Avalanche Accidents and Investment Risk

What does the assessment of avalanche risk have to do with how we make investment decisions?  At face value very little; the stark and exposed nature of the former could hardly seem further removed from detached and abstract financial markets.  Yet there are parallels.  Both situations require us to make complex choices in uncertain environments, potentially exposing us to significant threats.  Furthermore, in order to deal with the profound uncertainty we are prone to apply unsuitable mental shortcuts or be unduly swayed by our behavioural biases; often with damaging consequences.

I was drawn to the avalanche risk analogy by a recent article by Heidi Julavits in the New York Times Magazine where she described her experiences at an avalanche school[i].  This led me to research carried out by avalanche expert Ian McCammon[ii].  In a paper, published in 2004, McCammon details what he believes to be the main decision making errors that lead individuals to expose themselves to significant risk of an avalanche accident.

In the paper he reviews 715 recreational accidents between 1972 and 2003, and attempts to categorise the central causes of the incidents as a range of ‘heuristic traps’.  These are rules of thumb that are often useful in everyday life but deeply flawed if applied in an incorrect situation.

McCammon’s research was designed to understand why so many accidents occurred when (in hindsight) the risks were vividly apparent.  His description of this puzzle will resonate with anyone who has endured ‘difficult’ experiences in financial markets:

In hindsight, the danger was often obvious before these accidents happened, and so people struggle to explain how intelligent people with avalanche training could have seen the hazard, looked straight at it, and behaved as if it wasn’t there.”

I am not sure that McCammon’s use of the term heuristic is entirely accurate; most of what he describes are behavioural tendencies rather than rules of thumb.  Definitional matters aside, the similarities between some of the issues identified by McCammon and those experienced by investors are stark.

McCammon identified the following six ‘heuristic traps’:

Familiarity: In McCammon’s description, the problem of familiarity arises when we become complacent in our behaviour because of repeated exposure to a particular environment.  This can serve us well for the most part, but leaves us sharply vulnerable to change.

As investors we become rapidly conditioned to the prevailing regime; believing that the recent past will persist. Given how difficult it is to make forecasts about future states this tendency is understandable (our best guess about tomorrow might be what happened today). Yet shaping our behaviour and decisions based on what has worked in the past – whether it is about the economy, expected returns or correlations – can expose us to severe risks when the environment shifts.   

Consistency:  The notion of consistency utilised in the paper is similar to commitment.  Once we have made an initial decision or taken a particular view we find it difficult to reverse course and tend to ignore new information.

Perhaps one of the most powerful drivers of our investment behaviour is our inability to recant past views or acknowledge mistakes (as opposed to blaming others or circumstance).

Acceptance:  Acceptance is similar to signalling.  In a social context we engage in behaviour because of how we believe it will reflect on us; usually in order to gain respect or acceptance from others.

For professional investors in particular, the desire to be seen in a certain way by clients, colleagues, peers and superiors can have a profound impact on our behaviour.  It undoubtedly leads us towards complexity over simplicity (to prove our intelligence) and activity over passivity (there is a strong desire to be seen to be doing something).

Expert Halo:  We are often in thrall to experts and can overstate the value of their knowledge, or assume that their ability stretches far more broadly than it actually does.  We are also unwilling to challenge their views.

The randomness and uncertainty in financial markets makes identifying experts somewhat tricky.  We are frequently guilty of misjudging an individual’s circle of competence; if someone has outperformed the stock market for a number of years we are happy to hear them opine on any subject, no matter how tangential it may be, and may follow them into previously uncharted areas.  We are also quick to see expertise and skill, whilst dismissing the role of luck.   

Social Facilitation:  Related to the idea of acceptance detailed above; social facilitation refers to the manner in which our behaviour may change when in the presence of others.  In particular, McCammon states that our willingness to exercise our skills or take more risk is greater than if we were alone.

The idea of social facilitation does not simply mean that people are directly in our presence when making a decision; even the implied presence of others can influence our behaviour. Our actions are likely to alter if we believe that we are being evaluated or our performance being judged. Professional investors may, for example, make different decisions for their public funds then their personal accounts, other things being equal.

Scarcity:  The attractiveness of an achievement, activity or product tends to be greater when it is scarce.  The fear of losing an opportunity, particularly to another person, can often drive inappropriate risk taking behaviour.

Our willingness to chase performance or become involved in the latest fad or bubble is closely linked to McCammon’s notion of scarcity. We are often fearful of missing out on the outsized gains that have been (and may continue to be) enjoyed by others.

Whilst it is interesting to understand the potential behavioural drivers of the mistakes we make, it is also crucial to acknowledge that whether we are traversing terrain with avalanche risk or making an investment decision we are actively embracing risk and uncertainty.  That means there are always a range of potential outcomes (with a proportion of them negative).  In both domains there is a huge amount of randomness involved.  Although it is important to consider the robustness of our judgements, a negative outcome (even a deeply negative one) is just one of the paths across a distribution of possible results.  We can make prudent decisions and suffer from bad luck.

Towards the close of the paper McCammon states that:  “the challenge is to encode knowledge into simple, easily applied decision tools”.  This is the most vital point of his work.  Whilst the research is focused on ‘heuristic traps’ it is not a critique of quick and simple decision making, but rather a call to ensure that the quick decision tools utilised are appropriate for the situation.  In highly intricate and unpredictable environments, applying simple decision rules or heuristics can be invaluable. We just have to ensure that we are using the correct ones.

[i] https://www.nytimes.com/2019/12/31/magazine/avalanche-school-heidi-julavits.html

[ii] McCammon, I. (2004). Heuristic traps in recreational avalanche accidents: Evidence and implications. Avalanche news68(1), 42-50.