A Deeper Dive into Securities Exposed to the Global Trade Slowdown
For those who missed last week’s Factor Spotlight, my name is Chris Martin, Product Manager at Omega Point, with a follow-up post on Exchange Rate Sensitivity. If you did miss it, last week’s post is a potentially helpful primer for this week’s edition. All of us at Omega Point hope that you and your families remain healthy and safe as we collectively ride this out.
Today, we’ll dig deeper into Exchange Rate Sensitivity to better understand it’s characteristics in terms of Industry compositions and style exposures, and how these characteristics have changed during the recent Coronavirus pandemic and accompanying extreme macroeconomic environment. We’ll also provide our weekly market and factor update.
Exchange Rate Sensitivity (“ERS”) - Analyzing the Extremes
To better understand the characteristics of companies with extreme exposures to ERS, we create two portfolios based on Axioma US4-MH’s ERS factor. Below is our methodology for creating these extreme portfolios:
High ERS: all securities in the Russell 3000 with an exposure to ERS > 1.0, cap-weighted*
Low ERS: all securities in the Russell 3000 with an exposure to ERS < -1.0, cap-weighted*
We perform this process for two different dates - 12/31/2019 (i.e. stable market) and4/7/2020 (turbulent market) to see how recent market events have changed these high/low ERS portfolios and their characteristics.
* Note we make one manual adjustments to the cap-weighting scheme for each date. On 12/31/2019, Facebook is 50% of the high sensitivity portfolio and on 4/9/2020 Amazon is nearly 60% of the high sensitivity portfolio. Given we don’t want these mega caps to skew our analysis, we cap their weight at the next highest weight in each respective portfolio.
We see that the Low portfolio has much higher beta (Market Sensitivity), with both the High and Low having higher excess volatility and more liquid smaller cap names. We also see the Low portfolio has a lower Profitability exposure, lower Earning Yield exposure, and higher Value exposure. Although there are certainly some difference in Style exposures, the differences aren’t as big as one may expect given the extreme portfolio construction methodologies - we certainly saw big differences in Industry weights.
We see that the end of 2019 the High and Low portfolios are allocated across the industries we’d expect to be sensitive to exchange rates. The High portfolio is allocated in Industries that generally rely on exports, whereas the Low portfolio is allocated in Industries that generally import. Please note that this chart does not indicate that every asset within these Industries have either high or low Exchange Rate Sensitivity exposures, rather, this helps us understand the market cap weighted allocation across Industries for high and Exchange Rate Sensitivity exposure assets. For example, domestic shale producers in Oil, Gas & Consumable fuels likely wouldn’t have as low Exchange Rate Sensitivity exposure as they produce oil domestically as opposed to importing it.
We now move on to the High/Low portfolios for 2020, constructed after the risk model’s digestion of all the big recent market moves.
The High portfolio now has higher excess Volatility, and has a lower exposure to Market Sensitivity than the 2019 High portfolio - so although this portfolio is lower beta, there is higher excess volatility. The 2020 Low portfolio also picks up an interesting exposure to Dividend Yield, something we saw the opposite of in both High and Low portfolios in 2019.
We observe some very interesting changes in the industry exposures from Dec 31, 2019:
- The Software and Internet Retailers (AMZN) now represent 35% of the high ERS portfolio (from ~11% at the end of 2019).
- The biggest surprise are Oil & Gas and Biotech: Oil & Gas switching from ~10% of the Low portfolio in 2019 to ~5% of the High portfolio in 2020. Biotech switching from ~10% of the low Portfolio in 2019 to ~8% of the High ERS portfolio in 2020.
- We also see REITs and Electric Utilities make the Low portfolio, which is likely what is driving the positive Dividend Yield exposure above. We also see several Industries appear in the 2020 High and Low portfolios that we not in the 2019 portfolios - is this due to fundamental import/export strength or is something else going on?
We start by looking at Extraction Oil and Gas XOG — a security which is in the 2019/Low portfolio and in the 2020/High portfolio and whose Exchange Rate Sensitivity changed the most from 12/31/2019 to 4/9/2020:
We see a sharp increase in ERS exposure for XOG between March 19-23:
Below is XOG’s YTD performance from Exchange Rate Sensitivity, showing a large move between March 23-26:
Why would, XOG, with strong negative exposure to ERS suddenly switch to having strong positive exposure in such a short period of time?
