IRRBB Policy and Report Template (EVE, NII, Supervisory Outlier Test)

IRRBB Policy and Report: [Institution name] [Reporting date]

Interest rate risk in the banking book (IRRBB) policy and periodic measurement report for a bank: scope (IRRBB and credit spread risk in the banking book, CSRBB), risk appetite and limits, behavioural assumptions, economic value of equity (EVE) and net interest income (NII) methodologies, supervisory outlier test results, internal scenarios, hedging, model validation, disclosure and sign-off. It follows EU Directive 2013/36/EU (CRD) Articles 84 and 98(5), Commission Delegated Regulations (EU) 2024/856 (supervisory outlier tests) and 2024/857 (standardised methodologies), EBA guidelines EBA/GL/2022/14, and the Basel Committee standard "Interest rate risk in the banking book" (2016) with the shock recalibration of July 2024 (to be implemented by 2026-01-01), cited by name only. Replace every [bracketed] value and every row marked Example. Obtain the regulations and guidelines for the normative text. This is a document format, not financial, legal, regulatory or accounting advice.

Document control

Approvals

Revision history

Contents

• 1. Policy scope, governance and appetite
• 2. Risk definitions and perimeter
• 3. Behavioural assumptions
• 4. Measurement methodology
• 5. Hedging strategy and limits framework
• 6. Report summary
• 7. Repricing profile
• 8. EVE and NII results and supervisory outlier tests
• 9. Internal scenarios and CSRBB
• 10. Model validation and data quality
• 11. Disclosure, actions and sign-off
• 12. Glossary and references
• Standard and licence note

1. Policy scope, governance and appetite

1.1 Purpose and legal basis (as of 2026-10, check local law)

This policy sets how [Institution name] identifies, measures, limits, hedges and reports the risk that changes in interest rates and credit spreads reduce its economic value or its earnings from banking-book positions. Trading-book interest rate risk is governed by the market risk framework instead.

1.2 Roles

1.3 Risk appetite and limits

2. Risk definitions and perimeter

IRRBB has three components. Gap risk arises from the timing of repricing of assets and liabilities and the shape of the yield curve, covering parallel and non-parallel moves. Basis risk arises when instruments with similar repricing tenors are priced off different reference rates, for example a loan linked to a three-month benchmark funded by deposits whose rates follow the central bank deposit rate with a lag. Option risk arises from explicit options (caps, floors, swaptions held in the banking book) and from behavioural options embedded in client products: prepayment of fixed-rate mortgages, early withdrawal of term deposits and the ability of depositors to move non-maturity balances. CSRBB is the risk from changes in the market price of credit risk, liquidity and other characteristics of banking-book instruments that are not explained by IRRBB or expected credit losses.

3. Behavioural assumptions

Behavioural assumptions convert contractual cash flows into expected cash flows. They are the largest source of model risk in IRRBB. For non-maturity deposits, split balances into stable and non-stable parts and, within the stable part, a core portion whose rate does not move with market rates, and estimate the repricing profile of the core part from at least [ten] years of data including a period of rising rates. For prepayments, estimate a base conditional prepayment rate and its sensitivity to rate incentives. For term deposits, estimate early redemption rates and their dependence on the rate environment. Each assumption must be approved by ALCO, validated independently and reported with a sensitivity that shows the effect of a reasonable alternative.

4. Measurement methodology

4.1 Economic value of equity

EVE is the present value of banking-book asset cash flows minus the present value of banking-book liability cash flows, excluding equity. The change in EVE under a shock is the difference between the post-shock and the base present values. The internal measure uses a risk-free discount curve, includes commercial margins [or excludes them, state which] and runs off the balance sheet (no new business). For the supervisory outlier test the parameters of Delegated Regulation (EU) 2024/856 apply, including the treatment of commercial margins, the post-shock interest rate lower bound and the cap on the repricing maturity of non-maturity deposits; check the current text.

For a simple illustration, the change in EVE can be approximated from net repricing positions $P_i$ in time buckets with mid-point tenors $t_i$ (used here as a stand-in for modified duration) and bucket shocks $\Delta r_i$:

A positive net position (more assets than liabilities repricing in a bucket) loses value when rates in that bucket rise. Real measurement revalues every cash flow under the full shocked curve, including optionality and the lower bound, so the approximation is for reading the results, not for producing them.

4.2 Net interest income

The NII measure projects interest income and expense over a twelve-month horizon (and a [three]-year horizon for internal steering) under a constant balance sheet: maturing positions are replaced with the same type, size and margin. Shocks are applied instantaneously. The first-order approximation for the change in twelve-month NII from repricing gaps $G_k$ that reprice at month $m_k$ is:

4.3 Internal scenarios

In addition to the six supervisory shocks, the bank runs historical scenarios (for example the 2022 to 2023 hiking cycle replayed on the current balance sheet), hypothetical scenarios agreed with the economics team, basis risk scenarios (benchmark rates and deposit rates diverging) and a reverse scenario that finds the rate path breaching the EVE limit. Results are in section 9.

4.4 Standardised methodology

Where the competent authority considers an institution's internal measurement system unsatisfactory, it may require the institution to use the standardised methodology (CRD Article 84), and small and non-complex institutions may use the simplified standardised methodology, both set out in Delegated Regulation (EU) 2024/857. Even where the internal system is used, run the standardised methodology at least [annually] as a benchmark and explain material differences.

5. Hedging strategy and limits framework

The bank's structural position is driven by fixed-rate mortgages on the asset side and stable, low-beta deposits on the liability side. The strategy is to keep EVE sensitivity within limits while preserving the earnings stability provided by the deposit franchise. Pay-fixed interest rate swaps on the mortgage book reduce the long-end asset position; receive-fixed swaps on part of the core deposit replicating portfolio stabilise NII. Hedge accounting (portfolio fair value hedge or cash flow hedge) is chosen to reflect the economic hedge in the accounts, documented separately per hedge relationship.

