Reserve Bank of New Zealand Capital Review: Cost Benefit Analysis and External Expert Reports

MEMORANDUM FOR FSC FROM Financial Policy (Richard Downing) MEETING DATE 11 September 2019 SUBJECT Capital Review: Overview of approach to Cost Benefit Analysis and next steps FOR YOUR Information It is recommended that the Committee:

1. Note that the Financial Policy team is developing a draft Cost Benefit Assessment (CBA).
2. Note that the draft CBA includes a range of judgements and assumptions that will continue to be tested and refined during the final stages of the Capital Review.
3. Note that draft CBA information is scheduled to be provided to the Reserve Bank Board for their meeting on 19 October.
4. Note that the draft will continue to be developed and then revised to incorporate the impacts of final decisions in November. Background
5. This paper provides an overview of the draft CBA that the Financial Policy team is developing for the final stages of the Capital Review. The draft CBA is being tested and refined and should not be regarded as a robust finalised assessment at this stage. This paper also provides FSC an overview of the milestones over the next few months for the review and finalisation stages of the CBA. At this point the draft has been developed to reflect the December 2018 proposals to provide a base for the analysis. The final CBA will reflect the final decisions.
6. Documents published during the Capital Review have included commentary on the costs and benefits of higher capital. The draft CBA pulls together a number of strands of analysis from Capital Review consultation papers, the April 2019 Background Note, and other analysis carried out during the review. It also draws on material from submitters on the Capital Review.
7. All of this previous information has been combined in the draft CBA to compare the costs and benefits in one place. Where possible these have been quantified, with the analysis showing that the benefits exceed the costs under a range of plausible judgements about the key inputs to the analysis. The CBA will continue to be developed as the team test and refine a number of the assumptions and judgements underpinning the CBA. Part of this process includes input from John Yeasbley (NZIER), and QA across the FSPA and the Economics Departments. We will also provide a copy of the draft CBA to Treasury, after draft material has been discussed by the Reserve Bank Board.
8. A finalised CBA will be part of the Regulatory Impact Analysis that will accompany the final decisions of the Capital Review. Draft Regulatory Impact Analysis and CBA material will be provided to the Reserve Bank Board for their 19 October meeting.

3 Summary of the draft Cost Benefit Assessment The draft Cost Benefit Assessment compares the impact of the proposals on expected GDP and other key variables with a counterfactual representing a no change environment. 6. The draft CBA incorporates all of the proposals from the December 2018 consultation paper and is calibrated in line with the assumptions and judgements underpinning this paper and the April 2019 Background Paper. The central question considered in the CBA is a comparison of the benefits associated with an increase in the soundness of the financial system (the avoidance of lost output resulting from financial crisis) with the costs that could arise from higher interest rates (lower output/GDP due to higher interest rates). 7. To make the comparison, the CBA estimates outcomes across a range of variables if the proposals are implemented, and compares these with the outcomes that would exist in the absence of the proposals – the outcomes in the absence of the proposals make up the counterfactual for the CBA. The counterfactual incorporates the status quo policy settings, as well as a range of assumptions, including:  The probability of a financial crisis in the absence of higher capital.  The impact of a crisis.  The way in which the economy would respond to higher levels of capital, including through higher interest rates. 8. None of the inputs above can be known with certainty. The assumptions and judgements create uncertainty around the impacts of higher capital. The judgements are also important for describing the counterfactual, as the counterfactual also incorporates the probability of a crisis in the absence of the proposals, as well as an assumption about the impact of a crisis. The inputs above are used to create an estimate of Expected GDP for each of the situations below:  Expected GDP if the proposals are implemented  Expected GDP in the counterfactual without the proposed higher capital requirements. The draft Cost Benefit Assessment shows the benefits of setting CET1 capital requirements at 16% of Risk Weighted Assets outweigh the costs 9. The key conclusions of the draft CBA are:  The benefits of the proposals exceed the costs.  Benefits exceed costs for a range of plausible assumptions about the key judgements underpinning the analysis. 10. The result arises because the increase in soundness, and the associated reduction in the probability of a crisis, increases expected GDP by more than the reduction in expected GDP that is associated with higher capital levels and higher interest rates. In the draft CBA, interest rates are estimated to increase by 30 basis points. The net benefit from higher expected GDP also outweighs the impacts of other quantified costs and benefits, the largest of which is the direct cost of higher interest rates on borrowers.

MEMORANDUM FOR FSC FROM Financial Policy (Richard Downing) MEETING DATE 11 September 2019 SUBJECT Capital Review: Communications to support publishing the External Experts Reports FOR YOUR Information This paper recommends that the Committee: i. Note that Final Reports from the External Experts for the Capital Review will be published at the end of September. ii. Note that the communications plan to support the publication of the Reports will include a Media Release, a briefing for selected media in advance of release and Q&A to support key questions about the Reports. Purpose of this paper

1. This paper provides an overview of the final External Experts Reports for the Capital Review. Final reports were all delivered by the 31 August deadline.
2. The paper seeks agreement from FSC about key aspects of the communications plan to support the publications of the Reports, including bringing in some members of the media to the Bank for a meeting in advance of publication. Summary
3. The three External Experts for the Capital Review have provided their Final Reports.
4. The Final Reports show that all three External Experts were complimentary of the quality of the advice underpinning the Capital Review proposals. The External Experts also signalled their general comfort with overall direction proposed in the Capital Review. Nevertheless, each Expert has signalled a few areas where they think more analysis is needed (see Annex for a summary).
5. At the 20 August FSC meeting the Committee agreed that the Reports would be published late in September, subject to FSC considering a communications plan to support the publication. The Committee also emphasised that the External Experts Reports should be communicated first with the Reserve Bank Board and then with the Minister of Finance, prior to publication.
6. The External Expert Reports are being published in order to maintain a transparent consultation process. The Financial Policy team is working with the Communications team to develop a communication plan that incorporates the following steps:  Publish a Media Release, with supporting Q&A to proactively address key questions about the External Review processes.  Provide embargoed copies in advance to key media, with the opportunity for them to also request a briefing within this embargoed week to aid in their understanding.  Accept media requests for interviews.
7. We are also considering publishing a short Summary Document alongside the External Experts Reports. This would identify only the main and common themes, plus any feedback that may attract particular external commentary. It would include a holding line

2 response for each. It would also note that the Reserve Bank’s responses for all points raised will be published alongside final decisions. Report Findings 8. As part of final stages of the Capital Review, three External Experts were commissioned to independently review the Reserve Bank’s analysis and advice underpinning the Capital Review proposals. 9. The Final Reports show that all three External Experts were comfortable with the quality of the advice underpinning the Capital Review proposals. All three External Experts commented very favourably about the quality of the analysis. The Experts also pointed to the transparency of the process and commented that that the Reserve Bank had carefully considered the points raised by submitters during the first three consultation papers, noting that the Bank has not yet responded to submitters on the fourth consultation paper.; 10.Nevertheless, each reviewer raised a few points that they suggested be considered further during the final stages of the Capital Review. These points, and the response to￾date are summarised in Annex 1. 11.Of the points in Annex 1 raised by the External Reviewers, the comments by Professor Levine about gaps in the analysis of the incentives on banks and barriers to entry to new market participants may attract the most attention. Professor Levine notes that he does not think this gap would mean the capital requirements should be higher or lower than proposed, but rather there should be further analysis of these points. 12.It is fair to say that Professor Levine’s comments reflect his preferred paradigm about how to analyse capital regulation and its impacts on banks and the economy. As such, we see it is a valuable adjunct to our analysis, rather than as an alternative to the other analytical approaches used during the Capital Review. Risks and communications plan 13.There will likely be interest in the reports from the public. Publishing the documents promptly will keep the public and key stakeholders informed of the process. 14.The key risk is that some commentators may claim that the generally supportive tone of the reports means that the reports are biased. The range of issues identified by the External Experts should provide some assurance that this is not the case. The communications plan discussed below will also emphasise that the Experts remained independent from the Reserve Bank at all times. Key elements of the communications plan are:  Publish a Media Release. We are also considering publishing a short Summary Document.  Publish supporting Q&A to proactively address key questions about the External Review processes (see draft in Annex 2).  Provide embargoed copies a week in advance to key media, with the opportunity for them to also request a briefing within this embargoed week to aid in their understanding Accept media requests for interviews Recommendation 15.It is recommended that FSC note the key elements of the communication plan.

4 small, open economy the impacts of more capital might differ from what would be expected in a larger economy. of the relevance of international studies for New Zealand. Prof Miles Overall Summary: Concludes that the Capital Review was done with care and in an open minded way. He rejects the idea that the analysis errs consistently in favour of more equity – ie. the level of capital has not been biased upwards. If anything, Prof Miles suggests that we have taken a risk neutral perspective. Issue Response Inputs to modelling Much of Prof Miles’ analysis focuses on the modelling used in assessing the level of capital that is consistent with a 1 in 200 year risk appetite. He suggests that some of the variables have been set at levels that are too “pessimistic”, resulting in a higher level of capital to meet the 1 in 200 year threshold. However, he also notes that other variables are too optimistic, in particular his assessment is that the assumed cost of crisis is too low. Prof Miles concludes that these considerations tend to have offsetting impacts and that the results are not biased upwards. The CBA for the capital review will include a detailed description of each component of the modelling. All of the inputs are subject to a significant degree of uncertainty, so to reflect this the CBA will also include a detailed discussion of the sensitivity on the results to alternative input values, including illustrating the sensitivity with alternative scenarios. Impact on bank funding costs Prof Miles concludes that the cost of equity funding to banks in the RBNZ’s analysis is too high. This means the costs of higher capital will be overstated. The Financial Policy team has reviewed this topic in detail. Revised estimates for the impact on funding costs will be considered in the Cost Benefit Assessment. Escalating supervisory response Pro Miles suggests that the RBNZ should make it clear that a temporary dip in the level of CET1 below the (substantially higher) conservation buffer under the new regime is of significantly less concern than a similar size shortfall under the current regime where the conservation buffer is less than one third as large. This issue is relevant for the restrictions of dividends and distributions when a bank’s capital ratio enters the buffer. This will be addressed in the final proposal. Prof Miles’ conclusion is also relevant for the work planned for 2020 to develop a detailed proposal regarding the Escalating Supervisory Response as a bank moves further through the buffer. Countercyclical buffer (CCyB) Prof Miles suggests that the RBNZ consider making the CCyB part of the rest of the prudential buffer, to remove direct allocation of part of the buffer to the CCYB. Issues regarding the CCYB will be addressed in CCyb FSC paper scheduled for 24 September.

