COVID-19 doesn't have a political affiliation. It doesn't care who you voted for or who governs your state. Droves of scientists across government and academia, unbiased by politics, are researching the disease, constantly updating our understanding of how to best slow its spread and care for those who are infected. Simply put, COVID-19 should not be a political issue.
And yet... in a country with unprecedented political division, politics have played a painfully visible role in how we respond to this pandemic. In the data below, we can find both the impetus for and effects of policy at the national and state levels. I encourage you to dig through the data first to see what patterns emerge. I'll continue my commentary below the interactive chart.
The US is alone on the world stage in having almost no federal policy around how the pandemic is to be handled. There are no standard metrics or policies for when states or counties should restrict business openings. There has been no coordination to acquire essential equipment. There are no test centers operated by the federal government. Instead, policies have been set almost entirely at the state level.
This lack of federal coordination was deliberate. According to Vanity Fair, The White House did initially put together a plan, but by April, it had been scrapped for one simple political reason: The pandemic was disproportionately affecting blue states. As one insider remarked, "The political folks believed that because it was going to be relegated to Democratic states, that they could blame those governors, and that would be an effective political strategy."
If you look at the data, that's true. At the start of April, 9 of the 10 states with the highest per-capita caseload have Democratic governors Show it. The list of states isn't surprising. Most are home to huge urban centers -- densely-populated travel hot-spots that helped COVID take hold early and spread quickly.
In the absence of any meaningful federal policy, states had to step up. Unsurprisingly, the same states that were hit early took the virus seriously. Both citizens and policy-makers had a front-row seat to the tragedy unfolding before them. Blue state governors and mayors shut down cities and enforced mask and distancing policies.
Meanwhile, Trump criticized the shut-downs and spread misinformation, repeatedly claiming that the pandemic would end soon, either with a vaccine or by running its natural course. He discouraged the shut-downs, sowed doubt about the efficacy of masks, touted miracle cures, and demanded that schools open on time for the Fall.
The effect: A nation divided where Democratic and Republican leaders put forward entirely different policies and where levels of compliance were driven by political dogma.
Fast-forward to the first day of November. Of the 16 states afflicted with the highest per-capita caseload, all voted for Trump in 2016 Show it and 14 have Republican governors Show it.
Can this data tell us where we're going next? A few people have pointed out that when we look at deaths per capita, we see that blue states have been hit the hardest Show it. When we instead look at the rolling window of recent deaths, we see what is likely to be the next chapter in this tragic story Show it.
To give more context to these rolling averages, consider that in a typical year, over a two-week period, you can expect about 28 out of 100,000 people die. So if you're looking at deaths over 14 days per 100k, a value of 28 would represent a doubling of the baseline mortality rate.
States have different procedures for reporting. The biggest errors are introduced by a discrepancy in how states report the total number of tests, which affects the charts marked with a *. Some states report the true number of tests performed. Others report a sum of positive and negative tests with each score tracking unique people to prevent duplication. Here's a summary from the data provider:
In most states, the totalTestResults field is currently computed by adding positive and negative values because, historically, some states do not report totals, and to work around different reporting cadences for cases and tests. In Colorado, Delaware, the District of Columbia, Florida, Hawaii, Minnesota, Nevada, North Dakota, Rhode Island, Virginia, Washington, and Wisconsin, where reliable testing encounters figures are available with a complete time series, we directly report those figures in this field. In Alabama, Alaska, Arkansas, Georgia, Indiana, Kentucky, Maryland, Massachusetts, Missouri, Nebraska, New Hampshire, Utah, and Vermont, where reliable specimens figures are available with a complete time series, we directly report those figures in this field. In Arizona, Idaho, and South Dakota, where reliable unique people figures are available with a complete time series, we directly report those figures in this field. We are in the process of switching all states over to use directly reported total figures, using a policy of preferring testing encounters, specimens, and people, in that order.
For more details, please read the data descriptions provided by the COVID-19 API under "Historic Values for All States".
This was heavily inspired by these beautiful charts by Dan Goodspeed
© 2020 Rob Jagnow