Response to consultation Paper on draft Guidelines on loan origination and monitoring.
Go back
Several sub-sections of Section 4 are covered in previous questions. Our comments here therefore address the proposals under sub-sections 4.4 (“Credit decision making”) and 4.7 (“Remuneration”). We do not have material comments on other sub-sections.
Recommendation 1: Provide additional guidance on the definition of “small and non-complex credit facilities”.
The credit decision making sub-section does not reflect existing practices because it limits the “sole delegated credit authority for credit decisions” to “small and non-complex credit facilities”. While the draft guidelines offer neither a definition of nor a methodology to determine “small and non-complex”, we would deem that credit decision on retail and SMEs qualifies for the sole delegated authority exemption and would welcome additional guidance on this item.
Recommendation 2: Delete paragraph 82 of sub-section 4.7 (“Remuneration”).
The guidance on remuneration policies under sub-section 4.7 is not practical, especially the granular requirements listed under paragraph 82. It is unclear how bank would be expected to determine the adequate period that would link the variable remuneration of the staff involved in credit granting to “long term quality of credit exposures”. E.g. would a credit officer responsible for a mortgage portfolio be excepted to have a variable remuneration tied to the average duration of the portfolio, which could be above 15 years. The scope of the requirements in terms of staff is also unclear, e.g. at what level of seniority would the guidance be expected to apply. We recommend deleting paragraph 82 on the basis that the underlying objectives of ensuring the independence in credit decision-making and setting remuneration policies in line with credit risk appetite and strategies are effectively addressed by sub-section 4.4.1 and paragraph 81 of sub-section 4.7.
Recommendation 1: Differentiate the requirements along the revised categories in Question 1 (i.e. retail & SME; corporate) and reserve the most sophisticated requirements to material lending to large corporate.
The proposed “consumer” and “professional” categories are not adequate as explained through our responses. “Professional” would include all non-consumers as per the guideline definition, which will presumably cover a wide range of counterparties from partnership and SMEs to large international corporations. It is hard to see how the granular and prescriptive requirements on information collection and verification of paragraphs 93 and 94 are expected to apply to such a wide range of counterparties. For instance, it is unreasonable to expect the collation of information for SME to be at the same level as for a large corporate. SME generally do not provide business plans and financial projections as required under paragraph 93 e. and f.
Recommendation 2:
Allow flexibility and a simplified approach for lending to existing-to-bank customers vs. new-to-bank customers for retail and SME lending;
Allow the use of gross income (instead of only disposable income) for credit assessment of retail and SME counterparties;
Expand evidence to income (Annex 2) to include income surrogates.
In general, the guidelines assume the same level of availability, accuracy and reliability of information for all consumer segments in mature and in developing markets. For instance, the assessment of disposable income (paragraph 98) may not be practical and gross income should be allowed to be used. Pay slip and other income documentation may not reflect deductions for tax and other financial commitments as required in Annex 2. Income for self-employed consumers would normally be based on bank’s statements. A reliable credit bureau may not be available. Additionally, under Annex 2, evidence of income should be expanded to include income surrogates (e.g. net worth or Assets under Management) based assessment and derived income-based assessment where these approaches have proven to be predictive and reliable. Our concerns with this section would be mostly addressed by following our recommendation made under Question 1 to exclude non-EU originated loans and advances from the scope of the guidelines, as compliance with the requirements will be a challenge outside of the EU due to the specific nature of each retail and SME markets.
Recommendation 3: Remove the requirements to build a comprehensive view of all the borrower’s credit commitments (single customer view) for unsecured lending.
The general requirements suggest (paragraph 85) that banks should build a comprehensive view of the borrower’s financial position, including a view of all the its financial commitments. In the case of unsecured retail lending, this is often not necessary: expecting banks to build up such view would be costly without bringing much benefit to the loan origination process. Partial credit exposure measures, such as Debt-to-Income ratio (“DTI”) based on unsecured debt, can be sourced from information readily available in credit bureau markets. This makes DTI easily accessible and reliable. DTI has also proven to be predictive of the borrower’s ability to meet its financial obligations. All-encompassing measures, such as Debt Servicing Ratio (“DSR”), are more difficult to source and more prone to error due to assumptions required to calculate the counterparty total monthly obligations.
