Hot Topic: Understanding the 17 March 2011 Summary On 14 March 2011 the European Insurance and Occupational Pensions Authority (EIOPA) published the results of the fifth Quantitative Impact Study (QIS5) for Solvency II. QIS5 was conducted between August and November 2010 to assess the impact of the revised valuation rules and the Solvency II capital requirements on the European insurance industry. QIS5 is the final and most comprehensive road test of Solvency II s requirements and gives the best picture of the likely solvency position of insurers under Solvency II s requirements which come into force in January 2013. The findings of QIS5 will influence the European Commission in finalising its proposals, due to be published in June 2011, as to how Solvency II will be implemented. Whilst in many areas EIOPA s report indicates that the requirements tested in QIS5 were broadly appropriate, there are a number of significant areas identified where further changes are likely. Why are QIS5 results important? 68% of entities within Solvency II s scope participated in QIS5. The level of granularity required from the exercise coupled with the high level of participation means that the results should give the firmest indication yet of Solvency II s requirements. QIS5 has brought the capital implications of Solvency II into sharp focus and left insurers with important decisions to make. Those companies that address the impacts too late will face difficult questions from investors, analysts and the market and will find it progressively more difficult to rectify their capital situation before the 2013 deadline. What are the key messages? s? Capital position EIOPA s view is that the European insurance industry remains well capitalised under Solvency II s requirements. However, QIS5 has shown that Solvency II s capital requirements will, for many, be more onerous than the existing regime. Results show an average 12% reduction in insurers capital surplus compared to Solvency I. 15% of participants reported a deficit against Solvency II s standard formula Solvency Capital Requirement (SCR). The focus of QIS5 was on testing the standard formula SCR. In practice, insurers may instead use full or partial internal models to calculate their capital requirements. Whilst QIS5 sought to gather information from insurers planning to use internal models, comparatively few
submitted information on this basis (perhaps reflecting that, even where internal models are planned to be used, in many cases those models are not yet finalised). As a result EIOPA concluded that QIS5 did not allow conclusions to be drawn as to how internal model capital requirements compare with those under the standard formula. For groups, QIS5 highlighted an average 43% reduction in capital surplus compared to Solvency I when applying the standard formula to consolidated data. However, this position may be improved where internal models are used or where local rules are used for non-eea insurers in equivalent territories. All the figures above represent pan-european averages. In practice the impact on surplus capital of the move to Solvency II will vary between countries and between individual insurers within countries. Insurers may want to consider both the absolute impact on their solvency position of the move to Solvency II and also, taking account of the data published in the QIS5 report, how they anticipate their position will move relative to their peers. QIS6 is not on the cards EIOPA indicates that QIS5 will be the last fully comprehensive QIS exercise before Solvency II implementation. Some areas require further work but this work is to be done through specific tests and targeted work rather than another QIS exercise. Transitional arrangements EIOPA concludes that transitional arrangements will be needed in the following areas: Equivalence of third countries Hybrid capital and subordinated debt Discount rates on technical provisions However, EIOPA is concerned that too many transitional arrangements or over long transitional periods could jeopardise the timely implementation of Solvency II and have adverse effects on competition. For example, EIOPA has not suggested a need for transitional measures in respect of the calibration of the SCR. The draft Omnibus II Directive provides a framework for transitional measures in a wide number of areas 1. However, it appears EIOPA may support a narrower 1 See the PwC publication on Omnibus II available at http://www.pwc.co.uk/pdf/premium/pwc-summaryof-omnibus-ii.pdf scope and duration of transitionals. Whilst it will be for the Commission to determine the nature and extent of the proposed transitional measures, insurers should certainly not assume that transitionals will be brought forward to the maximum extent possible under Omnibus II. What is likely to change as a result of QIS5? Concerns remain around the calibration of certain elements of the standard formula SCR and in a number of areas participants found the calculations overly complex. In addition, concerns were raised over some aspects of the calculation of technical provisions and regarding the valuation of certain assets and other liabilities. A summary of some of the specific concerns highlighted is set out in Appendix 1. In particular, EIOPA has identified a number of areas where further work is needed and where it therefore appears more likely that changes could be made. Valuation of technical provisions: More work is required regarding the valuation of technical provisions, with areas of focus including: The illiquidity premium where further guidance is needed and consideration is being given to applying a premium to certain contracts only. Contract boundaries where some participants support using the definition being developed by the IASB. Expected profits in future premiums (EPIFP) where EIOPA is looking to ensure that they count as the highest quality capital to a prudent degree (for QIS5 all EPIFP were counted as Tier 1 capital). The extent to which technical provisions reflect expected future profits will be impacted by the conclusions reached on other areas such as the contract boundary. The risk margin where the full approach was found to be too complex and where further work is needed to enable simplifications to be consistently applied. Valuation of assets and other liabilities: The valuation of deferred taxes was the most difficult area and more work is required to achieve a consistent and appropriate approach. Other areas where difficulties were encountered included intangibles, participations, contingent liabilities, financial liabilities and employee benefits.
