Response to Submissions — Loss-absorbing Capacity

To: All Authorised Deposit-taking Institutions

On 8 November 2018, the Australian Prudential Regulation Authority (APRA) released a discussion paper, Increasing the loss-absorbing capacity of ADIs to support orderly resolution, for consultation. The discussion paper proposed changes to the application of the capital adequacy framework for authorised deposit-taking institutions (ADIs) designed to help facilitate orderly resolution in the unlikely event of failure.  

The global financial crisis demonstrated the widespread impact that financial institution failures can have on the broader financial system, and the disruption these failures can cause to the economy. As a result of a lack of residual financial resources at failing financial institutions, governments in some foreign jurisdictions had no alternative to using public funds to recapitalise those institutions. While government recapitalisation can be an important tool where needed, it shifts the financial burden onto taxpayers. To reduce the need for public funds when financial institutions fail, APRA has sought to bolster the amount of private funds available to support an orderly resolution.

The Australian Government’s 2014 Financial System Inquiry (FSI) recommended that APRA implement a framework for minimum loss-absorbing and recapitalisation capacity in line with emerging international practice, sufficient to facilitate the orderly resolution of Australian ADIs and minimise taxpayer support (Recommendation 3). APRA, as resolution authority, aimed to design an approach that supports its objectives in resolution, and also takes account of the particular characteristics of the Australian financial system. The Council of Financial Regulators (CFR) was consulted and were supportive of the approach.¹

The discussion paper outlined proposed changes to the application of Total Capital requirements designed to support orderly resolution. The proposed changes would increase Total Capital requirements by four to five percentage points of risk-weighted assets (RWA) for domestic systemically important banks (D-SIBs), to be implemented over four years.² For other ADIs, the need for additional loss absorbency would be considered as part of resolution planning. APRA anticipated that ADIs would primarily satisfy the increased Total Capital requirements by issuing additional Tier 2 capital.

APRA received seven submissions and met with a range of relevant stakeholders including ADIs, rating agencies and other market participants, and continued to consult with the CFR. Non-confidential submissions are available on APRA’s website.

Consultation feedback

The key issue raised during the consultation period was whether there would be sufficient capacity in debt markets to absorb the anticipated additional Tier 2 capital issuance. In particular, a number of submissions argued that the proposal may:

• be unachievable due to insufficient market capacity for Tier 2 capital instruments;
• increase risks for ADIs if Tier 2 capital cannot be issued on a consistent basis, particularly during periods of poor market conditions; and
• excessively increase funding costs for certain ADIs.

Feedback suggested there are circumstances in which the market capacity for Tier 2 capital would not be sufficient to absorb an increase in issuance equivalent to four to five percentage points of RWA. However, market capacity estimates provided by a number of stakeholders indicated that a significant proportion of the increase in requirements could be met through additional Tier 2 capital issuance, even after allowing for periods of unfavourable market conditions. Estimates of the market capacity for the D-SIBs’ Tier 2 capital ranged from $75 billion to $125 billion; currently around $40 billion of the D-SIBs’ Tier 2 capital is on issue.

A number of stakeholders advocated the introduction of a new instrument that would rank ahead of regulatory capital. They suggested that a new instrument would attract greater investor demand than regulatory capital instruments and that the anticipated issuance to meet loss-absorbing capacity requirements would not be bound by market capacity constraints.

Different potential structures for an alternative instrument were proposed. The most significant difference was whether the features of the new instrument would rely on contractual terms or legislation (i.e. statutory bail-in). However, as APRA does not have a statutory bail-in power, a proposal introducing a new instrument would, like Tier 2 capital instruments, require a loss absorption mechanism that relied on contractual terms.  

Some submissions suggested a new contractual instrument would be favourable for ADIs in terms of the overall cost of funding relative to the APRA proposal. The basis for this was that the new instrument would include certain features different to Tier 2 capital instruments, such as a later trigger point, not being subordinated and other terms that may attract a larger pool of investors. However, these features would make such instruments less consistent with APRA achieving its resolution objectives. Removing the relevant features is likely to reduce the stated market capacity benefits significantly.

APRA expects that instruments designed to support orderly resolution should include features that ensure losses are absorbed in a timely manner, limit the risk of ex-post compensation claims and have no impediments to loss absorption. Tier 2 capital instruments fulfil these important criteria for loss-absorbing capacity, whereas suggested alternative instruments are less satisfactory.

Further information on the key issues raised and APRA’s response is provided in Attachment A.

Increasing Total Capital to support orderly resolution

APRA has determined that the proposal outlined in the discussion paper, with some amendments, is the best course of action to support orderly resolution – requiring ADIs to maintain additional Total Capital.³ However, in recognition of the concerns raised around market capacity for Tier 2 capital, APRA will set the increase in Total Capital requirements at three percentage points of RWA, instead of four to five percentage points.⁴ D-SIBs will be expected to build Total Capital ratios ahead of complying with the higher requirement by 1 January 2024.

