Letters
29 August 2018
Hon. Yunus Carrim
Chairperson of the Standing Committee on Finance
New Assembly Building
Parliament Street
Cape Town
Dear Hon. Carrim
BANKS AMENDMENT BILL
The Banking Association South Africa (“BASA”) and its members welcome the opportunity to comment on the Banks Amendment Bill (“Bill”). This Bill essentially seeks to enable a state-owned company to register and conduct the business of a bank in
terms of the Banks Act.
BASA and its members support all new banks in South Africa insofar as they are subject to similar regulatory supervision enforced by the South African Reserve Bank as the primary supervisor under the provisions in the Banks Act, Act 94 of 1990 (“Banks Act”);
South African Reserve Bank Act, Act 90 of 1989 (“SARB Act”) and the Financial Sector
Regulation Act, Act 9 of 2017 (“FSR Act”).
We are of the view that this will ensure a level playing field in terms of prudential,
governance and risk standards which the entirety of the financial sector is subject to.
Furthermore, supervision for any new banks should be in line with the type of products
and services such banks offer. This should include the relevant prudential, market
conduct, credit and resolution requirements as this will be the best way to manage any
risks such bank/s may introduce to the financial system.
We would like to bring the matters below to your attention for consideration.
We look forward to engaging you further on issues raised and discussing a way forward.
Yours sincerely
C Coovadia
Managing Director
9 March 2018
Secretariat of the Basel Committee on Banking Supervision
Bank for International Settlements
CH-4002 Basel
Switzerland
baselcommittee@bis.org
Cc: SARB-banksup@resbank.co.za
Re: Discussion paper: Regulatory Treatment of Sovereign Exposures
We thank you for the opportunity to provide comment on the Basel Committee on Banking Supervision (BCBS) discussion paper that looks at the treatment of exposures to Sovereigns.
We believe the discussion paper is a good summary of the sources and channels of sovereign risk in the banking system, the holistic role of sovereign exposures and the existing regulatory treatment of sovereign exposures. As such, any revising of the existing regulatory treatment entails a risk of having unintended consequences and it is recommended that an impact assessment is conducted prior to making any changes.
We note that the Committee recognises that the specific roles of sovereign exposures may vary across jurisdictions due to the heterogeneity in banks’ business models, market structures and macro-financial balances.
In general:
- We are supportive of the differentiation between central banks, central government and other sovereign exposures and encourage the Committee to provide examples of subnational and public sector entities that satisfy the risk equivalence criterion for consideration as central government
- We do support the use of external ratings, but would recommend the introduction of a Standardised Credit Risk Assessment Approach (i.e. in line with the Bank asset class under the revised Standardised Approach contained in “Basel III: Finalising post-crisis reforms”). This would accommodate jurisdictions that do not allow the use of external ratings as well as subnational government and public sector exposures that are not externally rated. In terms of the granularity of the risk weight table, increased granularity could be accommodated when using external ratings. For a Standardised Credit Assessment Approach the granularity will have to be limited to avoid undue complexity.
- We would recommend that Sovereign bonds held for HQLA purposes should be subject to a 0% risk weight and be exempt from concentration risk assessment
- We do support risk weights that differentiate by at least: external rating agency ratings, local vs. foreign currency – where this may not be reflected in separate rating agency ratings, and entity type. We do prefer a more granular set, closer to that included in table 4.
- We acknowledge that the low default nature of sovereign exposures makes it difficult to model risk parameters (i.e. PDs and LGDs), but consider the removal of the Internal Ratings Based Approach in its entirety as unnecessary. The models employed by banks are equivalent to those employed by Rating Agencies in their rank ordering of sovereign entities. The issue lies with the calibration of the risk parameters and as such we wish to recommend the introduction of PD and LGD floors and/or ranges as means of ensuring appropriate capitalisation for the credit risk.
- Asset classification as set out in the document needs to be clarified in a “developing economy scenario” as:
- We are not sure that the distinction between central bank and central Government is clear enough or if even necessary?
- We would not recommend a distinction between central banks that target exchange rate stability vs. where this is not the case.
- We do agree that there is a difference between local and foreign currency assets to central banks.
Please see our response to the specific questions here:
Read The Banking Assocation South Africa’s response to BCBS Sovereign Exposures Discussion Paper
Yours sincerely,
Gary Haylett
GM, Prudential Division
Mr W Mostert
South African Reserve Bank
370 Helen Joseph Street
Pretoria
0002
E-mail: SARB-banksup@resbank.co.za
Wessel.mostert@resbank.co.za
Wessel
Re: Matters Related to the Standards for Interest Rate Risk in the
Banking Book
We thank you for the opportunity to provide comments on the proposed directive (15/8
dd. 25/07/17) issued in terms of section 6(6) of the Banks Act 94 of 1990.
We furthermore thank you for the ongoing dialogue and engagement with the industry
in our regular meetings, around the implementation of the Basel Committee on Banking
Supervision, Standards for Interest Rate Risk in the Banking Book (IRRBB), due
1 January 2018.
While several of the issues raised through the industry meetings have been dealt in the
proposed directive, there are still key items that require further clarification and are
contained in the BASA issues log, attached as Annexure A [#227616], updated post
our meeting on 7 August 17.
Download the full Letter to SARB re Basel Standards on IRRBB here.
25 April 2017
The Managing Director S&P Global South Africa
2 nd Floor, 30 Jellicoe Avenue
Rosebank,
2196
Attention: Mr Konrad Reuss D
Dear Mr Reuss
The Banking Association South Africa is the mandated representative of the banking sector, and following the recent negative sovereign rating action, our members have expressed their concern that the S&P Global national rating scale calibration for South Africa needs your urgent attention.
Applying the S&P Global ratings’ national and regional scale mapping tables for South Africa, published in June 2016, a long-term local-currency rating of BB+ maps to a national-scale long-term rating between zaA+ and zaA. However, with the negative outlook for the sovereign, we would assume a conservative interpretation of zaA.
The S&P Global publication General Criteria: National and Regional Scale Credit Ratings describes national scale ratings as providing “…a rank ordering of credit within the country.” National scale ratings provide further granularity on credit quality, however there appears to be a build-up of national scale ratings around zaA amongst banks and corporates, what S&P Global refer to as a rating compression.
Our concern is that investors may breach their mandates in the short-term and this would introduce a distortion in the market, whereby exposures to our members’ debt instruments may have to be reduced, whilst investors seek to change their mandates.
Although we believe the fundamentals have not changed as a result of the rating action, in the absence of a recalibration of South Africa’s national scale ratings, we are further concerned that the normal functions of the debt capital markets may be unnecessarily impacted.
We believe that these concerns could be ameliorated by S&P Global announcing their intention to recalibrate South Africa’s national rating scales. This action on its own should provide sufficient comfort to those parties forced to rebalance their portfolios. We would then also request S&P Global to recalibrate their national scale rating in respect of South Africa as soon as possible thereafter, and to provide clarity to the market regarding the likely timeframes that would be involved in completing this process.
We look forward to your urgent attention in this regard.
Yours faithfully
Mark Brits
Senior General Manager – Prudential
3 June 2016
Switzerland Secretary General
Basel Committee on Banking Supervision
Bank for International Settlements
Basel
CH-4002
Attention: Mr William Coen
Dear Mr Coen
Comment on the BCBS Consultative Document:
Standardised Measurement Approach for operational risk
The Banking Association of South Africa is grateful for this opportunity to provide feedback on the proposed Standardised Measurement Approach (SMA) published in the Basel Committee on Banking Supervision’s (BCBS/the Committee) consultative document number D355 titled “Standardised Measurement Approach for operational risk”. The work of the Committee to develop a standardised approach and simultaneously withdraw all current approaches including internal models for operational risk regulatory capital from the Basel Framework, is acknowledged.
Our response to BCBS 258 titled “The Regulatory Framework: Balancing Risk Sensitivity, Simplicity and Comparability” we believe is still relevant, and equally applicable to the operational risk discipline and in our commentary below, we shall highlight what we consider the unintended consequences of this proposal to be, and where possible, propose alternative approaches to counter these unintended consequences for your consideration.
Our feedback was prepared to focus on the specific questions posed in the consultative document. Where feedback spans more than one of these questions, for example the application of the SMA, feedback is shown under the “General feedback” section below. The remainder of the feedback is shown under the section “Specific questions feedback”.
General feedback
Withdrawal of internal modelling
(Refer paragraphs 5 and 6)
The observation that there was a “…diverse range of internal modelling practices subject to supervisory approval…” is cited as one of the reasons for the withdrawal of AMA. We would suggest that the supervisory colleges were not adequately tasked within their mandate to pursue enhanced information sharing between supervisors and that private entities such as the Operational Riskdata eXchange Association (ORX) based in Zurich, could have been leveraged better in pursuit of greater harmonisation.
The same approach could be adopted in the comment that “…internal modelling practices have exacerbated variability in risk weighted asset calculations…” or a more generic solution could be to limit the range of practices and reducing the relatively high 99.9th percentile requirement.
Furthermore, the Committee has adopted historical loss experience as an indicator of
future loss exposure for SMA which in our view is comparable to the AMA when not used in conjunction with forward-looking metrics such as scenario analysis.
What is the SMA measuring?
Our interpretation of SMA is that it is a proxy for a risk measure of the relative changes in the perceived operational risk profile of a bank. The absolute level of this “risk measure” is calibrated externally to the bank by the Committee using industry-wide data. Therefore, we believe that it is not possible for a bank to independently validate the SMA capital result as an appropriate reflection of the bank’s operational risk capital demand estimate, or perform credible Pillar II capital adequacy assessments for the SMA-derived Pillar I capital result.
We would appreciate further guidance on our interpretation and if correct, what validations can be performed?
Impact on Pillar II requirements
The consultative document is silent on the impact on Pillar II of the proposed SMA. We would welcome further guidance on the potential impact on aspects such as permissible stress testing and internal capital adequacy assessment approaches.
Superadditivity of SMA within a group
(Refer to paragraphs 4, 37, 38 and 39)
The application of the SMA within a banking group would apply concurrently at both the consolidated level as well as the subsidiary level. Due to the progressive increase of the marginal impact of the BI as the BI bucket increases, it follows that the SMA capital at a group level is higher than the sum of the SMA capital of the individual subsidiaries within the group.
This superadditivity of the SMA capital is counter-intuitive since our expectation is that risk measures should be sub-additive, whereby the group SMA is equal or less than the sum of the SMA capital of the subsidiaries. The superadditivity also creates capital management challenges because typically the entire group’s capital is allocated to subsidiaries.
With the proposed application of the SMA, the subsidiaries’ capital demand is potentially inflated above the SMA capital calculated for the subsidiary, on a stand-alone basis, due to this amount being allocated from the group. Therefore, the capital demand from a subsidiary could theoretically increase purely because the SMA is superadditive. To illustrate an unintended consequence, consider two similar sized banks, with similar product mixes and risk profiles which should in terms of the intention with the SMA have similar SMA capital. The one bank is a subsidiary of a larger banking group, and therefore would attract a higher capital demand simply because being allotted more capital from the group. This negatively impacts the competiveness of the bank belonging to the larger group due to having a higher return-on-equity hurdle ate, which is counterintuitive.
A potential solution is to calculate the SMA capital at a subsidiary level and that at the group level SMA be limited to the sum of the subsidiary level SMA capital.
Euro currency conversions
The calibration parameters used in the BI component, loss component and loss data thresholds are specified in Euros. For banks in non-Euro jurisdictions, a global standard to convert non-Euro amounts into Euro would assist with comparability.
We believe certain elements of these calculations should be harmonised and could include the exchange rate and frequency of conversion. In addition, guidance on a methodology to address different financial year-ends and the impact of specific exchange rate dates on the comparability of these results, would be useful when engaging the audit community.
In addition, the choice of currency conversion method should be designed so that any artificial currency conversion effect impacting the SMA capital of non-Euro banks can be addressed. As an example, where the SMA capital of a non-Euro bank changes from one reporting period to the next as a result of an appreciation or depreciation of the home currency, without a change to the local-currency-based risk profile.
Calibration
We understand that the Committee used the various Quantitative Impact Studies (QIS) conducted as part of the Basel III monitoring over the past years, to calibrate the SMA capital. Based on the experience shared by our members, we believe that several aspects of the submitted QIS data contain elements that may have been interpreted differently by different banks within the same jurisdiction and across jurisdictions.
Examples are:
(i) Differing interpretations and accounting standards applied across jurisdictions in respect of the supplied BI financial metrics;
(ii) Differing loss definitions used for supplied loss data due to lack of clarity on the permissible loss recoveries that can be considered in calculating loss amounts;
(iii) Differing approaches to split the trading and banking book P&L, which impacts the BI component, due to the BCBS D352 trading/banking book split only becoming effective on 1 January 2019 – therefore the supplied trading/banking book delineation is at best calculated on a best efforts basis which may be inconsistent across banks; (iv) Differing approaches used to convert non-Euro BI components, loss data and loss ata thresholds to Euro which could result in an artificial currency effect being present in upplied QIS data, and
(v) It is not clear how the QIS results supplied separately to host regulators by the consolidated group, and its subsidiaries are treated in the calibration process to avoid potential double counting of financial and loss data.
Due to the above sources of uncertainty, the SMA calibration performed by the Committee could include spurious inputs which may invalidate calibration results.
Risk sensitive metric
(Refer paragraph 2)
The proposed SMA is designed to be risk sensitive, however in our view it is based exclusively on historical financial metrics and loss data and lacks forward-looking risk sensitive information.
Banks may not have sufficient internal loss data and could depend on scenarios and other exposure ssessments to develop their profile. We believe that assessments of exposure should be captured within the SMA to better reflect the current risk profile and improve risk sensitivity.
This modification could create the right incentives to influence how operational risk is
managed within an organisation.
SMA methodology
(Refer paragraph 12)
The Committee has completed an analysis that has informed the changes to the Revised Standardised Approach (RSA) originally published in BCBS 291 to arrive at the SMA.
We would appreciate further information being disclosed on the SMA methodology as was previously disclosed by the Committee for the RSA. This disclosure would assist to:
(i) Gain a better understanding of these important methodology changes, and
(ii) Allow banks to further explore, enhance and/or customise the methodology to apply for purposes other than regulatory capital, for example Pillar II and capital allocation.
Where possible, guidance on the Committee’s approach to the frequency of future calibrations would also be appreciated.
Limited incentives to improve risk management
The specification of the SMA capital is in our view mainly based on financial statement items, with larger banks permitted to use internal loss experiences to make limited adjustments to capital derived from these financial items. As a result, banks’ internal risk measures, risk mitigation and risk transfer, apart from internal loss data, are therefore not used in capital calculations.
This disconnect between risk management practices and SMA capital requirements, may have the unintended consequences that senior management will place less focus and priority on further investment in operational risk management since the link between risk management inclusive of internal risk measures and the impact on capital requirements are weaker with the SMA than under the AMA.
Principles for the sound management of operational risk (PSMOR)
(Refer paragraph 42, BCBS 195)
We believe that the Committee is currently reviewing the PSMOR and would propose that this review be prioritised so that the publication of the updated PSMOR coincides with (or predates) the publication of the final SMA specification. This should provide banks with a holistic view of the standard at which operational risk management and measurement practices should be targeted which will assist in transitioning to the SMA.
Pari passu application
(Refer sections 4 and 5)
It is observed that within the BI and loss component, no distinction is made between businesses lines, products, processes, event types (notably conduct risk and related fines), expected versus unexpected losses, and diversified versus concentrated operations.
Therefore, for the same order size BI or loss components, these translate into identical capital requirements. This appears to be a departure from the approach for standardised approaches followed by the Committee for other risk types such as credit risk (e.g. BCBS D347) and market risk (e.g. BCBS D352, Section B) where standardised approaches amongst other are designed to capture idiosyncratic risk exposures according to product type, counterparty type and risk class within a bank instead of a one-size-fits-all approach.
While the proposed SMA may be a reasonable approach at a consolidated group level for a diversified banking group, care should be taken with the interpretation of SMA capital at a specialised subsidiary level, because SMA capital may not be comparable across subsidiaries.
Complexities relating to acquisitions and divestments
(Refer sections 4, 5 and 6.1)
In paragraph 43, bullet 2 notes that “Internal loss data are most relevant when clearly linked to a bank’s current business activities, technological processes and risk management procedures.” However, when a business or legal entity is acquired or divested from, the BI component will reflect the full effect only after three years. Assuming this business has good quality loss data, the loss component will reflect the full effect only after ten years.
We would appreciate additional guidance from the Committee in respect of permissible adjustments to inputs used in the BI and loss components due to material acquisitions and divestments, and retired products. For example, is not clear from the consultation how the SMA calculation at a consolidated group level will work when a bank with good quality loss data acquires a material entity that does not have good loss data. We are of the opinion that such further guidance would improve comparability because it could aid in preventing varying interpretations across jurisdictions.
Specific questions feedback
Question 1:
What are respondents’ views on the revised structure and definition of the BI?
We are of the view that the revised BI is an improvement on the version published in BCBS 291 due to the specific improvements cited in section 4.1 of BCBS D355, namely to avoid counterintuitive contribution to capital of some of the income statement items and certain bank business models. In our opinion the main drawback of the BI is that it primarily relies on business size as an indicator of risk. It also does not explicitly consider the forward-looking risk profile (as evidenced through operational risk management tools such as indicators and scenario analysis), nor risk mitigation or risk transfer measures that are in place. Undesirably the BI therefore appears to weaken the link between risk management practices and the capital impact compared to AMA; under AMA this link contributes materially to incentivising the appropriate senior management involvement and investment in risk management. Our relevant feedback and items for clarity are listed in the section “General Comments” above, with additional specific feedback below:
Consequences of improving controls after large losses
(Refer paragraphs 23, 35, Annex 1)
In the event that a large loss is incurred, the likely response of the bank would be to improve control deficiencies that contributed to the occurrence of the loss incident. Expenses incurred in improving controls are included under the “Other Operating Expense” (OOE) in the BI component. In addition, the loss amount would also be added to the OOE. Furthermore, the loss would be included in the loss component.
This treatment gives rise to two observations:
(i) Expenses incurred in improving controls could increase the SMA capital which is counterintuitive, and
(ii) Large losses are considered multiple times in the SMA capital calculation which seems overly penal and clarity is sought from the Committee on our interpretation.
In order to incentivise control improvements, we propose giving credit to any bona fide expenses incurred to improve controls. The bank’s external auditors can vet such expenses to avoid any abuse.
Simplicity and comparability
The construction of the SMA is not simple. It requires extensive data not dissimilar to hat needed for internal models. Furthermore, the different accounting standards and practices between certain jurisdictions can impact on the interpretation of the BI, reducing the benefit of comparability.
Guidance from the Committee on which financial statement items should be included or excluded for the BI, within the internationally adopted accounting standards would not only improve comparability but reduce the time it will take to achieve comparability.
Question 2:
What are respondents’ views on the inclusion of loss data into the SMA? Are there any modifications that the Committee should consider that would improve the methodology?
In our opinion the inclusion of the loss component does improve the risk sensitivity of he SMA, however the fact that it is purely historically focussed, and spans a relatively long historical time window of ten years are concerns due to it therefore not necessarily eflecting the current risk profile, and past losses having an impact on capital long after he loss date. Our feedback and items for clarity are listed in the section ldquo;General comments” above, with additional specific feedback below:
Definition of loss
(Refer section 6.2)
The text at bullet 4 of section 6.2 states that “Banks must not use losses net of insurance recoveries as an input for the SMA loss data set”. Although this statement specifies that insurance recoveries should not be considered, clarity is required on the types of loss recoveries that are permissible for the purposes of calculating the net loss used in the loss component.
Insurance recoveries are excluded in the loss component calculation. This recovery already occurred, and is therefore verifiable proof of working risk transfer that should be taken into account in the measurement of the risk profile as is intended with the loss component. To the extent that the intention is a forward-looking view of working insurance risk transfer, which may not continue to be as effective as was historically observed; it is our view that limited insurance offset be allowed to continue incentivising the exploring and development of effective risk transfer mechanisms.
Permissible loss adjustments
In paragraph 45, permissible loss adjustments of operational risk events is allowed, however further clarity on specific adjustments would be appreciated including those made within the reporting year such as non-insurance recoveries.
Reference dates
(paragraph 45)
In paragraph 45, bullet 1 notes “The bank must use either the date of discovery or ate
of accounting for building the SMA loss data set.”
We believe that to improve comparability only one option should be allowed. The unintended consequence could be that peer group banks even in the same jurisdiction would not be comparable. We propose the accounting date as the appropriate reference date in the loss component to align with the implied loss reference date used in the BI component. In the event that a loss has multiple associated impacts with each of its own accounting date, the earliest accounting date of all the associated impacts should be the reference date.
Large loss thresholds
(Refer paragraph 31)
Clarity is requested on how the large loss threshold of €10 million and €100 million, which are specified only in Euro, should be adjusted for non-Euro jurisdictions. Potential unintended consequences of having these large loss thresholds in the SMA specifications are “cliff” effects which could cause capital volatility as large losses roll out of the observation time window, and differing loss treatment in respect of classifying, accounting and reporting of large losses may reduce the comparability across peers.
10-year loss observation period for the loss component
(Refer paragraph 43)
The requirement to base the loss component on 10 years’ of data results in the loss component having an extremely long memory. Over such a long time period changes to the risk profile including changes to the internal control environment, business mix and business processes would most likely have taken place making older losses less relevant or even irrelevant to the current risk profile.
A further observation is that the 10-year loss period in respect of the loss component is not aligned with the 3-year period for which losses will be included in the BI component.
his could give rise to volatility in the SMA capital estimate due to large losses rolling out of the BI component but still being reflected in the loss component.
A potential solution is a shorter loss observation period that would be more relevant to the current risk profile. A further enhancement could be to introduce a weighting scheme that assigns a higher weight/relevance to recent losses. Both these proposals would improve the comparability of cumulative losses used in the BI and loss component thereby also improving the stability of the SMA capital.
Clarity on treatment of adopted loss data thresholds
(Refer paragraph 43)
It is stated that banks must apply a de minimis gross loss threshold for internal loss data collection of no higher than €10k. Where a bank adopted a lower threshold than €10k the Committee is requested to specify which threshold to use for purposes of calculating the loss component’s “Average Total Annual Loss”, i.e. the bank adopted threshold or the €10k threshold.
Clarity on timing losses in the context of the loss component
(Refer section 6.2 (e) and paragraph 45)
While the relevance of timing losses in the context of the BI component is understood, their treatment for the loss component is unclear as paragraph 45 requires loss impacts to be reported against a single reference date for the loss component.
It should further be noted that timing losses may reflect differently in the BI and loss components which may impact the ratio of these components used in the Internal Loss Multiplier.
Question 3:
What are respondents’ views on this example of an alternative method to enhance the stability of the SMA methodology? Are there other alternatives that the Committee should consider?
We note that the calibrated value of the parameter m has not been published yet, and depending on the final choice of m, the degree to which this method improves stability would vary. Therefore, we are in favour of this alternate method to enhance stability provided that the final value m is calibrated such that an acceptable balance is achieved between the SMA capital stability and responsiveness to the current risk profile.
Conclusion
As an industry, we appreciate the way the Committee has incorporated the emerging markets into their dialogue and extract great value from the publication of official frequently asked questions documents that form part of the overall Basel III package. Quantitative impact studies have aided in calibration and provided a useful starting point for regulatory reporting discussions and the interpretation of the Basel III package of reforms for our local market.
Our large South African banks have all adopted AMA for operational risk and through forums facilitated by ourselves and attended by our regulators, we have been able to share information and experiences within a competitive banking landscape.
Furthermore, in examining the approach taken by our members to the way they apply operational risk measures within their organisation and report operational risk to our supervisor, it is clear that there were a multitude of interpretations and approaches.
Through our forums, we have been able to develop agreed approaches on various aspects of the application of the AMA and general operational risk measures, with our regulator that has been invaluable in producing a more consistent result across our members.
We believe that advanced approaches for operational risk have merit, as our view of banking is that it is a complex environment deserved of complex methodologies. We appreciate the desirability of simplification, reiterate our concern that simplification could overstate the capital impact, reduce the incentive to migrate to more complex approaches and potentially make comparisons between banks academic.
We believe that our members and many other jurisdictions, having invested in the advanced approaches, would continue to adopt best international practice and further the discipline, despite the regulators in many jurisdictions no longer requiring banks to adopt AMA. It is our experience that a dual process of managing the risk in the bank using advanced approaches and reporting to the regulator on a different basis complicates the relationship between the bank and its regulator. This became abundantly clear in the migration from Basel I to the Basel II framework where many of the Basel II proposals were already adopted in banks. The introduction of the Basel II framework was a substantial adjustment in thinking for regulators but more of an alignment between banks that had evolved organically in their interpretation of these more advanced approaches. Regulating and supervising the banking sector on a lower standard than that used by the bank, can also impact on the ability of regulators and supervisors to attract talent and engage meaningfully with the industry.
We subscribe to the harmonisation of standards across the globe and have demonstrated our willingness to adopt pronouncements of the Financial Stability Board and their international standard setting bodies despite the difficulties that emerging markets have in adopting to these standards, more suited for global financial centres.
Varying accounting standards, inconsistent interpretation of definitions, different domestic frameworks, the size of the regulated entity and the use of hypothetical portfolios across jurisdictions still introduce important elements that will make simpler approaches difficult to compare or interpret.
We hope that our comments will add some value to your deliberations and remain available to the Committee should they wish to gain further insight on our submission.
Assuring you of our best attention.
Yours faithfully
Mark Brits
Senior General Manager – Prudential Division
Dear Sir
Financial Intelligence Centre Amendment Bill: B33-2015
1. Following on the public hearings on 11 November 2015, the Portfolio Committee afforded us an opportunity to make further representations on the Financial Intelligence Centre Amendment Bill (“the Bill”) after consultation with National Treasury (“NT”) and the Financial Intelligence Centre (“FIC”). The Committee also requested that the submission cover critical implementation and compliance issues under the existing FIC Act (and as amended by the proposed Bill) where the industry and NT/FIC are not in consensus, and for an indication of the estimated cost assessment of the impact of the Bill and new compliance requirements.
2. In line with this opportunity a delegation from BASA met with NT on 9 December 2015 to raise its “critical issues” under the Act and the Bill as suggested by the Committee, but unfortunately this (brief) initial meeting only served to highlight that much more consultation is required between the policy makers (NT), the regulators (FIC), the supervisors (SARB/FSB), and the wider affected financial sector, especially given the acknowledged breakdown in prior consultation processes on the Bill, and the wider issues in FICA that remain in contention between the parties. It was accordingly agreed that BASA would continue to prepare a submission to the Portfolio Committee as requested, but also that NT would schedule longer, more focused workshops on the critical issues between the affected parties in the 1st quarter of 2016. Given this extended consultation process NT would also liaise with the Portfolio Committee on the timelines for the Bill going forward.
Critical regulatory and supervisory compliance principles
3. As noted by NT during the Portfolio Committee hearings on 11 November 2015, “the reputational risk of noncompliance exceeds the cost of compliance”. This is because of the standardised international regulatory framework to combat money laundering, terrorist financing and the proliferation of weapons of mass destruction (“AML/CFT”), driven by the Financial Action Task Force’s (“FATF”) 40 Recommendations. The core underpin for South Africa’s anti-money laundering and combatting the financing of terrorism programme is, therefore, domestic compliance with these international FATF 40 Recommendations. This ensures a minimum level and standard of compliance across the world in order to improve the efficacy of the international focus on criminal proceeds and the financing of terrorist acts. Such uniformity is important, inter alia, for 2 very specific reasons:
3.1 The standardised international legislative, regulatory and supervisory framework, which is evaluated on a country basis by outside parties under agreed “Mutual Evaluation” protocols; significantly non-compliant countries are sanctioned by the international community, and change recommendations are made for less significant non-compliance (e.g. the 2009 Mutual Evaluation of South Africa, the 2012 revised 40 Recommendations and the imminent upcoming new Mutual Evaluation of SA, which have underpinned the urgency of the Bill).
3.2 Consistent compliance with this standardised regulatory and supervisory framework within multinational financial groups, which are expected to implement equivalent standardised internal compliance policies and procedures; in this context it is an accepted international supervisory requirement that such financial group policy is based on the regulatory requirements of the home country (i.e. where the holding company is), unless the requirements within a specific subsidiary/ branch host country are higher than these home country requirements, in which case the higher, local standards will prevail.
The Economic Freedom Front (EFF) march and demands today to the Reserve Bank and Financial Institutions
The Banking Association South Africa notes the demands made by the EFF during their march today (Tuesday, 27 October 2015) and recognise the right of parties to embark on peaceful protests to express their views.
We must emphasise we have full confidence in the South African Reserve Bank (SARB) in its role as the primary regulator of banks. The quality of their regulatory role held us in good stead through the financial crisis emanating from Europe and the USA. We also have confidence in the SARB’s management of monetary policy in South Africa.
The banking sector in South Africa is a significant funder of mortgages, SME’s, corporates, infrastructure and other areas of activity that contribute to economic growth. Banks conduct their business in a manner that protects depositors’ funds and within the extensive regulatory dictates they must follow.
We acknowledge the serious socio-economic problems, including poverty, inequality and joblessness, our country faces. We must emphasise the most sustainable manner to address these issues is the creation of an environment that gives investors the confidence to make long-term investments in South Africa that will create businesses and jobs. Banks can then finance corporates who invest and create such businesses.
We also urge creative partnerships between banks, government and other stakeholders to enable access to finance in a sustainable way to as many people as possible in our country. Such partnerships must include the creation of an environment for banks to broaden access, while making a profit and abiding requirements. That is the only sustainable route!
ENDS
For media enquiries or an interview with the MD please contact
Ms Thenji Nhlapo
Media & Communications Officer
Managing Director’s Office
The Banking Association South Africa
Tel: +27 11 645 6729
Cell: +27 76 791 6918
E-Mail: thenjin@banking.org.za
Dear Sir / Madam
Revisions to the Standardised Approach for credit risk
The Banking Association of South Africa (“BASA”) appreciates this opportunity to comment on the Basel Committee on Banking Supervision (BCBS) proposal to revise the BCBS standardised approach to credit risk to share our comments. Our response to the Committee?s specific questions has been captured on Annexure A; however we would like to provide some high level comments on the consultative document.
- We support a more risk sensitive approach to drive comparability
We support the need for a more risk sensitive approach to the calculation of capital requirements for credit risk exposures to aid comparability, but are concerned that the new proposals seek to deliver a one size fits all approach without recognising that exposures in the same asset classes can behave differently in different jurisdictions, because of the operation of national law, accounting approaches or market conditions.
The requirement to default “no data available” risk weights to 300% is exceptionally onerous. Such a threshold might be the appropriate long term approach to encourage proper risk based data collection however; we strongly feel that this risk threshold (if retained), should be phased in over a multi-year period. In addition, applying the proposed risk drivers on banks that are not yet subject to Basel III makes the framework biased against banks in emerging markets as these are primarily not on Basel III and would thus be subject to the punitive 300% risk weight.
- The impact of the changes will be wide ranging
The impact of the proposed revised standardised approach will be very wide ranging requiring banks to hold significantly higher levels of capital across the whole range of asset classes which will cause them to re-assess their business models, perhaps withdrawing from lending to elements of their current customer base that are more capital intensive. These proposed changes would likely result in more complexity in implementation as the required information may not be readily available in the level of detail required. The proposals may therefore lead to considerable IT investments.
- Credit ratings continue to have a role
We also support a simplified approach to the calculation of credit risk but do not support the complete removal of the ability for banks to take external credit ratings into account. For the majority of asset classes external ratings performed as expected as a rank-ordering predictor of default. Whilst agreeing that banks should not mechanistically rely on ratings they should be permitted to retain the ability to use them as part of their overall credit risk assessment process.
Given that external ratings use more inputs than the two risk-driver approaches being proposed, they should be retained to calculate capital and the proposed risk drivers applied only as an alternative when external ratings are not available.
- The proposed changes are too important to be rushed
We are concerned that the Committee is seeking to finalise its proposals with twelve months which we believe is too short a time frame for such a wide ranging revision.
We support the upcoming Basel monitoring exercise that will include QIS on credit risk SA. This will be used as an opportunity to complete in-depth analysis not only to gauge the impact but also to assess the appropriateness of the proposed risk drivers (as well as examine other risk drivers). However, sufficient time should be given to banks in doing the QIS since producing the needed data would likely be challenging.
In the attached annex we provide our responses to the questions contained in the consultation paper.
We thank you for taking our comments into consideration, and we look forward to future discussions on these issues.
Dear Sir / Madam
Re: Response to the second consultation on revisions to the Basel securitisation framework
The Banking Association of South Africa and its members appreciate this opportunity to comment on the Basel Committee on Banking Supervision (The Committee) second consultative document to revise the Basel securitisation framework.
We appreciate the further work which has been undertaken on the proposed capital framework for securitisation and welcome the changes to the hierarchy of approaches, the revised floor and clarification on the use of inferred ratings. The new proposal is an improvement on the old version in terms of the simplicity of the scheme, but still looks to penalise high-quality securitisations and we therefore believe that some sections of the revised document need further consideration.
We believe it is important contextualise the securitisation market in South Africa before addressing the questions raised in the consultation document, attached as Annexure A.
Total outstanding listed debt in the South African debt capital markets (including securitisation) as at 31 December 2013 was ZAR 1.7 trillion, of which South African Government debt accounted for ZAR 1.1 trillion. The remaining outstanding corporate and securitisation debt amounts to ZAR 692 billion and is broken down in Table 1 below.
Securitisation transactions represent a relatively small component of the overall South African debt market with Residential Mortgage Backed Securities (“RMBS”) making up the overall majority of issuance.
Securitisation transactions do provide originators with an alternative means of financing that contributes to the diversification of their funding sources, which is becoming increasingly important given the challenges with the implementation of the Net Stable Funding Ratio (NSFR) in South Africa. RMBS, in particular, is one of the few possible sources of liquid assets in South Africa that can be included in the assessment of the Basel III Liquidity Coverage Ratio (LCR).
Although there is a need to ensure the appropriate capitalisation of banks in respect of their exposures to securitisations, we believe that the latest proposed revision to the securitisation framework remains overly conservative and will lead to a reduction in the use of securitisation as an alternative means of financing. This we believe would further diminish the South African banking industry’s ability to comply with NSFR and LCR whilst hampering growth in segments such as residential mortgages.
It is also important to note that to date no investor has incurred any loss in respect of their participation in the South African securitisation market. It is our opinion that this is due to appropriate legislation governing the origination of debt as well as legislation governing entrance to the securitisation market and not only legislation governing the capitalisation banks’ exposures to securitisations.
We would like to draw The Committee’s attention to the following issues in addition to the specific questions contained in Annexure A.
COMMENTS ON: FUNDAMENTAL REVIEW OF THE TRADING BOOK – OUTSTANDING ISSUES
We thank you for the opportunity to provide comments on the third consultative paper on outstanding issues related to the fundamental review of the trading book- capital requirements. We have endeavoured to capture all the comments by our members below and would also like to inform you that we are also in support of the ISDA response document.
We fully support that this consultative document aims to contribute to a more resilient banking sector by strengthening capital standards for market risk and forms part of the Committee?s broader agenda to reform regulatory standards for banks in response to the financial crisis. We furthermore recognise the Committee?s incorporation of industry feedback since the previous round of consultation; in particular we welcome the continued development of the sensitivity based approach over the cash-flow based approach for standardised calculations.
1. Executive Summary
The new standardised approach, although much more aligned in terms of how banks actually measure their market risk exposure, requires much more data and system capabilities (close to the level of IMA approval). For South African banks, this is typically not a major issue; but in our African subsidiaries, that currently report only under the standardised approach, the implementation of the new proposed standardised approach is a significant challenge due to their systems and data capabilities not been able to deliver what is required under the new standardised approach without added cost and skills requirements.
We have a concern with the new regulations regarding the transferring of interest rate risk in the Banking Book (“BB”) to the Trading Book (“TB”), and the restrictions on the offsetting of this risk with similar risks in the TB prior to hedging it out. We will elaborate further on this below.
2. Proposed Rules on Internal Risk Transfers (IRT)
The IRT is a way of effectively managing risk within an organisation in the manner most efficient to the bank. Efficiency includes minimising the costs related to external trades, limiting exposures to external counterparties, as well as optimising risk management using specialist, centralised skills.
Risk transferred to the TB might diversify existing risks in the TB and the remaining risk is then managed within risk limits and appetite levels of the TB. This process allows for robust risk management for the entity and the industry as a whole, leading to lower market volumes and volatility. It also enables internal risk diversification before offsetting any remaining risks externally, which (in our view) is one of the key functions which banks are meant to fulfil in the economy.
The IRT process is however not necessarily used in isolation to manage interest rate risk in the banking book (“IRRBB”). Certain trades are externalised, for example when designated into a hedge accounting solution, or to comply with other accounting requirements.
As an illustration, when the bank offers market making services on derivatives to its customers, it is usual that the function in charge of mitigating IRRBB (Asset and Liability Management, ALM and/or Treasury) executes all or some of its risk mitigating derivative instruments with the TB of its own bank. Such IRT processes are efficient for the bank, since they minimise the number of desks facing the market and benefits from the diversification effects between trading book customers? transactions and ALM Treasury?s transactions, which effectively reflect risks originated from BB customers.
The TB desks are subject to a risk management framework, notably limits to their market risk exposures that de-facto binds them to materially offset the risks in relation to customers? transactions (including ALM Treasury?s transactions) with external transactions. Any residual risk is subject to the market risk capital charge applicable under current Basel III standards. Therefore, the risk of arbitrages of the boundary due to imperfect external offsetting process is extremely small, and we do not see the need for additional measures, subject to the trading desk being well governed via the market risk management framework.
We believe the proposed IRT rules only restrict banks? internal risk management practices, without clearly identifying the risk that is being addressed. The consultation document points to capital arbitrage as the main motivation, but it is unclear how this manifests through the current market risk management processes and measures in place and why this will not be addressed by the work being done by the Interest Rate Risk & Credit Risk in the Banking Book (“IRRCRBB”) Task Force.
We are also concerned that the stated need to prevent capital arbitrage could have unintended consequences for how banks operate and how they manage their risk, if implemented in accordance with current proposals.
Arbitrage opportunities exist in the boundary definition. Having a clear definition of BB and TB instruments, and what constitutes risk, can be transferred between the BB and the TB is key in eliminating capital arbitrage. Once risk is transferred to the TB however, we believe it should follow the same capital treatment as the rest of the TB and be seen as an external trade given that the transfer reflects risk from external counterparties in the BB.
Capital is currently held in the TB for the total risk being managed by the TB team, which is a reflection of total fair value risk in the entity. Given the nature of BB trades, a portion of the risk might not be able to be hedged in the market, but together with other risks for which there is not a liquid market, may prove a natural offset.
2.1 Specific commentary on options provided in the consultation document
Option 1 in the consultation document would basically oblige ALM Treasury to mitigate all its IRRBB with external counterparts, which would decrease the bank?s ability to benefit from diversification and natural hedging effects and would increase its counterparty risk and the related liquidity requirement.
Option 2, would have the very same detrimental impact as Option 1 since, with a portfolio limited to ALM Treasury?s transactions, there would basically be no possible diversification and natural hedging benefits.
Hence, both options would likely impact how banks are managing IRRBB with no immediately apparent benefits from a prudential and / or a risk management perspective.
While we understand the Committee?s intention to distinguish clearly between the Trading and Banking Book activities, we would like to outline the potential impacts and would also like to address concerns in an alternative way.
The currently proposed Options 1 and 2 will have a significant impact on bank?s risk management practice and capital requirements:
· Capital requirements will increase as any risk mismatch of TB and BB transactions will trigger capital charges under the TB environment. Currently, the overall position is measured after internal diversification. As a result banks can be charged for a risk that does not exist from an overall economic perspective, effectively meaning that capital requirements are being “grossed up”. This approach would not be in line with for instance credit risk, where risks are allowed to be pooled when capital requirements are determined
· Aiming to minimise capital charges in the TB, banks could be forced to centrally manage BB risk positions in a BB environment and only transfer risks to the eligible TB which might already has an external hedge available. As a consequence this could lead to risks not been managed optimally, but rather in a manner that minimises capital charges. Furthermore the BB could be forced to keep risks which are not in the BB?s interest
· Not allowing for internal risk diversification would synthetically increase market flows as all BB related hedge transactions would need to be externalised with the market. This effect increases costs for banks and will therefore have an impact on pricing and product offerings banks are able to make to BB clients (e.g. retail clients, fixed rate mortgages and corporate financing)
· Systemic risk is also likely to increase as a result of reduced diversification and increased market flows
· A possible change in market behaviour and introduction of moral hazards when banks do a “wash” trade between each other to move the risk to the traditional TB and circumvent the proposed increased governance
· Additional rise in counterparty risk exposure to banks with whom these trades would be executed
· In instances where some or all of the major banks are included in a single syndicated asset or liability structure, it would become impossible for those banks to externalise that risk without creating systemic risk within the monetary system. This is especially true for the South African market with a small number of participating banks.
· The proposed Options 1 and 2 do not address how any existing IRTs, which is currently used as an offset against the TB, will be handled and specifically do not clarify whether these offsets will have to be unwound and externalised.
We therefore recommend that banks should retain the ability to transfer BB risk to the TB that allow diversification and natural hedging effects, subject to the mandates and limits of the TB and governance standards that meet supervisory approval.
The concern about capital arbitrage can be mitigated if there is full transparency in the risk transferred without having to resort to the options proposed in the consultation document.
We furthermore recommend that such IRTs are accompanied by reporting requirements that show the details of the risk transferred as follow:
· IRT-transaction should be executed solely with an IRT-eligible TB-desk
· The IRT must be documented with respect to the BB interest rate risk being hedged and the sources of such risk. This information must be reported regularly to supervisors
· Banks must have clearly articulated and documented process in place for identifying and quantifying BB interest rate risk to be hedged through IRT?s. The methodology needs to be consistently applied, with any material changes in approach approved by the bank?s Asset and Liability Committee (“ALCO”) before implementation
· The IRT-transaction seen from the TB view is recognised in the TB market risk framework together with the other TB transactions
· The IRT-transaction seen from the BB view is recognized in the BB prudential framework.
· Each IRT-eligible TB-desk („the desk?) should:
o List the financial instruments („the instruments?) the desk can use to service its customers (including internal customers)
o Document the hedging strategy applied to the IRT in a transparent and comprehensive way
o Meet specific requirements on an on-going basis to ensure risks are appropriately offset
Examples of such requirements include:
· The desk routinely stands ready to and actually executes the instruments in both directions, and is willing and available to quote or enter into long and short positions in the instruments, in commercially reasonable amounts and throughout market cycles on a basis appropriate for the liquidity, maturity, and depth of the market for the instrument
· Trades between BB-desk and TB-desk should be on similar commercial terms to trades executed with external counterparties
· The bank has established and implements, maintains, and enforces written policies procedures, internal controls, analyses and independent reviews identifying and addressing:
o The financial instruments the desk is mandated to enter into Page 5 #154320 o The techniques and strategies the desk may use to manage the risks of its activity and the personnel responsible for ensuring that the actions taken by the desk to mitigate the risks
o The actions the desk take to mitigate promptly the risks of its financial exposure consistent with the limits; the products, instruments, and exposures the desk may use for risk management purposes
o The TB-desk should be subject to normal market risk management and monitoring practices and included in the bank?s overall traded market risk framework
o Limits applicable to the desk should be based on and be consistent with the nature and amount of the activity offered by the desk and include:
– The amount, types, and risks of its activities,
– The amount, types, and risks of the products, instruments, and exposures the desk may use for risk management purposes,
– The level of exposures to relevant risk factors from its financial exposure.
Moreover, we believe there should be continued interaction between the TBG and the TFIR on the issue of IRT. Given that the TFIR is doing a review of the prudential rules for IRRBB, we would assume that it would have some views as to the hedging activities that would be appropriate for banking book positions, and how these should be captured for regulatory capital purposes.
3. Revised Internal Model
Whilst the use of a base liquidity horizon of 10 days significantly simplifies the required system setup, the liquidity horizon scaling formula introduces a high level of operational intensity. For example, for a stressed ESR,S on the overall portfolio plus, say five components or asset classes (Interest Rate, Foreign Exchange, Equity, Commodity, Credit), it will be required to run Expected Shortfall for five different subsets of liquidity horizons (10-250, 20-250, 60-250, 120-250 and lastly 250). Additional calculations would also be needed to obtain ESF, C, ESR, C on the overall level.
With respect to liquidity horizons, “FX rates – liquid currency pairs” are grouped under the 10 day risk factor categorization and “FX rate (other currency pairs)” are categorized as 20 days with a footnote depicting which currency pairs are considered under this category. The footnote, however, lists many currency pairs that we are of the opinion should be considered as liquid, including but not limited to USDZAR. In addition we are appreciative of shorter liquidity horizons for FX and Rates risk factors, but it is questionable why currency pairs are being grouped in two separate currency pairs as there is a potential for FX triangular arbitrage.
On the matter of the Revised Standardised Model being considered as a possible floor to capital figures generated from the Revised Internal Model, we are concerned that a too conservative implementation could eliminate incentives for banks to maintain sophisticated risk systems. With the increase in complexity of both models, and the introduction of a mandatory model we believe that many banks might reconsider the use of the Internal Model – at great loss of an informative risk system that currently is the base of detailed risk analysis and comprehensive Stress Testing.
4. Revised Standardised Model
In terms of changes proposed to the standardised model and specifically concerning the capitalisation of basis risk, we believe further clarity would be required, specifically relating to the Disallowance Factor vs. the Correlation Method. Whilst we are in agreement that the Disallowance Factor method could possibly cause capitalization for basis risk where it does not exist, we perceive the Correlation Method to further add complexity to an already intricate model.
The treatment of non-linearity in Sensitivity Based Approach appears to have taken direction into the consideration of Vega and Curvature as inputs. Despite this, the use of a scenario-type approach (making use of spot-vol matrices) still appeals very much to us. We believe that the existing data availability and use of this risk measure in our existing risk management framework greatly supports the simplicity that is required under a Standardised Model. The treatment of Vega and Curvature also requires further development, most significantly from a data extraction, especially when considering the treatment of indices, which should be considered both in terms of considering the balance between simplicity and risk sensitivity as well as timelines for implementation.
We have to express concern regarding the increased complexity introduced by the Sensitivity Based Approach, and whether smaller banks with less sophisticated systems will be able to conform adequately to this model. It may be argued that these banks would be shielded against the full extent of the newly introduced complexity, since they typically trade in fewer asset classes and the trading is kept mostly to vanilla type activities. At the same time, the granularity of data required will still put notable strain on these banks.
5. Conclusion
We believe the Committee?s timetable is ambitious, but achievable. Whilst we are not in favour of postponing much needed regulatory reform, we believe that given the extent of the changes, it will be prudent to allow for more Quantitative Impact Studies. This will not only provide a platform for optimum calibration between models but also ensure adequate preparation in terms of operationally stable systems.
We hope that our comments have added value to your deliberations.
Should you require any further information, please do not hesitate to contact me.
Dear Sir/Madam
Comment on the BCBS Consultative Document Net Stable Funding Ratio Disclosure Standards
We welcome the opportunity to comment on the public disclosure templates for the Net Stable Funding Ratio and would appreciate further clarity on the following three aspects.
In Section 1, paragraph 7 states:
“The disclosure requirements set out in this document are applicable to all internationally active banks on a consolidated basis but may be used for other banks and on any subset of entities of internationally active banks to ensure greater consistency and a level playing field between domestic and cross – border banks”
The wording of the paragraph provides for disclosure of the NSFR to apply at the level of consolidation. This could be interpreted by the investor that there is fungibility across jurisdictions and therefore reliance can be placed on the consolidated number.
The NSFR common disclosure template in Section 2, paragraph 14 introduces a “no maturity” column without providing weightings from the final NSFR text. The term “no maturity” would naturally apply to perpetual products; however there are products without a specific contractual maturity date such as overnight deposits payable on demand or deposits that require a minimum notice period before repayment, that on any reporting day, a minimum remaining contractual period can be determined for purposes of performing the NSFR calculation and therefore should be included in the appropriate category e.g.
The South African regulator has elected to provide banks with the ability to access a contractual committed liquidity facility with the Central Bank to meet the liquidity coverage ratio. Is there an expectation that there would be any quantitative or qualitative disclosure of this facility in the NSFR and how would this be achieved?
I hope that our response to your consultative process will assist in your deliberations.
Should you require any further information please do not hesitate to contact me.
Dear Sir/Madam
Comment on the BCBS Consultative Document Operational risk – Revisions to the simpler approaches
We support the work done by the Basel Committee’s Working Group on Operational Risk and welcome the release of the consultative document on the revisions to simpler approaches for operational risk.
We shall address each question in turn.
Question 1: Are there any other weaknesses in the existing set of simple approaches that should be addressed by the Committee?
The most significant weakness in the current approach relates to the increasing capital requirement as gross income increases, whilst the operational risk exposure does not necessarily increase in the same ratio. This approach does not consider the maturity of the organisation’s risk management.
Question 2: Does a single standardised approach strike an appropriate balance across the Committee’s objectives of simplicity, comparability and risk sensitivity?
A single standardised approach should allow for increased comparability across banks as there will only be one operational risk capital requirement under the simpler approach to compare and interpret across banks.
However, the revisions are heavily slanted to capital management and there is a risk that the importance of day-to-day risk management and the incorporation of appropriate operational risk sensitivities may be diluted in this process. Banking is complex and the pursuit of simplicity and comparability, although desirable, should not be prioritised over sound risk practices. The emergence of risk management as a key risk discipline post the financial crisis, has provided much needed management support for the further development of this risk discipline.
The appropriate balance is difficult to achieve on a global scale and the BCBS could assist with its objective to improve the usefulness of this information to the end user, by determining who the targeted end user would be, to ensure that it is fit for purpose.
Question 3: Are there any further improvements to the BI that should be considered by the Committee?
Review mechanisms
One of the weaknesses of the existing standardized approach for operational risk is that there is no established framework to review its calibration. This is recognized in the consultative document which states that
“…no rigorous review has been made of the effectiveness of GI (or other potential indicators) as the proxy for the operational risk exposure of a bank and the adequacy of the calibration of the regulatory coefficients of the BIA and TSA”.
