ICB Implementation Team HM Treasury 1 Horse Guards Road London SW1A 2HQ. By

Transcription

1 ICB Implementation Team HM Treasury 1 Horse Guards Road London SW1A 2HQ By banking.commission@hmtreasury.gsi.gov.uk 12 March 2012 Government response to the Independent Commission on Banking The British Bankers Association (BBA) commented extensively in response to the two consultations undertaken by the Independent Commission on Banking (ICB) and has additionally made its views known to Government and to Parliament as appropriate. We have therefore read the Government s response to the ICB report with interest and welcome the opportunity to input further thoughts. We have previously expressed in principle support for the broad financial stability objectives set out by the ICB and would reiterate this for the avoidance of doubt. The ICB saw as the aims for its recommendations: A reduction in the probability and impact of systemic financial crisis; The maintenance of the efficient flow of credit to the real economy and the ability of households and businesses to manage their risks and financial needs over time; and The preservation of the functioning of the payments system and guaranteed capital certainty and liquidity for small savers including small and medium-sized enterprises (SMEs). The ICB recognised the range of banking reform measures being taken forward internationally, but nevertheless did not believe that the financial stability benefits relating to resolution, insulation of everyday banking services and curtailment of the implicit government guarantee could effectively be achieved without some measure of structural separation. It therefore proceeded with its recommendation for retail ring-fencing and outlined this based on a model in which services are mandated, permitted or prohibited for the ring-fenced bank; a model supported by provisions on legal, operational and economic links. While we see this as a model which can be made to work, the Government response itself made clear that there remain many detailed and complex aspects which require further consideration and analysis. In giving these issues further thought, it has crossed our minds that a good starting point would be to have a clearer image of what it is that we are trying to achieve from the introduction of the ring-fence model and how it fits within the policy objectives for recovery and resolution plans. If, as we believe, the objective is to enhance the prospect of an orderly resolution for both the ringfenced and non-ring-fenced part of a banking group, while at the same time placing emphasis on a continuity of service for those unlikely to be multi-banked, then this potentially guides the way in which we should think about the nature of mandated services. It also points to the need to appreciate that ancillary services are those which may be relevant to an appropriate service

2 2 provision from within the ring-fence but also may be relevant to the non-ring-fence and in both instances may be critical to a bank being able to provide an appropriate mix of services to customers within a normal banking relationship. The more that we give thought to this, the more we consider that a good way of assessing whether allocation within the model works is to think about the effect on a bank s customers whether household, SME, corporate or some other form of entity. This leads us to conclude that, should the dividing lines be inappropriately drawn, then the result may be one in which banks are unable to deliver services on a proper relationship basis, particularly if agency arrangements do not allow appropriate use of collateral and security or for relatively straightforward risk management to be provided from within a single banking relationship. We can therefore see considerable merit in the establishment of a principle which places emphasis on the maintenance of a single banking relationship. It is also the case that the banks are in the middle of a change programme variously instigated through business need and considerable reforms to the capital and liquidity regime, the regulatory framework and overlapping initiatives on recovery and resolution. Since these initiatives emanate from a variety of international, European and UK sources, both statutory and regulatory, there is very clearly a need to ensure that there is a strategic and economic coherence between the different initiatives. It is vital therefore that the Treasury and UK regulatory authorities coordinate in taking forward these differing initiatives to ensure that they remain broadly complementary. The Government response to the ICB, published by HM Treasury and BIS in December 2011, identified 20 or so areas requiring further analysis or consideration. In the first of the enclosed schedules we have outlined each of these in the left hand column and in the right hand column have set out our view on each, and in the process have drawn attention to particular circumstances and solutions which it may wish to consider. Examples of the types of issue raised include: The need to give further thought to the provision of services to SMEs and the cost that would arise if a bank were not permitted to provide foreign exchange and hedging services on a basis that takes account of cash deposits and other security and collateral; also the potential impact upon individuals of restrictions on structured retail deposits and fixed rate mortgages. Similar considerations for corporates, where the case for broadening the ring-fence to include commercial lending may be prejudiced by the cost of providing other services from outside the ring-fence, the inability to utilise security and collateral, and potential barriers in the way of global cash management services. The need to ensure that private banking, including investment management and custodian services, can be located entirely outside the ring-fence in order to keep services to high net worth clients within a single banking relationship, a good example of where it is helpful to think in terms of the primary nature of the service involved. The apparent inconsistency between the allocation of trade finance as a permitted service and the non-eea geographic prohibition, other practical considerations involving the geographical divide and the need to exclude Crown Dependencies from the prohibition. The need to ensure that treasury functions can undertake essential risk management, liquidity management and funding as envisaged in principle by the ICB, and whether this necessitates a review of the proposed limitations on economic links with the non-ring-fenced part of the group,

3 3 including the more restrictive large exposure limits proposed under the provisions on economic links and possible difficulties in respect of the prospective Net Stable Funding Ratio (NSFR). The need for an early understanding of the basis upon which non-eea branches are to be viewed as outside the scope of ring-fencing, the related issue of whether the Government is persuaded of the merit of a size exemption and concerns arising from governance considerations in respect of ring-fenced banks. Complex tax and pensions considerations that will need to be resolved as part of the introduction of the ring-fence structure, which may require primary legislation in their own right, and other transitional issues. We are not suggesting that any of these difficulties are insurmountable, but we are saying that it would be beneficial for the White Paper to give as clear an indication as possible of how they will be addressed in order to allow individual institutions to begin to plan the restructuring of their business. The second schedule sets out issues concerning the ICB s recommendations in respect of increasing banks loss-absorbing capacity based on the six areas outlined in the Treasury/BIS paper: The proposed additional equity buffer for large (and medium-sized) ring-fence banks and the prospect of this tying up with amendments to CRD IV. Confirmation that our members believe that recalibrating proposals for a leverage ratio would be unduly constraining in key areas and result in disincentives for moving towards the Advanced Internal Ratings Approach. Confirmation of our support in principle for the introduction of a bail-in regime as a complement to existing resolution tools based on appropriate long-term unsecured debt and on a basis aligned to any European or internationally agreed regime. The reasons why we believe that depositor preference may work against financial stability and the objectives of both ring-fencing and the bail-in regime. Our belief that there is considerable downside to specifying an explicit minimum level of primary loss-absorbing capital (PLAC) and the case for an exception for UK-headquartered G-SIBs having to apply these requirements across their global operations. Support for the notion of linking the assessment of the adequacy of an institution s recovery and resolution plan to any resolution buffer providing this is accompanied by a suitable level of objectivity about what would constitute grounds for concern. While the additional loss-absorbency proposals form a central part of the ICB s recommendations, they overlap considerably with initiatives underway in Europe and internationally and the confidence which can be placed in effective resolution will be enhanced if key aspects of additionally planned UK arrangements, such as the bail-in regime and other aspects of the crisis management arrangements, fully align to the broader regime expected to be put in place. The proposals also raise fundamental issues of scope, proportionality and international competitiveness. In noting the intention that the primary and secondary legislation related to the ring-fence be completed by the end of this Parliament in May 2015 and that banks will be expected to be compliant as soon as practically possible thereafter, we would make the point that PRA regulation

4 4 and authorisation will also be required before banks can begin their detailed implementation programmes. The Government, in keeping with the ICB, rightly acknowledges the appropriateness of the 2019 timeline for Basel III capital reform and other additional requirements. The BBA and its members stand ready to offer assistance to HM Treasury and BIS in the preparation of the White Paper and the subsequent process of giving statutory and regulatory expression to the ICB s recommendations. For further information on this submission please contact Paul Chisnall, Executive Director, BBA at paul.chisnall@bba.org.uk British Bankers Association March 2012

5 Schedule 1: Ring fencing Non-EEA branches Paragraph 2.15 explains that the UK branches of banking groups from outside the UK will generally be unaffected by ring-fencing provisions, subject to expectation that the home state prudential supervisor of branches of non EEA banking groups will give careful consideration as to whether it is appropriate to permit significant amounts of mandated services to be undertaken in a branch rather than through a UK subsidiary. Commentary The initial starting point for the UK prudential supervisor should be to consider whether it considers that it can place sufficient trust in the home supervisor s close engagement with the counterparty, whether bilaterally, in crisis management groups or supervisory colleges. If it can, then the question of whether the subsidiarisation of non EEA branches is necessary should become redundant. An early indication of the way in which the Government s thinking is developing will be important to foreign banks needing to know whether they are likely to be in or out of scope since an awareness of this may be a determining factor in whether they consider all or part of their UK operations to be viable going forward. Mandated services Paragraph 2.19 explains that the key questions are: - Whether there is a case for further services in particular the provision of credit to households and/or SMEs to be mandated within the ring-fence; - How SMEs should be defined for these purposes; and - How private banking services should be treated. Paragraph 2.20 asks whether requiring retail and SME deposits to be within the ring-fence but permitting matching assets i.e. lending including mortgages to individuals and SMEs to be placed outside the ring-fence could result in a ring- Our assessment is that the parameters of the ring-fence are broadly set in the right place since the core business of a ring-fenced bank will be the taking of deposits and the provision of overdrafts to individuals and SMEs. Ancillary to this will be the taking of deposits from and the provision of payments services to any customer within the EEA, including the provision of current accounts, savings accounts and those investment products which do not give rise to the ring-fenced bank being required to hold regulatory capital against market risk. Lending to consumers and businesses on a secured and unsecured basis, including mortgages and credit cards, trade and project finance, and the provision of advice from, and the sale of products on behalf of, non-ring-fenced banks absent an exposure for the ring-fenced bank will also be permitted.

6 Ring fencing 6 fenced bank with a mixture of assets and liabilities (including loans to large European corporates) that makes the bank or parts of it less attractive to buyers in a resolution. Paragraph 2.22 adds that the Government does not believe that the case has yet been made for mandating further services. At a conceptual level, we would say that this should broadly enable the ring-fence bank to provide the right mix of services to the customer base for which the Government is seeking to ensure a continuity of service in the event of the bank requiring resolution. The additional measures under consideration, including funding and leverage ratios, should each work to maintain a balanced book and there should be no reason why the permissive element of the ring-fence should result in additional complexity. We would agree that there is not a case for mandating further services. We would add that thinking about some of the more intricate decisions about the make-up of the mandated/permitted/prohibited model may flow more naturally if the Government could set out in clearer terms the overarching policy objectives for ring-fencing. If, for instance, the intention is to ensure continuity of service in the event of need for those who are unlikely to be multi-banked, but at the same time enable the on-going provision of banking services on a relationship basis and improve resolvability irrespective of ringfenced status, then this probably sets the scene for a more in principle approach to what needs to be mandated, what can be ancillary on the grounds that it is an activity that may be relevant or integral to a ring-fenced or non-ring-fenced bank - and what need to be prohibited. Turning specifically to SMEs, it needs to be appreciated that given the UK Companies Act definition of an SME takes us up to a medium-sized company with an annual turnover of 25.9m, we are talking about entities that may have complex banking needs. We would therefore see merit in further consideration being given to permitting foreign exchange and hedging transactions to be undertaken from within the ring-fence or otherwise allow monies

