VoxEU Column EU institutions Financial Regulation and Banking

Completing the Eurozone banking union: What must be done

Most commentators agree that a European banking union would end the ‘deadly embrace’ between creditors and governments. This column argues that a banking union would be welcome, but that current proposals are dogged with problems. To resolve these, we should stop discussing debt restructuring and instead enhance the borrowing capacity of the European Stability Mechanism. A programme to buy capital in financial institutions unable to raise it directly on the market should also be set up.

Most commentators argue that the banking union and a full diversification of banks’ assets across borders are the first steps to a safer Eurozone and the elimination of the ‘deadly embrace’ between creditors and governments. Then, a set of additional steps could follow – the ECB as lender of last resort, a credible implementation of a no-bailout rule, and, possibly, a plan for the restructuring of national debts, as forcefully stated by Eichengreen and Wyplosz (2016).

We think this view is overly optimistic. A successful implementation of the banking union depends crucially on how we deal with the legacy of the financial crisis, i.e. a rising geographical segmentation of the banking industry and an over-exposure to domestic risks, and on the design of the ‘other’ steps of the road map to a safer Eurozone, i.e. the role of the ECB and European Stability Mechanism as backstops of systemic financial disruptions. Because of these problems, the implementation of the banking union is far more difficult than previously estimated.

In this column, we look at the current debate in Italy over the sustainability of the financial sector to make suggestions about how the road map for the Eurozone stabilisation could move forward. The following points summarise our view:

  • Allowing (officially) the ECB to act as lender of last resort and the European Stability Mechanism to borrow from the market by issuing Eurobonds must be part of the design leading to an effective banking and capital market Union;
  • Eurozone-wide resources may be used to buy equity of financial institutions, not necessarily insolvent, that require new capital during the transition to a full banking union, following the experience of the Troubles Asset Relief Program in the US.

The Italian consensus

According to Italian banking lobbies and regulators, Italian banks are sufficiently solid. The recent changes in capital and leverage ratios, impairment provisions, risk weights, etc. imposed by the European Banking Authority and the ECB are too severe, have generated regulatory uncertainty, contributed to the high volatility of banks’ market valuations and, ultimately, have laid the ground for a credit crunch and future systemic crises. In his recent address to the ASSIOM FOREX Congress, the Governor of the Bank of Italy openly criticised the Bank Recovery and Resolution Directive, effective since January 2016, claiming that it unfairly affects banks’ debt retroactively. More importantly, Italian regulators, and the government, are particularly concerned about the possibility (discussed in European circles) that sovereign debt loses its risk-free status in the computation of regulatory capital, or that European banks may be forced to diversify their holdings of government securities. Italian banks hold about €400 billion (or 19%) of domestic public debt – this is why the Italian government is against any caps on their holdings.

The moderate confidence expressed by Italian regulators and banking lobbies appears to be at odds with market sentiment, pressures from European agencies, and financial data. Investors value the stock of non-performing loans no more than 50% of book value. Since the 2014 Asset Quality Review (AQR), Italian banks raised a substantial amount of capital to reach the CET1 capital targets imposed by European regulatory agencies (for example, the country average is now 11.5%, still 1.3 percentage points below the EU average). Even though the coverage ratio for non-performing loans is now close to the EU average, and the increase in non-performing loans is finally stabilising, serious problems remain in terms of risk exposure, stock of non-performing loans, and profitability. For example, while the non-performing-loans-to-loans ratio is 5% in the Eurozone, it is 16.7% for the Italian banks monitored by the European Banking Authority. These banks have return on regulatory capital of 5.1%, against 9.1% in the rest of the EU, and trade at half their book value, against an average of 0.8 for the Eurozone. In addition, domestic public debt as a fraction of total assets stands at 10.5%, against the EZ average of just 4.2%. Absent a Eurozone-wide systemic event, rather than insolvency risks, the biggest problems of Italian banks are the inability to reduce the stock of non-performing loans without a substantial recapitalisation, and the over-exposure to domestic risks.

To some extent, the arguments supporting the Italian consensus derive from the conflict between two objectives: designing the right institutions for the long run, and avoiding the possibility that their implementation could make the adjustment process painful and destabilising. A second concern is that new rules do not address systemic risk, while stripping governments of the power to rescue domestic banks. The Bank Recovery and Resolution objective to transfer risks from governments to markets is considered illusory. In fact, absent a central authority with the power to mobilise sufficient resources, should a systemic event materialise, country and domestic banks’ risks remain highly correlated.

Although some of these concerns have merit, they also reveal a dangerous departure from established principles of a sound economic integration and the role of a supra-national authority.

First, the case for reducing banks’ over-exposure to domestic public debt is strong. In fact, there is no scenario, unless countries’ public debts are consolidated at the central level, in which interest rate spreads within the Eurozone would totally vanish. In theory, it could be useful for the ECB to act as lender of last resort and commit to crisis prevention in case of a run on Eurozone sovereign debts. But this is true for liquidity problems, not insolvency and fiscal profligacy. Italian banks’ over-exposure to domestic government debt leaves them more at-risk and jeopardize the implementation of a full banking union. On the other hand, risk-adjusted banks’ profitability should not be negatively affected by a push toward higher diversification as high spreads correspond to more risk.

