Executive Summary

Due to government orders related to the COVID-19 pandemic, as many as 80 million U.S. students were out of school in 2020. Nationwide, school closures affected nearly all of the nation’s public and private schools. Preschool and child care centers in many communities have been affected with some continuing services to the children of essential workers. Nearly all of the nation’s college campuses closed for the spring term.

With the pandemic expected to endure until 2021, and perhaps longer, permanent school and child care center closures are unsustainable. Policymakers must prepare to make difficult decisions about how to reopen American schools while managing serious public health risks.

The good news is that children are at very low risk of serious illness or death from COVID-19. Indeed, children aged 1–14 are less likely to die of influenza than of COVID-19. Overall, Americans under the age of 25 represent 0.15 percent of all COVID-19 fatalities in the U.S.

Comparing the relative risk of death from COVID-19 vs. influenza or pneumonia. Estimating relative risk is highly dependent upon the ultimate lethality of COVID-19, which is unknown at this time. Nonetheless, a clear pattern emerges from what we know, in which those under aged 25 are at the lowest risk of death from COVID-19, relative to influenza or pneumonia. (Source: National Center for Health Statistics; A. Roy / FREOPP)

In addition, hospitalization rates among school-aged children associated with COVID-19 are sufficiently low to justify the reopening of schools. There have been reports of a significant but extremely rare Kawasaki-like inflammatory syndrome in children, but the syndrome appears to be manageable with medical intervention. Influenza pandemics carry significantly greater risks for children, and yet, the U.S. has not closed schools for influenza pandemics in generations.

There also appears to be very low risk of transmission of COVID-19 from children to adults. As we detail below, population-wide studies in Europe have found little to no evidence of children-to-adult transmission; indeed, children have generally received the virus from adults.

Some people have argued that, out of an abundance of caution, schools should not reopen until we have a vaccine. But as we discussed in an April 2020 FREOPP working paper on reopening the economy, we have no assurance that a vaccine will be available anytime soon. There are considerable technical hurdles to developing a vaccine for a novel virus. The world’s best scientists have been working on a vaccine for HIV for four decades, and have yet to develop one.

While the risks of COVID-19 in children are low and manageable, the harms of prolonged school closures are high. According to the American Academy of Pediatrics,

The importance of in-person learning is well-documented, and there is already evidence of the negative impacts on children because of school closures in the spring of 2020. Lengthy time away from school and associated interruption of supportive services often results in social isolation, making it difficult for schools to identify and address important learning deficits as well as child and adolescent physical or sexual abuse, substance use, depression, and suicidal ideation. This, in turn, places children and adolescents at considerable risk of morbidity and, in some cases, mortality.

Children from lower-income families have fewer opportunities to learn outside of school. Researchers have found that differences in outside of school learning opportunities contribute to the academic achievement gap between rich and poor children. The current situation is likely exacerbating this opportunity gap, particularly since poor children are less likely to have internet access at home.

Widespread school closures have other negative consequences for the nation’s children, and particularly those from low-socioeconomic backgrounds. For example, American schools provide food to more than half of the school aged population. Nearly 30 million children receive free or reduced-price lunch through the National School Lunch Program. While most children will not go hungry without free or subsidized meals, children from the poorest families could be affected by the lack of regular access to these services. Schools and child care centers also play a critical role in state child welfare systems and supporting children’s health.

Reopening schools is far from an outlandish idea. In May, several advanced nations reopened their schools with few problems, including Austria, Denmark, Finland, the Netherlands, Norway, Switzerland, Australia, New Zealand, and Singapore.

Policymakers must anticipate that significant numbers of school personnel and children may remain home based on real or perceived risks, according to national surveys. Four-in-five parents do not feel safe sending their children back to school next month or September. Nearly 40 percent report that they plan to keep their kids home until a vaccine is available, though this view may change if schools reopen with success.

But the likelihood that the pandemic will persist into — and perhaps well beyond — the 2020–21 school year requires policymakers to plan for the reopening of the nation’s schools. This is far from an outlandish idea. In May, several advanced nations reopened their schools with few problems, including Austria, Denmark, Finland, the Netherlands, Norway, Switzerland, Australia, New Zealand, and Singapore.

Summary of key policy recommendations

• The case for reopening schools is compelling. This is especially true for pre-school and primary schools, but it is also true for older children and young adults. The harms from school closures, combined with the limited risks to students themselves and our known experience in Europe, speaks in favor of reopening. However, it will be important for schools to take feasible and workable precautions wherever possible. When American students go back to school, students who live with high-risk individuals, such as the elderly and those with chronic disease, and/or are in regular close contact with high-risk individuals, could remain at home, as should high-risk teachers and staff. In addition, adults who work in schools should take precautions so as to not transmit the virus to each other, as they would do in other workplace setting.
• Schools should follow the American Academy of Pediatrics’ guidelines over the CDC’s. In reopening schools, the American Academy of Pediatrics’ guidelines are more practical than those of the Centers for Disease Control and Prevention, and better aligned with the latest scientific evidence. These guidelines include masks for older children; reasonable physical distancing; and testing and tracing protocols. Policymakers should lean into reopenings for younger children, i.e., non-adolescents, as they have exhibited the lowest risk of both infection and transmission.
• Despite present limitations, it is essential to improve virtual instruction. While virtual learning has proven more workable at the college level, it has been a failure for pre-school and primary school students. This is a big reason why school reopenings are essential. Nonetheless, some students will need to stay home. States should take the lead, not local school districts, in developing virtual schools and curriculums for primary and secondary students who need to stay home. It will be essential for states to realign school funding formulas such to accommodate alternatives to traditional schools.
• Microschools are a worthy alternative to pure virtual instruction, for those who cannot attend traditional schools. Microschools, sometimes called “pods,” are a promising option for students who require limited exposure to the traditional school setting for health reasons. Publicly-funded Education Savings Accounts can help parents provide their children with supplemental instruction where it is most needed. The Supreme Court’s recent decision in Espinoza v. Montana Dept. of Revenue has created much more flexibility for states to pursue this concept. Schools should transition to a mastery-based approach to measuring learning, so that we can compare performance of virtual and microschool models to those of traditional classrooms.
• Protect schools from frivolous lawsuits and relax licensing requirements. Congress should pass liability protection for child care providers and schools that reopen in good faith, requiring a clear standard of negligence or malice for legal liability. Policymakers should relax licensing requirements for child care, tutoring, and distance learning instruction so as to accelerate their utility during the pandemic, and waive teacher certification requirements to fill open positions.
• Avoid a one-size-fits-all approach to college reopenings. Colleges and universities should separate on-campus and off-campus students so as to reduce the risk of community transmission. Approaches to reopening will necessarily vary depending on local community infection rates and the availability of on-campus housing. Colleges should find creative ways to encourage students to self-report symptoms, and install temperature sensors outside of classrooms to immediately identify symptomatic students.
• Carefully target federal assistance. Before contemplating additional federal aid, Congress should ensure that the $30 billion in education funding authorized by the CARES Act is fully deployed. Policymakers may identify targeted needs not already addressed by existing relief.
• Set realistic expectations, and plan for problems. In particular, it will be essential for policymakers and educational leaders to keep an eye out for schools where adherence to public health guidelines is poor, and find solutions. Parents, teachers, and administrators must have realistic expectations that they will see cases of COVID-19 in their schools, just as we see them in the general population, and have plans in place to deal with these expected occurrences.

