Destroying The Economy Just to Save Some Old People?
Survival: Old vs. Young
From the beginning of the COVID-19 pandemic, there has been some discussion about whether the harm from the virus is dire enough to justify the harm we are doing to the economy through shut-down orders. As Stanford biologist Michael Levitt put it, “Look, this really is a disease that’s going to affect old people.” Are we sacrificing the young, who are hit hardest by the imploding economy, in order to save the old, who are going to die soon anyway?
I’m going to concentrate on deaths here. It’s what we have the best statistics for, and it is less complicated than the suffering and potential long-term consequences for people who get seriously ill from COVID but don’t die. This is not to say that these other considerations are unimportant.
There certainly is a disparity in deaths. As the top bar in the figure shows, almost 30% of the deaths are of people 85 and up (right side of bar), whereas 25-44 year olds account for less than 7% (left side) and the number of deaths under 25 years old is practically zero. This is even more lopsided if you consider the demographics of the country (bottom bar). Less than 2% of the country is over 85, and around 26% are 25-44. The 25-44 age group is dying at a rate less than one fiftieth of the 85+ age group.
Implicit in this train of thought is the idea that an old person’s life is worth less than a young person’s. Let’s stipulate that there is a little truth to that: if we had to choose between saving the life of a 30 year old or a 90 year old, I think most of us would choose the 30 year old. The 90 year old has lived a long life and expects to die relatively soon, but the 30 year old expects to live a long time, may have children that depend on him or her, may have parents upset to see their child predecease them, etc.
But if we are willing to use this argument to overlook the deaths of old people, maybe we should consider another question: how much more is the life of the young person worth?
Or to put it another way, maybe the small number of younger people killed by the virus is at least as much of a tragedy as the far larger number of old people.
Watching the Pandemic Kill
The statistics visualized in the figure are not from the official counts of COVID deaths. They are the difference between all deaths in the first 28 weeks of 2020 (all we have at this point) and the same period in 2019. This statistic catches people whose death was not officially attributed to COVID, those who died as an indirect consequence of the disease (for instance, not going to the hospital when feeling chest pains because of worrying about infection), and even people who were saved by the disease (for instance, due to fewer car accidents when people stay at home more). It is also complicated by year-to-year variations in deaths which are unrelated to COVID.
The same CDC data set also provides weekly deaths (see figure below), and so we can see the virus striking at around the 13th week of 2020. Until then, weekly deaths were similar to other recent years, but within a few weeks the statistic leaps to a much higher number. At that point, US society began reacting with people staying at home, schools closing, etc., with the result that deaths declined for the next 10 weeks or so, until a new rise in the most recent few weeks. The peak in deaths occurs at the same time as the peak of deaths officially attributed to COVID, giving us confidence it is primarily measuring COVID deaths. The total number of deaths of people of all ages (about 190,000) is a little higher than the official COVID count (about 150,000).
This figure gives us a way to compare COVID deaths to other causes of death. Here we see that, in the worst week, the virus increased the death rate for people 85 and up by about 50% (25,000 in 2020 divided by 16,000 in other years). It’s true that lots of old people die in an ordinary year, but COVID has brought a big increase in someone’s chance of dying.
This is true for the other age brackets, including 25-44 year olds. The absolute numbers are much smaller, but the percentage increase in 2020 is still big, about 30% (3700/2800). A younger adult only has a small chance of dying in the first place, but that chance has increased by a large fraction. The increase in danger is almost as great for 25-44 as for all of the older age brackets.
What do 25-44 year olds die of besides COVID? The CDC keeps track of that as well. Of the 140,000 people in that age group who die in a year, in 2017 some leading causes are accidents (48,000), suicide (15,000), cancer (15,000), heart disease (14,000), and homicide (9,000).
Actual and Avoided Deaths Among 25-44 Year Olds –
So far this year, about 14,000 more people in the 25-44 age group have died than in previous years. You can think of COVID as adding a new category of death that, all by itself, kills as many in this age range as suicide or cancer.
That number of deaths occurred in a society which took severe measures (though probably not severe enough) against the disease. Interestingly, deaths for 25-44 year olds did not drop much after about week 16, unlike all the older brackets. Is this evidence that this age group did not protect themselves as much as older people? If so, it might be a combination of not worrying about the disease as much and not being able to shelter in place due to work obligations.
Still, the data only goes to week 28. A whole year at the peak death rate would mean around 40,000 deaths (30% of 140). And even for this age group, we know that measures against the disease had some effect, because the death rate stopped going up. If we are going to wonder about the value of hurting the economy to fight the disease, the relevant comparison is the casualty rate in an open economy, not in the economy that we shut down. At the same rate of growth that we saw from weeks 13-15, another 3 weeks of growth would give a 60% increase in death rate, rising to a 90% increase after that. These too are probably underestimates; if the growth was exponential, death rates could have climbed much higher before leveling or declining.
In a year, its easy to imagine 100,000 or more people in their prime working and child-bearing years dying from COVID-19. The fact that an even larger groups of elderly people would die doesn’t stop this from being a very large number of younger people dying. For the 25-44 group, it is relevant to compare these deaths to the 58,000 the US lost in the Vietnam War.
Other Age Groups
So far, I’ve concentrated on the 25-44 age cohort, but its worth at least mentioning the others. The 45-64 group has lost about three times as many people as the 25-44 group. Speaking as a 55 year old, I like to think that, in the absence of COVID, this age bracket has some expectation of decades of life remaining. This is also true to varying extents of older age brackets. While it’s worse if my kids die than if I die, it’s still bad if I die!
The Bottom Line
A very small fraction of people dying of COVID-19 are under 45 years old. This has led to speculation that younger people are bearing the brunt of economic shock, all for the sake of protecting the majority of potential COVID victims, who are elderly. Here I’ve shown that deaths from rampant COVID would still add up to a big loss of 25-44 year olds, both in terms of their usual death rate, and as an absolute number. Economic and social sacrifices that this age group makes to stop COVID protects not only old people, but also themselves.
This is not to say that the economic shock is inevitable. Germany, South Korea, and a handful of other nations show that the US could do a much better job of protecting against the disease without destroying the economy. Here I’ve given an additional reason for why we should keep up the fight.
Data Sources
US Demographics: Age and Sex Composition in the United States: 2019, Table 1