Here’s another letter to Nick Horn:
Thanks for your follow-up letter in which you write that I, in this post , “totally miss the point of the New York Times article which DOES show positive correlation between covid and excess death of people between 25 and 44. These are real.” You conclude that I “callously discount them.”
Of course these deaths are real. But this reality doesn’t prove the op-ed  authors’ contention that Covid poses serious risks to persons in that age group. And my pointing out this fact doesn’t mean that I’m callous.
Overlook here the op-ed authors’ failure to note in the NYT that only 38 percent of excess deaths in this age group are attributed to Covid, and – to make your and the authors’ case as strong as possible – assume that all excess deaths are caused by Covid rather than some being caused, for example, by the lockdowns themselves or by whatever causes ordinary year-to-year random variation in deaths. That gives 12,000 Covid-caused excess deaths of Americans ages 25-44 from March through July (the period of the authors’ study). 12,000 is 0.014 percent of Americans in that age cohort. This percentage tells me that young adults, while not at zero risk from Covid (Who ever said that?!), are at only very low risk.
How can my conclusion here be squared with the authors’ claim that “Young adults are dying at historic rates”? Easily, for the authors’ claim – while sounding unambiguously ominous – is actually quite misleading. To see why, I’ll use a hypothetical example. While my numbers are made-up, they reveal starkly the error that leads the authors and their audience to leap mistakenly to the conclusion that Covid poses a serious risk to young adults.
Suppose an age cohort contains each year one million people. And in each year until 2020 one person in that cohort died. Now Covid arrives and takes the life of one person in the cohort who would not otherwise have died, so that in 2020 this cohort suffers the death of two persons. In 2020 excess deaths in this cohort are 100 percent of expected deaths, and the deaths of people in this cohort have doubled! These people are indeed “dying at historic rates.”
Yet clearly this fact doesn’t support the conclusion that Covid poses for people in this cohort a significant risk. Despite the huge Covid-caused percentage increase in excess deaths in this cohort – and despite Covid causing the rate of death in this cohort to be “historic” – Covid kills in this cohort only one in one million people. The chance that Covid will kill any randomly chosen member of this cohort is a minuscule 0.0001 percent.
Again, in fact using actual data – and even assuming that all excess deaths in the 25-44 cohort are caused by Covid, and even if we annualize the authors’ excess-death figure to get excess deaths in this cohort of 28,800 – the chance that Covid will kill any randomly chosen member of this cohort is a minuscule 0.0329 percent.
I leave to you the exercise of comparing this tiny chance of Americans ages 25-44 of dying from Covid with their chances of dying from automobile accidents, poisoning, drowning, or other familiar hazards the risks of which do not cause people to fly into fits of civilization-destroying hysteria.
Donald J. Boudreaux
Professor of Economics
Martha and Nelson Getchell Chair for the Study of Free Market Capitalism at the Mercatus Center
George Mason University
Fairfax, VA 22030