Temporary School Closures Reduce Income and Economic Mobility in Adulthood

classroom, school, education-2093744.jpg
CC0 Creative Commons

It will be years before we will know the long-term effects of learning loss from the COVID-19 pandemic school closures. From what we do know by examining the short-term effects of the pandemic on student outcomes, the closures were detrimental to student development. The National Assessment of Education Progress (NAEP) indicates that test scores in math and reading are continuing to decline despite interventions and tens of billions of federal dollars directed at helping students recover academically. Furthermore, school attendance has continued to experience the persistent effects of the closures with chronic absenteeism at all-time highs.

To help determine what we might expect, I estimate the economic effects of temporary school closures by following the outcomes of a large population of students who lost access to school during the nineteenth century. I find that losing access to school for even just one year can have profound effects on future income and economic mobility, and that these effects are larger for younger and more disadvantaged students. Compared to today, the effect of losing one year of instructional time is equivalent to losing more than two years of lifetime earnings for someone earning a median wage. This significantly undermines the opportunity and mobility of school-age cohorts as we continue to recover from the COVID-19 pandemic. 

Background on temporary school closures

Historical examples can provide policymakers with some indication of what to expect in terms of the long-term effects of losing access to instructional time, as well as which populations are most likely to be affected. For example, scholars have studied the effects of school closures related to summer instruction, the weather, differences in instructional timing, teacher strikes, student absences, and other pandemics.

However, many of the school closures examined in the existing literature lasted only a short period (e.g., a few days or months) rather than the many months that schools were closed to in-person instruction during the COVID-19 pandemic. Post-closure chronic absenteeism is making this difference even more profound. During the 2021-22 school year, at least a quarter of students in 27 states (and the District of Columbia) missed at least 10 percent of instructional days. Before the pandemic, only one state (and D.C.) met this threshold. Many of these studies only compare outcomes for specific ages or grades. This is also a problem with relying too much on the NAEP test scores to assess the consequences of the school closures since they only track progress for grades 4, 8, and 12.

A new study of the effect of school closures on outcomes in adulthood

I attempt to overcome these limitations in a working paper that examines economic outcomes in adulthood for cohorts of students who lost access to school in Virginia during the 1870s. The history of these school closures makes this study unique and relevant to today. In 1878, the Governor of Virginia vetoed a bill that earmarked revenue for schools, triggering a political impasse that lasted an entire year and led to state funding being cut by about 60 percent per student. However, local school districts responded differently to the funding cut; some closed all schools, others closed some schools, and some kept schools open. Across the state, about half of all enrolled students lost access to school for at least one school year.

To conduct the study, I hand-collected administrative data on school closures and student populations from every school district in Virginia during the 1870s. I then joined that information with individual records from the U.S. decennial census and tracked student populations affected by the closures (including those who weren’t enrolled in school) into adulthood. I was not able to capture outcomes for every student due to attrition (e.g., changing names or mortality). But I was able to capture outcomes for about 20 percent of the entire student population for more than 30 years after the school closures.

I found that students who were exposed to the closures were just as likely to be enrolled in school a year after they closed. However, they had incomes that were as much as five percent lower than their peers who were not exposed to the closures more than 30 years later. This is a considerable effect given that most of these students only lost access to one year of schooling. In today’s dollars, that amounts to lost earnings over the course of a working life of about $70,000 for someone making the median income. This is the equivalent of losing a little more than 27 months of pay over the course of a working life.

Figure 1. Predicted income rank of sons in adulthood based on income rank of fathers.

Source: Winfree (2023)

I also found that children who were exposed to closures had lower levels of economic mobility in adulthood relative to their peers. Figure 1 shows the predicted income rank of sons for those whose fathers were in the 25th percentile of income (Panel A) and the 50th percentile of income (Panel B). Everyone whose ages were between the red lines was eligible for school when the closures occurred. You can see that there was almost no difference in predicted income rank for those too young (age four and under) when the closures occurred. However, those children who lived in areas that closed all schools experienced a sharp decline in predicted income rank starting at about six years old when compared with their peers.

The effect was not the same for all children. In fact, I found that there was almost no effect of losing access to school for those whose fathers had the highest incomes. However, children growing up at the bottom of the income distribution were less likely to be upwardly economically mobile relative to their peers if they lived in a county that closed schools. These children were also less likely to be upwardly mobile relative to other children who grew up in the same county if they were eligible for school when the schools closed. In other words, the more economically disadvantaged students were to begin with, the more school closures impacted and limited them in adulthood.

I also found the largest effects on income and literacy for children who were younger (under the age of nine) when the closures occurred. This supports conclusions from across different fields of study that both access to other resources (associated with income) and age at which the closures occur (associated with their stage of development) are important determinants of how persistent the effects of the school closures might be. It also means that we may continue to see declines in test scores, without the right interventions, as younger children who experienced the pandemic closures get older.

Implications for policymakers

Since 2020, there has been $190 billion in federal appropriations sent to school districts around the country to deal with the fallout from the COVID-19 pandemic. That amounts to more than $4,000 per enrolled student in kindergarten through the 12th grade that has been allocated to schools. At least $26 billion has been earmarked specifically to help students recover academically. Providing effective oversight of how this money is being used by states, local governments, and school districts is critically important to the long-run recovery of students.

However, this analysis should also serve as a caution for the lasting consequences of closing schools, which could include chronic absenteeism, making the problem much worse for some students. Students are academically behind and may continue to be without intervention. If this is not corrected, it could shape the future economy and undermine opportunities for today’s younger generations of students.

About The Author