Ladies and gentleman,
Thank you for inviting me to speak here today in Dublin, which is an apt place to discuss the future of banking.
People in this country need no reminding of the dangers that an over-leveraged banking sector can pose to the real economy. And I expect you need no persuading of the importance of fundamentally strengthening our common approach to supervising and resolving banks in Europe.
Nevertheless, I would like to begin my remarks today by reiterating why we need banking union. Governments are in the midst of crucial negotiations about the shape and structure of banking union, and I think it is important to keep in mind the big picture of why this project is so important for our common future.
Thereafter, I will discuss two aspects about the future of banking.
The first is how the Single Supervisory Mechanism (SSM) will affect the supervision and the practice of banking in the euro area.
The second is how the process of structural change in euro area banking sector will affect intermediation, and what this will mean for Europe’s traditional universal banking model.
The rationale for banking union
Let me begin by briefly recalling why banking union – which includes the Single Supervisory Mechanism and the Single Resolution Mechanism (SRM) – is an essential complement to monetary union. There are three main reasons.
First, monetary union implies increasing integration between financial institutions and markets, be it through internationally active banking groups, bilateral trading exposures, or presence in the same market segments. We saw this process unfold in the euro area over its first decade. But such financial integration also creates what Dirk Schoenmaker has termed a ‘financial trilemma’: it is impossible to maintain the level of integration required for monetary union while practicing national supervision, without putting European financial stability at risk. 
Why? Because national supervisors have a national mandate and so it is rational that they take less account of externalities. This leads to the under-provision of European financial stability as a public good. One example of this is the promotion of so-called “national champions”, which may increase national welfare but whose failure can create spillovers for other countries. The incentives of a European supervisor, on the other hand, are fully aligned with its European financial stability mandate.
A second reason why banking union is essential to monetary union is that it supports the implementation of the single monetary policy. The even transmission of monetary policy across all member countries of the euro area requires a level of financial integration that ensures well-functioning cross-border money markets. Yet we have seen during the crisis that without a euro area-wide approach to financial governance, financial markets can end up re-nationalising.
A strong banking union can help mitigate several sources of this fragmentation. A single supervisor should enhance transparency and thus increase trust in cross-border lending, while supervisory ring-fencing or national asset-liability matching would no longer be relevant given the SSM’s European focus. In addition, an effective single resolution mechanism would help attenuate the infamous bank-sovereign nexus. Sovereigns would have less ability to intervene in failing banks, thus allowing bank risk to be better separated from sovereign risk.
The third reason that justifies the banking union has to do with the effects of unsupervised cross-border lending on real economic developments. Contrary to the “it’s mostly fiscal” view of the crisis,  financial sector developments largely explain the build-up of unsustainable current account and competitiveness positions in peripheral countries before the crisis.
The current account deficits in most peripheral countries were in fact led by huge capital inflows implying capital account surpluses. The exposures of banks from core to peripheral countries more than quintupled between 1999 and 2008. Competitiveness losses were simply the mechanism that connected the capital surplus and the current account deficit – that is, an appreciation of the real exchange rate caused by economic over-heating. As John Williamson explained long ago it is impossible to have “an immaculate transfer” from capital inflows to current account deficits. 
National supervisors found it impossible to contain these developments because they had to respect the single market rules and lacked the macro-prudential tools to offset the effects of large capital inflows. But by creating supervision at the level of market, banking union offers a possibility to better pre-empt such developments in the future – and therefore to better protect the real economy, growth and employment.
The SSM and the practice of banking in Europe
Let me now turn to the SSM in more detail and how I expect it to affect the supervision and practice of banking in Europe. There are six points I would like to highlight.
First, I expect the SSM to implement the most advanced supervision of banks. This means, of course, using all the modern methods to be forward-looking and risk-based, focusing on the viability of banks’ business models, the robustness of their balance sheets to shocks and the evolution of their liquidity and funding positions over time.
This is the basis on which we are currently designing the single supervisory model which will apply to all banks in the SSM – not only to those that are directly supervised by the ECB.
Second, the SSM will have to address the problems created by the heterogeneity in the way that banks calculate risk-weighted assets. Even though euro area banks have a healthy median Core Tier 1 capital ratio of above 12% of risk-weighted assets, investors seem to still have concerns about their robustness due to lack of transparency on how they calibrate their internal risk models, as well as differences in how models are validated across jurisdictions.
The SSM should create more homogeneity by imposing common principles about methods and models' parameters in order to improve the reliability of banks’ internal models. The Basel Committee is also discussing how to address this issue and some countries have recently introduced minimum floors to risk weights of some types of assets. The ECB/SSM will ensure that, both within and across countries, the same types of risks are given similar weights.