APRA expects that setting a requirement of an additional three percentage points of Total Capital is likely to result in ADIs increasing the issuance of existing forms of capital. However, this issuance would be less than was anticipated under the proposal outlined in the discussion paper, which should reduce the risk of unintended consequences or inefficiencies, particularly if periods of disruptive market conditions prevail during the implementation period.

If the D-SIBs meet the three percentage point increase in Total Capital requirements by issuing additional Tier 2 capital, the amount on issue is likely to increase by around $50 billion. Analysis conducted by the Reserve Bank of Australia (RBA) supports the view that globally the pricing for Tier 2 securities is closely related to that for bank senior unsecured bonds. This has meant that, in the past, issuance of Tier 2 securities into existing markets had little impact on their price. In the discussion paper, APRA estimated that the impact on total funding costs from the proposal would not be greater than five basis points. This estimate is well within the range suggested when the RBA estimates of the price impact are applied, which are based on historical market data.

APRA’s overall targeted calibration of an additional four to five percentage points of loss-absorbing capacity remains unchanged. Over the next four years, feasible alternative methods for raising the additional one to two percentage points will be considered in consultation with industry and other interested stakeholders.

Increasing Total Capital requirements for D-SIBs by three percentage points represents a meaningful step towards ensuring sufficient financial resources are available to support the orderly resolution of these ADIs, in line with international peers. This approach aligns with APRA’s resolution objectives, accounting for the distinctive features of the Australian financial system, the prevailing legal framework and institutional settings.  

 
Yours sincerely,  

John Lonsdale 
Deputy Chair

Footnotes:

¹ The CFR membership comprises the Reserve Bank of Australia (RBA) as chair, APRA, the Australian Securities and Investments Commission (ASIC), and the Treasury.

² The full increase was intended to apply to ADIs where the most appropriate strategy involves a whole-of-bank stabilisation and recapitalisation, such as the D-SIBs. For other ADIs, the adjustment would be set based on the outcome of resolution planning and would be no greater than that applied to D-SIBs; it is expected that an increase in requirements would only apply to a small number of ADIs, due to their complexity or the nature of their functions.

³ Paragraph 23 of Prudential Standard APS 110 Capital Adequacy allows APRA to determine higher prudential capital requirements (PCRs) for an ADI than the minimum PCRs.

⁴ APRA is consulting on revisions to the capital framework that may result in changes to the calculation of RWA or the measurement of capital. If adjustments are made APRA intends for the absolute amount of additional loss-absorbing capacity to remain the same as that calculated under the current capital framework. 

Attachment A — Response to Submissions 

Tier 2 market capacity

A number of submissions raised concerns about the capacity of the market to absorb the anticipated additional Tier 2 capital issuance. Some submissions argued that higher yields on Tier 2 capital instruments alone would not be sufficient to allow ADIs to meet the necessary increase in Total Capital requirements through the issuance of Tier 2 capital, and that market capacity constraints could be compounded by the Tier 2 market becoming unavailable during periods of disruptive financial market conditions.  

APRA received a number of market capacity estimates for the D-SIBs’ Tier 2 capital, ranging from $75 billion to $125 billion. These estimates generally indicated that the increase in requirements proposed in the discussion paper could be achieved by issuing Tier 2 capital during supportive market conditions; however, the amount to be raised was at the upper end of these estimates. Market capacity estimates that accounted for periods of unfavourable market conditions suggested the additional requirement could not be meet fully through additional Tier 2 capital issuance. There was a view that if there was a substantial increase in the issuance of Tier 2 capital instruments, due to the increase in Total Capital requirements, there would be a material price impact.  

To address this, APRA has decided to increase Total Capital requirements for D-SIBs by three percentage points, instead of four to five percentage points. D-SIBs will have over four years to meet the increased requirement. APRA estimates that the amount of D-SIB Tier 2 capital on issue will increase from around $40 billion currently to around $90 billion to meet increased Total Capital requirements.

Costs

Several stakeholders suggested the APRA proposal would increase overall funding costs by more than five basis points, exceeding the estimate in the discussion paper.

In the discussion paper, APRA acknowledged that overall funding costs would be affected by increasing Total Capital requirements, due to the higher cost of Tier 2 capital instruments compared to other forms of funding. APRA’s cost estimate relied on observed market pricing of Tier 2 capital instruments relative to senior unsecured debt. It was estimated that if Total Capital requirements were increased by five percentage points, spreads on Tier 2 capital instruments could widen by up to 80 basis points without overall funding costs exceeding 5 basis points in aggregate.⁵ Other stakeholders have noted that the impact on funding costs will depend on the pricing interaction of Tier 2 capital instruments and senior unsecured debt.

