Norges Bank


EU Commission proposal for a review of regulatory capital requirements

Felles brev fra Kredittilsynet og Norges Bank av 31. mars 2000 til EU-kommisjonen

Introduction and summary

This consultative statement has been jointly prepared by Norges Bank and Kredittilsynet. It includes views submitted by Norwegian trade organisations. A copy of the statements from the Norwegian Financial Services Association (FNH) and the Norwegian Savings Banks Association is enclosed.

EU directives in the capital adequacy sphere are implemented in Norway in relevant legislation as part of the EEA agreement. If the EU opts to amend the capital adequacy directives, Norges Bank and Kredittilsynet assume that the changes will be incorporated in the EEA agreement and thereby made applicable to Norway.

Norges Bank and Kredittilsynet take a very positive view of the fact that the Banking Advisory Committee and the Commission took steps to analyze the capital adequacy regulation concurrently with the Basel Committee's review of the capital adequacy framework. The process of putting a revised body of rules in place is time-consuming, and it was important to get started in view of the schedule planned for the entry into force of the revised Capital Accord. It is also important to bring European views to bear with greater weight vis-à-vis the Basel Committee. In our view a good job has been done by the Technical Sub Group (TSG) established under the Banking Advisory Committee.

The Commission's proposal starts out from the proposal of the Basel Committee, although the Commission's proposal is far more concrete than what has so far been produced by the Basel Committee. Our comments therefore largely correspond with those submitted to the Basel Committee.

Although the existing regulatory regime has in Norges Bank's and Kredittilsynet's view fulfilled its mission, the current rules display clear flaws. The chief objection is that they fail to measure credit risk precisely, i.e. the risk weights fail to reflect the actual likelihood of default on the part of the debtor. This can lead to mispricing and misallocation of credit, and may also impair banks' competitive position vis-à-vis the securities market as a source of finance.

Norges Bank and Kredittilsynet are concerned about the design of future capital adequacy rules in view of the need for financial stability. The stability of the banking system is crucial to the stability of the financial system. The banks are at the heart of the payments system, they are the principal funding source for a large number of enterprises and an important saving vehicle for large sections of the population. The capital adequacy framework affects bank behaviour. It should therefore be designed to provide banks with appropriate incentives and to establish a capital buffer of a size adequate to protect the system. The protection afforded by the capital adequacy rules constitutes the first line of defence against unforeseen future losses that may threaten the stability of the financial system.

Norges Bank and Kredittilsynet agree on the need for changes in the current capital adequacy framework, and support the main elements of the Commission's proposal.

The proposal for a new Accord and revision of directives entail in aggregate a far more complex and far-reaching framework than that existing today.

Where Pillar I is concerned, it is difficult from a Norwegian and European perspective to see that a further developed standard method based on external rating would do much to rectify the deficiencies of risk weighting under the current framework as long as so few borrowers in Europe are rated. Capital adequacy calculations based on banks' internal rating systems would probably ensure better correspondence between capital charges and the individual bank's risk profile, but would make it difficult to achieve a level playing field. With the long-term evolution of capital markets in Europe in mind, we would not oppose the use of external rating in the standard method. Assuming the Basel Committee and the Commission are in a position to arrive at precise and operational criteria that ensure uniform competitive conditions, we will support the use of internal rating for capital adequacy purposes. We believe this method should not be confined to the most sophisticated banks.

The introduction of Pillar II, the supervisory review process, is a logical consequence of the fact that different institutions have different risk profiles and that the authorities wish to ensure that institutions are capitalised according to their risk profile and strategy. The establishment of harmonised rules for authorities' intervention vis-à-vis institutions with deteriorating capital adequacy is a clear-cut improvement. Putting in place harmonised rules that empower supervisory authorities to impose capital charges on an individual institution in excess of the minimum requirement that is established generally in the respective country, is also a step forward. However, Pillar II will make heavier demands of the expertise and resources possessed both by most institutions as well as supervisory authorities. We assume that Pilar II can be applied in a simple manner to institutions of minor significance to the financial system and/or whose capital strength is well in excess of minimum requirements. Moreover, in our view Pillar II raises issues related to the distribution of responsibility between institutions and supervisory authorities. The proposal will assign to supervisory authorities a more active role in assessing and approving banks' risk management systems, overall risk, capital levels and power of intervention. This must not be allowed to undermine that it is the responsibility of bank management and owners to ensure prudent risk management and for taking appropriate action should problems arise. For its part, the banking industry is concerned about the higher costs of supervision entailed by Pillar II.