7 Ring fencing 7 on deposit to be used as collateral for services provided to customers on an agency basis. Without this then there is a very real cost to SMEs in particular of having their banking needs met from two separate entities. Since this is unlikely to be immaterial we would expect that this is something that HM Treasury would plan to build into its CBA. Thought also needs to be given to banks taking an equity stake in SME and other companies to which they lend. Enabling foreign exchange and hedging services in this way may also improve the feasibility of providing banking services to corporates as a permitted service. With restrictions as currently drawn, we see barriers in the way of ring-fenced banks being able to accommodate corporates on a competitive basis since although services may be bought in or contracted from another part of the group on an agency basis the separate provision of relatively straightforward services potentially has a determining effect. A better way of approaching this would be to ensure that corporate needs can be met irrespective of whether a bank is ring-fenced or non-ring-fenced. Without this, there is a risk that too severe a constraint is placed on banks ability to utilise deposited funds for corporate investment. Giving further thought to the rationale behind the mandated, ancillary and prohibited categories should enhance the ability of banks to meet their customer needs on a relationship basis without relying upon agency arrangements which may not be economically viable. This may also be the means by which we can reduce the prospect of a cliff-edge effect in which SME customer relationships have to be shifted from the ring-fence to the non-ring-fenced bank and back as a result of thresholds being crossed temporarily.

8 Ring fencing 8 There is in any case a need to think about SME thresholds in a way as to ensure that an overly legalistic approach does not result in smaller subsidiaries being split off from a banking relationship with a larger corporate group that otherwise chooses to bank with the non-ring-fenced bank. On private banking, we would make two recommendations: First, deposits held as part of an investment management or custody arrangement (as opposed to banking services) do not constitute a customer s principal bank account through which they conduct their day-to-day transactions and are therefore not a vital banking service. We therefore see a case for ensuring that these deposits are permitted to be held outside the ring-fence with the wealth management service to which they play a supportive role; without this an incidental part of wealth management would otherwise draw the activity back into scope; Second, it needs to be appreciated that wealth management services may be provided to individuals who would not of themselves meet a High Net Worth definition but instead need to be viewed as part of a family relationship. It may also be relevant to note that the way in which some HNWIs are holding and investing monies is changing and in some cases individuals are placing an increasing emphasis on money purchase pension schemes and SIPPs. Ensuring that services arising from private banking do not fall within mandated services would also fit with the nature of the financial services in which customers engage and also overcome any added complexities arising from the international nature of the client base.

9 Ring fencing 9 Such customers are articulate in their understanding of the basis upon which their wealth management services are provided and are capable of understanding that their banking services are provided outside the scope of ring-fencing. It also needs to be noted that a large part of the customer base is international in nature and more than capable of arranging for their wealth management to be undertaken in a competing financial centre. In trying to slot services into the appropriate category mandated, permitted or prohibited key factors will be keeping the banking relationship as simple as possible since this provides the most obvious means by which a customer s deposited funds can appropriately be used in order to provide extended services or better pricing based on collateral, security or netting arrangements. There is also a need to ensure that activities that would ordinarily be outside the scope of mandated services are not unwittingly drawn in as a result of unintended circumstances. Securities transactions for instance can sometimes fail and in the process generate a loan or overdraft ; these and other temporary, inadvertent facilities should be afforded a treatment based on the primary activity. Prohibited services Paragraph 2.25 sets out the rationale behind prohibiting certain services and paragraph 2.26 explains that the ICB took this as meaning that prohibited services should include any services to financial institutions or to non-eea customers, services that require firms to hold capital against market risk and derivatives Financial institutions: there is a need for a closer analysis of the exposures to financial institutions that constitute the type of risk from which ring-fencing is intended to insulate the ring-fenced bank. Deposit-taking would not appear to generate such a risk and global payments activity (with or without a financial institution as counterparty) should be permitted from within the ring-fenced bank, including transactions with a non-eea counterparty and