5 Annex 2: Draft Q&A How were the External Experts chosen? The three External Experts are all highly regarded experts in the field of economic and financial research and analysis. The RBNZ required that the External Experts be internationally recognised in the topic of bank capital and regulation and that they had no conflicts of interest from existing contractual arrangements with New Zealand banks or with the RBNZ. The External Experts were selected from a shortlist compiled by the Reserve Bank following conversations with contacts at the IMF, APRA, the Bank of International Settlements and the Bank of England. The Experts were selected due to their international standing, as well as a desire to have geographic spread, and a range of different academic, regulatory and banking experience covered among the Experts. Were New Zealand experts considered? The Reserve Bank considered a number of possible Experts, including some based in New Zealand. Ultimately the three Experts were considered to be the best candidates given their international standing, expertise and experience. What were the External Experts asked to do? The Terms of Reference for the External Experts are available here: https://www.rbnz.govt.nz/-/media/ReserveBank/Files/regulation-and￾supervision/banks/capital-review/Terms-of-Reference-Capital-Review-External￾Experts.pdf?revision=b1121938-7649-449e-af7e-fa26958197ef&la=en Were the External Experts given access to submissions on all stages of the Capital Review? Yes. The External Experts were given unredacted versions of all of the submissions provided to the Reserve Bank on all four consultation papers. The Experts were only asked to assess to Reserve Bank’s responses to submissions on consultation papers 1-3, as the Reserve Bank has not yet responded to submissions on paper 4. The Reserve Bank will respond to submissions on paper 4 as part of the publication of final decisions for the Capital Review. This is scheduled to take place on 2 December. Did the External Experts talk to anyone outside of the Reserve Bank? The Reserve Bank helped set up meetings between the External Experts and any external groups that the Experts wanted to meet. Each expert met with the New Zealand Bankers Association and one of the big four New Zealand banks (Westpac NZ). Two of the experts also met with consultants from Sapere Research Group, as part of their meeting with NZBA. Sapere Research Group produced research and analysis commissioned by the New Zealand Bankers Association. What steps did the Reserve Bank take to ensure that the Experts remained independent? The External Experts were given scope to address the work in whatever way they considered would be most useful for them. The Reserve Bank provided background material in response to the Experts’ requests, and answered a range of questions from the Experts throughout the process. The Experts were invited to seek meetings with any external party that they thought would help their analysis. The Reserve Bank did not provide any drafting suggestions to the Experts as they completed their reports. Any feedback provided by the Reserve Bank throughout the process focused an addressing questions from the Experts and seeking additional information in areas where the Reserve Bank wanted to further understand the perspectives raised by the Experts.

MEMORANDUM FOR Financial Stability Committee FROM Financial Policy (Author: Charles Lilly) DATE 10 September 2019 SUBJECT How much would the Capital Review proposals ‘level the playing field’ for mortgages? FOR YOUR Information It is recommended that the Committee:

1. Note that the December 2018 Capital Review proposals included an adjustment to the calibration of the internal ratings based (IRB) approach used by the four large banks, so that risk-weighted assets (RWA) calculated using IRB would be approximately 90 percent of the RWA calculated using the Standardised approach used by all other locally incorporated banks.
2. Note that simple comparisons of IRB and Standardised ‘risk weights’ can overstate the difference in capital required between the two approaches.
3. Note that, taking into account differences in the methodology, we estimate that IRB banks’ capital requirements for their mortgage portfolios would on average be 90 percent of the capital required of the Standardised banks, if the Capital Review proposals went ahead unchanged.
4. Note that, while it has a separate prudential objective, if the proposed 1 percent DSIB requirement is taken into account, the IRB banks’ mortgage portfolio capital requirements would be approximately 96 percent of the Standardised outcome. Purpose of this paper
5. This note provides the Committee with information on a point of contention in submissions on the fourth Capital Review consultation paper, namely the extent to which proposed changes to the risk-weighted assets (RWA) framework would ‘level the playing field’, or result in banks having broadly comparable levels of capital for similar credit risks. An earlier version of this paper was presented to the Banking Steering Group.
6. We are not seeking decisions from the Committee at this stage, this note is background for the upcoming decision paper on denominator proposals. We are continuing to work through submissions and the external experts’ comments on the denominator proposals before finalising our recommendations to the Committee. Introduction
7. The December consultation paper proposed changes to the calibration of the IRB approach (a higher IRB scalar) and the application of a RWA floor tied to the Standardised approach. The changes are aimed at mitigating the risks involved in retaining IRB while preserving its role in promoting a risk sensitive allocation of capital for accredited banks.

2 4. As calibrated in the December proposals, IRB RWA would still be the binding constraint, and since the IRB scalar is applied uniformly across different types of lending, the proposed combination would not change the relative capital allocation across different sectors (the minimum capital quantum would increase proportionately to current relative risk weights). Feedback received 5. The key point of debate in submissions and subsequent meetings with banks is the extent to which the proposals reduce the differences in average outcomes between the IRB and standardised approaches. On average, based on QIS data collected as at March 2018, we had estimated that the proposals would lead to RWAs calculated under the IRB approach reaching 90 percent of the comparable standardised RWA outcome. 6. As a generalisation, the IRB banks submitted that the proposal is excessively conservative and does not take into account the benefits (e.g. improved risk management and more effective capital allocation) and costs (e.g. operational expenses involved in developing and running IRB systems). IRB banks also submitted that, while calibrated to a 90 percent outcome at a relatively low point in the credit cycle, it is plausible that through a greater sensitivity to risk, IRB capital requirements could overshoot the Standardised RWA in more adverse conditions. suggested that we include a ‘ceiling’ relative to the Standardised RWA, in addition to the proposed floor. 7. Standardised banks submitted that the proposal does not go far enough, particularly in key markets (mortgages and small business lending) where they perceive that there would still be an unjustified difference in outcomes relative to the IRB banks. Comparing IRB and Standardised outcomes 8. Making direct comparisons of the minimum capital required for a given dollar of lending between the IRB and Standardised approaches can be difficult, because the frameworks take different approaches to the capital calculation. One cannot make simple, direct comparisons of mortgage ‘risk-weights’ (a term not defined in the IRB approach), e.g. 35% under Standardised vs. 30% under IRB: o The IRB approach uses ‘exposure at default’ (EAD) to measure the non-risk adjusted size of a credit exposure. EAD is modelled by IRB banks. For Standardised banks, ‘exposure amount’ is prescribed by the framework, and combines the on-balance sheet value of lending and a prescribed conversion of off-balance sheet amounts. o In addition to RWA, IRB banks calculate an expected loss amount (EL), which (net of eligible provisions) is deducted from their capital. The EL deduction is effectively an add-on amount of required capital for IRB banks which is not included in the RWA value. 9. The figures on page 4 illustrate the additive effect of the different aspects of the Capital Review proposals on the amount of Tier 1 capital each of the main mortgage lending banks would need for $100 of mortgage lending, relative to the status quo. The figures take into account differences in each banks’ LVR and investor/owner-occupier profiles, and IRB model outputs where relevant. ANZ

3 10. The different steps in the figures are as follows: Status quo 8.5% of current RWA for mortgages. ANZ overlay Supervisory adjustment to ANZ’s RWA, effective June 2019. IRB scalar Proposal to increase IRB scalar from 1.06 to 1.2. Increase to 15% Tier 1 Adds 6.5% of RWA to reflect new requirement (15% Tier 1). Takes into account RWA adjustments in previous two steps. Adjust for EAD Takes into account that EAD (IRB) is a more conservative measure of the lending exposure amount than in Standardised. Calculated using QIS data. EL deduction Takes into account that IRB requires a deduction from capital for expected losses. Calculated using Capital Satellite and Asset Quality surveys. DSIB buffer A further 1% of RWA is required of DSIBs. 11. The analysis shows that, excluding the DSIB requirement, Tier 1 capital required by the four IRB banks would on average be 90 percent of the Tier 1 capital required of the four Standardised banks. 12. Including the 1 percent DSIB buffer to the calculation, the differential between the IRB and Standardised banks closes to 96 percent. A DSIB buffer of 2 percent would result in IRB banks’ mortgage capital requirements being 103 percent of the Standardised banks’ requirements. 13. While the DSIB buffer aims to address a separate prudential objective than neutrality of the RWA methodologies, because the four proposed DSIBs are the four IRB banks it is a relevant factor to bear in mind when comparing capital outcomes between the two RWA approaches. 14. Two other factors are worth considering when thinking about the ‘gap’ in IRB and Standardised mortgage capital requirements: o The above estimates use data for June 2019, when we are arguably at the low point in the credit cycle. While IRB banks’ models aim to be ‘through-the-cycle’, they are still likely to exhibit some cyclicality. Therefore, during less benign times IRB capital requirements would be a higher proportion of the Standardised requirement. o The idea that a given loan has the same risk (and so capital needs) irrespective of which bank’s portfolio it’s in is intuitively appealing. However, the credit risk of a loan to a bank will depend on the other loans in the bank’s portfolio, due to diversification. For example, given the relative geographic concentrations in their respective portfolios, all else equal a marginal mortgage loan in Taranaki will be a riskier proposition for TSB than it would be to ANZ. In this way, greater diversification can justify a lower overall capital requirement.

