Seeking Alpha

Evaluating Stock Sectors' Tail Risks Amid Rising Trade Tensions

|
Includes: XLB, XLC, XLE, XLF, XLI, XLK, XLP, XLRE, XLU, XLV, XLY
by: Sankalp Soni
Summary

The worsening trade tensions have made it ever more important to allocate wisely among the various stock sectors.

VaR analysis has revealed that not all defensive sectors offer shelter from worsening global trade conditions.

While all economically-sensitive sectors pose a greater tail risk, certain sectors offer better risk-reward attributes over others.

Trade tensions have been on the rise since January 2018 as the US has targeted various trading partners including China, Europe, Mexico, and Canada. Certain sectors and industries are reliant upon the trading partners as export markets; others are reliant on imports from these trading partners. Trump's trade wars have caused notable downturns in the stock market, and as these issues do not seem to be going away any time soon, it is worthwhile to evaluate which sectors pose the highest tail (loss) risk in order to better position oneself for future trade-inflicted risks. In this article, Value at Risk (VaR) and related risk-reward metrics for all stock sectors will be calculated in order to determine which sectors pose the most and least downside risks amid ongoing trade tensions. Specific variations of VaR will be used to conduct a more thorough comparative analysis of tail risk. Not all defensive sectors necessarily offer desirable protection from trade tensions, while certain economically-sensitive sectors may offer attractive risk-reward attributes.

The SPDR Select Sector ETFs (issued by State Street Global Advisors) will be used as proxies for each stock sector. For each sector ETF, the monthly returns from January 2018 (beginning of trade tensions) to May 2019 have been calculated, based on which VaR calculations will be conducted. Keep in mind that while the research & analysis seeks to evaluate the downside risks as a result of trade tensions, other sector-specific factors will have also impacted sector performances over this time period.

Note: the Communication Services Select Sector ETF (XLC) was only incepted in June 2019 and, therefore, monthly returns of this ETF (and resulting Value at Risks) will be calculated from its inception as opposed to the beginning of trade war (January 2018).

One of the most common ways of calculating VaR is by calculating the required percentile (confidence level) of returns, based on historical returns. We will be calculating VaR using a 95% confidence level throughout this article. The results for the one-month Historical VaR (95%) for each sector ETF have been summarized in the table below. For simplicity, the Historical VaRs for each sector have been sorted from best to worst.

Select Sector ETF

Historical VaR (95%)

Utilities (XLU)

-3.90%

Real Estate (XLRE)

-6.92%

Communication Services (XLC)

-7.01%

Health Care (XLV)

-7.29%

Consumer Staples (XLP)

-7.88%

Financials (XLF)

-7.96%

Materials (XLB)

-8.38%

Consumer Discretionary (XLY)

-8.39%

Technology (XLK)

-8.43%

Industrials (XLI)

-10.70%

Energy (XLE)

-11.55%

The Historical VaR returns essentially reflect that there is a 5% probability the returns for the ETFs over a one-month time period could be lower than the VaR thresholds and that we can be 95% confident the returns will be above these thresholds. The results reveal that based on Historical VaRs, the three sectors that pose the most tail loss risk are Energy (XLE), Industrials (XLI), and Technology (XLK). Conversely, the three sectors that pose the least tail risk are Utilities (XLU), Real Estate (XLRE), and Communication Services (XLC).

Normal VaR is also commonly used as a methodology to determine VaR (and incorporated into Monte Carlo simulations), whereby the returns are assumed to be normally distributed (or at least closely), and the mean and standard deviation of returns are used as parameters for VaR estimation. However, such an assumption would lead to misleading results for returns that don't follow near normal distributions. Quick checks to determine whether a dataset follows a normal distribution is to assess its skewness and kurtosis. For a normal distribution, the skewness is 0 and kurtosis is 3. Furthermore, the mean and median of a dataset should be roughly equivalent for datasets that closely follow a normal distribution. The table below details the skewness, kurtosis, mean, and median returns for each sector ETF.

