top of page
Writer's pictureRobin Powell

2020 and the danger of extrapolation



  The contrasts between the first ten months of 2020 and the month of November were vivid. Consider:  

— At the end of October, the year-to-date total return of the S&P 500 was barely positive (2.77%), but was well ahead of the returns of the S&P MidCap 400 (-6.63%) and the S&P SmallCap 600 (-13.06%). In November, the 500 performed very well (10.95%), but the 400 (14.28%) and the 600 (18.17%) did much better.

— At the end of October, the cap-weighted S&P 500 had outperformed the average stock in the index (the S&P 500 Equal Weight) by 8.10%. In November, the equal-weight index outperformed by 3.35%.

— At the end of October, the best-performing sectors were Information Technology (22.13%) and Consumer Discretionary (19.76%), while the forlorn Energy sector had lost more than 50% of its value in 10 months. In November, Energy rose by 28.03%, well ahead of the erstwhile leaders.

— At the end of October, the best-performing factor indices were Growth (16.89%) and Momentum (15.97%), while value-based factors had all suffered double-digit declines. In November, the S&P 500 Enhanced Value Index rose by 19.82%, handily outperforming the growth and momentum indices.

— At the end of October, VIX stood at 38.02, a high level. In November, VIX fell to 20.57, as expectations of future volatility declined.

  To summarise, there were

market reversals on numerous fronts

.  

The significance of these developments

All this may be interesting as a matter of short-term market commentary, but its importance lies in what it illustrates about the fragile persistence of relative winners in equity markets and the risks of extrapolation from the past. These risks help explain two major long-term trends: the consistent underperformance of active managers and the consequent flow of assets to index funds.  

A dangerous assumption

Why, we are sometimes asked, do some investors remain loyal to active managers, despite all the evidence to the contrary? One obvious explanation is the natural human tendency to assume that the past predicts the future. In most of life, this isn’t a bad heuristic; the past does predict the future quite often. In some ways our routine daily lives depend on this assumption. Although COVID has kept me away from London for nearly a year, I’m confident that the cars there still drive on the left — because they always have. Investors often make an analogous assumption — they assume that good historical performance predicts good future performance. And November’s results, in a microcosm, show why this assumption can be dangerous. Almost 

any extrapolation based on performance through October turned out to be wrong

; past performance did not predict future returns. Those who assumed that it would are likely to be disappointed with their results — although perhaps not disappointed enough to forswear active management altogether.  





  Picture: Łukasz Łada via Unsplash  

bottom of page