Submission on the Mutual Capital Instruments Consultation Paper by Nelson Building Society and Southland Building Societies

BF\62578733\2 | Page 1

SUBMISSION ON THE MUTUAL CAPITAL INSTRUMENTS CONSULTATION PAPER INTRODUCTION This submission is made by Nelson Building Society (NBS) and Southland Building Societies (SBS), together the Submitters, in relation to the Reserve Bank of New Zealand's (RBNZ) consultation paper on Mutual Capital Instruments that qualify as CET1 capital (MCIs) dated 16 March 2022 (Consultation Paper). SBS is New Zealand's largest building society and the only one registered as a bank and subject to the capital requirements for banks set out in Bank Prudential Requirements (BPRs). NBS is a non-bank deposit taker (NBDT) and as such is currently governed by the capital rules in the Deposit Takers (Credit Ratings, Capital Ratios, and Related Party Exposures) Regulations 2010 (Deposit Taker Regulations). NBS is interested in having the ability to issue MCIs under the BPRs because issuing an instrument that qualifies as bank capital would be beneficial for ratings purposes. It also sees value in contributing to the development of policy for MCIs because it is conscious that these standards or something similar may become applicable to it as it grows and potentially becomes subject to bank capital requirements. The Submitters are grateful to the RBNZ for the work it has done in developing its policy for MCIs and in putting together the Consultation Paper. The Submitters recognise that mutuals are a small part of the banking sector and would like to thank the RBNZ for the time and effort spent consulting with us. Following the meeting between the RBNZ and our lawyers, Buddle Findlay, in December 2021, it appears that RBNZ and the Submitters have independently undertaken additional research and reached very similar conclusions on the best approach for an MCI in New Zealand albeit for slightly different reasons. In this Submission, we seek to set out our reasoning for supporting the RBNZ's preference for an MCI that reflects the approach taken in the United Kingdom (UK) and why we think the RBNZ should also accept, for prudential purposes, the provisions of the UK instrument relating to wind up. We also provide an industry perspective on some of the policy considerations and potential restrictions on MCIs to, hopefully, reinforce the preliminary conclusions reached by the RBNZ in its Consultation Paper and continue building on its understanding of the mutual sector.

BF\62578733\2 | Page 2 EXECUTIVE SUMMARY Primary Submission The Submitters agree with the RBNZ's conclusion that following the approach taken in the UK is a logical starting point for an MCI in New Zealand. We note that the core capital deferred shares (CCDS) in the UK include one feature that has not been adopted in the proposed MET1 instrument – a cap on investors entitlement to surplus assets to the par value of their shares. We are not clear on the reasons for the RBNZ taking a different approach, but we have a strong preference that this feature be accepted for prudential purposes as we think it achieves a better balance between the interests of investor shareholders and members. While we think the UK approach should be preferred, we also believe there is value in developing a framework that maintains enough flexibility to adapt features as we learn more about issuing mutual capital in the New Zealand market. In particular, we should not rule out the Australian approach at this early stage. Support for the Consultation Paper The Submitters strongly agree with the RBNZ on the following points made in the Consultation Paper: (a) The current inability of banks structured as mutuals to issue capital that qualifies as CET1 capital puts them at a disadvantage to non-mutual banks by limiting their potential to grow (and compete) and to manage their capital position. (b) It is consistent with international best practice to adapt capital requirements to be suitable for the mutual structure (and in fact encouraging diversity through adaption and proportionality is considered to enhance a regulatory framework rather than weaken it). (c) It is consistent with the RBNZ's financial stability objective to introduce an MCI because it enhances the ability of mutual banks to withstand periods of stress. (d) Introducing an MCI will positively impact competition and efficiency in the financial sector (although the RBNZ notes that this impact will be small, it has the potential to have a more significant impact over the long term). (e) Issuing capital (and the associated cost of capital) is not inherently inconsistent with mutual principles. While issuing capital could theoretically be inconsistent with member's interests (in the same way that a placement of new shares in a company could be inconsistent with the interests of existing shareholders), this is a very low risk. The boards of building societies are legally required to act in the best interests of their building society. This effectively requires them to ensure they act in the best interests of members who they are accountable to in a truly democratic manner (i.e. one member one vote). Accordingly, there are sufficient safeguards to ensure that capital is issued in accordance with mutual principles.