Calibration of the standard formula SCR: EIOPA s view was that the calibrations were generally accepted as appropriate by both supervisors and the industry. However, the exception to this is the non-life and health catastrophe risk sub-modules where EIOPA is already working with the industry to refine calibrations. Simplifications in calculation of standard formula SCR: QIS5 has highlighted a number of areas where the system could benefit from a less complicated design of the standard formula or from the use of further simplifications. Areas being considered include: - Calculation of the counterparty default risk sub-module - Calculation of the loss absorbing capacity of technical provisions and deferred taxes - Design of the non-life and health catastrophe risk sub-modules - The application of the look-through approach within the market risk module A number of participants queried the merits of the non-life lapse risk sub-module. However, in EIOPA s view this sub-module should be retained. Volatility A key concern of insurers is the level of volatility present in the calculation of both technical provisions and the SCR. EIOPA believes that the dampeners present in the calculation together with the ability of supervisors to flex their response to circumstances through the ladder of intervention goes some way to addressing this concern. However, EIOPA is planning to examine the issue of volatility, and its implications, further. Preparedness of supervisors At supervisory level, Solvency II is also a learning process. EIOPA and national supervisors are building together the very much needed expertise. However, a lot of work remains to be done in this area to ensure consistency of supervisory approach across Europe. Practical concerns Based on QIS5 submissions certain supervisors are concerned that some insurers are underestimating the volume of work required to prepare for Solvency II requirements. Overall larger insurers seem better prepared for implementation than smaller ones. Many insurers highlighted that the complexity of Solvency II requirements makes the transition challenging. Indeed the lack of expert resources, particularly actuarial resources was raised by many as a concern for ensuring a smooth implementation. Managing the transition to Solvency II alongside significant change to accounting standards (e.g. the IASB s revised insurance contracts and financial instruments standards) adds to the challenges faced by the industry. Some supervisors feel insurers have made significant progress on Pillar I partly thanks to QIS exercises. However, supervisors have raised concerns about the level of preparedness of firms to apply the Pillar II and Pillar III requirements. Not surprisingly the ORSA, in particular, was cited as an area of concern along with the development of risk management functions sufficient to comply with Solvency II. The QIS5 report also sets out a number of areas where participants seem to have completed the exercise in a different way from that intended by the Commission. The clarification of the requirements in these areas could have a significant impact on the results such as the treatment of hybrid or subordinated debt within own funds. We have listed some of the areas where participants seem to have made differing or incorrect interpretations of the technical specifications in Appendix 2.
What do I need to do? We expect insurers to consider the implications of the QIS5 results on their own capital structures. We expect insurers to review their investment and debt-financing strategies as certain asset classes and debt instruments will carry higher capital charges or be classified differently under the directive. Additionally, higher capital charges will mean certain products become unprofitable, leading insurers to stop writing certain classes of business or re-designing some products. The results may also impact some firms decision to seek approval for the use of an internal model. Contacts To gain a deeper understanding of how the QIS5 results impact your Solvency II programme please contact: Name Email Telephone Paul Clarke paul.e.clarke@uk.pwc.com 020 7804 4469 Philippe Guijarro philippe.guijarro@uk.pwc.com 013 1524 2260 Mike Vickery mike.p.vickery@uk.pwc.com 011 7923 4222 Kareline Daguer kareline.daguer@uk.pwc.com 020 7804 5390 Those groups not already doing so may look to streamline their group structures, Solvency II presents a real opportunity for companies to reduce capital and compliance costs, optimise available capital and align their group structures and we expect this to lead to an increase in M&A activity. Insurers might also wish to reflect on the QIS5 experience as they continue to develop their implementation plans for Solvency II. They should consider areas in which they encountered difficulties in preparing the required information and consider how process and data might be enhanced to facilitate production of the solvency data in a business as usual environment. In considering these points insurers should be aware of the areas where further changes in the requirements are likely. This publication has been prepared for general guidance on matters of interest only, and does not constitute professional advice. You should not act upon the information contained in this publication without obtaining specific professional advice. No representation or warranty (express or implied) is given as to the accuracy or completeness of the information contained in this publication, and, to the extent permitted by law, PricewaterhouseCoopers LLP, its members, employees and agents accept no liability, and disclaim all responsibility, for the consequences of you or anyone else acting, or refraining to act, in reliance on the information contained in this publication or for any decision based on it. 2011 PricewaterhouseCoopers LLP. All rights reserved. 'PricewaterhouseCoopers' refers to PricewaterhouseCoopers LLP (a limited liability partnership in the United Kingdom) or, as the context requires, other member firms of PricewaterhouseCoopers International Limited, each of which is a separate and independent legal entity.
