From Fragmentation to Financial Integration in Europe

Chapter 24. Structural Measures for the New European Architecture

Charles Enoch, Luc Everaert, Thierry Tressel, and Jianping Zhou
Published Date:
December 2013
  • ShareShare
Show Summary Details
Enoch Charles

As noted in other chapters in this volume, much has been accomplished in strengthening banks in Europe, and actions are in train to identify and address remaining gaps.

Nevertheless, there is a view—both in the United States and Europe—that measures taken so far are not enough. One response is to propose a qualitative jump in the prudential requirements imposed on banks. In the United States, much of the focus is on leverage ratios: while the average leverage ratio in the United States presently is around 3 percent, there are Congressional proposals to raise it up to 15 percent. In the United Kingdom, requirements up to 10 percent have been suggested. Meanwhile, the head of the Scientific Committee of the European Systemic Risk Board has argued that capital requirements of perhaps 30 percent might be needed to make the banking system safe.

Prospects for enactment of increases of such magnitude are at the moment not realistic. Banks lobby heavily against them; and, at a time of concern about lack of credit growth, policymakers listen to arguments that such imposts would be heavily burdensome and would serve to depress lending and economic activity. In any case, there is far from consensus that such levels would be appropriate and not counterproductive. More limited, albeit still ambitious, measures may well be enacted. For example, on July 9 joint proposals by the U.S. Federal Reserve Board, the Federal Deposit Insurance Corporation (FDIC), and the Office of the Comptroller of the Currency were published for a 5–6 percent leverage ratio for the eight largest banks by 2018.

Absent full protection of the banks, there is recognition that one day banks will fail again. Indeed, a regulatory regime that did not permit failure would be too tight for banks to properly undertake their activities.1 Thus, attention needs also to be on resolvability in case a bank does fail. In that connection, the Financial Stability Board has been working on developing the “Key Attributes” for a resolution regime, under which authorities would be able to resolve even the most complex banks should this become necessary. Within Europe, the prospective Recovery and Resolution Directive and its associated regulations will mark a fundamental step for enhancing the resolvability of banks.

Again, while progress in this regard has been impressive, the challenges ahead remain daunting. Resolving large cross-border institutions will be highly complicated. In addition, given the wide consensus that such resolution should in the future not be simply a public sector bail-out, significant policy and institutional change may be necessary. Work on increasing resolvability is proceeding on many fronts.

In this regard, policymakers are looking at bifurcating the institutional and regulatory regimes. The emerging legal and regulatory infrastructure may well be able to cover the great majority of bank failures, particularly in normal times. For the largest banks, more may be needed.2

As regards these largest banks, both within Europe and outside there is focus on the dual nature of banks, between “traditional” banking that provides essential deposit and payment services to the public, and investment banking that is deemed more risky and which is not as directly connected to basic financial infrastructure. Having the two activities linked within a single entity may lead to spillovers because the riskiness of investment banking may also deplete the financial soundness of the traditional banking part. Since it will be widely assumed that the traditional banking part may not be allowed to fail, there also will be an implicit subsidy on the risky part that will reduce its costs and allow it to undertake more activities than if it had to finance itself on a standalone basis. Moreover, if the joint entity does fail, the necessity of keeping the basic infrastructure operating will constrain the authorities in how they deal with the entity, potentially making it irresolvable except through the traditional public sector bail-out.

A separate argument for the toxicity of having both types of activity within a single organization draws attention to the respective cultures of the two types of institutions. Within the traditional bank, the emphasis may be on soundness and safety, and remuneration is relatively modest, while on the investment side short-term profit maximization is overriding, and remuneration for successful traders is a high multiple of that of their traditional-side colleagues. The resulting tendency for those who make their marks in the investment side of the banks to run the organization as a whole may lead to policies that jeopardize the traditional emphasis on soundness in the commercial banking side.3

In response to similar concerns in the 1930s, the Glass-Steagall Act in the United States required that commercial banking activity be undertaken by separate entities from those involved in investment banking. Notwithstanding that there had been extensive seepages across the boundaries, Glass-Steagall was not repealed until 1996. Although its repeal was followed by the major accumulation of banking sector risks that ultimately contributed to the financial crisis, there is no support for a full return to Glass-Steagall. Banking business nowadays is far more complicated and interrelated than in the 1930s, and the prospect of full separation of activities would be much more complicated and disruptive.

In the United Kingdom, following proposals by a commission headed by John Vickers, legislation is in train to require banks to separate their domestic retail banking from their other activities.4 Unlike the “Volcker Rule” incorporated in the Dodd-Frank Act in the United States, under which banks have to divest proprietary trading, such activities can remain within the same banking group under Vickers, albeit separately capitalized. The aim is to ensure “super resolvability” of the domestic retail bank. Only U.K. business is separated out. Thus, cross-border resolvability issues do not come into play.

Meanwhile, Michel Barnier, European Commissioner for Internal Markets and Services, established a high-level committee under Finnish central bank governor Erkki Liikanen to consider whether the various measures put in place to strengthen the banking system were sufficient or more would be needed. The Liikanen group offered a number of recommendations, the most far reaching of which is to require that banks above a particular size divide their activities between their commercial banking and their trading and investment banking functions into separately capitalized units (see Liikanen and others, 2012). Like Vickers, Liikanen would allow both types of activity to remain within the same banking group. Unlike Vickers, the Liikanen proposals cover cross-border as well as domestic operations, so issues of cross-border consistency come into play. Liikanen allows underwriting and trading business to remain with the retail business, while Vickers does not. Liikanen considered whether to make the proposal apply just to banks that had certain characteristics but determined that it should be mandatory for all banks in the EU, subject only to a minimum size threshold.

Given the perceived urgency of moving forward and political pressures to be seen to act, a number of European Union (EU) members, including France and Germany, have moved ahead to introduce proposals at a national level, even before the EC has prepared a common EU position. French President François Hollande had included in his election manifesto a commitment to segregate investment banking activities. In the end the French proposal—similar to Liikanen—avoided full separation of investment banks and preserved the so-called “universal banking model.” It is lighter than Liikanen, however, as it would permit banks to retain within the commercial banking side those commercial banking activities that are considered linked to the real economy, such as market-making, investment and financing services to clients, hedging, and asset management activities.5 The German proposal (recently passed into law by parliament) has been closely coordinated with that of France and is along similar lines, although again allowing limited extra activity to remain within the commercial banking side of the institution. Both the French and German governments have indicated that, if necessary, they will modify their proposals to be fully in line with EU requirements.

There is some support among policymakers and observers for moving in this direction, as part of a “belt and braces” approach for enhancing the soundness of the EU financial system. Even so, the proposals have also drawn criticism, including from the banks.6 The first priority in any case is that such measures should not be seen as substitutes for stronger capital and other buffers, but rather as complements and serving a somewhat different purpose. The second priority is that any transition be managed carefully, so as not to disrupt banking business in the meantime and not expose the bank to additional risk. A recent IMF paper argues for a full cost benefit analysis, including on cross-border costs of introducing Liikanen-type policies.7 The EC has undertaken that it will include an impact study when it presents its proposals in late 2013.

Three further issues remain. First are the various aspects of the cross-border impact of such separation. With differences in structures likely among EU members (not to mention also U.S. banks), there would be scope for regulatory arbitrage: if some activity is permitted within the commercial banking part in one jurisdiction and not another, and if that is deemed to give it a financial advantage, then such activity may gravitate to jurisdictions where such activities are permitted. This would not be consistent with the single market and could be an added element toward fragmentation. The actual impact cannot be determined before the EC proposals are presented, including the degree to which harmonization will be required within the EU and the extent to which existing proposals will be modified in line with these. The more the EC proposals are merely an umbrella under which various national proposals can be accommodated, the greater the scope for arbitrage. On the other side, the more the EC proposals accommodate national preferences for including particular activities within the commercial bank side of the operations, the less the need for banks to arbitrage, but also the lower the impact of the measure. In principle, it seems that the EC could determine a set of measures that are fully consistent with those proposed by the various national authorities, although in sum they would require banks to set up quite complex structures to meet all national requirements. An example would be if the tighter separation envisaged by Vickers only for U.K. domestic activities were to be applied on cross-border banking activities, as advocated in Liikanen.

There may be cross-border impact also from the application of a minimum size threshold as the requirement for a bank to segregate its business. A single minimum size affecting only the large banks would have a markedly diverse impact across Europe. A country such as France, with a small number of very large banks dominating the sector, would be heavily affected; Germany, with both large and small banks, less so; and countries with no large banks (particularly if a single metric is used across Europe) probably not at all. A single minimum size based on trading volume or market share, by contrast, would have even more impact across countries since, within the major banks’ universe, it would not affect smaller capital players.

The second issue relates to the possibility that segregating “risky” business from commercial banking activity may serve to push the former into the shadow bank sector. If, as argued by Liikanen, risky activities benefit from their cross-subsidization with commercial banking activities, then once this possibility is removed there will be less incentive to retail within an overall banking group, and such activities may be spun off. In answer to this, policymakers have stressed that the investment banking activities themselves will require a banking license, so spinning off out of the regulated banking sector will not be an easy option. In addition, if indeed there is spin-off into shadow banking, the response cannot be to seek to prevent this by restoring the cross-subsidization. Rather, it would be necessary to expand the perimeter of regulation and supervision so that soundness is maintained in the shadow banking sector, too.

