Sustainable Finance Note · Part 2 of 2 · 8 May 2026

The ESRS rewrite and the financial sector

Part 2: What you lose as a user

EFRAG's own user survey shows 67% of investor users expect data quality to fall. The ECB, EBA, ESMA and EIOPA converge on the same red line. The connectors that turned ESRS into a sustainable-finance backbone are quietly disappearing.

By Theodor Christensen  ·  8 May 2026

In Part 1 of this note I argued that the European Commission's May 2026 rewrite of the European Sustainability Reporting Standards delivers real producer-side relief to the financial sector: the fiduciary asset-management carve-out, the GHG accounting election, and the materiality re-engineering. This second note looks at the other end of the pipe: what financial firms lose when they consume their counterparties' ESRS data. The picture there is much less comfortable.

Banks, insurers and asset managers are unusual in the CSRD universe because the data they consume drives more sustainable-finance infrastructure than any other user community. The EU Taxonomy, SFDR, the EBA Pillar 3 ESG framework, Solvency II climate disclosure, the EU Climate Benchmarks regime and a thicket of national supervisory expectations all depend on what counterparties report under ESRS. When the Commission removes datapoints from the ESRS mandatory floor, the consequences ripple through that infrastructure in ways that are not visible from the producer-side cost-benefit analysis.

"None of the surveyed investors definitively anticipated benefits from the revised standards." EFRAG's own cost-benefit analysis, December 2025

The user-side picture: 67% expect quality decline

The most credible single piece of evidence on the user-side impact is the user-perception survey embedded in EFRAG's December 2025 cost-benefit analysis, the same study the Commission cites for the producer-side savings figures. The user findings are striking, and have received much less coverage than the cost numbers.

Sixty-seven per cent of investors and financial institutions surveyed anticipate a negative impact on information quality. Fifty-five per cent of all users (investors plus other users) expect lower information quality. Fifty-two per cent flag lower comparability. Forty-five per cent specifically flag loss of critical climate-standard data. Forty-three per cent flag loss of critical data in other environmental standards. None of the surveyed investors definitively anticipate benefits from the revised standards. The standards-setter's own user community has produced a clear majority view that the revisions will reduce the quality of the data they rely on.

The signal is corroborated by the formal February 2026 opinions of the European Central Bank, European Banking Authority, European Securities and Markets Authority and European Insurance and Occupational Pensions Authority. All four supervisors are broadly supportive of the simplification objective but worried about the same three things: the indefinite duration of the permanent reliefs, gaps in interoperability with IFRS S2, and the loss of cross-cutting linkages to other Union legislation. The shared phrase across the four opinions is "permanent blind spots." Their consolidated recommendation that permanent reliefs be time-limited to FY 2029, so they function as transitional rather than indefinite measures, has not yet been accommodated by the Commission in the published draft.

The connectors that disappear

The most visible loss is at the level of headline numbers and analytical bridges, not at the level of underlying data. Scope 1, 2 (location- and market-based) and 3 emissions, the energy mix, the high-climate-impact-sector breakdown of fossil-fuel consumption, transition-plan key features, GHG reduction targets, internal carbon pricing and anticipated financial effects of climate risks all survive in some form. What does not survive is the architecture that turned those numbers into investable signals.

GHG intensity per net revenue is gone as a mandatory item, deleted from the new E1-8 (formerly E1-6). Energy intensity per net revenue is gone from the new E1-7 (formerly E1-5). The single consolidated "total GHG emissions" headline is gone, as is the dual location/market-based total presentation. So is the explicit cross-reference in E1-1 to EU Taxonomy-aligned CapEx and CapEx plans. So is the disclosure of whether the undertaking is excluded from EU Paris-Aligned Benchmarks under Articles 12.1(d)–(g) and 12.2 of Regulation 2020/1818. So is the climate-incentive-remuneration disclosure that linked variable executive pay to GHG reduction targets. None of these were the heart of E1 in a strict accounting sense, but each was a connector that made the heart of E1 useful for capital-allocation, financed-emissions, or product-classification decisions.

"With the connectors removed, an investor reading an ESRS climate disclosure has the raw materials to build a thesis but is doing more of the carpentry alone."

The picture is similar across the social and biodiversity standards. ESRS S2's explicit identification of countries and commodities with significant child-labour or forced-labour risk is gone; the surviving disclosure in the new ESRS 2 IRO-2 §37(f) is much thinner. ESRS S1 loses the gender breakdown of training, the age distribution of management, the actual-uptake-of-family-related-leave metric and one of the five social-protection events. ESRS E4 Biodiversity loses an estimated 78% of mandatory datapoints and the Kunming-Montreal Global Biodiversity Framework-aligned plan disclosure. The biodiversity-conscious investor, and biodiversity is now demonstrably a top three issue in investor surveys, gets a markedly thinner data set.