10 Ring fencing 10 or other contracts which result in a requirement to hold capital against counterparty credit risk. Paragraph 2.27 explains that the Government agrees that certain services should be prohibited in order to meet the objectives of the ring-fence, particularly services which impede resolution and/or increase the ring-fenced bank s exposure to volatility in global financial markets. Paragraph 2.29 nevertheless identifies as requiring further analysis: - The characteristics of financial institutions that may be permitted counterparties of ring-fenced banks; - The characteristics of products prohibited by reference to their function eg derivatives or trading book assets; and - The approach to cross border activities and clients, and the treatment of UK Overseas Territories and Crown Dependencies. Paragraph 2.37 adds that prohibited services, such as clearing and settlement services of securities or market making, can also be of systemic importance and that ring-fencing can also enhance their resolvability by making them more separable and reducing their links to vital retail services. even between two non-eea counterparties. There are also categories of financial institutions and activities that should not be viewed specifically as being financial institutionorientated: consumer finance providers, insurance brokers, providers of mortgage indemnity insurance, investment trust providers, life assurance intermediates; also normal commercial lending to finance a financial institution s property and infrastructure; also activity such as the provision of third party mortgage administration services. We would like to see further analysis undertaken and consideration given to ring-fenced banks being allowed wider latitude to transact with other financial institutions where exposures are collateralised and risk effectively minimised. Products: there is a need to ensure that ring-fenced banks providing certain deposit-taking services (e.g. fixed rate and structured deposits) are able to match or hedge their risks appropriately in relation to providing an economic return on these products. To carry out these risk management activities effectively ring-fenced banks will be required to engage in certain derivative transactions with non-ring-fenced banks (whether within the same group or with an external non-ring-fenced bank).we consider this ability to be fundamental to the ring-fenced bank s ability to manage the risks associated with these deposits and, as such, believe these transactions should be permitted within ancillary services. Failure to permit ring-fenced banks to enter into these risk management arrangements with non-ringfenced banks would have a material effect on a ring-fenced bank s future ability to provide retail customers with a broad range of readily accessible and popular savings products. Permitting ring-fenced banks to manage risks on deposit products is no different in concept to their need to hedge risks on lending products. We see a case for the hedging of risk positions with non-ring-fenced institutions being permitted where dynamically managed

11 Ring fencing 11 subject to appropriate limits. Cross border activities and clients: while we have little difficulty with the broad objectives for the geographical prohibition, illustrated by the ICB as providing mortgages to American homeowners or a loan to an Australian energy company with no base or subsidiary in the EEA, there are issues which need addressing appropriately if significant difficulties are to be avoided: a. the basis on which the EEA status of the customer is to be determined, i.e. whether based on the domicile of the transacting customer or its parent rather or alternatively the location of the booking office. When this issue has been examined on previous occasions e.g. in respect of segmental reporting requirements, the conclusion drawn has been that location of booking office of the bank constitutes the only practical solution available. The ICB s final report suggests that this latter interpretation is what the Commissioners had in mind. b. The potential closure by some banks of deposits if the prohibition of non-eea customers extended to the holding of deposits located in the UK where client arrangements had been made on an outreach basis; c. in respect of trade finance where the contracting party is not the EEA importer or exporter but the non-eea bank of the non-eea exporter or importer (including any guarantees and collateral arrangements); and

12 Ring fencing 12 d. in respect of cash management services, if a ring-fenced bank is unable to provide services for a global corporate and its subsidiaries from entirely within or outside the ring-fence, or for that matter to enable UK SME and corporate clients to manage their non-eea cash needs from a single source. UK Overseas Territories and Crown Dependencies: bearing in mind steps already being undertaken to deal with complexities created by a bank having a physical presence or relationships with counterparties in UK overseas territories and crown dependencies, including RRPs, there may be a case for permitting these as EEA equivalent status for ring-fencing purposes. Deposits raised in the Crown Dependencies in particular are a major source of liquidity for the UK banking market and we would view the inclusion of Crown Dependencies within the permitted geographical scope for the ring-fenced bank as the right approach since there will also be a strong overlap with the provision of private banking and wealth management services to HNWI. This could play a significant part in enabling banks to match more closely the funding requirements and funding capacities of the ring-fenced and non-ring-fenced parts of a banking group. We further believe that further thought needs to be given to the geographical prohibition on non-eea activity since we continue to find it difficult to understand why the acceptance of non-eea deposits by a ring-fenced bank would add risk or complication. As explained in the attachment on the loss-absorbing schedule, we would also be concerned if the geographic criteria meant that ring-fenced banks could not access non-eea capital markets. This is a further area in which it may be helpful to think further about the rationale behind the

13 Ring fencing 13 limitation. If the issue is one of resolvability, then the answer may be to work toward a system of mutual recognition in which the right of the UK authorities to act appropriately under a resolution is established as a matter of course. Ancillary services Paragraph 2.42 confirms that the Government agrees that ringfenced banks should be allowed to undertake ancillary activities for the purposes of risk management, liquidity management and funding non-prohibited services and explains that it will consult further on limiting a ring-fenced bank s wholesale funding beyond existing regulation; further analysis, in conjunction with the FSA, will be undertaken on whether this is better achieved through the imposition of a backstop limit as suggested by the ICB or through the implementation and possible adaptation of existing liquidity requirements. Paragraph 2.43 adds that a number of workable models have been proposed for the rules governing ancillary activities, including building societies, insurance and utilities regulation, explaining that further analysis is needed to determine a model that is most effective for the particular characteristics of ringfenced banks (and that there may be a case for modernising There will be a need in particular to map out the broad parameters of the activities which need to be permitted in order to ensure that the treasury function can fulfil its risk management, liquidity management and funding duties. As the ICB recognised, this will involve the treasury function engaging with non-ring-fenced banks and non-ring-fenced financial institutions in a way which otherwise be prohibited for the ring-fenced bank. Where to pitch the limitations on wholesale funding involves a distinct set of issues. It also requires consideration of whether any limitations should distinguish between different types of wholesale funding. The UK bank levy, for instance, makes a distinction based on maturity, recognising that longer-maturity wholesale funding (which it defines as greater than one year remaining to maturity) is inherently more stable than wholesale monies with shorter residual maturities. (This also ties into issues concerning the crown dependencies and overseas territories since they provide a source of funding that has a maturity profile akin to UK core deposits.) While we can appreciate the need for transparency in respect of intragroup funding, we think the limits should be reviewed and certainly do not see a case for restrictions being more tightly drawn than those applying to third party finance. While we understand the motivation behind ensuring that ring-fenced banks do not become

