Address by Sylvia Cronin, Director of Insurance, at the Central Bank of Ireland Solvency II Forum

20 April 2015 Speech

Good morning ladies and gentlemen. In November 2012 I sat in the audience at a Central Bank of Ireland Insurance Forum where there was still a major degree of uncertainty around the future of the Solvency II project. I believe that the EIOPA initiative to introduce Preparatory Guidelines on Solvency II was instrumental in keeping regulators and industry focused on the Solvency II end-game. 

I would like to take this opportunity to acknowledge the work undertaken by Gabriel Bernardino, today’s guest speaker, and my EIOPA colleagues in achieving this. Over the last number of months the regulatory agenda has achieved a large degree of certainty and we have witnessed significant progress through:

  • The adoption of the Omnibus II Directive in March 2014.
  • The adoption of the Solvency II Delegated Acts in October 2014.
  • The publication of the first set of EIOPA guidelines in February 2015.
  • The publication of the first Implementing Technical Standards by the European Commission in March 2015.

This leaves just two further pieces of work in order to finalise the Solvency II framework: 

  • The second set of EIOPA Guidelines.
  • The publication of the remaining Implementing Technical Standards. 

It is expected that these elements will be completed by July 2015.

The Insurance Directorate is engaging with supervisors and industry to effect the changes required under the new supervisory framework and to prepare for the years ahead. 

As the 1 January implementation deadline approaches I’d like to take this opportunity to reflect on the experience gained through the application of the preparatory phase Guidelines. I shall outline our experiences as they relate to the FLAOR, Preparatory Submission of Information and the Internal Model pre-application process.

You will all have reviewed and provided input to the FLAOR documents submitted by your firms at the end of 2014. The FLAOR requirement was introduced to allow undertakings take an incremental approach to the preparation of their first ORSA in 2016. Of course, the nature of the ORSA requirement limits the scope of the feedback you can expect from the Central Bank; in a sense, the ORSA is a ‘self-portrait’ of your firms, and it would be useless to us as regulators if we told you exactly how to paint it. From our review of the FLAORs submitted thus far, we have noticed some general trends which I would like to draw your attention to. My remarks will provide guidance on how you should be involved in the creation of the ORSA document, that when done well, will be an invaluable source of information to help your firms to make sound strategic decisions. 

Before I continue, I should note that we are reasonably satisfied with the quality of the FLAOR reports which your firms submitted this year. Although these reports were to be prepared on a “best efforts” basis - many of you attempted to produce a full ORSA report, which bodes well for next year. The FLAOR reports we received have varied considerably in content within similar sectors. This variance is both a potential cause for concern and also a strength – as at the very least, it demonstrates that undertakings are making genuine attempts to produce reports that are firm specific. So what issues would I like to draw your attention to?

Firstly, Board ownership. For the ORSA to become a part of business as usual, it will not be enough for the actuary or risk function to present you and the Board with an essentially finished ORSA, which you then approve and file until the next ORSA comes around. The supervisory view is that this is not and will never be considered Board ownership. How could you possibly feel like you own a document into which you had no input? And how could the ORSA possibly be useful to you, if you don’t ensure the inclusion of the kind of risks, sensitivities, stresses and scenarios which you would find useful? 

The process by which your undertakings produce the ORSA is as important as the document itself. Why? Because the ORSA process represents a tremendous opportunity for your undertaking to integrate and join not just your existing infrastructure and processes in relation to risk management, but existing processes in capital management, and business planning too. From such integration comes improved information flow throughout your organisations, and from this, improved decision making at all levels, not least at Board level. By ‘improved’ decisions I mean decisions which are more risk focused. Decisions informed on the basis of comprehensive and accurate information. 

Such an approach is better for your business as it ensures closer alignment between decisions made at Board level, in terms of overall strategy and risk appetite of your firms, and the operations and decisions taken within your business units. This is the thinking which underpins the requirement for you to have a record of each ORSA as well a detailed ORSA policy. Our supervisory teams will take an active interest in these documents in order to evidence the extent to which the ORSA is embedded within your undertakings. Going forward, we expect to see better, more comprehensive information in relation to the ORSA process. For the ORSA to pass the so-called use test, Board ownership and evidence of Board ownership will be crucial going forward.

A key element of the ORSA is assessing your capital needs. The Board is uniquely positioned to define, describe and discuss these risks and their associated capital implications. In particular, the range of experience and perspective at the Board level can precipitate productive discussion on intangible, non-quantifiable risks and associated scenarios, discussions which can be clearly reflected in the ORSA.

Our experience thus far is that some of you are producing FLAORs which do not appropriately stress test what are, in our opinion, your key risks, and which ignore those risks that are difficult to quantify. This has been especially true of those of you with group ORSAs, which tend to be inadequately tailored to local entities. Examples of such risks include operational and reputational risks, group risk, pension risk and strategic risks. If these risks are not key risks for your firm – and this is certainly your judgement to make - then your ORSA must explain why this is the case. And if these risks cannot be quantified, then you should describe the mitigating actions you are taking. 

