Do green bonds offer higher yields than traditional bonds?

This post was originally published on 8 October 2019 at Enterprising Investor.

I need your help with something that puzzles me and so this blog post will be an experiment where I will ask you for your comments on this question at the end of the post.

When I look at ESG investing, there is a constant debate around the returns of ESG investments. Are they higher, lower or the same as comparable traditional investments? One argument that has been put forward is because ESG investing is essentially an approach where investors look at non-traditional sources of risk that are not always priced in the market, they might effectively have higher returns than traditional assets. 

The simplest way to check this conjecture is to look at bonds. There are plenty of issuers that issue traditional bonds as well as certified green bonds, i.e. bonds that explicitly help to enhance environment or social goals. Municipalities, states and governments as well as international organisations, such as development banks often issue bonds that are linked to specific projects. If the same issuer sells both traditional bonds and green bonds both types of bonds have identical credit risk from the perspective of the issuer, but the traditional bonds may have higher (lower) yield than the green bonds. The source of this higher (lower) risk in traditional bonds may be that ESG risks are perceived differently from traditional bond.  

There are studies like 2017 paper by the Bank for International Settlement that conclude  green bonds by the same issuer do indeed trade at lower yields (i.e. higher prices) to non-green bonds. However, while this study found green bond yields at issuance that were between 10bps (AAA-rated issuers) to 45bps (A- and BBB-rated issuers) lower than for non-green bonds of the same issuer, the variance of this premium was very high. With a standard deviation of 27bps between issuers the observed green bond yield premium was clearly not statistically significant.

On the other hand, there are studies on US corporate and municipal bonds that show green bonds trade at a yield premium to non-green bonds. A study by two researchers from the Sorbonne in Paris found an average yield premium of 8bps for green bonds vs. non-green bonds of the same issuer. Another study from two researchers at the neighbouring University Paris-Dauphine looked at bonds issued by French companies where legislative changes forced companies to become much more transparent in their ESG risks in 2015. They found no yield premium for green bonds or bonds of companies with lower ESG risks in the market. 

This result seems in accordance with a brand-new study by David Larcker and Edward Watts from Stanford University. They looked at municipal bonds issued in the US. The advantage of their study is the ability to look at bonds that were issued by the same issuer at virtually the same time. While some tranches of these newly issued municipal bonds were certified green bonds, others were not. The drawback of their study is that it is limited to US municipal bonds which are almost exclusively bought by taxable investors in the US directly or through intermediaries like mutual funds. Because municipal bonds are exempt from federal, state and municipal taxes for investors who live in the state or municipality issuing these bonds, they are particularly attractive to high income households in the US. They found that there is virtually no difference between the yields of green and non-green bonds (see our chart below) once the controlled pairs of bonds in the comparison sample are properly adjusted for fixed features such as callability terms and other specific tax differences. 

Yield differences between green and non-green municipal bonds issued with otherwise identical risks

Source: Larcker and Watts (2019).

These results sparked my interest and I had some discussions on this issue with my co-author Bill Fung (on a related project involving ESG Investing). The starting point of our discussion was simply why there is seemingly no observable risk premium or discount for green bonds?

In our discussions we found that these studies measure performance based on radically different assumptions about investor preferences in the green bond market. It seems that better understanding on how investors assess performance would provide important clues to determine how to measure this ever illusive green premium.  

And this is where we need your help.

The questions we are stuck with is that in our view it seems possible that there may be a structural difference between investors who buy green bonds and investors who buy traditional bonds. Does anyone know of studies that investigate why investors buy green bonds to begin with? Maybe you work at an organisation that invests in green bonds and are willing to share with us why and how you invest in these bonds? What goals do investors pursue with green bonds? Is it a feeling of moral superiority, risk management or other goals? And do you hold green bonds to maturity or actively trade in them to maximise periodic total return?

Similarly, does any of our readers know if there are systemic differences between issuers of green bonds and traditional bonds? Why do companies issue green bonds at all? Are managers of these companies incentivized differently?

As active participants in the Green Bond market, we would like to hear from our readers your opinion on how the green bond markets have evolved over time. Do you perceive improvements in depth and liquidity? 

I will try to collate your responses and come back to you with more focused questions.  Hopefully in time, together we can build a more accurate picture of the market structure of the green bond market beyond routine statistics.

Please write me an email with your answers and opinions to koi@jklement.com. My coauthor and I will collect your answers and summarize responses in a future post so that all our readers are able to learn from our collective wisdom.