Collaboration between data providers and users critical as expectations, methodologies and regulation of ESG scores and ratings evolve.
ESG data providers face serial challenges in their efforts to supply asset managers with the information they need to inform sustainable investment decisions and meet asset owner expectations.
In particular, they need to address issues around data quality, methodology transparency and the comparability of ESG scores across providers. These are evolving challenges because of the dynamics around the core task of providing quantitative input into the investment process.
To provide scores, ratings and other ESG-related data and analytics services, data providers typically sift through publicly available ESG-related information provided by corporates. From there, they analyse that information according to their individual methodologies and the specific issues and metrics they may be focused on – for example, carbon emissions – and generate outputs that sufficiently reflect input sources.
The asset manager then applies that score when assessing the corporate’s transition from carbon-intensive business models compared to its peers, albeit with considerable levels of variation between managers, reflecting the sophistication of internal models and plethora of inputs.
Providers also offer more granular scores, accounting for a number of issues within one statistic or rating, for example, a company’s overall ‘S’ score or ‘E’ score. The more issues crammed into one statistic, the more difficult it can be to understand how the underlying data has been weighted.
ESG ratings “vary strongly depending on the provider chosen”, according to a 2020 report by the Organisation for Economic Co-operation and Development (OECD).
This can be due to different frameworks, key indicators, methodologies and qualitative judgement, the report noted.
The process of translating information into a scoring system is challenging.
Data providers are having to condense a huge amount of information into a digestible statistic, which requires a lot of time and resources. The rapidly evolving landscape of end-user priorities makes it harder for providers to keep up, resulting in a data lag – meaning the score can become quickly outdated.
Simon MacMahon, Head of ESG and Corporate Governance Research at Sustainalytics, is increasingly seeing demand for scores relating to European corporate performance against the EU’s Taxonomy Regulation. Providers now have to work out how they translate alignment from corporates into a score.
“Impact-focused and activity-based data is a new area of data we’re seeing a lot of demand for,” he adds.
Pressure on the asset manager
Asset managers are having to commit a lot of time and money to validating ESG scores and applying them effectively within their investment process, says Michael Lewis, Head of ESG Thematic Research at DWS Group.
“To secure all the information we need, asset managers are having to buy more data to build a complete picture, which is expensive. Big asset managers have deep pockets and can buy what they need, but middle to small managers often find themselves with a narrower data set––it’s not a level playing field,” he tells ESG Investor.
This is particularly challenging when faced with complex ESG issues, like biodiversity, which refers to multiple nature-related issues, including water scarcity, deforestation, marine ecosystems and more.
With so many big impact issues falling under the biodiversity umbrella, there’s a lack of consensus as to which issues asset managers and asset owners want to see corporates prioritise in their public disclosures. This has a direct knock-on effect on biodiversity scores and ratings.
Until specific biodiversity issues and metrics are identified and agreed upon, corporate disclosures on biodiversity will remain vague and data providers’ scores will reflect that.
However, with the Task Force on Nature-related Financial Disclosures (TNFD) expected to be fully launched by 2023, Lewis is hopeful that specific “biodiversity hotspots” will be prioritised, enabling providers to streamline their resultant analyses and thus produce more usable scores.
Combining datasets, moving in-house
Asset managers typically rely on multiple data providers rather than using ESG scores in isolation. They must be willing to work with the underlying data in order to contextualise the scores they’ve purchased and enable comparability, Lewis says.
“Much of the ESG scores lack robustness and comparability [with other providers],” contends Alexandra Mihailescu Cichon, Executive Vice President of Sales and Marketing at RepRisk.
Asset managers are developing internal capabilities to understand ESG-related risks, impacts and opportunities independently, while contextualising the scores and ratings they buy from external vendors.
A recent report by Insight Investment Management determined that “no risk assessment can be 100% certain”. As a result, they don’t rely on a single data provider for their ESG-related information.
Invesco, too, leverages multiple internal resources to better supplement its ESG-related corporate research.
“Due diligence monitoring is done to ensure data providers are providing on time deliverables such as ESG data, research and recommendations. Invesco is constantly evaluating vendors to ensure our investment teams and clients are provided with the most current information,” the firm noted in its ‘Annual Stewardship 2020’ report.
