The Bracken column is named after Brendan Bracken, the founding editor of The Banker in 1926 and chairman of the modern-day Financial Times from 1945 to 1958.

There is a good deal of support for Jacques de Larosière’s February 2009 report on how to 'repair' the EU's financial supervisory and regulatory structures, balancing just enough centralisation with national safeguards to get widespread support. But he barely considers the key question of who will provide funds to tackle a banking crisis if the decisions to save or let a bank go affect more than one country. While funding is a domestic problem, it is also a domestic political tool to be shaped in line with national objectives, national priorities and national beneficiaries.

Policy makers, central bankers, ministers of finance, supervisors, ‘wise men’, and financial market practitioners are all focussed on strengthening the content and the institutional arrangements for banking supervision and financial crisis resolution. They are working through a great collection of international forums - G7, G20, G30, FSF, IMF, BCBS, IOSCO, IIF, and Jacques de Larosière’s committee. While the net for possible solutions is being cast very wide, some of the answers are a retread of ‘solutions’ that have been around for years. Some difficult issues have not been tackled, and their omission leaves holes in the proposals that are now firmly on the table. Specifically, the debate about the centralisation of financial services supervision (as creatively addressed by the Larosière Report to the EU in February 2009) must be accompanied by a parallel discussion on how the EU pays for cross border banking crises and supervisory funding.

We see three possible general models for future EU supervision and crisis resolution, depending on the stage of market developments:

• If the EU were to revert to predominantly national structures with few cross-border financial institutions, it would be sufficient to remain with the current model, including cooperative but decentralised supervision, decentralised crisis funding. This would include some improvements on the content, but no improvements in the decision making structures nor in the current domestic funding mechanisms.

• If, in the near future, the EU were to achieve a deeply integrated financial market with many large cross-border institutional structures, most would agree that the optimum solution would be a centralised supervisory regime with a centralised lender of last resort (LOLR) and crisis funding mechanism.

• If – as we think likely – the EU will increasingly see more market integration and more systemic cross-border institutions, but member states are not willing to adopt fully centralised supervision, it will need highly cooperative and semi-centralised supervision (but with supervisory decision making ultimately resting with the home), combined with de-centralised LOLR and crisis funding mechanisms, which are, however, incentivised to cooperate with an automatic funding key contribution, based on national interests.

In anticipation of the evolution of financial markets federalists usually argue for option 2, a central supervisor. However, there is a distinct lack of enthusiasm among the majority of the EU’s member states to give up their direct influence on the way supervision is performed in their territory. Living with the status quo (option 1) has not helped EU governments reach smooth outcomes - see the cases of Icesave or Fortis - and changes to rule-making powers by themselves are not likely to have much impact on national decision-making processes. We note the retrenching currently favoured in several member states, where the fear of imported losses politically outweighs the benefits of an open financial market. We offer the third model as both politically feasible and as a mechanism that allocates costs fairly and without prejudice to national interests.

Why is funding missing from the political debate?

The contribution to the EU debate by Jacques de Larosière and his report on “repairing” regulation and supervision goes a long way in the direction of option 3. However, his report omits concrete details or even a general approach to crisis funding. Instead it simply says that crisis funding should be dealt with, and non-binding criteria should be developed. In our opinion, the lack of binding arrangements on the funding of deposit guarantees, for crisis resolution, and of the resources needed by supervisors to be able to execute their practical supervisory tasks is the key reason why public authorities have not been able to cooperate effectively, cross border, in the lead-up to a crisis. Clear decisions on funding should therefore be part of the solution.

The lack of detail on funding in the political documents is understandable. Any ideas may easily run into opposition from member states. Funding mechanisms have been seen as a domestic political tool, shaped in line with national objectives and national priorities for the banking system. The right to decide is underpinned by the right and ability to pay, and vice versa. In the current cross-border environment, the allocation of costs in the existing domestic funding mechanisms is no longer well aligned with the allocation of benefits. However, any proposal on a new funding mechanism is viable only if the incentives following from a new funding-mechanism could align national interests with effective cross-border banking supervision.

The conflicts between member states when the Icelandic banks collapsed show what happens if banking institutions grow far beyond the ability and reach of their supervisors to oversee them, and public finances to stand behind them. A collapse of a truly large cross-border bank would not, hopefully, cause a full EU member state to default, as they might expect support from other countries with more stable financial frameworks. Moreover, we would expect the euro area to act to protect the value of its currency, and prevent the collapse of one of its members. Such coordinated action would probably also occur in the USA, and the UK benefits from being a systemic part of the euro area and the EU’s economy. This covers most of the home supervisors of the largest banks active in the EU.

However, what if banks continue to grow? The collapse of a full member state due to one (or more) of its banks failing is unlikely to happen now, but that does not mean that it could not happen in the future. The seeds of such an event are being planted in this crisis: the orthodox solution to a banking failure is to seek private solutions (with or without public support), which has so far led to ever-larger institutions being created.

Some suggestions to align costs and benefits

Paying for supervision

The supervisory authorities are diverse in competence, strength, independence, resources, and sophistication. Ongoing efforts by the European Commission and the Committee of European Banking Supervisors (CEBS) are eroding these problems, but miracles cannot be expected in just a few years. The suggestion in the Larosière Report to split ‘line-side’ (direct) supervision from policy formation, with line-side supervision becoming part of a network and policy becoming more centralized, could help. It does not, however, deal with the need to link funding to supervisory responsibility. Funding to a large extent equates to being able to hire competent staff, train them to European standards, and have them spend enough time on supervising each bank.

It should be possible to calculate the relationship between resources spent by supervisors and the assets under their supervision. If a consensus can be found on the level of resources which need to be spent in relation to assets under supervision, it would create a more level playing field and a better basis for mutual trust, i.e. that other supervisory authorities are equipped to, and actually do, the tasks entrusted to them. To ensure this, there should be a binding regulation obliging each member state to spend on actual line-side supervision an amount equal to some percentage of assets supervised on solo basis, plus some percentage of assets supervised on a consolidated basis. To ensure a level playing field between banks and full harmonisation on the content, such expenditure should be between minimum and maximum levels and the regulation should stipulate that this is to be funded proportionally by the banks whose assets are taken into consideration in this calculation. The funding of policy activities does not need to be harmonized, as long as the end-result (new regulation and policy) is well formed. In practice, the policy activities of supervisors and ministries of finance are rarely under-funded in those countries that have a large financial industry.

Who pays for crisis resolution?

We also need to consider the distribution of the costs of failure. Supervisory powers are allocated to the home state at legal entity level. Increasingly, some powers are allocated to the consolidated supervisor, at the group-wide level (This is the home supervisor of the highest entity in the group). The Larosière Report proposes to allocate at least part of the burden of crisis resolution on the basis of the home state responsibility. Though good in theory (i.e. linking responsibility to liability), this is not sustainable.

• The political and financial costs of failure are felt where the depositors and investors are situated. In large and complex banking groups, operating in multiple jurisdictions, we can no longer rely on the assumption that this is the home state.

• The home state liability system is asymmetrical and relationships are not bilateral: it is not politically feasible for member state A to commit to bail out or fund the bail out of citizens of member state B, while being unsure whether its own citizens will be bailed out by member state C.

• It seems fair to allocate liability to the banking supervisor only in so far as the supervisory tasks are badly performed. But bank failures have always occurred, and will always occur, even if banking supervision is performed well (though less often and with less damage); so this allocation of costs is imperfect.

• Head offices are located in particular member states for many reasons (e.g. historical association, company structure, tax reasons etc.) but the firms themselves have grown well beyond these starting conditions.

• Banking systems are structured in so many diverse ways – large state-owned or dominated banks designed to fund local SMEs, a fully foreign-owned banking system in various emerging European countries, and large banks in many western European countries with balance sheets that are many multiples of their home country’s GDP – that it is hard to construct a single regime to allocate financial responsibility consistently and fairly by allocating liability simply by reference to the home supervisor.

We suggest that the funding of ongoing banking supervisors in good times is different from crisis resolution and deposit insurance, and the financial link between these should be cut. Allocating blame and full liability to the home supervisor will not solve the issue of funding for crisis resolution and deposit insurance, as it does not align cost with benefit. Member states consider themselves bound politically and financially to take action only if it benefits (compensates or helps) their citizens, their voters, and their taxpayers, as shown by the pattern of recent interventions (Lehman, RBS, Fortis). The academic economic model on crisis intervention funding described by Charles Goodhart and Dirk Schoenmaker (‘Fiscal Burden Sharing in Cross-Border Banking Crises’ in the March 2009 edition of the International Journal of Central Banking) supports the assumption that member states need a more direct interest than just being in the EU to be able to support an intervention. Honest acknowledgement of this fact provides the clue to determining an equitable division of costs for rescue operations, and for dividing the costs of deposit insurance if a bank is allowed to fail.

How should they pay for it?

The general division of costs of bailing or failing a banking group on the basis of the financial interest of the citizens of member states would need to be laid down in binding regulations (also suggested by Goodhart/Schoenmaker). This entails a big leap over the current practice; which is to rely on good intentions and hard negotiations. No-one disagrees in general terms that, when a crisis occurs, information should flow freely, and there is a common need to find a resolution. This has been laid down in a variety of internal and public documents such as Memorandums of Understanding and statements of principles. But none of these are binding on their signatories. The only binding provisions are in the EU Banking Directives, and even these focus solely on information provision, without giving clear indications on what exactly should be shared and without any clarity at all about timelines. Member states, central banks and supervisors can agree on these principles, because this leaves them maximum freedom for manoeuvre when a crisis occurs.

In a crisis situation we think that the decision on bailing or failing should be made by “crisis colleges” composed of the ministries of finance, supervisors and central bankers of the member states whose citizens have a financial interest in deciding on the most cost effective solution for the financial group(s) involved. This decision-making process need no longer be sidetracked by a discussion on who will pay, as that will be identical regardless of the decision to bail or fail. If this is also laid down in binding regulations, it will help overcome the distrust currently hampering progress on improving cross border access to financial services.

We suspect that the reason so many eastern European member states did not agree to changes in supervisory responsibilities (in the Solvency II and Capital Requirements Directive context) was because they did not believe that they could, or would be allowed to, play a role in crisis resolution. They (probably rightly) assumed that the needs of their citizens would not figure highly in any rescue or failure of core parts of a group that had a presence in their countries. We agree that they should only abandon this point of view and move towards more consolidated supervision if and when the interests of their citizens can be fully integrated into the decision-making processes that precede any intervention into a banking group with subsidiaries or branches in their country. While increasing their costs in a crisis resolution, they would as a result actually have a seat at the decision-making table, which can help reduce costs for their citizens. The reduction of costs might simultaneously induce the big banking centres (the UK, Germany, or the Benelux) to give such a role to the smaller member-states, as it would remove barriers to competition to their industry without taking on board uncontrollable liability.

Regardless of which EU bank they do business with, residents of a member state should be covered by the deposit insurance of their own member state. This could be funded by annual contributions, levied by the banks on their customers (e.g. 10 euro per depositor per bank, under the same criterion as for the protection offered by the deposit guarantee directive). This would eliminate unfair competition between banks, create a first buffer for crisis in member states and eliminate any ‘free rider’ behaviour by the clients. Depending on the residency or citizenship of the depositor, the annual levy would benefit the fund of his or her member state. Third-country residents with a claim against the EU business of a banking group could be covered by the deposit guarantee scheme of the home state, unless this way of allocating costs can be expanded beyond the EU. The actual payment can be funnelled through a national system of payment, to the correct deposit insurer’s fund if a payment is received from a bank earmarking how many depositors it has from each member state.

