Today is a first anniversary and a second anniversary. The first anniversary of Joe Biden’s inauguration as President, and the second anniversary of the US reporting its first case of COVID-19.
One can therefore compare, on a temporally fair basis, how President Trump and President Biden have confronted COVID-19. Joe Biden and COVID-19 have just finished presiding over the US for the past year; for the year before that, Donald Trump and COVID-19 presided over the US.
Personally I think such a simple comparison is not hugely fair or edifying, even across equal time periods, because other important factors feed into the size and consequences of a pandemic. Nonetheless, some people do find comparisons like this interesting, and I’m confident one of them is Joe Biden, who declared that anyone responsible for 220,000 COVID-19 deaths “should not remain” President.
That benchmark in mind, I weigh the numbers of COVID-19 deaths reported under Trump and under Biden, according to Our World in Data’s summary of confirmed COVID-19 deaths.
On January 20, 2021, the US had experienced 412,893 confirmed COVID-19 deaths. Today it’s experienced 857,768. Taking the difference, the US suffered 444,875 confirmed COVID-19 deaths in Biden’s first year — and that isn’t counting the one-or-two thousand deaths to be assigned to today.
Basic arithmetic then shows that the body count for Biden’s first year is 8% higher than the body count for Trump’s last year. And it’s more than twice the 220,000 deaths Biden defined as disqualifying in October 2020.
By this confounded and approximate — but Biden-approved — metric, Biden’s COVID-19 record is already unacceptably poor and worse than Trump’s.
It’s easy to think of reasons why Biden faced a tougher pandemic than Trump, perhaps most obviously the emergence of faster-spreading variants, but Trump’s supporters could give opposing reasons why Trump faced a tougher challenge (like the lack of vaccines until Trump’s penultimate month in office, or street protests breaking out in every state during the pandemic’s first wave). Neither type of post hoc rationalization interests me here; both are a waste of time given the crudeness of raw body counts.
The one objection I entertain here is the objection that the death counts themselves are materially wrong. Undercounting of pandemic deaths has been a problem in most places, and the US is no exception. The CDC estimates that the US has suffered 968,036 excess deaths from February 1, 2020 through January 8 this year, more than 100,000 above the number of confirmed COVID-19 cases.
One can debate how many of those 968,036 excess deaths are down to SARS-CoV-2 infection itself as opposed to anti-pandemic restrictions, booming drug overdoses, or Americans buying more guns and shooting each other more often, but that’s basically a debate about how to separate indirectly pandemic-driven deaths from directly pandemic-driven deaths; it’s not credible to assume that overdoses and gun violence would have shot up by 30% year on year without the pandemic. In normal times the 968,036 excess deaths would, by definition, have been about zero, and the pandemic is what made the difference, directly and indirectly. If anything, the 968,036 estimate is too small, since it’s deflated by non-pharmaceutical interventions hitting mundane respiratory diseases like influenza.
In short, excess-death estimates imply that the US has underreported COVID-19 deaths, so one might compare excess deaths under Trump to excess deaths under Biden, instead of comparing inadequate counts of confirmed COVID-19 deaths.
The CDC’s weekly estimates of excess US deaths total 903,367 from January 12, 2020 through January 8, 2022. That total comprises 524,495 excess deaths through January 16, 2021 and 378,872 since then. The data’s weekly resolution makes it impossible to use January 20 exactly as a cutoff date, which isn’t ideal, but a bigger problem is that the latest counts are spuriously low. The data file explicitly warns that recent weeks’ data “are incomplete”, and reading the numbers confirms that.
Each of the 9 weeks of October 10, 2021 through December 11, 2021 has between 63,800 and 65,900 deaths attributed to it.
The next week, December 12–18, has an abruptly lower count of 62,047, hardly plausible in early winter with Delta then going strong and Thanksgiving-driven infections feeding into hospitals and morgues.
Only 57,281, 49,977, and 38,431 deaths respectively are reported for the 3 weeks after that.
Attributing only 378,872 excess deaths to the weeks after January 16, 2021 is therefore too generous to Biden. There’s room to argue about how to correct for reporting delay, but some correction is plainly needed.
To keep things simple, and to avoid grossly overestimating excess deaths under Biden, I just take the average weekly deaths for the final pair of fully reported weeks — 65,400 per week from October 10, 2021 through December 11, 2021 — and use them to re-calculate excess deaths in subsequent weeks.
That adds 53,864 deaths in the 4 weeks from December 12, 2021 through January 8, 2022. Last week (January 9–15) is missing entirely from the data file in deference to reporting delay, but with a baseline of about 61,000 expected deaths, my imputed count of 65,400 deaths implies 4400 excess deaths. Finally, this week, which isn’t yet over, adds 3100 deaths to date.
Adding 53,864 and 4400 and 3100 to 378,872 brings excess deaths under Biden up to 440,236. Despite the upward correction, that’s notably less than the estimate of 524,495 for Trump’s last year, reversing the result of comparing confirmed COVID-19 deaths. That might be explained by improving case detection over time; Trump could have benefited from worse test coverage and laxer attribution of COVID-19 deaths in 2020 compared to 2021.
At the same time, my estimate of excess deaths under Biden is, like the number of confirmed COVID-19 deaths under Biden, more than twice Biden’s red line of 220,000. That ought, maybe, to provoke some soul-searching.