Third, the SSM will lead to a harmonised treatment of non-performing exposures and provisioning rules, which at present vary between jurisdictions and are not directly comparable for investors.
The European Banking Authority (EBA) has recently released technical standards on supervisory reporting on non-performing exposures and forbearance, which will provide consistent indicators of asset quality for the whole EU. However, as it will take some time for banks to implement these standards, the SSM will adopt its own simplified standards, in particular for use in our forthcoming comprehensive assessment. We’ll go as far as the timeframe allows towards the full EBA definitions.
Fourth, as a result of these changes, I see that the SSM will create the conditions for further integration of the European banking market.
Unified supervision should create greater trust among banks, thus reducing the risk aversion related to cross-border lending. And it should alter the environment within banks, as cross-border banking groups will be able to optimise their internal management of capital and liquidity and reduce compliance costs.
On liquidity and capital, we expect liquidity to be more fungible within cross-border groups than in the past, and for capital needs related to stress test outcomes, macroeconomic and systemic buffers to be able to be met at the consolidated level, provided that capital is freely transferable between the group and individual institutions.
Capital needs related to idiosyncratic risks may still need to be met at the subsidiary level, but our approach is likely to leave some flexibility to the Joint Supervisory Teams that oversee each significant bank to tailor the approach to the relevant bank’s business model and statutory arrangements. A movement to subsidiarisation that we observe in other parts of the world has no justification inside the SSM perimeter.
On the cost of doing business, compliance costs for cross-border banks should be reduced by the development of the EU single rulebook and the single supervisory model within the SSM. This could mean, for example, that regulatory compliance functions no longer need to be duplicated in different euro area countries.
This potential to deepen banking market integration leads me to the fifth point, which is that I would not be surprised if the SSM would open a period of restructuring in the European banking sector, in particular through more mergers and acquisitions.
M&A activity has been very weak in the euro area since the crisis – from 2008 to 2012 the overall value of deals decreased fourfold to just €10 billion, with cross-border deals the most affected. Nevertheless, the weak profitability and excess capacity of the European banking sector suggests that efficiency gains could be achieved. The Herfindahl-Hischman concentration indicator for the euro area banks is at the 690 level, well below the range of 1000 to 1600 of acceptable concentration.
Naturally I expect this process of restructuring to be driven first and foremost by the incentives of the private sector to raise profitability and increase returns on assets. However, there may be collateral effects of the change in the institutional environment that may favour that movement.
Looking further ahead, a single resolution mechanism in Europe will create more possibilities to use consolidation as a resolution strategy. We see this in the US where, of the approximately 500 banks that the FDIC has taken into receivership since 2008, around 450 were resolved through “Purchase and Assumption” transactions – that is, selling parts of banks to other banks.
That said, there are also frictions that create obstacles to closer bank market integration.
These include different legal systems, corporate governance structures and tax regimes that exist in Europe, as well as different insolvency procedures. Further integration is needed in these areas if we are to gravitate towards a true single market in capital in the euro area.
Finally, the new SSM Regulation introduced an important change a change in macro-prudential policy-making, both in terms of instruments and decision-making.
As part of the Capital Requirements Directive IV/Capital Requirements Regulation, the ECB/SSM will gain a series of new macro-prudential instruments. These include, among others, the counter-cyclical capital buffer, the systemic risk buffer and the macro-prudential elements from Pillar 2. I am confident that they can play a role in curbing pro-cyclicality in credit developments where needed.
In terms of decision-making, a European authority will be able, if necessary, to apply prudential measures to banks in both borrowing and lending countries. This is important because, to the extent that macro-prudential measures mitigate credit booms by raising banks’ cost of equity, this effect can be offset if there are large inflows from abroad that lower banks’ cost of debt. In other words, the ECB/SSM has the tools to address such situations as we saw before the crisis, where supervisors lacked the tools to tame domestic financial booms being largely financed from banks in other euro area countries.
The future of intermediation in the euro area
Looking further ahead, I expect this period of structural change in the banking sector to also have a permanent effect on the structure of intermediation.
In the euro area, banks have historically played an important role in financing the real economy. Banks loans account for most of household borrowing and around 50% of non-financial firms’ external financing, which is very different from the US where around 75% of firms’ financing comes from capital markets (equity and debt securities). The importance of bank-based intermediation in the euro area explains the relatively large size of the euro area banking sector compared with the US – at 270% and 72% of GDP, respectively.
Overall, I expect the future of banking to involve some rebalancing away from such high levels of bank-based intermediation and towards more capital market-based intermediation.
On the banking side, I anticipate that over time there will be a gradual decline in the size of the euro area banking sector. Indeed, such a process of consolidation and resizing has already been on-going since 2008. In net terms the number of credit institutions has fallen by 9% since 2008, or around 600 institutions, while the total assets of the euro area banking sector have declined by almost 12%. The second largest reduction in the value of assets was recorded here in Ireland.
