Preliminary remarks. Rome, October Mr. Edgardo Palombini, Mr. Alessio Veccia, Ms. Manuela De Cesare.

Transcription

1

2 2

3 DEPOSIT GUARANTEE SYSTEMS: EFDI S FIRST REPORT Rome, October

4

5 Preliminary remarks This research is based on qualitative and quantitative information gathered from the deposit guarantee schemes, which participated in the European Forum of Deposit Insurers (EFDI) in Data included in the analysis primarily reflect and relate to data collected at the end of 2004, since they were received in 2005; however, data updated as of the end of 2005 are also included, when provided by the DGSs in the course of The main purpose of the whole research was to propose specific schedules regarding the principal features of deposit guarantee schemes in place in each EFDI member country. Each schedule is preceded by a short introduction and it is divided into six different sections that are highlighted in different colours in order to make reading and comparing schemes features easier. The results of such research have been collected and organised in the present Paper, released under the aegis of EFDI. It is composed of three main chapters. Chapter one carefully examines the basic topics on deposit guarantee matters and it analyses the characteristics of Directive 94/19/EC. Chapter two proposes an international overview based on statistical elaborations. Finally, chapter three consists of analytical schedules, one for each EFDI member DGS. In addition to this, the Paper also includes a glossary of frequently used terminology in the deposit guarantee field and some detailed tables in the annexes. The results of the research on DGS features were analysed and collected in the present conclusive Paper by Manuela De Cesare within the Research and Data Analysis Department 1 of the Interbank Deposit Protection Fund (Fondo Interbancario di Tutela dei Depositi - FITD). Special thanks go to Alessio Veccia for his precious contributions to the Paper, the EFDI Secretariat for the support in data collection activities and obviously all EFDI members for the efforts made in providing information. The EFDI Chairman and all his staff would like to dedicate this Paper, the first ever released under the aegis of EFDI, to our dear colleague Hans Jacobson, who unexpectedly died last year, and had given a great impulse to all EFDI activities with great enthusiasm and competence. Rome, October Mr. Edgardo Palombini, Mr. Alessio Veccia, Ms. Manuela De Cesare. 5

6 6

7 Table of Contents 1. Introduction Purposes and structure of the research Basic topics on Deposit Guarantee The European Directive...19 Guiding principles...19 Review of the deposit guarantee Directive International Overview Analytical Schedules for EFDI Countries Structure of the Schedule and preliminary remarks...47 Legal framework...49 Structure and Governance of the Scheme...49 Members...50 Coverage...51 Funding...52 Interventions EU Member States...54 Austria...55 Belgium...62 Cyprus...66 Czech Republic...72 Denmark...77 Estonia...82 Finland...89 France...93 Germany...99 Greece Hungary

8 Ireland Italy Latvia Lithuania Luxembourg Malta Poland Portugal Slovakia Slovenia Spain Sweden The Netherlands United Kingdom Non EU Countries Albania Armenia Bosnia and Herzegovina Bulgaria Croatia Iceland Macedonia Montenegro Norway Romania Serbia Turkey Recent developments: new EFDI members in References

9 Annexes...I Annex I - Frequently used terminology...iii Annex II - European Forum of Deposit Insurers... XI Annex III - Synthetic view on EFDI DGSs...XV 9

10 10

11 1. Introduction 1.1 Purposes and structure of the research This Paper analyses the main features of deposit guarantee in EU and non EU countries, and it focuses on those schemes that adhere to the European Forum of Deposit Insurers (hereinafter EFDI), which partecipated in two broad surveys carried out in This Paper consists of three chapters. The distinctive feature of the first chapter is its introductory and more general tone, since it outlines essential topics of deposit guarantee on the basis of the main literature on the matter and the Directive 94/19/EC. The second chapter gives an international overview on deposit guarantee through statistical elaborations and charts based on data collected from EFDI members in Finally, chapter three features the core research which is presented in analytical schedules and brief introductory notes proposed for each Deposit Guarantee Scheme (DGS) participating in the above mentioned surveys among EFDI members. Annexes have been added to the report including: a glossary of the frequently used terminology in the deposit guarantee field and some tables providing summarised views on schemes main features for each EFDI member country. 1.2 Basic topics on Deposit Guarantee The growing interest in deposit guarantee in regulatory and operative framework of present financial systems, as well as in research fields, is directly connected to the role that banks play in the economy. As a matter of fact, banks act as intermediaries in fund allocation; they are involved in the payment system and play an important function as agents in the transmission of monetary policy. Thus, banks are exposed to risks, namely liquidity and solvency ones, which may cause losses for depositors in the event of a default, in addition to a major threat to the whole financial system when actions to mitigate contagion risks are not taken opportunely. As a result, it is for stability purposes that in each country an appropriate financial safety net has been established, which usually consists of prudential regulation and 11

12 supervision, the lender of last resort function and deposit insurance. It is easy to understand how the interrelationships between safety-net participants are an important issue to discuss. A well organised safety net, based on prudential regulation and supervision, accomplishes the task of enhancing stability within the financial system, protecting the role that banks play in the intermediation activity, the payment system and monetary policy transmission. Conversely, it may increase risk, in particular moral hazard, which relates to the incentive for banks or parties protected by the guarantee to assume an excessive risktaking 2 attitude. In situations where depositors and other creditors know they are protected from losses, or they believe to be, or when they assume that a bank is too big to fail 3, they have definitely less incentive to access information needed to monitor banks behaviour; as a consequence, weak banks may attract deposits for high-risk activities while paying costs that are notably lower than otherwise possible. Moral hazard is inherent in deposit guarantee, as in other forms of insurance. The incentive to take excessive risks is greater where blanket coverage is granted as well as in implicit systems, but it is a threat also in explicit systems, based on limited guarantee 4 ; for the purpose of the effectiveness of the financial system, strong prudential regulations and supervision are essential, as the structure 5 of the insurance scheme itself too. 2 See Financial Stability Forum, Guidance for developing effective deposit insurance systems, September See Hefler R.T. (1999), What Deposit Insurance can do and cannot do in Finance and Development, March 1999, volume 36, n.1 Moral hazard is a term economists use to refer to anything (insurance or a government subsidy, for example) that encourages risky behaviour by leading financial risk takers to believe that they will reap the benefits of the risky investments they make while being protected from losses. (...) In designing and operating a safety net, countries need to balance two competing goals, assuring stability in the financial system when liquidity and solvency problems arise while minimising moral hazard. See also: International Monetary Fund Does Deposit Insurance increase banking system stability?, January This refers to large banks, which in the event of a crisis may have heavy repercussions on the financial system s stability of a country; as governments often extend the guarantee over the limits provided by the insurance system in such cases, this extension might be seen a) as not impartial towards small banks, and b) as a weakening of market discipline. On this topic see: Ketcha J. N., Deposit insurance system design and consideration. 4 Choices on limits and coverage may depend on the nature and quality of the safety net arrangements in a particular country. Countries with strong systems of bank examination, supervision and enforcement may rely heavily on such systems to curtail moral hazard that may result from adopting relatively high limits and broad coverage. In each case, choices made by different countries with respect to limits and coverage involve a trade-off of moral hazard against the possibility of bank runs, Working Group on Deposit Insurance - WGDI Option for addressing moral hazard. Draft discussion paper, November On the topic, HM Treasury, FSA, Bank of England A framework for Guarantee Schemes in the EU: a discussion paper, 2005: Guarantee Schemes must be designed in ways that seek to minimise moral hazard, while still providing enough protection for consumers to have confidence in financial institutions. 5 See Demirguc-Kunt A., Kane E. J. Deposit Insurance around the globe: where does it work?, World Bank and Boston College, July 2001: guarantee limits, exclusion of certain categories of deposits (e.g. interbank deposits) and depositors from coverage, coinsurance arrangements, mandatory membership, ex-post funding, private-public joint management as well as targeting surviving banks to cover losses, may be useful in reducing moral hazard and enhancing market discipline, to the extent that limiting coverage in a credible way ensures that identifiable groups of private individuals 12

