ESPREP-STT Reporting in Luxembourg: The Retired IRRBB Stress Test

ByRegReportingDesk

The ESPREP-STT reporting template that Luxembourg banks filed for years now sits in the CSSF archive. Circular CSSF 08/338, the instrument behind it, and the ESPREP-STT workbook itself have both been archived since 6 August 2025, and the CSSF catalogue records the circular as repealed. What that return measured, the interest rate risk in the banking book (IRRBB) stress test, moved to the harmonised European IRRBB templates from the 30 September 2024 reference date.

The measures largely carried over. The return captured the change in the economic value of equity (EVE) and in net interest income (NII) under prescribed interest rate shocks, together with the two supervisory outlier tests. The container is what changed: from an Excel workbook keyed to Circular CSSF 08/338 to XBRL templates filed under the EBA implementing technical standards on supervisory reporting. For any team reconciling a historical filing, answering a supervisory query on a pre-2024 reference date, or standing up the harmonised templates for the first time, the difference between the two regimes is worth keeping straight.

Related reading: our guide to the EBA IRRBB guidelines (EBA/GL/2022/14)

ESPREP-STT was a national reporting template. Its legal basis was Circular CSSF 08/338, published on 19 February 2008 on the implementation of a stress test to assess the interest rate risk arising from non-trading book activities. The circular was amended three times: by Circular CSSF 16/642 in 2016, by Circular CSSF 20/762 in 2020, and by Circular CSSF 24/849 in 2024.

Each amendment tracked the European standard. Circular CSSF 20/762 realigned the stress test following the adoption of the EBA guidelines on the management of interest rate risk arising from non-trading book activities (EBA/GL/2018/02). The reporting instructions that accompanied the template then pointed to the successor guidelines, EBA/GL/2022/14, which specify the criteria under Article 84(6) of Directive 2013/36/EU (the Capital Requirements Directive) for evaluating an institution’s internal IRRBB systems and for assessing credit spread risk in the banking book (CSRBB). The stress test itself was the mechanism the CSSF used to observe how sharply a rate move would erode an institution’s economic value.

The circular did more than name a test. It set the calculation rules the template relied on: the two parallel shock scenarios in paragraph 10, the six additional supervisory scenarios in paragraph 11, the modelling principles in paragraph 13, and the definition of a material currency in paragraph 13(j). Reading the archived instructions without the circular open alongside them leaves the field definitions half-explained, because the template deliberately referred back to the circular for the substance of each shock.

Who must report: the credit institutions and investment firms in scope

The CSSF marked Circular 08/338 as relevant for credit institutions and investment firms, and the template came in two variants for exactly that split. Credit institutions filed the workbook named ESPREP-BNNNN-YYYY-MM-STT.xlsx; investment firms filed ESPREP-PNNNN-YYYY-MM-STT.xlsx, where the four-digit NNNN was the institution identifier code. An institution reporting only on an individual basis completed the individual “IRRBB measures” worksheet (the N version); an institution that also reported on a consolidated basis completed both the N and C versions.

Scope also turned on whether an institution was an outlier. Outlier institutions, as determined under paragraph 12 of the circular, had to complete the “IRRBB outlier questions” worksheets. A non-outlier institution could still be asked to complete those questions on an individual or consolidated basis, but only on a specific request from the CSSF.

The harmonised regime that replaced ESPREP-STT reaches a wider population. From the 30 September 2024 reference date, the EBA implementing technical standards were extended to the full set of institutions already submitting COREP, excluding branches. Proportionality then applies by size class: large institutions (as defined in Article 4(1)(146) of the Capital Requirements Regulation), small and non-complex institutions or SNCIs (Article 4(1)(145)), and those that are neither. Third-country branches sit outside the general extension and are handled separately, which the caveats section below sets out.

What the ESPREP-STT report contains: EVE, NII and the two outlier tests

The workbook was built around a small number of worksheets: an “Identification” sheet, the “IRRBB measures” sheets in the N and C versions, and the “IRRBB outlier questions” sheets. The measures fell into three blocks.

The economic value block ran across items 1 to 10. Items 1 to 3 recorded the outcome and the change in EVE under the two parallel shock scenarios of paragraph 10 (a parallel shock down and a parallel shock up). Items 4 to 10 recorded the outcome and the change in EVE under the six additional interest rate scenarios of paragraph 11. Both the parallel-shock outcome and the six-scenario outcome were computed automatically from the currency-level figures the institution entered for each of its material currencies.