The advantage of the present regulation is its simplicity and the ease which it allows institutions to be compared within and between countries. A complex framework containing a series of options for implementation, including supervisory criteria for setting individual capital ratios, could easily breach the basic principle of a level playing field. Moreover, a complicated regime could mean that only a few persons employed by institutions and the supervisory authorities are familiar with the rules and with key factors bearing on the individual institution's capital adequacy. There may also be a possible conflict here between Pillar I (minimum capital requirements) and Pillar III (market discipline/transparency).

The proposed revision of the Capital Accord and the capital adequacy directives retains the existing regime in respect of market risk and the composition of own funds. The changes refer primarily to the credit risk area, but also entail a substantial development of guidelines and method for supervising banks' financial soundness. Norges Bank and Kredittilsynet believe that such a comprehensive review of the quantitative aspects of regulation should not be confined to the calculation base, but should be accompanied by a similarly thorough and necessary review of the regulation of capital structure. In our view the own funds directive entails an overly complex capital structure. A hierarchy of different instruments in the own funds could impede speedy solution of an acute solvency problem. Norges Bank and Kredittilsynet agree that there is no reason to amend the CAD at the present time.

We agree with the Basel Committee and the Commission that a necessary premise for revising the framework must be to ensure that it does not result in impairment of overall capital adequacy. We agree with the banking industry that risk management and lending practices have improved appreciably since the Accord was introduced in 1988. However, experience gained both in Norway and several other countries in the intervening period shows that the banking system was insufficiently robust to meet the losses that arose.

1. General observations

A comprehensive review of the capital adequacy provisions should be based on the objectives of the regulation. The main intention of the capital adequacy framework is to cushion institutions against the risk of unforeseen losses on their operations. Anticipated losses will be covered by ordinary loss provisions. In addition to acting as a buffer and as protection for depositors and other creditors, the capital adequacy rules form part of the safety net established to secure financial stability. Since the owners of the capital are the first to take losses, there is a strong incentive for long-term owners to ensure prudent conduct of business. For major, professional investors operating in the money and securities markets, bank's capital is a base for granting credit. In addition, the minimum requirements authorise the authorities to intervene in the operations of a bank before the entire capital is lost. The earlier intervention takes place, the greater the chance of finding a solution other than liquidation. In this way spread of contagion effects to other institutions and an economic setback can be avoided or contained. This protection of depositors and creditors also reduces the likelihood of panic responses to otherwise solid institutions.

In recent years a number of objections have been raised against the existing rules. Above all it has been pointed out that the rules neither enable sufficiently precise measurement of credit risk (i.e. risk weights fail to reflect the actual likelihood of default on the part of the debtor) nor recognise the value of modern risk management techniques. The consequences of these weaknesses could be mispricing and misallocation of credit.

The rules may also impact on the relationship between banks and the securities market as a source of finance. So long as the banks are required to hold 8% capital against loans to the financially soundest companies, direct borrowing in securities markets will probably be a cheaper form of funding for the latter. The fact that many banks in a number of countries have chosen to securitise assets is probably largely due to the capital requirements imposed on them. Moreover, the present rules do not explicitly cover risks other than credit and market risk.

The flaws in the present rules do not, in our view, impact equally on all types of institutions. Moreover, it will not be easy to put in place regulation that removes all the flaws mentioned and that at the same time fulfils the aims of the revision.

The banking system in Norway and several other European countries has not been sufficiently robust to avoid a banking crisis. Hence a central aim of the changes should be to ensure that the revision does not impair banks' overall financial strength. Leaving risk weights virtually unchanged while introducing increased acceptance of risk mitigation techniques will make this aim difficult to achieve. The Basel Committee and the Commission recommend the establishment of a separate capital requirement for other risks. These include operational risk, legal risk and reputational risk. In contrast to credit and market risk, these are not risks that banks assume in order to earn money. Norges Bank and Kredittilsynet support the establishment of a capital requirement to cover other risks. We would also point out that the capital adequacy rules do not cover risks such as concentration risk and risk associated with large exposures. Concentration risk refers to the risk associated with concentrating the loan portfolio in particular sectors and/or geographical areas, while risk attending large exposure refers to risk concentration in relation to individual borrowers. We would point out that both these risks could be substantial for a bank. While it is true that EU Directive 92/121 limits large exposures of credit institutions and investment firms, these do not impose explicit capital requirements in respect of the risk involved.

