The (Other) Color of COVID: Red vs. Blue During the Pandemic

(CNN recently hosted a special called the Color of COVID, highlighting the disproportionate burden borne by communities of color during this pandemic – definitely worth checking out. In this piece, I look at how the virus is progressing though communities based on their political affiliations.)

I’ll begin this article with a generic “I’ve got friends on both sides of the aisle” disclaimer – that’s what people do, right? As such, I’ve generally avoided political topics in my posts because I find that such conversations online typically lack depth, nuance and a curiosity to understand the ‘Other’s’ perspective. In the past week though, protests against Stay-at-Home directives are surging nationally with a decidedly Republican lean. I’ve also seen some of the aforementioned ‘friends’ in my own network questioning if things are as bad as being made out to be, or are we overreacting? I understand clearly the economic toll that this is taking on individuals, and disproportionately on those that can least afford it, but the potential cost of human life, should we re-open prematurely, doesn’t even need to be calculated to know it’s not worth it – or at least so I thought. So why is this opinion not shared by those protesting? Instead of blaming this perspective on consumption of one particular media source vs. another, I wondered, “Are Democrats and Republicans legitimately experiencing COVID-19 differently?”

It’s no secret that the virus first made its way through coastal urban areas that tend to lean Democrat, where it has been particularly devastating – but it’s since spread through the entire country, with no state unaffected… so how far behind are Republican enclaves? Thanks to the folks at the Center for Systems Science and Engineering at Johns Hopkins University, there is a detailed data set, updated daily, on the spread of COVID-19 at a county level in the US. And while the disease doesn’t differentiate between Republicans and Democrats, I’ve paired that data with the county level results of the 2016 election to understand, at a high level, whether or not there are actual differences of experience that may explain why some believe the country is overreacting.

1. To date, areas that lean Democrat have been impacted about 3 times as heavily as those that lean Republican.

As of April 18, there were 724,504 confirmed COVID-19 cases in the US. Of this total, 77% were recorded in counties that lean Democrat. If you live in one of these areas, you are three times more likely to know someone directly impacted by COVID. You are three times more likely to be someone who was diagnosed with the disease. Overall though, 55% of the population resides in such places, so is 77% of cases really that different or that far ahead from what those living in Republican areas are experiencing? Another way to look at this is the Infection Rate across the entire population. Seven hundred thousand plus cases correspond to an overall infection rate across the US of just about 0.20%. The infection rate in Democrat counties is 0.28%, compared to only 0.10% in Republican counties. That was the approximate rate of infection among Democrat counties on April 3, suggesting that Republicans, on average, may be two weeks behind Democrats on the experience curve.

2. The rate of change to infections is slower in Republican counties.

Republican areas are not only lagging in terms of the onset of cases, their growth curves are less steep compared to those experienced in counties that lean Democrat. One obvious reason for this is the density of urban areas which are over-represented in the Democrat base compared to the less dense rural areas across America that tend to lean Republican. However, there is also evidence to suggest that the earlier introduction of social distancing measures, relative to the number of cases in the community at the time, has also had a positive impact among Republicans.

A quick visual review of the cumulative cases in Democrat vs. Republican areas shows that the two groups appear to be on different trajectories.

At the same time, knowing the time lag for onset and the differences in size of the underlying populations, let’s instead look at the growth in cases after the 100th case was recorded. This view still shows different trajectories, but the difference is less pronounced, but still meaningful. (Side note: I should also caveat that testing availability was particularly poor earlier in the pandemic, so we have to assume more error in the early Democrat numbers.)

To provide additional comfort that this discrepancy is based on more than just the underlying population size and density of Republican counties, we can look at the rate of change in the number of cases, as a percent of the prior total, for both Democrats and Republicans. We expect these rates to be higher earlier in the pandemic before social distancing guidelines and as testing catches up. We also expect the Republican rate to mirror that of Democrat counties but to sit above the Democrat rate on any given day, assuming they are following similar social distancing guidelines, because of the lower base of the infected Republican population (simple math here – an increase of 1 on a base of 2 is 50% vs. 33% on a base of 3) at least until things normalize. And that was indeed the case until about a week ago. Now both rates are essentially equivalent at approximately 4%. These means that on any given day, about 22,000 new cases are diagnosed in Democrat counties compared to about 7,000 across all Republican counties. Without social distancing in Republican enclaves, this curve would lag the Democrat curve by several days, or look completely different, and we would likely not see the stabilization currently occurring.

We’re also seeing this impact by looking at the number of days it takes for the infection rate to increase tenfold. Republican counties mirrored Democrat ones early on in moving from 0.00005% (the approx. infection rate for the 100th case in Democrat areas) to 0.005% (the 10,000th such case). However, it took 3 days longer to then hit 0.05% (the 100,000th case in Democrat regions).

3. Many Republican areas have yet to be heavily impacted.

The 0.05% rate above has additional significance. It represents the 100,000th case in Democrat areas, which occurred on March 29. Republican areas, in the aggregate, hit that milestone 8 days later, on April 6 (representing about 80,000 cases among Republicans), and that lag has since grown to the two weeks mentioned earlier. By that March 29 date, almost all states had implemented Stay-at-Home orders as governors across the US prepared for a surge like that being experienced in New York and other hot spots. And while the aggregate rate of infection across the US is 0.20%, and even 0.10% among heavily Republican areas, there are many counties where that 0.05% mark has not been hit. Almost 50% of people living in Republican counties have not yet seen infections hit this threshold compared to less than 15% of the populace in Democrat areas. For these folks, it may feel like the predicted catastrophe hasn’t (yet) arrived... but they are bearing the financial cost.

Closing Thoughts

·        Republican areas are experiencing lower overall numbers of infections and overall rates of infection

·        While population and population density may be part of the reason, they’ve also likely benefited from relatively earlier social distancing practices

·        As such, while in the aggregate Republican enclaves are lag Democrat regions by about 2 weeks, the population in many areas are closer to 3 or more weeks behind – and are unlikely to ever catch up given the stabilization of new infections

However, this is all predicated on a continuation of social distancing practices, until the rate of new cases is 0% (for two weeks) – should these areas re-open as protests increase, the benefits achieved in the last few weeks could be entirely erased. We only need to think back to the rhetoric three weeks prior to March 29. At that point there were fewer than 500 confirmed cases in the US, the disease was being compared to the common flu, we were instructed to stay calm because it would go away, and we were told that life and the economy go on. What a difference 3 weeks makes. Although Republican areas have not been as heavily impacted, there is no reason to believe they cannot be – and we all need to be mindful of this.

Data Sources:

COVID-19 Dashboard by the Center for Systems Science and Engineering at Johns Hopkins University

MIT Election Data and Science Lab

US Census Bureau