This is where the two outlier tests are easy to conflate. The template kept them as separate flags. An institution reported “Yes” in cell K12 where the decline in EVE under the parallel shocks exceeded 20% of total own funds, as defined in Article 72 of the Capital Requirements Regulation, and “Yes” in cell K17 where the decline under the six supervisory scenarios exceeded 15% of Tier 1 capital, as defined in Article 25 of the same Regulation. The 15%-of-Tier-1 test on the six scenarios is the current supervisory outlier test; the 20%-of-own-funds test on the parallel shocks carried the legacy threshold forward, and both had to be answered.

The earnings block ran across items 11 to 16: forecasted NII over a twelve-month horizon under the baseline and under a parallel shock down and up (items 11 to 13), and forecasted earnings over the same horizon under the same shocks (items 14 to 16). An institution reported at least one earnings metric, either NII or earnings or both. Items 17 to 20 then captured the proportion of banking-book assets and liabilities subject to automatic optionality and to embedded behavioural optionality, per material currency. A short set of questions and, for outliers, a dedicated questions worksheet rounded out the file, asking for the main drivers of the worst-case results, the risk-free yield curves used, the measurement methods drawn from Annexes I and II of EBA/GL/2022/14, and the modelling of non-maturity deposits (NMDs).

Two conventions governed the numbers. The reporting currency was the currency of the institution’s own funds, entered with a single unit of “1”, so that one thousand euro was reported as “1000”; institutions caught by the ECB Short-Term Exercise on IRRBB could use the euro by derogation. Empty yellow cells were not permitted: a data point that did not apply took the value “.p”, and a data point that applied but was not reported took “.v” with a justification, while “0” was reserved for a value that was genuinely zero.

Reference dates, frequency and remittance deadlines for IRRBB reporting

The archived workbook carried the reference date in cell C16 of the “Identification” sheet in dd/mm/yyyy format and the level of consolidation in cell C18 as “N” or “C”. The archived Circular CSSF 08/338 set the legacy calculation and reporting calendar; the specific terms are set out in the archived circular text, which should be consulted directly for any pre-2024 period query. Institutions reporting the ECB Short-Term Exercise IRRBB for the same level of application and period were exempt from the corresponding CSSF reporting for that level and period.

Key dates for the transition and the current regime:

  • 19 February 2008: Circular CSSF 08/338 published.
  • 2016, 2020 and 2024: amendments (Circulars CSSF 16/642, 20/762 and 24/849) aligning the stress test with successive EBA guidelines.
  • 30 September 2024: first reference date for the harmonised EBA IRRBB templates, extended to the full COREP population except branches.
  • 31 December 2025: first reference date for the third-country branch national extension (the J 01.00 template only, filed annually).
  • 6 August 2025: Circular CSSF 08/338, the reporting instructions and the ESPREP-STT template archived.

For institutions within Article 21 scope, J 01.00 is quarterly for all institutions. Large institutions also report J 02.00, J 05.00 and J 08.00 quarterly; institutions that are neither large nor SNCIs report J 03.00, J 06.00 and J 09.00 quarterly; and SNCIs report J 04.00, J 07.00 and J 09.00 quarterly. The quarterly remittance dates are 12 May, 11 August, 11 November and 11 February. J 10.01 and J 10.02 are annual for large institutions, while J 11.01 and J 11.02 are annual for institutions that are neither large nor SNCIs and for SNCIs, with annual remittance on 11 February.

How ESPREP-STT reporting was submitted: from an XLSX workbook to XBRL on eDesk

ESPREP-STT was, at heart, a spreadsheet. Institutions completed the yellow cells of the Excel workbook and submitted it through the CSSF’s prudential reporting arrangements under the file-naming pattern ESPREP-BNNNN-YYYY-MM-STT.xlsx for banks and ESPREP-PNNNN-YYYY-MM-STT.xlsx for investment firms.

