Sustainability Ratings and Benchmarks: How ‘New’ Are They? 

By | March 28, 2023

Sustainable finance is a burgeoning sector, and the jury is still out on whether this is just a bubble, inflated by greenwashing practices, or a genuine step towards better ESG (environmental, social, and governance) conditions – or a combination of the two. Whatever the answer, sustainable finance is likely to remain high on the agenda of many policymakers, particularly in the European Union (EU). 

Despite the specific features of European capital markets, regulatory developments in the EU can be of interest to policymakers, practitioners and scholars in other areas of the world, if only because these and other stakeholders can learn from the successes and the failures that the European endeavor will inevitably bring about. 

The EU has been dealing with the regulation of sustainable finance ever since 2018, when the European Commission published its first action plan on the matter. From that moment on, the EU has witnessed a torrential increase of regulatory measures addressing disparate matters, from conduct-of-business rules to collective asset management, and from accounting principles to banks’ prudential regulation and supervision, to mention just a few.  

This titanic effort has delivered a brand-new area of European law, one whose complexity is comparable to that of traditional financial law. Our paper revolves around the relationship between these two fields and explores their similarities and differences. The analysis addresses just one of the many matters that sustainable finance involves: information intermediaries, with a specific focus on ESG ratings and benchmarks. It does so by addressing two simple research questions: how much does – and should – the law on sustainable finance mirror traditional financial law, when it comes to ESG ratings and benchmarks? This involves at least two sub-questions: (i) How “new” is sustainable finance compared to the “old” world, in the fields of ratings and benchmarks? and (ii) To what extent can the regulatory strategies that traditional financial law developed to support confidence in ratings and benchmarks be exported to the new world of ESG finance?  

Unsurprisingly, ESG ratings and benchmarks have gained traction as salient ingredients for an efficient allocation of funds towards sustainable assets. Just like other information intermediaries, those entities contribute to reducing market failures by simplifying the complexity of a broad set of information into the conciseness of symbols. However, while ratings and benchmarks tackle market failures such as asymmetric information and agency problems, they are in turn subject to their own market failures, mostly of the same kind as those they try to address.  

The relatively new world of ESG indicators displays many similarities with the original markets for ratings and benchmarks, but it also has some distinctive features. One element to consider in this comparison is that market mechanisms may not work in the same way as in traditional finance when sustainability considerations are included in investment decisions and hence into the dynamics of price discovery.  

In answering our research questions, we therefore analyze the market failures that justify the regulation of traditional information intermediaries, and we explore to what extent they also affect their correspondents in the world of sustainable finance. In the EU context, this exercise takes a different nuance when it addressed ESG ratings, where the analysis is essentially normative, compared to ESG benchmarks, where the approach is rather positive. The reason is that, under EU law, preparation and dissemination of ESG ratings are not a regulated activity, while ESG benchmarks qualify as “benchmarks” and are therefore subject to the whole regulatory framework that applies to that kind of indicator (Regulation (EU) 2016/1011 – Benchmark Regulation).  

For benchmarks, EU lawmakers just had to fine-tune the existing framework. In particular, they required benchmark administrators to disclose whether they include sustainability considerations when they determine their indexes. As our paper highlights, the Benchmark Regulation displays several features that make it particularly suitable to sustainable finance. For instance, the Regulation was first adopted in reaction to the LIBOR scandal, which highlighted the key role of input data in the definition of indices, as these are particularly prone to “garbage in, garbage out” dynamics. Here, the Benchmark Regulation identifies as the first-best choice, due to their higher objectivity, data on transactions that have already taken place. When this is not possible – as is typically the case with ESG-related information that does not refer to transactions – a sophisticated system of data governance makes up for the uncertainty on the quality of incoming information. Chief among the components of the legal strategies is the requirement that benchmark administrators adopt specific guidelines to ensure compliance with their methodology (a requisite from which smaller benchmark administrator can opt-out, but subject to the duty to explain the reason). These guidelines define the standards that input data shall satisfy and outline the use of discretion by the administrator when determining the benchmark based on those data (expert judgment; Article 11(1)(c) Benchmark Regulation). In addition to this, benchmark administrators prepare a code of conduct their data contributors shall adopt to maintain adequate levels of quality. This system of regulated self-regulation is particularly suitable for sustainability data, where the reliability of information is often too uncertain and originates from disparate sources that may be hardly regulated with generally applicable rules following a one-size-fits-all approach.  

As the Benchmark Regulation provided good regulatory answers to the general issues affecting the reliability of ESG data, EU lawmakers invested more into the creation of ad hoc labels for groups of benchmarks. Once again, second-degree market failures explain this strategy. As benchmarks themselves may be subject to asymmetric information, defining labels facilitates the aggregation and sharing of information concerning the quality, in terms of ESG impact, of the indices that decide to adopt those labels. The 2018 amendments to the Benchmark Regulation have added two labels characterized by the inclusion of objectives related to carbon emissions in the selection of the underlying assets. The two labels differentiate themselves based on the ambition of the targets, but they also share many features. Albeit with variable intensity, both types of benchmarks aim to reduce greenhouse gas emissions. Therefore, they do not rule out investments in assets that are linked to emissions but mandate an overall level of emissions that is lower than the investible universe and in line with a pre-defined decarbonization trajectory. The most ambitious label, dubbed the “EU Paris-aligned Benchmark,” identifies benchmarks selecting assets that, taken together, should allow the Paris Agreement target to keep the global temperature increase well below 2 °C (and possibly 1.5 °C) above pre-industrial levels to be reached (Article 3(1)(23b) Benchmark Regulation). The intensity of greenhouse gas emissions connected to the underlying assets shall be at least 50 % lower than that of the investable universe. The other ad hoc category of benchmarks falls under the “EU Climate Transition Benchmarks” label. In this case, the intensity of greenhouse gas emissions connected to the underlying assets shall be, less ambitiously, at least 30 % lower than that of the investable universe (Article 9 Regulation (EU) 2020/1818).      

As opposed to the publication of benchmarks, the preparation and dissemination of ESG ratings are not regulated under EU law. This is likely to change soon, as the European Commission has announced its intention to address ESG ratings and similar scoring systems that measure the sustainability of assets and firms. In this regard, the paper suggests that policymakers should be cautious when transposing rules from the “old world” to the “new” one.  

For ESG ratings, credit ratings seem to be the immediate reference for regulators, with the common label of “rating” being rather misleading and leading to the risk of an anchoring effect in the design of new rules. In particular, we explore the risk that the regulation of ESG ratings may incentivize excessive reliance on indicators whose reliability is subject to skepticism. At the moment, we do not see a risk of this kind as regards direct legal references to ESG ratings, as rules are always in clear in making such reference voluntary. However, we fear that a system of authorization and supervision that mirrored the EU regulation on credit rating agencies (Regulation (EC) No 1060/2009 – CRAR) may determine excessive reliance in an indirect form, by providing an implicit certificate of quality to providers of sustainability metrics in spite of the features of such indicators. Due to their multivariate nature, the assessments underlying ESG ratings are often more subjective than those supporting traditional indicators, which may make supervision of methodological robustness more problematic. Therefore, we see a higher risk of regulatory failures connected to authorization and supervision in the new world of sustainability, and we believe that a wiser regulatory strategy would be – at least in its initial steps – one that relies on transparency. A good example comes from the EU framework for financial analysts, which we believe could serve as a blueprint for sustainability ratings as well (Regulation (EU) 2016/958 on Financial Analysts – FAR). One notable element of this framework, which does not include a registration system for the dissemination of investment recommendations, is that it differentiates its requirements depending on the nature of the analysts, thus further reducing barriers to entry. In its baseline form, disclosure duties accompanying the publication of investment recommendations focus on a few essential elements, such as the need to clearly distinguish between fact and interpretations and to report on conflicts of interest (Article 5 FAR). Persons that present themselves on the market as experts must comply with more stringent rules, such as the duty to disclose their valuation or methodology, including the underlying assumptions (Article 4 FAR). Finally, the regime becomes even tighter for financial intermediaries, which must disclose additional items. Among these items, they must include a breakout that shows how many of their recommendations, calculated as a percentage over each class (such as “hold,” “buy” or “sell”), concerns issuers that have received investment or ancillary services from the analyst in the previous year (Articles 6(3) FAR). 

Overall, regulators are in a better position than the players of “The $64,000 Question” game show. The choice is not necessarily between doing nothing and quitting the regulatory game to avoid regulatory failures, on the one hand, versus going all in with the full menu of possible measures, on the other. Other intermediate models are available that can help address the conundrum. In this regard, disclosure appears to be the cheapest and least intrusive form that regulation can adopt in the first place to help tackle asymmetric information and agency costs in the market for sustainability ratings (and benchmarks). We believe other more intrusive measures should only be considered at a later stage, should major concerns on these intermediaries emerge.  


Matteo Gargantini is Assistant Professor of Business Law at the University of Genoa.  

Michele Siri is Full Professor of Business Law and the director of the Jean Monnet Centre of Excellence on European Union Sustainable Finance and Law (EUSFiL) at the University of Genoa.  

This post was adapted from their paper, “Information Intermediaries and Sustainability,” available on SSRN.  

The paper has received the Best Paper Award at the 2022 Annual Conference of the European Capital Markets Institute (ECMI). Financial support from ECMI is gratefully acknowledged. 

Leave a Reply

Your email address will not be published. Required fields are marked *