Rich Dell and Phil Edwards explore the nature of low-volatility offerings and the challenge created by the proliferation of idiosyncratic strategies
Low-volatility equities have attracted significant investor interest in recent years. This interest has clearly been driven in part by a desire to reduce equity-related volatility, but is also due to the attractive return characteristics offered by such strategies.
Traditional finance theory suggests that risk and return are inextricably linked, with higher risk assets generally commanding higher returns. However, there is now a large body of empirical evidence suggesting that high-volatility stocks have not earned superior returns relative to low-volatility stocks over long periods.
These observations have been well documented and frequently discussed. It is an increasingly accepted view that this ‘low-volatility anomaly’ is a function of investor behaviour; that is, the anomaly can be explained by behavioural factors. In particular, the following arguments have been used to help explain this phenomenon:
• Overconfidence – Investors as a group are consistently overconfident. This overconfidence is likely to have its biggest impact in terms of highly-volatile shares, which are perceived to have high-growth prospects.
• Representativeness – Investors are often seduced by stories of the fortunes made by individuals who have made speculative investments in new technologies that have yielded tremendous returns (for example, Microsoft), but these investors fail to appreciate that these results are rare and that most opportunities of this kind fail to deliver the expected returns.
• Lottery effect – Lottery participation provides an obvious parallel to investing in high volatility stocks. No rational individual would participate in a lottery for financial gain, as the expected return is a loss. The fact that millions of people do participate in lotteries is indicative of a willingness to act irrationally (and to overpay) in the hope, however small, of securing a large gain.
This is all very interesting and may well explain the existence of the phenomenon in the past, but shouldn’t smart investors be arbitraging this effect away? There are probably two reasons why this hasn’t happened.
First, the logical approach to gaining from investors’ irrational behaviour in this way would be to sell short high-volatility stocks and go long low-volatility stocks. The problem with this is that high-volatility stocks are by definition risky and unpredictable and the pain for the short seller if one stock does turn out to be the next Microsoft is likely to be extreme – a long investor can only lose 100% of their money whereas the short investor can lose much more than that.
Second, the nature of the investor universe is such that many participants are focused on managing risk relative to a benchmark index. Taking a significant position away from high-volatility shares is likely to create a high tracking error and therefore presents a potential business risk for professional fund managers.
It is likely that the combination of behavioural factors leading investors to favour high-volatility (or ‘glamour’) stocks, together with constraints on shorting and a continued focus on benchmark-relative risk will ensure the existence of the low-volatility anomaly into the future.
There are two obvious routes to constructing a low-volatility equity portfolio: building a portfolio of shares in companies that operate in relatively stable businesses and which will often have low financial gearing; or building a portfolio of shares in which the share prices themselves have low-volatility characteristics (that is, the individual shares are less volatile than the market).
The first type of strategy is described as ‘quality’. In broad terms, quality can represent dependability or predictability – the archetypal quality stock belongs to a strong, stable, well-managed and profitable company that is able to withstand external pressures, such as economic downturns (or financial crises). Such strategies rely on traditional fundamental research to assess the ‘quality’ of any potential investment in conjunction with valuation.
The second approach to building a low-volatility portfolio is labelled ‘systematic’, and this includes managed volatility or optimised minimum variance strategies. While quality strategies target companies that tend to exhibit lower volatility as a result of the nature of the underlying business, systematic strategies use quantitative techniques to directly target shares with low price volatility.
A third approach to reducing equity market volatility is simply to introduce some non-equity exposure to reduce beta. Such approaches might include long-only, equity-heavy strategies that have the flexibility to hold cash, bonds or other assets such as gold when they believe that equity markets are expensive, as well as long biased long-short strategies. It is important to note that these strategies do not necessarily benefit from the low-volatility anomaly discussed above: such approaches can invest in highly volatile stocks, while reducing portfolio volatility through non-equity holdings.
In recent years there has been a proliferation of low-volatility strategies. This has been particularly notable within systematic approaches (both actively managed strategies and ‘passive’ index approaches). Initially many ‘systematic’ strategies were variations of a simple optimised minimum variance approach. While there would be differences (such as the treatment of currency, stock or sector constraints, risk models, or the inclusion of a return forecast) they all relied on very similar inputs and in some ways were easily comparable.
However, many of the more recent strategies look to address the perceived issues surrounding optimisation. These issues include concentration of risk, arbitrary use of constraints, and high portfolio turnover. As a result a growing number of non-optimised systematic low-volatility strategies have been launched. Such strategies do not rely on risk models, optimisation or correlations; instead they define a universe of low-volatility stocks, perhaps by simply ranking stocks on some measure of volatility, and use that universe to build a portfolio. They can more easily control concentration and turnover, and often introduce a higher degree of portfolio manager oversight. This has some appeal. However the degree of volatility reduction is likely to be less than in optimised minimum variance strategies that explicitly target low-volatility portfolios.
This growth in different approaches presents a challenge for investors. While most systematic strategies target similar levels of volatility reduction (around 30% below the broad market) they can have very different approaches, which can lead to vastly differing portfolio characteristics. Even within ‘passive’ approaches, where MSCI, FTSE, Russell and S&P all have indices focused on delivering structurally low volatility, the differences are significant. Monitoring the relative success of such strategies represents a further challenge as they frequently have different investment targets and there is no universally accepted benchmark for low-volatility stocks.
Given this, it is critical for potential investors to know what the role of such a strategy is within their overall portfolio, and to have a good understanding of the differences between the relevant approaches.
As noted at the start, there has been significant interest in low-volatility equity strategies, especially since the financial crisis when these approaches have, in general, outperformed broad equity market returns by an impressive margin. To date, there has been particular interest in quality strategies as an intuitive and ‘traditional active’ approach to introducing a low volatility bias. Indeed, a number of ‘quality-biased’ strategies that Mercer rates highly have reached capacity and closed to new business. More recently, there has been an increasing interest in systematic strategies, for those investors able to get comfortable with the quantitative aspects of these approaches, and some investors considering a passive implementation. Interest in low-volatility strategies is expected to remain strong in the current environment.
Rich Dell and Phil Edwards are principals at Mercer Investments