The desire of many to know what drives assets’ returns is not new and after many decades of academic study a new investment paradigm emerged: Factor Investing.
This philosophy tries to capture risk premiums from academically-and-time proven factors in order to enhance risk-adjusted portfolio returns or achieve better diversification.
Academic studies go back to the 60s, when the most well-known model of stock returns, the Capital Asset Pricing Model (CAPM), became the foundation of modern financial theory. The CAPM, which was introduced by Lintner, Sharpe and Treynor, defines two drivers of securities’ returns: The idiosyncratic risk and the systematic risk. As investors can diversify away the idiosyncratic risk, the risk attributable to an individual security, in an efficient market no compensation would be granted for bearing such a risk. On the contrary, due to the fact that systematic risk cannot be diversified since it affects the market as a whole, investors are rewarded for bearing such a risk: the market premium. Therefore, the CAPM derived a linear relationship between expected returns and the exposure of the portfolio to the market risk (systematic risk) which was captured by Beta, defined as the covariance between the security’s returns and the benchmark’s returns divided by the variance of the benchmark’s returns.
Thus, in the CAPM context, a portfolio or a security obtaining returns higher than those of the market will be the result of greater exposure to market or systematic risk (higher beta). At the same time, any strategy that achieves returns above those expected according to its Beta will be generating alpha; an additional amount of return not explained by the exposure to the market factor and, therefore, attributable to the manager’s skill.
Ever since, finance has evolved through new asset pricing models which account for other factors apart from the market one. In 1976, Ross proposed the Arbitrage Pricing Theory (APT) which unlike the CAPM assumes that markets are not perfectly efficient. Besides, the APT explains securities’ returns not with one but with several factors, the number and nature of which are on the hands of the investor and vary across markets. The APT gave birth to multifactor models from which Fama and French created the 3-factor model in the early 1990s. This new model substantially improved the explicability strength of the CAPM by taking into account not only the market factor, but also the Size factor (large versus small capitalization stocks) and the Value factor (low versus high book to market stocks).
Later, in 1997, Carhart extended the model to four factors (4-factor model) with the introduction to another new factor: Momentum. The introduction of additional factors, such as profitability or volatility, results in deviations or extensions of the 4-factor model approach.
Factors and Smart Beta.
Factor-based investing is not new. Some of these investment strategies have been known for many years, such as Value, which was Graham and Dodd advocated in 1934. However, factor investing recent popularity is the result of several factors:
- The need to diversify among less-correlated stocks after seeing that traditional portfolios were not as well-diversified as thought in
- The difficulty in finding Alpha has led many investors to search the market for new ways to beat the
- Factor-based academic studies have demonstrated that most of the Alpha stated by active managers can be explained by systematic exposure to risk
- The creation of new vehicles to tilt portfolios to specific factors (Factor Indexes and ETFs).
This last point is maybe the most important. Nowadays, Investors have access to indices that track down particular factors through a cheap, systematic and transparent approaches by utilizing vehicles such as ETFs. In fact, the rise of factor investing is related to the emergence of “Smart Beta1”.
Index providers have introduced alternative weighting schemes such as Value-weighted indices or risk-weighted indices, which have enabled investors to expose their portfolios to factors by investing in funds or ETF’s that track down the desired factor index. This approach has recently become very popular and is better known as the “factor beta” or “smart beta approach” and aim at generating outperformance relative to broad-market cap-weighted indices (higher Sharpe ratio for example) by taking exposure to long-term rewarded risk factors beyond the market factor.
Having get this far, one might be wondering what a factor is. The term factor is nothing more than any stock characteristic that is statistically significant either in explaining risk or returns. Among the different existing factors, those with more acceptance are: Value, Quality, Low Volatility, Momentum, Size and Yield. These factors present further investigation and strong academic support that proves the existence of a premium.
For example, in 1992 Fama and French discovered that average small cap portfolio earned monthly returns of 1,47% which were above those of a large cap portfolio of 0,9%. In their paper (The Cross-Section of Expected Stock Returns) they also found that the outperformance went from July 1962 to December 1990.
The reasons to explain the existence of these premiums vary depending on the beliefs about the market and its participants. In this sense, two main camps can be found:
- The Market Efficient point of view: In an efficient market where investors are fully rational the existence of factor premiums that allow them to outperform the overall market are the result of exposure to systematic risks that cannot be diversified away. Proponents of this camp argue that, for example, the Value or Momentum premium existence is the consequence of investors demanding a higher return for bearing stocks which are more sensible to shocks in the economy.
- The behavioural point of view: On the other hand, the behavioural point of view defends that investors are far from rational, presenting emotional and cognitive biases, and, as a result, factors are able to earn excess Within this second camp, it also exists a subgroup that attribute the existence of these premiums to constraints and frictions that emerge from industry practice and regulation. In their opinion, investors being rational and anomalies can exist at the same time due to constraints such as the use of benchmarks or reputation concerns that lead managers to herd.
Besides, factor cyclicality, long-market periods during which factors underperform the overall market, is usually the reasoning utilized to justify why the outperformance of these strategies has not been arbitraged away. The length of this period is thought to last between two and six years what eventually may challenge the patience and conviction of many investors.
Already mentioned in the introduction to factor investing, the CAPM defined for the first time risk and made possible the evaluation of a manager performance relative to the market and the Value he added. In the CAPM framework, if a manger outperformed the market by increasing the market exposure (Beta) of the portfolio, investors were paying a high fee for an alpha that did not really exist.
Nowadays, with advances in factor investing and the emergence of factor indexes, assessment of managers’ performance can be more accurate and allow a better understanding and determination of portfolio returns, which are no longer evaluated relative to market exposure alone, but taking into account other factors as well.
In recent years, many research studies (Fama and French (2010)) or Ang, Goetzmann and Schaefer (2009) have displayed that major part of active returns (alpha) could be explained in terms of systematic factors. In 2013, Mok, Bender and Hammond (2013) concluded that average alpha of US institutional funds could be 50% explained by Fama and French factors.
If true, these excess returns cannot be considered as alpha any longer since they could be captured through systematic strategies, which are well known and can be mechanically implemented (by tracking down factor indexes, for example). Not being attributable to manager skill any more, these previously-unexplained returns are now explained and can be captured.
“Yesterday’s alpha is today’s beta”
Ferran Capella Martínez
Miembro del Servicio de Estudios y Publicaciones de la Bolsa de Barcelona, BME.