It’s important – actually, vital in both meanings here – to know what is being measured and how in order to understand what the measurement is telling us. This being a point that is being dismally ignored here.
Covid hasn’t reduced American life expectancy in the slightest. It has reduced the lifespan of many Americans – these two are not the same thing.
U.S. Life Expectancy Fell By 1.5 Years In 2020, The Biggest Drop Since WWII
Nope, age at death fell.
U.S. life expectancy fell by a year and a half in 2020, the largest one-year decline since World War II, public health officials said Wednesday. The decrease for both Black Americans and Hispanic Americans was even worse: three years.
The drop spelled out by the Centers for Disease Control and Prevention is due mainly to the COVID-19 pandemic, which health officials said is responsible for close to 74% of the overall life expectancy decline. More than 3.3 million Americans died last year, far more than any other year in U.S. history, with COVID-19 accounting for about 11% of those deaths.
The deaths and the ages at death are all entirely correct. But that’s simply not the same as life expectancy.
The thing not being understood here is that we use age of death of the current generation as a proxy for how long the next generation will live.
Sure, it’s a pretty good proxy. It’s also probably the best estimation we can make. But it is a proxy, it is an estimation. Which is where the problem lies here.
For that life expectancy number is “how long do we expect someone born today to live?”
Sure, we construct that estimate by asking how long someone who died today did live. But it’s in that difference that we have to be careful.
One thing we cannot do is measure the age of death of those who die and then claim that’s the number. Because of course there are some who were born at the same time who are not dead yet. This being true in ever decreasing portions until we get up to the age of 115 or so when there might be one left of any one birth year. It’s only after the entire birth year is dead that we know what the average lifespan of the whole birth year was.
Yes, OK, after 110 years the numbers aren’t going to change very much. But at 75 or 80 then yes they still can. If the 50% of the birth year that get past the average age of death have significantly longer lifespans than the previous generation then that average will rise.
But that’s not the major problem here with this estimation. We think that Covid is going to limit the lifespans of a baby popping out today in what manner? Quite, we don’t. Even assume that it remains endemic, as with ‘flu, forever. Kids will get it at some point as they age and they will when it’s a sniffle that doesn’t matter. That great mountain of new graves is from the already elderly who had no resistance as a result of previous infection, isn’t it?
It’s entirely true that Covid has reduced the lifespans of today’s elder generation. But it has reduced the life expectancy of the new generation not one whit.
Always remember that proxies are indeed proxies.