Solvency II faces another delay. But this time it feels different. A clear and credible alternative timeline has not yet been set, fuelling concerns for the credibility of the entire project. As the Directive enters what may turn out to be a critical stage, Ian Poynton, partner and Jonathan Goodliffe, solicitor, of Freshfields Bruckhaus Deringer LLP reflect on its historical path and implications for the future.
The Solvency II project has been over ten years in the making. Its aim of modernising the prudential regulation of insurance and reinsuerance across the EEA by applying a more market consistent and risk-sensitive regime is laudable. But the process has proved far more complicated than initially envisaged. Persistent delays have forced rule makers to adapt the regime to changing economic circumstances – each change raising new issues and with them more delays. The latest of these can be viewed as one in a series of delays, raising concerns that Solvency II may prove to be an over ambitious European project.
Solvency II will replace the existing Solvency I regime, which in general, only sets minimum prudential standards. Many member states have applied stricter or ‘super-equivalent’ requirements. This is often referred to as ‘gold-plating’. Much of Solvency II, on the other hand, is intended to apply fully harmonised standards. This leaves member states with much less room for manoeuvre and makes the process for agreeing the standards necessarily much more challenging.
The European Commission launched the legislative process with a view to adopting the Solvency II regime in July 2007. The Commission’s proposal announced that ‘the new solvency provisions are principles based’.
The Lamfalussy architecture
The Commission’s proposal applied the then four-level structure of the Lamfalussy financial services architecture.
Level 1 text: Level 1 was the directive proposed by the Commission, which was intended to contain only the main rules of the regime.
Level 2 text: It was to be supplemented by more detailed rules to be proposed by the Commission at Level 2.
Level 3 text: Non-binding standards and guidance to be adopted by the Committee of European Insurance and Occupational Pensions Supervisors (CEIOPS) at Level 3.
Level 4 text: The Commission would in due course review member state transposition of the rules and compliance at level four and take enforcement action if necessary. Experience suggests that action of this kind often is necessary, since many member states are generally late in transposing directives in the insurance field.
The timetable for the Solvency II Directive has regularly had to be put back because of the scale of the project and the difficulty of achieving enough agreement within Europe.
In 2004 the European Commission was working towards a target completion date of 2008. But the formal legislative process for adopting the directive was not launched until July 2007. The Level 1 directive was eventually adopted in November 2009. It provided for the regime to come into force from 1 November 2012, and was more detailed and prescriptive than contemplated by Baron Lamfalussy and his ‘committee of three wise men’. The Directive ran to 155 pages in the densely typed Official Journal.
An early victim of disagreement between member states was the ‘group support regime’. Originally proposed by the UK, this would have allowed insurance groups considerable flexibility in complying with capital adequacy requirements. They would have been able to maintain the higher of the two capital requirements under the Solvency II regime, the SCR, on a group basis rather than within every solo member of the group. But the proposed group support regime proved controversial and had to be dropped.
By November 2009 ‘principles based’ regulation had been found to be unsatisfactory. It had been succeeded, particularly in the UK, by a more ‘intrusive’ approach, which has since been developed within the government’s regulatory reform programme. CEIOPS was transformed from an association of insurance supervisors into a European regulatory authority in its own right, the European Insurance and Occupational Pensions Authority (EIOPA).
The Solvency II Level 1 text had to be adapted to the Lisbon Treaty and brought up to date. So in January 2011 the Commission proposed an ‘Omnibus II Directive’. Among other things this was intended to:
- give EIOPA enhanced powers within the regime (eg to resolve disputes between supervisors);
- insert a new ‘Level 2.5’ into the Lamfalussy structure. Level 2.5 as proposed by the Commission required EIOPA to draft implementing technical standards (ITS) and for the Commission to adopt those standards. ITS were to be used to cover technical areas of the regime that were not politically controversial; and
- provide for specified aspects of the Solvency II regime (including, for instance, governance and capital adequacy requirements) to be the subject of transitional or grandfathering provisions.
These changes have shifted regulatory power within Europe. Hector Sants, the chief executive of the FSA until mid-2012, remarked: “Going forward, the FSA and successor authorities will thus essentially be a supervisory arm of an EU policy-setting body”.
