Peter Kraneveld applies statistical theory to COVID-19 and concludes that investors should start thinking about events that have no mean regression
Financial economists like to think in terms of cycles and standard distributions: Gaussian curves (bell curves). Growth, interest rates and real estate are thought of as cyclical and short-term equity prices are said to be regressing to mean.
These are very handy assumptions, because they make things as predictable as seasons.
Now consider Moore’s law: “the number of transistors in a dense integrated circuit doubles about every two years.” The law suggests an upward curve that is neither cyclical nor Gaussian.
Intuitively, we suspect that there is a physical limit to the law. However, we will not know what the mean is until that limit is reached and we cannot reasonably expect the number of transistors to go down after it reached its zenith.
There will be no mean regression. Apparently, there are economic phenomena that have no cycle, no standard distribution and no mean. Imagine what this means for IT investments: a monotonous rise, following by sudden, unpredictable stagnation forever.
Many phenomena outside finance and economics act alike. Examples are floods, earthquakes, tsunamis, hurricanes, forest and prairie fires, wars, revolutions, states splitting or merging.
There are also phenomena such as computer break-ins, crime and terrorist attacks that can take so many forms that the mean is, well, meaningless.
These phenomena have no maximum and they do not disappear at the tail of the curve. In other words, no matter how extreme the most extreme of the event was, it is possible that a next event will be worse and no matter for how long such events do not occur, they can return. If they regress to anything, it is to infinity. The events are unpredictable.
A pandemic is such an unpredictable event. It is certain to occur at some time in the future, but it cannot be timed. Indeed, it can be argued that Bill Gates predicted COVID-19 in 2015, when he said “When I was a kid, the disaster we worried about most was a nuclear war… Today the greatest risk of global catastrophe doesn’t look like this… If anything kills over 10 million people in the next few decades, it’s most likely to be a highly infectious virus rather than a war.”
However, neatly applying the financial analyst’s adage “if you give them a rate, don’t give them a date”, his prediction was of no use. At best, it was a call to have large stocks of hospital equipment such as beds, gowns and masks. At worst it shows that Bill Gates gets more media time than other people, so that he has more chances of getting something right.
Statistical modelling gives clues on how to handle mean-less, unpredictable events. When the event occurs, it is important to take fast and drastic action to break the chain of transmission.
The first French hot spot of COVID-19 was In Creil. French citizens evacuated from China landed on a nearby military airport, were handled routinely and without requiring quarantine. How difficult would it have been to consider the evacuees as potentially contaminated, keep distance and hold them in quarantine for a while? It would have saved numerous lives and economic damage of a size that is still unclear.
This example illustrates the lop-sided effect of fast and drastic action. Its cost is next to zero compared to the consequences of not acting fast and decisive.
Now, people are returning from summer holidays and there is still no systematic testing at the airports. Isn’t it clear why there is a second wave? Or have decision-makers assumed a Gaussian curve, perhaps?
I can be like Bill Gates and predict that climate change will occur. The only question is whether it can be turned back. The prospects are bad. Voters allow politicians to prepare in a late, slow and insufficient way again. That is a recipe for death, disaster and incomprehensible economic damage.
As individuals, we can do too little, even if we buy a hybrid car, install sun cells on the roof and stop eating red meat. As investors, we should have changed our portfolios early, swiftly and radically already, with pressure on the big sinning companies as a nice bonus.
We need to stop thinking of disasters as having a standard distribution and start acting now.
Peter Kraneveld is an international pensions adviser at PRIME bv.