In the wake of the international financial crisis, the UK Government established an Independent Commission on Banking to examine options for the reform of the banking industry. The Commission published an Interim Report in April 2011 and its initial recommendations have been broadly endorsed by the Government. Implementation of the Commission's proposals will have consequences not only for UK banks but also for foreign banks conducting business in the UK, and for creditors, counterparties and other stakeholders.
Although the Commission proposes to issue a Final Report later in the year following a process of consultation, the approval of the key aspects of the Interim Report by the Government, combined with published responses from major US banks, means that it is now timely to analyse the proposals set out in the Interim Report and to examine their implications.
As the Interim Report points out, an ordinary business is subject to a variety of well-known and well- understood disciplines. The owners will prosper if the business does well, and will lose their capital if it fails. Lenders to the business will look for a reasonable prospect of repayment, and will thus limit their facilities to an appropriate percentage or multiple of its capital. The Government will not be on hand to rescue such a business, and ordinary insolvency procedures would take their course. Awareness of these consequences imposes constraints on shareholders, creditors and anyone else having a stake in the business. The combination of these factors acts as a brake to prevent undue or excessive risk-taking.
However, the financial crisis has demonstrated that these disciplines did not satisfactorily apply to banks. For example, (i) in the period leading up to the crisis, banks operated with exceptionally high debt-to-equity ratios, (ii) a substantial source of loan finance for banks is provided by its depositors, who are not usually in a position to understand its financial statements or to analyse the creditworthiness of the institution, (iii) a loss of confidence can create a run, since no bank holds liquid assets sufficient to meet its demand liabilities, (iv) the urgent need to prevent the collapse of major financial institutions meant that bailouts also rescued debt holders – who should have absorbed some of the risk – to the detriment of the taxpayer and (v) even before the crisis, banks had access to solvency assistance from the central bank in the form of its "lender of last resort" function. All of these considerations effectively meant that normal market disciplines did not properly apply to banks, and that the resultant losses were borne by the public purse, rather than by the stakeholders who were paid to assume those risks.
In many respects, the object of the recommendations is to restore some of the basic disciplines to the banking sector and thus minimise the risk of future failures in this sector. By the same token, the proposals are designed to facilitate any rescue that may prove necessary and to contain its public cost.
Financial Stability Reform
The Interim Report proposes various ways in which the structure of the UK banking system can be improved, making it easier and less costly to isolate and find solutions for the failure of a financial institution. The Commission explores various ways to reduce the probability and impact of bank failures by (i) increasing the loss-absorbing capacity of banks and (ii) imposing structural reform to create a degree of separation between retail banking (broadly the provision of deposit taking and payment and lending services to individuals) and wholesale/investment banking (providing large corporate customers and other financial institutions with a range of services).
The Commission stops short of proposing a strict separation of retail banking and wholesale/investment banking by requiring them to be in separate non-affiliated firms. It also dismisses the alternative approach of allowing banks to adopt any form of corporate structure and controlling risks through the use of very high capital requirements. Instead, the Commission opts for a "more moderate combination" of these approaches by way of:
- a moderate increase in capital requirements;
- measures to make a bank's borrowings more effective as a vehicle for loss absorption; and
- structural reform through internal ring-fencing within universal banks to insulate UK retail banking services from investment banking risk.
Each of these features requires separate examination.
Higher Capital Requirements
A key focus of the Interim Report was to explore ways to increase the exposure of the bank's owners and creditors to underlying risks of a bank's business. This could provide incentives for better monitoring and control which would lead to an overall reduction in the probability of bank failure. One such way would be to set a minimum equity requirement, although the Commission noted that this would involve a balancing act between the benefits of requiring more capital and the cost to society of imposing such a requirement.
This is not the first time that the banking industry has faced such a proposal. After the financial crisis, the Basel Committee on Banking Supervision introduced its framework for Basel III. This requires banks to have Tier 1 capital that is at least 7% of risk-weighted assets (RWA). The riskier the asset, the greater the amount of capital that the bank is required to hold against it. However, the Commission accepted that this may provide an incentive for banks to choose the riskier of the assets in any given risk-weight in order to maximise the return whilst holding the minimum capital required. Accordingly, the Commission noted that imposing an equity ratio to RWA alone may not provide a satisfactory solution. However, it also stated that imposing such a requirement would have a positive effect on the banking industry. The only question is -- what should that ratio be?
The Commission stated that a 7% equity ratio for systemically important banks is clearly too low, opting instead for a 10% minimum requirement on a Basel III basis, a ratio that should be agreed internationally. This would translate into at least a 3% 'SIFI surcharge' – an additional capital charge on systemically important financial institutions. It is important to note that this requirement is a minimum. It is entirely possible that, once the Consultation process has been completed, the actual percentage could be higher.
Loss Absorbing Capacity
As the Interim Report explains, the financial crisis revealed to the wider population that many banks were unable to bear the losses to which their risk-taking exposed them. Therefore, the Commission noted that reform is needed to ensure that banks can absorb the consequent losses if these risks go badly. It should be private stakeholders, not UK taxpayers, that bear the losses if things go wrong. A bank's enhanced ability to absorb its losses -- or at least, a significant proportion of them -- would have an array of obvious and positive effects. It would limit the adverse effect of failure on the rest of the financial system and the wider economy and, most importantly, this should serve to limit the cost of failure to the taxpayer.
The Interim Report points out one caveat on holding a minimum level of equity; that the debt of systemically important banks must also be made credibly loss-absorbing. There are a number of potential benefits to this. If bank insolvency could be avoided by converting debt instruments into equity, this would not only be beneficial for taxpayers, but could also be potentially beneficial for debt investors. The loss-absorbing debt instruments could also make up part of the additional capital that is required to be held under the new reforms.
The Interim Report proposes a number of tools for this purpose including contingent capital, bail-inable debt and depositor preference.
Contingent capital is debt that is designed to convert into equity or suffer write-down on some trigger, for example, a bank's common equity Tier 1 ratio falling below a certain level, while a bank is still viable. This would mean that in times of stress, it would be possible to re-capitalise the bank without requiring the shareholders to contribute further capital at a time when – for very obvious reasons – they may be reluctant to assume additional risk. Contingent capital carries an interest coupon which will usually be tax deductible and, hence, cheaper than equity.
It is easy to state the theory that debt should convert to equity or be written down when the bank approaches a 'trigger point'. But how should that trigger point be defined and when is it reached? Probably the most acceptable trigger from an investors' point of view would be based on a loss of capital – conversion or write-down would occur at the point at which Tier One Core Capital falls below the prescribed threshold. Other possibilities include a trigger by reference to the bank's quoted share price or CDS spreads. Some regulators have suggested that they would prefer a discretion to determine the trigger point themselves. In theory, there may be some attraction to this final option because financial crises may differ in their nature, effect and consequences, with the result that a measure of value judgment as to timing might be valuable in any given case. However, it is likely that the fixed income market would prefer an objective form of trigger. Quite apart from other considerations, regulatory discretion would make it difficult to price the instrument. So, a formula based on Tier 1 Core Capital or market conditions may be required. Banks would have to accept a corresponding obligation to make the relevant information promptly available to the market.
And what happens when the trigger is pulled? The options seem to be a conversion to equity or a debt write-down.
An equity conversion has a number of disadvantages. In particular, debt holders would effectively become the controllers of the bank overnight. They may not welcome such a change of status, not least because the investor's constitution or rules may prohibit exposures to equities or mandatory convertible instruments. In addition, investors holding significant tranches of debt may need to be approved as controllers of the banks for regulatory purposes, although this would only occur in the rare cases in which an investor is left with an equity stake of more than 10 per cent. The markets may find it easier to accept a debt write-down formula which, on some views, can be accepted simply as an accelerated insolvency procedure with equivalent consequences. It may be that this could be combined with a write-back, if the bank later reformed to financial health. Experience seems to be bearing out the write-down approach, although the market is as yet relatively shallow. Lloyds TSB Bank issued the first contingent capital bond in 2009 and this uses the equity conversion route. However, subsequent issues by other European banks have adopted the write-down approach.
