Discours de Robert Ophèle, Président de l'AMF - AFME Annual European Compliance and Legal Conference à Londres - 1er octobre 2018 (en anglais uniquement)

While MIFID II has been introduced into our supervisory framework just 9 months ago, the truth is that it is already fundamentally challenged by the consequences of the pending withdrawal of the UK. Allow me to focus on one aspect of MiFID II, linked to the functioning of markets and the type of interaction possible once the UK has withdrawn from the EU.

I will structure my address this morning around three main ideas:

• First, MiFID II has triggered major structural market changes and dramatically increased the volume of information provided to market participants and to supervisors. The first 9 months of implementation have highlighted just how rapidly markets can shift, how essential level 3 guidance is, and how difficult a task it can be to ensure data quality.

•  Second, the UK withdrawal from the EU will obviously force us to undertake a complete review of the key metrics and parameters that we use in our daily MiFID II implementation. No doubt this will be a challenging exercise, and throughout we will need to bear in mind the initial rationale behind this piece of legislation.
• Lastly, I will briefly touch upon the type of relationship the EU could have with the UK, from a MiFID II perspective, and what changes could be contemplated in order to achieve a well-balanced framework.

MiFID II is extremely broad in scope and ambitious in its policy objectives and it has already triggered major structural changes in EU markets, because markets and markets participants are very quick to adapt to any new regulatory environment, in some cases by circumventing certain provisions, and sometimes by moving their operation to a less demanding or a more efficient jurisdiction, both/be it inside or outside the EU.

Not only have we seen trading venues adapt swiftly, with now roughly 72 OTFs, the new type of trading venues introduced by MiFID II, as well as around 120 Systematic Internalisers (versus merely a dozen or so under MiFID I). But also some radical business moves: take the example of the prompt shift of European commodity derivative contracts from European venues to US venues [~2000 oil & gas derivative contracts moved from ICE Futures Europe to ICE Futures US] or the switch from regulated venues to OTC. We also face rapid developments in market microstructure, such as periodic auctions or RFQ systems as well as an important increase in trading flows through systematic internalisers.

Let’s be clear: MiFID II should not prevent innovation, but such innovation should not be designed for the sole purpose of circumventing MiFID II.

In such a rapidly evolving market structure, Level 3 guidance and a true convergence amongst NCAs on its implementation is absolutely necessary in order to achieve a fair and level playing field inside the EU. This is one of the key lessons to be learnt from the application in practice of MiFID II 9 months on: ESMA Q&As have tackled numerous topics which are essential for the interpretation and application of MiFID II requirements.

While Q&As are supposed to deal only with technical matters, in many cases they have a substantive impact on markets’ structures and any misalignment amongst NCAs on these topics can bear severe consequences (regulatory arbitrage, migration of trading volumes from the EU) since market players adapt fast to divergences between NCAs; there is a clear need for genuine supervisory convergence in order to guarantee a level playing field.

For instance, ESMA has provided key answers to questions on sensitive issues for market participants and where different interpretation of the regulation would trigger an immediate shift in the markets. Let me mention some:

• Can a trade be arranged, but not executed, on an OTF?
• Does the share trading obligation apply to the person executing the trade only or does it extend to the person from whom the trade originates? (Art. 23 MiFIR refers to “the trades [an investment firm] undertakes in shares...”).
• May direct electronic access to a trading venue be provided by a participant that is not authorized under MiFID II?

On all these issues we need a consistent regulatory and supervisory stance throughout the EU.

ESMA’s numerous opinions on pre-trade transparency waivers are also fundamental to reach convergence since they ensure that MiFID II is not circumvented and that trading venues all operate on an equal footing. This control is all the more important considering the increasing sophistication of waivers proposed by trading venues.

The importance of Level 3 calls into question whether governance around the production of Q&As and Opinions at ESA level is appropriate. There is a need to think again about how Level 3 guidance is produced:

• Should input from the industry be sought by NCAs when the topic addressed through an ESMA Q&A is material? (We think it should).
• Should there be transparency amongst regulators as to which NCA agrees with ESMA’s answer and intends to follow the guidance? (We think there should be) – Q&As are meant to be non-binding, but what’s the point in taking the time to work on them, if an NCA can then simply ignore them altogether?
• In cases where the guidance is material for the implementation of EU law, should we use stronger convergence tools? (we say yes).

The ESAs Review proposal is the right vehicle to tackle these issues. A number of amendments worthy of note are being taken to this end. The time is ripe to get it right.

MIFID II is also a sizeable IT challenge, indeed perhaps an unprecedented data project both for market players and market regulators; this data is at the very core of MIFID II implementation since all transparency requirements hinge largely upon it. But while the amount of data requested is enormous, the quality of this data still leaves much to be desired.

The Financial Instrument Reference Data System (FIRDS) run by ESMA manages no less than 12 million ISINs; 1.3 million LEIs have been issued worldwide; around 500 million transaction reports are exchanged between NCAs on average every month through the Transaction Reporting Exchange Mechanism (TREM).

ESMA and NCAs are monitoring the quality of the reference data we use for EU financial instruments carefully. But/Nevertheless we still receive too much inconsistent data, too many incomplete declarations. The difficulties encountered when implementing the ‘double volume cap’ in Q1 2018 have made it obvious where there is room for improvement. The MiFID II transparency regime will only operate fully when data is accurate and complete. It is in our common interest to reach such completeness; to this end, cooperation between NCAs and trading venues must be stepped up to improve data quality.

I also note that, at this juncture, a consolidated tape for equity instruments has yet to emerge; EU authorities will have to treat this issue in due time.

Looking back over the past 9 months, I believe there is a growing awareness that a number of legislative fixes will need to be considered in the short to medium term to correct certain deficiencies. It will render all the more legitimate a number of recalibrations or redrafts to be included in this exercise, some of which may actually stem from Brexit’s consequences.

MiFID II may be a complex piece of law, but that does not mean we should shy away from re-opening it and fixing deficiencies where evidence may show that we have gone too far or have generated unintended consequences.

Some fixes are already underway: regarding the application of the tick size regime to systematic internalisers in order to ensure a level playing field, level 2 amendments are ongoing and a Level 1 change is proposed by ECON through the Investment Firm Review legislative proposal.

Elsewhere, the AMF is strongly a strong supporter of any move to revise and re-calibrate:

• the disclosure of costs and charges, which needs to be better aligned with the corresponding rules set out in the PRIIPs Regulation,
• the inducement rules that currently constrain the financing of equity research, a particularly sensitive matter for our SME ecosystem.

In addition, and based on the experience accumulated with platform equivalence last year and with the UK withdrawal in mind, it may be time to consider amending Article 23 MiFIR by narrowing down the scope of the share trading obligation, for instance to shares of issuers established in the EU (i.e. excluding shares of third-country issuers having a dual listing in the EU).

Such a change would considerably simplify the EU framework, making it unnecessary for the Commission to invest time and effort in equivalence assessments of numerous third country frameworks. It would also reconcile the text with what was in all likelihood the genuine intention of co-legislators.

More generally, many of MiFID II key requirements hinge on quantitative thresholds that were calibrated to suit an EU28 integrated market including London as a dominant financial center concentrating large volumes of trading. Allow me to illustrate that dominance; earlier I mentioned the 500 million transaction report exchanged monthly through TREM: the UK currently sends 72 % of these reports to EU27 countries, while EU 27 countries send 11 % of these reports to the UK (implying that 17 % of these reports are sent between EU 27 countries).

In the light of this, it makes sense to question whether MiFID II’s quantitative calibrations will be relevant tomorrow when the UK is no longer in the EU, for instance:

• Do the 4% and 8% volume thresholds of the ‘double volume cap’ mechanism [Art. 5 of MiFIR] still make sense for the EU27 and for the UK taken separately? If the majority of ‘dark’ trading currently takes place on UK trading venues, one could logically expect that such thresholds will turn out to be too low when applied to the UK market alone, and hence probably too high when applied to the EU27.
• Will the thresholds used by classes of commodities to frame the Ancillary Activity Test for commodities brokers [Art. 2 of RTS 20] need to be revised once their UK trading activity in commodity derivatives are no longer part of the EU picture? Just about 100% of metals, oil and coal derivatives are traded in the UK, so the ancillary activity test for these types of commodities will need to be revisited.

Of course, to perform a proper assessment, we will need to take the potentially new EU markets landscape into account, with a number of trading venues possibly relocating in the EU27.

Such recalibration can be seen as the logical outcome of fragmentation brought about by Brexit. It requires us also to take a longer term view and prepare ourselves for a « steady state » relationship with the UK. This may prove to be somewhat of a challenge if the ‘no deal’ scenario becomes a reality, since we are likely to be drawn away from this long-term vision, to face the urgency of dealing with cliff-edge effects immediately after the UK leaves the Union.

I will now say a few words on the third-country regime of MiFID II / MiFIR and cooperation arrangements. This is particularly relevant in view of Brexit, as it seems clear from the Commission’s declarations that the third-country regimes that exist in most (not all) EU primary acts dealing with financial services will be the tools by default to address the UK – EU27 relationship post-Brexit in those fields of activity.

Here is yet another reason to stop and think about whether the third-country regime of MiFIR, which was conceived at a time when the UK was part of the EU, is still appropriate.