To help improve the quality of data needed by asset managers, and streamline the relevant scores issued by data providers, the Future of Sustainable Data Alliance (FoSDA) launched its initial recommendations last year to help foster best data practices.
Initial areas of improvement include moving away from singular datasets by focusing on combining datasets from providers and unlocking the potential of geospatial data to help identify and analyse far-reaching impacts.
Evolving public disclosures
Data providers are reliant on corporate public disclosures when generating their scores and ratings.
The volume and granularity of ESG-related disclosures from companies has improved over the last decade, which, in turn, enables data providers to deepen their level of analysis, according to MacMahon.
“Within [Sustainalytics’] main company universe of around 4,500 companies, about 80% of them now produce some kind of sustainability report,” he says.
“Corporate reporting of ESG raw data has constantly increased and improved over the past years, [allowing] us to conduct robust analysis of ESG risks and opportunities to provide meaningful and material ESG scores and signals,” agrees Kristina Rüter, Managing Director and Global Head of Methodology for ISS ESG, the responsible investment arm of the Institutional Shareholder Services (ISS).
However, she notes that there are “some quality as well as consistency issues with corporate disclosures”, which can vary substantially according to geography or industry.
These inconsistencies can lead to gaps in a provider’s understanding of the company, which means an end-score may not be wholly reflective of the corporate’s overall performance.
“For more mature issues, such as bribery, corruption and discrimination, we see more consistent reporting overall. Disclosures about more emergent issues, such as climate change and carbon emissions, are less satisfactory,” MacMahon explains.
The introduction of the updated Non-Financial Reporting Directive (NFRD) later this month will require large European corporates to include a non-financial statement which outlines their exposure to risks, impacts and opportunities across a number of ESG-related issues. This will streamline the information corporates are disclosing and, by extension, improve the quality of scores issued by providers.
The comparability and quality of corporate reporting is also being improved by the efforts of standards-setters and regulators to standardise ESG-related reporting.
Navigating bias
Relying on public disclosures also leaves providers open to corporate bias, which could impact the accuracy of the ESG scores produced, experts warn.
A sustainability report may be only an indicator of minimum or best practices. It doesn’t necessarily allow for providers to access the underbelly of a corporate’s ESG performance, but rather it’s a marketable tool and corporates want to highlight everything they are doing right, not the issues they are still contending with.
For example, in recent weeks global behemoth Amazon caused uproar following its denial of mistreating workers. This is despite the fact leaked reports highlighted that delivery drivers were reprimanded in May 2020 for urinating in bottles as they feared being penalised for taking breaks.
This wasn’t information Amazon publicly disclosed, yet it’s a vital piece of information for providers assessing Amazon’s social score.
Not all providers rely on the information provided by the corporates themselves, with RepRisk taking a risk-profiling approach that looks at corporates from the outside-in.
Capitalising on machine-learning technology, the data science firm searches for public information on a corporate specifically provided by stakeholders, NGOs, government agencies, media and regulators across 20 different languages. Information is then analysed and scored, listed publicly for clients on a daily basis so that the data doesn’t age and remains useful to asset managers.
“We’ve always focused on ESG through the risk lens,” Mihailescu Cichon says. “By looking at companies from an outside perspective, at what’s really happening on the ground, we’re avoiding treating our assessments as a tick box exercise.”
Regulation paves way for transparency
Many have called for regulatory action to be taken, in order to subject data providers to the same level of scrutiny experienced by asset managers and corporates.
“Regulation could provide us with data sets that are comparable, audited by third parties and free for all,” Lewis says.
“[ESG data] is a burden for investors right now because governments and accountants simply aren’t doing what they should be doing,” he adds. “We have no verification on the quality of the ESG data we receive.”
Regulation could ensure data providers are required to be fully transparent about their methodologies. This would help asset managers and asset owners understand exactly how public information has been translated into digestible data.
“Establishing minimum standards for disclosure, transparency and quality would introduce a common baseline, uncover greenwashing and low-quality approaches and enable investors to choose the provider with an approach that best matches their respective ESG investment strategies and targets,” Rüter says.
Regulators must walk a tightrope that improves the value of data providers, while encouraging the innovation needed for providers to contribute to solutions. In the meantime, greater collaboration and transparency between users and providers may shed greater light on ESG risks to investors.
