Ninety One: Lessons from a desk clean

 

One of the joys of an end-of-year desk clean is finding things you’d thought you had lost or that you hadn’t got around to reading. It is a reminder that what may appear important at the time of publishing can quickly become irrelevant, and that what may seem trifling can turn out to be quite interesting. My housekeeping turned up papers from the arenas of sport and politics, as well as from more familiar territory, with useful lessons for investors today.

In October 2018, Cliff Asness and Aaron Brown of AQR Capital Management published ‘Pulling the Goalie: Hockey and Investment Implications’. The paper’s title refers to a strategy in ice hockey in which the goalie is replaced by an outfielder, thus giving a numerical advantage in attacking players at the cost of leaving the goal undefended. Asness and Brown highlight that a number of papers have been published on this subject over the years, all of which agree that goalies should be removed earlier than is the usual practice.

 

Fans and commentators regard pulling the goalie as high risk. This is correct if risk is measured in number of goals conceded. However, in reality, the major risk for a coach is not scoring sufficient points. The research is clear: pulling the goalie does indeed result in more goals conceded, but it also results in more league points due to losses being turned into draws or even victories. In other words, pulling the goalie reduces the risk of losing a game.

Asness and Brown compare the misunderstanding around pulling goalies to focusing purely on the risk of an investment, rather than focusing on what the investment does to the risk (and return) of an overall portfolio.

Asness and Brown’s second investment lesson from hockey arises from analysis of why coaches don’t pull goalies earlier. They suggest that coaches are rewarded for being perceived as good coaches, rather than for winning. Consequently, pulling the goalie and suffering some bad defeats is worse than not pulling the goalie and suffering a bigger number of mild defeats (but missing the opportunity for turnarounds). In other words, coaches consider failing conventionally to be better for their long-term employment prospects. A contrarian investor may draw some parallels.

Peter Dolton and George McKerron of the University of Sussex used some huge datasets to track how the outcomes of football (soccer) matches affected happiness. Not surprisingly, they discovered that the negative effect of a loss on a person’s happiness was much larger than the positive effect of a win. Losing was particularly painful when expectations of a win had been high, though an expected loss still induced unhappiness.

It is pleasing that academics have discovered that loss aversion also applies to sports fans — mirroring the findings of Tversky and Kahneman, who informed us that people prefer avoiding financial losses to acquiring equivalent gains, a fact often used in investor presentations to justify selling losers. Of course, such irrational behaviour is music to the ears of contrarian investors.

 I copy here, without comment beyond that suggested in the subhead, a fragment of a speech by US President Herbert Hoover six months after the 1929 financial crash. It was highlighted in academic Dave Collum’s review of 2019.

‘We have for the first time attempted a great economic experiment. Possibly one of the greatest in our history. By co-operation between government officials and the entire community … we have undertaken to stabilise economic forces, to mitigate the effects of the crash, and to shorten its destructive period. I believe I can say with assurance that our great undertaking has succeeded to a remarkable degree.’ 

I was wrist-deep in an unguarded box of Celebrations when a colleague sent me a note from Macquarie Research, updating its view on the ‘value v growth and quality’ debate. Its conclusion: ‘Value only [my emphasis] runs when there is ample liquidity, strong reflation, co-ordinated policies, declining risks and, hence, a weakening USD’; and that its view on these factors ‘implies that if there is any value rally, it is likely to be earlier in the year and will likely imitate a dozen minor ripples rather than the several more substantive shifts that occurred over the last decade’. 

Perhaps the authors are right that value can only work when a huge number of variables come together. However, the long-term outperformance of the value factor suggests something rather more is at play. So let’s look at a factor that we regard as even more relevant: how much an investor is paying at present for value and growth/quality stocks. The charts in the paper, which show that growth/quality stocks remain extremely expensive relative to value stocks, speak for themselves.