4 Figure: Waterfall from status quo minimum required Tier 1 capital to comparable post￾Capital Review minimum required Tier 1 capital, aggregate of IRB and Standardised banks 0 1 2 3 4 5 6 0 1 2 3 4 5 6 Status quo ANZ overlay IRB scalar 15% Tier 1 Adjust for EAD EL deduction DSIB buffer End point Status quo 15% Tier 1 End point IRB banks Standardised banks Required Tier 1 capital per $100 lending s 18(c)(i)

MEMORANDUM FOR Financial Stability Committee FROM Financial Policy (Author: Charles Lilly) MEETING DATE 10 September 2019 SUBJECT Updated estimate of the interest rate impact of the Capital Review proposals FOR YOUR Information It is recommended that the Committee:

1. Note that the attached paper to the Banking Steering Group responds to feedback Dr. James Cummings provided on our estimates of the interest rate impact of the Capital Review proposals in his external expert report.
2. Note that, using a similar approach to Dr. Cummings’ research into Australian banks, we found that the Capital Review proposals could lead to an increase in banks’ lending rates of 26.5 bps, with sensitivity testing providing a range of 21 to 32 bps. These estimates are lower than our previous estimates and towards the lower end of the range we have publicly communicated (20 to 40 bps).
3. Note that Dr. Cummings has suggested a tax efficient means by which the large Australian-owned subsidiaries could raise the equity needed to meet the Capital Review proposals would be to issue shares in the subsidiary to New Zealand tax resident investors. This would allow the banks to realise the benefits of the imputation credits they receive on their New Zealand earnings, which the currently make little use of, without needing to relinquish their controlling stakes. We estimate that under a scenario where all of the equity required to meet the proposals is raised from New Zealand tax resident investors, the increase in lending rates required to meet banks’ cost of equity would fall to 15.5 bps from the 26.5 bps base case.

3 MEMORANDUM FOR Banking Steering Group FROM Financial Policy (Author: Charles Lilly) MEETING DATE 30 August 2019 SUBJECT Updated bank funding cost implications of the Capital Review FOR YOUR Information It is recommended that the Group:

1. Note that, in his Capital Review External Expert Report, Dr. James Cummings has recommended that Financial Policy review our current estimates of the impact of the proposals on banks’ interest rate margins. In his view we have implicitly used an excessively high equity risk premium (ERP) relative to other benchmarks and therefore overestimated the cost of equity and the extent to which banks would need to increase their lending interest rates to meet shareholder expectations.
2. Note that some of our existing estimates have used banks’ realised return on equity (RoE) as a conservative assumption for the cost of equity (CoE). CoE represents the returns that banks need to generate to satisfy shareholder return expectations; RoE is the return that banks actually achieve (and has been higher than CoE for Australian banks in the past 25 years).
3. Note that Dr. Cummings and his coauthors’ empirical research into Australian banks’ funding costs suggests that, on a comparable basis, changes in their capital ratios would lead to smaller interest rate impacts than we have so far estimated would result from our Capital Review proposals.
4. Note that in this paper we adapt Dr. Cummings’ research to data on the four largest New Zealand banks. In our base case replication of Dr. Cummings’ methodology, we find that the Capital Review proposals could lead to an average 26.5 bps increase in lending rates. This base case includes several conservative assumptions, including no fall in banks’ debt risk premium, and that banks operate at a 17% CET1 ratio post-Capital Review (2.5% above the proposed 14.5% CET1 requirement).
5. Note that sensitivity analysis of the model’s inputs in the base case suggests a range of lending rate impacts from 21 to 32 bps, which is towards the lower end of our publicly stated range of 20 to 40 bps (our other estimates are closer to 40 bps).
6. Note that if banks were to raise the additional equity needed to meet the proposed requirements in a tax-efficient manner (through issuance of equity to NZ-tax resident investors, e.g. via a local listing), the effect of the imputation system would be to reduce the estimated average impact on lending rates to 15.5 bps.
7. Note that in a benign, but plausible, combination of model inputs (half of the new equity being raised from NZ investors, a slightly higher market value for the New Zealand banks, and a 15.5% CET1 ratio post-Capital Review to account for some AT1) the estimated impact of the proposals on lending rates reduces to 15.1 bps.

7 15. While banks may set “aspirational” RoE targets that they seek to achieve to increase shareholder wealth beyond the CoE, when assessing what changes to interest rates would be economically justified as an outcome of the Capital Review we should only consider the CoE. However, if competition is imperfect, banks may reprice lending rates beyond what is needed to meet their CoE and seek returns in excess of what can reasonably be attributed to changes we are proposing to make to their funding structures. We must also recognise this as a possible outcome. Estimating the funding cost impact of the capital review proposals 16. In this section we apply New Zealand bank data to the CAPM-derived estimates of the ERP for Australian banks. Cummings and Nguyen (2019), henceforth CN, estimate the impact of hypothetical changes in Australian banks’ capital ratios on their funding costs using data from the period 1993-2017.7 17. Briefly, the CN methodology is as follows: a. Estimate an ERP for 16 listed Australian banks from the period January 1993 to December 2017 using both CAPM and DDM approaches b. Calculate two measures of financial leverage – market leverage, based on the market value of assets and equity, and book leverage, based on book values c. Run a panel regression of the ERP estimates on these leverage measures to establish the empirical relationship between banks’ leverage and their cost of equity d. Based on these regression results, evaluate the change in the weighted average cost of funding (WACC) in a hypothetical scenario where banks increase their leverage (increasing debt by 5% of RWA and decreasing equity by 5% of RWA), taking into account tax and imputation effects e. The outcome of the above steps (computed change in WACC) provides an estimate of the lending rate impact of a 5 percentage point decrease in banks’ CET1 ratios Estimating a market leverage multiple for NZ banks 18. Applying the CN methodology to determining WACC requires a before-and-after value for the market leverage multiple (the market value of assets relative to the market value of equity). The four Australian subsidiaries are not listed and therefore we are not able to directly observe how investors would value their equity as a standalone firm. However, we can roughly estimate a value with some simple valuation multiples. 19. Two valuation metrics for a listed company are the price-to-book (PB) ratio (the firm’s market capitalisation as a ratio of the book value of its equity) and the price-to￾earnings (PE) ratio (the current share price relative to the firms’ earnings per share (either current or expected)). While these ratios will vary significantly across industries, and for firms in a given industry with different business models, for mature firms in comparable industries these two ratios can provide a benchmark for valuation. We consider it a reasonable starting point to apply Australian bank PB and PE ratios to the large New Zealand banks.

7 As a side note, several submissions (WNZL, NERA, and Tailrisk Economics) directed our attention to CN and their methodology and results when discussing MM offset effects, and hence the impact of the proposals on lending rates. The analysis in this paper therefore assists with our engagement and response to the points made in these submissions.

10 Implications for WACC and lending rates 26. CN compute the change in the WACC that would result from Australian banks borrowing 5% of RWA and using the proceeds to return an equivalent value of equity to shareholders. CN’s WACC calculation takes into account the effect of tax and imputation credits: 𝑊𝐴𝐶𝐶 = 𝐸(𝑟𝐸 ) × [ 1 − 𝑡 1 − 𝑡(1 − 𝛾) ] × 𝑀𝑉 𝑜𝑓 𝑒𝑞𝑢𝑖𝑡𝑦 𝑀𝑉 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠

• 𝐸(𝑟𝐷) × (1 − 𝑡) × 𝑀𝑉 𝑜𝑓 𝑑𝑒𝑏𝑡 𝑀𝑉 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠 Where:  𝐸(𝑟𝐸 ) is the expected return on equity  𝐸(𝑟𝐷) is the expected return on debt  𝑡 is the corporate tax rate  𝛾 is the imputation “gamma”, the extent to which corporate tax paid by the firm is valued by shareholders through the franking system  𝑀𝑉 is the market value

27. To estimate the change in WACC for New Zealand banks under the Capital Review proposals we need to make a few more assumptions before we can plug in the values computed above: a. As in CN, 𝐸(𝑟𝐷) is set at 5%, which is also our risk-free rate. We similarly align with CN and make the conservative assumption that the expected return on debt doesn’t change in response to the Capital Review proposals (under MM we might expect the debt risk premium to fall somewhat, so one could instead have modelled this using a non-zero but declining debt risk premium before and after the Capital Review change). b. We set 𝛾 at zero given the low value of NZ imputation credits to the shareholder base of the Australian parent banks, who we assume would be the investors providing the additional capital needed to meet the proposed ratios (either through fresh raising of equity or the foregoing of dividends).8 c. 𝑡 is 28%.
28. Finally, to convert the change in WACC to an impact on lending interest rates, we divide the estimated change in WACC by the ratio of banks’ interest-earning assets to their total assets, 94.3%, to account for an assumption that the higher cost of funding is recovered through interest-earning assets only.
29. Table 5 summarises the outcomes of the base case specification, as well as a range of sensitivity testing of the model’s inputs. Results and sensitivity testing
30. In the base case, the change in leverage following the Capital Review increase the WACC by 25 basis points, which translates to a 26.5 basis point increase in lending

8 For example, according to their results announcement, in FY2018 ANZBGL’s NZ-resident shareholders were paid dividends of NZ$67m, around 1.5% of the total cash dividends paid by ANZBGL (AU$4193m). These dividends to NZ-resident shareholders included NZ imputation credits equivalent to about 11% of the gross amount of dividends paid (i.e. dividends were ~40% imputed for NZ shareholders). The NZ geography contributed 23% to ANZBGL’s total earnings. Together, these values suggest a value for 𝛾 for the NZ earnings of ANZBGL in 2018 should be no more than 72% (earnings payout ratio) x [1.5% (share of ANZ Group shareholders who could hypothetically access NZ imputation credits) / 23% (share of ANZ Group earnings generating NZ imputation credits)] x 40% (proportion of NZ dividend imputed) = 1.9%. Setting 𝛾 at 2% instead of zero would have only a small impact on WACC.