It should be said that a strategy which includes holding a large amount of contingent capital has not been tested before in the industry and it is unclear what effects it could have when a bank is nearing the 'trigger point'. The market would of course be aware that a bank will have issued contingent capital and may be concerned about a forthcoming trigger point. This could lead to the withdrawal of credit lines, widening CDS spreads and rating downgrades. This may lead to a speedier deterioration which would increase the likelihood of conversion, ultimately undermining financial stability.
Contingent capital is thus intended to preserve the bank as a going concern. It had been anticipated that a significant market would develop in these instruments. However, the Basel Committee on Banking Supervision and the Financial Stability Board are negative about the use of these instruments by global and systemically important banks. The prospects for contingent capital instruments are thus uncertain at the present time.
Bail-inable debt acts in a similar way to contingent capital. The debt is designed to convert into equity which would lead to a recapitalisation of the bank, or would suffer write-down. Similarly to contingent capital, the risk of debt converting into equity and diluting existing shareholdings should, in theory, act as an incentive for shareholders to do all they can to keep the bank properly capitalised and away from the bail-in trigger. The problem in using this tool would be in identifying those liabilities that would be covered by bail-in.
Because bail-inable debt would only be triggered on a bank's non-viability, there would be other mechanisms in place that would first absorb the losses before any bail-inable debt. Losses would first be borne by the shareholders. Contingent capital would then suffer write-down or convert into equity. Only after this would the holders of bail-inable debt bear any losses. Therefore, it is important that the holders of such bail-inable debt would be able to bear the associated risks.
The Interim Report observes that this form of debt may be especially appropriate to wholesale and investments banks. In view of the cross-border nature of the business of such institutions, the Interim Report emphasises an international approach to bail-in mechanisms, arguing that an international agreement would be a way of allowing global trading businesses to fail more safely.
In line with its main goal to protect the UK taxpayer, the Interim Report suggests giving depositors preferred status in the event of insolvency. This would be achieved by subordinating the claims of other senior unsecured creditors to those of depositors.
Deposits from individuals and some smaller companies are currently guaranteed by the Financial Services Compensation Scheme up to a limit of £85,000. At present, such deposits rank pari passu with other senior unsecured creditors. This means that losses can only be imposed on senior unsecured creditors to the same extent that they are imposed on retail depositors. However, retail depositors are not as well placed to assess and monitor a bank's creditworthiness or financial status as other senior unsecured creditors. A depositor preference mechanism could alleviate this problem, as well as creating a bigger buffer that would absorb losses before there was any effect on depositors.
Structural Reform – Ring Fences and Firewalls
The Interim Report identifies three problems with combining the activities of wholesale/investment banking and retail banking within large universal banks:
- Universal banks are large and complex institutions which means that, when they fail, society and public finance pays a high price. Different banking activities require different approaches by the authorities and it can be difficult for authorities to separate the bank in order to apply the appropriate tools to each part.
- The integrated system of universal banks means that it is much harder to stop failure spreading from one arm of the bank to the next, as well as to the wider financial system. Whilst integration may be beneficial in good times, it can heighten risk at times of general economic stress.
- Universal banks also have the comfort of knowing that their risk taking is subsidised, enjoying a de facto government guarantee. This factor – often referred to as "moral hazard" – is said to incentivise large financial institutions to take greater risks.
The Commission considers that a resolution to all of these problems could be achieved by isolating the UK retail banking activities within a universal bank and placing them into a separately capitalised subsidiary (a 'retail ring-fence'). This would make it easier for authorities to resolve the problems afflicting the retail bank without simultaneously providing a bailout for the investment banking business.
