On March 16, the Imperial College London published a coronavirus study that had an enormous impact on officials in the United States and UK. Disseminated widely and rapidly by the media, its predictions about the course of the outbreak alarmed officials into ratcheting up both voluntary and involuntary restrictions on daily activities to slow the pandemic, while also shutting down much of the economy. Its conclusions about the future were especially dire. Government officials in the United States were already beginning to move in the direction of more aggressive disease containment. The day before its publication President Trump declared the coronavirus outbreak a national emergency. But the study’s findings provided the official scientific rationale for the change in course that had already been brewing, and in the process added momentum to it.

The study performed an important service in this regard. Even the public already knew by March 16 how important it was to “flatten the curve” to slow the rise in the number of cases, so as not to overwhelm the health care system, but no one knew what it would take to do so, and officials still spoke as if washing our hands and coughing into our shirtsleeves would do the trick. The study demonstrated how this was insufficient, and how more aggressive measures were needed. I agree with the study in this regard. The study’s emphasis on social distancing, which has led to the closing of non-essential businesses and restrictions on small gatherings, will help to lower the anticipated April peak in COVID-19 cases. My concern is with what follows the April peak, for by then economic issues will start to rival in importance the public health issues. The two trillion-dollar stimulus plan likely to be passed by Congress cannot be renewed every month going forward. There’s not enough money. In addition, the severe economic dislocation resulting from aggressive public health measures such as lockdowns and “shelter-in-place” orders might cause public health problems in their own right, stemming from mass unemployment that causes people to lose their houses and their ability to buy food. True, the study’s authors specifically state that economic policy is outside their bailiwick. Theirs is a purely epidemiological analysis, they say. Yet policymakers who use the study’s findings do not have the luxury of ignoring the economic dimension. For this reason, policymakers would be wise to make a closer inspection of the study, and especially its simulation parameters and unwarranted assumptions. These reveal not only an alternative approach to the pandemic, but also reasons for more optimism.

The study distinguishes between disease mitigation, in which the rate of viral infection is slowed to prevent the health care system from being overloaded, and disease suppression, in which the pandemic is stopped dead in its tracks. With neither mitigation nor suppression, the study predicts 2.2 million deaths in the United States from COVID-19. With mitigation the number of deaths is only halved. Although the demand for critical care beds decreases with mitigation, it is still estimated to be eight times higher than the existing number of beds, leading to many deaths from lack of health care. In addition, many people during mitigation will still get the disease, and some of them will die, even with health care.

Mitigation in the study includes measures such as case isolation, household quarantine, and social distancing for people over age 70. In case isolation, symptomatic cases stay home for seven days. In household quarantine, both the patient and all household members remain at home for 14 days. The study assumes that case isolation and household quarantine will only be in place for three months, while social distancing for the 70 plus age group will only be in place for four months. But why so short a time? Most Americans already expect anyone with COVID-19 to stay home for seven days, and to always stay home for seven days. Nor would most Americans think it outrageous to compel an entire household to remain isolated for 14 days if a member caught the disease. Many households are stocking up on food for just this purpose, to hunker down for 14 days. They would think it natural to always compel a household to do so, including themselves, and not just over a three month period. Few people would expect case isolation and household quarantine to last only three months.

In addition, many people age 70 and over would be quite comfortable with social distancing for longer than four months, especially since they are most at risk. Many seniors are retired and do not need to risk the social contact that comes with employment. Only 15 percent of this age group works, and often in service areas or in white-collar jobs that can frequently be done in isolation. Many of their recreational activities can also be done in isolation. Half of all female seniors (and they compose a majority of this age cohort) live alone. The study’s assumption that social distancing among seniors would be difficult and intolerable beyond four months thus seems unwarranted.

