Substack

Tuesday, May 28, 2019

The note of caution on claims of alpha

All high-level decision-making (as an investor or a sportsperson or an administrator or a CEO) are essentially (well-informed) exercises in human judgement. They are very rarely exercises in logic spinning or theoretical reasoning. 

I have blogged earlier alluding to exercise of good judgement as mark of wisdom and perhaps the most important requirement of decision-making. But such exercises of judgement are inherently inconsistent. For instance, your judgement on the same issue can vary widely depending on a variety of factors, including your state of mind at the decision-making moment

We deeply under-estimate this reality and attribute outsized performances (or alphas) to some superior human trait, one which in case of superstars is deemed superior to that of everyone else. This is among one of the most misleading narratives of our times. Instead a more prudent approach may be to look at such performances from a Bayesian perspective.

Tim Harford points to Michael Blastland's recent book, Hidden Half, to highlight the potential role of good luck in explaining what can appear to be alphas. This is important,
... in a competition in which all the leaders are highly skilled, randomness may explain the difference between triumph and failure. Good luck plus skill beats bad luck plus skill any time.
In other words, for any sport or leadership position or market, there is a supply-side of such players or leaders or companies, which is hardly a handful but reasonable sized (in case of businesses, the numbers are likely to be reasonably large). From among them, one emerges a winner in the particular race. And, this is most likely the result of pure good luck. 

This assumes significance when we assess the true worth, as reflected in the executive compensation of Chief Executives and the so-called disruptive genius of internet companies like Facebook or Google or Amazon. It is not incorrect to argue that the confluence of eco-system enabling factors that created the conditions for social media, internet search, double-sided marketplaces like aggregators or e-commerce etc made the emergence of such behemoths inevitable at the turn of the millennium. The present set of winners emerged from among several equally placed competing companies. If there was perhaps one out-sized contributor to the winner's success, it was plain good luck. Much the same logic applies to today's superstar chief executives. 

But once they become successful, these leaders or companies or sportspersons enjoy the benefits of Mathew Effect (see this, this, and this) that is inherent to the dynamic of all the systems where they operate. This Mathew Effect raises insurmountable entry barriers, even on the same cohort of competitors when the next race (for an organisational position or a sporting event or market position).

One of the biggest concerns with present day capitalism is that its dynamic has amplified Mathew Effect, thereby entrenching winners. In fact, such Mathew Effect even conceals the deep deficiencies and incompetence in such winners and perpetuates a very inefficient market. In a broader sense,  there is so much evidence piling up that access to life's opportunities itself is increasingly dependent on the ovarian lottery

No comments: