Beta and Volatility - A Wolf in Growth and Value Sheep’s Clothing

As we near the end of a painful year that will also surely rank as among the most bizarre ever for the investment community, we might all be wondering what else 2020 can possibly throw at us? Well, don’t hold your breaths too long given that soon we may be experiencing a growth vs. value rotation within the mix, as we are beginning to see the characteristics of growth and value hit uncharted territory. In 2020, we’ve seen already seen a COVID-induced Volatility reversal, and a vaccine-induced Momentum crash, so why not add Growth and Value rotations to the rest of the fun we had in 2020?

For this week’s issue of Factor Spotlight, we evaluated the J.P. Morgan iDex U.S. Growth Index and J.P. Morgan iDexU.S. Value Index to dig deeper into this developing trend. What initially caught our attention here is that the spread between these indices has hit unprecedented levels since 2005, as seen in the chart below.

Given this, we investigated the characteristics of these indices to better understand what this massive spread might be telling us about the current market environment, as well as the possible implications for both value and growth investors.

Methodology for JP Morgan iDex US Indices

The JP Morgan iDex US Indices are constructed using the MSCI USA and the MSCI Barra Global Equity 3 risk model factor exposures for the Value and Growth factors. After applying various investibility screens to the MSCI USA universe, the top 100 securities by Value and the top 100 securities by Growth are selected and rebalanced monthly to form the constituents of the JP Morgan iDex US Value Index and JP Morgan iDex US Growth Index, respectively.

Two characteristics that make these indices unique are the chosen weighting scheme and the sector normalization. The indices are liquidity-weighted, which is an interesting alternative to the standard cap- or equal-weighting that we typically see with market indices. The sector normalization is applied to the factor exposures to reduce heavy sector bias that can be at times be evident within factor indices.

‍The table below shows the top 5 active sector weights for the indices relative to their Russell 1000 Style Index counterparts:

We can see that contrary to typical value and growth indices, the JP Morgan indices avoid the expected sector biases. In particular, the Growth Index is underweight the usual “growth” sectors of Information Technology and Communication Services and overweight the usual “non-growth” sectors of Industrials and Financials relative to the Russell 1000 Growth. Similarly, the Value Index is underweight the usual “value” sectors of Financials and Industrials and overweight the usual “non-value” sectors of Information Technology and Health Care. This indicates that the JP Morgan indices are gaining the appropriate growth and value tilts not just by taking major sector bets, but rather by representing true growth and value styles.

‍We can even illustrate this “pure” representation of growth and value by using another third party as validation for these indices. Though the indices were constructed using the MSCI Barra factor exposures, we can leverage Omega Point’s integration with many leading risk model providers and use the Axioma US 4 Medium Horizon risk model factor exposure to see just how much of a factor tilt is expressed in the JP Morgan indices. Again, we’ll look at the Growth and Value exposures of the JP Morgan indices relative to the relevant Russell 1000 style indices.

The JP Morgan indices range from about 0.2 - 0.85 overexposure to Growth and Value factors, respectively, relative to their Russell 1000 style counterparts. We also tested these indices against several other Russell Style Indexes (Mid Cap, 2000, and 3000) and observed similar overexposures.

The DNA of Growth vs Value

Now that we have a good understanding of the JP Morgan style indices, we can use these, along with other key factors from the Axioma US 4 Medium Horizon risk model to understand the drivers of the recent behavior in growth and value. We evaluated all of the style factors from the Axioma model and found that the Market Sensitivity and Volatility factors tell a lot of story when considering the behavior of growth vs value.

‍First, we can take a look at the Volatility factor. Intuitively, we know that growth generally has higher exposure to the Volatility factor, especially relative to value. We see that same trend in the chart below when looking at the Volatility factor exposures for the JP Morgan iDex US Growth and Value indices.

Of note here is that the levels of Volatility exposure within the two indices converges in 2016 and 2018 and even reverses in 2019. However, we now can see that the Volatility exposure spread of these indices, currently at 0.61 (Growth minus Value), is at the highest level that it has been since 2007! This is driven by Volatility having higher representation within growth and lower representation within value than ever before in this time period.

Given how the Volatility and Market Sensitivity factors tend to move with one another, we also took a look at the Market Sensitivity factor exposures within the JP Morgan indices.

