Contemplating a Recovery: adding the idiosyncratic volatility factor to the mix

I hope today's post finds you well and you are enjoying your Presidents' Day weekend. In last week’s Factor Spotlight, we sought to associate changes in fundamental factors—growth and profitability—with net momentum, and found that 77% of the time growth and profitability moved in the same direction, so did net momentum.

This week, we continue our multi-part series that focuses on positioning for a possible recovery by bringing volatility into the mix, by way of the idiosyncratic volatility factor. During our analysis we uncovered some promising correlations that I'm excited to share with you today. But first off, we'll go through a quick rundown of factor trends we're seeing over this past week.

‍US Market

‍Equities surged to a 10-week high with all sectors gaining as the U.S. consumer outlook brightened and positive developments in China trade talks overshadowed lingering concerns about global growth.

Factor Update

‍Here's an update on how some key factors have changed in our normalized return indicator over the past week:

• Market Sensitivity (Beta) has taken another dip this week to *+2.76* standard deviations above its historical mean.
• After hitting a peak of +2.7 SD just 2 weeks ago, Volatility continues its downward drift to settle at +1.65 SD.
• Profitability finally crept up to postive territory the week and lands at +0.07 SD.
• Momentum continues to drop and currently sits at -1.58 SD below the mean.

Idiosyncratic Volatility

‍This week, our analysis looks more closely at the idiosyncratic volatility factor. Before we dive in, it may be helpful to take a step back and examine how Volatilty differs from Market Sensitivity (Beta), as the two are commonly confused by even the most seasoned practitioners. We lean on Axioma’s definition of idiosyncratic volatility: the six-month average of absolute returns over cross-sectional standard deviation, fully orthogonalized to Market Sensitivity. That's quite a mouthful, now let’s recast that definition in simpler terms: Idiosyncratic vol is that portion of volatility not explained by Beta.

For example, if the stock of XYZ Company has a Beta with the market of 1.0, its shares can be expected to rise or fall in lock-step with the market. In the event the market rises by ten percent, we can expect XYZ’s share price to also rise by ten percent. In the unexpected case where XYZ outperforms the market, rising, say, fifteen percent to the market’s ten, we may attribute the excess returns to idiosyncratic volatility. Across a sector, where diversification tends to absorb company specific risks, idiosyncratic vol may thus provide a meaningful signal that the sector, as a whole, could outpace the market.

In our analysis, we stay with the three historical recovery periods examined last week—2009, 2011, and 2016—noting market rebounds during the brief periods 03/01/09 to 07/01/09, 10/01/11 to 11/01/11, and 02/11/16 to 05/01/16. And notably, we expand the number of GICS sectors examined, from nineteen to twenty-five.

Staying consistent with the methodology used last week, we measured the delta of sector exposures to Medium-Term Momentum, Growth, Profitability, and Idiosyncratic Volatility factors from the beginning to the end of each market rebound. We then highlighted in green all sectors which experienced a positive change to each factor and in blue all sectors that experienced a negative change to each factor.

Our analysis of the three periods uncovered 29 instances where Growth, Profitability and Idiosynchratic Volatility were either contemporaneously positive (19 instances) or negative (10 instances). Of these, there were 22 instances where Net Momentum followed the same direction (76% of the time). Isolating just the two most recent market rebounds, the percentage climbs even higher to 90%.

Similar to our analysis last week, this observation is also unlikely to be random and calls further attention to changing volatility in addition to growth and profitability factor scores so that we may better align to the sectors most likely to have increases in momentum (purely price-driven).

Below we highlight the delta of Momentum, Growth, Profitability and Volatility across 25 GICS sectors during the rebounds of 2009, 2011, and 2016.

Post-Global Financial Crisis: 3/1/09 - 7/1/09

Contemporaneous Net Factor Increase (G, P, V): Banks; Software & Services; Food & Staples Retailing; Health Care Equipment & Services; Insurance; Utilities.

Contemporaneous Net Factor Decrease (G, P, V): Energy; Technology Hardware & Equipment.

‍Post-Sovereign Debt Crisis: 10/1/11 - 11/1/11

Contemporaneous Net Factor Increase (G, P, V): Commercial & Professional Services; Consumer Durables & Apparel; Food & Staples Retailing; Pharmaceuticals, Biotechnology & Life Sciences; Real Estate; Retailing.

‍Contemporaneous Net Factor Decrease (G, P, V): Consumer Services; Internet Retailing; Semiconductors & Semiconductor Equipment; Technology Hardware & Equipment; Telecommunications Services.

Post-Factormageddon 2016: 2/11/16 - 5/1/16

Contemporaneous Net Factor Increase (G, P, V): Food, Beverage & Tobacco; Internet Retailing; Real Estate; Retailing; Telecommunications Services; Transportation; Utilities.

‍Contemporaneous Net Factor Decrease (G, P, V): Energy; Food & Staples Retailing; Pharmaceuticals, Biotechnology & Life Sciences; Semiconductors & Semiconductor Equipment.  

In next week’s Factor Spotlight, we will provide deltas on GICS sectors’ exposures to Medium-Term Momentum, Growth, Profitability, and Idiosyncratic Volatility since 12/31/18, and analyze their predictive effects using a model portfolio.

Regards,
Omer