Instruction No. 3/2020

Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 .................................................................................................................................................................................................. Topics Supervision :: Prudential Standards Mod. 99999911/T – 01/14

Index Text of the Instruction Annex I – Reporting Models 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 180 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 180 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 180 040 050 060 070 Issued debt securities 085 Deposits with no defined maturity: current retail 095 Deposits with no defined maturity: non-current retail 105 Deposits with no defined maturity: financial wholesalers 115 Deposits with no defined maturity: non-financial wholesalers 120 Deposits with defined maturity 130 Derivatives 140 Others 150 160 Contingent assets 170 Contingent liabilities 180 Annex III – IRRBB measurement methods Annex IV - Sophistication matrix for IRRBB measurement Annex V - Different levels of sophistication for each indicator and quantitative model Annex VI - Scenarios for the application of the "alert signal" outlier test

Text of the Instruction Subject: Update of the reporting of exposure to interest rate risk resulting from activities not included in the trading book and of the results of the 'outlier' tests assessed by the supervisor

This Instruction amends Bank of Portugal Instruction No. 34/2018, published in the 2nd Supplement of the Official Gazette No. 12/2018, of December 26, 2018 (Instruction No. 34/2018), with the objective of updating the standardized reporting of exposure to interest rate risk resulting from activities not included in the trading book and of the impact on economic value and financial margin resulting from a sudden and unexpected change of 200 basis points in the yield curve.

Among other aspects, through this amendment, updates resulting from the European Banking Authority Guidelines on the management of interest rate risk resulting from activities not included in the trading book (previously designated as Guidelines EBA/GL/2015/08 and now designated as Guidelines EBA/GL/2018/02), of July 19, 2018, are implemented.

Considering the provisions of Article 16(3) of Regulation (EU) No. 1093/2010 of the European Parliament and of the Council of November 24, which established the European Banking Authority, it is the responsibility of the Bank of Portugal, as the competent national authority, as well as of institutions, to ensure compliance with the guidelines and recommendations issued by the EBA, notably the aforementioned Guidelines EBA/GL/2018/02.

The provisions provided for in Instruction No. 34/2018 are thus updated for the purposes of applying the provisions of Article 116-A of the General Regime of Credit Institutions and Financial Companies, approved by Decree-Law No. 298/92 (“RGICSF”), which establishes that the Bank of Portugal must analyze the provisions, strategies, processes, and mechanisms applied by institutions to assess the risks to which they are or may be exposed. In particular, paragraph 5 of the same article is highlighted, according to which this analysis and assessment include the exposure of credit institutions to interest rate risk resulting from banking book activities, and measures are necessary, at least in the case of institutions whose economic value suffers a reduction corresponding to more than 20% of their own funds, following a sudden and unexpected change in the interest rate of 200 basis points or amplitude provided for in guidelines of the European Banking Authority (EBA) on the matter (hereinafter designated as the “outlier test”).

Consequent upon the entry into force of the aforementioned Guidelines, the principles underlying the calculation of the outlier test result are thus updated and new principles are introduced. In particular, a set of principles is clarified, including, among others, the need to calculate the change in economic value by currency, the definition of the associated significant currency, as well as the inclusion of non-performing exposures.

Point changes were also introduced to the reporting parameters, namely through a change in the disaggregation of exposure to deposits with no defined maturity and the introduction of the possibility of reporting exposure by time bands conditional on the scenarios applied to the yield curves, in order to respond to the new risk management requirements introduced by the Guidelines.

The new reporting will also allow institutions to communicate to the Bank of Portugal the result of the application of a second outlier test, designated as the “alert signal” outlier test.

Additionally, the reporting frequency will be changed to quarterly in cases where the institution calculates a variation in economic value of capital greater than 20% of own funds or 15% of Tier 1 own funds, depending on the respective associated outlier test, thus embodying the reporting obligation provided for in the new Guidelines.

The draft of this Instruction was subject to public consultation, in accordance with legal provisions.

In these terms, the Bank of Portugal, in the exercise of the competence conferred upon it by Article 17 of its Organic Law, approved by Law No. 5/98, of January 31, in its current wording, and by the combined provisions of paragraph 1 and paragraph 2 of Article 120, paragraph 1 of Article 121-A, and paragraph 1 of Article 196, all of the RGICSF, approves the following Instruction:

Article 1. Object This Instruction has as its object to amend Bank of Portugal Instruction No. 34/2018, published in the 2nd Supplement of the Official Gazette No. 12/2018, of December 26 (“Instruction No. 34/2018”), in order to update the specific rules and procedures applicable to the provision of standardized format information regarding interest rate risk resulting from activities not included in the trading book (also designated “Interest Rate Risk in the Banking Book” or, abbreviately, “IRRBB”).

Article 2. Amendments to Instruction No. 34/2018 1 – Articles 1, 4, 5, 6, 7, 8, and 10 of Instruction No. 34/2018 shall have the following wording:

“Article 1. Object This Instruction establishes the specific rules and procedures applicable to the provision of standardized format information regarding: a) […] b) The impact on economic value resulting from a sudden and unexpected change of 200 basis points in the yield curve (“outlier test”); c) The impact on the expected financial margin at one year resulting from a sudden and unexpected change of 200 basis points in the yield curve; d) The impact on economic value resulting from the application of the scenarios provided for in Annex VI of this Instruction (“alert signal” outlier test).

