How Can We Know If We’re Overreacting to COVID-19?

It may not seem like it now, but one day the threat of the COVID-19 virus will pass and life will more or less return to normal. When that happens, we as a society will be faced with a rather difficult question: did government respond appropriately to the crisis? How many lives were saved and how do they stack up against the harm caused by shuttering many of the nation’s businesses for weeks or months? Did we overreact or did we underreact, and if so, what should we have done instead?

Despite the fact that this is an extremely important question, as it will inform how we respond the next time a new virus emerges, the fact is that it’s going to be nearly impossible to answer with any degree of certainty. If the death toll ends up far lower than predicted, some will argue that it’s only because the preventative measures put in place worked, while others will claim they were never necessary in the first place. Conversely, if the damage is worse than expected, some will say that our very reaction made things worse, while others will argue that we didn’t go far enough. How will we know which of these claims is true?

Epistemology is the boringly-named study of truth and how we distinguish what is true from what is false. It’s a simple concept that turns out to be surprisingly complicated, but there are essentially two schools of thought. Rationalists believe that we can use logic and a priori reasoning to arrive at correct conclusions, while empiricists argue that only systematic observation of objective facts can get at the truth. As one might expect, both sides have a point.

In the hard sciences, such as physics, chemistry, and astronomy, the scientific method has emerged as a remarkable tool for solving the mysteries of the universe. The scientific method is fundamentally empirical, based on testing hypotheses against repeated observations. While the rationalists dreamed up plausible theories about a geocentric universe or hypothetical substances like phlogiston, empirical testing via the scientific method was able to dispel these myths and expose the truth.

In fact, the scientific method is so powerful and so useful that its reputation has rather exceeded its actual utility. While enormously successful in the hard sciences, the scientific method falls down in fields where consistent, repeatable testing is not possible. In short, fields involving people and their behavior.

The misleadingly-named social sciences are poor subjects for scientific testing, for the simple reason that people are different and will behave differently in the same situation. Even the same people will behave differently in the same situation if they are repeatedly subjected to it. Unfortunately, this hasn’t stopped experts and laypeople alike from trying to erroneously apply natural science methods to social sciences like psychology, sociology, and my own field of economics.

To give an example, a hundred years after the Great Depression, economists still remain divided on the cause of that crisis, and whether the policy measures taken to combat it made things better or worse. My own view, derived from the teachings of the Austrian School, is that the Federal Reserve’s expansion of credit is largely responsible for the crash, and that President Roosevelt’s heavy-handed response needlessly prolonged the crisis. However, because we can’t repeat the conditions of the Great Depression and test what would have happened without the New Deal or the National Industrial Recovery Act, there’s no way to say for sure. Even if we could repeat the conditions exactly, it would still prove little, because there’s no way to guarantee that people would respond in the same way to economic conditions.

The pesky thing about free will is that there’s no way to know for sure what people are going to do until they do it.

All of which brings us back to COVID-19. We are being told, rather forcefully, that we have to listen to the scientists. People who aren’t experts in medicine or epidemiology are being brusquely asked to sit down and shut up, and let the adults handle this problem. The trouble with this kind of reasoning is that the problem to be solved is not, at its core, a scientific one.

To be sure, science can tell us how the virus interacts with the human body and how it can spread through a population, but it can’t tell us what to do with that information, for the simple reason that other factors outside the discipline of medicine have to be considered. Science can’t tell us, for example, how many people would voluntarily self-quarantine absent government mandates. It can’t tell us to what extent people will wash their hands and follow the recommended sanitation guidelines. It can’t tell us which social gatherings would have been cancelled or altered without orders from on high. We can look at other countries like South Korea, whose response has been much less draconian, but at the end of the day you can’t expect Koreans to behave identically to Americans. Not only are they individuals with their own distinct personalities, but there are undoubtedly cultural differences, a different relationship to authority, and very probably a different immune response to the virus.

Without knowing all these variables, it is folly to make predictions about “what would have happened” if government had reacted differently, and it’s a big part of why I view the computer models being presented with skepticism. Such models make predictions which are not falsifiable, meaning that there’s no way to tell post hoc whether they were right or not. “Ordering that all restaurants close will save a million lives” is an alarming statement, but there’s no way to test whether it’s true or not because you can’t both close all the restaurants and leave them all open simultaneously to compare the results. Non-falsifiable claims are fundamentally non-scientific, because they can’t be tested. To make these predictions, the models incorporate a set of assumptions about human behavior, and we have no way of knowing whether those assumptions are even remotely accurate.

All this is beginning to sound rather fatalistic. If we can’t know empirically whether a policy response will have the intended effect, should we just give up and not even try? Of course not, but we must recognize that predictive models are necessarily flawed, and that we have to balance the opinions of doctors with those from a wide variety of other relevant fields. We keep being told that we shouldn’t question the experts; well, I’m an expert on economics, and there are some fairly predictable outcomes we can expect both from the enforced quarantines and the stimulus package recently signed by the president. You don’t need a crystal ball to know that halting the production of goods and services over a large portion of the economy for any significant period of time will increase poverty and unemployment, result in shortages, and put smaller firms out of business, leaving their larger competitors free to swallow them up. A greater concentration of market power in the hands of a few companies will result in less competition, higher prices, and lower quality goods. The increase in the money supply by the Federal Reserve will create inflation, reducing the buying power of everyone’s dollars.

These are all foreseeable outcomes that need to be considered alongside medical recommendations.

Economists are far from the only ones with contributions to make. Psychologists have warned of the mental health effects of prolonged periods of isolation and fear, as well as the aforementioned economic insecurity. Suicide rates are expected to rise, as indeed they already have. We also need to consider the fields of moral philosophy and law. What are the ethics of forcibly preventing people from seeing their friends and family? How is, for example, Bill De Blasio’s threat to permanently shut down houses of worship consistent with the First Amendment?

Both reason and history teach us that those in power will tend to use a crisis to expand their control over a population, and that this creates an incentive to exaggerate danger rather than downplay it. To forget this would be to ignore the entirety of political science from Plato’s Republic onward, something we can afford to do only at our own peril.

Scientific tunnel vision is not new. In the early 20th century, the public became so enamored with the theories of evolutionary biology that they allowed eugenicists to forcibly sterilize tens of thousands of Americans. Moral, ethical, and legal considerations were eclipsed by the narrow observation that selective breeding can increase the genetic strength of a population over time. Demands that objectors shut up and listen to the experts resulted in one of the blackest chapters in American history, one which only ended because war against the Nazis exposed the field of eugenics as a monstrous injustice.

I’m not saying we should ignore the scientists and doctors working to resolve this crisis—far from it. I’m saying that we have to recognize the inherent limits of their predictive powers, and weigh their recommendations against other considerations that also impact the health and prosperity of America’s citizens. Of course we should listen to experts, but we have to acknowledge that experts exist in a wide variety of relevant fields, and that no one is an expert on everything.

When this is all over, we may never know how many lives were saved by shelter in place orders or what the virus would have done under other circumstances. But what we can be sure of is that too little consideration was given to other aspects of human life, things that continue to matter even during a national emergency.