Regulatory Impact Statement - Capital Requirements for Non-Bank Deposit Takers

IN CONFIDENCE 1 REGULATORY IMPACT STATEMENT RBNZ CAPITAL REQUIREMENTS FOR NON-BANK DEPOSIT TAKERS Executive Summary

1. Two Regulatory Impact Statements (RIS) have been prepared and attached to this Cabinet paper – one on Capital Adequacy Requirements, and another on Related party Requirements.
2. Capital and related party requirements are key components of the prudential regime for NBDTs. Capital regulation ensures that owners have sufficient funds at risk in the institution and thus have strong incentives to manage risks prudently. Excessive related party exposures can undermine this objective if owners and other related parties extract benefits from the institution which could have the effect of reducing the net invested finds at risk. The risk from having insufficient capital levels as well as excessive related party exposures could potentially undermine the stability of the financial system if they lead to failure or weaken depositors’ confidence in the system.
3. This RIS supports the introduction of a new capital adequacy framework for NBDTs. Two key elements are considered. These are the appropriate measurement framework and the appropriate minimum level of capital. On the measurement framework, the Reserve Bank’s preferred approach is to introduce a single industry-wide framework based on an amended version of the bank regime, calibrated to the specific risks faced by the NBDT sector. On the minimum level of capital, the Reserve Bank’s preferred option is for all companies with a credit rating to be required to hold at least 8 percent tier one capital, and for smaller companies who do not hold a credit rating to maintain tier one capital in excess of 10 percent of risk weighted assets. Adequacy Statement
4. Treasury’s Regulatory Impact Analysis Team has reviewed this Regulatory Impact Statement and considers it to be adequate according to the adequacy criteria. Structure of the RIS
5. This impact statement concerns the development of an appropriate capital adequacy framework for non-bank deposit takers (NBDTs). It is made up of four sections containing: • an overview of the existing industry structure and an identification of the problem definition; • a summary of the objectives and assessment criteria against which the Reserve Bank has assessed various options;

IN CONFIDENCE 2 • a description of the options considered by the Reserve Bank and a summary of the analysis supporting the preferred option. This section includes an assessment of the potential impact of the policy, a description of likely benefits arising, and a qualitative analysis of the risks and costs associated with the proposed requirements; and, • a brief summary of implementation and review issues, including a description of the consultation that the Reserve Bank has conducted during the development of the policy. Problem definition and status quo Problem definition 6. Currently NBDTs must comply with various forms of capital ratios1 7. The capital adequacy framework is intended to address these problems. It will help ensure that risk is more clearly and consistently reported and will help reduce the likelihood of failure by ensuring that all NBDTs have to meet a minimum capital ratio. which are set out in their individual trust deeds. These requirements are generally not consistent nor transparent and it is difficult for depositors and expert advisers to make judgements about how well an NBDT is capitalised. In recent years many finance companies were poorly capitalised relative to the risks they were running and this contributed to the high rate of defaults over recent years. This has led to a loss of confidence in significant parts of the sector. Status quo 8. NBDTs represent a small percentage of the deposit-taking finance sector in New Zealand. The legislation defines NBDTs as organisations that offer debt securities to the public and that are in the business of lending money or providing other financial services. In practice, this includes building societies, credit unions, and finance companies. As demonstrated by figure 1, in April 2009, banks accounted for around 97 per cent of total assets in the sector. Finance companies accounted for around 2 per cent, while building societies and credit unions each account for less than 1 per cent of total assets.

1 A capital ratio or capital adequacy ratio is a measure of a bank’s capacity to absorb losses that might arise through credit, operational and market risk.

IN CONFIDENCE 3 Figure 1 Breakdown of total assets within the deposit-taking finance sector 9. The lending profiles of the sub-classes of NBDTs vary within the sector. Figure 2 provides a breakdown of the aggregate lending profiles for each type of institution, including registered banks, as at April 2009. This analysis is based on data from all firms with assets greater than $100m. Figure 2 Lending profiles of NBDTs

IN CONFIDENCE 4 10. The graph above shows that: • building societies and banks have similar lending profiles with residential housing making up the majority of loans, and the rest of the lending is diversified; • while credit unions also have significant exposure to the residential housing segment, their greatest exposure is in the consumer and personal financing segment of the household market. Credit unions are the least diversified in their lending profile; and, • the lending profile of finance companies is different from the other institutions. In particular, finance companies have negligible exposure to the residential housing segment of the market, and instead have significant exposures to the property and business services segment of the market. 11. Under existing arrangements, NBDTs are required to operate in accordance with the terms set out in a trust deed. The NBDTs compliance with the conditions of the deed is supervised by a trustee. Most NBDT trust deeds already have some form of minimum capital adequacy ratio requirement to provide a constraint on an NBDT’s capacity to take risks that are excessive in relation to its capital. However, there are a number of problems with the trustee-based regime: • the trustee measurement frameworks are often unconventional, at times complex and difficult to understand, and vary from trustee to trustee; • some frameworks do not adequately capture NBDTs risks or took too permissive a view of what should count as capital; • even when the flaws in a capital adequacy measurement framework are apparent it is often difficult for trustees to renegotiate trust deeds to effect improvements; and, • a competitive market for trustee services can place downward pressure on standards. 12. As a consequence it is difficult for the market to assess the risk of NBDTs, with many adopting risky strategies that were not apparent to depositors and which resulted in wide spread failures with large economic losses, as demonstrated in table 1 below. Table 1: Impact of failures in the non-bank sector Household financial assets Failures and suspensions Banks $90bn - Managed funds and trusts $51bn 4% Finance companies and SIs $9bn 30% Total $150bn 3% Source: RBNZ calculations based on December 2008 data, “SI” means savings institutions (essentially building societies).

