University of Chicago: There’s no Data to Determine if ESG Is Working

“People investing in ESG funds want their money to have an impact. They just don’t have a way to ask for that information,” says Jason Saul of the Center for Impact Sciences.

Illustration by II

Illustration by II

ESG investors want to see the results of putting their money into companies that claim they are helping to curb fossil fuel emissions or promote clean energy — but the data that is available rarely allows for that.

In a white paper titled “ESG: Right Thesis, Wrong Data,” Jason Saul, executive director of the Center for Impact Sciences at the University of Chicago’s Harris School of Public Policy, and Phyllis Kurlander Costanza, former head of social impact at UBS, argued that instead of using data that is based on corporate compliance with environmental, social, and governance mandates, investors need to develop their strategies around data that measures the impact of portfolio company initiatives.

The authors called existing ESG reporting frameworks “ESG 1.0,” which only focuses on companies’ adherence to codes of conduct and ethical guidelines around issues such as human rights violations and board diversity.

Typically, these questions are answered in a yes-or-no format: Does the company use child labor in its factories or have robust anti-corruption policies? Ratings agencies use the answers to derive their ESG scores. But as the ESG has matured, these scores are far too simple.

ESG 1.0 data originated in the 1990s when investors wanted to “avoid bad companies,” by negatively screening for “sin stocks,” shares of companies that produced goods like alcohol and firearms.

These designations of so-called good and bad companies were then turned into data to help investors identify other companies that fell into these categories. Saul told Institutional Investor that this simple process no longer satisfies impact-oriented investors.

“We’re trying to stretch an ESG 1.0 data regime to address the needs of ESG 2.0 investors,” Saul said. “That’s why everybody is so frustrated.”

But these guidelines don’t measure the impact of the “good” companies. This is where Saul and Costanza’s aspirational “ESG 2.0” comes in. ESG 1.0 was based on a concept called “materiality” — or “what really matters” to investors and auditors. But investors’ and auditors’ ideas about what matters has evolved over the years. While some investors see materiality as a question of compliance and risk, others are more concerned with the impact of their ESG investments and their capacity for positive value creation.

Saul and Costanza advocate for the industry to move from a standard of materiality to a standard of “intrinsicality” to determine the value of ESG data. In other words, ESG data should focus on the benefits that are intrinsic to value creation at the company and also provide a social impact.

For example, while ESG 1.0 data would measure the existence of a diversity initiative at a company, ESG 2.0 data would collect employee diversity statistics over time and aggregate changes in wealth distribution among underrepresented populations. The data would also measure how the initiative helped the company make more money.

In contrast, materiality — the basis for ESG 1.0 — considers “extrinsic” ESG value propositions. An extrinsic value proposition is an ESG initiative that is related to wider social good but isn’t related to the company’s bottom line or central functions. For example, a grocery store chain has an option for customers to donate to a national charity. While the initiative may have reputational benefits, it doesn’t improve the company’s margins.

According to the paper, companies that do well at disclosing their extrinsic ESG risks don’t perform better financially than companies that don’t.

While investors are increasingly more concerned about the intrinsic value of their ESG initiatives, they are still working with largely ESG 1.0 data, information that only measures extrinsic value.

To remedy this dissonance, Saul and Costanza suggested data agencies establish what they call an intrinsicality map. Today, agencies use an ESG materiality map, which evaluates the “relative importance of ESG data to companies,” the paper said. A materiality map measures financial and stakeholder materiality of ESG risks like climate risk, biodiversity, and waste transparency and focuses namely on the extrinsic value of these risks.

An intrinsic value map would identify and quantify the impacts of ESG that also contribute to value creation.

Saul and Costanza also suggested that the industry creates standards for impact, not just standards for compliance. ESG 1.0 largely uses administrative data to create ESG scores. Instead, the paper suggests the adoption of a “standardized taxonomy of social impacts” whereby companies report how they contributed to social outcomes like racial equity, education, housing, and financial inclusion.

At a higher level, ESG 2.0 can only happen if investors know how to voice their demands. “Most of the people investing in ESG funds — whether they’re high-net-worth, family offices, or pensions — actually want their money to have an impact. They just don’t have a way to ask for that information,” he said.