Response to submissions received on proposed implementation of Basel III capital adequacy requirements in New Zealand.

Transcription

1 Response to submissions received on proposed implementation of Basel III capital adequacy requirements in New Zealand. September 2012 This document sets out the to the main issues raised in submissions received on the November 2011 and March 2012 consultations. This document is not a consultation document.

2 2 Contents Introduction... 3 Capital ratios... 4 Common Equity Tier 1 Capital... 5 Additional Tier 1 Capital... 6 Tier 2 capital Capital issued out of consolidated subsidiaries that is held by third parties Regulatory deductions Loss absorbency at the point of non-viability Conservation buffer Countercyclical buffer Leverage ratio Other issues Transitional arrangements... 23

3 3 Introduction 1. In November 2011, the Reserve Bank issued a consultation paper on the implementation of Basel III in New Zealand. This consultation paper set out the Reserve Bank s proposals for implementing core Basel III capital measures relating to capital ratios, the definition of capital and the leverage ratio. The November consultation paper also noted that the Reserve Bank would consult on other elements of Basel III capital in In March 2012 the Reserve Bank issued a consultation paper on further elements of Basel III capital adequacy requirements in New Zealand. This paper set out the Reserve Bank s proposals for further elements of the framework: the operation of the conservation buffer; the countercyclical buffer; and minimum requirements to ensure that all classes of capital instrument fully absorb losses at the point of non-viability ( loss absorbency ). The paper also proposed transitional arrangements. 3. In May 2012 decisions were taken on some key elements of Basel III. These decisions were supported by the Reserve Bank s cost-benefit analysis, and took into account submissions received on the Reserve Bank s consultations and developments in the Basel III proposals of the Australian Prudential Regulation Authority (APRA). The Reserve Bank sent all locally-incorporated banks a letter providing an update on Basel III implementation in New Zealand, including details of the May 2012 decisions. 4. This feedback statement contains a summary of the main issues raised in response to the November 2011 and the March 2012 consultations, and the to these issues (including those responses already reported in May). Some minor or technical issues raised by just one or two submitters are not discussed in this paper.

4 4 Capital ratios 5. The Reserve Bank proposed to adopt the Basel III ratios as follows: Capital ratio requirements and buffers (as a percentage of risk weighted assets) Common equity Tier 1 capital Total capital Existing minimum ratios - 4.0% 8.0% New minimum ratios 4.5% 6.0% 8.0% Conservation buffer 2.5% New minimum ratio plus conservation buffer 7.0% 8.5% 10.5% Countercyclical buffer range 0-2.5% 1 6. Most submissions agreed with our proposal to adopt the Basel III minimum ratios and the conservation buffer. However a small number questioned whether the new ratios were necessary, given for instance the lack of bank failures in New Zealand. 7. Some submissions expressed concerns about the countercyclical buffer. These submissions and the are set out in a separate section of this paper on the countercyclical buffer. to submissions on the new minimum ratios and the conservation buffer 8. The Reserve Bank does not accept that a lack of recent bank failures in New Zealand is a sufficient reason not to adopt the Basel III ratios. Our cost benefit analysis took into account the probability of a bank failure, which although is very low, can result in substantial costs within the banking sector and the wider economy. As noted in our consultation paper, our cost benefit analysis suggests that tightening our existing requirement to the Basel III standards is easily justified. 1 While the countercyclical buffer is indicatively expected to vary between 0 and 2.5% of risk-weighted assets, no formal limit will be set on the maximum size of the buffer (i.e. the size of the buffer will be determined according to circumstances). See also discussion on the maximum size of the countercyclical buffer below.

5 5 Common Equity Tier 1 Capital Accumulated other comprehensive income and other disclosed reserves 9. One submission of substance was received on this issue. The submission did not support the proposal to retain the current treatment of revaluation reserves and foreign currency translation reserves as tier 2 capital. It was noted that the Basel III framework recognises these items in the definition of common equity as they are available to absorb losses on a going concern basis. Moreover as investors and rating agencies are likely to focus on common equity tier 1 (CET1) ratios the proposed treatment will disadvantage New Zealand banks. 10. The Reserve Bank accepts the proposed departure from the Basel III framework may have a noticeable (but minor) impact on the CET1 ratio of some banks. However one of our principles in implementing Basel III has been not to adopt Basel III measures when doing so would reduce the conservativeness of our standards. We do not consider there is a compelling case to take a different approach on the basis of the submission received. Supervisory approval for repayment of CET1 principal 11. Several submissions argued against the proposal to require supervisory approval for repayment of CET1 principal. The arguments presented included: that there are already sufficient protections and restrictions on the redemption of capital; the proposed requirement would be costly to administer; it would not be valued by the market; and it would be ineffective (i.e. could be gamed). 12. The Reserve Bank accepts that this requirement may not be effective. Also this requirement would be a departure from the Basel III framework and there does not seem to be a compelling New Zealand specific case to adopt it. The Reserve Bank therefore does not intend to require supervisory approval for repayment of CET1 principal.

6 6 Additional Tier 1 Capital Call-ability of Additional Tier 1 (AT1) instruments 13. Several submissions argued that the Reserve Bank should permit AT1 instruments to be called after five years, consistent with the Basel III framework. The main arguments in support of this view were: Non-common equity tier 1 instruments issued to date have included a call element by virtue of the call right being provided to an entity outside of, or upstream of, the regulated banking group. Allowing a call right would align with APRA s proposals. A call right would provide value to the investor and deliver lower pricing for the issuing bank. The risk that the call date becomes the effective maturity date can be managed through the supervisory approval process. 14. The Reserve Bank first advised banks of its response to this issue in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand. 15. The Reserve Bank s original concern had been that allowing a call right for AT1 instruments would represent a lowering of its capital adequacy standards. However, taking into account submissions the Reserve Bank now intends to allow AT1 instruments to be callable at the initiative of the bank after a minimum of five years, subject to the conditions set out in the Basel III framework as follows: i) To exercise a call option a bank must receive prior supervisory approval. ii) A bank must not do anything which creates an expectation that the call will be exercised. iii) Banks must not exercise a call unless: a. they replace the called instrument with capital of the same or better quality and the replacement of this capital is done at conditions which are sustainable for the income capacity of the bank; or b. the bank demonstrates that its capital position is well above the minimum capital requirement after the call option is exercised. 16. In coming to this decision we took into account that the conditions set out above are tighter than our current requirements and once adopted will provide sufficient prudential safeguards in respect of callable instruments.

7 7 17. As part of the supervisory approval process, the Reserve Bank intends to require that a bank cannot finalise its call of an instrument until the replacement instrument (if required) has been paid in. Perpetual non-cumulative fully-paid up preference shares 18. Several submissions argued against the Reserve Bank s proposal to restrict AT1 instruments to perpetual non-cumulative fully-paid up preference shares. The main arguments in support of this view were: The restriction could potentially stifle innovation and put NZ banks at a disadvantage. Concerns about the use of alternative instruments can be dealt with by placing a limit on the extent to which they contribute to tier 1 capital as currently envisaged. The restriction is not consistent with APRA s standards. Given the loss absorbency requirements, all hybrid instruments will be of the same quality at the point of default. Therefore it does not make sense to restrict AT1 instruments to perpetual non-cumulative fully-paid up preference shares. 19. The Reserve Bank s original concern had been that allowing additional types of instrument would represent a lowering of its capital adequacy standards. The Reserve Bank now accepts that removing the restriction will not materially lower the quality of regulatory capital providing the capital adequacy standards are clear that dividend/coupon payments are non-cumulative. In coming to this view the Reserve Bank took account of the Basel III loss absorbency and principal loss absorption (see paragraph 22 below) requirements, the various safeguards on the call-ability of AT1 instruments, and the restriction on the extent to which AT1 instruments can contribute to Tier 1 capital. 20. The Reserve Bank therefore does not intend to restrict AT1 instruments to perpetual noncumulative fully-paid up preference shares. Principal loss absorption 21. Within the Basel III framework, the following criterion applies to AT1 instruments: Instruments classified as liabilities for accounting purposes must have principal loss absorption through either (i) conversion to common shares at an objective pre-specified trigger point or (ii) a write-down mechanism which allocates losses to the instrument at a pre-specified trigger point. The write-down will have the following effects: a. reduce the claim of the instrument in liquidation; b. reduce the amount re-paid when a call is exercised; and

8 8 c. partially or fully reduce coupon/dividend payments on the instrument. 22. In the November 2011 consultation paper the Reserve Bank proposed not to adopt these criteria as it did not propose to recognise any instruments that were liabilities as AT1 instruments. However as noted above we now propose to remove the restriction that AT1 instruments can only be perpetual non-cumulative preference shares. As this potentially allows some AT1 instruments that are counted as liabilities, we now propose to adopt the criterion set out in paragraph 22 above. 23. APRA have proposed that the trigger point for this criterion is when the issuer s CET1 capital ratio falls below 5.125% of risk weighted assets. In proposing this trigger point, APRA have noted this is the point at which the full constraint on distributions applies under the capital conservation buffer regime. APRA s approach is consistent with the Basel Committee s guidance that states the trigger level for write-down/conversion must be at least 5.125% of CET We have not identified any reason not to adopt the same trigger point as APRA and so propose to do so. There are detailed implementation issues for this requirement that are similar to, or the same as, issues relating to the loss absorbency at point of non-viability criterion. On these issues we generally plan to take the approach described below in relation to loss absorbency at point of non-viability (see paragraphs below). However some issues relating to the loss absorbency at the point of non-viability criterion are not relevant to this requirement (for example, the trigger point for this requirement does not involve regulatory action and so the powers of the Reserve Bank and statutory manager that are relevant to loss absorbency at the point of non-viability are not relevant here). 25. We will invite comments on the details of this requirement as part of our consultation on proposed changes to the Banking Supervision Handbook. Restriction on contribution of perpetual non-cumulative preference shares without full voting rights to total tier 1 capital. 26. Some submissions did not agree that the Reserve Bank should retain the requirement that perpetual non-cumulative preference shares without full voting rights must not constitute more than 25% of tier 1 capital. It was argued that there will be little incentive to hold excess AT1 capital given the increased amount of common equity required and the increased focus by investors and rating agencies on the CET1 ratio. It was also noted that given the quality of non-common equity tier 1 capital will increase under Basel III the rationale for the restriction may not be valid.

9 9 27. The Reserve Bank accepts the quality of non-common equity tier 1 capital will increase under Basel III, however AT1 instruments will still be of lower quality than common equity. While banks may have little incentive to hold excess AT1 capital, the Reserve Bank does not consider this to be a good reason to remove the restriction. 28. Given the Reserve Bank now intends to remove the restriction on the types of instruments that can be recognised in AT1, we propose to alter the existing restriction on the contribution of perpetual non-cumulative preference shares without full voting rights so that it is a restriction on the contribution of AT1 instruments without full voting rights to total tier 1 capital. Dividend stopper arrangements 29. One submitter sought clarification on whether dividend stoppers that stop payments on other AT1 instruments would be permitted. 30. The Reserve Bank intends to be clear in capital adequacy requirements that dividend stopper arrangements that stop dividend payments on common shares or dividend/coupon payments on other AT1 instruments are not prohibited. This is consistent with guidance provided by the Basel Committee on Banking Supervision. 2 Supervisory approval for calling AT1 instruments Submission received 31. One submitter noted the Basel III requirement that banks should not expect their supervisor to approve a call if the instrument to be called will be replaced with a similar instrument that carries a higher credit spread. This submission argued that if the Reserve Bank intends to adopt a similar requirement, it should be clearly spelt out so investors can consider and price appropriately. 32. The Reserve Bank considers that ordinarily it may not be in the best interests of a bank to call an instrument and replace it with an instrument that is more costly. The Reserve Bank 2 Basel III definition of capital Frequently asked questions BIS, December 2011, p3.

10 10 will therefore not permit such action unless the bank concerned can satisfy the Reserve Bank as to the rationale. Tier 2 capital Supervisory approval for calling tier 2 instruments 33. Some submissions emphasised the importance of the Reserve Bank being reasonably clear ex ante about the circumstances in which supervisory approval would be granted. 34. The Reserve Bank had initially proposed not to adopt the Basel III requirement that banks must not exercise a call unless: a. they replace the called instrument with capital of the same or better quality and the replacement of this capital is done at conditions which are sustainable for the income capacity of the bank; or b. the bank demonstrates that its capital position is well above the minimum capital requirements after the call option is exercised. 35. Our rationale was that this approach should give the Reserve Bank suitable flexibility in forming a view on whether or not to grant supervisory approval for a call option. While submitters were not necessarily arguing for the Reserve Bank to adopt the above requirements, we have concluded that doing so would: provide more clarity to banks; be consistent with our approach for AT1 instruments; and would not limit the matters the Reserve Bank may take into account in determining whether to grant supervisory approval. 36. The Reserve Bank therefore intends to adopt the above restrictions on call options for Tier 2 instruments. 37. Also, as part of the supervisory approval process, the Reserve Bank intends to require that a bank cannot finalise its call of an instrument until the replacement instrument (if required) has been paid in.

11 11 Capital issued out of consolidated subsidiaries that is held by third parties 38. Few submissions were received on this topic. The issues raised were: It is unclear why the Reserve Bank is not recognising certain types of capital issued out of consolidated subsidiaries that is held by third parties (e.g. CET1 minority interests). The Basel III formula for recognising capital issued out of consolidated subsidies is very complex for what is likely to be a relevantly immaterial adjustment in the New Zealand context. It is unclear how the rules for minority interests would interact with the loss absorbency requirements. 39. The Reserve Bank now intends to recognise minority interests and other capital issued out of consolidated subsidiaries consistent with the Basel III framework. This includes ordinary shares issued by consolidated subsidiaries (but only if the subsidiary that issued the shares is itself a bank), and Tier 1 and Tier 2 qualifying capital issued by consolidated subsidiaries. We also intend to require that regulatory instruments issued out of consolidated subsidiaries to third parties are compliant with our loss absorbency rules. 40. While the formula for recognising capital issued out of consolidated subsidies is complex and the amounts on issue may be small at present, we do not consider these factors present a compelling reason for departing from the Basel III framework and taking a less conservative approach. Regulatory deductions Threshold deductions 41. Several submissions argued that the Reserve Bank should exercise the national discretion within the Basel III framework whereby certain items deducted from CET1 receive partial recognition in CET1 rather than being deducted in full. Reasons given were that exercising this discretion would not materially dilute capital quality, and that exercising the discretion would increase international harmonisation and alignment with the Basel III framework.

12 The Reserve Bank does not find the submissions on this topic compelling. The Basel III standard is to require the deductions in full. Allowing limited recognition is a concession. We note APRA are not adopting the concession. The Reserve Bank does not intend to adopt the concession either. Deferred tax assets Submission received 43. One submission argued that the Reserve Bank should allow netting of deferred tax assets and deferred tax liabilities, as permitted by the Basel III framework in certain circumstances. 44. The Reserve Bank intends to allow netting of deferred tax assets and deferred tax liabilities that: relate to taxes levied by the New Zealand Inland Revenue; arise as a result of deductable temporary differences as defined by NZ IAS 12; and may be offset under NZ IAS 12. Loss absorbency at the point of non-viability Write-off versus conversion option 45. The majority of submitters argued that the option to write-off rather than convert capital instruments should be available. The main arguments for this were: conversion is not possible for some banks (e.g. mutually owned banks); New Zealand banks do not have listed equity which makes conversion difficult; conversion into unlisted equity would be difficult to price; it would be undesirable to create minority interests in wholly owned companies; allowing the option of conversion or write-off allows banks to structure capital instruments in the most effective manner for their given capital structure; given APRA do not allow conversion to unlisted equity requiring this in New Zealand would mean capital instruments did not comply with APRAs requirements; the higher risk arising from the possible write-off of tier 2 instruments would be reflected in the price of the instruments. 46. The Reserve Bank first advised banks of its response to this issue in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand.

13 The Reserve Bank accepts many of the points made about the need to allow the option of write-off as well as conversion. Taking into account submissions and developments in APRA s proposals since our consultation, we now intend to allow write-off or conversion. Other write-off / conversion issues 48. Some submitters argued that write-off/conversion should be able to be done on a partial basis. Others argued that capital instruments should be able to be, or be required to be, ranked in order as to which is written-off/converted first (with AT1 being affected first). 49. Some submitters argued that instruments should be able to be converted into equity (or equity like instruments) of the parent, given that there is no listed bank equity in New Zealand. 50. One submitter argued that AT1 and Tier 2 instruments should convert into a class of equity that ranks prior to common equity, so as not to dilute the equity of common shareholders. Others argued that instruments should be able to be written back up if the bank returns to viability, again to reduce the dilution of shareholders. 51. The Reserve Bank considers it is reasonable to allow partial conversion or write-off. The extent to which conversion or write-off is required would be set out in a Reserve Bank direction or determined by the statutory manager. 52. The Reserve Bank has no objection to instruments being ranked in order as to which is written-off or converted first through the contractual terms of instruments. However we do not consider it is necessary to mandate such arrangements. 53. The Reserve Bank accepts that allowing capital instruments to be converted into shares of the parent would avoid the complexity associated with conversion when an entity does not have listed shares. However, to ensure legal certainty about the effectiveness of conversion to parent equity, we plan to require the instrument first be written-off in regard to the obligations of the New Zealand bank. The contract may however specify that an instrument that is written-off, may receive compensation in the form of ordinary shares in the parent entity.

