Does Every Little Bit of ESG Integration Help?

(Photo by iStock/AntiMartina)

Since it was launched in 2006, thousands of corporations and organizations have signed the “Principles for Responsible Investment” (PRI), committing to “incorporate ESG issues into investment analysis and decision-making processes,” more commonly known as ESG integration. In that time, ESG integration has been enshrined in thousands of pension fund and asset manager ESG policies, while regulations such as the EU Sustainable Finance Disclosure Regulation (SFDR) now require the practice of financial market participants. This has come at significant expense: ESG integration requires resources (most PRI signatories have dedicated ESG teams with as many as 80 specialists), data (most signatories work with one or more external data providers with the global annual market estimated at $1-2 billion), and training & education (many PRI signatories have sent their staff to ESG conferences or have worked with PRI or other organizations to train their staff on ESG concepts). The effort and expense associated with ESG integration is impossible to calculate, but my back-of-the-envelope guestimate would be in the range of $50-100 billion.

One might think the investment industry would only go to all this trouble and expense if there is clarity and consensus on the benefits that adopting such a new practice might bring. One would be wrong.

Why ESG integration?

The PRI preamble distills the purpose of ESG integration down to financial performance and societal impact: “We believe ESG issues can affect (financial) performance” and “we recognize that (ESG integration) may better align investors with societal objectives” (my italics). But it’s clear from words like “believe,” “can,” “may,” and “align” that the effects of ESG integration were untested when the first signatories made their commitments, especially for the second purpose: Not only does “may” keep options open, bets are hedged by the verb “align,” referencing the distinction between “impact-aligned” and “impact-generating” investments, even if that distinction was probably little made in 2006. The distinction points to the need for “additionality,” or causality, in order to have impact: The idea is that the investment should enable the impactful activity.

But if there is only “alignment,” does that mean there is no “additionality”? And can we really still call it “impact”?

While the PRI and the industry have adopted the principle of ESG integration as a doctrine, academics have treated it as a hypothesis, looking for data to test it and determine if it holds true, particularly the link between ESG integration and financial performance. And there appears to be consensus: ESG integration, in the aggregate, does not boost financial performance. (Though it doesn’t hurt financial performance either). This academic consensus doesn’t seem to have filtered through to practitioners, however. Or, if it has, they have taken the view that while it doesn’t help, it doesn’t hurt either, so we should continue the practice because of the societal benefits.

All of this is why the recent report “Does ESG integration impact the real economy? A theory of change and review of current evidence,” co-authored by researchers Julian Koelbel, Florian Heeb, and Anne Kellers at the Center for Sustainable Finance & Private Wealth (CSP) from the University of Zurich deserves wide attention and discussion. Commissioned by the Swiss government, the report analyzes the question that has been hanging over the market since 2006: Does ESG integration in fact have impact? Does it have societal benefits?

Four Conditions for Impact

As the report breaks down the question, there are four “assumptions” that need to be true for ESG integration to have impact, and the report indicates if and when these assumptions can hold. The report’s assessments can be summarized as follows:

1. ESG ratings reflect company impact. This assumption holds when ESG ratings focus on impact materiality rather than financial materiality;
2. Portfolio holdings deviate from the market benchmark by tilting towards ESG leaders and away from ESG laggards. This assumption holds for most dedicated ESG funds, which is still a small part of the market, but not for most large institutional investors (the majority of the market);
3. The market share of ESG investors is large enough to create an ESG premium. There is some evidence that an ESG premium exists, but it remains uncertain whether it is economically meaningful;
4. Managers consider the ESG premium to be large enough to justify additional investments or adopt improved practices. It seems unlikely that company managers will undertake impactful action as a result of ESG integration beyond the “low-hanging fruit.”

Based on these assumptions (and the evidence that they hold in practice), the authors conclude that the answer to the question “Does ESG integration have an impact on the real economy?” is: “Maybe a little bit.”

It should be noted that the report does not attempt to answer the question empirically, i.e. by analyzing a dataset or actual observations, but rather qualitatively, i.e. by reasoning through the logic based on our understanding of ESG issues, investing, finance, and markets. However, the authors have also reviewed the available research on this question, even if that research is quite limited.¹

I agree with the methodology of the report, outlining a theory of change and four assumptions to test if the theory holds, and I agree with the authors’ assessments of how likely it is the individual assumptions will hold. However, while I also agree with the high-level conclusion of the report—that ESG integration “maybe has a little bit of impact”—the word “maybe” reflects the unlikelihood that all four conditions be satisfied simultaneously.