Article 4. Reporting of the outlier test result and financial margin variation 1 - The entities referred to in paragraph 1 of Article 2 must report to the Bank of Portugal the impact of a sudden parallel variation of +/- 200 basis points (bps) in the yield curve on economic value. 2 - The entities referred to in paragraph 1 of Article 2 must report to the Bank of Portugal the impact of a sudden parallel variation of +/- 200 basis points (bps) in the yield curve on the expected financial margin at 1 year.

Article 5. Calculation methods 1 - When calculating the impact of the scenarios referred to in Articles 4 and 4-A on economic value in accordance with the previous article, the entities referred to in paragraph 1 of Article 2 must use one of the calculation methods contained in Annexes III and V of this Instruction, according to the sophistication level of the entity, assessed in accordance with Annex IV of this Instruction. 2 - When calculating the effect of the scenarios referred to in paragraph 2 of Article 4 on the expected financial margin at 1 year, entities must use one of the calculation methods indicated in Annexes III and V of this Instruction, according to the sophistication level of the entity, assessed in accordance with Annex IV of this Instruction. 3 – […]

Article 6. Complementary reporting The reports and calculations referred to in Articles 3, 4, and 4-A of this Instruction must be complemented with a report containing the following elements: a) A description of the method or methods used to calculate the impact of interest rate changes on economic value and financial margin, and the respective sophistication level in accordance with Annexes III, IV, and V of this Instruction; b) The risk-free yield curve(s) considered in letter u) of paragraph 1 of Article 4-B of this Instruction; c) A description of the hypotheses and assumptions assumed, both in the calculation of changes in economic value and in the financial margin resulting from the application of the scenarios provided for in Articles 4 and 4-A, and in particular regarding: i. the treatment given to elements whose periods of maturity or effective rate repricing diverge from contractual terms and regarding the treatment of elements with no defined contractual term; ii. the treatment of embedded automatic options (possible minimum and maximum limits of specific interest rates for instruments) and explicit automatic options. d) If the institution excludes commercial margins and other spread components in the calculation performed to determine variations in economic value in the outlier tests referred to in Articles 4 and 4-A, a description of the methods referred to in letter l) of Article 4-B of this Instruction.

Article 7. Reporting models 1 – The information required in Article 3, as well as the information from the calculations provided for in Articles 4 and 4-A, is provided according to the reporting models indicated below, contained in Annex I to this Instruction: a) […] b) […] c) IRRBB 03.00 – Outlier test results and application of the standard 200 bps shock on the expected financial margin at 1 year; d) IRRBB 04.00 – “Alert signal” outlier test results.

Article 8. Frequency 1 - Entities must report to the Bank of Portugal the information mentioned in Articles 3, 4, 4-A, and 6 of this Instruction, with a frequency: a) Quarterly, with reference to March 31, June 30, September 30, and December 31, if the calculation performed by the entity results in a negative variation in economic value greater than: i. 20% of own funds, according to paragraph 1 of Article 4, or ii. 15% of Tier 1 core own funds, according to Article 4-A. b) Semi-annually, with reference to June 30 and December 31, if none of the thresholds referred to in the previous letter are exceeded. 2 – […] 3 – The information with reference to: a) March 31 must be reported by May 31 of the same year; b) June 30 must be reported by August 30 of the same year; c) September 30 must be reported by November 30 of the same year; d) December 31 must be reported by February 28 of the following year. 4 – In order to ensure adequate monitoring of the evolution of interest rate risk exposure in the banking book of entities that exceed the thresholds referred to in letter a) of paragraph 1 of this article, the quarterly reporting obligation only ceases after 2 consecutive quarters in which the referred thresholds are not met.

Article 10. Exemption request for investment firms 1 – […] 2 – […] 3 – This exemption does not apply in cases where entities exceed the limits provided for in letter a) of paragraph 1 of Article 8, applying the reporting obligations provided for in that article.”

2 - Annexes I, II, III, IV, and V published in Instruction No. 34/2018 shall have the wording given in Annexes I, II, III, IV, and V to this Instruction, respectively.

Article 3. Additions to Instruction No. 34/2018 1 – Articles 4-A and 4-B are added to Instruction No. 34/2018, with the following wording:

“Article 4-A. Reporting of the “alert signal” outlier test results The entities referred to in paragraph 1 of Article 2 must report to the Bank of Portugal the impact of sudden variations in the yield curve resulting from the application of scenarios 1 to 6 on economic value as established in Annex V of this Instruction.

Article 4-B. Principles underlying the calculation of outlier tests 1 - When calculating the impact of sudden variations in the yield curve on economic value in accordance with Articles 4, paragraph 1, and 4-A, entities must take into consideration, in particular, the following: a) All positions of instruments sensitive to interest rates must be included;”

Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14

b) Operations in small trading portfolios, in accordance with Article 94 of Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012, must be included, unless their interest rate risk is captured in another risk indicator; c) All principal Tier 1 capital instruments and other perpetual capital instruments without call dates must be excluded from the calculation; d) Entities must reflect automatic and behavioral options in the calculation, adjusting the main assumptions of behavioral modeling to the characteristics of different interest rate scenarios; e) Pension plan obligations and pension plan assets must be included, unless their interest rate risk is captured in another risk indicator; f) Cash flows arising from interest rate-sensitive instruments must include any principal repayment, any principal revaluation, and any interest payments; g) Entities with a gross non-performing exposure ratio greater than 2% must include non-performing exposures as instruments sensitive to interest rates, net of impairments; h) The gross non-performing exposure ratio referred to in the previous subparagraph must correspond to the gross total of debt securities and loans and advances considered non-performing, in accordance with Commission Implementing Regulation (EU) No 680/2014 of 16 April 2014 establishing implementing technical standards with regard to reporting by institutions in accordance with Regulation (EU) No

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 IRRBB 03.00 - Results of the outlier test and application of the standard 200 bps shock to the 1-year expected net interest income Total and Significant Currencies Row Column 010 Amount 020 Change in the economic value of the banking book resulting from a parallel shift up in the yield curve after the standard 200 bps shock 030 Change in the economic value of the banking book resulting from a parallel shift down in the yield curve after the standard 200 bps shock 040 Estimated 1-year expected net interest income in a base scenario 050 Change in the estimated 1-year expected net interest income resulting from a parallel shift up in the yield curve after the standard 200 bps shock 060 Change in the estimated 1-year expected net interest income resulting from a parallel shift down in the yield curve after the standard 200 bps shock IRRBB 04.00 – Results of the "alert signal" outlier test Total and Significant Currencies Row Column 010 Amount 010 Change in economic value in a given shock scenario 020 Parallel shift up in the yield curve 030 Parallel shift down in the yield curve 040 Steepening of the yield curve 050 Flattening of the yield curve 060 Increase in short-term rates 070 Decrease in short-term rates 080 Magnitude of interest rate shocks by currency 090 Parallel shock 100 Short-term rate shock 110 Long-term rate shock

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 Annex II – Filling Notes

1. For the application of Article 3 of this Instruction, entities project and aggregate by time bands the future cash flows arising from the notional repricing of all balance sheet positions and off-balance sheet elements included in the banking book and sensitive to interest rates, namely: a) Assets, not deducted from Tier 1 core own funds, present in the banking book; b) Liabilities, including all non-interest-bearing deposits, and excluding the elements that make up Tier 1 core own funds and other perpetual own funds without call dates; c) Off-balance sheet elements.
2. The future cash flows arising from the notional repricing mentioned in the previous paragraph are defined as: a) Any repayment of the principal of an instrument; b) Any repricing of the principal of an instrument whenever it occurs on the nearest date on which the entity or its respective counterparty can unilaterally implement changes to the interest rate, or there is an automatic change in the rate of a variable rate instrument resulting from a change in an external benchmark; c) Any interest payment on the principal of an instrument that has not yet been subject to repricing or repayment.
3. Entities have the possibility to deduct the commercial margin or other spread components from the interest rate of the future cash flows arising from the notional repricing relative to the risk-free interest rate, in accordance with Article 4-B of this Instruction.
4. The IRRBB 01.00, IRRBB 02.00, IRRBB 03.00 and IRRBB 04.00 reporting models must be reported separately by "Total" and by "significant currency". A currency is classified as significant in accordance with points r) and s) of Article 4-B of this Instruction.
5. Significant currency reports must be expressed in the currency in which the instruments are denominated.
6. Without prejudice to specific indications, the reference exchange rate of the Bank of Portugal at the reference date of the report should be used for instruments contracted in currencies other than the euro for which it is necessary to aggregate the report on the "Total" sheets. IRRBB 01.00 - Distribution by time bands of future cash flows arising from notional repricing - pre-modelling
7. This model captures the projection and aggregation by time bands performed in accordance with the previous paragraphs, reflecting the contractual conditions of the instruments; that is, no modelling of future cash flows arising from notional repricing should be reflected.
8. Regarding fixed-rate instruments, all interest payments and total or partial periodic repayments of the principal must be allocated to the time bands corresponding to the time period between the contractual payment date and the reference date of the report (columns 030 – 210), disaggregated by type of instrument.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 the time period between the contractual date of payment and the reference date of the report (columns 030 – 210), disaggregated by type of instrument. 9. It is assumed that variable-rate instruments are fully repriced on the first interest rate repricing date. Thus, the entire principal must be allocated to the time band corresponding to the time period between the interest rate repricing date and the reference date of the report (columns 230 – 410), with no additional allocation of future cash flows arising from notional repricing in subsequent time bands. The interest payment components relating to the spread on a portion of the principal that has not yet been repaid must be allocated according to the contractual date of its repayment, and should be treated as fixed-rate instruments in accordance with the previous paragraph (and allocated in columns 230-410). 10. Instruments with no defined contractual maturity must be allocated to column 010 - Other elements. 11. Non-performing exposures must be included net of impairment, and considered as instruments with no defined contractual maturity for the purposes of reporting the IRRBB 01.00 model. 12. The disaggregation of deposits with no defined maturity must be carried out taking into account that: a) retail deposits considered to be held in a current account ("retail current") include non-interest-bearing accounts and other retail accounts whose interest-bearing component is not relevant to the customer's decision to hold money in the account; b) retail deposits considered to be held in a non-current account ("retail non-current") include retail accounts whose interest-bearing component is relevant to the customer's decision to hold money in the account; c) non-financial wholesale deposits include corporate customer accounts and other wholesale customer accounts, but exclude financial counterparty accounts. 13. Positions in derivative instruments must be calculated in accordance with Section 1 of Chapter 2 of Title IV of Part 3 of Regulation (EU) No. 575/2013, and allocated in accordance with the previous paragraphs. Columns 010 Other elements Elements with no defined contractual maturity that, by definition, cannot be allocated to a time band in accordance with contractual conditions. 030 – 210 Remaining contractual maturity of fixed-rate instruments in accordance with contractual conditions Elements allocated in accordance with 19 time bands in compliance with the instructions defined in paragraph 8. 230 – 410 Next repricing date of variable-rate instruments in accordance with contractual conditions Elements allocated in accordance with 19 time bands in compliance with the instructions defined in paragraph 9.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 Rows 010 Assets Instruments reflected in table F01.01 of Regulation No. 680/2014. 020 Debt securities Instruments reflected in table F01.01 of Regulation No. 680/2014. 030 Loans and advances Instruments reflected in table F01.01 of Regulation No. 680/2014. 040 Derivatives Instruments reflected in table F01.01 of Regulation No. 680/2014. 050 Other Other instruments reflected in table F01.01 of Regulation No. 680/2014 that are not classified as debt securities, loans and advances, and derivatives. 060 Liabilities Instruments reflected in table F01.02 of Regulation No. 680/2014. 070 Issued debt securities Instruments reflected in table F01.02 of Regulation No. 680/2014. 085 Deposits with no defined maturity: Retail current Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is classified as retail, and segmented as current in accordance with point a) of paragraph 12 of this Annex. 095 Deposits with no defined maturity: Retail non-current Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is classified as retail, and segmented as non-current, in accordance with point b) of paragraph 12 of this Annex. 105 Deposits with no defined maturity: Financial wholesale Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is financial, not being classified as retail. 115 Deposits with no defined maturity: Non-financial wholesale Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is non-financial, not being classified as retail, in accordance with point c) of paragraph 12 of this Annex. 120 Deposits with defined maturity Instruments reflected in table F01.02 of Regulation No. 680/2014. 130 Derivatives Instruments reflected in table F01.02 of Regulation No. 680/2014. 140 Other Other instruments reflected in table F01.02 of Regulation No. 680/2014 that are not classified as issued debt securities, deposits (with and without defined maturity), and derivatives. 150 Off-balance sheet elements 160 Contingent assets Instruments reflected in table F09.01 of Regulation No. 680/2014. 170 Contingent liabilities Instruments reflected in table F09.02 of Regulation No. 680/2014. IRRBB 02.00 - Distribution by time bands of future cash flows arising from notional repricing - post-modelling 14. This model captures the projection and aggregation of future cash flows arising from notional repricing that entities use for the calculation performed in accordance with Article 4, paragraphs 1 and 4-A of this Instruction by 19 time bands. Thus, any modelling of future cash flows arising from notional repricing relating to instruments whose effective maturity or interest rate repricing diverges from contractual maturities must be reflected.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 of future cash flows arising from notional repricing relating to instruments whose effective maturity or interest rate repricing diverges from contractual maturities. 15. Long positions (rows 010 and 160) and short positions (rows 060 and 170) must be offset against each other and produce a single long or net position per time band, given by row 180, which corresponds to the net position. 16. The classification by type of instrument carried out in the IRRBB 02.00 models must be consistent with that carried out for the IRRBB 01.00 model. 17. If the entity does not model the future cash flows arising from the notional repricing of an instrument, the time band allocation performed in the IRRBB 01.00 model must be consistent with that used in the IRRBB 02.00 models. 18. The institution must report the exposure provided for in the IRRBB 02.00 models according to the method it uses to determine the behavioural maturity of instruments (unconditional, in cases where the modelling of the behavioural maturity of instruments is independent of specific interest rate scenarios, or conditional, where the modelling predicts that the maturity of instruments is dependent or partially dependent on specific interest rate scenarios). If the institution uses conditional maturity estimation models, it must report the IRRBB 02.00 table by interest rate scenario considered in the outlier test provided for in Article 4-A of this Instruction. Columns 020 - 200 Maturity bands of notional repricing cash flows after modelling Elements allocated in accordance with 19 time bands after modelling Rows 010 Assets Instruments reflected in table F01.01 of Regulation No. 680/2014. 020 Debt securities Instruments reflected in table F01.01 of Regulation No. 680/2014. 030 Loans and advances Instruments reflected in table F01.01 of Regulation No. 680/2014. 040 Derivatives Instruments reflected in table F01.01 of Regulation No. 680/2014. 050 Other Other instruments reflected in table F01.01 of Regulation No. 680/2014 that are not classified as debt securities, loans and advances, and derivatives. 060 Liabilities Instruments reflected in table F01.02 of Regulation No. 680/2014. 070 Issued debt securities Instruments reflected in table F01.02 of Regulation No. 680/2014. 085 Deposits with no defined maturity: Retail current Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is classified as retail, and segmented as current in accordance with point a) of paragraph 12 of this Annex.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 095 Deposits with no defined maturity: Retail non-current Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is classified as retail, and segmented as non-current, in accordance with point b) of paragraph 12 of this Annex. 105 Deposits with no defined maturity: Financial wholesale Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is financial, not being classified as retail. 115 Deposits with no defined maturity: Non-financial wholesale Instruments reflected in table F01.02 of Regulation No. 680/2014, where the counterparty is non-financial, not being classified as retail, in accordance with point c) of paragraph 12 of this Annex. 120 Deposits with defined maturity Instruments reflected in table F01.02 of Regulation No. 680/2014. 130 Derivatives Instruments reflected in table F01.02 of Regulation No. 680/2014. 140 Other Other instruments reflected in table F01.02 of Regulation No. 680/2014 that are not classified as issued debt securities, deposits (with and without defined maturity), and derivatives. 150 Off-balance sheet elements 160 Contingent assets Instruments reflected in table F09.01 of Regulation No. 680/2014. 170 Contingent liabilities Instruments reflected in table F09.02 of Regulation No. 680/2014. 180 Net position Position given by the calculation performed in accordance with paragraph 15. IRRBB 03.00 - Results of the outlier test 19. This model captures the results obtained by institutions in the calculation of the standard 200 basis point shock on economic value and 1-year expected net interest income in accordance with Article 4 of this Instruction. 20. In the "Total" report, in particular, the result of the "outlier test" (rows 020 and 030) must follow the formula present in point t) of Article 4-B of this Instruction. Rows 020 Change in the economic value of the banking book resulting from a parallel shift up in the yield curve after the standard 200 bps shock Result obtained through the application of paragraph 1 of Article 4 of this Instruction. 030 Change in the economic value of the banking book resulting from a parallel shift down in the yield curve after the standard 200 bps shock Result obtained through the application of paragraph 1 of Article 4 of this Instruction. 040 Estimated 1-year expected net interest income in a base scenario Result of the estimation of 1-year expected net interest income, given by the difference between estimated 1-year interest income and estimated 1-year interest expenses of balance sheet positions and off-balance sheet elements included in the banking book and sensitive to interest rates. 050 Change in the estimated 1-year expected net interest income resulting from a parallel shift up in the yield curve after the standard 200 bps shock of the supervisor Result obtained through the application of paragraph 2 of Article 4-A of this Instruction.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Supervision Topics :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 060 Change in the estimated 1-year expected net interest income resulting from a parallel shift down in the yield curve after the standard 200 bps shock of the supervisor Result obtained through the application of paragraph 2 of Article 4-A of this Instruction. IRRBB 04.00 - Results of the "alert signal" outlier test 21. This model captures the results obtained by entities in the calculation of the "alert signal" outlier test on economic value in accordance with Article 4-A of this Instruction. 22. In the "Total" report, in particular, the result of the "alert signal" outlier test (rows 020 to 070) must follow the formula present in point t) of Article 4-B of this Instruction. 23. The magnitude of the interest rate shocks used for the calculation of the different interest rate change scenarios by significant currency used in the "alert signal" outlier test (rows 090-110) must be consistent with that established in Annex VI of this Instruction. Rows 010 Change in the economic value of the banking book 020 Change in the economic value of the banking book resulting from a parallel shift up scenario in the yield curve Result obtained through the application of Article 4-A of this Instruction. 030 Change in the economic value of the banking book resulting from a parallel shift down scenario in the yield curve Result obtained through the application of Article 4-A of this Instruction. 040 Change in the economic value of the banking book resulting from a steepening scenario of the yield curve, which corresponds to a decrease in short-term rates and an increase in long-term rates Result obtained through the application of Article 4-A of this Instruction. 050 Change in the economic value of the banking book resulting from a flattening scenario of the yield curve, which corresponds to an increase in short-term rates and a decrease in long-term rates Result obtained through the application of Article 4-A of this Instruction. 060 Change in the economic value of the banking book resulting from a short-term interest rate increase scenario Result obtained through the application of Article 4-A of this Instruction. 070 Change in the economic value of the banking book resulting from a short-term interest rate decrease scenario Result obtained through the application of Article 4-A of this Instruction. 080 Magnitude of interest rate shocks by significant currency 090 Parallel shock Magnitude of the parallel shock applied by significant currency, in accordance with Annex VI of this Instruction. 100 Short-term rate shock Magnitude of the short-term rate shock applied by significant currency, in accordance with Annex VI of this Instruction. 110 Long-term rate shock Magnitude of the long-term shock applied by significant currency, in accordance with Annex VI of this Instruction.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Annex III – IRRBB Measurement Methods

Cash Flow Modeling

Metric	Description	Risks Covered	Limitations of the Metric Used
Unconditional Cash Flows (assumes that the remaining maturity or repricing date of cash flows is independent of a specific interest rate scenario)

Based on Results: Shift Analysis: Repricing Shift The shift analysis groups all relevant instruments sensitive to interest rates into a predetermined number of time bands by remaining maturity or repricing date, which can be determined contractually or based on behavioral assumptions. It calculates net positions (“shifts”) in each time band. It shows the change in net interest income resulting from the variation in the yield curve, by multiplying each net position by the change in the interest rate.

Shift Risk (only parallel shift risk of the curve) • The metric shows shift risk only linearly. • It is based on the assumption that all positions in a given time band mature or reprice simultaneously. • It does not allow measuring basis risk and option risk.

Economic Value: Duration Analysis: Modified Duration / PV01 of Equity Modified duration shows the relative change in the net present value of a financial instrument due to marginal parallel variations of one percentage point of the yield curve. The modified duration of equity measures the institution’s banking book exposure to shift risk. The PV01 of equity is calculated from the modified duration of equity and expresses the absolute change in the value of equity resulting from a parallel variation of one basis point (0.01%) in the yield curve. The starting point is to group all cash flows of interest rate-sensitive instruments into time bands. For each type of instrument, an appropriate yield curve is selected. The modified duration of each instrument is calculated from the change in its net present value caused by a parallel variation of 1 percentage point in the yield curve. The modified duration of equity is calculated by multiplying the modified duration of assets by the value of assets and dividing the result by equity minus the modified duration of liabilities multiplied by the value of liabilities divided by equity. The PV01 of equity is calculated by multiplying the modified duration of equity by the value of equity (i.e., assets minus liabilities) and dividing by 10,000 to obtain the value change in basis points.

Shift Risk (only parallel risk) • Applies only to marginal variations of the yield curve. In the presence of convexities, it may underestimate the impact of more significant interest rate movements. • Applies only to parallel variations of the yield curve. • It does not allow measuring option risk and, at best, only partially detects basis risk.