1.2 Accounting rules

The valuation of assets and liabilities is based on accounting principles and valuation rules. Under the capital adequacy rules the same principles and rules underlie the calculation of statutory minimum capital requirements. Adequate, harmonised accounting principles are consequently a precondition for consistent calculation of equity capital. Flawed accounting principles could result in overvaluation of capital and weaken the stability of the financial system. Harmonisation of accounting rules must take into account debt instruments eligible in own funds as well as the institution's equity capital. A concrete example of an area where variations are likely in the international context is subordinated debt denominated in foreign currencies.

Harmonised accounting rules are also important in securing a level playing field across national borders. Differing valuation rules could substantially affect the size of equity capital which may in turn lead to competitive distortions.

Individual countries' practice at implementation of accounting rules could also affect the size of equity capital. Some countries, for example, apply depreciation and write-down rules less conservatively than others. Larger depreciation and write-downs entail lower book equity leading to competitive distortions and less comparability between countries.

Further development of the EU accounting directives and extensive harmonisation of these directives with the IASC's accounting standards is particularly important.

2. Scope of the rules

The trend in the direction of increasingly complex financial conglomerates and ownership structures necessitates a close review of the content of the existing consolidation rules.

2.1 Heterogeneous financial conglomerates

In Norway the capital adequacy requirements for credit institutions and investment firms also apply to insurance companies at company and group level and thus also to mixed groups. The intention has been to ensure that the rules do not provide incentives to organise a group in such a way as to exploit possibilities for double-gearing, lower capital charges or transference of risk to institutions in the group which are not subject to capital adequacy requirements. Capital requirements should be neutral in relation to where in the group risk arises.

The capital adequacy treatment of heterogeneous financial conglomerates is under review by the Mixed Technical Group. The Group's study is based on work done by the Joint Forum. In view of the rapid evolution of the market in terms of the establishment of mixed groups, and the strong market position of these groups, Norges Bank and Kredittilsynet believe it is important to rapidly put in place harmonised capital adequacy rules for heterogeneous financial conglomerates.

2.2 Exemptions

Under the current EU legislation capital requirements apply at the group, sub-group and company level. However, the Directive on the Supervision of Credit Institutions on a Consolidated Basis (2CSD) Article 3(7) and CAD Article 7(7) provides for limited exemption from the capital requirements for sub-groups and/or at the company level.

Norges Bank and Kredittilsynet agree with the Commission that the current wording of 2CSD to the effect that "steps must be taken to ensure that capital is adequately distributed within the group" and the wording of CAD to the effect that "measures must be taken to ensure the satisfactory allocation of own funds within the group" is too vague to ensure satisfactory allocation of capital within a group. Furthermore, we support the basic principle that capital should be available to cover risk wherever it arises. Norges Bank and Kredittilsynet consider that to fulfil this principle, the capital requirements must be met at all levels of the group, both on a solo basis and a layer-by-layer basis. Norges Bank and Kredittilsynet therefore support alternative b) Repeal of the waiver. We would moreover point out that competitive conditions can be affected since countries which do not implement the waiver provisions will face stricter regulation.

2.3 Horizontal groups

The Commission points out that horizontal groups are not regulated by current directives since parent companies are not encompassed by the consolidation requirement. This is either because the parent company is a non-financial undertaking or that it is a financial undertaking from a non-EEA country, but may also be because the group does not have a hierarchical ownership structure.

Norges Bank and Kredittilsynet support the proposal to include horizontal groups under the consolidation requirement. Of the two proposed consolidation methods we consider alternative b) requiring horizontal groups to establish a holding company to be preferable to alternative a) entailing horizontal consolidation of subsidiaries. In our view an advantage of establishing a holding company subject to consolidation provisions is that it would give a more clear picture of the group's overall risk exposure than alternative a). Norges Bank and Kredittilsynet would however point out that cases may arise in which it is not possible to set up a separate holding company. This applies for example in relation to collaborating savings banks. In such cases alternative a) with horizontal consolidation or other principles for consolidation, aggregation or deduction would be more appropriate.

2.4 Minority interests

Under the current regime credit institutions and investment firms that are subsidiaries must be fully consolidated. The Commission points to a problem with this provision, i.e. that where a subsidiary with minority owners has "surplus" own funds after meeting the capital requirement on an individual basis, this "surplus" will of necessity not be available to cover losses elsewhere in the group. The Commission proposes that full consolidation should continue to apply, but that a deduction be made in own funds in respect of minority interests in accordance with a method to be detailed in due course.

The Savings Banks Association points out that in the savings banks sector there is a tradition of collaborative ownership of companies. The Association goes on to say "This is often a strategic choice, and owners feel a joint responsibility for the subsidiary regardless of whether they are in majority or minority owners. Against this background consideration should be given to introducing a right of dispensation permitting conglomerates with majority-owned subsidiaries that are financial institutions to employ pro rata consolidation."