Sector ETF

Skewness

Kurtosis

Mean (%)

Median (%)

Technology (XLK)

-0.51

-0.99

0.95

0.06

Communication Services (XLC)

0.78

0.51

-0.24

-0.46

Industrials (XLI)

-0.20

-0.51

0.05

0.23

Materials (XLB)

-0.64

-0.83

-0.56

0.28

Consumer Discretionary (XLY)

-0.51

-0.20

0.94

1.99

Financials (XLF)

0.04

-0.21

-0.13

-0.98

Energy (XLE)

-0.39

-0.45

-0.75

1.55

Real Estate (XLRE)

0.03

0.66

0.94

1.11

Health Care (XLV)

-0.36

-0.39

0.54

1.06

Consumer Staples (XLP)

-1.01

0.25

0.15

1.63

Utilities (XLU)

-0.69

-0.72

0.87

1.60

The statistics in the table above exhibit that no sector ETF returns have skewness near 0 (except XLRE) or a kurtosis near 3. Furthermore, the Mean and Median returns are certainly not closely equivalent for any of the securities. Hence, the sector ETF returns do not follow a normal distribution, and thus Normal VaR calculations (and Monte Carlo simulations based on normal distribution assumptions) for our securities would yield unreliable/meaningless results.

Nevertheless, the statistics revealing the lack of normality of returns distributions are actually helpful in further evaluating the tail risks for our securities. Modified VaR is another useful measure for determining tail loss risk, which incorporates the mean, standard deviation, skewness, and kurtosis of the returns to determine the potential tail risks. The most attractive attribute of Modified VaR is that it does not assume a normal distribution, as it accepts that the shape of returns distributions will possess skewness and kurtosis that result in deviation away from the Gaussian assumption. This allows for more reliable predictions for potential tail risks going forward. The one-month Modified VaR results (with 95% confidence levels) for each sector ETF are summarized below, once again sorted from best to worst.

Select Sector ETF

Modified VaR

Utilities (XLU)

-4.08%

Real Estate (XLRE)

-6.29%

Consumer Staples (XLP)

-7.72%

Health Care (XLV)

-7.96%

Communication Services (XLC)

-8.40%

Materials (XLB)

-9.04%

Financials (XLF)

-9.12%

Consumer Discretionary (XLY)

-9.18%

Technology (XLK)

-9.28%

Industrials (XLI)

-10.71%

Energy (XLE)

-13.11%

The results reveal that based on Modified VaR, the Energy, Industrials, and Technology sectors still pose the highest tail risks. The potential losses implied by Modified VaR are greater than the tail risks implied by the Historical VaRs for these sectors. Hence, incorporating the Modified VaRs into risk-management strategies amid rising trade tensions will allow for more prudent and conservative portfolio allocation decisions, as this metric reflects higher potential losses.

Furthermore, the sectors posing the least tail risk according to this measure are Utilities, Real Estate, and Consumer Staples. These outcomes slightly differ from the results of the Historical VaR analysis, which suggested that Communication Services posed one of the least tail risks as opposed to Consumer Staples.

While Modified VaR offers us an effective method for estimating potential tail risk going forward, we should take this a step further by evaluating the Conditional Value at Risk (CVaR). CVaR, also known as Expected Tail Loss (ETL), is the average expected tail loss in the event that the minimum return threshold (determined by VaR) is exceeded. CVaR can actually be considered a more conservative measure than VaR because the measure is impacted by return outliers, consequently reflecting greater potential loss forecasts. It could be worthwhile taking into consideration these conservative measures, given the unprecedented and often unexpected nature of Trump's style of imposing new tariffs on trading partners. Moreover, it offers a more realistic forecast of the potential loss incurred in the event that tail risks materialize. Therefore, based on our Modified VaR thresholds, the CVaRs for our sector ETFs have been calculated and summarized in the table below.

Sector ETF

CVaR (or ETL)

Utilities (XLU)

-4.08%

Real Estate (XLRE)

-6.81%

Consumer Staples (XLP)

-8.32%

Communication Services (XLC)

-8.40%

Health Care (XLV)

-8.66%

Materials (XLB)

-9.11%

Technology (XLK)

-9.28%

Consumer Discretionary (XLY)

-9.64%

Financials (XLF)

-10.13%

Industrials (XLI)

-10.79%

Energy (XLE)

-13.11%

The CVaR results reveal that while the three sector ETFs posing the least tail risks remain the same, the Technology sector (XLK) is no longer one of the sectors posing the greatest tail risk and instead has been replaced by the Financial sector (XLF).

We can extend the use of Expected Tail Loss (ETL) for risk-reward analysis. The Rachev ratio is a metric that reflects maximum upside reward potential relative to maximum downside risk. More specifically, the upside reward is determined by Expected Tail Return (ETR) and downside risk is determined by ETL. Given that our ETLs had been calculated using 5% margins on the left tail (95% confidence levels), our ETRs will also be calculated using 5% margins, but on the right tail. The table below exhibits the ETLs, ETRs, and resulting Rachev ratios for our sector ETFs. The higher the Rachev ratio, the more desirable its risk-reward attribute, as it offers more upside potential relative to downside risk.

