Capital review - Next steps for the capital definition

Ref #7884923 v1.0 MEMORANDUM FOR FSO FROM Susan Guthrie DATE 17 Nov 2017 SUBJECT Next steps for the capital definition FOR YOUR Information and Agreement Background

1. The Capital Review was announced in March 2017. The second paper in the Review, released in July, addressed the question “What should qualify as bank capital?” Consultation on this paper closed on 8 September.
2. On 15 September FSO discussed the content of submissions to the second paper.1 A series of further work was agreed to by FSO: “The Committee agreed to the recommendations that Financial Policy undertake more work: 1) looking into the role of preference shares in the bank capital regime; 2) developing a CET1 instrument suitable for banks structured as mutual societies; 3) options for grandfathering; 4) looking into the pros and cons of a common equity regime in contrast to option four of the ‘numerator 1 Refer Documentum #7188187 for the minutes; Documentum #7174890 for the paper prepared for FSO. We recommend FSO: a) Note that on 15 September FSO asked for further analysis of five issues relating to the definition of capital. A sixth issue, based on bank feedback to Governors, was added in October. b) Note that further analysis has been done in relation to four issues. Two additional papers, one each on preference shares and a mutual society common equity instrument, are being presented at this FSO meeting. The analysis that has been done in relation to issuing contingent debt to parents is summarised in this paper. A paper on a local listing policy is due to go to the next BSG meeting and will then be brought to FSO. c) Note we believe it is premature to do further analysis of a common equity-only regime and grandfathering, but that it is important to keep options open in these areas in advance of further work. d) Note there are four in-principle decisions that we believe can be reasonably made, and announced, without further public consultation. e) Agree that four in-principle decisions be made before the end of the year. f) Agree that these in-principle decisions be announced as soon as an opportunity presents itself.

2 Ref #7884923 v1.0 paper’, and 5) local issuance. The next consultation paper could rule out contingent instruments as an option.” 3. In addition, Financial Policy was asked in October to look at a bank proposal in relation to AT1 contingent debt - namely to recognise such issues by banks to parents, but not to retail investors. 4. The purpose of this paper is to inform FSO of the work that has been done since 15 September, and to make various recommendations to FSO. Areas of Further Work 5. The six areas where further work was requested at FSO or by Governors are as follows: • the role of preference shares in the new regime; • recognising as common equity capital an instrument able to be issued by mutual societies; • allowing banks to issue contingent instruments to parents only; • requiring subsidiaries of foreign banks to list shares locally; • the pros and cons of a common equity-only regime; and • options for grandfathering contingent instruments. A re-look at the role of preference shares in the new regime.

1. The analysis that has been done on this topic since 15 September is outlined in a separate paper to FSO (documentum 7278623), but summarised here.
2. The second capital review paper outlined a package of preferred reforms. Option 4 included non-redeemable preference shares, with no contingent trigger, as AT1 capital and subordinated term debt (again, no contingent trigger) as Tier 2 capital.
3. Feedback from the banks was that they have no interest in, and investors have no appetite for, non-reedemable preference shares. Financial services firm Deutsche Craigs said, pointing to the experience pre-Basel III, that banks will easily find ways to offer redemption at a future date for ‘non-redeemable’ preference shares (for example, a sister company to the bank will agree to buy the shares from investors at a future date, giving effect to redemption in all but name).
4. This feedback warranted a fundamental rethink of the role of preference shares in the capital regime.
5. On balance, it appears that the most significant contribution preference shares can make is in the context of windup - the value contributed by investors is subordinate to all debt and thus acts as a meaningful buffer for depositors and other creditors. This suggests that preference shares should be accepted as Tier 2 capital, not Tier 1, irrespective of whether they are redeemable or not redeemable. Recommendation: Accept redeemable and non-redeemable preference shares as Tier 2 capital, but not Tier 1 (i.e. they form neither common equity nor AT1 capital).

3 Ref #7884923 v1.0 Recognising as common equity capital an instrument able to be issued by mutual societies. 6. The analysis that has been done on this topic since 15 September is outlined in a separate paper to FSO (documentum 7279410), but summarised here. 7. In their submissions, the banks structured as mutual societies requested that instruments that have capital qualities, but are unique to mutual societies, be accepted as common equity capital. 8. This option is envisaged in Basel III and several jurisdictions currently recognise such an instrument (Canada, the EU, Australia is currently developing policy in this area). 9. In principle, including such an instrument in the regime seems reasonable. In practice, if preference shares can, at most, qualify as Tier 2 capital, the instrument issued by mutual societies will need to differ materially from preference shares. 10. Based on a review of requirements overseas, it is possible to come up with a list of features that may, in combination, produce a mutual society instrument that is both common equity-like and likely to appeal to investors (i.e. for which there is a realistic market). Allowing banks to issue contingent instruments to parents only. 11. By way of a compromise with Option 4, one bank proposed that contingent debt issued to parents be permitted as capital (but not issues to retail investors). 12. There are several points that can be made here. Firstly, accepting contingent debt issued to parents means the capital regime administered by the Reserve Bank remains complex, both for supervisors and for the banks. Complexity arguably weakens the quality of the regulatory response to crises and imposes costs on banks and tax payers.2 Contingent debt instruments are complex and banks can make expensive mistakes. For example, just recently, 13. Secondly, an important factor in the New Zealand subsidiaries issuing so much contingent debt to their parents was a trans-Tasman tax benefit that will cease to exist from July 2018. It is not clear what, if any, commercial benefits will arise for the Australasian groups going forward, as a result of parents being able to purchase contingent debt from their New Zealand subsidiaries, instead of simply purchasing ordinary shares.

2 For example, see “The dog and the Frisbee” speech by Andrew Haldane from the BoE. Recommendation: The proposal merits in-principle support. We recommend that FP, advised by internal legal counsel, work collaboratively with the relevant banks to explore possible options. The options need to reflect decisions made with respect to preference shares. Any proposal arrived at will require public consultation. OIA s9(2)(ba)(i)

4 Ref #7884923 v1.0 14. Thirdly, allowing banks to issue contingent debt to parents would give a competitive advantage to banks with parent entities (assuming contingent debt is cheaper to issue than ordinary shares). Thus banks without parent entities would have an incentive to adopt a more complex corporate structure that includes a parent entity. 15. Hence the policy of allowing banks to issue to parents may have unintended consequences that weaken the regime and/or require a policy response. For example, at present, (and unlike many jurisdictions overseas) parent entities are not included in the banking group for capital regulation in New Zealand. This raises the prospect that the bank may be well-capitalised but the parent may not, having borrowed extensively to purchase the subsidiary’s ordinary share capital. If all local banks are incentivised to create parent holding companies, we may need to redefine what constitutes a group for capital regulation purposes (and that introduces trans￾Tasman complexity). 16. Fourthly, the regime that has operated in New Zealand since 2013 has largely been a regime of subsidiaries issuing to parents. 78% of the capital issued by the big four banks has been to parents. If we focus just on the more important AT1 capital, 90% of all issues by the big four banks have been to parents. 17. Under the existing regime, the big four New Zealand banks have acquired relatively poor quality capital compared to their parents. In contrast to their parent groups, the big four New Zealand banks have been accumulating less common equity, and more contingent debt, since Basel III was introduced. It is hard to see why, in the absence of any ceiling on issues of AT1 capital to parents, continuing to allow contingent debt in the framework will lead to a different pattern going forward. Thus, at a minimum, one would want to issue a ceiling on issues of AT1 capital to parents (but this introduces added complexity to the regime). 18. Contingent debt is lesser quality capital than ordinary shares, even though both instruments, when held the parent, expose the parent entity to loss. Compared to ordinary share capital, for example, there is some residual uncertainty that bank will

