Rikard Lundgren (SteenDier): The danger lurking in the fat tail
There are lies, damned lies and statistics – Mark Twain
Rikard Lundgren, CEO of advisory firm SteenDier, warns of the dangers of being seduced by high probabilities while ignoring the consequences of rare events.
A lesson about probability vs. consequences
In my last operational role, I was the chief investment officer of a seeder fund sponsored by Dutch pension fund manager APG. We were approached by over 1,500 hopeful fund managers looking for an investment. I interviewed 400, and eventually we invested in the strategies of 12 teams. Some strategies were based on statistical methods and models. After a very slick presentation by one such team, my only question was: “Your strategy is based on a statistical model that delivers positive returns 97% of the time. But what happens in the other 3% of the time?” The managers insisted that positive outcomes 97% of the time is a remarkable result and deserved our investment. This dialogue followed. “Let’s say you drove here and found an illegal parking space just outside my office. Everyone knows that Amsterdam has very few traffic wardens, so the risk of getting caught is low – let’s say 3%. So you parked illegally. But what you didn’t know is that here illegal parking is punishable by death. The risk is still only 3%, but how do you feel now about parking illegally?” “That’s different! You did not give us all the information!” “Just like you haven’t told me what can happen in the 3% when your investment model doesn’t work. If I don’t know the worst scenario, though highly unlikely, we won’t invest.” The wannabe fund managers hastily gathered up their presentations and left – perhaps to check on their car.
A better, more holistic understanding of risk
To make robust decisions we need to include the consequences of less likely events, the fat tails. We should never be satisfied with 97% positive outcomes without a full understanding of the consequences hiding in the much less likely 3%. Risk philosopher, options trader and author Nassim Nicholas Taleb wrote: “Never compare a multiplicative, systemic and fat-tailed risk with a non-multiplicative, idiosyncratic and thin-tailed one.” The investment team tried to sell us a strategy that delivered positive returns 97% of the time but they didn’t want to tell us about what consequences could be lurking in the fat and convex tail of negative return events.
Skin in the game
We accept far too many decisions based only on high probability outcomes, especially when they are taken for us by others. Taleb suggests that individuals or institutions that take decisions based on probabilities should always have skin in the game – like pilots who fly the aircraft in which we, the passengers, travel. This makes it more likely that they will apply a more holistic approach to risk. Even pilots may miss some 3% events, so we still need to think for ourselves. 97% of scientists say we are facing a climate crisis. Politicians say we must accept restrictions on our way of life, consumption, travel and investments. But what happens if they are wrong? What happens to relative prosperity if some countries include the less likely 3% scenarios while others ignore them?
Do 97% of experts act as pilots?
Have they sold their beachfront properties, stopped taking exotic holidays or eating meat? Do they only have green investments, shower less often, reduce indoor heating in winter and wear cardigans? As with the investment team, questions about consequences of the 3% are likely to meet resistance. But as good risk managers will agree, we need to insist on a holistic risk and consequences analysis to ensure decisions are robust also for the 3% scenarios. Ignoring the consequences of the unlikely leaves us ill-equipped for the damage this could do to us and our future.
The hopeful investment team’s strategy, that generated positive returns 97% of the time, included selling put options on equity markets. This generated steady fee revenues in normal markets but astronomical losses when equity markets fall dramatically – as they tend to do now and then.