However, there is no mention in the document if the proposed revised standardized approach would be subject to regular review. Such regular review is necessary to make sure the revised standardized approach remains reflective of banks’ actual risk profiles. Consideration should be given to providing appropriate review mechanisms to ensure that loss data characteristics across jurisdictions are appropriately incorporated.
Business models
The Business Indicator (BI) is driven by income statement components that appear to underpin the size and magnitude of a bank’s operations. This is intuitively appropriate for the rudimentary measuring of operational risk capital requirements.
Consideration of the structure of a bank’s business model may require further deliberation, particularly if a bank has a large “services component” and whether it is appropriate to compare it to a bank which is dominated by a large “financial component” when both have a similar BI. The question arises as to whether the BCBS regards the capital requirements for a bank whose business operations primarily result in fee income as equivalent to a bank whose operations primarily result in trading profit.
Furthermore, banks operating in high fee income regions are penalised more than banks operating in low fee income regions. Our view is that fee income is a relatively safe form of income and hence operational risk associated with it should be low. In South Africa, interest income is a significant contributor to revenue. The business indicator penalises banks with high interest income, however regions where interest rates are low may not recognise this as a concern.
The previous TSA tried to address some of these concerns in part by promoting business lines and beta factors, which we acknowledge could introduce the risk of mapping products and services to the incorrect business line, an onerous task in any bank and a risk that regulators and auditors would need to consider when discharging their obligations.
A potential solution to this could be to identify acceptable bands or levels for each component. If the component exceeds the band or level, an adjustment is made, i.e. if a services component is greater than say 50% of the total BI then the BI is adjusted by a factor of X or if the financial component is greater than 60% of the BI it is adjusted by a similar factor of Y.
The proxy for future losses
The development of the quantitative OpCaR model is based on each bank’s internal loss data. Internal loss data is by its very nature, backward looking and is not the best proxy for future losses, as confirmed in the introduction of the executive summary:
“In addition, the changing operational risk profiles of banks may render a calibration based on the past behaviour of variables unfit for the future.”
It is common practice e.g. for the control environment to be strengthened to prevent the recurrence of similar losses. We do recognise the challenges of including scenarios as well as business environment and internal control factors (BEICFs) in OpCaR. However, if the sole basis for operational risk capital is historical loss experience, we believe it underscores the need for a regular review of the revised standardized approach’s calibration.
Paragraph 7 of Annexure 2 notes that the OpCaR methodology does not include information relating to scenario analysis and BEICFs, and that exclusion of these elements may cause, in some cases, an underestimation of the OpCaR figures. Hence, it was decided to use a more conservative class of severity distributions in the OpCaR calculator.
However, it could also be the case that by excluding those same factors, OpCaR figures are overestimated. In addition, operational risk mitigants are not taken into account. Thus, we do not see the need to use conservative severity distribution assumptions in OpCaR.
Pillar 2
Annexure 4 provides an example of guidance regarding qualitative standards that should be observed by large internationally active banks under Pillar 2. It is not clear why there should be separate guidance, rather than just a reference to the Principles for the Sound Management of Operational Risk. Also, it is not clear what “under Pillar 2” means, if a bank fails to meet the standards would it be subject to an additional capital charge?
Mapping
Annexure 4, Section B, item 6 requires banks to include a mapping and reporting of gross income and loss experience by business line or activity. It is not clear why these are still necessary when gross income and business lines are no longer used in the proposed new TSA.
Absolute values
In Paragraph 13, both the interest component and financial component of the Business Indicator require absolute values. We would propose that the absolute value be extended to the services component to address the potential for netting (deductions) arising from different accounting treatments that could, in certain circumstances result in e.g. negative fee income in a particular year.
Question 4: What additional work should the Committee perform to assess the appropriateness of operational risk capital levels?
Calibration
In developing the new simple approach, the adequacy of operational risk capital requirement for banks was primarily assessed and analysed based on the 2010 QIS exercise. To ensure consistency with the 2010 data set, the appropriateness of the new proposed simpler approach and the resulting operational risk capital levels should be calibrated against the 2014 QIS exercise. The 2014 QIS data covers a longer time period than the 2010 QIS and the OpCaR model should be rerun. The results and deductions made from the 2010 QIS, pertaining to buckets and regulatory coefficients should be compared to that of the 2014 QIS to perform a validation of these proposals as provided for in the consultative paper.
In addition, we would propose that a sample of bank’s operational risk capital requirements should be taken across geographies, to assess the effect of the new proposed simpler approach against that of the current simpler approaches at a country level (demonstrating economies of scale). For example, where banks in emerging markets have small BI’s when compared to comparative banks in developed markets with large BI’s, would it make sense to have a differing percentage of operational risk capital requirements?
Domestically, both of these banks are of similar importance to the respective domestic financial system (domestic systemically important bank). It may be that banks with larger BI’s in developed markets play a more significant role in the wellness of the global financial system, however, should the BCBS not also consider the effect on that country’s financial system?
Migration to advance approaches
Banks may use more advanced operational risk methodologies for their in-house management of operational risk and continue to report on the standardised approaches. The incentive to migrate to AMA reporting stems from the recognition of improved risk management practices by the regulator and the reliance that the regulator can place on these advanced approaches is compensated with a lower capital requirement, reflecting a more accurate translation of the riskiness of the bank into capital. The reduction in capital may encourage banks of their own accord, to make the substantial investment required in operational risk systems and personnel.
The existing TSA framework therefore generates a capital requirement that is higher than AMA. However, regulators have national discretion to prescribe a minimum floor to the AMA capital benefit as applied in South Africa, which then limits the benefit derived from using AMA. The refinement of the proxy indicator will intentionally increase the capital impact of the new TSA which will directly translate into higher disallowance of AMA benefit and further reduce the incentive to migrate naturally to AMA.
We would propose that the BCBS reflect on providing guidance on national discretion that will ensure the AMA will continue to generate economic benefits commensurate with the investment in and promotion of the advanced approaches.
Research
As the operational risk discipline grows in importance, we would promote an ongoing commitment by the BCBS to research, review and analyse the latest developments in operational risk to ensure that the private initiatives of the industry, to enhance operational risk practices, remain relevant. Best practices in risk management should become the driver of behaviour rather than the pursuit of lower capital levels.
Question 5: Are there any other considerations that should be taken into account when establishing the size-based buckets and coefficients? How many BI buckets would be practical for implementation while adequately capturing differences in operational risk profiles?
Euro equivalence
Specifying the buckets in Euro introduces an additional risk that currency volatility will now influence operational risk capital requirements. This is not the case with the existing simpler approaches to operational risk. For stable currencies, this is less of an issue, however, for banks in emerging markets where the local currency is extremely volatile it will cause fluctuations in operational risk requirements based solely on the movement of exchange rates.
As an example when the local currency (ZAR) strengthens against the Euro, banks will be required to hold more operational risk capital and the average coefficient (beta) across buckets increases. This appears to be counterintuitive.
The introduction of exchange rates as a driver of operational risk capital may require global co-ordination between regulators to ensure that all banks with exposures to Basel jurisdiction apply the same exchange rate. It may be prudent for the BCBS to publish a table of appropriate exchange rates to use, for anchoring the lookup of capital coefficients for the range of currencies that apply in the different jurisdiction that subscribe to the Basel Accord.
Question 6: Are there any other considerations that should be taken into account when replacing business lines with size-based buckets?
Coefficients
Basel introduces a set of escalating coefficients from, 10% to 30%, based on the size of the bank that results in larger banks holding more capital. In addition, the method seems to penalises banks where there are multiple entities operating in one group.
Business
lines Will the BCBS promote further research on business line specific operational risk characteristics?
Question 7: Could there be any implementation challenges in the proposed layered approach?
At this early stage, the introduction of the layered approach does not seem to introduce complex implementation challenges however, over time the BCBS will need to ensure that jurisdictional idiosyncrasies are appropriately considered and managed.
Question 8: Do the issues of high interest margin and highly fee specialised businesses in some jurisdictions need special attention by the Committee? What could be other approaches to addressing these issues?
Following on from our comments in question 3 under the heading business models, where we address both scenarios, we would support this proposal. Failing to address e.g. the disproportionately high capital applicable to banks that are highly specialised with fee income businesses, might lead to a form of arbitrage, whereby fee structures are reconsidered.
Depending on the outcome of the analysis of the data from the 2014 QIS, it may be appropriate, to allow banks that are highly specialised in fee businesses to adjust the BI calculation above certain levels.
Question 9: What would be the most effective approach to promoting rigorous operational risk management at banks, particularly large banks?
