Tag: Basel Committee

BIS Proposes Enhanced External Auditor Responsibilities for Banks’ Capital Positions

Most major banks chose to use the so called advanced risk weighted approach to calculate the amount of capital they are required to hold as part of the Basel III frameworks against loans they write. This is an immensely complex (many would say over complex) and myopic approach to managing the risks in the banking system. But even then, who reviews and checks the approaches used, the calculations made and so the risk status of each bank?

Certainly the regulators do not have the time and resources to check each one in detail, so they tend to take the results on face value. So it was interesting to note that on 4 March, the secretary general of the Basel Committee on Banking Supervision (BCBS) suggested that entrusting external auditors with some responsibility for checking banks’ calculations of their risk-weighted assets (RWAs) would reduce the risk of errors and regulatory arbitrage.

External auditors would add another line of defense and enhance the accuracy and reliability of banks’ calculations of their Common Equity Tier 1 (CET1) ratios, a credit positive.

According to Moody’s, RWAs constitute the denominator of banks’ CET1 ratios and are commonly calculated by large banks themselves based on their own data. Because they are based on regulatory rather than accounting concepts, external auditors are not commonly entrusted with certifying whether banks’ calculations of RWAs are accurate.

RWA calculations are subject to the scrutiny of supervisors who have often expressed concerns about banks’ internal estimates of their RWAs. This lack of trust has led to many initiatives: in December 2017, the BCBS introduced input and output floors as a means to limit the benefit that banks can obtain from calculating model-based RWAs (the so-called Basel IV framework).

The input floors are minimum thresholds imposed on parameters used in internal models. The output floor adds an overlay of constraint whereby the maximum benefit from implementing internal models is limited to 72.5% of the calculation derived from the new standardized approach. However, this new framework will only apply starting in January 2022 and the output floor will be phased in through January 2027.

Along the same vein, the European Central Bank recently embarked on a multiyear endeavor aimed at checking the quality of banks’ internal models (referred to as targeted review on internals models, or TRIM) and instructed many banks to fix identified shortcomings which often resulted in additional RWAs.

Supervisors to some extent rely on auditors’ verification of financial statements to conduct their own prudential analysis. Beyond this, auditors vet banks’ internal credit risk models under International Financial Reporting Standard No. 9 (IFRS 9), and those models are closely linked to prudential capital models. External auditors already play a role in banks’ data and processes that are critical for the computation of RWAs.

Additionally, some supervisors already commission external auditors to undertake specific work as a means to complement and inform their supervisory work. Within the European Union, supervisors in an estimated 10 countries require external auditors to provide comfort on capital, solvency and other ratios.

However, the formal validation of RWAs and capital ratios remains outside the scope of the audit. There has been no official decision by the BCBS as to whether the relationship between supervisors and external auditors should be standardized. Entrusting external auditors with a mandatory responsibility to certify CET1 ratios in addition to their usual remit would be a bold step if this initiative were to be spearheaded by the BCBS in its role as global standard setter. It remains to be seen whether a sufficiently large consensus will emerge among BCBS members; some supervisors might be reluctant to empower external auditors with such a critical mandate.

To be sure, external auditors’ certification of CET1 ratios would not be a panacea. Calculating RWAs is a complicated process, particularly at banks that rely on risk-modelling techniques. External auditors are unlikely to be able to address all shortcomings that the supervisory community has wrestled with for years (e.g., accuracy and consistency of RWAs), even if the auditors’ involvement would add a useful line of defense to ensure the accuracy of RWAs.

APRA Gives APRA A Good Wrap

Wayne Byres, APRA Chairman spoke at the Financial Stability Institute (FSI) Executives’ Meeting of East Asia-Pacific Central Banks (EMEAP) – Basel Committee on Banking Supervision (BCBS) High Level Meeting, Sydney in Sydney.

He gave APRA a good wrap for is progress on the Basel based supervisory framework. His comments on their mortgage sector interventions were, well, interesting (I would say, too little too late!).

In 2014, we initiated a quite intensive supervisory effort to lift and reinforce lending standards. Our concern was that due to strong competitive pressures, policies were not suitably calibrated to the Australian environment at the time – one of high and rising house prices, high household debt, subdued household income growth and historically low interest rates. We issued additional supervisory guidance, and allocated significant resources to ensure lending policies were suitably aligned with it. Unfortunately, this provided evidence that strong incentives to grow profit and market share often saw lenders weaken and/or override policies in order to generate sales. Moreover, the dangers did not seem to be strongly called out by compliance and audit functions.

