Most writing about European sustainability reporting treats it as a single quarrel: for the rules or against them. The documents do not read that way. Signed opinions from the European Central Bank, the European Economic and Social Committee and the standard setter EFRAG, a Commission proposal and two directives, the latest adopted on 24 February 2026, show three separate arguments running at the same time. Is the regime too heavy? How much checking should a report receive? And what is a report for: investors alone, or also the people and places a company affects? Each argument has bodies on both sides, and each argument has reached a different point in law.
Of the three arguments, the law has settled two: Directive (EU) 2026/470 simplified reporting and closed the road to reasonable assurance, while double materiality, the argument that decides what a report is for, survived into the revised standards and is now fought over how it is applied, not whether it applies.
This is the fourth piece of the sustainability chapter of the CareerOn Industry Atlas. The first asked where the money is and found that every euro figure is a forecast. The second found that the law names roles, not firms. The third read what the first reporting companies actually did and what supervisors found. This one reads the arguments, because a candidate who walks into an interview in a reporting team will be asked which of them is still open, and the honest answer has changed since 2025.
Eight documents, read for what each side said
We read eight dated documents from three families. From European law: the Corporate Sustainability Reporting Directive of 14 December 2022, the Commission's proposal COM(2025) 81 of 26 February 2025, and Directive (EU) 2026/470 of 24 February 2026, the act known as Omnibus I. From institutional opinions: the opinion of the European Central Bank of 8 May 2025 and the opinion of the European Economic and Social Committee adopted on 18 June 2025. From the standard setter: EFRAG's cover letter of 3 December 2025, its draft amended ESRS 1 of November 2025, and the basis for conclusions that accompanied its advice in December 2025.
Two rules governed the reading. First, only an adopted act may state the rule. A proposal is printed as a proposal, an opinion as an opinion, and EFRAG's drafts as technical advice, which they remain until the Commission adopts them by delegated act. Second, nothing is scored. Each side is printed in the terms of its own document, and the piece does not rule on which side is right. We also refused four candidates: two news pages that described documents we could read directly, a factsheet marked draft that repeated the cover letter, and law-firm commentary on Omnibus I, which our series rules do not admit when the act itself is available.
One absence is printed rather than filled. No document from a member state or from civil society was admitted. Press coverage reports positions from both, but we did not read a signed document for either, so neither appears here as a side.
Three arguments, three different places in law
The starting point is the 2022 directive. It asks companies to report in two directions at once. In its own words, the relevant articles "require undertakings to report both on the impacts of the activities of the undertaking on people and the environment, and on how sustainability matters affect the undertaking." That is double materiality. The same directive set limited assurance for the first reports and left a path toward reasonable assurance, the stronger opinion an auditor gives on financial statements.
The Commission's proposal of February 2025 pushed on all three points. It proposed fewer companies in scope and fewer requirements. On assurance it was blunt: "The possibility of moving from a requirement for limited assurance to a requirement for reasonable assurance would be removed." The stated reason was cost: "there will be no future increase in costs of assurance for undertakings in scope."
Three arguments, and where the law left each one
| Argument | The 2022 rule | Proposed in 2025 | Where it stands |
|---|---|---|---|
| Simplification | Full reporting for the companies in scope | Fewer companies and fewer requirements | Settled by Directive (EU) 2026/470, adopted act |
| Assurance | Limited assurance, with a path to reasonable assurance | Remove the path to reasonable assurance | Settled by Directive (EU) 2026/470, adopted act |
| Double materiality | Report impacts and financial risks | Clearer instructions on applying it | Open on application; kept in draft ESRS 1, technical advice |
Sources 1 EUR-Lex (Publications Office of the European Union) · 2 European Commission · 5 EUR-Lex (Publications Office of the European Union) · 7 EFRAG
A year later, Directive (EU) 2026/470 turned two of those arguments into law. As our first piece read in the act, the scope narrowed and the path to reasonable assurance closed. Those two questions could be settled by a date, a threshold or a deleted power, and they were. The third could not. On double materiality the 2026 act did not remove the principle. It ordered better instructions for applying it, asking for "clear instructions on how to apply the materiality principle in order to ensure that undertakings are only required to report material information." The argument moved from whether to how.
Each side, in its own terms
The European Central Bank wrote as a body that uses the data. On assurance it took the side against change: "the ECB recommends the retention of the possibility for the Commission to further enhance the verification process, by adopting standards for reasonable assurance engagement." It did not ask for that step at once. It said the Commission "should consider doing so once sufficient time has elapsed following the first years of reporting under the CSRD." It would be wrong to print the bank as opposed to simplification as a whole. Its opinion welcomes streamlining the standards and asks to keep particular data points, so it is printed on the assurance argument and as a qualified voice on the rest.
The European Economic and Social Committee is easy to misread in the other direction. It is not a refusal. The committee takes note of simplification as a first step and sets conditions on it: simplification measures should strike "a balanced approach that supports businesses and investors, protects workers and serves the broader interests of society." That is a conditional side, and the exhibit prints it as one.
