COVID19 Precautionary Principle

Watched a short conversation between Taleb and Yaneer about the current situation with COVID19 pandemic and why precautionary principle applies here

Original thread:

I absorb information better by reading and writing. So here’s the loose transcription and note I take on it

Text in italics are my thoughts.

  • Opinions from people who don’t understand scale. Doctors are used to treating person by person. not a community, and the world at large.
  • Incerto, decision theory, asymmetry of uncertainty (domain of risk management).
  • Antifragility, complexity on the one side. Better to mistake a bear for a stone or a stone for a bear? You’d rather make a mistake on the one side than the other. When you have uncertainty and absence of evidence, it should lead you to more precaution.
  • Precaution principle applies to systemic, not idiosyncratic (individual) levels. You just put your seatbelt on, you don’t ask for evidence of a car crash. But at a systemic level, you should only worry about risks that are explosive, fat tail (not normal distribution, e.g. exponential). To identify that early. That tail risk.
  • To wait for that uncertain thing and to see it coming at you (certainty)
  • Jan 26th, got the warning out. Dismissed by many because it’s too costly. Airlines, govts. Recommended: constraining mobility, lowering connectivity. Basically distancing.
    • Just a big “told ya”. And now everyone is playing the blame game at the people in power / in charge for not listening (hindsight bias).
    • People don’t want to make mistakes. And the system does not favour people making smart bets. See outcome bias (Pete Carroll story that Annie Duke shares in her book. Good process bad outcomes)
  • Insurance costs money, precaution generates money. You hedge.
  • The precautionary principle should only be used in non trivial matters (unlimited downside, limited upside)
  • Risk requires understanding asymmetric evidence. Absense of evidence is not evidence of absence (black swan)
  • People who think strategically and understand the essence of what it means to face a catastrophe (why SG is well prepared for this, they have planned and hedged against this before. See Related)
  • Convexity: economic cost.
    • When the problem is exponential while the solution is linear, no matter how much we prepared we won’t be “ready”. But we could manage it to certain extent, reduce the damage.
  • There is problem in the minds of people who know enough statistics about it to be able to talk about it to humanity student but not enough to understand what these numbers mean and how it can be used in real life. Your grandmother who doesn’t know statistical framework, immediately understand “do not drink the water if you suspect it is poisoned”. You just don’t see reality, blocked by your “knowledge”.
    • Same idea: after one year of training you could learn enough martial art to get yourself into trouble.



  • Peter Ho
  • Black elephant
  • National scenario planning

The black elephant is a problem that is actually visible to everyone, but no one wants to deal with it, and so they pretend it is not there. When it blows up as a problem, we all feign surprise and shock, behaving as if it were a black swan.

Scenario planning helps to inculcate an anticipatory mindset in planners and policymakers so that they instinctively raise “what if” questions on the issues they deal with. It helps them to overcome their blind spots, and to confront or at least be aware of black elephants.

National scenario planning for SG

What a dream job.

Also published on Medium.

Leave a Reply

Your email address will not be published. Required fields are marked *