Mandated, not wanted: what ESMA has made of third-country clearing reporting

On 18 August 2026 the European Securities and Markets Authority (ESMA) opened a consultation on a framework for reporting how much European firms clear through recognised third-country CCPs. The draft itself is well crafted. What deserves attention is paragraph 15 of the same document: in it, ESMA notes that the supervisors it consulted consider the report unnecessary.

Sentences of that kind rarely appear in consultation papers. This one is deliberate, and it explains the entire exercise.

A paper that records the case against itself

Before drafting, ESMA asked supervisors about the usefulness, feasibility and proportionality of the mandate. The paper describes the response as “remarkably consistent across jurisdictions”. It came in three parts. First, the information required under Article 7d is not considered necessary for supervisory purposes. Second, existing reporting under Article 9 of EMIR already provides enough information, including on clearing at recognised third-country CCPs. Third, a second, parallel obligation would duplicate what exists, impose disproportionate cost and “deliver limited or no additional supervisory value”.

The clause that follows carries the entire justification for the exercise. The information could, the paper says, support a broader understanding of European dependencies on recognised third-country CCPs, “in particular in the current geopolitical context”. The word “geopolitical” appears exactly once in 51 pages. Nowhere does ESMA claim any supervisory benefit for itself.

Anyone wanting to know who was asked will find nothing. The paper gives no number, no committee, no period. It says “supervisors” and “across jurisdictions”, and nothing more. Publishing a verdict that blunt without saying who delivered it is hard to read as anything other than deliberate reticence.

Why the authority must deliver anyway

The obvious charge would be that a supervisory authority is manufacturing paperwork against its own judgement. It lands on the wrong party. Art. 7d(2) and (3) EMIR are not an invitation but an instruction: ESMA “shall develop” the technical standards and “shall submit” them to the Commission. Whether the obligation exists was settled by the Parliament and the Council, not by the authority. The paper says as much itself, noting that ESMA is confined to operationalising the obligation while preserving proportionality and avoiding unnecessary duplication.

ESMA's lack of enthusiasm for the mandate can be dated precisely. The standards were due with the Commission by 25 December 2025. They are still not there, and the paper explains why: in 2025 ESMA's Board of Supervisors ranked mandates by impact, urgency and data availability, and pushed this one back. That is not delay in the ordinary sense but a deliberate deprioritisation, by the authority's most senior body, of a mandate it was legally bound to deliver.

Three of five data categories made it into the template

That leaves the how, and here the draft repays study, because ESMA uses every inch of the discretion that Level 2 allows.

Article 7d lists five categories: the type of instruments cleared, average values by Union currency and asset class, margins, default fund contributions and the largest payment obligation. The template annexed to the draft contains three of them. Default fund contributions and the largest payment obligation have no field. ESMA intends instead to source those figures from an existing annual data request to the supervisors of Tier 1 CCPs. On this source it says something striking: it rests on cooperation arrangements and “does not constitute a regulatory reporting obligation”. Two of the five items required by the Regulation therefore depend on foreign authorities remaining willing to cooperate.

The second reduction matters more and has drawn little notice. The draft requires clearing members established in the Union to report “for financial instruments other than derivatives and securities financing transactions and for non-financial instruments”. For European firms, derivatives and securities financing transactions fall outside the new report altogether. That leaves securities, narrowly defined spot contracts and crypto-assets. The reasoning is sound: derivatives are already covered by Article 9 EMIR reporting, securities financing transactions in the regime under the Securities Financing Transactions Regulation. The full breadth applies only to clients outside the Union belonging to a group supervised within it.

Margins are treated with similar restraint. Only initial margin is to be reported, as an average of month-end observations over twelve months, aggregated per CCP. Whether variation margin should be added is one of the questions under consultation. For the format, the draft settles on plain CSV files; structured XML was considered and rejected as disproportionately complex given annual frequency and a small number of fields.

The figures that actually matter

How large is the dependency being measured? The reliable numbers come not from the draft but from the first annual report of the Joint Monitoring Mechanism, published on 6 July 2026. At the end of 2024, CCPs established in the Union accounted for roughly 20 per cent of notional outstanding in euro-denominated interest rate swaps, 33 per cent in forward rate agreements and 6 per cent in overnight index swaps. In exchange-traded derivatives, EU CCPs accounted for 30 per cent of open interest in futures on the euro short-term rate but only 3 per cent in Euribor futures. For scale: EU and Tier 2 CCPs together cleared some 82 trillion euros of notional in interest rate swaps.

On the effect of the active account requirement, in force since June 2025, the same report is more circumspect. Tier 2 CCPs still dominate clearing in the products concerned, with more than 50 per cent of the market by cleared volume, and in some cases 90 per cent or more. The requirement has “not yet materially altered” participants' core preferences. The cleanest measure of movement is the share of initial margin posted by EU clearing members at Tier 2 CCPs: it fell from around 58 per cent in the fourth quarter of 2024 to around 51 per cent a year later, driven by higher margins at European CCPs rather than by any retreat from London.

A fourth attempt at the same problem

The Article 7d report is not the first attempt to deal with this problem, and the sequence before it explains why the supervisors consulted sound weary. On 28 September 2020 ESMA classified the two large British houses, LCH Ltd and ICE Clear Europe, as Tier 2 CCPs, that is, systemically important to the Union. That was a classification, not an intervention: the same decision also recognised the houses, on the basis of a Commission equivalence decision, and that recognition has been extended several times since. Only EMIR 3 brought an instrument with teeth: Article 7a requires counterparties above the clearing threshold to hold an active account at a CCP authorised in the Union, and to clear a minimum number of representative trades through it.

How well that sequence worked is something the EU institutions say themselves. On 31 January 2025 the Commission extended equivalence for UK CCPs by three years, to 30 June 2028, arguing that the extension would let the agreed EMIR 3 measures begin to take effect and reduce exposure to UK CCPs. On 17 March 2025 ESMA aligned its tiering and recognition decisions with the same date. Five years after the first classification, the Union extended recognition on the grounds that the instrument meant to fix the problem had not yet begun to work.

That leaves the Article 7d report at the end of a line in which exactly one instrument was ever designed to reduce the dependency. Classification and recognition were preliminaries with no steering ambition; the account requirement was the intervention; and now comes a measurement. The order is not absurd, since enforcing an obligation eventually requires figures on how it performs. It does make it easier to see why the objection lands on this step: after three attempts, the question of purpose gets louder, not quieter.

The calendar defeats the purpose

The strongest objection to the new obligation concerns the calendar. Under the draft the first submission falls due no earlier than six months after entry into force and must then comprise separate reports for each calendar year from 2025 onwards. ESMA works the example through: if the regulation enters into force in May 2027, the first submission is due in January 2028 and covers 2025, 2026 and 2027.

Set against that is a line from the Joint Monitoring Mechanism report that settles the matter. Its own data on the active account requirement remains incomplete because the reporting designed for it only becomes available in the second half of 2026. One reporting regime will therefore measure the same exposure continuously from late 2026, while the Article 7d report delivers retrospective annual figures no earlier than the start of 2028. That is what the supervisors mean by limited additional value. The objection is about timing, not about substance.

One detail shows how far removed the exercise is from the authority's own agenda. In its press release ESMA presents the consultation as part of its simplification and burden-reduction agenda. Six weeks earlier the same authority had put a number on simplifying European transaction reporting: savings of up to one billion euros a year.

A drafting error in the law

For practitioners the draft contains a footnote of unusual candour. Article 7d asks, on its face, for “margins collected”. What is meant is margin posted. ESMA puts this down to the drafting history: the obligation was originally to fall on the third-country CCPs themselves, which do collect margin. During the EMIR 3 negotiations it was moved to European institutions and, in the paper's own words, “when this was done, the verb was not updated accordingly”. The draft therefore reads the provision against its wording.

A second ambiguity sits in the draft itself. The article on margin information applies, on its face, to every reporting entity, while the gap analysis in the same document requires margin only from a subset. Anyone responding to the consultation has a concrete point to raise.

Recommendations

Glossary