ESAs publish Consultation Paper proposing significant changes to the SFDR RTS

On 12 April 2023, the ESAs published a joint consultation paper seeking feedback on changes to the SFDR RTS in areas such as PAI disclosures, the SFDR DNSH test and product level disclosures. By way of recap, this review has been in response to a mandate the ESAs had previously received from the Commission to review the SFDR RTS and is separate from the broader review of the EU SFDR rules that was announced by the Commission in January of this year (which we think will involve a more fundamental review of the SFDR Level 1 requirements). Interestingly, the ESAs have noted that their proposals in this consultation paper go further than the mandate they had received from the Commission and their justification is that this is necessary to address the weakness / issues within the current framework. 

Overall the proposed changes (if adopted) are quite significant and will require firms to potentially revisit their categorisation of Article 9/8+ products (because of changes to the DNSH test and PAI indicators), uplift existing PAI and product disclosures (due to the inclusion of additional PAI indicators and changes to the product templates) and also provide detailed disclosures of any product level GHG emissions targets. 

The deadline for feedback is 4 July 2023 and the ESAs have not indicated when they expect to publish final rules.  

1. Various changes to PAI indicators 

New social PAI indicators

Taking account of the work that has been done at an EU level with respect to the first set of draft European Sustainability Reporting Standards (ESRS) under CSRD, the ESAs are proposing to introduce additional mandatory and voluntary social PAI indicators. FMPs will therefore have to report on the additional mandatory social indicators at an entity level, and also embed them within their DNSH framework for sustainable investments. 

The proposed new mandatory social PAI indicators are:

  • Amount of accumulated earnings in non-cooperative tax jurisdictions
  • Exposure to companies involved in the cultivation and production of tobacco
  • Interference in the formation of trade unions or election of worker representatives
  • Share of employees earning less than the adequate wage

Additionally, the ESAs are proposing to introduce the following voluntary social PAI indicators: excessive use of non-guaranteed-hour employees in investee companies; excessive use of temporary contract employees in investee companies; excessive use of non-employee workers in investee companies; insufficient employment of persons with disabilities within the workforce; lack of grievance/complaints handling mechanism for communities affected by the operations of the investee companies; and lack of grievance/complaints handling mechanism for consumers/end-users of the investee company

The ESAs have also noted that no social PAIs currently apply to real estate assets and they are seeking feedback on how the rules should be amended to enable these investments to duly take into account social factors – e.g. applying them to the entity managing the real estate asset.  

Changes to existing PAI indicators and calculations 

The ESAs are proposing to change a number of PAI indicators to mirror the datapoints that will be reported under the ESRS (e.g. number of days lost to work-related injuries, accidents, fatalities from work-related accidents, work related ill health and fatalities from ill health) or, they state to improve formulations (e.g. Operations and suppliers at significant risk of incidents of use of workforce qualifying as child labour instead of ‘operations and suppliers at significant risk of child labour’). Although the shift to mirror the ESRS standards will be helpful once corporates start reporting under the CSRD, the approach is still quite European centric and overlooks the fact that many international companies will not be captured by the CSRD for a while, or at all, and may instead report using international ESG disclosures standards such as the ones being prepared by the ISSB. 

Some PAI indicators are also being amended to ensure consistency with the broader EU Sustainable Finance package such as the EU Taxonomy Regulation and the Low Carbon Benchmark Regulation – e.g. as noted above, the ESAs are proposing a new tobacco exposures social PAI indicator (in line with the exclusions under the Climate Benchmarks Delegated Regulation) and are amending PAI 10 (Violations of UN Global Compact principles and Organisation for Economic Cooperation and Development (OECD) Guidelines for Multinational Enterprises) and 11 (Lack of processes and compliance mechanisms to monitor compliance with UN Global Compact principles and OECD Guidelines for Multinational Enterprises) to mirror the minimum safeguards in the EU Taxonomy Regulation (e.g. replacing the UN Global Compact principles with UN Guiding Principles on Business and Human Rights, including the principles and rights set out in the eight fundamental conventions identified in the Declaration of the International Labour Organisation on Fundamental Principles and Rights at Work and the International Bill of Human Rights).

The ESAs have also proposed new formulae for PAI indicators that don’t currently have associated formulae and proposing technical clarifications to certain indicators (e.g. specification of the metrics for the indicator on investments in companies without sustainable land/agriculture practices and for the indicator on investments in companies without sustainable oceans/seas practices). 