14 Ring fencing 14 the Building Societies Act). unduly reliant upon wholesale funding, it does not strike us as right that a parent would not be able to inject liquidity into a ring-fenced-bank encountering temporary difficulty. We therefore see a case for appropriate provision for this being built into the regime. The draft US Volcker rule creates an anomaly which many commentators view as wrong and we would highlight the need for the UK not to adopt a similar approach. This relates to market-making and the fact that trades in US Treasury Bills are permitted whereas trade in other Government debt may be prohibited. This is wrong on many levels and should not be emulated by the UK. There is also a broader need for some thinking to be done about how banks with a presence in both the UK and US will align Vickers and Volcker requirements. It also needs to be borne in mind that the Building Societies Act was designed for financial entities with a narrower scope than that envisaged for ring-fenced banks and was enacted before the Basel III ICAP and FSA ILAA regimes were put in place. Section 9a in particular is unduly restrictive and was not designed with larger, more complex institutions in mind. The right approach must be for the ring-fence rules to be drawn up on a basis consistent with the Basel III and FSA ILAA frameworks for risk management and liquidity management ensuring that the ring-fenced bank is capable of pursuing appropriate hedging strategies, funding through securitisation and covered bonds and managing its net position through the interbank market. Legal and operational links Paragraph 2.45 explains that the Government strongly supports the ICB s conclusion that a high degree of legal, While we understand the reason why the Government may wish to place restrictions on ring-fenced-banks owning or participating in other financial institutions there is a need to

15 Ring fencing 15 operational and economic separation is needed between the ring-fenced bank and the rest of a banking group for the ringfence to be effective and credible. It sees economic links in particular as needing to be carefully policed but considers that ring-fencing should allow some residual connection. On legal and operational links paragraph 2.47 explains that the Government will undertake further work to assess the operational separability necessary to give force to the ringfence resolvability and the appropriate restrictions on ownership of other financial entities. Paragraph 2.48 confirms that the Government agrees that a ring-fenced bank should be a separate legal entity; also that the Government agrees that equity participation in other financial institutions can create risks for the ring-fenced bank, impede supervision and hinder resolvability and separability. But it indicates a need for further analysis of the practical implications of placing limits on ownership or participation in other financial institutions undertaking prohibited activities. Paragraph 2.50 adds that the Government agrees that ringfenced banks need to be able to demonstrate operational separability and sets out four key principles for the FSA to act provide for circumstances which may be the exception to the rule. Instances which come to mind include bank interests in the Business Growth Fund and Big Society Capital, the ownership of the Reclaim Fund Ltd., interests in other, non-financial activities e.g. market information services, and possibly the National Loan Guarantee Scheme. These are examples only and it may be helpful to think through the way these may be impacted in order to ensure that the Government comes to the right answer. The survivability of operational subsidiaries in the event of recovery or resolution is already a focus for attention in the ongoing work on the design of recovery and resolution plans. This should very much guide the way in which this is looked at from the perspective of putting in place suitable arrangements for ring-fencing. The overriding principle should be what is needed to ensure resolvability and care should be taken to ensure that any ringfence statutory provision or regulation is compatible with Recovery and Resolution Plans being put in place under the current FSA initiative. On payment systems, while we agree with the proposal, we take the opportunity to note that it will most likely necessitate some form of national transition project. For the avoidance of doubt, we are working on the assumption that the Government is not suggesting that only ring-fenced banks should be members of payment systems; should this not be the case then we would wish to challenge any such exclusion. Payments systems need to be seen as supporting both ring-fenced and non-ring-fenced banks. We would also add that subject to suitable resolution planning we would wish to see group membership of payments systems continue and that there is also a relevance for Crown Dependencies here.

16 Ring fencing 16 upon in progressing its work on recovery and resolution plans: - independent capitalisation and funding for any operational subsidiaries; - effective service level agreements to support the continued provision of essential services in the event of failure of any part of the banking group; - the provision of services on an arms length basis so that such services are economically viable should part of a banking group fail; and - operational assets used by critical economic functions being, where possible, owned by the operational entity providing those services. Paragraph 2.53 adds that the ring-fenced bank should be a direct member of all payment systems it uses or have access via another ring-fenced bank as agent. Economic links The paper outlines the ICB s recommendations on the limitation of economic links between the ring-fenced bank and the rest of a banking group, with: The ICB recommended that all transactions between the ring-fenced bank and non-ringfenced part of the group be on a third party basis and subject to the same large exposure limits with: - Paragraph 2.54 recounting the ICB s principle that where a Aggregate exposures to a single counterparty limited to 25% of capital resources;