It is also worth mentioning that under the modernised Lamfalussy architecture, ‘non- controversial’ Level 2.5 technical rules may be adopted by means of ‘regulatory technical standards’ (RTS) as well as, or instead of, ITS. The most significant difference between the two is that the European Parliament is in a better position to influence RTS than ITS.
The regulation creating EIOPA also strengthened the status of Level 3 guidance by providing that it should be binding on member state supervisors on a ‘comply or explain’ basis.
It was hoped that Omnibus II would be agreed within a short interval. But it has proved highly controversial and is still not yet agreed (as of October 2012).
Proposed Level 2 and 2.5 rules and Level 3 guidance
In 2009 and 2010 CEIOPS consulted publicly on the policy to be covered by the Level 2 rules. It gave its final advice on the proposed content of the Level 2 rules to the Commission after the consultation process. This was followed by a quantitative impact study (QIS5) carried out by the Commission with CEIOPS.
The insurance industry and a minority within what was then still CEIOPS (which appears to have included the FSA) persuaded the Commission that the CEIOPS advice applied too strict a prudential standard in many significant respects. This was reflected in the technical specification for QIS5. One of the controversial areas related to the discounting of technical provisions in long-term business. It covered the extent to which the level of technical provisions might be reduced where illiquid assets are used to cover illiquid liabilities (the so called ‘illiquidity premium’).
The QIS5 technical specification was largely followed in the first draft of the Level 2 rules in October 2010. A revised version of these rules was produced in November 2011. These rules were composed on the assumption that Omnibus II would be adopted, which it still has not been.
This seems to have been one of the justifications for not officially publishing the draft Level 2 rules. Instead they have been circulated in draft to a limited number of ‘significant stakeholders’. The October 2010 version has since been published online by the Finnish parliament. The November 2011 version is more difficult – but by no means impossible – to obtain.
Meanwhile EIOPA is also consulting on draft Level 2.5 technical standards and draft Level 3 guidance. Some of these consultations are public and some are ‘pre-consultations’ addressed to the same list of ‘significant stakeholders’ used by the Commission for the Level 2 rules.
These factors have reduced the transparency of the policymaking process within the Solvency II project and thus the scope of the input which might otherwise have been provided.
Continuing disagreement on Omnibus II: policy
As Omnibus II is a Level 1 directive it needs to be adopted through the ‘ordinary legislative procedure’. This involves the Commission, the Council and the European Parliament.
Each of these bodies and each member state and/or MEP represented in those bodies may have a different perspective. Each body has published different versions of Omnibus II, articulating its own approach to what rules the final Solvency II package should contain and at what levels they should be set out.
Areas of disagreement have included the calculation of technical provisions for long-term products. In QIS5, discussion of this issue focused on the possible use of an illiquidity premium. More recently, the use of a Matching Adjustment and a Counter-Cyclical Premium have been put forward.
Another issue is how prudential regimes in countries outside Europe will be assessed, in terms of whether they are equivalent to the Solvency II regime. This in turn affects:
- the regulatory treatment of reinsurance ceded to firms in those countries;
- the regulatory capital European insurance groups operating in those countries need to maintain; and
- whether supervisors in Europe are to rely on group supervision exercised in those countries for insurance groups based in those countries, but with operations in Europe.
A third problematic issue concerns sovereign debt. Under the Solvency II rules as they stand, no capital charge is applied to sovereign debt issued by EEA member states. It seems to be generally accepted that this will need to change, but developing an alternative formula will be difficult.
Continuing disagreement on Omnibus II: legal issues
Apart from policy disagreements, there have been disagreements over whether specific rules should be set out in Levels 1,2 or 2.5. The Commission’s initial draft proposed that most of the regulatory material should be set out in the Level 2 rules and in ITS, where the Commission is in the strongest position to drive its agenda through.
By contrast, the Parliament has proposed more use of RTS, in relation to which it has more leverage. The Parliament has also proposed that much of the material, which the Commission and the Council were content to have set out in Level 2 rules, should be promoted to the Level 1 text. This includes, for instance, the rules for the Matching Adjustment and grandfathering.
Again this gives the Parliament more leverage. But there is a risk of the rules becoming rigid, in view of the difficulty of making changes to Level 1 text.
Experience suggests that financial regulation needs to be flexible, because sometimes rules adopted in year one will be problematic in year two. This is one of the justifications for giving supervisors such as the FSA power to adopt their own rules. Some existing rules within Solvency I would probably have been changed long ago if more flexibility had existed. An example of this is the current rules on admissible assets and exposure limits for investment of technical provisions.
It is fair to say that the ‘hard coding’ of rules into the Level 1 text has been justified on the basis that there is doubt over whether the delegated powers to make Level 2 or 2.5 rules are framed in wide enough terms in the Level 1 text. So there may be a concern that some Level 2-plus rules might be challenged on the grounds that they are ‘ultra vires’, i.e. outside the Commission’s delegated powers. But that issue might be addressed by widening the delegated powers within Level 1, as proposed by Omnibus II.
Level of harmonisation
There are shades of grey between the minimum harmonisation approach of Solvency I and the ideal articulated by Jean Monnet in the 1950s and recently endorsed by EIOPA: “A fusion of the interests of the European peoples and not merely another effort to maintain an equilibrium of those interests”.
But Solvency II does leave a number of regulatory matters to the discretion of member state supervisors. These include how surpluses are treated in with-profit funds under article 91. In other areas, harmonised rules at European level are unlikely to be in place until some time after the Solvency II regime comes into force. These areas include the supervision of risk concentrations within insurance groups and the regulation of finite reinsurance.
Some of the issues holding up agreement on Omnibus II, such as the calculation of technical provisions in long-term business, reflect the considerable differences between national insurance markets. The proposed matching adjustment, for instance, is an issue for only some member states, such as the UK, where the lack of an adjustment would significantly increase the price of many retirement products.
Delays to commencing Solvency II
When the timetable was set in 2009, 1 November 2012 was already a tight target to complete the project. Paul Sharma of the Financial Services Authority, however, originally claimed that it was “less likely to be delayed than the Olympics”.
The commencement was soon postponed to the end of 2012 to tie in with the financial year ends of most (re)insurers. In 2011 the concept of a ‘bifurcated commencement’ was introduced. This required member states to:
- have the rules in place for the commencement of the regime at the beginning of 2013; and
- for the commencement to be deferred to 1 January 2014.
The first point was soon, however, advanced to mid 2013.
What was hoped to be the final vote in the European Parliament on the Omnibus II Directive was until recently due to take place in November 2012. ‘Trilogue discussions’ between the Commission, the Council and the Parliament have not yet reached full agreement (November 2012).
A possible formula to resolve the issues for long-term business is to be subject to an impact assessment process, which is not due to end until March 2013. The Commission now proposes that the final vote on Omnibus II should be deferred until then. So it seems increasingly likely that the adoption date will be put off until January 2014, and the commencement date to January 2015. The Commission, however, has not yet officially announced this and there are suggestions that the regime might come into force for non-life business at an earlier point than for life business.
Impact of delays
These delays have had quite negative consequences. Solvency II was at one time seen as an example to the world in the prudential regulation of insurance. It may now, however, be seen as over-ambitious.
The uncertainty also affects the longer-term planning of insurance firms and those who invest in their products, whether they are policy holders or, for instance, private equity firms. Many insurance groups have invested heavily in complying with a regime that is repeatedly being deferred. Meanwhile they continue to bear the cost of complying with the existing regime.
The UK Parliament seems to be able to influence regulatory policy without getting involved in detailed drafting of the FSA Handbook (in relation, for instance, to with-profits products). There may be a case for adjusting the European legislative processes to achieve a similar result, and/or for addressing controversial issues within the project in a way that does not hold up the entire legislative timetable.
So, for instance, the Commission has recently asked EIOPA to consider whether capital requirements under Solvency II need to be adjusted to encourage insurers to invest more in infrastructure, small- or medium-sized enterprises and socially responsible investments. Arguably, this is as important as the long-term business issues, but because the capital requirements are set out in the Level 2 rules, the process to adjust them is less problematic in procedural terms.
The views expressed are the authors’ own.