1st Feb 2019

If we begin with certainties, we shall end in doubts; but if we begin with doubts, and we are patient with them, we shall end in certainties – Sir Francis Bacon

Fat tails

Perhaps the most challenging aspect of risk management is accounting for uncertainty. Unlike risk, uncertainty implies an inability to determine the likelihood and/or impact of a future event. As risk managers, our role is to transform uncertainty into a probabilistic assessment of risk. In truth, this isn’t always possible, and one must be prepared to wear some level of uncertainty.

Conventional statistics offer a false sense of security. Many investors have been conditioned to think about risks and probabilities in terms of normal distributions, or bell curves. These distributions imply that catastrophic events are extremely unlikely to happen. The reality is that these “rare” events – credit crises, sovereign defaults, currency devaluations, terrorist attacks, wars, and political upheaval, to name a few – occur with much higher frequency than statistics would imply. 100-year climatic events, for example, happen far more often than once a century. In financial markets, there is enough empirical support to suggest distributions that measure the likelihood of risk and their impact bulge around the edges, creating “fat tails”. Fat tails tells us that even though we might be capable of assigning probabilities to regular events, we’re especially bad at predicting the irregular.

Read More >