IN CONFIDENCE 5 13. At present, there is no requirement for trust deeds to meet a minimum capital adequacy standard. This will tend to generate capital ratios that are below socially optimal levels. An NBDT’s capital ratio will be set at a level that takes account of the private costs and benefits to the NBDT. It will ignore the costs of a default of an NBDT on the rest of the sector (through contagion2 14. Currently, the investing public has lost confidence in parts of the sector (particulary finance companies) and there is a risk of further losses of confidence if ostensibly well capitalised NBDTs run into trouble. In the absence of a prudential framework that the public can have confidence in, some depositors may refuse to invest in well capitalised NBDTs because traditional risk measures have proved so misleading and they do not have a reliable benchmark that signals prudent capital management. ) and on the broader economy if there is widespread failure in the sector. The effect of the deposit guarantee scheme will also generate socially sub￾optimal capital decisions. Even if it is fully withdrawn it will increase expectations of a future bailout of depositors should the sector get into trouble. If some of the future costs of failure are expected to fall on the Crown, NBDTs will run riskier strategies. 15. The underlying, high-level case for introducing a prudential regulatory regime for NBDTs was established prior to the implementation of the legislation. The primary concerns surrounded the absence of minimum entry requirements, inconsistency in governance and prudential requirements across NBDTs, inadequate official oversight of trustee supervision, inadequacies in public disclosures, and insufficient means for investors to assess and compare NBDT risk profiles. 16. In September 2008, the Reserve Bank of New Zealand Amendment Act (“the Act”) was passed. One of the key changes to the Act was the inclusion of a new section (Part 5D) which outlined the new responsibilities for the Reserve Bank as the prudential regulator NBDTs. The new legislation introduced a number of potential new regulatory requirements intended to: • promote the maintenance of a sound and efficient financial system; or, • avoid significant damage to the financial system that could result from the failure of a deposit taker. 17. In particular, Part 5D of the Act allows for the development of a prudential regulatory regime for NBDTs including: • capital adequacy requirements; • liquidity requirements; • related party lending restrictions; • governance standards; • risk management requirements; and, • credit rating requirements. 18. This regulatory impact statement supports the introduction of new prudential requirements for NBDTs, focusing specifically on the proposed capital adequacy requirements. The Reserve Bank is also introducing new credit rating requirements and

2 The transmission of a financial shock in one entity to other independent entities

IN CONFIDENCE 6 related party lending restrictions. The impact of these requirements is assessed separately. Objectives and Assessment Criteria 19. The Bank is required, under Part 5D of the Act, to put in place a capital measurement framework, and may also require a minimum capital ratio. The regulations for capital requirements seek to do this in a manner that is consistent with the broader objectives in the Act (as outlined in paragraph 9 above) and the principles the Reserve Bank is required to adhere to (in section 157F). 20. Specifically, Section 157F of the Act requires the Reserve Bank to take account of the following principles when carrying out its functions under Part 5D: 39. the desirability of consistency in the treatment of similar institutions, regardless of matters such as their corporate form; 40. the importance of recognising: i. that it is not the purpose of this part to eliminate all risk in relation to the performance of deposit takers or to limit diversity among deposit takers; and ii. that depositors are responsible for assessing risk in relation to potential investments and for their own investment choices; 41. the desirability of providing to depositors adequate information to enable them to assess risk in relation to potential investments and to distinguish between high-risk and low-risk deposit takers; 42. the desirability of sound governance of deposit takers; 43. the need to avoid unnecessary compliance costs; and, 44. the need to maintain competition within the deposit taking sector. 21. These principles sometimes push in different directions and the design of the framework has been a matter of balancing these principles. In practice, the Reserve Bank has had regard to the statutory principles, and sought to design a framework that is consistent with the following high-level considerations: • the need for the framework to apply to a diverse set of institutions ranging from finance companies with balance sheets of around $2 billion to credit unions with assets of less than $1 million; • the desirability of leveraging off the existing bank standardised regime where appropriate. This is based on well tested capital adequacy design principles and would allow risk to be compared across the financial sector and help maintain competitive neutrality across the banking and NBDT sectors; • the need for simplicity. The regime must be able to be understood and implemented at low cost by the smallest NBDT; • the need to have regard to lessons from the New Zealand NBDT sector experience of the last two years and relevant lessons from the international financial crisis;

IN CONFIDENCE 7 • the desirability of the structure and calibration of the framework reflecting sound risk measurement analytical principles; and, • the reality that the government deposit guarantee scheme will impact on the efficiency of the NBDT sector. 22. The Reserve Bank has assessed each of the options identified against these objectives below. It has also assessed its preferred option directly against the statutory principles set out in the Act (see paragraph 85). Alternative options and preferred option 23. There are two issues to consider in the structure of the capital adequacy requirements. The first relates to the choice of the framework for measuring the capital adequacy ratio (as a framework is required under the legislation), and the second is whether to have a Reserve Bank imposed minimum ratio and, if so, what that ratio should be. The choices as to the measurement framework and the minimum capital ratio can be made independently. Section 1: The measurement framework 24. A sector-wide capital adequacy measurement framework will provide a consistent basis upon which to compare the risk of NBDTs. A measurement framework can comprise a number of elements, including determinations on what is and is not eligible as capital, judgements on the appropriate treatments for different types of capital and different types of risk, and determination of different risk weightings for different asset classes. There are therefore a number of parts which can be independently calibrated, which results in a very wide range of options available to the Reserve Bank in designing the appropriate framework. To illustrate the main choices that were made and the logic behind them this section discusses four options: • Option 1: Simple leverage ratios; • Option 2: The capital adequacy regime as applied to New Zealand banks; • Option 3: Simplified version of the bank regime; and, • Option 4: Amended version of the bank regime. Option 1: Simple leverage ratios 25. A leverage ratio is simply the ratio of total assets of a financial institution to its capital. It makes no allowance for the different risk of different exposures and market and operational risks. 26. While this option has the advantage of simplicity it was not developed primarily because it does not capture differences in relative risk. For example, under simple leverage ratio government bonds would have the same treatment as a second mortgage on a property development, despite the fundamentally different risk profile of the respective exposures.

IN CONFIDENCE 8 27. In addition, the Reserve Bank notes that many NBDTs argued against such an approach during public consultation. Whilst many NBDT trust deeds contain simple leverage based measures for capital, there are also many that operate with significantly more advanced internal systems. As a result, a number of stakeholders considered that a regulatory requirement based on such a simple framework would represent a negative development. Option 2: The banking regime 28. The Reserve Bank’s capital adequacy framework that is applied to banks is based on the Basel II capital adequacy framework. This provides a set of structures and concepts broadly accepted in the international regulatory community. In particular, it requires banks to maintain tier 1 capital at a minimum of 4 per cent of risk weighted assets, and total tier 1 and tier 2 capital at a minimum of 8 per cent. 29. The Basel II framework is a complicated structure, which has two main variants. They are the ‘advanced’ framework, which applies to the four systemically important banks (which constitute 85 percent of the banking system). A ‘standardised’ model applies to four relatively small locally incorporated banks which constitute 5 per percent of the system. 3 30. Both frameworks have the following structure: • A set of rules that define what instruments count as capital. As noted above, the rules recognise two broad sets of capital: o Tier one instruments, which provide the best protection against default; and, o Tier two instruments (upper and lower), which provide much weaker default protection but some buffer against losses once a bank is in default. • A set of rules that define what items should be deducted from capital. • A set of rules and/or procedures for measuring a bank’s risk exposures. 31. The risks captured by the exposure measurement framework are off- and on-balance sheet credit risk, operational risk, and market risk. In the banking sector, credit risk typically accounts for more than 80 percent of total risk. 32. The ‘standardised’ model provides some recognition of the different risk of different credit exposures by assigning risk weights that roughly reflect relative risks of broad categories of loans. For example, a well secured residential mortgage loan will have a risk weight of 35 percent compared to a risk weight of 100 percent for other lending. This means that a bank has to carry a little more than a third of the capital of a standard loan to provide the same buffer against unexpected losses. 33. With the advanced model, banks measure risks much more finely using models laid down in the capital framework and inputs that banks have calculated themselves or are

3 The remaining 10 per cent of the banking system consists of branches of international banks that operate in accordance with requirements set by overseas jurisdictions.

IN CONFIDENCE 9 required to use by the Reserve Bank. The accredited banks have invested many millions of dollars in developing the models necessary to underpin their ‘advanced’ status. 34. The Reserve Bank considered that the advanced model would be too complex and too costly to adopt for NBDTs. Whilst the standardised model is significantly simpler than the advanced model, the framework still runs to some 70 pages and covers many areas that are not relevant to most NBDTs. Option 3: Simplified bank regime 35. The third option (put forward by many in industry during consultation) was to follow the standardised bank regime, with both tier one and tier two capital recognised, but with a single 100 percent risk weight applied to all exposures except housing, highly-rated corporates, and the government. 36. In the Bank’s view this model would be a backward step. With respect to capital it would mean that the minimum ratio for quality capital would be only 4 percent. The arguments for a higher level are set out below. It would also mean that tier two capital would be represented as having significant value when the Reserve Bank does not think it does, and when clear experience in the finance company sector is that low grade capital has done little to prevent failures and little to limit losses (although the argument applies more broadly, and is not just a reaction to recent events in this sector). 37. Moreover, with respect to risk differentiation using the risk weights in the standardised framework it would mean that: • Important differences in risk would be missed and it would be harder to identify the riskiest NBDTs. For example, with commercial exposures the same 100 per cent risk weight would be applied to all loans; • The framework would be in some respects less sophisticated than that already employed in some NBDT trust deeds; and, • It would be more difficult to make capital adequacy comparisons with the banks. Option 4: Amended bank regime 38. The final, and preferred, option is similar to option 3 above in that it builds on the existing bank regime, but with more targeted amendments designed to reflect the specific circumstances of the NBDT sector. A number of additional credit risk categories, which are distinguished by sector and/or by security cover, have been created in response to points raised in consultation. 39. In developing the NBDT framework the following key changes were made to the standardised bank regime: • the rules were substantially simplified with the regulations covering fewer than 20 pages compared to 70 for the standardised bank model and 130 for the advanced model;

IN CONFIDENCE 10 • only one level of capital, all of which counts as tier one in the bank regime, is recognised; and, • capital calibrations were updated, based on elements of the advanced model requirements where appropriate. 40. To calibrate the regime, the Reserve Bank drew on: its housing risk model; the models and risk parameters in the Basel II advanced modelling framework; its experience in accrediting New Zealand banks for advanced modelling status; and its experience in assessing risk in the NBDT sector. 41. The most significant difference between the bank regime and the proposed NBDT regime is the decision to recognise only one tier of capital. The Reserve Bank considers this to be appropriate for the following reason: • tier one capital is the only form of capital which is permanently and freely available to absorb unanticipated losses without the financial institution being obliged to cease trading. A higher level of tier one capital reduces the probability of an NBDT failing whereas tier two instruments primarily have value in reducing the extent of losses in the event of failure. A tier one capital requirements is therefore more directly targeted to part (b) of the purposes of Part 5D of the Act; • the international regulatory community is moving towards a greater emphasis on tier one capital within their banking requirements; • a single tier regime is much simpler than the current multi-tier bank regime which is more difficult to understand and which is more open to abuse and to disputes about interpretation; • tier two instruments and other subordinated debt have, in recent years, often been marketed to relatively unsophisticated investors who have not understood the nature of the risks of these instruments; and, • very few NBDTs that are currently operating normally have tier two capital. 42. The amended capital calibrations were subject to consultation with stakeholders. As a result of that consultation process, the Reserve Bank made a number of amendments to its initial proposals. A full description of those amendments is set out towards the end of this statement in the section on consultation. 43. For these reasons, the Reserve Bank considers that the amended version of the banking regime, featuring only one tier of capital, with a differentiated set of risk weights, represents the option best placed to meet the aims outlined above. 44. Annex 1 contains the risk weights for on balance sheet assets, as contained in the Reserve Banks proposed capital adequacy policy for NBDTs. Any credit equivalent, off￾balance sheet assets are included within the regime with a risk weighting of 100 per cent.

IN CONFIDENCE 11 Section 2: Minimum level of capital 45. As noted above, NBDTs are currently required to comply with a minimum capital requirement as set out by their trustee. Going forward, the Reserve Bank must consider whether it is sufficient for trustees to continue to be responsible for setting a minimum requirement that is consistent with the new measurement framework, or whether it is necessary for the Reserve Bank to determine a requirement through regulation. In the event that a regulatory minimum is considered necessary, there is then a second question surrounding the appropriate level. 46. In considering the appropriate approach, the Reserve Bank has analysed the following options: • Option 1: No minimum requirement set by regulation, but with trustee based requirements for each entity based on the Reserve Bank’s measurement framework; and • Option 2: Minimum requirement set by regulation: o Alternative 1: A 4 per cent minimum consistent with the tier one capital ratio under the bank capital adequacy framework; o Alternative 2: An 8 per cent minimum consistent with the total capital ratio under the bank capital adequacy framework; and o Alternative 3: A minimum ratio of above 8 per cent, for example, 12 per cent. Option 1: No minimum requirements 47. The Reserve Bank considers that there are a number of reasons why a robust minimum requirement is necessary. First, individual NBDTs do not have the incentives to take into account the risk that their increased probability of failure will have an impact on the rest of the sector. Recent events in the sector have heightened concerns about the risks of contagion. 48. Second, some NBDTs may still have the incentive, capacity, and inclination to represent themselves to depositors as safer than they really are. While the lessons from recent finance company failures will make depositors more sensitive to risk there will be a proportion of unsophisticated investors who do react to these signals and can be led into propositions that are riskier than they think. A robust minimum requirement reduces the scope for this kind of gaming. 49. Third, a robust minimum is required to offset the incentive effects of the deposit guarantee scheme and the sustained effect it will have on the likelihood of a government bailout even it is entirely withdrawn. 50. The main argument in favour of a trustee-based model was that the trustee would be in a better position to understand the risk profile of their clients and to require a minimum ratio that was consistent with that profile. 51. The Reserve Bank does not consider that such an approach is appropriate for the following reasons:

IN CONFIDENCE 12 • Trustee-imposed or agreed capital ratios have often not been effective in requiring finance companies to hold capital buffers commensurate with their risks and this has contributed to the high finance company failure rate over the last two years; • a Reserve Bank imposed minimum will have more credibility and will help restore confidence in the sector; • not all trustees may have the skills to translate their knowledge of individual NBDT businesses into a coherent and consistent framework for setting capital ratios; • individual trustee decisions may lead to inconsistencies in the treatment of NBDTs with different trustees; and, • a Reserve Bank minimum ratio will give NBDTs more certainty at an early point. 52. It should be noted that trustees would still be able to require a higher minimum if they thought that the risk profile of a particular NBDT justified it, regardless of the regulatory minimum having been imposed by the Reserve Bank. Option 2: Minimum requirement set by regulation 53. The preferred option is for the Reserve Bank to set a minimum required capital adequacy ratio. In considering the appropriate minimum there are, in effect, an infinite number of options along a continuous spectrum. However, for practical reasons, the Reserve Bank narrowed its analysis to an assessment of the following three alternatives: • Alternative 1: A 4 per cent minimum consistent with the tier one capital ratio under the bank capital adequacy framework; • Alternative 2: An 8 per cent minimum consistent with the total capital ratio under the bank capital adequacy framework; and, • Alternative 3: A minimum ratio of above 8 per cent, for example, 12 per cent. 54. In setting the minimum requirement, the Reserve Bank must reflect on a tension that is generated by two competing concerns. If the minimum is set too low then there is a risk that this will be interpreted as the supervisor’s view of an adequate capital ratio and encourage NBDTs with a higher ratio to reduce the amount of capital they hold. NBDTs may be regarded as ‘well capitalised’ when by any reasonable measure of investors’ expectations they are not. The lower the capital ratio the higher the probability of default. If, on the other hand, the ratio is set too high it may unreasonably reduce the range of business models that are permitted. While default rates will be further reduced, the benefits might be outweighed by a reduction in institutional diversity.

IN CONFIDENCE 13 Alternative 1: Four per cent minimum 55. The first alternative considered by the Reserve Bank is a minimum of 4 per cent which is the formal minimum for tier one capital under the current adequacy framework for banks. The evidence above suggests that the vast majority of NBDTs currently successfully operating would already be compliant with such a requirement. As such, the cost of compliance can be expected to be minimal. However, the Reserve Bank does not believe that 4 per cent is an adequate level of capital for banks or for NBDTs. In particular, it notes that: • a minimum of 4 per cent for NBDTs would create a strong risk that such a level would be perceived and represented as adequate and this would mislead depositors; • a 4 per cent requirement would be grossly inadequate compared to international standards4 • expected default rates with a 4 per cent ratio are many times higher than with an 8 per cent ratio; , which are at least 8 percent and, in some cases, heading towards 10 per cent; • the average un-weighted tier one ratio of New Zealand banks is currently about 9 per cent. A 4 per cent ratio would look very inadequate in comparison; and, • a minimum as low as 4 per cent would risk undermining existing NBDT capital ratios, which are in most cases significantly in excess of this level. 56. As a result, the Reserve Bank does not consider that a minimum ratio of 4 per cent would be likely to assist in restoring confidence in the sector. Alternative 2: Eight per cent minimum 57. The second alternative is to set a minimum ratio of 8 per cent in line with the total capital ratio under the bank capital adequacy framework and is consistent with banks actual tier one capital holdings, as outlined in paragraph 51. Given this market benchmark, this is the most natural minimum. Alternative 3: Twelve per cent minimum 58. Whilst the Reserve Bank was satisfied that a minimum of 8 per cent would be an appropriate solution, it also considered whether it would be preferable to adopt a higher minimum, for example 12 per cent. In considering such an approach, the Reserve Bank noted the following points: • the proposed framework already embeds several calibrations that take account of the systematically higher levels of risks associated with NBDTs and to increase the minimum ratio above that in the banking framework would be to risk an element of double counting;

4 See www.bis.org/publ/bcbsca.htm

IN CONFIDENCE 14 • while there is a policy intention to increase minimum standards that NBDTs must with respect to capital adequacy, it is not the intent to raise the bar so high that all NBDTs are pushed to a narrow range of outcomes, generating bank or near bank levels of creditworthiness. The Reserve Bank notes that many NBDTs (particularly in the finance company sector) will be challenged to meet the 8 per cent minimum and a material increase in this minimum might put the task beyond some of them; and, • while a higher minimum would further reduce the probability of default, this reduction would not be large. Most of the benefits from higher capital are captured by the move from a four to an eight per cent ratio.5 59. In light of these points, the Reserve Bank rejected the option of setting a higher minimum. As a result, its preferred option is to set a minimum requirement of 8 per cent, in line with the total capital ratio under the bank capital adequacy framework. Interaction with other regulations 60. The one exception to the industry wide capital requirement relates to those credit unions that are exempt from the requirement to hold a credit rating from an accepted credit rating agency. The initial cost of acquiring a credit rating will vary, but is expected to be at least $40-60k. There is then an on-going annual cost of around $25k to hold a credit rating. The Reserve Bank considers that it would be inappropriate to enforce such a cost on some of the smallest credit unions (who tend to be the smaller NBDTs). 61. As an alternative, the Reserve Bank has proposed that those entities be subject to a slightly higher minimum capital requirement to compensate for the lesser degree of outside scrutiny applied to unrated NBDTs. The appropriate level is a matter of judgement. In light of the diminishing benefits associated with each subsequent increase in the minimum level, the Reserve Bank considers that ten per cent would be an appropriate requirement. In practice, most unrated credit unions hold capital in excess of this level, so there should be minimal impact in the short-term. Impact of the preferred option 62. In considering the appropriate minimum, the Reserve Bank analysed, at an early stage of the policy development process, the impact of a preliminary version of the framework on capital adequacy against a wide sample (approximately 70) of NBDT balance sheets including all of the large ones. While not all of the information needed to precisely calculate each NBDT’s capital ratio was available (this is, of itself, instructive) the exercise provided a reasonable indication of the likely impact if the regime were applied

5 Using the Basel II corporate capital framework model the Reserve Bank calculates that moving from a 4 to an 8 per cent minimum requirement reduces the risk of failure about 10 times more than a movement from 8 to 12 per cent. Note that in practice these calculations are likely to overstate the disparity in benefits arising from movements from 4 to 8, and from 8 to 12 per cent because the Basel model assumes that underlying asset values are ‘normally’ distributed. In the real world extreme movements in asset values are more prevalent than this theoretical distribution assumes, and this increases the value of capital of these events on a financial institution’s solvency.

IN CONFIDENCE 15 around the middle of 2008. Since the survey was completed several calibrations have been eased which will make it easier for NBDTs to comply. Updating the analysis to June 2009 data suggested that: • Credit unions are largely comfortable with both capital and related party regulations. The bigger concern for this sector had been the credit rating requirement, although this issue is very likely to disappear for all but one or two credit unions once the NZACU puts in place a cross guarantee. This arrangement will enable participating members to be rated as a part of the group, thereby spreading costs. • The building societies are generally supportive of the regime and the capital and related party regulations. Most of the building societies remain conservative in their business practices, are mostly rated already, and are supportive of a capital regime that distinguishes clearly the different risk profiles in the NBDT sector. • While the majority of finance companies supported more regulation in the sector, a small number face challenges meeting the proposals (and would indeed face challenges meeting far less conservative requirements). Most of the concerns raised with the calibration of the capital and related party policies come from a few companies that are currently quite a distance from compliance. • Credit unions: Compliance with the proposed capital requirements is not an issue for credit unions, with the industry average tier one capital ratio sitting at 14.2 percent as

IN CONFIDENCE 16 at May 2009, with all credit unions above the 8 percent recommended minimum capital ratio. Figure 1: Tier one capital ratios6 • Building societies: Eight out of the nine building societies are already above the minimum required, and the industry average was 10.9 percent as at May 2009. The one building society that is below the proposed 8 percent, is close to 8 percent, and would not be expected to have any particular difficulties in meeting the requirement. • Finance companies: By number most finance companies are already above the 8 percent, but the industry average is skewed by some of the bigger players. Of the larger finance companies – that is, over 100 million dollars of assets – three companies would already be compliant or are unlikely to have issues in becoming compliant. On the other hand, three companies are likely to require substantial capital injections and / or restructuring, or have a longer path to compliance by way of exemption. One other finance company may also require a small capital injection to become compliant. Of the smaller finance companies, 10 of 13 are over 8 percent and therefore would be expected to be compliant. The remaining three non-compliant companies face significant challenges, but this is related to the poor quality of their books, and the consequent challenges they are likely to face in raising capital (rather than the capital regime per se). 6 Tier one capital instruments are primarily ordinary shares and retained earnings. Non-cumulative preference shares, which meet certain requirements that give them strong equity like characteristics, qualify as tier one capital but are restricted to a limit of 25 percent of that capital.

IN CONFIDENCE 17 63. The impact on NBDTs’ positions when the minimum capital requirement comes into effect in late (September) 2010 is uncertain, and will depend to a degree on how individuals NBDTs fare through the recession. Some will have the opportunity to build on their capital position by retaining earnings, but others will see their positions eroded through credit losses. Some are likely to fail, and some will have difficulty in complying with existing trust deeds and greater difficulty in complying with the new regime at least with their current balance sheet size. Compounding the risk to compliance are uncertainties about the availability of external equity funding to rebuild and strengthen balance sheets. 64. Whilst a handful of finance companies will need to raise significant additional capital to comply with the requirement, any costs associated with compliance will be reduced by the proposed transitional arrangements which include a one-year period for companies to achieve compliance and the potential for the Reserve Bank to issue exemptions in certain circumstances (as discussed below). Benefits from higher capital 65. It is difficult to quantify the benefits that might arise from the introduction of minimum capital requirements. As noted above many NBDTs will be compliant and the regime will not have a material impact (positive or negative) on them. 66. A small number of significant finance companies are likely to have to improve their capital positions. This will reduce the risk of individual failure and of a significant part of the specialist financing sector of the financial system failing at one time. 67. A further benefit is that higher capital ratios and more robust risk measurement will prevent a re-run of the ‘boom-bust’ cycle that took place in the sector in recent years. Table 2 outlines the capital ratios and losses associated with four recent high profile failures. Measured under the new framework these NBDTs were very weakly capitalised (despite their accounts tending to suggest that they were in a much stronger position). Table 2 Comparison of capital ratios for failed NBDTs NBDT Capital ratio NBDT framework % (estimate) Possible economic losses % Equity/assets last accounts % Bridgecorp 2-4 90 8 Dorchester 2 ½ - 4 Insufficient information 11 Hanover 1-3 30-50 11 St. Laurence Negative capital 40-60 19

IN CONFIDENCE 18 Potential costs associated with the preferred option 68. For the reasons outlined above, it is difficult to estimate the precise level of additional capital that NBDTs may seek to hold as a result of the introduction of the new capital adequacy requirements. Instead, the Reserve Bank has approached its analysis of the impact of the proposed approach by: • identifying a theoretical framework within which to analyse the issue; • considering real world departures from the that framework; and, • assessing specific arguments against higher capital. Theoretical framework 69. The Modigliani-Miller theory tells us that the overall cost of capital is constant regardless of the level of leverage (within reasonable bounds). To the extent that the new minimum capital requirements force NBDTs to hold ‘more’ capital (i.e. decrease their leverage), both the equity investment in financial institutions and deposits with those institutions will become safer and required rates of return will fall. 70. Furthermore, if it is assumed that NBDTs are faced with perfectly functioning capital markets, all firms operating in the sector will be able to access the required level of equity at the new appropriate market rate (i.e. the rate that provides investors with a fair return for the level of risk associated with the investment). To the extent that there is no impact on the overall cost of capital, this deleveraging effect should not therefore have any impact on the asset side of the balance sheet, and, as such, there will be no material change in the amount and price of loans, and hence economic activity. 71. However, this theoretical result rests on a number of assumptions that do not necessarily hold in practice. The Reserve Bank has therefore considered potential impacts that might arise through the application of the policy to the existing NBDT sector in New Zealand operating under current market conditions, including some arguments put forward by industry participants. Practical implications 72. The theoretical framework suggests that minimum capital requirements are likely to impact upon lending margins and rates of return, but not on the availability of loans in the NBDT sector. However, this outcome can only be expected if an assumption of efficient market conditions holds. 73. In practice, many NBDTs are not well connected with equity markets, while often managers, owners, depositors, and trustees have a poor understanding of underlying risk principles, relying instead on rules of thumb that may not have been right to begin with, but become especially misleading when circumstances change. It is possible that some NBDTs would not increase their capital but would react to an increase in regulatory capital by scaling back their lending – a viable and acceptable response. 74. Such an outcome could potentially lead to a reduction in lending in certain markets to a level below what might be efficient for the wider economy. To the extent that the markets affected can be classified as ‘commodity lending’, like residential property

IN CONFIDENCE 19 mortgages, then this would not matter because the loans would readily be replaced by other lenders. By contrast, if the NBDT were a very specialised lender, then the lending might not be as readily replaced. However, this risk needs to be assessed against the likelihood that the NBDT would abandon a profitable relationship where it has a strong franchise, and is likely to be low. Overall, in normal market conditions, there should be little impact on the supply of loans to the economy. 75. In current circumstances though, the risk that NBDTs’ capacity to lend is constrained is greater. Markets for risk in the financial sector have been dysfunctional internationally and there is less likelihood that capital will flow to NBDTs that need more, or that other financial institutions will step in to make up any shortfall. To some extent a reduction in lending capacity is a desirable part of the adjustment process to a less leveraged economy that is occurring in New Zealand and the rest of the world. However, to the extent that the introduction of the regime results in a real cost to the economy by placing excessive pressure on desirable lending relationships, then this is best handled by the exemption provisions in the regime. 76. As noted above, the Reserve Bank has the capacity to provide exemptions where certain conditions are met (including where compliance with regulatory requirements would be unduly onerous). This may include a longer adjustment path for NBDTs that need the capacity to appropriately service their clients funding needs that cannot be met elsewhere. As a result, the Reserve Bank considers that any costs arising in this manner can be effectively mitigated by retaining the flexibility to target the regime in where appropriate. Arguments against higher capital Sustainability of existing business models 77. According to the theoretical framework, requiring more equity will lead to a lower return on equity, leading some to argue that it will be impossible for firms to meet their existing required rates of return on capital, with the implication that otherwise sound entities may exit the industry. For example, some in industry have argued that with current capital levels they can meet their (generally high) target rate of return of, say, 30 per cent, but if they were required to increase capital by 25 per cent then this return would fall to 24 per cent. 78. There are several problems with this kind of argument. First it assumes that lending margins will stay as they are. However, if they are too low and do not reflect underlying risks then it is appropriate that they rise and for the most part it is clear that they have already done so. If the capital regime prompts some increases to reflect risks that were previously understated then that is an appropriate outcome. 79. More importantly it assumes that business models based around excessive expected rates of return are still appropriate or achievable. In general, high rates of return can only be achieved over time (in the face of growing competition) if the lender adopted a risky model and depositors were not fully aware of that risk and hence did not charge appropriately for it. This, broadly speaking, was the situation in many of the failures. 80. Looking forward, depositors are much less likely to invest in a highly leveraged NBDT, and certainly not at the low margins of the past. In the longer term, NBDTs with high profit expectations will have to change those expectations to rates of return which are

IN CONFIDENCE 20 more consistent with safer institutions. This will be part of the deleveraging process that is taking place around the world. 81. In the short run it is unlikely that NBDTs with credible lending opportunities and the support of their depositors will withdraw from the field. The failure of many competitors, the more circumspect approach taken by banks to lending with the consequent expansion is risk margins, possibly beyond levels that will be justified in the longer term, together with the value of maintaining a lending franchise should provide adequate incentive to stay in the market in most cases. Restriction of management discretion 82. The implementation of new regulations can distort decision making by market participants. One of the implications of the assumption that the cost of capital remains constant at all levels of leverage is that there is no generally accepted theory to dictate optimal leverage. In practice, many factors influence decisions over financial structure including, amongst other things, taxes, bankruptcy costs and agency costs. 83. It might therefore be credible to argue that the introduction of a minimum capital ratio may force some NBDTs to have a higher level of capital than they would otherwise choose to have. To the extent that such decisions were based on legitimate business factors, such an outcome could be considered socially sub-optimal. 84. However, an NBDT may want a low level of capital because this is a way of extracting value from implicit or explicit government guarantees. Having low capital is a risky strategy that pays off to the NBDTs’ owners if it is successful, but imposes some of the costs on the government when it is not. For similar reasons it is a way of extracting value from depositors who do not understand the risks NBDTs are running. 85. Further, individual NBDT decisions will take no account of the impact of the likelihood of failure on the wider economy. If it were a matter of the failure of one or two NBDTs then a failure would be of little wider consequence. If, however, conventions re-emerge that thinly capitalised business models are acceptable then there is a heightened risk that a shock will lead to widespread failures in the sector which could have macro-economic consequences. 86. A deleveraging of the NBDT sector should therefore lead to an increase in welfare because the minimum requirements off-set some of the existing distortions that generate excessive leverage in the economy. Reduction in diversity within the NBDT sector 87. A further concern surrounds a risk that the use of the bank capital framework as the basis for the NBDT regime might force all NBDTs to become ‘mini-banks’ with the same risk standard as a large bank (AA or AA- on the Standard and Poor’s rating scale). 88. In practice, this risk does not arise. Whilst the Basel advanced framework has been used to inform the proposed policy, the framework itself has been calibrated to a risk standard that is lower than that which applies to banks (BBB compared to AA). Most of the larger finance companies aspire to credit ratings which are not far from this standard. Furthermore, most smaller institutions that have, or are likely to have, lower credit

IN CONFIDENCE 21 ratings are already compliant with the proposed standards so the regime will not affect their balance sheet or risk rating. 89. More importantly, capital adequacy is not the only determinant of a financial institution’s risk. To varying degrees financial institutions have resources that reduce their overall risk and other risks that increase it, which are not captured by a capital ratio. The situation is captured in figure 3, which compares the position of a bank with an AA rating with a small NBDT with a BB rating. Overall risk is represented by the relative size of the bank or NBDTs risks (on the right hand side) compared to the resources available to absorb those risks (on the left hand side). The capital ratio is generated by on-balance capital versus the credit risks and market risks the firm faces. In the diagram these are about the same. The bank and the NBDT have the same capital ratio. However, the bank has more additional resources than the NBDT. It has a strong franchise that generates earnings that can absorb losses and a greater likelihood of support from external sources. 90. Markets are more likely to provide additional capital if needed because of the value of the franchise and governments are more likely to support a systemically important bank than a small NBDT. On the other hand the NBDT is likely to have greater risks that are not captured by the risk ratio framework. Some of this relates to additional operational risk that is not captured by the NBDT capital adequacy measurement framework. Generally operational risk, in relative terms, increases as the firm gets smaller because small firms cannot afford complex control systems and are less able to diversify their risks. The NBDT requirement for operational risk was calibrated to an average sized NBDT so it tends to understate the risk. Another important risk that is not captured by the capital ratio is the risk that market or competitive conditions will change. The banks strong franchise provides substantial protection that is not as available to the NBDT. Liquidity risk is another example of where the NBDT’s risk will be higher than the banks. Figure 3 Measuring Capital Adequacy and Overall Risk Small NBDT