11 rates. As documented above, this base case includes a number of conservative assumptions: a. No change in debt risk premium when banks reduce their leverage b. Banks use the additional equity to retire ‘average’ yielding debt rather than their most expensive debt9 c. Banks meet a 17% Tier 1 capital ratio only using CET1 capital d. Market valuations are based on long run average valuation multiples of Australian banks 31. Sensitivity analysis covers: a. Different CN model specifications b. A higher market value for New Zealand banks’ equity c. A lower risk-free rate of 3% d. Different MM offset assumptions (0% and 100%) e. A lower CET1 ratio of 15.5% (if banks issue 1.5% of qualifying AT1)10 f. A 25% higher market risk premium in New Zealand relative to Australia g. A scenario where the additional equity needed to meet the proposed capital requirements is raised from NZ-tax resident investors, who can take advantage of NZ imputation credits (e.g. if the Australian subsidiaries were to partially list on the NZX). 32. Most of these alternative ranges of inputs do not materially change the estimated impact of the Capital Review proposals on lending rates, with most producing outcomes in the range of 21 to 32 bps, compared to the 26.5 bps base case. 33. A 100% MM offset scenario, where we assume that the ERP is proportionate to the market leverage multiple, implies an impact on lending rates of 6.5 bps. The intuition behind this result being greater than 0 is that despite the CoE fully adjusting to lower risk, the proposals would still increase the WACC of banks because they are less able to take advantage of the tax shield on debt finance. 34. Another scenario with substantially lower impacts is one where banks raise the additional equity needed to meet the proposed requirements from NZ tax resident investors. As noted above, at present the Australian-owned banks may face a higher WACC than they otherwise might be able to achieve, because only a very small proportion of their ultimate shareholders are able to take advantage of imputation credits received on the banking group’s New Zealand operations.11 In the sensitivity analysis, we model a scenario where the full extent of the required increase in equity in the New Zealand subsidiaries is raised from NZ tax resident investors (e.g. through a partial divestment of the NZ subsidiary). Since New Zealand-sourced equity is more cost efficient than equity injected from the parent (or equivalently, built through foregone dividends) due to imputation, the estimated impact of the Capital Review on lending rates falls in this scenario to 15.5 bps. The most likely limitation to this

9 In this simplified model the conservative assumption is made that when banks increase their equity and retire their debt, the debt that is repaid is “average” priced debt, when we would expect that given the choice, banks would look to support their NIM by using new equity to retire their most expensive forms of debt. For example, retiring long dated market funding, instead of, for example, their ~$141b in deposits yielding <1%. 10 Making up more of the 17% Tier 1 ratio with AT1 instruments should in theory be accompanied with a higher cost of debt funding, marginally increasing the WACC and lending rate impact. We did not account for this in the calculation. 11 This raises another question, beyond the scope of this paper, of what barriers or competitive imperfections have prevented one (or more) of the four banks from seeing it worthwhile to obtain an apparent relative funding cost advantage over its peers by targeting a shareholder base that can take greater advantage of the New Zealand imputation system.

12 scenario occurring is the capacity of the New Zealand market to supply $23b in equity to meet the new requirements, rather than other considerations (e.g. APRA treatment of minority interests12). 35. Two final sensitivity tests are a combination of variations to the model. In an “adverse” calibration, relative to the base case we assume that there is no MM effect (the cost of equity is invariant to leverage), and that New Zealand’s market risk premium is 25% higher than in Australia. While these may appear to be implausible assumptions,13 the combination of the two only results in an estimated lending rate impact of 37.4 bps. 36. In a “benign” but fairly plausible combination of model inputs, relative to the base case we assume that banks have a higher market value than implied by the long run Australian data (we use a 14.5x PE ratio, a value suggested to us by ASB), that half of the additional equity needed to meet the new requirements is raised from New Zealand tax resident investors who can use imputation credits, and that banks only increase their CET1 ratios to 15.5% given the option to issue up to 1.5% of qualifying AT1 to meet a Tier 1 ratio of 17%. In this calibration, the total impact of the Capital Review on lending rates reduces to 15.1 bps.

12 A potential disadvantage to raising equity from minority shareholders is that capital contributed in this way is not necessarily recognised in the capital of the banking group in APRA’s Level 2 calculation. CET1 contributed to the New Zealand subsidiary from minority shareholders can only be counted in the Level 2 Group’s CET1 to the extent it contributes to meeting the minimum capital requirements associated with the NZ subsidiary’s contribution to Level 2 RWA. NZ subsidiary capital attributable to minority shareholders that is needed to meet RBNZ requirements, but which is surplus to the minimum APRA requirements, is not included in the Group’s capital. This is because this ‘excess’ capital contributed by the NZ minority investors would not be available to absorb losses elsewhere in the Level 2 Group. Our preliminary view is that it is not clear that the disadvantageous treatment of minority interest capital in a New Zealand subsidiary would be a material consideration: for the parent bank, APRA’s minimum requirements are the only constraint at Level 2, not the RBNZ’s. Raising the extra capital needed to meet RBNZ’s higher requirements from NZ minority shareholders means that a portion of this new external capital will not be recognised at Level 2. However, given the extra capital raised from NZ minority shareholders would not be needed to meet APRA’s Level 2 minimum CET1 requirements (since they are lower, and already being met), the non-recognition of some NZ minority interest capital at Level 2 has no detrimental impact on the ability of the banking group to meet its Level 2 requirements. This is only a preliminary view; for example, we have not fully thought through how the treatment of minority CET1 interests at Level 2 would support the parent’s ability to meet APRA’s recently announced higher Total capital requirements at Level 2. 13 For example, Dimson, Marsh and Staunton in the Credit Suisse Global Investment Returns Yearbook 2019 show there is no statistically significant difference in the market risk premium between Australia and New Zealand.

MEMORANDUM FOR FSC FROM Susan Guthrie, Financial Policy DATE 20 September 2019 SUBJECT AT1 capital FOR YOUR Decision Summary Note The purpose of this paper is to seek decisions from FSC about what should qualify as AT1 capital (this paper does not address the amount of AT1 capital we might require, nor does it address Tier 2. These topics will the subject of future FSC decision papers). Note A companion paper, providing a summary of feedback and responses to the issues raised, has also been prepared by way of background (refer paper #8525028) Note We have reviewed submissions, considered the final reports of the three external experts and engaged in follow-up discussions with banks, financial service providers and INFINZ. This engagement has resulted in proposed term sheets for AT1 instruments being provided by the domestic banks as a group and BNZ. Note A key concern for all banks is having access to Tier 1 capital alternatives that are cheaper than common equity. The recommendations in this paper address this concern by giving banks access to AT1 capital in the form of simple redeemable perpetual preference shares (‘RPPS’) - a conventional funding instrument that has an established history both in New Zealand and offshore, and has the loss absorbing properties of equity but is priced like debt. Note Unlike the current AT1 instruments, the proposed preference share will be able to be issued by most, perhaps all, domestic banks as well as the large four banks. This is in contrast to the current regime where only the large four, and Kiwibank using a questionable, complex, SPV structure, have been able to avail themselves of AT1 capital instruments. The proposal thus offers the prospect of a more level competitive impact. Note The proposal is strongly supported by the domestic banks, and the BNZ. WNZL is still investigating aspects of the proposal and is due to provide further feedback by Sep 27th. ANZ and ASB continue to advocate for the status quo (namely debt with contractual bail in) for a number of reasons, which are outlined briefly here, and in detail in the companion paper. Recommendations R1: We recommend accepting redeemable (and non-redeemable) perpetual preference shares (RPPS) as AT1 capital. This is a change with respect to an in-principle decision announced in December 2017. Previously, the Reserve Bank signalled that it would only accept RPPS as Tier 2 capital and non-redeemable perpetual preference shares as AT1 capital. R1: We recommend that only RPPS that do not have contractual conversion or write-off (‘contingency’) be accepted as AT1 capital. This reconfirms an in-principle decision made in December 2017.

2 Summary

1. This paper presents our recommendations for AT1 capital. In our view the recommended instrument offers, on balance, more net benefit than instruments that are currently accepted as AT1. A high level comparison of what is proposed with the status quo is provided in Table 1. Table 1: Significant cost advantage relative to CET1 Certainty as to going concern loss absorption Able to be issued by most NZ banks Complex or simple to administer, comply with Residual risk - inappropriate repayment yes yes Residual risk - signalling Interface with ESR Basel III compliant Marketable to local investors (retail and/or wholesale) Marketable to offshore investors Able to be sold to offshore parents APRA compliant (capital for Level 2 group if sold to 3rd parties) Unintended consequences for the wider ecosystem:

• different tax treatment for banks verus other business
• exposes banks to risk of miss-selling Current AT1 Proposed AT1 (RPPS) less certain more certain definitely yes yes possibly, but uncertain more complex less complex yes, higher risk than RPPS yes, lower risk than status quo complex, potentially adverse yes yes higher risk lesser risk yes no yes no qualified yes simple, complementary yes Introduction

1. Redeemable perpetual preference shares (RPPS) are a form of equity funding. The funding is available to the issuer indefinitely, unless voluntarily repaid (i.e. unless ‘redeemed’ on one of several specified ‘optional call’ dates set in advance in the contract terms).
2. RPPS differ from ordinary shares in that the dividend reflects a benchmark interest rate, rather than annual earnings, preference shares rank ahead of ordinary shares in liquidation and typically have no voting rights.
3. RPPS have the potential to absorb losses on a going concern basis because the dividends payable are fully discretionary to the issuer. No debt accrues if dividends are unpaid. Similarly the issuer is under no obligation to redeem the instrument.
4. With dividends payable are solely at the discretion of the issuer, any repurchase by the issuer is solely at the issuer’s discretion and no expiry date in the contract, a preference share can be considered, in theory, to be permanent capital. If the bank needs to retain the funding provided by the preference share, everything in the contract permits it to do so. Unsurprisingly, then, where the above conditions are met

3 accounting standards point to preference shares being accounted for as equity, not debt, on a bank’s balance sheet.1 5. From an investor’s perspective RPPS bear a strong resemblance to debt. Not only are dividends based on a benchmark interest rate, but if redeemed, the shares are repaid at face value rather than market value at the time the share is redeemed. RPPS provide low cost capital funding 6. Despite having the loss absorbing properties of equity, investors in RPPS require returns that are similar to fixed term debt. This is because the optional call dates act as an anchor for calculating the implied price volatility of the RPPS, and it is the implied price volatility that dictates the return required. Based on our discussions with the banks, Forsyth Barr and DeutscheCraigs, the returns demanded of RPPS are well below the RoE of banks. This is due primarily to the redeemable feature. 7. Recognising RPPS as AT1 capital thus has the potential to mitigate some of the lending rate impacts of increasing Tier 1 capital to 16% for the large banks, and 15% for all other banks, compared to having only CET1 or non-redeemable preference shares as AT1 capital. 8. We estimate that if systemic banks are permitted to recognise AT1 capital up to 1.5% of RWA (meaning 14.5% at least must be met with common equity), the average lending rate impact of increasing the Tier 1 capital requirement to 16% falls by 5 to 10 basis points (i.e. from an average of 32 basis points, for example, to 22 to 27 bps). RPPS can potentially be issued by most domestic banks 9. Currently RPPS are allowed as AT1 capital only if they have an added feature - contractual conversion to ordinary shares in the event a bank’s CET1 capital falls below 5.125%, or contractual conversion or write off in the event a point of non￾viability trigger is reached. 10. Contractual conversion is only possible for banks that take a particular corporate form or corporate structure. If we view corporate form as one dimension of competition, the current framework limits competition by giving only some banks access to Tier 1 capital other than owner equity and retained earnings. 11. Under the current regime only the large four banks, and Kiwibank using a questionable, complex, SPV structure, have been able to avail themselves of AT1 capital instruments. 12. Our recommendation is to accept simple RPPS as AT1 capital – i.e. RPPS that do not have contractual conversion or bail in. The RPPS will be able to be issued by most, perhaps all, domestic banks as well as the large four banks. The proposal thus offers the prospect of a more level competitive impact compared to the AT1 capital accepted today. Residual risk with RPPS 13. Because there are features that make a RPPS look like debt to investors, issuers can be under pressure to behave as if these shares are debt, endeavouring to pay

