How will Brexit change the map of global finance? Speech at the 2018 Europe – US Symposium of the Harvard Law School Program on International Financial Systems

Check against delivery.

1 Introduction

Ladies and gentlemen

It is a pleasure and a privilege to be at this Symposium – it’s the second such event I have attended, and I really must say how much I value the refreshing and inspiring interaction here.

That’s why I chose my topic – Brexit and the future of cross-border finance – so reluctantly. Because as important as this topic is, it does very much feel as though a wedge is being driven into the community of international policymakers and supervisors.

I will speak openly about the pending issues, because I think openness is crucial to maintaining a trusting relationship. I very much look forward to engaging in a frank and solutions-driven debate with you afterwards.

Brexit will change the map of global finance. That much is clear.

But whether these changes will be just modest amendments or seismic shifts is less clear – it is still too early to tell.

In the long term, Brexit may very well prove to be a tipping point. It might set in motion fundamental changes which impact on where banks do business and how international supervision works.

So there is good reason to think carefully about how we limit and regulate the cross-border provisioning of financial services after Brexit – because the decisions we take now may become the principles of international financial supervision in the future.

In my speech here tonight, let me raise two crucial questions – and try to provide tentative answers. The first question is: How fragmented will financial services – and their supervision – be in the EU and the UK after Brexit? The second question is important not just to the United States, but also to other third countries to the EU: Will we see changes in the general approach to the regulation of foreign firms in the EU?

2 Will Brexit fragment financial services and their supervision?

Let’s begin with the question about the future relations between the UK and the EU.

Brexit talks made substantial progress recently: so much so that agreement has been reached on a transition period;[1] moreover, a general free trade agreement between the EU and the UK also seems within reach.

The main question for me, then, is how far-reaching this free trade agreement can be. For example, an agreement without tariffs would be highly valuable and feasible.

There is, however, no clear case against non-tariff barriers like product standard regulations. Rules like these safeguard public interests such as food safety and financial stability. Public authorities need to have sovereignty to defend these standards – although, it is important that they not abuse this power for protectionist motives.

This is something we can see clearly in the financial services sector. Public interest in financial stability is the reason why a far-reaching free trade agreement in this area is rather unlikely.

That’s because the UK government has chosen to leave the single market and the customs union, and withdraw from the associated regulatory framework. Supervisors cannot allow a foreign bank to provide financial services in their market without having the possibility to curb behaviour that endangers financial stability.

That’s why I believe – and I’m sure many other supervisors would agree with me on this point – that a chapter on financial services in a free trade agreement (FTA) would be very limited, if that.[2]

And it looks like negotiations are moving in that direction. At its last meeting, the European Council of EU leaders laid down two important negotiation guidelines regarding financial services: the first is to assess the possibility of adding a chapter to a future FTA that covers some aspects of financial services. There would not appear to be much scope for liberalisation, however, because the current point of reference is CETA, the Comprehensive Economic and Trade Agreement between Canada and the EU. While CETA is one of the few Free Trade Agreements that has a section on financial services, the freedoms it provides are quite limited.

3 How to supervise cross-border banking after Brexit?

Whether there is an FTA containing a financial services chapter with limited freedoms or not, the pivotal question remains the same: how will each side ensure that the other side’s firms are sufficiently regulated and supervised?

That’s what the second negotiation guideline is all about. Even limited cross-border freedoms would be only possible if, say, the EU supervisor could ensure that the UK firm is subject to high supervisory standards in the UK. Short of asking these firms to apply for a full local licence, this can be done through the equivalence approach to regulation. The UK firm would be able to provide services if the EU supervisor could ensure that the UK rules are equivalent to those in the EU.

The EU already has that kind of approach in place – most notably, there is an equivalence agreement with the US on the supervision of central counterparties, or CCPs for short. But the EU approach is built on a law-by-law basis, where equivalence can only be granted when the specific piece of legislation has an explicit provision. There is, for example, no basis for an equivalence decision when it comes to the licensing of banks. The European Council has therefore asked the Commission and the Brexit negotiation team to analyse whether it would be prudent to enhance this approach.

These two guiding principles – an analysis of the possibility of adding a limited financial services chapter, and enhanced equivalence – combine to form a spectrum of three different types of potential negotiation outcomes:

  • At one end of the spectrum there is the scenario of a financial services chapter whose scope is similar to, or even wider than, CETA, with high supervisory standards ensured through enhanced equivalence.[3] I call this the "enhanced equivalence option".

  • At the other end of the spectrum there is the scenario of striking no deal on financial services, combined with the supervisory requirement of something known as "subsidiarisation" – that is, the establishment of independent subsidiaries in each other’s market. Equivalence would remain largely unchanged. I call this the "subsidiarisation option".

  • In between lie several combinations of quite limited financial services agreements – that are less encompassing than CETA – plus a less ambitious revision of the equivalence regime, ie an extension of the law-by-law approach to encompass other regulatory areas that are currently not out of scope.

Now this is already a lot to take in over a dinner speech – so I won’t discuss the merits of these approaches; and in any case, it is still too early to make a final call on which is the best option. But let me give you my take on what the next steps should be.

None of the remaining options for financial services are ideal. So we will have to make do with complicated, second-best solutions. And in the debate about the merits of each of these options, I am seeing considerable bias. That bias is that many people from the industry, and policymakers and journalists, too, weigh liberalisation and low costs for firms higher than they do financial stability and democratic accountability.

For example, if one were to follow the preferences of industry groups, a new equivalence regime would ensure a high level of liberalisation for financial services, and regulatory equivalence would be managed by a technical committee of EU and UK supervisors, with independent arbiters dealing with cases of conflict – these solutions have many labels, a prominent one being "mutual recognition".

I am rather sceptical about the mutual recognition approach and about similar approaches based on regulatory harmonisation through technical committees and independent arbitration mechanisms.

For one thing, a solution where technical committees impede regulatory divergence could quite clearly undermine the sovereignty of jurisdictions to set their own rules and the ability to safeguard financial stability.

Moreover, these approaches are built on a premise: that more liberalisation is always better, and that less liberalisation is costly to economic growth. I think it is far from clear that more liberalisation is always better, as it also comes with problems – increased financial instability for example.

Thus, while a smart system of enhanced equivalence might be sensible, we would first have to find a suitable approach.

On the other hand, the subsidiarisation option – ie requiring a foreign bank to establish an independent subsidiary – would not only be well suited to preserving financial stability and democratic accountability. It might actually also be less costly than a full-blown mutual recognition regime.

My point is this: There is no ideal option on the table, nor is any one of the three second-best solutions clearly better than the others. Since there will no longer be a single market for the UK and the EU, it is futile to try to tie policymakers’ and supervisors’ hands as far as possible through technical agreements – this would undermine sovereignty, potentially weaken financial stability, and from a societal perspective, it might not even be the most cost-efficient option on the table.

4 Three questions about post-Brexit cross-border supervision

Instead, we need to carefully analyse which divergence option is best suited. In this analysis we need to focus on three crucial questions:

  • The first concerns the scope of a financial services chapter, assuming there is one: Which financial services should be allowed to be provided across borders; and which should be allowed to only be provided through an independent, locally licensed subsidiary?

  • The second question is about equivalence: Can equivalence be expanded systematically without hurting democratic accountability, sovereignty and financial stability? As the approaches debated so far do not meet these criteria: How could this be done?

  • And the third question concerns subsidiarisation: Is this maybe a less costly option than it intuitively appears? Since it will play a greater role in any of the remaining, realistic scenarios, how can we best administer this – that is to say, what should we expect to be ring-fenced in both jurisdictions, and what shouldn’t?

You see, this remains an open-ended process. The key takeaways are that two sovereign regulatory jurisdictions will emerge, that free cross-border trade in financial services will be rather limited, and that the remaining options all require further analysis and thought.

No matter which direction we take, our decisions will have an indirect, albeit substantial impact on other third countries to the EU. But how? Let’s take a look at this second question of my talk.

5 Will Brexit alter relations with other third countries?

And, naturally, I will focus on the US case. Brexit could affect two avenues which give financial firms access to the EU market: namely, decisions of supervisory equivalence, and the licensing process. Let’s take a look at each one in turn.