Capital Ratio A/ (D+E) = 8% Overall risk (A+B+C)/ (D+E+F+G) = BB Credit rating

IN CONFIDENCE 22 Large Bank Capital ratio A/ (D+E) = 8% Overall risk (A+B+C)/ (D+E+G) =AA credit rating Assessment of preferred options against objectives 91. The Reserve Bank considers that the preferred options identified above are consistent with the principles set out in Section 157F of the Act, as noted in paragraph 18 above. In particular, the Bank notes that: • with respect to principle (a), a single regime has been applied to finance companies, credit unions and building societies;7 • with respect to principle (b), the important consideration was the calibration of the minimum capital ratio. The Reserve Bank believes that it has been set at a level which is credible but which is not so high as to unreasonably reduce the array of risk models that might be presented to investors. The Reserve Bank believes that the market will demand better capitalised NBDT investments and that the effect of the minimum ratio will be to help crystallise preferences around a new benchmark rather than constrain them. Another point to note is that a capital ratio is only one factor (others include the strength of the franchise; risk management systems and control) that impact on a financial institution’s risk, and in particular, how this risk is conveyed to investors through a credit rating. NBDTs will still be free to differentiate themselves with respect to these other factors; • with respect to principle (c), the Reserve Bank believes that the framework presents a material improvement from the capital adequacy measures

7 Whilst there is a high degree of diversity in the NBDT sector (as demonstrated by figure 1), a single regime can be applied to all businesses in the sector as differences in risk exposure are picked up in the risk weightings applied to different asset classes. Thus, a more risky business will be required to hold more capital in absolute terms, even when operating to the same minimum requirement, as a less risky business with a similar overall volume of assets.

IN CONFIDENCE 23 (typically simple leverage ratios) that are currently in NBDTs’ trust deeds and is better calibrated to depict NBDT risk than the standardised bank model; • principle (d) is not directly relevant to the design of a capital framework; • with respect to principle (e) the Reserve Bank believes that the framework will improve risk management because it will help direct management and depositor attention to the capital consequences of riskier business activities; • with respect to principle (f) the simplified framework will have a significantly lower compliance cost than the bank regime, with the framework document covering just 20 pages compared to up to 130 pages for the advance bank regime; and, • with respect to principle (g) the Reserve Bank believes that a robust capital regime will help to restore confidence in the NBDT sector, allowing a well managed and well capitalised NBDT to compete for funds from depositors who are comfortable with the risk profile of the NBDT. Implementation and review 92. It would be desirable to introduce the capital regulations in the third quarter of 2009 or very early in the fourth quarter. As noted in paragraph 59, the requirement to comply with the minimum capital ratio should be for the following year (ie September 2010), in order to give NBDTs time to comply – either by raising capital or restructuring their book. 93. All NBDTs will be required to be compliant with the minimum capital ratio at all times not just at the annual balance date. In practice, this will require NBDTs to be compliant with the requirements at each balance date (likely to be quarterly). Where an NBDT fails to comply with the requirements, it may, in the future, become subject to enforcement action. 94. As noted above, there is scope for individual NBDTs to apply for an exemption from this regulation. The Act allows the Reserve Bank to exempt any deposit taker, class of deposit takers, or trustee from compliance with all, or any specific, regulation(s) made under Part 5D. Under the legislation, to grant an exemption, the Reserve Bank must be satisfied that: • the exemption will be consistent with the maintenance of a sound and efficient financial system; and, • compliance with the relevant provision or provisions would, in the circumstances, require the deposit taker, class of deposit takers, or trustee to comply with requirements that are unduly onerous or burdensome; and, • the extent of the exemption is not broader than what is reasonably necessary to address the matters that gave rise to the exemption. 95. The Reserve Bank will unavoidably have to exercise a degree of judgement in assessing any applications for an exemption, which will need to be considered on a case-by-case basis. However, the potential to grant exemptions, where appropriate, can be expected to mitigate potential costs that may be incurred by some businesses, particularly where the

IN CONFIDENCE 24 transitional phase might not be sufficient to allow a business to enhance its capital or restructure its business in time to comply with the requirements. 96. As noted above, regulation of NBDTs is found in Part 5D of the Act dealing with requirements covering risk management, corporate governance, credit rating, minimum capital, related party lending and liquidity (being introduced variously in the course of 2009 and 2010). Additional legislation is also proposed to complete the NBDT regulatory regime, covering licensing, fitness and properness, ownership, and crisis management, and which will include enforcement powers such as directions and statutory management. 97. This additional legislation is currently in the course of policy development. It is hoped that it will be enacted towards the end of 2010. The two-step approach takes into account the effect of the recent enactment of the Financial Service Providers (Registration and Dispute Resolution) Act 2008. 98. As international standards develop over the next couple of years, the Reserve Bank will monitor the relevance of the regulations, as well as the experience of industry, including against the backdrop of the domestic and external environment. 99. The direct costs to the Reserve Bank arising from the operation of the new capital adequacy requirements (e.g. through on-going monitoring of compliance or enforcement action) are difficult to estimate precisely and are likely to fluctuate from year to year. The additional resources required to administer the NBDT regime in its entirety is expected to be around 4.5 FTEs. This, together with non-salary related costs and overheads is expected to result in an on-going cost of around $750k per annum. However, this cost will be covered by the Reserve Bank’s funding agreement with the New Zealand government, and as such, does not fall on the businesses captured by the regime. 100.The potential direct cost to businesses of complying with the capital adequacy requirements are difficult to forecast precisely. There will be an initial set-up cost incurred as companies put in place the systems to monitor the regime going forward, however, the relatively simple nature of the regime should restrict the scale of these costs. In addition, some of the larger firms are likely already to have in place similar systems, while we understand some sectors (e.g. credit unions) may be seeking to put in place a collective solution. Once in place, the on-going direct costs of monitoring compliance are expected to be marginal. 101.It is possible that the introduction of a strict minimum requirement could introduce a further potential cost if businesses feel compelled to adopt a significant buffer above the minimum level in order to reduce the risk of breaching the requirements and becoming subject to enforcement action. However, as the majority of NBDTs currently choose to hold capital in excess of the proposed minimum, the Reserve Bank considers it unlikely that many firms will feel it necessary to increase capital for this reason. Indeed, it may be the case that some very well capitalised firms will choose to adopt a more efficient financial structure following the introduction of the requirements.

