52023SC0204

ESG/sustainable investing or investing which takes ESG factors into account when making investment decisions, is becoming an important part of mainstream finance, with some estimates putting it at USD 40 trillion globally. 4 Notably, investment funds with sustainable characteristics or objectives have largely increased in number, size and the type of capital they attract. 5 In this context, an ESG investment ecosystem has developed, comprising investors, investees, data and other service providers who cater to the needs of these and other types of users. The offerings of such service providers can be broadly grouped into ESG data (raw or processed), ESG ratings 6 , other ESG assessment tools (such as screening tools, rankings, controversy alerts), and advisory and consulting services (such as regulatory reporting tools for the Sustainable Finance Disclosure Regulation (SFDR) 7 and EU Taxonomy Regulation 8 ).

The EU has put in place the building blocks for a sustainable finance framework 9 and stakeholders are now pointing out the remaining market inefficiencies and regulatory gaps, which have the potential to hinder the development of the EU sustainable finance market. Much stakeholder feedback in recent years has focused on ESG rating and data providers. ESG rating providers and data providers use information published by corporates; as such, the availability, accuracy and consistency of this data are key to the quality of the downstream products that depend on it. The advent of the Corporate Sustainability Reporting Directive (CSRD) 10 in the EU and the introduction of mandatory sustainability reporting standards for large companies are expected to improve the availability, accuracy but also consistency of ESG corporate data.

ESG ratings will however build on and complement that and continue to play an important and enabling role for the proper functioning of the EU sustainable finance market by providing important information for investment strategies, risk management and internal analysis by investors and financial institutions. They are also used by companies who seek to better understand sustainability risks and opportunities linked to their activities or those of their partners, and how they compare to their peers on these issues. R

ESMA 11 has estimated the number of ESG rating providers (offering subscriber pay or issuer pay ratings) operating in the EU to be 59, composed of a few large (mainly non-EU) providers and a large number of smaller (mainly EU) entities. According to ESMA’s assessment, the estimated market size of ESG rating providers based in the EU is EUR 300 million. Given that the majority of large ESG rating providers are based outside the EU (while offering services in the EU), this number likely largely underestimates the total size of the market. Annex 6 provides further detailed information on the ESG ratings ecosystem, including market players and users.

A number of papers have highlighted the importance and expected growth of the market of ESG ratings. It is expected to continue to grow substantially 12 in the coming years 13 . Growth and increase in demand for ESG ratings is driven by the changing nature of risks to companies, by growing investor awareness of the financial implications of those risks and by the growth in investment products that explicitly seek to meet certain sustainability standards or achieve certain sustainability objectives. 14 It is also driven by the change in strategies applied by investors, from negative/exclusionary screening to ESG integration, which is now predominant (USD 25.2 trillion in assets under management employing an ESG integration approach, while negative/exclusionary screening amount for USD 15.9 trillion in 2020) 15 . It is also linked to the sustainable finance legislation put forward by the EU, since 2018. A survey conducted by Ninety One 16 found that 88% of investment professionals use third-party ESG ratings as a part of their investment process, with 92% expecting to do so in the future. As a result, ESG ratings have an important effect on the allocation of capital by investors to companies. ESG ratings are one of the determining factors of whether company shares will be included in ESG-themed mutual or exchange-traded funds. With the growth of interest of investors in ESG ratings there is also an expectation that companies with a high ESG rating will be able to lower their cost of capital whereas the cost of capital of firms without ESG / with a bad ESG rating is expected to increase, which in turn is supposed to steer more capital into sustainable activities. Investors currently cannot learn much from the ESG ratings and cannot clearly separate ESG high performers from low performers, which hampers such process of capital allocation.

The issues facing ESG ratings are highly relevant for the Union’s long-term objectives and in particular the green transition, a key element of the Commission Strategic Foresight 17 , that requires adjusting the EU’s economic policies towards greater sustainability and more circularity. Investment into sustainable projects by both the private and the public sector will be key. It is estimated, at lower ends, that the twin transitions might need around EUR 650 billion euro annually. Targeted reforms and investments must address vulnerabilities at national and EU levels in order to increase the EU's resilience and facilitate the twin transitions (digitalisation and green transition).

At the international level, the International Organization of Securities Commissions (IOSCO) published a report containing a set of recommendations in November 2021 calling for the establishment of oversight on ESG rating providers, their operations and products 18 . This impact assessment carefully considers those recommendations. A number of jurisdictions have also started looking closely at this topic and are considering the next steps and whether and how to intervene (for example the United Kingdom, Japan and India) 19 . A more detailed overview of IOSCO recommendations and developments in other jurisdictions is included in Annex 7.

Stakeholders and market participants call for improvements in the ESG ratings market, highlighting a lack of transparency, reliability and comparability of ESG ratings, and raising the risk of greenwashing for all users of those ratings, including retail investors 20 . Member States 21 and MEPs have also called on the Commission to act on ESG ratings. Market participants have indicated 22 that a better understanding of what ESG ratings assess and increased reliability of ESG ratings would enhance the trust in and credibility of this fast-growing market, thereby facilitating progress towards the objectives of the European Green Deal.

Stakeholders have also pointed to the relevance of understanding how ESG factors influence credit ratings. Both the 2018 Sustainable Finance Action Plan and the 2021 Strategy on Financing the Transition to a Sustainable Economy acknowledged the importance of incorporation of ESG factors in creditworthiness assessments and asked ESMA for advice.

It should be kept in mind that credit ratings and ESG ratings play different roles and serve different purposes. Credit ratings assess the risk of default of a company and should capture all the risks that may affect creditworthiness - including sustainability risks where relevant. Credit ratings are used by investors to calibrate portfolio risk profiles (including for investment products) as well as for the calculation of prudential requirements under the EU’s Capital Requirements Regulation (CRR) and Solvency II, and play an important role in ensuring financial stability. Credit ratings are also used by the European Central Bank as one element in its open market operations.

ESG ratings have different purposes and usage. Some assess exposures to risk stemming from E, S and G factors, others impacts of the entity on the outside world, or still others may assess the compliance with international standards. There are currently no regulatory requirements in relation to the use of ESG ratings (as opposed to credit ratings) and they only play a limited, indirect role for financial stability. Due to the different functions of those ratings, the same company may receive a good credit rating and a bad ESG rating, or vice versa.

Credit Rating Agencies have been regulated since the financial crisis of 2008. They need to register with ESMA and are subject to a number of organisational requirements, as well as ongoing supervision by ESMA. The CRA Regulation has also introduced disclosure requirements in relation to methodologies and credit ratings as well as rating outlooks. In addition, in 2019, ESMA issued guidelines on disclosure requirements that include recommendations on how CRAs should present in their press releases which credit ratings have been impacted by ESG factors 23 .

The 2021 renewed sustainable finance strategy 24 acknowledged that credit rating agencies play an important role in the financial system by assessing the credit risk of financial and non-financial issuers. It indicated that stakeholders continue to express concerns around the lack of transparency on how credit rating agencies incorporate sustainability factors in their methodologies and that, subject to further assessment of the effectiveness of the existing measures by ESMA, the Commission would take action to improve transparency and ensure the inclusion of relevant ESG factors in credit ratings and credit outlooks, while ensuring methodological transparency.

Following the adoption on the renewed sustainable financne strategy, ESMA carried out the assessment of the impact of those Guideline 25 s and, in addition, reviewed how CRAs incorporate ESG factors in their methodologies. In addition, the last few years have brought changes and improvements as far as disclosure and incorporation of ESG factors in credit ratings are concerned, as some CRAs are now disclosing more that is required by the Regulation and ESMA Guidelines.

The remaining shortcomings relate to two areas: 1) the rigorousness of the inclusion of ESG factors in CRA’s methodologies, and 2) the clarity of the disclosures made by CRAs (whether and how ESG factors were considered).

Chapter 5.5 and Annex 8 summarises the evidence and analysis carried out by the Commission services and ESMA. It concludes that modification of Annex I of CRA Regulation together with changes to the delegated act on methodologies can address the remaining shortcomings.

2.Problem definition

The central problem this initiative aims to address is that the market suffers from deficiencies and is not functioning properly, with users’ and rated companies’ needs regarding ESG ratings not being met and confidence in ratings undermined.

This problem has a number of different dimensions, mainly the lack of transparency over methodologies of ESG rating providers and clarity on how they operate, in particular how they manage potential conflicts of interests. Consequently, ESG ratings do not serve their purpose and do not sufficiently enable users, investors and rated companies to take informed decisions as regards ESG-related risks, impacts and opportunities. Investors are reluctant to use these or they need to complement or verify the conclusions from ESG rating providers with further in-house analysis. In general only the larger asset managers are in the position to have such in-house department, smaller investors lacking expertise and financial means to do so 26 . The lack of trust in ESG ratings stems from the absence of transparency on methodologies and clarity over the operations of ESG rating providers. A number of scandals have happened in this market, but they are not the main reason for the lack of trust. However, given the growth of this market they are likely to become more common in the future in the absence of regulation and supervision of the market. Rated companies also face difficulties, to understand under what criteria they are assessed. Since decisions of investors and businesses are crucial for the transition to a climate-neutral and more sustainable economy, this ultimately hinders the market’s potential to contribute to the European Green Deal and achievement of UN Sustainable Development Goals.

The degree of information provided to clients and rated entities varies among providers. It is not possible for investors to assess and compare the value of ratings (as a product) without sufficient information on their scope and quality. This is the ‘information asymmetry’ market failure.

Market forces have not catalysed the creation of a standard or framework to be applied by ESG rating providers, at international, European or national level. There have been attempts by the market to establish voluntary, self-regulatory standards / codes of conduct mainly on their operations and transparency of methodologies, but none hast been successful. The standard that seems to have been the most successful was ARISTA 3.0, established in 2004, a voluntary European standard on the quality and transparency of methodologies and processes used by ESG rating agencies. This standard is seemingly non-existent, as ESG rating providers chose not to apply it anymore “due to lack of relevance, market awareness and recognition of the standard” 27 . Another standard that also got some traction at its launch was the Deep Data Delivery Standard 28 , which is meant to be used by asset managers as a quality assurance when contracting with ESG rating providers. A number of ESG rating providers initially joined it (MSCI, Vigeo Eiris, RepRisk, etc.). There is however no recent information as to whether this standard still exists, nor a list of members.

In the absence of requirements related to transparency of methdologies and objectives of ESG ratings, this situation is unlikely to improve and to meet the needs of users.

2.1.What is/are the problems

The deficiencies in the current ESG rating market identified in this section reflect the principal findings of the Commission’s Study on Sustainability-related ratings, data and research (henceforth abbreviated as “the Study”) 29 , targeted consultation 30 , ESMA analysis 31 , the IOSCO Report on ESG Ratings and Data Products Providers of November 2021 32 (‘IOSCO Report’) and other litterature review. Annex 2 provides a summary of all stakeholder consultations and their results.

It is useful to make a distinction between problems for users of ESG ratings and problems perceived by rated entities.

•Investors and companies observe issues with the reliability, accuracy and timeliness of ESG ratings;

•Rated companies are unsure how accurately ESG ratings reflect their actual performance and how to improve it.

2.1.1.Investors and companies observe issues with the reliability, accuracy and timeliness of ESG ratings

This problem analysis focuses on the primary intended users of ESG ratings :

·Investors and benchmark administrators increasingly use ESG ratings as part of their sustainable investment strategies to take into account risks and/or impacts linked to ESG issues. ESG ratings are often used for pre-investment decision-making (e.g., to inform a selection of equities to include in a portfolio or fund), post-investment analytics and attribution (e.g., to rate the ‘sustainability’ of an investment product or fund) or as a basis for investment screening or for company engagement (engagement between an investor as a shareholder of a company, and representatives of the investee company which can influence decisions taken by the company’s management).

·Companies use ESG ratings to look for and take account of operational risks and investment opportunities but also to see how they perform against ESG factors, compared to peers 33 ;

ESG ratings have an increasing impact on the operation of capital markets and on the confidence of investors in financial products. Investors however denounce a lack of ‘quality, effectiveness and impact of corporate ESG ratings’ 34 .

Investors and companies claim that ESG rating providers do not gather and process data and information in an accurate, timely or reliable manner 35 ; that their methodologies are opaque 36 and do not sufficiently take into account company context 37 ; that providers make errors (and are slow to correct them) and that engagement with multiple providers is time-consuming. There have also been companies suing ESG rating agencies for unfounded ratings, which has in turn created concerns about the accuracy of ESG ratings 38 .

ESG rating providers however, generally mention that companies do not publish sufficiently reliable data to enable appropriate comparability and analysis, while at the same time asset managers demand increased coverage, depth but also data quality 39 .

The Study noted that a number of ESG rating providers, including Bloomberg, CDP, FTSE Russell, ISS, MSCI, S&P SAM and Trucost, Workforce Disclosure Initiative, Vigeo Eiris and Sustainalytics 40 , utilise company verification as part of their data accuracy process. Company verification appears to be an effective tool used by ESG rating providers, but investors and companies find it difficult to navigate given the variety and lack of clarity over the processes used. Those providers may flag specific areas they want companies to review, including missing data, conflicting data from different sources or a large change in data between years. Increasingly providers are creating online platforms and individual company accounts as mechanisms for providing updates to companies and for companies to engage or ask questions. Some providers claim to have a continuous feedback system, but others only enable verification on an annual or semi-annual basis. Some providers don’t engage with companies, so that their ratings cannot be influenced by corporates. Therefore, there is a disconnect between what ratings providers say they offer, and the experience companies actually have with providers. The Study identified that only one ESG rating provider, out of 14 providers who responded to a question on data verification in the context of the Study, has detailed publicly available information on the data verification process 41 . Given the overall lack of information available publicly, it is difficult to understand whether ESG rating providers have effective methods in place for quality control and what those are, specifically how they deal with input data. There is no available evidence to support claims of functioning data verification.

Furthermore, information on methodology and frequency of rating updates is also limited and varies depending on the ESG rating provider concerned 42 . The majority of ESG ratings providers typically update ratings annually, but the frequency with which they undertake methodological updates varies from potentially daily to “periodically.” Updates may also be linked to feedback from companies, e.g. ISS QualityScore calls on companies to provide updated data as much as possible via their platform 43 . The Study noted clear differences in the updates to company ratings, which also affects the quality of ratings and whether information on companies is up-to-date or possibly outdated.

2.1.2.Problems from the perspective of rated entities – unsolicited ratings

The main problem from a rated entity perspective, where they are subject of ratings but not users/clients themselves, is that they find it difficult to understand whether and how they are rated, or how to improve their performance and may face risks of using incorrect/unreliable ESG ratings, where they can also be subject to greenwashing accusations.

Rated entities also face a large number of requests at different times from different ESG rating providers (e.g., via questionnaires), which creates administrative burdens and costs.

ESG rating providers have different level of engagement with companies. Although several ESG rating providers engage directly with companies to gather data and information (in addition to the ones publicly available), to facilitate data verification (via online platforms, direct exchanges, holding meetings or teleconferences), and to allow for error correction, grievance resolution, this is not always existent and practices including levels of engagement differ from one ESG rating provider to another. It is also not always clear to what extent the result of the engagement impacts the rating given to a company. Some providers don’t engage at all with rated companies and use only public data in order to remain objective. Few ESG rating providers publish some information on how they engage with issuers 44 . ESG rating providers may also organise frequent or annual exchanges/review process or allow companies to provide feedback using a specific application/portal, also prior to publication of the ESG rating. Some other ESG rating providers may have data verification process based on algorithmic checks, without human intervention 45 .

The large number of exchanges between ESG rating providers and companies, and the time necessary for these exchanges may give further advantage to larger companies that have the means and resources both in terms of personnel and budget to exchange with ESG rating providers and provide data 46 . This is reinforced by research that shows that large-cap companies have significantly higher ESG disclosure scores than mid-cap companies 47 . Research by Environmental Finance states that ‘companies complain about the growing number of questionnaires they are being asked to respond to, and the confusing array of standards they are being asked to meet, including the Global Reporting Initiative (GRI), the Sustainability Accounting Standards Board (SASB), the International Integrated Reporting Council (IIRC) and the Task Force on Climate-Related Financial Disclosures (TCFD)’ 48 . The Independent Research in Responsible Investment Survey (IRRI) of 2019 49 also stressed that completing standardized questionnaires from ESG rating providers is the most time-consuming activity for companies. This reflects the same result as in prior years. The Study noted that as a result much less time is spent on proactive and direct communications and answering focused questions from asset managers and financial analysts.

Finally, information which is made available to companies subject to ESG ratings is unclear and diverse 50 . It is also difficult to find information online on whether ESG rating agencies charge companies to see their own report, although it appears from larger ESG rating providers’ websites that access is free of charge 51 . Details of the information available to companies are also absent and it is understood from exchanges with stakeholders that it varies from one ESG rating provider to another. It also depends on the approach taken by the ESG rating provider, as some don’t engage with companies, seeing their role as advisors of financial institutions and stressing that the engagement should be made by the financial institutions which made investment decisions.

2.1.3.Consequences

There are several consequences of this problem which have one element in common – the ability of investors, businesses and other stakeholders to use and rely on ESG ratings for their decisions related to sustainability. Some of the consequences are common to all types of stakeholders while others are more specific to their individual needs.

The consequences of the problems from a user perspective are:

1.Investors are unable to take sufficient account of sustainability-related and other non-financial risks and opportunities in their investment decisions,

2.Investors are also less able to channel financial resources to companies and economic activities that address and do not exacerbate social and environmental problems,

3.Benchmark administrators construct benchmarks based on ESG ratings where they do not have fully clarity over how they were computed, 52

4.Companies cannot consider all potential risks and opportunities from their activities and channel investments accordingly.

The specific consequence from a rated entity perspective is that they may receive an ESG rating in relation to elements that are outdated, or incorrect, which can impact the terms of their access to finance.

The specific consequence from other stakeholders’ perspective is that non-governmental organisations, trade unions and other stakeholders are less able to hold companies accountable for their impacts on society and the environment.

The problems have negative consequences for the functioning of the EU sustainable finance market, leaving a gap in the EU sustainable finance framework that was established to provide more transparency and the tools for private capital to flow to sustainable investments that are urgently needed for the transition. The wider consequences are that the potential of the European Single Market to contribute to the objectives of European Green Deal, and to the achievement of the UN Sustainable Development Goals, is not fully utilised.

2.2.What are the problem drivers

There are two main but complementary problem drivers, further detailed below:

1.Lack of clarity on the charateristics of ESG ratings, their methodologies and their data sources (this includes how they are created and disclosure/transparency on these), and

2.Lack of clarity about and control of operations of ESG rating providers.

Users of ratings suffer from imperfect information, and it is not possible for them to assess and compare the value of ratings (as a product) without sufficient information on their scope and quality. ESG rating providers do not have incentives to be as transparent as users would like them to be, as the former perceive commercial value in their methodologies.

2.2.1.Problem driver 1: Lack of clarity on the characteristics of ESG ratings, the methodologies used and data sources

The first problem driver captures two main issues, (a) the lack of transparency on rating methodologies and (b) the lack of clear and consistent terminology, including the lack of understanding of what ESG ratings measure.

There is a notable demand for greater transparency of objectives, methodologies adopted and quality assurance processes in place by ESG rating providers that is currently not met 53 . Users fail to be able to make informed decisions and lack trust in the market. According to IOSCO’s report, the lack of transparency is “a key factor in encouraging users to build proprietary rating methodologies”, 54 even though external ESG ratings should in principle save time/cost to users. This creates a market inefficiency, which could produce consequences for sustainable investment at large. On the other hand, market participants should not entirely rely on external ESG ratings, as this may also lead to market inefficiencies, as it has happened in the past in the market of credit ratings.

Without sufficient transparency of the methodology adopted, it is not possible to assess how effective they have been in evaluating a company, or whether the selected criteria align with the sustainability objectives of the user 55 . A lack of transparency leads to a lack of understanding as to what the rating represents. This may lead to misuse of ESG ratings, if taken without making own judgements and additional analysis 56 . This is particularly true for passive investing, where investors do not actively manage individual securities but track a market-weighted index or portfolio that can be built based on ESG rating characteristics.

a.Lack of transparency on rating methodologies

oOverall lack of transparency of ESG rating methodologies

Investors, companies and other stakeholders face difficulties to understand how ESG ratings are computed and what criteria and assumptions are taken into account 57 58 . They find that ESG ratings lack understandability and are too complex 59 . At the same time, they welcome the diversity in the offering of ESG ratings 60 and what they seek most in an ESG rating is the depth and breadth of data coverage 61 .

The majority of publicly available methodology documents (which some ESG rating providers make available, while others do not) only cover their general approach to ratings, key sustainability-related issues or topics leveraged and some information on their scoring methodologies. The Study demonstrated that few providers disclose the underlying indicators that feed into specific topics or the actual weights of the sustainability-related topics or issues based on industry. Information is very diverse and in the absence of homogeneity it is difficult for users to understand and compare ESG ratings.

In October 2022, in light of the developments and discussions around ESG ratings and their transparency, a large ESG rating provider decided to publish, for free, on their website, methodology documents describing the calculations, data inputs, and processes followed to maintain ESG methodologies 62 . The methodology documents are divided between core documents and issue specific documents (e.g. biodiversity and land use, packaging materials and waste key, human capital development, business ethics, etc.). It clarifies that the methodology documents are not intended to provide an exhaustive description of all inputs, data sources, or derived calculations used to arrive at the final assessment. Rating reports of companies however, given the nature of their business model - user (investor pay) - are behind a paywall. Disclosures for other market participants are very diverse, ranging from a high-level summary / introductory document such as FTSE’s ESG Ratings and data model 63 , Sustainalytics risk ratings 64 , to more detailed information and paper from Refinitiv (20 pages+) detailing the breakdown of the components that go into their rating 65 .

Users of ESG ratings also find it difficult to understand what industry or sector classification was used and how it affects the rating. The Study identified that the majority of ESG ratings (with the exception RepRisk) use some level of industry or sector specification (such as SASB, GRI and/or (TCFD 66 ), but the degree to which rating provider methodologies cater to industry-specific issues varies 67 . One provider uses the same general bank of topics when scanning for controversies across all companies, irrespective of industry specifics. Hence, important industry-specific characteristics may be left out, which could also complicate comparison of companies from such industry based on ratings. Other concerns have been expressed with the fact that should companies subject of ESG ratings not respond in full to questionnaires sent to them by ESG rating providers, then they would get a bad rating for those questions left unanswered 68 .

There is also limited information on how certain key elements such as company controversies, press information or allegations are factored in an ESG rating and whether and how they are duly considered. The Study noted that the controversy scan process is updated more regularly (e.g., daily or weekly) than a company’s final ESG rating but the processes diverge. The majority of ESG rating providers have unique methodologies for assessing and incorporating company exposure to controversies into their scoring frameworks 69 , but information publicly available online is very limited or non-existent. Companies have also complained that some ESG rating providers still take into account controversies based on old and outdated information, impacting their final rating. In addition, ESG rating providers would not always inform companies, give clarity as to whether and how a controversy is being dealt with; instead, companies need to be proactive and check regularly what information is available on them online 70 .

Only a limited number of ESG rating providers 71 also make their ESG ratings publicly available. All other providers have established some form of paywalls, either through their own self-developed platform and/or via a paid Bloomberg subscription, which limit ESG ratings access to subscribers.

oConcerns and effects of biases and low-correlation of ESG ratings

The lack of understanding over ESG ratings by users and issuers is also affected by biases and concerns over low correlation between ESG ratings, which are well documented in literature 72 .

The first, and most commonly referenced concern, refers to company size bias, where larger companies with high market capitalisation may obtain higher ESG ratings while not necessarily demonstrating a better performance in economic, social or governance factors, because of the ability to dedicate more resources to non-financial disclosures 73 . Company size bias appears to be mainly based on larger companies having the resources to prepare and disclose more sustainability-related information than smaller companies 74 . Larger companies often have in-house sustainability teams and have the financial means to hire specialized consultants to support them develop targeted sustainability-related disclosures. A study 75 stressed that the way ESG rating providers measure company sustainability gives an advantage to larger companies with more resources, while not providing investors with adequate information needed to make decisions based on sustainability-related factors. Another study from Berenberg 76 compared the coverage of the three largest ESG rating providers to examine data gaps in 2020 and found that ESG ratings are inherently skewed towards large and more mature companies, while small and rapidly growing companies on average have lower ESG ratings.

Geographical bias is the second point of concern, that can also be very diverse in form. For example, where there is a geographical bias toward companies in regions with stronger reporting requirements. Other geographic biases can be where ESG ratings are influenced by the geographic location of the rated firm and the firm’s industry type 77 . Lastly, there can also be geographic biases where some ESG ratings would also be influenced by the country of origin of ESG rating providers 78 . European companies are also required under the CSRD and the EU taxonomy to establish sustainability reports and reports on specific ESG factors, while this might not be the case in other parts of the world 79 . ESG rating providers, because of their location, would also put emphasis on certain aspects more than on others (e.g. more weighs on social issues) or consider different dimensions 80 .

A third concern is industry bias, where differences in business models are not accurately captured in composite ratings. Some sectors may also be given low scores on average 81 .

The lack of understanding and clarity over methodologies has also raised concerns over low correlation (also referred to as divergence) of ratings, where companies may get different ESG ratings for what appears to be identical/similar dimensions. One of the likely explanations for such low correlations is the lack of transparency 82 and absence of clarity over what ESG ratings measure 83 , how they are computed and their specific objectives 84 . It is also because of the different ways ESG rating providers assess ESG elements and, depending on what they consider relevant, include those in their computations. Stakeholders understand ‘E’, ‘S’ and ‘G’ differently, and there is a subjective nature in the way sustainability and the materiality of ESG categories is assessed.

The main divergences are (i) scope divergence, for the selection of the sets of material categories, (ii) weight divergence, with the importance given to each ESG category, and (iii) measurement divergence, with the use of different indicators to explain the same attribute, or to measure the same ESG criteria 85 . The same study indicates that measurement (which includes definitions and indicators) contributes 56% of the divergence, scope 38%, and weight 6%. According to another study, the main source of divergence would be measurement divergence, explaining 50% of the divergence 86 .

In the absence of clarity to users, the latter may take ESG ratings at face value without digging more into the specifics. Divergences across ESG ratings that have different data points, assumptions and objectives, where stakeholders are clearly informed about these, should not be a problem, as they are effectively different ratings. As part of their job, institutional investors analyse the differences in ESG ratings across providers, to identify that which is best for their strategy and meets their clients’ demands. ESG ratings are used to supplement asset managers and asset owners in-house research, including engaging with companies directly, to form their own opinion on sustainability performance. Differences in approach across ESG rating providers also reflect investor diversity, needs and the varying focus or priorities across investors. However, low correlation between ESG ratings that have similar/identical data points and objectives should in principle not exist.

b.Lack of clear and consistent terminology and lack of understanding of what ESG ratings measure

There is a lack of clear and consistent terminology used among ESG rating providers and a need for a clearer and standardized definition of what an ESG rating or an ESG score means and what it effectively captures.

oLack of clear and consistent terminology

The notion of ESG rating is applied in an inconsistent manner by market participants, which creates confusion and a lack of clarity for users as to what is effectively in scope or not.

In its set of recommendations published in November 2021, IOSCO defined ESG ratings with a wide scope, as “the broad spectrum of ratings products that are marketed as providing an opinion regarding an entity, a financial instrument or a product, a company’s ESG profile or characteristics or exposure to ESG, climatic or environmental risks or impact on society and the environment that are issued using a defined ranking system of rating categories, whether or not these are explicitly labelled as ‘ESG ratings’”. This is the first definition established by an international institution. It is critical that this definition does not limit the scope to those services that are labelled strictly as ‘ESG ratings’, as otherwise services with the same characteristics could be omitted and the objective of this initiative to bring transparency to the market over those services would be hampered. Services that have the same attributes should be captured. This definition also refers to the notion of ‘opinion’ and it is important to clarify for the purpose of this work what this captures.

The market generally uses the term ‘ESG rating’ where there is a process involving an analyst, a human role in the process, but also some form of rating committee, overseeing the application of principles in the organisation. In such scenario, the ESG rating providers use both quantitative and qualitative data, with the rating as an outcome.

A number of ESG services providers 87 also offer so-called ‘ESG scores’, which they consider are different from a rating as there is no involvement of an analyst, of a person. ESG scores are based on automated quantitative analysis, set-up through an algorithm. Even if they could be considered as having some specific attributes, again because no analyst and personal judgement/opinion is given during the rating process and over input data, they are to fall within the broader definition of ESG ratings for the purpose of this impact assessment. ESG scores are effectively the same output and are used the same way and for the same purpose by users and should benefit from greater transparency over their methodology. Automated quantitative analysis is also based on some human intervention where choices have been made at the initial stages of the process on what data to take into account, under what circumstances, and on the respective weights. IOSCO in its set of recommendations also captures ESG scores in the definition of ESG ratings 88 .

In terms of object and actual use of ESG ratings, the majority of them are used in the context of investment decisions, by institutional investors, either to directly feed into the construction of the portfolio or as the input into the investment decision process, or via the construction of the benchmarks or as an input into the internal analysis or internal rating process. There are also ESG ratings used directly by companies, at their request, where they aim to assess risks or opportunities, or for contractual purposes, for example when the company asks for the assessment of its supply chain (e.g., EcoVadis 89 ).

oLack of clarity over what ESG ratings measure

There is confusion about coverage of E, S, G matters. Some ESG rating providers cover Environmental, Social and Governance criteria collectively, while others are very much specialised to individual ‘E’, ‘S’ or ‘G’ aspects. In general, larger rating providers cover the broad ESG spectrum (e.g., S&P, MSCI, Sustainalytics), while smaller providers focus on individual aspects or even topics within these (e.g., Carbon4 Finance with climate related risks).

Aggregated ESG ratings may be considered as also lacking clarity regarding what element – ‘E’, ‘S’ or ‘G’ or a subset of either – drives the final rating to the highest degree. Hence, an investor may choose to buy shares or bonds in a company based on a good overall rating, whereas such company could be only performing well in one aspect and have poorer performance on other aspects which may be of more relevance to such investor. Without sufficient clarity on the criteria selected, weights used, if arithmetic or geometric average is used (allowing or not compensation between dimensions), an investor may not be able to make sufficiently well-informed decisions and may be misled by the overall rating. Aggregated ESG scores may not provide relevant information and could be detrimental for prudent decision making. 90 This creates a potential for greenwashing.

ESG ratings and scores can also be grouped into several categories based on what they are measuring. The Study noted that the most common and widely used form is ESG risk ratings (e.g., MSCI), which measure a company’s exposure to ESG risks and management practices. There are also ESG ratings considering impact from companies on the environment and society (so-called ‘double materiality principle’): these (e.g., Impak Finance, Carbon4 Finance) measure the impact of an entity on the environment in general, on society and/or on some given metrics. Beyond those two groups there are a number of other ratings that measure aspects like disclosure, or a specific issue, for example supply chain (e.g., Ecovadis). The most prevalent issue-focused ratings are climate-related ratings, but even these look at different aspects, such as climate physical and/or transition risk (e.g., Carbon4 Finance), greenhouse gas emissions or carbon capture. As already mentioned above, diversity in offerings of ESG ratings is not a problem and is welcomed by users of ratings, but they also need clarity and be certain as to the object of ratings they buy. They may think they are buying some ESG rating looking into impacts on the environment/society, while effectively it only analyses the financial risks of a company 91 .

Each ESG rating provider is different and may also use terms like ‘topics’, ‘themes’ or ‘issues’ to indicate areas that they consider when evaluating sustainability or ESG risk exposure or impact. Issue selection is ultimately up to the rating provider, and this is where differences in methodological approaches, and ultimately differences in the final company rating, start to appear. Most users of ratings are unable to access the information at hand to know of all the themes/issues taken into account and are vulnerable to being misinformed 92 . In the current market, the ratings are presented as ESG ratings regardless of whether they cover only distinct parts or the full set of the ESG factors.

The Study stressed that once a set of issues has been identified, ESG rating providers will then select specific indicators, metrics or KPIs that measure performance against an identified issue. For example, one may have identified ‘climate change’ as a relevant issue to consider and use ‘carbon emissions’ as the corresponding indicator to evaluate a company’s performance on that issue. While there is some overlap, the number and type of issues and indicators varies by ESG rating provider and should always be considered on a case-by-case basis by the user of the rating. The most consistent measurements tend to be in the ‘Environmental’ category, such as greenhouse gas emissions, energy consumption, waste disposed, etc. 93 . Consistency in other topics, in particular ‘social’ tends to be more difficult. Units of measurement for each indicator can vary across ESG reports and surveys (sometimes depending on the country). Many ESG rating providers do not necessarily use all of the indicators collected on a given company in the final rating, instead, they will make a decision as to which indicators are most relevant for a company in their opinion, based on which industry or sector the company fits within. Given this choice from ESG rating providers, and the absence of transparency of methodologies, users may find it difficult to trace back characteristics of ESG ratings and exactly what criteria was selected, for what purpose.

2.2.2.Problem driver 2: Lack of clarity about, and control of, operations of ESG rating providers

ESG ratings providers currently do not have to comply with any rules in order to offer their services in the EU, including on their organisation/operations (i.e. there is no authorisation, registration or supervision) neither at national, European or international level, unlike other financial services actors, such as credit rating agencies 94 , benchmark administrators 95 , asset managers 96 .

Users, companies and other stakeholders face difficulties in understanding how ESG rating providers are structured, organised, and how the independence of their methodology and overall operations are guaranteed. This contributes to creating an overall lack of trust in the market but also potential risks, including of greenwashing and socialwashing.

One of the key issues relating to this concerns potential conflicts of interests, and the need for more robust governance arrangements.

The Study identified that there are four primary potential conflicts of interest, relating to ownership, product mix, separation of commercial and analytical teams, and methodology governance.

Regarding ownership, the Study mentioned that ownership raises a conflict of interest if not considered independent from the companies subject to evaluation. For example, if the parent company of the rating provider owns a portfolio of other companies rated by the rating provider, it could be perceived as influencing the rating provider. It also concerns ‘sister firms’ of ESG rating providers that may receive higher ratings from the affiliated ESG rating provider than they do from independent firms. An empirical study 97 also found that a rater may have ‘more incentives to provide favourable treatments to sister firms held by multiple blockholders than those sisters held by single blockholders. Second, it may also be the case that if a large shareholder of the rater is powerful and influential in terms of asset under management, it may create more incentives to assign higher ratings to the large shareholder’s holding firms'. This study also noted that ESG rating providers may have some incentives for inaccurate ESG ratings in order to establish or maintain a good reputation, or because of relationships of the ESG rating provider.

As regards the Mix of Products and Services Offered, some ESG rating providers rate companies and simultaneously offer paid advisory/consulting services to the same companies to improve ratings, either directly or via third party recommendations. This is perceived by companies and investors as a potential significant conflict of interest 98 and may create a bias towards those companies willing to pay for such advisory services (or perverse incentives to first give a worse rating and significantly improve it after advisory services are utilised). Some ESG rating providers also charge companies that are subject of ratings (but not users thereto) to see their own reports, again creating a perceived bias towards those companies with resources to financially engage.

