“About these matters there is no scientific basis on which to form any calculable probability whatever. We simply do not know.”
John Maynard Keynes,
The General Theory of Employment, Interest and Money
In 2019 how many investors were factoring into their models the risk of an infectious disease resulting from a virus which did not yet exist? Moreover, even if they had, what chance an investor would correctly anticipate the governmental response and the full recovery of the S&P 500 within six months? We venture very few.
Thinking about the problem of forecasting has been illuminated by a wonderful book by John Kay and Mervyn King entitled Radical Uncertainty: Decision-Making for an Unknowable Future published in March 2020. The basic thesis of the book is that market participants have misled themselves regarding risk and uncertainty. In essence, probabilities and probabilistic reasoning are faulty when it comes to uncertainty.
The authors describe how the debate about probabilistic thinking dates back to the invention of basic probability by Fermat and Pascal in France in the 17th century. They devised models to help gamblers in games of chance. Before that people thought that random events were either acts of God or were part of narrative suppositions. In 1921, Keynes questioned the standard thinking in his Treatise on Probability, followed by Frank Knight in Risks, Uncertainty & Profit. The latter made the distinction between risk and uncertainty (i.e. known knowns, and unknown unknowns). His successor at the University of Chicago, Milton Friedman, argued that the distinction was not valid and the Friedman view has been prevalent ever since, underpinning modelling in financial markets and other areas like climate change.
This reasoning led to what the authors call the Viniar paradox, named after David Viniar, the former Goldman Sachs CFO during the financial crisis who described the impact of Lehman as like experiencing several 25 standard deviation events in a row. This was clearly nonsense. What he really meant was the events were impossible within the framework of the Goldman risk model. The problem was the model.
Similarly, during the hunt for Osama bin Laden, Barack Obama was told that there was a 25 to 95% probability that the man in the house under surveillance was actually bin Laden. Obama is said to have replied, “so it’s a flip of the
coin isn’t it?”. In this sense, the probabilistic approach conceals rather than illuminates uncertainty.
Kay and King argue that we need to restore the distinction between risk and uncertainty. They propose that we distinguish between “resolvable” uncertainty (where more data allows you to formulate probability distributions) and “radical” uncertainty where this
is simply not possible. Covid-19 is a paradigm case of radical uncertainty and the book (written pre-covid) presciently identifies the risk of a pandemic caused by a novel virus. Covid is thus not a Black Swan in Nassim Nicholas Taleb’s sense of being an event which you cannot imagine (an unknown unknown). Rather a radical uncertainty can be one which you can describe but not define clearly in terms of its impact (a known unknown).
Investors face radical uncertainty every day and yet many still endeavour to pursue the Friedmanite course, seeking comfort in detailed financial models. They could be seen at pre-pandemic investor conferences making minor adjustments to detailed spreadsheets as companies presented themselves, using the opportunity to quiz management on the assumptions in the model. “What is your expected tax rate for the year ahead?”, being a common question.
As Kay and King argue, this type of behaviour reflects physics envy and is based on the notion that all you need is enough data and you can build a model and value stocks in a precise way. If NASA can send a probe to Mercury and predict where it will land after seven years and travelling five billion miles, surely one can predict a company’s profit in one year’s time?
The fallacy behind this reasoning stems from the distinction between stationary and complex systems. The laws of physics apply well enough for NASA to make predictions because the system has a stationary equilibrium. Conversely complex systems like financial markets are not stable since you cannot really understand the underlying structure – in any case it changes over time and crucially it changes when you interact with it. Think of GameStop raising $1bn to fund its transformation thanks to a short squeeze mounted by retail investors intent on humiliating hedge fund managers. Who predicted that?
So how should we deal with radical uncertainty? The authors recommend that we should accept that there are limits to our knowledge so that when we formulate a reference narrative as the basis of planning, we need to build buffers to control risk. We should try to think in scenarios of what could go wrong, based on an understanding of historical precedent. The parallel is the concept of modularity and how engineers
manage complex systems which allow a single part to fail but not affect the system as a whole. They also build in redundancy. For the last 20-40 years (the “era of optimisation”), redundancy and modularity have been regarded as signs of inefficiency. Yet in the long run, this has jeopardised the system.
At the same time, it is important to embrace uncertainty as a source of opportunity for further development. Uncertainty can be a good thing – life would be joyless if events were always predictable. For investors, some markets appear to offer more
volatility than others, and hence more optionality – although even here appearances can sometimes be deceptive. Think of apparently stable bookselling before the Internet or catering pre-covid.
The lessons for Marathon-London are twofold. First, try to assess the managers of companies and the culture they build around them as a guide to how they might respond in the face of radical uncertainty. Many of our more successful long-term investments have been in firms with a singular focus and dedication to maintaining competitive advantage which have exploited opportunities as they have presented themselves. Consider Flutter Entertainment, the world’s leading online gaming company which started out as a retail bookmaker in Ireland and was an early entrant in the online gaming market. Over the years the company built a strong position in the UK online market as well as seizing opportunities to enter new markets that were opening up, notably in Australia and more recently in North America. When we first invested in the company in 2002, we had no idea that the global market would evolve in this way and a discounted cash flow forecast in a spreadsheet made at that time would have woefully underestimated the potential. What we could say is that we had found a team who were dedicated to success with distinctive assets in a field ripe with opportunity.
Second, construct an investment process that recognises the limits of one’s knowledge and incorporates modularity. A highly concentrated portfolio predicated on a confident view of the future is vulnerable to radical uncertainty. As Mr. Viniar can attest, the danger of complex spreadsheet modelling is that it can build confidence without improving foresight. Sometimes we just need to be humble and accept, “we simply do not know”.
Learn more about our insights & how your Harbor representative can help.
Performance data shown represents past performance and is no guarantee of future results. All charts and data are for illustrative purposes only. Views expressed herein are those of Marathon-London Asset Management, LLP and may not be reflective of their current opinions or future actions, are subject to change without prior notice, and should not be considered investment advice. The information provided in this presentation is for informational purposes only.
The information provided in this article should not be considered as a recommendation to purchase or sell a particular security. The weightings, holdings, industries, sectors, and countries mentioned may change at any time and may not represent current or future investments. Investing entails risks and there can be no assurance that any investment will achieve profits or avoid incurring losses.
Harbor has engaged Marathon-London as a subadviser to one or more Harbor sponsored products.
