Is there a future for international banks?

Dirk Schoenmaker 25 August 2013



The international, centralised, business model of banks has come under pressure after the global financial crisis. Supervisors are leaving their traditional consolidated approach, under which a bank as a whole is assessed. Instead, they are moving towards a stand-alone approach, under which the national subsidiaries are supervised separately. McCauley et al (2012) document this move from the international to the multinational bank model.

While the new Basel III capital and liquidity requirements are much needed, the local application of these new requirements makes them extra costly. Global banks will find their extra capital and liquidity holdings, scattered across different countries. These local holdings will also hinder the effective use of these buffers in times of need.

In my new book, Governance of International Banking, I explore why supervisors take this national approach (Schoenmaker 2013). Coordination failure in international crisis management, as witnessed with Lehman and Fortis, appears to be the root cause. I then go on to look at the costs of this approach and lastly to propose a solution for effective international governance.

Coordination failure

In a multi-country model, Freixas (2003) shows that national authorities only consider the local systemic effects of a bank failure. Cross-border externalities are ignored. Building on this analysis, is what I call ‘the financial trilemma’. This states that the objectives of financial stability, international banking, and national financial autonomy are not compatible (Schoenmaker 2013). One objective has to give.

Figure 1 shows the different trade-offs of the financial trilemma. The diagonal represents the first objective of financial stability; the y-axis the second objective of international banking; and the x-axis the third objective of national governance. Before the crisis, we tried to combine international banking with a national governance approach. This led to a breakdown of financial stability, as witnessed by the dramatic failure to resolve the problems at Lehman or Fortis in a co-ordinated way. This is point A in Figure 1.

Assuming that we want a stable financial system, there are two equilibrium outcomes. The first is moving back to national banks combined with national governance (point C in Figure 1). That solution is discussed in the next section. The second equilibrium is international banking coupled with international governance (point B in Figure 1), which we analyse in the last section.

Figure 1. Banking integration versus governance

Costs of local capital and liquidity rules

Recognising that international coordination is likely to fail, supervisors have concluded that they must resolve the local operations of banks. They have left the integrated, consolidated approach to supervision and apply capital and liquidity rules at the local level. International banks then become a string of national subsidiaries within a multinational bank. That is also the approach under the new resolution plans. Notwithstanding the policy rhetoric on the single point of entry resolution approach (FDIC and Bank of England 2012), banks are privately told (for example, by the US authorities) that their resolution plans work on the assumption that there is no cooperation in case of resolution.

At the IMF, Eugenio Cerutti and others (2010) have done research on the impact of local capital requirements for a group of 25 Western banks operating in Central and Eastern Europe. Under a scenario of a 2% decline in GDP and a 2% increase in interest rates, these banks need to raise extra capital of €45 billion in case of ringfencing. The extra capital needed is only €25 billion without such ringfencing. Another example is a large European bank. If this bank has to maintain local liquidity pools, it needs €20 billion extra liquidity. At a one percent opportunity cost for holding liquid assets, the annual cost amounts to €200 million.

These local liquidity and capital holdings will be trapped in the national subsidiaries, as the national supervisors want to keep these extra safety valves at the national level, in particular when a crisis hits and liquidity and capital should be directed to where most needed. It feels like not being able to use the firemen and water resources of a neighbouring village, when the village’s fire brigade is fighting a raging fire.

How to keep international banks alive?

The solution is a supranational approach for supervision and resolution. The bottom line is to arrange an appropriate fiscal backstop through burden sharing (see also, Goodhart and Schoenmaker 2009). The Banking Union in Europe is moving into that direction with proposals for a European Resolution Board, backed up by the European Stability Mechanism. The Banking Union will save the international banking model for the Eurozone.
Figure 2 shows the improvement in resolution. The starting point is that a recapitalisation is efficient when the benefits (in the form of financial stability) exceed the costs. The solid diagonal in Figure 2 represents the line where benefits (B) and costs (C) are equal. The left dashed line measures the home country benefits.1

Basically, the home country only considers their local share of the benefits, ignoring the cross-border impact of averting bank failure. In the diagram area A – where there is no bank recapitalisation since the costs outweigh the benefit – there is an efficient outcome. Area B – where there is a recapitalisation – is, similarly, efficient. Area C, however, involves inefficiency. This is where the recapitalisation would be socially efficient considering spillovers but the costs for one nation exceed the one-nation benefit. Put simply, the home country will not do the recapitalisation.

In the supranational approach (all benefits in the home country and the rest of Europe are incorporated by the supranational body), area C, which indicates the area of inefficiency, is smaller under the supranational line than under the home country line.

Figure 2. Recapitalisation (recap) in a supranational setting

An illustration

To illustrate this, suppose the cost of recapitalising an ailing bank is 100, while the benefits are 150 (point X in Figure 2). In the current situation, only the home country benefits are taken into account: 80 (that is, 53% multiplied by 150). Faced with a cost of 100, the home country decides not to recapitalise. This is in area C in Figure 2. Although recapitalisation is the optimal strategy (benefits exceed costs, there is no recapitalisation. By contrast, decisions taken under the supranational approach would see that the European benefits are 114 (that is, 76% multiplied by 150). This 76% of European benefits includes the benefits in the home country (53%) and other European countries (23%). As these benefits now exceed cost, the European body recapitalises the ailing bank, which is the efficient outcome. This is in area B in Figure 2.

Concluding remarks

Obstfeld (2013) correctly observes that Europe may represent a possible solution based on the sacrifice of national autonomy, however this seems unlikely for the rest of the world. Nevertheless, it is worth laying down a long term vision for international banking. In such a world, the Bank for International Settlements could play the role of supervisor for global banks (obviously the so-called ‘globally systematically important banks’) and the IMF the role of resolution authority for these banks. The IMF already enjoys the necessary fiscal backstop of its member countries. The monetary stability mandate of the Bank for International Settlements and the IMF would then be broadened to a monetary and financial stability mandate.

They would thus become true global central banks.


Cerutti, E, A Ilyina, Y Makarova, and C Schmieder (2010), “Bankers Without Borders? Implications of Ring-Fencing for European Cross-Border Banks”, IMF Working Paper No. WP/10/247.

FDIC and Bank of England (2012), “Resolving Globally Active, Systemically Important, Financial Institutions”, a joint paper by the Federal Deposit Insurance Corporation and the Bank of England, 10 December.

Freixas, X (2003), “Crisis Management in Europe”, in J Kremers, D Schoenmaker and P Wierts (eds.) Financial Supervision in Europe, Cheltenham, Edward Elgar, 102-119.

Goodhart, C and D Schoenmaker (2009), “Fiscal Burden Sharing in Cross-Border Banking Crises”, International Journal of Central Banking 5, 141-165.

McCauley, R P McGuire, G von Peter (2012), “After the Global Financial Crisis: From International to Multinational Banking?”, Journal of Economics and Business 64, 7-23.

Obstfeld, M (2013), “Crises and the International System”, International Economic Journal 27, 143-155.

Schoenmaker, D (2013), Governance of International Banking: The Financial Trilemma, New York, Oxford University Press.

1 Indicated by αh = 0.53. Home country benefits are given by αh times B. The numbers for the home country benefits and the rest of Europe benefits are averages for the top 30 banks in Europe (Schoenmaker 2013).



Topics:  Global crisis International finance

Tags:  banking, Too big to fail, Eurozone crisis, international banks

Professor of Banking and Finance at the Rotterdam School of Management, Erasmus University Rotterdam; Non-Resident Fellow, Bruegel; and Research Fellow, CEPR


CEPR Policy Research