Moral Reasoning in a Pandemic

Image: Ventilators in storage (AdobeStock). 

Policy discussions during the pandemic have raised concerns for me, as a moral philosopher, about how policy analysts and policy makers are thinking about deaths from COVID-19 and the right way to combat them. The policy discussions I have in mind have ranged from broad issues about how and when to open the economy to more focused concerns about how Intensive Care Units in hospitals should allocate scarce medical equipment (including ventilators). I will here consider three areas of concern about how people are reasoning about what is morally right in the pandemic.

• • •

Interpersonal Aggregation 
How should we weigh the economic costs of keeping the economy shut down versus the lives lost to COVID-19 from opening it up? Speaking on the PBS evening News Hour June 18, economist Nick Bloom calculated that the experience of being shut in and suffering economic trauma could result in the loss of a year of life for a person. I do not want to second guess his estimate, but to ask about the use it might be put to in reasoning about what to do.

How should we weigh the economic costs of keeping the economy shut down versus the lives lost to COVID-19 from opening it up?

Bloom added up all the years of life that would be lost given 40 million jobs lost in the last three months of economic distress and came up with 40 million life years lost. The aim was to translate economic losses into the language of life. The question is how that translation helps us compare a closed economy to deaths caused by opening up the economy.

Bloom did not go so far as to compare this total number of life years lost with the deaths of other people from the virus. Indeed, he argued that people who are especially vulnerable to the virus should stay at home for their own safety while many others who are not so vulnerable could avoid the cost to them in terms of a life year lost by opening up the economy.

But some might be tempted to go further and conclude that the large total of life years lost (40 million) outweighs the deaths of individual people. But in translating economic loss into life years lost, we must not lose sight of the fact that if 40 million people each lose one year of life no one of those people will suffer a loss as great as someone who, for example, dies at the age of forty, thereby losing thirty years of life. It is a mistake in what is called “moral mathematics” to aggregate in an additive fashion small losses to many people, each of whom, for example, may die at age seventy rather than seventy one, and think of that summation of losses as greater than the loss of life to someone at age forty if he could otherwise have lived thirty more years. There is no one person who suffers the loss of 40 million years of life and there may be no one in the large group who dies at 40 rather than 70. This kind of “interpersonal aggregation” is morally misguided. 

Instead, it might be better morally speaking to use “pairwise comparison”: see if there is anyone among the 40 million who would suffer as much (or nearly as much) as a person who would die of COVID-19 when he would otherwise have lived much longer. If not, give priority to helping the person who would be worse off in dying rather than each of the many who would suffer a significantly lesser loss and be better off even with that loss than the person who dies. The moral idea is to give priority to the worse off.

The moral idea is to give priority to the worse off.

Consider an even more extreme version of interpersonal aggregation. Suppose someone calculated that each one of 1.2 million people would lose a month of life due to the closing the economy. That would mean a total of 100,000 years of life lost. But no person will suffer more than a month of lost life. Even more grotesquely, someone might use this kind of interpersonal aggregation to calculate how many months there are in an average lifetime and then figure out how many “lifetimes” would be lost even though no single person of the 1.2 million people would lose a lifetime—indeed, none would lose more than a month of life. (I have been told that this method was once used to argue that lower speed limits that made many people waste more minutes in traffic “cost lives,” once the minutes were added up to form units of minutes in an average life, and those lives lost could be weighed against lives saved from more traffic accidents occurring at higher speeds.)

• • •

Life Years 
If pairwise comparison is a better way to think morally, how should we do it? Consider the death of a single person and the expected number of years that person would lose if they died or have if they were saved. Some people think that we can determine how bad a death would be on the basis of expected years of (adequate) life that would be lost if that person were not saved. A death is worse if more years of life are lost, and in pairwise comparisons, we should allocate resources to avoid the worse outcome.

Suppose, for example, we must ration a resource—e.g., a drug or ventilator. Some people (e.g., the philosopher Peter Singer) think that medical personnel should save the person who (assuming equal chances and costs of achieving survival) would live the longest after treatment. Typically, this will mean saving the younger person: the idea is that it is worse for that person to die now because they have more years of life left. A March 2020 report in The New England Journal of Medicine on the pandemic in Italy described doctors there considering rationing so as to save younger rather than older people because more life years could be expected from a younger person. 