There are other questions about the study’s assumptions concerning mitigation. The study assumes only 70 percent of Americans will comply with case isolation, that only 50 percent will comply with household quarantine, and that only 75 percent of seniors will comply with social distancing, which is why the mitigation numbers are so disappointing. These estimates are too pessimistic. The compliance rate for these measures would likely be much higher. Driving is a good measure of what can be accomplished in this regard, for driving is practically a symbol of American freedom, and something that one would expect Americans to be loath to part with. Yet rates of driving appear to have already dropped fully by half in states with high COVID-19 case numbers, such as California and New York, while driving in states with few case numbers, such as Wyoming, have not, which is to be expected. When Americans feel a sense of urgency in their immediate environment, they reliably change their behavior.

As for those few infected households that refuse to accept the 14-day quarantine, most Americans would probably not see it as a violation of the country’s founding principles to force them to comply. Such enforcement is already taking root even in places like Kentucky and Fort Worth, Texas. This is the opposite of what one might expect, given the strong suspicion of government embedded in these regions, including among legislators.

Thus, mitigation may offer more possibilities than the study suggests, and the COVID-19 mortality rate may not be nearly as high as it predicts. In any event, rather than try mitigation, the study recommends suppression, which would include some combination of case isolation, home quarantine, closure of schools and universities, and social distancing of the entire population—all for a five-month period. The two suppression scenarios run in the study include, first, home quarantine, case isolation, and general social distancing, and, second, school and university closure, case isolation, and general social distancing. Curiously, no scenario in this part of the study includes all four modalities—for example, both home quarantine and school closure—which is surprising, since home quarantine for 14 days, as noted above, is something that most people would readily accept, along with school and university closure until the fall. True, we do not have official nationwide school and university closure until the fall as of this moment. Many districts plan to re-open at the end of April. But two weeks ago the plan had been to re-open in early April, and today there is growing expectation among the public that the schools will remain closed until fall. People sense government officials are trying to let them down gently, and it comes as no real surprise to them whenever the school opening date keeps being postponed. It is why Liberty University’s recent announcement that it would re-open its campus this semester received national attention. It went against the grain of what the public had already expected. Thus, for all practical purposes, schools and universities are closed until fall.

The addition of home quarantine to the policy mix that includes school closure would logically reduce the projected number of COVID-19 cases, which is already dramatically lowered through the other three suppression modalities. Indeed, even without home quarantine, the mix of school closure, case isolation, and social distancing of the general population lowers the number of new critical care cases dramatically after an April peak.

The study has other questionable assumptions and conclusions regarding suppression. First, as noted above, the study assumes that case isolation, home quarantine, and social distancing (at least for those 70 and above) would continue for no more than five months in a suppression strategy. In fact, they could go on for much longer, and with little public inconvenience or outcry. Second, the study seems to view school and university closure for five months as a continuous restriction. It doesn’t take into account the summer months, when school is out. On the one hand, this makes school closure an easy restriction, for the three summer months of closure come for free. On the other hand, summer may make the situation more difficult, and not less, given that students will no longer be bottled up inside institutions other than the occasional summer camp (a point that becomes relevant below). In any event, the fact that the study does not view school closure during summer as a significant epidemiological event is problematic.

Imperial College COVID-19 Response Team, March 16, 2020

Third, when combined with case isolation and general social distancing, the study judges school and university closure to be a more effective suppression strategy than home quarantine. But a close inspection of the graph (reproduced here from page 19 of the study) exposes an important nuance. Yes, the line that represents school and university closure (combined with case isolation and general social distancing) adheres very close to the x-axis, meaning few new COVID-19 critical care cases crop up after April, while the number of critical care beds available easily accommodates those that do. Yet the line that represents home quarantine without school and university closure runs only a little higher off the x-axis, and roughly tracks the line representing the current number of ICU beds available, which suggests the U.S. health care system could handle the scenario of suppression through home quarantine almost as well as it could the scenario based on school closure. This is important, as the public will likely accept home quarantine for 14 days. In contrast, school and university closure for the fall semester threatens social and economic havoc.

In addition, the line representing the suppression of critical care cases through home quarantine (along with general social distancing and case isolation) lifts off again when the five-month test period ends. But then, why would it end? As noted above, the public would accept a 14-day home quarantine strategy for good. The only major issue is the viability of general social distancing over a long period of time (to be discussed below).