The American systems of education are complex. Adapting these systems to address national public health concerns and accommodate individual student needs will require a careful and honest analysis of tradeoffs and potential unintended consequences. But the good news is that there are ways to reopen schools that keep children, teachers, and their communities safe.

Policymakers should also enable new flexible solutions, such as allowing groups of students and teachers to form microschools or “pods”— potentially combining virtual learning with traditional teaching in a small, in-person setting.

(Authors’ note: This is a working paper that will be continuously updated with new evidence and policy ideas. Revisions will be noted at the end of the paper. Please check back frequently.)

PART ONE: The Present Situation

The adverse consequences of school closures

The COVID-19 pandemic has caused unprecedented school closures around the world. As of May, an estimated 70 percent of children globally (or more than 1 billion students) have been affected by school closures.

In the United States, 48 states and the District of Columbia recommended school closures for the remainder of the school year as of May, affecting more than 50 million students. As of April, 20 states had closed child care centers except for those serving essential workers. Approximately 1,100 universities and colleges closed their campuses and transitioned to distance learning during the spring semester.

K-12 school systems and postsecondary institutions are currently developing plans for the 2020–21 school year. As national, state, and local policymakers and school leaders consider plans for the 2020–21 school year, growing evidence about COVID-19 and our current understanding of public and individual health risks should inform these decisions.

The nationwide closures of school and child care centers have had significant, negative consequences.

The child care sector

The child care sector has also experienced significant disruption, which may have lasting consequences. According to the Bureau of Labor Statistics, more than 330,000 child care workers have lost their jobs during the pandemic. A Bipartisan Policy Center survey found that 60 percent of child care providers were closed as of April.

A March 2020 survey of child care providers found that “30 percent would not survive a closure of more than two weeks without significant public investment and support,” to pay costs. Lengthy closures may permanently shrink the child care sector, and create new challenges for parents to find and afford care for their children. In April, the Center for American Progress warned that the pandemic may result in a reduction of 4.5 million child care slots.

During the economic lockdown, women have been laid off at a higher rate than men, in part due to their disproportionate participation in the hospitality and education sectors. Women will have a greater challenge re-entering the workforce unless childcare and schools reopen.

Child care already consumes a large share of working parents’ budgets. Working parents living below 200 percent of the poverty line spend approximately 35 percent of their income on child care. A long-term reduction in the supply of child care will increase costs and create new challenges for parents to find adequate care for their children.

During the economic lockdown, women have been laid off at a higher rate than men, in part due to their disproportionate participation in the hospitality and education sectors. Women will have a greater challenge re-entering the workforce unless childcare and schools reopen. Absent suitable child care, millions of mothers will have to stay home, foregoing income in the short term and decreasing their long-term earning potential, which will have a lasting impact on society, but particularly for female-headed and lower-income households.

K-12 education

In terms of K-12 student learning, widespread school closures have created significant disruptions that will have lasting effects for many students, and further undermine equal opportunity in American education.

Across the country, many school systems were unable to quickly transition to provide effective distance or virtual learning programs and instructional models. According to the Center for Reinventing Public Education, only “one in three districts expect teachers to provide instruction, track student engagement, or monitor academic progress for all students,” while schools were closed. A national survey of American parents found that 71 percent believed their children learned less than they would have had schools remained open. That survey also found that 40 percent of students had no individual interactions with teachers.

A national survey of American parents found that 71 percent believed their children learned less than they would have had schools remained open. That survey also found that 40 percent of students had no individual interactions with teachers.

For many students who are not receiving adequate instruction while schools are closed, the academic regression that occurs while students are out of school (often referred to as the “summer learning slide”) will continue so long as schools are closed. Based on available research evidence, the effects of this regression will be most acute for economically-disadvantaged children, and therefore will exacerbate the academic achievement gap. For example, Brown University researchers studied existing evidence about how time out of school affects student learning and projected the potential impacts of school closures during the pandemic. They found:

Under these projections, students are likely to return in fall 2020 with approximately 63–68% of the learning gains in reading relative to a typical school year and with 37–50% of the learning gains in math. However, we estimate that losing ground during the COVID-19 school closures would not be universal, with the top third of students potentially making gains in reading.

McKinsey modeled the effect of different potential pandemic scenarios and projected that low-income children will lose more than a year’s worth of learning if school closures and part-time schedules continue through the fall of 2020. In addition to economically-disadvantaged children, other student populations, including children with special needs and English language learners, suffer from school closures and the lack of specialized instruction outside of school.

Beyond these direct educational effects, widespread closures are having significant impacts on school systems. To date, more than 100 private schools have closed permanently due to the loss of revenue and families inability to afford tuition after the pandemic, according to the Cato Institute. These closures may increase the burdens on traditional public school systems as private school students enroll in public schools. (A coalition of organizations that support choice in education estimated that public schooling costs will increase by $15 billion if 20 percent of private school students enroll in public schools.) Moreover, many states are projecting revenue shortfalls due the pandemic and economic downturn.

Postsecondary education

At the higher education level, there is some evidence that keeping colleges online-only has negative consequences for student learning. A 2013 study on a principles of microeconomics course at a major research university examined the effect of randomly assigning students into in-person versus online sections of the class. While there was no difference in class performance for students with above-median GPAs, those with below-median GPAs who were assigned to the online sections performed significantly worse than those assigned to the in-person sections.

Online education certainly has great promise when designed thoughtfully, but student learning is likely to suffer from the hastily-constructed virtual curricula that colleges have created in the wake of the COVID-19 pandemic. The learning consequences are likely to be particularly negative for students with mediocre academic performance to begin with.

The pandemic is also exacerbating many colleges and universities’ financial challenges. For some colleges, the revenue drop is likely to force a permanent closure. While there are a number of colleges that were already on shaky financial footing before the pandemic, it is possible that the extraordinary circumstances will also force institutions that were otherwise financially healthy to shut down as well. Because far fewer students will be inclined to pay full tuition rates for online-only instruction, tuition revenue is likely to fall if colleges remain online for the fall semester. Auxiliary enterprises such as dormitories and dining halls, which account for a nontrivial 8% of revenue at public four-year institutions, also represent a potential loss of revenue if students are not on campus in the fall.

