The Past Speaks

Niall Ferguson (the historian, not the more famous Imperial College academic) has published a working paper on the impact of COVID-19 in which he talks about how we should use historical analogy to think about the pandemic’s likely economic and political impact. I have considerable respect for much of what Niall Ferguson has written elsewhere, particularly his early academic work, but the working paper on COVID is deeply flawed because Ferguson did not take advantage of the considerable literature in psychology and behavioural strategy on the pitfalls of analogical reasoning before putting pen to paper. The Stanford working paper gives the impression that Ferguson is not familiar with the concept of analogical distance, does not engage with the rational thinking movement, and has not read the literature that gives managers and other reasoners guidance on how to use analogical reasoning more effectively.

Anyone who wants to use historical analogy to understand COVID-19 faces a fundamental choice about whether to use an analogy that is high in analogical distance or one that is low in analogical distance. Dunbar (1997) observed that analogies vary in their analogical distance “(e.g., an atom is similar to a solar system versus a washing machine is similar to a dishwasher).” In the context of using history to try to think about the likely socio-economic impact of COVID-19, forming a historical analogy using, say, the 1968 Hong Kong Flu or the 1957 Asian Flu pandemics would be an example of low-distance analogical reasoning, while an analogy that connects COVID-19 to an event in another domain, such as military history, involves higher analogical distance. It is sometimes totally legitimate to use a high-distance historical analogy, but you need to justify the high distance. Ferguson, who does not even mention the 1968 and 1957 pandemics in his paper (!), frequently relies on a particularly high-distance analogy (the outbreak of the First World War in 1914) to understand COVID. A cynic might note here that Ferguson’s PhD thesis was on the First World War and that he has chosen to use this analogy simply because economies on mental effort and the need to read the historical studies on the pandemics of the twentieth century. More importantly, Ferguson does not really try to justify the use of this analogy.  Consider the following passage:

That is not to say that geopolitical analogies are always invalid, however, or that only the study of other global pandemics can help us understand this one. The lessons of the Black Death for 2020 are few and far between, except perhaps that the less integrated the world economy is, the slower a pathogen can travel; that social distancing always makes sense in a pandemic; and that people fleeing contagion are usually spreading it. Rather, we need to think of COVID-19 as one of those rare catastrophes that befall humanity at irregular intervals in history. In addition to pandemics, these include major wars, volcanic eruptions or earthquakes and extreme climatic events. Historians tend to
gravitate towards the study of such extreme disasters. Yet they seldom reflect very deeply on their common properties.

At no point in his paper, does Ferguson explain why we need to think of COVID-19 “as one of those rare catastrophes that befall humanity at irregular intervals in history.” Ferguson has chosen an analogy that directs our attention to his own published research and which, as a side effect, contributes to an intellectual climate in which decision-makers may be dramatically over-estimating the likely economic impact of the virus.

We are living through the coronavirus crisis and perhaps that means that all of us are inclined to over-estimate the long-term impact of the virus on a wide range of matters. Some people have predicted that the virus will change absolutely everything in our society and economy and will thus be regarded by future generations as a turning point of history. The view that the virus will result in permanent and fundamental change to our corporate governance systems and other core social institutions may be correct. I don’t know for certain because I can’t see the future. However, for reasons that  linked to the concepts of path dependency and regression to the mean, I think the opposite is more likely and that in five years things will be pretty much like they were in 2019.

That was the idea expressed recently by the economist Scott Sumner. I’ve long been a fan of Sumner’s work and he has had very intelligent things to say about the crisis whilst expressing a suitable degree of intellectual humility. One of the other social scientists who have guided my thinking about the crisis is the great Philip Tetlock, a guy who spends a lot of time thinking about how we can predict the future. His ideas reinforce my view that many people are overestimating the impact of the virus on social institutions. Ferguson’s historical analogies are, in my view, not helpful in calibrating our responses to the crisis.

There are other, more minor problems with Ferguson’s paper. For instance, he asserts that the COVID crisis has not witnessed an upsurge in patriotism and nationalism. “One important difference is that today’s crisis is happening without the offsetting boost to morale provided by patriotism.”

That’s not remotely true. The crisis has seen the sudden introduction of economic nationalist policies around the world (e.g. EU countries restricting exports of medical gear to other countries or the Canadian province of Quebec’s use of the crisis as an excuse for panier bleu economic nationalism). The crisis has also seen a surge in more benign forms of patriotism, at least here in the UK, where my street is still festooned with Union Jacks, England flags, and homemade signs that combined Union Jacks and rainbows and which salute “our N[ational] H[ealth] S[ervice]”. I could go on.

Instead of reading Nial Ferguson’s new working paper, busy decision-makers who want a historian to guide their thinking about the crisis should instead look at Going Viral, the excellent working paper by Dr Steve Davies that applies economic-history in thinking about this crisis. Steve isn’t a fellow of the Hoover Institution and thus doesn’t have the institutional prestige of a Niall Ferguson. Steve is also a committed classical liberal and is thus an adherent of a worldview that it is out of fashion right now. However, it is worth reading his paper, especially since Steve invested some time in learning about the pandemics of the twentieth century in the course of preparing his paper. Guess what, Steve talks about Hong Kong Flu, Asian Flu, and Spanish Flu.

During this crisis, I have been asked by several business people whether and how they should use economic history as a guide. My message to them is pretty much as follows:  if you hear someone using a historical analogy to try to predict the economic or political consequence of COVID, you need to set your bullsh*t detector to about 8 out of 10. Ask yourself about the incentive structure of the individual who is telling you this historical narrative. History can be a useful guide to what comes next, but be very critical of how people are using historical narrative ON you.

Guys, let’s think about the incentive structure of Niall Ferguson.