Reserve Bank of New Zealand Regulatory Impact Statement on Bank Disclosure Review

Regulatory Impact Statement Bank Disclosure Review Agency disclosure statement This Regulatory Impact Statement (RIS) has been prepared by the Reserve Bank. It provides an analysis of the options for addressing issues with the current bank disclosure regime. The current regime produces unnecessary compliance costs for banks and there are weaknesses in the information that is currently provided to market analysts and retail investors. The analysis contained in this RIS is of a quantitative and qualitative nature. Direct compliance cost impacts (savings) have been quantified, while other impacts, such as the potential for enhanced market discipline, are analysed qualitatively. Stakeholder consultation has confirmed that the RIS focuses on the main policy options, that the analysis comprises all main impacts and that the estimated impacts are broadly realistic. The policy options considered result in net compliance cost savings. They do not • Impose additional costs on businesses • Impair property rights, market competition, or incentives on businesses to innovate or invest, or • Override fundamental common law principles.

1 Executive Summary 1 The Reserve Bank undertook a fundamental review of the disclosure requirements for registered banks, which has been in existence for a number of years. Under the current regime, banks are required to publish quarterly a Key Information Summary (KIS), a General Disclosure Statement (GDS) and a Supplemental Disclosure Statement (SDS). While the KIS aims to provide a high￾level overview of a bank’s financial condition and is targeted at retail investors (depositors), the GDS contains detailed financial information and the SDS provides background information. These three documents combined are supposed to provide financial markets, analysts and the general public with sufficient information to take well-informed decisions and thus support market discipline. 2 However, consultations with all stakeholder groupings (e.g. banks, accountants/auditors, analysts and the general public) have confirmed that over the years some inconsistencies and duplication have emerged and there are indications that the current regime no longer meets the needs of its users. Market analysts, for instance, point out that some key information is currently not available or disclosed in a confusing way, while copies of some key disclosure documents, such as the KIS or the SDS, are hardly ever made use of. In this context, the requirement to have printed copies of the documents available in all branches seems ineffective, and overly onerous and costly. To summarise, the current regime produces unnecessary compliance costs for banks and there are weaknesses in the information that is being disclosed. 3 This RIS analyses different options for eliminating these unnecessary compliance costs and better aligning the information banks disclose with the needs of users. The preferred option envisages: • Abolishing the KIS • publishing a comprehensive annual GDS • and a briefer GDS in the off-quarters and half year compliant with NZ IAS 34 (the accounting standard for interim reporting). • Removing the requirement that banks have to keep printed copies of the disclosure documents and replacing it with one whereby documents are printed off upon request. 4 The analysis in this report shows that the savings for big banks are likely to be between $ 99,000 and $ 140,000 per bank per annum. These savings include the freeing up of staff time and reduced printing and external advice/accounting costs. Smaller banks’ current compliance costs are less, which means that their expected savings are also less. It is estimated that their savings will be closer to $ 21,500 per annum. 5 In addition to these compliance cost savings, there are likely to be further benefits in terms of labour efficiency gains and better aligning information disclosure with user needs, which is thought to improve decision-making by investors and support market discipline.

2 Adequacy Statement 6 The Reserve Bank confirms and attests: • that the principles of the Code of Good Regulatory Practice and the regulatory impact analysis requirements, including the Regulatory Impact Statement requirements, have been complied with; and • that this RIS meets the adequacy requirements

Status Quo and Problem 7 Comprehensive disclosure is a cornerstone of the Reserve Bank’s regulatory regime for registered banks in New Zealand. The Reserve Bank views market discipline as an important complement to regulatory discipline and the disclosure regime aims to ensure that the market has the information it needs to exercise that discipline. The overriding aim of market discipline is to contribute to the maintenance of a sound and efficient financial system as stipulated by Section 68 of the Reserve Bank Act. 8 Banks are currently required to publish quarterly: • a Key Information Summary (KIS) providing a high-level overview of the bank’s financial condition; • a General Disclosure Statement (GDS) containing detailed financial information on all aspects of the bank’s business, including its conditions of registration and compliance with prudential requirements and; • a Supplemental Disclosure Statement (SDS) containing background documents such as guarantee contracts, and, for branches, the financial statements of their overseas banking group. 9 The Bank’s disclosure requirements for registered banks are set out in the Registered Bank Disclosure Statement Orders in Councils1 (OiCs). Although these OiC’s have been amended over the years to take account of changes in international accounting standards – i.e. to allow for the use of the New Zealand version of the International Financial Reporting Standards (NZIFRS) – and to incorporate Basel II Pillar 3 requirements, their substance has not changed fundamentally. As a result, some inconsistencies and duplication have emerged between the OiC, and NZIFRSs and Basel II Pillar 3.

1 There are four Orders in Councils for disclosure statements: Full and Half year New Zealand Incorporated Registered Banks; Off quarter New Zealand Incorporated Registered Banks; Full and Half Year Overseas Incorporated Registered Banks; and Off quarter Overseas Incorporated Registered Banks.