The ESAs have also sought feedback on how the ‘current value of all investments’ metric should be used in PAI calculations for different asset classes. They seem to suggest that under the current rules and guidance, the denominator in the PAI calculations for each of the 3 asset classes (investee companies, sovereigns and supranationals or real estate assets) should be all the investments of the FMP, and not just the investments that relate to the specific asset class. They are seeking feedback however on whether the denominator should be limited to investments that relate to the specific asset class – which they note will have the advantage of the particular PAI indicator being focussed on the investments for which it is relevant but would in their view hinder comparability between FMPs and limit the relevance of the data disclosed. This is interesting feedback, as we don’t think the approach under the current rules / guidance is clear, and nor is it necessarily the case that the alternative approach would be less helpful / relevant. 

Finally, the ESAs are proposing firms to set out in the “explanation” column of the PAI disclosures, the proportion of data that is sourced directly from investee companies.  

Consideration of value chains in PAI disclosures

Whether value chains of investee companies should be considered in PAI indicators has not been clarified previously (excluding Scope 3 GHG emissions related PAI indicators). The ESAs are now proposing however that the contribution of investee companies’ value chains to the PAIs should be considered where the investee company is reporting impacts in its value chain according to the ESRS under its own materiality assessment (performed under the ESRS) and that in other cases, FMPs should include information on the value chains of investee companies where that information is readily available, e.g. in the public reporting of those investee companies. This will be a potentially significant endeavour for FMPs, where the companies are not disclosing this information under the ESRS.

2. Proposed changes to the DNSH test for sustainable investments 

The ESAs are proposing changes to the DNSH test as they are concerned about the lack of transparency and comparability in (i) DNSH disclosures published by FMPs; and (ii) the DNSH framework adopted by FMPs. They are also concerned about the disconnect between the use of environmental PAIs in the DNSH test and the DNSH criteria in the EU’s environmental Taxonomy. The ESAs also noted that a significant share of Article 9 SFDR financial products have exposures to fossil fuels, and in particular to coal activities. The ESAs note that the Taxonomy disconnect will only be resolved by changes to the Level 1 SFDR framework (and reference the Commission’s assessment of SFDR that was announced in Jan 2023) but have made the following proposals regarding SFDR DNSH disclosures:

  • maintain the status quo – noting that the SFDR and EU Taxonomy (for climate change mitigation / adaptation) regimes have only just become fully applicable and further change at this stage could result in market disruption, but also would preserve the weaknesses of the current framework;
  • require more granular DNSH disclosures – in particular the ESAs are proposing that the website disclosures of the financial product should set out quantitative DNSH thresholds for each PAI indicator, but FMPs will still have discretion in the setting of these thresholds, which the ESAs think could limit comparability and mean that certain significantly harmful activities are still included within the financial product (Note: the reference to “quantitative” thresholds is quite unhelpful, as to date the guidance has been that FMPs can decide if the PAI indicators have been respected in their DNSH framework, and for certain PAI indicators or certain asset classes, products etc., qualitative tests would make more sense);
  • provide an optional safe harbour for Taxonomy aligned investments – i.e. such investments would be deemed to be sustainable investments (as was initially proposed by the ESAs in 2021) but it is not clear from the paper if the entire investment would be regarded as an SFDR sustainable investment, or just the proportion that relates to Taxonomy aligned activities. We think the ESAs mean the latter, as they go on to state that FMPs could apply the SFDR DNSH test to the other activities;
  • shift to a single taxonomy-based system for DNSH – i.e. the Taxonomy DNSH criteria form the basis of the SFDR DNSH assessment. The ESAs have however noted that this would likely be a longer term consideration for the EU co-legislators, since the EU Taxonomy assessment is done at economic activity level whereas the SFDR assessment must be done at entity/investment level.
3. Treatment of derivatives 

The treatment of derivatives for SFDR purposes still remains an issue – the ESAs have previously confirmed that derivative exposures cannot be treated as making a positive Taxonomy alignment, but had not provided guidance on how they should be treated under the SFDR sustainable investment test, or in the numerator of PAI calculations. In the consultation paper, the ESAs are now unhelpfully proposing that:

  • derivative exposures should be included in the numerator of PAIs (i.e. they should be treated as having potential principal adverse impacts) if they amount to an equivalent long net exposure as in the ESA’s view the transaction would result in money going towards the adverse impacts covered by the relevant PAI indicators. However, where FMPs can show that the derivative does not ultimately result in a physical investment in the underlying security by the counterparty (or any other intermediary in the investment chain) the FMP would be allowed to consider that this derivative investment does not result in an adverse impact and should therefore be allowed to exclude it from the numerator;
  • derivative exposures should not be treated as sustainable investments (mirroring the approach for Taxonomy alignment) again if they amount to an equivalent long net exposure, and short positions (including via derivatives) should be used to reduce the long net exposure on a given Taxonomy alignment / SFDR sustainable investment issuer.

