Poul M. Thomsen | International Monetary Fund (IMF)


Euro Area , monetary union , banking , capital markets union , ring-fencing , EDIS.

JEL Codes:

A Financial Union for the Euro Area1

As the European policy machinery kicks into action again after the summer break, I welcome this opportunity to share some thoughts.

The question put to me is how much more deepening of the economic and monetary union is needed. This is a big one. There is no simple answer. My approach today will be to focus on the core issue of private risk-sharing. I want to discuss what it really means, what holds it back — and what concrete steps can and should be taken to push it into a higher gear. Recognizing that every roadmap needs a destination, I want to lay out a vision for a true financial sector union for the euro area.

It follows that I shall discuss the banking union and the capital markets union. But I will also try to advance the debate one step further by discussing a truly integrated financial markets Union — a union that can be more than the sum of its parts.

I will focus on finance today because — as I will explain — I see this as an area where meaningful progress is within reach—where progress can be made by taking small steps now. Much of this work, I believe, can be done within the confines of the current political consensus—we are not talking about Treaty change. We are talking about letting the technical experts hammer out mutually acceptable compromises. I will return to this.


But before I do so, let me first spend a few minutes reminding you of what we at the IMF see as the end-goals for a more-complete economic and monetary union. And here, of course, I will be reiterating a vision that we at the IMF have laid out many times over the last few years.

When we look to the medium term, envision a euro area architecture equipped with a full complement of public and private risk-sharing mechanisms. In the fiscal area, we advocate for the adoption of a modestly sized yet potentially powerful central macroeconomic stabilization capacity. In the financial sector, we support putting in place a common deposit insurance scheme and a common backstop for both bank resolution and deposit insurance.

As we have argued, such risk-sharing mechanisms need to be carefully designed to ensure incentive compatibility. And by this I mean that every step toward greater risk-sharing must be accompanied by parallel steps toward greater risk-reduction. The future configuration should be one that carefully balances individual responsibility with solidarity, both at the level of member states and at the level private financial institutions. There is in my view no other way forward.

Let me at the outset be very clear on the point about individual responsibility. No amount of progress on the common architecture in the coming months and years will fundamentally change the fortunes of euro area countries when the next major shock or downturn hits. I have said this often.

Resilience will continue to rest on fiscal adjustments and structural reforms at the national level. Here, unfortunately, we see far too much complacency. Let me be direct, progress on the banking and capital markets unions, and even on a central fiscal capacity, will not prevent some countries from being forced to undertake large pro-cyclical fiscal adjustments when the next shock or major downturn hits. When this happens, such countries will no doubt regret not having done more while times were good — regret having a much too procyclical stance during what is still a period of robust and strong growth throughout the Euro Area.

It is critical that all member countries respect the Stability and Growth Pact. It is critical that our friends in the European Commission enforce these rules consistently. Every club needs rules to thrive. Yes, steps are needed to simplify the SGP, but these must be matched by steps to ensure better compliance and stricter enforcement. We at the Fund detect support for the need to eventually make the fiscal rules less complex — something that we would strongly Support — but it does not seem to be a political priority at this juncture.

On fiscal risk sharing, in contrast, there is an ongoing dialogue on a common euro area budget, or some mechanism to support public investment in individual countries. While we see the recent proposals falling short of the full-blown countercyclical facility that many observers consider critical for the long-run stability of the eurozone, we are encouraged that discussions continue.

As many of you may know, earlier this year the IMF waded into this debate by tabling a proposal of our own for a central fiscal stabilization capacity. Let me emphasize that we very much view this as a proposal not just to establish an element of public risk-sharing, but also to improve compliance with the fiscal rules.

Our idea, quite simply, is that it should be possible to bridge the gulf between those who are calling for greater fiscal risk-sharing and those who worry about moral hazard and permanent Transfers — to explicitly link central support for economic stabilization to fiscal risk-reduction, that is, to better compliance with the rules.