The harmonised templates changed the format completely. IRRBB reporting under the EBA standards is filed in XBRL and submitted through the CSSF eDesk portal or its API, an entry point open for IRRBB submissions from the 30 September 2024 first reference date. The taxonomy tracks the EBA data point model: The IRRBB module was introduced in DPM 3.3 phase 3 (released 31 October 2023) for the advanced ad-hoc collection, and updated in DPM 3.4 (released 6 February 2024), the release that carried the full IRRBB module into production for the 30 September 2024 reference date. CSSF reporting distinguishes the accounting version from the preliminary/final version. For IRRBB, the applicable accounting version can be L, N or C depending on the institution’s reporting scope, while the preliminary version code “N” is provided separately. The move from a hand-completed workbook to a validated XBRL instance is the single largest operational change between the two regimes, and it is the reason the mapping work requires reviewing field definitions, shock scenarios and optionality treatments item by item, not simply copying values across.

Validation rules and the common rejection causes

The archived instructions required all yellow cells to be completed, permitted “0” only when the underlying value was genuinely zero, and required the outlier-questions worksheet where an institution selected “Yes” in K12 or K17. The “.p” and “.v” markers were not interchangeable, and using one for the other misstated whether a data point was inapplicable or simply omitted.

The harmonised templates are policed by formal validation. Submissions are checked against the EBA validation rules for the IRRBB module and against the CSSF plausibility checks and the ECB data-quality checks applied to prudential XBRL. Current submissions must use the applicable taxonomy, comply with the EBA validation rules and relevant CSSF/ECB checks, and carry the accounting and preliminary versions required by the CSSF Reporting Handbook for the relevant reporting scope. Because the EBA refreshes its validation rules on a quarterly cycle, teams filing the J-templates need to track the latest EBA validation rules update before each remittance, a discipline the old spreadsheet never demanded.

Caveats and interactions: proportionality, the ECB exercise and third-country branches

Several interactions sat around the ESPREP-STT return and still shape the harmonised one. Proportionality is the first. Under the harmonised templates the number of J-templates an institution files depends on its size class, so a small and non-complex institution completes a narrower set than a large institution. That was less visible in the single national workbook, which every in-scope institution completed in the same shape.

The ECB dimension is the second. Significant institutions supervised directly by the ECB were subject to an additional ad-hoc IRRBB data collection, based on the EBA decision EBA/DC/501 of 3 August 2023, for the 2023 Q4, 2024 Q1 and 2024 Q2 reference dates at the highest level of consolidation, with remittance dates aligned to the ECB Short-Term Exercise. The same Short-Term Exercise is what justified the euro reporting derogation described earlier. For the reference date 31 December 2023, only a limited sample of institutions participating in the Basel III monitoring exercise was drawn into the EBA collection.

Third-country branches are the third interaction, and the one most likely to be missed. Branches were excluded from the general extension of the harmonised templates, but a national extension applies the J 01.00 template to third-country branches on an annual basis, starting from the 31 December 2025 reference date. Finally, CSRBB is a related but distinct exposure: EBA/GL/2022/14 sets management and monitoring expectations for credit spread risk under Article 84(6)(c) of the Directive, while the IRRBB supervisory templates themselves stay focused on the EVE and NII measures.

Recent and upcoming changes for IRRBB reporting in Luxembourg

The decisive change is the migration itself. The harmonised IRRBB reporting requirements first applied under Commission Implementing Regulation (EU) 2024/855, which amended Implementing Regulation (EU) 2021/451 and applied from 1 September 2024; the first reporting reference date was 30 September 2024. Commission Implementing Regulation (EU) 2024/3117 subsequently recast the supervisory-reporting ITS and has applied from 28 June 2025, with current IRRBB reporting set out in Article 21. The CSSF Reporting Handbook records DPM 3.4 and the 30 September 2024 reference date as the point from which the ITS extended to the full population of institutions already submitting COREP, except branches.

The archival followed. On 6 August 2025 the CSSF archived Circular CSSF 08/338, the reporting instructions on IRRBB, and the ESPREP-STT template, and the catalogue entry for the circular records it as repealed by Circular CSSF 24/849. The earlier amending circulars, 16/642 and 20/762, were archived on the same day. The ad-hoc IRRBB collection under EBA/DC/501 remains the reference for the significant-institution data that bridged the 2023 to 2024 period.

Looking ahead, IRRBB reporting will keep moving with the EBA reporting framework rather than with a Luxembourg circular. Reporting teams tracking the EBA reporting framework roadmap should expect the IRRBB module to evolve alongside the wider COREP and FINREP updates, with taxonomy versions and validation rules published on the EBA’s usual release cadence.

Frequently Asked Questions

Do institutions still file the ESPREP-STT workbook for any reporting period?