According to an April 2020 survey of American college and university presidents, 86 percent listed “fall or summer enrollment” and 64 percent identified “long-term financial viability” as the “most pressing issues facing presidents due to COVID-19.” Permanent college and university closures will have serious ramifications for local economies in which colleges and universities are major sources of employment. But if college campuses are open in the fall for in-person instruction, many areas will hopefully avoid the worst-case scenarios of the COVID-19 recession.

Social and public health consequences of school closures

Beyond the immediate educational impacts, ongoing shutdowns of schools also significantly affects the social and public health of American children, and particularly disadvantaged kids. In a recent report recommending school reopenings, the American Academy of Pediatrics placed this issue in stark relief:

Beyond supporting the educational development of children and adolescents, schools play a critical role in addressing racial and social inequity…The importance of in-person learning is well-documented, and there is already evidence of the negative impacts on children because of school closures in the spring of 2020. Lengthy time away from school and associated interruption of supportive services often results in social isolation, making it difficult for schools to identify and address important learning deficits as well as child and adolescent physical or sexual abuse, substance use, depression, and suicidal ideation. This, in turn, places children and adolescents at considerable risk of morbidity and, in some cases, mortality. Beyond the educational impact and social impact of school closures, there has been substantial impact on food security and physical activity for children and families.

Widespread school closures, suspended sports and other extracurricular activities, and shuttered public parks and playgrounds have serious negative physical consequences for children.

Beyond supporting the educational development of children and adolescents, schools play a critical role in addressing racial and social inequity.

Many parents worry that children are spending too much time staring at screens, including as schools have transitioned to distance or virtual learning. Limiting interaction with peers may also negatively impact mental health and jeopardize the safety of some children. Anecdotal reports suggest that some health care providers are seeing an increase in child abuse incidents during the pandmemic. But reports of abuse are reportedly down which raises concerns that children are suffering abuse at home that is not identified at school or by others in the community.

Long-term school closures and social distancing may negatively affect children from poor households more than their peers. A 2015 study linked socioeconomic status with a range of risk factors for childhood obesity, including “less physical activity, poorer nutritional food consumption, more consumption of fried food and sugary beverages, and much more time watching television or playing video games.” Poor children are also at greater risk of child abuse or experiencing other violence. According to the American Psychological Association, “Poverty is a reliable predictor of child abuse and neglect,” and low socioeconomic status children “are more than twice as likely as their higher SES peers to have had three or more adverse experiences” of violence.

That school closures harm children should be relatively obvious. What may be less obvious is their harm to parents, especially single parents, many of whom lack alternative forms of child care and have effectively been forced to choose between their jobs and their children’s safety.

What we know about COVID-19 in school-aged children

The epidemiology of COVID-19, in terms of those at risk for serious illness and death, reveals a clear pattern around the world, as noted above. For children aged 5–14, the risk of dying from COVID-19 is approximately 1 in 2.5 million. For children under the age of 5 the risk is even lower. Americans younger than 25 represent a mere 0.15 percent of all U.S. COVID-19 deaths.

The risk of death from COVID-19 is far higher in the elderly than in younger Americans. According to data from the Centers for Disease Control and Prevention, as of July 29, 2020, those older than 65 are 25 times as likely to die of COVID-19 than those aged 25 to 54. Note that not all COVID-19 deaths reported elsewhere are counted by CDC, and that not all CDC-counted deaths were caused by COVID-19; some of these individuals died from other causes, but tested positive for SARS-CoV-2 antibodies. If we assume that there will be 200,000 total deaths from COVID-19 in the U.S., the odds of an individual under 25 dying from COVID-19 are around 2.5 per million, or 1 in 385,000. (Graphic: A. Roy / FREOPP)

Transmission of SARS-CoV-2 from children to adults is rare

Importantly, the evidence to date suggests that transmission of SARS-CoV-2 from children to adults is also rare. In a recent summary of international experience, Boston University epidemiologist Benjamin Linas put it succinctly: “Will someone in America contract COVID-19 from their sick child? Yes. Should I structure my life around such a rare occurrence? I do not think so.” Some researchers have speculated that children could readily transmit the virus to adults, but in the real world, it has been highly uncommon.

Importantly, the evidence to date suggests that transmission of SARS-CoV-2 from children to adults is rare.

In Iceland, a group of researchers conducted a population-wide survey of SARS-CoV-2 infection that was published in June in the New England Journal of Medicine. Kari Stefansson, one of the leaders of the group, said of the study, “Children under 10 are less likely to get infected than adults, and if they get infected, they are less likely to get seriously ill. What is interesting is that even if children do get infected, they are less likely to transmit the disease to other adults. We have not found a single instance of a child infecting parents.” (Emphasis added.)

Similarly, a wide-ranging study of COVID-19 by the Dutch National Institute for Public Health and the Environment found that, “within families, the virus is mainly transmitted from adults to children. Children are sometimes infected with the virus, but less often than adults. There were no families in which a child under 12 years old was the first patient in the family.” (Emphasis added.) In other words, children were not the ones passing SARS-CoV-2 to adults; but adults were the ones passing it on to children.

Some European countries have taken more effective actions than the US to contain COVID-19 and drive down new cases. In Europe, reopening schools has not led to a rise in COVID-19 cases. Denmark reopened schools on April 15, the first Western country to do so. But “you cannot see any negative effects from the reopening of schools,” said Peter Andersen of the Danish Serum Institute. Norway reopened schools on April 20, and cases there continue to trend downward. Austria reopened schools on May 18, and has not seen an increase in infections. Similar trends hold in Germany and Finland. A study of 541 students in northern France found zero cases of an infection transmitting from student to teacher.

Sweden has seen more population-adjusted deaths from COVID-19 than the U.S., and yet even in Sweden, opening schools did not lead to a higher infection rate among schoolchildren relative to Finland, which had a much lower overall death rate from COVID-19, according to a study jointly published by the two countries’ public health agencies.

An exception to this trend occurred in Israel, where an outbreak occurred at a high school in Jerusalem, infecting over 100 students and 15 staff members. Notably, teachers at the school did not wear masks and follow other Israeli Health Ministry guidelines, nor did students, and the school did not aggressively quarantine infected individuals. This experience highlights the value of ensuring that if schools reopen, they do so with proper protocols in place to help minimize the likelihood of transmission between students.