Each side, in the terms of its own document
| Body | Argument | Position | Legal state |
|---|---|---|---|
| European Commission | Simplification and assurance | For change: no future rise in assurance costs | Proposal |
| European Central Bank | Assurance | Against change: keep the possibility of reasonable assurance | Opinion |
| European Economic and Social Committee | Simplification | Conditional: simplify while keeping high standards | Opinion |
| EFRAG, cover letter | Simplification | For change: fewer mandatory datapoints | Technical advice |
| EFRAG, draft ESRS 1 | Double materiality | Keep both dimensions and how they interact | Technical advice |
| EFRAG, basis for conclusions | Double materiality | For change: a simpler assessment, tested with preparers | Technical advice |
Sources 2 European Commission · 3 EUR-Lex (Publications Office of the European Union) · 4 EUR-Lex (Publications Office of the European Union) · 6 EFRAG · 7 EFRAG · 8 EFRAG
EFRAG appears on both sides of the materiality argument, and that is not a contradiction. Its cover letter reports a large cut in requirements, and its basis for conclusions describes a simpler way to run the assessment. Yet its draft ESRS 1 keeps the principle intact: "Double materiality has two dimensions: impact materiality and financial materiality and the undertaking shall consider how they interact." The standard setter is simplifying the procedure while keeping the test that decides which disclosures apply at all. That is where the live argument now sits.
What the standard setter counted, and what it did not
Two figures circulate in this debate, and both need their context. The first is from EFRAG's cover letter: the amended standards cut the mandatory datapoints by 61% against the initial ESRS, and by 71% including the voluntary ones. That is a count of requirements in a standard. It is not a count of what companies disclose, and it does not tell a reader whether published reports will be shorter, better or merely different.
The second is from the basis for conclusions. EFRAG tested its draft with companies: "A total of twenty-eight preparers from different geographies and sectors, including high-impact sectors, participated in the test." The finding most often quoted from that test is cautious in the original. EFRAG explains "why several preparers stated that the results of the materiality assessment would not change by the application of the new provisions of the July ED." Twenty-eight preparers applying a draft rule is a simulation on a sample. It is not a measured effect on published reports.
What the standard setter counted, and what it did not
| Document | What it counts | Figure | What it is not |
|---|---|---|---|
| EFRAG cover letter | Mandatory datapoints against the initial ESRS | 61% fewer (71% with voluntary datapoints) | A count of what companies disclose |
| EFRAG basis for conclusions | Preparers who applied the draft rule | 28 preparers | A measured effect: several said their result would not change |
No admitted document measures what simplification has changed in the reports companies actually publish. Anyone who tells you it has made reporting easier, or emptied it of substance, is forecasting. The record at this date shows fewer rules on paper and a test in which several companies said their materiality result would not move.
Why the arguments split the way they did
The sides sort by who reads the report. A body that relies on the data for supervision or lending, as the central bank does, argues to keep data points and the route to stronger checking. A body weighing the burden on companies, as the Commission did in its proposal, argues cost. The standard setter sits between them: it cuts requirements while keeping the two-dimension test that decides which of them apply.
The law then did what law does well. It settled the questions that a legislator can settle with a date, a threshold or a deleted power. Scope and assurance were of that kind. Materiality is not. It is decided company by company, inside an assessment, and no directive can fix the answer in advance. The 2026 act acknowledged this by ordering instructions rather than an outcome, and by naming the risk that assurance providers might push companies to "dedicate excessive resources to the materiality assessment process." The argument has moved from Brussels into each reporting team.
The counter-case, kept in the piece
The thesis depends on documents that are not yet final. The draft ESRS 1 that keeps both dimensions is technical advice. If the Commission's delegated act adopting the amended standards dropped impact materiality, the third argument would be settled too, against the 2022 rule, and this piece would be wrong. Equally, a later act could restore a path to reasonable assurance and reopen the second argument. Neither has happened in the documents we read. The piece prints its reading date so that either change dates it rather than hiding it.
A second objection is that the arguments are not really separate: simplification and materiality are linked, because the materiality assessment is what decides how much a company reports. That is true, and EFRAG's own documents show the link. We still treat them apart because the law did. The 2026 act settled scope and assurance directly and handed materiality back to the standards.
What this means for your career
For a candidate, the open argument is the work. A double materiality assessment is what a reporting team does first, before any datapoint is prepared, and the record shows it is where judgement now concentrates. EFRAG's test is a reminder that companies run it differently and that a simpler procedure does not always change the result. Being able to run that assessment, explain its two dimensions and defend the outcome to an auditor is the skill the closed arguments have not touched.
For a hiring manager, the closed arguments set the scope of the job for the next years: limited assurance, a shorter list of mandatory datapoints, and the same assessment at the front of the process. That is a role that rewards precision over volume.
Two practice simulations on the CareerOn shelf sit on evidenced steps. ESG analyst: CSRD reporting and double materiality rehearses the assessment that the draft ESRS 1 keeps and that twenty-eight preparers ran in EFRAG's test. ESG analyst: non-financial reporting under CSRD rehearses the preparation of the disclosures that the 61% cut applies to. We place the second card on that preparation only. The cover letter counts requirements; it does not describe the gap analysis the card also rehearses, so we do not claim it does.
Two roles in this argument have no simulation yet: the statutory auditor or independent assurance provider whose opinion the assurance argument was about, and the supervisory enforcement examiner who reads the filed statement. Both remain recorded gaps in the CareerOn catalogue.
CareerOn view
The argument worth a candidate's time is the one still open. Learn to run and explain a double materiality assessment, because the law kept it and the standards now leave more of it to judgement. The arguments about scope and assurance will still be discussed, but in the documents we read they have been decided.
What we could not prove
We could not prove what the revised standards will finally say. The 2026 act orders a revision and does not write it, and EFRAG's drafts are advice until adopted. We could not prove any measured effect of simplification on published reports. We did not read a signed member-state or civil-society position, so none is printed. And we do not say who is right. Each side is printed as its document states it, with that document's legal standing beside it.