A Tour De Beta Across Sector SPDRs
Building upon a previous issue of Factor Spotlight": "A Tales of Two Betas", this week we return to the concept of Beta as a basic analytic measuring risk relative to the broad market. Omega Point now incorporates third-party risk model Betas for both reporting and portfolio construction purposes (but never fear, if you have your own beta calculation you would prefer to measure/manage, Omega Point supports this as well), and thus the functionality displayed and discussed in this week’s analysis is fully available to our clients.
This week we review the last 10 years (with exception for Real Estate) of both Historical and Predicted Beta using the MSCI Barra US Total Market Equity Trading Model of several SPDR Sector ETFs to better help us understand where we currently stand in terms of relative sector riskiness. We won’t be performing this analysis for every single Sector ETF, but more broadly will review more defensive sectors and then move onto cyclical sectors.
Heath Care is a unique sector during this timeframe, as the crisis was brought on by a novel virus. We accordingly don’t see dramatic jumps in the Beta of Heath Care, in fact, it is actually steadily decreasing. It’s interesting to see the recent divergence of Predicted Beta and the Historical betas: the risk model seems to be implying recent market behavior is quite different than the naïve historical beta measure, but of course only time will tell which one is correct.
The jump in Betas for Utilities is probably the most surprising beta jump of all the sectors. Typically associated with stable cash flows, utilities tended to have the lowest beta of all sectors, but even this sector’s low beta couldn’t survive COVID-driven forces. Utilities became more correlated to the broader market due to a flight-to-safety rally during the March sell-off, and these extreme data points are likely to keep the Utilities beta high, but we already see signs of this sector moving back towards the more expected low-beta nature of it’s historical performance.
Who could forget the toilet paper crisis of 2020? The Consumer Staples sector certainly did not - we see that even though Beta spiked during COVID, it never breached it’s historical highs nor stayed high beta for very long. The defensive nature of Consumer Staples held relatively true to it’s typical pre-crisis behavior.
Despite a major spike in Energy beta in March, Energy didn’t make any lasting dramatic changes to beta. Although we see upward beta trends beyond the big spike, the beta is still range bound within it’s 10 year history.
The biggest ‘winner’ during COVID, we see Technology making a very big drop in beta during March. Shutdown orders and a raging virus helped change the scope of Technology from a typically cyclical sector to more of a defensive sector - at least during this crisis. It’s interesting to see how Technology beta was at all time highs in early 2020 - now that the FANGM momentum is coming to an end, will we return to Technology as a high beta?
Overall Financials didn’t react as sharply to the COVID crisis, likely because of the actual or perceived Fed bailout. While beta certainly initially increased, it was still below 10 year highs and has been trending down since.
Below you can see the chart for Real Estate - note that this time horizon is different from the other sectors as XLRE is the newest Sector ETF created once Real Estate received it’s own official GICS sector in 2016.
From 2016 to the end of 2019 we see a general downtrend in beta, until COVID hit and spiked Real Estate’s beta where it has since maintained an elevated level. The current behavior of the Real Estate sector is quite different from history, likely due to the uncertainty around the real estate market. Commercial properties have obviously been hit hard by the ‘work from home’ times, and there are questions around the resiliency of the residential market as both the CARES act expires and temporary job cuts start to become permanent.
Overall we see that historical and predicted betas both tend to agree. Given the recent history used to calculate Historical Betas is the same data set that dominates the factor model used to calculate Predicted Betas, it makes sense that both beta measures tend to agree. While there are times when beta measures can disagree, it can be for many reasons besides “insight”. Even slight differences in beta calculation methodologies can lead to drastically different betas.
While we’ve looked at the 10 year timeseries of both Historical and Predicted Beta, the real question is “what will be the actual forward looking beta”. Will re-opening trade gain steam and shock certain sectors, or will vaccine news fade along with alarming macroeconomic data suggesting the worst is yet to come? The COVID shock to the markets made some dramatic impacts to some Sectors, but interestingly fewer than we may have expected.
There is still much uncertainty in the markets - although we see some short-term trends that suggest a certain story, the next few months are sure to be prone to more surprises. Competing narratives of government-mandated shut downs, positive vaccine news, spiking COVID numbers, expiration of the CARES act, macroeconomic updates, President-Elect Biden’s potential cabinet members (and their perceived agendas) all loom in the coming months and will likely see-saw the markets. Keeping an eye on changing betas and what factors/assets drive those changes will be a critical analytic for all investors to keep an eye on in the coming months.
US & Global Market Summary
US Market: 11/16/20 - 11/20/20
- Treasury Secretary Steven Mnuchin surprised the market with Thursday’s announcement that the Fed wouldn’t have authority beyond year end for to backstop corporate credit, as well as the Main Street and municipal lending programs. Fed Chair Jerome Powell had intimated the need for these programs to continue on Wednesday. This announcement coincided with Mnuchin’s request that the Fed return ~$455B of unused capacity to the Treasury.
- US retail sales increased less than expected in October, up +0.3% last month vs. +1.6% in September, making clear the impact of the latest surge in US COVID infections and declining household income as additional federal stimulus remains unlikely in the near term.
- Thursday’s unemployment claims report from the Labor Department showed at least 20.3 million people on unemployment benefits at the end of October. 12 million people will lose benefits next month as two federally-funded programs roll off after Christmas.
- Initial claims for state unemployment benefits increased 31,000 to a seasonally adjusted 742,000 for the week ended Nov. 14th, vs. the consensus forecast of 707,000.
- The housing market saw continued strength as existing home sales increased +4.3% to a total rate of 6.85mm units in October, the highest since November of 2005 (according to the National Association of Realtors).
Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)
- Value continued to rally as it shot up +0.41 standard deviations and now appears poised to earn an Overbought designation.
- Market Sensitivity and Volatility both saw continued strength as risk-on remained en vogue.
- Profitability remained neutral on the week, sitting around the mean at +0.07 SD.
- The decline in Growth continued, as it fell into Oversold space at -1.16 SD below the mean.
- Earnings Yield continued to revert after peaking at +1.53 SD above the mean on 10/8, crossing over into negative normalized territory.
- Momentum saw more pain, falling by -0.55 standard deviations to -2.15 SD below the mean, firmly in Extremely Oversold space.
- US Total Risk (using the Russell 3000 as proxy) declined by 36bps.
Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)
- Similar to the US, Volatility and Market Sensitivity both had another strong week, as Volatility entered Overbought territory at +1.13 SD above the mean.
- The rally in Value continued as it rose +0.3 standard deviations further into Overbought territory, currently +1.26 SD above the mean.
- Size continued to climb higher into Overbought space, now sitting at +1.55 SD above the mean.
- Growth experienced another gap down (-0.62 standard deviations), falling further into Oversold territory at -1.62 SD below the mean.
- Momentum continued to plummet, 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 38bps.