What banks specifically lose

Banks are the most exposed user group, because their lending books and credit-risk models depend on counterparty-level disclosure to a far greater extent than equity investors do. The deletion of the GHG intensity ratio is the single most consequential change for banks. The intensity ratio is the cleanest input to PCAF financed-emissions calculations across non-listed-equity asset classes, which is most of a typical loan book. Without a standardised intensity number, banks face higher data-vendor costs, more proxy modelling, and weaker comparability across portfolios.

The European Banking Authority's February 2026 opinion explicitly requested retention of the GHG intensity ratio as a mandatory cross-cutting metric, citing its centrality to Pillar 3 ESG and the green-asset ratio. The Commission has not yet accommodated this. If the final delegated act is published in Q4 2026 without restoration, banks will have to choose between modelling intensity from raw Scope 1/2/3 numbers (more variance, less comparability, more vendor dependency) or simply asking borrowers for the number directly under the bilateral relationship. The latter route runs into the new value-chain cap.

The value-chain cap is the second financial-sector pressure point. New ESRS 1 §66 limits the information that an in-scope undertaking may demand from a "protected" counterparty in its value chain to what is "necessary" under the future Voluntary SME Standard (VSME). For banks, the value chain is essentially the loan book. Procurement and credit teams that built broad ESRS-aligned questionnaires through 2024 and 2025 will need to retire those instruments and adopt the VSME-aligned approach for FY 2027. Whether the cap actually applies to banking relationships, which sit somewhere between a value-chain relationship and a financial-services counterparty relationship, is a live legal question. The Commission's 6 May 2026 explanatory note on the value-chain cap leans toward applying it broadly; banking associations will likely push back during the consultation.

What insurers and pension schemes lose

For insurers, the largest loss is the deletion of the topical anticipated-financial-effects disclosures across ESRS E2, E3, E4 and E5. Insurance underwriting and claims modelling for non-climate environmental risks, water stress, soil contamination, and biodiversity collapse, depended on these disclosures, in particular for property and liability lines exposed to chronic environmental change. With these requirements removed, insurers will rely more on national-supervisory disclosures (Solvency II climate stress under EIOPA), industry data, and bilateral information-gathering.

EIOPA's February 2026 opinion echoed the ECB and EBA emphasis on retaining forward-looking quantitative data, particularly for insurers exposed to physical climate risk.

Pension schemes, especially defined-benefit schemes that retain employer-side risk and so do not benefit from the new fiduciary carve-out described in Part 1, face a similar problem on the asset side. The biodiversity, water and pollution data they need to assess long-horizon natural-capital risk is being thinned at the EU mandatory level. They will need to fall back on TNFD-aligned voluntary disclosure or sector-specific frameworks.

What sustainable-finance product designers lose

SFDR Article 8 and Article 9 funds, ESG-labelled bonds and Climate Benchmark trackers all depend on issuer-level data inputs that ESRS provides. The deletion of the Paris-Aligned Benchmark exclusion disclosure is the single most damaging change for product designers. SFDR Article 9 funds that target Paris-Aligned Benchmark exposure lose a directly usable disclosure data point. The EU Platform on Sustainable Finance has called for restoration of both the Taxonomy CapEx-alignment cross-reference and the PAB exclusion disclosure, characterising their removal as a material weakening of the integrated EU sustainable-finance architecture.

Combined with the Commission's parallel "SFDR 2.0" proposal of November 2025, which deletes entity-level PAI reporting and the entity remuneration-policy disclosure to avoid CSRD overlap, the cross-cutting picture is one of weakening, not strengthening, even though each individual change is presented as simplification. A reasonable interpretation is that the EU is shifting from a maximally-comprehensive, prescriptive disclosure regime to a more selectively-targeted one that leans more heavily on supervisor enforcement and on voluntary frameworks (CDP, SBTi, IFRS S2, TNFD) for the granular content. This is not necessarily a worse equilibrium. IFRS S2 alignment and reduced regulatory fragmentation are real prizes, but it changes where competitive advantage lives in sustainable finance.

The asymmetry

Step back from the detail and the fundamental asymmetry of the file becomes visible. The simplification trades preparer cost for user data quality. EFRAG estimates EUR 4.7 billion of cumulative cost savings over 2027–2031, while 67% of investor users expect quality to deteriorate. The Commission's explicit intention is that simplification should not undermine "the objectives of the European Green Deal," but the data the Green Deal targets relied on — Taxonomy alignment metrics, Paris-Aligned Benchmark linkages, intensity ratios, climate-incentive-remuneration disclosure — is weakening at the same time that supervisors are still building disclosure regimes that depend on it.