Appendix 1: Problem areas identified by QIS5 respondents Area Significance Issues highlighted by respondents General Materiality Several respondents asked for more guidance around materiality and materiality thresholds. Regulatory Balance sheet Deferred taxes Technical provisions Contract boundary definition Liquidity premium Risk margin Segmentation Life insurance - Future discretionary benefits Non-life insurance - Premium provisions Reinsurance recoveries Standard Formula SCR Single equivalent scenario Loss absorbency of deferred taxes Participants found the valuation of deferred taxes under Solvency II to be difficult and they dealt with valuation issues in a number of different ways. This is an area where clarification and more guidance is likely to be issued to resolve the issues. Most participants felt that there should be clarification around the contract boundary definition. There is some strong support for aligning the definition to IFRS. Many participants felt that further guidance is needed as they found practical difficulties in calibrating economic scenario generators to varying discount rates. Applying the illiquidity premium to spot rates led to technical difficulties such as negative forward rates which need to be addressed before Solvency II implementation. More guidance is also needed as to how to assign contracts to buckets for the purpose of determining the extent of illiquidity premium to be applied. Some are suggesting a reform to the approach tested in QIS5 by, for example, applying a full illiquidity premium to some products and no illiquidity premium to all others. Participants found the full calculation as described in the technical specifications too burdensome and complex. The suggested simplifications were widely used. Supervisors raised issues about choice of different simplifications and felt more guidance was needed around the choice of method thatt insurers use. They also highlighted that in some cases the simplifications weree not sufficiently robust (for example where best estimates are negative). Participants requested more clarification on the methods to use for the inclusion of unavoidable market risk in the risk margin calculation. Many participants reported that the guidance on segmentation was not sufficiently clear in a number of areas. In addition, a widely held view was that the value of the second level segmentation of life contracts appears limited compared to its impact to the calculation. Many participants felt more guidance is needed on future discretionary benefits. Differing interpretations in this area can affect the results significantly. Several participants requested more guidance on how to calculate the best estimate for premium provisions. Participants found difficulties with calculating the probability of expected default of counterparties. In some cases it was difficult to understand the technical specifications or participants did not agree with the specifications. Some supervisors raised concerns about lack of rating for certain third-country reinsurers. The single equivalent scenario approach was rejected by the majority of participants and supervisors. It was found to be too complex and the costs of the calculation did not justify the benefits. This is an area where a change of approach might be considered. The majority of participants reported difficulties with the methodology for calculating the loss absorbency of deferred taxes. Overall, additional guidance on this area was widely requested. Since this adjustment can be material in the context of total SCR, the difficulties of interpretation and inconsistent treatments might have a material impact on the overall capital position of the industry.
