Four Facts About ESG Beliefs and Investor Portfolios

[What sustainable investors actually believe 3/5]

What drives retail investors into, or away from, sustainable funds?

Stefano Giglio, Matteo Maggiori, Johannes Stroebel, Zhenhao Tan, Stephen Utkus, and Xiao Xu connect what investors say they believe with what they actually hold in their paper "Four Facts About ESG Beliefs and Investor Portfolios".

They analyse ten waves of the GMSU-Vanguard survey fielded between June 2021 and December 2022, with around 2 000 responses per wave from US retail and retirement investors.

Their main conclusions include:

• Investors on average expected ESG investments to underperform the market by 1.4% per year over ten years, and the gap widened from 1% in June 2021 to 2% by December 2022.
• 45% of respondents see no specific reason to invest in sustainable assets, 25% cite ethical considerations, 22% value a hedge against climate risk, and only 7% expect outperformance.
• Return expectations are hugely dispersed, with a 4% standard deviation across investors, and they are mostly uncorrelated with their views on market returns, GDP growth, or disaster probabilities.
• Only 3.5% of investors hold any ESG fund, with an average portfolio share of 0.4%. Ethically motivated investors hold the most, and about half of them cite ethics as their primary motive.
• Financial considerations limit investment even among the ethically motivated: 4% of them hold ESG funds when expecting annual excess returns below -0.5%, against 12% when expecting more than 0.5%.
• Who is answering explains far more of the variation in ESG return expectations than when they answer, and observable demographics explain almost none of it.

The paper shows non-pecuniary framing alone does not move allocations, as even ethically motivated clients scale exposure with expected performance. Credible performance evidence and cost discipline remain central.

For analysts, the heterogeneity in beliefs and motives is the input equilibrium models of sustainable investing need, in the tradition of Pástor, Stambaugh, and Taylor, to generate a greenium.

We should however note the sample is naturally biased, as it covers one US asset manager's clients, older and wealthier than average, over an 18-month window dominated by weak ESG fund performance.