No new ESPREP-STT filings are made. From the 30 September 2024 reference date, IRRBB is reported through the harmonised J-templates, and the workbook, its instructions and the circular were archived on 6 August 2025. The archived files remain relevant only for reconciling or explaining a historical filing.

Which harmonised templates replace the old EVE and NII items?

The harmonised IRRBB module comprises J 01.00; J 02.00 to J 09.00 as applicable by size class; J 10.01 and J 10.02 annually for large institutions; and J 11.01 and J 11.02 annually for institutions that are neither large nor SNCIs and for SNCIs. The exact set filed therefore depends on the institution’s size class.

How did the ECB Short-Term Exercise relate to this return?

The Short-Term Exercise is a separate ECB data collection. EBA/DC/501 covered the EBA ad-hoc IRRBB collection, for which only a limited QIS sample was concerned at the 31 December 2023 reference date. Separately, ECB Banking Supervision required the IRRBB ad-hoc module from all Significant Institutions at the highest level of consolidation for 2023 Q4, 2024 Q1 and 2024 Q2, with remittance aligned to the Short-Term Exercise. Short-Term Exercise reporters were the institutions entitled to the euro reporting derogation under the legacy CSSF instructions.

Are third-country branches inside the harmonised IRRBB templates?

Branches were left out of the general extension of the harmonised templates. A national extension applies the J 01.00 template to third-country branches annually from the 31 December 2025 reference date, so a branch that files no other IRRBB template may still owe J 01.00.

What made an institution an outlier under ESPREP-STT?

An institution flagged itself as an outlier where the decline in EVE under the two parallel shocks exceeded 20% of total own funds, or where the decline under the six supervisory scenarios exceeded 15% of Tier 1 capital. An outlier had to complete the outlier questions worksheet explaining the drivers and its action plan.

Is credit spread risk (CSRBB) captured in these templates?

CSRBB is governed by the management and monitoring expectations in EBA/GL/2022/14 under Article 84(6)(c) of the Directive. The IRRBB supervisory templates report the EVE and NII measures rather than a standalone CSRBB return.

Key Takeaways

  • Circular CSSF 08/338 and the ESPREP-STT template have been archived since 6 August 2025; no new ESPREP-STT filings are made.
  • Harmonised IRRBB reporting applies from the 30 September 2024 reference date. The applicable quarterly templates have remittance dates of 12 May, 11 August, 11 November and 11 February; the size-class-specific J 10.01/J 10.02 or J 11.01/J 11.02 templates are annual, with remittance on 11 February.
  • The current EU supervisory outlier framework differs from the two legacy ESPREP-STT flags. The EVE threshold is a decline of more than 15% of Tier 1 capital under any of the six supervisory shock scenarios, while the NII threshold is a one-year NII decline of more than 5% of Tier 1 capital under either of the two NII supervisory shock scenarios.
  • Legacy file names were ESPREP-BNNNN-YYYY-MM-STT.xlsx for credit institutions and ESPREP-PNNNN-YYYY-MM-STT.xlsx for investment firms.
  • The harmonised IRRBB module comprises J 01.00, the applicable J 02.00 to J 09.00 templates, and the annual J 10.01/J 10.02 or J 11.01/J 11.02 templates according to size class. Luxembourg submissions are made in XBRL through eDesk or API; branches are outside the general Article 21 extension, while the CSSF national extension requires third-country branches to report J 01.00 annually from the 31 December 2025 reference date.
  • Keep the archived reporting instructions on file to answer any supervisory query on a pre-2024 IRRBB reference date.

Sources and References

Keeping the IRRBB return audit-ready after the switch

The ESPREP-STT template earned its retirement, but the exposure it measured did not go anywhere, and the supervisory outlier logic it carried lives on in the J-templates. For most Luxembourg institutions the live task is now the quarterly harmonised filing, with the next remittance falling on the 12 May, 11 August, 11 November or 11 February that follows the reference date. The one housekeeping step worth taking is to reconcile the final ESPREP-STT submission against the first harmonised J-template filing and keep the archived reporting instructions where the reporting team can reach them, so that a query on a pre-2024 reference date can be answered from the record rather than from memory.

Disclaimer: The information on RegReportingDesk.com is for educational and informational purposes only. It does not constitute legal, regulatory, tax, or compliance advice. Always consult your compliance officer, legal counsel, or the relevant supervisory authority for guidance specific to your institution.

Similar Posts