There has been a great deal of buzz among investors about a likely leadership change in momentum as we near the one-year anniversary of the market trough during the pandemic-induced downturn. As Bloomberg notes, the $2 trillion world of quantitative investing will have a potentially market-moving rebalance coming up over the next several weeks for momentum-based strategies that rely on the typical 12M-1M horizon for momentum calculations, with the mid-March 2020 market downturn starting to roll out of the horizon.
The impending momentum rotation has the potential to rock fundamental investor portfolios as rules-based strategies that rebalance close to or at month-end create systematic waves across the market. To help investors brace their portfolios, we will analyze two portfolios that represent “momentum flow” to see which sectors and style factors may be hardest hit. Through this analysis, we will also identify potential thematic baskets that investors can use to bet on the momentum rotation or simply use to hedge any exposure they have to names highly exposed to this event.
To run our analysis, we created two “momentum flow” portfolios to represent the names that are likely to experience heavy inflows and outflows based on momentum trading strategies. We used the following methodology to create these portfolios:
- Calculate 12M-1M returns for the Russell 3000 universe. The first return horizon is February 28, 2020 to January 29, 2021, which represents the most recent month-end momentum horizon using the 12M-1M return approach for calculating momentum. The second return horizon is March 31, 2020 to February 26, 2021, which represents the upcoming month-end momentum horizon. While we know that some of the systematic quant books will start trading on the momentum leadership change next week, as noted by Zerohedge, as a simplifying assumption, we’ve assumed that the bulk of the trade will happen at month-end.
- Normalize the returns for each horizon by calculating a square-root of market capitalization weighted z-score based on the 12M-1M return. To control for outliers and skewness in the distribution, the data was winsorized twice by truncating exposures greater than 5. This helped dampen the possibility that an individual stock or a handful of stocks, such as small caps and biotech names with outlier returns over the past year, might have a disproportionate impact on the normalization.
- Calculate a “flow” measure by calculating the delta between the current momentum exposures and the upcoming momentum exposures.
- Create a market-neutral momentum flow portfolio (“Momentum Flow - R3000”)by weighting the portfolio by the flow measure calculated in step 3. Here, the long side represents names that will likely experience in-flows due to the momentum composition change and the short side represents names that will experience out-flows.
- Create a market-neutral crowded momentum flow portfolio (“Momentum Flow - High Crowding”) to highlight names that are crowded and may be further bolstered by existing hedge fund positioning. To do this, we used the Wolfe Research QES US Broad risk model to screen the portfolio for names that have HF Crowding factor exposure > 1 on the long side and Short Interest factor exposure > 1 on the short side.
The resulting portfolios represents our projected thematic baskets related to the systematic momentum leadership change that is likely to occur over the next few weeks.
DNA of the Momentum Flow Portfolio
A look at the industry group breakdown of our Momentum Flow - R3000 portfolio illustrates that much of the impending momentum changes will build upon recent themes that have already been hitting the markets as certain sectors have been experiencing headwinds (and tailwinds) from positive sentiment around the economic recovery.
Industry Group Exposures - Momentum Flow - R3000 Portfolio
We see that the long side of our portfolio, which represents names that will receive heavy inflows, has high exposure to Energy, Consumer Services, and Banks. Energy and Banks in particular have been on a tear over the past several months and it seems that this trend may continue, while bringing Consumer Services, Capital Goods, and Retailing along for the ride.
On the short side, which represents the names that could experience heavy outflows, has high exposure to Health Care and Technology industries. This exposure also confirms the COVID themes at play here, as the popular stay-at-home names in these industries have struggled as of late and will seemingly continue to do so with the momentum rotation.
Using the Axioma US 4 Medium-Horizon risk model, we can understand the other style bets that exist within the portfolio outside of just momentum.
On the long side, we see strong overexposure to Market Sensitivity and underexposures to Size and Growth. This tells us that the names that will lead the momentum inflows will likely be high beta, low growth, small cap securities. The short side shows us slight overexposure to Size and slight underexposures to Volatility, Liquidity, and Growth. The names leading the momentum outflows will likely be low volatility, low growth large cap names that are lower on the liquidity spectrum.