Recommendation 4: Acknowledge that collateralised and guaranteed lending will reduce the creditworthiness assessment requirements.
The collection and documentation requirements should account for the impact of collateralised and guaranteed on lending to corporate. In the case of heavily collateralised lending for instance, which is common in certain structure transactions, comprehensive information on the borrower would not be a primary concerned. To some extent as well, the ownership structure and the place of the borrower within that ownership structure would also dictate the necessary amount of information to be collected and documented on the corporate borrower.
Recommendation 1:
Differentiate the requirements in line with the revised categories (i.e. retail; SME; corporate) and reserve the most sophisticated requirements to material lending to large corporate;
Allow flexibility and a simplified approach for lending to existing-to-bank customers vs. new-to-bank customers for retail and SME lending;
Allow the use of gross income (instead of only disposable income) for credit assessment of retail and SME counterparties;
Allow the use of automatic credit decision systems in addition to “credit decision-making body” for retail counterparties.
Concerns on section 5.2 mirror those highlighted under question 7 on section 5.1 but from a creditworthiness assessment point of view. The “consumers” and “professional” categories and the prescriptive nature of the requirements under each fail to deliver sufficient proportionality and flexibility to make the guidelines fir for purposes in many cases. In addition to issues highlighted under question 7 on the data collection and documentation, we believe that the selected issues below clearly demonstrate the unsuitability and disproportionate nature of the proposed requirements for credit assessment of retail and SMEs borrowers. Requirements to:
i. Perform a full-fledged traditional credit analysis for retail borrowers (paragraph 97 to 99) and no flexibility for simplified approaches (e.g. usage of behavior scorecard and credit bureau) for existing-to-bank retail customers;
ii. Perform a sensitivity analysis for each retail borrower, including variable such as income, interest rate, deferred payments and exchange rates (paragraphs 101, 110, 114 and 121);
iii. Perform a full-fledged traditional credit analysis for SME borrowers, including review of the current and projected financial position under possible adverse scenarios; assessment of the political, economic and legal environment in which the foreign counterparty of the institution’s client operates; assessment of the borrower’s risk profile vis-à-vis climate related risks (paragraph 126 to 130);
iv. Consider full financial projections, dividend distribution, projected capital and all the metrics listed under Annex 3 (paragraphs 131 to 151) for SME borrowers.
The materiality of individual exposures on retail and SMEs typically does not justify the above requirements and banks generally manage and monitor those exposures (excluding non-performing loans) at aggregate/ portfolio levels, including stress testing and sensitivity analysis. There is also no recognition under sub-section 5.3 of automatic credit decision systems which are commonly used in retail lending: paragraphs 182 and 183 assumes that credit decisions are taken by “the relevant credit decision-making body”. The final guidelines should allow for the use of automatic credit decision systems and more generally ensure the requirements retail and SME lending are practical and proportionate.
Recommendation 2: Remove the requirements to assess the borrower’s creditworthiness based on all its financial commitments (single customer view) for retail and SME counterparties.
We would also reiterate our comments on Question 7 on the use of partial credit measures for unsecured lending.
Recommendation 3: Acknowledge that collateralised and guaranteed lending will reduce the creditworthiness assessment requirements.
The draft guidelines do not recognise the specificities of collateralised or guaranteed lending. When banks are satisfied with the amount and nature of collaterals or the guarantee, the assessment of the borrower’s creditworthiness may not be a primary focus of the credit assessment. Also, in line with our response to Question 7, the ownership structure and the place of the borrower within that ownership structure may significantly influence the conduct of the creditworthiness assessment for corporate borrowers.
Recommendation 4: Remove duplication under section 5.2.7.
We believe that sub-section 5.2.7 on shipping contains the same requirements to consider factor such as supply and demand in the market twice under paragraph 171 c and also under paragraph 172. We would suggest deleting point c under paragraph 171 and keep paragraph 172 and replacing “future trade pattern” by “expected trade pattern”.
