10 Financial Integration and Stability
- Jörg Decressin, Wim Fonteyne, and Hamid Faruqee
- Published Date:
- September 2007
The integration of financial markets in the European Union poses evolving risks to financial stability and challenges to existing financial stability arrangements. Chapter 2 points out that financial integration carries risks of contagion and fundamental spillovers, as well as transition risks. Chapters 6–8 show that increasing integration has not diminished banking system risks because diversification gains have been offset by higher risk taking. Moreover, integration has increased banks’ exposure to EU-wide or euro-area-wide financial cycles and has increased the role of large, complex cross-border and cross-sector financial institutions, which can be an added source of systemic risk. As experience has shown, including in the European Union, financial crises can be extremely costly both fiscally and in terms of economic growth. Although full-blown crises may be unlikely, the costs of preventing or containing crises can be very substantial. Two key questions therefore are: How should cross-border risks be managed to ensure financial stability in an integrating financial market? And how should the EU financial stability framework be adapted to do so?
Issues relating to financial integration and the financial stability framework are intertwined, but reconciling them is a complex undertaking, particularly against the backdrop of complicated decision-making mechanisms and deeply entrenched legal and institutional legacies in the European Union. At a minimum, the financial stability framework ought not hamper financial integration. This means it needs to be consistent with a level playing field1 and must allow financial institutions to operate efficiently along business lines rather than national boundaries. In addition, there should be no artificial obstacles to cross-border mergers and acquisitions.
Although the global trend toward financial integration is straining financial stability arrangements everywhere, the challenges facing the European Union are unique because the European Union has set the objective of creating a fully integrated market; regional financial integration is proceeding faster than elsewhere; the supranational EU structure allows for a variety of solutions; and past initiatives, in particular the single passport, have created a unique basis for further integration. In this context, the European Union can have an important laboratory function from which other regions may benefit.
This chapter reviews the European Union’s current financial stability arrangements, identifies remaining challenges, and outlines some basic considerations and options for charting a course ahead. It finds that, notwithstanding major progress in the areas of regulation and supervision and the existence of arrangements for supervisory cooperation in normal times, there is not at present a pan-European framework sufficiently robust to deal with a major cross-border crisis, for instance, a crisis involving one or more of the European Union’s 46 largest LCFIs.2 To address the existing problems at their roots, aligning authority, responsibility, and accountability for financial stability will be essential, and reforming the financial stability architecture accordingly will require putting in place a number of fundamental foundation blocks. A comprehensive, multi-pronged approach might be best able to deliver the necessary reforms.
The chapter first discusses some basic requirements and challenges involved in building an EU financial stability framework and then outlines the main pillars on which such a framework should be built, reviewing what is now in place and what is missing. The chapter then discusses the need to chart a course for reform, touching upon process issues, fundamental obstacles, and some basic options for reform of the institutional architecture. The concluding section attempts to crystallize such a course for reform.
Building an EU Financial Stability Framework
Maintaining financial stability involves a complex set of institutions and arrangements. These include prudential authorities—regulators and supervisors—but also monetary and fiscal authorities. The working relationships among these authorities are also important. There must be mechanisms to prevent, manage, and contain crises; allow the orderly exit of failing institutions; and protect clients. The fiscal authority (the treasury) needs to be involved in financial stability partly because of its role as “solvency provider of last resort” but also because of its political responsibilities. The central bank’s involvement is crucial, even when it has no prudential responsibilities, because of its role as “liquidity provider of last resort.” Moreover, the actions of fiscal and monetary authorities affect the condition of the financial sector. Their involvement in financial stability monitoring can help raise awareness of the linkages between macroeconomic policy and financial stability and can also increase the effectiveness of public intervention in crisis situations.
An EU financial stability framework needs to be able to deal with the full spectrum of relevant risks, including systemic and cross-border risks. It also needs to operate at minimum cost to taxpayers and financial institutions, both in its regular functioning and in dealing with crises. This requires, among other things, rapid decision making and rapid implementation of remedial measures. Furthermore, an EU stability framework should not create additional obstacles to market integration or curtail market forces.3 A sustainable EU framework also requires accountability mechanisms that are robust over time (including during crises) and that foster national political acceptance for cross-border remedial and intervention actions.4
Notwithstanding major progress, a number of underlying factors complicate both the functioning and the further development of EU financial stability arrangements.
Lack of a Common Philosophy on Financial Stability
There has been some convergence in international thinking about financial stability over the past few decades, due in part to the emergence of international best practices,5 increasing cross-border contacts between decision makers, and the globalization of academic thought. Yet there is still no full common understanding within the European Union on the basic tenets of financial stability policy and basic elements such as the definitions of financial stability and systemic importance. The absence of a common philosophy shows up in EU countries’ positions in various debates, including “rules versus principles”; dealing with failing banks, hedge funds, and complex financial instruments; and the future of the EU financial stability framework. These differences have deep historical roots and often originate from divergent views on the respective roles of the state, laws and regulations, and market forces.
Diverse Institutional and Legal Legacies
Each of the 27 EU member states has its own institutional and legal framework, shaped by the country’s history, other institutions, past policies toward the financial sector, and the above-mentioned philosophical differences. Prudential authorities are set up differently and have different powers and different liability and accountability arrangements. For example, in some countries, the supervisory authorities can be sued if a bank fails; in others they are immune. In 13 of the 27 member states, a unified prudential authority has been established to oversee banking, securities markets, and insurance.6 In the other member states, these responsibilities are divided between specialized agencies (the central bank often being one of them) that deal with particular sectors and/or particular functions (regulation, supervision, licensing, conduct of business). In addition, the fundamental differences in member states’ legal systems, including in particular the civil laws governing financial products, have implications for the financial stability framework, for instance, the rules on mortgages or securities of claims (see also Chapter 1).
Institutional Mismatch between LCFIs and Prudential Authorities
Challenges to financial stability in the European Union arise at three basic levels: (1) financial institutions (including LCFIs), (2) individual countries, and (3) the integrating EU market (with the euro area being a subelement of this level). Before integration, institutions were at the lowest level, presented limited risks, and were relatively straightforward to deal with. They were active almost exclusively within their home country; institutional stability was supervised by a single national supervisor; and there was a risk of higher-order, country-wide financial instability only in cases involving systemic institutions or multiple institutions simultaneously.
The emergence of cross-border LCFIs has changed this. Such groups can be systemic in multiple countries, and their stability (or lack thereof) can become an issue even for the European Union as a whole. Moreover, these cross-border financial groups organize themselves increasingly without regard to national boundaries and legal structures, as they seek to minimize costs, allocate capital flexibly, and turn the integrated financial market into a business reality. They tend to centralize liquidity, risk, and asset-liability management and to ignore or downplay the distinctions between branches and subsidiaries in their day-to-day operations.
By contrast, the European Union’s basic prudential institutional setup remains country based and country oriented,7 and prudential authorities remain bound by the legal distinctions between branches and subsidiaries. This inherent dichotomy makes it challenging to deal with financial stability in cross-border groups and at the EU level. Furthermore, the prudential authorities’ ability to conduct effective oversight is constantly challenged by the fluidity of the banking landscape, which can shift responsibility for oversight from one jurisdiction to another with each change in corporate control.8
Misalignments of Authority, Responsibility, and Accountability
The European Union has sought to cope with this institutional mismatch by shifting responsibility for the regulation and supervision of cross-border financial groups to the home country, that is, the country in which the top entity in the group is licensed. However, this shift has not been complete and has introduced new challenges. Most important, the control that home-country authorities have gained over the foreign operations of LCFIs has not been accompanied by a commensurate degree of responsibility and accountability for financial stability in the host countries in which those LCFIs operate. For their part, host-country authorities may no longer have meaningful control over foreign institutions active in their market, but they remain responsible for financial stability within their borders.
This complicates financial stability management in countries that are host to a significant foreign bank presence. For such countries to use prudential tools in response to perceived threats to financial stability would require a level of cooperation from (perhaps multiple) foreign authorities that is difficult to achieve in practice. No formal mechanisms are in place to bring together all supervisors who have some degree of control over financial stability in a given country or to hold home-country agencies accountable. These tensions are illustrated in part by the difficulties facing some new member states as they attempt to design effective policy packages to contain the risks related to credit booms fueled in part by foreign banks.9
This mismatch between authority and responsibility at the institutional level implies potential differences of opinion with respect to the activities of a branch or subsidiary. The potential is greatest when dealing with LCFIs, which may have activities of systemic importance in some countries but not in others, and in crisis situations, when the country authorities with the greatest control over the situation (home-country authorities) do not necessarily face the greatest potential adverse consequences. In this context, it is only rational that host-country authorities seek to exercise as much power over foreign institutions as the rules allow, or even seek a rebalancing of the rules in favor of host-country control.
There are a number of committees at the EU level that bring together prudential authorities (Box 10.1).10 Nonetheless, no single institution or structure has the authority, responsibility, and accountability to safeguard EU financial stability. The focus of prudential authorities is clearly on the country and institution/group levels.
Complex Decision Making
Decision making, both in crisis situations and in general matters related to financial stability, is complicated by the potentially large number of actors and their divergent interests. Depending on its nature and scope, a crisis may involve central banks, other regulatory bodies, and ministries of finance, possibly from multiple countries. It may be difficult and slow to build consensus among the involved parties on basic issues such as the appointment of a lead crisis manager, appropriate restructuring or resolution measures, or the maximum acceptable cost of an intervention. In noncrisis times, this cumbersome decision making limits the adaptability of financial stability arrangements, and therefore hampers authorities’ ability to deal with rapidly evolving risks in financial markets.
