The most sophisticated test of ESG fund performance, yet
I think the debate about whether ESG investments have higher or lower returns than conventional investments will never go away. I am on record for stating that I do not think there is any return difference between the two approaches, while advocates of ESG investing often claim that ESG investments have higher returns, countering claims by advocates of conventional investments that ESG investing should lead to lower returns.
Now, a team from the University of Tasmania took another shot at the debate and looked at all US equity funds available to retail investors in the US between 1999 and 2022. Taking a shot is an apt metaphor because they used every gun in their arsenal to investigate if there was a return difference between ESG funds and conventional funds when adjusted for all kinds of fund characteristics.
First, they started with the common artillery gun in the form of a univariate test on performance. They sorted all the funds on their beta to the US stock market and then compared the performance of ESG funds with conventional funds. Below is a chart of the performance difference for different values of market beta. Note that the performance difference is small, typically in the range of 0.2% to 0.3% per year.
Performance difference between ESG funds and conventional funds
Source: Boateng et al. (2024)
However, critics of ESG investing may point out that most of the time ESG funds slightly underperform conventional funds, so constraining a portfolio to include sustainability criteria seems to cost at least some return.
This is when the researchers unpacked a rocket launcher in the form of a multivariate test designed to identify a systematic difference in returns across all funds. I spare you the details, but the result here was that indeed there seems to be a systematic difference in returns between ESG funds and conventional funds, yet the test was inconclusive as to why that difference emerges. It could well be that this is due to fund-specific differences in valuation, etc. Crucially, the researchers note that all conventional regression-based tests implicitly require that the number of time periods used for the regression analysis needs to be much larger than the number of funds in the sample (which is why most research first groups the many funds into clusters to significantly reduce the ‘number of funds’ analysed).
Then they unpack the full arsenal of sophisticated econometrics tools that allow them to account for non-normal distributions and for the number of funds to vastly exceed the number of periods. I do not pretend to understand the details of the maths involved here, so I just have to take their word for it.
And their verdict is: Drum roll…
…they find no statistically significant difference in the efficient frontier of ESG funds and conventional funds and correspondingly no statistically significant difference in returns. Which is probably going to disappoint both ESG enthusiasts and ESG critics. As I like to say about my work: I am an equal-opportunity insulter.
That is indeed a counterintuitive finding, as the biggest knock on ESG funds has been that anything that reduces the investable universe, thus lowering portfolio diversification, mathematically has to impair performance over time. The implication of that has been some fund houses and their customers essentially saying "we'll accept a bit of likely underperformance because we believe we're doing the right thing." Which is fine for experienced investors to voluntarily accept.
The problem is that under ERISA https://en.wikipedia.org/wiki/Employee_Retirement_Income_Security_Act_of_1974 in the US, ESG has been at odds with fiduciary duty for many large private and public pensions. ESG strategies, especially exclusionary ones, can conflict with ERISA's diversification requirement if they limit the range of assets and sectors available for investment, as fiduciaries cannot prioritize ESG goals for their own sake over financial performance. ESG factors can only be considered if they are material to risk and return. The CFA Institute used to echo the ERISA fiduciary duty concern, but in recent years has suggested that instead of exclusionary approaches, investors can use ESG scoring or engagement strategies that allow for a broader investment universe while still emphasizing improved ESG practices.
If a study such as this shows that the outcome of applying ESG criteria to *not* be harmful, then that's great news for ESG investors. Personally, I always viewed applying ESG criteria as a risk management tool rather than anything morals- or values-driven (which used to be called "ethical investing").
However, Trump I issued rules emphasizing that ESG factors must not sacrifice financial performance. The Biden administration's ERISA guidance has softened these rules, allowing fiduciaries to consider ESG factors if they are material to investment decisions, effectively reinforcing ESG's role in prudent investing. Trump II could reel those looser guidelines back in again.
Could there be a calendar time influence? I am not sure of the facts, but I shall pose the argument. Large investment funds went into ESG funds as being a noble thing. This pushed up those stock prices. More recently investors have taken money out of ESG funds, so the underlying stock prices fell.