To me, the perspective he paints is still far too narrow in terms of really tacking financial stability, Basel is but one element, not the universe!

Here is his introduction…

Translating prudential policy into prudent practice

I’ve been asked to speak this morning about the challenge of translating prudential policy into prudent practice. From a global perspective, that’s an important issue to focus on because policy reform will mean little if it does not translate into improved practices and behaviours.

Translating global reform proposals to improved banking practice is a four-step process:

  • global reforms have to be agreed as international standards;
  • agreed international standards then need to be translated into domestic regulation;
  • bank policies, systems and frameworks then have to be modified to comply with new regulations; and
  • actual banking behaviours and practices need to adjust to a new set of constraints on the way business is done.

The first of these steps is pretty much complete. A decade on from the financial crisis, the marathon international policy-making efforts to restore and protect the resilience of the global banking system have, by and large, reached the finish line. Greatly strengthened risk-based capital requirements, a supplementary leverage ratio, with additional buffers for systemically important banks, liquidity and funding requirements in the form of the Liquidity Coverage Ratio (LCR) and the Net Stable Funding Ratio (NSFR), tighter large exposure limits and new margining requirements for OTC derivatives, and an enhanced disclosure regime provide a comprehensive package of prudential policy reforms. The Basel Committee has worked long and hard, and should be commended for the final product.

The Basel Committee’s output – besides lots of paper! – are agreements on the design and calibration of minimum standards. Those agreements are critical to global financial stability. Unfortunately, they are not worth the paper they are written on if they are not translated into domestic regulation by member jurisdictions – the second step in the process I referred to earlier. To repeat the frequent exhortations of the G20 leaders, to achieve the desired objective of a resilient global financial system the reforms need to be implemented in a “full, timely and consistent manner”.

A quick look at the Basel Committee’s latest implementation monitoring report shows something of a mixed scorecard. While the core Basel III risk-based capital requirements and the LCR are largely in force around the world, there has been less progress in some other areas: many requirements are subject to lengthy transitional periods, implementation of the NSFR has been disappointingly slow in a number of jurisdictions, and margining and disclosure requirements can be described as a little patchy.

In short, not all commitments to act have been met by action. That is disappointing. While some delays and trade-offs are valid, on occasion it appears to reflect a regulatory version of the “first mover disadvantage” that supervisors often criticise the industry for – jurisdictions not wanting to do the right thing and move promptly because of a concern that other jurisdictions may not follow suit. This reveals a disappointing penchant to put the interests of banks and their shareholders above that of their depositors and the broader community – something that prudential supervisors must constantly guard against.

For our part, we’ve made good progress on the financial reform agenda in Australia. While the banking sector here escaped the worst of the crisis a decade ago, that didn’t mean APRA was complacent about the importance of building resilience. The core capital, liquidity and funding reforms of Basel III are all in place, with a conservative overlay in many areas. Moreover, we have done this without the extensive transitional periods that have been necessary in other parts of the world. I acknowledge, though, that we don’t have a perfect record and still have a few gaps to close, such as the enhanced Pillar III disclosure requirements. We are committed to addressing these as soon as we can.

China implements Basel Committee framework for controlling large exposures, curtailing bank risk

From Moody’s.

Last Friday, the China Banking Regulatory Commission published for public comment a draft regulation of commercial banks’ large exposure management in accordance with the Basel Committee on Banking Supervision’s framework. The draft regulation is credit positive for banks because it will quantifiably curtail the shadow-banking practice of investing in structured products without risk-managing the underlying exposures and will limit the concentration risk in traditional non-structured loan portfolios.

For the first time, regulators are introducing binding and quantifiable metrics to implement the look-through approach when measuring credit exposures of investments in structured products, as emphasized in a series of recent steps to tighten shadow banking activities.  A bank must aggregate unidentified counterparty risk in a structured investment’s underlying assets as if the credit exposures relate to a single counterparty (i.e., the unknown client) and reduce that aggregate exposure to below 15% of the bank’s Tier 1 capital by the end of 2018. Any investment in structured products above the capped amount must identify counterparty risks associated with underlying assets so that investing banks can manage the risks accordingly.