BF\62578733\2 | Page 3 (f) Any limitations on issuance of capital or caps on distributions should be a decision for the board, acting in the best interests of the building society (and effectively its members). Additional Feedback We also provide our views on other matters raised in the Consultation Paper which the RBNZ has requested feedback on, including the following: (a) Our inability to issue regulatory capital is more acute than the Consultation Paper suggests because our boards cannot simply decide to demutualise in order to improve our capital positions. Demutualisation is a process that must be driven by members whose interests are broader than optimising financial outcomes. (b) Even if the investor shareholders were entitled to have their capital returned ahead of the rights of members to residual assets, as is the position of the Mutual Equity Instrument (MEI), we believe the subordination requirement would be met. This is because the investors' claim to their paid in capital would still be the most subordinated "claim" in a liquidation. The members' right to residual assets is not a legal claim that can be made in a liquidation but a contractual right under the rules of a building society to the surplus after all legal claims have been satisfied. This view is consistent with the UK Prudential Regulation Authority's (PRA) acceptance of the PPDS instrument as CET1 capital. (c) The boards of mutuals are accountable to their members on a one vote per member basis (rather than investors proportionately to their share of paid-up voting capital) and market discipline must be viewed in light of this accountability structure. Therefore, proportionality as a criterion is of less relevance to capital in mutual banks for prudential purposes. (d) We understand from our UK counterparts and their lawyers that the inclusion of MCIs as eligible CET1 capital has not added any material complexity to the capital adequacy framework. It has only been seen as a positive, with no friction from the prudential regulator, and in practice is not materially more difficult to administer than ordinary shares in a company. (e) Restrictions on offering MCIs to retail investors should not be a matter for a prudential regulator as it does not change the prudential character of the capital. To the extent it is relevant it should be left for consideration by the Financial Markets Authority as conduct regulator (which is the same approach adopted in the UK). (f) We are agnostic on how an MCI is included in the BPRs, but see some merit in taking a similar approach to other comparable jurisdictions, such UK/EU, Australia and Canada who have all followed the same approach – MCI is a new component of CET1 capital, but its eligibility criteria is set out by reference to the criteria for ordinary shares and then setting out where the criteria for an MCI differs.

BF\62578733\2 | Page 4 THE PROBLEM DEFINITION The Consultation Paper provides an accurate description of how the inability of banks structured as mutuals to issue capital that qualifies as CET1 puts them at a disadvantage. The regulatory capital requirements (in the absence of an MCI) limit the ability of mutual banks to grow and their flexibility to manage their capital position (particularly in times of stress on the financial system). We note that the consultation paper appears to suggest that demutualisation is an alternative that mutual banks would prefer not to use. We would like to clarify that this is not simply a matter of preference, there are significant practical challenges with relying on this to improve a mutual banks capital position. In the absence of legislative change, demutualisation must be driven by members who have both financial and non-financial interests in owning their financial institution (discussed further below). The RBNZ has asked for feedback on how mutual principles interact with the eligibility requirements of ordinary shares as CET1 capital. In particular, it identified three eligibility criteria that may be inconsistent with mutual principles. We have provided our views on those eligibility criteria below: Criteria Application to mutual banks Subordination We do not believe that the entitlement of our members (or in the case of a credit union, a charity) to residual assets is a "claim" to the assets of a mutual bank. In the case of building societies, this is simply a contractual entitlement (provided for in our rules) that arises in the event that there are surplus assets in a wind up. It is not a claim that members could take to a liquidator. Therefore, even where investor shareholders have a claim to paid in capital that ranked ahead of members' rights to residual assets (eg the MEI) we believe this would still be the most subordinated "claim" and therefore consistent with the eligibility criteria for ordinary shares. However, we believe that the MET1 approach to distribution of assets on a wind up (and even more so, the CCDS approach) is more consistent with mutual principles because it is fairer for members. Proportionality It is possible for an MCI to meet this requirement in a way that is consistent with mutual principles. However, we note that participation in profits and losses is not central to market discipline in the context of a mutual bank. While investors will be incentivised to follow the financial position of the bank, the issue of mutual capital does not change the governance structure of a mutual. Instead, the governance structure remains with members on a one vote per member basis. They will still have the power to hold directors accountable for poor performance. As such, proportionality need not be a relevant consideration for MCIs. Distributions While we agree that it is a core tenet of mutuality that members are entitled to equal distributions, it is not inconsistent with mutual principles to recognise the legitimate cost of new capital. This cost should, of course, only be assumed where there is a reasonable expectation that taking on capital will yield a net benefit to members. In effect the directors should be satisfied that the return on capital should be greater than the cost.