IN CONFIDENCE 25 Consultation 102.A consultation paper was released on 19 December 2008. The Reserve Bank held three public meetings in Auckland, Wellington and Christchurch, as well as numerous private meetings with individual NBDTs (and prospective NBDTs) around the country. 41 written submissions were received. 103.There was broad support for the new framework, but many of the submissions raised concerns similar to the risks discussed above. 104.However, it should be noted that a number of submissions focused on the aspects of the regime that required them to hold more capital compared to banks while ignoring those elements that that gave them an advantage. 105.The Reserve Bank notes that its proposed approach requires NBDTs to hold twice as much tier one capital as currently required of banks under existing arrangements and the Basel framework. However, as noted in the assessment of options above, there is a shift internationally away from the two tiered Basel approach, with greater emphasis on high quality capital. The Reserve Bank will be working to close the gaps between these approaches in due course. 106.A number of stakeholders have focused on this perceived lack of competitive neutrality between NBDTs and banks. One of the primary facts cited in support of this view was a comparison of the risk weight of 150 per cent applied to the residual category (other lending) with that in the bank standardised model (100 percent). However, some respondents have not understood that the NBDT model has been largely calibrated using the advanced bank model. Whilst adopting the advanced bank regime was ruled out due to the excessive cost and complexity that it would create, the Reserve Bank has drawn on the regime where appropriate to inform the proposed approach. This is the most relevant model because the market for credit in New Zealand is dominated by the large banks that use the advanced model. 107.Broadly, and on average, loan portfolios with similar risks are treated similarly in the bank and NBDT sectors. Banks do get some capital concessions because they have spent considerable resources and time more finely differentiating risk than would be possible or necessary for NBDTs but that is appropriate because this work should mean that banks can better manage their risks. There are some specific cases where the banks have materially lower capital because the advanced bank models are not working properly. The advanced model regime is relatively new and the Reserve Bank has not, as yet, had the opportunity to address all of the teething problems that have come to its attention. By the time the NBDT regime comes into effect, however, these issues should have been addressed and the two regimes will be broadly competitively neutral. 108.Some respondents pointed out that the risk averaging process in the preliminary version of the regime had the effect of unnecessarily discriminating against some business lines and institutions. These arguments have merit and changes have been made in some parts of the framework. In particular, as a direct result of industry consultation, the Reserve Bank created a number of additional lending categories which attracted lower risk weights (100 percent compared to the standard 150 percent risk weight). These categories were:

IN CONFIDENCE 26 • personal loans of less than $40,000 secured with PPSR;8 • asset financing loans secured on self-propelled vehicles (tractors, aircraft etc), with a PPSR and a loan to value ratio of less than 60 per cent; and, • other loans secured by land and property with a loan to valuation ratio of less than 60 per cent. 109.In addition: • operating leases have been identified as a separate asset class with a risk weight of 175 per cent rather than being treated as an other asset with a risk weight of 350 per cent; • the risk weight for other assets has been reduced from 500 per cent to 350 per cent. This reflects the risk weight suitable for property; and, • an additional risk weight category has been introduced for residential housings loans. This reduces the risk weight in the 70-80 per cent LVR range from 75 percent to 50 percent to bring it in line with the bank standardised model number. 110.Finally, some risk weights for riskier loans were increased. These were: • an increase in the risk weight for the highest LVR loans; and, • an increase in the risk weights for other loans with a second ranking security or no security. 111.Once these changes were made the Reserve Bank conducted a more limited consultation with major players and parties with specialised expertise to check that the new structures and calibrations made technical sense. The changes were generally well received and no major flaws were uncovered although some parties still wanted lower risk weights in some of the risk classes. 112.The consultations have shown that many respondents did not have a full understanding of key elements of the technical framework underlying the Reserve Bank’s calibration of the regime. This is understandable because the NBDT sector cannot be expected to have a close acquaintance with the complex models that underpin the advanced bank risk measurement framework. The Reserve Bank has undertaken to conduct a series of seminars around the country to explain in a more accessible manner how the risk weight framework was developed.

8 Personal property securities register.

IN CONFIDENCE 27 Annex 1: Risk weights for on-balance sheet assets On-balance sheet assets Risk weight Cash (notes, coins, and gold bullion held on site) 0% Crown and the Reserve Bank (claims) 0% Public sector entities (claims) 20% Banks (claims) 20% Corporate short-term Corporate (short-term claim with rating grade 1) 20% Corporate (short-term claim with rating grade 2) 50% Corporate (short-term claim with rating grade 3) 100% Corporate (short-term claim with rating grade 4) 150% Corporate (short-term unrated claim) 150% Corporate long-term Corporate (long-term claim with rating grade 1) 20% Corporate (long-term claim with rating grade 2) 50% Corporate (long-term claim with rating grade 3) 100% Corporate (long-term claim with rating grade 4) 100% Corporate (long-term claim with rating grade 5) 150% Corporate (long-term claim with rating grade 6) 150%

IN CONFIDENCE 28 On-balance sheet assets Risk weight Corporate (long-term unrated claim) 150% Residential mortgage loans Residential mortgage loans (first ranking with loan-to-valuation ratio not exceeding 70%) 35% Residential mortgage loans (first ranking with loan-to-valuation ratio exceeding 70%; but not exceeding 80%) 50% Residential mortgage loans (first ranking with loan-to-valuation ratio exceeding 80%; but not exceeding 90%) 100% Residential mortgage loans (first ranking with loan-to-valuation ratio exceeding 90%; but not exceeding 100%) 125% Residential mortgage loans (first ranking with loan-to-valuation ratio exceeding 100%) 150% Residential mortgage loans (second or subsequent ranking) 150% Qualifying insured housing loans 20% Property development loans Property development loans (first ranking security with loan-to￾valuation ratio not exceeding 60%) 150% Property development loans (first ranking security with loan-to￾valuation ratio exceeding 60%; but not exceeding 100%) 200% Property development loans (second or subsequent ranking security, or no other security) 300% Personal loans Personal loans in respect of which a financing statement has been registered and fully secured under the Personal Property Securities Act 1999 100%

IN CONFIDENCE 29 On-balance sheet assets Risk weight Personal loans in respect of which a financing statement has not been registered and fully secured under the Personal Property Securities Act 1999 150% Other loans secured by first mortgage over land and/or buildings First mortgage over land and/or buildings (with loan-to-valuation ratio not exceeding 60%) 100% First mortgage over land and/or buildings (with loan-to-valuation ratio exceeding 60%; but not exceeding 100%) 150% Other loans not secured by a first mortgage over land and/or buildings Other loans not secured by a first mortgage over land and/or buildings in respect of which a financing statement has been registered and fully secured under the Personal Property Securities Act 1999 150% Other loans not secured by a first mortgage over land and/or buildings in respect of which a financing statement has not been registered and fully secured under the Personal Property Securities Act 1999 200% Operating leases 175% Equity holdings (not deducted from capital) 600% Other assets 350%