2018 was tough for investors thanks to broad declines in stock prices and increased volatility. The S&P 500 Net Return Index closed the year with a loss of 4.9% after a seemingly continued bull trend peaked out with a gain of over 10% at one point during the year.
The market volatility, as measured by the CBOE Market Volatility Index VIX, spiked in 2018 by an average of 50% compared with 2017 after two years of decline. In this turbulent market environment, risk-reducing strategies such as low/minimum-volatility ETFs have fared well in terms of both performance and raising assets.
In 2018, risk-oriented ETFs garnered closed to $65 billion globally despite flattish or slightly negative growth for the global ETF market.
Currently, there are around 110 ETFs available globally classed as “risk-oriented” and around 29 in Europe, according to Morningstar's classification. While most of these ETFs have been around for less than five years, those that invest in US equities are relatively well established and hold more than half of the assets invested in risk-oriented ETFs.
Performance Held Up
Looking at total returns for 2018, the performance discrepancy is notable across our sample of funds: The best performing low-volatility ETFs last year were Amundi MSCI USA Min Vol and X MSCI USA Min Vol, returning 0.57% and negative 0.19%, respectively. The iShares Edge S&P 500 Min Vol was the worst performer, with a loss of 5.36%, lagging the S&P 500 by 0.42%.
These differences are not surprising. While all low/minimum-volatility strategies share the same investment objective, which is to minimise volatility versus a market-cap index, they achieve this objective in various ways by employing different index methodologies.
Next, we put 2018 under a magnifying glass to examine how the eight low/minimum-volatility ETFs performed during the most volatile periods of the year. We identified five volatile months where the VIX stayed above 2018's average level: February, March, October, November, and December. We broke down each of these periods into two mini-periods, down and up, of fewer than nine days each on average. We then compared the daily moves of the ETFs with their parent benchmarks during these mini-periods.
In every period, all eight funds’ price moves were significantly less than their parent benchmarks' moves in both directions, upward and downward, by 40%-70%. However, the outperformance in market downturn was stronger than the underperformance in upturn: On average, the former was 4.2% while the latter was 3.1%. This is well in line with expectations, given the way these ETFs are constructed.
While the differences are minimal, SPDR S&P 500 Low Volatility ETF and Lyxor FTSE USA Minimum Variance ETF were the best performing funds in bear trends. Yet, such outperformance entailed greater underperformance in bull trends, both ranking at the bottom in market rallies. All in all, both funds achieved a 25% reduction in volatility
Different Index Methodologies
The two most-established risk-oriented beta strategies are low volatility and minimum variance. The low-volatility strategy aims to capture the low-volatility effect by overweighting stocks with lower price volatility and underweighting those with higher price volatility.
S&P 500 Low Volatility and MSCI USA Risk Weighted belong to this category. The index consists of the least volatile 100 stocks within the S&P 500 selected every quarter by ranking their past 12 months' volatility. As there are no constraints on sector weightings or turnover, the portfolio can end up with sector tilts.
The index heavily overweights the utilities, financials, and real estate sectors, and each sector contribution often exceeds 20%. Because this strategy centres around capturing the low-volatility effect of individual stocks, it does not necessarily result in the least volatile portfolio. The other strategy, minimum variance, focuses on reducing the overall volatility of a portfolio or the aggregate level of volatility.
Indexes such as S&P 500 Minimum Volatility, MSCI USA Minimum Volatility, and FTSE Minimum Variance are representative examples. Aside from price volatility, minimum-variance indexes consider correlations between the individual index constituents.
An optimisation algorithm is used to select and determine constituent weights in such a way that the aggregated portfolio ends up with a list of stocks, which, combined, achieve the lowest expected risk. A minimum-variance index can end up with a set of stocks that are very different from those in a low-volatility index.
The former can include some high-beta stocks because of their low level of correlation with other stocks in the index or because of their diversification benefits. While the above-mentioned three indexes employ broadly the same approach for index construction, the level of constraints and diversification differs.
The different starting universes and constraints among minimum-variance strategies lead to different stock selection and weighting, which in turn can result in varying short- and long-term returns.
The notable one-year return difference of 5.2% between iShares S&P 500 Min Vol ETF and X MSCI USA Min Vol ETF stems from the inclusion of high-beta stocks in S&P 500 Min Vol ETF, particularly technology stocks such as Amazon.com, Advanced Micro Devices, and NVIDIA, which lagged last year. These stocks were not included in the MSCI USA Min Vol ETF.
The highest individual beta in S&P 500 Min Vol was above 4.2 versus that of 1.8 in MSCI Min Vol. However, the similarity of index-construction methodologies between MSCI USA Minimum Volatility and S&P 500 Minimum Volatility results in portfolios with similar characteristics.
The one-year performance difference mentioned above is rather a short-term phenomenon that could also be attributable to differences in rebalancing timing considering that the majority of the funds’ return gap appeared in fourth-quarter 2018.
It also stems from the different starting universe: 620 holdings in MSCI USA versus 500 in S&P. Longer-term, looking at the three- to five-year track records of their underlying indexes, we can expect differences in performance between low-volatility and minimum-variance funds to be smaller.
Conclusion
We view both groups of risk-oriented beta strategies—low volatility and minimum variance—as suitable for investors who are looking to mitigate risk in their portfolios. Both strategies applied to US equities have fared well in delivering smoother market ride. Over the three- to five-year term, these ETFs have fulfilled their purposes by reducing volatility by 10%-20%. They have also meaningfully outperformed their Morningstar Category peers over the long term with higher risk-adjusted returns.
When investing in a risk-reducing strategy, investors should not pay too much attention to short-term performances. As the trade-off of a low-volatility strategy is underperformance during strong bull trends, one should consider longer-term investment horizons to fully benefit from this strategy.