Even with relying on school and university closure for only five months (which, as noted above, would probably happen anyhow for six months), the number of COVID-19 cases does not skyrocket again until October. This leads to one of the more irritating aspects of the study: It fails to take into account society’s capacity for innovation. For example, the red line in the study’s graphs, which represents the number of critical care beds in the U.S., remains constant between now and the rest of the year. This is inconceivable once the full force of America’s industrial sector comes online. It is more than merely plausible that the United States could double or even triple the number of fully equipped ICU venues between now and October. In the next two weeks alone, the U.S. Army Corps of Engineers will build 10,000 new critical care bed sites. Even the study’s primary author, Neil Ferguson, has recently revised his number of expected deaths in the UK downward, from several hundred thousand to twenty thousand, precisely because the recent aggressive public health steps taken by the government means there will be enough ICU beds to match the now-lower expected number of cases during the April peak. Nor does the study take into account the new and innovative anti-viral drug technology that will likely arise between now and October, if not to prevent COVID-19 then at least to treat it.

The question of closing schools this fall doesn’t even come into play, though, for the suppression strategy of case isolation, home quarantine, and social distancing for at least those above age 70 could continue for a very long time, with few complaints, and with little effect on the economy, until a vaccine comes along. The key question is whether the public will tolerate social distancing of the general population necessitated by this suppression strategy for longer than five months, while keeping the economy from cratering any further than it already will.

It is important to note that the study does not equate social distancing of the general population with a total lockdown, or with preventing people from going to work—which is what we have now in some cities and states. According to the study, social distancing for seniors is defined as lowering contacts in the workplace by 50 percent, which is easily doable, since many seniors are already retired and not working, and those who work typically do so in jobs amenable to teleworking. The study defines social distancing of the general population as reducing contact outside of the household, school, and workplace by 75 percent. What this means is that everyday socializing in the general population must be reduced by 75 percent. Social contact among the general population in the workplace is only expected to be reduced 25 percent, which is even less than the 50 percent reduction in social contact assumed for seniors in the workplace.

Yet sifting through these parameters and assumptions, one can discover another, more sustainable pandemic strategy that can save many lives without destroying our economy, which a strategy of total lockdown would threaten to do. The South Korean method that involves rigorous testing everywhere won’t work, since the United States will only have 27 million test kits by the end of this month. It is too late to use testing to confine the infection to specific regions, as the disease has already spread too far across the country. The major method left to us is social distancing, with the issue of school and university closure also a matter of debate, and home quarantine a given.

Another Option

To understand how general social distancing could work—and do so without destroying the U.S. economy—one must first understand the psychology of the American consumer who drives the American economy. At least in terms of discretionary spending, that person is mostly upper-middle class or higher. Low-income people obviously spend, but they do not have the same spending power beyond buying basic necessities, while lower middle-class people typically spend on practical consumer durables but not much more. While the middle classes drive consumer spending in Asia, it is the upper-middle class and rich that drive consumer spending in the United States, including leisure, luxury goods, entertainment, retail, and other services. To keep the consumer economy going in the United States, which is 70 percent of the American economy, these people need to keep buying.

Stores that sell basic goods are not at risk of bankruptcy or laying off workers. Everyone needs groceries and prescription medication. The companies at risk are those that sell everything else, especially to the upper-middle class and higher. But upper-middle class Americans have a peculiar psychology: They are extremely risk-averse, across their lives—for example, on how they invest, where they choose to live, the activities they pursue, and their social lives. With this temperament, upper-middle class consumers during the pandemic will avoid risk at all cost. A store telling them that it will provide them with masks, gloves, and some hand wash free of charge if they enter accomplishes little. Risk still exists, and so many will stay away. Not even “helicopter money” from the Federal government will change this behavior. Helicopter money will help low-income people buy basic necessities; it will not encourage the upper-middle class and the rich to jumpstart the consumer economy.

The online economy bypasses this risk, and those companies that sell products and services online will succeed and continue to employ workers. Every effort should be made to convert stores to this model, although most already have. Upper-middle class people will continue to patronize them while social distancing. This also includes stores that sell basic necessities.