BF\62578733\2 | Page 5 On the basis of our views outlined above, we do believe that it is possible for mutual banks to issue shares that would meet the eligibility criteria for ordinary shares without compromising mutuality. However, to reduce complexity and uncertainty we also believe there is a need to outline the specific eligibility criteria for an MCI.

BF\62578733\2 | Page 6 MUTUAL CAPITAL INSTRUMENTS – POLICY OPTIONS Option 1B: Mutual Tier 1 Capital Instrument (MET1) Like the RBNZ, the Submitters believe that following the approach taken in the UK is a logical starting point given the large building society industry in the UK and the acceptance the CCDS instrument has gained with the UK Prudential Regulation Authority (PRA) and market participants. The CCDS caps the entitlement of holders in a wind up to the par value of the shares (adjusted for multiple issues). We note that the MET1 instrument does not include this feature and we have a strong preference for including it as we believe this is fairer for our members. We set out below a number of scenarios that demonstrate why we believe this feature is fairer for members. While these examples are much more simplistic than those used in the Consultation Paper we believe they are useful to demonstrate that the CCDS would operate identically to the MET1 instrument, except in the event that a mutual bank is wound up having made a profit (i.e. surplus assets in excess of retained earnings and paid up capital). We believe it is unfair for investors to participate in profits in a wind up as they participate in profits via distributions, whereas members do not. Day One XBS has 1000 members and $100m in assets ($90m of deposits and $10m of retained earnings). It then takes on $10m in investment from 100 investors who each invest $100,000. Balance Sheet XBS Assets Liabilities & Equity Assets $110m Deposits $90m Retained Earnings $10m New Capital $10m $110m $110m

BF\62578733\2 | Page 7 MET1/CCDS: Shareholders are entitled to a share of residual assets in proportion to CET1 contribution (without limitation/up to par value) 100 shareholders get $7.5m ($10m capital - $2.5m (50% of losses)), or $75,000 each (distributed proportionately based on capital contribution, but in this example XBS shareholders invested equally. If say an investor invested $200,000, they would get a $150,000 return). 1000 members get $7.5m ($10m retained earnings - $2.5m (50% of losses)), or $7,500 each (distributed in equal shares). Result: Loss absorbed by shareholders and members based on each group's contribution to capital. MET1: Shareholders are entitled to share of residual assets in proportion to CET1 contribution without limitation CCDS: Shareholders are entitled to share of residual assets in proportion to CET1 contribution up to par value 100 shareholders get $12.5m ($125k each because in this example they all invested equally) 100 shareholders get $10m ($100k each) (capped at capital contributed) 1000 members get $12.5m ($12.5k each) (retained earnings pool + 50% of net profits) 1000 members get $15m ($15k each) (retained earnings pool +100% of net profits) Result: shareholders have their capital returned and participate in 50% profits. This is unfair as shareholders have participated in profits through dividends where members have not. Result: shareholders have their capital returned and do not participate in profits. This is fair as shareholders have participated in profits through dividends where members have not. One Year Later - Losses XBS is wound up having made losses of $5 million. Balance Sheet XBS Assets Liabilities & Equity Assets $105m Deposits $90m Retained Earnings $10m New Capital $10m (New Losses) ($5m) $105m $105m One Year Later - Profits XBS is wound up having made a profit of $5 million. Balance Sheet XBS Assets Liabilities & Equity Assets $115m Deposits $90m Retained Earnings $10m New Capital $10m Profit $10m Dividends ($5m) $115m $115m

