R.B. (Biff) Matthews is chairman and Doug McCutcheon is president of Longview Asset Management Ltd., a Toronto-based investment management firm.
Studies of investment returns reveal a startling paradox. Over time, the vast majority of professional investors will produce an after-fee return that is below that of the broader stock market. However, certain financial formulas relating to public companies can be used to produce substantial outperformance over the long run.
Why is this and how can we learn from it?
To back up a little, the distinction here is between two schools of investing – usually referred to as fundamental and quantitative. While the approaches are radically different, it is possible to combine the best aspects of each.
Fundamental investing is what most investors, including professionals, do. You examine a company, often in depth, predict how the business will do in the future, estimate whether the company is undervalued or overvalued by the market and buy or sell accordingly. Success depends on your ability to predict the company’s future profits better than the other market participants.
Quantitative investing typically involves a large portfolio and high turnover. The approach is somewhat counterintuitive. It does not rely on predictions of the future. Derived from academic research, and enabled by large amounts of data, it is based on the proven correlation between certain financial information about companies (referred to as factors) and higher investment returns. Most of this financial information is readily available on the internet.
To take a well-known example involving a value factor, many studies have shown that if you had, 50 years ago, simply measured the ratio of price per share to earnings per share (the P/E ratio) for each company in the stock index and then bought only the 10 per cent of companies having the lowest P/E ratio – and adjusted your portfolio once a year in accordance with this rule – you would have substantially outperformed the market over the long run. If you had added a few other factors that have also been proven to coincide with higher returns, you would have done even better.
So why doesn’t everybody invest using only quantitative factors? In part, because human nature gets in the way. As a species, we are not comfortable thinking in terms of probabilities. We would much rather listen to stories than study numbers. In addition, we are stunningly overconfident in our ability to predict the future. Most people believe that they can guess, better than most, how a business will perform. (About 90 per cent of people also think they are better than average drivers.)
There are also some valid reasons for adopting a fundamental approach. By focusing your portfolio on fewer companies, you are more able to understand both the underlying businesses and the management teams. And minimizing portfolio turnover, which is not possible with the quantitative approach, will effectively reduce the rate of tax you pay on your investment return. This saving is a guaranteed win, a rare find in the probabilistic world of investing.
To date, the quants have far outperformed most fundamental investors, at least on a before-tax basis. We expect that to continue. However, the fundamental versus quantitative question need not result in an either/or choice.
Fundamental investors can, without radically changing their approach, use tools from the quantitative toolbox, thus incorporating the best of both approaches. For example, they can screen for both value and quality factors in order to identify the companies on which to spend their valuable research time. And once the fundamental research is complete, they can again consider those factors before ultimately buying or selling. This brings more objectivity to the investment process.
Successful investing is all about putting the odds in your favour. Even small improvements in the odds make a huge difference over time. Using quantitative factors will greatly increase the chance of some fundamental investors outperforming the broader market over the long run.