Limit breaches are escalated by the second line to the ALCO chair on the day of detection; ALCO decides within [five] business days whether to hedge, accept with a time-bound waiver or change the strategy, and the board risk committee is informed at its next meeting. A breach of the supervisory outlier threshold is reported to the competent authority as required.

6. Report summary

Tier 1 capital is EUR 3,400 million (17.0% of a TREA of EUR 20,000 million). Base twelve-month NII is EUR 1,050 million.

Write the narrative for a board reader in four short paragraphs: what the position is and why (the structural drivers in section 7), how it changed since the previous report and which business or hedging decision caused the change, how close each metric is to its early warning and limit, and what management proposes. Avoid restating the tables; interpret them. When EVE and NII point in opposite directions, as they do here (EVE loses in rising rates while NII gains, and the reverse in falling rates), say which risk the bank prefers to carry and why that choice is consistent with the risk appetite.

7. Repricing profile

The positive position up to one year reflects floating-rate loans and short-dated liquidity assets; the negative positions from one to five years reflect the behavioural maturity of core deposits; the positive positions beyond five years reflect fixed-rate mortgages after the pay-fixed hedges. Within the first year the net position of +4,000 reprices as +2,500 in months 0 to 3 (mid-point month 1.5), +1,000 in months 3 to 6 (mid-point 4.5) and +500 in months 6 to 12 (mid-point 9).

8. EVE and NII results and supervisory outlier tests

8.1 EVE under the six supervisory shocks

The bucket shocks below are illustrative values for the example only; the real supervisory shocks are derived from the currency-specific parameters and formulas in Delegated Regulation (EU) 2024/856.

Worked check for the parallel up shock, using the formula in section 4.1 and the positions in section 7: 4,000 x 0.5 x 2% = 40; -2,500 x 2 x 2% = -100; -1,500 x 4 x 2% = -120; 1,800 x 7.5 x 2% = 270; 500 x 12.5 x 2% = 125. The sum is 215, and with the minus sign the change in EVE is -215.0, which is 6.32% of Tier 1 (215 / 3,400). The steepener gives -(-26 + 30 + 0 + 67.5 + 50) = -121.5. In the full calculation the parallel down and short down results are smaller than shown because the post-shock lower bound limits how far rates fall.

8.2 NII under the parallel shocks

The gaps used here are after deposit pass-through (BA-03). In a falling-rate scenario deposit rates close to zero cannot fall by the full amount, so the full calculation shows a larger NII decline than the linear approximation. For simplicity the summary in section 6 shows the linear result rounded to -59; a real report enters the full calculation and explains the difference.

8.3 Supervisory outlier test results

Being inside the outlier thresholds does not by itself mean the risk is acceptable: the supervisor also considers the internal measures, the quality of the assumptions and the trend. Record the margin to each threshold and the shock size that would reach it, and explain any quarter-on-quarter movement greater than [one] percentage point of Tier 1. If an outlier threshold is breached, notify the competent authority as required, present a remediation plan to ALCO within [ten] business days, and expect supervisory measures that may include reducing the exposure or using the standardised methodology.

9. Internal scenarios and CSRBB

CSRBB is measured by revaluing the fair-valued and amortised-cost securities portfolio under an internal spread widening of 50 bp for government bonds and covered bonds (calibrated on a historical one-year spread move), with spread durations from the position system.

The loss of 81 equals 2.38% of Tier 1 (81 / 3,400). Holdings match the liquidity buffer in the ILAAP Report (finance/ilaap-report); the Level 2B RMBS (200) are omitted from this simple illustration. A floating-rate coupon removes interest rate risk but not spread risk, so a real report includes them with a spread duration close to their weighted average life (at a weighted average life of about 4 years and the same +50 bp shock, this would add 200 x 4.0 x 0.50% = 4 of loss). The ICAAP Report (finance/icaap-report) holds internal capital of EUR 300 million for IRRBB (the worst supervisory EVE loss of 215 plus an add-on of 85 for basis and option risk set by the ICAAP methodology, 215 + 85 = 300) and EUR 80 million for CSRBB (this result, rounded), without diversification between risk types.

10. Model validation and data quality

The measurement engine and behavioural models are in the model inventory and are validated under the Model Risk Management Policy (finance/model-risk-management-policy), with findings recorded in Model Validation Report (ai-data/model-validation-report) format. Data quality checks reconcile the IRRBB positions with the general ledger each month, test completeness of contractual data (repricing dates, floors, prepayment options) and track manual adjustments.

11. Disclosure, actions and sign-off

The Pillar 3 disclosure under CRR Article 448 reports the EVE and NII changes under the supervisory shocks and the key modelling assumptions, including the average and longest repricing maturity of non-maturity deposits. Ensure that the figures in this report reconcile with the disclosure and the supervisory reporting templates for the same date.

12. Glossary and references

12.1 Glossary

12.2 References

Standard and licence note

Based on Directive 2013/36/EU, Commission Delegated Regulations (EU) 2024/856 and 2024/857, Regulation (EU) No 575/2013 and EBA guidelines cited by number (EU texts reused with acknowledgement of the source). The Basel Committee IRRBB standard and its 2024 recalibration are cited by name and chapter ID only; no text, shock table or formula from them is reproduced. The illustrative bucket shocks, methodology description, limits and all wording are original to this template. Fernhollow Bank and every figure are fictional and are not a calibration recommendation. Check the current official texts and your national requirements before use. Template licence: MIT.