Sector ETF

ETL

ETR

Rachev Ratio

Utilities (XLU)

-4.08%

5.47%

1.34

Real Estate (XLRE)

-6.81%

8.96%

1.31

Communication Services (XLC)

-8.40%

10.58%

1.26

Technology (XLK)

-9.28%

11.64%

1.25

Industrials (XLI)

-10.79%

11.35%

1.05

Consumer Discretionary (XLY)

-9.64%

9.86%

1.02

Health Care (XLV)

-8.66%

8.77%

1.01

Financials (XLF)

-10.13%

9.39%

0.93

Energy (XLE)

-13.11%

11.23%

0.86

Materials (XLB)

-9.11%

7.64%

0.84

Consumer Staples (XLP)

-8.32%

5.28%

0.64

Note: the results have been sorted by Rachev ratio (best to worst)

It is the sector ETFs that posed the least tail loss risks that offer the best risk-reward attributes, with Utilities and Real Estate topping the list. Unsurprisingly, these are also the sectors with the least global revenue exposure. Furthermore, while the Technology sector (XLK) posed one of the highest tail risks according to earlier VaR calculations, it also offers one of the highest Rachev ratios (1.25), implying that this is a sector worth taking risk in to try and profit from upside potential whenever signs emerge that trade tensions are cooling down.

Moreover, Modified VaR results had suggested that Consumer Staples (XLP) posed one of the least tail risks amid rising trade tensions; however, the Rachev ratio reveals that it offers the worst risk-reward ratio out of all sectors, whereby its downside risks strongly outweigh its upside potential. Additionally, the Energy sector (XLE), which consistently posed the highest tail risks from our VaR analysis, also offers one of the worst risk-reward ratios, confirming that is the worst sector to hold amid rising trade tensions. This is unsurprising given that no matter which goods are being tariffed, trade tensions threaten global economic growth conditions overall, and a worsening economic outlook suppresses demand for energy commodities such as Oil & Gas.

Bottom Line

As global trade conditions continue to worsen, effective risk management and pragmatic portfolio allocation are ever more important. Assessing the tail risks of the various sector ETFs using Historical VaR, Modified VaR, and Conditional VaR, and further evaluating their risk-reward attributes through Rachev ratio analysis has certainly helped reveal the best and worst sectors to hold exposure to amid rising trade tensions going forward.

The Energy sector posed the greatest tail loss risk, with a one-month Modified VaR (with 95% confidence) of -13.11%, while also offering one of the least attractive risk-reward attributes, with a Rachev ratio of 0.86. Whether Oil & Gas products are tariffed or not, mounting tariffs on other goods hurts global economic growth prospects, hence it makes sense to witness the Energy sector perform so poorly amid rising trade tensions. Though going forward, investors should also take into consideration factors such as the level of oil production by OPEC and other oil-producing nations, as supply dynamics will also impact energy prices and, consequently, the performance of the Energy sector (XLE).

On the other hand, the Utilities and Real Estate sectors appear to be the best sectors to hold exposure to, as their low global revenue exposures have resulted in comparatively limited Value at Risks, with one-month Modified VaRs (95% confidence) at -4.08% and -6.29%, respectively. Moreover, they also offer the best risk-reward characteristics, with Rachev ratios at 1.34 and 1.31 respectively. Therefore, the Utilities sector (XLU) stands out as the best defensive sector, over Health Care (XLV) and Consumer Staples (XLP), which hold more global revenue exposure.

It is also worth highlighting that out of the sector ETFs that posed the highest tail risks, the Technology sector (XLK) offered the best risk-reward potential, with a Rachev ratio at 1.25. Therefore, for investors that are seeking to allocate a specific portion of their portfolios to riskier assets to generate higher returns, the tech sector is certainly an area they should consider. Moreover, specific industries/companies within the tech sector may offer attractive returns due to high profit margins and limited exposure to countries such as China.

Apart from trade-related concerns and resulting tail risks, investors should also take into consideration other sector/industry-specific matters that influence performance going forward as well as fundamentals such as valuations. That being said, trade tensions do not appear to be going away anytime soon, hence investors need to be extra prudent about which sectors/industries they hold exposure to avoid costly tail risks.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.