In addition, the stress tests used to assess risks and documented within a sizeable number of the FLAORs were too benign in nature, and were not realistic or reflective of the firm’s historical experience. For example, if your undertaking has experienced claims variances in the region of 10% or even 20% year on year, a 5% shock to claims will not provide a sufficient picture of a “stressed” claims environment. The danger here is that without assessment and challenge from you and the Board, your ORSA will paint an overly optimistic and incomplete portrait of your undertaking’s financial resilience. 

I’d like to emphasise also that this discussion should extend to the fundamental assumptions which underpin your ORSA process, including in particular the business plans upon which the process relies and the time period which it considers. Certain reports we have reviewed so far typically consider time horizons and business plans over too short a time period, for example two years, or over time periods which are ill defined. The Board should challenge such fundamental assumptions and highlights the importance of Board involvement in the ORSA process from the initial planning stages onward and throughout.

While the Central Bank will continue to use the online reporting tool for low and medium low undertakings, this does not mean that we are not focusing on such undertakings. It is our intention to apply the current practice of thematic reviews for low and medium low undertakings to the ORSA, thereby reviewing a sample of ORSA reports each year. Moreover, such undertakings can expect supervisors to request, and to examine in detail, ORSA reports in advance of any engagement. High and Medium High firms can continue to expect their supervisors to review their ORSAs, providing tailored feedback as appropriate. Supervisory judgement is a cornerstone of the Solvency II regime. Where our expectations around the ORSA are not met you can expect increased supervisory oversight. Where we identify deficiencies in your ORSA process we will dedicate more resources to ensure our expectations are met. 

While your undertakings may have already begun the process of producing the FLAOR for 2015, I would hope that your undertakings will begin to address, as soon as possible, the issues I have raised here this morning. While the ORSA report, and the steps required to produce it, can be expected to change over time, as the process becomes increasingly embedded into your organisations, ORSAs which comply with Solvency II regulations must be produced in 2016. 

Other important lessons from the preparatory phase relate to data and reporting. The third of the Three Pillars of Solvency II - “Reporting and Disclosure” - is integral to the new regulatory framework and introduces very substantial reporting and disclosure requirements. 

A number of industry readiness surveys over the past year have consistently shown that this is the area of Solvency II implementation which requires the most work. Based on these surveys and our on-going engagement with industry; it is an area which gives us cause for concern. 

Some reports point to the difficulties that undertakings face in overlaying the extensive new reporting capabilities onto current risk and finance architecture. Much of this reporting architecture is on legacy systems which may not be able to provide you with sufficient data for a Solvency II risk-managed world. While an opportunity exists to overhaul and build more integrated systems, the early uncertainties around Solvency II implementation understandably caused many firms to delay decisions on making the necessary investment. This uncertainty is now well past and, together with the impending timelines, such decisions are now overdue.

Just to remind ourselves, those of you representing undertakings rated as High or Medium High impact under our PRISM risk-rating system, 33 in total, are subject to the so-called “Preparatory Phase”. This means that you must be ready to provide the narrative and quantitative information outlined in the 2013 Central Bank “Guidelines on Submission of Information”. This includes being ready to submit a set of Reporting templates in XBRL system language at the end of May this year in respect of 2014 year-end and again during November in respect of Q3 2015. 

Last month we invited firms, through your Compliance Officers, to take part in the External User Testing phase of our preparations. During this test phase undertakings can submit returns in a test environment in order to check the readiness of their own processes as well as ours. This testing phase is currently underway and will run up to 1st May. Somewhat surprisingly, of the 33 undertakings who must submit returns “for real” in May, only 17 firms elected to take part. This suggests that many firms have such faith in their preparations and systems that they don’t feel the need to take part; or the more likely explanation is that many firms are still struggling with their reporting plans. 

The offer still remains open until the end of this week.  

Looking beyond the Preparatory Phase all firms will need to be ready for full Solvency II reporting commencing with what are called “Day One reporting” from January 2016. During Q3 of this year we will again open for External User testing for all Central Bank regulated firms. I strongly advise that all undertakings test their systems at that point.

While the volume of quantitative reporting will clearly be a challenge for even the most highly-resourced organisations there are also signs that the narrative reporting is proving a challenge for some. A major 2014 report by EY surveying 180 European insurers across 20 countries found that there was very slow progress by firms on the specification and design of Regular Supervisory Reporting (RSR), the Solvency and Financial Condition Report (SFCR) and the ORSA report, with nearly 80% of respondents not meeting most requirements. These are key reports on which regulators will rely for much of their view on the level of risk and quality of governance and it is a concern that these, which require less system development, may prove to be difficult to produce.  

I appreciate that there is a lot of work done and much more to do when it comes to data and reporting under Solvency II. The Central Bank has been working very hard to make the changes required to take the reporting from you and I am glad to say that we are on track in this regard. We are also conscious of the need to support you in your plans and development work and the External User testing phase should be seen as a useful test.

Further lessons can be drawn from our experience of the Internal Model pre-application process. For many insurance undertakings and groups the adoption of Solvency II will include the implementation of partial or full internal models to calculate their Solvency Capital Requirements based on their risk profile. 