• Partial Modified Duration / Partial PV01 The partial modified duration of an instrument in a given time horizon is calculated in the same way as the modified duration described above, except that a parallel variation is not applied to the entire yield curve, but only to the segment corresponding to the desired time horizon. These partial indicators show the sensitivity of the market value of the banking book to a marginal variation of the yield curve in certain maturity segments. To each partial time indicator, a variation of different magnitude can be applied, through which the effect of the change in the shape of the yield curve on the entire portfolio can be calculated.

Shift Risk (parallel and non-parallel risk) • Applies only to marginal variations of interest rates. In the presence of convexities, it may underestimate the impact of more significant interest rate movements. • It does not allow measuring basis risk and option risk.

Cash Flows Partially or Fully Dependent on the Interest Rate Scenario (it is assumed that the remaining maturity or repricing of cash flows of options, instruments with embedded options, explicit options and — in more sophisticated approaches — of instruments whose maturity depends on customer behavior, is modeled conditionally to the considered interest rate scenario)

Based on Results: Focus on the Net Interest Income (NII) Component: • Change in NII The change in NII is a results-based indicator and measures the change in net interest income over a given time horizon (usually 1-5 years) caused by a sudden or gradual change in interest rates. The starting point is the allocation of all cash flows of interest rate-sensitive instruments into time bands (granular) (or, in more sophisticated systems, using the exact repricing dates of each position). The base scenario for calculations reflects the institution’s current business plan to predict the volume, price, and repricing date of future commercial transactions. The interest rates used to calculate future cash flows in the base scenario are obtained from forward rates, applying appropriate spreads or expected market rates to the different instruments. To assess the potential magnitude of variations in NII, banks use assumptions and models that allow them to predict the trajectory of interest rates, the amortization of assets, liabilities, and existing off-balance sheet items, as well as their potential replacement.

Results-based indicators can be differentiated according to the sophistication of the forecast of future cash flows: in simple amortization models, it is assumed that existing assets and liabilities mature without being replaced; in static balance sheet models, it is assumed that matured assets and liabilities are replaced by identical instruments; while dynamic and more complex cash flow models reflect the business response to various interest rate contexts in the size and composition of the banking book. All results-based indicators can be used in a scenario or in a stochastic analysis. Earnings at Risk (EaR) is an example of this latter analysis, measuring the maximum change in NII at a given confidence level.

Shift Risk (parallel and non-parallel), basis risk and, provided all cash flows are scenario-dependent, also option risk • Sensitivity of results to modeling and behavioral assumptions • Complexity

Economic Value: Focus on the Economic Value of Equity (EVE) • Change in EVE The change in EVE corresponds to the change in the net present value of all cash flows of assets, liabilities, and off-balance sheet items of the banking book resulting from a change in interest rates, assuming that all positions in the banking book are held to maturity. Interest rate risk can be assessed through ∆EVE for specific interest rate scenarios or through the distribution of ∆EVE in Monte Carlo simulations or historical simulations. Economic Value at Risk (EVaR) is an example of these latter, measuring the maximum change in the value of equity at a given confidence level.

Shift Risk (parallel and non-parallel), basis risk and, provided all cash flows are scenario-dependent, also option risk. • Sensitivity of results to modeling and behavioral assumptions • Stochastic indicators, which apply distribution assumptions, may fail to capture tail risks, nor the absence of linearization • The Monte Carlo simulation full repricing method is computationally demanding and can be difficult to interpret (“black box”) • Complexity

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 Annex IV - Sophistication Matrix for IRRBB Measurement

Institutions must apply, at a minimum, the level of sophistication indicated in Annex V corresponding to their categorization under the following categories:

• Category 1: entities classified as Global Systemically Important Institutions (G-SII) and other Systemically Important Institutions (O-SII) and, if applicable, other entities identified by the Bank of Portugal, based on the assessment of the size and internal organization of the entity, as well as the nature, scope, and complexity of its activities. • Category 2 – medium-large entities different from those included in category 1, which operate at a national level or develop considerable cross-border activities, are present in several business segments, including non-banking activities, and offer credit and financial products to corporate and retail clients. Specialized entities without systemic importance, whose business segments or payment systems hold significant market shares, or with significant financial exchanges. • Category 3 – small-medium entities that do not meet the classification criteria for categories 1 or 2, which operate at a national level or whose cross-border operations are not significant and that carry out their activity in a limited number of segments, mainly offering credit products to corporate and retail clients and having a limited range of financial products. Specialized entities, whose business segments or payment systems hold less important market shares, or with less significant financial exchanges. • Category 4 – all other national entities that are not very complex and of small size that do not fall into categories 1 to 3 (for example, with a limited scope of activities and whose business segments hold insignificant market shares).

If the complexity or diversity of an entity’s business model is significant, the entity must, regardless of its size, apply and implement risk measures that correspond to its specific business model and that adequately capture all sensitivities. All significant sensitivities to changes in interest rates must be duly captured, including sensitivity to behavioral assumptions.

Entities that offer financial products containing embedded options must use measurement systems that adequately capture the dependence of options on changes in interest rates. Entities with products that offer behavioral optionality to customers must use appropriate methods for modeling conditional cash flows to quantify IRRBB with regard to changes in customer behavior likely to occur in different interest rate stress scenarios.

