The impetus for establishing a banking union is rooted in the sovereign-bank-nexus, i.e., the vicious circle between financial sector and budgetary instability: A banking crisis will worsen the budgetary situation if, as is likely, the crisis leads to a recession and even more so if public assistance is needed to return the banking sector back to health. Likewise, a public debt crisis will create problems in the banking sector because banks tend to hold large volumes of public debt on their balance sheets (not least because they are encouraged or even required to by regulation) and also because refinancing costs of banks are closely correlated to those of the states in which they are headquartered.
Either can trigger a downward spiral, as we have seen in recent times: In the cases of Ireland and Spain, problems in the financial sector were the root-cause of a budget crisis in those countries which, prior to the financial crisis, had sound fiscal positions. In contrast, Italy is an example of a country where a basically sound banking sector has been hit by deterioration in the public debt position.
Conceptually, the idea of a banking union tries to break this vicious circle by weakening the connection between a national banking system and the public sector in its home jurisdiction:
The link from public-sector instability to financial sector instability is broken if banks have a diversified asset and funding base. In this case, problems in the public sector will only have a muted effect on the asset quality and funding costs of the respective domestic banking system. This would require a fully integrated banking market, with cross-border institutions and no home-bias in banks' asset portfolios and funding structures.
• The link from financial sector to budgetary instability is broken if the costs of stabilizing a banking system no longer fall exclusively on the home jurisdiction of the banks in trouble, but are instead shared across jurisdictions - either directly via common public budgets (like the ESM) or indirectly via joint / interconnected resolution funds and deposit guarantee schemes.
But the rationale for a banking union goes deeper than breaking the bank-sovereign nexus: Market integration, financial stability, and banking supervision at the national level just do not go together. More broadly then, the push for a Banking Union stems from three inter-linked and, if pursued successfully, mutually reinforcing motives:
• Maintaining financial stability on the basis of effective supervision and crisis management,
• preserving the Single Market for financial services, and
• avoiding competitive distortions.
The objectives are (i) to enhance financial stability by overcoming market fragmentation and (ii) to preserve the Single Market in financial services. This becomes even more pertinent in the face of mounting evidence that the re-nationalisation of Europe's financial markets is becoming entrenched as a result of market forces and regulatory action in the wake of the financial crisis. Such a re-fragmentation not only reduces the efficiency and competitiveness of Europe's financial markets, it is also inimical to financial stability.
A comprehensive banking union would comprise of, at a minimum, four elements:
• A single rule book, establishing materially uniform rules,
• pan-European banking supervision,
• a pan-European resolution regime including an EU bank resolution fund, and
• harmonised deposit guarantee schemes (DGS).
In September 2012, the European Commission presented its proposals for a Banking Union. While the four above-mentioned design elements are included in that proposal, the level of detail varies significantly: Whereas the Commission's proposals for a pan- European supervisory mechanism are fairly specific, the proposals on bank resolution and deposit guarantee schemes are less ambitious and essentially refer to proposals already in the EU's legislative process.
The different degrees of specification and detail on these four elements partly reflect the fact that progress has already been made on some of them - for example, successive versions of the Capital Requirements Directives, transposing the Basel capital requirements into EU law, constitute significant steps towards a single rule book. To a larger degree, however, the differences in specification reflect political opposition in member states that stands in the way of bolder concepts. This is as deplorable as it is dangerous because the four elements form an integral system. Separating one, such as supervision, from crisis management will distort incentives for authorities as well as financial sector participants. This in turn could make the EU's financial system less resilient.
There are a number of design issues with regard to the organisational and institutional set-up of a banking union that still need to be sorted out.
Authority for banking supervision
: Despite the October-summit agreement to clear all remaining legal hurdles by the end of this year, there is still considerable debate over the ECB as the EU banking supervisor of choice and the scope of its powers.
The communiqué of the June summit speaks only of a supervisory mechanism "involving the ECB". This was clearly motivated by the reputation the ECB enjoys and by the fact that the European Treaty (Art. 127.6) allows for the transfer of supervisory powers to the ECB, based on a unanimous vote by ECOFIN, which will make the legislative process easier.
However, there are a number of arguments against entrusting financial supervision to a central bank, most importantly potential conflicts of interest with the mandate of monetary policy as well as concerns over a concentration of power and question of how countries that are not members of the currency union should be represented in the decision making bodies of the supervisory mechanism.
The Commission proposal would give the ECB sweeping powers and full control in all areas of prudential policies. This is sensible conceptually and rational from the point of view of the ECB as financial supervision is prone to grave reputational risks. National authorities however tend to preserve as much power as possible and to limit the powers of any pan-European supervisor.