Norges Bank and Kredittilsynet support the Commission's proposal for full consolidation of subsidiaries with deduction made in own funds in respect of minority interests. Where the choice of deduction method is concerned, we are aware that the following possible solutions have been discussed by the consolidation group under the TSG:

Minority own funds will only be included as group capital up to an amount equal to:

a. the own funds requirement of the subsidiary.
b. their participation in the requirements of the subsidiary.
c. the portion of the requirements of the subsidiary not covered by the group's capital.

Of alternatives a)-c) Norges Bank and Kredittilsynet would prefer alternative b). Alternative b) is identical to pro rata consolidation, but has the advantage that the entire asset side of the subsidiary shows in the group balance sheet.

3. Minimum capital requirements

The main objection to the existing rules is their failure to provide for sufficiently precise measurement of credit risk. An argument advanced against the introduction of a more risk-based regime is its tendency to be pro-cyclical: a cyclical downturn or the effect of external shocks may be magnified by a requirement for higher capital. Higher capital requirements prompt a reduction in lending by financial institutions which may bring about a credit crunch. The result is a higher rate of company liquidations with the risk of forcing the economy in a downward spiral. Norges Bank and Kredittilsynet would point out that this problem may just as well be due to belated recognition of the increased risk as to insufficient initial pricing of credit risk and inadequate loss provisions. Nor will a regulation that are less sensitive to risk (like the present one) have a stabilising effect in the somewhat longer term, as witnessed in previous financial crises. In our view harmonised, more dynamic loss provision rules are needed. We therefore support the establishment of a regime that provides a better alignment between risk and capital charges.

3.1 Standard method - external rating

External credit rating is used both in country risk assessments and in connection with the major banks' funding activities. In our view competition among rating agencies could mean that considerations other than assumed credit risk might influence the individual rating agencies' risk classification, and that borrowers seeking a better rating will be able to indulge in a form of rating shopping. Criteria for classifying borrowers who have been rated by more than one rating agency also need to be established. In a longer-term perspective Norges Bank and Kredittilsynet believe that greater use of external rating will have a positive effect on the evolution of the European capital market.

Norges Bank and Kredittilsynet note that the proposal clearly encourages counterparties carrying the highest risk to refrain from being rated so long as they can achieve a lower weighting by not doing so. Moreover, it is not apparent that an external rating entails an obligation to employ the assigned classification. Hence the result may well be that the key premise of achieving rules that are more risk-based will only apply to assets attracting a lower weight as a result of the revision. This type of asymmetry will in our view be a major flaw in the regime. Norges Bank and Kredittilsynet would also point out that the proposed risk categories will continue to entail a very coarse-meshed system in which risk may vary widely within each weight category.

The Norwegian Financial Services Association (FNH) is of the view that both the standard method and the method based on internal rating lack an explicit technique for measuring credit risk. The Association considers that the weaknesses of the standard method in particular can be reduced by introducing four new risk categories in addition to the existing five categories.

3.1.1 Weighting of claims on sovereigns

Norges Bank and Kredittilsynet agree with the Commission and the Basel Committee that drawing a distinction between member and non-member countries of the OECD is not necessarily meaningful when it comes to assessing the likelihood of default by states.

The Savings Banks Association has doubts about external rating agencies' competence in assessing country risk, and calls for an initiative to be taken to study possible alternative solutions.

Although rating agencies have few and varying results to refer to as regards risk classification of states, Norges Bank and Kredittilsynet believe that in the absence of anything better, and so long as most states are rated, then using external rating to establish risk weights for states is acceptable.

3.1.2 Weighting of claims on banks

Assuming that the weighting of sovereigns is changed in line with the proposal, we agree with the Commission that the current weighting of claims on banks will no longer be meaningful.

Two alternative proposals are to hand. In relation to alternative 2 most claims on banks, including banks not subject to rating, will be assigned a 50% weighting. Only the very soundest banks will achieve the current 20% charge. Alternative 1 entails a 20% risk weight for banks in countries with a good rating.

The Commission states that alternative 1, in which claims on banks are assigned a risk category higher than the country's risk category, may contravene the Treaty of Rome in that it exposes banks to discrimination based on nationality. Against this background a possible third alternative is launched in which claims on institutions in the EEA area receive a common weighting.

Since banks are subject to supervision, capital adequacy requirements and have access to loans from a lender of last resort, they may warrant lower weights than other business activity. A weakness of alternative 1 is that it penalises the best banks, while the relatively speaking weaker ones come off better. The possible third alternative will also entail no improvement on the existing rules both because nothing is done to make the risk weights more risk-sensitive and because an artificial distinction is drawn between institutions within and outside the EEA area corresponding to Zone A and Zone B under the current rules.