All in all, as two researchers put it in the journal Pediatrics,

Almost 6 months into the pandemic, accumulating evidence and collective experience argue that children, particularly school-aged children, are far less important drivers of SARS-CoV-2 transmission than adults. Therefore, serious consideration should be paid toward strategies that allow schools to remain open, even during periods of COVID-19 spread. In doing so, we could minimize the potentially profound adverse social, developmental, and health costs that our children will continue to suffer until an effective treatment or vaccine can be developed and distributed or, failing that, until we reach herd immunity.

A more meaningful risk: Adult-to-adult transmission in schools

The relatively low risk of transmission from children to adults does not mean that schools should take no precautions. Indeed, the possibility of transmission from adults to other adults in schools remains. Younger teachers, especially those under 45, are at lower risk of serious illness and death. But older teachers and those with certain underlying cardiovascular, metabolic, or autoimmune conditions will need to take extra precautions, and in some cases stay home, in order to protect themselves.

The Israeli counterexample highlights another risk: that even if policymakers develop effective protocols and guidelines, the adherence to those guidelines can be inconsistent. We have seen this phenomenon already in the U.S., with respect to the controversy over wearing masks in enclosed public spaces.

PART TWO: Policy Recommendations

The importance of improving virtual classrooms and other outside-of-school learning options

While the case for reopening schools is compelling, some students and teachers and school staff do bear higher risk due to underlying health conditions and/or residential status, such as living with elderly family members. Moreover, many parents may not feel safe to send their kids back to school even if current evidence suggests there is a low-risk. As state and local policymakers work to reopen schools, they should also work to provide distance learning options for those students who are unable to physically attend school.

This will prove difficult. The spring 2020 experience with virtual learning, especially in the K-12 setting, has not been encouraging. In Boston, one in five public school students became virtual dropouts; i.e., they did not log into class when schools moved to all-virtual learning. There are wide disparities in children’s access to computers and broadband internet, and even those with sufficient access will face decrements in learning gains of up to 50 percent due to closures in the 2019–20 school year.

The spring 2020 experience with virtual learning, especially in the K-12 setting, has not been encouraging.

But it will remain necessary to improve the virtual learning environment and other outside-of-school options. Even when schools are reopened, many parents may choose not to send their children. According to a USA Today poll reported in May, 60 percent of parents with a child in K-12 school found that “6 in 10 say they would be very likely to pursue at-home learning options instead of sending their children back this fall,” and “nearly a third of parents, 30%, say they are “very likely” to do that.” A similar May 2020 poll published by the American Federation for Children, which supports education choice, found that 40 percent of parents planned to homeschool or use virtual schooling after school lockdowns ended. In June the American Enterprise Institute reported that “only one-fifth of parents feel safe sending their children back to school in August or September,” and lower-income families felt less safe than higher-income parents.

School employees may also reconsider their plans for the upcoming school year based on concerns about COVID-19. USA Today’s recent poll found that one in five teachers say they are unlikely to return to the classroom if schools reopen this year. At that rate, more than 600,000 public school teachers would be leaving their positions within months.

Teachers’ and parents’ perspectives about the individual and public health risks of COVID-19 may change as the public’s understanding of the available evidence changes, but policymakers must anticipate the potential for significant changes during the upcoming school year. For example, according to the National Center for Education Statistics, 3.3 percent of American K-12 students were homeschooled as of 2016. A potential tenfold increase in the homeschooling population would have significant implications for the K-12 education system, including sharp declines in traditional in-person school attendance.

Improving the quality and availability of virtual instruction or other flexible outside-of-school learning options will require an overhaul of how we think about public education.

• Realign school funding formulas to support virtual learning and outside-of-school learning options. State and district school funding formulas and compulsory education requirements are generally not designed to accommodate virtual learning or other outside-of-school learning options, such as microschooling and tutoring. States and localities will need to ensure that an appropriate portion of their education budgets is directed to virtual learning, and that virtual learning counts toward mandatory education requirements. This will require emergency changes to school funding formulas. For example, many school systems use “average daily attendance” as a mechanism for distributing funds to public schools. During the pandemic, school systems should transition to a student-centered funding model. Funds should follow children to the schools or learning settings of their personal circumstance to the extent possible. Moreover, consistent with decades of federal law and policy, state and local education programs should provide sufficient resources to children from low socioeconomic backgrounds and those with special needs. Implementing these changes will require sacrifices and overcoming significant challenges, such as teacher labor agreements and bureaucratic resistance. However, the unprecedented challenge of the pandemic and national shutdown of American schooling in the spring of 2020 demonstrates the urgent need for change.
• Close the digital divide for at home learning. States and school systems should take steps to close the digital divide in at home information technology by supporting access to hardware and reliable internet to the extent practicable. Many families live in areas without access to high-speed Internet and mobile internet including cellular devices may be the most practical option. States and school systems should invest resources to ensure that all students, particularly children living in low socioeconomic households, have access to necessary technology to learn at home.
• Establish statewide virtual schools. States should quickly establish options for full-time virtual schooling for children at home. The Florida Virtual School should serve as a model for such programs, and states that currently do not offer virtual education should establish virtual schools or allow parents to enroll their children in virtual learning programs. Importantly, the state education agency, local school districts, and school should implement programs to ensure that children have access to adequate information technology at home to facilitate distance learning. Teachers should make plans to integrate classroom instruction with distance learning and to transition between these two modes as required.
• Build safe alternatives to home learning for virtual students who need them. Some children may not find it feasible to work virtually, such as those with challenging domestic circumstances or those without high-speed internet access. States and school districts should consider creating such spaces where they can be of most use.
• Create emergency education savings accounts to support at-home learning. Several states have implemented publicly-funded education savings accounts (ESAs) that allow parents to control public funds that can be used on certain educated related expenses for their children. According to The Policy Circle: “As of 2020, over 20,000 K-12 students in the US were recipients of ESAs. Six states have ESA programs. As of 2018, 25 states have proposed legislation with the hopes of broadening educational offerings in their state.” ESAs allow parents to customize their education — including by purchasing instruction or educational services from a range of providers. For children who must learn at home during the pandemic, ESAs offer a practical option to allow a family to personalize their learning program and continue their education. (Federal and state policymakers could use 529 saving programs, which allow tax-free saving for certain K-12, postsecondary, and job training expenses through state-managed investments, as an existing vehicle for distributing aid to families while maintaining proper oversight to ensure that funds are spent appropriately.)

For those who need it, one practical option for providing out-of-home schooling is microschooling. Microschools provide the child care that many families require to work outside of the home, in addition to in-person instruction.

