By Tyler Cowen / Bloomberg Opinion

As the delta variant of covid-19 spreads, many parents are worried that their public schools will not fully reopen this fall. That would be a serious unforced error; and the mere possibility of it is evidence that America is not thinking rationally about risk.

Granted, the issue is complicated, with different risk factors and behavioral expectations for kindergartners compared to high-school students. And there are options between a pre-pandemic status quo ante and complete remote learning.

Still, it may be helpful to focus on one simple question (which does not yield a simple answer): How many student deaths are you willing to accept to have schools open again and operating at full capacity?

Is it 100? 300? 1,000? Or maybe zero? Might a few dramatic cases of “long covid” be enough to halt reopening?

It is hard for a politician or school board member to raise their hand and say: “I am willing to let [fill in the number] of children die to get schools fully reopened.” In other words, the very act of debating the question makes it hard to answer. And given the prominence of covid as a news story, the question will invariably be considered in explicit and emotional terms.

Of course, with or without covid, some number of children die at school. But it is surprisingly difficult to find out how many. In 2020, there were more than 50 million students in public elementary, middle or high schools, yet there is no systematic national database of student deaths at school. School shootings have claimed up to 75 deaths annually in recent years, and there are many other possible causes of death, such as traffic or sports accidents.

It’s entirely plausible that a few hundred students die each year for reasons directly related to school attendance. If suicides induced by school bullying but occurring off campus are included, the number could be higher still. Some 4,400 young people in America commit suicide in a typical year, and surely many of those deaths are attributable, at least partially, to events at school.

Adding up all these admittedly indirect chains of causation, it’s possible that school attendance leads to at least 2,000 deaths every year in the United States. And those have nothing to do with covid.

Fortunately, it is not customary in normal times to debate whether it is worth opening schools knowing that it could result in the death of perhaps 2,000 students. The true toll of opening schools is unknown, much less debated, and if there is a discussion it is over school shootings, which ought to be preventable (or at least limited) by measures other than closing schools.

This “head in the sand” approach is highly imperfect. Still, it is preferable to panicking and closing the schools every year.

It is difficult to calculate how many children have died of covid, but perhaps the best estimate comes from England, where it caused 25 deaths of people younger than 18 in the year ended in March. The final tally is certainly higher in the more populous U.S., but as of July seven states still were reporting zero covid deaths among children. This recent estimate suggests 358 deaths, though it is based on only 43 states.

Yes, it is worth considering whether school reopenings will lead to unacceptably high levels of covid in the nonschool population. It is also worth pointing out that covid is spreading very rapidly in states with low vaccination rates; without the schools playing a role. In any case, it does not justify focusing solely on the safety of children in discussions of school reopening.

Economists have long studied the tendency of people to assign more value to a “known life” than to a “statistical life.” When a baby is trapped down a well, for example, many millions of dollars will be spent trying to save her. Her photo will appear on the evening news and on social media. Yet when it comes to saving lives in the aggregate, such as by installing more and better smoke detectors, there is only modest interest.

Right now too many Americans are trapped: Because the pandemic has been so dramatic for so many, every life looks like a known life rather than a statistical life. We all need to start working our way back to a bit more emotional distance.

Tyler Cowen is a Bloomberg Opinion columnist. He is a professor of economics at George Mason University and writes for the blog Marginal Revolution. His books include “Big Business: A Love Letter to an American Anti-Hero.”