Although alternative 1 does not conform with principles of "risk-based" weights, alternative 2 entails substantial disadvantages for small countries with small banks. Based on the advice offered by the Norwegian banking industry, Norges Bank and Kredittilsynet express support for alternative 1. Based on the same arguments, we are in general in favour of a closer assessment of a possible alternative 3.

Norges Bank and Kredittilsynet would in all events point out the need for harmonised rules.

For short term claims on banks (a definition of 6 months proposed) the Basel Committee and the Commission propose, under alternative 2, a one-category lower risk weight than the weight applied to long-term borrowing. Hence as regards interbank financing there will be no difference for Norwegian banks between alternative 1 and alternative 2, apart from in the case of loans with a term between 6 and 12 months. Norges Bank and Kredittilsynet support such a proposal.

3.1.3 Weighting of claims on corporates

In order to reduce the competitive disadvantage between countries with varying incidence of external rating, the Commission will consider a set of criteria that can be used to determine risk weights for unrated undertakings. Norges Bank and Kredittilsynet are in general in favour of further development of the standard method, but find it difficult to adopt a view on the proposal until a draft version of such criteria becomes available.

3.1.4 Assets carrying high risk

Norges Bank and Kredittilsynet support the proposal to assign shares a higher risk category than other assets since shares are the most junior of obligations. A solution in which shares attract a risk weight higher than the corresponding weight given to senior obligations, with a cap of 150 per cent, seems reasonable.

The Commission considers that allowance should be made for concentration risk in the capital adequacy framework but envisages practical problems in implementing this in the framework. Norges Bank and Kredittilsynet support the Commission's view that capital charges for concentration risk should be considered under Pillar II of the Supervisory Review Process.

3.1.5 Short-term commitments

Under the current regime exposures granted but unutilised general loan facilities with an agreed term exceeding one year attract a conversion factor of 50% while those with a term shorter than one year are not subject to a capital requirement. Experience show that banks often adapt to the rules by structuring such products with a term of less than one year, and subsequently rolling the commitments. The Commission points out that some risk also attaches to short-term exposures and proposes two alternatives where either the conversion factor is raised to 20% for terms below one year or a common risk weight is assigned independent of term. Given that there is empirical support for the contention that losses on general loan facilities are lower for short maturities, it would be sensible to continue to provide incentives for banks to grant this type of loan on short terms. If not, Norges Bank and Kredittilsynet believe that the conversion factor should be independent of term.

3.1.6 Securitisation

Tranches of securitised loans issued in the international capital market are generally rated, and risk classification by external rating agencies should therefore be well suited to fixing capital charges on such assets. As the proposal from the Basel Committee and the Commission now stands, the weights on securitised claims could turn out to be lower than if the bank has a direct claim in its balance sheet. The proposal can probably be justified in terms of the fact that the loan portfolio underlying the securitised claims is broader based and therefore carries lower risk. At the same time a lower risk weighting of securitised claims will encourage asset sales by banks.

3.1.7 Criteria for approval of external rating agencies

Given the significance that the proposal recommends should be given to rating agencies in the capital adequacy context, Norges Bank and Kredittilsynet consider it to be particularly important that these agencies' assessments are independent and of a high professional standard. As a starting point the Commission lists the seven criteria which the Basel Commitee considers must be met by rating agencies in order for them to be employed for the purpose. While we have little knowledge and sparse experience of rating agencies' activities, we view the proposal as a sound basis for establishing such criteria.

Where approval of rating agencies is concerned, we would prefer this not be assigned to the national level. A solution based on national approval may be problematic in relation to the need for a level playing field, it may prove unnecessarily complicated and it could require competence and resources far in excess of those possessed by many supervisory authorities today. Norges Bank and Kredittilsynet therefore consider alternative iv) Subsidiarity or v) Centralised Recognition, which entail central level approval, to be preferable. However, the alternatives involving central level approval would require further clarification of who is to be assigned such authority. Norges Bank and Kredittilsynet support the idea of the Commission being made the central authority for approval.

3.2 Internal rating

The aim of the revision of the capital adequacy regulation is to ensure that capital charges to a greater degree reflect the institution's (credit) risk. Both the Basel Committee and the Commission are of the view that the use of the institutions' own internal rating systems (IRB) is more likely to ensure that capital charges correspond to the individual institution's risk than does the present system which is more risk-neutral. The use of internal rating systems will also help to ensure that regulation follows the market in terms of developing a best practice for risk management.

Norges Bank and Kredittilsynet share the view of the Basel Committee and the Commission that use of internal rating systems could help to meet the overarching premise that capital charges should be more aligned with the individual institution's risk profile. We would none the less point out that a number of practical challenges have to be dealt with before the method can be viewed as well-suited for regulatory purposes. Moreover, complex calculation systems do not automatically guarantee higher quality than simplified solutions.

Which institutions should be entitled to apply their own internal rating systems as a basis for calculating capital charges will depend on what is meant by the term internal rating and what minimum requirements must be met in order for the institution to be able to use its internal rating system. Most Norwegian institutions currently employ more or less advanced classification systems for internal monitoring and management of credit risk. However, only a minority of these could be expected to satisfy the requirements that are likely to be made of systems intended for capital adequacy purposes.

Norges Bank and Kredittilsynet share the view that banks normally have a broader and deeper knowledge of customers than do external rating agencies. A number of qualitative factors play a part in assessing an exposure which rating agencies are not necessarily able to capture in their analyses. Norwegian banks can also be assumed to have a deeper knowledge of macroeconomic and sector-specific factors in Norway than do international rating agencies, and the same will apply in the majority of small economies. Credit assessment of customers is likely to be more thoroughly grounded and documented by the banks than by external rating agencies, and the IRB approach will in all probability provide a more correct picture of the customer's credit risk than will external rating. A further advantage of IRB compared with external rating is that the bulk of the corporate loan portfolio will be classified.

Today's internal rating systems were not designed with capital adequacy calculations in mind. A potential problem is the temptation for institutions to establish IRB systems whose primary purpose is to minimise the capital requirement rather than to give as correct a picture as possible of the credit risk attending the individual customer. The authorities must be entitled to impose penalties on banks that systematically understate their risk.

Another central issue is whether use of internal rating should be confined to risk classification of claims on corporates, parts of the banking book or the entire banking book. Another important concern when formulating a set of rules is the relationship between quantitative and subjective assessments in the system, and how a level playing field should be secured both nationally and between countries. The aim must be to ensure that an internal rating regime produces the same capital requirements for the same debtor in all banks. If such standards are not achieved, the result will be differing national implementation of the rules with associated distortion of competitive conditions between institutions.

Consistency between calculations pursuant to the institutions' internal rating systems based on various risk categories and risk weights and pursuant to the current standard method is crucial to the suitability of internal rating systems for capital adequacy purposes. If it does not prove possible to achieve this consistency, arbitraging could arise with exposures being moved between the two calculation methods. Consistency will also be decisive in securing competitive equality between institutions using the standard method and those using internal rating systems.

The role of supervisory authorities will change appreciably with the introduction of internal rating in the capital adequacy framework. With the introduction of internal rating the authorities will be required both to evaluate the systems and in the event review the institutions' own assessments. It is important to ensure that the approval process does not put the authorities in a "position of responsibility" which could influence any subsequent review they may carry out.

In contrast to the Basel Committee the Commission believes that the right to employ internal rating systems should apply to a wider group of institutions, and not only be confined to the most sophisticated banks. Norges Bank and Kredittilsynet share this view.

3.3 Risk mitigation techniques

The current capital adequacy framework takes only limited account of risk mitigation techniques in connection with calculation of capital adequacy.

Norges Bank and Kredittilsynet are in principle in favour of greater acceptance of the use of risk mitigation techniques in the capital adequacy context, and consider it important to create incentives in the form of capital relief for the use and further development of such systems and techniques. The key issue is how the rules should be designed with a view to accommodating the techniques as far as possible. Norges Bank and Kredittilsynet would point out that the rules will become more complex as a result of this proposal.

The Commission presents two approaches to taking such techniques into account when calculating capital charges:

- Developing criteria for each individual technique (e.g. netting, credit derivatives etc.).

- Developing general principles for use across different techniques.

The Commission prefers the second alternative. In addition to the general principles it indicates the need for more explicit, supplementary guidelines. There will regardless be a need to clarify factors associated with individual instruments and techniques in order to avoid regulatory arbitrage.

Norges Bank and Kredittilsynet recognise the ability of general rules to cover a wide range of hedging instruments and techniques and their robustness to new methods. On the other hand general rules entail a risk of regulatory arbitrage since such rules might in our view inevitably gravitate towards those implemented in the most liberal country. Moreover, Norges Bank and Kredittilsynet believe that the task of interpreting general rules will necessitate substantial extra work both for institutions and the supervisory authorities. We base this view on the considerable effort put into interpreting the current simple regulation of risk mitigation. Moreover, a more complex body of rules will also necessitate increasing the complexity and scope of administration and oversight of compliance.

