Thursday, March 19, 2020

The Foolishness of the Coronavirus Panic and Trump's Economic Shutdown: A Darwinian Evolutionary Analysis

Is it reasonable to be thrown into a state of panic by the spread and virulence of the COVID-19 virus?  Does this justify a massive shutdown of the global economy as proposed by Trump and other political leaders?  A Darwinian evolutionary analysis of the coronavirus epidemic suggests that the answer to both questions is no.

My reasons for this conclusion have been summarized by various commentators--particularly, John Ioannidis and Richard Epstein.  Epstein has stated his reasoning in an video interview at Reason magazine.

At Trump's press conference on Monday afternoon, he announced the establishment of what might be called the Coronavirus Administrative State as recommended by a White House task force.  This would include economic lockdowns and business closings like those already ordered by mayors and governors in many American states and cities.  When Trump was asked how long this would be required, he said it might last until July or August.  This was a lie, because as indicated by later reporting here and here, the White House has already written out a plan for locking down much of the American economy for at least 18 months, which is in response to warnings from a team of epidemiologists at Imperial College London arguing that this is the only way to avoid massive global deaths from the virus--including 250,000 deaths in Great Britain and 1.2 million deaths in the U.S., or even twice that if the epidemic is not mitigated at all.

If you read that Imperial College study, you will see that it's based on a mathematical model that assumes a "reproduction number" (R) of 2.4 and an "infection fatality ratio" of between 0.25% and 1.0%.  A 2.4 reproduction number means that every person infected will pass the infection on to an average of 2.4 people, which creates an exponentially growing rate with a doubling time of 5 days.  An infection fatality ratio of 1% means that on average 1% of the people infected will die.  With these parameters, it's a simple mathematical exercise to project one or two million deaths in the U.S. and hundreds of millions around the world.

But do the existing data about the CORVID-19 epidemic and what we know about the evolution of viruses and of human behavior in response to deadly viruses support the assumptions behind this simple model?  The authors of the Imperial College study hardly ask much less answer this question.  The same can be said about similar studies, such as the essay by Tomas Pueyo that I cited in my previous post.

Consider the replication rate of 2.4.  This assumes that human beings will not respond to the epidemic in adaptive ways to slow down the replication.  That is, of course, a false assumption.  During the early stages of the spread of a disease, most people will not take precautions against it, because they are unaware of the danger.  But as soon as people learn about an epidemic and learn that they can reduce it by washing their hands, by "social distancing," and by isolating infected individuals and those most susceptible to deadly infections, they do so; and this reduces the replication rate.

Sometimes the replication rate for infectious viruses varies according to the seasons of the year.  One study of the coronavirus epidemic suggest that the pattern of the epidemic in China shows that the epidemic slows down when the temperature and humidity is high.  So it's possible that the epidemic could be slowed down in the Northern  Hemisphere by the coming of warmer spring and summer weather.  But other researchers dispute this.

A fixed replication rate of 2.4 ignores the fact that the most vulnerable people are hit first, and therefore replication will go down as the remaining pool of people is more resistant to infection.

A fixed mortality rate of 1% ignores the fact that viruses vary in their virulence, and that the most virulent viruses can be eliminated by natural selection.  As long as human beings are ignorant of the epidemic and not taking precautions against it, the most deadly viruses can reach their next victims before killing its host.  But once people adapt to slowing down the transmission of viruses, the most virulent viruses will kill their hosts before they have a chance to spread.  The weaker viruses will survive and reproduce at a higher rate than the stronger viruses, and consequently the viruses will become less severe by natural selection.

The mortality rate in an epidemic also often varies according to age groups.  In this case, the mortality for the coronavirus is much higher for the elderly--and particularly those with preexisting health problems--than for other age groups.  Some of the data indicate a fatality rate of 9.51% for those 80 and above, 5.33% for those 70-79, as opposed to 0.12% for those 30-39 and 0.09% for those 40-49.  That explains why the highest death rates have been in old age care facilities and in countries like Italy that have aging populations.

The conclusion from all this is that the most evolutionarily adaptive response to the coronavirus epidemic would be to isolate the most vulnerable groups--particularly, those over 70 with health problems.  But there is no reason to shut down large parts of our economy to protect younger, healthier people.  To do that--as Trump and others have started to do--will create more harm than good--the lives and livelihoods of hundreds of millions of people will be ruined.  Countries like South Korea and Singapore have brought the epidemic under control with massive testing and isolation of infected individuals, but without shutting down their economies.  That's the prudent policy.

There is an issue in political philosophy at stake here.  Some social commentators have said that the need for governmental suppression of economic activity to deal with the coronavirus pandemic shows the flaws in liberal political theory, because it shows that in times of emergency threats to human life, we have to sacrifice individual liberty and property rights for the sake of governmental intervention to protect public health.

That argument is false both in its interpretation of liberal political theory and in its assessment of the best public policy for the epidemic.  First, as Epstein observes, it overlooks the fact that liberal political theorists have always recognized that in time of emergencies when human survival is at stake, "the safety of the people is the highest law," and government must act to temporally set aside the normal laws of liberty and private property: "property rights are suspended in times of necessity, but only as long as the necessity lasts."

