Two Towers of Risk Perspective

I am attentively watching the novel coronavirus outbreak from its relative start in the first half of January, 2019 and I cannot help but think it is, beside a great tragedy, a deep well of risk-related and system complexity related lessons. It is also a large window showcasing the operations of governments and large institutions, as well as media and, lastly, the human mind.

When I talk about risk, I am mainly talking about payoffs, about the consequence. “Risk” can be easily aligned with probability of a given scenario taking place, however this is a fragile approach, because firstly the computation of the probability might be wrong and secondly one may underestimate the consequence simply because the “probability” of it happening is “very low”, hence it is a “low risk” event. Focusing on the payoff brings clarity into risk management; the truly important aspects begin to stand out The essentials that need to be secured and preserved are easy to discern from the expendables that can be “gambled with”, for the lack of a better term.

When it comes to risk, risk assessment and choosing how to respond, it is good to bear in mind there are several classes of risk that a given entity is exposed to. The basic division is on multiplicative risks and non-multiplicative risks. Multiplicative means that an occurrence of an event increases its chance of occurring again. When a person gets infected with an infectious disease and comes in contact with other people, he increases the chance of the disease multiplicating onto other people.

On the other hand, a non-multiplicative risk is a risk that is contained to either an individual, a small group of people or a location. It is, in other words, contained. When a persons has a car crash, he may subsequently cause a crash of numerous other cars, but in the end the number of crashing cars will be limited. Furthermore, if somebody crashes into your car and your car is later undergoing repairs, your chance of getting into a car crash has actually decreased. If you can’t drive, you can’t crash.
We can see that certain risks are sort of self-containing.

Secondly, the risks can be differentiated by their consequences (mild, causing discomfort versus serious – when you die it’s over) and time scale; some risks happen in a one-time event, while others are stealthily compounding over long periods, delivering pay-offs much later (months, decades, centuries).

These classes can then be combined. There are:

  • mild contained risks to individuals in a one-time event or a long period
  • serious contained risks to individuals a one-time event or a long period
  • mild multiplicative risks to societies over a time period
  • serious multiplicative risks to societies over a time period

Each class of risk (and this is only one of the possible divisions) requires a different kind of response. Moreover – and this is crucial – each class can be countered on various scales with various effectiveness. An individual does not have necessary tools to respond to multiplicative risks to society – governments do – but he/she is much better at managing risks of his/her own behaviour – much better than any government can.
I would therefore argue that governments and institutions should primarily focus their attention and resources to events threatening the survival of societies, whilst leaving it up to individuals to manage the contained risks emanating from their & their fellow citizens’ behaviour.
However, what see in the response to coronavirus outbreak is usually a government official choosing a response based on the individual-contained risk point of view, rather than the societal multiplicative risk. It is the argumentation that crossing a street is “statistically” more deadly than a virus outbreak, hence we should worry more about street crossing. I hope the reader is beginning to spot the nonsense.

The individual asks: What is the danger to me? What could harm me badly? What could harm people around me?
The government or institutional officer ask: What is the danger to the population? What could harm the society badly?
Note: by a population I mean that of a city, region, country or of the world.

You could think that both cases are the same and there is no difference, but you would be wrong. Both cases are different even though they sometimes tend to overlap, because life is dynamic and not static.
I think the greatest danger to population lies in multiplicative risks with serious consequences. Borrowing an example from Nassim Taleb’s video on Naive empiricism, terrorism and Ebola, if you wake up one day and read that 100 million people have died in say the past week, what could be the cause? Car crashes? Falls from ladders? Or a terror attack or a virus/bacteria spread? It is the multiplicative, high-consequence process that can deliver these results. That is why governments and certain institutions shall be primarily focused on these. For an individual, depending on specific circumstances*, the multiplicative process may be less relevant, but for the ensemble it will always prevail.

When it comes to nCoV, occurrence of which hasn’t been confirmed in Czechia at the time of writing, I can take precautions, I can alter my travel plans (I have) or scratch them entirely. I can save up money in case my workplace gets shut down (one should do this, of course, way before any emergency shows up), but I don’t see the point of worrying as an individual. I would be running the risk of worrying about coronavirus while stuffing my body with processed foods and sugars, for example, setting myself for a deadly outcome decades hence.
That being said, I believe the government should be worried and I am worried because they are not worried. Their response has been slow and inadequate, because – just like media and most people – they compare something unknown, unmeasured and possibly dangerous – to something well known and measured, so the can downplay the affair and stick to doing very little. They confuse the classes of risks and they do not understand that multiplicative, high-consequence risks are in today’s world, which is the most connected it has ever been, *the* risks that require quick and serious precautionary reaction, which cannot be – by definition – supported by data. When one makes a decision the data to back it up is not there and when the data finally shows up, it might be too late.

The divisions of risks and their importance to individuals and populations are not, I repeat, static – they change over time and place. The underlying principle in the paragraphs above is that survival is what matters – without survival, there is no second attempt, there is no round two. Once you focus on survival**, the risks attitude changes. Suddenly, taking certain risks that are not threatening with finality is less stressful, because the downside is limited while the upside is not, and at the same time saying no to risks that could bear high consequence gets easier, because it is simply not worth it.

* I cannot stress this enough – if a person is in Wuhan, then the multiplicative process in front of his eyes – the nCoV – becomes way more important than say diabetes. This shall be self-evident

** I will forever remember the unimaginative university professors who couldn’t make the distinction that if you survive, you can enjoy life, but when you are dead (or bankrupt – for companies), you can’t. They always understood survival as “barely surviving and staying there.”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

About Petr Klíma