In August, the European Commission sought feedback on two important sets of draft rules aimed at making energy markets more transparent and fair. They are being developed in line with Regulation (EU) No 1227/2011 on wholesale energy market integrity and transparency (REMIT), as amended in 2024, which promotes competition between participants by increasing transparency and preventing market manipulation.
The first set of rules is a proposal for a delegated regulation on insider platforms and registered reporting mechanisms. It includes strengthening the oversight of entities reporting energy market data to the EU Agency for the Cooperation of Energy Regulators (ACER).
The second set of rules covers a proposal to amend the existing Implementing Regulation (EU) No 1348/2014 on data reporting. It aims to improve the way in which energy market data is made available to ACER, allowing the institution to better monitor wholesale markets and detect potential abuses.
Eurelectric expresses concerns that the draft (Wholesale energy markets – data reporting rules (revision)) increases reporting burden without commensurate surveillance benefit. Exposure reporting is the clearest example. The notion of flagging prima facie unjustified trading behaviour implies a de facto hedge benchmark and the sanctioning of deviations from it. Forecasts are inherently uncertain and scenario-based: forcing a single reported figure would effectively prescribe the firms risk-management choices.
Furthermore, the Level 1 framework does not justify requiring market participants to report forecasts in the first place. ACERs mandate is not to monitor internal forecasts, which are anyway not determinative of trading behaviour. Eurelectric, therefore strongly opposes the inclusion of forecasts in exposure reporting and, at minimum, request that they be limited to ad hoc, nonbinding requests on duly reasoned grounds.
The new Algorithm ID requirement is unclear. Just as with human trading, Level 1 legislation does not justify granular metadata for strategy identification. The disclosure of strategy metadata is also commercially sensitive. Should any algorithm flagging be retained, a simple binary algorithmic trade indicator (using the Trader ID) would suffice. Double reporting must also be avoided: positions already reported under financial-market regimes, as well as intragroup transactions, should be explicitly excluded from exposure reporting, in keeping with the legislators simplification intent and to avoid costly development and the risk of overstated external exposures. Where genuinely relevant, ACER can request them ad hoc on a targeted basis. To preserve the confidentiality of bilateral contracts, we propose that ACERs accessibility be granted with the involvement of NRAs and limited to consultation without extraction of originals or copies.
Implementation should begin only after complete technical guidance is published (including final schemas, updated TRUM and examples). Stakeholder consultation on technical details, ACERs drafting thereof, and implementation by companies simply cannot fit into the proposed schedule for exposure reporting (be it 6 months or by April 2027). Exposure reporting – especially if reporting of forecasts is retained in spite of our opening considerations – requires substantial new IT, hence a minimum of 12 months from the publication of final guidance.
Eurelectric supports effective monitoring, but also expect legal certainty, proportionality, and collaboration that values data quality over quantity.
The constructive path forward is as follows:
- limiting reporting of forecasts to ad hoc
- nonbinding requests
- preventing duplication by excluding trades and positions already reported under financial-market reporting
- replacing the Algorithm ID with a binary algorithmic-trade flag until surveillance value and infrastructure are demonstrated
- consulting bilateral contracts with the involvement of NRAs and without original or copy extraction
- anchoring all timelines to the publication of complete technical guidance and providing at least 12 months for implementation
- committing to a genuinely iterative consultation with documented reasoning and measurable adjustments.
More information is avalable here
