Drivers of returns

SMART beta investing covers many approaches - some wildly different from one another. We use "beta" to refer to products that offer passive-like exposure, primarily through ETF (exchange-traded fund) vehicles. The use of "smart beta" implies that their long-term performance might beat that of their broader cap-weighted benchmarks in back tests.

Whether or not these claims are accurate, a growing body of research indicates that there is nothing novel driving the outperformance of smart-beta strategies. The "secret" to their performance is simply the choice of investment factors used to construct the portfolio. Factors such as value, momentum and small cap are important performance drivers of returns for any diversified equity portfolio.

There are several reasons for the emergence of factors at an aggregate level. Portfolio managers tend to disagree more about the individual stocks they pick than the investment styles they follow. Knowingly or unknowingly, managers prefer stocks with certain style characteristics such as value or quality, although they disagree about certain stocks. Therefore, at an aggregate level, stock-picking decisions wash out and are less relevant than the choice of style characteristics.

Capital asset pricing model

Since the 1950s, academics puzzling over the near-random fluctuations of individual securities have asked: "Are there common drivers that can explain at least some of the fluctuations of individual securities?"

Yes. These common drivers are - the factors.

In a way, portfolio managers first began using factor investing in the 1960s when they used the capital asset pricing model (CAPM) to value securities. This is a single-factor model where the expected return of a stock is entirely determined by its sensitivity to the market factor, or its market beta. Before that, any returns realised on a portfolio were fully attributed to the skills of the portfolio manager.

Portfolio managers using the CAPM valuation model divide the portfolio's active returns into two: one component, beta, describes the portfolio's sensitivity to the movements of the broad market; the remainder, which cannot be explained by market movements, is called alpha.

CAPM clearly shows that part of a portfolio's relative performance is attributable to exposure to the broad market, and not just to stock picking or alpha.

The idea that factors such as value or small caps can - in addition to the market factor - explain a large proportion of stock returns was first put forward by Fama and French in the 1990s.

What's the rationale behind that argument? Fama and French argue that value stocks or small caps are structurally riskier than other stocks, and so require a risk premium. Therefore, when estimating the expected return of a stock, you must take the value exposure and the small cap exposure of a stock into account. This is important as these exposures relate to the risk premia the stock is expected to earn.

This model was later extended to include another factor: momentum. Today, academics use the Fama-French-Carhart model as the standard to explain equity returns. However, even though it is now one of the key models used in quantitative finance, CAPM was rarely used during the first 15 years following its development.

That changed in 2009, when a group of three finance professors was asked to analyse the weak performance of the Government Pension Fund of Norway. They looked at the period from 2007 onwards including that of the Great Financial Crisis. The consultants discovered that the performance of the fund could be nearly fully explained by its exposure to equity risk factors such as value, momentum and size, as predicted by the Fama-French-Carhart model, or fixed income risks such as credit spreads and duration.

This report was seminal. It triggered a lot of interest in "factor investing" as an investment strategy, and is thought to have "given birth" to modern factor investing.

In response, index providers have launched a plethora of new indices such as smart beta indices, factor indices and style indices. Although these terms largely overlap, they all claim to capture the relevant risk premia.

What should investors do?

Firstly, it's important to note that these macroeconomic risks are non-rewarding risks and there is no risk premium attached to them.

Of course, there are times when a risk could pay off. To benefit from that, investors would have to "time" those periods when the risks are likely to pay off. However, we think trying to time risks, whether the oil price or the interest rate cycle, is a futile practice. Therefore, we believe investors would be better off neutralising their exposure to non-rewarding risks when assessing factor risk premia.

Certainly, the worst thing to do would be to leave exposures to non-rewarding risks unmanaged, like factor ETFs often do. This can lead to a lot of extra, uncompensated short-term volatility.

Portfolio managers - operating at the single stock level - can deal with these non-rewarding risks more efficiently. For example, they can create a portfolio of both value stocks and momentum stocks. This can provide the desired exposure to both the value factor and the momentum factor.

At the same time, it can help ensure managers are not only buying stocks with high oil price sensitivity, but also stocks with low oil price sensitivity. This will also help ensure that they are not only buying stocks with high interest rate sensitivity, but also stocks with low interest rate sensitivity.

In this way, portfolio managers can establish exposure to desired risk factors - value and momentum - while at the same time making sure that exposure to all non-rewarding risks is near zero. For an integrated portfolio solution operating on a single stock level, this strategy offers a much more stable method for earning risk premia than factor ETFs.

Risk premia

By now, the existence of risk premia in equity markets - in addition to equity risk premium - is a familiar concept. It is supported by academic research, as well as by the extensive experience of its practitioners.

Although there is no consensus yet on how many risk premia there are and how to define them, certain economically significant and persistent premia have been identified. Investors should not ignore these risk premia as they are a reliable source of risk-adjusted excess returns to their portfolios.

The Allianz Global Investors Best Styles approach, launched in 1999, is an integrated risk premium solution that has always applied the rules of factor investing, long before the terms factor investing or smart beta were coined. It has a proven track record of harnessing risk premia within equities in a manner that keeps performance stable, while remaining largely independent of the market and the macroeconomic environment. Solutions for a global institutional client base in Europe as well as in Asia and the US have been successfully implemented. W

Benedikt Henne and Klaus Teloeken are Co-Chief Investment Officers of Systematic Equity, Allianz Global Investors