One counterweight that does facilitate cooperation among EU authorities, particularly in times of crisis, is the personal relationships they have built through almost three decades of working together to assemble a financial stability framework. There are open channels of communication among supervisors and between supervisors and other decision makers. Nonetheless, against a background of incentives that have not been fundamentally aligned, personal relations may not guarantee adequate cooperation during a crisis.
Decision making is also complicated by the divergent interests of regulated entities. Regulatory capture is common in the financial industry, given the complexity of regulations and the resulting need for extensive consultations, as well as the industry’s resources and economic importance. Large institutions in particular are able to shape national prudential policies (Hardy, 2006). Because their interests are not always aligned across countries, the influence of these large institutions can make decision making more difficult. Another fault line runs between small, domestic institutions (which carry weight because they are numerous) and large, internationally active groups. Only the latter tend to have a strong interest in streamlined financial stability arrangements across countries.
Hurdles Facing the Financial Integration Process
Some aspects of the current financial stability arrangements generate or allow friction that can slow integration. Country-based responsibility and accountability structures and the decision-making challenges noted previously make it an uphill struggle to level the playing field for market participants across the European Union (see the discussion in this chapter on “Regulation and Supervision”). In addition, as discussed, host countries face reduced control over financial stability at the national level and have no means to hold home-country decision makers to account. In this context, national prudential authorities are wary of relinquishing supervisory control over their financial systems. Moreover, concerns about the competitiveness and interests of their own financial markets, institutions, and economy lead many national authorities to favor national control over their national financial system.11
Box 10.1.Key Bodies in the EU Banking Sector Stability Framework
European Banking Committee (EBC): Consists of high-level representatives of the ministers of finance of member states and is chaired by the European Commission. The ECB, the chair of the Committee of European Banking Supervisors (CEBS), and (optionally) national central banks may participate as observers. The EBC is a “Level 2” Lamfalussy Committee (see Box 10.2) that advises the European Commission on policy issues related to banking activities and on commission proposals in the banking area.
Committee of European Banking Supervisors (CEBS): Comprised of representatives of supervisory authorities and central banks, including the ECB, although only supervisory authorities have voting rights. Although the focus of CEBS, as a “Level 3” Lamfalussy Committee, is mainly on regulatory and supervisory convergence, it also plays a role in promoting supervisory cooperation and as a conduit and organizer for the exchange of information between supervisors on individual financial institutions, including in distress situations.
European Central Bank (ECB): The ECB’s main role in financial stability is monitoring, in cooperation with national central banks and supervisory agencies. It publishes an annual report on “EU Banking Sector Stability” and a twice-yearly Financial Stability Review for the euro area (both documents are prepared with the Banking Supervision Committee). It also advises on financial rulemaking within the Lamfalussy structure and participates in the Basel Committee on Banking Supervision, EBC, and CEBS (observer status).
Banking Supervision Committee (BSC): Brings together national central banks, banking supervisory authorities, and the ECB. The BSC plays a key role in the preparation of supervisory Memoranda of Understanding among EU supervisors. It also performs macro-prudential and structural monitoring of the EU financial system and analyzes the impact of regulatory and supervisory requirements on financial stability. Preparatory work is performed in four working groups: macro-prudential analysis, structural developments in the EU banking sector, crisis management, and credit registers.
Economic and Financial Committee (EFC): Comprised of deputy ministers of finance, the European Commission, ECB, and central banks. It provides high-level assessments of developments in financial markets and services and advises the Economic and Financial Affairs Council (ECOFIN) and the European Commission.
Financial Stability Table (FST): The EFC meets twice a year (in April and September) to discuss financial stability issues in a special configuration as the “Financial Stability Table,” in a group including the BSC and the Level-3 Lamfalussy Committees—CEBS, the Committee of European Securities Regulators (CESR), and the Committee of European Insurance and Occupational Pensions Supervisors (CEIOPS). The discussion of banking issues is based primarily on ECB reports, including its Financial Stability Review, and on ad hoc input from CEBS. The FST brings together the broadest group of actors in matters of financial stability (prudential, monetary, and fiscal authorities) and is the most likely forum for agreement on any form of policy coordination.
Financial Services Committee (FSC): Previously labeled the Financial Services Policy Group, the FSC is composed of representatives of the ministries of finance and the European Commission, joined by a representative of the ECB and the chairpersons of the Level 2 and 3 Lamfalussy Committees as nonvoting observers. The FSC discusses and provides guidance on cross-sector strategic and policy issues, especially technical and political aspects, and assists the EFC in preparing ECOFIN meetings.
Financial Conglomerates Committee (FCC): Created by the Conglomerates Directive (Directive 2002/87/EC), the FCC provides guidance to EU supervisory authorities on the implementation of conglomerate supervision.
There remain opportunities for prudential authorities to act upon these concerns in ways that run counter to the objective of a single financial market, despite efforts to limit the scope of such activities. For example, some authorities have discouraged certain forms of integration, most visibly market entry by foreign players.12 Less visibly, some have resisted hosting large-scale, foreign, branch-based operations or the optimization of cross-border banking operations within a single banking group.13 There have been cases where privatizations were delayed, ostensibly out of concern for the financial-stability implications of a shift toward foreign ownership. Resistance may take the form of moral suasion, which does not constitute a basis for a formal complaint. In this context, it will remain a difficult balancing act to reduce the scope for using financial stability policy tools to pursue unrelated objectives while safeguarding the ability of national authorities to effectively act on legitimate financial stability considerations in ways that are consistent with the single market.
Pillars of Stability
An effective financial stability framework should comprise five pillars (De Rato y Figaredo, 2007): (1) sound financial regulation; (2) efficient and effective supervision; (3) effective tools for corrective and preventive action and crisis management; (4) crisis-resolution mechanisms for dealing with failing institutions and related losses; and (5) a robust financial infrastructure. The optimal design of each of these pillars depends on the design of the others. Notwithstanding the challenges discussed above, there has been important progress in developing these pillars, both at the national and EU levels, particularly in the areas of regulation, supervision, and infrastructure. This section discusses progress in four of these pillars, although not strictly on a pillar-by-pillar basis.14
Sharing Information and Monitoring Financial Stability
The national foundations of prudential arrangements imply that, even though it is based on an increasingly harmonized accounting and reporting framework, information about the European Union’s financial system is collected locally, using different methodologies. There is no centralized store of prudential information, in part because of national confidentiality rules. As a result, information about a cross-border financial group is dispersed among the various national agencies involved in supervising that group.
Information-sharing arrangements have been put in place, but they fall short of fully addressing the problems, in part because they do not seek full information sharing.15 Supervisors lack access to each others’ source data, and they may be unaware of the existence of certain relevant data and therefore not request them. In this context, information asymmetries are inevitable, leaving no single authority with a full overview, at any given time, of all the risks within a group. Potentially most damaging, supervisors are in a position to control information during crisis situations, to the potential detriment of other involved parties. In fact, microeconomic research suggests that supervisors have strong incentives to withhold information in a crisis.16 In addition, at the EU level, no institution or committee has routine access to a comprehensive set of quantitative and qualitative supervisory information for the largest systemically important cross-border groups.17
Arrangements for information exchange within a decentralized structure will always be less effective than having all information centralized in a standardized format, so that it is subject to a full overview by the proverbial single pair of eyes. Most critical is the need to provide all relevant authorities, including the ECB, greater access to both quantitative and qualitative supervisory information on Europe’s LCFIs.18 The creation of such a centralized repository would offer major advantages and open up many possibilities: it would enhance both routine monitoring of financial stability and decision making under crisis conditions; it could greatly simplify the current cumbersome logistics of information reporting and sharing; it could become the basis for a framework for crisis management and resolution; and it would allow the establishment of a European group of off-site analysts to manage, analyze, and exploit the database. Ideally, current national confidentiality arrangements, which are incompatible with such a centralized repository, could be replaced by a full European confidentiality regime, covering all prudential authorities and allowing appropriate data protection and a free flow of information among prudential agencies.
Greater data availability would also facilitate the further evolution of the forums and structures that have been put in place to monitor EU and euro area financial stability, not only to improve this monitoring but also to facilitate coordinated action. At present, the bodies that could identify threats to financial stability, such as the Banking Supervision Committee (BSC) or the Financial Stability Table (FST), are not mandated to take action.
Regulation and Supervision
Legally, regulation and supervision remain national responsibilities. However, long-standing efforts toward harmonization and convergence have given both an increasingly European character. Since the late 1970s, the essence of financial regulation has been shifting toward the EU level, as a series of directives19 and a number of other instruments created a binding framework for national prudential regulation across the European Union.20
Directives are put into effect by being transposed into national law. In transposing the directives to their national regulatory systems, member states are legally bound to remain within the boundaries of the directives’ provisions. However, directives often leave significant scope for countries to impose stricter requirements on their domestic institutions, thus precluding a fully level playing field.
The European Union’s Financial Services Action Plan (FSAP, see Chapter 3) took legislative and regulatory harmonization to new levels, in particular with the Capital Requirements Directive (CRD) for banks and investment firms, the forthcoming Solvency II Directive for insurance companies, and the Markets in Financial Instruments Directive (MiFID) for financial markets.21 The CRD, which implements Basel II in the European Union, requires member states to limit some of the national discretion allowed by Basel II, thereby promoting regulatory convergence. This directive also consolidates many of the provisions of the earlier prudential directives and key provisions of the Financial Conglomerates Directive (FCD). Solvency II, which overhauls capital requirements for insurance companies but is not linked to a global initiative such as Basel II, is expected to achieve similar convergence in the insurance sector. Once these three directives are fully in effect, the basic body of regulations for the EU financial system will be harmonized to a significant extent.