In my experience, this is not only unlikely but extremely unlikely; I would perhaps gloss that “maybe” as “there is a vanishingly small probability.” After all, it is clear from the report (and from my own experience) that the four assumptions do not hold simultaneously today. And it also seems very unlikely that they will hold simultaneously at any point in the future. For example, what are the chances that we will be able to effectively capture “SDG impact materiality” in ESG ratings and that a significant part of the investment market will tilt towards ESG and impact leaders (perhaps thereby disregarding other relevant investment decision-making factors), and also that all this would prompt company managers to invest at some scale in SDG-like impactful programs beyond low-hanging-fruit? That likelihood is too small for ESG integration to be considered a useful tool to address societal problems.

Put simply: If all assumptions hold, we would see a “little bit of impact.” But this is a very big if. And while the report’s conclusion may seem pessimistic for ESG practitioners, even disheartening, to me it seems too optimistic. I would have articulated it differently: “ESG integration could theoretically have a little bit of impact in the right conditions but the likelihood that we will have all those conditions is exceedingly small.”

What kind of impact?

So much for the “maybe” in the conclusion. On to the phrase “little bit of impact”: A weakness of the report is that it does not provide a definition of “impact” that fully grapples with whether outcomes of corporate activity contribute in a positive way to sustainable purposes such as those identified in the SDGs. At times this makes for confusing reading; the term is sometimes used as a verb, as in the title and the main question of the report (“Does ESG integration impact the real economy?”). But in other places they use the term as a noun—as when saying that “ESG integration has impact”—such that the term is likely to be interpreted as “has environmental or social benefits.” In the world of ESG and impact, the UN Sustainable Development Goals (SDGs) have become a popular framework to determine if an investment is impactful or not, such that when the authors discuss whether or not ESG integration has “impact,” it is likely to be interpreted in that way, as something that helps us reach the SDGs.

This is more than semantics. The report’s conclusion, that ESG integration might have “a little bit of impact,” could mean one of two very different things: It could mean that ESG integration—if all conditions hold—can contribute to solving societal problems in line with reaching the SDGs. But it could also mean that while it can have effects in the real world, these would not be in ways that contribute to reaching the SDGs.

Perhaps more importantly, the authors do not really delve very deeply into the question of how much companies and investors can in fact contribute in terms of (SDG-like) impact in the first place. Clearly, in the context of the important question this report addresses, this is very much worth thinking about! But this is exceedingly difficult and (as far as I know) there are no good attempts at this that we can rely on.

The best way to think about the problem is to imagine a Venn diagram with two circles. For the first, picture the SDGs in their multi-colored variety—no poverty, zero hunger, clean water and sanitation, climate action, etc., etc.—and imagine all of the activities, programs, work, and effort that must go into meeting these goals: building and running schools, setting up sanitation programs, putting in place programs that reduce maternal mortality, getting food to hundreds of millions who live in poverty and can’t pay for it. If necessary, go to outlining goals and sub-goals, and scroll through the detail.

Bearing in mind that the SDGs were originally drafted as a roadmap for government policy and actions, on a not-for-profit basis, imagine the second circle filled with the products, services, activities and programs that companies can generate. And remember, companies are profit-seeking. This is not an indictment, simply an observation that for them to be or become involved in any SDG-related activity at scale, in the ordinary course of business, there needs to be a commercial or profit motive somewhere, at some point. And only when activities become commercially attractive to companies, investors can invest in them.

The two circles would certainly overlap. But by how much? Whenever I do this exercise, I conclude that the overlap is marginal. Going through the SDG subgoals, I simply identify too many where corporate activity is likely to play no role at all, or at most an insignificant one. This means that the “SDG-like” impact that companies can have is likely also relatively small. While I don’t want to suggest the CSP authors followed similar thinking, we do end up with the same conclusion: ESG integration can (maybe) have “a little bit” of impact.