14 Therefore the Reserve Bank will allow conversion into the common equity of the bank (whether listed or not), or write-off of the instrument with compensation provided in the form of common equity of the parent (whether listed or not). 55. The creation of a new class of equity and the write-up feature proposed in submissions do not feature in the Basel III framework, are not being adopted by APRA, and we are unaware of any convincing New Zealand specific reason why we should adopt them. Moreover, the Basel Committee has explicitly ruled out any temporary write-down mechanism as it could result in capital holders having a contingent claim that could rank in preference to any public sector injection. Consequently we do not support the creation of a new class of equity or write-up features. Terms and conditions versus legislation only 56. Only one submitter argued in favour of the option to implement the loss absorbency criteria in legislation only. This was on the basis that write-off is an expropriation of property rights and should therefore only be done through primary legislation. The majority of submitters supported the requirement forming part of the terms and conditions of the relevant instrument. Submitters considered this approach more dynamic, transparent and consistent with existing market practice. 57. The Reserve Bank intends to require that the loss absorbency requirements are included in the terms and conditions of regulatory capital instruments. Discretion powers versus statutory management 58. Submitters favoured using the existing mechanisms in the Reserve Bank of New Zealand Act 1989 (the Act), as these provide checks and balances on the exercise of the discretion. 59. Some submitters argued in favour of using the Reserve Bank s direction powers to trigger loss absorption, others supported using the powers of the statutory manager, and some supported both approaches. The argument in favour of using the direction powers is that it would allow the Reserve Bank to act at an earlier stage without needing to wait until statutory management is declared. The argument in favour of use of statutory management powers is that it would provide a strong signal that loss absorption will only be utilised as a last resort. This would make instruments more marketable and easier to price.

15 The Reserve Bank remains of the view that working within the existing mechanisms of the Act is appropriate. 61. The Reserve Bank agrees that application of the loss absorbency requirement should be available prior to statutory management. This allows conversion or write-off on a goingconcern basis and may make it easier for the bank to re-enter the market and continue trading. However, should a bank be placed into statutory management it is important that the statutory manager has a full suite of powers. This should include the power to activate the loss absorption requirement, particularly as extinguishing the claims of subordinated debt-holders forms part of the proposed approach to implementation of Open Bank Resolution. 62. The Reserve Bank therefore considers the loss absorption requirement should be able to be activated either: issuance of a direction by the Reserve Bank under section 113A(h) of the Act; or exercise by the statutory manager of the power under section 129 of the Act to exercise a contractual right of the bank. Trigger event quantitative versus non- quantitative 63. Submissions were split on whether the trigger point should be quantitative or based on the less precise non-viability criteria, however defined. The argument in favour of a quantitative trigger is certainty and transparency. Those who argued against a quantitative trigger argued: A quantitative trigger may become a de-facto automatic trigger for loss absorption. A quantitative trigger point is likely to be too insensitive because it would need to be set high enough to be guaranteed to be reached before a bank needs restructuring but this level is likely to be too high in a number of cases. It would be difficult to measure when a trigger point had been reached, particularly as capital ratios tend to be a lagging indicator of solvency. The same quantitative trigger is not appropriate for AT1 (going concern capital) and Tier 2 instruments (gone concern capital) or for conversion versus write-off in particular Tier 2 instruments should not be written off until a bank is clearly insolvent.

16 We do not favour a quantitative trigger because: there is no consensus in favour of it; it would be a departure from Basel III and from APRA s approach; and it would reduce the flexibility of the Reserve Bank to act. Implementation of a non-quantitative trigger event 65. The Reserve Bank sought views on whether the trigger event should be whenever the conditions of section 113(1) of the Act (or some subset of these conditions) are met. Submitters argued that some of the criteria in section 113(1) of the Act may allow statutory management to be imposed or a direction given prior to the bank being non-viable. 66. Several submitters argued that the trigger should be restricted to when a direction was issued or statutory management imposed on the basis that section 113(1)(a) or (b) of the Act is met, that is: the registered bank or associated person is insolvent or is likely to become insolvent; the registered bank or associated person is about to suspend payment or is unable to meet its obligations as and when they fall due. 67. Submitters considered that these requirements (section 113(1)(a) and (b)) most closely align with the non-viability criteria articulated by Basel and provide greater certainty to investors, and make pricing and rating easier (and more favourable for banks). Submitters argued that if the trigger is not to be restricted guidance should be issued on how the discretion would be used. 68. The Reserve Bank agrees that some of the criteria in section 113(1) are not relevant to the trigger event, in particular sub-sections 113(1)(f),(g) and (h). These sub-sections relate to failure to comply with a requirement of the Act, conviction for an offence under the Act, and failure to comply with a condition of registration respectively. 69. The other provisions of 113(1) are the grounds on which a bank may be placed into statutory management; being: insolvency, failure to meet financial obligations, conduct or circumstances of the bank being prejudicial to the soundness of the financial system, and imprudent behaviour. We consider it appropriate that loss absorbency may be triggered in any of these circumstances and that it is important that if a bank is placed into statutory management on any of these grounds that a statutory manager may trigger loss absorbency.