The Lamfalussy process put in place a more efficient and flexible EU-level regulatory structure (see Box 10.2 and Chapter 3) that can be used to adapt this body of regulations on an ongoing basis. Both Solvency II and MiFID are Lamfalussy directives, that is, principles-based framework directives that leave detailed regulation to the Level 2 and Level 3 committees, which thereby take on a quasi-regulatory function. In banking, the scope for regulation through the Lamfalussy process is more limited, because the CRD is not a Lamfalussy directive.
The FSAP and the Lamfalussy process together represent a major step toward greater convergence, not only in regulation but also in supervision. By overhauling the regulatory framework, the FSAP created a need for all member states to adapt their supervisory practices simultaneously. At the same time, Level 3 of the Lamfalussy process provides a forum for countries to adapt their national regulatory frameworks and supervisory practices in a convergent way.22 Accordingly, in the banking sector, the implementation of the CRD at the national level is being underpinned by guidance materials developed by the Committee of European Bank Supervisors (CEBS). Combined with the 2005 introduction of IFRS in the European Union, this has allowed significant progress toward harmonization of reporting requirements and formats, a long-standing issue for banks.23
Box. 10.2.The Lamfalussy Process
The “Lamfalussy process”—named after the chair of the EU advisory committee that proposed it, Baron Alexandre Lamfalussy—was established by the EU Council of Ministers in 2001 to facilitate decision making on financial sector legislation and regulation and to achieve faster progress toward harmonization by moving much of the discussion from the political level to “downstream” technical committees. To do so, it established a four-level EU financial rulemaking architecture for each of three sectoral pillars (banking, insurance, and securities). The levels are: legislative (Level 1); technical implementation (Level 2); exchange of information, cooperation, and convergence of supervisory practices (Level 3); and strengthened enforcement (Level 4). (See also Chapter 3.)
The Level 2 committees in essence act as regulators, putting in place secondary legislation with and through the European Commission that constitutes the basis for regulation at the national level. In doing so, the Level 2 committees take into account industry advice delivered through the Level 3 committees. This approach has the added advantage that secondary legislation can be modified relatively rapidly to adapt to changing circumstances, without having to go through the full legislative process. However, to date the role of the Level 2 committees has been relatively limited in the banking sector as, unlike MiFID and Solvency II, the CRD was not devised as a Lamfalussy framework directive. Instead, it contains an extensive regulatory framework established directly by the legislature.
Harmonized regulations are only effective if supervisors interpret and implement them in a harmonized way. Achieving this harmonization of supervisory practices—established practices as well as those related to new laws and regulations—is the main objective of the Level 3 committees. These committees bring together national supervisors and seek harmonization through (1) exchange of ideas and experience, (2) issuance of nonbinding interpretative guidelines and recommendations on regulations, and (3) standard setting in areas not covered by Level 1 or 2 legislation.
Despite Level 2’s limited role in the banking area, the Committee of European Banking Supervisors (CEBS) is the main actor in the effort to achieve harmonized implementation of the CRD. Its work in this respect encompasses, for example, common guidance on the supervisory review process (Pillar 2 of the new Basel Capital Adequacy Framework, or Basel II), as well as guidance for accreditation of rating agencies, guidelines on prudential reporting by banks, validation of internal ratings-based credit risk, and operational risk approaches. Common implementation of Pillar 3 of Basel II is being facilitated through a common framework for supervisory disclosure.1 Beyond the CRD, CEBS established guidelines on prudential adjustments (“prudential filters”) in the context of the introduction of the International Financial Reporting Standards (IFRS), in order to avoid the changes in accounting standards having undesirable effects on prudential indicators.
Since the end of 2005, cross-sectoral cooperation is being developed between the three Level 3 committees, under the label “3L3 work program.” This work focuses on improving and facilitating the supervision of conglomerates and on other issues of common interest.1 Supervisory disclosure, in this context, refers to a “comprehensive policy of transparency.” The basic idea is that transparency about supervisory practices will stimulate convergence of these practices. See CEBS (2005).
Nonetheless, progress toward convergence of regulation and supervisory practices under the Lamfalussy structure is being hampered by the need for unanimity among the 27 or more parties at the negotiating table, especially at Level 3. As a result, agreements tend to turn into compromises that can be far removed from best practice, gravitating in the direction of a collection of national practices that allows supervisors to continue business as usual.24 Similarly, at a higher level, extensive “customization” of harmonizing directives can also take place during the legislative process, as various interests seek to ensure that national customs are incorporated and/or accommodated rather than replaced. Even when harmonized regulations are adopted, the reality on the ground may fall short due to “goldplating,” the addition of national requirements to harmonizing directives.25 In a broader sense, goldplating also likely affects harmonized regulations and convergent supervisory practices adopted at Level 3.
Regulation and Supervision of Cross-Border Activities
In the current country-based but converging setup, the regulation and supervision of cross-border financial activities rely on a set of arrangements that allocate responsibilities between national authorities and organize cooperation between them. The Second Banking Directive, which entered into force in 1992, introduced the principles that constitute the single EU banking passport, namely, minimum harmonization, mutual recognition, and home-country control (see also Chapter 1 and Dermine, 2003). Home-country authorities became responsible for the EU-wide regulation and supervision of institutions they licensed, including their foreign branches and direct cross-border provision of banking services in other EU member states.26 Unlike branches, subsidiaries of foreign banks are seen as local corporate citizens and are therefore subject to local licensing and prudential oversight. A single passport was later also introduced for insurance and investment firms.
Home countries are also responsible for consolidated supervision of cross-border banking groups, which for subsidiaries adds a layer of supervision onto that exercised by host countries. The FCD,27 which applies to a relatively limited number of cross-sectoral financial groups, designated the supervisor of the top entity in such a group as the “coordinating supervisor.” It provides this coordinating supervisor with authority to plan and coordinate supervisory activities for the conglomerate as a whole. However, the FCD also confirms the authority of host-country supervisors to take certain actions.28 For banking groups, the CRD went further, designating the authority that licensed the top or main credit institution in a banking group as “consolidating supervisor”29 and giving it responsibility for coordination and overall supervision over the group. In addition, the CRD requires the introduction of written arrangements for coordination and cooperation between supervisors involved with a group, the exchange of information,30 and consultation in the application of sanctions.31 It provides scope for supervisors to go beyond minimum cooperation requirements, including by giving the consolidating supervisor a more important role and by delegating tasks and powers between home- and host-country authorities.
Solvency II is expected to go significantly further than the CRD in giving power to home-country supervisors. Proposals being discussed at the time of writing would turn the home-country supervisor of an insurance group into a “group supervisor,” who would supervise all the group’s EU branches and subsidiaries in coordination with host-country supervisors.
To implement supervisory cooperation in practice, member states have concluded a network of bilateral Memoranda of Understanding (MoUs) (Box 10.3). These MoUs lay down practical arrangements, including lists of contact persons, procedures for conduct of cross-border onsite inspections, and mechanisms for information exchange. Additionally, for some cross-border institutions, multilateral MoUs have established “colleges of supervisors” in which home- and host-country supervisors meet on a regular basis and discuss supervisory issues with regard to a specific LCFI.32
Box 10.3.Cooperation on Prudential Supervision and Crisis Management
Cooperation between supervisors in the European Union has been organized primarily by means of non-legally-binding Memorandums of Understanding (MoUs). To help implement the principle of a single banking license (and home-country control) introduced in 1992, an extensive network of bilateral MoUs was established between supervisors. In 2003 and 2005, multilateral MoUs were concluded among supervisors, central banks, and also ministries of finance (in 2005) on information exchange during crises. Supervisors have the option to go further than this general cooperation framework and agree to group-specific or regional MoUs.
Cooperation goes farthest in the Nordic countries. The Nordea Group, whose parent company is Swedish, has establishments in all Nordic countries. In terms of lending, Nordea Bank is the largest bank in Finland, the second largest in Denmark, the third largest in Norway, and the fourth largest in Sweden (Sveriges Riksbank, 2003). Supervisory cooperation between the Nordic supervisors is extensive and includes a college of supervisors for Nordea, established by a Nordea-specific MoU. In addition, a multilateral MoU on crisis management is in place (Nordic Central Banks, 2003).
Nordea has plans to transform into a European Company (SE), although these have been held up by tax and deposit guarantee issues. Under these plans, its subsidiaries would become branches of the Swedish parent bank. According to the home-country-control principle, the Swedish Finansinspektionen would become responsible for supervising major parts of the Danish, Finnish, and Norwegian banking sectors. Supervisory authorities from these countries argue that, in this case, it is justifiable for the role and power of the host-country authority to be broader than at present. Against this background, the Nordic supervisory authorities have launched investigations concerning the impact of the restructuring on cooperation among supervisors (Bank of Finland, 2004).
In the Benelux countries, the formation of the Belgian-Dutch Fortis banc-assurance group in 1990 drove the need for consolidated supervision at the conglomerate level. Against this background, the Belgian and Dutch banking and insurance supervisors concluded specific MoUs covering reporting requirements, the location of activities within the conglomerate, and the modalities of information exchange and consultation between supervisors (Banking, Finance, and Insurance Commission, 1996). Information exchange and cooperation on a practical level are considered to be smooth and effective, going beyond the specifics of the MoUs.