13 Many methods are available to mitigate moral hazard. All of them have both positive and negative aspects. Choice and practical use ultimately depend on institutional frameworks and specific features in different countries and require careful evaluation of the trade-off between aims and risks. Three main guidelines may be basically identified, also implementable jointly with the aim of attaining the most effective results: (i) monitoring by safety net participants, in a framework of strong prudential regulations and supervision; (ii) good corporate governance and sound risk management of individual banks, so that business strategies are consistent with safe and sound operations; and (iii) effective market discipline exerted by non covered depositors and other creditors 6. Besides this, specific features in a deposit guarantee scheme design are supposed to help in mitigating moral hazard, such as limited coverage, excluding certain categories of depositors, coinsurance arrangements, risk-based premiums etc. A recent analysis investigates the relationship between deposit insurance, debt-holder monitoring and risk taking for European banks; with reference to EU empirical evidence and on the basis of a stylised banking model, it appears that the introduction of an explicit guarantee system, which credibly limits the safety net to those explicitly covered, may result in a reduction in banks risk-taking (therefore, moral hazard) and in increased market monitoring of banks 7. (large depositors, subordinated debt holders or other banks) understand that their funds are at risk, and to the extent that this exposure to loss leads them to monitor the behaviour of both banks and safety-net managers. 6 See WGDI Option for addressing moral hazard. Draft discussion paper, November In a system of blanket guarantee, depositors have little or no incentive to monitor the performance of banks or discipline their risk behaviour, assuming that they have access to needed information. On the other hand, in an explicit, limited-coverage system, uninsured depositors and other unsecured creditors are aware that they are likely to sustain losses in the event of a bank failure and consequently they have a strong incentive to monitor bank performance and keep their funds in banks they regard as sound; such incentives may be reduced if depositors diversify their funds among a number of banks while staying within insurance limits. Moreover, a) effective depositor discipline as a result of limited coverage depends on whether limits are actually enforced when a bank fails, and b) strong accounting and disclosure regimes, as well as financially sophisticated depositors able to distinguish between sound and unsound banks (on the basis of available data) are also required. Bank owners also have different incentive under blanket or limited deposit guarantee systems, while seeking to control a bank s risk in order to protect their investments; such incentives are weakened if the bank is poorly capitalised or insolvent, as the liability in the event of a failure is limited to the amount of the investments. Bank managers and directors may exercise a restraining influence on excessive risk-taking activity of their bank, because of concerns about their professional reputation and potential liability. Supervisory authorities and deposit insurers may encounter serious difficulties in attempting to limit excessive bank risk in a blanket guarantee system adopted in a financial crisis; conversely, their actions are likely to be more effective in systems of limited-coverage insurance. Risky behaviour of banks and possible losses resulting from that, may be curtailed by various regulatory and supervisory actions. 7 See European Central Bank, Deposit Insurance, Moral hazard and market monitoring, Working Paper Series n. 302, February This research is based on empirical evidence regarding the European Union. It is argued that explicit deposit insurance may play a useful role as a commitment device of authorities to limit the safety net (i.e. protection of banks creditors against losses resulting from bank failures) to those explicitly covered under the deposit insurance. This 13

14 The introduction of an explicit deposit guarantee scheme (DGS) in a country can be seen as an expression of the support that government authorities intend to grant the banking system, in order to mitigate the widespread concern about negative effects and potential costs of massive and simultaneous deposits withdrawals (bank runs) and to reduce the risk of systemic failures. In this respect, it is believed that in every country at least a de facto deposit insurance system exists; this means that an implicit form of guarantee is in place, which usually derives from the necessity to manage the effects of a wide systemic crisis 8. The difference between implicit and explicit systems is due to the reliance of the latter on formal regulations adopted through banking laws, central bank rules or other regulations, on the basis of which the scheme s main features are outlined. The adoption of explicit systems seems to be more frequent as income level increases. It can be argued that effects of an explicit guarantee system deeply depend on the quality and effectiveness of banking and supervisory regulations 9. Generally speaking, relationships within the safety net, and a deposit guarantee scheme s features as well, result largely from each country s specific situation, in terms of financial history, institutional and legislative frameworks, the maturity of the economic and financial system, as well as cultural aspects and experience. points stands in contrast to the frequently reiterated argument that deposit insurance generates moral hazard and incentives for excessive risk taking by banks. If prior to the introduction of deposit insurance system, implicit guarantees were broad, the effect of introducing explicit deposit insurance on market monitoring and risk taking of banks may be positive. The reason is that all creditors of the bank aside from those insured under the explicit system may have a stronger incentive to monitor the bank. The model presented in such Paper suggests that profit maximising may increase or reduce their risk taking in response to the introduction of deposit insurance, depending on four factors. One, whether or not the explicit deposit insurance is credible in excluding all other creditors of the bank. Second, the charter value of the bank. Third, the share of liabilities aside from insured deposits on the balance-sheet of the bank. And fourth, whether or not the bank is too big to fail. As a result, the model highlights that the introduction of explicit deposit insurance, which credibly imposes limits to the safety net, will result in a reduction of risk for banks with low charter values, high share of non-insured liabilities and for banks that are not too big to fail. In the analysis, it is stressed that the limit to the safety net has to be credible for deposit insurance to have a beneficial effect. The credibility of the system may hold the key to reconciling the evidence presented in this paper with some of the evidence in the previous literature, which tended to find that deposit insurance increases moral hazard. This previous literature is largely based on data from developing or emerging markets or historical data sets from the 1920s, so it is plausible that weaker institutional structures, or the non existence of liquid markets for banks liabilities, make a limitation to the safety net less credible and subordinated debt holder or other uninsured bank creditors may not be able to perform the monitoring role that would have the risk reducing effect on banks in the presence of deposit insurance. As a consequence, deposit insurance may produce quite different effects from those shown in the Paper relating to EU countries. 8 See Demirguc-Kunt A., Karacaovali B., Laeven L. Deposit Insurance around the World: a comprehensive database, April If a formal regulation is not present for deposit insurance, we assume that the DGS is implicitly relying on the observation that every country establishes a de facto insurance system for banks. 14

15 As a consequence of such differences, deposit insurance is not offered to depositors homogeneously across countries but through specific structures, which are appropriate for the relevant institutional setting. Currently adopted schemes differ widely according to many aspects: (i) the function may be assigned either to a public or to a private (or joint) entity, independently established for such purpose or part of an already existing organisation; (ii) membership to the scheme can be mandatory or voluntary; (iii) the definition of what shall be considered eligible for coverage may vary a lot, as well as the level of coverage provided, and reimbursement can be in full or on the basis of a certain form of coinsurance; (iv) financial resources for payoffs can be collected via ex-ante contributions or by raising funds only in the event of a bank s failure (ex-post), with or without a risk-based system; and (v) according to the mandate, a DGS can be given only the task of reimbursing depositors or can be assigned a broader mandate to participate in data collection, crisis management and supervisory activities in the banking sector, varying from pay box to risk minimiser. Whatever the institutional framework 10, a deposit guarantee scheme has the main aims of promoting consumer confidence and protection, as well as enhancing financial stability, both of which ultimately depend on the DGS effectiveness in paying out compensation in the event of a member bank s failure. As far as regards the first point, a scheme has the specific task of protecting less financially sophisticated small depositors from the risk of losses they might incur when a member bank fails. Due to the close bond existing between banks in a financial system, depositors are granted the existence of a mechanism enabling them to have quick and immediate access to their covered funds, even in the event of their bank s failure 11 ; this 9 On this topic, see: Cull R., Senbet L., Sorge M., Deposit insurance and bank intermediation in the long run, BIS Working Paper n. 156, The institutional framework for EU guarantee schemes consists of three main elements, i.e. the Directive, the national legislations and rules, as well as the network of cooperation and information sharing between schemes and with supervisory authorities. See HM Treasury, FSA, Bank of England, A framework for Guarantee Schemes in the EU: a discussion paper, October Deposit insurance without an effective insolvency system can produce negative effects and enhance moral hazard, to the extent that costs for the pursuit of risky strategies are perceived on a smaller scale and excessive risk taking is encouraged as a result. In order to limit these negative effects, whether dealing with a smaller, deposit-based bank or a lager, more market-based bank, the goals of an insolvency process must focus on promptly returning insured funds to depositors, maintaining critical bank functions, and ensuring public confidence in the impartiality of the resolution process. Speed is a fundamental element in an effective bank insolvency process, see Krimminger M. H., Deposit Insurance and Bank Insolvency in a changing world: synergies and challenges, International Monetary Fund Conference, 15