Recommendation 1: Exclude wealth lending, private banking lending and other fully collateralised lending for retail and SME lending from the loan origination procedures requirements.
Please refer to our response to Question 1 for details.
Recommendation 1:
Adopt a more principle-based guidance that would allow banks to explain their pricing and factors of influences;
Recalibrate the requirements away from “transaction” level toward client and portfolio levels;
Remove the granular guidance on costs considerations.
The guidelines put emphasis on cost considerations when setting loan pricing. While the cost of capital and cost of funding are key elements of pricing, other non-cost factors often have a significant influence in setting the price. This includes for instance:
i. prevailing market conditions at the time of the transaction, e.g. prevailing risk appetite and positive or adverse sentiment on a specific industry or segment; the demand-supply balance between central bank liquidity and available pool of bank investable loan assets;
ii. pricing mechanism and competition (banks bidding for a specific deal);
iii. whether the transaction is highly commoditized (e.g. retail products), vanilla (e.g. cash loan to corporate) or whether banks provide significant value-add service (e.g. advice on optimal capital structure, maiden market transaction, tailor made funding solutions); and
iv. the overall profitability of the relationship with the counterparty.
While the guidelines briefly touch on the profitability of the relationship under paragraph 188, the other factors are absent from the section. We believe this is a gap that should be addressed in the final text. At the large or syndicated loan level, the decision to provide lending facilities is assessed by banks as part of the overall relationship, and loans may be provided to establish or maintain a broader banking relationship. Banks have various tools to track the reality against these relationship expectations and should be able to justify why lending/pricing decisions were taken. For businesses such as retail, banks would typically monitor the profitability at the product portfolio level instead of the transaction or counterparty level.
The pricing will also be significantly influenced by guarantees, collaterals and more generally by the corporate ownership structure for corporate lending. These elements should be included in the guidelines as key considerations under the pricing section. This is consistent with our comments under Questions 7 and 8, where we have stressed the impact of guarantees, collaterals and of the ownership structure on the assessment of the borrowers’ creditworthiness and on data collection and documentation and are integral with the Bank’s risk management.
Recommendation 2: Add RoRWA and RoTA as risk-adjusted performance measures.
We would expect Banks to be able to explain their risk-adjusted performance measures rather than having to use one uniform measure and suggest that the risk-adjusted performance measures listed under paragraph 188 should add return on risk-weighted assets (“RoRWA”) and return on total assets (“RoTA”), which are commonly used by banks to assets business lines profitability. Many banks do not take administrative cost and fully-loaded Economic-Value-Added (“EVA”) or RoTE (“Return on Tangible Equity) to the individual loan or client level, for example in our case, the transmission mechanism used to drive this is a target RoRWA figure which is driven by the RoTE and based on a forward-looking corporate plan.
Recommendation 3: Remove granular guidance around monitoring and review of transaction below costs.
The guidance provided on monitoring under paragraph 190 may not be proportionate depending on the granularity of the loan portfolio in question. Banks tend to review low-returning (below cost of capital/funding) on either a relationship basis or portfolio basis. Only for the largest loan sizes would transaction-level monitoring lead to the right level of management focus on pricing. Profitability or margin analysis are typically conducted at the portfolio and/ or business line levels, especially for high volume transaction businesses, such as retail lending or SMEs lending. The proposed guidance also raises concerns around consistent application, as banks and supervisors may have different interpretations of “below cost”.
Recommendation 4: Clarify the definition of “pricing”.
The guidelines do not clearly define “pricing”, which could therefore be interpreted as all-in pricing (including fees) or interest income only. This should be clarified in the final guidelines.
Recommendation 1:
Allow exemption for rotation requirements for valuers of immovable property collateral for smaller portfolios and where it is justified;
Be explicit that valuation of movable property collateral (section 7.2.2) is not mandatory;
Revise paragraph 225 on conflict of interest to specific areas within the control of the institutions.
