Possible further changes to the Capital Requirements Directive – Comments on the Commissions Service Staff Working document

 

European Commission              
Directorate-General Internal Market and Services 

markt-h1@ec.europa.eu  

Date: 16.04.2010

Finance Norway (FNO) was established in 2010 by the Norwegian Savings Banks Association and the Norwegian Financial Services Association and represents some 180 financial institutions operating in the Norwegian market.  


 

Dear Sir/Madam;

Finance Norway welcomes the opportunity to comment on the consultative document of March 2010 regarding further possible changes to the Capital Requirements Directive (CRD). When it comes to the details of the proposed framework, we fully support the views presented by the European Savings Banks Group (ESBG) and the European Banking Federation (EBF). The intention behind presenting some views separately is to highlight the Norwegian – and smaller market – perspective that we believe should be taken adequately into consideration when amending the directive.

The need for reform

The recent financial crisis has documented the need to improve risk management in the financial sector. We therefore support the ongoing work to avoid future situations like the one that has now dampened the economic outlook for nearly two years.

However, we are deeply concerned with the general approach of the proposed new regulatory environment, and would caution especially against a regime based on increased focus on regulatory quantitative “parameters”, especially when it comes to liquidity risk. In our view, historic experience – mainly from the 1960-1980’s - shows that quantitative requirements over time tend to direct economic behaviour in a way that results in large imbalances in the economy. When markets adjust – sooner or later, the final consequences of the regulation may be substantial unpredictable and undesirable effects. Norwegian experience from the late 1980s and early 1990s is a good example of such mechanisms. Liberalisation of financial regulations first allowed scope for increased private consumption, with steeply rising house prices, and later – as interest rates rose and regulations concerning banks’ capital tightened – resulted in major financial problems for the banks. Later experiences with liquidity requirements show clearly how difficult it is to set parameters that yield the desired results at industry level.   

Experience also shows that compulsory quantitative requirements tend to make market participants lean back on the minimum requirements and invest resources in finding ways to reducing the regulatory burden. If such behaviour implies that less effort is put into risk management within banks, it is not certain that the overall result will be a system with less total risk.  Regulation can also distort the level playing field, in a worst case scenario in a way that gives rise to unregulated (“grey”) markets.

We would caution especially against an overly tight and prescriptive one-size-fits-all framework which does not take account of institutional, market or country-specific factors.
 
The leverage ratio

A leverage ratio, as suggested, represents a quantitative regulation which may well reduce the banks’ focus on actual risks, as institutions may be compelled, unintentionally to draw on resources from the work on risk-based management.

It seems highly inappropriate, irrespective, to impose the same required leverage ratio on all types of institutions. If achieving such a ratio is intended to ensure that risk is limited to a defined level, it must address the differences in the institutions’ business models. A common required leverage ratio will imply extra costs for low-risk businesses, like mortgage banks, and  (relatively) lower costs for institutions providing more inherently high-risk-services, like investment banks.   

The Norwegian banking crisis in the early 90-ies was met with both a qualitative and quantitative risk-based approach. Norwegian banks are sound today, and none experienced solvency problems during the last financial crisis. The liquidity challenges that arose were in essence caused by the international market failure.

Liquidity requirements

The recent financial crisis has documented the need to improve liquidity risk management and regulatory liquidity risk standards.  FNO therefore supports the introduction of a quantitative standard as a supplement to the existing qualitative approach. This standard should consist of a short term ratio that focuses on the adequacy of a financial institution’s liquidity buffer and a long term ratio that focuses on the structure of its funding.

In addition to our general concerns about too strong a focus on quantitative regulations, we are, however, concerned by the proposed calibration and definitions, which we find far too conservative in the light of national experience and market size, and the likely negative consequences for the traditional relationship-driven banking model.  We fear that a regulation such as the one indicated will have a negative impact on both the financial system and the broader economy.

Our comments are presented in the following as replies to the individual questions presented in the consultative document.

Question 1: Comments are sought on the concept of the Liquidity Coverage Requirement and its likely impact on institutions' resilience to liquidity risk. Quantitative and qualitative evidence is also sought on the types and severity of liquidity stress experienced by institutions during the financial crisis and – in the light of that evidence – on the appropriateness of the tentative calibration in Annex I. In particular, we would be interested in learning how the pricing of banking products would be affected by this measure.