This process has seen euro area banks become significantly less leveraged, with the average loan-to-deposit ratio falling from 144% in 2008 to 120% in mid-2013. I expect that this ratio will continue to fall and converge towards 100% as banks seek to further consolidate their balance sheets and develop safer, more sustainable business models.
And as a natural response to this, firms are likely to pursue new avenues for disintermediation. This brings me to development of capital markets.
The crisis has already boosted disintermediation in the euro area, in particular for larger firms and those located in countries with more developed corporate bond markets, such as Germany and France. Our data shows that these firms have been able to offset reduced access to bank finance with bond issuance in the last 12 months. Going forward, we can predict that the European capital markets will develop further.
The first area will be the deepening of corporate bond and equity markets. This has obvious advantages in terms of diversifying the financing mix for euro area firms and allowing them access to a larger pool of non-bank investors. At the end of June 2013 insurers, occupational pension funds, money market funds and investment funds together had assets worth almost 16 trillion euros – that is only a trillion euros less than the sum of euro bank deposits.
Moreover, deeper capital markets integration has been shown to dissipate the impact of local shocks by spreading losses across the market. In an old seminal paper, this risk sharing mechanism is estimated to explain the absorption of two-thirds of shocks in the US in the nineties. 
A second area where improvements can be expected relates to securitisation markets in Europe, in particular asset-backed securities (ABS) for SME lending. SMEs are particularly vulnerable to banking sector stress, and as they employ nearly 70% of EU workers and contribute around 60% of EU value added, this creates a concomitant vulnerability for the European economy.
Securitisation may be one way to stimulate credit to SMEs, as banks would feel more confident that they could shift risks to non-bank investors. However, we all know about the risks potentially related to extreme cases of securitization. After all, the “originate and distribute“ model was one the proximate causes of the crisis. The function of monitoring borrowers and managing credit risk that justifies the existence of financial intermediaries was lost in the process. Also, complex products with multiple layers led to significant losses and, therefore, we should not allow the repetition of that same experience. Nevertheless, it is useful to recall that the default rate of ABSs in the EU was just 1.4% from mid-2007 to the first quarter of 2013, compared 17.4% in the US. For senior tranches of European SME ABS, there have been no defaults.
Third, the euro area would greatly benefit from deepening debt capital markets for long-term financing, in particular for infrastructure projects. Investment in the euro area is currently still lower than in 2007, and there are signs that this may have reduced the euro area’s growth potential. This means that developing new ways to structure investment financing – for example, through the Commission’s project bonds initiative – must be a priority to avoid that the crisis leaves lasting damage to the euro area economy.
However, while I see some advantages for more disintermediation in the euro area, I do not think we should go too far in the opposite direction and replicate the US model. A predominance of capital markets in financing the economy opens opportunities but brings also more volatility. A balanced funding mix between banks and capital markets is best for financial stability – and close bank-firm relationships in several countries are simply part of the fabric of the euro area economy. But this implies that we need to have profitable, viable and stable banks in the euro area.
For this reason, I expect Europe to keep its universal banking model, with some limits on high-risk proprietary trading as we have seen in the recent German and French legislation, inspired by the Liikanen Report. Radical separation of retail and investment banking functions, eliminating the European concept of balanced universal banks would go too far, especially when we know that Europe’s universal banks have faired relatively well throughout the crisis and returned to profitability more quickly than others.
Let me conclude.
I have outlined today some of the changes that I expect banking union, and the SSM in particular, to prompt in the supervision and structure of euro area banking. These will have fundamental importance for the future of our economies.
As we are a bank-based economy, there are few issues that are more vital to growth and job creation in the euro area than a well-supervised and stable banking sector. In passing this responsibility to the European level, we are therefore taking a major step towards common ownership of our economic outcomes.
In this way, banking union can be seen as the first step in developing a real EMU – that is, a union where banking, fiscal, economic issues are decided in common, backed by the democratic mandate of the people of Europe.
Thank you for your attention.
Dirk Schoenmaker (2011) “The financial trilemma” in Economic Letters, 111, pp. 57-9
See “The European Crisis and the role of the financial system” speech by Vítor Constâncio, Vice-President of the ECB, at the Bank of Greece conference on “The crisis in the euro area” Athens, 23 May 2013. http://www.ecb.europa.eu/press/key/date/2013/html/sp130523_1.en.html
See Williamson, J., “Comment”, in Bergsten, F. (ed), International Adjustment and Financing: The Lessons of 1985-1991, Institute for International Economics, 1991, p. 243.
B. Sørensen and O. Yosha (1996), “Channels of interstate risk sharing: United States (1963-1990)”, The Quarterly Journal of Economics, November, pp. 1081-1111