A number of EU countries, including the United Kingdom and the Netherlands, have initiated so-called supervisory colleges for the supervision of cross-EU financial conglomerates under their jurisdiction. These colleges bring together representatives of the supervisory agencies of the countries where the conglomerate is active and address supervisory issues concerning the group. Joint supervisory plans are designed and implemented, and valuable contacts are made among the agencies involved.
The merger in 2000 between the Bank Austria Creditanstalt and Germany’s HypoVereinsbank spurred closer cooperation between Austrian and German supervisory bodies. Supervisors from the two countries concluded MoUs in the areas of banking and securities supervision (BaFin, 2004) and work closely together in practice.
The delegation options offered by the CRD and the Solvency II proposals reflect an ongoing trend in supervision of cross-border groups in the direction of more home-country control. There is room for this trend to run further in the current context, for example through greater use of institution-specific MoUs such as those for Nordea and Fortis and an enhanced role of home country supervisors within supervisory colleges. In addition, more extensive use of the European Company Statute would foster a greater degree of branch-based cross-border banking and therefore a de facto streamlining of supervision in favor of home countries.33 This could lead to a situation in which supervision of cross-border EU financial institutions becomes the responsibility of a selected handful of national supervisors.
Despite the important progress made toward improving supervision of cross-border financial groups, significant problems remain. Indeed, the recent “Francq Report” (Financial Services Committee, 2006) identified supervisory convergence and cooperation, the cost-efficiency of the EU system, and cross-border supervision as the main areas in which further progress was needed.34 For various reasons, banks organize the bulk of their cross-border retail operations through subsidiaries rather than branches,35 and so most cross-border institutions face at least two levels of supervision. In fact, a conglomerate with multicountry activities in banking, insurance, and securities could be dealing with dozens of supervisory agencies across the European Union. Current information-sharing and cooperation agreements are not sufficient to overcome the inefficiencies of the present system. For example, a coordinating supervisor under the FCD or a consolidating supervisor under the CRD must continually coordinate the gathering and dissemination of a broad range of information across all the countries and financial sectors in which the financial group is active.36 The reporting burden on the conglomerate or banking group can also be heavy, potentially extending beyond national reporting requirements (notwithstanding the principle that the consolidating supervisor’s first recourse for information should be national supervisors, not the institution itself).37
By itself, the trend toward greater home-country control cannot be expected to resolve all problems. To be effective, voluntary forms of delegation of powers to home supervisors would require the agreement of all host supervisors for a certain group, something for which they currently have few incentives,38 and nonbinding arrangements such as MoUs risk breaking down during crisis times. Ultimately, significant further devolution of supervisory powers to home countries seems feasible only if host-country authorities receive adequate assurance that host-country financial stability considerations will be effectively addressed. Reliance on national prudential authorities to supervise cross-border financial institutions also raises issues of horizontal consistency, given that supervisors in different countries can have different approaches to supervising groups headquartered in their respective countries.
At present, crisis prevention is an underdeveloped area, both in EU member states and at the EU cross-border level. The CRD and other directives provide the baseline standards with which banks must comply and oblige member states to impose appropriate sanctions in cases of noncompliance, but there is no agreed toolkit for early action against problems in banks. Progress toward harmonization is hampered, among other factors, by differences in prudential authorities’ legal powers with regard to shareholders, as well as EU limitations on these powers.39 Member states use different tools, apply different triggers for remedial action, offer different degrees of flexibility in the use of sanctions, and keep a sharper eye on domestic considerations than on those in other member states.
The divergent incentives faced by home- and host-country supervisors when problems emerge with regard to a cross-border group constitute a potential source of tension.40 Home-country authorities may seek to delay providing information or taking crisis-management actions in the hope of avoiding reputational effects, political fallout, and capital losses to the institution.41 Also, home-country authorities using a risk-based approach to supervision may not always have an incentive to keep small host-country authorities informed in a way that meets the host’s needs. Therefore, host-country authorities may seek to retain as much intervention authority as possible.42
An important element in building host countries’ confidence in home-country control is reassurance that home countries will take action early, when the cost of a problem can still be limited. This argues in favor of a harmonized, effective, and transparent system of early remedial action against delinquent, unsafe, or unsound institutions. Such a system would also contribute importantly to a level playing field by promoting “equal remedial treatment” for all EU banking groups.
Crisis management remains essentially a national responsibility, albeit increasingly supported by cross-border arrangements for coordination and information exchange.43 The CRD and the FCD provide a basic framework, but neither directive includes detailed crisis-specific provisions or deals with the nonsupervisory aspects of crisis-management (such as central bank involvement and liquidity support). The FCD prescribes the gathering and exchange of information, but leaves full power to the host-country authorities to use their own crisis-management tools when needed, thus leaving a significant degree of ambiguity as to who has final authority to take expedient action in case of problems. The CRD gives the consolidating supervisor responsibilities for planning and coordinating supervisory actions in emergency situations,44 and obliges the consolidating supervisor to alert all supervisors concerned as soon as is practicable when an emergency situation arises that could jeopardize the stability of the financial system in any member state.
Broad efforts to test and improve crisis-management mechanisms have built on these basic arrangements and developed the nonsupervisory aspects of cross-border crisis management. The second “Brouwer Report” (Economic and Financial Committee, 2001) concluded that the framework of MoUs on supervisory cooperation in Europe was functional but did not adequately address crisis management. Two multilateral MoUs agreed since then have brought significant progress. The first, signed in 2003 between the ECB, banking supervisors, and EU national central banks, sets out high-level principles but focuses primarily on information sharing.45 It gives the home-country authority responsibility for informing other involved supervisors and for making most crisis-management decisions. To address systemic crisis-management issues that may include fiscal burdens, the July 2005 MoU involved ministries of finance along with national central banks, the ECB, and EU banking supervisors. The focus of this second memorandum remains on information sharing, although it also encourages the development of crisis-management tools.
The operational modalities laid out in the MoUs have been tested in a series of “war games” managed by the ECB, which simulated stability shocks to financial institutions. These games identified potential weaknesses in the coordination of responses to a crisis and conveyed lessons for managing systemic liquidity and avoiding breakdowns in payment systems. The games also helped to familiarize decision makers with each other and to establish channels of communication among them. However, the current arrangements have yet to be tested in a real life situation, when real money is at stake and incentives diverge.
Emergency Liquidity Assistance (ELA)
Emergency liquidity assistance (lender-of-last-resort support) remains a responsibility of national central banks, both within and outside the euro area (Eurosystem). There is no clear EU-wide liquidity support mechanism, and arrangements, including collateral requirements, differ across countries. Within the Eurosystem, a national central bank deciding to provide liquidity support needs to inform the ECB and the other central banks ex post for small operations and needs the ex ante consent of the Governing Council of the ECB for large operations. The national central bank and the national treasury carry the risks involved in ELA operations. For cross-border banks, host countries are responsible for ELA, as well as liquidity oversight, for both subsidiaries and branches. These arrangements are based on the facts that liquidity support interacts with monetary policy, that such support may be needed to maintain the stability of domestic markets and payment systems, and that providing liquidity support requires the ability to provide unlimited amounts of money in local currency. Within the Eurosystem, national responsibility is driven by the possibility that ELA operations could generate losses for the central bank that may need to be compensated by the fiscal authorities.46
Current arrangements for liquidity support have some inherent contradictions.47 ELA to branches is the responsibility of the host country. However, the host-country authorities have no direct access to comprehensive supervisory information about these banks and therefore no means to accurately assess the risks involved in a potential ELA operation. In particular, they may not be able to distinguish between liquidity and solvency problems, a distinction that is crucial in liquidity support operations. Moreover, the prohibition on ring fencing may render liquidity support to a branch ineffective and costly, as funds could easily flow back to foreign parts of the group. In these circumstances, host countries may be hesitant to provide liquidity support. Most probably, the head office or interbank market parties would be approached first, or liquidity support would take place only in coordination with the home-country authorities. Liquidity support to local subsidiaries, by contrast, would remain on the subsidiary’s own balance sheet and would benefit local creditors, although here too there are some risks the support would be redistributed across the group unless the subsidiary is ring fenced. Support from the parent would likely be explored as the first option.
Crisis resolution, as discussed here, refers to dealing with one or more failing financial institutions. Depending on the circumstances, resolution requires that the failing institution be returned to health, restructured, or wound down and liquidated. In the latter case, arrangements are needed to allocate losses and compensate insured creditors, particularly depositors in a bank and policyholders in an insurance company.
Resolution of failing financial institutions remains fundamentally a national responsibility, and there are major philosophical differences among EU member states in this area. Some countries have a tradition of preventing all bank failures in order to preserve confidence in the financial system. In other countries, concerns about moral hazard led policymakers to signal that financial institutions should not expect bailouts. Many countries maintain “constructive ambiguity” about their intended approach to bank insolvencies, thus seeking to avoid moral hazard while maintaining flexibility in action. In practice, however, “open bank” support has been common, and most European countries have been reluctant or unwilling to close even small insolvent banks.
There has been only very limited harmonization of crisis-resolution frameworks for banks, and in many countries these frameworks are underdeveloped, which can result in higher than necessary resolution costs. Speed, specialist expertise, and a focus on the interests of depositors and the general welfare are essential for efficient bank resolution. This means that banks should be subject to specific bankruptcy rules, allowing specific solutions and quick insolvency proceedings that cannot be unduly slowed by shareholders (shareholder rights often stand in the way of cost-minimizing resolution proceedings). Ideally, prudential authorities should be able to trigger and control the resolution process (depending on the case, the deposit insurance system, the central bank, and the finance ministry may also need to be involved). Judicial review should be ex post because judicial delays can significantly increase the costs of a bank failure. In EU countries, however, bank insolvency typically falls under general insolvency laws, which vary widely between member states.48 Most countries have some bank-specific provisions and exemptions (Hüpkes, 2000), and in Italy the provisions of the general bankruptcy law apply only insofar as they are compatible with the special rules for bank insolvency outlined in the banking law. However, no EU country goes as far as the United States, which has a fully separate insolvency regime for banks.49 Some prudential authorities appear to lack appropriate powers to take a failing bank out of business in an orderly way or to control the resolution process.
This underdevelopment of national crisis-resolution frameworks risks pushing up the costs of a bank failure. Conventional bankruptcy proceedings are unsuitable to extract the maximum value out of the typical financial assets on a bank’s balance sheet. They usually impose a stay and suspend the bank’s operations. During this period, asset decay occurs (Beck, 2003), positions in financial markets can no longer be managed and lose value (Herring, 2004), and illiquid assets may have to be sold at deeply discounted, distress values. Even under the best of circumstances, the nature of banking (transforming liquid liabilities into illiquid assets) means that banks typically have a liquidation value that is well below their value as a going concern. This tends to make recapitalization and restructuring a less expensive solution than asset-by-asset liquidation, even from a direct-cost perspective.50 However, conventional bankruptcy proceedings do not always provide the flexibility to allow such lower-cost solutions. With respect to LCFIs, these problems are compounded by the focus of bankruptcy law on the rights of claimholders, which may be at odds with the need to take into account broader economic and systemic considerations.
Failures of cross-border banks are handled within national insolvency frameworks, on the basis of the principles established by the 2001 Reorganization and Winding-Up Directive.51 This directive introduced a hybrid universalist/territorialist approach for handling the insolvency of a cross-border financial group:52 the parent company and its EU branches are considered a single estate subject to home-country insolvency proceedings (universalist), whereas subsidiaries are subject to host-country insolvency proceedings (territorialist). The directive also prescribes equality of treatment for claimants from different countries, imposes information requirements, and establishes some limited minimum standards for the winding-up legislation of member states. While legally consistent, the territorialist approach to subsidiaries is increasingly at variance with the operational organization and economic unity of cross-border banking groups and the resulting interdependence between subsidiaries and parent company. Untangling an integrated group into different estates can cause significant loss of value, may limit the options of each of the receivers individually, and would be time consuming. Moreover, in cases of multisector conglomerates, the sectoral segmentation of European insolvency law would hamper a consistent and effective resolution (Hadjiemmanuil, 2005).
Limited harmonization of deposit insurance schemes has been achieved through the Deposit Insurance Directive (94/19/EC), but these schemes still vary widely from country to country, including with respect to the definition of deposits, the amount of coverage,53 co-insurance, risk-based premiums, and funding (see Garcia and Nieto, 2005). This variation affects the incentives of consumers and banks, with implications for both financial stability and financial integration. Depositors may seek the most generous coverage or may seek to avoid dealing with a foreign deposit insurance system. Some arrangements impose more discipline on banks than others, and the variety in arrangements creates uncertainty and cross-country disproportionalities in the costs of bank failures. Moreover, increasingly mobile banks54 have an incentive to seek coverage by generous unfunded, and thus inexpensive, schemes in normal times as well as an incentive to “jump ship” at the prospect of having to contribute to cover the costs of a major insolvency.55 Perhaps most important, EU deposit insurance systems have generally not been designed to deal with systemic problems, including the failure of one or more LCFIs. They lack the capacity; the level of coverage they offer is insufficient to deflect political pressure in favor of a general bailout of depositors if not of the institution; and lengthy repayment periods mean they do not offer depositors liquidity insurance.56
Overall, this implies that there is no cost-efficient way to liquidate an LCFI, even at the national level.57 In light also of systemic, economic, and social considerations, the least-cost policy option available in an LCFI insolvency scenario will likely involve ensuring continuity in much of the operations of the failing institution. However, this may not be possible without a solvency support package that includes taxpayer money or public guarantees.58 Other sources of funds could be tapped, including existing shareholders, a “White Knight” acquirer or new shareholder, other private sector parties that have a stake in the institution’s survival (for example, creditors), the central bank, and a funded deposit guarantee scheme. However, private sector funding will require at least a prospect of profitability and may only address part of the financing need by purchasing specific assets. Central bank funding is generally only appropriate for liquidity support,59 and the use of deposit guarantee funds would need to be strictly circumscribed so as to not endanger the fund’s primary objective of protecting depositors. Moreover, the diminishing leverage of national prudential authorities over internationally active financial institutions is reducing their ability to exert moral suasion to engineer a private-sector-led solution.
The situation would be significantly more complex with the failure of a cross-border LCFI. The basic philosophical differences between countries, information asymmetries, complex decision making, and the territorialist approach to subsidiaries under the Winding-Up Directive would make it very difficult to agree on a resolution strategy that minimizes collective costs. Just as at the national level, cost efficiency may well require public solvency support. However, no decisions have been made at the EU level on how to finance solvency support to a cross-border financial institution. In the absence of cost-sharing arrangements, home-country authorities may be reluctant to use taxpayer money to the benefit of host countries or may lack the capacity to do so: banking groups that are “too large to fail” can also be “too large to save” due to the high costs relative to the home country’s resources.60 Without home-country assistance, host countries would have to decide whether they want to bail out the local operations and carry the associated costs themselves. However, host countries may not be able to restore their local branch or subsidiary as a going concern, because the subsidiary may not necessarily constitute an operationally and financially independent unit. Moreover, the assets and liabilities of a branch would be under the receivership of the home-country authorities and would have to be purchased from the estate to make continuation possible. Liquidity and solvency support by the host-country authorities may be impeded or ineffective because ring-fencing may take time or may be impossible.61 By the time the appropriate safeguards are in place, the support may come too late to allow a cost-minimizing resolution process.
In this context, the incentives of the different national authorities diverge when problems emerge. Their incentives are geared toward minimizing the costs to their own treasury and citizens by maximizing a troubled bank’s assets and minimizing its liabilities in the country. Hence, a resolution process could result in a “chaotic scramble for assets” (Herring, 2003). In the absence of ex ante cost-sharing mechanisms, incentives for cooperation will be determined by countries’ initial estimates of the costs for their jurisdiction. Countries that expect “their” costs to be relatively low, for instance because the local establishment is solvent, will not be interested in participating in a collective solution that would be more costly to them. Valuable time can be lost trying to reach an agreement on cost-sharing, given the practical, political, informational, and other constraints that would present themselves in an LCFI failure, and the cost of resolution will increase accordingly.62 Time can also be lost obtaining the necessary clearance from the EU Commission, as public solvency support to a failing bank amounts to state aid under EU competition rules.
Moral hazard is an inherent problem under current arrangements. In the absence of an operational framework to handle the failure of a (cross-border) LCFI in a cost-efficient way that imposes losses on shareholders and uninsured creditors, there is a widespread perception among market participants and regulators that a troubled LCFI would receive some form of direct or indirect solvency support,63 and that such support is likely to also benefit shareholders and/or uninsured creditors.64
Charting the Course Ahead
“If one does not know to which port one is sailing, no wind is favorable.”
There is general agreement that Europe’s financial stability framework needs to be adapted, but views differ considerably on priorities as regards substance, process, and sequencing. The fundamental point of debate is the optimal degree of centralization of financial stability oversight. Proponents of a decentralized approach argue that proximity is important and that responsibility for financial stability should rest with the national fiscal authority that must pick up the bill when things go wrong (an argument cited by Goodhart, 2004b, and Goodhart and Schoenmaker, 2006). The main counterargument is that financial stability should be managed where the risks lie, and this is increasingly at the cross-border level. To make this possible, some view a move toward joint fiscal responsibility in the form of a burden-sharing agreement to be the cornerstone of the reform process.65 Others consider discussion of burden sharing to be premature, raising moral hazard and potentially forestalling more practical steps forward. A variety of more ambitious reform proposals have been floated outside policymaking circles.66
In the 2005–10 White Paper, the European Commission, with relatively broad support among market participants and policymakers, opted for a gradual, evolutionary reform process. The main arguments in favor of this gradual approach are that the full potential of the FSAP and the Lamfalussy process must be exploited, that the full impact of these initiatives will be evident only with time, and that there is a certain degree of reform fatigue among authorities and market participants alike. More generally, a gradual approach makes agreement more likely to be reached, reduces the risks of large mistakes, and provides more continuity.
On the other hand, the gradual, evolutionary approach may be too slow to keep up with the integration of markets. Moreover, the absence of agreement about basic principles and about “the port to which one is sailing” increases the risks of inconsistencies, gaps, and misconceived reforms.
Fundamental questions also remain about how worst-case scenarios would be handled and who would pay the bill. Reforms that risk putting national authorities at a disadvantage or increase uncertainty with respect to these worst-case scenarios may be impossible. Finally, financial crises may not await the completion of the evolutionary process.
In view of these limitations and the critical challenges of reform, rather than wait for the completion of the evolutionary process, it may be preferable to adopt a reform strategy that provides a greater sense of direction, facilitates coherence, and allows faster progress. Such a strategy might grow out of a multipronged approach, involving: (1) identifying and analyzing remaining gaps, such as frameworks for remedial action and crisis resolution, and urgently pursuing work on addressing them; (2) debating and seeking a convergence of views on fundamental “philosophical” issues with respect to basic objectives and principles of a financial stability policy, including in such areas as supervisory powers, crisis resolution, protection of depositors, management of systemic risk and systemic crises, and bank shareholder rights; (3) considering all the options, even if such options seem infeasible in the short to medium run; (4) continuing ongoing efforts to make progress in convergence of regulation and supervisory practices, improve cross-border cooperation, and increase the efficiency of the existing arrangements; and (5) seeking to identify and address fundamental obstacles that hamper the reform process itself. In its implementation, such a multipronged strategy could be pragmatic and build on existing efforts. Yet, by fostering debate about fundamental and long-term issues, it could provide a sense of direction.
Consensus on faster and more far-reaching reforms will not be easy to achieve, especially when they touch upon the sensitive issue of the institutional architecture. It may help to focus the debate in the first place on what the financial stability framework needs to deliver, namely financial stability maintained in a cost-efficient and integration-compatible way, and only in second order on what kind of institutional reforms may be needed to make this possible.
Addressing Fundamental Obstacles
Pending the construction of the common stability framework, the problems caused by misaligned incentives need to be reduced. This may require a dual approach. On the one hand, the incentives of prudential authorities need to be realigned, toward a more EU-wide perspective. For example, they could be given a European mandate requiring them to work toward common EU objectives, such as minimizing the collective costs of a crisis.
Other steps in this direction could include the removal of potential sources of mistrust; extensive cross-border dialogue through various channels and forums; and work toward a common financial stability philosophy. On the other hand, the potential impact of misaligned incentives needs to be reduced through sound institutional and procedural arrangements, ex ante agreements on core issues, and extensive information sharing to minimize information asymmetries. Decision-making procedures that do not require unanimity might be particularly helpful in many cases.
Another fundamental obstacle is reluctance among member states to cooperate more and, in some cases, to delegate certain regulatory and supervisory powers to an agency outside their national jurisdictions. To relinquish some measure of control, member states must have confidence that appropriate action will be taken when needed, that they will be kept informed and involved, and that outside agencies will respect their interests. They must also be appropriately represented in decision making and implementation. Reconciling all these conflicting interests efficiently will require giving the relevant prudential authorities joint responsibility and joint accountability for those aspects of financial stability that they control, at all three levels of the system: the EU-wide level, the country level, and the institutional level. In particular, home-country supervisors should be given explicit responsibility and held accountable for financial stability issues arising from their banks’ activities in host countries.
A third obstacle to strengthening financial stability arrangements and nurturing financial integration is the lack of agreement on basic principles and procedures for crisis management and resolution. Such principles and procedures are needed to reconcile financial stability and moral hazard considerations and thereby to put in place a credible (which implies cost efficient) market discipline policy. They should focus on minimizing the costs of resolving crises and should include the option of using taxpayer funds in ways that do not benefit shareholders and impose losses on uninsured creditors. Minimizing collective costs in a crisis requires a mechanism to compensate countries that incur disproportionate costs and is also a prerequisite for sustainable cost sharing. A harmonized framework for early preventive and corrective action is needed, to minimize the use of the resolution framework and to build trust between countries.67
Adapting the Institutional Architecture
Once there is agreement on what the financial stability framework should deliver, an institutional architecture must be designed to turn this into reality. This in essence involves making choices on a number of criteria: a centralized or decentralized framework, uniform or diverse treatment for similar regulated entities, a sector-specific or an integrated approach, and reliance on new or existing institutions, among others. These choices present themselves for each of the five pillars of the financial stability framework (financial regulation, supervision, crisis prevention and management, crisis resolution, and financial infrastructure). However, internal consistency is an imperative, and so these choices are not independent. From the integration policy point of view, decisions on centralization and uniformity are the most important.
Centralization is more consistent with the single market and with increasing cross-border risks. It also provides greater uniformity. In particular, centralized regulation could deliver a single rule book. On the other hand, proximity argues in favor of a decentralized approach toward supervision of small local institutions and other local/national aspects of financial stability. Most proposals to centralize financial stability arrangements are therefore hybrid proposals that include some decentralized elements. There are a number of prominent issues that need to be resolved to allow a centralized approach. These include fiscal responsibility, accountability, financing, organizational structure, links with national institutions, and consistency with national laws. Moreover, there are limits on the extent to which centralized and decentralized elements can be combined in a consistent framework. Some proposals, such as the lead supervisor (discussed below), focus on centralizing financial stability functions at the institutional (LCFIs), not at the EU level, within an architecture that remains decentralized at the country level.
Uniformity and harmonization level the playing field, but there are also arguments in favor of diversity—that is, different prudential regimes existing alongside each other—because the needs of various institutions, and the risks they represent, are not always the same. In particular, the optimal regulatory and supervisory framework for small local banks is different from that for cross-border LCFIs.
Proposals to adapt the institutional architecture entail various degrees of centralization, uniformity, and departure from existing arrangements. Four categories of proposals are discussed here.
Enhancing Existing Structures
Proposals. A variety of proposals have been floated to improve and support aspects of the existing prudential framework while leaving basic arrangements unchanged. These include enhancing the structures for information exchange, coordination, and cooperation, for example, by means of colleges of supervisors; instituting qualified majority voting in Level 3 committees; and giving regulators and supervisors European mandates alongside their existing national ones.
Benefits. These measures would enhance efficiency and effectiveness and have the potential to soften the impact of many of the existing problems. Qualified majority voting at Level 3 could improve the speed and quality of rulemaking. European mandates could help align incentives.
Drawbacks. Few of the existing problems would be fully eliminated. For example, a European mandate alongside a national mandate might be of limited effectiveness in aligning incentives.
Feasibility. These measures may be relatively easy to implement compared with alternative proposals, although ensuring accountability and enforceability of European mandates might be challenging.
Designating a Lead Supervisor
Proposal. The home-country authorities could be given responsibility for regulation and supervision of both branches and subsidiaries throughout the European Union, with host-country authorities performing supporting or delegated tasks and fulfilling an advisory role (see, for example, EFR, 2004 and 2005).
Benefits. This would simplify and streamline supervision of LCFIs, establish a single point of contact, and facilitate cross-border operations.
Drawbacks. The mismatches between authority, responsibility, and accountability could be worse than under present arrangements. Horizontal consistency in supervision between groups would not be automatic. The effectiveness of supervision may be reduced when lead supervisory authority must be transferred, for example as a result of mergers or acquisitions or changes in the location of group headquarters.
Feasibility. This measure fits within existing institutional structures, but requires resolution of the problem of applying home-country regulations in host-country legal systems. Depending on the arrangements, other fundamental problems (including the need for more harmonized remedial action and a cross-border crisis-resolution framework) might arise. Lead supervision implies a loss of control for host-country authorities, which many would find difficult to accept (see Bednarski and Bielicki, 2006).
Box 10.4.The Case for a European Banking Charter
To resolve the question of different regulatory and supervisory needs for small, domestically operating financial institutions and large cross-border institutions, Čihák and Decressin (2007) propose introducing a European Banking Charter. The European Banking Charter would be a full-fledged European regulatory and legal framework for financial institutions and would exist alongside national regimes. It would be designed to be attractive to EU financial institutions that are heavily engaged in cross-border business. To be effective, it would have to be a true, fully integrated regime, without any country-specific elements. A key feature of the proposal is that the regime would be optional: financial institutions would be free to choose between operating under national banking charters or the European Banking Charter. This would facilitate the development of a supervisory infrastructure that is attuned to market developments, would promote competition among regulators with a view to avoiding excessive regulation, and would create space for financial innovation. The authors also argue that the Charter could be favorable to Europe’s taxpayers if it comes with efficient and effective EU-level crisis-prevention, -management, and -resolution mechanisms.
Institutions chartered under the European Banking Charter would compete on a truly level playing field within the European Union, at least from the perspective of prudential policies and practices. Although there would remain many problems to be resolved related to supervision, the key advantages of the proposal are that it focuses on large cross-border institutions and bypasses the need for extensive harmonization of national legislations.Source: Based on Čihák and Decressin (2007)
Establishing a 28th Regime
Proposal. There could be a pan-European prudential “28th regime” functioning alongside national regimes, focusing on the specific needs of cross-border LCFIs. This could be made optional (Box 10.4) or mandatory for LCFIs.68
Benefits. A specific EU-wide legal and regulatory system addresses the urgent need for a financial stability framework for systemically important LCFIs. It would create a level playing field for LCFIs without full harmonization and would allow efficient and effective supervision of LCFIs.
Drawbacks. An optional regime may not resolve existing problems if a significant number of LCFIs opt out. A mandatory system could fail to level the playing field between LCFIs and other banks.
Feasibility. This measure would leave the present institutional structure largely intact. By focusing on LCFIs, instead of on negotiating a harmonized system for all banks, it would deal with the most urgent issues and would facilitate agreement. There would need to be significant legal and institutional changes, especially if the 28th regime would cover all pillars of the financial stability framework. In particular, there would need to be solutions to remedial action and crisis resolution, including burden sharing.
Developing a European System of Prudential Agencies
Proposal. Existing or reformed national prudential agencies could be brought together in a regulatory and supervisory system with a cross-border superstructure, along the lines of the European System of Central Banks. The superstructure could act as a single regulator, include other pillars of the financial stability framework (for example, deposit insurance and crisis management), exercise quality control over supervision by the national entities, and provide information-sharing and coordination mechanisms. It could also act as a single point of contact and lead or consolidating supervisor for LCFIs.69
Benefits. This superstructure would combine many of the advantages of centralization, including potentially a single rule book, with many of the advantages of decentralization. The system could be designed to ensure that interests of host countries are adequately taken into account.
Drawbacks. Depending on the arrangements, decision-making structures could be heavy.
Feasibility. Depending on the extent of the superstructure, there may need to be significant legal changes, and perhaps a treaty change. Fundamental questions, including on burden sharing, would need to be resolved. This would require a degree of political support that presently seems absent. Implementation would be facilitated by the possibility of incorporating existing national agencies.
Financial stability arrangements in the European Union will need to adapt in line with the ambition for, and progress toward, a single financial market.70 Most urgent is the need to develop a framework that effectively contains the risks posed by LCFIs while simultaneously reducing the costs and operational constraints placed on these institutions by current arrangements.
Although EU countries generally feature strong national prudential policies and practices, these systems are challenged with the need to deal not only with LCFIs, but also increasingly with EU-wide risks and country-level risks in countries with a significant foreign bank presence. Moreover, aspects of the current financial stability arrangements generate friction that slows integration: creating a level regulatory and supervisory playing field throughout the internal market remains a major struggle; current arrangements leave significant, albeit declining, scope for prudential authorities to resist financial integration; and misalignments between authority, responsibility, and accountability provide national authorities with incentives to do so.
Aligning authority, responsibility, and accountability for financial stability is key to addressing these challenges at their root. An integrated financial market requires financial stability arrangements that foster joint responsibility and joint accountability for financial stability, at the level of the EU, individual countries, and institutions (LCFIs). In parallel, the potential impact of misaligned incentives should be reduced through sound institutional and procedural arrangements and extensive information sharing. Achieving those objectives would allow financial stability arrangements to be streamlined while promoting confidence in host countries that their interests will be taken into account. For efficiency reasons, this streamlining would have to reflect the reality that the operational differences between branches and subsidiaries have become much narrower.
Joint responsibility and accountability in time will need to be reflected in agreement on a number of foundation blocks, the presence or absence of which will determine the feasibility and effectiveness of any reform of the institutional architecture for financial stability. These foundation blocks, in essence certain specific elements that the pillars of an integration-compatible EU financial stability framework will need to deliver, include:
Much greater information sharing, including with the ECB: Ideally, such reinforced information sharing would take the form of a central repository of prudential information supported by a European confidentiality regime.
A common rule book of regulatory standards and supervisory practices: this will build confidence between national authorities and stimulate financial integration by promoting cross-border banking.
Rules and tools for harmonized crisis prevention and remedial action based on early intervention and expeditious decision-making procedures: these should be designed around the objectives of limiting the risks and costs of crises and should be incorporated in the common rule book in order to enhance trust between national authorities. The tools need to include the ability to impose mandatory capital increases in cases of near-insolvency.
A credible and cost-efficient, EU-level exit framework for systemic banks, supported by further harmonization and improvements in national crisis-resolution frameworks: this requires better and more harmonized national bank insolvency laws; single bankruptcy proceedings for cross-border groups, including operationally integrated subsidiaries; greater convergence in national deposit insurance schemes; agreement on how to provide solvency support to failing systemic banks while avoiding moral hazard; and removal of the need for ex ante approval by the European Commission for financial support.
A shared commitment, enshrined in prudential authorities’ mandates, to minimize the (EU-wide) collective costs of a crisis, combined with arrangements to share any related fiscal costs according to agreed principles.
These foundation blocks will need to be designed to work together smoothly. Coming to agreement on these foundation blocks and on the way they will work together will require convergence toward a set of principles that amount to a sort of common “financial stability philosophy.” Such an increased basic understanding will in turn facilitate agreement on concrete reforms, as well as on coordinated actions in times of crisis.
A variety of institutional arrangements—ranging from new, more centralized, prudential institutions to setups developed from existing institutions—could be compatible with these foundation blocks. To stay with the building analogy, various designs of houses could be built based on them. The spectrum of feasible and effective institutional architectures is also likely to evolve over time, partly as a function of progress on the foundation blocks. What is important is that the debate place due emphasis on what the architecture has to deliver, not on what is deliverable under the existing architecture. The debate also needs to remain open, precluding no reasonable options even if those options are not achievable under current circumstances.
Focusing on these fundamental building blocks and maintaining an open debate about long-term issues should provide a helpful sense of direction as well as permit a reform strategy that can be pragmatic in its implementation. These efforts should be part of a multipronged reform strategy that should, to the extent possible, avoid allowing progress in one area to be impeded by obstacles encountered in another. In this regard, existing efforts to improve regulation, supervision, coordination, and cooperation should be continued and strengthened where possible, with a particular focus on the systemically important LCFIs.
There are no easy solutions or quick fixes to address the European Union’s financial stability challenges. However, considering what European (financial) integration has already achieved, the foundation blocks of an adapted financial stability framework do not seem out of reach.
Walter (2003) argues that the regulatory environment is central to the evolution of the financial services industry and that a reasonably level regulatory playing field is therefore essential for European financial integration.
A mapping exercise carried out by the BSC shows that a set of 46 EU banking groups with significant cross-border activity accounted for 68 percent of EU banking assets in 2005 (ECB, 2006a; and Trichet, 2007). Although European banks tend to have a strong international orientation, the top 30 conduct 78 percent of their business in Europe (Schoenmaker and van Laecke, 2007). See also Chapters 6 and 7.
The analysis in Chapter 4 suggests that market forces could strengthen financial institutions in the European Union, as competition among banks and bank efficiency are currently both lower than in the United States.
For a discussion of the risk of policy reversals caused by financial crises see, for example, Rajan (2006).
In banking, Basel Committee guidance has been particularly important.
Nonetheless, the Maastricht Treaty (Article 105(6)) envisages the—thus far unused—possibility of transferring a degree of prudential authority to the ECB, through a unanimous decision by the European Council.
For example, in the event of a merger between Barclays, based in the United Kingdom, and ABN Amro, based in the Netherlands, responsibility for consolidated supervision would fall to the U.K. Financial Services Authority (FSA), even though the FSA is unfamiliar with ABN Amro’s operations. Meanwhile, the Dutch central bank’s base of knowledge about ABN Amro and its operations could be expected to diminish over time.
For a discussion on the role supervisory and prudential policies can play in containing the risks related to episodes of rapid credit growth, see Hilbers and others (2005).
Lannoo (2002) provides an extensive although by now somewhat dated overview of EU and EEA fora for cooperation in financial regulation and supervision.
Barros and others (2005) suggest that political obstacles could be a significant factor in explaining the relatively limited number of cross-border bank mergers and acquisitions in the European Union.
To reduce the scope for national interests to influence the prudential approval process for mergers and acquisitions, the European Commission has prepared a proposal for a Directive that aims to standardize approval procedures and reduce supervisors’ scope for discretion. At the time of writing, the proposed Directive was not yet approved by the EU Council.
Some prudential authorities apply the “stand-alone principle,” which requires that a subsidiary be able to operate independently from its parent (see Bednarski and Bielicki, 2006). This precludes any significant degree of operational integration between parent and subsidiary.
Financial infrastructure—which comprises payment systems, payment instruments, and securities clearing and settlement systems—is not discussed here. However, a good overview of relevant issues from the perspective of a euro area member state is given in Bank of Finland (2005). See also Chapter 5 and Kazarian (2006).
CEBS’ guidelines, for example, specify that information exchange should be “proportionate and risk-focused, to avoid unnecessary information flow,” that information flows will naturally be asymmetric, and that supervisors need to exercise restraint in order to avoid disproportionate or redundant flows of information (CEBS, 2006b). The CRD distinguishes between essential and relevant information, the former of which should be provided by supervisors on their own initiative, the latter upon request. In its 2006 Annual Report, CEBS notes that an analysis of barriers to effective information sharing showed that legal and practical issues may still be an impediment to an easy exchange of information.
Financial crisis management in the European Union is a real-world specimen of what the microeconomic literature labels a “mechanism design problem.” This literature shows that when a large number of agents with little knowledge about each other’s preferences find themselves with different preferences across a limited set of projects on which they need to agree, no social choice functions exist that are implementable in dominant strategies and that are ex post efficient. See Čihák and Decressin (2007).
The ECB, for example, has no access to confidential information about financial groups. The information shared in the BSC is typically at the aggregate level.
These include the First and Second Banking Directives, the Own Funds Directive, the Solvency Ratio Directive, the Large Exposures Directive, the Directive on Consolidated Supervision, and the Directive on the Annual Accounts of Banks.
Upon joining the European Union, new member states were obliged to accept these rules, as part of the “Acquis Communautaire,” the body of EU legislation in existence at the time.
CRD: Directive 2006/48/EC and Directive 2006/49/EC; MiFID: Directive 2004/39/EC.
Separate from the Lamfalussy process, the European Commission has created an electronic helpdesk to assist countries in interpreting the CRD, thus contributing to harmonized implementation of its provisions.
CEBS has introduced standardized formats for capital reporting (COREP) and consolidated reporting (FINREP) by banks, although these standardized formats only partially harmonize the reporting requirements themselves. Reporting is to take place using XBRL (Extensible Business Reporting Language), which will facilitate consistent compilation of data across the European Union.