Our proposed fund seeks to encourage saving by member states with limited fiscal space — to encourage them to build buffers — by requiring regular contributions from all euro area countries while at the same time making transfers conditional on compliance with the fiscal rules. By having transfers triggered automatically by a readily observed cyclical indicator linked to unemployment, our proposal explicitly sets out to smooth macroeconomic shocks.

Our analysis shows that a fund financed by a relatively modest annual contribution from member states — say, about one-third of one percent of euro area GDP — could act as a powerful stabilizer. It would not substitute for national fiscal responsibility, nor completely offset large shocks, but it certainly would help. Moreover, in the event of large shocks there will always be a role for ex post risk sharing, through official assistance subject to policy conditionality, provided by the ESM.

Now we understand, of course, that Europe is not a political union, and agreeing architectural change is a complex business.

From such a perspective, then, our proposal might appear politically unrealistic, despite explicitly making access to transfers conditional on obeying the rules. But even if it takes time to build consensus, we are convinced that greater fiscal integration in the eurozone will eventually only happen by bridging the fundamental divide between risk-reduction and risk-sharing. Our proposal is a way to build this bridge.


Let me pause for a brief reality check.

Not long ago, there had been signs that Brexit and broader geopolitical challenges from both East and West might breathe new life into European efforts to take major, fundamentally political steps to advance the eurozone’s shared architecture. I still hope that this will be the case, but it appears that major, pathbreaking initiatives are not likely any time soon.

We see undercurrents of discontent and euro-skepticism feeding into poll results across the union. We see basic challenges to the European project, challenges that governments new and old cannot afford to ignore. We see at best a limited consensus on what comes next. The June summit delivered an outcome that fell short of the elevated hopes of only a few months earlier. I think it is right, therefore, for me to focus on goals that I consider achievable even in the current political context.

My basic premise today is that it is vital that scarce political capital be used well: to find achievable wins and lock in meaningful progress, wherever possible.

At the Fund we have long argued for more public and private risk-sharing – both are essential. Today, as I mentioned, I would like to focus on facilitating more private risk-sharing through the financial markets.


Private risk-sharing is mentioned often but explained rarely. What does it mean?

There may be different definitions, but I refer, quite simply, to the diversification of risk exposures by financial intermediaries in the euro area, across both national borders and economic sectors. I mean moving to a union where equity capital in one jurisdiction can support prudent risk-taking in another jurisdiction; where deposits in one country can fund sound lending in another; and where financial and nonfinancial firms alike can issue equity and debt into a European capital market, to a European investor base. I mean making another push to achieve the original vision of a European banking and capital markets union, that of finance without borders.


Let me start with banking.

Everywhere in the world, banks sit at the heart of finance. Even in the jurisdictions where capital markets are the most developed — and here I refer of course to the United States — banks exert an influence that extends far beyond their relative share of financial sector assets.

Modern banking groups are complex animals. Some units of the group focus on the bread-and-butter business of retail banking, accepting demand deposits and making loans, from relatively standardized consumer credit lines or residential mortgages to more bespoke commercial real estate or SME financings.

Some focus on trade finance, or automotive finance, or credit cards. Some units focus on buying loans and bundling them into special purpose vehicle structures, earning securitization fees and raising wholesale funds. Some units focus on managing proprietary trading portfolios of securities and derivatives. Some focus on managing segregated portfolios on behalf of their clients. Others may underwrite insurance products. Yet others may provide depositary services to mutual funds and pension funds or prime brokerage support to hedge funds.

And somewhere in the group structure there will often be a centralized treasury unit issuing commercial paper and channeling intra-group liquidity to where it is needed most.

I could go on, but I think my point is made. In the real world, there are no neat dividing lines between banking and the capital markets. In the real world, bankers often run nonbank finance. Let us be clear: banking is the pivot of finance.

It is appropriate, therefore, that Europe has pursued banking union in advance of capital markets union. Anything else might have put the cart before the horse.


Having explained why I want to start by discussing banking, let me spend a few minutes recollecting the past. While this will be familiar ground to most of you, the tenth anniversary of the global financial crisis should be a time to pause and reflect on the road traveled.

In its old-pre-crisis configuration, cross-border banking integration in the EU relied heavily on passporting, whereby any bank incorporated in any member country enjoys permission to establish branches in any other member country. Under the old model, however, there was a dangerous disconnect between cross-border banking activity and its oversight. Back then, the responsibility for official oversight of cross-border bank branches within Europe rested almost entirely with the national supervisor in each group’s home country.

But this model, built around supervision as a national competency, allowed some national supervisors to favor their own so-called “national champions,” tolerating excessive asset expansion and risk-taking both at home and abroad.

What developed was a macro-significant flow of funds within the euro area, where of course there was no currency risk.

The highly rated, too-big-to-fail, so-called core banks enjoyed strong liquidity for at least three reasons. First, they were seen as a preferred destination for European retail savings and wholesale funds. Second, they increasingly attracted petrodollars post-9/11. Third, they actively borrowed in the U.S. credit markets, including to raise dollar funding to finance their large U.S. securities Portfolios — one can think of this as a transatlantic wholesale funding pipeline. At the fringes of this great funding game, cheap euro liquidity was also channeled to euro area countries hungry for more credit. There, it helped finance cheap lending on increasingly lax terms, lending that in many cases inflated unsustainable local asset price bubbles. That was Europe’s old banking union.


Then, of course, came the day the music stopped. To be precise, that day was ten years ago tomorrow. We all remember, it was a Monday. In New York, Lehman Brothers filed for bankruptcy, an event that would send shockwaves around the globe.

With a force none could have predicted, wholesale funding markets froze all over the world, trade financing collapsed, goods and services trade fell off a cliff, banks started to drop like dominos, and fiscal coffers were forced open.

Almost immediately as Lehman went down, European authorities rushed to intervene its European subsidiaries, before liquidity and collateral could be spirited away to the parent. Then AIG very nearly collapsed. Then the U.S. money fund industry suffered a run.

It kept getting worse. By the end of September 2008, Wachovia, the fourth largest banking group in the United States, required government money, while in Dublin the authorities issued a blanket guarantee to the entire Irish banking system. In mid-October, the U.S. government announced a blanket guarantee of its own, and moved to force-feed taxpayer capital into the nine largest U.S. bank holding companies. All across Europe there was a chain reaction of emergency taxpayer support.

Everywhere, money ran for the perceived safety of home.

Parent funding of cross-border bank branches within Europe disappeared overnight, precipitating credit contractions, asset price collapses, and deep recessions in the host jurisdictions. And so, in the space of just a few weeks in late 2008, the old European banking union gave way to a harsh new reality of home bias and cross-border fragmentation, with most national authorities preferring to host highly capitalized cross-border bank subsidiaries rather than cross-border branches. Much of this reality remains in evidence today.


What did Europe do in response?

The short answer is: a lot. Leaders came together to mandate the construction of a new and better banking union. In the relatively short period since that apex decision in June 2012, Europe has traveled a great distance. I find it truly remarkable to note that, as recently as six years ago, there was no Single Supervisory Mechanism or Single Resolution Mechanism. A lot has changed.

Today, the European Central Bank has a formidable banking supervision arm, and the Single Resolution Board is an emerging piece of the European institutional set-up.

There can be little doubt that the operationalization of the SSM has fostered a step improvement in the quality of banking supervision across the euro area. Joint supervisory teams assigned to all major European banks comprise both national and ECB personnel, always headed by an ECB bank supervisor. This, among other important new supervisory features, greatly limits the scope for the bad old “national champion” approaches.

And we have seen from the various bank failure cases over the last year or so that the fledgling SRB is learning to implement the new rules embodied in the SRM Regulation and the Bank Recovery and Resolution Directive — the BRRD — without triggering major adverse spillovers. Where once taxpayer bail-outs were commonplace, today there is a growing insistence that bail-in needs to happen, that private stakeholders must be made to pay the price for risk-taking gone bad.

Taking these positive observations, let us say that the cup is half full.


But the cup is also half empty. There is a lingering impediment to a truly integrated banking union, an issue that has become better understood over time. It is called ring-fencing.

What does ring-fencing mean?

It means, in a nutshell, that countries do not yet fully trust the new supervisory and resolution arrangements, despite all the progress made. Quite rationally then, national authorities seek to protect their economies and their taxpayers by ensuring that banks maintain sufficient capital and liquidity within their own jurisdictions. The problem, however, is that such defensive measures also tend to limit the free flow of capital and liquidity across borders, and thus act as barriers to a fully integrated banking union.

The SSM and SRM are by construct dependent on national resources and fragmented national rules, which limits their freedom of maneuver. Member states, in turn, neither fully trust the ability of the SSM to ensure prudent behavior in banking, nor the ability of the SRM to properly charge back the costs of bank failure to the banking industry.

Given these doubts, national authorities still erect barriers to protect their economies and taxpayers. Especially in the host jurisdictions, they take steps to ensure that if banks within their borders fail, then they fail with enough remaining buffers to limit any potential need for fiscal support.

But this self-protection, stemming from national fiduciary duty, stands in the way of a single banking marketplace. It acts as an additional impediment to cross-border mergers and acquisitions. It leaves large banking groups asking why they must pay large sums as resolution levies to the Single Resolution Fund, yet still not enjoy some of the fruits that the new banking union was supposed to bring. Arguably, such groups charge their customers more than they would if they were better able to move their capital and liquidity across jurisdictions.


So: what remains to be done?

To bring some structure to the debate, I would like to propose two axes of attack: on the one front, a push to scale back member states’ ability to ring-fence; and, on a second front, an effort to reduce their incentives to do so. Let me elaborate on each of these in turn.


First, on the ability to ring-fence. This has much to do with taking further steps to enhance banking supervision.

The ability of national authorities to maintain barriers to the cross-border flow of bank capital and liquidity stems from various gaps and provisions in the EU rulebook. Examples of such barriers include the inability to allow capital relief at the subsidiary level even when there are explicit solvency guarantees from the parent in another EU jurisdiction, and the refusal in some EU jurisdictions to relax large exposure limits for cross-border intragroup transactions.

Not only does this fragmentation along national lines hinder private risk-sharing across the euro area, it also reduces the SSM’s ability to push for further risk-reduction in banking. In important areas such as the proper classification and provisioning of legacy assets, overseeing banks’ corporate governance, imposing penalties for non-compliance, or overseeing major acquisitions, the ECB still lacks the direct powers it needs to conduct fully effective banking supervision.

But the good news is that many of these impediments, both to cross-border flows and to strong oversight, can be fixed at a technical level, in many cases by amending the so-called single rulebook — the package comprising the Capital Requirements Regulation, the Capital Requirements Directive, the BRRD, the Deposit Guarantee Schemes Directive, and the attendant technical standards. As I mentioned at the outset, much of this work can be done within the confines of the current political Consensus — we are not talking about Treaty change. We are talking about letting the technical experts hammer out mutually acceptable compromises.

Nonetheless, even in these relatively technical areas, let no one underestimate the inevitable resistance from national and private vested interests. Change will not be easy, and one can expect many battles to play out here in Brussels and other capitals.

The key, I believe, will be for all parties to remember that a more-integrated banking market will offer substantial benefits for all — it is not a zero-sum game.


Let me now turn to my second axis of attack: a set of concrete steps to reduce countries’ incentives to ring-fence.

Principally, in my view, reducing such incentives requires steps to further improve and harmonize bank resolution arrangements. We can agree that bank failure needs to be handled within a unified, transparent, and predictable framework, one that protects financial stability while minimizing costs to the economy, surviving banks, and taxpayers. We can agree that bail-ins must indeed become the norm, and bail-outs the systemic exception. The challenge is how to get there.

Solutions are already in train. For the larger, more complex banks where contagion can be a real concern, there should be enough loss-absorbing liabilities — the BRRD’s “minimum requirement for own funds and eligible liabilities” or MREL — so that undercapitalization can be remedied by equity write-downs and debt-equity conversions without burning senior unsecured debt or uninsured deposits.

And in cases where the least-costly solution to protect insured deposits of a failed bank requires some public support, such support should come from the common resources of the Single Resolution Fund.