Some of the requirements for valuers under sub-sections 7.2 and 7.3 will be onerous or not possible to comply with for practical reasons. Importantly, they could go against reasonable and prudent practices. For instance:
i. The need to ensure adequate rotation of valuers on immovable property collateral will not be practical in some markets. This may be due to the small size of the portfolio or the limited availability of reputable valuation offices in developing market. It would be less prudent to force banks to rotate as expected in paragraph 214 and use a less skilled/ qualified valuers;
ii. The valuation requirement for movable property collateral (sub-section 7.2.2) would depend on the type of collateral and this should not be mandatory. For example, this should not apply to vehicle loans;
iii. The requirement for institutions to mitigate any conflict of interest for the valuers (paragraph 225) is onerous. The assignment of the valuer is decided by the valuation agency and the bank would not know if the valuer assigned was related to the buyer or seller of the property. We would
propose that this requirement be more specific to what is within the bank’s control. For immovable property collateral, the market practice is for a sample check or the use of 2 valuations for the same property.
Recommendation 1: Make the requirements under sub-sections 8.1, 8.2 and 8.4 more compatible with retail and SME lending.
The general requirements under sub-sections 8.1 (“General requirements for credit risk monitoring framework”) and 8.2 (“Monitoring of credit exposure and borrowers”) and 8.4 (“Monitoring of covenants”) are not all suitable for retail and SME lending. Paragraph 231 mandates the monitoring of group of connected clients. Paragraph 234 requires the monitoring and data infrastructure to support a “single customer view”. While those requirements can reasonably be expected for material credit exposure to corporate given that EU banks must comply with the large exposure framework, it should not apply for retail and SME counterparties. For those counterparties (excluding non-performing exposures), we monitor retail and SME exposure at the portfolio level (e.g. credit cards, personal loans, mortgage) for portfolio management actions or at the client/ product level for wealth management lending. Other requirements, such as the monitoring of qualitative factors (paragraph 238), the continuous monitoring of the borrower’s financial situation and repayment (paragraph 240 and 241) or the monitoring of collateral insurance (paragraph 252) are also more applicable to large corporate. Repayment performance, for instance, feeds into the behaviour scorecard of the borrower as part of the data collection strategy for retail and SME. The ongoing monitoring of collateral insurance impractical, as retail customer will usually not comply with bank’s request once the mortgage loan is granted.
Recommendation 2: Differentiate the requirements in line with the revised categories in Question 1 (i.e. retail & SME; corporate) and reserve the most sophisticated requirements to material lending to corporate
Consistent with the approach adopted through the other section, the monitoring section put forward two different sets of requirements based on the “consumer” and “professional” categories. As mentioned, those categories are not adequate and fail to deliver the necessary proportionality. Sub-section 8.3 (“Credit review of professionals”)’s requirements will not be in line with industry practices for SMEs exposures managed under retail segment, as banks commonly rely on behavior scores. Annual credit review of SME borrowers is not needed with regular repayments over a fixed term (e.g. short-term business installment loans, government guaranteed installment loans, loan against property). The requirements under sub-section 8.3 should apply to “corporate” counterparty only.
Recommendation 3: Reserve the “single customer view” requirements for the “corporate” category only.
The requirements to monitor the single customer view for “consumer” is not feasible and will not bring material benefit to the credit monitoring process. It is not market practice to have a single customer view for retail counterparties. There are system and infrastructure constraints as no single system have the capacity of processing and monitoring customer exposures across the various retail products effectively. Refer to our comments and recommendation under Question 7 for details.
5. What are the respondents’ views on the requirements for governance for credit granting and monitoring (Section 4)?
Most of the proposed guidance is adequate if implemented in a proportionate manner by supervisory authorities. The specific proposal on remuneration is not practical.Several sub-sections of Section 4 are covered in previous questions. Our comments here therefore address the proposals under sub-sections 4.4 (“Credit decision making”) and 4.7 (“Remuneration”). We do not have material comments on other sub-sections.
Recommendation 1: Provide additional guidance on the definition of “small and non-complex credit facilities”.
The credit decision making sub-section does not reflect existing practices because it limits the “sole delegated credit authority for credit decisions” to “small and non-complex credit facilities”. While the draft guidelines offer neither a definition of nor a methodology to determine “small and non-complex”, we would deem that credit decision on retail and SMEs qualifies for the sole delegated authority exemption and would welcome additional guidance on this item.
Recommendation 2: Delete paragraph 82 of sub-section 4.7 (“Remuneration”).
The guidance on remuneration policies under sub-section 4.7 is not practical, especially the granular requirements listed under paragraph 82. It is unclear how bank would be expected to determine the adequate period that would link the variable remuneration of the staff involved in credit granting to “long term quality of credit exposures”. E.g. would a credit officer responsible for a mortgage portfolio be excepted to have a variable remuneration tied to the average duration of the portfolio, which could be above 15 years. The scope of the requirements in terms of staff is also unclear, e.g. at what level of seniority would the guidance be expected to apply. We recommend deleting paragraph 82 on the basis that the underlying objectives of ensuring the independence in credit decision-making and setting remuneration policies in line with credit risk appetite and strategies are effectively addressed by sub-section 4.4.1 and paragraph 81 of sub-section 4.7.
6. What are the respondent’s views on how the guidelines capture the role of the risk management function in credit granting process?
No comments.7. What are the respondents’ views on the requirements for collection of information and documentation for the purposes of creditworthiness assessment (Section 5.1)?
The requirements for collection of information and documentation are too prescriptive and granular to fit with all business models across different markets. The impact of guarantees, collaterals and ownership structure on the requirements should be considered for lending to corporate.Recommendation 1: Differentiate the requirements along the revised categories in Question 1 (i.e. retail & SME; corporate) and reserve the most sophisticated requirements to material lending to large corporate.
The proposed “consumer” and “professional” categories are not adequate as explained through our responses. “Professional” would include all non-consumers as per the guideline definition, which will presumably cover a wide range of counterparties from partnership and SMEs to large international corporations. It is hard to see how the granular and prescriptive requirements on information collection and verification of paragraphs 93 and 94 are expected to apply to such a wide range of counterparties. For instance, it is unreasonable to expect the collation of information for SME to be at the same level as for a large corporate. SME generally do not provide business plans and financial projections as required under paragraph 93 e. and f.
Recommendation 2:
Allow flexibility and a simplified approach for lending to existing-to-bank customers vs. new-to-bank customers for retail and SME lending;
Allow the use of gross income (instead of only disposable income) for credit assessment of retail and SME counterparties;
Expand evidence to income (Annex 2) to include income surrogates.
In general, the guidelines assume the same level of availability, accuracy and reliability of information for all consumer segments in mature and in developing markets. For instance, the assessment of disposable income (paragraph 98) may not be practical and gross income should be allowed to be used. Pay slip and other income documentation may not reflect deductions for tax and other financial commitments as required in Annex 2. Income for self-employed consumers would normally be based on bank’s statements. A reliable credit bureau may not be available. Additionally, under Annex 2, evidence of income should be expanded to include income surrogates (e.g. net worth or Assets under Management) based assessment and derived income-based assessment where these approaches have proven to be predictive and reliable. Our concerns with this section would be mostly addressed by following our recommendation made under Question 1 to exclude non-EU originated loans and advances from the scope of the guidelines, as compliance with the requirements will be a challenge outside of the EU due to the specific nature of each retail and SME markets.
Recommendation 3: Remove the requirements to build a comprehensive view of all the borrower’s credit commitments (single customer view) for unsecured lending.
The general requirements suggest (paragraph 85) that banks should build a comprehensive view of the borrower’s financial position, including a view of all the its financial commitments. In the case of unsecured retail lending, this is often not necessary: expecting banks to build up such view would be costly without bringing much benefit to the loan origination process. Partial credit exposure measures, such as Debt-to-Income ratio (“DTI”) based on unsecured debt, can be sourced from information readily available in credit bureau markets. This makes DTI easily accessible and reliable. DTI has also proven to be predictive of the borrower’s ability to meet its financial obligations. All-encompassing measures, such as Debt Servicing Ratio (“DSR”), are more difficult to source and more prone to error due to assumptions required to calculate the counterparty total monthly obligations.
Recommendation 4: Acknowledge that collateralised and guaranteed lending will reduce the creditworthiness assessment requirements.