• Enable the establishment of microschools. For those who need it, one practical option for providing out-of-home schooling is microschooling, sometimes called “pods.” For example, a teacher could provide a microschool setting for a group of children to limit exposure to the traditional school setting. At a microschool, a child could learn in a variety of ways including through teacher instruction, virtual learning, or hybrid-homeschooling. Importantly, microschools would provide the child care that many families may require to work outside of the home in addition to in-person instruction. Prenda Schools are one prominent example of a network of microschools across the United States, and recently reported growing interest in their program. Other innovators in this field include Acton Academy, SchoolHouse, Selected for Families, and Learning Pods.
• Relax licensing requirements for tutoring, distance learning instruction. Enabling widespread homeschooling and distance learning with public support during the pandemic will require significant flexibility. Relaxing licensing rules to allow teachers to provide tutoring and distance learning services would benefit students. For example, the many school teachers who may choose to resign instead of returning to the classroom for the 2020–21 school year could find new arrangements as tutors, distance learning coaches, and even microschool providers if policies are reformed to offer this flexibility.
• Provide additional services to children who must remain at home. Schools and school districts should ensure that core services, such as the National School Lunch Program, remain available to children from the poorest families who are unable to attend school in person. For example, policymakers could develop or expand flexible approaches to providing food subsidies to the most disadvantaged children who must remain at home, such as the Brookings Institution’s recommendations for electronic benefit transfer programs to assist affected children during the pandemic.
• Consider blended models of virtual and in-person learning. Some school districts, for example in California and New York, have asked families to opt into a blended system, in which students will experience a combination of virtual and in-person learning in any given week. For example, in one proposed system, one cohort of students will attend school in person on Mondays and Tuesdays, and learn virtually for the balance of the week. Another cohort will start the week virtually and end the week in-person on Thursdays and Fridays. Wednesdays will be a “cleaning” day where the campus is closed, so that classrooms and facilities may be fully cleaned and reset for the next cohort of students. This method allows in-person class sizes to remain small to permit social distancing in the classroom, and ensures that cohorts can be kept together to minimize interactions between groups of students. This blended model can be used to introduce some in-person learning, especially in communities where rates of viral transmission remain high.

The CDC’s recommendations are outdated

In May 2020, the Centers for Disease Control and Prevention published its long-awaited guidance for reopening workplaces and schools. Unfortunately, that guidance (which is now months old) does not derive from the best and most recent scientific evidence regarding pediatric health and SARS-CoV-2 transmission, and makes contradictory demands upon schools that will prove impossible to implement, especially in resource-constrained communities.

In general, the CDC guidelines ignore the fact that risks to children from COVID-19 are far lower than they are for older adults. For example, the CDC recommends that schools consider closing playgrounds, despite the fact that SARS-CoV-2 transmission is an especially low risk outdoors. Indeed, giving children time outdoors or even holding classes outdoors would likely reduce their already low risk of transmission from classmates. In addition, as the CDC acknowledges elsewhere, transmission by touching a surface, such as a play structure, that an infected person has also touched “is not thought to be the main way the virus spreads.” The American Academy of Pediatrics recommends, in contrast to the CDC, that pre-K and elementary schools should “utilize outdoor spaces when possible…reducing classmate interactions [and] play…may not provide enough COVID-19 risk reduction to justify potential harms.”

As the AAP notes, the CDC “recommends that schools ‘space seating/desks at least 6 feet apart when feasible.’ In many school settings, 6 feet between students is not feasible without limiting the number of students. Evidence suggests that spacing as close as 3 feet may approach the benefits of 6 feet of space, particularly if students are wearing face coverings and are asymptomatic.”

The difference between 3 feet and 6 feet — for a school — is critical, as it may mean the difference between having enough physical space to educate most students, or only some. But for this to be wise, it will be important for older children to wear masks, have effective public health systems to test symptomatic people, and isolate positive cases Ideally, we would have the testing capacity to also regularly survey the asymptomatic population, but this capability does not yet exist.

Any school attempting to adhere to all of the CDC’s guidelines will find it impossible to keep their doors open to a meaningful proportion of students. Fortunately, the American Academy of Pediatrics has developed a more coherent and pragmatic set of recommendations.

After the CDC received considerable pushback from schools who explained why CDC guidelines would be difficult to implement, CDC director Robert Redfield said, “I want to make it very clear that [it] is not the intent of CDC’s guidelines to be used as a rationale to keep schools closed.” Intent or not, it is simply a fact that any school attempting to adhere to all of the CDC’s guidelines will find it impossible to keep their doors open to a meaningful proportion of students.

Fortunately, the American Academy of Pediatrics has developed a more coherent and pragmatic set of recommendations, highlighting both the “need to account for…vulnerable populations, including those who are medically fragile, live in poverty, have developmental challenges, or have special health care needs or disabilities,” while reopening school for those who most benefit from it. Our own recommendations align with the AAP’s, and expand upon them.

Reopening preschools and child care centers

Prior to the pandemic, roughly one-in-three children under four were enrolled in an organized child care program. Among married couples with children under age 6, 60 percent had both parents working. Reopening the economy will require reopening and expanding child care options for working parents.

Policymakers should relax child care regulations to increase supply during the pandemic. Waiving certain licensing requirements is a commonsense way to increase the supply of available child care spots.

To ensure that critical child care services are available, policymakers should implement a series of reforms to give providers and parents greater flexibility to find adequate care arrangements during the pandemic. This will include easing certain regulatory requirements for providers, facilitating home-based daycare programs, and giving families greater choice to use federal- or state-government resources for child care by hiring the provider of choice. In addition, policymakers should provide liability protections for preschool and child care providers or risk a severe contraction of supply:

• Relax child care regulations. Policymakers should relax child care regulations to increase supply during the pandemic. For example, state governments in Connecticut, New York, Tennessee, Hawaii and Nebraska included measures to relax child care regulations as a part of their emergency COVID-19 response orders. Waiving certain licensing requirements is a commonsense way to increase the supply of available child care spots. Moreover, there is good evidence that easing regulations will not undermine child care quality. (The Mercatus Center at George Mason university reviewed the impact of a variety of child care regulations and found that child-staff ratios and group size limits did not affect the quality of care provided.)
• Expand home-based and microschool child care options. One promising strategy to maintain child care services for parents who need care is to increase the availability of home-based care in informal settings. Home-based providers could provide care to a smaller number of children and, therefore, may be better positioned to follow pandemic guidelines. But government rules can prevent home-based care. For example, common regulations include the elimination of all tripping hazards and chipped paint, and extreme sanitation standards, that are impractical for many home-based child care providers. Policymakers should ease rules to allow parents, friends, and neighbors to find workable solutions for their children.
• Increase parents’ choices for how public child care benefits are used. Policymakers should give parents greater flexibility for how government funding and support for child care is used, including by placing children in home-based care or microschool settings. Giving parents new options could increase the hours of care provided than some traditional child care centers, particularly during the pandemic if many centers remain closed. (For example, FREOPP compared the hours of care provided by the federal Head Start program with state-based and private preschool or child care programs. In 37 states, the per-child cost of the Head Start program is more than the average cost of full-time child care for a 4-year-old, while Head Start providers are only required to provide 448 hours of care per year. Letting parents use their child’s share of Head Start funding to purchase an alternative provider could double the hours of care provided and increase parents’ earnings by 20 percent or more.)
• Provide liability protections for child care providers. Policymakers must also address questions of legal liability during the pandemic which could contract the supply of child care providers. Liability insurance for accidents already poses a barrier to entry for providers. The prospect of parents suing if a child or loved one contracts the coronavirus could significantly increase these costs and cause many centers to remain closed and prevent new providers from opening.

