Why sustainable business needs better ESG ratings
by Beth Stackpole.
ESG data is noisy — and may not help firms protect the planet. Here’s what to keep in mind as you measure and invest. ESG ratings are measured differently across agencies, and can therefore be an inconsistent measure of performance. Some experts are increasingly skeptical that ESG scores have any direct correlation to more sustainable business.
In a recent Q&A, Lenora Suki, product manager for sustainable finance solutions at Bloomberg LP, summed up the situation: “Right now, information about sustainability is often nonfinancial and so diffuse and diverse that it’s difficult for decision-makers to digest and synthesize into strategic and operational decisions.”
This isn’t to suggest that it’s time to write off ESG. All signs point to consumers wanting to purchase products from companies that align with their values, whether that involves protecting the environment, empowering women, or addressing societal inequities.
And investors are ever more enthusiastic about ESG investing.
According to the Global Sustainable Investment Alliance, global sustainable investment reached $35.3 trillion at the start of 2020, a 15% increase over the past two years. In addition, the same report found that sustainable investment assets under management comprised more than a third (35.9%) of total assets under management in 2020, up from 33.4% in 2018.
A majority of business leaders are fully onboard with sustainability measures, while also aware their efforts may not have significant impact. In 2016, a United Nations/Accenture joint survey found that more than three-quarters (78%) of CEO respondents believed corporate efforts would contribute to Global Goals, adopted by the United Nations as a universal call to action to end poverty and protect the planet.
Only a fifth of CEOs responding to a 2019 United Nations/Accenture survey felt businesses are actually making a difference in the worldwide sustainability agenda.
In the most recent survey (2019), only a fifth of responding CEOs felt that businesses are actually making a difference in the worldwide sustainability agenda. Nevertheless, almost half (48%) remain committed to implementing sustainability as part of their overall operations.
The disconnect between accelerating ESG activity and confidence in the results should serve as a wake-up call for companies and investors alike, according to Kenneth Pucker, SM ’90, the former COO at Timberland.
Challenges include discrepancies in ESG measurements, ongoing data quality problems, and the pervasive greenwashing that can happen as companies try to spin their data in a positive light, said Pucker, who in June published “Overselling Sustainability Reporting” in Harvard Business Review.
“ESG ratings firms take self-reported data from companies on their corporate social responsibility [CSR] activities, add their own information and weightings, and mix it in a caldron to come out with a rating for a company,” said Pucker, now a senior lecturer at the Fletcher School at Tufts University.
“A problem is garbage in, garbage out,” Pucker said. “The reporting is not complete, results are mostly unaudited, and they are not comparable, so ESG ratings often use bad data that’s unaudited, extrapolated, and interpolated.”
Why ESG ratings diverge — and what to do about it
The MIT Sloan Sustainability Initiative has taken up the cause with the launch of the Aggregate Confusion Project, aimed at improving the quality of ESG measurements.
ESG ratings are used by individual investors and fund managers to guide investment strategies by understanding potential sustainability risks to business performance. Companies also leverage ESG ratings as an internal benchmarking tool to help improve sustainability performance and for potential public relations opportunities.
The problem, as defined by the MIT project, is that ESG data is noisy. That leads to a divergence in ratings from the independent agencies that evaluate and assign ESG ratings to firms, according to Florian Berg, a research fellow working on the challenge. In turn, that leads to confusion and a much lower probability that ESG ratings have a direct correlation to financial performance, undermining their utility as an investment tool, Berg said.
Berg, along with research affiliate Julian Koelbel and MIT Sloan professor Roberto Rigobon, has published three papers on ESG ratings divergence and its impact. Using a common taxonomy and various mathematical modeling techniques, the MIT team was able to fit the different rating agency approaches into a consistent framework to better understand the ratings differences.
In the initial paper, the researchers identified the following takeaways:
There’s ambiguity around ESG ratings. The research found the correlation among six prominent ratings agencies (KLD/MSCI Stats, Sustainalytics, Vigeo Eiris/Moody’s, RobecoSAM/S&P Global, Asset4/Refinitiv, and MSCI) was on average 0.61. In comparison, mainstream credit ratings from Moody’s and Standard & Poor’s are correlated at 0.99.
