A Closer Look at the Death Toll of 2020

By Robert VerBruggen

Tuesday, January 05, 2021

 

The year 2020 is, mercifully, behind us. But far into the future we will be asking ourselves what the hell just happened.

 

One of the most important questions, of course, is what we accomplished through our reaction to COVID-19. As individuals, we retreated to our homes to protect ourselves and stop the spread of the virus. Governments mandated lockdowns, masks, social distancing, and more. The economy predictably cratered. All this certainly had consequences for our mental and physical health, our substance abuse, our kids’ education. Was the cure worse than the disease?

 

A new study from Casey B. Mulligan, who recently served as chief economist for President Trump’s Council of Economic Advisers for about a year, is a small step toward answering that question. It focuses specifically on the issue of who died, and why, through October 3 of last year. We’ll have better analyses once our slow-moving federal bureaucracy gets more data together, but the study paints a broad-brush picture of the death toll from COVID — and from our response to it.

 

The paper compares last year with previous ones, using a statistical model to predict what mortality would have looked like in 2020 if preexisting trends had continued. Things actually got off to a good start, with deaths below expectations, possibly thanks to a mild flu season. But from March through early October, “excess” deaths totaled a shocking 250,000.

 

How much of this was just COVID itself? We do have an official tally of COVID deaths, of course, but many worry that it’s wrong somehow — that folks who died “with but not of” COVID were included, or that officials had a financial incentive to misclassify non-COVID deaths as COVID, or that many genuine COVID deaths were misclassified in the other direction. So Mulligan digs into the data to see how believable the official numbers are.

 

At the very beginning of the pandemic in late March and early April, mortality among the very old (age 85 and up) shot up considerably more than one would expect from the official COVID statistics. This likely means that the health-care system hadn’t yet figured out how to identify these deaths consistently. Starting in May, though, the excess deaths and the official tally begin to line up closely for this group, implying the official numbers are more or less right for that period.

 

But something unsettling happened with working-age men. They saw a gradual increase in excess deaths after the pandemic set in, above and beyond the official COVID tally. This is a pattern that can’t be waved away as undercounted COVID deaths. These are, in all likelihood, largely deaths resulting from our response to the virus.

 

Overall, the study estimates that, of those 250,000 total extra deaths, 30,000 were from causes other than COVID itself. Here’s how the trends look for the population at large:

In time, the government will tally up the details from all the death certificates filled out in 2020, and we’ll know which kinds of deaths increased most. But Mulligan provides strong evidence that “deaths of despair” — i.e., deaths from drug overdoses, alcohol abuse, and suicide — were a significant factor.

 

For one thing, demographic groups that already had high rates of these deaths tended to see the biggest increases in non-COVID excess deaths last year. And for another, the early data we have on overdoses suggest they rose markedly, at least in the first half of 2020. This is true both in federal data running through May and in local data from several counties that go a bit later.

 

So, was the cure worse than the disease? If we take the question literally and go by this study’s estimates, the disease killed 220,000 while the cure killed “only” 30,000, so it isn’t even close — or at least it wasn’t as of early October, when the paper’s data run out.

 

But if we’re trying to figure out what we gained from the decisions that killed 30,000, the real question isn’t how many people did die of COVID, but how many didn’t die of COVID thanks to those decisions. If the lockdowns and other measures lowered the COVID death toll by even, say, one-fifth, they probably reduced the overall number of deaths — though the people the measures killed were younger, and everyone else suffered a reduced quality of life while the measures were in effect too.

 

There’s a ton more to learn about all this. All the way back in May, I compiled a very long list of studies that sought to determine how effective the lockdowns had been at stopping the virus, and the conclusions were scattered. Since then there’s been even more research — and some places with seemingly effective responses have gone on to suffer badly in subsequent waves. We won’t know where everyone ends up until the pandemic is over, and even then there will be intense arguments over what effect the lockdowns had and whether the benefits were worth the cost. It’s very hard to tell what might have happened if individuals and governments had behaved differently, and very hard to weigh costs and benefits that are so varied and uncertain.

 

The question of who died and how is much simpler, but no less important. On that question, at least, the answers are quickly coming into focus — and nothing about the picture is pretty.