Condivergence: Fooled by randomness and other expert mistakes

This article first appeared in Forum, The Edge Malaysia Weekly, on September 7, 2020 - September 13, 2020.
-A +A

Is the Covid-19 pandemic a Black Swan event or a Grey Rhino? If so, why didn’t the best experts see it coming and the best nations cope that well, recording the highest death rates?

A Black Swan event is defined as a low-probability event with high impact. A Grey Rhino is defined as an event with a high probability that everyone knows is there, but neglects when it will charge at you. It is like the elephant in the room that no one wants to talk about.

In the World Economic Forum’s Risk Report 2020, published in January, infectious disease was not included in the top 10 risks, but was number 10 in terms of impact. Risk consultant Willis Towers Watson’s survey report on extreme risks for insurers did not include pandemics as one of the top 15 risks. Of the 30 risk categories, the pandemic was classified as a one-in-100 year event and the risk uncertainty was classified as medium, but “crushing” in impact and global in scope. Even the best experts were unable to see the pandemic risks coming.

Nassim Taleb, author of The Black Swan: The Impact of the Highly Improbable, considers that specialist experts tend to be less wise than the crowd’s common sense. If the models used by the experts are wrong, their predictions are likely to be wrong, with very harmful consequences. Specialists tend to develop “tunnel” vision, ignoring what is not predicted in their models. And if many (but not all) experts agree that their model is best, then they will not only not predict what is coming, but tend to ignore or even condemn “outlier” experts who argue that the majority are wrong. Worse, they continue to repeat the same mistakes over and over again.

A good example is the use of “average risk”.  The risk of dying from accidents was less than 1%, or 6% of all deaths in the US. As we all know, when Covid-19 appeared, there was a belief that the mortality rate would be like SARS, at roughly 1%, and maybe even lower for young people. So, those who believed that the risk was 1 in 100 ,and that it would be lower for the young, were willing to party and ignore washing of hands, wearing of masks and practising social distancing.

What they forgot is that Covid-19 is infectious (interactive between two or more parties) and that if you enter an area where there is a super-spreader, the probability of being infected is significantly higher than 1%. Indeed, where the UK health service failed badly, leading to high mortality rates, was to move elderly patients out of hospitals into care homes without testing them first, resulting in these homes becoming deadly clusters of infection and deaths.

Taleb is right in that “common sense” would tell people not to go into areas where there are concentrated risks, such as those where there is high crime. Even though, on average, for a whole country, the risks of being robbed is low, the chances are extremely high in a known high-crime area at night, if one is alone and flashing money.

Asians who have been following standard operating procedures (SOPs) during lockdown were therefore horrified that not only were many young people in the US, the UK and Europe not wearing masks, but they were attending parties and political rallies as if the pandemic risks were already over. We do not blame the young, but the politicians who allow such folly to happen should know better.

The point I want to make is that we are taught to recognise average risks from the Bell Curve, which assumes that risks are balanced. The average represents the whole. If we throw a coin many times, it is likely that it will come up heads half the time, and tails, half the time. The basic assumption is that each act of throwing the coin is independent of each other. If, however, the coin is “loaded or biased”, then the statistical curve will not be a balanced bell, but skewed with fat tails, meaning that a low probability event could generate more than its share of losses or gains. Skewed curves follow what is technically known as power laws.

Natural and social scientists have discovered that skewed curves occur more frequently in nature and human behaviour than is commonly acknowledged. The world is more unequal than equal. Natural selection, for example, is better predicted with a power law. In a pristine natural forest, one species of trees may be more common than other trees. Man is an example of a dominant species, and other species are disappearing. However, dominant species also attract predators and disease, so they either have to adapt (develop immune systems) or yield to more resilient species. Information scientists who study networks discover that over time, the largest activities concentrate in a small number of hubs, exhibiting winner-take-all characteristics. In network industries, such as media, banking or aircraft manufacturing, only a handful of companies dominate. Concentration occurs but competition from diverse sources and newcomers erode that concentration. If, however, the dominant players capture or influence the laws and rules of the game, then the bias or dominance is retained. If 1% of the rich are able to capture the politics, then some are definitely more equal than others.

The second simplifying assumption that is flawed is “randomness”. The idea of random behaviour comes from “atomism”, which believes that the world is made up of atoms that are independent of each other and therefore behave randomly. Burton Malkiel’s famous book, A Random Walk down Wall Street, was extremely influential in finance theory. It argues that asset prices behave randomly and, therefore, one cannot outperform the market. This led to the efficient market hypothesis — an important myth underpinning neoliberal market ideology.

However, experts like Warren Buffet consistently outperform the market indices by investing in value stocks. He particularly likes those companies that exhibit monopolistic or oligopolistic performance. In other words, if you invest in companies that “capture” their business, and everyone believes that you have the winning model, you can grow so large as to affect the market itself. The current glove stock boom is the belief that, in the short run, rubber glove manufacturers are able to exploit frictions in the market because competitors cannot enter the industry fast enough to lower prices during the period of high demand arising from the pandemic.

Individualism (the belief that the individual has total free will independent of everyone else) is a core element of liberalism. As the right-wing Hoover Institution’s paean to Milton Friedman argues, “Individualism is its (liberalism) creed; collectivism and tyranny its enemy. Society is a collection of individuals, and the whole is no greater than the sum of its parts”.

But what if everything is interactive with each other? The Greeks thought that everything can be reduced to very basic particles called atoms. This introduced “reductionist” thinking, because you break a complex problem into its parts and then study the parts in depth. The classical scientists in the West, beginning with Galileo and Descartes, explained natural behaviour in simple mathematical or linear terms, paving the way for the Scientific Revolution. Every machine is explained as a whole that is the sum (simple addition) of its parts.

But what is life and complexity? Life is more than the sum of physical atoms, because humans, animals or even plants are able to reproduce themselves and seem to have consciousness that physical laws cannot explain. If everything interacts with each other, then the whole could be larger or smaller than its parts. The creeper vine in my garden seems to know how to stretch its tentacles to my roof.

Individualism was expressed by the French philosopher Rene Descartes (1596-1650) as: “I think, therefore I am.” But every person lives in a society and interacts daily with others, so that society thinks in ways we have not been able to decipher well, even with advances in social sciences.

Society and civilisation have grown because we are enriched by interaction with different peoples, cultures, geographies and natural environments. Life is definitely a whole that is more than the sum of its parts. Any interactive relationship is non-linear and complex. Interactivity explains momentum trading in stocks. We are influenced by those we trust. If they believe the market is going up, it becomes self-fulfilling.

The right wing ideologue who argues for individualism and claims that “the whole is the sum of its parts” is like a dead clock that is right twice a day; the rest of the time, it is wrong. If we believe in science and humanity, then we should embrace openness and diversity. We cannot turn back the clock, because “we connect, therefore we are.” It is from the richness of interactivity that we discover our interdependence, interconnectivity and social good or bad. Life changes by means of interactive loops between opposites.

That is the true meaning of condivergence: We are both growing and dying at the same time.


Tan Sri Andrew Sheng writes on global issues from an Asian perspective

Save by subscribing to us for your print and/or digital copy.

P/S: The Edge is also available on Apple's AppStore and Androids' Google Play.