In this case, however, the necessity is not clear enough to justify the sacrifice of liberty and property for security.  In this case, the extreme measures for shutting down much of the economy would not serve the public good.

If John Locke is right that human beings have natural rights to life, liberty, and property; that they will consent to government as long as it secures those rights; but that when government infringes on those rights, they have the natural right to the "executive power of the law of nature" allowing them to rebel, then we might predict that as people see that their governments are threatening their lives, liberty, and property by shutting down economic activity in ways that are unnecessary for handling this epidemic, then people will engage in civil disobedience.  For example, how long will it be before owners of bars, restaurants, and other businesses shut down by government defy the closure orders and open up for business?

In the U.S., Trump hopes that mailing out checks for a few thousand dollars to Americans will prevent civil unrest.  How likely is that to work as the months go by, and more and more people invoke their natural right to rebel?


MNewsham said...

They are already rebelling by going on Spring Break and insisting on holding large gatherings in church, but I thought that was stupidity, rather than people's natural right to rebel.

W. Bond said...

A brief note on your sources. First, I like Epstein. I own several of his books, read his column, listen to his podcast. He gave a lecture, available on Youtube, "Natural Law in Ancient and Modern Guise" which is outstanding. If you read his column on this topic, however, there is an aside on the HIV epidemic where he gets the history and biology/Medicine completely backwards, I write as someone who was caring for those patients at a public hospital in training pre and post 1996. If he so misunderstands that story, I see no reason to trust his hastily thrown together thoughts on this one.

Second: Ioannidis. An interesting case. To date, he is most famous for a single paper. After the piece this week, he may be best known for this one, which has been widely dismissed this week in the medical world as risibly myopic. Many wise rebuttals were instantly written. Here, however, you have not a polymath Law Professor towards the end of his career, but a Professor of (among other things Epidemiology) who is fellowship trained in I.D. at the peak of his powers.

Yet, he writes (this week!): "If we assume that case fatality rate among individuals infected by SARS-CoV-2 is 0.3% in the general population — a mid-range guess from my Diamond Princess analysis — and that 1% of the U.S. population gets infected (about 3.3 million people), this would translate to about 10,000 deaths. This sounds like a huge number, but it is buried within the noise of the estimate of deaths from 'influenza-like illness.' If we had not known about a new virus out there, and had not checked individuals with PCR tests, the number of total deaths due to 'influenza-like illness' would not seem unusual this year. At most, we might have casually noted that flu this season seems to be a bit worse than average. The media coverage would have been less than for an NBA game between the two most indifferent teams."


W. Bond said...

(continued) Many studies that estimate flu mortality, looks at excess death rates during the flu outbreak, and attribute those deltas to seasonal influenza. This is different than seeing pts directly die from the disease. So, these macro numbers and comparisons are often misleading. Ask practicing physicians how often they see pts requiring prolonged critical care from seasonal influenza? Very rare. In Kirkland 1/6 of a nursing home died in one week. This just doesn't happen during the seasonal flu outbreak -- including all respiratory infections, bacterial pneumonia, etc. Such an occurrence alone would prompt a public health investigation. In the initial case series from China, 30% were infected health care workers. Never would have been confused with the flu. In Italy, 20% of healthcare workers have become infected. Not the flu. The CFR is one thing, the % requiring oxygen, hospitalization, and prolonged critical care is another. The CFR goes up significantly, by the way, if we are one ICU nurse short. Anyone following this outbreak closely could never have made his statement. A grand international medical delusion? No one who can be taken seriously could write the above quoted statement. How then, does he get this so wrong? As you may know, there is a type of radical Evidence Based Medicine that did not develop the RCT, but has elevated the RCT to the only form of knowledge in clinical Medicine. Curiously, they tend to disparage not only common sense and observation, but also basic science in Medicine, even advocating for the the study of pathophysiology, etc to be dropped from curriculum. Ioannidis is the smartest of this school, but comes from this school. He focuses on the Cruise ship data, while ignoring China, Italy, Kirkland (and everywhere else) precisely because he is blinded by his own adherence to a single method. Here he is like the proverbial drunk looking for his keys under the streetlight. Or, if you prefer, he is like Eryximachus in the Symposium.

In any case, he demonstrates a profound lack of the virtue of practical wisdom. Calling for better data is one thing, but here it is a cop out. Prudence requires acting wisely in the face of uncertainty, risk, and partial, emerging information.

Last, my personal 2 cents. It has been clear for at least one month that two seemingly (but not actually) contradictory statements are likely both true. Any individual's absolute risk of becoming critically ill from COVID-19 will be low. At the same, time the risk that this could overwhelm are medical resources is non-trivial. The CFR goes up in this case, not just from COVID, but from acute MI, hip fractures, perforated viscus, and every other non-COVID condition. I suspect it is difficult, if indeed not natural, for people to hold these two thoughts at the same time. Most responses are either panic or dismissal, where the truth lies in between. Social distancing as a mitigation strategy is for your neighbor, but hoarding behavior means people are doing it for themselves. This epidemic will likely continue to unfold unevenly across the country. When one adds in that it takes 1-2 weeks often for the ARDS to present after the initial infection, I do agree with you that many will likely grow restless.

Larry Arnhart said...

W Bond,

Thanks for these clarifying comments, which do raise some good objections to what I have written. I will have to think more about this.