On the Separation of Commercial and Analytical Teams, the Study concluded that providers selling multiple products – such as ESG ratings, ESG data, ESG benchmarks - may require an elevated and appropriate separation between departments to avoid potential conflicts of interest. Most ESG rating providers have introduced some rules on the separation of their departments 99 , but information and details on how such separation is effective, benefits from a ‘Chinese wall’, a business term used to describe a virtual barrier erected to block the exchange of information between departments in a company 100 is very limited but also unclear 101 . Furthermore, a conflict of interest may arise if product and service provision teams have the potential to be influenced by teams from the commercial part of the business.

Finally, regarding Methodology Governance, the Study highlighted that a conflict of interest may arise if an ESG rating provider does not have processes in place to ensure that the methodology is appropriately applied and analysts are not influenced by outside parties.

Another aspect which also seems to lack transparency and clarity concerns fee schedules, rates and terms from ESG rating providers. In its overview of findings from its Call for Evidence, ESMA concluded that “transparency issues seem to exist regarding the minimum term and fee rates of ESG data and rating contracts, leading to different users contracting for the same ESG products under differing contractual terms” 102 . Little information is available publicly, on the websites of sustainability assessment providers (that include rating providers) about fee schedules and how much they charge for various products. No information is also available specifically on ESG rating fees. Some providers indicate some information but only with very large ranges for their fees related to various products and services they provide 103 without specifying further reasons/criteria for the differences and how users/issuers will be charged for various products and services. Some providers also indicate that fees are negotiable and that discounts may be available where clients subscribe to multiple products or additional locations 104 .

Some other concerns have been raised with regards to the handling of confidential information by ESG rating providers in the absence of any clarity on how they treat such information 105 .

Robust governance is needed in the operations of ESG rating providers to give certainty and clarity for users over the way ESG rating providers are structured and provide their services.

2.3 How likely is the problem to persist?

There is little evidence to suggest that the current trends and functioning of the ESG rating market will change.

Due to the expected strengthening and growth of the ESG rating market, driven by booming interest in sustainable investment as well as expectations on asset managers, banks and insurers to disclose ESG-related information on their portfolios, the vulnerabilities that exist within the current market-based system are likely to grow. This may amplify the risk of a potential high-impact or high-visibility occurrence of greenwashing. This already creates reputational problems and risks for the ESG rating market and leads to inappropriate investment decisions or allocations to e.g. companies that do not contribute to greening the environment / tackling climate change, thereby reducing the trust of investors on which this market relies. This may also eventually lead to increasing costs for issuers as well as for society generally.

Section 5.1 (describing the baseline scenario or the so-called “no policy change” scenario) provides a more detailed description of the expected evolution in the case of no new policy initiative by the European Union.

3.Why should the EU act?

3.1Legal basis

In this Impact Assessment, the combination of three Articles of the TFEU were considered as legal bases. Firstly, Article 50(g) TFEU on coordinating safeguards required of EU companies by Member States with a view to making such safeguards equivalent throughout the Union, and/or Article 53(1) TFEU on the taking-up and pursuing of activities by self-employed persons. Then and in addition, Article 114 TFEU, lex generalis, that applies as well with the objective of establishing or ensuring the functioning of the internal market.

3.2Subsidiarity: Necessity of EU action

The objective of this initiative, to improve clarity over the characteristics of ESG ratings but also operations of ESG rating providers, cannot be sufficiently achieved by Member States acting independently and action at EU level is necessary for the proper functioning of EU capital markets.

There is currently no EU regulatory framework for ESG rating providers. Member States do not regulate the activities of ESG rating providers or the conditions under which they deliver their products or services. Each ESG rating provider follows its own rules and clarity over what they do and how they do it is missing. It can be said that the ESG rating providers market is a black box.

The market of ESG rating providers is global, and the biggest providers operate in a number of Member States. They may have a headquarter in one Member State and subsidiaries in others or in many cases have their headquaters outside the EU and their subsidiaries within the EU. ESG rating providers also provide cross-border services.

Although Member States could individually take action to strengthen the reliability and comparability of ESG ratings, such measures are likely to be significantly different between Member States, which may create diverging levels of transparency, barriers for market participants and challenges for those operating across borders (as this is particularly the case for this market, with users in a number of Member States), in addition to limiting comparability between ratings. For example, an ESG rating provider would potentially have to register with and meet different requirements of national authorities, creating extra costs. Moreover, different and potentially even inconsistent requirements could result in divergent operations of the same ESG rating provider across the EU; this could even lead to divergence of ratings even within the same provider. This would cause issues and confusion for rated companies and rating users alike and would ultimately lead to an uneven protection of investors in different Member States. Some users could also be tempted to choose an ESG rating provider located in a given Member State because of the absence of or lower criteria set, creating a fair competition issue.

On the other hand, with the complete absence of rules on the operations of ESG rating providers, the situation as it stands today and issues relating to transparency would continue and could even worsen, as further explained in Section 5.1.

3.3Subsidiarity: Added value of EU action

As attention to sustainable investing and ESG ratings grows in jurisdictions around the world, it becomes essential for the EU to engage with its partners on the basis of a coherent and comprehensive European approach. The initiative would not replace national legislation as there is currently no legislation in place in any Member State regulating the functioning of ESG rating providers.

The general objective is – by improving clarity over the characteristics of ESG ratings but also operations of ESG rating providers – to better exploit the potential of the European Single Market to contribute to the transition towards a fully sustainable and inclusive economic and financial system in accordance with the European Green Deal and UN Sustainable Development Goals.

The functioning of the internal market would be improved with further clarity over operations from actors that are increasingly important for channelling finance.

4.Objectives: What is to be achieved?

4.1General objectives

The general objective of this initiative is to better exploit the potential of the European Single Market and to contribute to the transition towards a fully sustainable and inclusive economic and financial system in accordance with the European Green Deal and UN Sustainable Development Goals. This can be achieved by improving the ability of investors to make informed decisions regarding the sustainability of investments as well as by helping companies to understand their sustainability performance. Since ESG ratings and underlying data are used for investment decisions and allocation of capital, the general objective of the initiative is to ensure that ESG ratings enable investors and rated companies to take informed decisions with regard to managing ESG risks and the impacts of their investments or operations. This can be achieved by enhancing the reliability and quality of information about ESG ratings.

At the same time, it is crucial to foster trust and confidence in the operations of ESG rating providers by ensuring that the market operates properly and ESG rating providers prevent and manage conflicts of interest. Reliable and quality ESG ratings will also contribute to the integrity of the financial ecosystem and investor protection.

4.2 Specific objectives

There are two specific objectives, which relate to the two problem drivers:

·Increased clarity on ESG ratings characteristics (what they mean and what objectives they pursue), the methodologies and the data sources or estimates used to obtain the ESG ratings

·Provide increased clarity on the operations of ESG rating providers as well as ensure the prevention and mitigation of risks of conflicts of interest at ESG rating providers’ level

5.What are the available policy options?

5.1What is the baseline from which options are assessed?

The market of ESG ratings has expanded rapidly, following the growth in demand for ESG ratings, in particular from investors. However, the market is not functioning well.

The baseline scenario for the purposes of this impact assessment is one of ‘no policy change’ in the market of ESG ratings. The present trend of increased usage of ESG ratings means that the problems will grow accordingly in the next 5-10 years. The new requirements imposed on financial institutions and market participants with the sustainable finance legislation (such as the CSRD, SFDR, Taxonomy Regulation) will continue to put pressure on the market and lead to an increased demand for ESG ratings but also other sustainability assessment tools.

In the survey conducted for the Study and the IRRI Survey 2019 106 , the vast majority (75%) of asset managers anticipated an increased demand for both in-house and external ESG ratingsRespondents consider the decisive determinants of this growth to be demand from investors for ratings for their investment decisions, further standardisation of information disclosed by companies and other market participants, and growth in demand from companies for ratings including on future strategies. The majority of respondents expect to increase their usage of ESG ratings.

Not only the use of ESG ratings is expected to grow but also the role they play in investment strategies. Changes of investment trends are already visible. In 2020, ESG integration became the dominant approach, surpassing exclusionary screening-based strategies and became the largest sustainable investment strategy globally according to Global Sustainable Investment Review (GSIR) 107 with a combined USD 25.2 trillion in assets under management employing an ESG integration approach, also being the most commonly reported strategy in most regions 108

While coming directly from the market, demand for ESG data and ratings is also reinforced by sustainable finance legislation, such as the CSRD, SFDR, Taxonomy Regulation or EU Climate Benchmarks legislation, and initiatives under the revised Sustainable Finance Action Plan.

Notably, the entry into force of the CSRD and the introduction of mandatory sustainability reporting standards for companies should impact the baseline scenario by increasing the availability, standardisation, quality and reliability of corporate sustainability data. This will have a positive impact on the quality of ESG ratings, which relies on the quality of available data. Unfortunately, the impact on correlation amongst ratings is harder to predict. According to a research paper published in 2021, “greater ESG disclosure actually leads to greater ESG rating disagreement” 109 . Where ESG rating providers appear to disagree most is on the assessment of firms ‘outcome metrics’, which refer to actual changes achieved (e.g., the percentage of women employed by a company), since evaluating outcomes is more subjective 110 . As such, it is possible that ESG ratings divergence may increase with the implementation of the CSRD. However, as the authors themselves note, this dynamic may be related to the early phase of ESG disclosures, when there is little consensus among analysts on ESG performance metrics. Indeed, this dynamic of divergence may neutralise or reverse in the future. In any case, greater disclosure will reduce the issues around data collection.

Continued uncertainty for users of ESG ratings.

According to the Study, half of asset managers and half of benchmark administrators use corporate data and carry out their own assessments, whereas the other half rely on third party ESG ratings. This has been confirmed by the targeted consultation: only 17% of respondents stated that ESG ratings are a decisive input for their investment decisions; the majority use them either as a starting point for their internal analysis (32%) or as one of many sources of information (30%). This means that the improvement of the quality and availability of corporate reporting will have positive consequences for those users who carry out their assessments internally. However, users continue to face problems linked to third party ESG ratings, which can result in bad investment decisions and even greenwashing. Without any action, a continued lack of trust in ESG ratings can be expected. 111 .

Continued risk of misalignment of investments and greenwashing 112

The risk of greewashing and misalignment of investments will continue to incease with the growth of ESG funds and of the usage of ESG ratings. As long as ESG ratings and their bases remain insufficiently transparent, investors may believe that a company they invest in has some positive impact/consideration of the environment, while in reality environmental/social impact is not the objective of the rating or some key elements have not been factored in (e.g. controversies). For example, a rrecent article by Bloomberg - and their analysis of MSCI’s ESG rating for McDonald’s 113 - highlights potential risks and consequences associated with the lack of or sub-optimal transparency on ESG ratings that may lead to some form of greenwashing or socialwashing 114 .

Given the increasingly important role the ESG ratings play in the ecosystem 115 , the lack of reliable and credible ESG assessments and possible greenwashing can lead to misalignment of investments with Green Deal objectives. Misalignment of investments may in turn undermine the stability of the financial system, if in the future corrective actions will be necessary.

5.2Description of policy options

In order to focus the impact assessment on a limited number of key variables and improve its readability, this section presents different policy options for two main variables:

·increased clarity on operations of ESG rating providers as well as the prevention of risks of conflict of interest at ESG rating providers level, and

·the need to improve clarity of ESG ratings characteristics (i.e. what they mean and what objectives they pursue, the methodologies and the data sources or estimates used for their development).

These variables are most important in terms of (i) addressing the identified problems and (ii) the principal determinants of costs.

Scope is also an important dimension as entities fall under the scope of the initiative need to be defined precisely. The purpose of the initiative is to cover all ESG rating providers that operate in the EU. Therefore, the term “operate in the EU” must be defined, as well as the term “ESG ratings”. The analysis of the scope is done in a separate Chapter 5.5.

Policy options are analysed for two dimensions, i.e. entities (ESG rating providers) and products (ESG ratings).

1.Regulatory treatment of ESG rating providers:

oOption 1 – Industry code of conduct

oOption 2 – Registration and light supervision

oOption 3 – Authorisation, principle-based organisational requirements, transparency requirements and risk-based supervision

2.The extent of transparency requirements on ESG ratings and their methodologies

oOption 1 - Minimum disclosure requirements for the public

oOption 2 –Minimum transparency disclosures to the general public and more comprehensive disclosures to clients of ESG rating providers and rated companies

Overview how dimensions address the specific objectives:

		Dimensions
		Regulatory treatment of ESG rating providers	Transparency on ESG ratings
Specific objectives	Increase clarity and oversight over the operations of ESG rating providers	X	N/A
	Increase clarity of ESG ratings characteristics	N/A	X

5.2.1Regulatory treatment of ESG rating providers

Option 1 – Industry code of conduct (voluntary)

The Commission would encourage the development and adoption of a voluntary industry code of conduct regarding operations and disclosures by ESG rating providers. The code of conduct would be developed directly by the industry, but should in that case ideally be built on the IOSCO recommendations 116 .

The code of conduct would cover the following recommendations:

·adopting and implementing policies and procedures designed to help ensure the issuance of high quality ESG ratings, including defined methodologies,

·adopting and implementing policies and procedures designed to help ensure their decisions are independent, free from political or economic interference,

·identification, or appropriate management, mitigation and disclosure of potential conflicts of interest,

·adopting and implementing policies and procedures concerning public disclosure and transparency on the scope and objective of providers’ rating product(s), as well as their methodologies and processes,

·adopting and implementing policies and procedures designed to address and protect all non-public information received from or communicated to ESG rating providers by any entity.

Providers could be asked to apply the code of conduct on the comply or explain basis.

Option 2 – Registration and light supervision of ESG rating providers

The option entails the establishment of a central registry of those ESG rating providers falling under the scope of this initiative. Under this option providers would be required to register at EU level as well as provide basic information (such as entity name, domicile, registration number, scope of activities, type of services provided, type of business model, policies on the prevention of the conflicts of interests, policies on ensuring the quality of ESG ratings, etc.) and update the information as necessary.

Given ESMA’s role and experience as a European supervisor, e.g., of credit rating agencies, it would be logical for ESMA to also manage the central registry for ESG rating providers. Only registered entities would be permitted to provide ESG ratings. A simplified sanctioning regime will be applied in case of non-compliance with the registration requirements. ESMA would be given the task of enforcement.

Information to be provided for the registration purposes will be specified in Level 2 measures. This option doesn’t include on-going supervision of registered entities.

Under this option, the information made available on the central registry would provide clarity to markets on the operations of ESG rating providers. However, this option does not offer greater transparency or clarity over characteristics/methodologies of ESG ratings.

Option 3 – Authorisation, proportionate organisational requirements and risk-based supervision of ESG rating providers

This option requires in-scope ESG rating providers to undergo an authorisation process in order to offer their services in the EU. Authorisation will be subject to meeting proportionate and principle-based requirements.

These requirements should encompass the following:

·the provider’s rating/scoring processes and decisions are independent, free from political or economic interference,

·providers have procedures for ensuring the issuance of high quality ESG ratings, including the use of transparent and defined methodologies and ensuring the quality of data used in the rating/scoring process,

·the provider’s rating/scoring business is properly separated from other businesses to avoid potential conflicts of interest and if ancillary services are provided, they do not influence the role of analysts while preparing an ESG rating,

·providers have governance and internal control policies, including identification, for appropriate management, mitigation and disclosure of potential conflicts of interest or undue influence; addressing the specificities of company pay model and user-pay model,

·providers have policies and procedures designed to address and protect all non-public information received from or communicated to them by any entity,

·providers have policies on the disclosure of key information concerning their business model and their policies,

·Providers have rules in place regarding record keeping,

·Providers have rules in place for receiving, investigating and retaining records concerning complaints made.

Details of the information to be provided for the purpose of authorisation will be specified in delegated acts, to be developed by ESMA.

Given the cross-border nature of activities of ESG rating providers and the importance of ensuring a consistent EU supervisory approach, authorization and supervision will be carried out by ESMA.

This option subjects in-scope ESG rating providers to ongoing (EU) supervision, based on a risk-based and data driven approach and proportionate to the size of the entity. This approach has been undertaken by ESMA in relation to all directly supervised entities. 117

This option would also provide for a sanctioning regime and the application of supervisory measures by the supervisor (ESMA). This regime would be framed in the legislation, it would provide for fines, public notices in case of non-respect of the obligations.

Authorization and supervision should not be too burdensome for new entrants and smaller providers such as to prevent them from entering or/ remaining in the market.

To mitigate potential concerns about loss of access to the market providers, providers would be allowed to continue operating on the condition that they notify ESMA and become authorized within a pre-determined period of time, the transitional period 118 . The length of the transitional period will be adapted to the size of players, giving smaller players more time to complete the authorisation and in this way easing the burden for them. Similarly, a transitional period for new and small entrants to the market could be foreseen, requiring them to notify ESMA and then giving them a period of time to complete the authorisation process.

The mitigation measures for smaller providers are discussed in Annex 5 and include: (i) transition period (ii) adjustment of supervisory fees to the size of the provider (iii) risk-based supervision (iv) exemption for smaller and innovative providers from certain organisational requirements which may not be proportionate to the nature, scale and business model of the provider.

5.2.2Level of transparency requirements on ESG ratings and their methodologies

Option 1 - Minimum transparency for the general public only

This option requires all ESG rating providers to publicly disclose key information about ESG rating characteristics and methodologies, clearly indicating at least the following:

·high level overview of the rating methodologies used (and changes thereto), including whether analysis is backward-looking or forward-looking,

·high level overview of data processes (data sources, including if they are public or non–public, and if they are sourced from CSRD reports, estimation of input data in case of unavailability, frequency of data updates),

·information on the ratings’ objective (e.g., whether it assesses risks or impacts (or both), is it a compliance assessment or supply chain assessment or other type of rating),

·clearly marking whether the rating is assessing risk or impacts or some other dimension,

·application of the materiality perspective,

·the rating’s scope – i.e., is it an aggregated rating (aggregating E and S and G factor), rating of individual factors or specific issues (e.g., transition risks),

·in case of aggregated ESG rating, weighting, i.e., percentage of the different ESG factors (e.g., 33% Environment, 33% Social, 33% Governance), and the explanation of the weighting method,

·within the E, S or G factors, specification of the topics subject to the ESG rating/score, also looking at the topics from the CSRD (e.g., climate, biodiversity, pollution, water and marine resources, working conditions, ethics and corporate culture),

·information on whether the rating is expressed in absolute or relative values,

·reference to the use of Artificial Intelligence (if applicable) in the data collection and/or rating/scoring process,

·high level information on pricing methods.

Option 2 - Minimum transparency disclosures for the general public and more comprehensive disclosures to clients of ESG rating providers and rated companies

This option would require all providers to disclose more granular information about ESG rating characteristics and methodologies, in addition to public disclosures covered in Option 1. This option would require providers to disclose the following additional information to their subscribers and rated companies:

·a more granular overview of the rating methodologies used (and changes thereto), including whether the analysis is backward-looking or forward-looking, which metrics have been selected as relevant, the main KPIs and weighting method, any potential shortcomings of methodologies, policies for the revision of methodologies, last date of the revision;

·a more granular overview of data processes, including

omore detailed explanation of data sources used – e.g., public or otherwise, mentioning whether derived from CRSD reports,

owhere data is unavailable, indicate the use of and explain any estimation methodology,

othe policies for updating data and revising historical data, date of last updates of data,

odata quality controls,

oengagement with rated companies.

·If applicable, an explanation of any AI methodology used in the data collection and/or rating process.

Precise disclosure requirements would be developed by ESMA as a Level 2 measure.

For both Option 1 and Option 2, It is not the objective of this initiative to require ESG rating providers to disclose intellectual capital, intellectual property, know-how or the results of innovation that would qualify as trade secrets as defined in the Trade Secrets Directive 119 .

5.3Scope, Definition of ESG ratings and types of providers

5.3.1Entities to be covered/geographical coverage

Figure 1: Share of respondents by nature of establishment according to the mapping exercise carried out by ESMA

The market of specialised ESG rating providers is global, with the biggest providers having headquarters outside of the EU (Bloomberg – USA, MSCI USA, Refinitiv- UK, RepRisk – Switzerland, S&P – USA, Morningstar – USA, Moody’s – USA).

According to the mapping exercise carried out by ESMA in the context of the Call for Evidence 120 (see Figure 1 above), there are 59 entities providing ESG ratings in the EU, either to European investors or companies. Half of the ESG rating providers that responded to ESMA Call for Evidence indicated that they are an independent legal entity without a group structure. The other half are either parent companies or subsidiaries of a group structure. Many providers have more than one entity in the EU, and many have headquarters outside the EU. Out of 59 entities, 14 have no presence at all in the EU, yet they provide services to European investors and/or companies.

It is crucial that the initiative covers all providers operating in the EU and not only those having a legal entity established in the EU. In addition, even if the group has a legal entity in the EU, the EU may not be the location where the development of ESG ratings takes place. The market is global and so are the operations of providers that may have data specialists in one location, analysts and sales teams in others.

Firstly, this initiative should create a level-playing field for all ESG rating providers operating in the EU. In order to achieve that the initiative cannot leave out any providers, no matter what size or location, as long as they operate in the EU.

Secondly, the term “operate in the EU” needs to be clearly defined. For the purpose of this initiative, the term “operating in the EU” is defined as providing services to European regulated financial institutions and other regulated financial market participants as well as to European companies in scope of the CSRD. There are the following reasons for such a definition:

·Two thirds of ESG rating providers offer their product to professional investors, i.e. financial institutions and financial market participants, mainly asset managers and benchmark administrators.

·One third of providers offer their products to companies (meaning that it is a company which requests and pays for the ESG rating), Therefore, in order to capture that group, the term “operating in the EU” needs to include ESG ratings requested by European companies, both listed and large non-listed ones (as defined by the Corporate Sustanability Reporting Directive).

From an economic point of view, it is also crucial to bring within scope all ESG ratings providers operating in the EU, including also small providers.

In line with the responses to the public consultation, leaving out small providers would be contrary to the objective of this initiative i.e. ensuring that ESG ratings enable investors and rated companies to take informed decisions with regard to managing ESG risks and impacts of their investments or operations). Exclusion of smaller providers could negativelly impact them, as they could be regarded as less trustworthy and their ratings as of less quality and reliability. Therefore, there is no other option that could be addressing the objectives of this initiative, apart from including under the scope of the initiative all providers that operate in the EU, no matter what is their location or size. At the same time, this initiatve foresees mitigating measures for smaller providers.

5.3.2Alternative coverage of geograhical scope

We have analysed alternative options on the geographical scope, namely the differentiation of requirements for the entities located in the EU as opposed to those located outside the EU. We have concluded that the option limiting the geographical scope to the territory of the EU would lead not only to an unlevel playing field, but also would not reach the objectives of the initiative, and as a result doesn’t constitute a viable option.

As explained in this Impact Assessment, European providers are mainly smaller in size and thus their market share is rather limited. If non-EU providers, which are larger and have bigger market share, could continue providing their services in the EU without being subject to the transparency requirments or organisational requirments, problems with the lack of transparency and problems relate to conflict of interest would continue to influence ESG ratings used by European investors. Therefore the objectives of this initiative would not be reached.

Summing up, there is no other viable option than the one presented in the Impact Assessment, to create a level playing field for all ESG rating providers operating in the EU. In order to achieve that, the initiative cannot leave out any providers, no matter where they are located, as long as they provide ESG ratings to European regulated financial institutions and other regulated financial market participants as well as to European non-financial companies.

5.3.3Types of providers

There are two major groups of ESG ratings providers: specialised entities providing ESG ratings to professional investors and companies (see description in Chapter X on Problem Definition and Annex 6), as well as financial institutions developing ESG ratings for their own purposes, principally for internal analysis or marketing.

It is important to make a distinction between ESG ratings that are made available for commercial purposes, and those ESG ratings that are developed in-house for internal use. Financial institutions use their own proprietary models and don’t provide ESG ratings commercially to other entities. Therefore, the key problem that this initiative aims to address (i.e. information asymmetry between providers and users of ESG ratings) does not materialize in their case. Furthermore, financial institutions are regulated entities and are already subject to sectoral legislation.

The Commission, following outreach activities, academic literature and the Study, has identified problems mainly in relation to the operation of specialised entities offering ESG ratings, while evidence of any problems in relation to the transparency and use of in-house ESG ratings developed by asset managers or benchmark administrators is rather anecdotal. In addition, asset managers are already subject to authorisation, supervision and disclosure requirements under EU sectoral legislation, i.e., Directive 2011/61/EU (Alternative Investment Fund Directive - AIFMD) 121 , Directive 2014/91/EU (Undertakings for Collective Investment in Transferable Securities (UCITS)) 122 and Regulation (EU) 2019/2088 on sustainability‐related disclosures in the financial services sector (SFDR) 123 . Given the absence of strong evidence of shortcomings in this market and extensive existing legislation on the operations of investors, the development of ESG ratings by asset managers, where the latter are not marketed/sold and where they are for in-house use, is out of the scope of this Impact Assessment. At the same time, it is also important not to disincentivise asset managers from developing their own ESG ratings.

Summing up on scope, this initiative will target the specialised entities providing ESG ratings to the public or to subscribers and will not cover financial institutions or other market participants developing ESG ratings for their own purposes. The development of ESG ratings for own purposes and own use is out of the scope of this analysis.

5.3.4Defining ESG ratings

As discussed in Chapter 2.2 on Problem Drivers, there is a lack of clarity around what ESG ratings are, what they mean and what they capture. There is no universally accepted definition of ESG ratings; in addition, market participants use different terms – such as ESG “rating”, “score” or “valuation”, or they refer to ESG “data” as a term encompassing data and ratings/scores/valuations.

As described in Chapter 2.2, there are two main distinctive types of products - opinions formed by analysts (usually referred to as ratings) and scores, as well as products mixing those features. Scores are typically based only on a pre-established statistical system or model, without any additional substantial rating-specific analytical input from a rating analyst. Most ESG ratings constitute a mixture of the two, with more or less involvement of analysts. There are, however, also pure scores as well as pure ratings provided in the form of an opinion with a role for the lead analyst.

The first attempt at defining ESG ratings has been undertaken by IOSCO (please see Chapter 1.1), widely agreed between practitioners and supervisors. It therefore forms the basis of the definition of ESG ratings for this initiative. Furthermore, this initiative will define the term ‘ESG ratings’, both for legal purposes and in order to define the scope of the initiative.

As explained above, it is crucial that the definition recognises the specificities and diversity/complexity of ESG ratings: ESG ratings can differ in their processes (involvement of an analyst or Artificial Intelligence / algorithmic driven or combination of both), in their scope (e.g. looking into the full ESG spectrum or only a subset or individual factors) and in what they measure (risks or impacts, compliance or double materiality).The introduction of a definition of ESG ratings is necessary to define the scope of legislation and application of its requirements.

At the same time, the purpose of this initiative is not to create a new label or a new product, but to ensure that existing products having similar features and performing similar functions fall under the scope of this initiative.

5.4Options discarded at an early stage

The following options were considered but discarded at an early stage:

5.4.1Registration and supervision at national level,

The market of ESG rating providers is global therefore any national system would be inappropriate and inefficient due to the fact that some providers have subsidiaries in more than one Member State, or even outside the EU, as well as provide services cross border.

·Harmonising methodologies of ESG rating providers,

There are several arguments against the harmonisation of methodologies. Both providers and users are against the harmonisation of methodologies, as they welcome a variety of approaches. Next, ESG ratings pursue different objectives and cover different areas so their methodologies cannot be harmonised. It is also crucial that ESG rating providers have full control of the methodologies they use and are independent in their choice. Lastly, defining the objectives and methodologies would affect innovation in this market and new ways of measuring ESG risks and impacts

5.4.2Setting minimum requirements on the content of ESG ratings,

As explained above, users welcome the diversity of asessments and do not call for regulating the content of ESG ratings. In addition, the ESG ratings pursue a variety of objectives and respond to demands from the professional investors. The existing classification systems for sustainability factors has been developed for corporates and not for the use by ESG rating providers. In addition, such an option would affect the inovation in this market.

5.4.3Detailed templates for disclosure requirements.

Given the variety of objectives that ESG ratings pursue and methodologies used, a number of templates would need to be developed, undermining the objective of this option. The introduction of standardised templates for disclosures might have some negative effects on innovation. In addition, the requirement for the use of templates would create additional costs and burdens for smaller providers and new entrants. ESG rating providers would need to implement new procedures, and potentially renew or adapt their existing IT and reporting infrastructure.

Annex 9 describes each of these options in detail and the reasons for discarding them at an early stage.

5.5CRAs – ESG factors in credit ratings

As explained in Introduction, Credit ratings are out of scope of this Impact Assessment, but given the interest of stakeholders in the developments in the area of credit ratings, Annex 8 summarises the evidence and analysis carried out by the Commission services and ESMA.

The main options analysed in Annex 8 are related to two problems:

·Varied degree of rigorousness of the incorporation of ESG factors in the assessment of creditworthiness. Furthermore, there is heterogeneity in the identification and definition of ESG factors across CRA methodologies.

·Insufficient disclosure in relation to ESG factors that have influenced changes to credit ratings. ESMA’s assessment suggests that although there has been an increase in the overall level of ESG disclosures in CRAs’ press releases since the introduction of the Guidelines on Disclosure Requirements applicable to credit ratings in 2019, there is room for further improvement.

5.5.1Possible policy options:

A.Specify in the CRA legal framework that in the case of the incorporation of ESG factors in the creditworthiness assessment, CRAs must do this in a systemic way and develop relevant methodologies:

·Option 1 – via the targeted amendment of the CRA Regulation 124 and subsequent amedments of the delegated act on methodologies 125

·Option 2 – via the amendment of the delegated act on methodologies 126

B.Require that disclosures made by CRAs include a clear explanation as to whether and how ESG factors were considered within the CRA’s methodologies:

·Option 1 – via the amendments of Annex I to CRA Regulation and subsequent amendments of ESMA Guidelines,

·Option 2 – via the amendment of ESMA Guidelines.

5.5.2Analysis:

1)The main area that can be further improved relates to the rigorousness of the inclusion of ESG factors in CRA’s methodologies. The general principles regarding methodologies set up by the Credit Rating Agencies Regulation already apply to the incorporation of ESG factors in credit ratings, whereas precise details concerning the conditions which methodologies should meet are specified in the deledated act on the methodologies for CRAs 127 . Our assessment is that changes to the CRA Regulation would need to be accompanied by changes to the delegated act. It is also possible to amend the delegated act without amending the Regulation itself. Given time and resources needed to amend the Regulation and then the delegated, in our assessment it is more effective and efficient to amend the delegated act directly.

2)As regards the disclosure requirements, the Credit Rating Agencies Regulation includes a number of requirements in that respect 128 . The regulation requires that CRAs disclose information to the public on their methodologies, models and key rating assumptions which they use in their credit rating activities. With a view to ensuring transparency, disclosure of any material modification to the methodologies and practices, procedures and processes of credit rating agencies need be made prior to their coming into effect, unless extreme market conditions require an immediate change in the credit rating. Specific disclosure requirements are provided for in the Annex I to the CRAR Regulation. They have been further detailed in ESMA Guidelines. The CRA Regulation provides the Commission with the mandate to update the Annex I in view of the market developments.

5.5.3Conclusions

Changes to Annex I and ESMA Guidelines will be sufficiently effective to address remaining shortcomings in relation to disclosures and changes to the delegated act on methodologies – in relation to the incorporation of ESG factors in CRAs methodologies (for more explanation see Annex 8).

6.What are the impacts of the policy options and how do they compare?

The first part of this section analyses the impacts of the policy options on the regulatory treatment of ESG rating providers. The second part of this section analyses the impacts of the policy options on the transparency requirements for ESG ratings.

6.1Options on regulatory treatment of ESG rating providers

6.1.1Impacts of the Policy Options

Option 1: Industry-led Code of Conduct for operations of ESG rating providers

Under this option, the Commission would encourage the development and adoption by the industry of a voluntary code of conduct regarding operations and disclosures by ESG rating providers taking into account IOSCO Recommendations 129 .

Expected benefits: An industry-led code of conduct would be expected to partially address each objective: clarity on ESG ratings, and clarity on providers’ operations and prevention of risks of conflicts of interest. The extent to which each objective is addressed would depend on both the comprehensiveness of the code of conduct, and the extent to which providers adopt it. Ultimately, greater clarity on ESG ratings and on providers’ operations would benefit the customers of code-adherent providers. They would face lower costs for understanding the products and make more informed choices. In addition, this approach might allow greater flexibility to adapt to changing markets than regulatory options.

Potential drawbacks: This option would only partially address either objective, since there would be no incentive for the Code of Conduct to be as rigorous as required; Existing market pressures would determine the rigor and comprehensiveness of the Code, and those pressures have proven insufficient to address the problem. Moreover, the adoption of the Code of Conduct would be voluntary. Some providers might choose not to adopt it, or multiple industry Codes might arise. Both of these cases would undermine an increase in clarity across the market for users or rated entities.

Who is impacted and how:

·Users of ESG ratings: Users of providers who adopt the Code of Conduct may have lower due diligence costs when selecting providers. Reliable and transparent ESG ratings can contribute to reducing information asymmetry between companies and investors but also avoid important costs for users when they have to analyse a large number of companies directly in order to get an accurate understanding of the ratings 130 . The magnitude of this benefit is not possible to quantify given that each user may have different expectations, resources, and costs of due diligence, but it would be rather limited. Users would also benefit from increased transparency from at least those providers adopting the Code of Conduct, enabling more informed choices. Overall, limited potential benefits to users would be expected relative to the baseline.

·Rated entities: As with users, entities rated by, or wishing to be rated by, providers adopting the Code of Conduct would benefit from lower costs of due diligence when selecting providers. The magnitude of this benefit is not possible to quantify given that each rated entity has different expectations, resources and costs of due diligence. However, in case of a more rigorous Code of Conduct the benefit would be greater overall as it would likely meet the needs of more rated entities. Overall, limited potential benefits to rated entities would be expected relative to the baseline.

·ESG rating providers: ESG rating providers would be free to choose whether or not to participate in the development of an Industry Code of Conduct, and whether or not to implement it. Therefore, any costs would be voluntary. This option would be expected to result in minimal costs to ESG rating providers, arising from resources they may voluntarily dedicate to the industry development of a Code of Conduct, and from the implementation of the resulting standards if they adopt them. However, given that an industry-led Code of Conduct would by nature be expected to be less stringent than regulatory requirements, and compliance would not necessarily be externally verified, these costs would be lower than for any of the other options for this dimension. This option could also result in some limited benefits to providers stemming from the industry collaboration and sharing of best practices. Overall, a mixed effect of limited costs and indirect benefits to providers would be expected relative to the baseline.

·EU and national authorities: There would be minimal costs to EU authorities stemming from engagement with the industry regarding the development of the Code of Conduct. There would be no resources needed for supervision. At the same time, public authorities would not receive information for monitoring developments in the sector.

