Volatility is Partying Like It’s... 2009?
Volatility has been on the run! For a factor that typically has abysmal performance and a very negative premia over time, Volatility is turning out to be 2020’s comeback kid. For all of the managers who actively tilt their portfolios away from this factor or manage ‘low volatility’ products, this recent trend from the Volatility factor could really be throwing a wrench into things.
An Unexpected Resurgence
Volatility certainly deserves its reputation as a negative performer, since January 2007 its cumulative performance was just shy of -57%.
However, since the COVID-driven market rebound in early April, Volatility is at long last on an upward trajectory. On a YTD basis, Volatility is up over 6%. 2009 - the midst of the Financial Crisis - was the last time that this factor had a run of positive performance.At that time, Volatility was up over 10% YTD in Sept 2009 before sliding back and finishing the year with annual performance of over 4%.
A key question is whether the drivers of Volatility’s run this time around are the same as they were in 2009? A look at its composition may help provide an answer.
To do this, we used the Russell 3000 universe and built a market-neutral high-minus-low (HML) Volatility portfolio, where the long side is equal-weighted across all stocks with a Volatility exposure greater than 1 and the short side is equal-weighted across all stocks with a Volatility exposure less than -1. This Volatility HML portfolio doesn’t track the factor as well as a factor-mimicking portfolio (FMP) might, but the benefit is that the HML allows us to see sector composition of the factor.
Below is a chart of the Volatility HML portfolio sector weights at three points in time - two periods where Volatility experienced extended positive performance (Aug 2020 and Sept 2009) and one period where Volatility was behaving with the usual negative performance (Jan 2020).
Interestingly, the composition of the portfolio is different across all three periods. For the current period (blue), we can see that the portfolio is long Financials, Energy, and Health Care and short Communication Services, Information Technology, and Industrials. This implies that currently, Financials, Energy, and Health Care are high Volatility sectors, whereas Communication Services, Information Technology, and Industrials are low Volatility sectors.
Rewinding to pre-COVID days at the beginning of 2020 (purple), we see that Health Care was the dominant high Volatility sector whereas Industrials was the dominant low Volatility sector.
Going all the way back to when Volatility reached its YTD high point in Sept 2009 (gray), Financials, Health Care, and Consumer Discretionary were the high Volatility sectors and Information Technology was by far the dominant low Volatility sector.
While there are some similarities in the sector composition - namely for Financials, Health Care, and Information Technology - the magnitude of the sector weights is quite different across all three periods. Furthermore, sectors like Communication Services, Consumer Discretionary, and Industrials play different parts in the Volatility story across time. These are all indications that the drivers of the Volatility trend change over time. While Volatility’s DNA today is not completely different from 2009, there are some differences that may render the current run in Volatility different from the past.
Have Correlations Changed?
Another method for comparing the Volatility drivers today vs 2009 is to analyze correlations. In Omega Point, common ETFs are screened on-the-fly to highlight the ones which have the highest correlations with specified factors. In our case, we can use the correlations of ETFs vs the Volatility factor to tease out differences in the market environment today vs 2009.
Below are the most correlated ETFs to the Volatility factor in 2009 (top) and 2020 (bottom).
A striking difference between the two periods is that the magnitude of correlations was much higher in 2009 compared to now. In 2009, the top ETFs all had correlations above 80%; however, today, the top ETFs barely break 60% correlation to the Volatility factor. This tells us that Volatility, or possibly even the broader market, is likely more diversified today than it was in 2009.
One thing that stands out as a surprising similarity between the two periods is the prevalence of small cap, mid-cap, and value ETFs. ETFs with these styles are in the list of most correlated ETFs in 2009 and 2020, indicating that there may be size and value influences in the drivers for Volatility’s positive performance.
Where Does Volatility Go From Here?
One thing is for sure - we can add Volatility’s positive behavior to the list of unusual occurrences of 2020. Volatility is broken and this has left investors scrambling to reposition for this new normal. Will Volatility become the new shining star of the factor world? Or is this just a temporary dislocation that will revert once the chaos in the markets subside? Time will tell, but in this time of market turmoil and the hunt for alpha, perhaps this dislocation opens up a new opportunity for investors and product creators alike to capitalize on.
In next week’s Factor Spotlight, we’ll continue to analyze Volatility and dig into the “runaway Volatility” phenomena that might be responsible for this factor’s superstar performance in 2020.
US & Global Market Summary
US Market: 8/17/20 - 8/21/20
- Major US stock market indices ticked higher for the week powered by technology stocks and positive reports on the U.S. economy that were better than expected. The S&P 500 hit an all-time high and erased the last of its historic losses from the coronavirus pandemic.
- Data from IHS Markit showed U.S. manufacturing activity hit its highest level in 19 months in August, while services were at their highest level in 17 months even with new COVID-19 cases remaining high across the United States.
- Home sales surged in July to a record month-over-month spike of 24.7%.
- Apple rose 8.2% this week and became the first publicly traded company in the U.S. to reach a market valuation of $2 trillion.
- The U.S. dollar gained against the euro for the first week since mid-June following data showing a strong uptick in U.S. business activity versus Europe.
- China’s Commerce Ministry on Thursday said that Chinese and U.S. trade envoys will hold a meeting by phone “in the near future” to discuss an agreement aimed at resolving their tariff war but no details on timing were given.
Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)
- Volatility continued to to rally from its trough on 8/4 and once again seeing the biggest normalized gains for the week.
- Market Sensitivity saw yet another week of solid gains as it continues its approach towards the mean.
- Momentum hit a peak of +1.95 SD above the mean on 8/6 and then saw a steep reversion, falling 0.4 standard deviations in the past week.
- Growth continued to drop into negative territory
- Size was the week’s biggest loser and headed into negative territory.
- US Total Risk (using the Russell 3000 as proxy) declined by a whopping 253bps.
Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)
- Growth was the week’s strongest performer countering last week’s decline, but still resides in Oversold territory.
- Profitability continued its strong march into positive territory over the last several weeks.
- Market Sensitivity saw slight gains as it continues its climb back towards the mean.
- Momentum was the week’s biggest loser continuing its fall from a peak of +2.58 SD on 8/5.
- Earnings Yield, Exchange Rate Sensitivity and Size were all grouped in negative territory for the second week in a row.
- Global Risk (using the ACWI as proxy) had a steep slide on par with US Risk movement, declining by 260bps.