3 10 In particular, the Bank’s disclosure regime imposes requirements on banks that go beyond the NZIFRSs and Basel II Pillar 3 for the half-year and ‘off-quarter’ reporting periods. For the half￾year reporting period, banks are currently required to produce financial reports on the basis of NZ IFRSs full-year reporting requirements. Normally any entity reporting interim financial results (commonly at the half year) would do so on the basis of NZ IAS 34 Interim Financial Reporting, 2 which requires significantly less information than the full-year reporting standards. The following table illustrates the additional requirements due to the current OiCs when taking the NZIFRSs and Basel II Pillar 3 requirements as the benchmark (minimum) requirements. Requirements Annual Half year Off quarter Minimum requirement NZ IFRSs NZ IAS 34 Interim Financial Reporting

OiC additional - accounting Not much NZ IFRSs full-year compliance Spelt out in OiCs, not directly linked to NZ IFRSs OiC additional – Basel Pillar 3 such as capital adequacy Pillar 3 such as capital adequacy Pillar 3 such as capital adequacy OiC additional – other prudential Conditions of Registration, connected exposures and other prudential information Conditions of Registration, connected exposures and other prudential information Conditions of Registration, connected exposures and other prudential information 11 NZ IFRS contains the New Zealand-specific Appendix E specifying ‘additional disclosure requirements applicable to financial institutions’, including banks. Appendix E includes requirements that had previously been located in other reporting standards such as NZ IAS 30 and FRS 33. Over time significant overlap between Appendix E and the OiCs requirements has emerged, while some other requirements imposed by Appendix E are of limited value. Overall, a number of Appendix E requirements have become superfluous, raising the possibility that it would be more efficient to cut Appendix E as a whole, and preserve any useful Appendix E items that are not already duplicated in the current OiCs by adding them to the OiC requirements. 12 Although the current off-quarter reporting is largely due to Reserve Bank requirements, it should be noted that this arrangement ensures that banks are exempt from the continuous disclosure requirements under the Securities Act which would otherwise apply. The off-quarter requirements, therefore, already represent simplification and a lower compliance burden for banks. (Indeed, the Bank’s disclosure regime is designed as a ‘one-stop shop’ to minimise the overall compliance burden on banks.) However, it appears that some off-quarter reporting requirements have become superfluous and provide no or very little added value. 13 As mentioned above, banks are currently required to publish quarterly a GDS, KIS and a SDS, with the off-quarter GDSs being a shorter version of the on-quarter GDSs. There is substantial feedback from stakeholders to suggest that some of the disclosed information, especially the KIS and the SDS, is of little or no usefulness and perhaps obsolete.

2 This standard sets the minimum content of an interim report, including condensed financial statements and selected notes. It is to provide an update on the latest complete set of annual statements, and focuses on new activities and events, and does not duplicate information previously reported.

4 14 While banks have to display printed copies of the KIS in all of their branches, we understand from the banks that there is hardly ever a request for a copy of the document. Website hits are infrequent and it is not clear how many of those hits are from retail depositors, the target audience. However, it could be argued that when making investment (deposit) decisions it is rational for retail depositors to rely on the analysis provided by market commentators and financial journalists instead of carrying out their own. But, being a more sophisticated readership, professional analysts have different information needs and also do not consult the KIS. 15 What’s more, their feedback suggests that the current disclosure regime does not always match their requirements. Inconsistencies in the way in which information is reported or categorised across banks and a lack of historical information seems to impinge upon the quality of the analysis market commentators can provide and, through the decisions of retail depositors, on market discipline. 16 Quarterly disclosure requires documents to be prepared, or at least updated, and to have director / senior level sign-off, even when their information has not changed. This can make disclosure unnecessarily costly. For example, there appears to be minimal change in the SDS from one quarter to the next, yet the document requires costly and time consuming preparation and high￾level sign-off. Moreover, as with the KIS, there is no noteworthy demand for the SDS. Potential users either do not value the information sufficiently or they can obtain it elsewhere. 17 Implementation of the NZ IFRS and Basel II capital requirements has also led to some terminological issues in that the OiCs still refer to old standards that have been superseded. Hence, banks may still have to prepare information to a standard or format that has become redundant. This leads to confusion about what information to provide and in what format. 18 Somewhat related to this issue is the difficulty for market analysts and commentators to compare information across banks and to establish historical comparisons. Often banks interpret requirements differently and there is a lack of standardisation in the way information is disclosed. This has in the past led to avoidable errors in published analysis. The two main areas where a lack of standardisation impedes analysts’ ability to draw comparisons are residential mortgage lending and the breakdown of mortgage lending by loan-to-valuation ratios (LVR). Examples of the current difficulties include: • The term ‘residential mortgage lending’ may refer only to lending for owner-occupied house purchases or also include lending secured against a residential property for investment purposes or small business loans. • The use of a number of different terms all sounding similar but in fact referring to different things. Examples are “housing lending”, “residential mortgages”, “real estate lending”, “retail mortgages” etc. It appears that “retail mortgages” generally excludes business lending while “residential mortgages” may not but the distinction is not always clear in disclosure statements. • Sectoral credit risk concentration may include a line for some form of housing lending but this is not always so and sometimes only a total personal lending figure is provided.