Stakeholder views: A small proportion of the responding ESG rating users and rated companies (7% and 11%, respectively) expressed support for this approach, 30% of ESG rating providers supported this measure, citing the advantage of limited market intervention. However, a larger proportion did not consider such an approach to be sufficient to address the lack of clarity on ESG ratings and lack of clarity and operational oversight.

6.1.2Option 2: Registration and light supervision

This option would mean the establishment by ESMA of a central registry of ESG rating providers. Under this option providers would be required to register at EU level as well as to provide basic information (entity name, domicile, registration number, scope of activities, type of services provided, type of business model, policies on the prevention of conflicts of interest, policies on ensuring the quality of ESG ratings etc.) and changes thereto. A simplified sanctioning regime would be applied by ESMA in case of non-compliance with the registration requirements.

Expected benefits: Enforced registration of ESG rating providers would be expected to partially address the objective of increased clarity on their operations. Providers would be required to provide basic information. Ultimately, greater clarity on providers’ operations would allow users to make more informed choices of provider and provide some level of reassurance that operations are conducted appropriately.

Potential drawbacks: This option would not fully address the objective of increased clarity on operations and the prevention of risks of conflicts of interest at ESG rating providers level since there would be no oversight on the operations after initial registration. Thus, aside from requiring information on quality assurance and conflict of interest management policies from providers, there would be no assessment of the sufficiency of such policies, nor of their continued adequacy over time. Moreover, there would be some cost to the supervisor when setting up the registration procedure and sanctioning regime, which would be passed onto providers as fees. There would also be some costs to providers when providing the necessary documentation for the registration process.

Who is impacted and how:

·Users of ESG ratings: Users would be expected to benefit from lower costs of due diligence when selecting providers with the operations and conflict of interest management that they deem appropriate. Reliable and transparent ESG ratings can contribute to reducing information asymmetry between companies and investors but also avoid important costs of analysing a large number of companies directly 131 . The magnitude of this benefit is not possible to quantify given that each user may have different expectations, different resources, and thus different costs of due diligence. Prospective users would also benefit from greater clarity as it would help them make more informed choices of provider. However, users would have limited assurance that providers’ operations are indeed sound and free of conflicts of interest, since there would be no ongoing oversight. Overall, limited potential benefits to users would be expected relative to the baseline.

·Rated entities: Similarly to users, rated entities would experience many of the same impacts as users of ESG ratings, as described above. However, as is the case with users, rated entities would have limited assurance that providers’ operations are free of conflicts of interest, since there would be no ongoing oversight or assessment of the sufficiency of providers’ governance. Their investors would also have only limited assurance that the ratings are free from conflicts of interest. Overall, limited potential benefits to rated entities would be expected relative to the baseline.

·ESG rating providers: This option would be expected to result in negligible or low one-off compliance costs to ESG rating providers, stemming from the registration process 132 . The costs from the registration procedure and sanctioning regime would be reflected in fees, which would be low 133 and proportionate based on provider size. This option could also result in some indirect commercial benefits from the positive perception of being a registered provider; however, such potential benefits are not possible to quantify. Overall, a mixed effect of limited costs in the short term and potential benefits in the long term would be expected relative to the baseline.

·EU and national authorities: There would be limited costs for public authorities stemming from the registration procedure and enforcement / sanctioning regime. These limited costs would mostly be passed onto providers as fees. As a result, public authorities would have basic information that can be used to monitor who is present in the market, but these would be also limited.

Stakeholder views: Most stakeholders responding to the targeted consultation (82%) expressed support for a registration or authorisation requirement. The proportion was very similar for users and for rated entities, while 50% of responding ESG rating providers supported the requirement.

6.1.3Option 3: Authorisation, principle-based organisational requirements and risk-based supervision

This option would entail ESG rating providers being authorized by ESMA subject to proportionate rules for their operations and ongoing proportionate supervision. The organisational requirements would be principle-based and focus on independence, preventing or mitigating conflicts of interest, data quality assurance, disclosure of key information about business models, and appropriate treatment of non-public information.

Expected benefits: This option would fully meet the specific objective of increasing the clarity of providers’ operations and ensuring the prevention of risks of conflicts of interest. Reliable and transparent ESG ratings can contribute to reducing information asymmetry between companies and investors but also avoid important costs of analysing a large number of companies directly 134 . This would lead to a reduction of due diligence costs for users and rated entities and enable them to make more informed choices of ESG rating providers. Providers have been asked to quantify this, and while noting that there would be a reduction of due diligence costs, they noted that it is difficult to estimate this reduction in advance. 135 Further clarity on the organisational policies of providers and the prevention or mitigation of conflicts of interests would also be beneficial to the market itself, by fostering increased trust, reliability and useability of ESG ratings providers, with proportionate supervision offering an additional layer of security for users of ratings.

Potential drawbacks: This option could entail relatively high costs for ESG rating providers (detailed below). The main driver of costs is mostly related to authorisation for both smaller and larger ESG rating providers. Further one-off adjustment costs are likely to arise in relation to the need to upgrade IT systems and strengthen internal procedures, but estimates are difficult to obtain. In terms of supervisory fees, they would be borne primarily by large providers (around 94% of the total). Smaller entities would bear only about 5% of the costs of supervision for ESMA. In order to prevent large cost impacts on smaller providers, the rate of supervisory fees would be very progressive, ranging from 0.37% to 2.4% of annual revenue. The ultimate impact of this increased cost on the market and availability of ESG ratings is not possible to quantify. The proportionality of this option and related requirements would need to be carefully ensured, and consider the impact on smaller players.

Who is impacted and how:

·Users of ESG ratings: Users would be expected to benefit from lower costs of due diligence when selecting providers given that there would be public information about business models, as well as regulatory oversight. The magnitude of this benefit is not possible to quantify given that each user currently faces different costs of due diligence. However, of all options considered in this dimension, the benefit would be greatest under this option, since it would provide the most reassurance to users regarding the soundness of providers’ operations. Overall, a net benefit to users would be expected relative to the baseline.

·Rated entities: Similar to users, rated entities would benefit from lower costs of due diligence when selecting providers; The magnitude of this benefit is not possible to quantify given that each rated entity currently faces different costs of due diligence. However, the benefit would be greater under this option than under all other options in this dimension, since it would provide the most reassurance to rated entities regarding the soundness of operations of providers. On the other hand, ESG rating providers may pass through part of their increased costs to rated entities in the form of higher prices, depending on the providers’ business model. Overall, a net benefit to rated entities would be expected relative to the baseline.

·ESG rating providers: This option would be expected to result in higher costs to ESG rating providers, from the combination of one-off costs of authorisation and ongoing costs of supervision (see Annex 3 and Annex 10 for further details). The costs stemming from initial authorisation would be around EUR 68 000 – 108 000 per provider, implying an expected total one-off cost of EUR 4 – 6.4 million 136 . Moreover, costs would arise from the introduction of ongoing supervision, which would lead to approximately EUR 6.7-10.6 million per year in total (from the provision of required documents and interaction with the supervisor during the supervisory review process). In particular, costs 137 for smaller providers would range from approximately EUR 61-82 thousand per year 138 , while costs for larger providers would range from approximately EUR 284-490 thousand per provider per year. Additionally, ESG rating providers would pay supervisory fees to finance the cost of supervision by ESMA. The fees would be proportionately distributed across providers based on their revenues and are expected to range between 0.37% of revenues for smaller market players and 2.4% for larger ones based on a preliminary assessment by ESMA. Both supervisory fees and interaction with the supervisor would hence be proportionate to provider’s size, mitigating the risk of disproportionate impact on smaller providers. The requirements and process for supervision will be further operationalised via Level 2 measures to be adopted by ESMA. ESMA will carry out a further analysis of the impacts of such level 2 measures during their development, which will give a more precise assessment of the costs. Other mitigating measures for smaller providers, such as transition period and possibility of the exemptions from certain organisational requirements that would not be proportionate to the nature, scale or business model of small provider, are described in Annex 5. There could be further adjustment costs depending on the provider’s existing structure and need to align further with the organisational requirements (e.g., separation of businesses and potential changes to internal governance). However, this option would also be expected to result in indirect commercial benefits to providers from the positive perception of being regulated and supervised, and from the expected growth of the market and greater confidence in ratings.

·EU and national authorities: ESMA would be tasked with granting authorisations and with supervising the sector on an ongoing basis, which would require further resources. The total annual cost increase is estimated at approximately EUR 5.7 million. This cost would however not be borne by EU budget and would instead be passed onto ESG rating providers as fees 139 . The authorisation process and ongoing supervision would at the same time allow public authorities to have a better overview of the sector and use such information in monitoring activities.

Stakeholder views: In addition to the support for a registration or authorisation requirement mentioned under the previous option, a large majority of stakeholders responding to the relevant question in the targeted consultation support a legislative intervention for ESG rating providers, including 81% of users of ESG ratings, 86% of rated companies and 62% of ESG rating providers. However, in targeted interviews providers have indicated their concerns about potential costs linked to this approach.

6.1.4Comparison of options

6.1.4.1Comparison with regards to effectiveness:

Option 1 (Code of conduct) will be only partly effective as it would not fully address the specific objectives of increased clarity on the characteristics of ESG ratings and on the operations of ESG ratings providers The extent to which it will meet the specific objectives will depend on both the comprehensiveness of the code of conduct, and the extent to which providers will adopt and implement it. One of the main reasons for a lack of effectiveness could be the voluntary nature of the code. Experience indicates that codes of conduct generally suffer from low effectiveness. It is also important to remember that there were already at least two industry attempts to develop a code of conduct which failed due to a low take-up.

Option 2 (registration and light supervision) will be more effective. It will meet partly the objective of providing more clarity of operations of ESG rating providers by requiring that providers register and disclose a certain level of information about their business models. Users or rated companies would be able to consult the registry to better understand providers’ operations. However, this option only provides for enforcement regarding registration requirements. The lack of clear organisational requirements and their enforcement may be ineffective in ensuring the prevention or mitigation of conflicts of interests. This will depend on the impact of higher business model transparency, the pressure exerted by market participants and the role of reputational effects.

Option 3 (authorisation, organisational requirements and supervision) will be most effective in reaching the objectives referred to above. This option would fully meet the specific objective of increasing clarity and ensuring that providers manage conflict of interests. Furthermore, oversight will help to improve the quality and reliability of ESG ratings. This would increase overall trust in ESG ratings and contribute to better allocation of ESG investment.

6.1.4.2Comparison with regards to efficiency (cost-effectiveness):

Option 1 (Code of conduct) would lead to minimal voluntary costs for ESG rating providers who choose to participate in the development and implementation of the Code of Conduct. Overall, this option would be somewhat efficient as it would entail minimal costs but obtain limited benefits.

Option 2 (registration and light supervision), would result in negligible to low compliance costs for all ESG rating providers to be registered. Overall, this option would be somewhat efficient, since despite slightly higher costs compared to Option 1, there would be a potential for slightly higher benefits.

Option 3 (authorisation, organisational requirements and supervision) would result in potentially significant costs to ESG rating providers particularly in the short term, from the combination of one-off costs of authorisation and ongoing costs of supervision. Additionally, in order to limit the negative impact on smaller providers, a number of mitigating measures as described in Annex 5. are foreseen Overall, this option may be less efficient than other options in the short term, but in the long term the benefits would be expected to outweigh the costs, since this option would also result in the highest benefits to rating users and rated entities, as well as providers themselves and society at large. It would lead cost savings from lower due diligence costs for users and rated entities. Increased quality and reliability would increase overall trust in ESG ratings and contribute to better allocation of ESG investment, thus serving the objectives of the EU Green Deal and various UN Sustainable Development Goals, to the benefit of society at large. It would also fuel growth of the ESG ratings market, benefiting providers themselves and potentially improving coverage of rated entities, including those who currently find it difficult or too costly to be rated.

6.1.4.3Comparison with regards to policy coherence:

Option 1 (Code of Conduct) would not contradict EU law but would only be partially aligned with the overall EU Sustainable Finance Strategy and the objective to bring further clarity on the operations of financial market participants. The EU has introduced rules on the operations of the vast majority of financial market participants (e.g., credit rating agencies, benchmark administrators, investors, banks, insurance companies, green bond verifiers) with the objective to ensure a capital market union and financial stability. Those pieces of legislation require the registration / authorisation of market participants in order to offer their services in the EU. They are also subject to rules regarding their operations, in particular for credit rating agencies, benchmark administrators and green bond verifiers, rules on their input data and transparency of methodologies. A code of conduct that could also be facilitated and supported by the Commission and followed and applied by market participants would provide additional transparency and clarity than what is currently the case but it would also come with a number of risks. More specifically, it is unsure what would be the take up of this code, and how many market players would adhere and abide by it and its rules. The absence of clear and strict requirements to be applied, including oversight, would also not achieve the full objectives of this initiative.

Option 2 (registration and light supervision) would not contradict EU law and would bring additional transparency and clarity on the market of ESG rating providers, but it would not follow the approach taken for other pieces of legislation on financial services e.g., credit rating agencies, benchmark administrators and green bond verifiers for what concerns in particular supervision. In particular, in those other relevant frameworks, financial market participants are subject to ongoing supervision in order to ensure that their activities respect established rules on an ongoing basis, not just at the time of registration/authorisation.

Option 3 (authorisation, organisational requirements and supervision) would be aligned with the approach taken for other financial market participants and relevant legislation, e.g. EU Benchmark Regulation, Credit Rating Agencies Regulation and EU Green Bond Standard proposal that provide for an ongoing supervision and a number of organisational requirements, processes and documents concerning governance structure to be complied with (e.g. rules on input data, adopt, implement and enforce adequate measures to ensure that the ratings issued are based on a thorough analysis, prevention and management of conflicts of interests, record keeping requirements). A number of ESG rating providers also have in their structure other activities such as benchmark administrators (e.g., MSCI), credit rating agency (e.g., Moody’s) or green bond verifiers (e.g., ISS) and some synergies can be expected. They also have experience with dealing with authorisation process and ongoing supervision, which could help reduce overall costs linked to the introduction of a regulatory framework on their ESG rating activity.

Table 1 below provides a high-level summary of how the previously described options compare

Table 1: Comparison of the policy options to reduce the vulnerability of the EU and its operators

	Effectiveness	Efficiency (Cost-effectiveness)	Coherence
Option 1 – Industry Code of Conduct for ESG rating providers	+/-	+	+/-
Option 2 – Registration and light supervision	+	+	+
Option 3 – Authorisation, principle-based organisational requirements and risk-based supervision	++	+/-	++

Legend: ++ = Positive + = Slightly positive +/- = Mixed effect

0 = no effect - = Slightly negative -- = Negative

Conclusion: Based on the above assessment and comparison of options, Option 3 is the preferred option. While Option 3 implies higher costs for the sector, it is expected to result in significantly higher benefits for users and broader society that justify its selection.

6.2Options on transparency requirements

6.2.1Impacts of the policy options

6.2.1.1Option 1 - Minimum transparency towards the public

This option would require all providers to disclose publicly key information about ESG rating characteristics and methodologies. Precise disclosure requirements would be developed in delegated acts, the scope of which would be framed in the legislation, however standardized disclosure templates would not be foreseen.

Expected benefits: This option would be expected to partially address the needs of users and rated entities for greater transparency on ESG ratings’ objectives, methodologies and underlying data. This would lead to a reduction of due diligence costs for users and rated entities and a more informed choice of ESG rating provider. Anecdotal evidence 140 suggests that an asset manager can require the work of 6-8 full time employees (FTEs) just to understand the methodologies and data processes behind an ESG rating service, a cost which would be expected to be reduced with minimum transparency requirements based on responses from asset managers. Similarly, anecdotal evidence 141 from rated entities suggests 2 or more FTEs are needed to deal with the rating processes of different providers, while greater transparency would allow the entity to make a more informed choice of provider and avoid multiple processes.

It would also improve the protection of investors against potential greenwashing, including on social and governance factors. This is expected to lead to better, science based, data-driven decisions that facilitate an orderly transition to a sustainable economy. Indirect benefits stemming from greater transparency would include an increase in ratings quality and fitness for purpose. These would increase overall trust in ESG ratings and contribute to better allocation of ESG investment, thus serving the objectives of the EU Green Deal and various UN Sustainable Development Goals, to the benefit of society at large. According to EFAMA statistics, by the end of Q1 2021, asset managers in Europe applied an ESG investment approach to approximately EUR 11 trillion of assets 142 . Even a small improvement in the allocation of such an amount of investments could have a positive impact on ESG objectives.

Potential drawbacks: This option may not fully address the objective of increased clarity across the market, since the limited disclosure requirements would not cover all relevant methodological information. As for rating providers, they would face moderate costs of compliance which are described below. There could be some impact of the costs of compliance on the market and availability of ESG ratings at least in the short term, but how much is difficult to quantify. In the medium term, however, greater confidence in ESG ratings and higher expected usage would be expected to lead to market growth and attractiveness.

Who is impacted and how:

·Users of ESG ratings: Users would be expected to benefit from lower costs of due diligence to understand ESG ratings and their methodologies. Anecdotal evidence outlined above demonstrates that FTE costs would be expected to decrease to some degree thanks to the requirements for minimum transparency and clarity. Users would also better understand the limitations of ESG ratings and thus use them appropriately. A large asset manager has indicated that this could also help when considering their clients’ questions/request or media attention focusing on ESG ratings. Prospective users would also benefit from greater transparency from providers, as it would help them make a more informed selection. On the other hand, ESG rating providers may pass on part of their increased costs to users in the form of higher prices, depending on the providers’ business model. Overall, a net benefit to users would be expected relative to the baseline.

·Rated entities: Similarly to users, (prospective) rated entities would benefit from lower costs of due diligence to understand and make informed choices regarding ESG ratings and their methodologies; the magnitude of this benefit is not possible to quantify given that each rated entity has a different maturity in the area, different resources, needs and costs of due diligence. Anecdotal evidence from rated entities suggests 2 or more FTEs are needed just to deal with the rating processes of different providers, while greater transparency could allow them to save up a part of these costs. On the other hand, ESG rating providers may pass through part of their increased costs to rated entities in the form of higher prices, depending on the providers’ business model. Overall, a net benefit would be expected relative to the baseline.

·ESG rating providers: This option would be expected to result in moderate costs to ESG rating providers, stemming from the implementation of the new disclosure requirements. The one-off costs due to collecting relevant data and describing characteristics and methodologies of ESG ratings and operations of ESG rating providers have been estimated in the range of EUR 3 700 – 5 000 per provider, and approximately EUR 220 – 295 thousand in total 143 . Ongoing costs related to annual updates of the information and compliance checks are expected in the range of EUR 6 500 – 16 200 per provider and approximately EUR 380 – 955 thousand in total. These costs would vary depending on the provider’s internal governance, product complexity, number of rating products and existing level of transparency. However, this option would also be expected to result in moderate indirect commercial benefits to providers from the greater transparency and the positive perception of compliance with regulatory requirements.

·EU and national authorities: Depending on the option chosen on regulatory treatment of ESG rating providers, there would be either limited or more thorough enforcement of rules on disclosures. This would require staffing on ESMA side, which is accounted for in the estimate under Option 3 on regulatory treatment 144 .Costs of supervision would be passed on to rating providers via supervisory fees, that would be adapted to their size and revenues.

Stakeholder views: Almost all stakeholders (98%) who call for legislative intervention in their reply to the targeted consultation have expressed support for minimum disclosure requirements. Variations among stakeholder groups are outlined below.

The vast majority (94%) of respondents believe there should be minimum disclosure requirements in relation to providers’ methodologies, but they are less aligned on standardisation of templates with only a slight majority (51%) supporting the latter. Users desire transparency on methodological items (covering both high-level and the individual rating level) such as whether it reflects risks or impact (or another factor), the materiality assessment for each sector and rated company, which metrics have been selected as relevant, and what weight has been attributed to each of these, whether AI has been used or not, the integration of national or company specificities, changes to rating methodologies, reasons for changes to ratings, whether controversies are included or not, whether analysis is backward-looking or forward-looking, etc.

Almost all responding rated entities (93%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, and the majority (56%) believe these should be done via standard templates. In terms of transparency, rated entities want similar disclosures as investors, but point out that if these are not made public, they need to be communicated at least to the rated entities themselves. Rated entities seem to expect more granular methodological disclosures than investors do.

The large majority of responding rating providers (88%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, but unlike the other respondents, only 38% respondents believe these should be done via standard templates. In terms of transparency, they seem to generally agree on the need for methodology disclosures – at least to users. One respondent cautioned that commercial secrets should not be made known to competitors as that would discourage innovation.

6.2.1.2Option 2 - Minimum transparency towards the public and more comprehensive disclosures to clients of ESG rating providers and rated companies

This option would require that providers disclose, in addition to the public disclosures under Option 1, additional granular methodological information to their subscribers and to rated companies.

Expected benefits: This option would achieve all of the benefits outlined above under the expected benefits for Option 1. It would address the specific objective to a greater degree, meeting most of the needs of users and rated entities for greater transparency on ESG ratings’ objectives, methodologies and underlying data. The higher granularity of disclosures, as evidenced in responses of asset managers, would lead to a considerable reduction of due diligence costs for users of ESG ratings and enable a more informed choice of ESG rating provider, as well as more efficient use of ratings as a result of better understanding of the links between methodologies and individual ratings.

Potential drawbacks: Under this option, rating providers would face relatively larger costs of compliance (described below). However, of all the costs assessed arising from this initiative, the costs of complying with transparency requirements would be the lowest for providers, as many of them already currently provide information to users in varying degrees and formats. The ultimate impact of those costs on the market and availability of ESG ratings is not possible to quantify. The proportionality of this option and related requirements is essential. The precise disclosure requirements would have to be carefully calibrated in Level 2 to minimise this impact on smaller providers. If providers were to exit the market, users and rated entities would also ultimately suffer from lower competition among providers.

Who is impacted and how:

·Users of ESG ratings: Users would be expected to significantly benefit from this policy option as they would better understand ESG ratings and their limitations. This will make them more likely to use ESG ratings more efficiently and select rating products best suited to their needs. Users also expect to save up costs they currently expend on due diligence to understand the offered ESG ratings and their methodologies. The magnitude of this benefit is difficult to quantify given that each user has different resources, needs and costs of due diligence. However, this benefit is expected to be rather large, as illustrated by the following hypothetical example: assuming that each asset manager in the EU would save just 0.5-1.3 FTEs due to the additional transparency, it would translate into cost savings of approximately EUR 110-290 million annually (more details available in Annex 3) 145 . The total savings would likely be higher considering other user groups would also benefit. Such benefit is expected to be significantly greater than under Option 1, as users report that they need more granular information on methodologies to be sufficiently sure in their choice and use of ESG ratings. Indeed, asset managers indicate that even when their providers make information available publicly, they still have to make further enquiries to gain sufficient understanding of the products. This additional need would be largely met with this policy option, so asset managers expect a significant reduction in time spent on due diligence and reaching out for more information from providers. On the other hand, ESG rating providers may pass through part of their increased costs to users in the form of higher prices, depending on the providers’ business model. Overall, a significant net benefit would be expected relative to the status quo.

·Rated entities: Similarly to users, (prospective) rated entities would benefit from lower costs of due diligence to understand ESG ratings and their methodologies and make informed choices; the magnitude of this benefit is not possible to quantify given that each rated entity has a different maturity in the area, different resources, needs and costs of due diligence. Anecdotal evidence from rated entities suggests 2 or more FTEs are needed just to deal with the rating processes of different providers, while greater transparency would allow the entity to make a more informed appropriate choice of provider and avoid multiple processes. Moreover, the benefit would be significantly would be greater under this option than under Option 1. On the other hand, ESG rating providers may pass through part of their increased costs to rated entities in the form of higher prices, depending on the providers’ business model. Overall, a significant net benefit would be expected relative to the status quo.

·ESG rating providers: This option would result in relatively higher costs of compliance to ESG rating providers than under Option 1, including both one-off costs (collecting relevant data and describing characteristics and methodologies of ESG ratings and of their operations) and ongoing costs (annual updates of the information and compliance checks) of disclosure requirements. The one-off costs have been estimated in the range of EUR 7 500 – 15 000 per provider, and approximately EUR 440 - 880 thousand in total 146 . The ongoing costs have been estimated in the range of EUR 13 000 – 29 000 per provider, and approximately EUR 770 thousand to 1.7 million in total. Stakeholders further observed that these costs may significantly vary depending on the required granularity of disclosures and the scope and number of products, sectors, entities, or instruments covered. In terms of positive impacts, this option would be expected to result in at least moderate indirect commercial benefits to providers from the greater transparency and the positive perception of compliance with regulatory requirements, to a greater extent than under Option 1.

·EU and national authorities: Depending on the option chosen on regulatory treatment of ESG rating providers, there would be either limited or more thorough enforcement of rules on disclosures. This would require staffing on ESMA side, which is accounted for in the estimate under Option 3 on regulatory treatment 147 . Costs of supervision would be passed on to rating providers via supervisory fees, that would be adapted to their size and revenues.

Stakeholder views: As explained under Option 1, the vast majority of stakeholders (98%) who call for legislative intervention in their response to the targeted consultation have expressed support for minimum disclosure requirements. Additional interviews supported that around a third of stakeholders who expressed an opinion on the content of such disclosure requirements called for very comprehensive and granular disclosures, some requesting fully public methodologies. For these respondents – and particularly asset managers and rated entities, Option 2 would likely be preferable to Option 1 as the former goes further in terms of the disclosures. Providers’ concerns on commercial sensitivity would also be addressed, as the more granular disclosures would only be required towards the client base and rated entities.

6.2.2Comparison of options

6.2.2.1Comparison with regards to effectiveness:

Option 1 (minimum transparency) would partially address the needs of users and rated entities for greater transparency on ESG ratings’ objectives, methodologies and underlying data. It would ensure the availability of general information allowing investors to understand and compare the objectives of ESG ratings provided by various providers. It would also improve the protection of retail investors against potential greenwashing. Retail investors and the public at large will be given explanations about the objectives and general characteristics of ESG ratings. It will not, however, meet the needs of subscribers and rated companies to have a more in-depth understanding of the methodologies and data used. Option 2 (minimum transparency to the public and additional disclosures to clients and rated entities) would fully address the specific objective by meeting the needs of users and rated entities. It will ensure that subscribers and rated companies will receive sufficient explanations about the methodologies and data used by ESG rating providers and consequently ensure that they take informed investment decisions.

6.2.2.2Comparison with regards to efficiency (cost-effectiveness):

Option 1 (minimum transparency to the public), would result in moderate costs for all ESG rating providers in order to meet the disclosure requirements. However, there would be direct benefits for rating users and rated entities, stemming from lower due diligence costs when selecting providers and more appropriate usage of ratings. Providers would also have some indirect commercial benefits from the additional transparency. Moreover, the ultimate result of better allocation of ESG investment, thus serving the EU Green Deal and various UN Sustainable Development Goals, would benefit society at large. Overall, this option is considered to be cost-effective.

Option 2 (minimum transparency to the public and additional disclosures to clients and rated entities), would result in potentially important costs for all ESG rating providers in order to meet the disclosure requirements, particularly in the short term. However, it would also result in the highest direct benefits for rating users and rated entities, stemming from lower due diligence costs when selecting providers, more efficient engagement with providers, more efficient investment processes and more efficient use of their data budgets. Providers would also have some indirect commercial benefits from the additional transparency, likely to a greater degree than under Option 1. Moreover, the ultimate result of better allocation of ESG investment, thus serving the EU Green Deal and various UN Sustainable Development Goals, would benefit society at large. This impact would be expected to be greater and also materialize quicker under Option 2 than under Option 1. Considering the size of the ESG investment ecosystem, this benefit should not be underestimated. Overall, this option would be less cost-effective in the short term than Option 1, but over the long-term benefits would be expected to outweigh the costs.

6.2.2.3Comparison with regards to policy coherence:

Option 1 (minimum transparency to the public) would be aligned with EU law, and follow the approach taken for disclosures related to financial institutions e.g., credit rating agencies, where the latter have to publicly disclose the methodologies, models and key rating assumptions it uses in its credit rating activities. It would also follow the approach taken in the Benchmark Regulation, where benchmark administrators should publish or make available (to the public) information on the key elements of the methodology, details of the internal review and the approval of a given methodology, as well as the frequency of such review, etc. This option would have to be limited to information that is necessary, not to hamper the business model / commercial activities from ESG rating providers. One should not have to disclose information that would directly have negative competitive effect or affect their intellectual property rights.

Option 2 (minimum transparency to the public and more comprehensive disclosures for clients of ESG rating providers and rated companies, organisational requirements and supervision) would not be contrary to EU law and would be aligned with the overall sustainable finance agenda and the objective to bring transparency and trust to financial markets, where market participants need granular information to take informed decisions.

Minimum disclosure requirements would not achieve that objective as it would not help in particular those professional investors and companies that need detailed information for their investment strategies, looking into risks/impacts but also opportunities. This option, also linking with considerations for ESG rating providers to report as to whether they have considered CSRD disclosure requirements for the creation of ESG ratings will help provide more comparability and trust to the market.

It could also be considered that differentiating the disclosures by the type of recipient (users / rated companies) would go beyond some legislations that generally limit the disclosures to the general public. However, given the horizontal nature and impacts of this initiative (see further interactions with other sustainable finance initiatives in Annex 4), the diversity and complexity of the market, that has different business models (user / subscriber pay and issuer pay), which is different from other financial services markets that generally have one type of offering (e.g. the market of credit rating agencies is issuer based only, hence disclosures are targeted to this type of client only), such differentiation in terms of who received what information is necessary. Users of ESG ratings have different needs from companies that are the subject of ratings. Similarly to the option above, this option would have to be limited to information that is necessary to inform market participants.

Table 2 below provides a high-level summary of how the previously described options compare

Table 2: Comparison of the policy options to reduce the vulnerability of the EU and its operators

Effectiveness

Efficiency

(Cost-effectiveness)

Coherence

Option 1 – Minimum transparency to the public

+/-

Option 2 – Minimum transparency to the public and more comprehensive disclosures for clients of ESG rating providers and rated companies

+/-

Legend: ++ = Positive + = Slightly positive +/- = Mixed effect

0 = no effect - = Slightly negative -- = Negative

Conclusion: Based on the above assessment and comparison of options, Option 2 is the preferred option.

7.Preferred option

7.1Summary of preferred option

On the basis of the comparison of options carried out in Chapter 6.2. the preferred option would combine Option 3 on authorisation, proportionate organisational requirements and risk-based supervision of ESG rating providers together with Option 2 on the transparency requirements.

The preferred options provide for the authorisation, proportionate organisational requirements and risk-based supervision of ESG rating providers and minimum transparency to the general public plus more granular information to users and rated companies.

The preffered option presents the most efficient and effective combination of options in relation to meeting the objectives of the inititative.

Key elements of the preffered options:

·authorisation

·ongoing supervision

·organisational requirements

·disclosure requirements

Authorisation is necessary to ensure that ESG rating providers meet a certain number of quality requirements (organisational requirements) before they can offer services to clients or rate companies under the scope of this initiative.

The authorisation in combination with the organisational requirements ensures addressing the problem about the lack of clarity about and control of operations of ESG rating providers and addresses the problem of conflict of interests in relation to ownership, product mix, separation of commercial and analytical teams, and governance and quality assurance. The authorisation, not combined with the organisational requirments, would not be effective in meeting the objective of ….

In order to limit the burden for regulated entities, organizational requirements will be principle-based. This will give the regulated entities the flexibility to adapt the requirements to the specificities of their organisations and should reduce the burden on small and medium-sized providers. In addition, we foresee a number of mitigating measures for smaller providers, described in Annex 5 “SME Test”.

Ongoing supervision requirements are necessary for ensuring that providers fulfil their obligations, both in terms of how they operate, whether they meet the organisational requirements, but also in terms of what they disclose to the general public as well as users.. Ongoing supervision is a necessary condition for the proper enforcement.

Disclsoure requirements: The effectiveness and impact of disclosure requirements depend on their enforcement and therefore its effectivenss depends on other key elements like n authorisation, organisational requirements and supervision of ESG rating providers.

Summing up, only the combination of all key elements will facilitate meeting the objectives of the initiatitve in the most effective and efficient way.

Supervision

ESG rating providers falling within the scope of this initiative would be subject to ongoing supervision by ESMA.

ESMA has adopted a risk-based and data-driven approach when conducting supervision and prioritises its supervisory activities according to the level of risk identified and outcomes it targets. A risk analysis is used to identify areas of possible non-compliance and to assess the potential significance of the issue. Risks are identified by collecting and assessing information from different sources like periodic and ad hoc notifications submitted by the entities, registration application files, meetings/calls with entities, data analysis, on-site inspections/investigations, etc. To identify key priority areas for supervision, ESMA considers the result of its supervisory risk assessment, and key market trends in the respective industries and capital markets more broadly.

The result of the risk analysis is a ranking / mapping of the supervised entities and the risks related to their functioning. The supervisory work programme is prepared on the basis of this risk analysis. The outcome of the risk analysis is taken into consideration when monitoring the work programme, as emerging risks may trigger deviations or adjustments to the initially planned work programme. ESMA can use different supervisory tools to address risks, e.g., engagement with supervised entities, the development of policy tools, conducting investigations and requesting supervised entities to implement a remedial action plan. Where breaches of the regulation have been identified, ESMA will adopt timely enforcement actions, that could range from the issuance of public notices to the imposition of fines and the withdrawal of registration/authorisation.

7.2Legislative approach

The legislation would lay down the organisational requirements, authorization process and transparency requirements to which ESG rating providers would be subject, as well as grant ESMA the necessary supervisory powers. The “level 2 measures” will operationalize the organizational and transparency requirements laid down by the level 1 measure, by specifying information needed for the authorisation process and further specifying disclosure requirements. ESMA will be responsible for the elaboration of these Level 2 measures and will be required to carry out a further analysis of the impacts of such level 2 measures, which will give a more precise assessment of the costs stemming from the level 2 measures.

7.3Combination of options, discarded at early stage

We have pre-analysed the possibility of alternative combinations of options on the regulatory treatment and transparency, and decided to present and analyse the most feasible and impactful ones in order to keep the structure of the Impact Assessment as simple as possible.

Option 1 (code of conduct) cannot be combined with continuous supervision, as codes of conduct are by nature voluntary. The positive impacts of the voluntary ESMA Guidelines on CRAs stem from the fact that ESMA Guidelines supplement existing level 1 and level 2 legislation as well as from the fact that ESMA has a range of supervisory tools in relation to CRAs. ESMA Guidelines are part of a comprehensive legal and supervisory framework. A soft approach towards ESG rating providers, either in the form of a recommendation or guidelines could not have similar effect as ESMA Guidelines supplementing level 1 and level 2 legislation.

As for option 2 on regulatory treatment, it foresees requirements in relation to registration, including the enforcement of those requirements. We could have combined it with Option 1 on transparency, however, given the limited scope of the enforcement foreseen (only in relation to the requirements for registration), it would not have the required impact since it would not be subject to ongoing supervision. We have therefore decided not to present such an option, for the sake of avoiding repetitive arguments.