I have no major concern with regard to the provision of investment services to retail investors. There is no third-country equivalence regime in MiFID II, but the Directive harmonizes to a large extent how each Member State must regulate third-country entities wishing to serve retail clients in each EU Member State separately (as you know, there is no EU passport for services provided to retail clients). In France, we have opted to require the establishment of a branch, and, if my understanding is correct, all Member States of the EU27 have taken a similar stance.

When it comes to third-country entities providing investment services to professional clients and eligible counterparties, things are quite different, since MiFIR does provide an equivalence regime.

In the absence of equivalence decision taken by the Commission regarding the UK under MiFIR [Art. 47 – not the platform equivalence necessary for the trading obligation for shares and certain derivatives, in Art. 23 & 28 MiFIR], UK firms will access EU27 clients on a Member State per Member State basis, subject to national third-country rules⁽¹⁾. In such a situation there is no EU harmonization, and no passport.

Harmonization kicks in if the Commission takes an equivalence decision towards a third country. In which case, third-country firms have access to the whole of the Single Market, but they are not supervised in the EU, have no obligation to have any legal presence in the EU and no MiFID/R rules apply to them whatsoever. Instead, we fully defer the supervision of these firms’ operations in the EU to third-country supervisors, and to third-country rules providing those have been assessed as equivalent to our rules.

To me, the key question here is whether this equivalence regime for wholesale services is fit for purpose. Overall my main concern is that this regime may lead to situations where third-country firms would obtain more favorable treatment than EU firms, putting the latter in a weakened position and, in some cases, endangering the high standards of market integrity and investor protection that we have built over time in the EU since the crisis.

One could argue that, since the MiFIR third country regime has never been applied, we lack the benefit of experience to amend it appropriately. This argument would be mistaken. The US fierce reaction to our EMIR 2.2 reform which could retroactively affect equivalence agreements recently granted, demonstrates the difficulty in changing the rules on equivalence after a number of assessments and decisions have already been previously made. If we agree that there are shortcomings in the architecture of the third-country regime – and I think that awareness of this is gradually growing– we should correct them now.

This is why the proposal by the Commission to introduce a number of targeted amendments to the equivalence regime, as part of its proposal to reform the prudential treatment of investment firms (“IFR”) and is much welcome despite my feeling that these could be further complemented.

As it stands, this regime is based on full substituted compliance towards third-country rules and supervisors: this will raise some practical issues which could have direct effects on investor protection and market integrity in the EU.

As regards conduct of business, transparency, reporting and trading obligations, third-country firms operating in the EU under the equivalence regime should be required to apply them. Going one step further, the relevance of MiFID II Level 3 demonstrates, in my view, just how essential it will be that the Commission – if and when it undertakes equivalence assessments under MiFIR in the future – does not limit its analysis to Level 1 principles alone.

Rather, it would seem justified that it takes full account of third-countries’ actual supervisory practices in applying MiFID II-equivalent rules. When doing that, the corpus of EU Level 3 measures (e.g. ESMA guidelines, opinions and Q&As), and in particular those critical Q&As and opinions issued to date by ESMA, should offer a useful benchmark to identify detrimental misalignments between the EU27 and a third country’s supervisory practice.

When I say that some MiFID rules should apply to third-country entities whose country has been declared equivalent, allow me to offer some examples by way of illustration:

• I believe that such third-country entities, when undertaking a trade in the EU, must be required to comply with the trading obligations for shares and derivatives under MiFIR (if the MiFIR text is not amended, this will not happen).
• They should also be held to the reporting obligations imposed on EU firms – transaction reports, post-trade disclosures, trade order data or financial instrument reference data, depending on the services they provide – or else significant trading activities in the Single Market will remain unmonitored and we will be treating third-country firms more favorably than EU investment firms.
• If third-country entities do not report under MiFIR, I fear that we may not be able to apply and enforce our transparency thresholds, liquidity assessment and double volume cap requirement properly, which is only possible if we have a comprehensive picture of the volumes traded in the EU.

Let there be no misunderstanding: the EU27 should be open to third-country firms and existing third-country regimes should be used whenever possible to reach that objective. My concern is to ensure that European regulators are in a position to monitor trading activity in the EU. It is not about undermining the philosophy of equivalence or forcing local presence.

Lastly, a word on cooperation, since the existence of cooperation agreements with third-country supervisors is a mandatory precondition for third-country regimes to become operational. We will need an MMOU signed both by the EU27 regulators and the UK FCA in order to cover all cooperation arrangements that are required by financial services legislation, for instance in the field of delegation and outsourcing. The industry must be fully aware that there is a strong commitment from ESMA and all EU27 NCAs to have an MMOU with the UK FCA in place. It is upon this assumption that market players should work, when putting their contingency plans in place.