Regardless of our ultimate position on the policy of saving younger rather than older people, we should understand that an argument for the morality of saving younger patients need not be based on pairwise comparisons of life years expected from point of treatment. An alternative argument focuses on the fact that the older person will have had more life years even if he dies than the younger person will have had if he dies. On this view, if the younger person is not helped, he would wind up being the worst off, and so it would be most reasonable to save him rather than the older person who, even if not saved, would still be better off than the not helped younger person. In other words, it is morally more valuable (all else equal) to provide a benefit to someone who will have had less of that thing. This way of thinking, which conflicts with the focus on years of life lost, captures our intuitive thought that often a younger death is worse because the person will have had less of the good of life.

We should understand that an argument for the morality of saving younger patients need not be based on pairwise comparisons of life years expected from point of treatment.

This way of making pairwise comparisons has practical consequences. Notice that if doctors think in terms of helping people based on expected life years they would have to help a sixty year old whose life would be extended 20 years by treatment in preference to a thirty year old whose life would be extended 15 years by treatment. By contrast, concern for helping the person who will have had the fewer life years if he dies would favor helping the younger person because he will have had only thirty years of life if he dies whereas the older person will have had sixty even if he is not helped. The death of this younger person is worse not because they lose out on more years than the older person but because they will not have lived very long.

• • •

Love and Masks 
In his final daily address about the pandemic to the people of New York, Governor Andrew Cuomo said that “love does win” and gave as an example “I wear a mask for you and you wear a mask for me.” The point of Cuomo’s example was that the sorts of mask it is now recommended that we wear on the streets do not primarily protect the wearer from getting COVID-19. Rather they reduce the chances of transmitting the virus to others.

As is well known, many people in the United States refuse to wear masks. This would probably not be so if they had been told that their wearing a mask directly protects them. As it is, if someone doesn’t wear a mask, they can still be a free rider: they get the benefit of protection from others wearing masks, while not wearing a mask themselves.

Is it really true that wearing masks to protect each other is an act of love? One reason for thinking otherwise is that a person may wear a mask in an effort to encourage a practice that protects the person who wears the mask: If I wear a mask, I may thereby encourage others to do so, which in turn will protect me. In this case, I am wearing the mask out of concern for self, not for others, so wearing it need not be an act of love. But there is another important concern about treating wearing a mask as an act of love. As Cuomo emphasized in earlier addresses to the public, if I do not wear a mask, I increase risks to others. This, as Cuomo said, is simply disrespectful to others. I think this way of putting the matter is more accurate.

Is it really true that wearing masks to protect each other is an act of love?

Acts of love typically seek to benefit others, making them better off than they would otherwise be. Wearing a mask does not make someone else better off than they would otherwise be. Instead, it avoids making them worse off by spreading a disease to them. If I wear a mask, I am not a benefactor to others as if I saved someone from drowning; rather, I avoid being a threat to them. Conservatives and libertarians typically emphasize not violating the negative right people have not to be harmed. Why then might so many of them object to wearing masks?

Those philosophical beliefs do not deny that there should be restrictions on freedom to prevent some from imposing harm on others even if the harm would be unintended. And sometimes you can only avoid harming someone by doing something, such as waiting until someone leaves a space before entering it or swerving one’s car to avoid hitting someone. If the risk of someone’s spreading disease were extremely low and this person were the only one who posed the risk, there might be an objection to requiring a mask. However, we know that if everyone who presents this small risk goes maskless some people will be infected; to avoid this occurring, all who present only a small risk may have to pay a small cost, doing their fair share to reduce the risk.

Finally, what should be done if many people actually refuse to wear masks that would primarily protect others, not themselves? An N95 mask would protect to a high degree the person who wears it. Nonmedical personnel are strongly discouraged from wearing them primarily because there is a shortage and priority should be given to medical personnel. But perhaps production of these masks for those in the general public should be ramped up. Those who refuse to act for the sake of others (and indirectly for their own sake) may wear a mask or other device that is intended to protect them directly and those who cannot rely on people wearing masks for the sake of others could engage in self-defense by wearing an N95 mask or similar protection. For those concerned that people staying at home  or even maintaining distance sufficient alone to prevent spread of the virus will prevent an economic resurgence, the question is whether the economic cost of providing N95 masks or similar self protection more generally would be smaller than the economic costs of not opening up the economy to avoid serious illness or death.

Whatever the best policy turns out to be, the important point is that those on the political right and left should be able to agree that there is a duty not to harm others and wearing a mask to fulfill the duty may justifiably be required of most people, even if love or simply saving lives could not be similarly required.