7.4Impacts of preferred option

This initiative is expected to bring transparency and clarity to the operations of ESG rating providers hence benefitting users of those ratings but also rated companies and other interested stakeholders. It will contribute to the objectives of the EU Green Deal and sustainable finance strategy.

The initiative will have significant direct and indirect economic impacts on the relevant parties. First, ESG rating providers will be subject to disclosure and other operational requirements, implying costs. Then, ESG rating users and rated entities will benefit from more transparency on ESG rating characteristics and operations of ESG rating providers. Public authorities will also be impacted, in particular with the introduction of supervision of ESG rating providers, even if within existing bodies. These impacts have been described under each option in Chapter 6. This initiative can have a number of indirect impacts on other key economic, social and environmental factors, as further described below. As these impacts would not vary widely among the different options, nor be expected to be significant, for ease of reading they have been described here only once for the preferred option.

7.4.1Economic impacts

First, the initiative would be expected to have a positive impact on the functioning of financial markets and the conditions for investment, particularly in relation to ESG investing, considering the important role that ESG ratings play in this ecosystem. The preferred option should enable investors, and rated entities, to understand and make informed choices regarding ratings. Indeed, ESG ratings are expected to gradually increase in quality and fitness-for-purpose, thus enabling investors and rated companies to take informed decisions regarding ESG-related risks, impacts and opportunities. Users will also better understand the limitations of ESG ratings, so they use them appropriately. The higher trust in ratings and their higher fitness for purpose should lead to more efficient capital allocation and more sustainable investment. Additionally, the initiative is expected to benefit rating users and rated entities, by reducing the costs of gathering information or the need to use extra providers, thus decreasing their costs of doing business.

Then, the initiative may have an impact on the general competitiveness of the ESG ratings providers. The cost of doing business for these companies will likely increase, at least in the short term, due to additional costs to comply with requirements on transparency and operations, but increased trust in ESG ratings could enhance market growth eventually leading to more profitability. Similarly, the capacity to innovate in this sector may be temporarily reduced due to the increased cost of business, but the opportunities provided by a larger market are expected to incentivize greater innovation in the medium term.. Finally, the new requirements potentially have a bigger impact on smaller providers, with some adminstrative burden and costs of organisational changes but it could also give more visibility to their operations. The increased costs of providing ESG ratings may affect the market structure and share of individual providers, reinforcing consolidation trends It is not possible to estimate to what extent an increase in additional costs could affect the market shares of existing players, as it is not possible to disaggregate the effect of the new requirements from the market trends: in the last years we have seen quite a few mergers and acquisitions and they are likely to continue.

However, in order to mitigate impacts on smaller providers, this initiative will include a number of mitigating measures such as: transition period for small providers, adjustment of supervisory fees to the size of the provider, proportionate supervision and the possibility for smaller and innovative entities to ask for some exemptions from a wide range of internal organisational measures under certain conditions (please see Annex 5 SME Test).

The initiative may have an impact on providers operating from third countries as they would have to meet the minimum requirements to be able to provide their services in the EU. However, the preferred option would be origin neutral as it would apply equally to domestic and third country operators on this market. In this instance, it was not possible to aim for regulatory convergence or common regulatory approaches with other jurisdictions, since there are currently no other binding regulatory frameworks, only a set of recommendations from IOSCO (Annex 7). Hence, the initiative could potentially impact the competitiveness of European providers in non—EU markets, where they would need to compete with unregulated third country providers. Given the niche and mostly local nature of most (smaller) European providers, this impact is not expected to be significant.

Finally, the potential impacts of the initiative on the operation and competitiveness of SMEs have been discussed in detail in Annex 5.

7.4.2Impact on UN Sustainable Development Goals (SDGs)

This initiative is expected to directly or indirectly contribute to the achievement of all SDGs (with possible exception of SDG 16) and, in particular, to the achievement of SDG 12 (responsible production and consumption). Annex 3 provides more detail.

7.4.3Environmental Impacts

No direct environmental impacts and no significant harm, either direct or indirect, are expected to arise from the implementation of the preferred option. This initiative would be expected to produce indirect positive impacts on several environmental factors such as climate, use of resources, biodiversity, sustainable production and consumption, etc. This initiative is expected to increase confidence in ESG ratings and improve market functioning, which could lead to greater use of ESG ratings. Through more transparency on methodologies, competitive pressure is expected over time to make ESG ratings more relevant in capturing factors that mostly drive environmental performance of companies. As investors increasingly value and trust ESG ratings, companies will be also incentivised to improve both their sustainability transparency and performance.

7.4.4Social Impacts

As regards social impacts, similarly to indirect environmental impacts outlined above, this initiative would be expected to have indirect positive social impacts on working conditions, employment, health and safety. Additionally, the initiative should have a positive impact on governance and civic participation. Civil society organisations could use ESG ratings and would also benefit from greater transparency on the products and providers, as well as greater confidence in the ratings.

7.4.5Impact on fundamental rights

The preferred options respect the rights and principles set out in the Charter of Fundamental Rights of the European Union, in particular those in Article 16 (freedom to conduct a business). The free movement of persons, services and establishment constituting one of the basic rights and freedoms protected by the Treaty on the European Union and the Treaty on the Functioning of the European Union is relevant for this measure.

7.5Application of the ‘one in, one out’ approach and REFIT

A large part of the costs arising from this initiative can be categorised as administrative costs relevant for the “one in, one out” offsetting. Total recurring administrative costs subject to one in, one out offsetting are estimated in the approximate magnitude of EUR 7.5-12.3 million. Meanwhile, total one-off administrative costs subject to one in, one out offsetting are estimated in the approximate magnitude of EUR 4.5 – 7.3 million.

These include the following:

1.Costs of disclosures on methodologies and operations of ESG rating providers: collecting and disclosing data is expected to give rise to costs in the range of EUR 440 - 880 thousand one-off and EUR 770 thousand to 1.7 million on a recurring basis 148 .

2.Costs of authorisation: approximately EUR 4 – 6.4 million one-off cost for preparing the required documents.

3.Costs of complying with supervisors’ requests: approximate total magnitude of EUR 6.7 – 10.6 million per year on a recurring basis. This cost item is expected to be proportionate based on provider size (as confirmed both by ESMA and interviews with providers).

Further detail is available in Annex 3. There are no administrative cost savings, as it is a new legislation not amending previous EU rules. For the same reason, this is also not a REFIT initiative.

8.How will actual impacts be monitored and evaluated?

The Commission will ensure that the actions selected in this IA contribute to the achievement of the policy objectives with a blend of specific monitoring elements designed to measure efficiency in implementation and progress towards specific objectives, but also, other monitoring tools contributing to the general objectives. Those elements and tools are further described below. To monitor progress towards meeting the specific objectives, the Commission services will explore the possibility of organising periodic surveys of investors, companies but also of ESG rating providers.

ESMA, that under this initiative would be acting as supervisor of ESG rating providers, would be the legitimate actor to organise the surveys and take stock of the developments and highlight potential issues of concerns, also liaising with relevant national authorities in the Member State where ESG ratings are used and where ESG rating providers are located and have their operations.

The surveys would gather data on (1) users’ and rated entities’ perceptions of the evolution of information available, in particular with respect to the specific objectives identified in section 4 and whether those objectives are being met, and (2) the costs and benefits of ESG rating regulation and oversight from an ESG rating providers perspective. Such surveys will be dependent on the availability of financial resources.

Progress towards meeting the specific objectives can also be monitored through analyses of:

-the number of ESG rating providers providing services in the EU;

-the number of ESG ratings made available;

-the growth in use of ESG ratings by users and issuers;

-the number of ESG ratings having financial risks and/or impact objectives;

-information made available to the public, online, by ESG rating providers, in particular development of eventual booklets/information papers;

-the extent to which ESG ratings/scores that measure the same constructs vary between different rating agencies, and in particular whether divergence declines over time;

-whether correlation of ESG ratings with similar objectives has increased;

-amount of ESG investment based on ESG ratings;

-market share of ESG investment based on ESG ratings;

-the number of scandals linked to ESG ratings;

-Whether ESG ratings continue to suffer from biases (size, location, industry).

Monitoring progress towards meeting the general objectives is by definition considerably more complex, since it is methodologically difficult to distinguish the impacts of the proposed intervention on ESG ratings from other possible causes. Nevertheless, the Commission services propose to monitor progress regarding the general objectives by:

-Monitoring trends in investments in companies that carry out sustainable economic activities; 149

-Engaging with supervisors and other relevant stakeholders to assess whether concerns of greenwashing involving ESG ratings are being reduced.

It may also be possible to use the proposed surveys of investors, companies and ESG rating providers to gather evidence about whether stakeholders perceive that ESG rating providers are more accountable for their activities.

In addition, different stakeholders, including civil society organisations, rating agencies, business organisations and public authorities are likely to publish reports that monitor developments in this area, which will be a useful complement to the monitoring carried out by the Commission services.

The indicators above and surveys will help evaluating whether the preferred policy option is successful in achieving the specific objectives. These indicators will then serve as a basis for an evaluation that should be presented at the latest five years after the entry into force of the present initiative. The draft proposal will further contain a commitment to evaluate the impacts of the new legislative act. The Commission will start monitoring the implementation of the preferred policy option after the entry into force of the initiative.

Annex 1: Procedural information

·Lead DG, Decide Planning/CWP references

This Impact Assessment Report was prepared by Directorate C "Financial Markets" of the Directorate General "Directorate-General for Financial Stability, Financial Services and Capital Markets Union" (DG FISMA). The Decide Planning reference of the "Initiative on ESG Ratings" is PLAN/2021/12801. This initiative is part of the Commission’s renewed sustainable finance strategy adopted in July 2021 150 .

·Organisation and timing

Several services of the Commission with an interest in the initiative have been involved in the development of this analysis. Four Inter-Service Steering Group (ISSG) meetings, consisting of representatives from various Directorates-General of the Commission, were held in 2022.

The first meeting took place on 31 January 2022, attended by representatives from CLIMA, CONECT, COMP, ECFIN, EMPL, ENER, ENV, ESTAT, GROW, INTPA, JRC, JUST, MOVE, REGIO, TRADE, LS and the Secretariat General (SG).

The second meeting was held on 19 September 2022. Representatives from CLIMA, COMP, DEFIS EMPL, GROW, ENV, INTPA, JRC, JUST, TRADE, MARE, and the Secretariat General (SG) were present.

The third meeting was held on 17 October 2022 and was attended by ECFIN, CLIMA, EMPL, GRO, INTPA, JRC, BUDG, JUST, TRADE, CNECT, REFORM and the SG.

The fourth meeting was held on 10 November 2022 and was attended by ECFIN, CLIMA, COMP, ENV, REFORM, TRADE, ESTAT, RTD, SJ and the SG.

This was the last meeting of the ISSG before the submission to the Regulatory Scrutiny Board on 16 November 2022. The meetings were chaired by SG.

DG FISMA has considered the comments made by DGs in the final version of the IA. In particular, it has added a significant number of sources in the problem definition section to demonstrate market failures, further detailed disclosure requirements to which ESG rating providers would be subject to, clarified the scope of the initiative and the proportionality principle and application of the requirements towards smaller ESG rating providers. DG FISMA also integrated some additional considerations over the integration of ESG factors in credit ratings worthiness asessment. The discussion on possible options, analysis of impacts and the preferred option takes account of the views and input of different DGs.

·Consultation of the Regulatory Scrutiny Board

The Impact Assessment report was examined by the Regulatory Scrutiny Board on 14 December 2022. The Board gave a Positive Opinion with reservations.

·Evidence, sources and quality

The impact assessment draws on a combination of desk research, external studies, studies carried out on behalf of the Commission, meetings with stakeholders, a targeted public consultation, academic research papers and other. The material used has been gathered since the First Sustainable Finance Action Plan in May 2018. This material includes but is not limited to the following:

1.A study commissioned by DG FISMA, “Study on Sustainability Related Ratings, Data and Research” (consortium led by SustainAbility Institute, ERM) provides information about the trends and business models in the market for sustainability rating agencies and data providers, and issues in the market such as lack of transparency, biases, lack of engagement with companies, etc. 151

2.External studies: “Environmental, Social and Governance (ESG) Ratings and Data Products Providers”, IOSCO, November 2021 152 ; “ESG Investing and Climate Transition”, OECD, October 2021 153 , etc.

3.A Call for Evidence on Market Characteristics of ESG Rating and Data Providers in the EU conducted by ESMA. The call for evidence was published on 3 February 2022 with a deadline of 11 March 2022. The purpose was to develop a picture of the size, structure, resourcing, revenues and product offerings of the different ESG rating providers operating in the EU. The call was mainly addressed to three target groups: ESG rating providers; users of ESG ratings; and entities subject to rating assessment of ESG rating providers. ESMA published its findings in a letter to the Commission on 27 June 2022. 154

4.A call for evidence and an online targeted consultation conducted by DG FISMA. The call for evidence and consultation document were published on 4 April 2022 and the deadline for responses was 10 June 2022. A total 168 organisations and persons responded to the targeted consultation. Individual companies were the biggest single group of respondents (59%), followed by business associations (23%) and non-governmental organisations (9%). The remainder represented public authorities, academic and research institutions, trade unions and citizens. The consultation consisted of over 200 questions divided into 2 main sections; the ESG ratings section covered the use of ESG ratings, current dynamics and functioning of the market, and the need for and potential costs of EU intervention; the credit ratings section covered the extent of incorporation of ESG factors in credit ratings, the disclosures on ESG factors in credit ratings, and the need for and potential costs of EU intervention. A Feedback Statement providing an overview of the contributions to this public consultation was published on the Commission’s website on 3 August 2022. No campaigns or similar action with major impact on statistical results on the targeted consultation have been identified.

5.Bilateral meetings with various stakeholders between April and October 2022, including meetings with 14 different ESG rating providers, Meetings with users or associations representing, among others, rating users, meetings with other stakeholders such as academics, NGOs, public bodies and supervisory authorities.

We also analysed existing international recommendations and codes for ESG rating providers, such as the IOSCO recommendations. 168 organisations and persons responded to the targeted consultation supporting this initiative, mainly investors, ESG rating providers and listed companies.

The material used to inform this impact assessment comes from the market participants as well as other reputable and well-recognised sources such as academic studies. Findings were cross-checked across the different information sources to avoid biases caused by outliers or vested interests by authors.

·Glossary of companies’names mentioned in this Impact Assessment

Term or acronym	Meaning or definition
As you Sow	As you Sow, a non-profit organization, www.asyousow.org .
Bloomerg	Bloomberg LP, www.bloomberg.com .
CAER	Now ISS ESG Australia, www.issgovernance.com
Carbon4 Finance	Carbon4 Finance SAS, www.carbon4finance.com
CDP	CDP Worldwide, a non-profit organization, www.cdp.net
Ecodes	Fundacion Ecologia y Desarrollo, www.ecodes.org
EcoVadis	Ecovadis SAS, www.ecovadis.com
Ecovalores	Ecovalores, www.ecovalores.org
Environmental Finance	Environmental-Finance.com, published by Field Gibson Media Ltd, www.environmental-finance.com
FactSet	FactSet Research Systems Inc., www.factset.com
FTSE Russell	FTSE International Ltd., www.ftserussell.com
GES	GES International, now acquired by Sustainalytics
Greeneye	Greeneye, www.greeneye.co.il
Impak Finance	Impak Finance Inc., www.impakanalytics.com
Imug	Imug rating GmbH, www.imug-rating.de
Inrate	Inrate AG, www.inrate.com
ISS	Institutional Shareholder Services Inc., www.issgovernance.com
KOCSR	Korea CSR Research Services
Oekom research	Oekom Research AG, www.oekom.de
Moody’s	Moody's Investors Service, Inc., www.moodys.com
Morningstar	Morningstar, Inc., www.morningstar.com
MSCI	MSCI Inc., www.msci.com
Ninety One	Ninety One UK Limited, www.ninetyone.com
Refinitif	Refinitiv Limited and Refinitiv US LLC, www.refinitiv.com
RepRisk	RepRisk AG, www.reprisk.com
Russell Investments	Russell Investments Group, LLC., www.russellinvestments.com
ShareAction	Fairshare Educational Foundation (trading as ShareAction), www.shareaction.org
Solactive	Solactive AG, www.solactive.com
Sustainable Value Investors	Sustainable Value Investors, www.sustainablevalueinvestors.com
Sustainalytics	Sustainalytics, a Morningstar company, www.sustainalytics.com
Sustainometric	Sustainometric, www.sustainometric.com
S&P	Standard & Poor's Financial Services LLC, www.spglobal.com
Trucost	S&P Global Trucost, www.spglobal.com/esg/trucost
Vigeo Eiris	Now integrated by Moody’s ESG Solutions France SAS

Annex 2: Stakeholder consultation (Synopsis report)

1.Introduction

At an early stage, the preparation of this initiative was informed by a stakeholder consultation supporting the Renewed Sustainable Finance Strategy. The overall feedback from the Consultation on the Renewed Sustainable Finance Strategy showed a strong support for EU action on ESG ratings. The main challenges to be addressed, according to the stakeholders, include a lack of standardisation of ESG reporting, data and metrics, and a lack of transparency by ESG data providers regarding methodologies and aggregated underlying data used in their analysis. Concerns were also raised about potential conflicts of interest. 155

Following on from the Renewed Sustainable Finance Strategy, the Commission’s consultation strategy aimed to collect evidence and stakeholder views relating to the ESG ratings market. The main topics covered by this strategy were the usage of ESG ratings and dynamics of the market, the functioning of the market (and the nature and extent of existing problems), stakeholder views on potential intervention to remedy such problems, and potential costs of intervention for various stakeholders. More detailed information is provided below on the stakeholder groups that were consulted, the various consultation activities, and the results thereof.

2.Stakeholder groups

The main stakeholder groups concerned by this initiative are the following:

1.Users of ESG ratings, who can be further subdivided into:

oInvestors, who often use ESG ratings as part of their investment process, for instance for pre-selection of investments, or for weighting in a portfolio, etc.

oNon-investor companies, such as companies who want to assess the risks in their supply chains, or who want to assess and improve their own ESG performance. Note that in the latter case, this group overlaps to a certain extent with another stakeholder group, ‘Rated entities’.

oOther, e.g., civil society, public sector entities, academics, etc. who use ESG ratings for their research or other aims like tracking trends, calling for accountability.

2.ESG rating providers,

3.Rated entities, who have one or more solicited or unsolicited ESG ratings.

All of the above stakeholder groups were consulted in the preparation of this impact assessment.

3.Consultation activities and other sources

The following consultation activities were carried out to support this impact assessment.

1.A study commissioned by DG FISMA, “Study on Sustainability Related Ratings, Data and Research” (‘the Study’), in order to understand the state of the play of the sustainability-related products and services market. It established an inventory and classification of market actors, sustainability products and services available in the market and analysed the use and perceived quality of sustainability-related products and services by market participants. A consortium led by SustainAbility Institute, ERM, combined desk research with stakeholder engagement; they engaged with senior-level of the major sustainability-related product and service providers, asset managers, asset owners and other industry participants across 14 Member States. Market participants were invited to contribute to the study via structured interviews and through an online survey launched in April 2020. In total, 236 respondents provided input, 156 of which 198 through the survey and 38 through in-depth interviews. All the respondents were from market participants active in Member States and 125 of the respondents were domiciled in a Member State. All relevant stakeholder groups were covered by the Study. 157

2.A call for evidence and an online targeted consultation conducted by DG FISMA, in order to confirm the findings of the Study, and get detailed feedback on the issues in the market as well as understand the expectations from regulatory responses. The call for evidence and consultation document were published on 4 April 2022 and the deadline for responses was 10 June 2022. A total 168 organisations and persons responded to the targeted consultation. Individual companies were the biggest single group of respondents (59%), followed by business associations (23%) and non-governmental organisations (9%). The remainder represented public authorities, academic and research institutions, trade unions and citizens. Regarding the respondents’ role in the ESG rating / credit rating market, 28.6% respondents play a role as ESG rating users (investors), 42.3% as ESG rating users (non- investors), 19.6% as ESG rating users (others), 29.2% as rated companies, 12,5% as ESG rating providers, 5.4% as CRAs, 18.4% as credit rating users, 3% as auditors, 2.4% as supervisors and 39% as other. 158 A Feedback Statement providing an overview of the contributions to this public consultation was published on the Commission’s website on 3 August 2022. No campaigns or similar action with major impact on statistical results on the targeted consultation have been identified.

3.Targeted informal outreach in the form of bilateral meetings with various stakeholders between April and October 2022, in order to get in-depth feedback on issues in the market and the consequences of potential intervention. This outreach included:

oMeetings with 14 different ESG rating providers. This targeted outreach focused notably on costs and their proportionality, and a balance sample of small and large providers was targeted. Some of the stakeholders provided a confidential written follow-up based on an interview guide shared prior to the calls, which was used in the assessment of costs (as described in Annex 3);

oMeetings with users or associations representing, among others, users of ESG ratings. Directly and indirectly, over 20 asset managers have contributed;

oMeetings with other stakeholders such as academics, NGOs, public bodies and supervisory authorities.

The results of the various activities outlined above are described in the next section, under the relevant stakeholder groups.

4.Results of stakeholder consultations

Summary of call for evidence

25 responses were received to the call for evidence which summarized the context, the Commission’s initial understanding of the problems in the market, and the possible courses of action. 12 of the responses were from individual EU citizens, while the remainder were from companies and business associations (11) and NGOs and foundations (3). Several responses were found to be irrelevant or uninformative with respect to the issues at hand and are thus not considered further in this impact assessment.

Among the informative contributions, most were in favour of potential action at the EU level and highlighted some points that were echoed in the consultation results presented below. Briefly, these were: (i) the importance of comparable and reliable ESG data being used for ESG ratings and the expected improvement following the CSRD’s entry into force; (ii) the need for transparency of objectives, methodologies and data used by ESG rating providers; (iii) the need to avoid potential conflicts of interest at the level of ESG rating providers. Respondents also commented on a number of other aspects they considered problematic, such as the (lack of) dialogue between rated entities and the raters, the (insufficient) accounting for regional or other specificities of rated entities and the (lack of) appropriate materiality thresholds in assessments of specific ESG aspects. All contributions were taken into account in this impact assessment.

Analysis of consultation results by stakeholder group

The Study surveyed asset managers, asset owners, benchmark administrators and rated companies. Views of each stakeholder group were sometimes presented individually in the Study, and sometimes in aggregate. The Consultation took a similar but slightly more granular approach, separating the investor users, non-investor users, rated entities, providers and others. Stakeholder feedback collected via these two and other activities are presented below under the relevant heading based on the grouping of the Consultation. The Consultation results are not per se representative of the market, however this Impact Assessment is based on a large number of different sources, indicated in Annex I.

Views of ESG rating users (investors)

Dynamics of and problems in the market

The Study concluded that asset managers use sustainability-related data all across the investment decision process, including for portfolio selection, index construction, risk management, voting practices, and engagement with companies. In the survey conducted for the Study, and in the IRRI Survey 2019 159 , asset managers reported that they multi-source sustainability-related products and services from on average 2 rating and data providers. The Russell Investment 2020 survey 160 also confirmed this and added that increasingly asset managers combine external services with in-house views.

Asset owners surveyed for the Study also valued ESG ratings highly, although there was a wide variance in usage of these and other sustainability-related services (including the approach to in-house versus external) by asset owners of different sizes and types.

In the survey conducted for the Study, 75% of asset managers anticipated an increased demand for both in-house sustainability-related research capacity, and external sustainability-related data, rating and research products and services, at broadly equivalent levels. In the Study, rating users who identified as investors or benchmark administrators highlighted, among others, the need for greater transparency and focus on sector-specific material issues. Specifically, they noted that there is a lack transparency of the methodologies deployed and lack of understanding as to what the rating or data represents. In addition, investors called for data standardisation (lack of which can lead to bias against smaller companies), data quality and greater comparability among ratings.

In the Consultation, the conclusions of the Study were largely confirmed. First, all investor users who expressed an opinion expect the market to continue to grow, although 85% do not think the market is functioning well today. They consider that the market suffers due to the lack of transparency of providers’ operations, lack of explanation of what ratings measure, lack of common definitions and use of different terms for the same products, lack of comparability and reliability of ratings, and lack of supervision over the functioning of the market. Responding investors are divided on the current quality of ESG ratings, with around a third each considering the quality in the poor range, neutral and in the good range. However, almost all consider it an issue that providers do not communicate and disclose relevant underlying information, and the majority consider it an issue that there is a lack of transparency on the methodology and objectives of ratings. Indeed, almost all investors consider that there are significant biases in the methodology used by providers.

Respondents are more divided on whether providers using very different methodologies or having different objectives (i.e., assessing different sustainability aspects) constitutes a problem. However, 74% of investors agree on the fact that the current level of correlation among ratings assessing the same sustainability aspects is not adequate.

74% consider the market to be prone to potential conflicts of interest, with the main risk (among others) cited by almost all respondents being providers offering paid advisory services to companies they assess.

Over half of investors consider the fees charged by providers are not proportionate to the services provided. Moreover, 70% consider that the information provided about fees is not transparent or clear enough, calling for more information to be made available on: notice periods for contract changes (including fees), additional charges for data re-use, fees of each aspect of a service (e.g., per dataset and other product). A few go on to suggest that cost-based pricing mechanisms or FRANDT principles be imposed on providers, and that EU authorities collect and publish information on licensing practices of providers.

Need for intervention

A large majority of investors responding to the relevant question of the Consultation (69%) consider legislative intervention is needed to address the problems outlined above, including a supervision regime. Almost all responding investors (95%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, but they are more divided on whether these should be via standard templates, with only a slight majority (52%) supporting the latter. Investors desire transparency on methodological items (covering both high-level and the individual rating level) such as whether it reflects risks or impact (or another factor), the materiality assessment for each sector and rated company, which metrics have been selected as relevant, and what weight has been attributed to each of these, whether AI has been used or not, the integration of national or company specificities, changes to rating methodologies, reasons for changes to ratings, whether controversies are included or not, whether analysis is backward-looking or forward-looking, etc. Investors made the point that disclosures on these aspects could be similar to those made by CRAs. Investors that are not in favour of standardised templates for disclosing information consider that it may be premature to introduce a mandatory, close-ended template for providers as it is a new market. They also consider that a close-ended template would not fit given the diversity of rating products and that it would hinder innovation.

From the data side, investors are interested in the sources of data (e.g., public or not, voluntary or mandatory disclosures), the data collection processes, how missing data is dealt with and the methodology for estimation, the practice for updating data and revising historical data, data quality controls, etc. In relation to governance, internal control and management of conflicts of interest, respondents mention that providers should publicly disclose their policies, including the level of engagement with rated entities, and whether ratings are solicited or not. In order to increase reliability, investors would, among other measures, like to have more information on the qualification of analysts, shortcomings of methodologies, etc.

Two thirds support the same regime applying to all providers, regardless of size, and almost all support the same regime applying to all providers who rate European companies, regardless of the provider’s location (e.g., extra-EU). When asked what kind of supervision on ESG providers’ operations would be appropriate, many investors responded that it could be made similar to the supervision of CRAs, by ESMA.

Views of ESG rating users (non-investors)

Dynamics of and problems in the market

The relevant views of non-investor rating users on problems in the market were not described separately in the Study; thus, they have already been summarized in the previous sub-section. In the Consultation, the views of non-investor users were also mostly aligned with those of investor users.

First, almost all non-investor users who expressed an opinion expect the market to continue to grow, although 85% do not think the market is functioning well today. They consider that the market suffers from the same shortcomings as those identified by investor users, namely the lack of transparency of providers’ operations, lack of explanation of what ratings measure, lack of common definitions and use of different terms for the same products, lack of comparability and reliability of ratings, and lack of supervision over the functioning of the market. Responding non-investors are also divided on the current quality of ESG ratings, with around a third each considering the quality in the poor range, neutral and in the good range. However, the majority consider it an issue that providers do not communicate and disclose relevant underlying information, and that there is a lack of transparency on the methodology and objectives of ratings. Indeed, a large majority consider that there are significant biases in the methodology used by providers.

For non-investor users, the use of different methodologies and objectives by different providers seems to be a bigger problem than it is for investors. Further, 88% of non-investors consider the current level of correlation among ratings assessing the same sustainability aspects to be inadequate.

86% consider the market to be prone to potential conflicts of interest, with the main risk (among others) cited by almost all respondents being providers offering paid advisory services to companies they assess. However, unlike investors, non-investor users are divided on whether or not the fees charged by providers are proportionate to the services provided.

The majority do consider, nevertheless, that the information provided about fees is not transparent and clear enough, calling for more information to be made available on: notice periods for contract changes (including fees), additional charges for data re-use, fees of each aspect of a service (e.g., per dataset and other product), criteria for setting fees.

Need for intervention

A large majority of non-investors responding to the relevant question of the Consultation (83%) consider legislative intervention is needed to address the problems outlined above, including a supervision regime. Almost all responding non-investors (95%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, and differently from investor users, the majority of non-investor users (68%) support disclosures via standard templates. Non-investor users, similarly to investors, desire transparency on methodological items (objective of rating, whether it is absolute or relative, key assumptions, metrics and weights) and data collection (sources, exchange with companies, estimation). In order to increase reliability, investors would, among other measures, like to have more information on providers’ resourcing, time spent per company, shortcomings of methodologies, etc.

As for investor users, 60% of non-investor users support the same regime applying to all providers, regardless of size, and all support the same regime applying to all providers who rate European companies, regardless of the provider’s location (e.g., extra-EU). On the other hand, the remaining responding non-investors consider that a lighter regime for smaller providers is necessary to support opening up the market. Too strict rules for small players could result in market concentration of ESG rating, which may disadvantage issuers in terms of market power.

Views of ESG-rated entities

Dynamics of and problems in the market

In the Study, rated entities reported frustrations that providers do not gather and process data and information in a timely, reliable or efficient manner; that providers’ methodologies are opaque and do not sufficiently take into account company context; that providers make errors (and are slow to correct them) and that engagement with multiple providers is time-consuming. The majority of rated entities said sustainability exposures and practices are only moderately reflected by providers, and overall were frustrated by a lack of transparency and comparability across providers.

In the consultation, rated entities’ responses broadly confirmed the frustrations observed in the Study. They were also broadly aligned with users’ responses to the consultation, with the minor differences being mentioned in this section where relevant.

First, like investors, almost all rated entities who expressed an opinion expect the market to continue to grow, although 88% do not think the market is functioning well today. They consider that the market suffers due to the lack of reliability of ratings, lack of transparency of providers’ operations, lack of explanation of what ratings measure, lack of common definitions and use of different terms for the same products, and lack of supervision over the functioning of the market. On the current quality of ESG ratings, half of responding rated entities consider it good or very good. However, almost all consider it an issue that providers do not communicate and disclose relevant underlying information, and the majority consider it an issue that there is a lack of transparency on the methodology and objectives of ratings. Indeed, almost all rated entities consider that there are significant biases in the methodology used by providers.

For rated entities, as for non-investor users, the use of different methodologies and objectives by different providers seems to be a bigger problem than it is for investors. Also, for rated entities, the current level of correlation among ratings assessing the same sustainability aspects is even more problematic than for investors; 95% of rated entities consider it inadequate.

Similarly to investors, 78% of responding rated entities consider the market to be prone to potential conflicts of interest, with the main risk (among others) cited by almost all respondents arising where providers offering paid advisory services to companies they assess. Rated entities also consider it a risk when providers charge companies to see their own (rating) reports. However, unlike investors, rated entities are divided on whether or not the fees charged by providers are proportionate to the services provided.

Need for intervention

A large majority of rated entities responding to the relevant question of the Consultation (86%) consider legislative intervention is needed to address the problems outlined above, including a supervision regime. Almost all responding rated entities (93%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, and the majority (56%) believe these should be done via standard templates. In terms of transparency, rated entities want similar disclosures as investors, but point out that if these are not made public, they need to be communicated at least to the rated entities themselves. Rated entities seem to expect more granular disclosures about methodology than investors do, and a few also call for standardized information requests from rating providers (to reduce the burden on rated entities) 161 , enhanced engagement with rating providers or even an obligatory right of recourse allowing rated entities to respond to providers’ rating decisions. Similarly to investors, ESG-rated entities that are not in favour of mandatory templates consider that the ESG rating market is still developing, and so it may be premature to have a standardised format for disclosing methodologies. They consider that ESG rating providers should indeed disclose their full methodologies, but enforcing a standardised approach could reduce innovation and narrow the range of insight available to users of ESG ratings. In relation to conflicts of interest, various rated entities call for Chinese walls, legal entity separation or even prohibition against providers rating companies to whom they also offer advisory services.

Just over half of the responding rated entities support the same regime applying to all providers, regardless of size, while all support the same regime applying to all providers who rate European companies, regardless of the provider’s location (e.g. extra-EU). When asked what kind of supervision on ESG providers’ operations would be appropriate, many rated entities responded that a risk-based supervision by ESMA would be appropriate.

Views of ESG rating providers

Dynamics of and problems in the market

The Study concluded that ESG rating providers agree with other stakeholders (in particular investors) about potential conflicts of interest, the need for clearer definitions and consistent use of terminology, and the need for greater transparency of governance and methodologies. In addition, ESG rating providers complained that “companies do not publish sufficient reliable data to enable appropriate comparability and analysis, while at the same time asset managers demand increased breadth, depth, and quality of data.”

In the Consultation, ESG rating providers’ responses mostly confirmed the findings of the Study and were partially aligned with others’ responses; the main differences in views were on the extent of the existing shortcomings of the market. Like investors and rated entities, all rating providers who expressed an opinion expect the market to continue to grow, although 71% do not think the market is functioning well today. However, only a slight majority consider that the market suffers from a lack of transparency of providers’ operations. ESG rating providers seem to be concerned most with the lack of clear explanation of what ratings measure, and lack of supervision/enforcement over the functioning of the market. On the current quality of ESG ratings, just over half of responding rating providers consider it good or very good. However, the majority consider it an issue that providers do not communicate and disclose relevant underlying information, and that there is a lack of transparency on the methodology and objectives of ratings. Indeed, most providers consider that there are significant biases in the methodologies.

Unlike other respondents, most ESG rating providers consider the level of correlation between ratings assessing the same sustainability aspects to be adequate.

Similarly to other respondents, 75% or responding providers consider the market to be prone to potential conflicts of interest, with the main risk (among others) cited by all respondents arising when providers offering paid advisory services to companies they assess. On the other hand and as may be expected, rating providers are mostly not concerned by the fees charged in proportion to services offered. Need for intervention

A majority of rating providers responding to the relevant question of the Consultation (62%) consider legislative intervention is needed to address the problems outlined above, but (unlike the other respondents) they are divided on whether there should be a authorisation regime. Some ESG rating providers consider that authorisation would be appropriate to ensure the quality of their ESG ratings, whereas others consider that EU authorisation would be costly and could slow the development of the market. The large majority of responding rating providers (88%) believe there should be minimum disclosure requirements in relation to providers’ methodologies, but, again unlike the other respondents, 50% of rating providers believe these should not be done via standard templates. In terms of transparency, providers seem to generally agree on the need for methodology disclosures – at least to users, but one cautions that commercial secrets should not be made known to competitors as that would discourage innovation. Another provider calls for a differentiation between the regimes governing ratings and scores. In relation to conflicts of interest, rating providers were rather strongly in support of various measures, from disclosures on conflict-of-interest prevention to mandatory separation of activities to even prohibition of the ‘issuer-pays’ business model.