1 Refer IAS 32.

4 dividends in all circumstances (rather than exercise the right not to pay), for example, or redeeming on a scheduled call date rather than extending.2 14. RPPS are thus not risk-free from a financial stability perspective:  Funding that has the potential to absorb losses may be paid out inappropriately by banks (e.g. in hindsight the redemption may leave the bank with insufficient capital) (‘inappropriate repayment’ risk); and  By exercising its discretion to not redeem a bank experiencing stress may signal bad news, prompting a widespread loss of confidence in the bank, thus making bad news worse (‘signalling risk’) 15. Measures that mitigate against these risks are available but none reduce the risk of inappropriate repayment or signalling risk to zero:  We could aim for the capital position to be no worse after redemption than it is today (this approach suggests a tolerance for banks actively using the conservation buffer). This would imply requiring banks to issue replacement capital in advance of (or simultaneously with) redemption; or  We could aim for the capital position to be above regulatory requirements after redemption (suggesting no tolerance for banks remaining in the conservation buffer for any length of time). Banks would be required to produce reliable capital projections showing adequate capital. 16. These measures are similar what currently exists under Basel III. The difference is the condition will relate to a higher level of capital. For example, taking the second mitigating condition, given the minimum ratio that will apply is 16% Tier 1 (not 8.5% as under Basel III) the consequences of the accounting measures overstating the actual capital position may not be as severe as they would otherwise be. Hence we see either of these possible mitigating measures as providing reasonable protection from inappropriate repayment risk. 17. We have been discussing these mitigating measures with banks. All have agreed that conditions are necessary and indeed expected by investors. Mitigating measures addressing inappropriate repayment risk are envisaged in the terms sheets for RPPS proposed by BNZ and the domestic banks. 18. The mitigating measures outlined above would potentially have implications for dividends payments from RPPS, not just redemption, and this in turn would have implications for dividends payable on ordinary shares. This aspect was no surprise to banks and many said that this simply mirrors current practice – if there is a prospect Tier 1 capital will fall below the regulatory minimum (16% if the proposal is adopted) investors will not expect dividends on ordinary shares to be paid. 19. In terms of mitigating signalling risk, one approach is to aim for well-functioning secondary markets in bank funding instruments. Tools to help foster well-functioning markets have not been included in the consultation thus far and are not being proposed at this stage, but for completeness include listing capital instruments. A specific policy may not be required as every bank has indicated that any RPPS sold locally would be listed. This is an area of future policy research.

2 Banks will be incentivised to redeem the preference share even if they are low on capital, because it may signal bad news to other creditors.

5 Marketability of RPPS 20. Views among banks vary as to the marketability of RPPS. Some (e.g. BNZ, NAB) say there is existing demand offshore for appropriately structured RPPS.3 In contrast ANZ and ASB say the RPPS would be strictly a local product. All agree that there is strong local wholesale and retail demand for higher yielding bank funding instruments. Preference shares versus debt as AT1 21. All banks if given the choice would prefer to be allowed to recognise redeemable perpetual debt as AT1 rather than RPPS. This is because redeemable debt would be cheaper to issue offshore than RPPS and the investor base may be somewhat wider. 22. However we do not propose accepting legal-form debt as AT1 capital – just RPPS. This is because in our view RPPS offers more certain loss absorption than debt and, unlike debt, RPPS provide Basel III-compliant AT1 without having the complexity introduced by contractual bail in. 23. In our view the potential barriers to loss absorption on a going concern basis are greater for legal form debt than preference shares. These barriers relate to signalling risks, potential IRD claims, and residual fiscal risk. 24. Providing appropriate statutory powers exist, perpetual preference shares can be Basel III AT1 compliant without the need for complexity in the capital framework.4 Perpetual preference shares do not require contractual bail in. In contrast, in order to be Basel III compliant, perpetual debt would need to have contractual bail in when the CET1 ratio fell below 5.125%. Excluding contractual bail in 25. The RPPS we propose has no contingent terms. The instrument begins life as equity and remains equity throughout and at no point is the RPPS required to convert into an ordinary share. This is consistent with the December 2017 in-principle decision to only recognise simple capital instruments – i.e. none with contractual conversion to debt or write off (‘contractual bail in’). 26. The December 2017 in-principle decision to reject contingent bail in reflected a number of concerns, all of which we retain. We have concerns that contractual bail in:  brings significant added complexity to the regime for no added benefit above and beyond what can be achieved by the RPPS, the ESR and statutory powers;  is unreliable in terms of absorbing losses;  could in some cases undermine the effective operation of the ESR; and  has adverse unintended consequences including creating an uneven playing field between banks of varying corporate form and between foreign and domestic banks. 27. The interaction of contractual bail and the ESR is an important issue. Our analysis at this stage is limited by the fact that the ESR is as yet undeveloped. Hence our views at this stage are preliminary rather than fully developed. However, as the companion paper illustrates, there are complex, potentially adverse implications for the ESR.

3 Our own research indicates that perpetual preference shares issued by foreign banks can be listed in the US market, for example, and globally there are exchange traded perpetual preference share funds (for example some managed funds offered by Invesco). 4 Refer Basel Committee on Banking Supervision Press Release 13 January 2011 “Final elements of the reforms to raise the quality of regulatory capital issued by the Basel Committee”.

PAPER FOR Financial Stability Committee FROM Susan Guthrie (Financial Policy) DATE 20 September 2019 SUBJECT Background paper regarding AT1 capital FOR YOUR Information Summary Note: The purpose of this paper is to alert the Committee to the substantive content of submissions, external experts’ views and on-going feedback from banks in relation to AT1 capital, and to respond to the issues raised. Note: This paper is a companion to the decision paper provided to FSC about AT1 capital (refer paper #8524944). Note: Neither paper addresses Tier 2, nor the quantum of capital. Note: The feedback we have received in relation to what should count as AT1 capital can be usefully grouped into several key themes: • the pros and cons of accepting instruments with contractual ‘bail in’ (contractual conversion to ordinary shares); • the relative merits of perpetual preference shares and perpetual subordinated debt as AT1 capital; and • the pros and cons of allowing redeemable perpetual preference shares (‘RPPS’) as AT1 capital. Note: based on a detailed analysis of the feedback we have received we have a number of recommendations for FSC. These recommendations include one departure from the in-principle decisions announced in December 2017 – we now recommend accepting redeemable perpetual preference shares (‘RPPS’) as AT1 capital. Background

1. In-principle decisions relevant for AT1 capital were announced on 19 December
2. These decisions were to: a. not recognise instruments with contractual conversion or write-off (‘contractual bail in’); b. include non-redeemable perpetual preference shares as AT1; and c. include redeemable preference shares as Tier 2.
3. Final decisions on these issues are due to be announced in December 2019. Feedback provided through the consultation process
4. Capital instruments were discussed in 30 submissions provided in response to the capital ratio consultation paper released in December 2018 and mentioned in the final reports of all three external experts.

2 2 4. Providers of the 30 submissions included banks; parent banks, academic Martien Lubberink and the consulting group Sapere whose submission he co-authored; INFINZ; law firms Russell McVeagh, Chapman Trip and Buddle Findlay; and financial services groups Forsyth Barr and Harbour Asset Management. 1 An earlier consultation, held in 2017 specifically about the definition of capital, elicited 15 submissions. 5. In addition to this feedback, we have held further meetings with each of the large four banks, and the domestic banks as a group. This engagement has resulted in proposed terms sheets for redeemable perpetual preference shares and redeemable perpetual bonds, provided by BNZ and the domestic banks. 6. The external experts commented on capital instruments, focusing on contractual bail in. All three experts support the Reserve Bank’s in-principle decision to not recognise instruments with contractual bail in, with Dr Cummings suggesting we revisit the decision if/when large local banks list some of their ordinary shares. Overview of feedback 7. The majority of submitters argued for the status quo for AT1, which in practice means advocating for three instrument features – accepting contractual bail for AT1 instruments; accepting legal form debt as well as preference shares for AT1 capital; and accepting redeemable instruments as AT1 capital. 8. Submitters did not necessarily express a view on each of these features explicitly, but such a taxonomy provides a useful means of describing and explaining the views expressed and providing responses to those views. 9. In some cases views about contractual bail in were expressed in terms of the TLAC policies promulgated by the Financial Stability Board. As TLAC was only the vehicle by which views were expressed about contractual bail in, we have focused our commentary here on those views, not TLAC per se. Moreover TLAC compliant funding departs from Basel III regulatory bank capital in several ways – for example, in terms of minimum maturity. As the focus here is on bank capital we have not included discussions about aspects of TLAC other than contractual bail in. Issue 1: Should contractual bail in be accepted in AT1 instruments? 10. The December 2017 in-principle decision to recognise only non-contingent instruments reflected four concerns: a. Complexity: Contingent instruments introduce complexity to the regime, which makes the regime costly to comply with, and administer, and vulnerable to arbitrage; b. Uncertainty: the limited international evidence that exists suggests contractual bail in cannot be relied on to deliver going concern capital. Moreover, in New Zealand, the performance of contractual bail in instruments is further

1 The capital definition-related content of these 30 submissions, and the external experts’ reports, are compiled in an Excel workbook for easy reference “2019.09 Capital definition related submissions”, documentum #8517283.

