American parents and educators alike are split on whether schools should reopen in the fall. Advocates point to flattening coronavirus-infection curves in many states and studies downplaying the role of children in spreading the virus. Opponents focus on rising cases in other states as well as reports of serious health issues for children coming down with a Kawasaki-like disease.

The debate won't be settled before the big decision must be made. But we do have some historical data that sheds light on the long-term costs of keeping kids out of school to avoid a pandemic. It comes from a long-forgotten polio epidemic that struck the U.S. just over a century ago.

At the end of the 19th century, poliomyelitis — better known as "infantile paralysis" — was a rare, little-understood viral disease that overwhelmingly affected children, not adults. It was what one historian of medicine has described as a "fine-print medical curiosity" known mostly in Europe.

In the first decade of the 20th century, isolated outbreaks started appearing in the U.S. during the summer. Then it exploded in 1916 for reasons that remain a mystery, even today.

The epidemic began with a handful of Italian children in Brooklyn and spread outward from there. Public health officials erroneously blamed Italian immigrants for spreading it. They also blamed cats, triggering a cruel campaign that killed tens of thousands of animals in a futile attempt to arrest the disease.

Approximately a quarter of the children afflicted died, largely because some of the later treatments — the scary but effective "iron lung" — would not become available for another decade or so. Parents watching their children suffer from pain and paralysis felt understandably terrified and helpless.

The mystery surrounding the disease and how it was transmitted only fueled anxieties. "The principal thing known about poliomyelitis is that it is one of the most baffling diseases studied by scientists, and that they really know very little about it," the New York Times wrote.

New York City closed down, but the disease still spread throughout the East Coast, and eventually the entire nation. At least 27 states faced major outbreaks before the epidemic faded in November. Faced with the danger to school-age children, cities and towns hit hard by the disease kept schools closed when summer ended.

Even after schools reopened, some parents kept their kids at home. In New York City, for example, it's estimated that upward of 200,000 children —roughly a quarter of the total — did not return until several weeks after the larger school system reopened. Similar conditions prevailed in other cities and towns, with even small outbreaks triggering school closures.

By the time the epidemic subsided, 6,000 children across the country had died and thousands more remained paralyzed. Most would make partial or full recoveries, though the disease returned in subsequent summers.

A century later, the economists Keith Meyers and Melissa Thomasson chose this first polio epidemic to study whether keeping kids out of school, even for a few weeks, had long-term consequences. This research, conducted before the Covid-19 pandemic materialized, was what they described as a "natural experiment" on the "effects that epidemiological induced panics and large-scale school closures have on the educational outcomes of school-age children."

More specifically, they speculated that closures might have prompted some kids to drop out of school. Testing this hypothesis required running six different regressions that took population samples in different locales. Some of these places sustained significant outbreaks; others did not. Data on the severity of the outbreaks — specifically, polio deaths — became a proxy for delays in reopening schools.

What they found was striking. Even though schools delayed reopening for a few weeks or a month at most, the effects persisted across the lives of one group of children in particular: those who were between the ages of 14 and 17 when the epidemic hit. By contrast, children younger than 14 or older than 17 did not exhibit this pattern.

Why? At the time, the legal working age was 14. The authors speculated that when schools failed to reopen, many students went out and got a job — and never came back.

The study found that a 1% increase in the polio morbidity rate led to a 6% reduction in average educational attainment. Simply put, the bigger the polio problem in any given area — and the more disruptive the school closings — the worse the long-term effects for this particular age group. Cities and towns that escaped polio didn't follow this pattern.

The study has several potential implications for our current dilemma. Much of the debate about reopening schools has focused on fears that children may "fall behind" in the curriculum. In this age of standardized tests and lock-step learning, that's an understandable anxiety, but the evidence from 1916 suggests it is likely overblown. Students seem to have caught up back then.

Rather, the real damage done by closing schools (or shifting to online learning) is that students in high school or college may drop out and enter the workforce — or even worse, join the ranks of the unemployed. In fact, there is already considerable evidence that students forced into remote learning did precisely that, disengaging from online classes at high rates and focusing on other duties, like taking care of siblings while parents continue to work during the outbreak.

Students who dropped out in 1916 probably suffered few long-term consequences. At the time, only about 10% of the population even bothered to finish high school. Formal secondary education wasn't as important for success in life, and college even less so.

We live in a very different world now, where completion of both these levels of education have become increasingly essential for building the skills necessary to secure well-paid employment over the course of a career. Another year of online education in high schools and colleges may well push many students to drop out, with consequences that could reverberate for decades.

There are no easy answers to the question of whether schools and colleges should resume in-person classes in the fall. But the long-term effects on young adults' education needs to be factored into the final decision.

Stephen Mihm, an associate professor of history at the University of Georgia, is a contributor to Bloomberg Opinion.