Much attention has been given in recent years to whether or not sustainable investors can expect to receive higher or lower returns for investing with their conscience. Most of the focus so far has been on equity investors. But what about sustainable bond investors? How will their returns differ from those of the broader bond market? LARRY SWEDROE investigates the latest evidence.
While most of the concerns over global warming and the economic risks of climate change have focused on equity risks, academic researchers have also begun to investigate the impact on corporate bonds and the implications for sustainable bond investors.
For example, Michael Halling, Jin Yu and Josef Zechner, authors of the 2020 study Primary Corporate Bond Markets and Social Responsibility, found that good environmental (E) and social (S) performance has been rewarded with lower credit spreads even when controlling for bond ratings and various firm characteristics, such as net book leverage, size, industry and profitability. The effect was strongest for low-rated bonds; for highly rated issuers (i.e., AAA or AA), the aggregate E and S score was insignificant. An important observation was that they found evidence that the explanatory power for spreads had decreased in recent years, the likely explanation being that in late 2015 Moody’s and S&P announced they would take ESG dimensions more explicitly into consideration when determining credit ratings, thereby reducing the information content in the respective E and S scores.
Tinghua Duan, Frank Weikai Li and Quan Wen, authors of the 2020 study Is Carbon Risk Priced in the Cross-Section of Corporate Bond Returns?, found that bonds of high carbon emission (CE) firms were riskier on average than those of low CE firms as indicated by a higher bond market beta, downside risk, higher illiquidity and lower credit ratings. They also had more negative cash flow surprises and deteriorating creditworthiness in the future. And despite their greater risk, the bonds of high CE firms significantly underperformed the bonds of low CE firms. The low carbon premium effect was economically significant: Corporate bonds in the lowest carbon emissions intensity (CEI) quintile generated 1.7 percent (t-stat = 2.62) higher returns per annum than bonds in the highest CEI quintile. Like Halling, Yu and Zechner, they found that the effects were mostly in high-yield bonds and that the low carbon premium declined in the post-2015 period—indicating the market was becoming more efficient, recognising the risks from carbon emissions.
Conflicting forces
Given the heightened investor interest in sustainable strategies, keep in mind that investor preferences lead to different short- and long-term impacts on asset prices and returns. Firms with high sustainable investing scores earn rising portfolio weights, leading to short-term capital gains for their stocks — realized returns rise temporarily. However, the long-term effect is that higher valuations reduce expected long-term returns. The result can be an increase in green asset returns even though brown assets earn higher expected returns.
There is, in other words, an ambiguous relationship between carbon risk and returns in the short term. Over time, as the markets develop a better understanding of carbon risk and the unexpected component falls relative to the expected component, we should expect a positive relationship between returns and carbon risk.
Latest research
Peter Diep, Lukasz Pomorski and Scott Richardson contribute to the literature with their September 2021 paper, Sustainable Systematic Credit, in which they assessed the direct investment relevance of ESG measures for U.S. and European corporate bonds. They began by noting that there have been substantial flows into sustainable investment grade (IG) and high yield (HY) strategies, with inflows of $79.2 billion into ESG-labeled corporate IG strategies and $31.6 billion into ESG-labeled HY strategies from January 2020 through January 2021, across both U.S. and European markets. The authors explained: “While this growth is exceptionally large relative to history, it is still small relative to the respective size of the U.S. (EU) corporate bonds market of nearly $10T ($3.5T). With the continued rise of asset ownership in sustainable-linked investments and regulatory pressure to ensure investments are linked to sustainable objectives, this is a trend that is likely to continue and perhaps even accelerate.”
To measure an issuer’s ESG score relative to industry peers, they used MSCI ESG scores (designed to provide users a holistic assessment of companies’ ESG risks), which ranks scores from 0 (worst) to 10 (best in class). To capture an issuer’s greenhouse gas emissions, expressed in terms of carbon dioxide (CO2) equivalents, they used Trucost data on Scope 1 (direct emissions from production) and Scope 2 (indirect emissions from consumption of purchased electricity, heat or steam) emissions. Their sample covered the period December 2012-April 2021. Following is a summary of their conclusions:
There was a consistent negative relation between ESG and credit spreads, implying that higher scores of sustainability were associated with lower credit spreads, suggestive of an ex-ante pricing impact of ESG/sustainability. The negative relation generally extended to the component measures E, S and G.