3 3 complicated and potentially hindered by the trans-Tasman structure, and unlisted nature, of systemic banks; c. Unintended consequences: When contractual bail in is present not all banks can issue AT1 capital nor report the same capital value for a given amount of instrument issuance. This is because of the way different corporate forms interact with the tax system. The inclusion of contractual bail in thus incentivises banks to adopt a common corporate form, rather than innovate and compete through corporate form (for example, adopting mutual structures); and d. Redundancy: Resolution of a non-viable bank does not require contractual bail in when statutory powers exist (i.e. the latter can be considered sufficient). Unlike Australia, New Zealand legislation imparts statutory powers to bank regulators and thus contractual bail offers is unnecessary. Moreover, as will be explained below, allowing contractual bail in the framework could complicate and/or impede the use of supervisory powers to prepare banks for resolution. 11. The recent proposal to introduce the Escalating Supervisory Response (ESR) introduces an important new consideration to the issue of contractual bail in. Upon reflection we have formed the view that: a. an effective ESR has the potential to deliver more capital reliably, and in a timely way, to going concern banks than contractual bail in; and b. accepting contractual bail in when an effective ESR exists, has the potential to impede the exercise of the ESR. 12. An example where the ESR could outperform contractual bail in is when a viable banks needs more ordinary share capital. Under the ESR in theory banks could be instructed to issue new share capital when CET1 remains quite high (for example 12% of RWA), as part of their recovery steps. At this level of capital it is likely there will be willing equity investors and new share capital will be forthcoming. In contrast, to the best of our knowledge, no banks globally have been willing to issue contractual bail in instruments with a CET1 trigger of 12%. And this should be no surprise – the theoretical justification for AT1 contractual bail in is that it provide new ordinary share capital at a time when no willing providers of new ordinary share capital can be expected – i.e. when the bank is close to failing. In other words, the ESR has the potential to increase the amount of high quality in going concern banks sooner, and more reliably, than contractual bail in. 13. The above example can also be used to illustrate how contractual bail might impede and lessen the effective operation of supervision and the ESR. Imagine a bank that following an unexpected loss has CET1 capital of 12%. When weighing up whether or not to acquire newly issued ordinary shares in the bank new third party investors can be expected to consider the implications of any debt with contractual bail in the bank may have issued. If this debt has a trigger of, say 9%, the third party investors will be concerned that one further relatively small shock could see their newly acquired stake significantly diluted. So despite the bank being clearly viable, and its new equity needs relatively modest, the bank may struggle to find genuine new equity because of the presence of existing debt with contractual bail in.

4 4 14. It is also possible to imagine other situations where the inclusion of contractual bail in could impede the effective operation of the ESR. For example, a bank may be approaching a condition of non-viability and the supervisors have required, and are overseeing, an orderly transfer of assets to other banks or entities. If, at this point, some creditors are bailed in contractually, the transfer process may be challenged by the new shareholders who have an interest in retaining as many ‘good’ assets as possible. The interests of the new shareholders will not necessarily align with what is needed, at this time, to ensure stability of the financial system. 15. The above examples are illustrative only. The ability to comprehensively analyse the implications of contractual bail for the ESR, at this stage, is limited by the fact that the ESR is as yet undeveloped. Hence our views at this stage are preliminary rather than fully developed. However the key point here is that there are complex and potentially complex adverse implications for the ESR as a result of having capital instruments with contractual bail in. Despite its importance, this issue has not been raised, nor analysed in depth in the submissions. 16. In terms of the length of written text we received about capital instruments, the issue of contractual bail in was the most commented on. No new arguments of substance were made by those who support contractual bail (i.e. essentially the same issues that were canvassed in the earlier consultation were raised again). These views are briefly outlined below. 17. A common theme amongst those who support contractual bail in is that the status quo provides Tier 1 capital that is cheaper for banks to issue than ordinary shares and thus, by retaining the status quo, the economic impacts of raising Tier 1 capital can be lessened. It is important to note, however, that we are proposing an AT1 instrument without contractual bail in that will provide funding that is as cheap as what is currently accepted as AT1 capital. We can deliver on the low cost AT1 capital the banks want without accepting the complexity, uncertainty, unintended consequences and redundancy that accompanies contractual bail in. The feedback we have received is that redeemability, not contractual bail in, drives investor required returns from AT1 capital. 18. In some submissions the authors seem to have mistakenly assumed that the only option being proposed for Tier 1 capital was ordinary shares – these authors did not discuss the relative merits of preference shares with and without contractual bail in, for example. Feedback from Martien Lubberink, Sapere, and INFINZ 19. Associate Professor Lubberink’s personal submission focused on debt instruments that have contractual bail in. Lubberink also formed part of the Sapere consulting group, and appears to have played some role in the INFINZ submission (he is a member of INFINZ and represented INFINZ in a meeting with the Reserve Bank). 20. Lubberink claims that instruments with contractual bail in have been seen internationally to be “loss absorbing”. In making this argument Lubberink uses “loss absorbing” to mean banks have boosted their reported regulatory common equity capital (‘CET1’) by exploiting how CET1 is calculated. The regulatory value of CET1

5 5 reflects balance sheet values for debt capital, not market values. If the market value of debt capital is less than its book value, and retained earnings are used to buy the below-par listed debt capital, retained earnings will fall by less than the fall in the value of debt on the balance sheet (and thus CET1 will go up). In other words, as a result of repurchasing its own ailing listed debt capital instruments, a struggling bank can achieve a one-off boost to reported CET1 capital. 21. We do not find Lubberink’s arguments persuasive. This sort of ‘recapitalisation’ could be achieved by a bank repurchasing any of its listed debt that is trading below the value that applies when CET1 is calculated – it is not unique to debt with contractual bail in. Nor, moreover, does the ‘recapitalisation’ appear to alter anything about the underlying financial strength of the bank. In fact, one could argue that by spending retained earnings buying back what is otherwise perpetual funding, using the instrument to ‘recapitalise’ the bank has weakened the financial position of the bank. Feedback from law firms 22. Legal firms expressed the view that the contractual bail in instruments issued currently are legally certain. However our concern is not with a lack of legal certainty around conversion – rather a lack of certainty about the economic outcome of what appear to be largely certain legal terms. 23. Our concern about the economic outcomes relates to the recapitalisation of foreign￾owned subsidiaries. Given the terms and conditions of the instruments the Australasian banks have sought to issue under Basel III, there is uncertainty as to whether new shares will be issued in the New Zealand subsidiary when the contingent instruments are triggered. This uncertainty reflects several factors, including a lack of any firm basis for the price needed to convert debt into equity (a problem generated in part by the absence of listed share prices for New Zealand’s systemic banks). 24. There is also one area of legal uncertainty and that relates to the current requirement to ‘write off’ AT1 preference shares if conversion to ordinary shares fails. The means to give effect to ‘write off’ of a preference share (mandatory redemption for zero consideration) have, to the best of our knowledge, not been tested in the courts in New Zealand. Feedback from Australian-owned banks and their parents 25. At some point all of the large banks have called for contractual bail in, claiming benefits from aligning with APRA’s definition of AT1. ANZ has been particularly adamant in expressing this view in bilateral meetings whereas, at the other end of the spectrum, the BNZ has offered up prospective term sheets for simple AT1 instruments that have no contractual bail in. 26. The banks have said that aligning with APRA means they can issue capital to third parties, and have it recognised by APRA as group capital, and that this optionality is of value to them. In particular the banks have said alignment offers the following potential benefits:

6 6 a. a more diverse source of funding for the Australasian group as a whole during business as usual times; b. to the extent the NZ subsidiary uses the funding optionality, the subsidiary develops a capability for issuing in wholesale markets and a marketing presence among global investors; and c. in the event that the parent is unable to fund itself offshore, the NZ subsidiary could issue capital that is globally acceptable under its own name. 27. We do not find these arguments persuasive: a. the first of the banks’ arguments presumes the NZ subsidiary has access to investors not available to the parent, but this seems unlikely in most situations (there may be cost advantages in using the subsidiary rather than the parent to access the local market, but we would expect these to be relatively modest); b. if the local subsidiary is already sourcing non-capital funding from third parties offshore – as the large four do – the enhancement of capability and marketing presence is at the margin. Also, this enhancement will occur if subsidiaries issue any globally-accepted instrument, the benefit is not dependent on contractual bail in being included. The feedback we have received from BNZ and NAB is that instruments that omit contractual bail in, such as redeemable preference shares, are marketable offshore; and c. it is not necessary to offer contractual bail in in order to attract funding to the subsidiary when the parent has lost market access. The AT1 instrument we are proposing is conventional and well understood globally and are therefore in our view able to deliver standalone funding from diverse sources. 28. Consistent with these potential benefits being relatively modest, nearly 90% of AT1 issuance by the big four was to parents. If the funding optionality provided by the current regime was so potentially beneficial, why was it not exploited more? Feedback from domestic banks 29. The domestic banks’ views appear to have undergone some evolution during the consultation process. In their written submission the domestic banks called for the inclusion of both complex and simple instruments but our view is that, based on recent conversations, the domestic banks do support a regime made up of simple capital instruments. The domestic banks have provided us with proposed term sheets for AT1 instruments that omit contractual bail in. Feedback from external experts 30. The external expert’s comments about capital instruments were confined to contractual bail in features. All three experts provided qualified support for the in￾principle decision to opt for a simple framework that omits contractual bail in. a. Professor Miles’s view is that “the proposal to exclude contingent debt, at least initially, from the capital buffers required of banks has merits.” 2

