Sustainable Fixed Income Investing: Does ESG Actually Matter for Bonds?

Book: Systematic Fixed Income: An Investor’s Guide Author: Scott A. Richardson, Ph.D. ISBN: 9781119900139 Publisher: John Wiley & Sons, 2022

You can’t write about investing in 2022 without talking about sustainability. That’s basically how Richardson opens Chapter 10, and he’s right. ESG is everywhere now. But here’s the thing: does it actually matter for bond investors? Or is it just a feel-good checkbox?

Richardson walks through the evidence carefully. And the answer is… complicated.

The Rise of Sustainable Investing

First, the scale. The UN Principles for Responsible Investing (PRI) launched in 2006. By 2020, about $100 trillion in assets were committed to PRI. That’s not a niche thing anymore. The Global Sustainable Investment Alliance reported $35.3 trillion invested sustainably as of 2020, which was nearly 36% of total managed assets.

Most of this started in equities. But the fixed income world is catching up fast. There are now thousands of sustainable indices tracking both stocks and bonds. The direction is clear: asset owners want sustainable options.

Do ESG Scores Actually Predict Anything in Bonds?

Richardson draws heavily on research by Diep, Pomorski, and Richardson (2021). They looked at four corporate bond universes: US investment grade, US high yield, European IG, and European HY. For ESG data, they used MSCI scores (0 to 10 scale, measured relative to industry peers).

ESG and credit spreads: Without controlling for credit risk, better ESG scores are associated with lower credit spreads. That makes sense on the surface. But once you add a proper default probability model, the relationship gets much weaker. Only US IG showed a statistically significant link after controlling for fundamentals.

So ESG scores kind of overlap with what you already know from credit analysis. They’re not adding much new information about spreads.

ESG and future returns: This one is pretty clear. There’s basically no relationship between ESG scores and future credit excess returns. The test statistics are nowhere near significance across all four bond universes. Whether you control for expected returns or not, ESG scores don’t predict who will outperform.

ESG and risk (volatility): Here’s where it gets slightly more interesting. Higher ESG scores are associated with lower future volatility of credit excess returns. But again, the economic magnitude is small. After controlling for existing risk measures, a one-standard-deviation improvement in ESG scores is linked to only a 2-5% reduction in future volatility. Statistically significant in most cases, but not exactly a massive edge.

The Honest Takeaway on Direct Investment Relevance

Richardson is pretty straight about this. The direct investment relevance of ESG measures for bond investors is modest. The sample period is limited (2012-2020), they only tested one data provider, and the measures are broad aggregates. It’s possible that more targeted measures could show stronger results. But he also warns against data mining: don’t keep testing different measures and samples until you find something that works and then call it a discovery.

Building Sustainable Portfolios Without Wrecking Returns

Here’s where it gets more practical and more optimistic. Even if ESG doesn’t directly predict returns, can you build a portfolio that’s more sustainable without losing money?

Richardson argues that systematic approaches are uniquely suited for this. Think about it: you’re trying to maximize risk-adjusted returns while hitting multiple sustainability targets simultaneously. A discretionary manager juggling all that in their head? Good luck. A systematic process can handle these trade-offs in the optimizer.

The research tested several sustainability levers on a global high-yield portfolio:

1. Static screens removing controversial weapons, fossil fuels, and tobacco companies
2. Tactical screens excluding the bottom 10% of ESG scores
3. Sustainability tilts making the portfolio score 10% better than the benchmark
4. Carbon reduction keeping carbon intensity at least 25% below the benchmark

The combined impact on breadth? About 18% of market cap and 11% of issuers removed for US HY. That’s meaningful but not devastating. There are still enough liquid issuers across sectors to do proper security selection.

And the key finding: the expected return frontier for sustainability exclusions and tilts is relatively flat. You can make significant improvements in sustainability without much reduction in expected returns. This works because the systematic portfolio wasn’t specifically targeting low-ESG issuers in the first place, and there are enough remaining issuers to maintain good diversification.

For carbon intensity, you can achieve up to about 70% reduction before the return penalty becomes steep. That’s because carbon emissions are concentrated in a small number of industries.

The actual backtest results? The cumulative returns of the regular systematic portfolio and the sustainable systematic portfolio are nearly identical. Richardson mentions there are three lines on the chart but they’re so close together you can barely tell them apart.

Governance: The Part of ESG That Actually Works

One area where sustainability measures do show investment relevance is governance. Specifically, measures like aggregate accruals (a proxy for management quality and financial reporting discretion). Companies with higher accruals tend to have lower future profitability, more negative analyst revisions, and lower future credit excess returns. This is well-documented in academic research going back to the 1990s.

So if you’re looking for ESG factors that actually help bond returns, look at the G more than the E or S.

Country-Level Sustainability: Still a Work in Progress

Richardson closes the chapter by discussing sustainable government bond investing, which is much harder. Corporate bond universes have thousands of issuers, so excluding a few doesn’t hurt much. But government bond indices are extremely concentrated. The US and Japan account for 63% of a typical developed-market government bond index. If your sustainability criteria exclude one of them, your tracking error goes through the roof.

For emerging markets, the same problem exists with countries like Brazil, Mexico, Indonesia, and Turkey dominating the index. So government bond sustainability tends to use tilting rather than exclusions.

The UN Sustainable Development Goals (SDGs) offer one framework for country-level sustainability, with firms like Bridgewater Associates and APG Asset Management building investment processes around them. But there’s no consensus yet on how to measure and weight country sustainability. Different asset managers produce very different scores for the same countries.

The Bottom Line

Sustainability in fixed income is real and growing. But the honest assessment is:

• ESG scores don’t predict bond returns in a meaningful way
• ESG scores have a small relationship with risk, mainly through governance
• A systematic approach can integrate sustainability constraints without significant return sacrifice
• Country-level sustainable investing is still in its early stages
• The governance component of ESG has the strongest investment case

Richardson’s main point is encouraging: you can do well and do good, up to a point. You just shouldn’t expect sustainability to be an alpha source on its own.

Previous: Liquidity and Trading in Bond Markets Next: Systematic Fixed Income Portfolio Examples