5 19 Similar problems exist in the area of liquidity disclosure and the maturity ladder, where banks use different time buckets and insufficiently short-dated time bands. Exposure to different industry sectors is also difficult to assess across banks, as not all banks use the standard ANZSIC sector classification when disclosing the information, and of those that do, some do not use the latest 2006 version of the ANZSIC codes. 20 The following table summarises the issues we are dealing with in this RIS. Periodic Reports Issues Annual General Disclosure Statements (GDSs) There are inconsistencies and overlaps between the OiCs and NZ IFRSs. A material portion of the additional OiC requirements is driven by Basel II Pillar 3 capital adequacy disclosures. But among other requirements, some are of doubtful value, or are in fact inconsistent with NZ IFRSs. Half-year GDSs – based on full-year NZ IFRSs There are quite a few areas where the RBNZ’s imposition of full-year NZ IFRSs requirements on half-year disclosure is not warranted – such as full disclosure of accounting policies, risk management policies, detailed notes on tax expense, deferred tax, bonds and notes, management remuneration, retirement benefits etc. Some of these notes are lengthy but can be more or less copied from one period to the next, as little usually changes (although these clutter up the disclosure for its users). But some notes are costly to produce even though they are only one table (such as key management remuneration). Off-quarter GDSs The off-quarter OiC requires slightly more information than NZ IAS 34 Interim Financial Reporting. Some of that extra information is of value, but some may not be. Key Information Summary (KIS) It appears that the current content of the KIS is not useful, and no one uses it. For a short document, it is often filled with legal and technical jargon and hard to comprehend. Supplemental Disclosure Statement (SDS) A low-benefit item reserved for legal documents (e.g. guarantee contracts). No apparent demand for it. 21 The exact size of the overall compliance burden and the part thereof which has become redundant is difficult to measure accurately. At a very basic level, page count estimates might serve as a very rough first approximation – although it should be borne in mind that the compliance burden is not necessarily proportional to the number of pages in a disclosure document. The current page count of the quarterly disclosure documents is shown in the middle column (baseline) of the following table. Eliminating the SDS would bring the page count of the annual reports down by about 100 pages to approximately 100-130 pages. It is estimated that producing half-year GDS reports in line with NZ IAS 34 would reduce the page count to about 30-50 pages per publication, similar to the size of the current off-quarter GDS. A revamped KIS would continue to be about 10 pages long, meaning that half-year and off-quarter disclosure documents could be expected to be between 40 and 60 pages. Abolishing the KIS would reduce all page counts by approximately another 10 pages.

6 Baseline Minus SDS, new KIS and GDS (IAS 34) Annual 90-120 (GDS) + 100 (SDS)

• 10(KIS) = 200-230 90-120 + 10= 100-130 Half-year 80-130+100+10=190-240 30-50 + 10 = 40- 60 Off-quarter 30-50+100+10=140-160 30-50+10 = 40- 60 22 Banks spend considerable time preparing information that appears to be superfluous or duplicative. Although there is considerable variation in the estimates we received from banks in terms of the financial cost of this, a rough estimate suggests that it is well into the tens of thousands per year, and possibly as high as $ 65,000. 23 This figure does not include all the costs of the current regime. The costs of compiling and publishing superfluous information are not included. Moreover, to the extent that the information currently provided falls short of matching the needs of its users, there may be additional costs in terms of the quality of market analysis and consequently depositors’ investment decisions and market discipline. Measuring these costs in a meaningful way would be extremely difficult, if not impossible, and is beyond the scope of this RIS. 24 In summary, the current disclosure requirements are outdated in parts, which causes an unnecessary compliance burden for banks, and do not always match the needs of users, who therefore may not be able to fulfil their role as agents imposing market discipline as well as they should. Objectives 25 The overriding objective of the disclosure regime continues to be the facilitation of market discipline to support financial market stability and efficiency. 26 This is to be achieved in an efficient manner so that the information banks have to provide is not unnecessarily burdensome and a balance is drawn between the value of the information that has to be disclosed on the one hand, and the costs of producing and publishing it on the other. Expressed in economic terminology, this means disclosing information up to the point at which its marginal benefit is equal to its marginal cost. (It is of course illusory to expect to be able to measure MB and MC.)

7 27 The specific objectives of the present exercise are to enhance the efficiency of the disclosure regime by eliminating duplication and redundant disclosure requirements (i.e. by reducing the compliance burden), clarifying definitional issues and better matching the provision of information with the needs of its users, without however adding unnecessary compliance costs. 28 This is to be done by amending existing disclosure requirements as appropriate and within the existing basic framework. Thus a couple of high-level constraints apply: the NZ IFRS are taken as a given, except for NZ IFRS 7 Appendix E3 (referred to earlier in this RIS). The Financial Reporting Act 1993 requires audited financial statements to be published based on generally accepted accounting practices and it would be unusual, retrograde and ultimately detrimental for New Zealand to implement its own, idiosyncratic practices. Equally, the disclosure leg to the Basel II capital standards is known as Pillar 3, and it represents a widely-accepted international norm. It is difficult to see how non-compliance with its broad outlines would be in the long term interest of New Zealand banks, especially at a time when there is a global push for increased transparency and comparability of information. Hence, adherence to the NZ IFRS and Basel II Pillar 3 are a key objective that any amendments to the current disclosure regime must meet. Consultation 29 This proposal has been informed by substantial and multifaceted consultation with all key stakeholders as well as the general public. The underlying motivation for reviewing the disclosure regime came from feedback received from banks and accountants as well as anecdotal evidence about the questionable usefulness of some of the currently disclosed information. 30 The Reserve Bank established an industry user group (IUG), which brought together six big banks operating in New Zealand and Reserve Bank officials, and met twice (and corresponded on many occasions). The first meeting in November 2009 focused on the objectives and scope of the review and on eliciting initial industry feedback. The group met again in March 2010. The main purpose of that meeting was for the Bank to communicate and seek feedback on its initial analysis and to sketch out some high level policy options. Both group meetings proved useful for closing information gaps and discussing the detail of some options, such as revamping the KIS. 31 In January 2010 the Reserve Bank met with representatives of most of the main audit firms in New Zealand to gather their views on the disclosure regime. Throughout March and April meetings were held with the Securities Commission and the FRSB, and feedback was sought from consumer associations and financial journalists, and the three main credit rating agencies. 32 In August 2010 the Reserve Bank then published a consultation document seeking submissions from all stakeholders. The consultation document contained a detailed account of Reserve Bank thinking and concrete policy options for improving the disclosure regime. Altogether feedback was sought on 41 questions (as well as on any other issues submitters wished to raise). All key

3 Appendix E is not part of the Reserve Bank’s disclosure regime. However, since Appendix E adds additional New Zealand-specific disclosure requirements to those required by IFRSs, the New Zealand Accounting Standards Review Board (ASRB) is able to abolish it while keeping NZ IFRSs compliant with IFRSs.

8 stakeholders and stakeholder groups replied, including all major banks, three audit firms, three government agencies and two individuals. There was substantial agreement with the Reserve Bank’s analysis and its proposed way forward, especially amongst banks, but some differences of opinion emerged on a small number of issues. The main policy options were subsequently amended to reflect this feedback where possible. 33 After an initial analysis of the compliance cost savings, banks were again consulted to ascertain whether they agreed with that analysis. They generally agreed with the analysis but the smaller banks argued that their compliance cost savings were lower than calculated, while one of the bigger banks thought that its compliance cost savings from freeing up staff time would be significantly higher. 34 The options and the analysis of the compliance cost savings take into account all the feedback that the Reserve Bank received as part of the consultations. Options considered but rejected at an early stage 35 Before going into the detail of the options seriously considered for the present proposal, it is worth recounting other potential options that were discarded at an earlier stage after preliminary analysis found them wanting. It should be added that stakeholder consultation confirmed that none of these options warrant further consideration. 36 Maintaining the status quo: As argued above, there are substantial issues with the current regime. To briefly recollect the issues raised in the problem definition, they are as follows: • Overlaps and inconsistencies between the OiCs and NZ IFRS requirements • The disclosure of superfluous information • Disclosure documents not meeting the needs of its target audience (e.g. KIS) • Confusion about definitions and difficulty of comparing between banks 37 Given the significance of these issues, the a priori assumption is that the status quo is no longer tenable and should be changed as long as any such changes can be shown to lead to a net benefit compared to the status quo (or baseline to be precise). 38 Removal of the disclosure regime: The removal of the Reserve Bank’s disclosure regime would most certainly require banks to disclose information on the basis of the NZ IFRS and the Securities Act from which banks are currently exempt. Under such a scenario, banks would have to publish a registered prospectus for public debt offers and be subject to continuous disclosure requirements. The latter means that banks would have to inform the market of any material changes in their positions by updating their prospectuses. There are practical issues with this. Foreign owned banks listed on foreign stock exchanges are already subject to continuous disclosure abroad. Moreover, if a prospectus update was required, any deposits, including payments made to a bank’s own customers, would have to be placed in trust until a revised prospectus was issued. Indeed, continuous disclosure in accordance with the prospectus regime

9 was one of the original drivers for the Reserve Bank’s disclosure regime in the first place. Rather than reducing compliance costs, this option has all the potential of actually increasing them. 39 But the biggest drawback of this option is that it is unlikely to meet some of the key objectives of the disclosure regime. Reliance on the NZ IFRS and continuous disclosure requirements via prospectuses would not provide the market (and the Reserve Bank) with capital adequacy information in line with Pillar 3 requirements. The Reserve Bank’s disclosure regime in general is geared towards the release of information to the market on the risks of failure of a bank. This is so that market discipline can play an effective role in maintaining a sound and efficient financial system. NZ IFRS information on its own however has a slightly different emphasis in that it is primarily concerned with presenting a true and fair view of a bank’s historical position and performance at a specific point in time. While some of this information is useful for assessing a bank’s risk of failure, it is not sufficient on its own. 40 There are therefore strong reasons to believe that the abolition of the Bank’s disclosure regime would actually increase the compliance burden for banks, lead to the loss of valuable information and undermine market discipline, and thus adversely affect soundness and efficiency. The removal of the disclosure regime is not considered to be a realistic option. 41 Separate publication of financial reporting and prudential disclosures: A further, perhaps more realistic, alternative is to retain the financial reporting standards (NZ IFRS) and to require additional prudential information to be published separately. 42 This would mean annual financial reports based on NZ IFRS and half-yearly reports based on NZ IAS 34 plus Pillar 3 information. The latter would be in the form of an annual qualitative report and quantitative information disclosure at the half-year, with further brief quantitative updates in the off-quarters. If deemed useful, a revamped KIS could be published on a quarterly basis. 43 The attraction of this option is that it would align New Zealand with international practice. However, it would mean the release of two separate documents in spite of the strong linkages between the two sets of information. For instance, the NZ IFRS require an analysis of impaired assets, while Pillar 3 requires a breakdown of this analysis into separate Basel II credit risk exposure categories. Disclosing information in two separate documents could have implications in terms of transparency and user-friendliness. 44 Banks confirmed that they were not attracted to this option in initial discussion and again in their submissions when they concurred that this option should be dismissed. The Reserve Bank also considers it more helpful to disclose information in one document. It should also be stressed that this option is not about the content of information to be disclosed per se, but the vehicle for disclosure. It was decided to not pursue this option any further. Option under consideration