BF\62578733\2 | Page 8 While including this feature means that shareholders would not share proportionately in surpluses (which is an eligibility criterion for ordinary shares), the Basel Committee on Banking Supervision is clear that it is appropriate to adapt the standards to take account of the mutual structure. As noted in the section above, proportionality is less relevant for an MCI because market discipline operates differently in mutual banks. This approach (capping distributions to shareholders on wind up) has been accepted in the UK/EU, Australia and Canada. Option 1A: Mutual Equity Instrument (MEI) While we believe the UK approach to MCIs should be preferred, we also believe there is merit in also having the option to issue a simplified instrument. In Australia, the initial Mutual Equity Interest was a more complex instrument similar to the CCDS and APRA have subsequently developed a simplified instrument due to practical complexities with the initial approach.1 While we have not been able to ascertain the exact nature of these difficulties, this does demonstrate that there may be value in maintaining some flexibility in approach to adapt as we learn more from our own experiences. We think a simplified alternative MCI should directly follow the Australian approach rather than the terms of the proposed MEI. In particular, this would involve two features that differ from the MEI, being: (a) that distributions may be calculated by reference to paid up capital; and (b) that shareholders' entitlement to surplus assets is capped at the par value of the shares. These features would allow for a simpler instrument that better protects the interests of members (when compared to the MEI). It would still be loss absorbing on a going concern basis and we believe the absence of proportionality is acceptable on the basis of the reasons outlined above. 1 APRA Discussion Paper: Common Equity Tier 1 capital instruments for mutually owned ADIs (26 July 2017) p10.

BF\62578733\2 | Page 9 POLICY CONSIDERATIONS Consistency with Eligibility Criteria for Ordinary Shares We believe that consistency with the eligibility criteria for ordinary shares should only be a relevant consideration if the criteria have the same impact for prudential purposes in the context of the mutual structure. In particular, the following eligibility criteria should be adapted to better suit the mutual structure with no material impact for prudential purposes: (a) Distributions: eligibility criteria for distributions should focus on loss-absorbency characteristics, such as distributions being non-obligatory and non￾cumulative. The requirement that distributions should reflect financial performance is less relevant in the context of the mutual structure because market discipline operates differently. (b) Proportionality: proportionate participation in surplus assets in a wind up has limited value for prudential purposes in the context of the mutual structure because market discipline operates differently (and should therefore not be included as a criterion for an MCI). In the appendix to this submission, we set out our assessment of the two policy options with our proposed amendments against the eligibility criteria for ordinary shares. Consistency with Mutual Principles We were pleased to see that the RBNZ have included consistency with mutual principles as one of its evaluation criteria for their policy options. As we have outlined above, we think it is possible to issue capital that is consistent with mutual principles and that enabling mutual banks to do this has the potential to improve outcomes for our members. The RBNZ also noted in the Consultation Paper that our boards have a legal duty to act in the best interests of their members. Accordingly, any decision to issue a MCI will only be made if it is in the best interests of members. In terms of issuing capital, the role of the board is analogous to that of a company board making a share placement, which the board should not do if it prejudices existing shareholders. The main difference is that for members, their interests are not just financial, but also effected in intangible benefits such as contributions to their communities and generally more personal service. They do not put value on their rights to surplus assets in a wind up Ultimately the board is elected by and is accountable to its members. There are sufficient checks and balances in the mutual governance structure to protect members interests, which should alleviate the concerns of the RBNZ about the risks of mutual banks raising capital. Consistency with Capital Review Principles The Consultation Paper states that the MET1 capital instrument would be inconsistent with the following capital review principles:

