In April 2026, the EBA opened a consultation on simplifying IRRBB reporting. Before anyone in finance or risk starts celebrating a lighter workload, it's worth asking what a supervisor who now benchmarks bank models in near-real time is actually willing to give back.
Spoiler: not what you're hoping for.
The heaviest reporting build of the 2024 cycle might, on paper, be about to get lighter. In April 2026, the EBA opened a consultation (EBA/CP/2026/07) that explicitly invites input on simplifying IRRBB reporting in a future round. Before anyone in a bank’s finance or risk function celebrates, it is worth asking what, exactly, the supervisor might be willing to give back.
Start with what was built. Since the 30 September 2024 reference date, EU banks have filed dedicated IRRBB templates quarterly: the J-templates, J 01.00 through J 08.00. They carry the outlier test results, ΔEVE under six scenarios and ΔNII under two, against Tier 1 capital. They carry sensitivity breakdowns by currency and instrument. They carry the slotted repricing cash flows, which means the supervisor effectively receives the bank’s gap profile every quarter.
And then the layer that changed the game: the behavioral parameters. NMD core proportions. Weighted-average and maximum repricing maturities, reported against the 5-year cap. Pass-through rates. Prepayment and early-redemption rates.
Read that list again as a supervisor. Banks are not just reporting their risk numbers. They are reporting the assumptions that produce the numbers.
That design choice, assumptions alongside outcomes, is what lets supervisors benchmark models rather than just exposures. And within eighteen months of the first filing, the benchmarking machinery was visibly running.
The EBA’s heatmap workstream, a January 2024 roadmap followed by implementation reports in February 2025 and January 2026, delivered exactly the kind of findings that only assumption-level data supports: wide dispersion in NMD maturities and betas, constant-spread margin modeling except on NMDs, divergent CSRBB perimeters. The January 2026 report confirmed the 5-year NMD cap as the supervisory default, with deviations subject to approval.
SREP runs on the same fuel. The ECB published a dedicated IRRBB and CSRBB methodology in December 2024; the quarterly SOT flow means outlier status is visible in near-real time, not annually; and a dedicated EU-wide IRRBB stress test has become largely redundant, because the supervisor already holds continuous data that the 2017 exercise had to collect by hand. IRRBB inspections and targeted ALM reviews sit on ECB priorities through 2026-28.
Here is the uncomfortable arithmetic for anyone hoping the consultation heralds a rollback. The entire EU supervisory apparatus for interest rate risk, the heatmap, the SREP methodology, the continuous outlier monitoring, now runs on the J-template data. Simplification that removed the behavioral parameters would not trim the framework; it would unplug it.
Supervisors do not hand back data that works.
So what does EBA/CP/2026/07 realistically open? Proportionality refinement, not retreat. The reporting framework already tiers its demands: SNCIs and “other” institutions file reduced template sets. The plausible landing zone is more of that: recalibrated thresholds, trimmed granularity at the edges, perhaps consolidation of items that proved redundant in practice. Relief flows to the smaller end of the 4,500-institution population. The assumption-level core, the part that powers everything, stays.
There is a general lesson here about how the EU framework evolves. The direction of travel, visible in every revision since 2018, runs from principles toward prescriptions: guidelines hardened into technical standards, a modeling convention hardened into a quasi-limit, ad hoc data collections hardened into quarterly law. Reporting that has become the engine of supervision does not reverse down that road. The realistic best case is that it stops getting heavier.
For practitioners, the implication is architectural. If you have been running the J-templates as a quarterly fire drill, spreadsheets, manual reconciliations, heroics, on the theory that the requirement might shrink, the consultation is not your reprieve. Treat the reporting stack as permanent infrastructure. That means investing in the unglamorous things that only pay off over a long horizon: automated production from the same data that feeds internal ALM, reconciliation between reported SOT figures and internal measures, assumption data that is governed at source rather than assembled at deadline.
There is a quiet upside to that conclusion. A bank whose reported assumptions are generated by the same machinery that runs its risk management has nothing to fear from a supervisor who benchmarks models. The banks that should worry are those for which the J-templates describe a model that exists mainly in the templates.
The consultation is worth engaging with; proportionality is a real negotiation, and the burden case deserves to be made with evidence. Just do not confuse the negotiation over how much detail with a negotiation over whether. That question was settled the first time the heatmap turned reported assumptions into supervisory findings.
EU banks have filed J-templates quarterly since September 2024, and the templates carry more than risk outcomes. They carry the behavioral assumptions behind them: deposit core proportions, repricing maturities, pass-through and prepayment rates. That data now feeds the EBA's heatmap, the ECB's SREP methodology, and near-real-time outlier monitoring, machinery that a rollback would have to dismantle, not trim. EBA/CP/2026/07 is realistically a proportionality exercise: relief for smaller institutions, recalibrated thresholds, trimmed granularity at the edges. The assumption-level core stays. If your reporting stack still runs on quarterly heroics, this consultation is not the signal to wait it out. Build it as permanent infrastructure now.
Mirai ALM & Liquidity connects the same assumption data that drives your internal risk management to automated J-template production, so what you report and what you manage are never two different models. Learn more
For the full picture, including the Basel BCBS 368 framework, the deposit-modeling debates behind the SVB failure, and how the EU, UK, and US regimes each address it differently, read the complete guide "IRRBB: A Complete Reference, From Zero to Expert."
To see how Mirai Regulatory Reporting automates IRRBB measurement, EVE and NII stress testing, and regulatory reporting across these frameworks, learn more here.