10 45 This leaves two options that can be considered as genuine alternatives. The purpose of this section is to describe and analyse these two options against the baseline, i.e. the status quo plus any foreseeable changes going forward. 46 Option A involves redesigning the KIS and making it the only off-quarter publication plus some changes to the GDS half-year and annual reports. The half-year GDS would be in line with NZ IAS34. Option B proposes the same changes to the annual and half-year GDS but a briefer GDS in line with NZ IAS 34 in the off-quarters, and removing the KIS altogether. Baseline 47 The status quo has already been explained above. The most likely sources of change to the status quo would be new international accounting standards or capital adequacy reporting requirements, and changes to the domestic regulatory environment not initiated by the Reserve Bank. 48 Although it can never be excluded that some amendments may be made to international accounting standards and capital adequacy reporting requirements, bearing in mind the emergence of a new capital adequacy regime as proposed by the BIS and dubbed Basel III, it is unlikely that there will be any fundamental developments affecting the status quo any time soon. Likewise the establishment of the Financial Markets Authority and other ongoing regulatory developments in financial markets are not likely to have a significant impact on the Reserve Bank’s existing disclosure regime. It is therefore reasonable to assume that the status quo can also serve as our baseline. Option A 49 Under this option, the KIS will be redesigned to be the only off-quarter document, the half-year GDS brought in line with NZ IAS 34 and the full-year GDS modified to eliminate inconsistencies and overlaps between the OiCs and the NZ IFRS: • Annual – New KIS; modified GDS • Half-year – New KIS; streamlined GDS (based on NZ IAS 34) • Off-quarter – New KIS only 50 A main feature of this option is the redesign of the content and structure of the KIS to make it more useful for retail depositors and so that it can serve as the only off-quarter disclosure document. This will involve expanding the KIS to include more financial statement items (Profit and Loss and the Balance Sheet) and certain ratios currently reported in the off-quarter GDS. 51 The other key change under this option affects the half-year GDS, which will be substantially trimmed down. The full-year NZ IFRS and comprehensive additional prudential disclosure requirements will no longer apply and be replaced with either something similar to the current off-quarter disclosure requirements or based on NZ IAS 34. The latter, i.e. NZ IAS 34 based disclosure, is strongly favoured by the industry. Option B

11 52 This option proposes to remove the KIS altogether and to publish a brief GDS based on NZ IAS 34 in the off-quarters. Changes to the full-year and half-year GDS are the same as under Option A. Essentially this option consists of the following: • Annual – modified GDS (as under A) • Half-year – streamlined GDS (NZ IAS 34; as under Option A) • Off-quarter – brief GDS (based on NZ IAS 34)

53 The rationale for removing the KIS altogether is that for the 14 years that it has been in place it does not appear to have delivered what it was intended to achieve. It is largely ignored by its target audience, retail investors, who are more likely to draw on information and analysis provided by market commentators, who are, however, by and large more sophisticated users with different information needs. And financial journalists canvassed have confirmed that they do not currently make use of the KIS. 54 A variant of Option B which was also considered would have further reduced off-quarter disclosure requirements. However, it would have meant non-compliance with NZ IAS 34 in the off-quarters and was therefore rejected by the banks, who favour NZ IAS 34 compliant disclosure if off-quarter disclosure is retained. Issues common to both options 55 Options A and B differ in regards to the main, high-level issues such as which disclosure documents to retain or how to modify them. But there are also a range of issues which are dealt with in exactly the same way under both options. Listing them as separate options would make this RIS unwieldy and produce no additional benefit. But that is not to say that there are no choices to be made and alternative ways in which the problem can be addressed. These choices have been analysed but their solutions do not differ across the two main policy options presented here. 56 For instance, the issue of insufficient standardisation discussed in the problem definition above is dealt with in the same way under both options. But the two most noteworthy changes to the existing regime that do not differ between Option A and B are the proposals to abolish the SDS and the Reserve Bank’s support for the removal of Appendix E of the NZ IFRS. 57 Both options propose to abolish the SDS given the lack of demand for this document. Unlike the KIS, where a revamped version of that document can be contemplated and indeed is one of the key differences between the two options for change, it is difficult to see how the SDS could be amended to make them more relevant. Its information is often largely static, i.e. does not change very much from quarter to quarter, and mostly available from other sources, e.g. the financial statements of a NZ branch’s overseas banking group. 58 Moreover, it is proposed to eliminate duplication between the current OiCs and NZ IFRS as well as unnecessary reporting requirements. Separately from the Reserve Bank’s disclosure review, the ASRB is expected to remove Appendix E from NZ IFRS 7 in relation to registered banks.

12 (This follows consultation by the Financial Reporting Standards Board for its own purposes on removing Appendix E, which the Reserve Bank strongly supported.) Options A and B have both been developed on this basis, and this has helped remove overlaps, at the same time as avoiding the loss of some valuable information by transferring a few useful elements of Appendix E into the OiCs.

59 This section analyses the costs and benefits of options A and B against the baseline. Where at all possible, the analysis is done quantitatively and impacts are monetised. It should be stressed, however, that we have had to rely on stakeholder estimates and that these estimates are often incomplete and vary widely across banks. Any assumptions are clearly spelled out and we report ranges rather than point estimates. However, it is important to keep in mind that the quantitative analysis is illustrative and should be treated with a degree of caution. A fuller understanding of the costs and benefits (or pros and cons) of the two options requires taking into account the qualitative analysis which complements the quantitative part. Costs and benefits Costs 60 The costs of the proposal are mainly in the form of the loss of availability of information and the added compliance burden of any new or modified reporting requirements that go beyond existing obligations. 61 Taking the latter first, cost increases due to new obligations may occur in the following areas: • A prescribed set of time buckets (to replace the bank’s own choice of time buckets) in the ageing analysis of past due assets, and in the interest rate re-pricing schedule. • The requirement to include agriculture as a separate row in the sectoral breakdown of credit exposures. • In cases where a bank’s disclosure statement includes a variety of different figures in the general area of mortgage lending (see paragraph [17] above), a reconciliation to be provided between those figures. • A requirement to include an “on demand” time band in the maturity analysis of assets and liabilities for liquidity purposes. • Banks that are accredited to use internal modelling approaches to calculate their capital requirements under Basel II will in future have to disclose their solo capital adequacy ratio using the Basel II framework rather than the superseded Basel I framework.

62 The consultation asked banks to submit ‘hard’ information on the likely impact of the proposed changes on compliance costs. As alluded to above, the information provided was somewhat sketchy and estimates varied widely across banks. Moreover, the information supplied generally did not distinguish between the impact of new requirements and the reduction due to the elimination of existing ones, and referred to the combined impact of the proposed changes instead. Any increases in costs as a result of new requirements are therefore already included in the calculations presented in this RIS (see ‘benefits’ section).

13 63 Both options contain measures generally aimed at eliminating reporting requirements that have become redundant, have never matched the needs of their users in the first place or are superfluous as they are currently being reported more than once. Hence, any proposals to discontinue the disclosure of certain types of information should not result in the loss of valuable information: the same or comparable information will either continue to be available from other sources or, where that is not the case, it is not currently being used by recipients. Second round negative impacts on market participants’ ability to make well-informed decisions, and thus exercise market discipline, or efficiency are therefore unlikely. Benefits 64 Benefits take the form of a reduction in compliance costs and the better servicing of the information needs of users. Better matching information disclosure to the needs of users should lead to better informed decisions by market participants, enhance market discipline and support the soundness and efficiency of the financial sector. Quantifying this benefit is extremely challenging and beyond the scope of this RIS. A qualitative analysis appears more appropriate. This qualitative analysis is additional to the quantitative cost benefit analysis of the reduction of compliance costs. 65 Compliance cost reductions fall into three different categories: a reduction in the cost of buying in external expertise, e.g. advisers and accountants, a reduction in the costs directly associated with disclosing the information, e.g. printing, and the freeing up of staff time for other purposes. It is the latter in particular which has the potential for productivity gains. 66 Freeing up staff time means that they can spend more time contributing to a bank’s core output functions, thus enhancing labour efficiency / productivity.4 In the long run, this is likely to be the most significant benefit for banks. 5 It should be pointed out that any such effect rests on a number of assumptions, the most important one of which is that staff spend the extra time available to them on productive activities. 6 67 The potential for efficiency gains points to the need to distinguish between eliminated disclosure requirements that actually free up bank staff time for other tasks, and changes to the regime that affect the way in which or to whom information is disclosed, while leaving the actual work of compiling the information unaffected. The latter may lead to financial savings, as does a reduction in the demand for external expertise, but it has no direct effect on labour efficiency. 68 The reduction in the costs of having to buy in external expertise, printing and distribution of disclosure material can be estimated in monetary terms with the information provided by banks in

4 These effects may not be negligible. The European Commission calculated in 2005 that reducing the compliance burden across the EU by 25 percent would increase the EU’s GDP by c. 1.7 percent (REF here) 5 It should be stressed that we are primarily talking about a level increase in productivity and not an increase in a bank’s productivity growth rate as such. 6 Assuming that staff levels do not change, if staff simply extend their break times or there are no opportunities for them to contribute to productive tasks then there won’t be a positive effect on labour efficiency.

14 their submissions to the consultations. Although more difficult, the reduction in the amount of staff time spent on disclosure requirements can be analysed through use of the so-called Standard Cost Model. The labour efficiency gains can in theory be modelled but data limitations and the costs associated with such an analysis put it beyond the scope of this RIS. 69 The Standard Cost Model, originally developed in the Netherlands, is a standard tool for measuring the costs of reporting requirements. Although not difficult to use as such, obtaining the necessary data can be challenging. Calculating the costs associated with meeting reporting requirements (excluding any disclosure costs such as printing costs) requires data on the time employees spend complying with the requirements, the costs to the bank of employing those employees (e.g. wages, overhead costs), how often a year the tasks have to be carried out and the number of entities in the sector. Calculating the reduction of compliance costs: , (or CC = PxQ, where P = time x tariff; Q=frequency x number of affected units) ; and For the sector Compliance costs (unnecessary) = PxQ, where P = time x tariff; Q=frequency Per institution 70 While information as regards frequency and number of affected units is fairly straightforward to obtain, the same cannot be said when it comes estimating the tariff, employee costs for the bank, and time bank staff spend complying with information and disclosure requirements. Banks were asked in IUG meetings and in the consultation on the policy options to provide us with the necessary data. 71 Of the eight banks that made a submission to the consultation document, four expect compliance cost savings other than from printing and external accountants to be minimal. They argued that the changes would merely spread the workload more evenly across the year or that they would continue to collect the information as part of the overseas disclosure requirements of their parent. One bank said that the changes would lead to significant compliance cost savings but it did not quantify the savings. The remaining three banks provided some data, with one bank providing data on staff time savings from changing the half-year requirements and the other two focusing on off-quarter time savings. As all banks favoured Option B over Option A, their data seemed to refer to savings from adopting Option B as against the baseline. 72 As all banks had previously concurred that the proposed changes would lead to real cost savings and given that the focus of our analysis are the compliance costs imposed by Reserve Bank requirements, and not internal processes or overseas requirements which are outside the scope of influence of the Reserve Bank, we have used the quantitative data provided by some banks to extrapolate across other banks.

15 73 Based on this data and Reserve Bank assumptions and estimates, initial calculations of the compliance cost savings were made. Banks were then again consulted on those initial calculations and specifically asked whether they agreed or disagreed with the calculated savings, could provide further estimates of cost savings and whether there were any other impacts not yet captured by the analysis. All banks who replied to this specific consultation said that they found the calculations broadly realistic, although some smaller banks felt that the figures might overstate their expected savings. One of the big banks argued the opposite, saying that it expects significantly higher savings from freeing up staff time. The calculations have subsequently been amended to reflect this feedback. 74 The following assumptions underlying the compliance cost savings are based on stakeholder and Reserve Bank estimates: Option B • hourly tariff of $ 100 to take account of non-wage labour costs, overheads and the mix of staff seniority involved in compiling disclosure information; • savings from moving to NZ IAS34 based disclosure for half-year reports, no KIS: 200 hrs per bank; • savings from NZ IAS34 disclosure and no KIS in off-quarters: 20 hrs per bank; and • reduced accounting fees $ 15,000 per institution per annum • reduction in printing costs $60,000 per institution per annum 75 For six banks who also made up the industry user group7 , this leads to the following average annual compliance cost savings: • Compliance cost savings at the half-year: 1002006 = $ 120,000 • Savings in the off-quarters: 201002*6 = $ 24,000 $ 144,000

76 When adding the financial savings from having to print fewer disclosure documents and reduced external accounting fees, the benefits for these six banks increase by a further $ 450,000. • Savings on accounting fees: 150006 = $ 90,000 • Printing costs 600006 = $360,000 • Compliance cost savings from above = • Total annual savings for big 6 $594,000 $144,000 This gives an estimated average compliance cost saving per IUG bank of $ 99,000 pa. 77 But this central scenario may overstate the savings for the two smaller of the six IUG banks. So far, we have treated all six IUG banks equally, although clearly there are differences between

7 Westpac, ANZ/National, BNZ, ABS, Kiwibank and SBS.

16 them, not least in terms of size which may affect the number of printed disclosure documents that they need to produce. Differences in organisational structures, information collection etc also have an impact on the number of hours staff spend preparing disclosure information. It appears less burdensome for the two smaller of the six banks which made up the industry user group to comply with current disclosure requirements so that their savings may be smaller than the average figure of $ 99,000 per bank. 78 And savings for branches of overseas registered banks and other smaller players in the sector might be lower still, since they currently print fewer copies of the disclosure documents and their less complex accounting needs and internal structures render these functions less costly, too. Feedback from this segment of the sector confirmed this view. Based on that feedback, we assume that accounting and printing costs are both significantly less than those incurred by their bigger competitors, at $ 5,000 and $ 4,500 respectively. We further assume time savings for staff are 50 percent of those for bigger banks at both the half-year and the off-quarters. Thus we calculate the savings for smaller banks as follows: o Compliance cost savings at the half-year: 100100 = $ 10,000 o Savings in the off-quarters: 10010*2 = $ 2,000 o Reduction in accounting fees = $ 4,500 o Reduction in printing costs = $ 21,500 $ 5,000 79 On the other hand, one of the big four banks estimates significantly higher time savings from the proposed changes. According to this bank, its time savings are likely to be closer to 650 hrs per annum. Thus its estimated overall savings will be approximately $ 140,000 per annum. 80 Thus we expect direct compliance cost savings for the bigger banks operating in New Zealand to be between $ 99,000 and $ 140,000 per bank per annum. For some smaller players in the sector the savings could be significantly lower, at approximately $ 21,000 pa. However, it should be borne in mind that their existing compliance costs are also likely to be significantly lower. Additional benefits 81 But these figures do not include all the expected benefits from the proposed changes. The calculated savings in staff time free up capacity within the institutions, and while the staff costs approximate the value of this staff time as measured by its opportunity cost, they do not include any further, second round efficiency gains. Such further efficiency gains would come from redirecting the extra capacity that becomes available towards productive business activities. For example, some banks in their submissions pointed out that the proposed changes mean that boards would have less to read and in general spend less time on signing off on disclosure documents. If this extra time was used to increase efficiency or innovation, then that would bring further benefits. At least one bank stressed that they view the extra time available to the board as a key intangible impact. 82 Nor do the figures calculated above take account of the potential to help strengthen market discipline through better-informed investment decisions on the part of retail investors. It is

17 generally held that retail investors lack a sufficiently sophisticated understanding of financial markets to exercise market discipline and that this is what has contributed to unwise investment decisions. Simply providing them with information on which they can make their investment decisions may not be enough. The information failure, which the current disclosure regime already addresses, is compounded by transaction costs to do with analysing and interpreting disclosure information. There are strong indications to suggest that retail investors (i.e. consumers) rely on market commentators to do this for them. Assuming that improving the information available to commentators enhances the quality of their analysis therefore could be reasonably expected to lead to better investment decisions on the part of retail investors/depositors that better reflect their risk return preferences. This should contribute to market discipline, a better allocation of capital and strengthen allocative efficiency overall. This in turn will contribute to one of the Reserve Bank’s core objectives, namely to promote financial system stability and efficiency. 83 These impact should materialise gradually and over a longer period of time, but it is extremely difficult and beyond the scope of this RIS to quantify them in a satisfactory way. Nevertheless, it is important to be aware of these further benefits to appreciate that the overall benefits are likely to be bigger than those calculated and monetised above. Option A 84 All banks preferred Option B to Option A. An NZ IAS 34 compliant GDS was favoured over a revamped KIS, not least because it offers greater clarity. It is also highly questionable that a new KIS would be used by retail investors, and the absence of the GDS in the off-quarters might provide financial analysts with insufficient information in the off-quarters. In other words, the benefits from Option B are arguably higher than those offered by Option A. 85 We estimate that the requirements under Option A would result in lower compliance cost savings. In addition to the streamlined NZ IAS 34 compliant GDS at the half year and annual stage, Option A would also require the publication of a revamped KIS. It is only in the off-quarters Summary of impacts Estimate of direct compliance cost savings for big banks $ 99,000 – 140,000 Estimate of direct compliance cost savings for smaller banks $ 21,500 Additional benefits (not quantified) • Better matching of information disclosure with user needs • Potential for better informed decision-making by investors • Contributes to market discipline • Potential for labour efficiency gains for banks

18 where a new KIS might offer some more flexibility than Option B. On the whole, however, it is likely that the savings would be less as compared with Option B, since throughout the year information would need to be collected for two types of document. 86 These strong indications of lower benefits and compliance cost savings, coupled with a clear preference by stakeholders for Option B (and a lack of hard data) suffice to conclude that the net benefits calculated and described for option B outweigh those that could reasonably expected from option A. Efforts to obtain hard, more robust data in order to carry out the same level of detailed analysis as done on option B are considered to be disproportionate. Conclusions and recommendations 87 This RIS has provided a proportionate analysis of different policy options for relieving banks of unnecessary reporting obligations and better matching the information banks disclose to the needs of its users. The analysis has shown that option B (see above) would lead to compliance cost savings for big banks of between $ 99,000 and $ 140,000 pa. Reflecting their smaller size and lower compliance costs overall, other banks are likely to benefit less, with the smaller players possibly by approximately $ 21,500 pa. In addition to these compliance cost savings, there are further benefits in the form of efficiency gains, extra clarity as regards the information that needs to be disclosed and disclosed information better meeting user needs. 88 Option B was found to produce the highest net benefit while the alternative, option A, would in all likelihood result in lower benefits and compliance cost savings. It should be pointed out that stakeholders, as far as they expressed a preference, unanimously favoured option B. Thus it is recommended that the disclosure regime is amended as outlined in option B. Implementation 89 Most banks have expressed the wish for the changes to be implemented before the next off￾quarter disclosure date at the end of Q1/2011. The Reserve Bank considers that an early introduction would produce valuable benefits and publicly indicated that it plans to work towards an implementation date of March 2011. 90 Although this is a slightly ambitious timetable as the new Orders in Council have to be drafted and the proposed changes published in the Gazette, the Reserve Bank considers that it is entirely achievable. However, the Reserve Bank notes that any delays in the upcoming procedures could mean that the benefits will not be reaped until the following quarter. Monitoring, evaluation and review 91 The success of the new disclosure regime will be monitored as part of the Reserve Bank’s ongoing communication and information exchange with the banking sector and other

19 stakeholders. A review might be carried out in due course, once enough time has passed for the changes to be implemented, or if any unexpected issues should emerge. Annex 1 Information summary: • What is the problem under consideration and why is intervention by the Bank necessary? • Current disclosure regime is outdated, leading to inconsistencies, duplication and unnecessary compliance costs • Also does not always meet the requirements of its users • Reserve Bank is responsible for bank disclosure regime through Orders in Council • Bank disclosure is aimed at overcoming information asymmetries to strengthen market discipline, a cornerstone of the Reserve Bank’s regulatory framework • What are the objectives? • To eliminate unnecessary compliance costs • To better align disclosure information with the requirements of its users • To ensure NZ continues to abide by international regulatory standards • What is the preferred option and what other policy options have been considered? • Preferred option consists of reducing the number of disclosure documents (only a quarterly modified GDS is retained, compliant with NZ IAS34 for the interim accounting periods); abolishing the requirement to retain printed copies of disclosure documents in branches; facilitating sign-off and updating of information; and providing market analysts with more consistent and user-friendly information • Further options considered were the baseline/status quo and an option that would have kept the KIS in all quarters and an annual and half-yearly GDS • Preferred options shown to produce greater net benefits and favoured by stakeholders • More radical options such as abolishing the disclosure regime altogether or publishing financial reporting and prudential disclosure information separately were rejected at an earlier stage. Stakeholder consultation confirmed that they also did not view them as realistic alternatives. • Who are the affected groups? • Banks, auditors, financial analysts, credit rating agencies, general public/retail investors • What are the monetized costs and benefits in NPV terms (low, high and best estimate) and by affected group? • For the big banks between $99,000 and $140,000 per bank per annum • For smaller banks approximately $21,500 • Other key non-monetised costs and benefits by affected group • Market analysts should benefit from better information • The general public/retail investors from better financial analysis • This should strengthen market discipline and the stability and efficiency of the financial system • There might be labour efficiency gains for banks from freeing up staff time. One bank in particular stressed the importance of the freeing up of board time that the changes would produce. • Key assumptions/sensitivities/risks • Underlying data for compliance costs savings calculation was provided by banks

20 • There was some variation in the estimates between banks and there is a risk that our estimates and extrapolation means that for some banks the savings are under/overestimated. However, we believe that the risks are pretty even either way and the reporting of ranges rather than point estimates increases the likelihood of being in the right ballpark. • We have assumed a tariff of $ 100 ph per employee. Higher/lower values would affect the savings. • Banks were specifically asked to comment on the underlying assumptions and whether our approached captured all important impacts. They generally agreed with the analysis and any refined estimates they provided are fully reflected in the final calculations. • (If no monetization/quantification) Why is a monetized CBA impractical or disproportionate? • Other information