The draft regulation’s measure of the unknown client limit will discipline banks’ implementation of the look-through approach to their investment portfolio to address opaque bank investment categories such as “investment in loans and receivables” that have been originated by other financial institutions. For the 16 listed banks that we rate, which account for more than 70% of the country’s commercial banking-sector assets, total investment in loans and receivables was slightly more than 100% of Tier 1 capital as of 30 June 2017. This implies a forced look-through approach will be applied to more than 85% of this segment of banks’ investment portfolios (see exhibit).

For the concentration risk in traditional non-structured loan portfolios, the draft regulation reiterates the current rule limiting a bank’s loans to a single customer to 10% of the bank’s Tier 1 capital, and extends the limit to include non-loan credit exposure to a single customer at 15% of Tier 1 capital. For a group of connected customers, either through corporate governance or through economic dependence, the draft regulation caps a bank’s total credit exposure at 20% of Tier 1 capital.

For a group of connected financial-institution counterparties, the draft regulation caps a bank’s total credit exposure at 100% of Tier 1 capital by 30 June 2019 and steadily lowers the cap to 25% by the end of 2021. In the case of credit exposures between global systemically important banks (G-SIBs), the cap is 15% of Tier 1 capital within a year of the bank’s designation as a G-SIB.

Bank Stress Testing Is Maturing But Diverse

The Basel Committee’s Working Group on Stress Testing has published a 66 page report “Supervisory and bank stress testing: range of practices“. APRA is mentioned several times through the report, most notably about the limited disclosure of results here, compared with some other countries. Also the scope and purpose of these tests vary considerably, and the extensions into macroprudential differs.  So the approach, and outputs of stress testing are very different.

The report sets out a range of observed supervisory and bank stress testing practices with the aim of describing and comparing these practices and highlighting areas of evolution. The level of data reported on supervisory stress tests reflects the differing objectives and areas of focus across supervisors.

It draws on the results of two surveys completed during 2016: (i) a survey completed by Basel Committee member authorities (banking supervisors and central banks), which had participation of 31 authorities from 23 countries; and (ii) a survey completed by 54 respondent banks from across 24 countries, including 20 global systemically important banks (G-SIBs). Case studies, and other supervisory findings.

There are two fundamental types of supervisory stress tests: (1) those in which the supervisors collect data from the firms and then use their own models and scenarios to assess the performance of the firms under stress (referred to in this report as either “supervisor-run” or “top-down” tests), and (2) those in which the supervisors issue scenarios and guidance to the firms, which then run their own models and report the results to the supervisor (the “institution-run” or “bottom-up” tests).

A number of authorities now use stress tests specifically for capital adequacy assessment or to inform macroprudential policies such as the countercyclical capital buffer. In some cases, stress testing frameworks aim to address multiple objectives. For example, in Canada the macroprudential stress test outcomes inform ongoing supervisory work, such as capital adequacy assessments and risk identification, prioritisation, and measurement, as well as financial stability policy initiatives. Stress testing can also be used to facilitate communication with relevant domestic and foreign parties regarding the stability of the financial system.

In the US, supervisory estimates of post-stress capital ratios for each bank under adverse and severely adverse economic and financial conditions are publicly released along with detailed information on losses and revenues. In the EU, for all regular EU-wide stress test exercises hitherto, the EBA Board of Supervisors decided to publish the quantitative results at the bank-level and provide comprehensive granular data for several types of portfolios on a regular basis (eg sovereign portfolios and risk weights from internal models).

In contrast, for a number of countries, only aggregate-level results are published; often this is in the context of an FSAP review. For example, the Australian Prudential Regulation Authority (APRA) has traditionally disclosed only the aggregate results.

APRA do not impose capital requirements directly based on stress test results, but see stress testing as a strong tool to inform supervisors’ judgments of capital adequacy. Stress testing is also expected to be reflected in banks’ capital decisions, helping banks to set target surplus thresholds and fostering greater understanding of the dynamics between capital and risk.

APRA has used stress tests results to evaluate the adequacy of banks’ recovery planning. In particular, a bank’s management actions in a stress test scenario should be consistent with and linked closely to a bank’s recovery plan in order to be credible.

In 2014 and again in 2017, the Australian and New Zealand supervisory authorities completed a coordinated banking industry stress test. Although both countries had worked together for previous banking industry stress tests, the level of engagement increased in 2014 with close coordination and collaboration on scenario design, templates, analysis and outcomes.

There is a strong link between the banking sectors in Australia and New Zealand; the four major Australian banks have significant exposures to New Zealand and their subsidiaries dominate the New Zealand banking system.

APRA was responsible for the overall coordination and execution of the stress test. There was a common scenario and set of reporting templates covering Australia and New Zealand. Timing of all stages was closely coordinated. The Reserve Bank of New Zealand (RBNZ) provided challenge and input into the scenario and determined the specific economic parameters for New Zealand, which focused on additional agricultural risks. The RBNZ was responsible for analysing the results for New Zealand banks. Each authority engaged directly with the banks within their jurisdiction on queries and feedback throughout the process.

Ongoing engagement and communication was critical to the success of the exercise. Particular consideration was given to issues that differed between the two jurisdictions. At the highest level this involved ensuring that the economic parameters between Australia and New Zealand were realistic and consistent in a stressed environment. For example, there was discussion and challenge as to the relationship between interest rates in Australia and a corresponding level for interest rates in New Zealand.

Here is a summary of their overall observations:

  • In recent years, there has been significant advancement and evolution in stress testing methodologies and infrastructure at both banks and authorities.
  • Supervisory authorities and central banks continue to devote more resources to enhance the stress testing of regulated institutions, with most supervisory stress testing exercises being carried out on at least an annual basis. This is resulting in significant progress in how the exercises are performed and how they are incorporated into the banking supervision process.
  • Banks have been making improvements to their governance structures, with banks’ boards, or delegated committees of boards, taking active roles in reviewing and challenging the results of stress tests, in addition to providing oversight of the overall framework.
  • Banks are increasingly looking to leverage the resources dedicated to stress testing frameworks to inform the risk management and strategic planning of the bank. Stress testing frameworks are increasingly integrated into business as usual processes.
  • Key challenges that remain for banks include finding and maintaining sufficient resources to run stress testing frameworks, and improving data quality, data granularity and the systems needed to efficiently aggregate data from across the banking group for use in stress tests. For national authorities, greater coordination of stress testing activities across authorities is needed, eg via the exchange information on stress test plans and results through supervisory colleges.

Microprudential use of supervisory stress tests

  • Supervisory stress test results are primarily used by supervisory authorities for reviewing and validating the Internal Capital Adequacy Assessment Process (ICAAP) of banks and their liquidity adequacy assessments.
  • Since the global financial crisis, an increasing number of countries assess capital ratio levels under adverse scenarios and use the resulting assessment for evaluating capital adequacy or required capital. However, a wide variety of practices exist.
  • Certain supervisor set capital add-ons for banks by using rules-based methods based solely on stress tests (eg by benchmarking against formal hurdle rates). Those authorities which require add-ons based on stress tests mostly employ a combination of formal process and supervisory judgment. A few jurisdictions have a more rules-based treatment.
  • It is less common for supervisors to use outputs from stress tests for other purposes than those related to capital/liquidity assessments. Nevertheless, some supervisors review and challenge banks’ business plans or banks’ recovery plans on the basis of stress test findings.

Macroprudential use of supervisory stress tests

  • Stress tests are increasingly used to calibrate macroprudential measures and supervisory policy changes. Other macroprudential uses are early warning exercises to identify potential weaknesses of the system and enhance crisis management plans.
  • Macroprudential stress tests are increasing in importance as a way of assessing the financial resilience of banking systems, as they can allow for a more direct assessment of feedback loops, amplification mechanisms and spillovers. These important effects are most frequently assessed via top down approaches.

 

Basel Committee’s low risk weight for covered bonds is credit positive for issuing banks

From Moody’s.

Last Thursday, the Basel Committee on Banking Supervision accredited covered bonds with low risk weights, closely following a precedent set by European regulation. A low risk weight is credit positive for issuing banks’ sales of covered bonds outside the European Union (EU) given the global application of the Basel rules.

In Australia, covered bonds funded about 10% of all outstanding mortgages in 2016, up from 6% five years earlier.  More will be funded this way once the new rules are in place.

Existing EU covered bond issuers will benefit from the Basel IV regulation because their covered bonds become a more attractive investment for non-EU bank investors. Amid rising minimum requirements for regulatory capital, risk weights applied in the calculation of banks’ stock of risk-weighted assets have gained importance in their investment decisions. European covered bond issuers can diversify their investor base and potentially reduce their funding costs as demand for their bonds increases. However, some issuers may be incentivised to increase their share of covered bond issuance in foreign currencies, thereby increasing their exposure to foreign-currency fluctuations given that cover pool assets typically are denominated in euros or other local European currencies.