After considering submissions APRA still expects the overall funding cost impact from increasing Total Capital requirements to not exceed five basis points, as outlined in the discussion paper, particularly given the lower increase of three percentage points to Total Capital requirements; the RBA’s analysis aligns with these estimates.   

Eligible instruments

Some submissions suggested a new instrument be introduced, with similar features to instruments in other jurisdictions designed to meet the Financial Stability Board’s (FSB) Total Loss-Absorbing Capacity (TLAC) standard. Internationally, such instruments are generally subject to statutory powers to impose losses in resolution and are subordinated to liabilities needed to maintain critical functions. These submissions suggested that a new instrument would attract a larger pool of investors than Tier 2 capital instruments. Options proposed by stakeholders included adopting a statutory bail-in approach or introducing an instrument with contractual clauses to affect loss absorption.

APRA does not have a statutory power to write-off or convert the interests of creditors. Expanding APRA’s powers to allow for a statutory bail-in approach would require legislative change, which is ultimately a consideration for government and would need to go through the parliamentary process to be implemented. Pursuing this approach would raise significant policy trade-offs for government to consider, including determining the scope of such a power, the associated effect on credit ratings, impact on funding costs and ultimately the economy.  

In relation to a new contractual instrument, APRA acknowledges that this may be attractive to certain investors and could provide benefits in terms of market capacity and overall costs. However, the success of any additional loss-absorbing capacity lies in its role in facilitating an orderly resolution by APRA. There are features of the proposed new contractual instrument that are inconsistent with ensuring APRA’s objectives in resolution can be met. These features include:

• Trigger point – a later trigger point than regulatory capital, i.e. after the point of non-viability.  

  APRA does not consider there would likely be any difference between the point of non-viability and resolution. A non-viability trigger event is likely to present the final opportunity to stabilise the critical functions of a distressed ADI in an orderly manner. As such, APRA requires instruments designed to absorb losses to do so at this point and no later, well before the ADI is at risk of becoming insolvent;
   

• Subordination – no subordination to senior creditors.  

  APRA considers subordination to be a pre-requisite for loss-absorbing capacity. Subordinating the additional loss-absorbing capacity to liabilities needed to maintain the ongoing function of the ADI supports the continuity of services and limits the possibility of successful litigation against actions taken by authorities in resolution or the risk of ex-post compensation claims;⁶ and
   

• Certainty – the absence of the ability to write-off the instrument and no provision that the instrument must absorb losses before the use of public funds.

  APRA considers these to be important features for ensuring losses can be absorbed in resolution and limiting the risk of operational complexities frustrating an orderly process. Including the option to write-off instruments designed to absorb losses and terms that ensure instruments absorb losses ahead of the use of public funds are necessary features in achieving APRA’s resolution objectives.

Given the above, designing a new contractual instrument in a way that both meets APRA’s objectives and appeals to investors, sufficient to provide market capacity benefits, is unlikely to be possible. For these reasons, subject to taking into account concerns on capacity as noted above, APRA believes that increasing the issuance of existing forms of capital – rather than creating new contractual instruments – is the most effective means of increasing loss-absorbing capacity.  

Calibration

Some respondents indicated that APRA’s proposed approach of increasing loss-absorbing capacity by an overall targeted calibration of four to five percentage points was appropriate, providing analysis that the proposed increase was broadly in line with other jurisdictions. Some acknowledged and supported APRA accounting for the differences in the definitions of capital and calculations of RWA when determining calibration. In contrast, one submission queried the methods used to inform the calibration, suggesting that APRA’s estimate exceeded the level needed to facilitate orderly resolution and was above internationally comparable ratios.

APRA notes that while the calibration was determined with reference to developments in other jurisdictions, the final amount is an assessment of the financial resources needed to support orderly resolution. APRA does not see a case to adjust the overall targeted calibration from the proposed four to five percentage points of RWA outlined in the discussion paper.

Figure 1 illustrates the increase in Total Capital requirements by 1 January 2024. Setting a requirement to increase Total Capital by three percentage points does not lower APRA’s overall targeted calibration for loss-absorbing capacity of an additional four to five per cent of RWA. It ensures that significant progress can be made in increasing loss-absorbing capacity over the next four years, while the most feasible alternative method for raising the additional one to two percentage points of loss-absorbing capacity is further considered by APRA. 

Figure 1 - Changes to the D-SIBs' capital structures

Footnotes:

⁵ This analysis did not consider the potential for a lower cost of senior unsecured debt due to reduced issuance and greater creditor protection as a result of increased subordinated issuance. 

⁶ Some submissions suggested that features of the Canadian approach to loss-absorbing capacity could be applicable to an Australian contractual approach. However, the Canadian approach is based on a statutory bail-in power. Although it is the case that the Canadian authorities have not required eligible instruments to be subordinated to operational liabilities, APRA notes that this is out of line with approaches taken in other jurisdictions, which have required subordination in accordance with the principles of the FSB TLAC standard.