A huge pile of research points to an array of simple numerical stock strategies that boosts returns over the long term. Such methods range from low-ratio value strategies to momentum-oriented schemes.
If the studies are to be believed, it should be easy as pie to make a small fortune on Bay Street. (Even when you don’t start with a large one.) All you have to do is to select a strategy and stick with it, letting the cold, hard numbers determine your buying and selling for you.
Picking a good long-term approach is one thing. But actually following it for a long period is quite another. It turns out that there are more than a few devils in the practical details.
I was recently reminded of a big one when I watched Tobias Carlisle’s informative presentation to the UC Davis MBA Value Investing Class on quantitative techniques. He discussed James Montier’s suggestion that numerical methods might represent a ceiling to potential returns, rather than a floor, for those who tinker with them.
Mr. Montier looked at how well a wide variety of simple models in many non-financial fields, ranging from medicine to criminal recidivism, worked. In nearly every instance, the models outperformed the experts.
For instance, he pointed to research by Leli and Filskov on diagnosing brain damage. For this purpose a model was developed that correctly identified 83 per cent of new cases, whereas professionals working from the same data got it right only 63 per cent of the time.
Thankfully, when you arm the pros with the model their performance improved. But here comes the kicker: They still only managed to get it right 75 per cent of the time. Yup, despite knowing the model’s results, they failed to match it, let alone beat it.
Such findings are widespread in many fields. A meta analysis by Grove et al showed that simple models beat the experts in 64 of 136 studies, while another 64 cases basically resulted in a tie. A mere eight clearly went to the humans, and in those cases there were indications that the experts had access to more information than was provided to the models.
Based on such wide-ranging observations, Mr. Montier suggested that finance likely suffered from similar problems. In other words, investors might get the best performance from following simple stock screens to the letter, rather than using them as a short list to pick and choose from. The extra intelligence brought to bear on the problem might actually hurt returns.
Given my propensity to use screens, such considerations have been much on my mind for several years. It’s also why I tend to equally weight portfolios (by dollar value) when screening, and generally opt for all the stocks a screen uncovers, rather than picking and choosing.
Mind you, the temptation to be more selective is quite strong and there are more than a few institutional pressures that can push people, such as your humble scribe, down such a path. After all, when someone asks you for your favourite stock, he or she generally doesn’t want to be given a list of 20.
As it happens, money manager Joel Greenblatt ran an unintended experiment that suggests that stock investors suffer from the problems Mr. Montier warned about.
Mr. Greenblatt is the author of The Little Book that Beats the Market, and he espouses a mechanical technique called magic formula investing. His method screens for stocks with high earnings yields and high returns on capital. It buys, say 20 or 30 such firms and holds them for a year. (In practice, purchases are spread out over the year.) Although there is a little math and accounting involved, it isn’t a very complicated approach.
Mr. Greenblatt manages money based on the method and offers investors two types of accounts. In one type, investors follow the method automatically. In the other, they start with his list and then pick individual stocks they want to buy.
Before jumping to the results you should be aware that the accounts have only been up and running for a few years, which is too short a period to yield a definitive answer. Nonetheless, so far things don’t look good for the humans.
From May 1, 2009, to March 31, 2012, the strictly mechanical accounts achieved total gains of 84 per cent (after expenses) versus 63 per cent for the S&P 500. That’s not bad. But the average returns for the accounts that were more selective – those investors who bought some of the magic formula stocks but not all of them – only gained 59 per cent. The extra effort was rewarded with substantial underperformance.
Such tentative evidence bolsters James Montier’s case. Simple numerical strategies may well represent a ceiling to returns for most investors. It’s something that’s well worth considering the next time you screen for stocks.