Second, EU-centralised regulation is important to address bad behaviour of banks and insufficient supervision. Even though the problems of Italian banks are largely due to the deep recession, it is hard to deny the responsibilities of their management and, ultimately, of their weak governance. Exactly because controlling shareholders – often with empty pockets – tried their best to avoid new equity issuance that would have diluted their control, Italian banks, over the last years, sold substantial amount of bonds to small investors who trusted them and believed, or were led to believe, that there was an implicit state guarantee. Therefore, moral hazard is not just a secondary issue that concerns only economists and alike and, without the rules imposed by the European Banking Authority and the ECB, the Italian banking system would be in significantly worse shape. In addition, in Italy (as well as in other Eurozone countries) there exists a large constituency with a strong ‘domestic bias’ – for example, public opinion is strongly influenced by the fears of local stake/stock-holders losing control of banks; the concern that domestic savings would finance foreign, and not domestic, borrowers, and by a general aversion to the very idea of a capital market union (i.e. a system characterised by cross border ownership and assets).

What can be done?

Although the Bank Recovery and Resolution Directive rules and the reduction of the over-exposure to domestic public debt are necessary, their implementation in the absence of some complementary measures would require too much time and effort. Along the transition, banking activity could experience serious troubles possibly jeopardising the final full recovery of the Italian economy. The very recent institution of the government-sponsored Atlante Fund, a privately managed investment vehicle that should buy non-performing loans and capital of troubled financial institutions, exemplifies these risks. In fact, its main shareholders, the largest Italian banks and insurance companies together with the state sovereign fund, stated that they plan to invest in shares and assets they believe are ‘under-valued’ by other investors and European institutions. The risk is that the fund exacerbates the exposure to domestic risks of financial institutions without addressing their under-capitalisation.

Below is a list of complementary measures that could speed up, or save, the of the banking union.

  • First, we should stop discussing debt restructuring and instead, as suggested by Pisany-Ferry (2016), enhance the borrowing capacity of the European Stability Mechanism to provide guarantees for effective fiscal stabilisation in case of severe recessions in countries with limited fiscal capacity. Eurobonds could resolve the uncertainty about the ability of Eurozone institutions to manage macroeconomic risks, as the ECB unconventional policies, in a time of low inflation and zero interest rates, become ineffective.
  • Second, a programme to buy capital in financial institutions unable to raise it directly on the market, using Eurozone-wide resources, should be set up. This recapitalisation fund, following the experience of the Troubled Asset Relief Program in the US, would use public resources to buy non-voting shares, with senior status, in financial institutions, conditional on the full implementation of the Bank Recovery and Resolution Directive and the diversification of the exposure to sovereign risk. The recapitalisation fund should sustain all banks, not necessarily insolvent ones, that need to raise equity along the transition to the banking union without triggering the Bank Recovery and Resolution Directive clauses, and supplement the European Stability Mechanism that can already inject equity in systematically important institutions in a crisis, but only after existing stakeholders are penalised according to the Bank Recovery and Resolution Directive rules.

Eurozone countries with more solid banks could oppose these two tools, arguing that taxpayers’ money would be used to subside banks in the Eurozone periphery. However, the alternative is a continuing instability of the monetary union, whose dissolution would be very costly for all member countries. Since the idea of a EU recapitalisation fund is novel, we spend the remaining part of this column justifying its proposal, and refer to Pisany-Ferry (2016) and other commentaries in Baldwin and Giavazzi (2016) for convincing arguments in favour of enhancing the borrowing capacity of the European Stability Mechanism.

  • First, recapitalisation has the benefits of breaking the ‘deadly embrace’ between sovereigns and banks without violating state-aid rules as any banks can rely on it. In particular, if banks obtaining funds through recapitalisation must comply with tough rules imposed by the EU supervising authorities and reduce exposure to domestic risks, then the risks faced by recapitalisation are limited, and the moral hazard and the deadly embrace problems would be reduced quickly.
  • Second, given the high leverage of financial institutions, the resources required to buy equity in Eurozone banks are far smaller than those required to buy troubled assets directly. Even though banks will require more capital along the transition to a full banking union, they could find it expensive because stock prices tend to be low when uncertainty is high. To speed up banks’ recovery, and avoid losses of taxpayers’ money, some constraints should be considered till banks repay public funds: limits to dividend distributions, caps to management compensation, increasing the number of independent board members. In addition, the repurchasing price of recapitalisation stake should increase over time.
  • Third, the institution of recapitalisation could postpone the creation of the European Deposit Insurance Scheme to a time when banks are safer, more capitalised, and less exposed to domestic sovereign risk.


Baldwin, R. and F. Giavazzi (2016), How to Fix Europe’s Monetary Union, A VoxEU.org eBook.

Eichengreen, B. and C. Wyplosz (2016), “Minimal conditions for the survival of the euro” in R. Baldwin and F. Giavazzi, How to Fix Europe’s Monetary Union, A VoxEU.org eBook.

European Banking Authority (2015), “EU wide Transparency Exercise”, report, November.

Pisany-Ferry, J. “The Eurozone’s Zeno paradox and how to solve it”, in R. Baldwin and F. Giavazzi, How to Fix Europe’s Monetary Union, A VoxEU.org eBook.

1,155 Reads