Ladies and gentlemen,
Thank you very much for inviting me to speak at ICMA today.
It is good news that the euro area capital markets are now starting to show some signs of de-fragmentation. The ECB’s financial integration report, released last week, shows improvements in integration of money markets, bond markets and equity markets. Indeed, price and quantity based indicators of financial integration point out that, although still worse than before crisis, the integration in European financial markets has significantly improved.  Yet I think there is a risk that, as we celebrate receding fragmentation, we miss the key question. And that is: what are we trying to achieve via de-fragmentation?
I sometimes perceive a tacit assumption that getting back to the status quo ante, where prices on euro-denominated financial assets converge, would constitute success. Impairments to monetary policy transmission would diminish. Access to finance for firms and households would improve. Even the infamous bank-sovereign nexus could turn into a strength as bond prices on banks’ balance sheets rise.
Yet we must keep in mind that convergence is not the same thing as integration. We saw substantial price convergence in the euro area in its first decade, only to be faced with a sudden fragmentation of financial markets when the region was hit by the crisis. Convergence is a welcome process; but it does not in itself guarantee deep and resilient financial integration.
Indeed, convergence may even be a sign of trouble ahead. Narrowing spreads in that first decade may have been a sign that markets were underpricing risk and that financial imbalances were building up. In this sense, convergence was probably a leading indicator of financial instability and hence financial de-integration.
In short, if we see the return of price convergence as “job done”, we risk making no progress at all. We may even be planting the seeds of the next crisis. The objective of de-fragmentation therefore has to be more than this. We have to be aiming at genuine financial integration. But what is this?
In the Eurosystem, we define financial integration as a situation whereby there are no frictions that discriminate between economic agents in their access to and investment of capital. But while this is useful as a theoretical benchmark for measuring financial integration, it is perhaps too maximalist to form an agenda for policymakers.
I think for policy purposes it is more useful to think in terms of creating a single market in capital. The single market in goods and services was created for, and is still primarily geared towards the EU-28. And indeed, a single market in both, goods and services and in capital, has benefits in its own right for an economic area even without a common currency. But in a monetary union, as I will argue, the creation of a single market in capital is not only beneficial, but essential. And what matters, as I will further argue, is not the intensity but the quality (in other words, the composition) of capital market integration.
To structure my remarks, I would like to begin by thinking about the objectives we want to achieve from such a single market, and then consider the policies that are necessary to get there.
What is the objective of a single market in capital?
In the context of monetary union, there are two objectives of a single market in capital. The first is to improve what economists call allocation – that is, credit is allocated efficiently and without reference to location. The second is to improve diversification, i.e. financial markets are integrated in such a way as to help companies and households cushion local shocks. In this sense, it is complementary to the single market in goods and services, which also raises allocative efficiency and may make local shocks less likely through trade integration.
There are some academics, however, who argue that the single market in goods and services may in time make local shocks more likely by increasing regional specialisation.  If this is correct, completing the single market in capital becomes even more essential for the euro area: the greater the risk of local shocks, the greater the need for shock absorption through capital markets. In other words, the single market in capital is not a luxury, but an essential stabilising force within monetary union. It is essential for the euro to be stable, and for it to deliver its full effects in terms of growth and jobs.
So how do we currently fare in achieving the two objectives of the single market, allocation and diversification?
Looking at credit developments in the euro area, in my view we still have a way to go to separate allocation from location. There are two ways in which I see that location still matters for credit: via the location of borrowers and of lenders.
First, in the euro area today it is the location of borrowers, rather than their creditworthiness per se, that matters most for access to finance, in particular for SMEs.
Credit supply in the euro area as a whole is most likely sufficient to meet aggregate credit demand. The problem is that cross-border frictions and high levels of uncertainty prevent credit markets from clearing. In a true single market, a creditworthy SME that could not borrow from a domestic bank would borrow directly from a bank located in another member state instead. Thus, the more efficient supplier would take market share at the expense of the less efficient one. But the mismatch between credit conditions in the euro area core and periphery shows this is not the case.
Part of the explanation is that banks’ credit assessments are influenced by the health of local sovereigns – rather than just the characteristics of the firm. But this is not the whole explanation. Even if sovereign risk were zero across the euro area, most banks are not structurally set up to provide cross-border lending. Direct cross-border loans to non-financial corporations account for just 7.5% of total loans to firms. And local affiliates of foreign banks represent on average only around 20% of national markets, and much less in larger countries.