By expanding parents’ options for small-group child care, we can do more to get parents back to work while protecting public health. Many of the changes we propose were worth enacting in the pre-pandemic environment. Now, they are urgently needed to reopen the economy and provide children with the care they need.

Reopening K-12 schools for in-person learning

• Lean into school reopenings for younger children. The risks of infection and transmission are low for all school-aged children, but they are especially low for elementary school students. Schools and policymakers should understand this, and recognize the value in reopening schools on time especially for younger children.
• Modify the school calendar to make up for lost time. State and local officials and school leaders should consider options to adjust the school calendar to account for lost school days and hours during the 2019–20 school year. The average public school calendar in the United States includes 180 days with 6.64 hours of school per day. Many states and local governments recommended school closures in March which persisted through June, which includes roughly three months of lost school days (or nearly a third of the school year) for many schools. Moving forward, state and local officials could reopen schools as early as feasible, reduce scheduled vacation days during the upcoming school year, and plan for an extended school year next summer.

Schools and policymakers should recognize the value in reopening schools on time, especially for younger children.

• Implement school-level public health protocols. Both the AAP and CDC guidelines for reopening schools include common-sense recommendations for how schools can minimize public health risks. These recommendations include promoting better hygiene (including washing hands), enhancing cleaning protocols, and providing access to personal protective equipment and face coverings for school personnel and students and encouraging their use (mandatory for staff, high school students, and middle school students). States and local school districts should also study and identify best practices for preventing outbreaks in the school, such as by implementing temperature checks and school entrances and prepare for testing and tracing if necessary. While SARS-CoV-2 can spread in asymptomatic individuals, temperature checks have value in that thermometers are readily available for daily use and provide instantaneous results. States should consider creating exemptions or waivers from so-called “seat time” requirements for time taken to implement these safety and public health protocols.
• Waive teacher certification requirements to fill open teaching positions. Anticipating potential widespread teacher resignations when schools reopen, states should relax teacher certification requirements for public school teachers to give school leaders flexibility to hire replacement teachers. Teacher quality is a significant factor affecting student achievement; however, licensing is not an effective predictor of teacher quality or effectiveness. New hiring flexibility will allow school leaders to fill teacher vacancies under the current challenging circumstances. States could also explore creative partnerships with schools of education and alternative teacher certification programs to expedite teacher training and to bring new teachers into the classroom by fall.
• Support the private school sector. Widespread closures of private schools due to declining enrollment and parents’ inability to pay tuition during the current economic downturn has the potential to exacerbate public schools’ challenges in the fall. The availability of private schools increases equal opportunity, particularly when public support is provided to help low-income parents choose their children’s school. Through the federal Title I program and many state special education programs, private schools already receive public funding. In its recent in Espinoza v. Montana Dept of Revenue, the U.S. Supreme Court ruled that state Blaine Amendments that prohibited public funds from being used at private schools were unconstitutional. This decision paves the way for new programs that allow parents to use public funds to choose private and religious schools for their children. Using public funding to sustain the private school sector during the pandemic and related economic downturn will prevent overcrowding and related challenges in the public school system, particularly when there are public health reasons to reduce class sizes, and when public schools may face teacher shortages.
• Provide remedial instruction, tutoring and other specialized support to at-risk students. State policymakers, school districts, and school leaders should anticipate significant learning loss for many students, due to the ongoing school closures. Economically-disadvantaged and children with specific needs, including special education students and English language learners, may be most affected. Providing these students with the support they need must be a priority. For example, state policymakers could develop and implement programs similar to Florida’s “Reading Scholarship Accounts,” which provide $500 in funding to an account to purchase reading instructional materials and tutoring services. States, school districts, and schools should implement a range of immediate and long-term strategies (including planned summer school offerings in 2021) to address pandemic-related learning loss and the effect of widespread social promotion in 2020.
• Transition to a mastery-based approach to measuring student learning and reconsidering social promotion. Providing personalized instruction to all students affected by the 2020 school closures should encourage states, districts, and schools to transition to mastery-based approaches to measuring students’ achievement and progress, particularly as students recover learning lost during the pandemic at their own pace. Moreover, states, school districts, and schools should reconsider their approach to social promotion to ensure that children at risk of falling further behind during the pandemic master necessary skills to succeed in subsequent grades.

Colleges, universities, and other postsecondary institutions

As of September 4, 2,248 colleges and universities had announced their plans for the fall semester. The plurality of those (44%) are planning for a fully or primarily online semester, according to the Chronicle of Higher Education, while 28% will hold classes fully or primarily in person. The remainder are pursuing a hybrid approach or considering alternative models.

Brown University President Christina Paxson has written that “the reopening of college and university campuses in the fall should be a national priority.” Purdue University President Mitch Daniels argues that “failure to take on the job of reopening [college campuses] would be not only anti-scientific but also an unacceptable breach of duty.”

The challenge of housing college students during a pandemic

Students at colleges with large residential student populations tend to skew younger; 78% of undergraduates at private nonprofit four-year universities are under the age of 25. Among this age group, the risk of hospitalization from COVID-19 is quite low.

However, there is reason to be concerned that while college students themselves are at low risk of serious illness or death from COVID-19, they are able to increase the spread of the coronavirus to their relatives and other off-campus acquaintances.

On residential campuses, many institutions have proposed limiting the number of students in dormitories in order to reduce the potential spread of COVID-19. Several universities are planning to limit dormitory capacity to 50% or lower. While this may help to limit viral transmission on campus, it also means that many students will need to study from off-campus locations. And while elite colleges often have enough on-campus housing for all students, this is actually quite rare for non-elite institutions. At the typical public four-year institutions, for example, only 25 percent of students live on-campus.

One approach being tried by certain elite universities is to reduce the number of students on campus by class year. For example, Yale is only allowing freshmen, juniors, and seniors on campus in the fall, and only sophomores, juniors, and seniors on campus in the spring. Stanford will allow only freshmen, sophomores, and new transfer students on campus for the fall and summer quarters, and only juniors and seniors on campus for the winter and spring quarters. This on-off approach is suboptimal, because it will deprive students of more fulsome opportunities to interact with their classmates in the classroom and on campus, but it does minimize intra-cohort discrimination.