Yet 25 percent of Americans do not use the Internet, whether because they cannot afford the service or because they lack access to broadband. Many of these Americans must still physically go to the store, but doing so risks violating social distancing in the general population. For social distancing to work, the Imperial College study says there must be a 75 percent reduction in social contacts outside of home, school, or work. Requiring low-income people to get their basic necessities by physically patronizing stores would likely violate this rule right off the bat. CVS, for example, sells drugs, but will only deliver them by mail if a customer registers online. As Congress ladles out money, it is essential to ensure that every American who consumes basic needs independently, outside of an institutional caregiver, has Internet access through a computer or some communication device, while also ensuring that the stores that service them can deliver food and prescription drugs to their residences. Otherwise we will have the paradox of an economy improving from upper-middle class people spending on luxury goods and a pandemic worsening from low-income people forced to venture out for shopping for basic goods.

The upper-middle class will patronize stores in person, but the stores must account for the mindset of the risk-averse and organize shopping in a way that satisfies their concerns. That means, for example, curbside delivery service (something that Best Buy has recently implemented); curbside consultations; pre-visit consultations through phone or Internet; better Internet product visuals; special traffic rules that let shoppers sign up on the Internet for a specific time slot, allowing that store to restrict the flow of customers so that only a handful can shop at a time. This would require stores to expand their hours or even shift to 24-hour schedules and spread workers out (which would actually help workers to social distance, too). Customers would also have to be reassured that they could avoid touching anything, including doors, while in the store, and that employees would also abide by social distancing rules.

Plane travel will require stringent rules to tempt upper-middle class consumers to use them in the near future, but those rules are manageable. Passenger planes must travel at 60 percent capacity to break even; planes will have to fly at that capacity to space passengers out and preserve social distancing. Americans Airlines has just recently announced its intention to improve social distancing on its planes along these lines. Screens will have to be placed between passengers. In addition, airlines will have to administer the new FDA-approved COVID-19 virus test that yields results in 45 minutes to all boarding passengers, and prevent those who test positive from boarding. Special head gear beyond masks for passengers would ease jitters. All these steps may seem ridiculous, but the psychology of sufficient numbers of upper-middle class consumers will demand it. So will the economy. Much of the consumer economy today (and therefore the economy in general) is organized around the sale of non-essential items—for example, leisure, entertainment, retail, restaurants, and travel for fun. This is the Achilles’ heel of a consumer economy geared not to the vital purchases of all income groups, but to the caprices of the upper-middle class and the wealthy. It is not the economy we might wish to have, but it is the economy we do have, and we must satisfy the peculiar consumer psychologies that come with it.

Economically speaking, finding a short-term treatment for COVID-19 is just as important as finding a long-term vaccine, and for the same reason: Fear moves upper-middle class people. It is why the anxiety grew dramatically in the United States not after the initial deaths of very elderly and debilitated nursing home residents in Washington state, but when middle-aged people without pre-existing disease started dying. At that point the fear became palpable, and the economy, at least in regard to non-essential purchases, began to collapse. Finding a treatment for COVID-19 won’t prevent a person from getting the disease, but the goal is to prevent a person from dying of the disease, to dissipate the mass fear stalling the consumer spending that drives the economy.

The same is true for the antibody test that shows whether one has had COVID-19. Many cases of the disease are asymptomatic, and so many upper-middle class people wonder if they have already had the disease. If they discover through the test that they have, they will feel immediate relief and re-start their old spending habits. Every upper-middle class person I’ve spoken with over the last few weeks has confessed this point to me. In terms of getting the economy moving again, the antibody test, which will emerge soon, should be made not only available but ubiquitous. For many upper-middle class people it will have the same salutary effect on their psychologies as a vaccine.

The 75 percent reduction in social contact outside the home, school, or workplace among the general population is the Imperial College study’s stiffest requirement for suppression, but it is achievable. To some degree, social distancing complements a longstanding trend in American upper-middle class behavior. Over the past few decades, the public square has shrunk, while upper-middle class people have recreated many of the amenities of public life inside their homes: in-home entertainment centers, in-home gyms, in-home bars, in-home game areas, in-home offices, and in-home restaurant-grade kitchens. Socializing in upper-middle class culture often takes the form of conversation through social media. In addition there exists a range of services from coaching to therapy to yoga to financial counseling that upper-middle class people purchase online. This foundation will allow for effective social distancing and strong consumer activity to exist simultaneously.