The four categories indicated in the sophistication framework of Annex V reflect the categorization of entities established above. The different categories reflect different sizes and structures, as well as the nature, scope, and complexity of the institutions' activities, with Category 1 corresponding to the most sophisticated entities.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Annex V - Different Levels of Sophistication for Each Indicator and Quantitative Model

IRRBB Indicators and Modeling	Indicative Supervisory Expectations Regarding IRRBB Indicators and Modeling According to the Entity's Sophistication Category
Cash Flow Modeling	Metric
Unconditional Cash Flows (assumes that the remaining maturity or repricing date of cash flows is independent of a specific interest rate scenario)

Based on Results: Shift Analysis: • Repricing Shift Time bands recommended in the document “Principles for the Management and Supervision of Interest Rate Risk in the Banking Book” of the Basel Committee on Banking Supervision (BCBS) (April 2016). Shift based on the evolution of the size and composition of the banking book, resulting from the corporate response to different interest rate contexts. Including expected commercial margins in accordance with the interest rate scenario.

Time bands recommended in BCBS standards. Application of standard shocks. Yield curve model with maturities corresponding to the time bands.

Time bands recommended in BCBS standards, with application of partial duration. Partial duration calculated by instrument type and by time band. Application of standard shocks and other stress and shock scenarios on interest rates. Yield curve model with maturities corresponding to the time horizons.

Partial duration calculated by operation and by time band. Application of standard shocks and other stress and shock scenarios on interest rates. Yield curve model with maturities corresponding to the time bands.

Other stress and shock scenarios on interest rates. Yield curve model with maturities corresponding to the time bands.

Economic Value: Duration Analysis: • Modified Duration / PV01 of Equity • Partial Modified Duration / Partial PV01

Cash Flows Partially or Fully Dependent on the Interest Rate Scenario (it is assumed that the remaining maturity or repricing of cash flows of options, instruments with embedded options, explicit options and — in more sophisticated approaches — of instruments whose maturity depends on customer behavior, is modeled conditionally to the considered interest rate scenario)

Based on Results: • Net Interest Income (NII) Standard shocks applied to results in a static balance sheet. Based on time horizons recommended in BCBS standards.

Standard shocks and other stress and shock scenarios on interest rates for the yield curve applied to results, reflecting a static balance sheet or simple assumptions about the future development of activity.

Standard shocks and other stress and shock scenarios on interest rates for the yield curve and between market reference rates applied separately to results forecast in the business plan or in a static balance sheet. Inclusion of expected commercial margins in accordance with the interest rate scenario.

Comprehensive stress and interest rate scenarios, combining variations in yield curves with changes in basis and credit spreads, as well as changes in customer behavior, used to predict business volumes and results in order to quantify the difference compared to the underlying business plan. Inclusion of expected commercial margins in accordance with the interest rate scenario.

Economic Value: • Economic Value of Equity (EVE) Application of standard shocks and other stress and shock scenarios on interest rates for the yield curve, using the time horizons recommended in BCBS standards; yield curve maturities correspond to the time horizons.

Indicator calculated based on operations or cash flows. Application of standard shocks and other stress and shock scenarios on interest rates for the yield curve and between market reference rates separately. Adequate maturities in yield curves. Full evaluation of optionality.

Comprehensive stress and interest rate scenarios, combining variations in yield curves with changes in basis and credit spreads, as well as changes in customer behavior. Adequate maturities in yield curves. Full evaluation of optionality. Scenario analysis complemented by Monte Carlo simulations or historical simulations in portfolios with significant optionality. Daily update of risk factors.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14 Annex VI - Scenarios for the Application of the “Alert Signal” Outlier Test

Chapter 1 - Interest Rate Shock Scenarios and Magnitude of Shocks

The six shock scenarios applied to interest rates for the measurement of EVE within the framework of the “alert signal” outlier test are: (i) parallel yield curve up shock; (ii) parallel yield curve down shock; (iii) increase in the slope of the yield curve, which corresponds to decreases in short-term rates and increases in long-term rates (steepener shock); (iv) decrease in the slope of the yield curve, which corresponds to an increase in short-term rates and a decrease in long-term rates (flattener shock); (v) increase in short-term interest rates, which corresponds to a maximum of the up shock at the shortest time point of the yield curve, decreasing in amplitude to zero at the maximum point (short rate shock up); and (vi) decrease in short-term interest rates, which corresponds to a maximum of the down shock at the shortest time point of the yield curve, decreasing in amplitude to zero at the maximum point (short rate shock down).

The six shock scenarios mentioned above must be applied to interest rates to calculate parallel and non-parallel shift risks for EVE. These scenarios are applied separately to IRRBB exposures in each of the currencies in which the entity has significant positions.

The magnitude of the shock for the six interest rate scenarios is based on a historical interest rate record. More precisely, to capture the environment and interest rate cycles of each jurisdiction, a historical time series for various maturities, between 2000 and 2015, was used in order to calculate parallel, short, and long shocks to the yield curve for a given currency. However, deviations from the aforementioned 16-year period are permitted, provided they better reflect the idiosyncratic circumstances of a given jurisdiction.

Table 1 presents the values for parallel, short, and long shocks applicable to interest rates of certain currencies. The shocks allow capturing the heterogeneity of economic contexts in various countries. These must subsequently be used to calculate shocks at different points of the yield curve, in order to create the 6 interest rate shock scenarios in accordance with the methodology described below.

Table 1. Specific Magnitude of Interest Rate Shocks, R̅ Shock Type, Scenario in Basis Points (bps)

Currencies	Parallel	Short	Long
ARS	400	500	300
AUD	300	450	200

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14

BRL 400 500 300 CAD 200 300 150 CHF 100 150 100 CNY 250 300 150 EUR 200 250 100 GBP 250 300 150 HKD 200 250 100 IDR 400 500 350 INR 400 500 300 JPY 100 KRW 300 400 200 MXN 400 500 300 RUB 400 500 300 SAR 200 300 150 SEK 200 300 150 SGD 150 200 100 TRY 400 500 300 USD 200 300 150 ZAR 400 500 300

Chapter 2 - Calibration of shock magnitude for other currencies

With regard to the calibration of the magnitude of shocks on interest rates for currencies other than those specified in Table 1, entities must apply the following process:

Step 1: Calculation of the average daily interest rate Entities must collect a 16-year time series of daily "risk-free" interest rates for each currency c, at maturity terms of 3 months, 6 months, 1 year, 2 years, 5 years, 7 years, 10 years, 15 years, and 20 years. They must then calculate a global average interest rate for each currency c across all observations in the time series and for all maturity terms. The result is a single measure per currency.

Step 2: Application of global shock parameters Entities must apply, for each currency c, the global shock parameters to the average interest rate, as indicated in Table 2.

Table 2. Base parameters for global interest rate shocks Parallel: 60% Short: 85% Long: 40%

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14

The application of the global shock parameters from Table 2 to the average interest rates calculated in Step 1 results in a revised interest rate shock, per currency and for the different segments of the yield curve, i.e., for parallel, short-term, and long-term shocks.

Step 3: Application of maximum and minimum limits The proposed calibration for calculating the magnitude of interest rate shocks may result in shocks that are too small for some currencies, and too large for others. In order to ensure a minimum level of prudence and harmonization, a minimum limit of 100 bps and maximum limits (expressed as Δ̅j (tk)) of 500 bps for the short-term shock, 400 bps for the parallel shock, and 300 bps for the long-term shock have been established.

The change in the "risk-free" interest rate in scenario j and currency c, at the midpoint of the time band tk, can be defined as follows:

|Δ ̅j (tk)| = max{100, {|Δ ̅, (t )|,Δ ̅ }}

where Δ̅j = {400, 500, 300}, for j = {parallel, short, long}, respectively. By applying the maximum and minimum limits to the shocks calculated in Step 2, rounding to the nearest 50 bps, the final set of interest rate shocks per currency is obtained, as indicated in Table 1.

Step 4: Adjustments for other currencies not indicated in Table 1 Since certain jurisdictions may have undergone significant economic changes in the period between 2000 and 2015, the calculations performed in Steps 1 to 4 may not be the most appropriate for these. This is the case, in particular, if interest rates during the first years of the period differ considerably from interest rates in the more recent years of the considered period.

For currencies not referred to in Table 1, the time series to be used to calculate the average interest rate according to Step 1 are determined based on the following principle: if the average interest rate calculated according to Step 1, in the period from 2000 to 2006, is greater than 700 bps, data from the last 10 years (i.e., from 2006 to 2015) are used; otherwise, the complete time series with data from 2000 to 2015 must be used.

The use of this principle allows identifying contexts with high interest rates and periods of significant structural changes before the financial crisis. Furthermore, this principle allows detecting currencies that exceed the maximum limit (700 bps x 0.6 = 420 bps > 400 bps) in the first years of the considered period and encourages greater consideration of more recent interest rates in the considered historical period.

Table 3 shows the results of applying Steps 1 to 4 to the currencies of Member States of the European Union that are not covered by Table 1. The magnitudes of shocks on interest rates for other currencies can be obtained in a similar manner through the application of the method described in this section.

Table 3. Specific magnitude of shocks on interest rates, R̅type of shock, scenario for other currencies of EU Member States in basis points (bps)

Currencies	Parallel	Short	Long
BGN	250	350	150
CZK	200	250	100
DKK	200	250	150
HRK	250	400	200
HUF	300	450	200
PLN	250	350	150
RON	350	500	250

Chapter 3 - Parametrization of the magnitude of shocks on interest rates

Taking into account, for currency c, the specific magnitude of instantaneous parallel, short-term, and long-term shocks on "risk-free" interest rates, the following parametrizations of the six interest rate shock scenarios must be applied:

(i) Parallel shock for currency c: a constant parallel rise or fall across all time horizons:

Δ , (tk) = ± ̅ ,

(ii) Short-term interest rate shock for currency c: greater rise or fall at the midpoint of the shortest maturity. This shock, obtained through the formula Sshort term ( ) = − , where = 4, decreases to zero in the long term of the yield curve1 and is the midpoint of period k:

Δ , (tk) = ± ̅ , ∙ Sshort term ( ) = ± ̅ , ∙ −

(iii) Long-term interest rate shock for currency c: this shock only applies to rotation shocks. The shock is largest at the midpoint of the longest considered maturity of the yield curve, and is related to the decay rate factor of the short-term shock, where Slong term ( ) = 1 - Sshort term ( ):

Δ , (tk) = ± ̅ , ∙ Slong term ( ) = ± ̅ , ∙ (1 − − )

1 The value of in the denominator of the function − controls the decay rate of the shock.

Annex to Instruction No. 3/2020 BO No. 1/2020 2nd Supplement • 2020/02/14 Topics Supervision :: Prudential Standards .................................................................................................................................................................................................. Mod. 99999911/T – 01/14

(iv) Rotation shock for currency c: involves the application of rotations in the yield curve (i.e., changes in the slope of the yield curve), where a shock is applied to long-term and short-term interest rates, by applying the following formulas:

Δ , (tk) = - 0.65 ∙ |Δ , ( ) | + 0.9 ∙ |Δ , ( ) | Δ çã , (tk) = + 0.8 ∙ |Δ , ( ) | - 0.6 ∙ |Δ , ( ) |