A change in the ESG score from the lower quartile to the upper quartile was associated with a 17 basis points lower credit spread, corresponding to 10-15 percent of the typical credit spreads in the sample.
After controlling for default probabilities directly, there was only modest evidence that ESG measures explain cross-sectional variation in credit spreads — part of the ESG information that matters might be subsumed by measures of default probabilities.
There was little evidence that ESG measures are related to future credit excess returns.
There was a marginal increase in ESG scores over time (stronger for IG markets).
The composite ESG measure, and component measures, showed muted correlations with the carry, defensive, momentum or value factors in bonds. Thus, there is the potential for ESG/sustainability measures to provide a diversifying source of return predictability.
Diep, Pomorski and Richardson also analyzed how ESG measures can be incorporated in an investment process to help achieve the joint objective of maximising risk-adjusted returns and a sustainability target: “Survey evidence suggests a range of design choices that are typical for sustainable investors: (i) static tilts to remove issuers engaged in generally accepted non-sustainable business models (e.g., controversial weapons, tobacco or coal), (ii) dynamic tilts to remove issuers that score poorly relative to their peers across ESG measures, (iii) tilting toward issuers that score favourably relative to their peers across ESG measures, and (iv) reducing the carbon intensity of the portfolio via under-weighting heavy greenhouse gas emitters.”
In developing sustainable corporate bond portfolios, they adopted two additional ESG-related objectives. First, they required the final portfolio to yield an ESG score that was at least 10 percent better than the benchmark. Second, they required their portfolio to have at least 25 percent lower carbon emissions than the benchmark. Using these measures, they found:
There was a reduction in investment breadth because of the exclusions. However, there were still enough sustainable issuers to be able to engage in security selection and still provide benchmark-relative performance.
There were only modest distortions (lower expected returns and possibly higher tracking error from a reduced investment opportunity set) from incorporating such ESG objectives.
These findings led the authors to conclude: “It is currently possible to have significant improvements in the ESG and carbon emission profile without meaningfully changing portfolio attractiveness.”
With a nod to the aforementioned issue of conflicting forces, they added: “In the presence of ESG-aware or ESG-motivated investors, it is unclear as to whether securities with higher ESG scores will earn lower or higher future excess returns, and when those return patterns will be realized. If investors shun securities with poor ESG attributes these securities will experience low returns whilst investors remove these securities from their portfolios. However, going forward those securities may earn a high return as compensation for the relative investor neglect. The dynamics for this depend on the mix of investor types and their trading decisions. Unfortunately, this can greatly complicate empirical analysis on the relation between ESG or sustainability measures and future returns due to the impact of in-sample flows into or out of securities with high or low ESG scores. Specifically, it may be the case that flows into securities with favourable ESG scores experience positive returns whilst that net flow is occurring, but once a new equilibrium of holdings has been reached those same securities will experience a lower future return.”
So what does this mean for sustainable bond investors?
The above findings provide good news for sustainable bond investors in that the evidence shows that tactical exclusions (such as removing corporate issuers that score poorly on dimensions of ESG relative to peers), sustainability tilts (such as ensuring the overall portfolio is superior to the benchmark along the ESG dimension) and economically meaningful reductions in carbon intensity can be achieved without sacrificing much return potential.
This finding is consistent with Greg Richey’s finding in his 2017 study on sustainable investments and their impact on equity risks and returns, Fewer Reasons to Sin: A Five-Factor Investigation of Vice Stocks. Richey found that the excess returns of sin stocks were well explained by their exposure to the five Fama-French factors (beta, size, value, investment and profitability).
In other words, the research shows that sustainable bond investors can “have their cake and eat it too” — they can pursue their sustainability objectives exclusively, at least using intelligent design that takes into account factor exposures, without a detrimental impact on the likelihood of achieving their financial objectives.