A retail ring-fence could also be designed so that any surplus capital above the minimum requirement could be transferred between the wholesale/investment banking arm and the retail banking arm and vice versa, thereby generating support to each other at times of economic stress. In turn, an improvement in the authorities' ability to resolve a bank without taxpayer support would substantially reduce any expectations by market participants of a government bail-out.
Whilst acknowledging the benefits of full separation, such as a guarantee of separability in a crisis, the Commission noted that such a proposal may ultimately be more costly than is necessary to address the problems outlined.
Similarly, the Commission dismissed other structural reform ideas, such as 'narrow banking', 'full reserve banking' and 'limited purposes banking' as the benefits of introducing such reforms do not appear to outweigh the costs and the risks. Subject to the completion of the further consultation process, the Commission and the Government itself have clearly settled for a ring-fence solution.
How will this work in practice?
The Commission breaks down activities into three broad categories:
- those which must take place within the ring-fence, such as insured deposits since these are explicitly guaranteed under the terms of the Financial Services Compensation Scheme;
- those that may take place within the ring-fence, such as services typically required by individuals and small and medium-sized enterprises (SMEs); and
- activities which must not take place within the ring-fence, for example the provision of capital markets services, trading and hedging services.
The real problem of implementing such a key proposal may lie in the definition of those activities which must be ring-fenced. The Commission noted that the FSA authorisation scheme which provides definitions of specific financial activities would be a good basis for the design of a ring-fence.
The Commission stressed that this method of separation is only an illustration and that, as an alternative, separation could also occur on the basis of customer type. However, the Interim Report identifies that this could limit customer choice as certain customers could only receive services from certain entities.
In order to effectively enforce the ring-fence to ensure that it delivers a material improvement in the resolvability of banks, the Commission noted that a retail ring-fence could require additional rules such as:
- if a subsidiary seeks a licence from the regulator to conduct retail deposit-taking, that subsidiary can only conduct activities which are permitted to take place in a retail ring-fence;
- under no circumstances can the parent company transfer capital out of the retail entity if it would result in a drop below the minimum regulatory capital ratio described;
- the retail subsidiary cannot own equity in other parts of the group, thus effectively insulating it from risks and losses flowing from the non-ring-fenced business;
- intra-group exposures by, or guarantees from, the retail subsidiary will be treated as third party exposures for regulatory purposes;
- cross-defaults between the retail subsidiary and the rest of the group may also need to be limited;
- the retail subsidiary must have access to operational services which will continue in the event of insolvency of the rest of the group; and/or
- the retail subsidiary and the rest of the group must enter into separate master netting agreements.
What are the practical implications for UK banks?
As will always be the case for reforms of this kind, the devil will lie in the detail of the eventual legislation, and a full analysis must await the publication of that document. Nevertheless, it is possible to state some preliminary views about the consequences of ring-fencing for the UK banking industry.
First of all, ring-fencing must be taken to its logical conclusion. This has a number of implications:
- first of all, if an entity applies to the Financial Services Authority for permission to accept deposits, then it will only be granted other business permissions to the extent that these relate to permitted activities within the ring-fence;
- secondly, the preservation of the required capital base within the retail subsidiary necessarily implies a restriction on transfers of capital or loans to the parent institution;
- thirdly, the retail subsidiary would be barred from giving guarantees in respect of any obligations of other group entities;
- fourthly, any permissible exposures of the retail bank in respect of the remainder of the group would have to be treated on an arms length basis and risk-weighted accordingly. Techniques designed to reduce the capital assigned to such risks – such as credit risk mitigation and netting agreements – would have to be justified independently;
- fifthly, any facilities contracted by the retail subsidiary may be guaranteed by the parent institution, but the converse would not be true;
- sixthly, to reinforce the firewall, it is suggested that defaults in the parent bank or wider group should not trigger cross-default clauses in any debt documentation contracted by the retail subsidiary; and
- finally, the retail subsidiary will need to establish independent systems which can continue to operate smoothly notwithstanding difficulties which may afflict the parent institution or the rest of the group.
What are the consequences of these requirements?
Clearly, the retail and investment sides of a banking group will have to be separately funded. The limitation of recourse to the retail business is likely to increase the cost of borrowing for the investment banking side. That is, of course, one of the objectives of the ring-fencing exercise. The retail subsidiary will also need to establish its own funding arrangements but, if necessary, these could be granted by the parent institution.
Nevertheless, it is necessary to retain a sense of proportion. The Interim Report relates only to the retail deposit business of UK banks to the extent to which that business is carried on within the UK itself. Consequently, a "universal" UK bank could continue to carry on its retail deposit-taking business in other countries in which it is authorised to do so.
How will this affect foreign banks in the UK?
The treatment of foreign banks that conduct retail banking activities in the UK is not an issue that the Commission discusses in-depth. Notably, the Commission stated that it does not wish to impose global separation on UK-headquartered banks as authorities in other jurisdictions should be free to act according to their own views and needs.
However, the Interim Report does make clear that any entity seeking to conduct retail banking in the UK would need to place those activities in a separate subsidiary from their wholesale/investment operations. The Interim Report confirms that the rules would not apply to a global universal bank based in London which does not conduct UK retail activities. However, it is envisaged that these rules would apply to any entity seeking authorisation from the UK regulator to conduct retail banking in the UK.
The treatment of UK branches of banks incorporated in the European Economic Area (EEA) gives rise to possible problems in this area. Under the laws applicable within the EEA, a banking entity incorporated in one member state of the EEA is entitled to establish a branch or provide services into other member states using its home licence (the so-called "passport" system). The Commission accepts that these branches would be primarily regulated by their home authorities who would also be responsible for running their deposit insurance scheme. It would not, therefore, be possible for the UK to impose ring-fencing or firewall requirements on EEA banks which are passported into the UK since this would infringe obligations under the EU Treaties.
One effect of the clash between the UK proposals and EU rules could be that, in theory, a UK bank could try to avoid the new regulatory reforms by transferring ownership of their UK business to an EEA entity and branching back into the UK under the passport system. However, the Commission appears unconvinced that this would happen on a large scale due to the reputational and political risks that this could incur. In addition, the legal framework for the transfer of a UK banking business does not readily accommodate such a transfer, and there would accordingly be significant practical obstacles to such a move.
Nevertheless, the passporting system potentially creates a competitive imbalance in this area.
How has the Interim Report been received?
The Chancellor of the Exchequer, George Osborne, recently gave the proposals his endorsement in the annual Mansion House speech. The Chancellor stated that the Interim Report's final proposals (due in September 2011) will be judged against the following conditions:
- all banks should be allowed to fail without affecting vital banking services;
- failures should not impose costs cost on the taxpayer;
- the proposals should apply across the entirety of the diverse sector;
- the proposals must be consistent with EU and international law.
The Chancellor agreed that there was a need for further capital requirements on systemically important banks, and agreed that the Commission's proposals on ring-fencing should be agreed internationally.
More recently, a number of large banks have also now formally responded to the proposals contained in the Commission's Interim Report. Whilst the majority of banks have embraced the need for reform and are generally positive about an effective resolution regime, the ring-fence proposal has proved unpopular and has produced significant criticism from all major players in the banking sector. However, the majority of banks have welcomed the additional resolution tools proposed by the Commission, such as depositor preference and bail-inable debt.
In a lengthy response, The Royal Bank of Scotland states that ring-fencing will make UK banks even riskier "in the eyes of creditors and rating agencies", and argues that such a proposal will produce significant costs that will affect shareholders, consumers, the Government and the economy as a whole. However, on the basis that ring-fencing may be recommended by the Commission in its Final Report, RBS considers a variety of ring-fence designs, including the UK Defined Liquidity Group (DLG) framework. RBS states that this model is the best design for a ring-fence as regulators are already familiar with it, meaning a quick implementation without the "significant disruption to potentially millions of customers who would have to be moved to new sort codes and account numbers". Failing this, RBS considers the next best approach would be to impose a narrowly defined ring-fence which would include only retail and SME customers as this would leave enough diversity in the business inside the ring-fence to enable it to have reasonable prospects of stability and competitive success.
HSBC is likewise unenthusiastic about ring-fencing and considers that there would be a number of practical issues that would remain to be resolved if such a proposal was to be recommended. HSBC also rejects the idea that a ring-fence will be beneficial for all banks. The Commission's argument that a ring-fence will curtail a perceived government guarantee will not benefit HSBC because it is too large a bank to be bailed out by any single government. Instead, HSBC points towards a geographic ring-fence via subsidiarisation, which would focus on business which is of direct relevance to the UK.
Barclays is sceptical about the net contribution of a ring-fence solution and does not believe it to be necessary. It states that, in a resolution context, the retail ring-fence adds only marginal value yet would significantly increase the overall operational costs of the ring-fenced entity. Therefore, Barclays argues for operational subsidiarisation which it believes is a significantly more cost-effective way of ensuring continuity of operations.
Lloyds Banking Group states that ring-fencing could make a significant contribution to improving the resolvability of universal banks, but it may come as a cost to economic growth. Lloyds is negative about a ring-fence that is narrowly defined and would entail the separation of retail funding sources from the uses of those funds, stating that both subsidiaries would ultimately be dependent on wholesale markets which could be vulnerable in a crisis. Instead, Lloyds sets out a number of activities that must be in the ring-fence, such as a current accounts, savings accounts and mortgages, and a number of activities that can be in the ring-fence, such as corporate deposits, consumer credit and business credit. Overall, Lloyds suggests a multi-pillar approach comprising subsidiarisation, better regulation and capital requirements.
In relation to the higher capital requirements proposed by the Commission, RBS suggests increasing the required Core Tier 1 ratio for the ring-fenced entity to 10.5%. Lloyds on the other hand have stated that a bank's loss absorbency should be measured with reference to total capital rather than just Core Tier One capital. HSBC describes the process of identifying the SIFIs that are to hold higher amounts of capital as "flawed" and that it "runs the risk of creating greater market distortions and moral hazards than it is seeking to address". It also argues that the capital reforms that have been proposed by the Basel Committee will already impose a "meaningful cost" on the economy. Barclays also considers the 10% minimum capital ratio to be too high, in particular because of the existing Basel requirements that have already more than doubled the Core Tier 1 requirements. Barclays believes that requiring banks to hold more capital than the international standards will "have only marginal benefit to financial stability" and will drive up the costs of credit.
The UK banks thus place significant emphasis on a cost/benefit analysis in relation to the Commission's proposals.
Overall, the Interim Report has been welcomed as a measured approach to reform in the UK banking industry, although the banks themselves are unenthusiastic about the ring-fence proposal, believing that costs may outweigh benefits. The real battle now will be how the ring-fence will be defined, as separating a universal bank will not be an easy task. Some banks have argued that giving a broad definition to 'retail activities' could encourage rather than counter excessive risk-taking due to the implied taxpayer guarantee of retail banking, while others are keen to include all banking book assets within the ring-fence. Naturally, it will be a challenge for the Commission to find a single approach that fits all banks.
The Interim Report is, at best, vague on how structural reform will be implemented, and it does not give any in-depth discussion on how such businesses should be split. In fact, the process of identification of the 'systemically important' banks that will require some form of separation under the ring-fencing proposals is particularly unclear. In a report published on 20 July 2011, the Treasury Select Committee commented on the Interim Report's lack of depth on particular issues such as corporate governance. However, the Commission will be able to add further detail to these proposals when the Final Report is released in September.