Test win, historical probability and Donald Rumsfeld

Surprises keep piling up, after the sudden regime collapse in Afghanistan, on Monday came the Indian test victory where the wagging tail created a hurricane. Though it is highly preposterous to equate a life and death crisis to a game, the surprise/shock elements need to be further debated. When in markets such surprises start occuring on a very regular basis we term it as that it has entered into a highly volatile zone and in such an environment the expectation of reward goes up commensurately. This in fact is fine, the problem occurs when an apparent lack of activity dulls our senses in believing that things are under control and we start making bets which are unadjusted for risks. We would today discuss some thought models useful for handling surprises or shocks.

In 2002, the then US Defence Secretary presented his views on the above phenomenon as he saw it manifesting it in the real world. He said that possible future states and the associated risks/uncertainty can be divided and understood with the help of a "known unknown" matrix. Rumsfeld said that there are known knowns, these are things which we know we know. Then there are known unknowns, these are things we know that we don't know. And finally there are unknown unknowns, these are things we don't know that we don't know. This is quite a mouthful, let us explain in some detail.

Known knowns are simple to handle, we know the event and we know the probability associated with them. Simplest example is a roulette table or a dice throw, the probability of any future state can be correctly modelled as the possible states are known in advance. Then comes the known unknown, where we know the state but then there is no way to know the exact probability of its occurrence. Let's park it for some time. The final is unknown unknown, it is actually the sea of ignorance, we don't know what can hit us. The only clear example I can recollect from history might be the impact of Spanish arrival on the shores of Inca/Mayan civilization. Prior to the arrival the locals were not even aware of the force of nature that is going to hit them in future. In current context I can't pinpoint what can be possibly be unknown unknown. The endeavor of modern science and technology is to bring as much of this unknown to the known category with the advancement of knowledge.

Now we will come to the important and relevant part, in finance and in life in general, we operate in the domain of known unknowns. We know the event can happen but are not sure of its probability. The whole quest is to assign or fit a probability to the event. We are no longer in Las Vegas where the russian roulette has six chambers, here the number of chambers can be infinite. The common refuge hence is history, how many times the event has happened in history. How many times a tailender has scored a 50 on the fifth day, that will be our best estimate of the same occuring today. How many times an entity which produced stellar profits has turned out to be fraud. How many times a small riot turned into a full blown civil war etc. The problem is that these historical references are incomplete. That is why many once in a hundred year events happen on a yearly basis.

Readers can now sense the drift of the argument, historical probability can not be trusted fully, adjustments and revisions need to be done. Author Margaret Hefferman introduces us to some solutions in her book Uncharted: How to Map the Future. Hefferman stresses on the importance of thinking in terms of scenario building. Distill down the possible states in future to 4-5 scenarios, taking into account not just historical data but the current subjective assessments also. For example, let's say the possible question is the "state of European union in 10 years". Then the spectrum of scenarios will be "complete collapse of EU" to "EU as the sole world super power". She writes that once we leave the false dichotomy of black and white and embrace the scenario approach, a wider, more robust discussion can happen. When we see the different Central banks publishing fan charts for their estimates on interest rates and currency, the same approach is at play. We are living in times where history is getting created daily, it is better to be circumspect about what we know and what we don't, surprise can come from anywhere.