“Radical uncertainty” indicates regulators should stay away from “Bayesian subjective probabilities”

Sir, John Kay in “The other multiplier effect, or Keynes’s view of probability”, August 15, writes that “the largest and most famous Dutch book… a set of choices such that a seemingly attractive selection from it is certain to lose money for the person who makes the selection… would be the collection of ingenious structures products RBS acquired when it bought ABN Amro”. 

Forget it! That book, as a Dutch book, does not even come close to Basel II regulations. Those regulations, which offered a world without bank crises, set the bank capital requirements lower when the perceived risk were lower, and thereby doomed the banks to overdose on perceived risks, and create extremely dangerous and obese exposures to what was, and is, officially deemed as “absolutely not risky”. 

Kay mentions the possibility that one has to use “Bayesian subjective probabilities… because if they did not, people would devise schemes that made money at their expense”. That might or might not be true, but, at least, when it comes to bank regulators, the best is for them not even to engage in any sort of risk arbitration. One single capital requirement for any bank asset is the only rational response to any “radical uncertainty”.