Area Market risk - Look-through approach Market risk - volatility Market risk - geographic diversification Market risk - Interest rate risk sub-module Market risk - Spread risk sub-module Counterparty default risk Life and non-life lapse risks Health underwriting risk Non-life future premiums Non-life CAT risk Undertaking specific parameters (USP) Risk mitigation techniques Significance Issues highlighted by respondents Many participants raised concerns about the complexity and cost of applying a look-through approach to structured credit products, collective investment schemes and investment funds. Participants raised issues around availability of information to apply the look-through approach to these products and requested guidance on the application of the proportionality principle to this approach. Some participants felt that the omission of volatility stresses from the equity and interest rate risk sub-modules was a significant omission. Some participants felt there was lack of recognition for geographical diversification within an asset class. A few participants claimed the design of this sub-module was too penal and others found it was too complex. Participants raised various concerns around this sub-module touching on its calibration, consistency and complexity. It was the most commented upon sub- module within market risk. Counterparty default risk attracted significant comment on the difficulty of applying the full calculation and whether the methods are proportional to its relative minor impact for many participants. Even the simplifications where found to be too complex by some participants. Lapse risk caused difficulties for both life and non-life insurers. Life insurers strongly objected to the requirement to model the risk at policy level. Non-life insurers noted that in many cases they did not have systems and processes in place to model the risk and it was perceived by many as being immaterial and hence the effort involved in calculating the stress was judged to be superfluous. Key areas of concern for this module were segmentation (as mentioned above), the split between disability / morbidity sub-modules, lapse and CAT risk. While the introduction of future premiums and contract boundaries made sense from a theoretical point of view, the difficulties encountered in calculating them outweighed the benefits. Non-life catastrophe topped the list of complaints with comments on methods, calibration, data availability and the effort required to calculate the risk charge. EIOPA will carry out further work to address the issuess raised. Very few participants responded to the USP part of the QIS5 exercise, making it very difficult to draw meaningful results. Participants raised concerns over lack of appropriate data to develop USPs and some respondents suggested additional USPs. Participants found risk mitigation techniques other than proportional reinsurance difficult to apply within the standard formula. Further guidance is likely to be issued to address some of the concerns raised. Minimum Capital Requirement (MCR) MCR calculation Participants raised concerns over the calculation of the Absolute MCR (AMCR). For a number of insurers AMCR is above SCR. As a result, it is not clear how the concept of ladder of intervention will work in that context. In addition, participants that calculated the Composite MCR found that this approach was very penal to insurers with very small non-life portfolios. High Medium Low
Appendix 2: Errors or inconsistencies in interpretation of QIS5 technical specifications Area Regulatory balance sheet Observation Participations EIOPA expressed surprise at the large number of participations valued on a mark to model basis. This method is not permitted for subsidiaries and should only be used as a last resort for other participations. Intangibles Some participants valued intangibles on a cost basis which is not in line with QIS5 specifications. If fair value measurement was not possible QIS5 required intangibles to be valued at nil. Deferred taxes It seems participants did not follow IAS12 correctly in many cases. There was substantial variation in the way deferred taxes were recognised and valued and some participants did not calculate deferred taxes at all. Areas where accounting valuation differs from Solvency II valuation In general QIS5 required assets to be valued on an economic basis. However, where a cost based approach is used in the financial statements (e.g. property) this caused difficulties. In some cases such assets were valued on a historical cost basis on the grounds of immateriality. Own funds Treatment of hybrid and subordinated debt in own funds Participants reported hybrid and subordinated debt in Tiers 1, 2 and 3 within the without transitional measures scenario. However, supervisors in countries where these instruments were reported are of the view that, without transitional measures, these items do not qualify as basic own funds. The technical specifications appear to have been interpreted very widely and optimistically by participants during the QIS5 exercise. As a consequence, QIS5 results on this area are not very reliable. However, EIOPA highlights that transitional measures over subordinated debt are a crucial element to a smooth transition as they represent a material component of own funds. The incorrect treatment of hybrid and subordinated debt under QIS5 is also likely to have an impact on the results of tiering and limits within own funds. Ring-fenced funds It was noted that in some cases participants concluded there was no ring-fencing required whilst the local supervisor and other parts of the industry believed ring-fencing was required. It is clear that both participants and supervisors need more guidance about the characteristics of a ring-fenced fund which in turn should lead to more consistency of application. There are also concerns over the way calculations of capital have been carried out where there are ring-fenced funds. Participants appear to have found the recording of the impact of ring-fenced funds in the spreadsheets very difficult. Restricted reserves Participants appear to have been confused about identifying restricted reserves. Some reported items in error. It was also the case that items reported as restricted reserves were in fact ring-fenced funds in nature, including with-profits funds and reserves for guaranteed funds. It appears that more guidance is needed over the definition of restricted reserves to ensure consistency of approach. Expected profits in future premiums (EPIFP) A significant number of participants did not complete the calculations and it attracted severe criticisms by many insurers. EIOPA acknowledged that the data from QIS5 cannot be extrapolated safely. More work will be needed on this area and the debate is ongoing as to whether the full EPIFP will be classified as Tier 1 under the final rules. Technical provisions Contract boundaries The different interpretations of the contract boundaries may have led to the incorrect calculation of technical provisions. Illiquidity premium Many supervisors reported inconsistent application of the illiquidity premium to buckets Segmentation There may have been different interpretations of the required segmentations due to unclear guidance SCR Internal Model Solvency II valuation rules when applying internal model Some participants seemed to believe they could apply IFRS when valuing assets and liabilities if using an internal model it is clarified that internal models should apply the Solvency II approach to valuation Replacing standard formula parameters and internal modelling Participants mentioned using specific parameters to account for their specific risk profile. It is clarified that the use of undertaking specific parameters is restricted to a specific set of parameters. Replacing the standard formula parameters with specific ones is not considered internal modelling and does not comply with Solvency II requirements regarding internal model.