2 Miles, David (2019). An Appraisal of the RBNZ Proposals on Bank Capital. 30 August 2019.

7 7 b. Dr Cummings, who has published extensively on contractual bail in instruments, took a similar stance but went one step further and recommended monitoring developments overseas and reviewing the role of contingent bail in at a later period: “The decision by the Reserve Bank to remove the contractual loss-absorption mechanisms from non-common equity capital instruments is reasonable, taking account of the uncertainty about the way in which the mechanisms can be used to recapitalise unlisted banks that are wholly owned subsidiaries of foreign parent banks. However the decision by the Reserve Bank places greater reliance on the powers of a statutory manager to restructure the claims of preference shareholders and subordinated debt holders…There may be limited sources of new equity available to a statutory manager to support the restructuring of a failed bank. To address this concern the Reserve Bank should continue to monitor the performance of the Basel III loss-absorption features in other countries and assess whether the mechanisms can be adapted to suit New Zealand’s circumstances. The monitoring should take account of the experience of the contractual loss-absorption features in countries with a significant presence of foreign-owned and unlisted banks. The contractual features that incorporate conversion to common equity may become feasible in New Zealand if one or more of the large banks opts to lists common equity on a public stock exchange.”3 c. Professor Levine concludes that, in terms of the in-principle decision to remove contingent debt from the capital framework,“given current conditions in New Zealand I found the arguments in favour of this conclusion prudent and persuasive”.4 5 Issue 2: should legal form debt be accepted as AT1? 31. Both domestic and large banks argued for the inclusion of legal form debt as well as preference shares in AT1 capital, pointing to the cost advantages of perpetual debt over preference shares when sold offshore, and the potentially wider investor base. 32. If sold locally it seems likely that preference shares will cost banks the same as perpetual debt. However preference shares will be more expensive than perpetual debt if sold to offshore investors who ascribe no value to the imputation credits that accompany preference shares (the cost difference is due to the imputation credit having to be augmented by an equivalent cash payment to offshore investors). 33. In responding to banks views it is helpful to consider why Tier 1 alternatives to common equity are being contemplated at all. Perpetual preference shares and perpetual debt offer the prospect of cheaper Tier 1 funding that common equity

3 Cummings, Dr James (2019). External Review of the Reserve Bank of New Zealand’s Capital Proposals. August 2019. 4 Levine, Ross (2019). A Report on the Reserve Bank’s Capital Review, commissioned by the Reserve Bank of New Zealand. August 2019. 5 Professor Levine is also of the view that, because uninsured debt-holders have incentives to limit risk-taking “well-designed Tier 2 capital can enhance incentives and complement Tier 1 capital.” ibid

8 8 (although this is dependent on the instruments being redeemable, a feature which is discussed in detail in the next section). Allowing alternative instruments to count as Tier 1 capital thus lowers the potential lending rate impact of increasing Tier 1 capital requirements. 34. A further benefit of allowing some alternative instruments in Tier 1 is that permitting wholly-owned subsidiaries of foreign banks to tap into local and/or international capital markets offers the prospect of these banks being able to raise capital independently of their parents (which is potentially of value if the parents lose access to markets).6 35. The in-principle decision to only include preference shares reflected concerns about potential barriers to loss absorption on a going concern basis being greater for legal form debt than preference shares. 36. In theory both perpetual preference shares and perpetual debt can be structured to absorb losses on a going concern basis. They achieve this by having fully discretionary and non-cumulative dividend or coupon payments, and the issuer having no obligation to redeem on an optional call date. However neither instrument absorbs losses on a going concern basis with certainty. 37. Failing to pay dividends or coupons, or redeeming on the pre-announced call date, can signal bank distress to depositors and senior creditors, potentially prompting a loss of confidence in the bank. Thus at the very point they act to absorb losses, preference shares and debt have the potential to exacerbate stress, rather than alleviate it. 38. Presented with this ‘signalling’ risk, banks have an incentive to pay dividends and coupons, or redeem, when the financial condition of the bank (which may be known only to management) dictates otherwise. Hence, in practice these instruments may not in practice be as loss-absorbing on a going concern basis as first appears. 39. In our view signalling risk is greater with legal form debt that preference shares. This is because latter are accounted for as equity on bank balance sheets, rather than a liability and, as a result, more typically understood to be at risk of loss. 40. Related to the above point, we believe there is potentially more residual fiscal risk with perpetual debt, were it to be accepted as AT1 capital, than preference shares to the extent AT1 capital is able to be sold to local retail investors. 41. There is also potentially added complexity in the event losses are eventually imposed on holders of perpetual debt. It is possible, depending on the ownership structure of the bank and whether the investor is related or not, that the loss may be deemed taxable income to the issuer, creating uncertainty about the loss absorbing value, in practice, of debt capital. In contrast, there is no prospect that investor losses on

6 Feedback we have received from BNZ and NAB is that appropriately structured perpetual preference shares are marketable offshore (ANZ has a very different view, saying they are strictly a domestic product). Our own research indicates that perpetual preference shares issued by foreign banks can be listed in the US market, for example, and globally there are exchange traded perpetual preference share funds (for example some managed funds offered by Invesco).

9 9 preference shares (accounted for as equity in the accounts) could give risk to an IRD claim. 42. It is also important to note that, providing statutory powers exist, perpetual preference shares can be Basel III AT1 compliant without the need for complexity in the capital framework. Perpetual preference shares do not require contractual bail in in order to comply.7 In contrast, in order to be Basel III compliant, perpetual debt would need to have contractual bail in when the CET1 ratio fell below 5.125%. 43. Views among banks vary as to the marketability of preference shares. Some (e.g. BNZ, NAB) say there is existing demand offshore for appropriately structured (redeemable) perpetual preference shares. ANZ and ASB have told us that preference shares are strictly a local product. All banks agree that there is strong local wholesale and retail demand for higher yielding bank funding instruments. Issue 3: should perpetual AT1 instruments be redeemable? 44. An in-principle decision made in December 2017 was to recognise only non￾redeemable perpetual preference shares as AT1 capital. The rationale was that redeemability introduces two risks: the risk that perpetual funding will be repaid (in order to avoid signalling bad news) when the financial condition of the bank suggests the funding should be retained; and related to this, the risk that bad news will be exacerbated by a loss of confidence in the bank when the option to extend the term beyond the optional call date is exercised (‘signalling’ risk). The in-principle decision was, in effect, to reject any instruments that absorb losses on a going concern basis with uncertainty. 45. The feedback we have received from a wide variety of sources is that redeemability lowers the cost of funding considerably compared to ordinary shares. We estimate that if systemic banks are permitted to recognise redeemable preference shares up to 1.5% of RWA (meaning 14.5% at least must be met with common equity), the average lending rate impact of increasing the Tier 1 capital requirement to 16% falls by 5 to 10 basis points (i.e. from an average of 32 basis points, for example, to 22 to 27 bps). 46. Given we are interested in the lending rate impacts of higher capital, we have an interest in lower cost alternatives to common equity and thus redeemability.8 We have thus weighed up the risks that come with redeemability and considered the potential mitigating measures we might introduce to keep these risks at tolerable levels. 47. In terms of mitigating the risk of inappropriate repayment there are several options available:

7 Refer Basel Committee on Banking Supervision Press Release 13 January 2011 “Final elements of the reforms to raise the quality of regulatory capital issued by the Basel Committee”. 8 When there is an optional redemption date, the implied price volatility for the instrument, which is a key driver of an investor’s required return, is based on the optional redemption date. The further the date, the higher the implied volatility and required return.

10 10 a. We could require the capital position to be no worse after redemption than it was before. This would imply requiring banks to issue replacement capital of the same or better quality in advance of (or simultaneously with) redemption. This approach would be consistent with allowing banks to use the conservation buffers for periods of time, and not penalising them for being in the buffer. b. Alternatively we could aim for the capital position to be above regulatory requirements after redemption. Banks could be required to produce reliable forward capital projections showing adequate capital after redemption for example (note, projections provide no guarantee of future capital positions). This approach would penalise banks for entering the conservation buffer (they would not be able to redeem the preference shares) and thus reflects a stricter approach to the minimum requirements than above. c. If, in practice, if bank boards have zero tolerance for entering the conservation buffer they will voluntarily issue replacement capital in advance of redemption if there is any risk the buffer will be breached. 48. These mitigating measures are similar in principle to the protections that exist in the current regime (under BS16 banks cannot redeem a capital instrument without Reserve Bank approval). However currently what the Reserve Bank would consider in making its decision is not always stated explicitly (for example, is it the minimum CET1 requirement that must be met or something higher). The term sheets for perpetual (redeemable) preference shares could state these requirements explicitly for example. In our discussions with banks there was general agreement that mitigating measures such as those outlined above would be appropriate and acceptable. 49. Note, the mitigating measures outlined above would potentially have implications for dividends payments from preference shares, not just redemption, and this in turn would have implications for dividends payable on ordinary shares. Preference share dividends would only be able to be paid if the bank is likely to have capital above regulatory minimums afterwards. If the preference share dividends cannot be paid, neither can ordinary share dividends (this is one sense in which preference shares are ‘preferred’). Hence ordinary share dividends cannot be paid if doing so would leave the capital position below regulatory requirements. This is similar to the present situation where dividends on ordinary shares cannot be paid if distributions have not been paid on AT1 instruments. This tying of ordinary share dividends to minimum capital requirements, via the proposed preference share, would need to be reflected in the Escalating Supervisory Response (which applies to all banks, not just those that have issued AT1 capital). 50. In terms of mitigating signalling risk, one option would be to aim for well-functioning secondary markets in bank funding instruments. In well-functioning markets investors collect, analyse and act on news about issuers in a timely and efficient way, thereby informing (through prices emerging from trades) all stakeholders in banks (including depositors and other creditors). In this way the implications of adverse news would not ‘sneak up’ on creditors and depositors – they would be alerted to it by capital instrument pricing movements - and the risk of surprise from non-redemption for creditors and depositors, and a disorderly loss of confidence, would be mitigated.

11 11 51. Tools to help foster well-functioning markets have not been included in the capital review consultations and are not being proposed at this stage, but for completeness here include having mandatory listing of AT1 and/or Tier 2 capital and potentially extending the current cap on related party holdings of AT1 and Tier 2 capital to parents. 52. Banks have indicated that in all cases any perpetual redeemable preference shares issued locally would be listed. This reflects strong local investor preferences for liquid, and thus listed, instruments. Some submitters indicated support for the listing of capital instruments as a means to mitigate signalling risk. 9 Others supported listing for reasons of local capital market development and instilling discipline among banks.10 Conclusion 53. We have reviewed submissions, considered the final reports of the three external experts and engaged in follow-up discussions with banks, financial service providers and INFINZ. This engagement has resulted in proposed term sheets for AT1 instruments being provided by the domestic banks as a group and BNZ. 54. A key concern for all banks is having access to Tier 1 capital alternatives that are cheaper than common equity. The recommendations listed below, and outlined in the decision paper, address this concern by giving banks access to AT1 capital in the form of redeemable perpetual preference shares - a conventional, long-established, funding instrument that has the loss absorbing properties of equity but is priced like debt. 55. A high level comparison of what is proposed with the status quo is provided in table form in Appendix 1. 56. In the decision paper we make a number of recommendations. These recommendations are as follows:

9 In their submission, Russell McVeagh pointed to the mitigating effects on signalling risks of listing capital instruments: “Signalling effect: The second benefit is the positive signalling effects of continuous disclosure that result from New Zealand banks issuing contingent capital instruments. An issuer of debt instruments that are quoted on NZX is subject to the continuous disclosure obligations of the NZX Listing Rules. Contingent debt instruments are more price sensitive than senior bonds, and so issuers of these instruments effectively are subject to enhanced disclosure obligations. Market analysts may also review and provide commentary on these instruments. This is likely to lead to a more informed market which would be a positive development.” (RmV reference: 3773306 v1). 10 For example, refer the submission from INFINZ.