In such ways, individual member states will be protected from having to shoulder the costs of banking failure alone. And that, in turn, will help soften the need for ring-fencing and self-protection. So the path to borderless banking, I would say, rests also on steps to achieve a better bank exit framework.

We are not there yet. The recent experience showed that the system still permits bank liquidations under national procedures that vary widely. It also reminded us that the EU state aid rules still allow national governments to provide taxpayer money to banks in liquidation subject to burden-sharing requirements that are less exacting than the BRRD rules. I am tempted to call this a loophole.

So, in exit policy as in bank supervision, there is still plenty of technical work to be done. Rather than diving into the details, let me simply note that the IMF has tabled a wealth of concrete recommendations over the summer. I commend the work of our Financial Sector Assessment Program team that recently completed its evaluation of the euro area’s new framework. Its reports are rightly being seen as essential reading for those involved in further upgrading bank resolution in Europe.

I would, however, like to emphasize one specific action that goes beyond technical change and requires political support: to secure the much-discussed common backstop to the Single Resolution Fund.

It is encouraging that the June summit requested detailed proposals in this area. To be frank, I will not argue that such a backstop will be an immediate game-changer for bank resolution. A few back-of-the-envelope sums show that it would not be called upon in anything but a severe crisis, and no such crisis is in our baseline. But it is symbolic, as one further step toward curbing national incentives to ring-fence, and it rightfully belongs in a well-designed resolution toolkit.

Ideally, the backstop should be large and automatic, allowing the Single Resolution Board to draw on it at short notice and without condition. I urge European policy makers to press forward and get this done.

I noted earlier that the IMF also strongly supports early adoption of the European Deposit Insurance Scheme, or EDIS. Here, however, reaching agreement will need to overcome multiple obstacles, including at the political level.

Some of the strongest banks in the euro area fear that their contributions to EDIS will end up financing losses at weak banks elsewhere. In addition, some governments in the member countries with the strongest balance sheets worry that EDIS could become a system of permanent state-to-state fiscal transfers through the back door—transfers that are expressly forbidden under the Treaties — despite it being in the first instance a privately funded scheme. In principle, both concerns can be addressed by risk-based insurance premia, properly calibrated. Nonetheless, my own personal view is that progress on EDIS will require parallel progress on reducing the pervasive home bias in banks’ holdings of home-country sovereign securities. And that, in turn, is a complex and contentious area, one that I cannot do justice to today.

The important thing is, I do not see roadblocks to common deposit insurance as barriers to progress in other areas.


Let me now turn relatively briefly to nonbank finance. This, of course, is an extremely broad area, spanning everything from financial market infrastructure to securities underwriting to insurance, with much in between. In some areas, such as mutual funds eligible for marketing to retail Investors — the so-called UCITS Regime — Europe is a world leader. In others, such as harmonized financial reporting, there remains more work to be done.

I cannot do justice to the multifaceted challenge of developing and integrating nonbank finance in Europe in the minutes I have remaining. So, after briefly noting that initiatives range from corporate insolvency harmonization to SME securitization, let me just focus on one specific and important part of the topic: regulatory oversight of the securities and derivatives markets.

I mentioned at the outset that these markets are uniquely well developed in the United States. I will not claim to be an expert on how this came to be. Perhaps it has to do with the long U.S. history of banning commercial banks from engaging in securities underwriting, under the Glass–Steagall Acts, which is in sharp contrast to the European model of universal banking. Perhaps it reflects the strong federal powers vested in the U.S. Securities and Exchange Commission, for the securities market, and in the Commodity Futures Trading Commission, for derivatives.

Whatever the answers, it seems clear to me that the European capital markets union endeavor could benefit from further study of the U.S. example as part of its ongoing search for a guiding framework.

Allow me to offer one related observation. I think it is important to bear in mind some basic differences between banking and capital markets oversight. In banking, the mandate is to ensure safety and soundness, meaning, to reduce the risk of failure; this is pursued through regulations and, more importantly, through prudential supervision to proactively limit risk-taking ex ante. In the capital markets area, in contrast, there is no mandate to reduce the likelihood of failure; here, the focus must be on ex post punishment for breaches of conduct-of-business rules focused on truth, transparency, and disclosure.

