s great. I can walk around. There’s nobody outdoors.”

The externalities from our behaviors mean that we really need compliance for a lot of these things. And that also now leads us to: How do we perceive the laws being passed, the guidance being given by both the CDC and those in political power?

And our idea was: To what extent would we observe differences in the behavior of these individuals? Given that the political party in power at this point is a Republican president, and you have a Democratic House. 

We wanted to explore two individuals that are exposed to the same kind of risk. Because what we don’t want to be capturing is just that if you’re Republican and live in a rural area, you’re just less likely to be affected by COVID-19 because there’s no density, [and it’s] less likely for individuals to come from abroad. So in that sense, it wouldn’t be that crazy to believe that they move around more than people in cities, where there’s high density and there’s more interaction from abroad. 

We want to control for all of that and generate this experiment where we’re saying two individuals that have the same likelihood to be exposed are now behaving differently because one perceives it as a lower risk because “those in my political party view it as not that big of a deal.”

The competing argument that a lot of people would make is just saying, “Republicans might have different preferences, or risk preferences, than Democrats.” They might be a lot more risky, right? And as a result, you would say, “Oh, Republicans aren’t going to follow the rules as much as Democrats because they’re willing to take on more risk.”

So we try to look at these two arguments. And what we end up finding is that yes, after you try to control for all these other characteristics—the economic characteristics in the area, the actual risk of contracting COVID-19—we still observe a difference between high-Trump areas versus low-Trump areas.

Now, then you would say, “Well, is this just risk preferences?” Well, no, because if we look at the time series over the pandemic, at Republican news media like Fox News, as the president starts taking this more seriously, we actually see that the behavior of these high-Trump areas, these high-Republican areas, they become more conservative. They’re doing more social distancing.

So that would go against the view that it’s just a preference argument. It’s more that the perception seems to be changing as those in power, those of your political party, start taking it more seriously.

So we use this event. We can say, “Hey, there was CPAC.” CPAC is the Conservative Political Action Committee gathering. At that event, certain high-level Republican senators, like [Ted] Cruz from Texas, got exposed to someone with COVID-19. This was reported the following week. So when that information came out, and you heard that Cruz started doing self quarantine, then we start seeing that in these Republican areas, we start getting a little bit more social distancing. 

They’re kind of catching up. Democrat areas in the same time period are still becoming more conservative, they’re social distancing more. It’s just that Republican areas are now doing it at a higher rate because they’re catching up, because they’re taking it seriously.

And in that sense, it’s not surprising, but it’s worrying when we think about policy responses that partisanship affects compliance, given that we’re in these free societies—we don’t have the National Guard or the military coming out and standing on the streets, making sure we don’t leave our apartments. We rely on compliance by the electorate.

Bipartisan reactions would seem to facilitate more of this compliance. Rather than having the two parties opposed, you would think that bipartisanship would allow for more compliance, because now everybody’s thinking is on the same page.

We’re not saying that the Democrats were right, or Republicans were right. We’re just saying that it’s very interesting that despite being exposed to the same risk, we still see different perspectives. It could be that Democrats overreacted to the risk, or it could be that Republicans underreacted. We don’t want to really take a stand on that because in order to take a stand on that, you have to be really specific on the assumption. So we’re not taking any normative stance on who was right or wrong.

Our main point is: If you want to think about policy and compliance in this voluntary, free society that we’re in, information and how we perceive that information matters. And the fact that we have either news sources or political leaders from the parties that have these diametrically opposed views on subjects, for whatever reason, seems to affect the behavior of individuals [in ways] that have real consequences—in this case, in human lives, because we end up observing certain mortality rates.