As stated above, a Liquidity Coverage Ratio/Requirement (LCR) could have positive effects as a standard (not requirement). However, the definition/calibrations suggested do not  take national differences into account to the necessary extent. This is a major weakness as the liquidity requirements are meant to be fulfilled in each single currency separately. Another clear weakness is the lack of recognition of the fact that individual institutions and markets are hit differently by shocks.

The money market for Norwegian kroner is highly dependent on a well functioning US-dollar money market, and liquidity reserves has to a large extent to be based on holdings of foreign securities that are accepted by the Norwegian Central Bank as collateral for loans. The need to hold such assets has again increased banks’ dependence on borrowing in markets abroad. Therefore, the financial crisis hit the Norwegian money markets relatively hard.

The dependence on foreign markets must be seen in light of the fact that the Norwegian money and securities markets are small. In addition to being a consequence of a small economy, this reflects the central governments budget surpluses for the last few decades, which has resulted in net public wealth of 140% of GDP. The main reason that the government nevertheless issues a limited volume of government bonds is to maintain a risk-free yield curve in Norway.

On the demand side, we have witnessed an increasing demand for Norwegian government bonds from foreign investors who now hold about 64 percent of the total outstanding volume. Life insurance companies and pension funds own about 20 per cent.  A considerable part of issued government debt is thus held by long term investors, which reduces the liquidity of the securities.

The proposed narrow definition of what are considered “best” highly liquid assets implies that the demand for the limited number of Norwegian government bonds will increase even more. As a result, an even larger part of these bonds will remain in the balance sheet of banks due to regulatory herd behaviour and with possible adverse price movements as a consequence. Clearly there is also a danger that the sale of government bonds in such a regime will be perceived as a liquidity problem in the selling bank. 

We consider it unlikely that the volume of outstanding public debt, denominated in Norwegian kroner, will be sufficient in the foreseeable future to meet the demand that the suggested LCR would trigger. The definition of “high liquid assets” must generally take account of the fact that the national markets differ in respect of what assets are actually available and considered as liquid; otherwise the main effects of the regulation will be adverse. A situation where highly liquid assets are limited to assets that withdraw liquidity from the market on issuance must also be avoided. This is the case for Norwegian government securities, where the government does not need to borrow in order to spend, and the liquidity paid in is being withdrawn from the market into government accounts at the central bank.

In economies where deep and active markets for government debt do not exist, covered bonds represents assets where credit risk is well separated from both risk associated with the issuer and risks associated with the overall performance of the issuers’ industry.  Covered bonds should therefore be treated as liquid assets nearly equivalent to government bonds. In Norway, the issuing of covered bonds is regulated by the Financial Services Act, which gives the bondholders a preferential claim over the cover pool in case of bankruptcy. Thus the “haircuts” to be applied should be far less than the proposal indicates.

We fully agree that the central banks should not be regarded as lenders of first resort. However, there are again market differences that have to be taken into consideration. In the case of Norway, for the last 20-25 years the money market has been highly dependent on the central bank’s lending facilities. This reflects the government’s budget surpluses and the choice to keep central government liquidity with the central bank. Macro liquidity in Norway therefore fluctuates significantly on a daily basis and withdrawn liquidity (taxes etc.) from the money market has to be sterilized by either a reduction in central bank reserves or by borrowing from the central bank. The system has resulted in structural central bank lending over longer periods of time. Therefore, borrowing from the central bank can not automatically be regarded as unfortunate behaviour on the part of banks; it may be a result of public choices/decisions, and reserves in the central banks should be considered as highly liquid, irrespective of whether they take the form of deposits or available credit bases on standing credit lines.

Central banks, for their part, should be urged to make their criteria for lending clear and transparent, to avoid uncertainty about availability  in times of stress.

In the absence of information from impact studies, we find it inappropriate to suggest what impact level an LCR would have on prices. However, it seems apparent that the effect would be higher prices, even if one takes into consideration the negative effects on overall economic activity as a result of the reduced credit-multiplier.

Question 2: In particular, views would be welcome on whether certain corporate and covered bonds should also be eligible for the buffer (see Annex I) and whether central bank eligibility should be mandatory for the buffer assets?

