To answer that credibly, we need a proper vision of markets. It would be premature to suggest rules to survive in hostile territory, without a reliable map.

One map of financial markets assumes they accurately reflect economic fundamentals, captured in deductive models such as CAPM.  This suggests the best way to survive is to spread risk significantly (in the case of CAPM by diversifying across the full asset spectrum). Buffett described such investment techniques as the ‘Noah’s Ark school of investing’, and suggested that people attracted to the idea should concentrate on captaining arks, rather than managing money.

A more useful map is well encapsulated in a metaphor developed by the Belfast-born Brian Arthur, Professor at the Santa Fe Institute. His El Farol Bar Problem goes as follows. Imagine 100 people in a local community decide independently each Thursday evening whether or not to go to a pub offering Irish music. There is no sure way to tell the numbers coming in advance - the evening is enjoyable if things are not too crowded, specifically if fewer than 60 people are present. Notice that there is no deductive answer – potential attendees are propelled into a world of induction, trying to infer from past experience just how many people will actually show up in future. In this situation, commonality of expectations may have perverse effects. If nearly everyone believes few will go, nearly everyone will go! If practically all believe most will go, few will go, again invalidating the belief.

Asset management is primarily an exercise in induction, conceptually more challenging than the deductive map suggests. The territory is hostile, and the linkages between reality and participants complex.

Arthur suggests that in computer simulations of the problem, with attendees following their own rules and hunches to guess the likely numbers who will attend the following Thursday, the mean attendance actually converges on 60! To be clear, not all inductive situations gravitate to such a happy equilibrium. When customers of Northern Rock saw unanticipated queues outside its branches, people quickly joined in, rather than leaving their banking transactions for another day! Herd expectations can also lead to euphoria or panic, not equilibrium. As in the Northern Rock case, customers may flock to a bank should they sense it is in trouble, which can actually precipitate its collapse. Equally, if everyone thinks stocks are on the up, everyone buys and you may generate a dangerous bubble. This type of outcome has been repeated over history in the many bubbles and crashes that pervade financial markets. 

Now, what has all this got to do with asset management? Well, potentially quite a lot. Asset managers are fallible, uncertain about the economies and securities they are trying to understand, and about how other managers are thinking about these uncertainties. As in El Farol, their own actions may actually affect the reality they are trying to understand. Sometimes the consensus may imply an equilibrium – at other times it can ensure boom or bust.

So, asset management is primarily an exercise in induction, conceptually more challenging than the deductive map suggests. The territory is hostile, and the linkages between reality and participants complex. We shall look at some credible rules to guide investors under these circumstances in our next post – Noah’s Ark may not always be the right boat to board.