12 12 Recommendations R1: We recommend accepting redeemable (or non-redeemable) perpetual preference shares (RPPS), with suitable protections in the contract terms, as AT1 capital. This is a change with respect to an in-principle decision announced in December 2017. Previously, the Reserve Bank signalled that it would only accept RPPS as Tier 2 capital and non-redeemable perpetual preference shares as AT1 capital. R1: We recommend that only RPPS that do not have contractual conversion or write-off (‘contingency’) be accepted as AT1 capital. This reconfirms an in-principle decision made in December 2017.

13 13 Appendix 1 A high level comparison of what is proposed with the status quo is provided in Table 1. Table 1. Significant cost advantage relative to CET1 Certainty as to going concern loss absorption Able to be issued by most NZ banks Complex or simple to administer, comply with Residual risk - inappropriate repayment yes yes Residual risk - signalling Interface with ESR Basel III compliant Marketable to local investors (retail and/or wholesale) Marketable to offshore investors Able to be sold to offshore parents APRA compliant (capital for Level 2 group if sold to 3rd parties) Unintended consequences for the wider ecosystem:

• different tax treatment for banks verus other business
• exposes banks to risk of miss-selling Current AT1 Proposed AT1 (RPPS) less certain more certain definitely yes yes possibly, but uncertain more complex less complex yes, higher risk than RPPS yes, lower risk than status quo complex, potentially adverse yes yes higher risk lesser risk yes no yes no qualified yes simple, complementary yes

MEMORANDUM FOR FSC FROM Susan Guthrie, Financial Policy DATE 20 Sep 2019 SUBJECT Public workshop project FOR YOUR Decision We recommend that FSC: Note: Kantar NZ was commissioned by the Reserve Bank to deliver three public workshops in August. These have now been completed and Kantar has provide three products for us: a final written report, a power-point presentation and a collection of brief videos (‘vox pops’) in which each participant provides their final feedback to us. Note: the key take-outs from the project are: a. once participants had sufficient knowledge to assess the policy’s impact on themselves and others, the level of engagement increased significantly; b. the participants themselves developed a ‘short cut’ way of thinking about the inherent trade￾offs, viewing bank capital as analogous to insurance. In terms of communicating to wider audiences, who have no opportunity to acquire new knowledge, insurance-type language may be helpful; c. when people are given the time and support to understand the policy by and large (Auckland youth being an exception) they express support for it; and d. Aucklanders aged under 30 were primarily concerned about the possible adverse impacts of the policy on rent and other living costs, and mortgage costs, with the benefits of higher capital being of less importance to this group. Note: This purpose of this paper is to make recommendations about i) the communications plan for this project, and ii) further work with Kantar. Agree: That we adopt the following forward plan: a. time any public statements about the workshop project to coincide with the announcement of final decisions; b. accompany any public statements we make about the workshop project with Kantar’s written report (having first liaised with Kantar to make sure no intellectual property belonging to Kantar is being inadvertently released); c. prepare for a possible OIA request in advance by identifying what, if any, further material could reasonably be released; and d. based on the expertise they have developed as a result of the workshop project, contract Kantar to provide a limited number of further outputs that we can use when communicating final decisions (for example, a short written narrative describing the policy that can be used to communicate with general audiences);

2 Background

1. Kantar NZ was commissioned by the Reserve Bank to deliver three deliberative workshops with members of the public on the topic of increasing bank capital. The workshops were held in August and involved 44 participants in total. Kantar has delivered a presentation of their findings on 6th September, provided us with an archive of video vox pops prepared by each participant outlining their final views, and supplied a comprehensive written report.
2. Kantar held the workshops in Auckland, Napier and Christchurch. A younger demographic was used in Auckland (primarily ranging in age from 18 to mid-30s), a mixed demographic in Napier and an older demographic in Christchurch. There was also diversity in terms of ethnicity and socio-economic status.
3. The workshops were ‘deliberative’, meaning they departed from traditional focus groups and did not seek to simply gauge current viewpoints. Each workshop took four hours and in this time participants were given an opportunity to acquire new knowledge and to deliberate the policy with each other. As a result of these processes, participants arrived at informed, and in many cases new, views.
4. The research project was qualitative, rather than quantitative. The research provides key insights into how the public engage with and think about the proposal, rather than quantitative findings. The insights provided by the research include: a. which aspects of the capital proposal matter most to people; b. what factors drive people’s viewpoints; and c. what types of engagement are most helpful for people in terms of understanding and forming a view about the policy.
5. Due to the limited sample size and the deliberative nature of the workshops, the results cannot be relied on to make strong claims about the level of overall current ‘public support’ for the policy. Unlike participants in the workshops, the general public typically have had no opportunity to spend an intensive four hours developing a deeper understanding of the capital proposal. Hence the views eventually formed by participants cannot be considered as representative of the current views of the general public.
6. Each workshop followed a common format. The workshop progressed through four stages beginning with technical information about bank operations (Stage 1), making linkages from bank insolvency to personal, social and fiscal risk (Stage 2), reviewing conflicting perspectives on the issue as expressed in submissions (Stage 3) and reviewing the personal trade-offs the policy presents for five constructed personals (Stage 4).
7. Kantar and the Reserve Bank worked collaboratively to provide the material used in the workshops. Key take-outs
8. In terms of the key take-outs, we find that: a. once participants had sufficient knowledge to assess the policy’s impact on themselves and others, the level of engagement increased significantly. Most participants were fully engaged by the end of four hours, and actively debating the pros and cons of the policy;

3 b. the participants themselves developed a ‘short cut’ way of thinking about the inherent trade-offs, viewing bank capital as analogous to insurance. In terms of communicating to wider audiences, who have no opportunity to acquire new knowledge, insurance-type language may be helpful; c. when people are given the time and support to understand the policy by and large (Auckland youth being an exception) they tend to express support for it; and d. Aucklanders aged under 30 were primarily concerned about the possible adverse impacts of the policy on rent and other living costs, and mortgage costs, with the benefits of higher capital being of less importance to this group. Next steps 9. We view the workshop project as a useful adjunct to the other stakeholder engagement that has been taking place. We thus see the insights it provides as an input into final decisions, alongside other feedback, external experts reports and so on. We thus believe we should make public statements about this project at the same time as the overall feedback is being discussed and responded to – namely when final decisions are being announced. 10. We have very satisfied with Kantar’s performance on this project and value the communications insights they provided. We believe there would be merit in contracting Kantar to prepare some material we could use to communicate final decisions. 11. In order to conduct the workshops effectively Kantar themselves had to develop a good working knowledge of the capital proposal. This coupled with their first-hand knowledge gained from their direct interactions with the 44 participants around what successful relatable messaging and tone looks like means they are in a good position to prepare communications resources that will be effective with a general audience. Recommendations 12. We recommend FSC approve the following: a. time any public statements about the workshop project to coincide with the announcement of final decisions; b. accompany any public statements we make about the workshop project with Kantar’s written report (having first liaised with Kantar to make sure no intellectual property belonging to Kantar is being inadvertently released); c. prepare for a possible OIA request in advance by identifying what, if any, further material could reasonably be released; and d. based on the expertise they have developed as a result of the workshop project, contract Kantar to provide a limited number of further outputs that we can use when communicating final decisions (for example, a short written narrative in relatable content and tone, plus imagery/graphics to help tell the story that can be used to communicate with general audiences);

Ref #8672748 v1.3 MEMORANDUM FOR FSC Chew Session FROM Financial Policy and Financial Systems Analysis (Primary Author: Paula Hontalba) MEETING DATE 24 September 2019 (FSC Chew Session) SUBJECT Potential impact of Capital Review on agri sector FOR YOUR Information Background

1. This note was used to brief the Minister of Finance (MoF) on the potential impact of the capital proposals on the dairy sector. Our key messages were: a. Irrespective of the Capital Review, banks have formed the view that current dairy debt levels are unsustainable and banks have in recent years tightened their lending standards and increased their margins. The purpose of this memorandum is to brief you on the potential impact of our proposed higher capital requirements on the agricultural sector. We recommend that you: • Note that the agri sector, particularly dairy, remains highly indebted. However, the dairy sector’s financial position has improved over the last 3 years, as evidenced by the decline in the proportion of high-LVR debt and increase in the proportion of dairy debt that is being paid down (rather than on interest only terms). • Note that the impact of higher bank capital on the agri sector will depend on many factors aside from capital requirements, including banks’ target sector allocation, competition, farmers’ access to other forms of financing, and the transition period. • Note that based on our estimates of an average lending spread increase of 20 to 40 bps and based on the current composition of the big four banks’ risk weighted￾assets, we estimate that on average, the farm lending spread could increase by 30 to 70 bps. High LVR loans are likely to experience higher repricing. • Note that we plan to cover off sectoral impacts, particularly on the agri sector and small businesses, as part of our Response to Submissions and the Regulatory Impact Statement (RIS). • Note that Financial Policy team have been meeting with various regional Federated Farmers groups to explain the Capital Review and how it might impact farmers, and to seek farmers’ views on the proposals. • Discuss how the Reserve Bank could potentially respond to points raised by banks, and concerns raised by the rural sector.

s(18)(c)(i)

10 Ref #8672748 v1.3 Next steps 26. As part of our Response to Submissions and the Regulatory Impact Statement, we plan to address the concerns raised by submitters regarding the potential impact of the capital proposals on specific sectors, such as agri and small businesses. In particular, we are doing some scenario analysis on how different transitions could affect different banks and sectors.9 27. We have also been engaging with members of Federated Farmers from various regions. Once we have completed our meetings with Federated Farmers, we will prepare a separate information paper for FSC. 9 Refer to “CRISP” spreadsheet model (#8476512).