The upshot of this lack of contestability in national markets is that domestic banks gain significant market power. And recent research suggests that this is negative for allocation: where banks face limited competition in their domestic markets, and firms are more dependent on them, financing constraints for SMEs have been higher.  This suggests that a more integrated retail banking market and more balanced financing mix would better support growth in the euro area.
The second way in which location can affect allocation – the location of lenders – matters to the extent that non-economic factors influence banks’ business decisions.
One dimension to this issue is the connection between local banks and local interests. There is intrinsic value for retail banks in being physically present in local markets as it lowers the costs of monitoring. Yet we also know there is a risk that local banks can become engaged in so-called “connected lending”, which tends to be inefficient.  Research suggests that in this context retail banking integration – in the form of cross-border M&A – can improve allocative efficiency by increasing the distance between the main shareholders and management of a bank and the vested interests in the country where the bank operates. 
Another dimension is the connection between national banks and national interests. There may be allocative inefficiencies that arise from trying to build up or defend “national champions” in the context of an integrated EU financial market.  One example of this phenomenon during the crisis was the decision of several national authorities to merge failing banks with other domestic banks rather than resolve them, which probably increased concentration. Indeed, in the US the FDIC has resolved around 500 banks since 2008; the equivalent figure for the euro area is around 50.
This underlines that reaping the allocative benefits of a single market in capital is linked not only to having a more integrated banking system, but also having more arms’ length governance of it – that is, having European rather than national supervision and resolution. European banks should finance the European economy. This is the essence of the European project and it benefits us all. Renationalising savings may provide short term protection but in the longer term it would be harmful to growth and jobs.
Turning to diversification, we see a similarly incomplete picture as for allocation. Most studies on risk-sharing within the euro area suggest that financial markets in the euro area do not provide much cross-border insurance. Indeed, a recent study by Bruegel found that, during the crisis, credit and capital markets absorbed only about 10% of the shock to GDP in net terms.  By contrast, in the US it has been estimated that capital and credit markets smooth around two-thirds of a given shock. 
What might be the reason for the lack of risk diversification in the euro area? In my view, on top of the lack of cross-border bank lending, the problem stems from the underdevelopment of two cross-border risk-sharing channels: private ownership of financial and non-financial firms and private insurance against banking crises.
First, private ownership structures in the euro area do not offer much possibility for risk sharing. For one, European firms tend to have a preference for debt financing over equity , which has so far been less suitable for absorbing losses – that is, until the adoption of the new bail-in rules for banks. And there is a pronounced home bias in the holding of that equity, meaning losses are not shared across borders. Only 44% of equity issued in the euro area is held by other euro area residents. Thus, equity market integration falls well short of what we might expect in a single market in capital.
The low level of cross-border ownership of retail banks also matters for risk-sharing. This has an equity aspect: whereas cross-border banks, in particular those that operate via branches, can in principle offset losses in one jurisdiction with gains made in other jurisdictions, national banks are exposed to concentrated risks. And it has a funding aspect: cross-border banks are less likely to be exposed to “sudden stops” as intragroup funding acts as a shock-absorber. This is, for example, one explanation why banks in the Baltics, where foreign ownership was high, did not see as high capital outflows as banks in the euro area periphery. 
The second channel for potential risk-sharing – private insurance against banking crises – was virtually non-existent in the euro area when the financial crisis struck. Although bank-financed resolution funds existed, they were not uniformly funded ex ante, and there was no mechanism for cross-border lending between authorities. Thus, losses that exceeded national funds ended up on sovereign balance sheets, which equated to a sizeable percentage of GDP.  For the euro area as a whole the total commitment to bank rescue packages amounted to 26% of 2008 GDP.
Insofar as these fiscal outlays raised government borrowing costs and threatened market access, the effect of insufficient insurance against banking crises was also to over-burden fiscal policy when it was the most needed to support growth, thus making the euro area even more unstable at a national level. This experience contrasts unfavourably with that in the US, where private insurance against banking crises was organised federally and proved quite cost effective. The FDIC restructured bank assets amounting to almost 1.9 trillion dollars from 2007-13 and recorded losses of just 33 billion dollars (which will ultimately be recouped from the US banking sector).