Crowding Could Bolster the High Momentum Flow Names
In analyzing the Momentum Flow - High Crowding portfolio, we see very similar sector themes at play, along with some new ones.
Industry Group Exposures - Momentum Flow - High Crowding Portfolio
The momentum flow theme, once screened for high crowding, shows even higher exposure to Energy, Consumer Services, and Retailing on the long side. Interestingly, Banks no longer appear on the top industry group exposures which indicates that institutional investors may be shifting their attention elsewhere despite the strong performance of the banking sector over the past several months.
Turning to our crowded names with high expected outflows, we again see Health Care and Technology at the top of the list. However, we also see Food, Beverage, & Tobacco, along with Retailing as top exposures. This tells us that stock selection within Retailing may be challenging for investors on both the long and short side as institutional investors have crowded into these names all while the upcoming momentum leadership change is poised to have a heavy impact in this industry.
A look at the style exposures for the Momentum Flow - High Crowding portfolio confirms that the same style themes are at play. However, we see that these exposures are more extreme in many cases.
The momentum inflows for the most crowded long names will target even higher beta and less growth, while the outflows will likely be concentrated in much larger cap names with lower volatility, less growth, and less liquidity as compared to the first momentum flow basket for the broad universe.
Turning Systematic Headwinds into Tailwinds
Using the above analysis, we have identified the sectors and styles that are likely to be heavily impacted by the upcoming momentum rebalance, and in doing so, we have also created potential thematic baskets that investors can use to bet on the momentum rotation or use to hedge any exposure they have to this event. If you are interested in seeing names included on our Momentum Flow Portfolios or understand how you can use these thematic baskets to hedge your risk to this impending market event, please don’t hesitate to reach out to us!
Next week, we’ll continue our deep dive into the momentum rotation by evaluating other factors at play in this ‘momentous’ systematic event.
US & Global Market Summary
US Market: 03/08/21 - 03/12/21
- The broader US market enjoyed a strong positive move over the course of the week, with the Dow hitting its fifth consecutive record high on Friday. The Nasdaq took another tumble on Friday after rebounding by ~6% over the past three trading days.
- Optimism abounds around reopening, as President Biden signed the $1.9T fiscal stimulus bill on Thursday and economic data gave reason to believe the economy has started to pick up speed.
- The Labor Department also reported that first-time filings for unemployment insurance in the week ended March 6 were at 712,000, lower than consensus expectations of 725,000.
- Yields continued to rise in the bond market, with yield on the 10Y Treasury hitting a 13-month high at 1.6%.
- The University of Michigan’s consumer sentiment index rose to 83, the highest reading since the pandemic started.
- The US CPI print was in line with consensus expectations at +0.4%, seemingly supporting the case for ongoing dovishness out of the Fed.
- Oil prices fell, with Brent and WTI down slightly this week after surging by ~10% over the past two weeks.
Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)
- Value continued its rapid climb, albeit at a slightly decelerated pace, as it was up another +0.46 standard deviations and headed towards Extremely Overbought territory.
- The ongoing rebound in Market Sensitivity saw it up another +0.33 standard deviations.
- Volatility saw a subtle uptick as it bounced back from a recent bottom of -1.82 SD below the mean on 3/9.
- Momentum fell by 0.63 standard deviations as it moved further into negative normalized space, sitting at -0.64 SD below the mean.
- Growth continued its freefall, down 0.63 standard deviations. This chart showing normalized return for Growth over the past 12 months shows a) how rapidly the factor has fallen out of favor as of late, and b) that normalized return currently matches the 3/23/20 trough of -2.25 SD below the mean.
- US Total Risk (using the Russell 3000 as proxy) ticked up by 27bps.
Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)
- Earnings Yield was again the week’s top global factor, up a +0.87 standard deviations and firmly in Overbought territory at +1.62 SD above the mean.
- Value continued to climb, and is also now classified as Overbought at +1.34 SD above the mean.
- Momentum fell by another 0.42 standard deviations and earned an Oversold label at -1.07 SD below the mean.
- Global Growth continued to suffer, down another 0.72 standard deviations and garnering an Extremely Oversold designation at -2.07 SD below the mean.
- Global Risk (using the ACWI as proxy) saw an increase of 44bps.