As noted above, this option is not available to institutions without significant on-campus housing capacity, such as public community colleges. Two in five community college students live at home with their parents, making the risk of inadvertent virus transmission from campuses to at-risk individuals higher at these schools. Community colleges also have more students over the age of 35 (14%, versus 6% at public four-year colleges).

Older students, along with younger students who have preexisting health conditions, may be unable to return to campuses under any circumstances while the pandemic is ongoing.

The good news is that we have far more experience with effective virtual instruction at the collegiate level, relative to primary and secondary schools.

This combination of factors — the need to reduce the proportion of students on campus, and the need to create learning environments for at-risk individuals and staff — requires colleges and universities to continue allowing students to complete their coursework from home.

The good news is that we have far more experience with effective virtual instruction at the collegiate level, relative to primary and secondary schools, through organizations like Coursera, edX, and Western Governors University. Here are some ways for traditional colleges to benefit from that experience:

• Invest in online instruction. Even as in-person instruction resumes, colleges will have to develop instructional programming to serve students who are at-risk themselves, or who have no choice but to live at home with at-risk relatives. It will become imperative for institutions to invest in developing quality online courses, along with HyFlex models that allow students to switch back and forth between virtual and in-person instruction. While developing these courses will add additional costs, institutions should look at it as a long-term investment that will pay off even after the COVID-19 pandemic has subsided. As of Fall 2018, 35% of undergraduates and 40% of graduate students were taking at least one of their courses online, and those numbers are likely to grow over the coming years.
• Form regional consortia to share courses and best practices. Currently, institutions are at varying points of development in their online courses. Some public university systems, such as the University of Maryland and Arizona State University, have well-developed online programs that their in-person branches should take advantage of. Other state university systems lag far behind. Universities should consider forming regional consortia, perhaps based around athletic conferences, to allow students attending one university to enroll in online courses at another. Colleges should relax any restrictions that apply to transfer credits from other universities in order to give students as much flexibility as possible.
• Promote alternative methods of credit accumulation. Some students, particularly those at the beginnings of their college careers, may find it worthwhile to earn college credits through AP or CLEP exams. Not only are these exams far cheaper than typical college tuition, but students can study for them at home, making them a good option to maintain progress toward their degrees for students who are forced to remain at home during the fall semester. To facilitate this, colleges should temporarily waive any restrictions regarding the maximum number of AP/CLEP credits that students may apply toward their degrees.

For students who are able to take part in on-campus instruction, post-secondary institutions will need to build on the “test, trace, isolate” paradigm by protecting older students and staff, and finding creative ways to encourage students to comply with public health guidelines.

• Promote physical distancing on campus. Several universities have floated ideas to ensure that physical distancing is possible on campus. These include limiting class sizes, holding the largest classes online, and limiting large social gatherings. “We will forgo the concerts, convocations, and social occasions that ordinarily enliven campus life,” says Purdue’s Mitch Daniels. “It will be a quieter fall without fraternity parties, but first things first.” Several universities plan to end the fall term before the Thanksgiving holiday, which will keep students off campus during the peak of the winter flu season and minimize travel back and forth between campuses and homes.
• Test frequently, and isolate infected people. Colleges and universities should do everything possible to regularly test their communities and isolate infected individuals. The ideal approach would be to test all students once each at: the beginning of each semester; the end of each semester; and, for schools where the vast majority of students are on-campus (see below), during each departure and return from campus. On these campuses, if someone does leave, he or she should be re-tested and confirmed negative before re-entering. If resources prove insufficient to test all students on campus, institutions should prioritize students from places with a high degree of community spread, along with those displaying COVID-19 symptoms. Where possible, colleges should designate special residential facilities to house newly infected people and their contacts.

Many college students will be reluctant to come forward with symptoms, if they feel self-reporting will deprive them of class time and social opportunities.

• Reward students for adherence. One key challenge to infection control on college campuses will be rewarding student participation, rather than punishing the infected. Many college students will be reluctant to come forward with symptoms, if they feel self-reporting will deprive them of class time and social opportunities. Sherry Pagoto and Laurie Groshon of the University of Connecticut conducted focus groups of college students to research their attitudes toward on-campus public health measures, and found that students were skeptical that such measures were enforceable. The students suggested that schools offer prizes, such as food or clothing with college insignia, as a reward for self-reporting.
• Deploy contact tracing and all available diagnostic technologies, including temperature sensors. Colleges can also encourage students to make use of privacy-preserving contact-tracing apps to make contact tracing more efficient. Governments can help institutions cope by subsidizing testing equipment, thermometers, and masks. While thermometers cannot detect asymptomatic COVID-19 infections, they have two important attributes. First, not all students may self-report symptoms, as noted above; placing laser-guided thermometers at the entrances to classrooms and dormitories as a requirement for attendance can identify these cases. Second, unlike serologic or PCR-based testing, laser-guided temperature sensors are widely accessible today, and therefore available for daily use. After the 2003 SARS-CoV-1 crisis in Asia, most Asian countries began widely deploying such temperature sensors, and they are likely to have had a role in the Pacific Rim’s relative success at early detection of COVID-19.
• Protect faculty and staff. While most college students are younger and thus at far lower risk of severe symptoms from COVID-19, 37% of tenured or tenure-track college professors are over the age of 55. Extra steps will be required to protect these individuals, who are at greater risk of hospitalization or death if they contract the virus. Institutions should consider temporarily replacing older faculty members with younger faculty, adjunct professors, or graduate instructors for the coming year, particularly for larger courses.

The ongoing process of reopening college campuses

We recommend that colleges test all students at the beginning of the semester, but not all institutions have implemented universal testing. The University of Illinois at Urbana-Champaign is a leader in testing, conducting over 10,000 tests per day on average and over 218,000 since the beginning of July. This is equivalent to three times the university’s student body. Illinois resumed in-person classes on August 24 and requires every student and employee to be tested for COVID-19 at least twice per week. President Timothy Killeen predicts that “we will face outbreaks, there’s no question. What we’re going to do is have actionable tests so that we quickly squelch down the outbreak.”

Many universities have been able to increase the number of COVID-19 tests performed by implementing pre-arrival testing. Pennsylvania State University shipped free testing kits to students coming from areas with ongoing COVID-19 outbreaks; these students were not allowed to return to campus until they tested negative. As of September 3, pre-arrival testing accounted for approximately 21,000 of the 31,000 student COVID-19 tests Penn State has completed.

In addition to mass pre-arrival testing, some universities have implemented more out-of-the-box policies to control the spread of COVID-19. In assigning roommates in dormitories, Purdue University will pair students who have already recovered from COVID-19 (and are therefore unlikely to contract it again) with students who have preexisting health conditions that may put them at greater risk of complications if they contract the disease. The University of Arizona ran tests for COVID-19 on wastewater from twenty campus buildings, which identified a potential outbreak in one of the dormitories. This allowed the university to isolate two asymptomatic students before they spread the disease to others.

Not all universities have been able to follow through on their plans. The University of North Carolina at Chapel Hill, which had originally planned to hold classes in-person, shifted all undergraduate classes online less than a week after the semester began. In the first week of classes, the university reported a test positivity rate of 13.6%. Lack of widespread testing was almost certainly a factor in the outbreak; when the university announced it would shift to online instruction, it had performed fewer than 3,000 tests on a student body of over 30,000.

Sudden changes in direction may become more common as the semester progresses. New York Governor Andrew Cuomo has issued a mandate to shut down college campuses and move classes online if a school reports either 100 individuals or more than 5% of the student body as infected with COVID-19 over a two-week period. As of writing, one school, the State University of New York at Oneonta, had to shift online after a campus outbreak. Outside New York, institutions such as Temple University and the University of Minnesota have postponed the start of in-person instruction by two weeks.

At least one school, Unity College in Maine, has decided to move permanently to a partially-online hybrid model due to the COVID-19 pandemic.

Deploying unspent federal assistance

Decisions about how and when to reopen American schools are largely the responsibility of state governments, local school districts, and communities. Federal policymakers should play a limited role in guiding state and local education policy and supporting reopenings by providing funding and offering regulatory relief, not dictating how schools reopen.

In March, Congress provided more than $30 billion in education stabilization funding in the CARES Act, including $3 billion in aid to be distributed by governors, $13 billion to state education agencies to be distributed through existing federal elementary and secondary grant programs, and $14 billion for a higher education emergency relief fund.

The Government Accountability Office recently reported that only $83 million of the $17 billion education stabilization fund appropriated by the CARES Act for K-12 education had been spent as of May 31, or less than 1 percent of authorized funds.

The CARES Act also authorized the Secretary of Education to waive certain federal regulatory requirements during the pandemic. The law also requires states and districts to pay employees to the extent possible during school closures and disruptions. The CARES Act also provided more than $300 million for competitive grant programs to encourage new K-12 and workforce training models in response to the pandemic.

Congress is currently considering new spending packages for education. However, the Government Accountability Office recently reported that only $83 million of the $17 billion education stabilization fund appropriated by the CARES Act for K-12 education had been spent as of May 31, or less than 1 percent of authorized funds.

Federal, state, and local policymakers should prioritize spending currently available resources to support school reopenings (including purchasing necessary personal protective equipment and other assets necessary to protect public health) while also ensuring that the educational needs of disadvantaged children and students with special needs are met if schools are closed or they must remain home. Any additional federal support should prioritize districts that genuinely intend to reopen after acquiring protective equipment and developing health protocols.

Federal policymakers can play a constructive role by providing regulatory flexibility to state governments and school districts. The CARES Act already provides the Secretary of Education broad discretion to waive certain regulations.

Future federal legislation could authorize the state education agencies and school districts to provide Title I and IDEA funding assistance directly to parents of eligible children who remain at home to support at-home learning, tutoring, and other services broadly consistent with federal law. Similarly, regulations governing federal preschool and child care programs could similarly be relaxed during the pandemic to allow funds to be provided directly to eligible parents during the pandemic if centers remain closed to access needed services.

Finally, any additional funds that Congress appropriates to aid college students should be distributed directly to students rather than funneled through institutions, which have too often been excessively slow in disbursing funds allocated under the CARES Act.

Setting realistic expectations — and planning for them

As the outbreak in an Israeli high school shows, we cannot assume that all schools and all students will follow public health protocols with alacrity. Many will understandably feel that wearing masks all day is burdensome, and others will be afraid to report symptoms out of fear of being sent home. It will be important for states to closely monitor adherence to these protocols, identify problems, and provide enough flexibility for districts to experiment with ways to improve adherence.

Equally important, the ability to operate schools safely depends on the infection rate of the surrounding community. Communities that are effective at driving down infection rates will be far more successful at opening and operating safer schools. The importance of community-wide efforts to control spread and achieve very low rates of community transmission are as important as efforts within schools to reduce and prevent transmission.

It is also essential to understand that as schools reopen, some children will test positive for COVID-19, for the same reasons that they test positive in the general population. School administrators and policymakers should fully expect that some of their students will test positive. As the AAP notes in its recommendations, “it is critically important to develop strategies that can be revised and adapted depending on the level of viral transmission in the school and throughout the community and done with close communication with state and/or local public health authorities and recognizing the differences between school districts, including urban, suburban, and rural districts.”

Every school should do its best to conduct temperature and symptom screening of every child, teacher, and staffer on a daily basis, in order to identify potential positive cases of COVID-19. Again, citing the AAP, “School policies regarding temperature screening and temperature checks must balance the practicality of performing these screening procedures for large numbers of students and staff with the information known about how children manifest COVID-19 infection, the risk of transmission in schools, and the possible lost instructional time to conduct the screenings…Parents should be instructed to keep their child at home if they are ill. Any student or staff member with a fever of 100.4 degrees or greater or symptoms of possible COVID-19 virus infection should not be present in school.”

Teachers and staff should wear masks, as well as older children, such as those in high school and middle school, with accommodations for those with health needs, the deaf, and other special populations. School health staff, in particular, should be trained in the proper wearing of masks, and be able to offer that instruction to students and teachers.

Schools will need to have a clear plan for screening students, teachers, and staff, and a clear plan for sending individuals home who screen positively. Parents should conduct temperature tests each day at home, in order to identify symptomatic children and keep them at home.

Nonetheless, because SARS-CoV-2 can spread in asymptomatic individuals, in a country with 80 million students, we must expect a manageable number of COVID-19 cases. It will be important to take those cases seriously, and prepare for their emergence, while understanding them in the context of overall risk.

The experience of other countries shows us that we can reopen schools while promoting public health.

That context, above all, includes the well-established harms of keeping schools closed: that our most economically and socially fragile populations fall further behind their peers in educational and economic attainment. As an international group of researchers wrote recently in Nature, “Many actions, such as closing schools…impose large and visible costs on society, but their benefits [to reducing spread of infection] cannot be directly observed.”

The experience of other countries shows us that we can reopen schools while promoting public health. America has the potential to do this well, if we take our responsibilities seriously. We must commit to driving down local infection rates and investing in the policies outlined above: planning well, managing thoughtfully, and implementing successfully the reopening of our schools.

America should strive not merely to do as well as other countries, but to exceed them: by deploying our talent for innovation, and developing new tools for educating children at home and — especially — at school.