Efforts to achieve a common reporting form are illustrative in this regard: the standardized reporting format for the new solvency ratio introduced by CEBS in 2006 included a core set of data to be used “almost uniformly” throughout the European Union, plus standardized supplemental information details which provide “sufficient flexibility to meet the needs of different supervisory authorities.” See CEBS (2006a). The Inter-Institutional Monitoring Group (IIMG, 2006) also pointed out that, since Level 3 committee members represent national authorities, the results from their work may be driven more by “consensus” than “best practice.”
The IIMG defines goldplating as “regulatory additions made while implementing in national law rules which were adopted at European level under a maximum harmonization regime.” As an example, McCreevy (2005) cites a case of a national regulatory authority that reportedly added 300 pages of guidance to a 10-page directive.
Host countries nevertheless retain responsibility for enforcing local liquidity requirements on foreign branches, can require certain minimal financial reporting, and are allowed to take measures to protect depositors in emergency situations.
Directive 2002/87/EC. With regard to the implementation of the FCD, the European Commission is assisted by the Financial Conglomerates Committee (FCC) and the Mixed Technical Group (MTG), both chaired by the Commission. The FCD gives a number of criteria to define a financial conglomerate, mostly of a quantitative nature (Articles 2.14 and 3).
Article 12 (2). Prior consultation with other involved supervisors is required unless the situation is considered urgent.
Notice the differences in terminology and applicability between the FCD and the CRD. The former uses the term “coordinating supervisor” and the latter “consolidating supervisor” for what is essentially the same basic function. The FCD applies to a limited number of cross-sector conglomerates; the CRD to banking groups. Many, but not all, of these conglomerates are also banking groups.
The CRD distinguishes two types of information to be exchanged: “essential” and “relevant.” Supervisors are required to exchange the former on their own initiative; the latter upon request. The directive defines “essential” as having the potential to “materially influence the assessment of the financial soundness” of an institution or its subsidiary and provides the examples of adverse developments in credit institutions or other entities of the group and major sanctions and measures taken by the supervisors with regard to the institution. The term “relevant” information is not strictly defined, leaving significant scope for differing interpretations. However, the CRD specifies that the importance of the subsidiary in the financial system in the host country must be taken into consideration.
The obligation of prior consultation is relaxed, however, in case of urgency or when consultation could jeopardize the effectiveness of the measure. In these cases, the supervisor shall without delay inform the other supervisors.
To help implement the CRD’s provisions on supervisory cooperation, CEBS has issued common guidelines on cooperation between consolidating supervisors and host-country supervisors (CEBS, 2006b). In addition, in 2006 CEBS launched a pilot project on operational networking, providing infrastructure to support enhanced exchange of information and experiences between supervisors involved with a sample of 10 cross-border banking groups.
Such MoUs have been concluded for Nordea, Fortis, and Sampo.
The European Company Statute enables the establishment of a European public limited-liability company (Societas Europae, or SE) that facilitates the cross-border presence of multinationals or even the movement of the SE’s head office between member states.
The Francq Report recommended that ECOFIN supply the political impetus for the process of supervisory convergence. It also recommended a more systematic use of tools aimed at fostering a European supervisory culture, greater use of nonbinding mediation mechanisms and voluntary delegation between supervisors, and work on streamlined reporting requirements and data-sharing arrangements.
See Dermine (2003). A notable exception is the cross-border operations of EU banks in the United Kingdom, which reflect London’s position as a financial center. These operations are very large in terms of assets and are primarily conducted through branches. Crossborder subsidiaries of EU credit institutions in other EU countries held three times as many assets as similar cross-border branches. See European Central Bank (2006a).
As described in the CRD (Art. 12 (1), (a) through (h)), it covers identification of the group structure, strategic policies, risk management, adverse developments with regard to the group, and supervisory measures undertaken against the group.
This principle is enshrined in both the FCD (Art. 11 (2)) and the CRD (Art. 130 (2))
In addition to the other considerations discussed in this chapter, supervisory institutions are often funded by fees levied on the institutions they supervise. This means that the trend towards home country supervision may deprive some host supervisors of their funding base.
In the Panagis Pafitis case, the European Court of Justice ruled that a provisional administrator of a bank could not impose a capital increase without the approval of the existing shareholders, because this would be a violation of shareholder rights as defined in the Second Company Law Directive (see Hüpkes, 2000).
See also Walkner and Raes (2005) on incentive issues.
Leeway for delaying crisis actions and the provision of information exists within the current framework, in part due to relatively general exemptions and to the scope for interpretation left by legal texts (for example, Arts. 130–132 of the CRD).
For example, by making use of the scope offered by Art. 12 of the Conglomerates Directive.
For an in-depth discussion of current arrangements, see ECB (2007b). An early but comprehensive analysis of the challenges involved in crisis management in the EMU context is provided by Prati and Schinasi (1999).
In particular, Articles 129–132.
Cooperation with non-EU authorities is outside the scope of the MoU, the full content of which has not been disclosed to the public.
Boot (2006) puts this argument into a broader context.
See also Vives (2001).
Bankruptcies of insurance companies and investment firms are also usually handled under general bankruptcy law.
The United States has long had special bankruptcy arrangements for banks (see Hüpkes, 2003). The current regime was put in place by the 1991 Federal Deposit Insurance Improvement Act (FDICIA).
However, if a large part of the bank’s liabilities are uninsured, liquidation might be cheaper from the point of view of direct resolution costs, even in the case of a significant gap between the liquidation value of assets and their going-concern value (see Hoggarth, Reidhill, and Sinclair, 2003).
Directive 2001/24/EC on the Reorganization and Winding-Up of Credit Institutions. Similar principles were laid down for insurance companies in the 2001 Directive on the Reorganization and Winding-Up of Insurance Undertakings (Directive 2001/17/EC).
See Hadjiemmanuil (2005) for a broader discussion.
The amount of coverage even differs within countries, in some cases because of the existence of multiple deposit insurance systems, but also between domestic banks and branches of banks from other EU countries that are covered by their home-country systems.
The mobility of banks is increasing due to increased opportunities for cross-border mergers and acquisitions and due to the possibilities offered by the European Company Statute.
In some funded systems, there is ambiguity about whether contributions should be seen as deposits or as insurance premiums, that is, as a claim upon the fund or as a nonrefundable expense. In the latter case, it can act as an impediment to bank mobility and the transformation of cross-border subsidiaries into branches.
See Kaufman (2004) for a discussion on the importance of liquidity insurance.
This is the case even beyond the EU context. Hüpkes (2005) notes that it “seems impractical to put an LCFI into liquidation” (p. 196), the U.S. FDIC has special arrangements that it intends to use in an LCFI crisis, and there is a general scarcity of precedents of LCFIs being liquidated in developed economies (see OECD, 2002). See also Fonteyne (2007b) and Hüpkes (2000, 2003, 2005).
Fiscal burdens can occur at different stages of the crisis-management and -resolution process: it can result from a decision to provide solvency support to a troubled financial institution, losses on lender-of-last-resort operations, shortfalls in the deposit insurance system, a political decision to compensate depositors (and perhaps other creditors) over and above the compensation provided by the deposit insurance scheme, and the provision of government guarantees or compensation for losses or negative equity as part of a purchase and assumption by a private acquirer.
To some extent, the central bank’s seigniorage income (Goodhart and Schoenmaker, 2006) or valuation gains could be used to finance solvency support, but these are in essence also fiscal resources, and their overuse could leave the central bank with a weakened balance sheet that could impede its ability to fulfill its primary duty of conducting monetary policy.
See, for example, Dermine (2005).
Ring-fencing of branches is not permitted within the European Union. Subsidiaries can normally be ring-fenced, but nonetheless, and despite legal limitations, the operational integration within banking groups can provide significant scope to move assets around during the initial phases of a crisis.
Freixas (2003) presents an analysis of “improvised cooperation” to organize a bailout, and argues that the creation of EMU has made it more costly to bail out transnational institutions.
This is illustrated, for example, by rating agencies taking the probability of official support into account when rating a financial institution and by statements of academics and insiders. The EFR (2005) remarks that “an implicit or explicit understanding that, in the case of a major financial crisis, fiscal resources will be made available to limit the damage to the overall economy” has been a historic element underpinning financial stability arrangements in all member states.
In this regard, Prati and Schinasi (1999) argue that, with certain important caveats, public transparency about the mechanisms for resolving systemic crises may help reduce moral hazard.
For example, the European Commission (2006b) points out that burden sharing needs to be addressed before any far-reaching reforms of deposit insurance schemes can be considered.
For an overview of some positions in these debates, see Fonteyne (2007a).
One option that has been proposed is some form of Structured Early Intervention and Resolution framework. See Nieto and Wall (2006); Mayes, Nieto, and Wall (2007); and European Shadow Financial Regulatory Committee (1998).
The possibility of a mandatory 28th regime for multijurisdictional institutions is discussed briefly in EFR (2005).
A proposal for such a system, a European System of Financial Supervisors (ESFS) with a European Financial Authority (EFA) as the superstructure, is developed by Schoenmaker and Oosterloo (2006). In this proposal, the EFA plays a relatively limited role, as it would not be in charge of supervision of pan-European groups. Rather, the national supervisory agency in the country where the group’s headquarters are located would act as lead supervisor. Emphasizing process as much as institutional setup, Ingves (2007) proposes a European Organization for Financial Supervision (EOFS) that would gradually take on increasing responsibilities in the supervision of pan-European banks, working together with national supervisors.
Lamfalussy (2004) described the existing institutional setup as a “mind-boggling patchwork.”