16 contributes to maintaining confidence in the banking system and to minimising the risk of financial contagion, which may spread the crisis causing depositors to withdraw their funds even from safe and solid banks (contagion runs). By participating in the financial system s safety net, a DGS is supposed to contribute to keeping financial stability in stressful situations, as it may prevent depositors from engaging in bank run. This function is fulfilled insofar as the scheme helps to maintain public confidence in banks and to extend tools that are usable for banking crisis solutions, provided that banks risk levels are assessed by a strong prudential supervisory activity and an appropriate regulatory framework and sound accounting and financial reporting regimes exist 12. Where an effective control is in place, a DGS may offer a valid exit-mechanism for 28 May On this topic, see also Hefler R.T. (1999), What Deposit Insurance can do and cannot do in Finance and Development, March 1999, volume 36, n.1: The role of deposit insurance is to stabilise the financial system in the event of bank failures by assuring depositors they will have immediate access to their insured funds even if their bank fails. Thereby reducing their incentive to make a run on the bank. By discouraging bank runs, deposit insurance can prevent panic from spreading through a financial system. Such panic can threaten healthy banks as well as troubled banks (. 12 See Financial Stability Forum Guidance for developing effective deposit insurance systems, September The FSF Working Group on Deposit Insurance has developed guidance for those countries that are considering the adoption or the reform of an explicit, limited-coverage deposit insurance system (DIS). The first step in designing a deposit insurance system is to identify the public-policy objectives that the system is expected to achieve and these objectives must be well understood. In conjunction with the identification of public-policy objectives, policymakers should conduct a situational analysis, based on such factors as the level of economic activity, current monetary and fiscal policies, the state and structure of the banking system, public attitudes and expectations, the legal framework, prudential regulatory, supervisory, accounting and disclosure regimes. Moreover, an iterative self-assessment methodology should be used, as a tool to design, implement, modify and continually assess a DIS. After policymakers have completed a situational analysis, as part of a self-assessment process, attention should turn to issues such as the mandate and structure of the DGS to be implemented or modified. Mandate and powers: there is no a single mandate or set of mandates suitable for all deposit insurers, ranging from paybox systems to those with broader powers and responsibilities called risk minimiser, with a variety of combinations in between. Whatever mandate selected, it is critical that there is consistency between the stated objectives and the powers and responsibilities given to the deposit insurer; specifying the mandate clarifies the role of deposit insurance within the financial safety net and clarity of the mandate reinforces the stability of the financial system and contributes to sound governance and greater accountability. Structure: whether deposit insurance function is assigned to an existing organisation or not, it is vitally important to set clearly the responsibility and accountability of each safetynet function. Basic governance arrangements: there is a variety of forms that can be used by a DIS; however, it should reflect the mandate and the degree to which the deposit insurer is legally separate from the other financial safety-net participants. Human resources arrangements and indemnification systems are linked to the form of governance too (use of direct staff, access to resources of other safety-net participants and/or reliance on outside service providers). Interrelationships among safety-net participants: when the functions are assigned to different organisations, issues relating to information sharing, allocation of powers and responsibilities and coordination of actions among different functions are more complex and need to be addressed clearly and explicitly. Membership to the scheme should be mandatory, in order to avoid adverse selection and what is an insurable deposit should be clearly defined. It is preferable to apply deposit insurance on a per depositor per bank basis. Funding: a DGS should have available all funding mechanisms necessary to ensure the prompt reimbursement of depositors claims after a bank failure; member banks should pay the costs of deposit insurance since they and their clients directly benefit from having an effective DIS. In order for a DGS to be effective, it is essential that the public be informed about its benefits and limitations, so that credibility can be maintained and strengthened. Then cross-border issues should be considered, along with issues of resolution, reimbursement, recovery and set-off rights. Cooperation among the various financial safety-net participants, both before and after a failure, is definitely essential. 16

17 failed institutions, granting that depositors have quick access to their covered funds, so that stability and trust in the banking system are maintained. Thus, a DGS may contribute to the financial system stability provided that it is part of a well structured safety net and it is credible. Credibility means the scheme is adequately designed, correctly implemented and properly understood by the public. However, it is worth noting that a deposit guarantee system may be able to deal with only a limited number of simultaneous failures, but definitely not with a systemic crisis by itself. As far as membership to the scheme, it is argued that in general this should be on a compulsory basis so as to avoid that only (or mainly) riskier banks are attracted, giving rise to adverse selection problems. A risk-based premium system, besides providing a further tool to restrain moral hazard, may address this problem as well; but to the extent that it might not consider all kinds of risks linked with banking activity, it is necessary for membership not to be a bank s discretionary decision. The Basel Committee has recently focused on several principles that supervisory authorities should follow in dealing with weak banks, including: speed, cost efficiency, flexibility, transparency and cooperation, consistency and the avoidance of moral hazard 13. As previously mentioned, a notable and complex aspect linked to the particular function that DGSs play is the establishment up of relationships with the supervisory authorities in the financial system, which basically depends on the specific context. It can be argued that prudential supervision and deposit guarantee are complementary under certain aspects, as they pursue common targets See Basel Committee paper Supervisory guidance on dealing with weak banks: report of the task force on dealing with weak banks, Speed means that supervisors should act promptly. Experiences from many countries show that when not dealing with the problems promptly, they grow rapidly making the eventual resolution efforts more difficult and more expensive, with the possibility of becoming more widespread and systemic; cost-efficiency should lead to a least-cost criterion to guide the supervisor in making choices between alternative actions consistent with achieving the supervisory objectives; flexibility means that, although legislation frequently adopts a rules-based approach, it is also helpful if the legislator allows the supervisor to exercise discretion in the deployment and timing of supervisory tools; consistency: consistent and well-understood supervisory actions will not distort the competitive environment. Such an approach will also minimise confusion and uncertainty in times of crisis. Similar problems in different banks that are, large or small, private or state-owned, should receive similar treatment; transparency and cooperation is important, as inadequate or incorrect information from banks increases uncertainty for everyone involved. It can also lead to misplaced supervisory action and add to the costs of solving the problems. The banks and the relevant authority should aim for a high degree of information sharing and transparency about their intended actions; last but not least, avoiding moral hazard, i.e. supervisory actions should not create incentives for banks to act in a manner that would incur costs that they do not have to bear entirely. On this topic, see also Andrews M., Josefsson M., What happens after supervisory intervention? Considering bank closure options, IMF Working Paper, January

18 While defining the relationships within the safety net, there is the need to reconcile such requirements as: clearly defining the mandate assigned to each participant, in terms of tasks, powers, responsibilities, coordination of the different functions and effective information sharing 15. Note that the need for close coordination and information sharing, which are both in any case essential normally as well as in times of crisis, may vary significantly depending on whether a single entity performs all the safety net functions or these are assigned to different organisations. The DGS s form of governance reflects its mandate and the extent to which it is legally separate from the other safety net participants. Provided that information sharing is essential to the effectiveness of a scheme, its information needs vary according to its mandate (from pay-box to risk minimiser); the main source of data is usually the supervisor, due to its power to collect all accurate data needed to carry out its tasks. Furthermore, a clear allocation of powers and responsibilities within the safety net help to prevent overlaps and delayed interventions, which could happen if it is not clearly specified which institution has the power to initiate liquidation proceedings of a bank. Given the general background briefly outlined, it is worth analysing the European Union regulatory framework in detail: it is based on Directive 94/19/EC, which through minimum standards has given crucial help in spreading and establishing deposit guarantee as a factor of financial stability and growth in Europe since its introduction. 14 Deposit insurance programs, to a certain degree, increase the need for governments to supervise and regulate banks. However, it is maintained that even in the absence of a deposit insurance programme, given the unique role played by banks in a nation s economy, there is a need for the prudential supervision and regulation of banks. While the specifics of bank supervision and regulation will vary from nation to nation given their institutional, cultural, historical and legal differences, the basic goals are generally quite similar, namely: maintaining public confidence in the banking system, protecting depositors funds, fostering an efficient and competitive banking system and insuring compliance with banking laws and regulations. In this case, bank supervision, examinations and regulations provide effective mechanisms for limiting excessive risk-taking by banks. See Ketcha J. N., Deposit insurance system design and consideration. 15 See FSF Guidance for developing effective deposit insurance systems ; Kahn C. M., Santos J. A. C., Allocating bank regulatory powers: lender of last resort, deposit insurance and supervision, Bis Working Papers, n. 102, 2001; Boccuzzi G., Interrelationships between deposit insurers and supervisory authorities in the Italian and international experience, FITD Working Paper series n. 9,

19 1.3 The European Directive Guiding principles Deposit guarantee is regulated at the EU level through the Directive 94/19/EC of the European Parliament and of the Council of May 30 th 1994, later implemented by each Member State in the national regulatory framework. Through this Directive, the European Union pursues the target of guaranteeing depositors of all credit institutions, while safeguarding and enhancing stability in the banking system as a whole. On the basis of the minimum harmonisation principle, deposit guarantee has been a key factor in the completion of the single market. As a matter of fact, there is a sort of solidarity bond between banks operating in the same system, so that DGSs need to act efficiently for the benefit of depositors and need to be managed effectively in order to support supervisory authorities in times of stress of the banking system resulting from a crisis. The Directive is based on minimum harmonisation, meaning that only main deposit guarantee elements are harmonised, while the discretion of each EU Member legislative framework is allowed on all the other features not explicitly mentioned. This is the reason why the DGS design varies significantly all over Europe and among the 25 Member States, as later stressed in the course of the present paper. Harmonisation refers to main aspects of DGS such as mandatory membership for all credit institutions, exclusion of interbank deposits from coverage, minimum coverage level equal to euro 16 when deposits become unavailable, and information to be provided to depositors as an essential part of the guarantee itself. 16 Until December 31 st 1999 Member States, where the level of coverage was less than euro when the Directive was adopted, had the right to retain the maximum amount laid down in their guarantee schemes, provided that this amount was not less than euro. Art. 7, paragraph 5 of the Directive states that the amount referred to in paragraph 1 shall be reviewed periodically by the Commission at least once every five years. If appropriate, the Commission shall submit to the European Parliament and to the Council a proposal for a Directive to adjust the amount referred to in paragraph 1, taking account in particular of developments in the banking sector and the economic and monetary situation in the Community. The first review shall not take place until five years after the end of the period referred to in Article 7 (1), second subparagraph. 19

20 The Directive provides for each Member State to ensure within its territory the introduction of and official recognition of one or more deposit guarantee schemes, to which every type of credit institution is required to belong 17. Notwithstanding this, an EU country may be allowed, under specific conditions 18, to exempt a credit institution from the obligation of joining a deposit guarantee scheme, informing the Commission accordingly; this option is available provided that the bank at hand belongs to an other insurance system, which protects the bank itself ensuring its liquidity and solvency and grants depositors a coverage almost equivalent to the one offered by the DGS. Moreover, European regulations provide for the procedures to follow when a credit institution does not comply with the obligations it is supposed to observe as a member of a scheme (measures may vary from the application of sanctions to the exclusion from the system and the withdrawal of the license) 19. The Directive also states that an officially established and recognised scheme in an EU country extends its guarantee to depositors of a member bank s branches operating in other Member States as well 20. Directive 94/19/EC thus extends to deposit guarantee such general principles as Home Country Control and recognition in host Member States of the legislation applicable to the credit institution s head office, which along with all its branches forms a sole entity subject to sole recognition and to centralised supervision within the EU. 17 See Directive, art. 3, paragraph Same art. as in previous note: the system must be in existence and have been officially recognised when this Directive is adopted; the system must be designed to prevent deposits with credit institutions belonging to the system from becoming unavailable and have the resources necessary for that purpose at its disposal; the system must not consist of a guarantee granted to a credit institution by a Member State itself or by any of its local or regional authorities; the system must ensure that depositors are informed in accordance with the terms and conditions laid down in Article See Directive, art. 3, paragraphs See Directive, art. 4, paragraph 1. As to limits provided for in the Directive, a temporary derogation had been in force until December 31 st 1999, on the basis of which depositors of EU Member banks in other EU States were protected up to the limit and scope of guarantee provided for in the host country scheme (export prohibition clause). See art. 4, paragraph 1of the Directive Until 31 December 1999 neither the level nor the scope, including the percentage, of cover provided shall exceed the maximum level or scope of cover offered by the corresponding guarantee scheme within the territory of the host Member State (topping-down). This clause caused a temporary derogation of general purpose principles in the Directive, namely minimum harmonisation and mutual recognition. Reasons for its introduction were enhancing gradual convergence to the provided harmonisation and avoiding market perturbations possibly deriving, in the very beginning phase, from differences in coverage provided by different countries within EU. The Commission, while rewieving the clause as stated by the Directive itself, considered it no more important due to an essential insignificant impact on competition of the remaining differences between coverage levels. The clause and related limitation were abolished as of 12/31/

21 Provided that in an EU context every type of credit institution is bound to participate in the home Member State s DGS, branches of such institutions are not obliged to adhere to the host Member country DGS too; even though, they may decide to join it voluntarily, under certain circumstances expressly mentioned in the Directive 21. Conversely, branches established in EU Member States by banks with head offices outside the EU shall belong to a system in their home country granting a cover equivalent to that provided for in the Directive; if the home country does not have a DGS or, in case it has, but it fails to have a coverage deemed congruent and equivalent, EU Member States may require these branches to join the DGS active within their territories. Voluntary membership available to EU branches relies on the topping up mechanism. According to this, the branch may extend the coverage offered to depositors up to the maximum granted in the host Member country, if this exceeds in term of level and/or scope (including the percentage) the guarantee provided in the Member State where the credit institution is authorised and to which DGS has to consequently belong 22. To this extent, host Member States having a guarantee above the EU minimum level must ensure that their DGS can offer supplementary coverage to foreign EU branches, upon request. Branches applying for voluntary membership are subject to all obligations incumbent on members of the host country scheme. In case of non compliance with such rules, the home country s competent authorities are notified and they cooperate with the DGS in order to take all appropriate actions to ensure that these obligations are observed, including the exclusion of the branch and the withdrawal of the supplementary cover if necessary. 21 That is the so called clause on topping up; see Directive art. 4, paragraphs 2-5. In particular: 2. Where the level and/or scope, including the percentage, of cover offered by the host Member State guarantee scheme exceeds the level and/or scope of cover provided in the Member State in which a credit institution is authorised, the host Member State shall ensure that there is an officially recognised deposit-guarantee scheme within its territory which a branch may join voluntarily in order to supplement the guarantee which its depositors already enjoy by virtue of its membership of its home Member State scheme. The scheme to be joined by the branch shall cover the category of institution to which it belongs or most closely corresponds in the host Member State. 3. Member States shall ensure that objective and generally applied conditions are established for branches' membership of a host Member State's scheme in accordance with paragraph 2. Admission shall be conditional on fulfilment of the relevant obligations of membership, including in particular payment of any contributions and other charges. Member States shall follow the guiding principles set out in Annex II in implementing this paragraph. 22 Topping up Agreement refers to the agreement between an EU bank, which has a branch in another EU country, and the deposit guarantee scheme operating in the host country. It contains provisions on supplementary coverage provided by the host country scheme to the branch, as the bank wants to extend its home country scheme s protection in the host country. 21

22 Reasons for topping up may be various, even though this option was mainly considered important at the time Directive 94/19/EC was introduced in order to prevent distortions of competition between institutions. Nevertheless, in very few cases voluntary membership and topping up have been activated since 1994 and only a limited number of agreements between schemes are effective under the current framework. This low interest in topping up is partly due to its nature, being a derogation of the general principle of home country control, but it is argued that there are also other factors as well. One of these relates to the substantial convergence in deposit guarantee conditions within Europe, which is due to the minimum harmonisation rules imposed on the level and scope of coverage by the Directive. In fact, most Member States have compulsory deposit guarantees amounting to between and euro 23, while the extent of their scope of coverage basically depends on how Annex I of the Directive was implemented in the national legislation. It is believed that banks do not regard deposit guarantee as a competitive factor, or at least they do not consider it important enough for their competitiveness to bear the cost of a supplementary cover. This attitude also results from such factors as: (i) low public awareness about the differences between guarantees and (ii) stability in the financial markets within EU. Furthermore the burden of additional information requests from the host Member States DGS might raise confidentiality concerns sometimes. Finally, it is worth noting that major banks are supposed to attract business in host markets on the basis of their standing, so that their branches have little interest in asking for topping up. As far as Annex II of the Directive is concerned, the home and the host DGSs shall bilaterally establish appropriate rules and procedures when a branch applies to join the host member scheme for supplementary cover. The purpose of this Bilateral Agreement (BA) is that of cooperation between schemes and mutual information exchange for paying compensation to depositors at that branch, since both schemes are responsible for pay-outs 23 Coverage is considerably higher only in four countries within the EU: Denmark (around euro), France ( euro), Italy ( ,38 euro) and the United Kingdom (around euro, with coinsurance). Among other European Area Counties, that are non EU Members, only a notably higher level is provided in Norway, equal to around euro (the explicit guarantee system was introduced in 1961, through two separate schemes for commercial banks and savings banks respectively; actually, a guarantee system for savings banks had been introduced on voluntary basis before that date, in 1921, becoming mandatory in 1924, but this is not considered a pure deposit guarantee system. It is worth noting that the two above mentioned schemes merged recently). 22

23 when a crisis arises and an insured institution is no longer able to fulfill its obligation towards depositors 24. As noted before, few bilateral agreements have been concluded so far in the EU 25 ; quite often, such few documents were signed not considering necessary the existence of a previously signed topping up agreement between the countries or a specific request or need of the branch at hand. They are usually rather general in nature, basically in line with the guidelines in Annex II on information sharing and cooperation between deposit guarantee schemes. Most Member States seem to agree with the current provision in the Directive, where the need for bilateral agreements only comes after a topping up agreement has been concluded, arguing that such BA is likely to facilitate a prompt reimbursement of depositors claims. Nevertheless, some EU Members assert that in any event such a generally stated agreement might not facilitate crisis management, because every pay-out situation is different and its specificity is the result of the systems involved. However, the vast majority of Member States seem to prefer that the current general system based on guiding principle is maintained, rather than introducing an EU model of agreement. Finally, a further factor is worth a remark: information requirements, which are imposed by the host DGS to all its members, might negatively affect the interest in topping up by foreign branches in certain situations. 24 Annex II of the Directive includes the guiding principle that shall apply both to the drawing up of those procedures and in the framing of the membership conditions applicable to such a branch (as referred to in Article 4 (2)) : a) the host Member State scheme will retain full rights to impose its objective and generally applied rules on participating credit institutions; it will be able to require the provision of relevant information and have the right to verify such information with the home Member State's competent authorities; b) the host Member State scheme will meet claims for supplementary compensation upon a declaration from the home Member State's competent authorities that deposits are unavailable. The host Member State scheme will retain full rights to verify a depositor's entitlement according to its own standards and procedures before paying supplementary compensation; c) home Member State and host Member State schemes will cooperate fully with each other to ensure that depositors receive compensation promptly and in the correct amounts. In particular, they will agree on how the existence of a counterclaim which may give rise to set-off under either scheme will affect the compensation paid to the depositor by each scheme; d) host Member State schemes will be entitled to charge branches for supplementary cover on an appropriate basis which takes into account the guarantee funded by the home Member State scheme. To facilitate charging, the host Member State scheme will be entitled to assume that its liability will in all circumstances be limited to the excess of the guarantee it has offered over the guarantee offered by the home Member State regardless of whether the home Member State actually pays any compensation in respect of deposits held within the host Member State's territory. 25 As of December 31 st 2005 there are 7 bilateral agreements, between: i) Denmark and Sweden; ii) Finland and Sweden; iii) Germany and Luxembourg; iv) Ireland and the United Kingdom; v) Italy and Slovenia; vi) Luxembourg and Sweden; and vii) The Netherlands and United Kingdom. 23

24 The EU Commission firstly reported in on the operation of the topping up clause, since this was stipulated in the Directive. Being aware that the provision had been used in a very limited way, the Commission carried out the analysis taking into account whether to abolish the clause, propose amendments, or even investigate the possibility of a topping up offered by the home country scheme, rather than by the host DGS as in the present framework. Arguments for keeping unchanged the topping up provision (i.e. the current regime of host-country responsibility) were considered prevalent, mainly with respect to the planned EU enlargement process and the likely role of such clause in crossborder establishment of branches by new Member States banks. Furthermore, it was argued that there was neither clear evidence of the assumed greater attractiveness for banks of a home-based topping up system, nor anything specific that could actually prevent banks from organising such a system on a voluntary basis, in case they would find it more opportune. Currently, the topping up provision is again under analysis within the wide-ranging review of the Directive that the EU Commission began undertaking in early 2005 and it is still underway in To this extent, the Commission sent a consultative working paper on deposit guarantee schemes to Member States in mid July 2005; it also carried out an impact assessment on the minimum coverage level, whose results were published in November Further steps will be a communication of the Commission, which is expected in autumn 2006, and a conclusive report with a DGS overview to be issued by the end of December 2006; these two steps will follow a broad consultation with EU Members, on the basis of a detailed survey distributed to every DGS in May 2006 and all related scenario analyses. 26 As provided for in art. 4, paragraph 5 of the Directive: 5. The Commission shall report on the operation of paragraphs 2, 3 and 4 no later than 31 December 1999 and shall, if appropriate, propose amendments thereto. 24

25 Review of the deposit guarantee Directive In the course of 2005 the EU Commission began to undertake a thorough analysis relating to various aspects of Directive 94/19/EC on deposit guarantee schemes, as part of the revision activities required by the Directive itself in primis for the periodical assessment of the minimum harmonised guarantee level 27. Such a broad task for reviewing DGS features is a consequence of various discussion on supervisory convergence in recent times and specific requests from Member States and other authorities; in any case, there is no common view at the moment on whether and to what extent DGS features need to be further harmonised 28. Also, there is no consensus on which additional aspects could/should be converged. In support of likely amendments, the review argued that the present framework based on the Directive seems to reflect only a partial picture of the present European banking system, which has been modified since the Directive entered into force in It is worth noting that such review in the deposit insurance field is just one part of a more extended analysis involving more general topics such as solvency, financial stability, crisis management, supervision and relationships between home and host authorities, as well as the other types of guarantee systems existing in the EU (for investment and insurance, respectively). The reasons why the Commission decided to extend the review of the Directive beyond the assessment on the minimum coverage level, thereby investigating the possibility of further harmonisation, may be as follows: (i) the need to adequately cater to all crossborder deposit protection risks; (ii) the necessity to avoid any competitive distortions in the single market, possibly relating both to costs of the DGS to banks and to the guarantee granteed to depositors; (ii) the need to avoid that deposit insurance creates incentives or disincentives for banks to change the location of their company s registered office in order to minimise costs connected with it; and (iv) the possibility of easing crisis management 27 See Directive, art. 7, paragraph As aforementioned, the Commission s timing consists of several steps of analysis, including consultation activities, an impact assessment, broad surveys and some discussion tables involving BAC, European Banking Committee and Financial Stability Table, BCE and CEBS, as well as experts from Member States, deposit guarantee schemes and EFDI. 25

26 while harmonising the Directive provisions and improving information sharing arrangements. It is argued that deposit guarantee may not be a determining factor in any of the above mentioned issues but it is one of the factors involved in the decision making process of market participants, as a recent experience suggests 29. To this extent, the EU Commission is convinced that effort should be made to avoid that national DGSs either act as a barrier or an incentive to the decision of market participants concerning the location of corporate headquarters. The revision activity undertaken by the Commission is focused on such aspects of the current Directive as follows: (i) the minimum harmonised level of coverage, (ii) the definition of deposits, (iii) topping up arrangements, (iv) funding mechanisms, (v) the division of home and host country responsibilities, relating to information sharing agreements and crisis management procedures, and (vi) the asset of relationships between national central banks and deposit guarantee schemes. The minimum harmonised guarantee and the definition of deposits As mentioned previously, this topic constitutes the main focus of the revision activities, on the basis of the Directive s provision that addresses that the Commission reviews the minimum coverage level periodically, at least every five years by the end of the transitional period (December 31 st 1999). 29 This refers to Nordea, the largest group in the Scandinavian region, which decided to change the group s legal structure by creating one European Company - "Societas Europaea" (SE), based on a European Company Statute in force since late This is the last step of a restructuring process that began in 1998 when Merita Bank (Fin), Nordbanken (Swe), Unibank (Den) and Christiania Bank (Nor) merged forming a fully integrated international financial group, which is the leader in the internet banking sector. Adopting the Statutes of the European Company, Nordea has the aim of establishing a one-bank structure, with one legal entity conducting business in all local markets through branches, with benefits in terms of improved operational efficiency, reduced operational risks and enhanced capital efficiency. The changed structure will reduce complexity and support the bank s philosophy of business decision-making power being close to the customers and employees. Moreover, difficulties relating to four different national legislations (Denmark, Finland, Norway, Sweden), with many payment systems and supervisory authorities, would be overcome, as well as value-added tax on intragroup operations. To the extent that adopting this simplified legal structure implies the changing of subsidiaries into branches, from a deposit guarantee point of view this means that all newly created branches exit their respective host country schemes and join the home country one (i.e. the State where the holding has its headquarters, Sweden in this case). 26

27 To this extent, the Commission firstly carried out an impact assessment 30 in 2005 in order to: (i) describe the current situation, in terms of deposit distribution and coverage levels; (ii) compare it with the situation in place in 1995, taking into account such factors as rising prices and averages on deposits (both as numbers and volumes); and (iii) perform an analysis on possible modifications of the minimum guarantee level so as to outline some alternative scenarios for the definition of the theoretical optimal level of guarantee 31. It is worth noting that parameters used in 1994 for deciding on the level of coverage and risk allocation are still deemed to be sound arguments: the aim is to find out an equilibrium point between a too high and a too low level of coverage, since the former could lead banks to excessive risk taking and depositors to potential negligence, while the latter would leave an excessive portion of deposits unprotected over the minimum threshold, facing the risk of producing financial instability as a result of depositors run to withdraw their deposits in times of crisis. On the basis of such arguments and considering the median value of EU coverage levels in 1994, the Commission s initial proposal addressed a minimum harmonised guarantee threshold equal to euro 32 ; later on, such minimum level was raised up to euro during the legislative procedure. The mentioned impact assessment, whose results had been summarised in a Report, pointed out how the average coverage level has decreased in real terms since 1995, mainly as a result of the erosive effect of inflation in those countries where the guarantee was 30 The Report on the minimum guarantee level of Deposit Guarantee Scheme Directive 94/19/EC of the European Commission, with the impact assessment results, was published in November 2005 and it is available on the website of the EU Commission - DG Internal Market - Banks ( 31 As far as regards the data gathering exercise carried out in spring 2005 for the purpose of impact assessment, most Member States could provide aggregate information only, not being generally able to supply data on deposit distributions. To overcome this problem and to estimate likely effects of adjustments on the minimum guarantee level and to outline alternative scenarios, deposit distributions (both insured and guaranteed) have been proxied by means of a related macroeconomic variable, whose distribution is readily available: household disposable income (part of which goes to household savings and part of these savings is placed in deposits), defined as the difference between the total household income and direct taxes, compulsory fees and fines. This methodology has allowed only a preliminary and rough estimation of the amount of insured deposit distribution; it has not provided any information on the distribution of the number of deposits, which is required in its turn to derive the guaranteed deposit distribution as well as to perform any scenario analysis of different deposit guarantee levels. To obtain a view on the distributions of the number of deposits the minimum, maximum or average number of deposits in each bracket has been estimate by dividing the bracket total amount of insured deposits by the upper bound, the lower bound or the average bracket value as appropriate (assuming that the amount of insured deposits is uniformly distributed inside each bracket, i.e. that each bracket is composed of a number of deposits with an equal amount). 32 In 1994, excluding those States with very high levels of coverage (Germany and Italy) and those without any guarantee at all (Greece and Portugal), euro was approximately the median level between coverage levels provided in Europe at that time. Nowadays, the median on current coverage levels in all EU member countries is equal to euro. 27

28 higher than the minimum harmonised provided for by the Directive upon its introduction, and as a consequence of the general economic growth since On the contrary, an increase in the average guarantee level has been recorded in nominal terms; such opposite variation is due both to the end (12/31/1999) of transitional regimes (enabling some Member States to keep guarantee levels in a range between euro and euro) and to upward adjustments in some other countries (especially in UK) after It also highlighted how the levels of coverage expressed in real terms have converged over time in the EU-15 area, while the recent enlargement process (May 1 st 2004) has somewhat inverted this phenomenon as a result of the very different economic significance that deposit guarantee assumes in the ten new Member States. In fact, both EU-15 countries and new Member States provide similar levels of deposit protection; notwithstanding this, such guarantee levels differ in terms of economic significance (GDP per capita and guaranteed deposits / insured deposits ratio) and deposits are de facto much more protected in the new Member States than they were in 1995 in the original 15 EU Member States 33. As far as regards the guaranteed/insured deposits ratio, a progressive increase was observed in EU-15 countries, especially in some of them, while such ratio seems to have substantially higher levels in the ten countries that recently joined the European Union. Moreover, the EU enlargement produced a slight but significant increase in the differences between Member States relating to the percentage of eligible deposits that is covered by the DGS. In its Report, the Commission proposes three different scenarios in order to test the impact of a potential revision of the minimum guarantee threshold: moving towards a fully harmonised level set at euro, adjusting the existing guarantee levels in accordance with either national inflation trends or economic activity development trends (this latter scenario is in terms of national per capita GDP). 33 In the 10 new Member States the amount of insured deposits is lower than in the other 15 EU countries, although the level of guarantee is high when expressed on a GDP per capita basis. The limited data available about guaranteed deposits compared to the total amount of insured deposits suggests that the proportion of deposits which are guaranteed can be very different in the various Member States. Estimates based on household disposable income, in the absence of data about the distributions of deposits, would suggest a confirmation of the notion that the proportion of guaranteed deposits compared to insured deposits has generally fallen over time and differs from country to country. 28

29 Given the current framework, where the guaranteed/insured deposits ratio (calculated on the basis of aggregated data and estimated distribution) ranges between minimum values close to zero to maximum values around 100%, it does not seem to be evident whether the two aims of the Directive - not to leave too great portion of deposits without protection in the interest both of consumer protection and of the stability of the financial system - have been achieved since the regulations entered into force in In this respect, it is stressed that any quantitative analysis, like the discussed impact assessment, can not easily explain whether qualitative targets have been met, since such objectives seem not to be clearly expressed firstly in the Directive itself; therefore, further clarifications would be useful before making any decision on how (and if) to create any possible further harmonisation 34. Besides the minimum coverage level, there are other aspects relating to it, which are currently under analysis for the purpose of investigating whether or not there is room for further harmonisation. Such issues are those relating to coinsurance and the possible introduction of a de minimis clause. The first topic refers to the percentage of deposits provided for in article 7, paragraph 4 of the Directive, to which Member States may limit the guarantee provided for either in paragraph 1 (the minimum harmonised level) or in paragraph 3 (the possible higher or more comprehensive cover) of the same article; this limitation is allowed as long as the percentage guaranteed is equal to or exceeds 90% of aggregate deposits until the amount to be paid under the guarantee reaches the amount referred to in paragraph 1. Up till now 34 As from the Report of the Commission: the objective of covering a percentage of insurable (i.e. insured in the present research) deposits in all countries within a common (harmonised) interval by means of guarantee levels which might also instrumentally differ in nominal terms in each Member State might represent one viable option for the future in line with the spirit of the Directive. Such a possibility requires however a refinement of the work done to estimate the deposits distributions, which at the moment are - although useful for the present report - not sufficiently accurate for such a policy decision. Further work would therefore be useful on this as well as other alternative scenarios, combined with quantitative analysis as to the cost implications of possible changes to the current guarantee levels... One solution could, once more, be the definition of different coverage levels compatible in all countries with a common range in the insured / insurable deposits ratio (i.e. guaranteed over insured deposits in the present research), although the resulting nominal differences between guarantee levels across Member States might not then fulfil the internal market objectives in the Directive. 29

30 coinsurance, even though introduced for the purposes of restraining moral hazard, has been little used in EU Member States 35. As far as regards the de minimis clause, introducing such a rule in the Directive would take into account the often disproportionate administrative costs, which may be associated with the reimbursement of very small sums; to the extent of avoiding such burden, a number of DGSs would like the introduction of a new clause in the Directive setting a threshold (e.g. 20 euro) below which repayments would not necessarily be honoured. Other Member States have raised doubts as to whether this kind of rule would ensure both the equal treatment of depositors and the compliance with the Directive s primary aim to protect less financially sophisticated depositors. Finally, the Commission has been investigating the possibility of proposing amendments or clarifications to article 1 of the Directive, regarding the definition of deposits Coinsurance is applied in 12 out of 25 EU Member States. Generally speaking, coinsurance is the mechanism by which depositors are not protected for the full amount of their deposits up to the coverage level provided for in the legislation, but only for a certain percentage of them. As a result, depositors can bear losses, whose maximum threshold is set at 10% by the Directive. In 4 out of 12 States a double mechanism exists: depositors are reimbursed in full up to a certain amount, then 10% coinsurance is applied up to the coverage level provided for in the regulation. Note that two different value may be defined as far as regards the coverage provided by a DGS, depending on whether coinsurance is applied or not: the level of coverage and the payout limit, i.e. the maximum protection granted in the event of deposits being unavailable under the national law, not applying coinsurance or once it is applied respectively. 36 See art. 1 of the Directive:. «deposit shall mean any credit balance which results from funds left in an account or from temporary situations deriving from normal banking transactions and which a credit institution must repay under the legal and contractual conditions applicable, and any debt evidenced by a certificate issued by a credit institution. For the purpose of calculating a credit balance, Member States shall apply the rules and regulations relating to set-off and counterclaims according to the legal and contractual conditions applicable to a deposit; «joint account» shall mean an account opened in the names of two or more persons or over which two or more persons have rights that may operate against the signature of one or more of those persons; «unavailable deposit» shall mean a deposit that is due and payable but has not been paid by a credit institution under the legal and contractual conditions applicable thereto, where either: the relevant competent authorities have determined that in their view the credit institution concerned appears to be unable for the time being, for reasons which are directly related to its financial circumstances, to repay the deposit and to have no current prospect of being able to do so. The competent authorities shall make that determination as soon as possible and at the latest 21 days after first becoming satisfied that a credit institution has failed to repay deposits which are due and payable; or a judicial authority has made a ruling for reasons which are directly related to the credit institution's financial circumstances which has the effect of suspending depositors' ability to make claims against it, should that occur before the aforementioned determination has been made; «credit institution» shall mean an undertaking the business of which is to receive deposits or other repayable funds from the public and to grant credits for its own account; «branch» shall mean a place of business which forms a legally dependent part of a credit institution and which conducts directly all or some of the operations inherent in the business of credit institutions; any number of branches set up in the same Member State by a credit institution which has its head office in another Member State shall be regarded as a single branch. 30

31 Such provision, which is quite general in fact, was implemented by Member States in various and often more or less extended ways, also as a result of the total or partial adoption of the exclusions provided for in Annex I of the Directive 37. In practice, this means different methods and procedures in the definition and calculation of deposits, which might result in differential treatment between depositors in different Member States, affect the volume of deposits covered under each DGS and produce further possible implications on funding and costs too. Proposing the harmonisation of all national provisions on this matter seems to be definitely unrealistic, although more clearness on certain aspects (such as unequivocally defining covered vs. eligible deposits) could help to overcome difficulties of interpretation between countries. Topping up arrangements As previously noted, the provision concerning topping up has been little used since the introduction of the Directive more than ten years ago. The rationale for this behaviour could be due to the fact that different guarantee levels do not seem to influence depositor s choices significantly. Also the low competitiveness in the retail sector and in the perceived complexity of the current framework may be another explanation. No data are available on any depositors pay-offs carried out under a topping up agreement, thus no possible related analysis and assessments have been or may currently be carried out. When the Commission decided to introduce topping up arrangements in the original proposal for a directive, its opinion was that the Home Country Principle needed to be tempered by a clause allowing EU banks branches to participate voluntarily in the host country DGS, provided that the coverage was higher (or wider in scope) than the guarantee 37 There are two different kinds of exclusions in the Directive deriving from art. 2 and Annex I, respectively. On the one hand, Article 2 lists all the funds that must be excluded from coverage (deposits made by other credit institutions on their own behalf and for their own account; all instruments which would fall within the definition of 'own funds' in Article 2 of Council Directive 89/299/EEC of 17 April 1989 on the own funds of credit institutions; deposits arising out of transactions in connection with which there has been a criminal conviction for money laundering as defined in Article 1 of Council Directive 91/308/EEC of 10 June 1991 on prevention of the use of the financial system for the purpose of money laundering); on the other hand, Annex I lists those deposits that may be excluded from protection, as provided for in Article 7, paragraph. 2: Member States may provide that certain depositors or deposits shall be excluded from guarantee or shall be granted a lower level of guarantee. Those exclusions are listed in Annex I. As far as regards this last provision, none of the 14 cathegories of deposits listed in Annex I is granted a lower level of coverage in any Member 31

32 offered in the home country. The idea was that such provision would prevent distortion of competition between banks and differences in protection levels unlikely acceptable by depositors, who above all would be enabled to benefit from the higher coverage level of the host country. It is logical that topping up is relevant only in those Member countries where the national DGS provides a level or a scope of coverage that is greater and vaster than the minimum provided for in the Directive. In any case topping up is an option that depends on a bank s discretion, and the issue of arbitrage opportunities may consequently arise; to this extent, credit institutions may decide whether or not to use topping up agreements on the basis of the costs of such arrangements at present or in the foreseeable future. As a result, risks could be concentrated in countries where a cost-free or cheaper guarantee is offered 38, with important implications on financial stability. At the present time, the Commission does not seem in favour of abolishing the topping up provision; through the review of the Directive and related analysis, it would check whether Member States are interested in future changes to the current arrangements, and it would evaluate carefully the implications of any possible amendment. Funding mechanisms Funding is one of the main DGS features that is not subject to any harmonisation rule in the Directive 94/19/EC. In the recitals, it states that the cost of financing must be borne, in principle, by credit institutions participating in the DGS, while decisions on the scheme s organisational layout and funding method are left to a Member State s individual discretion. Moreover, the financing capacity of such deposit guarantee schemes must be in proportion to their liabilities, so as to avoid any possible threat to the banking system s stability of the Member State concerned. State. Exclusions may vary significantly from Member State to Member State (for details, see the relevant table in the Annexes to this Paper). 38 This refers basically to ex-post schemes. 32

33 As a result of a Member State s individual discretion that the Directive allows on such and several other aspects, existing deposit guarantee schemes are widely heterogeneous in Europe. In practice, one of the main reasons why funding mechanisms can be differentiated depends on whether or not a fund to be managed exists. On the one hand, guarantee schemes are referred to as ex-ante systems when a fund exists and member institutions are called to pay periodical contributions to build up the fund, in addition to a possible fixed entry fee, where required. Such regular payments are calculated by the DGS on the basis of a bank s assets, volume of deposits (either total, insured or guaranteed) or on other parameters as of a certain date. Pre-funded schemes have the merit of ensuring that money is quickly available to reimburse depositors claims in the event of a crisis, without risks to other banks provided that the crisis is not system-wide and taxpayers are not consequently required to supplement the scheme s insufficient funds. On the other hand, in ex-post schemes there is not a fund to be managed in normal times and member institutions are required to make funds available only after a crisis or insolvency event occurs. The main drawback in the management of the crisis is that the funds might take a longer time to be accessed. DGSs based on intermediate funding methods can be referred to as mixed 39 schemes. While reviewing the Directive, the EU Commission has highlighted various aspects deemed to be worthy of further analysis; in order to achieve this target, it has been carrying out broad consultations with EU Members since the beginning of the revision activities. The vast majority of such aspects are important particularly for ex-ante systems. Generally speaking, differences between countries in the costs of deposit guarantee schemes, which are due to the funding mechanism in use, might constitute an element of competitive distortion and, if significant enough, such differences might give rise to arbitrage opportunities. 39 As far as EU Member States, there are ex-ante schemes in 14 countries, ex-post schemes in 5 and mixed ones in the remaining 6. All the other 12 EFDI member countries, which are outside EU, are ex-ante funded. 33

34 One of the elements relating to the discussion in focus is whether there is room for harmonising the funding of DGSs. Doubts on the matter arise from all parties concerned 40. On the topic, the EU Commission carried out a survey in May-June 2006 in Member States. Such data gathering excercise, although quite vast in nature, was supposed to basically focus on funding arrangements, in order to assess, by means of a specific scenario analysis, the likely cost implications of harmonising funding within the EU. Work is still underway and the Commission is going to release a communication based on the survey results in late Another topic, which is important for the purposes of the review, concerns the role that differences in financing methods between countries may play in the event of cross-border operations. Moreover, increasing cross-border activities and integration in the EU banking sector draw more and more the attention to the responsibilities regime between home and host guarantee schemes, as addressed in the following section. Finally, whether deposit guarantee schemes should be funded on risk adjusted premiums or not is under analysis; as such practice could contribute to restraining moral hazard, it seems to be desiderable in principle. Currently, few schemes in the EU calculate premiums on the basis of the level of risk associated with their member institutions 41. The division of home and host country responsibilities When a Directive on deposit guarantee schemes was initially proposed, the EU Commission decided to tightly link its proposal to the home country principle, deemed consistent with the general approach of mutual recognition adopted for the banking sector and more generally for the Single Market after The opinion expressed on the matter 40 Harmonisation is very difficult on this topic and there are both advantages and disadvantages. Those in favour of looking for harmonisation of some aspects of funding argue that such an agreement between Member States could prevent them from managing schemes that could not be able to bear their own costs in the long term. Moreover, common rules could neutralise some unjustified competition elements and help in creating a more regular playing field. 41 As explained in chapter 2, monitoring systems to assess the risk level of member banks and/or risk-based premium systems are used in 16 DGS (10 EU Member States and 3 non EU countries), corresponding to 38% of the sample (42 EFDI member schemes). 42 Relating to the opinion expressed by the BAC in that time, the breach of the home country approach in an area so closely linked to banking supervision as deposit guarantee schemes would set a potentially dangerous precedent in the creation of the Single Market in banking services. 34

35 by the BAC was decisive; it highlighted the potential dangerousness of applying a different approach in an area so closely linked to banking supervision such as deposit guarantee. Since branches of EU banks are supervised by the authorities of the Member State where the head office is located (i.e. the bank s home country) and branches form a single unit along with the credit institution to which they belong, the home country principle implies the participation of a bank (and all its branches in other EU countries) in the home country DGS. As a consequence, an effective crisis management basically depends on timely information sharing and cooperation between all the authorities in the financial safety net. It is worth noting that such needs become wider and more complex when dealing with cross-border and systemically relevant banks, as well as with banks benefiting from a supplementary coverage provided by the host country DGS (topping up). The increasing cross-border integration in the European banking system underlines the importance of having an appropriate division of responsibilities between home and host deposit guarantee schemes, since large movements of guaranteed deposits may be reported from one DGS to another as a result of specific EU regulations 43. Such problems have been recently highlighted in relation to the adoption of the European Company 44 (SE) Statutes by a well-known Scandinavian bank that, changing its subsidiaries into branches, caused them to exit the host countries DGS and enter the scheme of the country where the SE main office is located. As a consequence, and consistently with the home country principle, responsibilities for supervising such branches are not assigned to the host Member State authorities anymore, but they are given to the 43 As regards the Statutes of the European Company for example (see below). 44 The European Union adopted the Statutes of the European Company (Societas Europaea - SE) in October 2001; the regulation entered into force by the end of The creation of the Statutes means in practice that companies established in more than one Member State may merge and operate through the EU on the basis of a single set of rules and a unified management and reporting system, avoiding the need to set up a costly and administratively time-consuming complex network of subsidiaries governed by different national laws. Once formed, the SE will be able to operate on a Europeanwide basis and be governed by community law directly applicable in all Member States. References to two pieces of legislation: Council Regulation (EC) No. 2157/2001 of October 8 th, on the Statute for a European Company; Directive 2001/86/EC of October 8 th 2001 complementing the Statutes for a European Company with regard to the involvement of employees at the European Company. Up till now an official register for European Companies does not exist yet; 14 medium and large size companies have adopted the Statutes, all belonging to the North European area (Austria, Germany, Sweden, The Netherlands), among which are 4 Shell companies, or rather an empty legal packaging under construction. The proceedings are still incomplete in some countries, even though the transitory period ended on October 8 th Actually, even in countries where the Statutes have been formally implemented, as in Italy, there are some aspects missing in the fulfilment set of rules and such situation prevents an SE from establishing its headquarters in the country. 35