This again underscores the importance of European rather than national governance arrangements in a European banking market.
How can we achieve a single market in capital?
This discussion reveals that there is still some way to go to build a genuine single market in capital in Europe, both in terms of allocation and diversification. This is not about reverting to the pre-crisis situation: this is about rebuilding the system as it should have been to avoid the crisis. It also reveals that having such a single market would be extremely beneficial for the euro area, if not indispensable. And the conclusions that emerge in terms of policy are relatively clear: there is a need for greater governance integration in the euro area, greater retail banking integration, and greater capital market integration.
So how are we progressing in implementing this agenda?
Governance integration is clearly the area where the most rapid progress has been made, at least in the banking sector. We now have the first two elements of a genuine Banking Union: the Single Supervisory Mechanism (SSM) and Single Resolution Mechanism (SRM), with the Single Resolution Fund (SRF). A Single Deposit Guarantee Scheme should come one day, but it may be less urgent now that financial sector reform and bail-in rules have made insured deposits safer. These developments have been much discussed so I will not dwell on them much today.
The SSM and SRM (at least after 8 years, when the SRF is fully mutualised) will be indifferent to geographical factors in their supervision and resolution strategies. Banking Union can also be expected to improve risk-sharing, in particular via the new resolution framework, which offers two new channels for private insurance.
First, the new bail-in regime shifts the costs of bank failure where it ought to be shifted, that is, onto private shareholders and creditors. Whereas bail-out is concentrated in the home government , bail-in can spread losses across jurisdictions, to the extent that bank equity and debt is partially held abroad. Hence the importance of a strict and uniform enforcement of bail-in rules, compliant with the Bank Resolution and Recovery Directive (BRRD). Second, the resolution fund allows resolution costs to be shared across the euro area banking sector, as the fund is financed ex ante and ex post by contributions from the whole sector.
In the transition period while the resolution fund is being built up, however, it is important that we have in place an enhanced borrowing capacity. The credibility of the FDIC in times of systemic stress in part depends on the fact that it can immediately access a credit line of $100 billion from the US Treasury, which can be expanded to $500 billion under federal law.
Let me insist that a backstop is not a transfer system between taxpayers: any outlays are recovered by additional levies on the banking sector in the future. Thus, it is still private insurance; the only transfer is an intertemporal one among banks.
Retail banking integration
Governance integration in turn provides a more permissive environment for retail banking integration, although we have yet to see much progress here.
As until the crisis banks integrated mainly through wholesale markets, it is unlikely that they will gravitate straight from this to direct cross-border retail lending. More probable is an intermediate phase where cross-border M&A expands – indeed, this is what we saw in the US after the Riegle-Neal Act on interstate banking in 1994. As I said, physical presence matters for banks in local markets, especially given cultural differences in the euro area, and this is also beneficial from a risk-sharing perspective as banks have equity in the game.
M&A activity has been 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. The SSM nonetheless increases the potential to reap economies of scale within the single market, for instance by allowing banks to optimise their internal management of capital and liquidity. The mere presence of the SSM could induce restructuring as banks seek to reduce margins in anticipation of increased euro area competition. This will be compounded with the competitive pressure created by the expansion of e-banking.
That said, policy should acknowledge the trade-off between the benefits of retail banking integration and the systemic dangers of size.
On the one hand, there are benefits to size: banks need a certain capacity to be able to diversify idiosyncratic risks by lending to a broad enough range of firms. This is why the three biggest originators in each euro area country represent on average 80% of the outstanding amount of the SME loans. On the other hand, we have to beware concentration – that is, simply replacing large national banks with larger European banks may end increasing banks’ market power. It may also contradict the new regulation aiming at reducing systemic risk, in particular the leverage ratio.
I am confident that the new framework can steer us towards the right balance. Banking Union and the regulation against “too big to fail” should encourage banks to reach optimal size relative to the European market – that is, large enough to operate across borders and diversify risks but small enough to be resolved with the resources of the Single Resolution Fund. Regulatory initiatives on banking sector structural reform should support this process by distancing deposit-taking and trading activities, thus making large and complex banks easier to resolve.