17 Ring fencing 17 ring-fenced bank is part of a wider corporate group its relationships with entities in that group should be conducted on a third party basis and that it should not be dependent for its solvency or liquidity on the continued financial health of the group, which should be ensured through both regulations and sufficiently independent governance. - Paragraph 2.55 recounting the specific recommendations in supporting the principle: all transactions to be treated for regulatory purposes no more favourably than those between third parties; ring-fenced banks being subject to regulatory requirements including capital, large exposures, liquidity and funding on a solo basis; assets being sold to and from the ring-fence at market value; dividend payments being conditional upon the board satisfying itself that the ring-fenced bank has sufficient financial resources; and the board of the ring-fenced bank having a majority of independent non-executives of whom one is the chair and no more than one sits on the board of the parent or another part of the group and which has a duty to maintain the integrity of the ring fence at all times. Secured exposures i.e. the gross value of the collateral between the ring-fenced bank and the rest of the group limited to no more than twice the level of third party unsecured limits and comprised of assets only of the highest quality. The secured lending limits proposed by the ICB would appear to be based on US regulation of intra-group transactions and would impose lower exposure limits and more demanding collateral restrictions between ring-fenced and non-ring-fenced parts of the group than between a group entity and an outside third party. A requirement for assets only of the highest quality of itself would exclude this from being a meaningful funding channel. This may therefore work to create a bias towards third party funding which may add in a layer of complexity and risk. Liquidity considerations are clearly an important part of the formula in creating the right set of rules to enable ring-fencing to be put in place in such a way as to enable the design of a successful banking model. As a minimum, exposures between the ring-fenced and nonring-fenced part of a banking group should be on terms no more restrictive than those with outside third parties. A further practical consideration is whether the introduction of the retail ring-fence would impact upon the ability of banking groups to meet future requirements for a Net Stable Funding Ratio (NSFR) which requires banks to match more of their long term assets with long term funding.

18 Ring fencing 18 - Paragraph 2.56 recounting a number of details governing arm s length transactions including: exposures from the ring-fenced bank to the rest of the group and vice versa being subject to the same large exposure limits as for third parties and secured exposures between the ring-fenced bank and the rest of the group being no more than twice those unsecured third party limits; no forms of unlimited guarantee or joint liability arising out side the large exposure limits; and the ring-fenced bank not being party to agreements containing cross-default clauses or similar arrangements - Paragraph 2.56 additionally explaining that the Government believes that the ring-fenced bank should not be dependent on the financial health of the rest of its corporate group for its solvency or liquidity and that it will consult further on the regulatory and governance requirements necessary to implement this. - Paragraph 2.57 acknowledges a need to undertake further work on the extent to which existing regulation between third parties needs to be supplemented with additional rules to ensure the economic independence of the ring-fenced bank (with paragraph 2.58 explaining that certain types of

19 Ring fencing 19 contractual arrangement such as cross-default clauses and intragroup guarantees being likely to require additional rules and limits to ensure that intragroup asset transfers and secured lending are conducted on an arm s length basis); Limits on intragroup exposures (with paragraph 2.59 explaining that the Government agrees that the ring-fenced bank should meet at a minimum capital and liquidity requirements on a solo basis, that the large exposures regime should apply intragroup and that specific restrictions on funding from the rest of the group may be appropriate to the extent that the rules on wholesale funding are not sufficient; paragraph 2.60 explaining that the relationship between the ring-fenced bank and the rest of the corporate group may be asymmetrical; and paragraphs 2.61 and 2.62 signalling further analysis: to ensure that limits on intragroup exposures are optimal; and on the case for a secured funding limit including how this should be calibrated, the definition of appropriate collateral and appropriate haircut. De minimis exemptions Paragraph 2.68 explains that the Government believes that the case for de minimis exemptions from ring-fencing, in particular for very small firms, should be reviewed. Paragraph 2.69 suggests that exemption could be based on: While on the one hand the Government is keen to avoid the risk that depositors and other customers are not confused, given the relatively high level of deposit protection now in place - 100,000 and the separate ability for the regulator to take a close interest in RRPs it may be that a relatively high exemption threshold should be set, whether in terms

20 Ring fencing 20 - Bank size, with paragraph 2.70 adding that exemption based on institutions being viewed as small enough to be allowed to fail without government support, may be dependent upon any exempt institution having a high degree of resolvability and/or meeting higher prudential standards; - The amount of either mandated or prohibited services, with paragraph 2.71 adding that it can be argued that the provision of a small amount of prohibited services by a ringfenced bank may have little bearing on resolvability or financial contagion to vital banking services; the Government however is concerned that this may undermine the objectives of the ring-fence and does not currently believe there to be a strong case for such an exemption; or - Individual transaction size and, though paragraph 2.72 does not see a strong case for such an exemption, the Government will undertake further analysis. of the ring-fence applying only to G-SIBs or by reference to another existing threshold. There would appear to be some logic, for instance, in adopting the 20bn of relevant aggregate liabilities allowance adopted for the UK bank levy given that this was considered by the Treasury at the time of the introduction of the bank levy to balance the probability that the failure of a bank could pose a systemic risk against the relative burden imposed. It may be, however, that concern over the prospect of systemic risk arising from institutions described in the second Turner paper as too similar to fail may dictate that any numerical exemption be set at a considerably lower level. Thresholds aimed at achieving the right cut based entirely on pragmatic grounds may be 7 billion (the ceiling for being able to utilise the small and local distribution facility within the Dormant Bank and Building Societies Accounts Act 2008 (also drawn up by the Treasury) or a 2bn- 3bn threshold which would exclude a smaller but significant number of institutions. There may also be a case for exemption based on customer numbers or a de minimis amount of mandated EEA activity for a principally non-eea bank (overlapping with the NNWI treatment and whether or not the bank is open to general retail deposit-taking). The other way of looking at the need for de minimis exemption is to consider whether a single limit can provide a simple and effective means for scoping out organisations that would otherwise be inappropriately caught, including private banks, investment managers, niche banks with only a small number of non-eea customers and overseas banks with only a low volume of UK deposit-taking or other mandated services. Issues concerning small amounts of mandated services can most likely be dealt with in this