As we had described in last week’s post: Exchange Rate Sensitivity is calculated by regressing a security’s stock price return on the SDR currency basket. On the day the Fed announced QE infinity, XOG’s stock return was negative and the USD depreciated, indicating a positive beta. But in this case, decreasing oil prices is the bigger driver of its negative return, not the weakening of the USD. This is different from what we’d normally expect - traditionally these companies are net importers and thus normally have a negative exposure to Exchange Rate Sensitivity.
In times of such dramatic market shifts, we need to be careful in our interpretation of factors we normally expect to behave a certain way. Over this time period, oil crashed, the Fed made an emergency rate cut, unleashed potentially unlimited Quantitative easing, Congress passed a $2T stimulus, and the relative strength of USD plummeted. This is not a time of normal global importing and exporting, as supply and demand is shocked, albeit disproportionally across the world and industries.
We’ve seen some big changes in the normally docile ERS factor driven by outlier events. It remains to be seen if securities with extreme changes in ERS remain at their current levels, if they start to trend back to their longer term averages, or if they will swing wildly again with the appearance of more unexpected Quantitative easing events from the world’s fiscal authorities. Specifically we know that the factor return for ERS can make major moves if the relative strength of the USD changes dramatically.
As a steward of capital, making sound investments while managing risk is that much more complex when historical factor relationships are out-of-whack. Monitoring your portfolio specifically for large changes in ERS is heavily advised. Furthermore, identifying securities in your portfolio that have seen large unintuitive changes in their exposure to ERS could open up investment opportunities if you are positioned for a rebound in the company’s long-term relationship to the factor.
US & Global Market Summary
US Market: 4/10/20 - 4/16/20
- The S&P 500 advanced higher for the second week in a row after seeing a +2.68% gain on Friday (not captured in above chart). Investors were encouraged by news that new infection rates appeared to be peaking and various governments were starting to plan reopening their economies.
- COVID-19 cases topped 688,000 in the US, while news about potential treatment from Gilead (GILD), rapid testing from Abbot (ABT), and a potential vaccine from Johnson & Johnson (JNJ) gave investors reason for optimism.
- New jobless claims continued to soar, with 5.5 million workers applying for unemployment insurance over the past week.
- U.S. home-building activity plummeted in March, with housing starts down 22.3% over last month.
- Several banks kicked off 1Q20 earnings season last week, including JPM, C, BAC, and PNC. While initial results caused these banks’ shares to fall (profits fell as they shored up loan-loss reserves), it remains clear that the Fed and other central banks will continue to enact policy that will help cushion the shocks to the financial system.
Factor Update: Axioma US Equity Risk Model (AXUS4-MH)
- Value was the biggest winner this week, as it crossed over the mean and headed into positive territory after being nearly 3 SD below the mean in Feb.
- Market Sensitivity saw continued strength on a normalized basis, as it crossed into Overbought territory at 1.52 SD above the mean. This factor was at -3.45 SD below the mean on March 18th.
- Volatility now sits exactly at the mean after seeing a +0.53 standard deviation boost over the week.
- Growth also saw a recovery to the tune of +0.39 standard deviations as it headed away from Oversold territory.
- While Earnings Yield (another proxy for Value) still saw strength this week, it slowed down on a normalized basis after vaulting up one full standard deviation last week.
- Medium-Term Momentum teetered at the edge of an Oversold designation before seeing a slight boost this week.
- Size continued to fall from a peak of +3.2 SD above the mean on 3/18, and continued its trajectory towards exiting Overbought space.
- Weakness in Profitability persisted, as it fell 0.73 SD and remains Extremely Overbought (recent peak was +6.47 SD above the mean on 3/20).
- US Total Risk (using the Russell 3000 as proxy) decreased by 66bps.
Factor Update: Axioma Worldwide Equity Risk Model (AXWW4-MH)
- Worldwide Volatility saw another +1 SD move, and is now within knocking distance to an Overbought designation.
- Market Sensitivity also saw a +1 SD move, and now sits in Overbought territory after moving almost +2.5 SD in two weeks.
- This week’s Factor Spotlight: Exchange Rate Sensitivity saw a 0.67 SD move but is still in extreme oversold territory.
- Value also continued to recovery internationally, heading into positive normalized territory for the first time since mid-January.
- Earnings Yield saw a slight decline as it remains in Extremely Overbought space, now sitting at +2.25 SD above the mean.
- Profitability continued to fall down to earth, while still hovering over +3 SD above the mean.
- Size continued to revert, falling to +1.46 SD above the mean and becoming merely an Overbought factor.
- Global Risk (using the ACWI as proxy) declined by 56bps.