Outside the EU, lower risk weights for covered bonds will foster the development of covered bond markets and encourage the bonds’ use as a funding tool. The additional funding source will make non-EU banks less reliant on deposits and the sometimes volatile unsecured wholesale funding market. Additionally, the covered bonds will provide an opportunity to improve their asset-liability matching, particularly for mortgages, which typically have 20- to 30-year maturities, versus five to 10 years for covered bonds. In the EU, banks investing to fulfil liquidity coverage requirements, for example, typically absorb about one third of primary covered bond market issuance.

Covered bond markets outside the EU include Australia, Canada, New Zealand, and Singapore, where the bonds have had a less relevant, but growing, role in financing local mortgage markets. In Australia, covered bonds funded about 10% of all outstanding mortgages in 2016, up from 6% five years earlier, but significantly less than in Sweden, where covered bonds finance about 55% of all outstanding residential loans, according to the European Covered Bond Council’s HypoStat 2017. Once Basel IV rules are implemented, we expect that non-EU banks will become more active investors in their domestic covered bond markets, thereby facilitating domestic mortgage funding.

Lower risk weights reflect covered bonds’ status as the only bank debt that cannot be bailed-in and that has a proven track record of sound credit and liquidity. Basel IV regulation stipulates certain requirements that issuers must fulfil to achieve low risk weights for their covered bonds beginning January 2022. The Basel IV requirements include that the covered bond issuer be subject by law to special public supervision designed to protect bondholders, that the value of the cover pool backing the covered bonds is restricted to a maximum loan-to-property value ratio of 80%, and that the covered bonds are protected by at least 10% over-collateralisation.

2017 G-SIB Banks Revealed

The Financial Stability Board (FSB), in consultation with Basel Committee on Banking Supervision (BCBS) and national authorities, has identified the 2017 list of global systemically important banks (G-SIBs), using end-2016 data and the updated assessment methodology published by the BCBS in July 2013. One bank has been added to and one bank has been removed from the list of G-SIBs that were identified in 2016, and therefore the overall number of G-SIBs remains 30.

FSB member authorities apply the following requirements to G-SIBs:

Higher capital buffer: Since the November 2012 update, the G-SIBs have been allocated to buckets corresponding to higher capital buffers that they are required to hold by national authorities in accordance with international standards. Higher capital buffer requirements began to be phased in from 1 January 2016 for G-SIBs (based on the November 2014 assessment) with full implementation by 1 January 2019. The capital buffer requirements for the G-SIBs identified in the annual update each November will apply to them as from January fourteen months later. The assignment of G-SIBs to the buckets, in the list published today, determines the higher capital buffer requirements that will apply to each G-SIB from 1 January 2019.

Total Loss-Absorbing Capacity (TLAC): G-SIBs are required to meet the TLAC standard, alongside the regulatory capital requirements set out in the Basel III framework. The TLAC standard will be phased-in from 1 January 2019 for G-SIBs identified in the 2015 list (provided that they continue to be designated as G-SIBs thereafter).

Resolvability: These include group-wide resolution planning and regular resolvability assessments. The resolvability of each G-SIB is also reviewed in a high-level FSB Resolvability Assessment Process (RAP) by senior regulators within the firms’ Crisis Management Groups.

Higher supervisory expectations: These include supervisory expectations for risk management functions, risk data aggregation capabilities, risk governance and internal controls.

9  Compared with the list of G-SIBs published in 2016, the number of banks identified as G-SIBs remains the same. One bank (Royal Bank of Canada) has been added to and one bank (Groupe BPCE) has been removed from the list. Two banks moved to a higher bucket: both Bank of China and China Construction Bank moved from bucket 1 to 2. Three banks moved to a lower bucket: Citigroup moved from bucket 4 to 3, BNP Paribas moved from bucket 3 to 2, and Credit Suisse moved from bucket 2 to 1.

10  The bucket approach is defined in Table 2 of the Basel Committee document Global systemically important banks: updated assessment methodology and the higher loss absorbency requirement, July 2013. The numbers in parentheses are the required level of additional common equity loss absorbency as a percentage of risk-weighted assets that each G-SIB will be required to hold in 2019.

A new list of G-SIBs will next be published in November 2018