We would also point out that the complexity of risk mitigation techniques in general requires that sufficient focus has to be given to management and control of operational risks arising with the use of such techniques, including special risks associated with legal factors and documentation risk.

The proposed methods raise a number of technically complex issues at the same time as the presentations and discussions contained in the paper are of a relatively general nature. This limits the possibilities for concrete feedback on the methods proposed.

Norges Bank and Kredittilsynet support the proposal to allow capital relief on the use of hedging techniques even where an uncovered risk is present due to the hedging instrument having a shorter term than the hedged exposure. In our view the framework should stimulate prudent risk management.

The central concern in the work ahead is to meet the requirement of a level playing field.

3.4 Capital requirements for other risks and interest rate risk in the banking book

3.4.1 Capital requirements for other risks

A capital requirement for other risks will serve as a buffer against risks which are not explicitly covered elsewhere in the capital adequacy framework but which none the less make a substantial contribution to the overall risk in credit institutions and investment firms. The current capital adequacy rules for credit risk are relatively crude, and are assumed to implicitly cover other risks. A more precise measurement of credit risk, as entailed by the proposed revision of the capital adequacy rules, may lead to the reduction of the buffer in the existing regime.

However, several problems are attached to setting a special capital requirement. Other risks constitute a highly complex group of risk sources. Moreover, it may be difficult to draw a clear distinction between other risks on the one hand and credit and market risk on the other. Moreover, the significance of other risks may vary widely between different areas of activity.

The Commission believes that the introduction of a capital charge, even in a crude format, will focus attention on such risks and encourage better management and control. Norges Bank and Kredittilsynet share this view and would point out that fine-tuning the capital charge via the supervisory review might encourage improved management of such risk.

The Commission proposes formulating the standard method for determining the capital charge as a linear function of the institution's size and earnings. Norges Bank and Kredittilsynet consider this to be an acceptable approach in light of the fact that institutions' own systems are currently insufficiently developed to quantify other risks.

As regards the indicators in the linear formula, consideration could be given to splitting the income element into several components. This could capture business areas' differing levels of exposure to operational risk.

Norges Bank and Kredittilsynet would also point out that residual operational risk may attend asset securitisation. Capital requirements could reasonably be imposed in respect of such risk, and we would request the formulation of a harmonised standard with this in mind. The absence of international rules governing the operational aspect of securitisation has resulted in a variety of rules among countries. In our view these requirements need to be harmonised to ensure a level playing field and to reduce the risk of regulatory arbitrage.

3.4.2 Capital requirements for interest rate risk in the banking book

Interest rate risk can represent a substantial risk for financial institutions. Based on the premise of establishing a capital adequacy framework more aligned with the individual institution's risk, Norges Bank and Kredittilsynet agree that capital requirements should also be applied to interest rate risk in the banking book. CAD requires capital backing for interest rate risk in the trading book. Setting a separate requirement for such risk in the banking book reduces the opportunities for arbitraging between the two books. This too calls for the establishment of an appropriate capital requirement.

Our general view is that treatment of interest rate risk for positions in the banking and trading books should be harmonised. Norges Bank and Kredittilsynet support the proposal to define the banking book as those positions which are not included in the trading book.

The Commission proposes that institutions with little or no banking book activity should be exempted from calculation of interest rate risk in the banking book. It is made clear that exemption is not intended to exclude small institutions whose interest rate exposure is mainly via positions in the banking book for associated capital charges. Norges Bank and Kredittilsynet therefore consider it most appropriate to base the criterion for such exemption on the size of the banking book relative to the institution's total activity. A possible refinement of this criterion would be to consider interest income's share of the institution's total income in addition to the first-mentioned criterion.

The Commission further proposes that capital backing should only be required for interest rate risk in excess of a given threshold. The threshold is defined such that only institutions whose capital is reduced by more than X per cent in the event of a parallel shift of N base points in the interest rate curve in all currencies should be subject to capital charges. Any capital charge thereby triggered equals that part of the change in the institution's capital that exceeds Y per cent. Institutions whose interest rate exposure in the banking book is below this threshold will thus be exempted from this capital charge. Norges Bank and Kredittilsynet support the principle of establishing such a free zone.

Norges Bank and Kredittilsynet support the Commission's proposal to apply the duration method as the standard method for calculating interest rate risk in the banking book. We also support the proposal to empower supervisory authorities to approve the use of the maturity method by institutions with a simple risk profile and negligible interest rate exposure. Norges Bank and Kredittilsynet are in favour of this harmonisation of the principles for calculating capital charges for interest rate risk in the trading and banking book.