The standard setting bodies have responded to the G20 and FSB by producing an unprecedented series of banking and broader financial reforms. In South Africa, bank management recognises the importance of operational risk and the introduction of higher capital requirements for operational risk is well positioned, despite the many new regulatory requirements that compete for both management time and bank resources. The large South African commercial banks adopted AMA from 2008, with the implementation of Basel II.
The need to translate into appropriate action, within such a rapidly changing environment, operational risk managers would benefit enormously from BCBS guidance, based on quantitative and qualitative analysis of the range of available practices.
Certainty in this regard, could aid in driving convergence to the minimum standards for operational risk management practices, corporate governance and appropriate incentive frameworks that in the process would avoid misinterpretation, the wasting of resources and ensure more consistent risk management practices across the industry.
Competition between banks and the moral suasion of regulators will only serve to increase the desire to comply. In the absence of tangible minimum standards the natural evolution to a higher AMA standard of operational risk may be constrained.
In addition, the regulators would benefit from agreed reporting definitions that would standardise the information collected on operational risk from the banks.
In concluding, we hope that our comments offer sufficient substance for the BCBS to continue its work in the area of operational risk and we thank you for the opportunity to participate in this process.
Should you require any further information, please do not hesitate to contact me.
Dear Mr Quan Tan
Comments on: The revised Standardised Approach to Market Risk – Update on revised Accord texts
Our member?s thank you for the opportunity to provide comments on the two alternative frameworks proposed in your paper of 10 March 2014.
1. Executive summary In general:
1.1 We support the Sensitivity Based Approach (SBA) for the General Interest Rate Risk (GIRR) and Credit Spread Risk (CSR) for the following reasons:
o Simplifies the original proposed calculation;
o Possible leverage off established risk infrastructure in which the sensitivities are already calculated in an existing risk framework or could easily be extended in this regard;
o SBA has a more logical and meaningful risk interpretation relative to the cash flow mapping alternative, i.e. the sensitivities could be utilised appropriately to manage market risk.
1.2 We welcome the addition of the allowance of full netting-off of either sensitivities or adjusted cash flows (depending upon the final Standardised Approach) at an appropriate instrument level.
Notwithstanding the above, we are still of the view that the consultative document (issued for comment by 31 January 2014) and the revised Standardised Approach for Market Risk proposal (issued for comment by 16 April 2014) remains a complex model for small emerging market banks to implement.
Building a new standardised model that requires substantial cash-flow data inputs for each asset will not be cost effective, let alone trying to find resources capable of building such a model. Considering that the modelling approach bears little resemblance to current trading desk management techniques, we propose that these banks, at the discretion of their national regulator, be allowed to retain the use of the existing Standardised Approach.
Banks in the South African context are typically on an internal model and to make a standardised approach so complex that it can be used to calibrate a model, undermines the robustness of the model licence award process. You either have a model that is reviewed by the regulator in the region or use a simplified standardised approach that is capital expensive but simple to implement. Making it more complex to achieve similar results in the two models is confusing as, under the framework for internal models, on any specific desk you will need to be able to switch between the two as the floor removes any incentive to have an internal model.
Some emerging market banks with sub branches in other remote jurisdictions will not have any internal models approved and hence will be required to go via the simple standardised approach as the have simple portfolios.
2. Technical Comments and Feedback
2.1 Annex 2 – Sensitivities Based Approach – Draft Accord
Section 2 Derivation of notional positions
Line 52 – 56 (Page 1of 27)
With regards to the SBA model, the draft proposal indicates that Market Rates or PAR PV01?s are to be used. The use of Zero PV01?s may be more suited given that all of the exposures will need to be bucketed into a vertex of 10 points. We question if there is any specific motive or benefit that The Committee has specifically selected PAR-based bucket exposures and request clarification on this.
2.2 Annex 2 – Sensitivity Based Approach – Draft Accord
Section 2.2 Other asset classes
Line 75 and 106 (Page 3/27) – Treatment of Options and Warrant
Both paragraphs prescribe a similar treatment; with the exception that paragraph 106 prescribes an additional treatise on default risk. The main topic of this discussion pertains to the market risk framework, in this context these paragraphs are treated as equivalents.
The market risk treatment suggests a two-phased approach to the market risk of option positions: (i) The delta position of an option position is treated as Equity Risk for which section “(D) Equity Risk” of the document applies, and (ii) non delta risks are captured through the scenario matrix approach discussed in section “(I) Options non-delta Risk”
The most plausible reason for the two-phased treatment is to offset equity positions which are held as hedges for option positions. If this is not the case, then a single phase treatment – which entails the recalculation of the value of an option position under different conditions – describes market risk more accurately.
By having the two-phase treatment, two species of systematic error may be introduced. These are (i) double counting and (ii) unlike comparisons.
2.2.1 Double counting Double counting entails that the risks that are accounted for as equity risk is accounted for again as non-delta risk. Section (3) of “(I) Options non-delta Risk” describes the process of stripping out of the effect of delta from the scenario matrix. The method entails that the specific delta for a set of perturbations be “stripped out” for that scenario. Mathematically this may be written as
Apparently the intention is to account for it as equity risk; however the equity risk treatise is of a (potentially) different scale. The risk weights for the various buckets are between 30% and 70%, and can only coincidentally offset the general 40% or 50% perturbation of the scenario matrix. It is therefore virtually unavoidable to encounter at least a portion of double counting for option positions.
2.2.2 Unlike comparisons
By not comparing similar instruments will become problematic when dissecting an option position into equity and non-delta risks, with the intention of combining these risks later on.
By not comparing similar instruments will become problematic when dissecting an option position into equity and non-delta risks, with the intention of combining these risks later on.
In its simplest form, and ignoring offsets, the equity risk of an option assumes a market perturbed by the risk weight. Since only net positions (after longshort offsets) are noted, this shift is to be interpreted as to be in the direction of the most adverse effect. It could be either up or down; the true direction is obscured.
To some degree it makes sense to compensate for this move by eliminating it from the scenario matrix, but only insofar the direction and magnitude of the spot perturbation are matched.
Currently equity risk shifts are to be added to the worst performing non-delta position, even if one implies an equity shift in one direction while the other implies an equity shift in the other direct, or even a stagnant underlying market.
Please refer Annexure 1 for a detailed example.
2.3 Annex 2 – Sensitivities Based Approach – Draft Accord
Section 3(a) General Interest Rate Risk Line
114 (Page 8of 27)
With reference to the addition of the allowance of full netting-off of either sensitivities or adjusted cash flows (depending upon the final Standardised Approach standard) at an appropriate instrument level, we would suggest the allowance for netting off at a unique „risk driver? level as well, in order to avoid capital requirements for perfectly hedged (i.e. no basis risk) exposures.
As an example, consider the simple case of a 1 year interest rate swap with quarterly resets; this position is fully hedged through a replication strategy consisting of Forward Rate Agreements (3×6, 6×9, 9×12) with reference to the same floating rate, the hedged position would have no market risk exposure, yet would attract capitalisation (differing instruments).
The allowance for netting of exposures that are priced off a unique risk driver (e.g. the Rand Swap curve) would recognise the effective hedging across instruments more appropriately.
3. Conclusion
The complexity as well as associated implementation difficulty of the new standardised approach is not a good fit for emerging markets and we would request that the BCBS consider the discretionary use of the current Standardised Approach under local Regulator supervision.
We trust that our comments have added value to your deliberations. Should you require any further information, please do not hesitate to contact us.
ANNEXURE 1
Example:
Consider a long call option on the equity price of a small emerging market company (bucket 9). To simplify matters, assume there is only one option on a notional of ZAR100.00, and that there are no interest and dividends, and that we operate in a flat volatility environment; the implied volatility being 25% for this particular option. Assume the option matures in 3 months and it is struck at-the-money. The current fair market price of the option is R4.98, and the delta is 0.525.
The capital requirement for non-delta risk is ZAR1.24. Added to the capital requirement of ZAR36.74 equals ZAR37.99, which is almost 8 times the value of the option.
Clearly, the maximum loss for this particular option is its market value, ZAR4.98.
If one was to use the delta implication accounted for by the equity risk, i.e. ZAR36.74 as the delta risk for the option matrix, the non-delta risk would be summarized as follows:
The worst performing scenarios are 3 and 4, which are both positive.
Adding ZAR31.76 to the drawdown for equity risk (ZAR36.74) equals ZAR4.98, which is precisely the premium of the option.
Naturally, accounting delta risk in this way is also not a good idea as the option matrix now suggests that delta risk is the only risk (because all the P/L outcomes after delta risk are positive).