5 Ref #7884923 v1.0 be recapitalised by contingent debt. The parent entity incurs a loss on its ordinary share investment as soon as net assets in the subsidiary decline (as soon as a loss occurs in the bank). In contrast, contingent instruments remain as a claim on the bank until such time as the bank becomes non-viable and enters statutory management (this reflects our view that even contingent debt that is triggered by the CET1 ratio falling below 5.125% will, in fact, not trigger until the bank is non-viable). It is possible that, due to unexpected legal impediments, and/or impediments coinciding with political pressure, the contingent debt may not in fact write off (or convert into ordinary shares), but remain as a claim on a rescued bank. In contrast, had the bank been entirely funded by ordinary share capital, the parent’s investment would have eroded completely before government intervention. 19. Contingent debt also introduces extra risks for issuing banks. The financial implications of contingent debt are less certain for issuers because the instruments are vulnerable to tax policy changes in a way that ordinary shares are not (distributions on contingent are tax deductible, whereas dividends paid on ordinary shares are not). Also, the requirement to make regular interest payments on contingent debt may place more pressure on the New Zealand subsidiaries to generate income (compared to ordinary shares), for example, leading them to take more risks than they otherwise might. Requiring subsidiaries of foreign banks to list shares locally; 20. Analysis on this policy option has been done, but BSG is yet to review the report that has been prepared. Once BSG have had a chance to provide feedback, a paper will be brought to FSO. A common equity-only regime. 21. Given our recommendation is to permit redeemable and non-redeemable preference shares as Tier 2 capital, alongside subordinated debt as Tier 2 capital, at present our recommendations are incompatible with a common-equity-only regime. 22. However, we concede that having a common equity-only regime remains an option that should, at this point, remain on the table for consideration. If our recommendation that a common equity instrument be developed for mutual society banks is adopted, all banks would be able to participate in a common equity regime, and thus a common equity-only regime is feasible. Moreover, a common equity-only regime would be consistent with the principle that New Zealand adopt a conservative regime relative to peers. 23. If, following further consultation on preference shares, banks reject the proposed treatment of preference shares as Tier 2 capital and/or reject the inclusion of (non￾contingent) debt as Tier 2 capital, then it would seem appropriate at that time, to do an in-depth analysis of a common equity-only regime. 24. Analysis of the pros and cons of an equity-only regime would also need to factor in any changes being proposed to the measurement and aggregation of bank risk, and Recommendation: Do not include as capital instruments that have a contingent trigger, including those instruments issued to parent entities.

6 Ref #7884923 v1.0 thus should not precede decisions made in relation to the forthcoming denominator paper. Options for grandfathering contingent instruments. 25. Nothing further has been done in this area, as it seems reasonable to wait until policy on the denominator side has been finalised, and submissions have been received on the issue of the appropriate capital ratio for New Zealand. Areas where announcements can be made without further public consultation 26. While FSO requested further work be done in six areas, in our view there are also important aspects of the capital definition that require no further analysis, or public consultation, in order for FSO to make an in-principle decision. 27. These in-principle decisions would be followed by subsequent analysis, and public consultation, aimed at giving effect to the decisions (for example, in relation to Handbook redrafting). 28. The in-principle decisions that we believe can now be made by FSO without further analysis or public consultation are listed below: • Contingent debt and contingent preference shares will not be recognised as Tier 1 capital (including instruments issued to parent entities), acknowledging that the regulation wording used to give effect to this decision, and the grandfathering policy, will be subject to further consultation; • Redeemable and non-redeemable preference shares (with no contingent trigger) will be acceptable as Tier 2 capital, acknowledging that the regulation wording used to give effect to this decision, and the grandfathering policy, will be subject to further consultation; • Ideally the regime will recognise as common equity capital appropriately configured instruments issued by mutual society banks, acknowledging that the options need to be explored with the banks concerned; and • Capital issued through a SPV will only be recognised as capital if the SPV is part of the consolidated banking group for capital reporting purposes (this prevents a recurrence of the Kiwibank issue), acknowledging that the regulation wording used to give effect to this decision, and the grandfathering policy, will be subject to further consultation; Recommendation: Do not rule out the option of moving to a common equity-only regime, but do no further analysis at present. Recommendation: do further analysis of the grandfathering options once policy relating to the measurement and aggregation of bank risk has been decided, and submissions on the capital ratio has been received.

7 Ref #7884923 v1.0 29. In addition to making these in-principle decisions, an outline of the timetable for further consultations relating to the definition of capital could be released at the same time (for example, a schedule for public consultation on the wording to be used to give effect to the in-principle decisions). Recommended Timetable 30. We propose that FSO make in-principle decisions as outlined above, and announce these decisions. The submissions on the numerator paper are on the RBNZ website already. We could add a final decision paper to the website to reflect the in-principle decisions, note there are some areas of further work, and provide information about how these decisions fit in with the broader capital review.