The capital asset pricing model (CAPM) dates back to 1964 and has long been the centerpiece used to explain the relationship between risk and return. According to the theory, higher risk should lead to higher returns. Empirical findings, however, contradict this notion. Figure 1 depicts the risk-return profile of ten portfolios sorted on the volatility of historical returns. This clearly shows that the equity market has generally not rewarded investors for taking on more (volatility) risk.
Figure 1 | Long-term risk-return profile of ten volatility-sorted portfolios
Source: Robeco, CRSP. Figure shows average, annualized returns and volatilities of 10 portfolios sorted on past 36-month return volatility. The investment universe covers all common stocks traded on NYSE, AMEX and NASDAQ exchanges with valid market capitalization and return data from 1926 till 2020. Portfolios are equal weighted and portfolio returns are from January 1929 to December 2020.
Low Volatility effect confounds risk-based school of thought
The CAPM assumes a linear relationship between the risk (market sensitivity, i.e., beta) and returns of financial securities. However, numerous studies have illustrated that low beta stocks counterintuitively outperform their high beta peers on a risk-adjusted basis. This was pointed out as far back as the 1970s in a seminal paper that demonstrated that less volatile stock portfolios generated higher returns than riskier counterparts.1
The efficient market hypothesis suggests that low-risk stocks must exhibit other risks that are not captured by their market betas, and this explains their long-term returns. However, attempts to identify these risks have been few and far between. They also pale in comparison to the behavioral finance explanations of the phenomenon.
Risk-based theories that explain the low volatility effect have largely been disputed within the academic field.
Low volatility stocks are typically found in defensive sectors and have more predictable cash flows, leading them to exhibit lower valuation uncertainty. Thus, they portray bond-like characteristics, while investors are also likely to use them as replacements for bonds given that they typically pay out dividends. Despite these features, Robeco research concluded that interest rate risk does not account for the long-term added value from low volatility strategies.
In general, risk-based theories that explain the low volatility effect have largely been disputed within the academic field. On the other hand, research from the behavioral school of thought is far more significant on this front.
Investor behavior drives Low Volatility premium
Behavioral biases and constraints offer more convincing reasons for why low volatility stocks have the potential to generate higher risk-adjusted returns than their high volatility counterparts. But in contrast to other factor premiums that are driven by irrational investor behavior, the low volatility anomaly is premised on ‘rational’ investor behavior. Some of the research that explores this premise is outlined below.
Within the investment industry, relative returns often supersede absolute returns as a yardstick for performance or manager aptitude. Low volatility investing can therefore be unpopular due to how markedly different low volatility portfolios can look when compared to benchmarks. This results in higher tracking errors (relative risk) that are not palatable for some investors, especially when short-term underperformance in up markets is a possibility.2 Thus, the desire to keep up with the markets against which portfolios are measured incentivizes investments in high volatility stocks.
The focus on relative performance gives rise to so-called agency issues according to research.3 Investment professionals usually have option-like incentive contracts. They typically seek to maximize the value of these by targeting high portfolio returns, which can cause them to be more attracted to higher-risk stocks.
Another paper states that asset managers are motivated to invest in profit-maximizing, high beta stocks.4 Consequently, they may be willing to overpay for stocks that outperform in up markets, which tend to be highly volatile in nature, and underpay for those that outperform in down markets, which typically have low volatility characteristics.
One academic study also highlights how leverage constraints contribute to the low volatility effect.5 Based on their risk appetite, investors may be able to enhance their returns by leveraging a low volatility portfolio. This may allow them to increase their return potential without taking on additional risk. But due to leverage (or borrowing) constraints, they tend to overweight riskier investments in search of higher returns, therefore lowering their expected returns.
The lottery ticket effect is another documented reason for the low volatility phenomenon.6 Many investors participate in the market to gamble and this steers them towards high-risk stocks due to their upside potential, while their downside risk is limited to the investment amount. In this scenario, the investors are willing to pay a premium for the risk instead of being compensated for it.
The low volatility premium has been persistent from as far back as the 1930s.
Why hasn’t Low Volatility been arbitraged away?
In our view, the low volatility effect is one of the most persistent market anomalies. In 2008, the style became more widely accepted as its watershed moment arrived with the global financial crisis, when it provided downside protection amid the broad-based sell-off. That said, the anomaly has been observed over a long time period and is closely linked to behavioral biases. Indeed, we have observed that the low volatility premium has been persistent from as far back as the 1930s. We believe there are a few reasons why it has not been arbitraged away.
Firstly, due to the importance of relative performance measures within the investment industry, investors typically choose not to deviate significantly from the benchmark, while they simultaneously aim for higher returns than those delivered by it. This dilemma incentivizes them to prefer more volatile stocks compared to their low volatility peers.
Secondly, low volatility ETF investments have increased over time. But even though large amounts of capital are currently invested in low-risk strategies, or those targeting specific defensive sectors, these are balanced against significant assets in high risk or high-risk targeting ETFs.7
Lastly, the lack of leverage constraints and relative performance measures make it attractive for hedge fund managers to exploit the low volatility anomaly. Although they have no leverage constraints and their performance is measured in absolute terms, their option-like incentive structure tilts their preference towards riskier stocks. This helps to keep the low volatility anomaly alive.8
In the next paper of this series, we will discuss the value factor through a behavioral finance lens. In the previous article, we touched on momentum.