The collection and documentation requirements should account for the impact of collateralised and guaranteed on lending to corporate. In the case of heavily collateralised lending for instance, which is common in certain structure transactions, comprehensive information on the borrower would not be a primary concerned. To some extent as well, the ownership structure and the place of the borrower within that ownership structure would also dictate the necessary amount of information to be collected and documented on the corporate borrower.
8. What are the respondents’ views on the requirements for assessment of borrower’s creditworthiness (Section 5.2)?
The requirements for assessment of borrower’s creditworthiness are too prescriptive and granular to fit with all business models across different markets. The impact of guarantees, collaterals and ownership structure on the requirements should be considered for lending to corporate.Recommendation 1:
Differentiate the requirements in line with the revised categories (i.e. retail; SME; corporate) and reserve the most sophisticated requirements to material lending to large corporate;
Allow flexibility and a simplified approach for lending to existing-to-bank customers vs. new-to-bank customers for retail and SME lending;
Allow the use of gross income (instead of only disposable income) for credit assessment of retail and SME counterparties;
Allow the use of automatic credit decision systems in addition to “credit decision-making body” for retail counterparties.
Concerns on section 5.2 mirror those highlighted under question 7 on section 5.1 but from a creditworthiness assessment point of view. The “consumers” and “professional” categories and the prescriptive nature of the requirements under each fail to deliver sufficient proportionality and flexibility to make the guidelines fir for purposes in many cases. In addition to issues highlighted under question 7 on the data collection and documentation, we believe that the selected issues below clearly demonstrate the unsuitability and disproportionate nature of the proposed requirements for credit assessment of retail and SMEs borrowers. Requirements to:
i. Perform a full-fledged traditional credit analysis for retail borrowers (paragraph 97 to 99) and no flexibility for simplified approaches (e.g. usage of behavior scorecard and credit bureau) for existing-to-bank retail customers;
ii. Perform a sensitivity analysis for each retail borrower, including variable such as income, interest rate, deferred payments and exchange rates (paragraphs 101, 110, 114 and 121);
iii. Perform a full-fledged traditional credit analysis for SME borrowers, including review of the current and projected financial position under possible adverse scenarios; assessment of the political, economic and legal environment in which the foreign counterparty of the institution’s client operates; assessment of the borrower’s risk profile vis-à-vis climate related risks (paragraph 126 to 130);
iv. Consider full financial projections, dividend distribution, projected capital and all the metrics listed under Annex 3 (paragraphs 131 to 151) for SME borrowers.
The materiality of individual exposures on retail and SMEs typically does not justify the above requirements and banks generally manage and monitor those exposures (excluding non-performing loans) at aggregate/ portfolio levels, including stress testing and sensitivity analysis. There is also no recognition under sub-section 5.3 of automatic credit decision systems which are commonly used in retail lending: paragraphs 182 and 183 assumes that credit decisions are taken by “the relevant credit decision-making body”. The final guidelines should allow for the use of automatic credit decision systems and more generally ensure the requirements retail and SME lending are practical and proportionate.
Recommendation 2: Remove the requirements to assess the borrower’s creditworthiness based on all its financial commitments (single customer view) for retail and SME counterparties.
We would also reiterate our comments on Question 7 on the use of partial credit measures for unsecured lending.
Recommendation 3: Acknowledge that collateralised and guaranteed lending will reduce the creditworthiness assessment requirements.
The draft guidelines do not recognise the specificities of collateralised or guaranteed lending. When banks are satisfied with the amount and nature of collaterals or the guarantee, the assessment of the borrower’s creditworthiness may not be a primary focus of the credit assessment. Also, in line with our response to Question 7, the ownership structure and the place of the borrower within that ownership structure may significantly influence the conduct of the creditworthiness assessment for corporate borrowers.
Recommendation 4: Remove duplication under section 5.2.7.
We believe that sub-section 5.2.7 on shipping contains the same requirements to consider factor such as supply and demand in the market twice under paragraph 171 c and also under paragraph 172. We would suggest deleting point c under paragraph 171 and keep paragraph 172 and replacing “future trade pattern” by “expected trade pattern”.