Finally, there is the question as to whether all the disruption that would be involved in enforcing Liikanen-type proposals actually would enhance resolvability.8 Much of recent bank distress has emerged from the core of commercial banking, as banks overinvested in local real estate lending or in government bonds. On the other side, where authorities have been prepared to close an institution that had no commercial banking function, in particular Lehman Brothers, this was in hindsight a significant contributor to setting off the global financial crisis.9

There is no single answer, and no quick solution. Liikanen-type proposals do not of themselves ensure resolvability, let alone bank soundness. The problem of excessive lending for real estate needs to be handled at a prior level. The legacy issue of bank purchases of bonds of governments they previously considered safe needs to be resolved by fiscal adjustment so that government finances are restored to soundness, as well as, in time, possible mutualization of debt obligations, so that bank holdings are of institutions whose soundness is not in doubt.

There is also the issue that, even when segmented, resolving a large cross-border bank will be complicated and difficult. In this connection, the recent agreement between the FDIC and the Bank of England may show a way forward for the large banks in the EU. Under this agreement, banks would be required to establish holding companies with the banks as subsidiaries under them. In the event of resolution, the authorities would work on a “Single Point of Entry” basis on the holding company, allowing the bank subsidiary to continue to function while eliminating shareholders and others subject to “bail in” at a holding company level. If the triggers are set at a sufficiently high level, if intervention is mandatory, and supervision is strong enough to identify the deterioration in the bank’s position at an early enough stage, then resolvability will be more assured. But such an approach requires that banks are organized nationally as subsidiaries of holding companies, and that all the group’s borrowings are conducted at the level of the holding company. This proposal clearly has merits, but it implies a cross-border structure for the banks that does not cover all affected banks at the moment and may not be the best suited for all types of cross-border banking.

In short, although full cross-border analysis of the costs and benefits of enforcing structural measures in line with those proposed by Governor Liikanen is still awaited, there may be a case for such measures as an added instrument in the “toolkit” for establishing and managing a sound financial system. Cross-country consistency will need to be assured, and introduction of such structural measures should not be at the expense of ongoing work to strengthen banks’ soundness and increase resolvability. Many of these qualifications, however, are short-term and operational in nature. In the longer term, if the structural changes can be implemented in Europe in a way that ensures consistency without just, in effect, validating a patchwork of disparate initiatives, there could be merit in a European approach in this area to reduce the riskiness and enhance of resolvability of the Union’s largest banks.


    ChowJ.and J.Surti2011Making Banks Safer: Can Volcker and Vickers Do It?IMF Working Paper No. 11/236 (Washington: International Monetary Fund).

    GambacortaLeonardoand Adrian VanRixtel2013Structural Bank Regulation Initiatives: Approaches and Implications” (Basel: Bank for International SettlementsApril).

    LiikanenErkki and others 2012Report of the High-Level Expert Group on Reforming the Structure of the EU Banking Sector” (Brussels: European UnionOctober). Available at:

    ViñalsJoséCeylaPazarbasiogluJaySurtiAdityaNarainMichaelaErbenovaand JulianChow2013Creating a Safer Financial System: Will the Volcker, Vickers, and Liikanen Structural Measures Help?IMF Staff Discussion Note No. 13/4 (Washington: International Monetary Fund).

That said, a number of policy innovations also serve to reduce the risk of crisis. See, for instance, Chapter 20 of this book on developments in the design and implementation of macroprudential policies.

The Financial Stability Board has identified 28 globally systemically important financial institutions, for which additional capital buffers are proposed. As noted below, the European Commission (EC) is considering the thresholds above which banks should become subject to a special regulatory regime.

This argument is put forward most strongly by Paul Volcker, former Chairman of the Federal Reserve, as well as by John Reed, who oversaw the merger of Citibank with Travelers to create Citigroup.

Work is also in train in the United Kingdom, as part of the banking bill, to “change the culture of banking” through instituting more individual accountability for the heads of failing institutions.

Under Liikanen, the provision of hedging services to clients would also remain part of the retail banking arm.

A Bank for International Settlements study looked at the implication of Liikanen-type proposals for banks’ business models (see Gambacorta and Van Rixtel, 2013).

Viñals and others (2013) provide a decision tree framework that can be followed to determine whether to adopt structural measures such as those under consideration in the first place.

There are differences among bankers as to how far Liikanen-type proposals would actually disrupt existing banking structures, with some banks claiming that they already have quasi-Liikanen segregation in place.

For empirical evidence on the link between trading activities and banks’ vulnerability, see for instance, Chow and Surti (2011).

    Other Resources Citing This Publication