BF\62578733\2 | Page 10 (a) The capital framework should be practical to administer, minimise unnecessary complexity and compliance costs. (b) The capital framework should be transparent to enable effective market discipline. In relation to the first principle, we agree that there may be some additional complexity for a resolution authority or liquidator in a wind up. However, we understand from our UK counterparts and their lawyers that in practice the market has not found these instruments difficult to understand and, administratively, they are not materially more complex than an ordinary share in a company. We believe that including eligibility criteria for an MCI in the BPRs should be a relatively simple exercise, following the approach of comparable jurisdictions (such as the UK), which would not materially add to the complexity of the regime. Therefore, we believe that the additional complexity would be limited to the wind up of a building society which is a one-off event (which we do not think has ever happened in New Zealand) that would not, in practice, add materially to the complexity of the capital regime or compliance costs. In relation to the second principle, we believe that these instruments would have a very limited impact on the status quo in terms of transparency and market discipline. While there are some minor complexities in how mutual shares operate (in relation to tax and wind-up), as noted above, market discipline must be considered differently in light of the mutual structure. The overall governance and accountability framework for a building society's board remains materially unchanged by issuing capital. Specifically, investor shareholders will not obtain any meaningful control over the board and therefore, whether or not a mutual bank issues capital, it is the members that will hold the board accountable for its performance, consistent with the democratic principles of building societies. Costs and Benefits We are confident that there will be a clear net benefit of introducing an MCI that qualifies as CET1 capital. The Consultation Paper notes that there would be a small positive impact on both financial stability and efficiency in the financial system as a result of: (a) enabling mutual banks to improve their financial position to better absorb losses during times of stress and potentially providing a means of injecting capital as a resolution tool; and (b) creating a level playing field, where banks structured as mutuals are given the same opportunity to grow and compete. We agree – although we are hopeful that an enabling regulatory framework could help reinvigorate the mutual financial sector. In the UK 23% of all residential mortgages and 18% of all retail deposits are held with building societies.2 We don't see why this kind of market share would not be possible in New Zealand over the longer term, especially as co-operative ownership is gaining momentum internationally. We believe that an enabling regulatory environment has the potential to have a significant positive impact on financial stability and efficiency if we take a longer-term view. 2 See BSA - About Us | The Building Societies Association.

BF\62578733\2 | Page 11 We also think that certain counterarguments for, and costs of, introducing an MCI described in the Consultation paper should not be given much weight, specifically: (a) demutualisation is not a realistic alternative for introducing an MCI; (b) introducing an MCI would strengthen the capital adequacy framework rather than weaken it. Taking a proportionate approach and adapting rules to suit different organisational structures is not only consistent with BCBS standards, it is encouraged by them as having a positive impact on the financial sector; and (c) while capital raising introduces additional compliance costs, such costs will only be assumed where there is a reasonable expectation of a net benefit to members.

BF\62578733\2 | Page 12 RESTRICTIONS Caps on issuance and distributions The Consultation Paper discusses restrictions both on the proportion of mutual ordinary shares that can make up a mutual bank's CET1 capital and on the proportion of profits that can be distributed to shareholders. The rationale for such limitations appears to be consistency with mutual principles – in particular, protecting the interests of members. MET1 The RBNZ appear to have concluded that such limitations would not be necessary for the MET1 instrument. We strongly support this conclusion. We agree that protecting member interests should be paramount. However, as noted above, boards of building societies are (effectively) legally required to act in the best interests of their members. Again, we note that in this sense their core obligations are no different from a company board to its shareholders when making a placement of shares. The key difference is that the interests of our members are not purely financial. Our boards are in the best position to assess whether issuing capital would be in the best interests of members. For these reasons, we believe it would be preferable to avoid prescriptive limitations on issuing capital and to allow our boards (who have fiduciary duties to the entity and must have regard to the members to whom they are accountable) the flexibility to decide whether to issue capital on a case-by-case basis. Equally, we believe that the level of distributions should not be subject to any prescriptive limitations, but rather left to the discretion of our boards. The Consultation Paper also notes that banks have existing internal processes that require them to assess the impact of a range of anticipated distribution rates, which arguably makes caps unnecessary. We agree with this observation. Boards of building societies would not make distributions without first considering their impact and determining whether either making them or the amount of the distributions was in the best interests of members (and balancing their interests against those of investor shareholders). While it may be appropriate to cap distributions to protect member interests, as with limitations on issuance, we believe that such limitations should be developed by the board (potentially in consultation with their membership) and be capable of being adapted over time based on the needs of the mutual bank. In that sense, we can draw close analogies to a company policy on the percentage of profits it distributes to shareholders, which can be adapted over time. In many ways, the broader considerations that mutual banks must consider as core principles of our mutuality are increasingly similar to what banks must now consider as ESG factors. The board of a company must also balance shareholder profits against the interests of other stakeholders.