At the same time, supervisors have new tools to help ensure that size does not equate to systemic risk, such as the “systemic risk buffer”. The leverage ratio will also be an important safeguard. And supervisors should be aided in this task by greater market discipline. As bail-in removes the implicit state guarantee for bank debt, we can expect shareholders and creditor to exert more scrutiny over the sustainability of bank business models.
Capital market integration
While Europe is and will remain a bank-financed economy, I see achieving a greater role for capital markets as central to a more efficient and financing mix. This brings me to the issue of capital market integration.
It is likely that the ongoing deleveraging of the banking sector will provide an impetus for capital market development in Europe. And indeed, we are already seeing capital markets complementing banks in the euro area: as bank credit to corporates has fallen, it has been approximately matched by issuance of corporate bonds.
Capital market access, however, is not evenly distributed in the euro area. There are two ways in which the playing field is not level. First, corporate bond issuance is strongly concentrated in non-distressed countries, where there has been no decrease in the net flow of bank loans . And while in principle firms from distressed countries can issue in those markets, in practice it is legally complicated, due to issues of governing law. While the original buyer and seller may agree on governing law, it becomes more complex for investors within a chain of financial intermediaries to understand which legal regime applies.
This points to a key missing piece in the single market in capital which is a harmonised framework for cross-border securities trading. The Commission has made good progress here in recent years, in particular in harmonising the rules needed for the transparency and integrity of securities markets – notably market abuse.
We are also seeing progress in dismantling operational and technical barriers to integration, that is, payment and cross-border securities settlement. TARGET2-Securities (T2S) will go live in June 2015 and will be the common securities settlement platform of 24 Central Securities Depositories (CSDs), both inside and outside the euro area, for settlement in central bank money.
But despite these encouraging signs, the integration of European corporate bond and equities markets is still hindered by lack of harmonisation in key areas. One area is laws relating to rights in securities,  which prevents investors for being able to assess the investment risk in another jurisdiction on the same basis. A key step forward here would be a Securities Law Directive, and the Commission is working on this issue.
Another area is differences in insolvency law across jurisdictions. While full harmonisation of insolvency law in the EU is not realistic for the near-term, it makes it all the more necessary to initiate the process leading to it. An intermediate step could be to develop an equivalent of the BRRD to cover systemically important non-bank financial institutions. Moreover, the ECB is in favour of adopting a comprehensive regime for the resolution of CSDs.
Still another area where more harmonisation would be helpful is corporate governance. For example, withholding tax and relief collection procedures for intermediated securities held by non-resident investors remain diverse and fragmented, although the Commission has issued a recommendation to simplify procedures in these areas. Further tax regime integration at the EU level is therefore a key part of further capital market integration.
The second way in which the playing field is not level for capital market access is between smaller and larger firms. Even with a more integrated European capital market, most SMEs do not disclose enough financial information to directly tap capital markets. Thus, relationship lending through banks will continue to play a key role in SME financing. Capital markets can nonetheless play a role, if in a more complementary way, in helping small firms to secure access to finance at various stages of their lifecycle. There are two channels here.
The first channel is for younger firms and relates to deepening venture capital markets in Europe. Venture capital investment in Europe is consistently much lower than in the US and rates of returns are worse.  This is in part a vicious circle reflecting the fact that the industry is fragmented across Europe: similar to bank lending to SMEs, successful venture capital depends on a large deal flow to cover the majority of investments that will fail. As most net new job creation comes from young firms, and young firms tend to be more sensitive to changes in investment opportunities,  this is missing part of the single market in capital that has substantial economic ramifications.
The second channel is for more established firms and relates to reviving high-quality securitisation markets in Europe. A deeper asset-backed securities (ABS) market would help SMEs to diversify their funding sources by tapping the funds of non-bank investors. Moreover, to the extent that ABS allows banks to partially remove SME credit risk from their balance sheets, smaller, less diversified originators should be better placed to enter the market. In this way, ABS could mean that SME lending is less dependent on a handful of large banks in each country, which would increase the contestability of domestic markets. The ECB is working with the Bank of England on the causes and roadblocks of the impaired securitisation market and will propose ways to deal with them. 
Let me conclude.
The main point I have tried to illustrate today is that de-fragmentation must have an aim, and that aim goes deeper than price convergence. True financial integration implies a single market in capital, where there is efficient allocation and effective diversification. It implies sizeable cross-border holdings of debt and equity; a European market for banking M&A; and direct cross-border exposures from banks in one jurisdiction to firms and households in another. And as we are still some way from reaching this goal, current favourable market developments are no reason to slow down the reform agenda.
I am convinced that laying the foundations of true financial integration will require concerted action and new pieces of legislation extending far beyond the domains today covered by Banking Union. I trust that the remarkable job accomplished under the leadership of Michel Barnier will be taken forward by the new Commission and Parliament.
But this is not all. Ultimately, the sustainability of financial integration depends on fiscal and economic integration as well. Remember that one cause of resource misallocation before the crisis was closed product and services markets that generated excessive rents and distorted prices signals. Remember that the collapse of cross-border lending during the crisis was aggravated by diverging fiscal positions of sovereigns. Hence, a single market in capital requires not just a banking union, but a fiscal union and economic union as well. We should not lose sight of these linkages.
Thank you for your attention.
See the discussion on the new SYNFINT indicator in the press release on the report “Financial Integration in Europe”, April, 2014, http://www.ecb.europa.eu/press/pr/date/2014/html/pr140428.en.html
See for example Krugman, P. (1991), “Increasing returns and economic geography”, Journal of Political Economy, 99(3).
Ryan, R., O’Toole, C. and McCann, F. (2014), “Does bank market power affect SME financing constraints”, Journal of Banking and Finance, January.
According to Cuñat and Garricano (2009), the Spanish cajas whose chief executives were appointed on the basis of political standing, rather than banking experience, extended more loans to real estate developers and fared substantially worse both before and during the crisis. See Cuñat, V. and and Garicano, L. (2009), “Did good cajas extend bad loans? The role of governance and human capital in cajas’ portfolio decisions”, October.
Giannetti, M., and Ongena, S. (2009), “Financial Integration and Firm Performance: Evidence from Foreign Bank Entry in Emerging Markets.” Review of Finance 13, 181-223.
An example of this view is Véron, N. (2013), “Banking nationalism and the European crisis”, October.
Van Beers, M., Bijlsma, M. and Zwart, G. (2014), “Cross-country insurance mechanisms in currency unions: an empirical assessment”, Bruegel Working Paper 2014/04, March.
Sørensen, B. and Yosha, O. (1996), “Channels of interstate risk sharing: United States (1963-1990)”, The Quarterly Journal of Economics, November.
See for example Fatica, S. et al (2012), “The debt-equity tax bias: consequences and solutions”, European Commission Taxation Papers, Working Paper No.33-2012, July.
Gros, D., and Alcidi, C. (2013), “Country adjustment to a ‘sudden stop’: Does the euro make a diﬀerence?”, European Commission Economic Papers 492, April.
Bénassy-Quéré, A. and Roussellet, G. (2012), “Fiscal sustainability in the presence of systemic banks: the case of EU countries”, CEPII Working Paper, 2012-05, March.
Given that ESM direct recapitalisation of banks, even when available, will be limited in size and subject to very strict conditions.
As shown in a recent ECB publication, non-financial corporates debt security, bank loans and quoted share net issuance has been around 66 billion in non-distressed countries and negative at (– 93) billion in distressed countries in 2013. Non- stressed countries' are defined as Belgium, Germany, France, Netherlands and Austria. Distressed countries are categorised as Ireland, Greece, Spain, Italy and Portugal. See “Financial Integration in Europe”, ECB, May 2014.
At present, the legal position that an account holder “owns” as soon as securities are booked to his account differs across euro area countries: in some jurisdictions the holder enjoys full and unshared property rights, whereas in others the holder receives a position that is in comparison “inferior”, i.e. a shared or indirect property interest or a mere claim against the account provider.
Veugelers, R. (2011), “Mind Europe’s early-stage equity gap”, Bruegel Policy Contribution 2011/18, December 2011.
Adelino, M., Song, M. and Robinson, D. (2014), “Firm age, investment opportunities and job creation”, NBER Working Paper 19845, January.
Bank of England and European Central Bank, “The impaired EU securitisation market: causes, roadblocks and how to deal with them”, joint paper, April 2014. See also the communication from the Commission on Long Term Financing of the European Economy, COM(2014) 168 final, March 2014.