Indexology^® Blog

In a prior blog series,[1] we explored the relationship between environmental (E), social (S), and governance (G) scores and future stock performance. In all three cases, the results showed that top quintile portfolios outperformed the bottom quintile portfolios. However, a deeper analysis revealed that the spread between Q1 and Q5 portfolios was the highest for the G-ranked portfolio (1.68%) and the lowest for the S-ranked portfolio (1.34%) over a long-term investment horizon (January 2001-December 2017). In medium-term periods, such as five-years, the spread returns for the E-ranked portfolio was the highest (2.70%), while the S-ranked portfolio performed the worst (-0.63%).

Therefore, ESG-minded market participants should be aware of their investment time horizon and the return expectations associated with that time horizon.

According to the RobecoSAM scoring process, the relative weights of the ESG score components vary by industry due to materiality. For example, as shown in Exhibit 1, the E dimension warranted a higher weighting in the electric utilities industry compared to the banking or pharmaceutical industries, while the G dimension carried the highest weighting in pharmaceuticals. The total ESG score was then calculated after applying these component industry-specific weights for E, S, and G.

Given the return information derived from component scores and future stock performance analysis, investors may wish to alter the weights of ESG components. To illustrate whether reweighting matters, we calculated two hypothetical ESG portfolios by reweighting the E, S, and G scores for each security and re-ranking the universe (see Exhibit 2). For example, we constructed an ESG – S Light portfolio in which each security in the universe received a weighting of 40% in E, 20% in S, and 40% in G. We also constructed a 50% E and 50% G portfolio in which S was eliminated altogether.

By reweighting, market participants can incorporate the return expectations of ESG components into portfolio construction and overweight the component that is the most economically meaningful to them. Exhibit 3 shows the annualized returns for the three hypothetical portfolios formed using different component weighting combinations. Consistent with previous findings, the bottom Q5 portfolios for all three scenarios performed the worst. It is worth noting that the spread between the top Q1 portfolio and the bottom Q5 portfolio was the highest for the E&G portfolio.

In Exhibit 4, we display all three scenarios and their performance across different investment horizons. Up to this point in our analysis, one of the consistent takeaways was that avoiding the worst quintile and instead opting for the top three quintiles had economic benefits, ranging approximately 200 bps.

Our prior analysis indicated that, while ranking by overall RobecoSAM ESG score could lead to positive returns, the three underlying components had different relationships with future stock performance. Therefore, investors may wish to alter the weight of each component in the overall ESG score.

We found that integrating ESG into the investing process can be a bespoke, highly customized effort because different investors have different views of which subcomponent is material to them.

[1]   Exploring the G in ESG – Part I
Exploring the G in ESG – Part 2
Exploring the G in ESG – Part 3

The Dow Jones Commodity Index (DJCI) was up 2.9% for April 2018 and up 3.7% YTD, while the S&P GSCI was up 5.0% for the month and 7.3% YTD. Precious metals was the worst-performing sector in the indices and energy was the best.

Exhibit 1 depicts the month-to-date and YTD performance of the sector indices in the S&P GSCI.

Precious metals was down 0.4% for the month and flat for the year, weighed down by a strengthening U.S. dollar. Livestock’s decline of 8.9% YTD erased the gains it earned in 2016, when it finished the year up 8.4%. The loss was driven by price declines in lean hogs, which make up about 32% of the sector and were down 15.1% YTD because of ample global supplies.

The S&P GSCI Energy was up 6.5% for the month, and all the energy commodities were positive for a second consecutive month. Petroleum prices increased as U.S. drilling declined, a decline was seen in global oil stocks, and Saudi Arabia’s Crown Prince, Mohammad Bin Salman, announced to the press in March that Saudi Arabia will be working with Russia on a deal to extend control over major exporters over a period of one to two decades. In April, Brent crude was the best performer, up 8.7%, followed by heating oil, up 6.9%. Natural gas was down 0.1%, due to warming weather conditions.

Of the 24 commodities tracked by the S&P GSCI, 17 were positive in April. Exhibit 2 depicts the April performance of the single commodity indices.

Aluminum was the best-performing commodity in the indices, up 14.9% in April, after proposed sanctions on Russian aluminum producer Rusal were announced. Sugar was the worst-performing commodity in the indices, down 5.2% for the month and off 22.1% YTD, due to a global surplus.

Exhibit 3 depicts the performance for sugar and aluminum since index levels were rebased to 100 on April 30, 2008.

It can be seen in Exhibit 3 that the S&P GSCI Aluminum has reverted back to its late-2014 levels, when demand exceeded supply, while the S&P GSCI Sugar has reverted to its 1999 levels. To understand the concept of index levels, a hypothetical portfolio of USD 100 tracking an index based at 100 would increase by USD 20 if the index levels increased to 120 and decrease by USD 20 if the index level declines to 80.

The European Central Bank (ECB) announced last Thursday, April 26, 2018, that it would maintain its monetary policy and bond-buying program, as growth in the eurozone slowed in the first quarter. The ECB corporate bond purchases have pushed yields in the region to their lowest since the financial crisis. Inflation targets in the region are not expected to be reached and monetary stimulation could continue for longer than expected. Corporate bond yields are dramatically lower than comparable U.S. and UK markets. Investors in the region have speculated on the effect a cut in monetary stimulus would have on the asset class. Today’s announcement continued the uncertainty and the likelihood that eurozone corporate bond yields are not poised to rise anytime soon.

The U.S. investment-grade corporate bond market has seen yields getting progressively higher with the U.S. Fed Rate hikes over the past year. The S&P 500^® Investment Grade Corporate Bond Index, which is designed to measure the performance of U.S. corporate debt issued by constituents in the S&P 500 with an investment-grade rating, yielded 3.85% as of April 25, 2018—rising 74 bps year-over-year. Meanwhile, the S&P Eurozone Investment Grade Corporate Bond Index has seen its yield rise only 1 bps in the past year, going from a yield of 0.77% to 0.78%. The eurozone index yield sank to lows of 0.50% back in November 2017. The option-adjusted spread (OAS), which measures the spread over a risk-free rate (usually a treasury/government bond), for both indices has tightened in the past year between 8 bps and 16 bps, showing the effect the ECB’s bond purchasing program has had on eurozone corporates. German bunds have largely been range-bound in the sub 0.10% level, while U.S. Treasury yields have risen nearly 100 bps in the past year. Despite concerns that the asset class could be subject to a correction once the ECB switches monetary policy, and despite the attractiveness of higher rates in the U.S., a weaker-than-hoped-for European economy may keep corporates below the 1% level for longer than expected.

In April, the S&P 500 (TR) gained 0.4%, ending its first consecutive monthly loss in almost two years, but the index is still down 0.4% year-to-date (ending April 30, 2018.)  Mid caps are also down for the year, -1.0%, after the S&P 400 (TR) lost 0.3% in April.  However, Small caps pushed into positive territory, up now 1.6% year-to-date, from the S&P 600 (TR) gain of 1.0% in April.

Overall, 6 of 11 sectors gained in large and mid caps while 8 of 11 gained in small caps.  Energy led the gains across the size spectrum with total returns of 9.4%, 13.4% and 16.6%, respectively in large, mid and small caps that more than tripled the next best sector’s returns (S&P 500 Consumer Discretionary gained 2.4%, S&P 400 Utilities gained 3.9%, and S&P 600 Telecommunications gained 5.9%.) It  was the S&P 500 Energy’s 17th best month on record since October 1989, and it gained most since Sep. 2017.  Consumer Staples posted its 28th worst month in history, losing 4.3%, making it the 3rd consecutive monthly loss and worst 3-month loss (-12.5%) since the 3 months ending in Feb. 2009.

Energy’s outperformance not only propelled small caps to outperform large caps (since smaller energy companies rise more with oil) but drove value to outperform growth.  The S&P 500 Value has 12.5% more energy than the S&P 500 Growth, which has nearly none. While the value outperformed growth across all sizes for the first time in 2018, the mid and small caps had a much greater premium (respective 1.8% and 1.3%) than the large cap premium at 0.2%.  The mid cap premium was the most since Nov. 2016 and the small cap was most since Sep. 2017.  It is also the first time large cap value outperformed growth for 4 of 6 months since the second half of 2016.

As shown at the end of 2017, when growth and large caps outperform as much as they did in 2017 (that was the most since 1999,) the trend reverses.  That’s what seems to be happening now.

In a previous blog, we performed preliminary exploration of rising interest rate exposure of the S&P 500^® Low Volatility Index. In this blog, we continue the analysis to see if there is a relationship between the magnitude of interest rate change and magnitude of active return of the low volatility index relative to the S&P 500. To do so, we run a regression line by plotting the historical monthly excess returns (y-axis) against the monthly interest rate changes (x-axis).

Looking at the trend line in Exhibit 1, there is a downward sloping, negative relationship between the degree of interest rate movements and the excess return of the low volatility index relative to the S&P 500. The regression equation, also shown in the chart, confirms the negative relationship.

The regression equation has a slope coefficient of -3.07 and an r-squared value of 8.8%. The coefficient indicates that for every 1% change in interest rate, the excess return of the low volatility index is expected to change by -3.07 times. For example, if interest rates rise by 1%, the relative return is expected to be -3.07%. Conversely, if rates decline by 1%, the excess return is expected to be 3.07%.

The r-squared value is the trend line’s “goodness of fit” to the data; in essence, it is the explanatory power of interest rate movements on excess returns. We note that the r-squared value is relatively low; however, the coefficient to interest rates is statistically significant. Ensuring that coefficients are statistically significant when it comes to factors that have low explanatory power, such as macroeconomic factors, on equity performance is especially critical. In this case, the t-stat of the interest rate change coefficient is -5.61, which is significant at the 99^th percent confidence interval.

Combined with the findings in the first blog, we can conclude that, historically, the S&P 500 Low Volatility Index tends to be negatively affected by rising interest rates. In a subsequent blog, we will explore an alternative low volatility index strategy that is designed to reduce interest rate exposure while still preserving low volatility properties.