ESG tilting provides some benefit to bond portfolios during COVID-19

This article is based on a report by the Bloomberg Quantitative Research team that appeared first on the Bloomberg Terminal. Existing subscribers can access the full version of “ESG and the corporate bond index performance during the COVID-19 crisis” on Terminal here.

Careful analysis reveals that high ESG (and particularly Governance) ratings provided a small benefit, consistent with longer-term historical analysis.

ESG has become an important consideration in portfolio construction over the last several years. Beyond conservational, ethical and social objectives, the investment rationale for higher ESG-rated companies presumes that such companies are better prepared to deal with anticipated as well as unexpected large-scale risks that the future may bring. The coronavirus crisis fast-forwarded us into the future and the world has to deal with a disruption of vast proportions. Should it be expected that highly ESG-rated companies weathered the violent market reaction better?

In that respect, we have analyzed the performance of bonds of high-ESG vs low-ESG rated issuers during the onset of the crisis in March 2020 controlling for all other important risk characteristics. While we identify a small outperformance of high-ESG portfolios, such outperformance was dwarfed by the magnitude of the bond (negative) returns.^[1] Although the outperformance was larger than the historical average, its magnitude was consistent with historical precedent as we reported in previous publications on the Bloomberg Terminal, where we established that since 2007 high-ESG bond portfolio have outperformed low-ESG portfolios by a small amount. In related research available on the Terminal, we also found evidence that downside risk tends to be slightly lower in high ESG portfolios.

In Figure 1, we report the performance of the US Corporate index, as well as the performance of each ESG Rating bucket for the period starting on February 28 and ending on March 20, when credit markets bottomed out.

 

Click image to expand

All ESG rating buckets had losses exceeding 10%, but the lower-rated buckets had significantly bigger losses than the higher-rated buckets. A naïve analysis might attempt to attribute this difference in returns to the ESG factor, however a more careful look reveals that there is a strong relationship between the ESG rating and the credit rating (higher index rating numbers represent a worse credit quality). Hence, it is reasonable to assume that most of the return differentiation can be attributed to the difference in credit risk exposure.

In order to isolate the impact of ESG ratings during the crisis we follow the same approach as in our long-term studies. We split the US IG Corporate Index into buckets with the same sector and rating, and within each bucket, we select a subset of bonds with the highest ESG rating and another subset with the lowest ESG rating, making sure that the DTS of the high-ESG subset and the low-ESG subset is equal to the DTS of the bucket. We then combine all high-ESG subsets into a high-ESG portfolio and all low-ESG subsets into a low-ESG portfolio, using the bucket weights. We now have a high-ESG portfolio and a low-ESG portfolio that have the same risk profile with the Corporate Index. By comparing the performance of these two portfolios, we can isolate the impact of the ESG rating.

The average monthly outperformance of the high-ESG portfolio relative to the low-ESG portfolio from January 2007 to March 2020 has been 3.8 bps. The outperformance from the beginning of March until March 20 was 12 bps, larger but not sufficiently so to be statistically significant. As a matter of fact, there is no reason to expect that the “E” and “S” scores would have any impact on performance during the coronavirus crisis. For example, it is not clear how the dramatic drop of in energy demand would affect clean energy companies relative to their traditional energy counterparts. However, one could argue that proper governance should have some impact on how well a company might cope with the disruptions brought about by the crisis. We therefore repeat the above analysis by focusing only on the “G” rating.

The monthly outperformance of the “high-G” portfolio vs the “low-G” portfolio since the January 2007 is shown in Figure 2.

We see that after the 2008 market crisis, the “high-G” portfolio is showing a small but steady outperformance. The average monthly outperformance is about 2.5 bps which accumulates to about 4% over the period of our study. The outperformance during the onset of the coronavirus crisis (beginning of March until March 20) was 45 bps. While this is almost 20 times higher than the monthly average, it is still not statistically significant. Indeed, since 2010 outperformance of similar magnitude is recorded about once every year. Nevertheless, it is encouraging to see that the behavior of high ESG portfolios during the coronavirus crises was consistent with our prior analysis and our expectations.

 

Click image to expand

Bloomberg offers two major families of ESG bond indices, the Bloomberg Barclays MSCI Sustainability indices (exclude bonds from issues with an ESG rating below BBB and/or that are experiencing severe ESG controversies), and the Bloomberg Barclays MSCI ESG-Weighted indices (reweight all bonds in the index according to their ESG rating and the ESG rating trend). The tickers for the indices derived from the US Corporate index are SUCMTRUU for the Sustainability and UCEWTRUU for the ESG-Weighted version. In Figure 3 we plot their year-to-date difference in total return vs. the US Corporate index, using the COMP <GO> terminal function.

 

Click image to expand

For the first two months of the year, we see the differences in return between the three indices were negligible – as expected due to the very tight tracking errors. However, by the end of February, we note that both ESG indices began to outperform the US Corp, precisely when the market began to come down. Moreover, the relative outperformance peaked at about 1% precisely around the market bottom on March 20. Of course, as discussed previously, this could be driven by small credit risk differences between indices rather than by their ESG score differences. Regardless, ESG tilting during the onset of the coronavirus crises mitigated some of the losses of bond portfolios, both in its own right but also because ESG tilting tends to bias portfolios towards issuers with higher credit rating.

For follow up on the Bloomberg family of ESG indices as well as methods for constructing ESG-tilted portfolios and custom indices using the Bloomberg portfolio and index construction tools, please contact your Bloomberg Index specialist.