Bifurcated vs. federal
: A two-tier supervisory system that limits EU supervision to large, multi-jurisdictional institutions would create scope for competitive distortions and regulatory arbitrage. Worse, it would ignore the important lesson of the recent financial crisis that smaller, regional banks are at least as likely to cause systemic crises as large ones. Hence, the European supervisory system should be federal. For practical reasons, small and home-market oriented institutions would continue to be supervised by national authorities, but these would be subject to the final say of the EU-level authority, which would directly supervise systemically relevant financial institutions that operate on a pan-European basis.
Single Rule Book
: Arguably, the single rule book should be the most easily achievable element of banking union. Over the past few years, the EU has made considerable progress in establishing a harmonised framework for banking regulation and supervision.
However, EU members have recently veered off that course and allowed for greater national discretion. Moreover, actual day-to-day supervisory practices have never been as closely aligned as the rule-books suggest. Clearly, both issues will need to be addressed to achieve a truly single rule book.
Resolution regime and fund
: Effective bank resolution regimes are needed to ensure that even the largest and most complex financial institutions can be wound down in an orderly way. A resolution fund to provide bridge financing, financed predominantly but not exclusively by contributions from the financial industry, would be a useful element of such a regime. While some EU member states have set up such funds at a national level, a pan-European scheme has yet to be established as member states cannot agree on a financing mechanism or resolution authority, both of which would inevitably infringe on national sovereignty.
Deposit Guarantee Schemes (DGS)
: DGS play an important part in maintaining depositors' trust in the stability of the banking system. Historically, DGS were developed in response to specific market structures and it is thus no surprise that the design of such schemes varies substantially across the EU. Given the complexity of bringing very different schemes together, and given the fact that DGS are of limited relevance in dealing with failures of large cross-border banks, it would probably be best that instead of aiming for a common supra-national scheme efforts be directed at ensuring that all national schemes are equally robust and equipped to meet potential demands. Beyond such minimum harmonisation (as is indeed sketched out in the current legislation), existing national DGS could remain in force and be complemented by a limited re-insurance scheme, which would kick in if national DGS were exhausted and the state in question was incapable of backing the system up.
The political process
The realisation of a full EU banking union will prove difficult. Some countries see a banking union as an integral part of the new institutional framework for a more stable European Monetary Union and a step towards closer economic union with tougher disciplines on economic and fiscal policies as well as towards closer fiscal and political union.
They are therefore asking for a well-designed, comprehensive and consistent framework. Other countries, however, see banking union in a more narrow context, namely in the context of the debate about direct ESM assistance to individual banks, which necessitates taking banking supervision for those banks out of the hands of national authorities and transferring it to the EU.
The establishment of supra-national structures and institutions is evidently in conflict with national sovereignty. Financial supervision is inextricably linked to the exercise of sovereign power. More importantly, supervision creates a latent fiscal liability which may become real in the event of a systemic crisis. As the recent crisis has shown, in a systemic crisis fiscal resources may be required to restore confidence in the financial system. This is why, ultimately, the issue of how to organise financial supervision cannot be separated from fiscal liability. Supra-national supervision also threatens vested interests, in this case of national supervisory authorities bent on preserving their powers. Similarly, supra-national arrangements, especially those for supervision, would also disrupt the relationships between national authorities and banks in any given country, which are often marked by regulatory capture and, in times of crisis, a tendency for regulatory forbearance.
Cross-border burden sharing raises potential distribution conflicts. Those countries and institutions that expect to be net payers will be wary of committing their resources to preserve the stability of other country's financial systems. This is particularly true as long as it remains unclear whether the envisaged institutional arrangements for supervision will be strong enough to ensure effective discipline on risks accumulated in the financial systems of banking union members 
. A banking union therefore pre-supposes elements of a political union.
In addition, the banking union plan puts the spotlight on a fundamental issue: that of how institutional arrangements for the euro-area can be reconciled with those for the EU-27 as a whole. Specifically, the question is whether it is more important to strengthen the stability of the euro area or to safeguard and maintain the Single Market for financial services.
The relative importance of either objective has a bearing on the institutional design of a banking union: Stressing the latter objective favours a strong role for the EBA to ensure the consistency of rules and supervision in the EU-27, whereas emphasis on the former objective gives the ECB a prominent role and strives to integrate crisis management systems at the EU level.
As with so many other institutional arrangements in the EU, the design of the banking union - at least its initial design - will be the result of what is politically possible and not necessarily what is required to put Europe's financial system on a firmer footing. It is likely that instead of a consistent, integrated design, an incoherent system will be established that leaves out pan-European crisis management instruments - ignoring the fact that supervision and crisis management are inextricably linked. EU-level supervision will remain weak and dependent on the support of national supervisors which have little incentive to cooperate or to share problems in their banking sectors at an early stage. If this were indeed the result, EU- leaders would have wasted an important opportunity to build a more unified and stronger Europe.