Banking union and the financial architecture of the Eurozone: A retrospective

The Eurozone crisis has brutally exposed the inadequacy of institutional financial arrangements in the Eurozone. Modalities of banking union are being discussed now and the late June 2012 Eurozone summit agreed to give the ECB supervisory powers and, once this common supervisor is in place, it will allow direct capital injections into banks using the European Stability Mechanism. The European Commission will release a more concrete proposal on 11 September.

Other necessary elements of a banking union are a common Resolution Agency (RA) of troubled banks and a common Deposit Insurance Fund (DIF). Indeed, the central bank is the liquidity authority but it cannot be at the same time the solvency authority since bank restructuring may have fiscal consequences. The fact that a monetary union needed common stability and regulatory facilities is not news. This was pointed out, for example, by Folkerts-Landau and Garber (1994) as well as in an early CEPR report (Chiappori et al. 1991), and several authors, academic networks, and think tanks have insisted overtime on the issue. Twenty years ago I proposed the following architecture:

1) The ESCB (European System of Central Banks) should perform a Lender of Last Resort function if the stability of the European financial and payment system is to be preserved.

2) The Lender of Last Resort function of the ESCB needs to have associated supervisory powers, although, perhaps, the ESCB need not have them in the exclusive.

3) The concern for a potential misuse of the Lender of Last Resort facility by a ESCB with supervisory powers is legitimate but not overwhelming. Indeed:

4) A potentially optimal structure could be for the ECB to have authority in liquidity matters while another European agency has authority over solvency matters (and perhaps deposit insurance). In this arrangement both agencies would have supervisory powers but the ECB would have the primacy." (Vives 1992)

Ten years later, after essentially no progress had been made, I concluded that "Financial regulation has traditionally been changed and developed in response to crisis. The examples of the great depression of the 1930s, the S&L's crisis, or the recent international financial crisis come to mind. The question is whether the EU will wait for a major crisis to develop an adequate stability and supervisory frame, or whether it will be able to overcome the political obstacles to relinquish national control and move ahead with the needed reforms." (Vives 2001b)

After 20 years regulation seems to be moving now in the indicated direction after, or better in the mist, of a major crisis when sovereign default looms large.

The optimal financial architecture design is the outcome of a delicate balance between the trade-offs arising when assigning regulatory functions to different institutions. The academic literature can contribute to clarify the issues at hand. Monetary policy may conflict with the Lender of Last Resort function, with the latter function abused with inflationary consequences, while at the same time it may show informational synergies. More seriously, the Lender of Last Resort function may display informational synergies with macroprudential supervision as well as conflict. For example, with an ECB without supervisory powers a national govern¬ment can pressure the central bank to provide help for ailing national institutions based on information privy to this national authority. On the other hand, the so-called Bernanke's put was perceived as provoking asset inflation with its deleterious macroprudential consequences. Furthermore, concern for the reputation of the central bank as supervisor may encourage a lax attitude, using excessively the Lender of Last Resort facility so that bank crises will not put its supervisory capacity in question. Other trade-offs may be present between supervision functions. For example, the authority responsible for the authorisation and supervision of an entity may be reluctant to accept that the entity must be closed since closure would reflect unfavourably on the regulator's capacity. The reasons for separation of powers in regulation range from providing appropriate incentives to regulators, improving the measurability of tasks and focus of the regulatory agency, introducing competition between agencies to generate information, to making capture and collusion between regulator and industry more difficult. (See e.g. Goodhart and Schoenmaker 1995 and Vives 2001a and the references therein.)

The UK attempted to resolve the trade-offs with the now failed FSA model. In this model the central bank is weakly responsible for the stability of the financial system and has some (weak) monitoring capacity; while the FSA authorises and supervises financial institutions, is responsible for supervision of banking, insurance and securities, and for solvency problems of individual institutions. A MOU between the UK Treasury, the Bank of England (BOE) and the FSA was supposed to regulate what to do in a crisis but the process failed miserably in the Northern Rock case. Competition policy was left the province of the general competition authorities. The UK is now moving to a 'twin peaks' structure where the BOE is in charge of both macro- and micro-prudential supervision and a new agency (FCA) will deal with conduct regulation including competition, market integrity, and consumer protection. This new framework seems to strike a reasonable compromise and provides insights into a desirable Eurozone architecture.