36 home country ones instead. It is also argued that the transfer of a substantial amount of guaranteed deposits between DGSs could give rise to repercussions in both home and host country systems, as far as regards interventions in crisis times, cooperation and information sharing. Relating to the current regime of responsibilities, the feasibility of making the host country DGS entirely responsible for the guarantee of deposits held by cross-border branches operating in its territory could be in theory investigated, mainly for consumer protection needs; in practice, this change would require a non convenient reversal of a fundamental principle on which the Single Market in general, and the Internal Market for financial services in particular, are based. The question of whether the current mix of homehost responsibilities is appropriate or not, is open, with all relevant implications on information sharing needs. Along with the responsibilities issue, other discussions have been carried out since the beginning of the revision activities, namely those relating to possible new provisions on entry-exit 45 rules and on the transferability of contributions between like schemes, up to the hypothesis of creating a specific DGS for systemically significant banks. As far as regards the transferability of contributions, it has been argued that such contributions should not be considered something like the property of a bank, to be withdrawn from a fund and transferred to another just in the event of the bank s head office moving to another country; it seems to be more appropriate if relevant schemes would sign an agreement in order to make available to the new DGS a portion of the contribution paid by such bank to the old scheme. The matter is still open. Apart from the specific strategy that the Commission will decide to adopt at the end of the review process, cross-border operations and related implications are definitely a new topic for the Directive on deposit guarantee, since they were not significant when the Directive was adopted in Careful analysis of every aspect is therefore needed, since it definitely involves interrelationships between all safety net players. 45 Entry-exit rules, where provided, are usually very restrictive and paid premiums are generally not reimbursed. This is a very important factor in cross-border operations. 36

37 It is worth considering a further development of the analysis on the matter of responsibilities, in order to appropriately provide for the need of more detailed information sharing agreements between a supervisory authority and DGSs in the same system, as well as between Member States. Such purpose should be achieved apart from the fact that the underway revision activities and the consultation in itinere may or may not finally support substantial amendments to the current framework. 37

38 2. International Overview Before expanding the analysis to the specific features of each deposit guarantee scheme (DGS) involved in the research, it is worth outlining a more general panorama through some statistical elaborations on the collected data 46. The following graphs help compare the main DGS features, two thirds of which have been based on EU principles. This international overview is based on data relating to 42 deposit guarantee schemes, which correspond to 37 countries (25 EU Member States and 12 non EU countries) since more than one DGS exists in some EU countries 47. The first aspect analysed in Chart 1 is the legal form of the scheme, which is private in 27% of the sample, public in 57% and joint in the remaining 16%. Joint forms may be the case for schemes initially established as private organisations on a voluntary basis and later reformed as public and mandatory bodies, often in the process of implementing the EU Directive. Chart 1 Nature of the Scheme 57% 27% 16% Private Public Mixed Deposit guarantee is managed by a separate entity in 89% of the sample, while in 11% of cases it is managed by an already existing organisation, usually the Central Bank or the Banking Association (Chart 2). 46 Unless othervise specified, all data used in this statistical overview refer to December 31 st Note that levels of coverage/payout limits are expressed in euro according to the relevant exchange rates as of the end of In terms of EFDI membership, 4 schemes are in Austria, 6 in Germany, 3 in Italy and 3 in Spain. Some of them are not included in the analysis due to missing information. 38

39 Chart 2 DGS as Independent Institution 89% 11% No Another important topic is whether or not the scheme may benefit from public contributions. Chart 3 highlights how the vast majority of DGSs does not rely on public funds (81%), while a small 19% may get resources from the state. This information does not contradict the previous data on the DGS nature, since it just means that also public schemes collect resources from member institutions for both operating expenses and interventions. Chart 3 Public Contributions 81% 19% No Although public contributions are generally not provided, the banking system might face a shortcoming of resources when dealing with interventions of big banks or in case of multiple crises arising simultaneously; for this purpose, supplementary guarantees from the Government or the use of extraordinary public funds may be provided as well, in order to prevent the crisis or crises from also involving the other banks in the system. 39

40 Depending on the role a DGS plays in the financial safety-net, as well as on the relationships with the other authorities within it, supervisory and regulatory responsibilities may or may not be assigned to the scheme. Usually DGSs are not given supervisory powers (81%), as highlighted in chart 4; only 19% have this sort of powers on member institutions, mostly regarding non-separate entities so that deposit guarantee functions are managed directly by the Supervisory Authority. Chart 4 Supervisory Power 81% 19% No This information on supervisory powers may be read together with what Chart 5 shows, where 59% of the schemes have no independence with respect to intervention decisions, meaning that interventions are usually subject to the authorisation of the Central Bank or another supervisory authority. Chart 5 Independent Intervention decisions 59% 41% No In line with the two previous features, chart 6 investigates whether or not DGSs use monitoring systems of each to assess member bank s level of risk, often targeted to adjust 40

41 contributions on the basis of the resulting risk level. Most of the schemes in the survey (62%) do not use any monitoring system, while 38% analyse members risk profile and generally apply risk-based contributions as well. Such percentage relates to 16 DGS established in 10 EU Member States and in 3 non EU countries. Chart 6 Monitoring system and/or Risk-based contributions 62% 38% No DGSs are usually distinguished on the basis of the funding mechanism they use. Among the 37 countries participating in the survey, 70% build and manage a fund that is promptly usable to cover losses as they arise (ex-ante schemes); conversely, 14% have only virtual funds, i.e. members are asked for contribution only after the intervention s decision is made (ex-post schemes); finally, the remaining 16% of the sample rely on a mixed form of funding. All non EU countries have ex-ante funded DGSs, although in one case the scheme has been very recently reformed so that it would be better defined as mixed; as far as the 25 EU countries, there are ex-ante DGSs in 14 Member States, ex-post DGSs in 5 and mixed funding in the rest. Chart 7 Funding Mechanism 70% 14% 16% Ex-ante Ex-post Mixed 41

42 In addressing EU Member States, it is worth analysing a topic related to cross-border operations and cooperation between DGSs, namely Bilateral Agreements (BA) concluded for topping up purposes. In fact, there are relatively few topping up agreements in place and substantially no data on depositors payoffs under such agreements, upon which an analysis may be made. As Chart 8 shows, 27% of the sample signed a BA, relating to 10 of 25 EU countries. Underway agreements are not included in the graph. Note that the chart has used data from the entire set of 37 countries, including EU members as well as the 12 non EU EFDIadhering countries where the Directive itself has not for the most part been implemented yet. Chart 8 Bilateral Agreement 27% 73% No Before concludig this international overview, a comparison between different DGS levels of coverage and payout limits is presented. Results are shown in two charts since countries are divided into two groups: EU Member States and non EU countries. Note that figures shown in both charts are payout limits rather than coverage levels; of course, the two amounts are equal when coinsurance is not present. Average and median values are calculated taking into account the level of coverage provided for in each country, or the payout limit if coinsurance is applied. As far as the first group (Chart 9a), the average level of guarantee in the EU Member States is around euro, not much over the minimum harmonised level ( euro) provided for in the Directive. The median is equal to euro. 42

43 Only Estonia, Latvia and Lithuania provide coverage below the EU minimum, since they recently joined the European Union (on May 1 st 2004) and there are transitional arrangements underway to enable compliance with that figure by the end of Chart 9a Level of coverage - payout limit in EU-25 (in euro, as of the end of 2005) Unite d Kingdo m The Netherlands Sweden Spain S lo ve nia Slovakia P ortugal P o land Malta Luxembo urg Lithuania Latvia Italy Ireland Hungary Greece Germany France Finland Es to nia Denmark Cze c h Republic Cyprus B elgium A us tria Note that the Germany s level of protection in the chart ( euro) is that granted by mandatory schemes established while implementing the Directive; nevertheless, previously existing voluntary systems guarantee nearly complete coverage to the depositors of their member banks, as explained further in this Paper. In non UE countries (Chart 9b) the average level of protection granted is euro, with Norway excluded from such calculation since it provides an extraordinarily higher level of coverage (around euro) that might generate inconsistent results. If Norway were included, this average value would be euro. Similar remarks may be made regarding the median level, which is euro without Norway and up to euro with Norway. 43

44 Moreover, it is worth highlighting that few changes recently occurred as far as regards Bulgaria, Romania and Serbia. Bulgaria and Romania are in the process of joining the EU; they shall become EU Members on January 1 st As a result, their coverage levels changed in and they are going to be raised up to the EU minimum level by the end of Serbia significantly increased the level of coverage, which was very low at the end of 2004 (around 62 euro), up to the equivalent of euro since July 1 st Note also that in Armenia a different level of coverage exists, depending on whether deposits are denominated in drams rather than in foreign currency (around euro and euro, respectively). Chart 9b Level of coverage - payout limit in non EU countries (in euro, as of the end of 2005) Turkey S e rbia Romania Norway Montenegro Mace do nia Iceland Croatia B ulga ria Bo s nia and Herzego vina Arm e nia A lba nia As noted above, level of coverage may differ from payout limit due to the presence of coinsurance. Coinsurance is the mechanism by which depositors are not fully covered up to the level of coverage provided for in the national legislation; instead, depositors suffer losses on the basis of a certain percentage of their deposits. According to the Directive, the extent to which depositors can be exposed to losses in the event of a crisis may not exceed 10%. In the EU, such a mechanism is applied in 48% of countries (12 out of 25). 44