In fact, we had in our pre-application process one of the highest numbers of internal models in Europe. 

We have now reached the end of the pre-application process and have entered the formal application process. It is therefore timely to note some of the key lessons learnt to date, many of the lessons apply to all undertakings. 

Undertakings applying for internal model approval and Actuaries in standard formula undertakings use a range of models. The Board and senior management need to understand these models, their limitations and the sensitivity of its outputs.  

In particular, the Board need to ensure that the risks of inaccurate output of a model are communicated in a transparent manner, and ensure the users of the model are aware of the risks that may lead to the models being used incorrectly or inappropriately. You need to heed the warning lessons from the not so distant past. The over reliance on complex Banking models without proper understanding of their limitations is considered as one of the causes of the financial crisis in 2007-2008.

Robust Model Governance is another critical aspect of the internal model approval process.  The feedback from our reviews is relevant not only to internal model undertakings but all undertakings involved in the ORSA process. 

  • The senior management and the Board should assess and understand the stability of the risk measures and models over time and there should be appropriate validation of results in place. 
  • The process for removing or adding scenarios and stress tests should be comprehensive and challenged. 
  • There should be robust governance around expert judgement. 
  • Where the risk profile of the undertaking changes significantly the model or Standard Formula must be reassessed for continued suitability. 
  • And last but not least, where an undertaking seeks internal model approval the model must be used in practical business decision and strategic decisions and key issues discussed at the Board level should be understood in the context of the Internal Model results. This requires Boards of subsidiaries of international groups to be satisfied in their own right with the appropriateness of group model methods and assumptions in respect of the local business. 

Another important area involves data. Assessment of completeness, accuracy and integrity of data is a task that we perform not only during assessment of internal modes but also as part of our day-to-day supervisory review process, such as reserve reviews.

Here I would like to stress the following points:

  • We require more detailed rationale when an undertaking removes outliers from the data, and we also require evidence of how the events removed can be expected not to re-occur. 
  • We also require the undertaking to carry out sensitivity tests on key assumptions. 
    In addition, where an undertaking uses external data to support their internal analysis we request that the Board should satisfy themselves that such data is appropriate.

Looking forward into 2015 and beyond

In the context of the current challenges it is important for the Boards to understand their solvency capital requirements, whether they are going to calculate them using internal model, undertaking specific parameters or standard formula. In each case they have to bear in mind not only existing risks facing the organisation but also emerging risks.  

Going forward we also envisage a more robust challenge of data used and meaningful segregation of data to expose any underlying trends or data limitations as they emerge.

As the ORSA process becomes more embedded in the risk management process, we expect a deeper insight of the risk profile of an undertaking and a more robust challenge of the solvency capital requirements.  As risk discussions become more prospective in nature this in turn places key judgements and expert opinions at the heart of those risk discussions. 

What will happen once Solvency II comes into effect?

The internal model approval or decision to select a standard formula for the SCR calculation is not the end of the journey. As I stated earlier it is only the beginning. The Central Bank will need to ensure that both standard formula and internal models remain fit for purpose in the years to come. The internal model or standard formula should appropriately reflect the risk profile of the undertaking. 

Sharing our observations, conducting robust interactions with senior management and the boards, and increasing standards will ensure a deeper insight into risk, and a smoother transition into Solvency II.

Today’s event has provided the opportunity to reflect on the work completed to date. It also provides a timely opportunity to remind ourselves of the purpose of the new legislation. 

The aim of the Solvency II regime is twofold: 

  1. To ensure the financial soundness of insurance undertakings.
  2. To protect policyholders. 

Most of my speech has focussed on the impact of solvency on  insurance undertakings - so I will briefly touch on the importance of consumer protection. 

At European level EIOPA is charged with promoting transparency and fairness in the market for insurance products. The Central Bank continues to contribute and influence the EIOPA agenda in the area of Consumer Protection. Nevertheless, a major challenge at national and European level is to ensure that the consumer protection framework remains fit for purpose. My colleagues in the Consumer Protection Directorate will continue to monitor and respond to the changing consumer environment with the aim of embedding good consumer-based practices within Irish insurance undertakings.  

In conclusion, the development of a more fully harmonised regime is a huge step forward for the European insurance market.

The future challenge for EIOPA, and national supervisors, is to ensure that Solvency II is implemented and embedded in a consistent way throughout the EU. Your challenge will be to provide direction and leadership to achieve best practice in your undertakings as we move towards implementation. A major lesson drawn from our supervisory engagement during the Preparatory Phase is that full engagement from the board is crucial for the successful implementation of Solvency II. In this regard our expectation is that undertakings will ensure there are sufficient resources and internal capability in place to deliver the Solvency II project. Over the coming months we will continue to challenge undertakings to ensure that both the Board and senior management are fully engaged in the Solvency II project.

Solvency II has been described as the marathon of marathons. The entry into application is now close and collectively we need to focus our efforts to achieve this target.

And remember it is only the beginning of the next new exciting journey.