Test the healthy and gather data we don't already have

On April 2, the federal government announced that about 6.6 million people filed for unemployment benefits that week. The week earlier about 3 million Americans had filed. Until theses last two weeks, the worst week for unemployment filings was 695,000 in 1982.

And the economic pain is surely just starting. If current rates persist, upwards of 25% of the workforce soon could be unemployed, a number greater than was reached during the Great Depression of the 1930s.

But ignoring the coronavirus outbreak could lead to the death of millions of Americans. The pandemic forces us to make cruel choices among unhappy alternatives.

To see the kind of trade-offs we face, it is useful to think about the determinants of this epidemic. The number of people who will be infected over any period of time depends on the number who are currently infected and the rate at which infected people transmit the virus to susceptible people.

This transmission rate, in turn, is determined by biological factors — the chances an infected person will pass this particular virus to a susceptible person during a meeting — and by economic factors — the number of susceptible people an infected person meets and the conditions under which those meetings take place.

Public policy can do little about biological factors, but it can do a lot about the economic factors. One policy, which we think of as the current approach, is to limit meetings for almost everyone by imposing mass quarantines, ostensibly for a limited amount of time. At the end of the day, a "shelter in place" is simply a mass quarantine policy. And this policy will indeed reduce infections as long as it continues in place.

But this policy also has costs. A substantial fraction of economic activities require, or have enhanced value from, people being close to each other. To understand the value of social proximity, ask yourself a simple question: Why would people choose to pay the high rents of downtown Minneapolis (or the insanely high rents of midtown Manhattan), if not to be close to other people? Why would firms choose to pay the insanely high rents they pay to be in urban centers, if they did not value their employees being near each other?

Indeed, the last three decades have witnessed a rise in the market value of social proximity precisely at the time the internet was supposed to be freeing us from the usefulness of being near each other. Densely populated large cities have grown more rapidly in terms of economic output than sparsely populated rural areas.

It is precisely because social proximity is so valuable that mass quarantines are so costly.

While it is fashionable to say that the internet has made remote work a reality, the data say advances in technology have made it more valuable for people to work and live close to each other. The idea that this epidemic will finally usher in a new substantially online economy is magical thinking.

If we do not use the limited amount of time of a mass quarantine wisely, at the end of it we will still be almost where we were two weeks ago — a small fraction of the population infected having the ability to infect almost everybody else. So we will have no option but to extend the mass quarantine, again for a "limited period of time," and the carousel of hope and despair will continue to spin.

This policy will be astronomically costly.

So should we now immediately return to business as usual before the pandemic hit? Such a policy would be predicated on the hope that the infection rate is low, and a relatively large number of people have already acquired immunity. But absent data showing that this hope is true, such a policy runs the risk that millions of lives will be lost.

A wise use of the breathing room provided by mass quarantines would be to put in place the infrastructure to allow us to mimic the policies of countries such as South Korea, Taiwan, Singapore and Hong Kong. These countries have thus far controlled the pandemic at much lower economic cost. While the details differ, the common thread is that these countries, first, have aggressively employed measures of low economic cost to reduce transmission rates such as encouraging hand washing and mask wearing in public.

Second, theirs are aggressive but targeted quarantine policies. They quarantine people displaying symptoms, aggressively trace the people they have contacted, test their contacts, and then quarantine those who have the virus (and sometimes those who have just had contact with those who test positive), regardless of whether they are symptomatic or not.

It is a test, trace and isolate policy. These countries have not generally engaged in mass quarantines or shut down factories, shopping malls or restaurants.

Given the huge costs associated with non-targeted shutdowns, the needed testing and tracing infrastructure simply has to be priority one during the mass quarantine period. Put simply, a limited mass quarantine period makes sense only if we use the time it buys us to radically change the facts on the ground once this limited time is over — that is, if we use this time to implement an aggressive policy of expanding both testing capacity and treatment facilities.

In the meantime, as long as testing resources are scarce, we should not be testing those who have symptoms. We should simply assume that they are infected and treat them as if we knew they were. Instead, we should be testing the people whom those with symptoms have contacted who are currently asymptomatic.

Obviously, it makes sense to test health care workers, especially those involved in treating patients with coronavirus symptoms.

Another wise use of testing resources is to acquire the data needed to determine the benefits and costs of continued mass quarantines. For our state in particular, these benefits and costs depend critically on how many Minnesotans have been infected already.

Right now, we are all flying blind. We do not know if 700 or 7,000 or 700,000 Minnesotans have been infected. Testing a random fraction of the population would go far in allowing us to make evidence-based state policy, even if the tests are imperfect. (Abigail Wozniak of the Federal Reserve Bank of Minneapolis is leading such an effort.)

In terms of tracing, the state government should be engaging employees or contractors to trace all possible contacts. Innovators in Taiwan and Singapore have already developed apps that help in tracing and we should be quickly introducing them here.

The policies we advocate are based not just on abstract theories that we have developed. These are policies already being successfully implemented now. The ancient advice that we should mimic what successful people and countries are already doing is a sound one.

V.V. Chari and Christopher Phelan are professors of economics at the University of Minnesota and advisers to the Federal Reserve Bank of Minneapolis.