Again, we see that the Market Sensitivity exposure spread of these indices hit the lowest point in the period since 2007 coming out of the March 2020 downturn, with a -0.79 exposure spread in May 2020. Though some of this extreme negative spread is caused by lower Market Sensitivity exposure within the Growth Index, we see a much larger portion of the negative spread is caused by a major spike in this exposure within the Value Index.

Value is effectively suffering from a massive runaway beta effect, caused by shifting demand for “stay-at-home” vs “brick-and-mortar” stocks impacted by COVID-19. The last time we saw beta exposure levels within value this high was in the thick of the Global Financial Crisis. Interestingly, the period following the Credit Crisis was a time that was very favorable for value investing. Perhaps the recent peak in the Value Index’s Market Sensitivity exposure and subsequent downtrend is a signal that greener pastures are coming for value investing, similar to what we saw coming out of the Credit Crisis.

Implications on Stock Picking Opportunity

The last step in our analysis takes us look at the predicted risk characteristics of the JP Morgan style indices. Intuitively, we know that we are in a very challenging time for active stock pickers, but we can use idiosyncratic risk as a proxy to quantify this effect.

Looking at the JP Morgan iDex US Growth Index vs the Russell 1000 Growth, we can see that the idiosyncratic component of predicted risk hit one of the lowest points since 2007 in April 2020. Interestingly, the specific risk of this index has bounced back sharply coming out of the March 2020 market downturn and is now sitting at 25%.

The same chart shown for the JP Morgan iDex US Value Index vs the Russell 1000 Value shows paints a tougher picture for value.

We know that stock picking opportunities tend to be a bigger challenge in value as compared to growth, which can be quantitatively observed, especially for recent periods, in the very low levels of idiosyncratic risk for the Value Index in comparison to the Growth Index. The Value Index had a low point of 4% in April 2020 and currently has a level of only 10%. However, again, we see an uptrend in the level of idiosyncratic risk, similar to what we saw post-Global Financial Crisis, that may signal easier stock picking opportunity ahead for the value world.

‍While it is impossible to say with certainty what might lie ahead for growth vs value, we do know that we are seeing major changes in the factor and risk tendencies with these particular market indicators. Growth and value investors need to be aware of these changing dynamics and take steps to ensure they are not blind-sided by the beta and volatility factors lurking in their portfolios.

US & Global Market Summary

US Market: 11/23/20 - 12/04/20

• Stocks rose higher during a historic market rally that saw major indexes at record highs along with a rise in Treasury yields, reflecting expectations for additional fiscal stimulus.
• Gains were broad-based and many leaders delivering powerful gains, including Qualcomm, CrowdStrike (CRWD), and Advanced Micro Devices (AMD).
• The unemployment rate declined to 6.7% from 6.9%, but the participation rate also declined, indicating that more people exited the labor force.
• The U.K. will be giving doses of the Pfizer (PFE) and BioNTech (BNTX) coronavirus vaccine within the next few days. An FDA panel will review the Pfizer vaccine, with emergency approval expected soon after.
• Oil extended its gains to a nine-month high after OPEC and its allies agreed to increase production more gradually than previously planned, and the dollar hit a new two-and-a-half-year low against major currencies.

Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)

• Volatility was the most impressive 2-week mover and sits comfortably within positive territory well above its flat position on Nov 16.
• Value and Market Sensitivity continued their climbs deeper into positive territory over the past 2 weeks.
• Size and Growth showed welcome gains but still sit within negative normalized territory.
• The pain continues for Momentum as it slides yet deeper into Extremely Oversold space.
• US Total Risk (using the Russell 3000 as proxy) declined by 54bps.

‍Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)

• Similar to the US, Volatility and Market Sensitivity both continue their strong runs to sit atop the 2=week leaderboard on both lists.
• The rally in Value continued as it rose +0.38 standard deviations further into Overbought territory.
• Size dropped slightly breaking it’s climb into Overbought space.
• Growth experienced yet another gap down (-0.18 SD), falling even further into Oversold territory.
• As with the US list, Momentum continued its freefall falling by -0.63 standard deviations and landing deep in Extremely Oversold space at -2.37 SD below the mean.
• Global Risk (using the ACWI as proxy) decreased by 57bps.

Regards,
Alyx