The financial sector sits exactly at this fault line. As preparers (Part 1), financial firms benefit from the cost reduction, the fiduciary carve-out, the GHG accounting flexibility and the simpler materiality regime. As users, the same firms face a thinner pipe, less comparability, and a particular loss of the connectors that made ESRS data actionable for capital-allocation decisions.

The likely outcome over the next eighteen months is a partial accommodation. The Commission will probably restore some of the items the supervisors have flagged, most plausibly the GHG intensity ratio and the Paris-Aligned Benchmark exclusion disclosure, both of which have direct supervisory significance. It is less likely to time-limit the permanent reliefs, because doing so would reopen the political settlement that delivered Omnibus I. The fiduciary carve-out, the GHG accounting election and the materiality re-engineering will almost certainly stick.

"EUR 4.7 billion of cumulative savings, against 67% of investor users expecting quality to deteriorate. The financial sector sits exactly at this fault line."

What investment, credit and product teams should be doing now

• Build robustness against intensity-ratio loss. Even if the EBA recommendation lands and intensity ratios are restored in the final delegated act, plan for the possibility that they are not. PCAF-aligned financed-emissions calculations will need stronger reliance on raw Scope 1/2/3 disclosures, more proxy modelling, and probably more vendor data spend.
• Re-document Article 8 and Article 9 fund methodologies. Fund methodologies that reference the Paris-Aligned Benchmark exclusion disclosure as an input will need to either find a workaround in the surviving E1 disclosures or transition to a self-contained methodology. The SFDR 2.0 trilogue running through 2026 will affect this in parallel and should be tracked together.
• Engage in the public consultation. Closes 3 June 2026. The investor-side joint statements (IIGCC / PRI / Eurosif / EFAMA / UNEP FI) are the most effective amplifier for restoration of high-value items. Single-firm submissions are heard but rarely move the dial.
• Track the four supervisor opinions. The ECB, EBA, ESMA and EIOPA February 2026 opinions are publicly available and constitute the single best forecast of which items will be restored in the final delegated act. They are also a defensible basis for internal investment-policy positions during the FY 2026 transition.
• Plan for FY 2027 transition risk. Year-on-year comparability between FY 2026 disclosures (under existing standards) and FY 2027 disclosures (under revised standards) will be poor, especially for issuers that take advantage of phase-ins or change their GHG accounting boundary. Investment processes and credit models that assume comparable time series will need to be parameterised against this.
• For sustainable-finance strategy teams: the strategic question is whether the EU's integrated sustainable-finance architecture — Taxonomy, SFDR, ESRS, CSDDD, Paris-Aligned Benchmarks — survives the simplification programme as a coherent whole. Each of the five legs is being simplified or narrowed in parallel. Firms that have built market positions around mining and analysing very granular ESRS data will need to refresh that positioning.

The bottom line

The ESRS rewrite is genuine simplification, not pure deregulation. The 61% datapoint reduction, the materiality re-engineering, the value-chain cap, the fiduciary carve-out and the GHG accounting flexibility together address most of the legitimate complaints made about the 2023 standards. The producer-side wins, set out in Part 1 of this note, are real.

But the simplification is asymmetric, and the asymmetry falls hardest on the financial sector when it acts as a user. EFRAG's own user survey signals a clear majority view that the revisions will reduce data quality and comparability. The four supervisor opinions of February 2026 converge on a recommendation — time-limit the reliefs and restore the intensity ratios — that the Commission has not yet accepted. The next eighteen months are the moment when the final shape of the rule is set: public consultation closes 3 June 2026, the Commission is expected to adopt the final delegated act in Q4 2026, application is from FY 2027 with optional early adoption for FY 2026.

Financial-sector teams should treat the period between now and adoption as the window in which the supervisor red lines either become binding amendments or are absorbed into the final text. Prepare for the simplification as published, but be ready to refit the data-collection, financed-emissions and product-classification infrastructure if the supervisor recommendations land.

In one sentence: the financial sector gets meaningful relief as preparers, but the simplification quietly deconstructs the data architecture that made the EU sustainable-finance regime distinctively powerful. Whether that trade-off is worth it depends on which side of the fence you are on, and how much weight you give to the EFRAG user survey and the four supervisor opinions over the political imperative of regulatory burden reduction. For most readers of this note, both sides apply at once.