Unlike the other respondents, 39% of ESG rating providers responding to the relevant question in the Consultation (and unsurprisingly, many of the smaller providers) were in favour of a lighter regime tailored to smaller providers. The responding providers listed several barriers to entry for smaller providers, such as the initial investment needed, difficulty collecting all relevant data, the large coverage needed to be competitive, the switching costs and market recognition/client base of incumbents, the difficulty to differentiate due to a lack of transparency in the market, and the fact that clients (investors) themselves are very cost-conscious due to their own decreasing margins.

However, similarly to other respondents, providers do not support tailoring the regime based on the provider’s location (e.g. lighter regime for extra-EU providers). Also similarly to other respondents, providers gave the CRA regulation as an appropriate example of a supervision regime.

Other views

We also examined the views expressed by respondents to the consultation not falling into the stakeholder groups above. The views were in general aligned with those of the main stakeholder groups, and as such are not repeated here.

Annex 3: Who is affected and how?

a)Practical implications of the initiative

The preferred option would primarily entail obligations for ESG rating providers to:

(I)notify ESMA and apply for authorisation to continue commercializing their ratings in the EU;

(II)meet minimum organisational requirements for an authorisation (e.g. if the business does not already meet them, this requires putting in place relevant policies and procedures and ensure separation of business activities);

(III)publicly disclose key information about ESG rating characteristics and methodologies and provide additional granular disclosures to their clients and/or rated entities;

(IV)provide information required by the supervisor during supervisory review.

The preferred option also tasks ESMA with processing requests for authorisation and supervising the market.

b)Summary of costs and benefits

I. Overview of Benefits (total for all provisions) – Preferred Option
Description	Amount	Comments
Direct benefits
Reduction of due diligence costs for rating users and rated entities	This will be reflected in fewer resources needed by users or rated entities to get the same level of understanding of the ESG ratings in the market and their methodologies, as well as clarity on the sound operations of the ESG rating provider. It is challenging to estimate the size of this impact, but we can consider the following hypothetical example: assuming that each asset manager in the EU would save 0.5-1.3 FTEs due to the additional transparency, it would translate into cost savings of approximately EUR 110-290 million 162 . The total savings would likely be higher considering that other user groups would also benefit.	Who benefits: ESG rating users (investors, companies and other users), rated entities Users and rated entities will have an improved understanding of the ESG rating characteristics and their methodologies, and of the operations of ESG ratings providers.
More informed choice of ESG rating provider and more effective engagement with ESG rating provider	Not possible to quantify but can be partially covered by estimate in first row; part would be in addition.	Who benefits: ESG rating users (investors, companies and other users), rated entities Differences between ESG ratings (objective, factors/KPIs considered, weights, etc.) become more explicit and users – or potential users - will be better able to compare rating products of different ESG rating providers, thus being able to make more informed choices of the products best suited for their purposes. Similarly, entities wishing to be rated can make more informed choices of rating provider. In terms of operations, users and rated entities can select the ESG rating providers with the business model and management of conflict of interest that they deem sufficient. Users and rated entities can also engage with ESG rating providers more effectively, being able to challenge certain methodological choices as needed. Users would better understand the limitations of ESG ratings and increased granularity of disclosures would allow them to prioritise additional analyst focus into cases where they consider data quality to be low. This would therefore increase efficiency of their investment processes.
Indirect benefits
Increase in quality of ESG ratings	Not possible to quantify directly but see final row for size of investment pool which could benefit.	Who benefits: ESG rating users (investors, companies and other users), rated entities Quality of ESG ratings is expected to gradually increase as a result of greater oversight and transparency, as well as more effective engagement of users and rated entities with ESG rating providers. Market pressures would be expected to lead to a reduction in rating biases and in risks of conflict of interest. Further, users such as asset managers would make more efficient use of their external data budgets by selecting the right providers, and even in some cases using fewer providers than they otherwise do to ensure benchmarking for quality. The precise savings would depend on each user.
Higher fitness for purpose	Not possible to quantify but see final row for size of investment pool which could benefit.	Who benefits: ESG rating users (investors, companies and other users), rated entities ESG ratings are expected to gradually become more fit for purpose to enable investors and rated companies to take informed decisions as regards ESG-related risks, impacts and opportunities. Users will also better understand the limitations of ESG ratings so they use them appropriately. A deeper understanding of the comparability and differences that lead to lack of correlation between different providers, could potentially expose the redundancies in using multiple providers, or conversely justifying the use of different providers in different contexts. This in turn would help with efficiency in external data budgets of users.
Higher trust in ESG ratings	Not possible to quantify, but see final row for size of investment pool which could benefit.	Who benefits: ESG rating users (investors, companies and other users), rated entities, society at large Higher (perceived) quality, reliability and accuracy are expected to lead to higher trust in ESG ratings, which is also ultimately expected to fuel growth of the ESG ratings market. The growth in the market could also lead to an increase in coverage of rated entities, including those who currently find it difficult or too costly to be rated. 163
More sustainable EU economy	According to EFAMA statistics, by the end of Q1 2021, asset managers in Europe applied an ESG investment approach to circa EUR 11 trillion of assets. An improvement in allocation of these EUR 11 trillion could have a significant positive impact on ESG objectives.	Who benefits: society at large The higher trust in ratings and their higher usefulness/fitness for purpose, could lead to better allocation of capital with regards to its contribution to ESG objectives or avoidance of ESG risks and more sustainable investment. Similarly, as investors increasingly value and trust ESG ratings, this would be expected to encourage rated entities to adopt sustainable practices. Moreover, with the greater transparency of rating methodologies, rated entities would be enabled to do so.
Administrative cost savings related to the ‘one in, one out’ approach*
n/a	n/a	n/a

II. Overview of costs – Preferred option
			Users and rated companies		ESG rating providers		Supervisors
		One-off		Recurrent	One-off	Recurrent	One-off	Recurrent
Authorisation, principle-based organisational requirements and risk-based supervision	Direct adjustment costs	n/a		n/a	One-off adjustment costs may arise e.g., due to the need to upgrade IT systems or strengthening of internal procedures and familiarisation with the new rules (e.g., legal and consultancy costs). Data that would allow to provide an estimate were very limited and the impact would greatly depend on the readiness and complexity of the provider.	Recurring adjustment costs may arise, e.g. running a strengthened internal control and compliance function and maintenance of IT systems. Data that would allow to provide an estimate were very limited and the impact would greatly depend on the readiness and complexity / governance of the provider. Possible further cost related to changes to the business models, where ESG rating providers do not sufficiently separate business activities or do not have policies in place against conflict of interest. These would be very provider-specific and we do not have sufficient data to estimate the size of impact or number of providers impacted.	n/a	n/a
	Direct administrative costs	n/a		n/a	Applying for authorisation: preparing and submitting relevant documents would imply an approximate magnitude of costs of EUR 68 000 – 108 000 per provider and EUR 4 - 6.4 million in total.	Costs are expected to arise from ongoing interaction with supervisor and responding to their information requests and estimated at an approximate total magnitude of EUR 6.7 – 10.6 million 164 . Costs of interaction with the supervisor are expected to be proportionate between entities, as risk-based supervision focuses more effort on larger players in a given market 165 . .	n/a	n/a
	Direct regulatory fees and charges	n/a		n/a	n/a	Supervisory fees – sliding scale ranging between 0.37% of revenues for smallest providers and 2.4% of revenues for the largest providers 166 .	n/a	n/a
	Direct enforcement costs	n/a		n/a	n/a	n/a	Costs for ESMA to assess applications for authorisation. (reflected in the supervisory fee detailed above)	Ongoing supervision of ESG rating providers by ESMA will imply further resource needs. The total annual cost increase is estimated at approximately EUR 5.7 million annually 167 . Such costs would be fully financed through supervisory fees paid by ESG rating providers (detailed under “Direct regulatory fees and charges”).
	Indirect costs	n/a		Possible pass-through of fees and other recurring costs to users of ESG ratings or rated entities, depending on business model.	n/a	n/a	n/a	n/a
Minimum transparency towards the public and more comprehensive disclosures to clients of ESG rating providers and rated companies	Direct adjustment costs	n/a		n/a	Part of the potential upgrade of IT systems (see above) may also be attributed to disclosure obligations. 168 Providers were not able to clearly separate the two costs. Further costs would stem also from the need to become familiar with the new rules (e.g. legal and consultancy costs).	Part of the strengthening of the compliance function or IT maintenance (see above) may also be attributed to disclosure obligations. 169 Providers were not able to clearly separate the two costs.	n/a	n/a
	Direct administrative costs	n/a		n/a	One-off cost of disclosing information: collecting relevant data and describing characteristics and methodologies of ESG ratings and operations of ESG rating providers. If relevant policies/frameworks do not exist internally, these may have to be prepared from scratch. Total estimated in the approximate range of EUR 7 500 – 15 000 per provider and EUR 440 - 880 thousand in total 170 .	Ongoing cost of disclosures: annual updates of information on ESG rating and operations of ESG rating providers and related compliance checks. Estimated in the approximate range of EUR 13 000 – 29 000 per provider and EUR 770 thousand to 1.7 million in total.	n/a	n/a
	Direct regulatory fees and charges	n/a		n/a	n/a	n/a	n/a	n/a
	Direct enforcement costs	n/a		n/a	n/a	n/a	n/a	Additional resource needs for ESMA in order to enforce the rules on disclosures. There would be part of the estimate presented above and are paid for by the supervisory fees detailed further above in the table)
	Indirect costs	n/a		Possible pass-through of disclosure costs to users of ESG ratings or rated entities, depending on business model.	n/a	n/a	n/a	n/a
Costs related to the ‘one in, one out’ approach
Total	Direct adjustment costs	n/a		n/a	n/a	Adapting business models to comply with rules to avoid conflict of interest and other obligations.
	Indirect adjustment costs	n/a		Possible pass-through of costs by ESG rating providers to users of ESG ratings or rated entities.	Possible further costs related to update of IT systems or organisational changes.	Possible further costs related to strengthened internal control and compliance function and maintenance of IT systems.
	Administrative costs (for offsetting)	n/a		n/a	One-off cost in the approximate magnitude of EUR 4.5 – 7.3 million. These consist of the one-off cost of disclosures: approximate total of EUR 440 - 880 thousand and the cost of applying for authorisation: approximate total of EUR 4 - 6.4 million.	On-going administrative costs in the approximate magnitude of EUR 7.5-12.3 million. These consist of the on-going cost of disclosures (approximate total of EUR 770 thousand to 1.7 million) and costs of ongoing interaction with supervisor (approximate total of EUR 6.7-10.6 million).

c)Relevant Sustainable Development Goals

III. Overview of relevant Sustainable Development Goals – Preferred Option
Relevant SDG	Expected progress towards the Goal	Comments
SDG 12 – Responsible production and consumption Specifically, item 12.6: Encourage companies, especially large and transnational companies, to adopt sustainable practices and to integrate sustainability information into their reporting cycle.	This initiative would be expected to encourage and enable companies to adopt sustainable practices. The usage of ESG ratings is expected to increase, as noted in the evidence collected for this impact assessment. The objective of this initiative is to increase confidence in ESG ratings and improve the functioning of the market. This could lead to even more widespread use of ESG ratings. As investors increasingly value and trust ESG ratings, companies will also be incentivized to improve both their sustainability transparency and performance.	The UN Indicator for this SDG is 12.6.1 Number of companies publishing sustainability reports. While this initiative could indirectly lead to an improvement of this indicator for the reasons mentioned, it would be very difficult to isolate its effect from the other regulatory initiatives (in the EU and across the globe) directly targeting sustainability reporting (such as the CSRD, discussed in Annex 4). Thus, an attempt to quantify this specific impact of this initiative is not considered feasible or proportionate.
All SDGs (with possible exception of SDG 16, as it is unlikely that this initative will help to improve institutions and access to justice ).	All SDGs are expected to be served (indirectly) through the general objective of this initiative. The general objective is to ensure that ESG ratings enable investors and rated companies to take informed decisions as regards ESG-related risks, opportunities and impacts and foster trust and confidence in the operations of ESG rating providers. Since ESG ratings are used by investors for the allocation of funds and by some companies to analyse their ESG performance and take corrective actions, this, in turn, will contribute to reaching a wider objective of meeting of European Green Deal and UN Sustainable Development Goals. Some ESG ratings or scores also assess alignment with SDGs and greater trust and transparency on these ratings could encourage investment into progress on SDGs.	All SDGs are impacted in some way by at least one of the E, S and G factors; thus, more efficient ESG investment would be expected to contribute to progress in all areas. However, this contribution would likely only be incremental, and an attempt to quantify it is not considered feasible or proportionate.

Methodology for estimation of costs

Cost estimates in this impact assessment were obtained through the targeted consultation and subsequent interviews with ESG rating providers and subject to further plausibility checks and assessment using contextual information collected in this process 171 .

With regards to the costs of disclosures, respondents to the targeted consultation estimated the number of hours of work that making such disclosures would require, both on one-off and recurring basis. This was corrected for outliers and then multiplied with data on non-wage labour costs 172 , including 25% overheads, based on Eurostat Structure of earnings survey and Labour Force Survey. Considering a rather large variation between the individual responses, a range was constructed around the estimate, reflecting uncertainty around some of the estimates and actual granularity of disclosures which will be specified further through delegated acts. The estimate from the consultation was further checked with ESG rating providers during subsequent interviews, which broadly supported the range, but highlighted dependency on the granularity 173 . The figures have been adjusted to reflect a business as usual (BAU) factor of 30%, based on stakeholder replies from the consultation and contextual information which pointed at higher than originally considered BAU. However, the actual BAU factor would vary between providers and their business lines, depending notably on what internal procedures are already in place.

The cost estimate for the more comprehensive disclosures is built on further extrapolation from the estimate for Option 1 based on the qualitative and limited quantitative information provided in interviews with ESG rating providers on the dependence on granularity of disclosures. While evidence was more limited, providers broadly indicated that more granular disclosures would be approximately 2-3 times more expensive than more limited disclosures. It is safe to assume that there would be some overlaps in the data updates between the two sets of disclosures, which is reflected in multiplying the less granular estimate by two for the ongoing disclosures.

Despite similar effort for collecting cost information, both through a targeted consultation and targeted interviews, responses (especially quantitative) regarding cost of applying for an authorisation and supervision were rather limited in the consultation and in the follow-up interviews. Moreover, figures varied significantly. For the cost of ongoing interaction with the supervisor, both information from ESMA and from the stakeholders reflected the expectation that intensity of supervision would be proportionate to entity size and hence should not significantly negatively impact smaller providers. Since most stakeholders who provided quantitative information on this element typically referred to the number of FTEs (“Full Time Equivalent”) in a compliance department, which is assumed to cover both work on disclosures and interaction with the supervisor, the estimate for the latter was obtained by subtracting the cost of disclosures (as estimated above) from the cost of running a compliance department to address this attribution issue. No BAU factor was applied to the cost of interaction with the supervisor, as this cost is assumed to fully arise due to the new legislation.

With regards to costs of registration vs. authorisation, a limited number of responses had to be manually re-classified 174 based on the contextual information from the interviews, which pointed at overall low cost of registration (some providers stating it would be negligible if not requiring setting up an entity in the EU and others pointing at costs of thousands or lower tens of thousands EUR). Estimates also varied between smaller providers, who had a tendency to estimate costs as lower, and larger players who expected higher costs of applying for an authorisation. A BAU factor of 20% is applied to gathering information and providing documents required for an authorisation process, considering that a vast majority of such costs would be due to this new obligation but some of the documents (e.g. internal policy documents) would be in place at least for a certain share of ESG rating providers.

The size of expected resource needs for ESMA and corresponding supervisory fees for the sector that would cover these costs were estimated by ESMA and made more precise as a result of deliberations between ESMA and DG FISMA about the intensity of supervision and other key parameters. The size of the budget and fees has been calibrated by ESMA based on the number and size distribution of ESG ratings providers. To derive the range of supervisory fees, ESMA used the information on revenues from the sale of ESG ratings, where these were available. Where not available, ESMA has extrapolated the number of entities in each size bucket proportionately (in the absence of information that would suggest a different distribution.

Annex 4: Interactions with other sustainable finance legislation and initiatives

In addition to the CSRD, SFDR and the EU taxonomy Regulation, this initiative also interacts with the work on Green Bonds Standards and EU climate benchmarks.

Title	Corporate Sustainability Reporting Directive (CSRD)
Legislative or non-legislative?	Legislative
Brief description of policy or legislation	On 21 April 2021, the Commission adopted a proposal for a Corporate Sustainability Reporting Directive (CSRD), which would amend the existing reporting requirements of the Non-Financial Reporting Directive. The proposal A.extends the scope to all large companies and all companies listed on regulated markets (except listed micro-enterprises) B.requires the audit (assurance) of reported information C.introduces more detailed reporting requirements, and a requirement to report according to mandatory EU sustainability reporting standards D.requires companies to digitally ‘tag’ the reported information, so it is machine readable and feeds into the European single access point envisaged in the capital markets union action plan The CSRD was agreed by co-legislators in June 2022 and should be published by the end of 2022 in the Official Journal. The CSRD envisages the adoption of EU sustainability reporting standards. The draft standards are to be developed by the European Financial Reporting Advisory Group (EFRAG). The standards will be tailored to EU policies, while building on and contributing to international standardisation initiatives. The first set of standards are to be adopted by June 2023.
Interaction with ESG ratings	The main element from the CSRD is the creation of mandatory EU sustainability reporting standards. Standardising information to be disclosed by large companies should help fill gaps in data availability, improve the quality but also reliability of corporate sustainability data. Since all ESG rating providers use some form of public data for their analysis, the standardisation of the latter should have a positive impact on the overall quality and reliability of ESG ratings, which cannot be reliable if the underlying data is of poor quality or is unavailable. ESG rating providers, given the availability of more and better data may also be less prone to using other sources of data that are difficult to compare and are by nature less reliable. The impact on correlation amongst ratings is however harder to predict. According to a research paper published in 2021, “greater ESG disclosure actually leads to greater ESG rating disagreement”. 175 In particular, ESG rating providers appear to disagree most on how to assess ‘outcome metrics’ disclosed by firms. As such, it is possible that with the advent of reporting required by the CSRD, divergence in ESG ratings increases at first. However, as the authors themselves note, this dynamic may be related to the early phase of ESG disclosures, when there is little consensus among analysts on ESG performance metrics and may neutralize or even change signs (i.e. greater disclosure leads to lower divergence) in the future. In any case, greater disclosure reduces the issues around data collection and divergence of underlying data used by ESG rating providers. Lastly, given that more companies will be required to report under the CSRD than required by the NFRD, there may be more demand for ESG ratings on and by these companies that may want to have some external party look at new possible risks and opportunities.

Title	European Single Access Point (ESAP)
Legislative or non-legislative?	Legislative and non-legislative
Brief description of policy or legislation	The ESAP will provide seamless, EU wide (cross border) access to information published in relation to financial services, capital markets and sustainability published by entities or about entities operating in the financial sector. For this, the ESAP legislative package lays down the foundations of ESAP by organising the collection of information by collection bodies, addressing metadata and IT formats, and mandating minimum requirements in terms of quality, functionalities, governance and monitoring. The Commission proposed to make ESMA responsible for the governance of ESAP. Where appropriate, ESMA would work in close cooperation with the European Banking Authority (EBA) and the European Insurance and Occupational Pensions Authority (EIOPA), consult the national collection bodies (NCAs, OAMs) and the Securities and Markets Stakeholders Group.
Interaction with non-financial reporting	The availability of data via ESAP, hence easier access by stakeholders, can also serve as sources of information / input data for ESG rating providers, reducing use of estimations but also contributing to better quality of ratings overall.

Title	Credit Rating Agencies Regulation (CRA)
Legislative or non-legislative?	Legislative and non-legislative
Brief description of policy or legislation	The first set of rules, which entered into force at the end of 2009, established a regulatory framework for CRAs and introduced a regulatory oversight regime, whereby CRAs had to be registered and were supervised by national competent authorities. In addition, CRAs were required to avoid conflicts of interest, and to have sound rating methodologies and transparent rating activities. In 2011, these rules were amended to take into account the creation of the European Securities and Markets Authority (ESMA), which supervised CRAs registered in the EU. ESMA has become a direct supervisor of CRAs. A further amendment was made in 2013 to reinforce the rules aimed at enhancing market competition and addressing weaknesses related to sovereign debt credit. The “level 1” legislation is supplemeted by delegates acts and ESMA Guidelines.
Interaction with ESG ratings	Investors take into account a variety of factors for the assessment of financial risks: Value at risk (VaR), stress testing, market risks, credit risks, liquidity risks, interest risk. Credit risk is an important component of the financial assessment as it replies to the question whether the company or project is financially sound and what the likelihood is that the company will meet its debt obligations. Some professional investors carry their own credit risk assessment but many rely on the assessment provided by the specialised entites called Credit Rating Agencies. The CRAs are required to take into account all factors relevant for the assesement of creditworthiness, and to explain the reason for changes of credit ratings by accompanying each change with the press release. Some changes of credit ratings can be caused by ESG factors. How CRAs take ESG factors into account is explained in Annex 8. Some CRAs have also entered the market of ESG ratings and started to offer ESG ratings, scores or valuations. However, due to the regulatory requirements, the activities of credit ratings are separated from other businesses.

Title	Sustainable Finance Disclosure Regulation (Regulation EU 2019/2088 on Sustainability-related disclosure in the financial services sector) (SFDR)
Legislative or non-legislative?	Legislative
Brief description of policy or legislation	The SFDR was adopted by co-legislators in spring 2019 and was published on 9 December in the Official Journal. It is already in force and most of the provisions apply from 10 March 2021 and complement corporate disclosures by creating a comprehensive reporting framework for financial products and financial entities. Compliance with sustainability-related disclosures is expected to have considerable behavioural effects on financial firms, and indirectly on the business models of companies that are being invested in. Different investment strategies may entail investments in economic activities with different levels of environmental performance. On 4 February 2021, the European Insurance and Occupational Pensions Authority, the European Securities and Markets Authority and the European Banking Authority (collectively the ESAs) jointly submitted to the Commission a first batch of seven draft regulatory technical standards (RTS) under Articles 2a(3), 4(6) and (7), 8(3), 9(5), 10(2) and 11(4) of the SFDR. On 22 October 2021, the ESAs jointly submitted to the Commission a second batch of six draft regulatory technical standards under Articles 8(4), 9(6) and 11(5) SFDR. The Commission decided to bundle all 13 draft regulatory technical standards in a single delegated regulation. The SFDR together with its RTS that will apply from 1 January 2023 aim to trigger changes in investment decisions by financial market participants that produce financial products and end-investors that purchase them, and financial advisers that provide investment recommendations, towards more ESG considerations by improving the quality of information from the point of view of ESG positive and negative impacts. The Regulation is expected to also have effects on business models of investee companies. The SFDR lays down rules for sustainability-related disclosures toward end-investors, for both outside-in sustainability risks and inside-out adverse sustainability impacts. It does so in relation to: 1.the integration of sustainability risks by financial market participants and financial advisers in all investment processes, 2.financial products that pursue the objective of sustainable investment or have environmental or social characteristics, and 3.adverse impacts on sustainability matters at entity and financial products levels, i.e. that or, in certain circumstances, whether financial market participants and financial advisers consider negative externalities on environment and social justice of the investment decisions/advice and, if so, how this is reflected at the product level. In terms of legal framework, the SFDR is a directly applicable Regulation which introduces additional disclosure requirements to the existing elements of relevant sectoral legislations (AIFMD, UCITS, Solvency II, IORP II, national pension rules, IDD and MiFID II), via a self-standing text (lex specialis) providing full harmonization, cross-sectoral consistency and regulatory neutrality as well as joint convergence by ESMA, EIOPA and EBA (ESAs). Instead of amending all these existing sectoral legislations in an identical way, the SFDR comes on “top” of existing rules in order to impose sustainability disclosure obligations in the same way. This way consistency and regulatory neutrality across all relevant institutional investors' sectors is ensured. In addition, COM has revised Level 2 measures through Delegated Acts (under current empowerments in UCITS, AIFMD, MiFID II, Solvency II, IDD) to explicitly clarify how product manufacturers and financial advisers should integrate sustainability risks and consider principal adverse impact in the investment and advisory processes (that is in the areas of organizational requirements, operating conditions, risk management and target market assessment) as part of their duties towards clients/customers/beneficiaries. On 30 September, the ESAs proposed amendments to the RTS in relation to the information that should be provided in pre-contractual documents, on websites, and in periodic reports about the exposure of financial products referred to in Article 8(1) to (2a) of the SFDR and in Article 9(1) to (4a) of the SFDR to investments in fossil gas and nuclear energy activities.
Interaction with ESG ratings	On 22 October 2021, the ESAs jointly submitted to the Commission draft RTS developed under Article 8(4) and 9(6) of the SFDR. The draft RTS specify information required in Articles 5 and 6 of the Taxonomy Regulation on “how and to what extent” financial products referred to in Articles 8 and 9 of the SFDR invest, or are invested, in environmentally sustainable economic activities within the meaning of the Taxonomy Regulation. Financial market participants that offer financial products referred to in Articles 8 and 9 of the SFDR must comply with the respective requirements set out in the SFRD and in Articles 5 and 6 of the Taxonomy Regulation. In this context, financial market participants and financial advisers use a number of different sources of information for their needs under the SFDR and to be able to meet their disclosure requirements. ESG ratings including data that is used to reach the final rating is used as an important element to fill data gaps and comply with the disclosure obligations. Where investors do not have clarity over the methodology used and how a specific ESG rating was achieved they may risk providing inaccurate disclosures that give a misleading idea of the degree of sustainability of the financial product. In addition, certain ESG passive funds track ESG benchmarks which rely heavily on ESG ratings. In such case, where data is taken at face value, some misallocation of capital may also happen with risks of greenwashing.

Title	Taxonomy Regulation: Regulation (EU) 2020/852 on the establishment of a framework to facilitate sustainable investment, and amending Regulation (EU) 2019/2088.
Legislative or non-legislative?	Legislative
Brief description of policy or legislation	Regulation (EU) 2020/852, or Taxonomy Regulation (TR), establishes criteria for determining whether an economic activity qualifies as environmentally sustainable in the EU. The TR applies to: (i) measures adopted by Member States or by the Union that set out requirements for financial market participants or issuers in respect of financial products or corporate bonds that are made available as environmentally sustainable; (ii) financial market participants that make available financial products; and (iii) undertakings which are subject to the obligation to publish a non-financial statement or a consolidated non-financial statement pursuant to Article 19a or Article 29a of the NFRD. The TR is centred around six environmental objectives: climate change mitigation, climate change adaptation, sustainable use and protection of water and marine resources, transition to a circular economy, pollution prevention and control, and protection and restoration of biodiversity and ecosystems. In order to qualify for inclusion in the EU Taxonomy, economic activities will need to: (a) substantially contribute to at least one of the six environmental objectives, by complying with robust and science-based technical screening criteria; (b) do no significant harm to the remaining environmental objectives; and (c) respect minimum social safeguards. With respect to climate change mitigation, the TR recognises three different types of environmentally sustainable economic activities: (i) low-carbon, which in and of themselves contribute substantially to one of the six environmental objectives; (ii) transitional, which are consistent with EU and international environmental goals but for which there are no technologically and economically feasible low-carbon alternatives; and (iii) enabling, which enable other activities to make a substantial contribution to at least one of the environmental objectives and, at the same time, are environmentally sustainable themselves (according to three specific criteria). The technical screening criteria will be developed in two batches: a delegated act on the two climate-related objectives – which was adopted by the Commission at the beginning of Q2 2021- and a delegated act on the remaining four environmental objectives which is still being developed (date of adoption unknown). An additional delegated act complementing the TR has also been adopted in July 2021. It specifies the application of the requirements to publish the information that undertakings under the scope of the NFRD would have to disclose in their non-financial statements or consolidated non-financial statements.
Interaction with ESG ratings	Article 8 of the EU Taxonomy Regulation requires financial and non-financial undertakings under the scope of the CSRD to include in their sustainability reporting or consolidated sustainability reporting information on how and to what extent their activities are associated with economic activities that qualify as environmentally sustainable. In particular, these undertakings will be required to report the share of their turnover, capital expenditure, and operating expenditure that is Taxonomy-aligned. Article 5 and 6 of the EU Taxonomy Regulation requires financial undertakings to disclose the proportion of taxonomy-aligned investment for all their Article 8 SFDR products and Article 9 SFDR products. In some specific cases, where the information is not publicly available, financial and non-financial market participants can rely on specific additional sources to complement company reported data for their Taxonomy reporting obligations. However, these ‘additional sources’ do not include ‘ESG ratings’: -‘Equivalent information’ (as defined in SFDR Delegated Act): financial undertakings can use ‘equivalent information’ for their product-level Taxonomy reporting for companies that are not subject to the Art.8 Delegated Act requirements within the TR. This would typically include companies not subject to the NFRD. Equivalent information should be considered information that provides the same content and level of granularity as that provided by the actual regulatory reporting of companies and should be provided by the companies themselves or by third-party ESG data providers. The ESAs are currently developing guidelines on what can be considered ‘equivalent information’ and it has not been decided yet if ESG ratings will be eligible. -‘Estimates’ (as defined in Art.8 DA): the use of ‘estimates’ is permitted under the entity-level disclosure requirements under Art.8 DA for the DNSH MSS assessments of third-country companies. However, the use of ESG ratings by financial undertakings has been observed on the market since January 2022. In fact, some market participants have been using ESG ratings to cope with their taxonomy product-level disclosures in order to cope with the lack of company Taxonomy data. Users of ratings reported important lack of comparability in the scores provided by ESG rating providers, mainly due to lack of transparency in the methodology used.

Title	European green bond Regulation
Legislative or non-legislative?	legislative
Brief description of policy or legislation	The European Commission proposed its Regulation on European green bonds in July 2021, with the aim of creating a new gold standard for green bonds, known as the European green bond standard (EuGB). The proposal must be approved by the EU co-legislators before this new standard can enter into application. The EuGB would be a voluntary standard, meaning that it would be available to all green bond issuers on a voluntary basis, and issuers would still be free to make use of market-based standards if they prefer. The standard would be usable by all types of bond issuers and bonds (including corporate bonds, sovereign bonds, project bonds, covered bonds, asset-backed securities, etc.). The standard is based on a strong link with the EU Taxonomy Regulation, as it requires any issuers that wishes to make use of it to allocate 100% of the proceeds of the green bond to finance Taxonomy-aligned economic activities, before maturity of the bond. By standardising and clarifying the issuance of high-quality green bonds in the EU, the standard aims to support issuers in their transition towards climate neutrality. The strong link to the EU Taxonomy will give issuers a reliable way to demonstrate that their bonds meet the environmental and climate-related goals. It will also help provide investors with certainty about the environmental credentials of the bond. The EuGB is aligned with market best practice on disclosure and external review. In particular, issuers will need to obtain a mandatory pre-issuance review, confirming alignment of the bond with the standard, and a mandatory post-issuance review, confirming allocation of proceeds to Taxonomy-aligned activities. Issuers will be supervised by national competent authorities to ensure they comply with the disclosure and external review requirements. External reviewers of European Green Bonds will have to be registered and supervised with ESMA, which will improve transparency and reliability for investors. This will ensure that they comply with certain conditions, such as: -reviewer and staff are of sufficient competence and experience -Specific organisational requirements: compliance function, separation of business lines -Disclosure and record-keeping requirements The standard will also be open to issuers and external reviewers based outside the EU.
Interaction with ESG ratings	Issuers of future EU GBS bonds are required to disclose the Taxonomy-alignment of projects funded by their bond. For that purpose, they may seek for ESG ratings in particular to obtain underlying data that would serve their needs. The information disclosed in ESG ratings provide additional information to investors, including green bond investors, allowing them to better assess the overall non-financial performance of companies. Finally, some ESG rating providers also have some activities as EU green bond verifiers.

Title	EU Climate Benchmarks: Regulation (EU) 2016/1011 as amended by Regulation (EU) 2019/2089 and the respective Delegated Regulations
Legislative or non-legislative?	Legislative
Brief description of policy or legislation	Regulation (EU) 2016/1011 as amended by Regulation (EU) 2019/2089 creates two new types of EU climate benchmarks (EU Climate Transition and EU Paris-aligned benchmarks) and lays down a number of Environmental, Social and Governance (ESG) disclosure requirements for all benchmarks (with the exception of interest and foreign exchange benchmarks). The minimum standards for the construction of the EU Climate Benchmarks and the exact scope and content of the ESG disclosure requirements have been further specified in the three delegated acts that were published in the Official Journal on 3 December 2020, and which entered into application on 23 December 2020. 176 On the disclosure front, the delegated acts require benchmark administrators to explain, using a set template, which ESG factors they have taken into account when designing their benchmark methodology. They shall also explain how those factors are reflected in the key elements of that methodology, including for the selection of underlying assets, weighting factors, metrics and proxies. In addition, benchmark administrators shall explain in the benchmark statement, using a standard template, how ESG factors are reflected in each benchmark or family of benchmarks they provide and publish. Finally, benchmark administrators shall disclose information on the alignment with the objectives of the Paris Agreement.
Interaction with ESG ratings	The Delegated Regulations lay down a list of ESG factors to be disclosed by benchmark administrators, for those benchmarks that pursue ESG objectives, depending on the type of underlying assets concerned (e.g. equity, fixed income, sovereign). Information on the ESG factors should be made at an aggregated weighted value of the benchmark, not for each individual constituent (company). In order for benchmark administrators to be able to disclose such information at an aggregated weighted value, they should still be in possession of the data from each of the constituent. They will have to source the information directly from companies (e.g. via their annual reports) or obtain this information from external rating/data providers. The Delegated Regulations also lays down as an option for benchmark administrators the possibility to disclose information on ESG ratings of the benchmark. Better information over the methodologies used and objects of the ratings should help benchmark administrators to have clarity over the constituents of the benchmarks and the objectives they pursue. Improved quality and reliability of the ratings used by benchmark administrators will foster confidence from benchmark administrators, and eventually improve the picking up of those ratings as a reliable tool to design meaningful benchmarks. For users of benchmarks, it will help them to ensure that they identify the right benchmarks that are aligned with their investment strategy and that will allow them to implement that strategy, possibly avoiding risks of greenwashing. In turn, the positive impact of this initiative on the quality of ESG benchmarks will improve the confidence of investors in ESG benchmarks, thus further supporting sustainable investing.