9. What are the respondents’ views on the scope of the asset classes and products covered in loan origination procedures (Section 5)?
The loan origination procedures are not suitable for wealth and private banking lending.Recommendation 1: Exclude wealth lending, private banking lending and other fully collateralised lending for retail and SME lending from the loan origination procedures requirements.
Please refer to our response to Question 1 for details.
10. What are the respondents’ views on the requirements for loan pricing (Section 6)?
The proposed requirements would not be compatible with loan pricing and monitoring practices. The one-size-fits-all approach does not consider key difference between syndicated loans to large corporate and bilateral lending to smaller corporate, SME or retail clients.Recommendation 1:
Adopt a more principle-based guidance that would allow banks to explain their pricing and factors of influences;
Recalibrate the requirements away from “transaction” level toward client and portfolio levels;
Remove the granular guidance on costs considerations.
The guidelines put emphasis on cost considerations when setting loan pricing. While the cost of capital and cost of funding are key elements of pricing, other non-cost factors often have a significant influence in setting the price. This includes for instance:
i. prevailing market conditions at the time of the transaction, e.g. prevailing risk appetite and positive or adverse sentiment on a specific industry or segment; the demand-supply balance between central bank liquidity and available pool of bank investable loan assets;
ii. pricing mechanism and competition (banks bidding for a specific deal);
iii. whether the transaction is highly commoditized (e.g. retail products), vanilla (e.g. cash loan to corporate) or whether banks provide significant value-add service (e.g. advice on optimal capital structure, maiden market transaction, tailor made funding solutions); and
iv. the overall profitability of the relationship with the counterparty.
While the guidelines briefly touch on the profitability of the relationship under paragraph 188, the other factors are absent from the section. We believe this is a gap that should be addressed in the final text. At the large or syndicated loan level, the decision to provide lending facilities is assessed by banks as part of the overall relationship, and loans may be provided to establish or maintain a broader banking relationship. Banks have various tools to track the reality against these relationship expectations and should be able to justify why lending/pricing decisions were taken. For businesses such as retail, banks would typically monitor the profitability at the product portfolio level instead of the transaction or counterparty level.
The pricing will also be significantly influenced by guarantees, collaterals and more generally by the corporate ownership structure for corporate lending. These elements should be included in the guidelines as key considerations under the pricing section. This is consistent with our comments under Questions 7 and 8, where we have stressed the impact of guarantees, collaterals and of the ownership structure on the assessment of the borrowers’ creditworthiness and on data collection and documentation and are integral with the Bank’s risk management.
Recommendation 2: Add RoRWA and RoTA as risk-adjusted performance measures.
We would expect Banks to be able to explain their risk-adjusted performance measures rather than having to use one uniform measure and suggest that the risk-adjusted performance measures listed under paragraph 188 should add return on risk-weighted assets (“RoRWA”) and return on total assets (“RoTA”), which are commonly used by banks to assets business lines profitability. Many banks do not take administrative cost and fully-loaded Economic-Value-Added (“EVA”) or RoTE (“Return on Tangible Equity) to the individual loan or client level, for example in our case, the transmission mechanism used to drive this is a target RoRWA figure which is driven by the RoTE and based on a forward-looking corporate plan.
Recommendation 3: Remove granular guidance around monitoring and review of transaction below costs.
The guidance provided on monitoring under paragraph 190 may not be proportionate depending on the granularity of the loan portfolio in question. Banks tend to review low-returning (below cost of capital/funding) on either a relationship basis or portfolio basis. Only for the largest loan sizes would transaction-level monitoring lead to the right level of management focus on pricing. Profitability or margin analysis are typically conducted at the portfolio and/ or business line levels, especially for high volume transaction businesses, such as retail lending or SMEs lending. The proposed guidance also raises concerns around consistent application, as banks and supervisors may have different interpretations of “below cost”.
Recommendation 4: Clarify the definition of “pricing”.
The guidelines do not clearly define “pricing”, which could therefore be interpreted as all-in pricing (including fees) or interest income only. This should be clarified in the final guidelines.
11. What are the respondents’ views on the requirements for valuation of immovable and movable property collateral (Section 7)?
Some of the requirements for the valuation of immovable and movable property collateral would not be fit for all business models across different markets.Recommendation 1:
Allow exemption for rotation requirements for valuers of immovable property collateral for smaller portfolios and where it is justified;
Be explicit that valuation of movable property collateral (section 7.2.2) is not mandatory;
Revise paragraph 225 on conflict of interest to specific areas within the control of the institutions.
Some of the requirements for valuers under sub-sections 7.2 and 7.3 will be onerous or not possible to comply with for practical reasons. Importantly, they could go against reasonable and prudent practices. For instance:
i. The need to ensure adequate rotation of valuers on immovable property collateral will not be practical in some markets. This may be due to the small size of the portfolio or the limited availability of reputable valuation offices in developing market. It would be less prudent to force banks to rotate as expected in paragraph 214 and use a less skilled/ qualified valuers;
ii. The valuation requirement for movable property collateral (sub-section 7.2.2) would depend on the type of collateral and this should not be mandatory. For example, this should not apply to vehicle loans;
iii. The requirement for institutions to mitigate any conflict of interest for the valuers (paragraph 225) is onerous. The assignment of the valuer is decided by the valuation agency and the bank would not know if the valuer assigned was related to the buyer or seller of the property. We would
propose that this requirement be more specific to what is within the bank’s control. For immovable property collateral, the market practice is for a sample check or the use of 2 valuations for the same property.
12. What are the respondents’ views on the proposed requirements on monitoring framework (Section 8)?
Some of the proposed requirements on monitoring framework is not suitable for retail and SME lending.Recommendation 1: Make the requirements under sub-sections 8.1, 8.2 and 8.4 more compatible with retail and SME lending.
The general requirements under sub-sections 8.1 (“General requirements for credit risk monitoring framework”) and 8.2 (“Monitoring of credit exposure and borrowers”) and 8.4 (“Monitoring of covenants”) are not all suitable for retail and SME lending. Paragraph 231 mandates the monitoring of group of connected clients. Paragraph 234 requires the monitoring and data infrastructure to support a “single customer view”. While those requirements can reasonably be expected for material credit exposure to corporate given that EU banks must comply with the large exposure framework, it should not apply for retail and SME counterparties. For those counterparties (excluding non-performing exposures), we monitor retail and SME exposure at the portfolio level (e.g. credit cards, personal loans, mortgage) for portfolio management actions or at the client/ product level for wealth management lending. Other requirements, such as the monitoring of qualitative factors (paragraph 238), the continuous monitoring of the borrower’s financial situation and repayment (paragraph 240 and 241) or the monitoring of collateral insurance (paragraph 252) are also more applicable to large corporate. Repayment performance, for instance, feeds into the behaviour scorecard of the borrower as part of the data collection strategy for retail and SME. The ongoing monitoring of collateral insurance impractical, as retail customer will usually not comply with bank’s request once the mortgage loan is granted.
Recommendation 2: Differentiate the requirements in line with the revised categories in Question 1 (i.e. retail & SME; corporate) and reserve the most sophisticated requirements to material lending to corporate
Consistent with the approach adopted through the other section, the monitoring section put forward two different sets of requirements based on the “consumer” and “professional” categories. As mentioned, those categories are not adequate and fail to deliver the necessary proportionality. Sub-section 8.3 (“Credit review of professionals”)’s requirements will not be in line with industry practices for SMEs exposures managed under retail segment, as banks commonly rely on behavior scores. Annual credit review of SME borrowers is not needed with regular repayments over a fixed term (e.g. short-term business installment loans, government guaranteed installment loans, loan against property). The requirements under sub-section 8.3 should apply to “corporate” counterparty only.
Recommendation 3: Reserve the “single customer view” requirements for the “corporate” category only.
The requirements to monitor the single customer view for “consumer” is not feasible and will not bring material benefit to the credit monitoring process. It is not market practice to have a single customer view for retail counterparties. There are system and infrastructure constraints as no single system have the capacity of processing and monitoring customer exposures across the various retail products effectively. Refer to our comments and recommendation under Question 7 for details.