21 Ring fencing 21 way; issues concerning small amounts of prohibited services through other means whether in terms of a reappraisal of the geographical limitations or an appropriate definition of permitted activity within ancillary services. Governance Paragraph 2.64 explains that governance will be central to ensuring the economic independence of the ring-fenced bank but adds that, where the ring-fenced bank contains most of a banking group s business, the delivery of the group s strategy on a day-to-day basis may be complicated by separate board arrangements adding that the ICB recognised this tension and that flexibility may be appropriate in such circumstances. The paper identifies a need to consult on the best approach to achieving the appropriate balance, including through the composition, duties and objectives of the ringfenced bank s board. Paragraph 2.65 adds that disclosure provides an important means of maintaining a clear and credible separation between ring-fenced banks and the rest of their group and that the Government will undertake further analysis and consult on whether and how disclosure requirements should vary depending on the nature of the ring-fenced institution. The expectation that the ring-fenced bank will have a separate, independent board to the main group is of significant concern given the duplication that this could involve and the practical issue of securing and receiving approval of sufficient people with the necessary levels of knowledge and experience if independence were defined too rigidly. We would therefore concur with there being a need to review this in light of the differing business models of the major UK banks in particular. A fundamental question that does not yet appear to have been addressed is the question of to whom would the board of a whollyowned ring-fenced bank subsidiary of a global bank report? Would this be to the group board and if so is the expectation that there be specific accountabilities attached to this in respect of the group s shareholders? This ties into a broader consideration of the duties of directors, including NEDs, and we see a case for careful thought to be given to issues arising from director duties and liabilities both generally and specifically in respect of a ring-fenced-bank. There is a need in particular to ensure that we do not unduly create duplication and conflicts of interests for the directors of a ring-fenced bank and the parent. There will need to be clarity over where decision making responsibility lies; commercial decisions relating to the operation of the Group (including the ring-fence) should be taken by the parent Group board, rather than board of the ring-fenced bank. Responsibility specifically placed on the directors of the board of the

22 Ring fencing 22 ring-fence bank should be limited to ensuring the separate operability and resolvability of the ring-fence bank. While the intention is for ring-fenced banks to have their own culture and identity it is also important they fit within the strategic direction provided by the Group board. Disclosure is an understandable consideration within a group context; it would be a matter of concern however if the Government were to conclude that private banks or mutuals should meet stock exchange listing requirements which they were otherwise not subject to. They already complete substantial prudential regulatory returns and meet other reporting requirements according to their circumstances. So requiring them to meet additional exchange based listing compliance requirements in this area would seemingly conflict with the Government objectives for financial services regulation to be proportionate and open to competition. Tax and pensions Paragraph 2.66 explains that there are other circumstances in which the ring-fenced bank may be liable for obligations of the broader group and cites VAT and deficits of group pension funds as two particularly significant examples. Such liabilities could endanger the economic independence of the ring-fenced bank were e.g. a wholesale bank to collapse and leave the ring-fenced bank with a large liability for the group. The ICB recommended removing these involuntary obligations or otherwise mitigating their impact. A question for the Treasury may be the basis upon which they have put a cost on the VAT and defined benefit pension scheme (DB) consequences of the introduction of retail ringfencing as this may tell us something about the way in which they see these specific complications being resolved. In respect of VAT, the most immediate point in issue is whether, under EU tax and state aid laws, the Treasury can remove the existing joint and several liability provisions under VAT grouping. In the event that they cannot, or chose not to, it may then become necessary to consider available options to mitigate the potential risk of the ring-fenced bank from becoming liable for debts of other VAT group members. There are other tax issues to

23 Ring fencing 23 Paragraph 2.67 explains that removing joint and several liability may lead to new costs for banks or result in existing liabilities being brought forward. E.g. where banks pension funds are in deficit, pension trustees may require this to be addressed more quickly if recourse to the ring-fenced bank is reduced. It adds that a) the relevant costs should be seen as an inevitable consequence of the curtailment b) that different ways of addressing such liabilities will have different costs and c) that the Government will seek to mitigate their impact to the extent that such obligations can be met without undermining the objectives of the ring-fence. consider, including tax bases, allowances and ongoing reliefs, including the treatment of deferred tax assets (and consequences, if any of activity moving from a bank to a non-bank since non-ring-fenced entities may not be deposit-taking entities). In respect of pensions, issues arise as a result of: Depositor preference: pension schemes are unsecured creditors and so depositor preference would result in subordination to protected deposits in a way which does not apply under current FSCS arrangements. Funding / volatility: a pension scheme's liabilities are inherently volatile, impacted by factors such as changes in mortality, gilt yields and inflation. Where the entity inside the ring-fence is legally responsible for DB pension scheme funding, the tension between that volatility and the objectives of the ring-fence needs to be addressed. Funding / regulatory regime: the Pensions Regulator requires DB pension schemes to be funded on a prudent basis, with deficits removed as quickly as the sponsor can reasonably afford. There is an inherent tension between those expectations and the proposed enhanced capital requirements where banks subject to those requirements may have less scope for providing cash and/or security to fund the pension scheme. Contingent assets: many DB pension scheme sponsors have in recent years used noncash assets to provide an income stream and/or security to pension schemes. Scheme trustees will need to understand matters such as: where such contingent assets will sit following the imposition of a ring-fence; whether they might be compromised when needed by the exercise of powers under the Banking Act 2009; how such assets are