Annex 5: SME test

(1) − Identification of affected businesses

This initiative does not specifically target SMEs, but they would be affected in the following ways.

First, European ESG rating providers that are SMEs would be directly affected, as they would fall under the scope of this initiative. We note that SME ESG rating providers currently appear to make up for a large part of the market, at least in numbers (around 45 out of 59 providers are estimated to be SMEs), where the market has a few large ESG rating providers.

As such, ESG rating providers would face certain costs, which may be sizeable for SMEs in particular, hence the importance of introducing mitigation measures (further described below, in section 4). Some of the options which would have created an increase in costs for small providers such as the requirement to have standardised disclosure templates, or the need to comply with a prescribed methodology have been discarded at an early stage.

The preferred option, providing for authorisation, organisational requirements and ongoing supervision of ESG rating providers would lead to important costs for providers. Feedback from stakeholders points to the high impact of costs related to compliance, authorisation and supervision for smaller providers and new entrants. Smaller providers usually need to invest in development and research in order to attract new users and to establish their reputation, and additional costs may be detrimental to them. On the positive side, small providers that would comply with the new requirements would benefit reputationally and commercially from the additional transparency and regulatory oversight; this benefit could be even higher for SMEs, as being authorised could give them additional publicity and trust.

Second, users of ESG ratings that are SMEs (e.g. smaller asset managers or other financial institutions) would be indirectly affected, benefiting from the increased transparency on methodologies and objectives of ratings, but also clarity on ESG ratings and providers’ operations. Smaller asset managers have much less resources to fully investigate and understand the ESG ratings they use and to do complementary checks. Therefore, they are more likely to use and rely on ESG ratings without carrying out their own internal assessments. The increased transparency and clarity would enable them to do so, and the oversight would increase confidence in the ratings.

Third, SME corporates that are rated by ESG rating providers would also be indirectly affected, benefiting from the increased transparency and clarity on ESG ratings and providers’ operations leading to less time spent by rated corporates on understanding the results of ratings of their own companies. It could also help SMEs to use ESG ratings as a check of their sustainability and could result in SMEs taking actions to improve their management of ESG risks and impacts. Furthermore, this initiative could help correct some of the biases in ESG ratings that may currently negatively impact rated SME corporates, as evidenced in the Study.

(2) − Consultation of SME stakeholders

Using data collected in the stakeholder consultations described in Annex 2, this impact assessment has analysed the ESG rating providers active in Europe and identified those that qualify as SMEs. First, we considered their feedback to the consultations (12 out of 19 ESG rating providers who responded were SMEs). The feedback of the SME ESG rating providers to the consultation is reflected under the relevant stakeholder group in Annex 2 since the SMEs’ feedback was mostly aligned with the overall stakeholder group. Of note is that SME ESG rating providers were more favourable toward a legislative intervention, than their larger competitors. Then, the Commission services reached out to a number of SME providers for further detailed feedback on the initiative. We discussed in particular policy options and their impacts on these providers. We also reached out to several SME ESG rating providers who had not responded to the consultation. The feedback received from this additional outreach is summarized in the next sub-section.

Similarly, feedback from SME rating users and rated corporates to the consultation is reflected in Annex 2 within each respective stakeholder group. Almost a quarter of the individual rating user companies responding to the consultation were SMEs, and around 10% of individual rated companies responding to the consultation were SMEs. Their feedback was mostly aligned with the respective stakeholder group’s overall feedback, and they agreed on the shortcomings of the ESG ratings market. Of note were the facts that proportionately more SMEs found the ESG ratings market to be uncompetitive compared to large companies. Also, rated SMEs collectively tended to rely as much, if not more, on their ESG ratings to assess their sustainability impacts, than do their larger peers. Despite this, proportionately fewer rated SMEs had access to their own ratings, compared to large companies.

(3) − Assessment of the impacts on SMEs

The Commission aimed to adequately and proportionately assess the impact on SMEs by additionally reaching out to SME ESG rating providers after the targeted consultation, as described above in item (2). For the purpose of the assessment of impacts, discussions were organised with these providers about the policy options and their potential costs and benefits.

In terms of impact, the biggest impact on the operations of SME ESG rating agencies will stem from the requirements related to the authorisation and supervision.. However, the cost of authorisation and supervision will be adjusted to the size and revenue of companies. According to our assessment, the one-off costs stemming from the authorisation could amount to EUR 68.000 177 and the annual on-going cost of supervision for smaller providers could be in the range of EUR 61-82 thousand per provider, as compared with the costs for bigger providers EUR 284-490 EUR per provider. Supervisory fees would be proportionate to revenues and are expected to be in a lower range of 0.37% for smaller providers 178 . The cost of business adjustments is expected to be low as most of small providers focus on one or two services and thus, they do not face the same conflicts of interests as bigger providers with more complex business models. On the benefit side, smaller providers can gain more visibility and recognition once they receive the authorisation and are subject to the supervision. Investors will have more trust in their operations and increased transparency can help them attract new users. As regards costs of disclosure requirements, most of SME providers have not considered them to be substantial. They have also recognised the need for increased transparency as well as benefits that it would lead to.

From the users’ side, SMEs (as asset managers or rated corporates) are also more resource constrained and hence the ability to more easily understand characteristics of ESG ratings and their methodologies and ability to compare them will benefit them proportionately even more than other users.

(4) – Minimising negative impacts on SMEs

First, it is important to note that in the targeted consultation, respondents were split as to views on the exclusion or introduction of a lighter regime for smaller ESG rating providers, for different reasons.

Some argued that excluding smaller market players would not meet the objectives of this initiative, to bring more transparency and clarity over ESG rating providers overall. Smaller ESG rating providers have also differentiated themselves from larger ESG rating providers by offering more targeted, specific ESG ratings and they specialise on certain topics 179 . Others worried that it could lead to an un-level playing field, and even negatively impact smaller providers instead of helping them, by leading to a perception of ‘lower quality’ of their ratings, as they would not be subject to any rules or oversight of their activities. Others still suggested that the rules overall should not be burdensome to begin with and should be proportionate by nature (e.g., principle-based) and could be phased in, by being first applicable to larger than smaller providers.

However, and in order to mitigate costs and impacts of new requirements as described under (3), we propose the following mitigating measures:

a.transition period: To mitigate potential concerns about the administrative burden, once new rules enter into force, providers would be allowed to continue operating on the condition that they notify ESMA and become authorized within a pre-determined period of time, the so called “transition period”. The length of the transition period will be adapted to the size of players, giving smaller players more time for completing the authorisation and in this way easing the administrative burden for them. The transition period could range from 12 to 36 months, based on the assumption that the authorisation of one entity takes up to 6 months 180 .

b.adjustment of supervisory fees to the size of the provider. As described in Chapter 6.1.1, under Option 3, fees would be proportionately distributed across providers based on their revenues and are expected to range between 0.37% and 2.4% of revenues based on a preliminary assessment by ESMA, with fees for smaller providers closer to the lower range of 0.37% of their revenues.

c.proportionate supervision: ESMA has adopted a risk-based and data-driven approach when conducting supervision 181 and prioritises its supervisory activities according to the level of risk identified as well as the significance/size of supervised entities. A risk analysis is used to identify areas of possible non-compliance and to assess the potential significance of the issue. As a result, both supervisory fees and interaction with the supervisor would hence be proportionate to provider’s size, mitigating the risk of disproportionate impact on smaller providers.

d.provide the possibility for smaller and innovative entities to ask for some exemptions from a wide range of internal organisational measures, if the provider is able to demonstrate that those requirements are not proportionate in view of the nature, scale and complexity of its business and the nature and range of issue assessed by its product (e.g. data driven rater 182 or innovative forward-looking rater 183 ).

(5) Results of discussions with SMEs

The feedback from SMEs among ESG rating providers was mixed, depending on the provider and the relevant policy option.

For the disclosures on methodologies and on operations, SME providers recognised the need for the transparency and agreed on the objectives sought by this initiative. On the other hand, some suggested tailoring the information to be disclosed to the size and particular business model and not requiring a standard template for disclosures in order to minimize the burden on smaller providers (e.g. by reducing the need for large changes to the IT systems to create standardized disclosures). Others agreed on the need for the transparency and haven’t considered it to lead to a substantial increase in costs. For the oversight, most of the SME providers considered authorisation and ongoing supervision to constitute a substantial cost. Those costs were seen to be more ‘binary’ in nature, meaning even a low level of supervision would necessitate (potentially costly) internal processes and structures to be put in place. Nevertheless, some suggested that less intensive supervision of their operations (e.g. more flexibility in staffing requirements, less frequent information requests or reviews, etc.) would be more proportionate, and also in line with a ‘risk-based’ supervision. The providers pointed to similar examples in the CRA supervisory regime.

Annex 6: The Ecosystem of ESG Ratings

Who uses ESG ratings

Investors and companies are the prime users and recipients of ESG ratings, given that the majority of ESG ratings are provided by subscription. Academics constitute another group of users 184 . Few individual citizens and consumers directly use ESG ratings, but more use such information indirectly, for example when considering the advice or opinions of financial advisers or non-governmental organisations.

What is an ESG rating

IOSCO in its report on a set of recommendations published in November 2021 185 has defined ESG rating as referring “to the broad spectrum of ratings products that are marketed as providing an opinion regarding an entity, a financial instrument or product, a company’s ESG profile or characteristics or exposure to ESG, climatic or environmental risks or impact on society and the environment, that are issued using a defined ranking system of rating categories, whether or not these are explicitly labelled as “ESG ratings””. It is the first definition of ESG ratings from an international organisation. The notion of ESG ratings is not otherwise clearly framed, be it at international, European or national level and various terms are used today in the market without general consensus and clarity (ratings, scores, valuations).

In addition, ESG ratings do not constitute a homogenous group. They differ as to what they assess (aggregated ESG, only individual E, S or G, or even specific indicators within each letter), from which perspective they assess (risks to the company only, double materiality, impacts only, compliance with international principles), and how they make an assessment (best in the class or in absolute terms, quantitative or qualitative assessment).

ESG ratings providers also have different business models (large, for-profit providers such as MSCI and S&P, boutique providers such as Carbon4Finance or non-for profit providers such as CDP) and revenue models (user-pay (investors) model, company-pay model, mixed model or public funding).

In general we can divide ESG ratings into 4 categories, depending on their purpose:

–assessing risk (financial perspective) (e.g. MSCI),

–assessing impacts (e.g. Carbon4Finance),

–assessing compliance with international principles and guidelines (e.g. Standard Ethics),

–assessing supply chain sustainability risks (not used for direct investment purposes) (EcoVadis).

Some ESG ratings assess companies relative to their peers, i.e. corporate sustainability performance – for the best in the class (e.g. LSEG/Refinitiv), others in absolute terms (e.g. Carbon4Finance).

Many ESG rating providers offer ratings of both corporates and portfolios. Some providers rate countries.

The rating process also differs; on the one hand there are providers who use fully automated quantitative assessments based on KPIs and data without the involvement of the analyst (quite often referred to as scoring) 186 . On the other hand, there are providers whose rating processes resemble that of the credit rating process – based on the analysis of both quantitative and qualitative information, with the involvement of the analyst and rating committees 187 . Both practices fall within scope of this work. Fully automated practices should not be considered as different given that they are based on criteria set by the relevant company and provide a number, a rating of a company. Yet other providers apply a fully qualitative process, based on non-public information and engagement with companies, based solely on the analytical judgement on the forward-looking perspective, and delivered as an opinion from the ratings agency (e.g. S&P ESG Valuation).

Depending on the process and business model, as well as whether the companies being rated are listed or not, some ESG rating providers use only public data, others use a mixture of public and non-public and a few use only non-public information 188 .

Who develops ESG ratings

There are two major groups of ESG ratings providers: the first is specialised entities and the second is financial institutions and other participants (mainly asset managers and benchmark administrators but also some banks or analysts in investment firms). The first group is currently unregulated and not subject to any oversight. Some big groups have regulated legal entities within their structures, like credit rating agencies and thus fall under the scope of the Credit Rating Agency Regulation or benchmark administrators that fall under the Benchmark Regulation. The regulated entities are separate legal entities and they don’t provide ESG ratings. The second group is subject to extensive regulation and supervision, in particular the AIFMD, UCITS and SFDR.

Another difference between those two groups is that specialised entities develop ESG ratings as a business model, that they commercialise their ratings for the use of professional investors or rated companies, whereas asset managers, benchmark administrators or other financial institutions develop ESG ratings for their own purposes. In the case of asset managers, in-house ESG ratings of own investment funds are used mainly for their investment decisions. According to the Study, “This is also how asset managers consider that they add value for their clients and differentiate themselves in the market by adding proprietary classifications and methodologies. They as a result can generate independent and differentiated insights into sustainability-related topics ahead of the market.” In the survey conducted for the Study and the IRRI Survey 2019 189 , 75% of asset managers anticipated an increased demand for both in-house and external ESG ratings.

In the case of benchmark administrators, in-house ESG ratings serve as a basis for the construction of sustainability benchmarks. EU law also gives the possibility for benchmark administrators to disclose ESG ratings in the benchmark statement over benchmarks that pursue ESG objectives 190 . Anecdotal evidence also suggests that some asset managers may use in-house ESG ratings in contracts with benchmark administrators for the development of customised benchmarks. They however do not directly sell the ESG ratings themselves. As such asset managers’ and benchmark administrators’ ESG rating activities fall outside the scope of this impact assessment.

According to KPMG estimates in the IOSCO Report, worldwide there are around 160 ESG data and rating providers. ESMA estimated the number of ESG rating providers operating in the EU (in June 2022) to be 59 191 .

As far as the structure of the market is concerned, there is a small number of very large non-EU entities and a large number of significantly smaller EU entities.

Recent years have also seen consolidation, with mostly large players acquiring smaller and specialized targets. 192 The Study noted that “The business model and governance structure for how multiple products and acquisitions are managed across a single provider varies by firm, and the structure becomes more complex with each new acquisition.” According to the Russell Investment 2020 survey 193 , the three largest ESG rating and data providers based referenced by asset managers globally are MSCI ESG Research (42% of asset managers surveyed subscribe to their services), Sustainalytics (37%) and ISS-oekom (31%).

Moreover, with the exception of ISS, which was acquired by Deutsche Börse AG in November 2020, the major providers are all currently headquartered in North America or the United Kingdom. This includes leading providers MSCI, S&P, Moody’s, Fitch’s, and CDP. All these groups have operations in the EU, due in part to the acquisitions of previously well-established EU headquartered providers such as Sustainalytics, Vigeo Eiris, Oekom and SAM. The leading providers all sell products to EU-headquartered investors and include EU companies in their coverage.

Specialised ESG rating providers broadly fall into three categories:

·For-profit large providers that offer multiple sustainability-related products and services, as well as non-sustainability-related products and services 194 ;

·For-profit boutique providers that offer speciality sustainability-related products and services 195 ;

·Non-profit providers that offer sustainability-related products and services. 196

In addition, they use different revenue models:

–user pay model – subscription fee is paid by asset managers, benchmark administrators and asset owners, ESG ratings are used for investment purposes.

–company pay model – fee paid by the company which requests a rating on its own operations or on the supply chain, some ratings are private hence not made available to the general public, company decides if it wants to publish ratings. Some ESG ratings are used for investment purposes, some for other purposes such as to understand operational risks or seek opportunities.

–mixture of two – requesting client (a buyer organization) has a fixed cost (based on its revenue) and rated company in the supply chain pays a fee which depends on its size

–revenue from re-selling data and ratings to other ESG rating providers

–public funding – CDP is partly funded from public money (including EU funds) and partly from fees from commercial activities

The Study clarifies that the majority of ESG rating or score providers have organised their business model defining the level of granularity of information to stakeholders based on the relation to them, i.e. whether they are rated companies or clients/investors. The level of granularity also differs depending on the providers, notwithstanding the relationships. According to the study, those that are ready to pay a – unique – or subscription fee are given additional level of details compared to the general public and information on their website, which is either not present or high-level 197 . Many ESG rating providers also mentioned in the context of the Study that they do provide additional information to clients (subscribers) and to companies who pay for access to different databases 198 .. The additional information provided to companies could come in the form of benchmarking tools 199 to improve performance and/or specific paid services to understand how to improve ratings.

How is data verified

Data sources utilized by ESG rating providers fall into a few major categories: data directly from the rated company, unstructured data from alternative sources (such as the internet, literature, media reports, satellite images), data from public databases, and third-party data that has already flowed through a different provider 200 . Though the primary source of information identified by most ESG rating providers is self-disclosed company data (both public and non-public), they also commonly utilize data from all other types of sources with distinctions depending on the methodology, approach and product or service offered. Where providers are not able to get direct corporate data, they will often estimate data. Based on the findings from the Study, the bulk of data estimation occurs in specific issue areas, such as carbon emissions, and in estimates of revenue that are affected by a particular sustainability-related issue. ESG rating providers have various multi-step approaches to assuring the quality of the datasets they use, both in terms of errors, and timeliness of data updates, but companies and investors alike have expressed frustrations at low data quality, inaccurate information and absence of clarity as to how estimations (if any, where data is not available) are effectively made 201 .

Annex 7: Global review of ESG rating initiatives

The topic of ESG data and ESG ratings has been on the European Commission’s radar for several years and the EU was perhaps the first in considering the possibility of regulating (part of) the market as far back as 2018 in the first sustainable finance action plan, in addition to legislation aiming to improve comparability and reliability of sustainability reporting. The market has continued to evolve rapidly since then and other jurisdictions have also increased their focus on this market, particularly but not only on the reporting and ESG data aspects, with numerous regulations in place and on the way. 202 We have reviewed various past and ongoing (private and public) initiatives to understand the global landscape for ESG rating providers in particular, given the focus of this initiative. This provides an important backdrop for the rest of this impact assessment given that ESG rating providers, as well as ESG rating users or ESG-rated entities, are often active in multiple jurisdictions.

International

In November 2021, IOSCO published a report 203 highlighting concerns related to ESG ratings and data provision and issuing recommendations to address those concerns. IOSCO defined ESG ratings with a wide scope: “the broad spectrum of ratings products that are marketed as providing an opinion regarding an entity, a financial instrument or a product, a company’s ESG profile or characteristics or exposure to ESG, climatic or environmental risks or impact on society and the environment that are issued using a defined ranking system of rating categories, whether or not these are explicitly labelled as “ESG ratings””. Specifically, IOSCO identified the following concerns:

a)there is little clarity and alignment on definitions, including on what ratings or data products intend to measure;

b)there is a lack of transparency about the methodologies underpinning these ratings or data products;

c)while there is wide divergence within the ESG ratings and data products industry, there is an uneven coverage of products offered, with certain industries or geographical areas benefitting from more coverage than others, thereby leading to gaps for investors seeking to follow certain investment strategies;

d)there may be concerns about the management of conflicts of interest where the ESG ratings and data products provider or an entity closely associated with the provider performs consulting services for companies that are the subject of these ESG ratings or data products; and

e)better communication with companies that are the subject of ESG ratings or data products was identified as an area meriting further attention given the importance of ensuring the ESG ratings or other data products are based on sound information.

To address these concerns, IOSCO issued the following recommendations to various actors:

·Regulators could consider focusing more attention on the use of ESG ratings and data products and ESG ratings and data products providers that may be subject to their jurisdiction.

·ESG ratings and data products providers could consider adopting and implementing written procedures designed to help ensure the issuance of high quality ESG ratings and data products based on publicly disclosed data sources where possible and other information sources where necessary, using transparent and defined methodologies.

·ESG ratings and data products providers could consider adopting and implementing written policies and procedures designed to help ensure their decisions are independent, free from political or economic interference, and appropriately address potential conflicts of interest that may arise from, among other things, the ESG ratings and data products providers’ organizational structure, business or financial activities, or the financial interests of the ESG ratings and ESG data products providers and their officers and employees.

·ESG ratings and data products providers could consider identifying, avoiding or appropriately managing, mitigating and disclosing potential conflicts of interest that may compromise the independence and objectivity of the ESG ratings and ESG data products provider’s operations.

·ESG ratings and data products providers could consider making adequate levels of public disclosure and transparency a priority for their ESG ratings and data products, including their methodologies and processes to enable the users of the product to understand what the product is and how it is produced, including any potential conflicts of interest and while maintaining a balance with respect to proprietary or confidential information, data and methodologies.

·ESG ratings and data products providers could consider adopting and implementing written policies and procedures designed to address and protect all non-public information received from or communicated to them by any entity, or its agents, related to their ESG ratings and data products, in a manner appropriate in the circumstances.

·Market participants could consider conducting due diligence or gathering and reviewing information on the ESG ratings and data products that they use in their internal processes. This due diligence or information gathering and review could include an understanding of what is being rated or assessed by the product, how it is being rated or assessed and, limitations and the purposes for which the product is being used.

·ESG ratings and data products providers could consider improving information gathering processes with entities covered by their products in a manner that leads to more efficient information procurement for both the providers and these entities.

·Where feasible and appropriate, ESG ratings and data products providers could consider responding to and addressing issues flagged by entities covered by their ESG ratings and data products while maintaining the objectivity of these products.

·Entities subject to assessment by ESG ratings and data products providers could consider streamlining their disclosure processes for sustainability related information to the extent possible, bearing in mind jurisdictions’ applicable regulatory and other legal requirements.

Various jurisdictions, including the ones highlighted below, took IOSCO’s recommendations into account when looking into the ESG data and ratings area.

United Kingdom

In June 2021, the FCA published a consultation paper "Enhancing climate-related disclosures by standard listed companies and seeking views on ESG topics in capital markets" 204 . The paper also addressed the role of ESG data and ratings providers and identified various policy issues for ESG rating providers and areas of potential harm in the FCA’s view. Some of these areas of concern were:

·the "hardwiring" of ESG ratings into investment processes;

·a lack of transparency of methodologies and interpretability of ESG ratings;

·the need to ensure good governance and management of conflicts of interest; and

·the costs for issuers of meeting providers’ data requests.

In that paper the FCA identified various policy options to address these concerns, such as “guidance” for users of third party ESG data and ratings (including in relation to risk management, outsourcing arrangements, due diligence and use of ratings in benchmarks and indices), “soft regulation” for ESG data and rating providers (encouraging voluntary standards), or formal regulation of ESG data and ratings providers, focusing on transparency, governance and managing conflicts of interest. The paper called for views from market participants on these options until September 2021 and the FCA’s final policy was announced for the first half of 2022. Following this, in October 2021 the FCA's Perimeter Report 2020/21 205 and the government’s "Greening Finance: A Roadmap to Sustainable Investing" 206 echoed similar ideas.

In December 2021, the FCA's CEO and the Economic Secretary to the Treasury (EST) discussed the FCA's Perimeter Report, including the issue of ESG data and ratings. A record of this meeting stated "The EST noted that there have been discussions on ESG ratings and potentially bringing the providers into the perimeter. The CEO noted that there is support for the government’s work to consider bringing ESG ratings providers into the perimeter."

In June 2022 the FCA published the feedback statement based on the 2021 consultation, where it stated “We see a clear rationale for regulatory oversight of certain ESG data and rating providers – and for a globally consistent regulatory approach informed by the recommendations on ESG data and ratings developed by the International Organization of Securities Commission (IOSCO) in 2021…We therefore support the Government’s consideration of bringing ESG data and rating providers within our regulatory perimeter. As noted by the Economic Secretary to the Treasury and our Chief Executive Officer, ‘common baseline standards that support innovation would allow the UK to lead globally’.” As such, at this stage it appears likely that regulating ESG data and rating providers is in the regulatory pipeline of the UK, focusing on “enhancing transparency of ESG ratings and data products and their methodologies, assisting users in interpreting what they aim to measure and how, and promoting strong governance, conflicts management, and systems and controls to underpin data and rating products and give the market confidence that they are objective, independent and free from bias, determined as the result of a systematic process, and of reliable quality.”

Most recently, on 30 March 2023 the UK government published it is green finance strategy 207 , incuding plans to regulate ESG rating providers. The public consultation foresees giving a new mandate to Financial Conduct Authority (FCA) to regulate ESG ratings firms due to concerns over a lack of transparency, conflicts of interest and poor governance in the sector. The consultation is open until the end of June 2023.

Japan

Japan’s financial regulator, Financial Services Agency (FSA) published a report in June 2021, “Report of the Expert Panel on Sustainable Finance," recommending that the FSA promote discussions on codes of conduct for ESG evaluation and data providers. Then, in February 2022 established a Technical Committee for ESG Evaluation and Data Providers, composed of market participants and academics. The discussions and conclusions of the Technical Committee were summarized in a report published in July 2022, and also formed the basis of the draft Code of Conduct for ESG evaluation and data providers 208 published at the same time.

The report particularly highlighted the importance of dialogue between investors, rated entities, and ESG evaluation and data providers, so the Code of Conduct aims to enhance said dialogue. Importantly, the Code of Conduct proposes to cover not just ESG evaluations on an issuer and/or instrument basis, but also the provision of data related to such evaluations. It is designed as a principles-based, voluntary ‘comply or explain’ code, applicable to all ESG evaluation and data providers that either participate in, or provide services to participants in, the Japanese financial markets. The Code of Conduct is intended to cover both ‘subscriber-pays’ and ‘issuer-pays’ business models, although certain provisions differ in their detail based on the applicable business model. The high-level principles are listed here for reference.

First, “ESG evaluation and data providers should ensure the quality of ESG evaluation and data they provide. The basic procedures necessary for this purpose should be established.” This includes measures such as establishing consistent methodologies, assessing evaluation results against those methodologies, and ensuring data quality even when the collection is outsourced. Second, providers should “secure necessary professional human resources to ensure the quality of the evaluation and data provision services they provide, and should develop their own professional skills”, although the FSA acknowledges the current shortage of expertise in this rapidly expanding market.

Third, providers should “establish effective policies so that they can independently make decisions and appropriately address conflicts of interest that may arise from their organization and ownership, business, investment and funding, and compensation for their officers and employees, etc. With regard to conflicts of interest, providers should identify their own activities and situations that could undermine the independence, objectivity, and neutrality of their business, and avoid potential conflicts of interest or appropriately manage and reduce the risk of conflict of interest.” For example, firewalls and suitable compensation schemes could be used to manage conflicts of interest. FSA also points out that providers with more complex evaluation methodologies may need to provide more disclosures and have more stringent measures in place to assure clients that the evaluations are free of such conflicts.

Fourth, providers should “recognize that ensuring transparency is an essential and prioritized issue, and publicly clarify their philosophy in providing services, such as the purpose, approach, and basic methodology of evaluations. Methodologies and processes for formulating services should be sufficiently disclosed.” FSA notes that due consideration can be given to intellectual property while ensuring transparency. Disclosures should also include information on the data collection, such as data sources and use of estimates. FSA notes that the Code assumes general publication of disclosures, although it acknowledges that each provider should make a judgment as to what is disclosed to the public versus to the provider’s direct clients and/or rated entities. Indeed, the fifth principle ensures data protection: providers “should establish policies and procedures to appropriately protect non-public information obtained in the course of business.”

Sixth and last, providers “should devise and improve the way they gather information from companies so that the process becomes efficient for both service providers and that companies or necessary information can be sufficiently obtained. When important or reasonable issues related to information source are raised by companies subject to evaluation, ESG evaluation and data providers should appropriately respond to the issues.” Finally, an appendix includes expectations that investors understand the characteristics and differences of the evaluations they use, and publicly clarify how they have used them in their investment decisions.

The draft code was open for public comments until September.

India

India is another jurisdiction that has been active in sustainable finance topics in general, and ESG ratings in particular. First, in May 2021 the Securities and Exchange Board of India (SEBI), India’s financial regulator, issued revised sustainability reporting requirements “Business Responsibility and Sustainability Report”, BRSR, requiring listed companies to disclose the financial implications of, as well as plans to mitigate or adapt to, ESG risks. BRSR filing will be mandatory from the financial year 2022-23 onwards for the top 1000 listed companies by market capitalization, while other companies can voluntarily issue BRSRs. Second, in October 2021 SEBI proposed new disclosure norms for mutual funds that invest using ESG criteria.

Most recently, in a consultation paper launched in January 2022 209 , SEBI took note of the increased need for sustainable finance in the coming years and the growing role of ESG rating providers in channeling such investments. SEBI mentioned potential risks to “investor protection, the transparency and efficiency of markets, risk pricing, and capital allocation” stemming from the lack of regulatory oversight of those providers. Following discussions with market participants, SEBI concluded that there is ambiguity about the wide range of products offered, inconsistency in disclosures and transparency of the methodology and rating process, potential conflicts of interest, and lack of India-specific providers. It thus proposed a regulatory framework to address these issues and achieve better interpretability, comparability, and reliability.

In brief, the proposed regulatory framework includes an accreditation scheme to ensure that a provider has adequate capital, infrastructure, and skilled manpower to conduct its activities. Relatedly, any listed entity, mutual fund, alternative investment fund or index provider using an ESG rating would need to use a SEBI accredited provider. SEBI proposed that only registered credit rating agencies (CRA) and research analysts (RA) with a minimum net worth to ensure adequate capitalization would be eligible to be accredited, and they would need to have specialists in data analytics, sustainability, finance, IT and law. The accreditation would be subject to review every two years, but could be revoked at any time.

As for the product scope of the framework, SEBI proposed to limit it to ESG impact ratings, ESG corporate risk ratings and ESG financial risk ratings. According to SEBI, ratings covering only the environmental aspect, for example, should not be called ESG ratings by the providers and would not fall under the framework. While SEBI considered the possibility of standardizing the rating scale for ESG ratings, it concluded that it would be premature at this stage. The SEBI proposal requires disclosure of rating methodologies including definitions, KPIs and weights of each component, data sources and estimates used and an annual evaluation of methodologies against results. It also requires providers to have and implement policies to ensure proper and consistent application of rating methodologies and rigorous assessments by qualified rating committees. The SEBI proposal includes measures to prevent conflict of interest, such as requiring a dedicated policy, separating rating personnel from other personnel, remunerating rating personnel in a way that avoids conflict of interest and disclosing any potential conflict of interest between rating activities and other activities. Lastly, the SEBI proposal looks at the existing business models in the industry and considers mandating the prevalent one for all, i.e. the subscriber-pays business model.

The consultation closed in March 2022 and as of the date of the conclusion of this impact assessment, no feedback statement has been published by SEBI.

Other

A number of other jurisdictions such as Switzerland, Nigeria, Hong Kong, Indonesia, Singapore, and Thailand have been taking action in the ESG area, usually around enhancing the reporting of listed companies, without (yet) intervening in the ESG ratings market specifically.

Currently, ESG disclosures are not regulated in the US, but in June 2021 the House of Representatives passed an ESG Disclosure Simplification Act that would require listed companies to disclose quarterly and annual ESG metrics. The act also assigned authority to the Securities Exchange Commission (SEC) to define the metrics and require the necessary disclosures. However, the act needs to pass through the second chamber, the Senate, in order to become law and as yet, no such vote has taken place or been scheduled.

Private initiatives

ESG rating providers themselves have made individual and collective efforts to address certain issues. In addition to internal codes of conduct at many, if not most, providers, there are (at least) 2 instances of industry standards that were established.

First, the standard currently known as ARISTA was supported with initial funding by the European Commission in 2004 and has been an industry-specific quality standard for ESG rating and research organizations for over 10 years. In 2004, a number of independent Responsible Investment Research (RIR) Groups working for customers in the financial services industry formed the AICSRR (Association for Independent Corporate Sustainability and Responsibility Research). AICSRR was later renamed Association for Responsible Investment Services (ARISE). The standard was formerly known as CSRR-QS (Corporate Sustainability and Responsibility Research Quality Standard) and was relaunched in 2012 as ARISTA (Responsible Investment Research Standard).

The purpose of the standard is to encourage independence, objectivity and professionalism; quality management systems; transparency; assurance processes and research characteristics viewed as best practices. The standard aims to form a basis for further verification procedures in addition to the in-house assurance processes. The standard covers various aspects of the company such as the organisational and managerial structure, its research processes and its assessment processes. In terms of products and services, the standard covers the collection of corporate sustainability and responsibility (CSR) data and related research, analysis, evaluation, rating, ranking, screening, and risks and opportunities assessments. In 2012, 10 groups had been certified against the ARISTA standard, including Vigeo Eiris, EthiFinance, imug, oekom research, GES, Ecodes, CAER, Greeneye and KOCSR, with Ecovalores and Inrate preparing to follow their example. However, as of 2022 this does not appear to be in common use.

Second, another market-based standard was the “Deep Data Delivery Standard” 210 , developed in 2016. Investment professionals and academics from across the world developed a list of 10 principles for deep data sets; it was intended as a public good that any asset manager or asset owner could use in contracting third-party data providers.

The first five principles outline key characteristics of the data and related processes such as scope, identification, and machine readability, while standards six to ten help ensure the integrity of the data providers. To meet this standard, deep data sets need to be delivered:

1)with a minimum of 5 years historical data on at least 30 independent indicators per data set (e.g., credit rating, ESG data) whereby any data point that is not delivered as reported at the respective point in time should be flagged as backfilled;

2)with 98% value weighted market coverage, where a market (e.g. equity index) is claimed to be covered;

3)with an assurance that ratings will be reconsidered for at least 8.25% of the companies covered in the average month of the following year;

4)including considerate, accurate identifiers (e.g. ISINs) for 99% of the firms covered in every month of current and historical data coverage;

5)in machine readable format (e.g., CSV, XML) and with proper documentation of the data structure;

6)with an assurance of individual rating independence meaning that none of the rated entities in the respective market (e.g., equity index) financially contributed to their rating or paid for access solely to their own rating;

7)with an assurance of organizational rating independence meaning that whenever rating agencies win entities as new clients which they also rate an independent analysis is conducted if these new clients receive, statistically significant, higher ratings than in the year before and any biases found in this analysis will be addressed within 12 months;

8)with an assurance that all research or rating reports in the following year will indicate names and office locations of all analysts substantially involved in the analysis as well as the extent to which their data sources exceed those self-reported by the rated entity;

9)with an assurance that all research or rating reports in the following year will include a logbook detailing any errata, where applicable, as well as the dates and roles of participants in communication with the rated companies;

10)accompanied by the ratio of the rating agency’s research costs to total cost or the ratio of research head count to total head count in the most recent financial year.

A number of providers were certified once against the standard after 2016, but as of 2022 it too does not appear to have become a widespread industry standard.

Annex 8: Incorporation of ESG factors into credit ratings

Background

The provision of credit ratings was regulated after the financial crisis of 2008.

In the period leading up to the financial crisis in 2008, credit rating agencies (CRAs) failed to properly appreciate the risks in more complex financial instruments. For instance, structured finance products backed by risky sub-prime mortgages were issued with incorrect ratings that were far too high. During the subsequent euro area debt crisis, certain countries were faced with abrupt bond selloffs and higher borrowing costs following a downgrade of their credit rating. In addition, investors relied excessively on external credit ratings, without carrying out their own due diligence.

In response, the Commission made proposals to strengthen the regulatory and supervisory framework for CRAs in the EU, in order to restore market confidence and increase investor protection. The new EU rules were introduced in three consecutive steps.