BF\62578733\2 | Page 13 MEI As noted above, we propose that if an alternative instrument were to be made available, it should directly follow the terms of the Australian Mutual Equity Interest. We note that APRA imposed restrictions (despite strong opposition from the industry) on both issuance and distributions on the basis of preserving mutuality (our arguments above in relation to MET1 apply equally here), but also because these instruments are untested. In terms of the instrument being untested, we do not believe this is a strong argument. Our legal advisors believe we can determine how these instruments would operate in a wind up with a high degree of legal certainty. To the extent that the concern relates to market discipline, as noted above, the introduction of investor shareholders does not materially change the governance or accountability structures from the status quo and therefore there is no increased risk to financial stability. As such, we believe there is no justification to impose limitations on a mutual bank's ability to manage its capital position, even where it is issuing a simplified instrument. Restrictions on offers to retail investors The Consultation Paper notes that issuing MCIs to retail investors has given rise to difficulties in other jurisdictions where investors did not fully comprehend the difference between the MCI and savings products (which were insured). While we acknowledge that particular care must be taken in how the risks of investment are disclosed to retail investors (particularly those who are also our depositor members), we think that our existing market conduct regulation provides a sufficient safeguard for this. The retail market could provide a valuable source of funding and, as mutuals, we would like to retain the possibility of giving our existing members the opportunity to invest. Any retail issues must be undertaken in close consultation with the Financial Markets Authority to ensure compliance with the existing framework. However, we do not think there is a prudential reason to place restrictions on issuances to retail investors. Ultimately, we believe any restrictions for retail investors is a matter for the Financial Markets Authority as conduct regulator (rather than the RBNZ as prudential regulator). Based on our discussions with UK building societies and their lawyers, this is certainly the approach taken in the UK.

BF\62578733\2 | Page 14 AMENDING THE BPRS: ELIGIBILITY CRITERIA AND ADDITIONAL REQUIREMENTS The Consultation Paper proposes to include a MCI as a new component of CET1 capital, rather than amending the eligibility criteria of ordinary shares. We are agnostic about how it is included in the BPRs, but suggest it might be useful to follow the approach taken in the UK (using the European Union framework), Australia and Canada, whereby eligibility criteria is provided by reference to the criteria for ordinary shares and then describing the exceptions to those criteria for an MCI. We propose that, like these jurisdictions, eligibility criteria should describe necessary characteristics rather than being prescriptive about the terms of the instrument. For example, the UK/EU prudential requirements are broader than the terms of the CCDS and the PRA have accepted that at least one instrument with different terms, the profit-participating deferred share, satisfies the eligibility criteria for CET1 capital. As a general proposition, we think there is no need for any differential treatment between the MCI and ordinary shares (for example, in relation to requiring approval for any repurchase) and think that differential treatment would add unnecessary complexity to the regime. While arguably there is no difference between an MCI and ordinary shares with rights additional to the basic rights set out in the Companies Act 1993 (e.g. to appoint a specified number of directors), we think that it would be reasonable for the RBNZ to require a legal opinion attesting that an instrument complies with the eligibility criteria for an MCI and do not believe that this would introduce a significant burden for banks structured as mutuals. As suggested in the Consultation Paper, this requirement could easily be incorporated by reference to the notification requirements for AT1 and T2 instruments.

BF\62578733\2 | Page 15 APPENDIX BPR110 requirement Australian Mutual Equity Interest’s characteristics CCDS instrument’s characteristics Permanence The instrument would have no specified maturity date. A mutual bank could redeem the instrument through a non-compulsory offer to purchase at the prevailing market price (although this may require an amendment to the mutual bank’s constitution or rules). ✔ The instrument would have no specified maturity date. A mutual bank could redeem the instrument through a non-compulsory offer to purchase at the prevailing market price (although this may require an amendment to the mutual bank’s constitution or rules). ✔ Subordination Upon winding up or liquidation, investors would be entitled to a claim on the surplus assets capped at the principal amount of the investment. After which, they would not participate in the distribution of any residual surplus assets which would be distributed equally amongst members on a ‘per member’ basis. While the instrument would rank ahead of the members' entitlement to residual assets, it would be the most subordinated claim against the mutual bank. ✔3 On a winding up or liquidation the instruments would rank (i) junior to all other claims (e.g., members’ deposits, wholesale debt instruments, T2 and AT1 instruments), and (ii) pari passu amongst themselves and members in aggregate (but capped at the par value of the instruments) at which point all residual assets would be distributed equally amongst members on a 'per member' basis. 4 ✔ Proportionality5 Proportionality would not be achieved as currently understood, though we do not think this is a relevant consideration for prudential purposes in the context of the mutual structure where market discipline operates differently. Investors would receive the principal of their investment in full proportionately with other investors before any residual assets are distributed to members equally on a ‘per member’ basis. N/A Investors’ claim to surplus assets would be proportionate to total CET1 capital contributed but capped at the par value of the instrument (so that distributions in a wind up would be proportionate unless the mutual bank was wound up having made a profit). Again, we note that this criterion is of limited relevance to a CET1 instrument for mutual banks, so this approach is appropriate from a prudential perspective. N/A 3 We have changed this from a ✘ to a ✔ based on our view that the subordination requirement is met even if the investor shareholders' claims rank ahead of the members' entitlement to residual assets. We believe the CCDS approach to wind up is preferable on the basis of maintaining equity between investors and members, rather than for prudential purposes. 4 This has been amended to reflect the cap on investor shareholders' entitlement to surplus assets. 5 Both instruments partially meet this eligibility requirement, but this is of less relevance for an MCI.

BF\62578733\2 | Page 16 Distributions Distributions would be loss absorbing on a going concern basis as they are non-obligatory and non-cumulative. The board must not make distributions unless all legal and contractual obligations are met and payments on more senior capital instruments have been made. Distributions may be linked to the principal paid at issuance. Caps may be set (for example through provisions in the rules of a building society) to preserve retained earnings and members’ equity interest in the mutual bank. ✘6 A separately published, board-approved distribution policy would set out the board’s expectation for distributions. The distribution policy would be indicative only, and distributions would be at the discretion of the board. This would be similar to how companies provide shareholders and potential investors’ earnings guidance based on key performance indicators. The distribution policy could be amended by the board to meet the needs of the mutual banks (in the same way a bank structured as a company could). Caps may be set (for example through provisions in the rules of a building society) to preserve retained earnings and members’ equity interest in the mutual bank. ✔7 Voting rights Investors would become members of the mutual, and subject to its rules, including one vote per member, regardless of the number of instruments held by the investor. ✔ Investors would become members of the mutual, and subject to its rules, including one vote per member, regardless of the number of instruments held by the investor. ✔ Variable and uncertain value Because investors would receive their principal in full (before any residual surplus assets are distributed) the instrument’s value would essentially remain fixed at the principal value of the investment. The only circumstances where the value would decline (subject to the ability to adjust par value to account for losses) would be when retained earnings had been depleted. In this case, further losses would continue to detract from CET1 capital, which would at this stage only consist of the MCI. In such a case, an investor would receive less than the face value of their investment if there were any surplus assets to distribute. ½ 8 Investors’ claim to surplus assets would be contractually capped according to a predetermined formula specified in the terms of the instrument. But the value of the investors’ claim would be variable and uncertain (but capped at the par value of the instruments), and hence loss absorbing on a going concern basis. ✔ 6 This does not meet the eligibility criteria for ordinary shares, but in the context of the mutual structure we think this is appropriate and that the criteria should focus on the loss-absorbing characteristics of the instrument. 7 The CCDS would meet the distributions criteria for ordinary share. Where caps are set by the board in a policy, this is still consistent and no different from what the board of a company could do. The RBNZ may permit building societies to include caps in their rules as has been the approach of building societies in the UK. This type of contractual cap may be technically inconsistent with the distribution criteria, but should no raise any prudential concerns. 8 It is possible to create a variable value by adjusting the nominal value of the instruments to account for losses (this was a feature of the PPDS issued in the UK).