17 The Reserve Bank therefore intends that the trigger for write-off or conversion will be on the grounds that any one of the criteria in section 113(1)(a)-(e) of the Act are met. We plan to provide further context on when a loss absorbency direction may be issued on such grounds through guidance in the Banking Supervision Handbook. Application of loss absorbency requirements to AT1 and Tier 2 instruments 71. AT1 instruments: one submitter argued that there was no benefit from writingoff/converting AT1 instruments given that the only AT1 instrument we allow is preference shares which do not form a liability of the bank. 72. Tier 2 instruments: one submitter argued there should be no loss absorbency requirement for Tier 2 instruments. While the Australian owned banks would still issue tier 2 instruments with loss absorbency features (to met APRA requirements), exempting Tier 2 instruments would provide a small concession to the much smaller NZ owned institutions who are likely to have difficulty raising loss absorbent Tier 2 instruments. 73. AT1 instruments: for instruments classified as equity in the bank s balance sheet, conversion into common shares would not increase the net assets of the bank. However conversion may encourage the parent (if one exists) to issue new equity to the subsidiary to avoid dilution of the existing shareholders that could result from conversion. Also conversion ensures that should there be a Crown capital injection, the AT1 instruments are not preferred to Crown capital. Finally the Reserve Bank intends to remove the restriction that only non-cumulative perpetual preference shares can qualify at AT1 instruments and so the assumption that all AT1 instruments will not be liabilities may not be valid. 74. Tier 2 instruments: The Reserve Bank does not consider the argument provided justify a departure from the Basel III framework. Conversion ratio 75. One submitter supported the approach of not requiring a capped conversion ratio, but argued that the New Zealand rules should not prevent such caps given that APRA rules provide for a cap on the conversion ratio for instruments issued by NZ subsidiaries of Australian parents.

18 Another submitter argued requiring conversion to be based on market price at conversion would provide strong incentives for the existing shareholders to put more capital into the bank prior to a conversion event being triggered. 77. The Reserve Bank intends to permit caps on the conversion ratio but does not intend to require such caps. If holders of capital instruments will receive few shares on conversion, then they will discipline banks by demanding higher coupon rates for greater risk taking. On the other hand if the capital holders receive a relatively large number of shares on conversion, shareholders will apply discipline given the threat of dilution. Tax consequences 78. Some submitters noted that write-off or conversion of subordinated debt may trigger a tax liability for the bank which would reduce the value of any new capital created. Reserve Bank response 79. The amount of the instrument recognised as capital will be the amount of common equity that would be generated from a conversion or write-off, net of any costs including any tax liability. Conservation buffer Prudent management of costs while operating within the conservation buffer 80. The Reserve Bank will require banks operating within the conservation buffer to submit a capital plan for rebuilding the buffer. Some submissions were concerned that the Reserve Bank would expect this plan to show prudent management of costs (such as staff bonuses). It was considered that this expectation went beyond the Basel III framework that envisages banks will be able to conduct business as normal while operating in the buffer, and imposes constraints only in relation to distributions (not bank operations). 81. The Reserve Bank first advised banks of its response to this issue in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand.

19 The Reserve Bank considers that constraints on discretionary staff bonuses as envisaged by the Basel III framework may be unworkable and is not consistent with our relatively nonintrusive supervisory approach. However we do consider that prudent management of staff bonuses is appropriate when banks are operating within the conservation buffer. Nevertheless, taking into account submissions, we do not now intend to require prudent management of costs from banks operating within the buffer. Restrictions on distributions while operating in the conservation buffer 83. Most submissions received on this issue disagreed with the proposal to fully restrict distributions to the extent necessary to restore the buffer, and maintained that the Reserve Bank should align with the Basel III framework whereby restrictions apply in proportion to the extent to which the bank is operating within the buffer. The submissions noted the proposed approach would result in banks holding excess capital given the consequences of operating within the buffer, and would significantly impact on the market for AT1 instruments, potentially making such instruments unmarketable. 84. The Reserve Bank first advised banks of its response to this issue in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand. 85. The Reserve Bank remains of the view that a bank operating within the buffer should restore the buffer as soon as possible. However we accept that in many circumstances banks will be incentivised to do so through market discipline. While there may be merit in requiring banks to restore the buffer as soon as possible, on balance, we accept the concerns raised by industry and now intend to adopt the Basel III partial restriction requirements without modification. Countercyclical buffer Notice period 86. Many submissions considered the appropriate notice period would depend on the size of the countercyclical buffer, while several submissions argued that more than 12 months would be needed. Some submissions argued that 12 months is an unreasonably short notice period for co-operatives and mutuals.