In the Eurozone the ECB has taken up in practice in the current crisis the Lender of Last Resort function of helping solvent but illiquid banks. A corollary of this function is that the ECB should have also supervisory powers but that those need not be exclusive. A possible design building on the received academic literature could be the following. The ECB should be responsible for the systemic stability of the financial system and macroprudential regulation and supervision. It could also authorise financial institutions. A RA with supervisory powers should be in charge of solvency matters, closure decisions, and also deposit insurance for individual institutions. This would correspond to a model close to the successful US FDIC except that the RA should be closely coordinated with the EU competition authority in charge of controlling state aid.

However, there are several further issues that need to be resolved. Should the ECB supervise all banks in the Eurozone or only the systemic and cross-border ones? How should banks outside the Eurozone be supervised? How to structure the financing and the ex ante burden sharing agreements backing up the RA? Who should regulate markets?

In the long term it would be better for the ECB and the FDIC to supervise all banks in the Eurozone in order not to create two tiers of institutions. Recall that if many small banks correlate their strategy they may become a systemic problem. Banks outside the Eurozone could be supervised by the current supra-regulator EBA with general remit to preserve and deepen the single financial market. To this the market-wide regulator in the EU could be added to build a European Financial Services Authority. In the Eurozone, where there is no single treasury, the RA/DIF should be financed with levies on banks, with a backstop agreed among the governments before a crisis strikes. Levies or insurance premia on banks should be calibrated to the perceived risk positions of institutions according to market indicators such as credit-default spreads. Flat premia would merely induce cross-subsidisation of risky banks by safe ones. Following the FDIC model, the agency should be bound by a prompt corrective-action procedure to avoid the regulatory forbearance that we have witnessed so many times in banking crises, from the savings and loan crisis in the US in the 1980s to Japan in the 1990s, with Spain as the latest example. The agency should limit taxpayers' exposure by wiping out shareholders and subordinated debt holders if needed in a restructuring procedure.

The banking union project is a long-term one and it cannot imply the ex post mutualisation of banks debts. Indeed, insurance cannot be arranged once a crisis is ongoing, because solvent countries and banks should not pay for insolvent ones except in so far this is needed to preserve systemic stability. A European DIF would address the next crisis, but not this one, while a European RA could start functioning with funds from the European Stability Mechanism (ESM).

Banking union is necessary but not sufficient for monetary union to survive. There is no European deposit insurance fund that could sustain a run on deposits in Italy if it was feared that this country would leave the Eurozone. Indeed, to cope with sovereign risk a high degree of political and fiscal integration is needed.

Chiappori, P, C Mayer, D Neven and X Vives (1991) “The Microeconomics of Monetary Union”, in Monitoring European Integration: The Making of Monetary Union, CEPR.

Folkerts-Landau, D and P Garber (1994), "The ECB: a Bank or a Monetary Policy Rule", in M Canzoneri, V Grilli and P Masson (eds.), Establishing a Central Bank: Issues in Europe and Lessons from the US, Cambridge University Press.

Goodhart, C and D Schoenmaker (1995), "Should the Functions of Monetary Policy and Banking Supervision be Separated?", Oxford Economic Papers, 46:539-560.

Vives, X (1992), "The Supervisory Function of the European System of Central Banks" in Giornale degli Economisti e Annali di Economia, October-December, 51(9-12):523-532.

Vives, X (2001a), “Central Banks and Supervision with Application to the EMU” in A Santomero, S Viotti and A Vredin (eds.), Challenges for Modern Central Banking, Kluwer Academic Publishers (USA), 95-113.

Vives, X (2001b), “Restructuring Financial Regulation in the European Monetary Union”, Journal of Financial Services Research, 19(1):57-82.