As for the workplace, the study expects only a 25 percent reduction in social contact at work to accomplish satisfactory disease suppression, which should not be hard to achieve, especially if workers, including unions, are flexible—for example, in the spreading out of shifts. A large number of workers in the service economy can simply work from home. Given that very large companies with more than 10,000 employees now employ almost 40 percent of Americans, and that another 15 percent of Americans work for government, many options exist for social distancing. Not all businesses are small restaurants, bars, and pubs with inflexible schedules, small manpower bases, small physical spaces, and tiny profit margins to tide them over during an economic downturn. Even the enormous chain restaurants can usually compensate for financial losses in one region by making more money in others. Finally, someone will have to deliver all these goods to the houses of the upper-middle class and rich. Domino’s Pizza just announced that it will hire 10,000 delivery people. More Uber drivers will also be needed. This will provide temporary work until the pandemic ends.

Yet the above logic applies only to people between the ages of 30 and 70 (with social distancing in the above 70 age group taken into account in the first part of this essay). It does not apply to young people, who tend to have the opposite sensibility. Rather than being averse to risk, they seem almost to court it. As the pandemic rages, young people are still partying on the beach and in off-campus fraternity houses that remain open despite colleges having been shut down. In Baltimore, where I live and go on solo walks, while carefully social distancing, I see young people all the time playing, laughing, hugging, and kissing in the parks and on the campuses of closed schools, both uptown and downtown, to this day. How can social distancing be achieved in this group to get to the necessary 75 percent reduction in social contact? It cannot.

This is where the Imperial College study becomes unreasonable in its assumptions. It argues that the most effective suppression strategy demands school and university closure, which causes the line representing new critical care cases to curve down after April toward zero, before climbing again, dangerously so, as students return to school in the fall and the weather cools. Thus, policymakers face a fork in the road in late April, depending on whether the critical care case numbers going forward actually do, in fact, decline. I do not believe they will plummet quite so much, as students out of school congregate. They find each other. Keeping them from doing so is like fighting the winds or reversing the tides. It is impossible, especially when both parents work, and even more so when a single parent does. These young people will go home after their fun and infect their parents or next-door neighbors, which increases the caseload. Much of the ground gained through social distancing among the general population of adults risks being lost. Early statistics out of China suggest that 80 percent of transmission occurs through the household.

If the study is correct, then school and university closure would have to continue through the fall semester, assuming no treatment for COVID-19 is found by then. The study assumes only a five month school closure; to prevent the fall spike school closure would have to be extended. The spike might be less intense than predicted if home quarantine is added to the policy mix. As noted above, the study does not combine school closure and home quarantine in its test case. If school closure is needed in the fall it will require college students to learn online, and younger students to be homeschooled.

But who will homeschool the children when the parents are working? Even more relevant, who will be at home to prevent the children and teenagers from going outside, finding their friends, and getting infected? The study assumes that contacts in the community will increase by 25 percent with school closure. This seems like a low number, given the determination of young people to find each other. It is why more than half of the coronavirus infections in New York are between 18 and 49. Young people even throw coronavirus parties. “If I get corona, I get corona,” one teen declared at a party last week. At least in school, the authorities can force some kind of social distancing on young people in the classroom. Without school, young people will be free to roam. The school closure strategy in the fall really pins all our hopes on finding some kind of treatment for COVID-19, since implementing closure would be economically disastrous as well as medically useless.

The alternative strategy, and one more likely to be needed, I believe—as the critical care case numbers probably will decline after the April peak, just not quite as much as expected—is to let students return to school in the fall. However, for college students it must be a one-way trip. No going home on weekends or holidays. This is key. College students will congregate no matter what. Great effort at social distancing should be tried in college—for example, through spacing in classrooms, converting to single-occupant dorm rooms, using alternative housing nearby, such as now-vacant hotels, setting up Quonset hut-like rooms for students on vacant campus land, or even using now-unused cruise ships to house students if the college is on a coast. But most likely these social distancing efforts will not work. The students will find each other; they will infect one another. Fortunately, their mortality rate is almost negligible, and, most important, they will not be going home to infect their parents if they remain at college.

Other adjustments will have to be made for the fall semester. Commuter schools will need to move to totally online platforms, or simply not offer classes for that semester, to prevent students going to and from home. In addition, colleges in general will have to craft an online alternative for students with pre-existing disease, and for those students who would prefer to stay at home and not take the risk of being in close contact with other students. Finally, teachers will need physical protection against infection from students—for example, teaching by video-feed.

Keeping the colleges open is important for an economic reason. In many towns and cities, a college or a hospital is the major employer. The biggest employer in Baltimore, for example, is Johns Hopkins, which has both a college and a hospital. Colleges serve as an economic lifeline for many towns—another downside of the global economy in which our manufacturing base was long ago off-shored. Not to have the students in college would destroy the delicate eco-system that enables many of these towns to survive. With hospitals busy treating COVID-19 cases and colleges busy educating students, and with much of this activity funded by the Federal government, these towns can survive economically.

As for younger children, kindergarten can be cancelled for the fall semester. Children who are kindergarten age or younger can remain at home. If they are in daycare, however, because a single parent works, or because both parents work, a big problem arises, as the children can become infected in daycare and then return home to infect their parents. The same problem exists for pre-college students attending school. Global capitalism gave us open borders and the easy transmission of the COVID-19 virus; it also created a wage scale that demands both parents work, even if one parent would prefer to stay home and raise the children. The practice leaves society little wiggle room.

If a technological solution to the pandemic has not been found by the fall, then one alternative is to switch to a suppression strategy that includes home quarantine, case isolation, and general social distancing, which will lead to a spike in critical care cases, according to the study, but one that might be managed with the new ICU beds and ventilators that will have come along by then. Another alternative, if society is desperate and the lives of hundreds of thousands truly are on the line, is to separate children from their parents for a month or two, to prevent the spread within the household. In other words, for a short period of time children can live at daycare. Hopefully, it won’t come to that.

As for pre-college youths, no obvious solution exists. A key point of the strategy I have outlined is to keep young people as far apart from adults and seniors as possible, and for as long as possible. In summer, when young people are roaming, without school to occupy them, rather than see the line representing the school closure suppression strategy adhere close to the x-axis, I expect it will rise somewhat, as young people, many of whom will be asymptomatic, infect their elders. Even summer camp will have little positive effect, as summer camp for young people is not a one-way trip; nor is it structured around a classroom, where young people can be made to social distance. On they contrary, young people play with each other in close proximity. Fortunately, summer weather is warm, which may suppress the coronavirus. Yet on the return to school in the fall, unless an enormous number of boarding schools are suddenly built for young people, there will be going back and forth between home and school, which will be a problem.

As for the small bars and restaurants in NYC closing down, and which have received an enormous amount of press attention, probably nothing can be done to save them. Older adults and seniors happily order take-out, but many twenty-somethings and thirty-somethings want to do more than just eat when they go to these venues. They want to “see and be seen;” they want to find each other. It is the reason why so many NYC bars and restaurants exist. Since they are non-essential, as so much is in the American consumer economy, and veritable petri dishes, it is probably wise to let them close for the foreseeable future. Curiously, but not surprisingly, the more expensive restaurants that cater to middle-aged, upper-middle class, and rich adults who are content with take-out seem to be holding on.

But there is hope—and one that I am very confident in holding. According to the study, after the April peak, things improve, and we gain time to strategize and work out the kinks in our suppression strategy. We have until the fall.

During that time, the kind of technological innovation that is the hallmark of capitalist societies can hopefully work its magic. I believe that the discovery of COVID-19 treatments, coupled with improvements in screening, will get us through this crisis, or at least help us muddle through until a proper vaccine is developed. Innovation will be our salvation, as it has been countless times before in our past.