24 Ring fencing 24 treated in a bank's resolution and recovery plans; and whether the imposition of a ringfence impacts a bank's ability to use assets in this way. Employer covenant: the strength of a bank's covenant (being its ability to support its DB pension scheme financially) is a critical driver to negotiations between pension scheme trustees and scheme sponsors on long-term plans to clear scheme deficits. The imposition of a ring-fence will impact on that covenant with potentially different effects depending on whether the sponsor is inside the ring-fence (where covenants might be said to improve due to higher capital requirements) or outside (where there is likely greater risk exposure). Pension Protection Fund: changes to insolvency ratings and any decrease in the provision of contingent assets may have an impact on schemes' PPF levies (which are usually funded directly by the scheme sponsor i.e. the bank). Restructuring: since many DB pension schemes are run on a group-wide basis, the separation of retail and investment banking operations may lead to schemes being 'segregated' (separated into financially insulated sections) or de-merged; without specific provision the separation would also crystallise any deficit. Any restructuring requirement should recognise the practical difficulties of separation and include a materiality threshold. Implementation Table 6A identifies that the structural reform which ring-fencing involves will need to be implemented through a combination of In addition to relevant issues identified above, including tax and pensions considerations, in mapping out the required actions the Government will need to give thought to the following

25 Ring fencing 25 primary legislation, secondary legislation and regulatory rules. transitional considerations: Moving customer and/or supplier contracts: where a Part VII approach is unlikely to be practicable (not least as Part VII transfers are meant to deal only with deposit taking businesses). What number of accounts are we talking about which will need new sort code and account numbers? What will be the impact on direct debits (in theory an automated process) and standing orders (a manual process)? External funding arrangements: which will be by rated legal entity and/or at a group level. How are these to be split or treated if the issuer changes and/or the issuer s business are split across the ring-fence? Will all existing programmes need to be rerated and/or become obsolete? Hedging activities: similar considerations apply to in-force hedges. Will ring-fencing give rise to defaults or other implications under ISDA agreements? Regulatory permissions: will existing permission be split appropriately and grandfathered (including with respect to Approved Pensions)? Will fresh approvals be required; will existing rule waivers continue to apply? Branding: the introduction of the ring-fence should be brand-neutral from a statutory perspective and not oblige customer disruption and/or unnecessary rebranding expenditure. Accounting considerations: there is a need to ensure that a transfer of assets solely for the purposes of ring-fence and non-ring fence allocation do not trigger an accounting event as this would result in a need to capture valuations unrelated to the underlying economic activity; transfers need to be able to take place without the crystallisation of gains and losses in the institution initiating the transfer and related regulatory and tax

26 Ring fencing 26 effects neutralised. British Bankers Association 12 th March 2012

27 Schedule 2: Loss absorbency Equity requirements Paragraph 3.9 explains that the Government supports the ICB recommendation that large ring-fenced banks should be required to have an additional equity ring-fence buffer and that calibrating this at 3% of RWAs is proportionate (and clarifies that this should be in addition to the baseline capital conservation buffer of 2.5% and any other countercyclical capital buffer but not in addition to any G-SIB surcharge, with only the higher of a ring-fenced bank s buffer and the G-SIB surcharge applying. Commentary The ability to implement the ring-fence buffer may be constrained by European legislation. The initial draft of CRD IV, however, permits national supervisors to use the Pillar 2 process to impose a wide range of measures, including additional capital, when a certain type of institution is or might be exposed to similar risks or pose similar risks to the financial system. This broadens the existing Pillar 2 process which has until now focused on individual institutions rather than groups of institutions. It might be possible for the UK to use this flexibility to impose the ICB surcharge. Paragraph 3.11 adds that the Government supports setting a smaller ring-fence buffer for smaller banks (with no buffer for the smallest). Recent amendments to the CRD IV text suggested by the Danish Presidency open the way for national supervisors to impose a systemic risk buffer of up to 3% CET1 within Pillar 1 or in excess of 3% subject to ex ante approval by the Commission. Leverage ratio Paragraph 3.19 explains that the Government agrees with the ICB that RWAs should remain the primary determinant of loss-absorbing capital and that this should be combined with a mandatory, minimum leverage ratio as a backstop for banks and also confirms that the Government supports the ICB s recommendation that all ring-fenced banks should meet this requirement on a solo basis. Paragraph 3.20 explains that the ICB s view was that any increase in the minimum equity-to-rwas ratio There is indeed a concern amongst banks and building societies that in certain areas the introduction of a leverage ratio could result in the ratio and not the capital requirement driving the constraint. This would act as a disincentive for banks and building societies to move to the Advanced Internal Ratings Based (AIRB) approach and therefore is not in keeping with broader regulatory objectives. The effect is particularly accentuated in the case of prime residential mortgages where we estimate that the leverage ratio could be the constraining