As stated in connection with question 1, the Norwegian market will be dependent on the eligibility of Norwegian covered bonds as a buffer. The proposal implie eligibility for bonds not issued by the individual institution itself. We can not see the rationale behind this restriction, as credit risk is separate from the issuer’s position, but emphasise that it should be explicitly clarified that this restriction does not limit the eligibility of covered bonds issued by separate legal entities.

According to the proposal, ”high quality liquid assets” must be both traded in an “active and sizable” market and be central bank eligible. We find this combination of requirements both unnecessary and unfortunate. “And” should be replaced with “or”, because:

  • Central banks’ eligibility criteria are set in the light of the individual central bank´s credit risk perspective, and cannot at the same time automatically be seen as appropriate for a bank’s liquidity risk limitation purposes.
  • Central bank eligibility gives access to the highest form of liquidity, regardless of the size of the market in which the eligible asset is traded. Possibly with a haircut equivalent to the haircut stipulated by the central bank.

We also believe the proposed combined conditions will represent competitive disadvantages for institutions located in the smaller non-euro markets because these markets are generally less “active and sizable”. Central banks may base their requirements on an assessment of the relevant assets that are available in both domestic and international markets. Therefore, general acceptance of assets with central bank eligibility as being eligible for the buffer also could be used to lessen the burden due to the lack of national ”deep and active markets”.

Question 3: Views are also sought on the possible implications of including various financial instruments in the buffer and of their tentative factors (see Annex I) for the primary and secondary markets in which these products are traded and their participants.

In the absence of a “deep and active” market for government bonds, Norwegian financial markets will be dependent in coming years on the market for covered bonds as the basis for low-risk references. The inclusion of covered bonds in the buffer on fair terms is accordingly essential for the sound development of such a market.   

Question 4: Comments are sought on the concept of the Net Stable Funding Requirement and its likely impact on institutions' resilience to liquidity risk. Quantitative and qualitative evidence is also sought on the types and severity of liquidity stress experienced by institutions during the financial crisis and – in the light of that evidence – on the appropriateness of the tentative calibration in Annex II. In particular, we would be interested in learning how the pricing of banking products would be affected by this measure.

A Net Stable Funding Requirement (NSFR) with the proposed 100 per cent floor will reduce the banks’ ability to transform short term deposits into longer term credits, and the role of banks would virtually be limited to distributing liquidity. Less maturity-transformation will clearly result in lower liquidity/funding risk.

The transformation function is a key element of banks´ services to the rest of the modern economy – a task that is of vital importance. If this function is disturbed, coupled with a negative impact on banks capacity to lend, as new deposits/funding can not be used to the same extent for lending to consumers and businesses, a clear negative impact on overall economic activity must be expected.  This comes at a time when banks’ lending capacity is already hit by market and regulatory requirements for increased capital ratios.  These effects, together with the danger of replacing bank lending with borrowing from other (less regulated) sources, must be clearly explained to policymakers before decisions are made. It would be a paradox if the regulation had the consequence of introducing new unregulated players to take on the function of maturity transformation.

The suggested composition/calibration of the NSFR does not take into consideration that the impact on funding will depend upon the source of stress: institution-specific problems or market-wide turbulence. In the latter case, one should recognise that a “run-down of deposits” is likely to take the form of customers redistributing deposits to reduce exposure to individual financial entities. The total amount of deposits will thus remain relatively stable. 

Question 5: Comments are in particular sought on the merits of allowing less than 100% stable funding for commercial lending that has a contractual maturity of less than one year. Is it realistic to assume that lending is reduced under liquidity stress at the expense of risking established client relationships? Does such a differentiation between lending with more and with less than one year maturity set undesirable incentives that could discourage for instance long term funding of non-financial enterprises or encourage investment in marketable securities rather than loans?

Our general view is that banks should be given the possibility to tailor their funding to their particular circumstances and experience. This will be in line with the principles behind the Basel II framework which encourage banks to improve risk measurement and management based on their specific business model and experience. This will also reduce the risk of regulatory failure due to herd behaviour.

A natural consequence of the proposed “clear cut” 1-year horizon for the NSFR is larger price differences between  short- and longer-term lending. Marketable securities as an alternative to deposits will increase competitiveness for both banks and banks´ customers, hereunder perhaps especially as an alternative to standard deposits.  