Similarly, in the event of a crisis intervention, temporarily or permanently funded by the taxpayer, the costs should be equally distributed between the member states in proportion to the protection their citizens/residents gain as a result of the action. Whether this is the location of assets or the nationality of the creditors and depositors is a matter of preference, but as assets cannot vote and financial assets are increasingly dematerialised and difficult to locate, we favour a residency criteria as a clearer indicator of national interest. The interests of the residents may be more difficult to quantify, but the recent crisis, as well as the already fully analyzed crises in e.g. the Nordic countries in the 1990s, should give ample indicators as to the direct and indirect costs to a country’s residents if a banking group is bailed or failed.

The alignment between national interests and funding will remove thresholds to information sharing. Currently public authorities are not incentivised to share information in a crisis; but with this change they would no longer have (costs) reasons to sabotage the best outcome in negotiations over which state should pay. As payment would be outcome-neutral, and the only variable would be the amount of the total payment, there would be good incentives to cooperate and fully disclose information between the various public authorities. Part funding the rescue of a foreign bank would also be defensible by local politicians as citizens would see direct and tangible benefits.

We also suggest that the division of costs should basically be the same regardless of whether it concerns liquidity assistance, solvency assistance or deposit insurance payout, and related costs of allowing a bank to fail. This includes direct costs and indirect costs, to be defined in the regulation. This should prevent the system being gamed or frustrated by different stakeholders in the liquidation of banks (as we see in the private sector, where parties holding varying positions in shares, senior debt and subordinated debt of a failed institution can make resolutions difficult).

The home member state of the group should be responsible for the payment in the first instance (as they are now), with a binding commitment that the funding will be equitably honoured by the other member states rapidly and without hesitation. This will allow the continuum of decision-making, as identified in the Larosière report, to take centre stage when moving from normal times to crisis times.

Constructive clarity

Where does this leave the orthodox theories of constructive ambiguity and moral hazard? Clear rules for sharing the burden of funding banking supervision, crisis resolution and deposit insurance do not sit easily with the need to ensure that market operators understand that they are playing in a risky arena. The Larosière Report excludes ambiguity about the role of the state, but defends ambiguity on who should be saved and why, in order to prevent banks anticipating this and embracing risky behaviour because nothing can happen to the bank in any case. We suspect that, ever since Northern Rock was bailed, and policy-makers have acknowledged the disruption caused by the Lehman bankruptcy, all banks strongly suspect that they need to be truly insignificant to be allowed to fail. This is deeply regrettable in so many different ways, but for now at least it is a fact of financial life.

This will only change if there is no ambiguity at all. It should be replaced by a doctrine of constructive clarity as to which banks, or preferably which legal entities or businesses within a banking group, would not be allowed to fail, and how that will be guaranteed (e.g. ranging from prohibitions to expand certain businesses or other government controls if a business is too systemic). Such institutions may need to operate within a tighter regulatory regime, with closer supervisory oversight, and may be required to meet specific social objectives.

If the funding issue cannot be dealt with in a clear and equitable way, we see no financially stable way in which the large and complex financial institutions we have now can be allowed to operate across borders without a high level of local supervision and interventions. If they are too big too fail, they start to become too big to be bailed by their home member state.

We offer one option to equitably divide the costs of banking supervision and bank failure - no doubt there are more. But whichever option is chosen, it should be clear and binding. The lack of binding commitments on the quality of supervision, and interventions has led to a regrettable lack of trust, between supervisors, and of the public in their banks and their supervisors. It is essential that this is reversed. Funding will be key.

The views expressed are those of the authors and do not necessarily represent the views of their organisations. Roel Theissen is an advisor on banking supervision and a senior lecturer at the Erasmus University Rotterdam, and Alan Houmann is the Director for European Government Affairs with Citi. Both writers have worked at the Committee of European Banking Supervisors on secondment from De Nederlandsche Bank and the UK Financial Services Authority, respectively.

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter