1. Introduction

The fast increase in cross-border banking claims between 1999 and 2007 had raised hopes that the single financial market was becoming a reality. But the financial crisis proved that it was a foggy and cloudy dawn, with many dangerous spots still in the shade.  Although banking markets were rapidly integrating, institutions kept them pretty ring fenced along national borders; the regulatory framework aimed at harmonizing the actions of national supervisors left in fact very large degrees of freedom to its implementation in member states; measures and resources to recover or resolve banks in trouble were national with no institutional framework to mutualize them. Banks, consequently, were international in life, but national in disease and death, even though the consequences of such casualties could not be contained within national borders, precisely because their activities were large, spread across several countries and because of the perverse interaction between banking and sovereign debt.

The Banking Union is the response to this geographical mismatch between markets and the rules overseeing them. As convincingly argued by Schoenmaker (2012 and this issue), it is impossible to have international financial integration and financial stability if supervision remains national. And it is impossible to break the perverse link between banking and sovereign balance sheets if the resolution and guarantee framework and funds are not integrated and mutualized. The objective of the Banking Union is therefore to guarantee an integrated and resilient European banking sector.

As reported below and in the Numbers section of this issue, there is only limited evidence of a retrenching of banking activities within national borders as a consequence of the crisis (Figures 4, 5, 8 and 9). In other words the crisis has only partially discouraged EU banks from operating in other EU countries through branches and subsidiaries. Will the Banking Union strengthen this pattern by providing an effective levelled playing field to finally achieve the transition of the integrated financial market from its shady dawn to a full sunshine?

This issue of European Economy addresses this question by mostly dealing with the Single Supervisory Mechanism (SSM). The SSM is a daunting task of institutional harmonization and supervisory centralization the Eurosystem is undertaking. It involves: defining and implementing a coherent harmonized legal and regulatory framework for banks, based on the CRR/CRDIV; building up an effective central supervisory apparatus, defining its legal framework, governance and procedures; coordinating the operations of national competent authorities (NCAs) within a single rule-book in a coherent arbitrage between the union and national legal frameworks; implementing a thorough assessment of the balance sheets and activities of the 122 large systemic banks, subject to the central supervision by the ECB, which account for 25 trillion in assets (80% of the euro area total).

The complexity of hammering a single supervisory mechanism across such a broad and diverse legislative and institutional space has produced a yet imperfect, even though we deem extremely necessary, scheme. Our bottom line is that the SSM is a fundamental step forward towards a single European banking market. But there are still many grains of sand in the wheels. Part of them will be blown away with time, fine tuning and adaptation, part are structurally there to stay.

Transient conflicting items should be removed and smoothed out as rapidly as possible, as they could render the transition to a fully functioning SSM painful, with conflicts of power and unclear signals given to stakeholders, especially the supervised entities and the markets.

Yet some of the issues have inherent structural problems that will be lifted with difficulty, perhaps smoothed through practice, but still persistent. A first grain of sand comes from the interface between the single supervisory mechanism and national supervisors. The crucial role that national supervisors still play in overseeing the operations of their home based banking activities in joint supervisory teams, even for the 122 banks under the direct supervision of the ECB, raises a key issue of information sharing and of the national implementation of the directives from the single supervisor. A second grain is fed by the potentially conflicting objectives of micro and macro prudential regulation. Finally, sand is produced by the interaction between countries which are inside and those outside the union, especially for what concerns the supervision of banks based in both groups of countries.

These impediments provide a serious challenge to the full implementation of the SSM and these early years of transition are especially delicate as national and central supervisors practice their joint exercise and fine tune the difficult balance of powers that it involves. The second and third pillars of the Banking Union will only be addressed superficially here, for questions connected and pertaining to the SSM. This does not mean that supervision and resolution and deposit guarantee can be seen as separate matters. In our view they constitute intricate and inseparable parts of a whole. A key problem in the first stages of the Union is precisely the different pace of implementation that the three pillars are following, with many ingredients of the third, but also of the second pillar, yet to be defined, primarily at a political level. All the same, the SSM in itself poses a set of conceptual and policy issues that require a careful and deep assessment, hence our choice to adopt a targeted focus of analysis In the end the SSM and the full banking Union are crucial preconditions for achieving an effective integrated banking market, and for making such market financially stable and sustainable both from a micro and macro perspective. But the attritions in its mechanism may constitute a still important obstacle to cross border and multinational banking and a challenge to maintaining financial stability in the euro area.

2. A daunting institutional endeavour

The Institutions section of this issue of the journal and the very thorough piece by Ignazio Angeloni discuss all the main ingredients of the SSM. Its implementation has already been largely undertaken at light speed, since the political decision to launch the banking union in June 2012. The SSM at the ECB has assumed supervisory powers since November 2014. It is useful to briefly recall here its main ingredients so as to substantiate the discussion of the following sections.

The first step has been the adoption of SSM Regulation and the incept of the preparatory work to actually set up the SSM: establishing a governance framework and operational procedures and internal arrangements for the functioning of the SSM; setting up the supervisory structures and recruiting staff; carrying out a comprehensive assessment of the conditions of the 122 banks supervised by the ECB, through stress tests and an asset quality review, which led to the recapitalisation of banks with capital shortfalls; developing common methodologies and a Supervisory manual to standardise procedures across the Union.

The second step has been the harmonisation of the legal framework in terms of capital requirements and other matters. Discrepancies between member countries were large because of large degree of “options” and “discretion” national authorities could enjoy in applying the Capital Requirement Directive (CRD) IV and the Capital requirements Regulation (CRR). This gave rise to a regulation specifying the legal obligations for the banks under the SSM and a guide for the supervisory teams on how to treat individual cases.

A final step has been the development of the SREP (Supervisory Review and Evaluation), an SSM specific methodology for assessing and measuring risk of individual banks. This is a very comprehensive assessment exercise, on top of the initial Asset Quality Reviews and of the Stress tests, including the assessment of the business model of internal governance and risk management, and of risks to capital and to liquidity and funding.

This complex exercise of supervisory and legal harmonization still faces some limits, as very lucidly discussed by Ferrarini in this issue. In particular, the fact that the SSM operates in the Eurozone, whereas the regulatory framework applies at the EU level, introduces an element of geographically asymmetric sovereignty, implying that the single supervisor “is subject to EU prudential regulation and national law provisions, often unduly limiting its supervisory discretion”. Therefore, the SSM, always according to Ferrarini, can be defined as a “semi-strong” form of centralization, as it necessarily relies on supervisory coordination with National Competent Authorities. The need for coordination is further enhanced by the fact that the ECB has limited sanctioning powers on national banks, as it lacks locus standi in front of national courts. This limits the ECB ability to impose administrative sanctions on banks, as well as to achieve the effective enforcement of its rulings, without the full support of the NCAs.

Moreover, the fact that the SSM is focused on the Eurozone implies that the ECB has to continuously interact and cooperate with other EU authorities, like the European Banking Authority (EBA) and all the European Supervisory Authorities, forming the European System of Financial Supervisors (EFSF).

3. Grains of sand in the wheels: potential conflicts and problems in implementation

A number of frictions arise from the institutional design of semi-strong centralization combined with a geographical mismatch in sovereignty between supervision and regulation.

3.1 Central vs local supervision 

The first core question is the interaction between the central (SSM) and the peripheral supervisors (NCAs), as discussed by several contributions to this issue. In other words, how far the incentive system inbred in the institutional design of the SSM may favour or discourage an effective cooperation between NCAs and the central authority. The difficulties of a decentralized system of supervision of banks operating in Europe were debated well ahead of the crisis, which, however, was the tilting event forcing European institutions to embrace some form of centralization.

In complex environments, like the international financial market, there are pros and cons for centralizing relevant activities such as banking supervision. The possibility that centralization also involves cons may not seem obvious prima facie and it is worth discussing it together with its benefits.

Supervising a bank involves (at least) three main activities: collecting information, processing this information, and consequently acting (or don’t). In principle one could argue that centralizing all these activities at the European level could not “make things worse” than having all or some of them decentralized to national authorities. The simple argument is that, in the worst case scenario, a supranational entity, that has been attributed the authority to perform all the activities previously performed by local supervisors, can always mimic and replicate the outcome that a decentralized system would implement.

This argument is at the same time simple and wrong, because it does not account for the actual incentives to perform the three activities of supervision that systematically take us far away from an ideal first best environment. Since the process of supervising a bank is certainly not perfect (or efficient as economists would say), the effects of some sand in the wheels are debatable and it is thus important to identify the sources of inefficiency and properly understand their interactions^[1]This is a general principle well known in economics. In a first best environment, i.e. one in which decisions are efficient, adding any type of inefficiency, for example in terms of the organization … Continue reading. This is particularly important when considering a hybrid centralization framework as the one currently operating in Europe.

The presence of externalities of supervisory acts that may spill-over to other countries is certainly a strong plus in favour of centralizing the supervisory process. For example, in the event of distress of a bank, the home country supervisor in charge would likely primarily care about the consequences of supervisory acts on the home country, but not on those foreign countries in which the bank operates or where it might generate systemic spillovers. Furthermore, the systemic impact of a supervisory decision to act or not to act on a distressed bank in a large home market may be much smaller than the impact generated on a relatively small foreign market, where the bank operates with a large market share (Hüttl and Schoenmaker in this issue present a full characterization of the relevant cases.)

The consequences of these externalities in a decentralized system of supervision are lucidly discussed by Dell’Ariccia in this issue: higher supervisory standards in one country not only make that banking system more stable, but they also benefit foreign banking systems. If national supervisors fail to account for this external positive effect, they act sub-optimally, with consequences that are more relevant the more internationally integrated are national banking sectors. This may easily lead to “under-supervision” if supervision is perceived as a burden for banks which reduces their profitability (at least in the short run) and supervisors care about local banks’ profits. In this environment, independent supervisors may end up in a race-to-the-bottom, lowering supervisory standards in order to provide domestic banks with competitive advantages. The endpoint may well be a very weak banking system that the very same countries evaluate as suboptimal. Centralization may eliminate these perverse incentives and sort countries out of this prisoners’ dilemma.

At the same time, there is no reason to centralize the supervision of small non systemic banks, precisely because in this case there are no international externalities. The Banking Union is designed accordingly, leaving this responsibility to NCAs.

Transferring supervisory responsibility to a central supranational authority that does care for all involved countries internalizes these externalities, thus apparently eradicating the issue from its roots. However, this fully desirable outcome of centralization should not be taken for granted. In fact, consider the SSM as an application of a specific type of centralization of supervision. The ECB performs its supervisory tasks within the SSM, and still relies on information on banks’ activities that is at least in part produced by local competent authorities. Deprived with their supervisory powers on large banks, these authorities may have limited incentive to acquire information effectively, for example failing to investigate a national champion potentially in trouble or not transferring the full information to the central body. By avoiding to collect precise information on the bank that would be transferred and used by the ECB or simply by slowing down the process of information acquisition, national authorities may be still in the position to affect the supervisory process even if this has been advocated by the ECB.

This issue is investigated in Carletti et al. (2015) and Calzolari et al. (2015), who show that indeed centralization does not come with only pros even in very simple and realistic environments. Clearly, an obvious solution would be to centralize also information acquisition, thus avoiding to rely on biased national authorities. However, information is a complex object. Indeed, dealing with information, both in terms of acquiring and transferring it, is far from being a simple task, and centralizing this process is not necessarily an optimal solution. For example, several theoretical and empirical analyses have convincingly shown that information acquisition benefits a lot from (geographical and cultural) proximity, in our case between the authority supervising and collecting the information and the bank, the object of supervision. The quality of information and its timeliness are difficult to specify ex-ante with full detail. And the transmission of information is also problematic when one deals with “soft”, i.e. not-easily codifiable, information. These difficulties in dealing with information show that we may expect that the SSM will continue to rely on locally generated information, thus leaving the door open to significant agency issues.

A second possible issue with centralization is “regulatory or supervisory capture”. On the one hand it is generally believed that taking supervisory powers away from local inbreeding allows supervisors to act more independently. This is probably true when dealing with large but still national banks that may invest and target many resources to affect the supervisory process at the national level. However, the effect for large and cross border banks is less obvious, because in a banking union these banks have to deal with one single supervisor rather than many national and independent supervisors.

With a metaphor, centralizing the storage of valuable data in a cloud storage service does not necessarily guarantee a safer storages system even if companies offering cloud computing are specialized and use safer and better technologies. In fact, these large cloud storage systems are clearly of larger value for hackers who may concentrate and target more resources in breaching these systems rather than hacking a few computers of a single company.

What will be the effect of banking union in terms of pressure exerted by large cross-border banks on the ECB is of course too early to state, but European institutions will have to devote a great deal of attention in limiting concerted pressure efforts of large cross-border banks. It is also worth mentioning that larger international liabilities may strengthen a national regulator’s commitment not to bail out a troubled bank, and this would act as a disciplining device that is significantly weakened in presence of the centralized supervisor of a banking union.

This is a general principle well known in economics. In a first best environment, i.e. one in which decisions are efficient, adding any type of inefficiency, for example in terms of the organization of the decision process, is necessarily detrimental. This is no longer always the case in a second best environment. When decisions are in any case suboptimal (i.e. second best), additional distortions may actually improve the overall outcome. In these cases though the optimal design of the environment must rely on details and their possibly complex interactions.

3.2 Insiders vs. outsiders

Another, often mentioned, limit of centralized supervision is the lack of flexibility and the reduced ability to tailor the standards of supervision to countries’ economies and their banks. Although tailoring supervision has the flip side of a more likely capture of supervisors by large banks, it is clear that centralization works best for countries with similar and similarly developed financial systems, as discussed by Dell’Ariccia in this issue.

The decision of limiting centralization to homogenous national banking systems opens the door to another significant issue, namely how to deal with “outside” countries that are not currently part of the SSM, and their banks. This is addressed in Ferrarini, Hüttl and Schoenmaker and De Haas in this issue.

The SSM enforces mandatory participation for Euro-area countries, and contemplates voluntary and subsequent entry by other European Union members, currently Bulgaria, Croatia, Czech Republic, Hungary, Poland, Romania, Denmark, Sweden and the United Kingdom. The presence of these two groups, ins and outs, calls for an investigation of the actions of the SSM that may affect the out-countries and their incentive to join the SSM.

In light of the previous discussion, it is clear that both these dimensions depend on the level of financial integration and of financial flow between the two groups of countries. Scandinavian countries are deeply involved in outgoing flows in terms of banking activities towards SSM participating countries, and new accession countries are instead dependent on inward flows of banking activities from SSM countries (Hüttl and Schoenmaker in this issue). Hence, both these groups will likely benefit from the enhanced stability generated by SSM although on different dimensions. It is thus conceivable that for the “outs” the urgency to enter the SSM declines with its implementation. These countries may thus prefer to remain “outs” also to avoid losing their independence in supervising banks operating within their boundaries, even though there are large asymmetries in the costs and benefits of joining for individual countries depending on the extent and the direction of their links with the Union. The case is even more special for the UK because, although the international financial integration of its banking sector is significant, it is more related to third countries than to European Union countries.

A significant opportunity for all out-countries from joining the SSM would currently be the participation in the SSM’s governing bodies and the timely access to precise and complete information regarding the foreign activities of financial institutions that operate in their countries. Since these benefits largely depend on the actual functioning of the SSM, which is currently in its infancy, it is reasonable to foresee that we will not have many new accessions in the near future. And also that the possible negative externalities of the “outs” on the Eurozone will not be mitigated by the Single Supervisory mechanism^[2]Also, it should be taken into account that the Single Resolution Mechanism has a broader geographical application than the SSM, as it involves all countries in the European Union except Sweden and … Continue reading.

3.3 Microprudential vs. Macroprudential

Last, but certainly not least, a smooth functioning of the SSM requires to disentangle the possibly conflicting relationship between microprudential (MIP) and macroprudential (MAP) supervision. As argued by Angeloni in this issue, the main aim of the SSM is “to ensure the safety and soundness of European banks, both individually and as a system”. As such, “the SSM possesses both microprudential and macroprudential powers”, although the latters are shared between the ECB and national authorities of the member states.

In principles, MIP decisions aiming at addressing institution specific concerns may conflict with MAP objectives. In particular, system-wide MIP interventions, such as the Supervisory Review and Evaluation Process (SREP), can affect aggregate credit supply and therefore have first order MAP effects. Indeed, prescriptions may even go in opposite directions across the economic cycle: for example, during a downturn, with growing stress in banking markets, MIP might prescribe an increase in capital requirements, whereas MAP may suggest a reduction. Disentangling the relationship between MIP and MAP requires the definition of a clear and transparent ordering of possibly alternative policy objectives. Alessandri and Panetta make a strong case that MAP goals should precede MIP objectives, since “MAP authorities internalise the trade-off between capital and credit, whereas MIP authorities operating on individual institutions do not”. Aggregate welfare is therefore maximized by first addressing MAP concerns and then MIP problems.

However, this ordering introduces an additional reason for potential institutional conflicts between the ECB and the national authorities. Under the SSM, the ECB has full MIP responsibilities and, through the European Systemic Risk Board, retains some MAP powers to adjust the policy stance adopted by individual national authorities. In the case of MAP interventions, member states are left nonetheless a number of degrees of freedom, since the ECB has only the right to increase countercyclical capital buffers if it deems it necessary, but not to reduce them. In theory, one could even envisage a situation in which the ECB forces some MAP decisions, that are not taken by individual member states through, system-wide MIP interventions. A clear institutional setting and a strong coordination between the ECB and the national MAP authorities is therefore crucial, so that banks can foresee the supervisory stance that they are likely to face.

4. Impact on cross border and multinational banks

What will be the impact of the SSM on the European banking industry and on the European economy in general? With still so many inherent uncertainties and frictions in the SSM and even more in the implementation of the SRM and in the coordination of national deposit guarantee schemes, answering such a question at this early stage of the of the path to the Banking Union is a bit like tea-leaf reading. But it is an exercise that is worth trying, if anything, to uncover and address the possible pitfalls that may lay ahead.

A first issue is that the set-up of the SSM is instrumental to the implementation of the second and the third pillars of the banking union, which involve an increasing mutualisation of the resolution funds and, in perspective, probably also of the national deposit guarantee schemes. While with the SRM, in equilibrium, the burden of saving weak banks will be left to investors and to the industry itself, the SSM is certainly a precondition if some form of mutualised fiscal-back stops to banking crisis will ever be set up, and a proper European Guarantee Scheme will be forged in the future.

Even if we look backward, we can say that the launch of the banking union has been instrumental to the implementation of any serious structural mutualised fiscal fund like the ESM in the past. Therefore there is no doubt that the SSM has enhanced the capability of the European banking system to build up adequate weapons to face banking crisis. A first result of the SSM has been to strengthen the European banking industry with respect to what would have been otherwise, i.e. fragmented supervisory and resolution institutions and mechanisms.

A second issue relates to the financial trilemma. As convincingly argued by Schoenmaker (2012 and in this issue), it is impossible to have international financial integration and financial stability if supervision remains national. The objective of the SSM is therefore to guarantee an integrated and resilient European banking sector. A counterfactual exercise of what might happen without the SSM clearly points towards a retrenchment, a balkanization of the European banking industry.

If we take a step backward, a careful reading of the data during the crisis provides a very nuanced overall picture. It is difficult to conclude that either multinational banking (as defined by the activities of branches and subsidiaries based in foreign countries) or cross border-banking (as defined by cross border loans and deposits) seriously retrenched during the crisis with respect to domestic banking activities. They declined in absolute terms, but not much more than domestic banking.

Financial intermediaries have four major ways to expand abroad: direct lending to foreign non-banking clients, interbank lending to foreign banks, purchase of foreign assets such as government bonds, and setting up a foreign subsidiary, possibly capable of funding locally its lending activities (multinational banking). In all major euro area countries, in the aftermath of the financial crisis the share of assets of branches and (less so) of subsidiaries of banks from other members of the euro area registered a sharp contraction (Figures 4 and 5). On the other hand, even during the crisis, in all major countries the number of bank branches from other euro area countries continued to increase, although at a slower pace (Figures 1 and 2), while that of subsidiaries was relatively stable (with the visible exception of a decline in France, but which started in 2001, long before the crisis, see Figure 3).

Also for cross-border banking the picture is quite muddled. The value of loans to both other financial intermediaries and the real sector remained broadly stable (with the only noticeable exception of those made by Italian banks to banks in other euro area member states during the peak of the sovereign debt crisis; Figures 8 and 9). In contrast, there is some more evidence of a partial retrenchment of cross-border banking activities coming from the share of deposits from banks of other euro area member states in Italy and in France (Figures 10 and 11).

The evidence, even though rather blurred, shows therefore a structural resilience of cross border and especially multinational banking within the EU during the crisis. We still live in a fairly integrated banking market. The banking union, with all its problems and limits, will strengthen this pattern, although possibly not in a neutral way for all forms of international banking.

As argued by De Haas in this issue, cross-border banking certainly benefits from the SSM, but this may somehow crowd-out multinational banking. The renewed trust on the conditions of the balance sheet of foreign banks provided by SSM supervision will favour a revival of cross-border interbank lending. A second result of the SSM may therefore be a reversal of the recent increase in multinational banking and an expansion of cross-border lending.

A third issue relates to the average stance of supervision. As argued by Dell’Ariccia in this issue, supranational supervision may have tighter standards than national supervision. In the transition period, this may cause a drop in credit supply, hampering the still weak European recovery. In addition, since national authorities had different supervisory styles before the SSM, the centralization of supervision may have heterogeneous effects across the euro area. In other words, the tightening of the supervisory framework will be different in each country, depending on the initial distance from the ‘supervisory frontier’. And the playing field scenario can become even more uneven if, as argued above, national authorities can thwart the activities of the SSM by limiting its access to information.

However, in the long run banks will adapt to the new standards, the SSM will improve its ability to collect information, and the effect of the higher standards of supranational supervision are likely to prevail, reducing the possibility of opportunistic behaviours and regulatory arbitrage and, in turn, the likelihood of individual bank defaults and financial crises. A third result of the SSM should therefore be to increase the resilience of the banking sector, at least in the long run.

A related issue is that of the possible differences between the supervisory stance that will be faced by the 122 banks directly supervised by the ECB and that of all the other smaller financial intermediaries. Despite the centralization of bank supervision at the ECB level, a common SREP and a unique Guide to Banking Supervision, as explained by Angeloni in this issue, it is unlikely that a small bank in Finland will be supervised by the Finnish authorities in the same way as a small bank in Spain will be supervised by the Spanish authorities. The SSM will therefore have an asymmetric effect: it will level the playing field for the 122 large banks supervised from Frankfurt, but it will leave a more uneven playing field between small and large banks, because of the dual system that will emerge within each country. While a different treatment of Evli Pankki Oyj in Finland and Caja Rural de Villar in Spain is unlikely to cause bilateral competition concerns, this is not the case for the relationships between Nordea, a large multinational bank operating in Finland and supervised from Frankfurt, and the small Evli Pankki Oyj. The tighter standards of supranational supervision will add to the increasingly different burdens faced by large banks, possibly reducing their competitiveness with respect to smaller financial institutions.

This effect can be balanced by the fact that large European banks can capture the regulator, at least to a certain degree. In this sense, the likelihood for a small bank supervised at the national level that it will be let go bankrupt is still higher than that of a large bank supervised from Frankfurt. But the difference between the likelihoods of these two events is still smaller with the SSM than it was when all banks were supervised nationally. All in all, a fourth result of the SSM seems therefore be to that small banks will face a relatively more favourable environment than without and before the SSM^[3]Of course, the large confusion in the sharing of power between national and supranational authority during  the transition period can make this case rather unclear, as shown by the institutional … Continue reading.

The next important issue is the lack of flexibility of a common supervisor with respect to national specificities. As argued by Dell’Ariccia and above, centralization will have a smaller impact on countries with similar financial systems. This is certainly true in the transition period, when the structure of the financial system is a given.

But in the long run the financial system responds and adapts to new regulations. Financial systems across Europe will therefore become more similar as a result of the SSM. This, in turn, may have two additional effects. First, since the structure of the financial system affects that of the nonfinancial sector (Cetorelli and Strahan, 2006), the SSM may also have a sizeable impact on the real economy, levelling the playing field for firms’ access to the banking market and reducing the impact of national characteristics on the characteristics of European firms (e.g., size and leverage). Second, since banks are by and large the most important players in the European financial sector, a more similar banking system will de facto mean a more similar financial system. Non-bank intermediaries will face similar conditions in each European country, becoming themselves more similar across borders. The overall financial sector will become more uniform across Europe, and integration among similar players will become easier. During the transition period, the lack of flexibility of the SSM may indeed be a relevant issue. But in the long run it seems likely that the single supervisor will increase the convergence of the financial sectors of the European countries. This may be the fifth result of the SSM.

Finally, unified supervision may remove some hidden constraints that limit international bank integration and internal (within firms) capital markets (Focarelli and Pozzolo, 2001). As argued by Ferrarini in this issue, this may foster consolidation in the European banking sector. Bank profitability has not yet recovered after the crisis but, as shown also recently in the January 23rd issue of the Economist, there is no clear link between the drop in returns and bank size. Low profitability is pushing the whole sector towards a rather slow but relentless reorganization, mostly aimed at reducing operational and international diversification. However, as argued by De Haas, banks are not retreating uniformly from foreign countries and they are maintaining their closer and more strategic affiliates. The levelling of the playing field and the convergence of the European banking and financial sectors favoured by the SSM may trigger a cross-border consolidation process where more efficient and operationally focused banks acquire the activities of less efficient and excessively diversified banks abroad. This may be the sixth result of the SSM.

Summing up we are in a challenging new era for European banking. The ark of the Banking Union will probably provide help in sailing through difficult waters. But details need to be observed carefully, to prevent vicious waters sinking in unchallenged. We hope this issue of European Economy will be helpful to avoid them being even unnoticed.

---

References
Carletti, E., Dell’Ariccia, G., and Marquez, R., (2015). Supervisory Incentives in a Banking Union, IMF unpublished manuscript.

Calzolari, G., Loranth, G., and Colliard, J. E., (2015). Monitoring a multinational bank, Working Paper.

Cetorelli, N., and Strahan, P. E., (2006). Finance as a Barrier to Entry: Bank Competition and Industry Structure in Local U.S. Markets, American Finance Association, vol. 61(1), pages 437-461, 02.

Faia, E., and Weder, B., (2015). Cross-Border Resolution of Global Banks, SAFE Working Paper No. 88. Available at SSRN: http://ssrn.com/abstract=2572882 or http://dx.doi.org/10.2139/ssrn.2572882
Focarelli, D., and Pozzolo, A. F., (2001). The patterns of cross-border bank mergers and share-holdings in OECD countries, Journal of Banking & Finance, Elsevier, vol. 25(12), pages 2305-2337, December.

Schoenmaker, D., (2011). The financial trilemma, Economics Letters, Elsevier, vol. 111(1), pages 57-59.