The Commission proposes that institutions be permitted to employ internal risk management models to calculate capital charges for interest rate risk in the banking book. We are in favour of this proposal. EU Directive 98/31/EU currently permits the use of internal models to calculate capital charges for interest rate risk in the trading book. Which parameters should be changed in the model to enable them to be used to calculate interest rate risk in the banking book should in our view be based on a thorough examination of factors specific to interest rate instruments included in the banking book.

4. Pillar II - supervisory review of capital adequacy

The purpose of the Supervisory Review Process is to review soundness and strategy for capitalisation in banks and investment firms and to ensure that the particular institution's own funds is consistent with its risk profile. The supervisory authorities will be secured a legal basis for determining individual capital charges in cases where they consider the institution to be insufficiently capitalised. The justification for individual minimum requirements is that institutions' risk profiles differ and that the general minimum requirements fail to capture this disparity.

Norges Bank and Kredittilsynet support the principles underlying the Supervisory Review Process. The proposal could promote soundness in the financial industry via more effective supervision at the same time as the institutions have an incentive to develop and improve their risk management processes.

Norges Bank and Kredittilsynet would none the less point out that implementation of Pillar II will be resource- and competence-demanding both for the supervisory authorities and for the banks under supervision. The resources needed will in part depend on the number of institutions under supervision and their complexity.

Further, the supervisory authorities may find it difficult to determine how risk profiles or possible flaws in the risk management process can be assigned higher capital charges. Flaws in the risk management process require a qualitative assessment which cannot be directly translated into a quantitative charge. At the same time persuasive arguments will be required of the supervisory authorities if a higher capital charge is to be imposed on banks with well-founded, complex risk management systems.

The Commission states that the Supervisory Review Process is also relevant in relation to other risks and internal rating. Norges Bank and Kredittilsynet would point out that substantial progress has yet to be made in developing methods for quantifying other risks, including operational risk. Any capital charges for such risks must therefore be based on crude methods as described in the Commission's proposal. In order to secure a reasonable degree of consistency between the capital charge and the bank's own risk management the capital charge in respect of, for example, operational risk should not be assessed independently of Pillar II and the internal control environment in the institution as a whole.

The Commission proposes that the Supervisory Review Process should be implemented in legislation through legally binding articles in Directives and through guidelines for practising such provisions. Norges Bank and Kredittilsynet support this approach and have no comments on the proposed provisions and guidelines. We agree with the Commission that individual adjustments will be needed, including adjustments resulting from disparities between different countries' legal basis and supervisory methods. Even so, harmonised standards should as far as possible be established. A higher degree of individual adjustment could easily give rise to differing competitive conditions and impede comparison between banks in different countries.

Norges Bank and Kredittilsynet would point out that Pillar II entails a significant change in supervisory method. The proposal will entail a more active role for the supervisory authorities in assessing and approving banks' risk management systems, overall risk, level of capital adequacy as well as the power of supervisory intervention. This must not be allowed to undermine that it is the responsibility of bank owners and managers for proper risk management and for taking appropriate action should problems arise.

5. Pillar III - market discipline

The Commission proposes the introduction of a requirement to publish information of relevance to assessing a bank's capital adequacy. This proposal is in keeping with the Basel Committee's proposal. The reasoning is that the market will have a disciplinary function in that it will reward banks which manage risk effectively and have a strong capital ratio, but will penalise banks whose risk management is poor and capital ratios low.

The proposal calls for the furnishing of information that is of material significance to analysing the adequacy of a bank's capital and which permits assessment of the bank's capacity to withstand future losses.

Norges Bank, Kredittilsynet and the Norwegian banking associations support the proposed introduction of Pillar III. Greater transparency will promote better information to decision-makers, and will benefit the financial system as a whole. It may be noted that an information requirement operates in Norway which is in all essentials in keeping with the proposal. Some information will, however, be confidential or market-sensitive and will not be publishable.

The Financial Services Association requests that individual capital requirements should not be published. The justification is that such publication would in effect turn the supervisory authorities into rating agencies, and that national supervisory authorities' differing practice of the rules would result in the information not being readily available to players who lack thorough knowledge of the rules.

Norges Bank and Kredittilsynet would point out that although publication would promote market control, it could trigger an accelerating negative effect on banks' financial position. On the other hand, publication would facilitate comparison between institutions.


Bjørn Skogstad Aamo

Svein Gjedrem
General Director Governor
Kredittilsynet Norges Bank

Publisert 31. mars 2000 09:19