20 We remain of the view that a notice period of up to 12 months is appropriate. A longer notice period runs the risk that the countercyclical buffer is built up too late. In the context of a credit boom it is likely that capital can be raised reasonably easily and even if it could not the bank would simply face restrictions on distributions (there would be no breach of conditions of registration). Also, ordinarily we would expect dialogue with banks and public signalling of our unease about credit developments to begin in advance of the notice period commencing. 88. Ownership structure is a matter for banks to choose taking into account the various costs and benefits. For some kinds of ownership structure there will be heavy reliance on retained earnings to raise capital and it is up to the banks in this position to manage such risks. The Reserve Bank does not accept that the existence of cooperatives and mutuals provides a reason to have different countercyclical buffer requirements for different banks. Maximum size of the buffer 89. Several submitters considered a maximum size for the buffer should be set in the policy to enhance the transparency of the regime and to inform the decision about an appropriate length for the notice period. Some submitters considered the maximum size for the buffer should be no higher than 2.5% of risk-weighted assets. 90. The Basel III framework allows national authorities to set a buffer in excess of 2.5% of riskweighted assets. We remain of the view that no formal limit on the maximum size of the buffer is appropriate for New Zealand for two key reasons. First, because we have limited practical experience on which to calibrate the buffer, there is no strong basis for selecting a stipulated maximum. Second, retaining an open-ended maximum may be useful in conditioning bank incentives against excessive credit growth. Removal of the buffer 91. There was a mix of views on whether market expectations would allow banks to run down the countercyclical buffer once it is released by the Reserve Bank.

21 This issue remains the subject of considerable international debate. We appreciated views from NZ market participants. Indicators for determining when to apply the countercyclical buffer 93. Several submitters requested more clarity on the range of indicators that the Reserve Bank would use to calibrate a countercyclical buffer in the event of excess credit growth. It was argued this would enhance the transparency of the decision-making framework and create a more credible regime. It was also noted that enhanced transparency should include prior communication with the banks and the public. 94. We agree with the points made in submissions. Our intention is to provide further details on the indicator framework that will be used to guide the use of the countercyclical buffer. Application to non-bank deposit takers 95. Some submissions questioned whether the buffer would also be applied to non-bank deposit takers. 96. As noted in our March 2012 consultation paper the buffer could potentially be extended to other lenders (such as non-bank deposit takers) in the future. However extending the application of the buffer is not currently a priority given that banks account for the vast majority of lending and because additional work would be needed to address a range of practical issues involved in extending the regime. Leverage ratio 97. Several banks agreed with our position not to adopt the leverage ratio in our capital adequacy standards. However one submission noted that rating agencies and investors will wish to see leverage ratios given the comparability it provides for analytical purposes. 98. One submission put forward several arguments in favour of the leverage ratio including: ease of measurement; it provides a back-stop and prevents errors in the risk weighted asset

22 22 approach from bringing a bank down; international consensus on its usefulness; the banks don t like it and banks wish to reduce their capital; leaving it out could leave the Reserve Bank open to justifiable criticism. 99. Our view is that ease of measurement and international consensus on usefulness are not necessarily compelling reasons to adopt the ratio and do not outweigh our concerns about how it can give a misleading picture of risk (i.e. it is not risk-based and also a bank that improves its liquidity position could as a result record a lower leverage ratio) We do not consider a back-stop is necessary, but we do consider this kind of back-stop would undermine our existing risk based approach to capital requirements. We do not accept the absence of a leverage ratio requirement would allow banks to reduce their capital below the appropriate risk-based level Overall we remain of the view that a leverage ratio requirement is not appropriate for New Zealand banks. 3 Other issues Internal capital adequacy assessment process (ICAAP) Submission received 102. One submission queried whether the ICAAP would still need to cover off the risk of raising capital during times of stress and the need to absorb losses during times of stress, given the new conservation buffer requirements The Reserve Bank does not require banks to hold regulatory capital for risks identified through the ICAAP process. We are not planning any changes to our ICAAP requirements as part of Basel III implementation. However, as always, banks may expect supervisory scrutiny of ICAAPs. 3 Further explanation of the Reserve Bank s view on the leverage ratio is provided in the Reserve Bank s public submission to the Basel Committee s consultative proposals to strengthen the resilience of the banking sector, 14 April This submission is available on the Reserve Bank s website here:

23 23 Conservative adjustments for internal models banks 104. Some banks questioned whether the Reserve Bank s conservative approach to the implementation of the internal models approach within the Basel II framework should remain given the increased conservatism introduced by Basel III. On a related issue some banks considered the Reserve Bank s conservative Basel II adjustments should be given effect through higher capital ratios rather than through more conservative risk-weights to facilitate meaningful comparison of capital ratios across jurisdictions As Basel III builds on (rather than replaces) the Basel II framework, the Reserve Bank remains committed to ensure the Basel II framework is applied and calibrated appropriately in New Zealand. Our approach to Basel II has resulted in an appropriately more conservative outcome in the calculation of risk-weighted assets compared to most other jurisdictions, taking into account New Zealand circumstances The Reserve Bank is not the only banking supervisor that has imposed conservative adjustments to risk-weighted assets. Banks may disclose comparative information to provide investors with more information on how to interpret head-line capital ratios. Transitional arrangements Phase out of non-qualifying instruments 107. Some submitters considered that more time is needed to replace non-qualifying instruments. The main arguments presented in support of this view were: An accelerated timeline would disadvantage New Zealand banks relative to global peers. The timeline proposed would put a strain on the domestic market s ability to fund new instruments over a relatively short space of time. Compounding this difficulty is the fact that Basel III requirements make regulatory capital instruments less attractive to investors. Alignment with Basel III or APRA timelines should be the default, unless the benefits of early adoption can be demonstrated.

24 The Reserve Bank first advised banks of its response to this issue in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand Although banks did not generally provide any quantitative detail about the cost of the proposed timetable for phasing out non-qualifying instruments, the Reserve Bank accepts submitters consider more time is needed to replace non-qualifying instruments. The Reserve Bank also notes that APRA are now prepared to adopt the Basel III transitional arrangements for capital instruments issued out of subsidiaries and held by third parties, and that these arrangements are relevant to some capital instruments issued by some New Zealand banks The Reserve Bank therefore intends to adopt the following timetable for de-recognition of all non-qualifying capital instruments issued by New Zealand banks. Year commencing Percentage of instruments that may be included in regulatory capital 1 January January January January January January We also intend to require that instruments with a call and step-up on or after 12 September 2010 and before 1 January 2013 will not be recognised for transitional arrangements from 1 January 2013, and instruments with a call and step-up after 1 January 2013 be fully derecognised from the date the call and step-up apply. Early adoption of the conservation buffer and countercyclical buffer framework 112. Several submissions argued that early adoption of the conservation buffer and countercyclical buffer framework (from 1 January 2014) would make it challenging and expensive for banks to comply, and that there were no grounds to implement these buffers ahead of the timing proposed by APRA (from 1 January 2016) or the Basel III timeframe (three-year phase-in from 1 January 2016). Some submissions also argued that implementation of the countercyclical buffer framework from 1 January 2014 does not allow sufficient time given the additional policy work required (e.g. in relation to the range

25 25 of financial indicators and other evidence to determine when application of this buffer would be appropriate) The Reserve Bank first advised banks of its response to this issue, in respect of the conservation buffer, in its 8 May 2012 letter providing an update on Basel III implementation in New Zealand The Reserve Bank confirms that it will implement the conservation buffer and countercyclical buffer framework from 1 January 2014 (for the countercyclical buffer this means that from that date the Reserve Bank could announce that the requirement would apply from some future date). In coming to this decision the Reserve Bank took into account: that banks are reasonably well placed to meet the buffer requirements; that for some banks the phase-out decisions noted above will make it easier to comply with the 1 January 2014 date; and our long-held expectation that unless there are good reasons to delay we will implement elements of Basel III ahead of the Basel Committee s timetable as some other jurisdictions are planning The Reserve Bank acknowledges that further work is needed on the countercyclical buffer framework in relation to financial indicators as well as governance and accountability arrangements for the operation of the buffer. However we consider this work can be undertaken in sufficient time to allow the implementation of the countercyclical buffer in 2014.