Implications and Recommendations

The report offers some recommendations for investors that, while useful, are mostly focused on the idea of making all four assumptions hold. Yet as the authors note, doing so would not only be “demanding,” but the impact—in the unlikely scenario that all assumptions hold—would be insignificant (“a little bit”).

To sum up, since we already knew that ESG integration does not affect financial performance, the report suggests that ESG integration does not have impact today, and will not likely have it in the future either. My recommendations to the investment industry would therefore be a little bit more blunt:

1. Revisit the PRI principles. Determine what the main objectives of the PRI were, and evaluate if the principles and the work conducted by signatories since 2006 have gotten us closer to meeting those objectives. If not, redraft the principles.
2. Revisit ESG and responsible investment policies and beliefs. If you truly aim to contribute to solving societal problems, do not count on ESG integration but look for tools that have an evidence-based track record of delivering impact.
3. Re-educate your ESG teams. Focus on the type of investing that can have impact: active ownership, investing in new technologies or emerging markets infrastructure – anything that contributes to the SDGs but isn’t receiving funding today.
4. Work with governments. Support them in developing effective policy to address societal issues, and invest with them, for example through blended finance structures, also in light of point three.
5. Revisit your range of ESG Funds. Carefully review the prospectuses and slide decks on mentions of “impact” and “SDGs.” If these claims rely mostly on ESG integration in your investment processes, they will be challenged sooner or later.

The report similarly has implications for regulators who “want to create conditions that enhance the impact of ESG integration.” Because the strength of ESG integration lies in its scale, the authors assert that even uncertain or small impacts may add up to a meaningful effect. However, with the likelihood of creating these conditions ranging from “maybe” to “rather unlikely” and with—if we have those conditions—“uncertain and small impacts” as outcomes, my perhaps somewhat skeptical conclusion is that policy makers should not put many of their eggs in this basket.

Instead, my recommendations for governments and policy makers would be more hard-nosed:

1. Don’t count on ESG integration as a tool in your toolbox to address societal challenges. Satisfying the four conditions is exceedingly difficult if not impossible, and even if you succeed the impact will be insignificant (and may not be the type of impact you need or expect).
2. Stick with the basics. If you want to marshal private capital or corporate efforts in order to meet the SDGs or the aims of the Paris Agreement, use tried-and-tested tools such as blended finance, public-private partnerships, guarantees, and subsidies. Or simply invest in new technologies or programs yourself.
3. Revisit ESG and sustainable finance regulations. Many say they intend to “shift capital” towards SDGs and to protect consumers, and then go on to require ESG integration by financial firms (amongst other things). We now know that continuing to require financial firms to comply with many of these regulations doesn’t make sense: It doesn’t achieve those goals but does create massive reporting burdens with questionable benefits and is potentially a distraction from more meaningful work.

Is there also good news here?

Yes. While the conclusion from the report may be disheartening to many, and the recommendations listed above may come across as cynical, we have to recognize that we’re in the middle of a process of innovation. Most innovation is a gradual process, and trial and error is vital: As Matt Ridley explains in How Innovation Works, successful innovations are usually preceded by a “pre-history characterized by failure.” “Innovation is … a collective, incremental, and messy network phenomenon.” There are legions of people working to direct finance and investments to where they can be most impactful. So the good news from this report is that it represents a very small but crucial step in our collective learning for this incremental and messy undertaking of contributing to the SDGs through company and investor “impact.”

In other words, the good news is that this report should encourage those legions of people to (re)direct their efforts so that they can be more consequential, and will allow them to be more effective every day.

1 The research that was reviewed is referenced in the report. Pedro Matos, in another survey of relevant research on ESG (ESG and Responsible Institutional Investing Around the World) said “I am aware of almost no academic research in top-ranked finance journals on how ESG investing is contributing to the achievement of the SDGs.” The CSP authors might have referenced “When Can Impact Investing Create Real Impact?” (2013) in their references; while not peer-reviewed, in line with the CSP conclusion it opines that for equities bought in the secondary market, “it is virtually impossible for a socially motivated investor to increase the beneficial outputs of a publicly traded corporation by purchasing its stock.” This paper is also referenced in the recent article “ESG Investing Isn’t Designed to Save the Planet,” by Ken Pucker and Andrew King that was published after the CSP Paper and that also agrees with the conclusion of the CSP report.

Read more stories by Harald Walkate.