Factor Investing

Saturday, 30 August 2014

There has always been a heated debate between academics claiming that the market is efficient and it is immensely difficult for investors to outperform indices in the long run and those who argue that there are plenty of inefficiencies to be exploited. It would be reasonable to conclude that the market is mostly efficient, however there is a set of methods that help investor select stocks with above-average performance.

These methods are based on a certain logic and aim to select a portfolio of stocks that meet certain criterion. There has to be a fundamental reason as to why such portfolio should return more than the market in the future and why selected shares are priced attractively. A selection criterion for such a portfolio is called a factor.

A simple example of factor investing is a selection of shares with intrinsic value higher than their market price. This method had been pioneered by Benjamin Graham and Warren Buffet, arguably one of the most successful investors of all times. Even in academia Asness et al. and Sloan et al. conclude that the combination of cheap and quality shares has a good chance of delivering a solid return.

The chart below shows the performance of a portfolio comprised of shares meeting the cheap and quality criteria (the exact details of the selection algorithm will be discussed in the future article). The portfolio is rebalanced every single year (thus new cheap and quality shares are bought every year and old ones sold). What is the main factor that influences the performance of our portfolio? It is Mr. Market (as Benjamin Graham called it) - when the stock market declines our portfolio will follow. Apart from this logical correlation, it is important to note that all our shares are similar in their parameters and there is a significant correlation between them. These shares are popular among value-oriented investors similar to Warren Buffet. While they might be able to beat the market in the long run, they are also prone to following the crowd in the short-term as they react to macroeconomic events, geopolitical risks, etc.

As a result, they unwillingly cause prices of cheap and quality shares to move in synchronicity with indices. The fact that the individual shares in a factor portfolio behave similarly enables us to enhance our overall return. We can do this by constructing a multi-factor portfolio.

Factors have lower correlation among than individual shares in a single factor portfolio, while delivering higher return than the overall market. Thus a diversified portfolio has a much better characteristics and higher return potential than a portfolio with one prevailing factor (picking just cheap and quality shares).The factor-based approach also provides us with a better opportunity for risk-management. As the stock market itself, various investment styles (factors) experience bull and bear markets. There are times when cheap and quality shares lag the market (late 90s) and also times when they outperform it.

Simple risk-management techniques allow us to improve our risk/reward ratio. We can decrease our allocation in a factor in times of negative momentum and then rebuild our position later, when the negative momentum reverses. This helps us diminish the influence of bad periods on factor portfolio. Factor-based approach to investing can offer attractive returns. The problem is the practical side of constructing a truly diversified multi-factor portfolio. Retail investor does not have resources to access historical database with correct price and fundamental data of companies she would like to trade. She is therefore unable to test the performance of factor portfolios and use risk-management techniques to enhance portfolio risk/return characteristics.

Recently, a new web service QUANTPICKER has been launched to help retail investors examine strategies they would like to implement. A simple, user-friendly tool allows one to backtest individual factors, mix them up and apply different risk-management techniques to a portfolio. Weights of individual components of a multi-factor portfolio can be easily exported as CSV files every month. This allows a retail investor to simply construct a modern, diversified portfolio that favorably compares with professionals.