If you read the second paper you cited, directly from the abstract, you'll see that they explicitly said that 1918 was "starkly different from today": > Our re

If you read the second paper you cited, directly from the abstract, you'll see that they explicitly said that 1918 was "starkly different from today": > Our results are not inconsistent with an emerging literature that finds negative short-run effects of COVID-19-related school closures on learning. The situation in 1918 was starkly different from today: (1) schools closed in 1918 for many fewer days on average, (2) the 1918 virus was much deadlier to young adults and children, boosting absenteeism even in schools that stayed open, and (3) the lack of effective remote learning platforms in 1918 may have reduced the scope for school closures to increase socioeconomic inequality. Furthermore, from the paper: > Another important contrast is that school closures in the 1918-19 pandemic were substantially shorter than current COVID-19 related school closures, potentially limiting their effects. In our sample of 1918-19 school closures, the average closure length was 36 days, and some cities decided to make up for missing school days by extending the school year. Closing schools is just one of the many measures society has taken to steal the prime years of life from the young to give to the elderly.