One question that arises, I think, is whether Europe needs a centralized financial markets agency. At present, the European Securities and Markets Authority — ESMA — is mostly a standard-setting body, with direct regulatory authority only over the rating agencies and trade depositaries. In every other area of the European securities and derivatives markets, regulation and enforcement remain national competencies.

Has this resulted in a push for national champions and a race to the bottom, as was once the case in banking? Is there therefore a case for a “super-ESMA” vested with broad, pan-European regulatory powers? To me, the answers to these questions are not obvious — but they seem important, and well worth further study.

As the thinking advances, I would like to inject one cautionary note: despite my general advocacy for strong financial sector oversight, there remains a role for a light touch in some areas. There is a concern doing the rounds that, as the country with the deepest financial markets and financial market expertise exits the EU, a certain heavy-handedness might take root. To some extent I share this concern.

Market regulation must be fit for purpose, and proportionate to the risks. In the securities and derivatives space, Europe must guard against any perception that public support would be forthcoming in a crisis—intermediaries must be allowed to fail. This will ensure that market discipline remains in the front seat. There can be no room for moral hazard.


Colleagues, let me now wrap up.

My basic premise today has been that scarce political capital must be used well.

I have identified finance as a key area where we should push — meaningful progress is within reach. Here, I believe, much of the necessary work can be done within the confines of the current political Consensus — we are not talking about Treaty change. We are talking about letting the experts hammer out compromises to take forward the banking and capital markets union projects. But I am under no illusion that this would still encounter strong resistance from vested interests.

In banking, the task is to further strengthen supervision and resolution. To be concrete, the task is to systematically remove the remaining national fragmentation from the single rulebook. This will maximize the effectiveness of the new framework in controlling excessive risk-taking, and will ensure robust risk-sharing when banks fail.

And that, in turn, will reduce the need for individual countries to protect themselves with ring-fencing measures that also act as barriers to cross-border flows — I believe there can be a virtuous circle between less fragmentation and more trust.

In nonbank finance, I have focused on one aspect today: ensuring robust oversight of the securities and derivatives markets. Here the mandate must be to ensure truth, transparency, and disclosure. The question can be asked if there is a case for a “super-ESMA” with pan-European regulatory powers. But one must also guard against overreach, and allow intermediaries to fail, so that market discipline can flourish.

In my vision for a European financial union, the current preference for subsidiarization and other defensive measures at the national level will recede over time as fragmentation is reduced. Eventually, this will allow a return to a banking model centered on cross-border branching. While this might sound like a case of “back to the future,” prudence will be embedded in a way that bears no semblance to the bad old days of forbearance and arbitrage that took Europe to crisis a decade ago.

  • 1.

    Speech given by Poul M. Thomsen at the conference „Ten years after the start of the financial crisis: contours of a new normal“ jointly organised by Belgian Financial Forum (BFF) and SUERF at the National Bank of Belgium, on 14 September 2018.

About the authors

Poul M. Thomsen

Poul M. Thomsen, a Danish national, has been Director of the European Department at the International Monetary Fund since November 2014 supervising the Fund’s bilateral surveillance work for the 44 countries in the Department, its policy dialogue with EU institutions, including the ECB, and its program discussions with European countries with Fund supported programs.
As Director Mr. Thomsen also has the responsibility for the Fund’s outreach activities in Europe and its interactions with European senior officials. Before taking up his current position, Mr. Thomsen had, as Deputy Director of the European Department, the primary responsibility for the Fund’s programs with European countries affected by the global financial crisis and the subsequent crisis in the Euro Zone, including as mission chief for Iceland, Greece and Portugal and as supervisor of the programs for Romania and Ukraine. Before the global financial crisis, Mr. Thomsen gained extensive knowledge of countries in Central and Eastern Europe, having worked on the region continuously from 1987 to 2008, including as mission chief for multiple countries in the region, head of the Fund’s Russia Division during the 1998 Russian financial crisis, and Director of its Moscow Office from 2001 to 2004.

More on these topics