This blog post, co-authored with Dirk Schoenmaker, was published today by Bruegel.
Update (April 10): the same post was also published by the Peterson Institute.
The European Commission recently published a consultation on the operation of the European Supervisory Authorities (ESAs), the three EU financial regulatory agencies established in 2011. This debate is timely. In the short term, Brexit will force a relocation of the European Banking Authority (EBA), which is currently based in London. Separately, but also as a consequence of Brexit, the European Union needs to rethink its capital markets oversight framework. Following the recent pooling of euro-area banking supervision at the European Central Bank (ECB), it should further separate prudential oversight from the protection of savers, investors and market integrity, thus implementing an institutional concept known in financial regulatory debates as “twin peaks.” In this blog post, we tentatively explore how to bring these strands together in the European Union’s next legislative steps.
The three ESAs are the EBA, the European Insurance and Occupational Pensions Authority (EIOPA), and the European Securities and Markets Authority (ESMA). This sectoral approach of having separate bodies for banking, insurance and securities is a historical legacy. The European Union had set up sectoral committees in the 1990s and 2000s: first came the Forum of European Securities Commissions (FESCO) in 1997, followed by the so-called Lamfalussy Level-3 Committees of the 2000s. After the financial shock of 2007-08, the Larosière Report of February 2009 crystallised a consensus to move one step further towards EU-level policy integration and a “single rulebook.” However, given widespread lingering differences among member states, the Larosière Report opted for the path of least resistance for institutional design, namely to keep the sectoral scope and intergovernmental governance of the three Lamfalussy Committees while transforming them into fully-formed EU agencies with their own legal personality.
In fact, the apparent institutional parallelism between the three agencies was slightly misleading from the start. Not only do they cover different sectors, they also have different roles. Notably, ESMA was quickly granted directly binding powers over several market segments: it is the sole supervisor of credit market agencies and trade repositories,[1] and it also has authority to recognise third-country Central Counterparties (CCPs) for their operation in the European Union. In 2015, ESMA created a dedicated Supervision Department to handle these tasks. The EBA and EIOPA have no comparable competencies.
Almost a decade later, after major developments like banking union and Brexit, it is time to re-examine the ESA architecture. It should be reframed in a long-term vision for EU financial oversight, and this vision should bebased on the ‘twin peaks’ concept. Twin peaks is shorthand for the separation of conduct-of-business supervision from prudential supervision, originally proposed by Michael Taylor in a 1995 pamphlet and since adopted by jurisdictions including Australia, the Netherlands and the UK.
There are three main reasons to support a twin peaks model.
First, Europe’s banks and insurers are often linked in financial conglomerates, which warrants integrated supervision of banking and insurance. Based on end-2015 disclosures, we calculate that bank-insurance conglomerates account for 49% of the aggregate assets of the EU’s 30 largest banks, and for 40% of the assets of its 30 largest insurers. In a crisis, problems may emerge in any part of a financial conglomerate, even though the nature and maturity of liabilities are structurally different between banking and insurance. Since the European Union does not appear to envisage a tighter regulatory separation between banking and insurance, it should move towards more integrated prudential supervision of both sectors. In the same spirit, the European Union should amend its capital requirements legislation to reverse the so-called Copenhagen compromise, which allows for double-counting of the same capital for banking and insurance regulation purposes and does not fully comply with the global standards set by the Basel Committee on Banking Supervision.
Second, the EU27 urgently needs to upgrade the supervision of its capital markets after Brexit. An empowered EU-level markets supervisor – in practice, a reformed ESMA with additional direct authority – is needed to adapt quickly to this new reality, as we argued in an earlier blog post. This would help to revive the EU’s capital markets union, whose rationale is sound but which so far has delivered only limited results. A twin peaks design enhances the importance of the markets supervisor by granting it authority over most aspects of the protection of investors’ and savers’ interests.
Third, prudential supervision and conduct-of-business supervision (including markets supervision) require different mindsets, skills and approaches. Prudential supervision requires staff trained in economics, finance and/or accountancy, while conduct-of-business supervision is more behavioural and legalistic. Conduct-of-business supervision involves policing the conduct of financial institutions in the markets (insider trading, market abuse, disclosure, etc) and towards clients (adequate information provision, duty of care, know your customer, etc). But on these issues, long and sometimes painful experience demonstrates that prudential supervisors often prioritise the interests of supervised entities over those of their customers.
Overlapping tasks between the prudential and markets authorities typically result in the failure to protect savers. This is one of the lessons of the debacle of securities misselling in Italy, with major negative consequences for individual households, for the reputation of banks, for that of public authorities, and for financial stability.
The recommendation for twin peaks is a long-term aspirational goal. Indeed, the treaty basis for ECB banking supervision expressly prevents its extension to insurance. Current national arrangements vary widely. For example, Cyprus, Greece, Italy, Luxembourg, Portugal, Slovenia and Spain have separate insurance supervisors. Moreover, in many member states, conduct-of-business supervision of banks and insurers is at least partly located in the prudential authorities. Nevertheless, defining a long-term goal is important for making the right decisions on short-term issues.
The most urgent priority is the relocation of EBA. Uncertainty about this will have damaging consequences for EBA staff morale and makes recruitment of qualified experts near-impossible. To remedy this, we recommend that the European Union announce a decision on the new location as soon as the late spring or summer of 2017.
We simultaneously caution against recently broached suggestions to merge EBA with either EIOPA or ESMA, which we view as counterproductive. An EBA-EIOPA merger would not achieve twin peaks, since neither agency is actually a supervisor. Instead, it would generate unwieldy governance complexities given the absence of a counterpart to banking union in the insurance space. An EBA-ESMA merger would be even worse, going squarely against the twin-peaks vision and creating damaging conflicts of interest between the conduct-of-business and prudential mandates of the resulting entity. It is also worth recalling that EBA is needed for technical rule-making and supervisory convergence, as long as the banking union framework does not cover all EU member states. EBA is the right venue to balance the interests between the “ins” and “outs” of banking union: as long as there are “outs,” its competencies cannot be directly assumed by the ECB.
Given all these considerations and the operational imperative to come to a decision quickly, we suggest that the 2017 decision on EBA should be only about relocation, and not about any further reform at this stage. Even though the choice of a new location can be expected to be a political one, rather than policy-driven, it is worth noting that having EBA in Frankfurt, even without a merger with EIOPA, would facilitate joint work with both EIOPA and ECB banking supervisors. It would also ease an eventual merger in a hypothetical long-term scenario in which all EU countries would have joined the banking union. Alternatively, EU leaders may opt for locating EBA in a non-euro-area member state, as a signal that the post-Brexit dominance of euro-area banks (which represent about nine-tenth of the EU27’s total banking assets) will not result in the “outs” being ignored in the preparation of banking regulations.
While EBA reform can wait, the reform and further empowerment of ESMA should be an immediate priority for EU leaders. This would give an important sense of direction for EU financial markets policy at a time when many financial firms need to determine their operational response to Brexit by end-2017 or early 2018. In essence, ESMA’s governance and funding should be reformed to make it more independent and accountable. Its direct authority should be extended to include registration of all EU-critical market infrastructure (in liaison with prudential authorities for CCPs which are systemically important for financial stability, including presumably a legislative empowerment of the ECB and Single Resolution Board in that area); accounting enforcement; audit firm supervision; wholesale market activities of investment banks (under the EU Markets in Financial Instruments legislation); and most conduct-of-business regulatory relations with market authorities outside the European Union.
In accordance with the subsidiarity principle, most other tasks currently lodged in national authorities, such as authorisations for initial public offerings, fund management registrations, and the conduct-of-business supervision of smaller investment and insurance intermediaries to protect retail investors and savers, should remain there for the foreseeable future. However, national authorities should operate under ESMA’s umbrella to ensure EU-wide consistency.
The pooling of conduct-of-business supervision under ESMA’s oversight also implies that an increasing number of financial firms (including banks, insurers and CCPs) will be subject to “double supervision” by both conduct-of-business and prudential authorities. Special interests will surely denounce this duality as unacceptable complexity. But given the different objectives of the two types of supervision, it should be viewed as not only unavoidable, but actually desirable.
The European Commission’s review of Capital Markets Union, scheduled for June 2017, provides an appropriate opportunity to trigger a decision-making process that could result in EU decisions of principle on ESMA reform and further empowerment later this year. Of course, the finalisation and implementation of the corresponding legislation would inevitably take longer.
As for insurance, the EU Solvency II legislation, in force since last year, allows insurers to use their internal models for capital purposes. Given the strong cross-border nature of the large European insurers, EIOPA should become responsible for the approval and monitoring of these internal models. EIOPA might thus get direct supervisory tasks vis-à-vis the large insurers in the European Union.
These decisions would support the long-term vision of a Twin Peaks regulatory framework for the European Union. The completion of that framework would require treaty change, to merge ECB banking supervision with a still-to-be-created EU level of insurance supervision; and an extension of banking union to all EU member states, to allow the duplication between ECB Banking Supervision to be eliminated. None of these two conditions can be expected to be fulfilled any time soon. In the meantime, the swift relocation of EBA, elevation of ESMA into a vigorous conduct-of-business supervisor at the EU level, and gradual build-up of EIOPA direct authority, would together represent a consistent and effective European approach.
[1] Disclosure: one of the authors (Nicolas Véron) is an independent board member for a trade repository supervised by ESMA. More details are available here.