The first set of rules, which entered into force at the end of 2009, established a regulatory framework for CRAs and introduced a regulatory oversight regime, whereby CRAs had to be registered and were supervised by national competent authorities. In addition, CRAs were required to avoid conflicts of interest, and to have sound rating methodologies and transparent rating activities. In 2011, these rules were amended to take into account the creation of the European Securities and Markets Authority (ESMA), which now supervises CRAs registered in the EU. A further amendment was made in 2013 to reinforce the rules aimed at enhancing market competition and addressing weaknesses related to sovereign debt credit.

The CRA Regulation 211 provides the following definition of a credit rating:

“credit rating means an opinion regarding the creditworthiness of an entity, a debt or financial obligation, debt security, preferred share or other financial instrument, or of an issuer of such a debt or financial obligation, debt security, preferred share or other financial instrument, issued using an established and defined ranking system of rating categories”.

Credit Ratings assess the risk of default of a company and should capture all the risks that may affect creditworthiness - including sustainability risks where relevant. Credit ratings are used by investors to calibrate portfolio risk profiles (including for investment products).

Credit ratings are also important for the calculation of prudential requirements under the EU’s Capital Requirements Regulation (CRR) and Solvency II and are used by the European Central Bank as one element in its open market operations; with the ECB recognising the shortfall in integration of climate risks in credit ratings and introducing additional requirements on climate risks in its market operations as of October 2022.

Changes in credit ratings or rating outlooks can have important impacts on financial markets, financial stability and the broader economy.

As far as links with the market of ESG rating providers are concerned, some large groups have within their structures, both a legal entity registered with ESMA as a Credit Rating Agency, as well as entities that provide ESG ratings. However, due to the legal requirements, CRAs cannot provide ESG ratings.

ESG factors in credit ratings

ESG risks, starting from climate- and environment-related risks are an increasingly material source of risk for all economic actors and are likely to be more and more relevant to assess companies’ creditworthiness. Credit ratings are defined as opinions about factors that may affect creditworthiness, which may also include sustainability factors, among many others.

The CRA Regulation requires CRAs to incorporate in their assessment all driving factors relevant for the assessment of creditworthiness. CRAs should therefore consider these risks and integrate them in a systematic and rigorous way where relevant for a creditworthiness assessment.

Both the report of the HLEG on Sustainable Finance 212 , 2018 Sustainable Finance Action Plan 213 and the 2021 Strategy on Financing the Transition to a Sustainable Economy 214 acknowledged the importance of incorporation of ESG factors in the creditworthiness assessment and asked ESMA for advice.

In 2019 ESMA provided its Technical Advice to the Commission 215 stating that it is not advisable to explicitly mandate the consideration of sustainability characteristics in CRA’s credit assessments, and concluded the following:

“While it is clear that the CRAs contacted are considering E, S or G factors in their credit ratings, the extent to which each factor is considered varies by asset class, according to the importance assigned to that factor by a CRA’s methodology for the purpose of assessing credit worthiness. As a result, while ESG considerations can be a factor, credit ratings should not be understood as providing an opinion on sustainability characteristics of an issuer or entity”.

“In addition, given the specific role credit ratings continue to occupy within the financial system the Technical Advice concludes that it would not be advisable to amend the CRA Regulation to more explicitly mandate the consideration of sustainability characteristics in CRA’s credit assessments. Although it could be useful to update the CRA Regulation’s disclosure provisions, to provide a more consistent level of transparency around how CRAs are considering ESG factors in these assessments and ensure the CRA regulatory framework keeps pace with ESG developments in other areas.

Finally, given the trajectory of EU financial market legislation towards integrating sustainability assessments in the operational and decision-making processes of financial market participants, it may be relevant to assess whether there are sufficient regulatory safeguards in place elsewhere for those non-credit rating products that will fill the need for such sustainability considerations.” 216

Further to the above mentioned, in 2019 ESMA issued Guidelines on disclosure requirements 217 that include recommendations for how CRAs should present which credit ratings have been impacted by ESG factors. In 2021, ESMA carried out an assessment of results of Guidelines 218 and, in addition, reviewed how CRAs incorporate ESG factors in their methodologies.

Based on the recent ESMA assessments 219 as well as the results of the targeted consultation and meetings with stakeholders, two main problems have been identified:

Incorporation of ESG factors

The CRA Regulation requires CRAs to incorporate all driving factors relevant for the assessment of creditworthiness. In order to assess the level of integration of ESG factors into CRAs’ credit rating methodologies, ESMA has carried out a thematic review of their methodologies and rating process. ESMA’s assessment included not only CRAs’ credit rating methodologies, but also research papers, internal guidance documents, and rating committee packages.

Based on ESMA thematic review 220 , since 2019, CRAs have developed two types of approaches to integrating ESG factors in their credit rating process:

·ESG-specific credit methodologies: Three CRAs have introduced formal ESG-specific credit methodologies, which are either of a transversal or a sector-specific nature. These methodologies tend to apply in combination rather than as standalone with other credit rating methodologies that are already in place.

·ESG-specific research or white papers: Two CRAs have integrated the consideration of ESG factors into their credit rating process by reference to published research or white papers. However, these do not adhere to the same governance, validation and transparency rules as the ESG-specific methodologies.

Across CRAs’ ESG methodologies, research and white papers, there are different degrees of rigorousness. In general, CRAs’ ESG-specific documentation only loosely establishes how ESG factors are to be assessed. This gives CRAs significant leeway to use "expert judgement" in their application. Only one CRA in the sample had taken the additional step to introduce more specific quantitative measures and indicators to be considered when assessing the credit impact of ESG risks.

Another conclusion refers to heterogeneity in the identification and definition of ESG factors across CRA methodologies and frameworks. There is currently no common definition of ESG factors. The situation may change as there are a few ongoing initiatives at the European and global levels: introduction of the EU taxonomy for sustainable activities can contribute to a greater standardisation of environmental impacts, the work of EFRAG on the European Sustainability Reporting Standards and the work initiated by the International Sustainability Standards Board to deliver comprehensive sustainability-related disclosure standards may foster more comparability across CRAs.

Although these ESG-specific methodologies and research or white papers were new, their introduction did not induce significant changes at the level of ratings. The absence of impact on credit rating levels suggests that any introduced changes were mainly a formal codification of what had existed practice. CRAs echoed this view when this observation was raised with them.

Regarding the application of methodologies: CRAs diverge in the systematic consideration of ESG factors in their credit rating processes and have different degrees of detail in the assessment of ESG factors. Inputs in the rating process are mainly qualitative and provided by issuers. Rating committee documents show a high level of expert judgment that is not always linked to methodologies. Combined, these findings may be indicative of the general low level of rigorousness of methodologies.

Transparency

Transparency around credit rating actions is a key objective of the CRA Regulation and plays an important role in the context of enhancing investor protection and promoting properly functioning financial markets. The CRA Regulation includes a number of requirements relating to the disclosure and presentation of credit ratings. Their purpose is to enable investors to understand the key drivers and assumptions of the credit rating, any limitations underpinning the credit rating, as well as references to where other relevant information can be found. Improving transparency around credit rating actions has been a key objective of ESMA’s supervisory and policy engagements for a number of years. As part of this work, in 2019, ESMA issued the Guidelines to improve the disclosures around credit rating actions.

The goal of the Guidelines is to ensure greater transparency around the consideration of ESG factors where they are material to a credit rating action. In particular, the Guidelines provide that where ESG factors have been key drivers behind a change to the credit rating or rating outlook, CRAs should identify the relevant factors, elaborate on their materiality, and include a reference explaining how ESG factors are considered as part of the credit rating process, for example within their methodologies. Prior to the publication of the Guidelines, ESMA conducted a public consultation and received input from various market participants. The feedback ESMA received from this process outlined that market participants were generally supportive of the need for greater consistency regarding CRA’s disclosure practices, as well as supportive of ensuring transparency concerning the relevance of ESG factors to credit ratings. ESMA published the final Guidelines in July 2019 and has been considering them for the purposes of its supervision since 30 March 2020.

ESMA also analysed the possible impact of the Guidelines on ESG disclosures since their entry into force in April 2020 221 . In this regard, ESMA’s assessment of CRAs’ ESG disclosures practices is based on the application of natural language processing (NLP) techniques to more than 60,000 press releases reported to ESMA by these three CRAs over the period from January 2019 to December 2020, thereby accounting for over 90% of the total credit rating activity in this period.

ESMA findings show that there has been a notable (c. 60%) increase in the reporting of ESG factors after the introduction of the Guidelines, concentrated in environmental and social disclosures, and to a lesser extent governance (see Figure 1 below):

Figure 1: Press releases (%) with meaningful ESG considerations

Figure 2a below shows that the Guidelines had a positive effect on ESG disclosures by large CRAs. It is worth noting that 3 large CRAs account for 91% of the market.

Figure 2a: Average ESG Score across CRAs

During fall 2021, ESMA carried out an additional assessment of disclosures made by smaller CRAs. Adding smaller CRAs has not changed the overall assessment: there is an overall increase in ESG disclosures (c 50%). However, disclosures of small CRAs are more volatile, resulting from the smaller number of reported rating actions. It is important to remember that not all rating are caused by the sustainability factors.

Figure 2b: Average ESG Score across CRAs

One of the key findings is that the improvement in ESG disclosures is not uniform across all CRAs in the sample. In fact, the overall improvement is mainly driven by one large CRA that significantly increased the reporting of ESG factors in its press releases after April 2020.

Figure 3: ESG Score Heterogeneity

ESMA is of the opinion that some heterogeneity might be expected given the differences in the respective CRA methodologies. It will however also take further steps, using its supervisory powers, to address the observed gap in the level of ESG disclosures across CRAs.

Results of the targeted consultation:

Use of credit ratings

Credit ratings are used for investment decisions, but they are not decisive factors, more so one of many sources of information (36.8%) or as a starting point (10%).

For the majority of users it is either is very important (46%) or important (26%) to understand to what extent individual credit rating actions have been influenced by sustainability factors.

Respondents agree that it is important that credit rating agencies are transparent about the factors that drive changes in credit ratings. Where an ESG factor impacts the credit risk of a company/security, they expect that factor to be incorporated into the credit risk assessment. It is important for users to understand whether the factors have been considered to ensure they are not considered twice or not at all. However, respondents also stressed that credit ratings should remain credit ratings and measure only credit risk.

Transparency

Almost half of respondents (49%) cannot find information about the extent to which CRA’s methodologies or the rating process incorporate sustainability factors to be sufficiently well disclosed.

Users look for that information both on public websites as well as reports available to subscribers.

The majority of respondents (55%) are of the opinion that the level of disclosure differs depending on individual CRAs. There are significant differences depending on the maturity of the CRA provider in term of ESG understanding. The degree of disclosure depends also on the maturity of the methodology adopted by the CRA: the higher the degree of maturity, the more transparency.

Many respondents claim the level of disclosure has improved (38%), but only 15% claim that it is sufficient, whereas 24% claim that it remains insufficient despite some improvements.

As far as the incorporation of ESG factors in the methodologies and rating process are concerned more respondents are of the opinion that it is sufficient (37%) than insufficient (21%).

Opinions of CRAs

7 CRAs replied to questions about CRAs: All replied that they explicitly incorporate ESG factors into their methodologies. They all incorporate all types of sustainability factors: E, S and G. They all have a framework or document describing how they incorporate ESG factors in the credit rating process. They all have improved disclosures on ESG factors and follow ESMA Guidelines on disclosures.

The majority (5 out of 7) consider the current trends are sufficient to ensure that CRAs incorporate relevant ESG factors in credit ratings. Only one CRA sees the need for further detailing of ESMA Guidelines on disclosures.

In the case of the EU intervention, the majority of CRAs identified the risk of excessive importance being attributed to ESG factors as compared with other factors relevant for the assessment of creditworthiness. Others claim that it will bring a more coherent approach to the incorporation of ESG factors into credit ratings.

In a scenario where there would be no EU intervention, 4 CRAs claim that market trends will be sufficient to meet investors’ demands for information on the impact of ESG factors on credit ratings, whereas 3 CRAs are of the opinion that the existing gap between approaches of CRAs to the incorporation of ESG factors in credit ratings will grow.

Opinions on the need for the EU Intervention

There is an almost even split between those who consider that the current trends in the market are sufficient to ensure that CRAs incorporate relevant ESG factors in credit ratings (42%) and those who disagree (40%).

As far as disclosures are concerned, the majority (68%) don’t consider market trends and ESMA Guidelines to be sufficient.

Out of those who don’t consider the current situation sufficient, the majority favours a soft approach (55%): further detailing ESMA Guidelines (37%) or further supervisory actions by ESMA (18%). Only 24% are in favour of the legislative intervention.

Possible policy options

It is important to remember that CRAs are independent in the choice of methodologies and factors used for the assessment of creditworthiness. The principle of independence is embedded in the EU legislation as well as IOSCO principles.

There are 2 policy dimensions examined below:

1)incorporation of ESG factors in the methodologies and rating process

2)disclosure to the public

Policy options regarding dimension 1) incorporation of ESG factors in the methodologies and rating process:

Option 1 – Specify in the CRA Regulation that in the case of the incorporation of ESG factors in the creditworthiness assessment, CRAs must do this in a systemic way and develop relevant methodologies via the targeted amendment of the CRA Regulation 222 and subsequent amendments of the delegated act on methodologies 223 ;’ or

Option 2 – Amend the delegated act on methodologies in order to set conditions that methodologies need to meet in the case of the incorporation of ESG factors in creditworthiness assessment 224

Policy options regarding dimenasion 2) disclosures to the public:

Option 1 – Require CRAs to include a clear explanation as to whether and how ESG factors were considered within the CRA’s methodologies via the update of Annex I to CRA Regulation and subsequent amendments of ESMA Guidelines; or

Option 2 – via the amendment of ESMA Guidelines.

Analysis:

The main area that can be further improved relates to the rigorousness of the inclusion of ESG factors in CRA’s methodologies. The general principles regarding methodologies set up by the Credit Rating Agencies Regulation already apply to the incorporation of ESG factors in credit ratings, whereas precise details concerning the conditions which methodologies should meet are specified in the delegated act on the methodologies for CRAs 225 . Our assessment is that changes to the CRA Regulation would need to be accompanied by the changes to the delegated act. It is also possible to amend the delegated act without amending the Regulation itself. Given time and resources needed to amend the Regulation and then the delegated, in our assessment it is more effective and efficient to amend the delegated act directly.

As regards the disclosure requirements, the Credit Rating Agencies Regulation includes a number of requirements in that respect 226 . The regulation requires that CRAs disclose information to the public on their methodologies, models and key rating assumptions which they use in their credit rating activities. With a view to ensuring transparency, disclosure of any material modification to the methodologies and practices, procedures and processes of credit rating agencies need be made prior to their coming into effect, unless extreme market conditions require an immediate change in the credit rating. Specific disclosure requirements are provided for in the Annex I to the CRAR Regulation. They have been further detailed in ESMA Guidelines.

The CRA Regulation includes also a number of disclosure requirements relating to the issuance of credit ratings 227 . The purpose of these requirements is to ensure a sufficient level of transparency around the characteristics of the credit ratings themselves. This transparency enables the user of the credit rating to understand the main reasons for the credit rating, any limits or uncertainties underpinning the credit rating as well as where further information can be found to facilitate their own due diligence. This information is typically disclosed through the issuing of a publicly disclosed press release. Those transparency requirements have been subject to further clarifications by ESMA.

In order to further improve the transparency the Commission could amend the Annex I of CRA Regulation by including specifying that CRAs need to disclose the information on ESG factors that have influenced the credit rating change.

CRA Regulation lays down the empowerment for the Commission to update the Annex I via a delegated act in view of market developments. More specifically, such a mandate is set out in Art 37 of CRA regulation :

“In order to take account of developments, including international developments, on financial markets, in particular in relation to new financial instruments, the Commission may adopt, by means of delegated acts in accordance with Article 38a and subject to the conditions of Articles 38b and 38c, measures to amend the Annexes, excluding Annex III.”

At the same time, large CRAs have continued improving their disclosures, and disclose now more than it is required by ESMA Guidelines.

Positions from various stakeholders

Results of the targeted consultation show mixed views. There is almost even split between those who consider that the current trends in the market are sufficient to ensure that CRAs incorporate relevant ESG factors in credit ratings (42%) and those who disagree (40%). As far as disclosures are concerned, the majority (68%) don’t consider market trends and ESMA Guidelines to be sufficient. Out of those who don’t consider the current situation sufficient, the majority is favour soft approach (55%): further detailing ESMA Guidelines (37%) or further supervisory actions by ESMA (18%). Only 24% are in favour of the legislative intervention.

The majority of CRAs (5 out of 7) consider the current trends are sufficient to ensure that CRAs incorporate relevant ESG factors in credit ratings. Only one CRA sees the need for further detailing of ESMA Guidelines on disclosures.

Majority of respondents (55%) to the targeted consultation were in favour of the soft approach, as compared with only 24% in the favour of the legislative intervention

On the other hand, the ECB has expressed its support for a regulatory intervention, as it relies on comprehensive credit ratings for its monetary policy and financial stability mandates.

Conclusions

Incorporation of ESG factors in methodologies

Taking into account the legal framework, where the requirements in regard to methodologies are specified in RTS and based on ESMA advice, it seems sufficient to review RTS on methodologies to ensure more rigorous incorporation of ESG factors in the methodologies. This will be in line with the proportionality principles. The preferred policy option would therefore be Option 2.

Disclosure requirements

In order to introduce stronger requirements, the Commission could explore using the delegation foreseen in the CRA Regulation to amend and update the Annex I. The preferred policy option would therefore be Option 1.

Annex 9: Options discarded at early stage

The following options have been considered but discarded at an early stage for the reasons described.

·Regulatory treatment of ESG rating providers

Description of option:

Require registration and supervision at the national level for providers falling into the scope.

Reasons for discarding this option:

The market of ESG rating providers is global. The biggest providers operate in a number of Member States. They may be headquatered in one Member State and have subsidiaries in others, or may be headquartered outside the EU and have EU subsidiaries.

Evidence gathered by ESMA indicates that among those located in the EU, the majority of entities had their headquarters or legal domicile in three Member States: Germany, Italy and France. In eight Member States there are either one or two providers. The remaining Member States host no providers (see below diagrams).

While the Commission is pursuing the European Green Deal and developing European Sustainability Reporting Standards, it would highly be inefficient for each Member State to set up a national supervisory framework for the supervision of one or two entities, that are often operating on a European, if not global scale.

Any national system would be inappropriate due to the fact that some providers have subsidiaries in more than one Member State, as well as provide services cross border, the national competent authorities would need to closely co-operate to ensure coherent enforcement. Therefore, any system of national supervision would need to be supplemented by the EU level co-ordination. As a result, administrative costs would be even higher, as in addition to costs of setting up and running national supervision, there would be the costs of an EU level co-ordination mechanism.

·Harmonising methodologies of ESG rating providers

Description of option:

This option envisages the harmonisation of methodologies used by ESG rating providers, and they would agree on common definitions and objectives.

Reasons for discarding this option:

Firstly, the results of the targeted consultation show 65% of respondents welcome variety of types of ESG ratings. This has been confirmed during exchanges with various stakeholders.

Next, currently ESG rating providers use own classifications and definitions of Environmental, Social and Governance factors. There are no global or european standards developed for the use by ESG rating providers.. Although there are some classification systems developed recently for corporates, like the Taxonomy Regulation, or being developed currently, like European Sustainability Reporting Standards, they have been developed with different objectives in mind. The Climate Delegated Act covers 2 of 6 environmental objectives related to impacts (climate change mitigation and adaptation) and only 9 sectors (forestry, manufacturing, energy, water and waste, transport, buildings, ICT, research and financial services). The European Sustainability Reporting Standards will be addressed to corporates, whereas data reported by corporates constitutes only a small fraction of data used by ESG rating providers. Anecdotal evidence suggests that ESG data reported by corporates constitutes circa 10% of all data points used by ESG rating providers. ESG ratings providers use data from many sources, not only coporates, so there would be the need to assess the suitability of ESRS for the use by ESG rating providers.

Thirdly, ESG ratings pursue different objectives and cover different areas. Subsequently, the harmonisation would require development of the classification system of objectives. The next dimension that would need to be harmonised is the weightings of E, S and G factors in the case of aggregated ratings, and separately methodologies for disaggregated or topical ratings. As a result several types of methodologies would need to be developed – for different classes (topics), objectives, and types – aggregated or disaggregated and topical. The last remaining dimension would be related to rating method- opinion vs algorithmically score.

The next important question is who would have such an expertise and reputation as to judge which approach, which methodology is best, given that the current variety of approaches reflects differences in views on how to assess the complexity of ESG factors. If decisions on the content of methodologies were to be taken at the political level, that could impact the reliability of methodologies and ratings. For that reason, in a similar market of credit rating, CRAs are guaranteed the independence of methodologies they use.

The situation may change in a near future, as there are a few ongoing initiatives at the European and global levels. The work of EFRAG on the European Sustainability Reporting Standards and the work initiated by the International Sustainability Standards Board to deliver a comprehensive sustainability-related disclosure standards may foster more convergence. The harmonisation and standardisation of sustainability data reported by companies is expected to have positive impact. Currently, ESG rating providers face problems due to the gaps in the data reported by corporates, different formatting and different classification used by reporting companies. As a result ESG rating providers need to standardise data themselves and look for the data from different sources as well as use estimations when no data is available. This leads to differences already at the input level. If input data is harmonised, the results should be more comparable.

Next, users welcome diversity in products that are offered and are not calling for the harmonisation of methodologies. They also welcome different objectives pursued by ESG ratings. The variety of objectives and methodologies is an answer to different and changing needs of users. The results of the targeted consultation show 65% of respondents welcome variety of types of ESG ratings.

Finally, defining the objectives and methodologies would affect innovation in this market and new ways of measuring ESG risks and impacts.

·Setting minimum requirements on the content of ESG ratings

Description of option:

The proposal could specify the minimum content of ESG ratings

Reasons for discarding this option:fann

As explained above, users welcome the diversity of asessments and do not call for regulating the content of ESG ratings (the results of the targeted consultation show 65% of respondents welcome variety of types of ESG ratings). In addition, the ESG ratings pursue a variety of objectives and respond to demands from the professional investors. As explained above, the existing classification systems has been developed for corporates and not for the use by ESG rating providers. In addition, such an option would affect the inovation in this market.

·Requiring the use of detailed templates for disclosure requirements

Description of option:

The Commission could propose the use of disclosure templates by ESG rating providers. This would allow for additional comparability of information by users of ESG ratings.

Reasons for discarding this option:

The market of ESG ratings is a new and growing market with providers offering variety of assessments/opinions. Users welcome that diversity and are in general sophisticated institutional investors. The introduction of standardised templates for disclosures might have some negative effects on innovation and development as it would force providers to limit to providing information on given elements and not disclose additional information that in fact is useful to users of ESG ratings. There is a risk that the nature and substance of the methodology would be lost. Templates would need to not be giving an exhaustive list of information to be disclosed and should also take into account the different business models and objectives pursued by ESG ratings. Rating methodologies by their nature differ from one provider to another and information to be made available should inform of the specificities. Therefore it is likely that it would require development of a number of different templates, which would likely undermine the overall objective of this option. Simplified templates would result in omitting specific information, whereas detailed and long templates would be burdensome.

The CRA Regulation also does not provide for requirements on templates. The Benchmark Regulation does require the use of templates for what concerns the explanation in the benchmark statement and benchmark methodology on how ESG factors are reflected. It is therefore in a clearly defined situation, contrary to the market of ESG ratings which is very diversified and complex. The template also gives the possibility for benchmark administrators to provide for a hyperlink where all details can be found on their website.

Requiring the use of templates would create additional costs and burdens for smaller providers and new entrants. ESG rating providers would need to implement new procedures, and potentially renew or adapt their existing IT and reporting infrastructure.

Annex 10 Additional explanations and changes introduced following the Opinion of the Regulatory Scrutiny Board

Question 1

How do investors use ESG ratings in investment decisions? How do the ESG ratings affect the allocation of capital ?

Recent years have seen an important growth in sustainable investing, and specifically in the use of ESG ratings, and this sector is expected to continue to grow in the coming years (studies find annual growth of ESG data (including ESG ratings) to be 28% 228 ). Almost all respondents to the Commission targeted consultation, from investors to issuers and ESG rating providers also mentioned the growing importance of ESG ratings in the market for sustainable investing. This is driven by increased demand for ESG investments and regulatory obligations.

ESG ratings are used by professional investors 229 who wish to incorporate ESG factors into their investment decision-making processes, mainly by screening or assessing companies in their various funds and portfolios. ESG ratings are also used as a framework for corporate engagement and stewardship.

The Study 230 commissioned by DG FISMA found that ESG ratings are the most frequently referenced source of information that professional investors rely on to assess ESG performance (55%). Another survey conducted by Ninety One 231 found that 88% of investment professionals use third-party ESG ratings as a part of their investment process, with 92% expecting to do so in the future.

In the Commission targeted consultation supporting this initiative, investors were asked a question on how decisive ESG ratings are in their decision processes. They indicated that ESG ratings are important, used mainly as a starting point in their analysis but also further down in their decision process. They highlighted that they also use other, similar, ESG tools to complement their analysis such as controversy scores or assessments, exclusion lists but also United Nations Sustainable Development Goals alignment assessments. In follow-up meetings, they stressed that one of the reasons for using these other means is the lack of trust in and clarity of the ESG ratings’ market. In their responses to the consultation but also in follow-up meetings, investors also highlighted that ESG ratings are important for assessing risks and opportunities.

As a result, ESG ratings have an important effect on the allocation of capital by investors to companies. ESG ratings are one of the determining factors of whether company shares will be included in ESG-themed mutual or exchange-traded funds.

We have provided additional explanations, in the Problem Definition and Annex 2 on the ESG ratings ecosystem, on the concrete use of ESG ratings by investors from responses to the Commission consultation and other sources, and on the importance of ESG ratings in the investment decision-making process. We will also add additional data sources in these two parts on the expected growth of the ESG ratings market.

Question 2

Are there viable options on the geographical scope, type of ESG rating provider and on defining (or not defining) ESG ratings? Are there alternative combinations of the options on regulatory treatment and transparency beyond the ones presented?

(1)Geographical scope

We have analysed alternative options on the geographical scope, namely the differentiation of requirements for the entities located in the EU as opposed to those located outside the EU. We have concluded that the option limiting the geographical scope to the territory of the EU would lead not only to an unlevel playing field and a competitive disadvantage for EU providers, but also would not reach the objectives of the initiative, and as a result doesn’t constitute a viable option.

As explained in the Impact Assessment, European providers are smaller in size, and yet need to compete with larger non-EU providers. Many of the smaller providers and start-ups have been acquired by larger providers as the market is undergoing consolidation. If, in addition to the ongoing consolidation, only European providers were to be subject to new requirements, then they would face additional costs and would be placed at a disadvantage vis-à-vis non-EU providers. At the same time, users of ESG ratings would continue to face problems of lack of transparency as well as uncertainty as to the quality and reliability of ESG ratings, as large non-EU providers would not be subject to the requirements.

A similar approach has been taken in other financial services legislation, including in the area of sustainable finance. For example, in the Credit Rating Agencies Regulation, where the non-EU CRAs that want their credit ratings to be used in the EU need to meet the requirements of EU legislation.

Summing up, we have concluded that there is no other viable option than the one presented in the Impact Assessment, to create a level playing field for all ESG rating providers operating in the EU. In order to achieve that, the initiative cannot leave out any providers, no matter where they are located, as long as they provide ESG ratings to European regulated financial institutions and other regulated financial market participants as well as to European non-financial companies.

(2)Type of ESG rating providers

As explained in the Impact Assessment, there are two major groups of ESG ratings providers: specialised entities providing ESG ratings to professional investors and companies, as well as financial institutions developing ESG ratings for their own purposes, principally for internal analysis or marketing. The latter use their own proprietary models and don’t provide ESG ratings commercially to other entities. Therefore, the key problem that this initiative aims to address (i.e. information asymmetry between providers and users of ESG ratings) does not materialize in their case. Furthermore, financial institutions are regulated entities and are already subject to sectoral legislation.

Therefore, the development of ESG ratings for own purposes is out of the scope of this analysis. The focus of the initiative is on the specialised entities whose business model is to develop ESG ratings and provide it to professional investors as well as companies.

We have clarified in the IA that the development of ESG ratings by regulated financial institutions for own purposes is out of scope of this initiative. Therefore we do not see any alternative options related to the type of ESG rating providers.

(3)Defining ESG ratings

The intention of this initiative is indeed to introduce a clear definition of what is to be understood as an ESG rating. It would be primarily based on the definition from IOSCO’s set of recommendations published in November 2021, where ESG ratings are referred to as “the broad spectrum of ratings products that are marketed as providing an opinion regarding an entity, a financial instrument or a product, a company’s ESG profile or characteristics or exposure to ESG, climatic or environmental risks or impact on society and the environment that are issued using a defined ranking system of rating categories, whether or not these are explicitly labelled as ‘ESG ratings’. It is crucial that the definition recognises the specificities and diversity/complexity of the ESG ratings market: ESG ratings can differ in their processes (involvement of an analyst or pure Artificial Intelligence / algorithmic driven), in their scope (e.g. looking into the full ESG spectrum or only a subset) and in what they measure (risks or impacts).

We have clarified that we will introduce a definition of ESG ratings in the legislation.

The introduction of a definition of ESG ratings is necessary to define the scope of legislation and application of its requirements. This will bring further clarity to all market participants and more trust in this market as only ESG ratings subject to authorisation can be marketed in the EU.

(4)Alternative combinations of options

For example, a combination of the light regime (option 1 on a code of conduct or 2 on light registration) with the transparency requirements would not be effective, as in the case of options 1 and 2, no continuous supervision is foreseen. Absent ongoing supervision, the efficiency of the transparency requirements would be very limited. Option 1 (code of conduct) cannot be combined with continuous supervision, as codes of conduct are by nature voluntary. The positive impacts of the voluntary ESMA Guidelines on CRAs stem from the fact that ESMA Guidelines supplement existing level 1 and level 2 legislation as well as from the fact that ESMA has a range of supervisory tools in relation to CRAs. ESMA Guidelines are part of a comprehensive legal and supervisory framework. A soft approach towards ESG rating providers, either in the form of a recommendation or guidelines could not have similar effect as ESMA Guidelines supplementing level 1 and level 2 legislation.

We have added explanation that in order not to multiply the combination of options, we have not analysed those combinations that are clearly ineffective and would not have much impact.

Question 3

How is coherence between dealing with ESG factors by credit rating agencies and the ESG ratings ensured?

As explained in the Impact Assessment, credit ratings and ESG ratings play different roles and serve different purposes. Credit Ratings assess the risk of default of a company and should capture all the risks that may affect creditworthiness - including sustainability risks where relevant. Credit ratings are used by investors to calibrate portfolio risk profiles (including for investment products) as well as for the calculation of prudential requirements under the EU’s Capital Requirements Regulation (CRR) and Solvency II, and play an important role in ensuring financial stability. Credit ratings are also used by the European Central Bank as one element in its open market operations.

The initiative on ESG ratings follows a similar but lighter approach – the requirement of the registration/authorisation for providers of ESG ratings, supplemented by organisational requirement as well as ongoing supervision by ESMA, and the introduction of transparency requirements in relation to methodologies.

Given the less critical role ESG ratings play as regards financial stability, and given the fact that the market is still growing and undergoing changes, the Impact Assessment proposes proportionate - principle-based - organisational requirements, rather than prescriptive requirements specified in delegated acts.

In order to ensure transparency, the Impact Assessment proposes to develop a definition of ESG ratings and introduce disclosure requirements in relation to the objectives of ESG ratings (ESG ratings are much more diversified as compared to credit ratings which are precisely defined), as well as in relation to methodologies. Given the different business model applied by the majority of ESG rating providers (subscriber pay model) as compared with the business model of CRAs (issuer pay model), the approach to the transparency of the product, i.e. rating, cannot be the same. Whereas the CRA Regulation requires public disclosure of credit ratings, such a requirement cannot be introduced for ESG ratings that are provided for a fee to subscribers.

Given the global role of ESG rating providers, similar to that played by credit rating agencies, and given ESMA experience with the supervision of the latter, the Impact Assessment proposes the supervision to be carried out by the same authority, i.e. by ESMA. This would facilitate coherent supervision of both types of entities, credit rating agencies and ESG rating providers.

Question 4

What is the evidence underpinning the impact analysis?

(1)Clarification of evidence used in the Impact Assessment

We have gathered a significant amount of evidence from different sources, including a Study commissioned by DG FISMA, a Commission consultation but also outreach activities and exchanges with stakeholders. ESMA also supported the Commission work via a call for evidence published in 2022 and provided a mapping of ESG rating providers operating in the EU. We also analysed existing international recommendations and codes for ESG rating providers, such as the IOSCO recommendations. 168 organisations and persons responded to the targeted consultation supporting this initiative, mainly investors, ESG rating providers and listed companies.

In order to understand better the problems which this market faces in practice, and following the targeted consultation, we have undertaken a significant review of academic literature 232 , market analysis and outreach activities with a large number of key stakeholders in the ESG ratings market. We held bilateral meetings with various stakeholders between April and October 2022, including meetings with 14 different ESG rating providers, meetings with users or associations representing, amongst others, rating users, meetings with other stakeholders such as academics, NGOs, public bodies and supervisory authorities.

We have carried out an additional review and streamlining of the evidence gathered from our research and outreach activities, and we have included additional references to such evidence throughout the Impact Assessment where necessary, including from ESMA’s mapping exercise and responses to the consultation, in particular in the impacts section.

(2)Assessment of impacts

The Impact Assessment aims to capture and assess the impacts stemming from both Level 1 and the introduction of delegated acts / level 2 measures. Level 2 measures will operationalise the requirements relating to authorisation and transparency to the public.

The Impact Assessment states that ESG ratings providers will be required to be authorised by ESMA in order to provide ESG ratings to European regulated financial institutions and companies within the scope of the CSRD who solicit ESG ratings. The key requirements for authorisation will be specified in Level 1, which will be further operationalised by Level 2 measures to be developed by ESMA.

Since Level 1 measures will be principle-based and details will only be set at Level 2 (regarding transparency and authorisation only), some of the cost figures are based on estimates or only indicate possible ranges.

When developing recommendations for Level 2 measures, ESMA would be asked to further assess the impacts, including costs, in order to make these assessments more precise. The Commission will carefully monitor ESMA’s work.

We have further clarified in the Impact Assessment that an additional cost assessment for the authorisation requirement will be done by ESMA when drafting the Level 2 measures.

(3)Main factors driving the costs

In the Impact Assessment, two types of costs are analysed (1) related to authorisation and supervision and (2) related to transparency measures. Authorisation and supervision requirements are the main drivers of costs, accounting for about 90% of the costs of this initiative, while the costs of complying with transparency measures are expected to be significantly less burdensome for the sector.

One-off costs estimated in the Impact Assessment are mostly related to authorisation for both smaller and larger ESG rating providers. Further one-off adjustment costs are likely to arise in relation to the need to upgrade IT systems and strengthen internal procedures, but estimates are difficult to obtain.

In terms of supervisory fees, they would be borne primarily by large providers (around 94% of the total). Smaller entities would bear only about 5% of the costs of supervision for ESMA. In order to prevent large cost impacts on smaller providers, the rate of supervisory fees would be very progressive, ranging from 0.37% to 2.4% of annual revenue.

(4)Market growth and competition

Recent years, even without any regulatory framework, have seen a consolidation of the ESG ratings market, with mostly large – non-EU - players acquiring smaller and specialized market players. According to the Russell Investment 2020 survey, the three largest ESG rating and data providers globally are MSCI ESG Research (42% of asset managers surveyed subscribe to their services), Sustainalytics (37%) and ISS-oekom (31%). Consolidation is to be expected for this type of market given its large economies of scale.

Regulation will cause additional costs for ESG rating providers, but measures have been identified to make sure that they are proportionate and that any distorting effect will be limited. We held numerous meetings with stakeholders and in particular smaller ESG rating providers to discuss potential implications of the initiative on their activities. In the assessment of options and in Annex 4 on the SME test, we are proposing a number of mitigating measures (transition period, adjustment of supervisory fees, proportionate supervision). Nevertheless, the costs of regulation could increase barriers to entry or have an additional effect on consolidation.

At the same time effective regulation that ensures a high level of quality can also help competition. It avoids situations where one player may operate on the basis of lower costs or quality, can exploit the lack of transparency and still attract customers, while others invest a lot, resulting in an unlevel playing field.

And the introduction of a regulatory framework is expected to lead to reduced costs for investors. The lack of trust has led to investors additional spending on due diligence activities, checking the data from ESG rating providers and complementing these with in-house ESG ratings. Investors will be able to make better informed decisions as they will have more information on how ESG ratings are computed and what they measure. They will also have more confidence in the overall operations of ESG rating providers.

We have added further evidence from responses to the Commission targeted consultation and exchanges with stakeholders as to how they expect the competition to evolve in the coming years.

Changes introduced in the Impact Assessment:

·Scale of the problems identified: we have provides for additional information and sources on the problems identified and how important they are for investors, for example ‘ratings that don’t rate - the subjective world of ESG ratings agencies’, ‘ESG investing, issues and challenges’. The problem definition section already contains a large number of references to literature and other sources of information showing the significance of the problems.

·Allocation of capital: we have added references to additional sources and literature providing information on the use of ESG ratings and on how important they are for professional investors in their investment decision making process and their implications for companies (e.g. survey conducted by Ninety One finding that 88% of investment professionals use third-party ESG ratings as a part of their investment process, with 92% expecting to do so in the future).

·Success of this initiative (monitoring): we have added further concrete KPIs and specified who should be responsible for carrying out the surveys and check the progress in the implementation of this framework.

·Option combinations: we have emphasised and highlighted the key elements underlying the various options. We will also add that we have pre-analysed the possibility of alternative combinations of the options on the regulatory treatment and the options on transparency to illustrate why we have decided to pursue our analysis only for the most feasible and impactful ones and combination thereof, which allows us at the same time keeping the Impact Assessment as simple as possible.

·Impacts: we will clarify that the analysis in the Impact Assessment aims to capture and assess the impacts stemming from both Level 1 and the introduction of delegated acts / level 2 measures. Level 2 measures, to be developed by ESMA, will operationalise the requirements relating to authorisation and transparency to the public and, at this stage, ESMA would be asked to further assess the impacts, including costs, to further narrow down the cost ranges provided in the IA.

(1)

Strategy for financing the transition to a sustainable economy , COM/2021/390 final.

(2)

COM/2018/097 final.

(3)

European Green Deal .

(4)

OECD (2020), OECD Business and Finance Outlook 2020: Sustainable and Resilient Finance, OECD Publishing , Paris.

(5)

European Investment Funds Study 2022 - Sustainable funds: from niche to mainstream .

(6)

See IOSCO set of recommendations , definition of ESG ratings : “the broad spectrum of ratings products that are marketed as providing an opinion regarding an entity, a financial instrument or a product, a company’s ESG profile or characteristics or exposure to ESG, climatic or environmental risks or impact on society and the environment that are issued using a defined ranking system of rating categories”.

(7)

Regulation (EU) 2019/2088 on sustainability‐related disclosures in the financial services sector, OJ L 317, 9.12.2019, p. 1–16.

(8)

Regulation (EU) 2020/852 on the establishment of a framework to facilitate sustainable investment, OJ L 198, 22.6.2020, p. 13–43.

(9)

These building blocks are: 1) a classification system, or ‘taxonomy’, of sustainable activities, 2) a disclosure framework for non-financial and financial companies, and 3) investment tools, including benchmarks, standards and labels.

(10)

Directive (EU) 2022/2464 as regards corporate sustainability reporting .

(11)

ESMA Call for Evidence on ESG Ratings

(12)

Environmentalanalyst, ESG data market could exceed $1.3bn in 2022 , annual growth of 28% in the ESG data market (which includes ESG ratings)

(13)

Environmental Finance, ESG data market (including ESG ratings) ‘to more than double, to $5bn' .

(14)

According to the OECD ESG Investing: Practices, Progress and Challenges , “in the US, the current level of ESG investing is now over 20% of all professionally managed assets, at over USD 11 trillion”, while in Europe, “industry data related to a broader range of ESG practices suggests the level is over USD 17 trillion.” According to a report in the Financial Times , research carried out by PwC has predicted that “ESG funds will experience a more than threefold jump in assets by 2025, increasing their share of the European fund sector from 15 per cent to 57 per cent.”.

(15)

Global Sustainable Investment Review 2020 .

(16)

See Ninety One survey here .

(17)

Commission Strategi Foresight Report 2022

(18)

IOSCO Recommendations .

(19)

The United Kingdom is working on the introduction of a code of conduct (as a first step) and a regulatory framework with oversight of ESG rating providers in the longer term. India also published a consultation in January 2022, to establish a regulatory framework for ESG rating providers. The aim is to improve transparency and comparability concerning the metrics used in ratings. Japan was the first country to introduce a (draft) code of conduct for ESG data and ratings providers in July 2022.

(20)

For example, see ‘Ratings that don’t rate - the subjective world of ESG ratings agencies’

(21)

For instance see a Position paper from the French and Dutch financial market authorities: Call for a European Regulation for the provision of ESG data, ratings, and related services .

(22)

Outreach activities, responses to the consultation from the Commission.

(23)

CRAs and sustainability

(24)

Renewed sustainable finance strategy

(25)

ESG disclosures in Credit Rating Agencies

(26)

"What is the value of ESG ratings for responsible investors in allocating capital towards sustainable companies? Evidence from European institutional investors. " Huys, Mona .

(27)

See the Study, p. 26.

(28)

‘Deep Data Delivery Standard’ .

(29)

(30)

See Commission targeted consultation here .

(31)

ESMA Call for Evidence .

(32)

IOSCO Report on ESG Ratings and Data Products Providers .

(33)

Marius Banke Stephanie Lenger and Christiane Pott, “ESG Ratings in the Corporate Reporting of DAX40 Companies in Germany: Effects on Market Participants”, 2022.

(34)

Wong, Brackley, & Petroy, “Rate the Raters 2019: Expert Survey Results”, 2019.

(35)

Evidence gathered from responses to the Commission consultation on ESG ratings.

(36)

The Economist, 2019; Financial Times, 2020a,b.

(37)

Mackintosh (2018); Berg et al. (2020).

(38)

Isra Vision, a German manufacturer of camera technology, sued ISS ESG, for assigning it the lowest “D-” ESG rating. Consequently, the Munich District Court banned ISS ESG from rating the ESG of Isra Vision.
“Heavy flows into ESG funds raise questions over ratings” by Billy Nauman, Financial Times, March 4, 2020.

(39)

Rate the Raters 2020: Investor Survey and Interview Results , p. 20.

(40)

See for example ISS Data verification , MSCI data verification process .

(41)

Refinitiv, Environmental, Social and Governance scores .

(42)

Exception those using surveys (e.g. CDP, SAM CSA) that publish specific updates to methodology as well as annual ratings.

(43)

ISS QualityScore FAQ .

(44)

E.g. MSCI ESG General FAQs for Corporate Issuers , Sustainalytics, The esg risk rating: frequently asked questions – for companies .

(45)

E.g. Refinitiv.

(46)

Samuel Drempetic, Christian Klein and Bernhard Zwergel, ‘The Influence of Firm Size on the ESG Score: Corporate Sustainability Ratings Under Review’ , Journal of Business Ethics, 2020.

(47)

N. Tamimi and R. Sebastianelli, ‘Transparency among S&P 500 Companies: An analysis of ESG Disclosure Scores’, Management Decision, Volume 55, Issue 3, January 2017.

(48)

Peter Cripps, 'ESG Data Files - part one, continued: reported data,' Analysis, Environmental Finance, 3 July 2019.

(49)

Extel and SRI-Connect, Independent Research in Responsible Investment Survey 2019, SRI-Connect, 2019. Approximately 50% of participating companies were from EU member states.

(50)

See for example ‘ The ESG Risk Rating: - Sustainalytics’ .

(51)

MSCI ESG General FAQs for Corporate Issuers .

(52)

Huys, Mona, op.cit.

(53)

IOSCO Report, p.27.

(54)

Tang, Yan and Yao, 2022, op.cit.

(55)

ESG ratings should be treated as a starting point and should not be taken for granted (Bos, 2020).

(56)

Tang, Yan and Yao, 2022, op.cit. :“The issue of limited transparency in the ESG rating processes ranges from indicators and algorithms to the qualitative proprietary assessment techniques applied”.

(57)

IOSCO report, p. 4; stakeholders’ feedback to Commission consultation.

(58)

ESMA letter on ESG ratings, p. 3.

(59)

The results of the targeted consultation show 65% of respondents welcome variety of types of ESG ratings.

(60)

ESMA letter on ESG ratings, p. 18.

(61)

MSCI ESG Methodologies .

(62)

FTSE Russell, ESG Ratings and Data Model Product Overview: Integrating ESG into Investments .

(63)

Sustainalytics ESG Risk Ratings- Methodology Abstract .

(64)

Refinitiv, Environmental, Social and Governance (ESG) Scores from Refinitiv , April 2020.

(65)

https://www.sasb.org/ , https://www.globalreporting.org/ , https://www.fsb-tcfd.org/ .

(66)

CDP, ISS-ESG, SAM CSA, Sustainalytics, and Vigeo Eiris provide highly customized, question-level differences on issues and metrics within their frameworks depending on company sector. MSCI, ISS QualityScore, FTSE Russell and Refinitiv include the same starting universe of topics or metrics with refinements based on industry.

(67)

Berg, Koelbel, and Rigobon, “Aggregate Confusion: The Divergence of ESG Ratings”, Forthcoming Review of Finance, 2019.

(68)

For example, MSCI ESG Controversies and Global Norms Methodology .

(69)

Leaders Arena, why are ESG controversies so controversial & what can companies do about them?

“As with ESG Ratings in general, the ESG controversies approaches by leading providers table highlights key differences between agencies. Simply put, what is considered as a controversy by one agency, may not be by another, and severity grading also varies. It is also often unclear the impact of a controversy on the overall company ESG score”.

(70)

Sustainalytics ESG Ratings , CDP Scores , Standard Ethics Corporate Rating , ESG Ratings Climate Search Tool .

(71)

Giese et al., 2019; Ratsimiveh et al., 2020, Kotsantonis and Serafeim, 2019.

(72)

Melas, Nagy, and Kulkarni, “NN Investment Partners and ECCE’s study”, 2016.

(73)

Samuel Drempetic et al. (2020), op.cit.

(74)

Ibid.

(75)

Matthias Born, Katharina Raatz, Dr Rupini Deepa Rajagopalan and Justus Schirmacher, ‘ESG Ratings: The Small and Mid-cap Conundrum’ , Research paper, Berenberg, May 2020.

(76)

Gyonyorova et al., 2021.

(77)

Aaron Chatterji, Rodolphe Durand, David Levine, and Samuel Touboul, “Do Ratings of Firms Converge? Implications for Strategy Research”, Strategic Management Journal, 2016.

(78)

Hawley, 2017.

(79)

Chatterji, Durand, Levine and Touboul, 2016, op. cit.

(80)

Fabio Alessandrini and Eric Jondeau, “ESG Investing: From Sin Stocks to Smart Beta”, Journal of Portfolio Management, 2020. “These scores may suffer from industry biases. For instance, the energy sector is given a particularly low score on average according to the environmental criterion”, p. 6.

(81)

Dimson, Marsh and Staunton, 2020, Lopez, Contreras and Bendix, 2020.

(82)

Commission targeted consultation - Lack of transparency and clarity most important factor for low-correlation.

(83)

ESG investing, issues and challenges

(84)

Berg, Koelbel, & Rigobon, 2019, op. cit.

(85)

Chatterji, Durand, Levine, & Touboul, 2016, op.cit.

(86)

E.g. Bloomberg

(87)

See IOSCO Report , p. 10.

(88)

See information from Ecovadis on ESG ratings of supply chain .

(89)

2DegreesInvesting - Do we speak the same language? A market survey on the future of ESG ratings .

(90)

‘The ESG Mirage’ , Bloomberg, 2021, “This critical feature of the ESG system, which flips the very notion of sustainable investing on its head for many investors, can be seen repeatedly in thousands of pages of MSCI’s rating reports.”.

(91)

ESMA Call for Evidence, page 20, section 37, 63% of respondents argued that the level of methodological transparency is still not satisfactory.

(92)

See the Study, p. 98.

(93)

Subject to Regulation (EC) No 1060/2009 on credit rating agencies, OJ L 302 17.11.2009, p. 1.

(94)

Subject to Regulation (EU) 2016/1011, OJ L 171, 29.6.2016, p. 1-65.

(95)

Subject to Directive 2011/61/EU, OH L 174, 1.7.2011, p. 1-73.

(96)

Tang, Yan, and Yao, 2022, op. cit.

(97)

Example of ESG rating provider offering consulting services, ISS Corporate Solutions, “Sustainability Solutions,” .

(98)

For example, MSCI, How MSCI manages conflicts of interest .

(99)

See definition of Chinese Wall on Investopedia .

(100)

Sustainalytics, managing potential institutional conflicts of interest policy abstract , 6 pages.

(101)

ESMA Call for Evidence , p. 17, Section 25.

(102)

For example, MSCI indicates that the ‘fees for each of MSCI ESG Research’s services range from $5,000 to $2,000,000 per year, depending on the specific services selected as well as the business units, location or use case.’.

(103)

See information from MSCI here .

(104)

See IOSCO Environmental, Social and Governance (ESG) Ratings and Data Products Providers Final Report, p. 37.

(105)

IRRI Survey 2019 .

(106)

http://www.gsi-alliance.org/wp-content/uploads/2021/08/GSIR-20201.pdf

(107)

CFA Certificate in ESG Investing curriculum Oct. 2021 version

(108)

Dane Christensen, George Serafeim, Anywhere Sikochi, “ Why is Corporate Virtue in the Eye of The Beholder? The Case of ESG Ratings ” , Harvard Law School Forum on Corporate Governance, 2021.

(109)

UNPRI - ESG rating disagreements: why is corporate virtue so subjective? .

(110)

See further information here .

(111)

Recently, ESG ratings came under an increased scrutiny due to critical media reports (e.g. Greenwashing accusations against French care home provider Orpéa).

(112)

Cam Simpson, Akshat Rathi, and Saijel Kishan, “ The ESG Mirage ”, Bloomberg, December 10, 2021.

(113)

See footnote above. “BlackRock and other investment funds use ESG ratings to justify a “sustainable” label on stock and bond funds. MSCI ESG Benchmark is constructed on the basis of MSCI ESG ratings. MSCI ESG ratings assess only ESG-related risk and opportunities (finanical materiality) and measure companies not against universal standards but against their industry peers. This means that all invesments labeled sustainable on the basis of MSCI ESG ratings or any passive funds using MSCI ESG Index direct the flow of capital into companies that are best among their peers in managing the ESG related risks and opprtunities to themselves, with no consideration of their impact on the environment or society.”.

(114)

What Should Everyone Know about ESG Ratings? Kornelia Fabisika and Sophie-Dorothee Rotermund

(115)

IOSCO - Environmental, Social and Governance (ESG) Ratings and Data Products Providers .

(116)

ESMA has adopted a risk-based and data-driven approach when conducting supervision and prioritises its supervisory activities according to the level of risk identified and outcomes it targets. A risk analysis is used to identify areas of possible non-compliance and to assess the potential significance of the issue. Risks are identified by collecting and assessing information from different sources like periodic and ad hoc notifications submitted by the entities, registration application files, meetings/calls with entities, data analysis, on-site inspections/investigations etc. To identify key priority areas for supervision, ESMA considers the result of its supervisory risk assessment, and key market trends in the respective industries and capital markets more broadly.

(117)

The precise period of time is yet to be determined, but it would be designed to allow sufficient time, that could also take account of the size of market players.

(118)

Directive (EU) 2016/943 of the European Parliament and of the Council of 8 June 2016 on the protection of undisclosed know-how and business information (trade secrets) against their unlawful acquisition, use and disclosure, OJ L 157, 15.6.2016, p. 1–18.

(119)

ESMA Call for Evidence on ESG Ratings .

(120)

OJ L 174, 1.7.2011, p. 1–73.

(121)

OJ L 257, 28.8.2014, p. 186–213.

(122)

OJ L 317, 9.12.2019, p. 1–16.

(123)

Regulation (EC) No 1060/2009 of the European Parliament and of the Council of 16 September 2009 on credit rating agencies

(124)

Commission Delegated Regulation (EU) No 447/2012 of 21 March 2012 supplementing Regulation (EC) No 1060/2009 of the European Parliament and of the Council on credit rating agencies by laying down regulatory technical standards for the assessment of compliance of credit rating methodologies

(125)

Idem

(126)

Idem

(127)

Section D of Annex I or CRA Regulation – Rules on on the presentation of credit ratings and rating outlooks

(128)

Environmental, Social and Governance (ESG) Ratings and Data Products Providers - Final Report .

(129)

Van Duuren, Plantinga, & Scholtens, 2016.

(130)

Van Duuren, Plantinga, & Scholtens, 2016.

(131)

As confirmed by a majority of interviewed stakeholders. There were limited outliers, which mainly pointed at a situation where establishment of a legal entity in the EU would be required.

(132)

No precise figure was received from ESMA for the size of fees, but as the range of ESMA’s activity is much more limited compared to Option 3, it can be reasoned that its budget needs and supervisory fees covering them would be also significantly smaller (especially when levied on the market actors of the same size).

(133)

Van Duuren, Plantinga, & Scholtens, 2016.

(134)

As confirmed by a majority of interviewed ESG ratings providers, both through interviews and written feedback.

(135)

Commission services are reflecting on possible measures to mitigate the impact of the authorisation requirement, notably on smaller players. To ensure market players can continue operating, it is foreseen that they will be allowed to stay in the market after notifying ESMA, on the condition that they would complete the authorisation process within a pre-deterrmined time period. This would allow both market players (and the supervisor) to spread the costs in time and become compliant with the rules gradually.

(136)

Costs of provision of required documents and interaction with the supervisory authority during the supervisory review process.

(137)

The ongoing costs for smaller providers identified here are similar in size to the estimated compliance costs for external verifiers in the impact assessment on the European green bond standard. This also suggests that expected costs are proportionate.

(138)

Relative size of the fees is described under impacts on ESG rating providers. Passing costs of supervision onto supervised entities is a typical practice in the financial sector.

(139)

Based on targeted interviews with asset managers.

(140)

See footnote above.

(141)

EFAMA Asset Management Report 2021 .

(142)

While stakeholder replies varied in this regard, it seems safer to assume that to obtain the total costs of the first year of disclosures, one has to add the ongoing cost element presented below to this one-off cost element.

(143)

The estimate was provided as an aggregate, which does not allow to separate out the cost of supervising disclosures from the cost of supervising operations.

(144)

For instance, one medium-sized asset manager indicated that they currently need up to 6-8 FTEs just to understand the methodologies and data processes behind an ESG ratings, a cost which would be expected to fall significantly with sufficient transparency and clarity. Responses of other asset managers have mostly supported a rather significant effect of more comprehensive disclosures.

(145)

(146)

The estimate was provided as an aggregate, which does not allow to separate out the cost of supervising disclosures from the cost of supervising operations.

(147)

For the first year, it is safer to assume that these costs have to be added together (ongoing + one-off element).

(148)

This will be done in close cooperation with the Platform on Sustainable Finance, which will monitor trends regarding capital flows towards sustainable investments as set out in Article 20 of the Taxonomy Regulation.

(149)

Renewed sustainable finance strategy

(150)

Study on Sustainability Related Ratings, Data and Research .

(151)

IOSCO - Environmental, Social and Governance (ESG) Ratings and Data Products Providers .

(152)

ESG Investing and Climate Transition, OECD .

(153)

ESMA Call for Evidence on Market Characteristics of ESG Rating and Data Providers .

(154)

Consultation on a renewed sustainable finance strategy - summary of responses (europa.eu) .

(155)

Types of market participants: 50 asset managers, 4 NGOs, 11 asset owners, 5 benchmark administrators, 9 data providers, 5 grant funded research providers, 10 CRAs, 24 ESG rating providers, 8 sell-side brokers, 45 experts and 65 companies.

(156)

Study on sustainability-related ratings, data and research .

(157)

As means of clarification, respondents could choose more than one stakeholder group.

(158)

IRRI Survey 2019 .

(159)

Russell Investment 2020 Survey .

(160)

Stephanie Mooji, “ The ESG Rating and Ranking Industry; Vice or Virtue in the Adoption of Responsible Investment? ”, University of Oxford, 2017. Based on interviews with 20 companies in the Netherlands, Germany and the UK, companies receive approximately 30 requests a year on average from ESG rating and ranking providers. And for companies, addressing multiple individual requests for data and information on sustainability-related performance from stakeholders can be time intensive and costly.

(161)

This example is based on the responses of more than 20 asset managers. While most of them only provided qualitative information and there were some outliers, most expected rather large savings from greater transparency on ESG rating methodologies. Indicative expected savings in FTEs from those who provided figures pointed at approximately 1.5-4 fewer FTEs being needed for due diligence on ESG ratings, reflecting the large resources decidated to this presently. In the hypothetical example, we conservatively work with just one third of this estimate to account for smaller or less ESG-focused asset managers who would likely have lower cost savings. We work with an approximate number of asset managers in the EU based on EFAMA and Statista figures. Note that these cost savings can’t be linked to EU legislation and are hence not considered to be relevant for “one in, one out” off-setting.

(162)

Note that in the first years of application, it is possible that we see a temporary decline in the market due to the effect of new rules (adaptation of providers, new costs, etc.).

(163)

This estimate is subject to relatively larger uncertainty as the Commission services have attempted to get estimates for these costs, but only few entities provided specific figures and the intensity of supervision (i.e. number of supervisory reviews in a certain number of years or how many additional documents would be asked for by the supervisor) is not yet fully clear. Some of the figures that have informed this estimate were derived by subtracting estimated cost of disclosure from the total recurring compliance that some providers indicated rather than a self-standing figure for dealing with the supervisor.

(164)

The effort level assumed in this estimate based on stakeholder responses greatly varies by size and corresponds to about 0.8-1.1 FTEs per smaller provider vs almost 4-7 FTEs for larger providers.

(165)

Based on a preliminary assessment provided by ESMA. The figures are subject to uncertainty as ESMA’s estimation only considered providers where revenue information was available. This estimate covers also the cost of enforcing disclosure rules and the cost for ESMA to grant authorisations to market participants, where the latter may be charged through a one-time authorisation fee.

(166)

This figure is a result of estimation by ESMA calibrated based on available data on the market, as the intensity of supervision and hence costs for the supervisor are expected to differ based on provider size, consistent with the risk-based supervision approach. The figure includes labour costs, overheads as well as relevant IT costs. As data on revenues from ESG ratings were available only for some providers, assumptions were used to extrapolate the estimate to the whole market in order to estimate total resource needs for ESMA and the corresponding supervisory fees to cover these costs.

(167)

According to some of the interviewed ESG ratings providers.

(168)

According to some of the interviewed ESG ratings providers.

(169)

We note that the one-off costs as estimated from stakeholder feedback are lower than recurrding costs. Our assumption is that, if we wanted to obtain the total cost of preparing disclosures for the first time, we would need to add the one-off cost element to the recurring cost detailed in the next cell.

(170)

Such assessment had to deal with some conflicting information provided by stakeholders concerning e.g. the ratio of one-off and ongoing costs of disclosure. Some more detailed responses by stakeholders in follow-up to the interviews also revealed certain degree of double-counting. Responses may have been influenced also by complexity of governance of the providers, their product range or business model.

(171)

We used those for labour category “professionals” as this seems to most closely reflect the nature of the work expected to arise from the initiative. The same source was also used for the other estimates, where hours of work or FTEs per year were used and monetised.

(172)

There were several outliers at this stage as well, mostly among larger providers. In some instances, it was however not clear if the estimates do not include double counting between cost of disclosure, cost of ongoing supervision or broader adjustment costs. Where the figure provided corresponded rather to total labour cost increase in a typical year due to the intervention, it was used to derive approximate costs of dealing with a supervisor, where data were more limited (see below).

(173)

This had to be done also since the consultation has not clearly distinguished between registration and authorisation, which has likely contributed to a greater disparity in the responses. Another reason was that a more detailed estimate for costs of registration received as follow-up to an interview revealed assumptions that rather matched the intended contents of Option 2 (authorisation and supervision).

(174)

Christensen, Serafeim and Sikochi. 2021, op.cit.

(175)

Commission Delegated Regulation (EU) 2020/1816 of 17 July 2020 supplementing Regulation (EU) 2016/1011 of the European Parliament and of the Council as regards the explanation in the benchmark statement of how environmental, social and governance factors are reflected in each benchmark provided and published (OJ L 406, 3.12.2020, p. 1).

Commission Delegated Regulation (EU) 2020/1817 of 17 July 2020 supplementing Regulation (EU) 2016/1011 of the European Parliament and of the Council as regards the minimum content of the explanation on how environmental, social and governance factors are reflected in the benchmark methodology (OJ L 406, 3.12.2020, p. 12).

Commission Delegated Regulation (EU) 2020/1818 of 17 July 2020 supplementing Regulation (EU) 2016/1011 of the European Parliament and of the Council as regards minimum standards for EU Climate Transition Benchmarks and EU Paris-aligned Benchmarks (OJ L 406, 3.12.2020, p. 17).

(176)

Based on input provided by respondents the range could be between EUR 68.000-108.000, with smaller providers being closer to the lower range due likely to lower complexity of their business and processes.

(177)

According to ESMA assessement, fees could range between 0.37% (for smaller) and 2.4% (for large providers) of revenue of providers.

(178)

E.g., Carbon4 – climate transition risks

(179)

Based on input provided by ESMA

(180)

ESMA risk-based and data-driven approach when conducting supervision

(181)

Small group of quants develop a data-driven ESG ratings model, which has very broad firm coverage from day one.

(182)

Innovative, but not revenue-producing business that takes more than seven years from founding to finalize its ESG ratings model.

(183)

E.g. Stanford University, ESG ratings: A compass without direction .

(184)

IOSCO, Environmental, Social and Governance (ESG) Ratings and Data Products Providers Final Report , p.47

(185)

E.g. LSEG/Refinitiv, Bloomberg

(186)

E.g. MSCI, S&P

(187)

Cadwalader, Wickersham & Taft LLP, European Union: ESG Ratings: A Call For Greater Transparency And Precision

(188)

IRRI Survey 2019 .

(189)

Commission Delegated Regulation (EU) 2020/1816, OJ L 406, 3.12.2020, p. 1–11.

(190)

ESMA Call for Evidence on ESG Ratings

(191)

E. Avetisyan and K. Hockerts, “The Consolidation of the ESG Rating Industry as an Enactment of Institutional Retrogression”, Business Strategy and the Environment 26, no. 3, March 2017.

(192)

Russell Investment 2020 survey

(193)

E.g. MSCI, S&P, Moody’s

(194)

E.g. Sustainometric, Solactive, RepRisk, Truvalue Labs (prior to October 2020 acquisition by FactSet), Sustainable Value Investors, Carbon4 Finance, ISS (prior to the November 2020 acquisition by Deutsche Börse AG

(195)

E.g. CDP, ShareAction WDI, As You Sow

(196)

The Study noted that for example ISS, MSCI and Vigeo Eiris maintain company portals to engage companies, receive updates and provide company reports and methodologies (in more detail than what is publicly disclosed). This enables companies to review the underlying indicators they are assessed on.

(197)

For example, FTSE Russell’s ESG Ratings and data model are available to subscribers through a web interface .

(198)

Are ESG ratings really necessary? , GreenBiz, Richard Mattison, “If you are a company going through this assessment, you'd log into a portal and see all of this [ESG] information with explanations and help. They get free benchmarking tools to allow them to compare themselves with their peer groups. The companies find this very useful, because it allows them to understand the leading topics of note from our perspective and how they compare with their peers on a number of different elements.".

(199)

See the Study, and also CEPS ‘a critical look at the ESG market’ .

(200)

Reply from AFEP to the Commission Consultation on ESG ratings .

(201)

Jurisdictions looking at these topics include US, Russia, China, Japan, Singapore, India, Brazil, Chile.

(202)

ESG Ratings and Data Products Providers .

(203)

FCA - Enhancing climate-related disclosures by standard listed companies and seeking views on ESG topics in capital markets .

(204)

FCA Perimeter Report 2020/21 .

(205)

Greening Finance - A Roadmap to Sustainable Investing .

(206)

Mobilising Green Investments, March 2023

(207)

FSA - The Code of Conduct for ESG Evaluation and Data Providers .

(208)

Securities and Exchange Board of India -Consultation Paper on Environmental, Social and Governance (ESG) Rating Providers for Securities Markets .

(209)

Deep Data Delivery Standard .

(210)

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:02009R1060-20190101

(211)

https://finance.ec.europa.eu/publications/high-level-expert-group-sustainable-finance-hleg_en

(212)

2018 Sustainable Finance Action Plan

(213)

https://finance.ec.europa.eu/publications/strategy-financing-transition-sustainable-economy_en

(214)

https://www.esma.europa.eu/esmas-activities/sustainable-finance/cras-and-sustainability

(215)

ESMA Technical Advice to the European Commission on Sustainability Considerations in the credit rating market .

(216)

https://www.esma.europa.eu/esmas-activities/sustainable-finance/cras-and-sustainability

(217)

https://www.esma.europa.eu/policy-activities/sustainable-finance/cras-sustainability

(218)

ESMA was mandated to (i) assess the implementation of the updated guidelines aimed at improving disclosure of information on how sustainability factors are taken into account in credit ratings and outlooks; (ii) review how sustainability factors are incorporated by CRAs in their methodologies. ESMA has delivered its assessment of the application of guidelines in September and October 2020. The assessment of the incorporation of ESG factors in methodologies has been concluded in May 2022.

(219)

https://www.esma.europa.eu/sites/default/files/library/newsletter_june_2022.pdf

(220)

https://www.esma.europa.eu/document/text-mining-esg-disclosures-in-rating-agency-press-releases

(221)

Regulation (EC) No 1060/2009 of the European Parliament and of the Council of 16 September 2009 on credit rating agencies

(222)

(223)

Idem

(224)

Idem

(225)

Section D of Annex I or CRA Regulation – Rules on on the presentation of credit ratings and rating outlooks

(226)

Examples from Section D of Annex I of CRA Regulation:

I.A credit rating agency shall ensure that at least:

(a) substantially material sources, including the rated entity or, where appropriate, a related third party, which were used to prepare the credit rating or rating outlook are indicated together with an indication as to whether the credit rating or rating outlook has been disclosed to that rated entity or related third party and amended following that disclosure before being issued;

(b) the principal methodology or version of methodology that was used in determining the rating is clearly indicated, with a reference to its comprehensive description; where the credit rating is based on more than one methodology, or where reference only to the principal methodology might cause investors to overlook other important aspects of the credit rating, including any significant adjustments and deviations, the credit rating agency shall explain this fact in the credit rating and indicate how the different methodologies or these other aspects are taken into account in the credit rating;

(c) the meaning of each rating category, the definition of default or recovery and any appropriate risk warning, including a sensitivity analysis of the relevant key rating assumptions, such as mathematical or correlation assumptions, accompanied by worst-case scenario credit ratings as well as best-case scenario credit ratings are explained;

(d) the date at which the credit rating was first released for distribution and when it was last updated including any rating outlooks is indicated clearly and prominently;

(e) information is given as to whether the credit rating concerns a newly issued financial instrument and whether the credit rating agency is rating the financial instrument for the first time; and

(f) in the case of a rating outlook, the time horizon is provided during which a change in the credit rating is expected.

When publishing credit ratings or rating outlooks, credit rating agencies shall include a reference to the historical default rates published by ESMA in a central repository in accordance with Article 11(2), together with an explanatory statement of the meaning of those default rates.

II.A credit rating agency shall accompany the disclosure of rating methodologies, models and key rating assumptions with guidance which explains assumptions, parameters, limits and uncertainties surrounding the models and rating methodologies used in credit ratings, including simulations of stress scenarios undertaken by the credit rating agency when establishing the credit ratings, credit rating information on cash-flow analysis it has performed or is relying upon and, where applicable, an indication of any expected change in the credit rating. Such guidance shall be clear and easily comprehensible

(227)

Environmental Finance, ESG data market (including ESG ratings) ‘to more than double, to $5bn' .

(228)

ESG ratings are sold by ESG rating providers to professional investors. Retail investors are typically not purchasing or using these, although from exchanges with stakeholders it appears that they start to pay closer attention to ESG ratings and may consider basing their investment decisions on them.

(229)

Study on sustainability-related ratings, data and research

(230)

See Ninety One survey here .

(231)

See Section 2 on problem definition.

Term or acronym	Meaning or definition
CMU	Capital Markets Union, a plan of the European Commission to create a single market for capital. The aim is to get money – investments and savings – flowing across the EU so that it can benefit consumers, investors and companies, regardless of where they are located.
CRA	Credit Rating Agency
CSRD	Corporate Sustainability Reporting Directive
DG FISMA	European Commission Directorate-General for Financial Stability, Financial Services and Capital Markets Union
Double Materiality	Requirement for undertakings to report both on the impacts of the activities of the undertaking on people and the environment, and on how various sustainability matters affect the undertaking.
ESAP	European Single Access Point
ESG	Environmental, Social and Governance
ESMA	European Securities and Markets Authority
EU	European Union
IOSCO	International Organization of Securities Commissions
KPIs	Key Performance Indicators
SFDR	Sustainable Finance Disclosure Regulation
SMEs	Small and medium-sized enterprises
UN	United Nations

Choose the experimental features you want to try