The finding suggests that ESG ratings do not properly reflect ESG performance, making it difficult for decision-makers to identify outperformers and laggards.
Divergence in ratings also hampers the motivation of companies to improve their ESG performance, the paper states, because there are mixed signals from ratings agencies about what to focus on and what is valued in the industry.
There are three factors driving ratings divergence:
- Scope divergence occurs when ratings are based on different attributes — for example, one rating service includes carbon emissions or labor practices while another does not — which leads to ratings inconsistencies.
- Measurement divergence happens when agencies measure the same attributes, but do so using different raw data, which results in different assessments.
- The last factor relates to weights divergence, which emerges when ESG ratings agencies take different views on the relative importance of attributes.
They also identify something they call the “rater effect,” namely, a firm receiving a high score in one category is more likely to receive high scores in other categories, particularly from that same rater.
As part of their ongoing research, the team is trying to understand the effects of ESG-driven investment flows on stock price and company behavior. They are also developing smarter ways to aggregate ESG factors into composite indices for use by portfolio managers and passively managed funds.
ESG ratings and stock performance
A subsequent paper depicts in detail a noise-correction procedure designed to tackle the bias caused by noisy ESG data and ratings divergence. Using a variety of modeling techniques, the research team was able to demonstrate ESG ratings actually have a much higher impact on stock returns than previously anticipated.
Without the correction procedure, standard regression estimates of ESG ratings’ impact on stock returns were biased downward by about 60%, Berg said.
“We found the link between financial performance and ESG performance to be stronger because it’s not overshadowed, hidden by noise,” Berg explained.
The practical takeaway from the research: “Relying on the scores of several complementary ratings yields better results,” Berg said.
Addressing ESG shortcomings
Next steps for the Aggregate Confusion project are to work with industry partners to gather data, solicit feedback, and create a set of best practices for use by ESG ratings providers as well as the companies reporting CSR data for disclosure purposes.
Improving the measurement of raw data and formalizing best practices is all well and good, but Pucker is convinced it will take global standards and formal audits to demonstrably bolster the utility of ESG ratings.
Pucker explained the environmental impacts of sustainability reporting and ESG investing have been oversold, hampered by lack of oversight into what goes into sustainability reports as well as huge gaps in data tracked to generate a complete picture of carbon footprint.
As an example, Pucker cited Scope 3 emissions, which comprise all the upstream and downstream emissions generated by suppliers and customers. Figures from CDP, a global aggregator of corporate carbon emissions data, show that only about half of companies disclosing CSR data report on Scope 3 emissions even though that constitutes the bulk of their greenhouse gas impact.
More pointedly, Pucker contends the problem is bigger than noisy data or ratings discrepancies. At its core, ESG investing may deliver better financial returns for investors, but it’s not designed to deliver the intended impact for solving environmental problems. “Asset managers have incentives to drive equity returns, not solve water problems in Egypt or deforestation in Malaysia,” Pucker said. “It’s not what they are paid to do.”
Pucker believes impact investing is better positioned to make more of a material difference. Estimated by the Global Impact Investing Network to be a $715 billion market, impact investing targets companies or institutions that have specific intentions to build products and services that generate social or environmental impact — for example, one that is actually developing technology that solves water problems.
This type of investment in companies pursuing climate tech to solve transportation, agricultural, energy, and land use challenges is more likely to deliver change, Pucker said.
More work to be done
Regardless of that debate, Pucker said the industry has a lot of work to do to get better at CSR reporting and make ESG ratings a more effective tool.
Making reporting mandatory, creating comprehensive standards on what to report, increasing transparency in what is reported, and enacting some form of industry or regulatory governance is one piece of the equation. Efforts like the Aggregate Confusion Project for improving data quality and ESG measurements is another critical step.
“All those things will help the correlation of ratings converge among different raters, but to me, that’s not a leveraged solution for impact,” Pucker said. “Ultimately, addressing environmental and social externalities needs effective regulatory policy.”