In a November 30, 2022, speech on “Inflation and the Labor Market,” Federal Reserve chairman Jerome Powell blamed most of the 3.5 million estimated shortfall in the U.S. labor force on premature retirements. He also blamed a large portion – between 280,000 and 680,000 – on “long Covid.”
In a footnote, however, Powell acknowledged a far more somber factor: an estimated 400,000 unexpected deaths among working age people.
It’s easy to blame these deaths on Covid-19. The virus is of course one significant cause. But it’s not nearly the only cause, especially among young and middle-age workers. We need better government data transparency to make a full assessment. Until then, we can proceed with others who track mortality for a living – life insurance companies.
The Great Divide – 2020 vs. 2021
In 2020, Covid-19 took many lives, even among select groups of middle-age people, specifically those with comorbidities such as diabetes. In 2020, Covid did not take very many lives of healthy young and middle-age people – for example, the types of people who are employed at large and mid-size companies and who have group life insurance. As you can see in the chart below, group life insurance benefit payments in 2020 were barely higher than in 2018.
In 2021, however, group life payments exploded by 20.7% over the five year average and by 15% over the acute pandemic year of 2020. Why would healthy young and middle-age people suddenly begin dying in large numbers in 2021 when they’d navigated 2020 with relative success?
Especially when we consider that in 2021, the U.S. administered 520 million Covid-19 vaccine doses. Shouldn’t healthy people employed in good jobs with good benefits, now protected with vaccines, have fared better in 2021 than in 2020? Surely, overdoses and suicides have risen in recent years. But those causes of death are less prominent among the group life cohorts in general, and the latest data confirm these were not drivers of the group life surge. Curiously, two of the largest spikes in 2021 came from deadly automobile accidents and non-automobile accidents.
Let’s look at a few of these young adult age groups in more detail. In the charts below, we’ve broken out total all-cause deaths into three groups – 30-34, 35-39, and 40-44. Eyeballing the age group charts alone shows that factors other than Covid-19 itself must have driven large portions of the mortality spike in young and middle-age workers. (We are using official statistics, which likely overstate Covid mortality and understate non-Covid mortality. It’s the best we’ve got for now.)
The most important overall point is that 2021 was far worse for young and middle-age people than 2020.
Another key point is that 2022 was also worse than 2020, though not as bad as 2021.
Mortality rates in 2022 were still dramatically higher than the pre-pandemic baseline.
Covid-19 hit hard in 2020, especially for the old, vulnerable, and comorbid. In other words, Covid-19 took many of the most unhealthy from us in 2020. In principle, therefore, a smaller number unhealthy people might have been susceptible to Covid-19 in 2021 and 2022. High mortality years are often followed by low mortality years. After two successive high mortality years, the third year is even more likely to be low-mortality. For 2022 to be as bad, or somewhat worse, than 2020, is thus a big surprise. Last year’s milder Omicron variants make 2022’s stubbornly high mortality rate even more baffling.
All-cause mortality is crucial to understand whether public health policies are working. All-cause numbers can also help expose faulty reasoning when overly narrow, overly complicated, or overly clever analyses miss or hide important signals. For example, an analysis which purported to show lockdowns reduced Covid deaths but which neglected to show other deaths rose even more, would not reflect the totality of the policy’s effects. Likewise, a chemotherapy which shrinks tumors but kills patients may be successful in its narrow task yet fail the larger mission. Most analysts and health authorities studiously ignored all-cause over the last three years. The all-cause figures above show our Covid policies were far from successful.
For other purposes, however, it’s helpful and even necessary to drill down on specific causes. Important signals can also be lost in large groupings – Simpson’s paradox, for example, is a common statistical illusion. (Few have dug deeper, with as much specificity, as John Beaudoin, an engineer from Massachusetts who gained access to his state’s digital death records for the last eight years. He shows that specific causes of death spike and fall at important moments and periods. CDC data is not organized with such granularity. More on Beaudoin’s analysis in coming weeks…)
We know that recent years saw an upswing in drug overdoses and suicides, which accelerated with the pandemic lockdowns. Although these troubling trends cannot explain the enormous and unprecedented all-cause mortality seen above, we should attempt to account for them. Likewise, although Covid-19 did not cause all these record deaths, it was a significant factor.