Question 6: Views are sought on possible implications of inclusion and tentative "availability factors" (see Annex II) pertaining to various sources of stable funding for respective markets and funding suppliers. Would there be any implications of the tentative required degree of coverage for various asset categories for respective bank clients?

Experience from Norway is clearly that the various forms of funding are less likely to diminish than the proposed NSR seems to assume. In particular, the proposed significant haircuts on deposits do not correspond with Norwegian experience neither from the last financial crises or the Norwegian banking crises in the early 1990s. The Norwegian government decided not to make state guarantees for deposits during the recent financial crisis. Nevertheless, we did not experience outflows of any significance from well established banks.

Norway is dominated by traditional relationship-driven banking models with a large SME sector that is not able to access the bond market itself. Furthermore, time deposits only account for a minor part of the overall deposit structure. Our experience is that in periods of stress depositors, who typically are also borrowers, will act cautiously and not move deposits out of “their” bank. 

We caution against being overly prescriptive in setting behavioural overlays that should apply to outflows and inflows given the difficulty in defining stable vs non-stable sources of funding. We also believe that a key explanation can be found in the existence of a well funded deposit insurance scheme with a high coverage.

Question 7: Do you agree that all parameters should be transparently set at European level, possibly in the form of Technical Standards by the EBA where parameters need to reflect specific sub-categories of retail deposits?

Ideally, the regulation for financial institutions should be common to the whole EEA. However, in practice regulations must take into account national differences in markets, bank structure, and the different types of services provided to customers.   

Question 8: In your view, what are the categories of deposits that require a different treatment from that in Annexes I and II and why? Please provide evidence relating to the behaviour of such deposits under stress.

Norwegian experiences during both the financial crisis and the earlier Norwegian banking crisis show clearly that deposits covered by guarantee fund should be treated as stable funding, with an accordingly high “availability factor”.

Question 9: Comments are sought on the scope of application as set out above and in particular on the criteria referred to in point 17 for both domestic entities and entities located in another Member State.

Application of the regulation should be at the level on which the regulated entities manage their liquidity in the different jurisdictions, combined with special treatment of items that are more/less available due to connections within a financial group.   

Question 10: Should entities other than credit institutions and 730K investment firms be subject to stand-alone liquidity standards? Should other entities be included in the scope of consolidated liquidity requirements of a banking group even if not subject to stand-alone liquidity standards (i.e. financial institutions or 50K or 125K investment firms)?

Generally, regulations must be made and implemented in a way that does not distort the competition between types of financial institutions, and especially not in a way that gives incentives for establishing new types of unregulated service providers (“Grey-market”). However, regulations must also be implemented in a way that takes into consideration type of business and inherent risk. 

Question 13: Do stakeholders agree with the conclusion that for credit institutions with significant branches or cross-border services in another Member State, liquidity supervision should be the responsibility of the home Member State, in close collaboration with the host member States? Do you agree that separate liquidity standards at the level of branches could be lifted based on a harmonised standard and uniform reorganisation and winding-up procedures?

Seen from the perspective of effectiveness, of both banks and supervisors, it would be highly desirable if one could place the responsibility of supervision on the institutions’ home states. However, this approach requires that liquidity standards are imposed in a way that does not give rise to competitive distortions.

Question 14: Comments are sought on the merit of using harmonised Monitoring Tools, either in the context of Supervisory Review or as mandatory elements of a supervisory reporting framework for liquidity risk. Comments are also sought on the individual tools listed in Annex III, their quality and possible alternatives or complements.

Generally, we find the proposal of harmonised Monitoring Tools to the best of the proposed tools when it comes to reducing the risk of future liquidity stress scenarios, both in separate institutions and market-wide. Harmonised monitoring will make it easier to benchmark own exposure to the guidance/recommendations from supervisors, and in principle well in line with a risk-based approach.  

Question 15: What could be considered a meaningful approach for monitoring intraday liquidity risk?

Intra-day liquidity risk is dependent on several factors, spanning from customer- and institution-specific differences to the specifications of the national infrastructures. In this context, we find it difficult, perhaps impossible, to suggest one common approach to monitoring intraday liquidity risk.


On a general level, Finance Norway supports the underlying intentions of the proposal. However, the proposal as it now stands should be reconsidered in light of the negative consequences it will have on the banking industry and the economy as a whole.

Yours sincerely,

Jan Digranes
Director

Per Erik Stokstad
Deputy Director