A new equivalence framework could be more systematic. Yet its precise calibration still remains to be seen. Again, one example can be found in the world of CCPs. The EU is currently revising the European Market Infrastructure Regulation, or EMIR for short. The consensus in the Council of Ministers and the Commission clearly points in one particular direction, which is to move away from a quite flexible, outcome-driven equivalence approach to one where EMIR requirements have to be met more precisely by foreign CCPs. To give you just two examples, EU supervisors[4] would have clearer compliance expectations and would review an equivalence decision at more regular intervals. Ultimately, where equivalence cannot be achieved this could lead to an entity being required to relocate to the EU.

For third-country CCPs, this enhanced equivalence is attractive, as it makes the decision more predictable and transparent, whilst ensuring the possibility of providing clearing services across borders without the need to operate a fully licensed subsidiary.

But where US CCPs are concerned, this approach may not be appropriate or necessary; because we already have a robust equivalence decision. Given the thorough process of mutual assessment that we carried out, there is a case for sustaining this outcome – for example, through a bilateral US-EU agreement on CCP equivalence. It is important that authorities on both sides of the Atlantic enter into pragmatic dialogue.

One important question, however, remains unsolved, namely that of how to resolve an internationally active CCP. And a lot of work still needs to be done on this score.

The second avenue via which Brexit could affect how financial firms gain access to the EU market are higher demands for licensing foreign bank branches and subsidiaries. Take, for example, the discussion about a draft law proposing to introduce EU intermediate parent undertakings, or IPUs for short. Similar to the US IHCs, foreign banks would have to bring their EU operations under a single holding company. This will most likely become law; and for good reason, because we need to have a better, consolidated understanding of what is going on at international banks doing business in Europe. What’s more, we need to be able to act reliably in times of turmoil. Otherwise, we would not be able to fulfil our mandate of ensuring financial stability. But to be clear, this instrument is by no means intended to make market entry more difficult for foreign banks – and the Bundesbank will continue to advocate fair access for foreign banks.

To sum up: from the perspective of US banks and those from other third countries, Brexit will impact on how they can do business in the EU single market. Both the new IPU requirements and a potential enhanced equivalence approach mean a more systematic approach to ensuring fair, transparent and stable rules for the provisioning of cross-border financial services. While this may imply some new requirements, it might end up streamlining the rules and promoting clear supervisory dialogue. I am confident that US banks will not face any new barriers of any substance – that would not be in the EU’s interest.

6 Conclusion

Ladies and gentlemen, as Brexit changes the map of global finance, we need to think carefully about how we limit and regulate the international provisioning of financial services – because the decisions of today may become the principles of international supervision of tomorrow.

The EU and the UK will become two sovereign regulatory jurisdictions, between which free cross-border provision of financial services will be limited, relative to today. We need to carefully analyse which divergence option will be best suited – in terms of financial stability, democratic accountability, and cost efficiency.

No matter which direction we take, these decisions will have an indirect, albeit substantial impact on the US and other third countries to the EU. While potentially burdensome at the outset, in the long run the changes in licensing and equivalence might provide a sounder footing for future cross-border activities – on the one hand setting strict expectations, but on the other hand also providing greater transparency and certainty.

I am confident that we can expand upon our strong partnership in this new environment.

Thank you for your attention. I wish you a fruitful and successful symposium.


  1. I.e. a period during which the UK is no longer part of the EU but will remain within the single market and the customs union; this 21-month phase, which starts on 30 March 2019, creates time: time for policymakers to develop the rules of a new partnership; and time for businesses to adapt to the new reality strategically and thoughtfully – rather than hastily.

  2. The options still on the table are a "no deal" scenario and one with quite limited freedoms for financial services, as set out in CETA (the Comprehensive Economic and Trade Agreement between Canada and the EU). These scenarios would mean two things for financial transactions between the UK and the EU. First, the "no deal" scenario would mean that the EU and UK would trade under rules set by the World Trade Organization – where coverage of services sectors is thin at best. Services providers would then have to apply for comprehensive licenses in both jurisdictions and have all the necessary elements of a fully functioning bank ready in both places. And second, even a CETA-like deal would most likely not mean far-ranging freedoms to provide financial services in the respective foreign jurisdiction – and thus force services providers to apply for comprehensive licenses in both jurisdictions.

  3. This could be achieved either through a general revision of the EU’s equivalence procedure and its scope, or by agreeing a bilateral arrangement with the United Kingdom.

  4. The European Securities and Markets Authority (ESMA) is most likely to become responsible for EMIR-related decisions of equivalence.