This is an unhelpful position and in our view is contradictory to the treatment of derivatives for PAI calculations, as (despite the greenwashing concerns they raise in the paper) the ESAs seem to have decided that derivatives can have negative ESG impacts but not positive ones. The ESAs have also confirmed that netting should never be used to reduce the figures to below 0 in all cases. 

4. Proposed changes to SFDR product level disclosures

"Simplification" of product-level disclosure templates

The ESAs acknowledge that the financial product disclosure templates are excessively long and complex, and so are also consulting on certain changes to language, layout and structure of these templates to enhance comprehensibility, in particular for retail investors. Accordingly, the phrasing of the questions posed in the templates has been tweaked across the board to render them more straightforward and to reduce the use of technical terms.

GHG emissions reduction targets at product level 

Ironically given their stated objective of simplifying disclosures, the ESAs propose to introduce a whole new layer of complexity by requiring disclosure in pre-contractual documents, on websites and in periodic reports on GHG emissions reduction (i.e. decarbonisation) targets (albeit the ESAs do note that GHG emissions reduction target disclosure requirements should not create any additional burden for products without such targets - these products simply have to disclose this up front). 

Financial products with GHG emissions reduction targets will have to provide in pre-contractual disclosures a narrative explanation about the way the target will be achieved, indicating whether the financial product (a) divests from investments with particular GHG emissions levels and invests instead in companies with lower GHG emissions; (b) invests in companies that are expected to deliver GHG emissions reductions over the duration of the investment; (c) engages with investee companies to contribute to their GHG emissions reduction. They will also have to indicate the share of the investments of the financial product covered by the GHG emission reduction target, as well as whether the reduction target is compatible with limiting global warming to 1.5 degrees. Critically, financial products must disclose the baseline GHG emissions of the financial product, as well as the final and intermediate decarbonisation targets and the corresponding years for those targets to be achieved.

Moreover, in periodic disclosures, financial products must disclose measurements of progress to date, when available. If target(s) have not been met, an explanation and details of the corrective actions planned to meet the target must be specified. More details on the steps planned to achieve the GHG emissions reduction target must be included in a new section of the website disclosure.

Where data on the investee companies’ GHG removals and storage and/or the purchase of carbon credits are not readily available, financial products must disclose details of the best efforts used to obtain the information either directly from investee companies, or by carrying out additional research, cooperating with third party data providers or external experts or making reasonable assumptions.

GHG emission reduction disclosure requirements would also apply to products having GHG emissions reduction as their investment objective under Article 9(3) SFDR. But such products would be required to show the relevant benchmark disclosures when they passively track a Paris-Aligned Benchmark (PAB) or Climate Transition Benchmark (CTB) by providing a hyperlink to a description of the benchmark methodology.

‘Dashboard’ of key information 

The ESAs also suggest amending the front-end of the pre-contractual and periodic disclosure templates to remove the existing summary box and replace it with a "dedicated dashboard” of key information to complement the more detailed information in the body of the disclosure (with the intention of drawing specifically retail investors' attention to the most vital information and mitigating risk of information overload). 

The dashboard requires a narrative summary (of max 250 characters, including spaces) of the environmental and social characteristics of the product and the percentage of the product's investments that promote such characteristics (for Art 8 products) or the sustainable investment objective (for Art 9 products). The dashboard also require disclosure of (a) the minimum commitment of sustainable investments; (b) the minimum commitment of Taxonomy aligned investments; (c) whether or not the product considers PAIs (which is presumably always the case for Art 9 products but the option to state 'No' is confusingly included nonetheless); and (d) whether or not the product has a GHG emissions reduction target. 

One potentially helpful side effect of the dashboard being used is that the "asset allocation" diagram may be removed (given that the dashboard would provide information up front about the product's sustainable and Taxonomy-aligned investments). This should mitigate some of the issues firms have encountered in the present iteration of the template in trying to divine the E/S split of investments.

The joint consultation paper is available here.

The press release is available here.