Welcome to the latest edition of KPMG s Actuate Newsletter

Transcription

1 ACTUATE NEWSLETTER APRIL 2013 Welcome to the latest edition of KPMG s Actuate Newsletter Actuarial Contents The risks and rewards of DTAs on economic balance sheets Gordon Gray & Debbie MacDonald Page 2 Solvency II - A European View Michael Rallings & John Jenkins Page 6 Since our last newsletter distribution a number of regulatory changes have dominated the industry - such as the delay in the implementation of Solvency II and the split of the Financial Services Authority (FSA) into the Prudential Regulatory Authority (PRA) and the Financial Conduct Authority (FCA). Delays in the implementation of Solvency II have resulted in many companies marking time in their development plans in this area. However, we note that are still important steps being taken by EIOPA and national supervisors to ensure that a pan-european risk based solvency regime comes into being. In advance of full Solvency II implementation, EIOPA is planning to issue interim measures which it expects all national supervisors to comply with, by ensuring firms meet the specified outcomes. The interim measures are expected to cover aspects of Pillar 2 and Pillar 3. Further, as expected, the PRA has decided to replace the ICA regime with a Solvency II overlay, i.e. an ICAS plus regime. A number of firms are also now engaged with the PRA and the FCA on thematic reviews, and a number of Section 166 skilled persons reports are being commissioned by the PRA. It is clear that there is significant work to be done within the industry to get to grips with these latest developments and thematic reviews. In this edition of our newsletter, the first article delves into the world of ICA plus and tax. It looks at how firms can potentially reduce their economic (ICAS plus) capital requirements by revisiting the tax approach. It also highlights the importance for actuaries of understanding, challenging and managing any deferred tax asset. Our second article looks at Solvency II across Belgium, Germany, Netherlands and the United Kingdom and provides a quick snap shot of how firms have progressed with the implementation of internal model or standard formula approaches. It also highlights some challenges and views across technical provisions, risk margins and wider issues related to SII. We trust you will find the articles of interest. Should you wish to discuss any of the articles further, please get in touch with your usual KPMG actuarial contact or use the contact details on the back of this newsletter. Additionally please pass this publication freely amongst your colleagues. If you did not receive this newsletter directly, and would like to do so in future, please contact Isobel Gaston (isobel. gaston@kpmg.co.uk) to have your name added to our distribution list. John Jenkins 1 Actuate newsletter / April 2013

2 The risks and rewards of DTAs on economic balance sheets Gordon Gray & Debbie MacDonald Gordon Gray Debbie MacDonald If recognised to best advantage, tax can be an important counter cyclical adjustment to economic balance sheets, and can potentially reduce economic (or Solvency II) capital requirements by 23%. However, the valuation of deferred tax assets (DTAs) is a matter of judgement. For the UK life sector, these judgements may become increasingly material over the coming years as deferred tax liabilities reduce as a consequence of the reform of the life tax rules. Some companies are finding that their economic and SII capital requirements are more onerous than they would ideally like. For such companies, revisiting tax may be a valid and fruitful line of enquiry. Market consistent? Internally consistent would be a start The economic base balance sheet and the post stress balance sheet (whether drawn up under Solvency II or otherwise) will contain current and deferred tax balances. A market consistent model for deferred tax asset valuation might resemble option pricing models. Such an approach would be novel and there could be some quite complex challenges in developing such a model (and we don t explore this idea further here). SII requires that deferred tax is computed using the IFRS rules for deferred tax, as set out in IAS 12. IAS 12 is not a particularly modern accounting standard - it does not permit discounting for example, and bears little resemblance to market consistent principles. While deviation from IAS 12 may be possible (or even desirable) for economic balance sheets not drawn up under SII, IAS 12 remains a useful guideline. It at least has the merit of being an internationally recognised framework for deferred tax valuation. Unlike Deferred Tax Liabilities (DTLs), DTAs cannot be recognised automatically and must be tested for recoverability. Under IAS12 a view is needed on what the probable future taxable profits are. Probable is interpreted as more likely than not. Almost by definition, this requires anticipation of value which is not on the SII balance sheet. This is an interesting and fundamental point as it implies that the valuation methodology used to draw up the SII balance sheet needs to be departed from. The same may apply to an economic balance sheet. For example, if a liability comprises a BEL plus a risk margin then the BEL could arguably be the measure of the more likely than not liability, implying that the risk margin release will be a source of future profits (see example 1a). If one was feeling mischievous, it could thus be argued that all (or most) deferred tax assets are actually contradictions. The concept of probable future taxable profits effectively invokes a super balance sheet which is wider in scope than the valuation balance sheet. A purist may consider this internally inconsistent, but IAS 12 and commercial pressure may demand a more pragmatic view on this point of principle. There is no clear framework on how to determine future taxable profits and so the matter is largely one of judgement. 2 Actuate newsletter / April 2013

3 Example 1a and 1b: Supporting loss absorbency of deferred tax A pensions only company has a pre-tax SCR of 1,000. The risk margin is 300 and contract boundaries have inflated the SCR by 100. Future new business will contribute profits of 10 p.a. Assume, for the sake of illustration that the tax rate is 25% now but will fall to 22%. There is a DTL of 125 which will run off linearly over the next 10 years (due to the phased transition to taxation on an IFRS basis). Pre-tax SCR / potential LADT (i) DTL (ii) Release risk margin (iii) Relax contract boundary (iv) Future new business Unrecognised tax credit SCR % reduction Now Gross 25% % In 5 years time Gross 22% % Comment The release of risk margin (ii) can be anticipated on a more likely than not basis and the same basis can be applied to contract boundaries (iii). A 5 year time horizon is taken for future new business (iv). By combining the effect of the items listed, the methodology gives a good result at present and with minimal reliance on additional modelling (a 23.75% reduction to the SCR). The position could be optimised by assuming more years of new business (for example). In 5 years time, the same methodology will give a poorer result (a 15.4% reduction) due to the falling tax rate and the run off of DTLs (i). This could perhaps be improved by investigating and modelling further sources of future profits. April 2013 / Actuate newsletter 3

4 DTA prudent or optimistic? The use of DTAs to increase economic available capital or reduce the required capital needs to be subject to validation. There are also implications for risk management, governance and stakeholder management. It may be legitimate to use the most optimistic assumptions possible to support DTAs so long as there is suitable understanding of the associated risks. Suitable qualitative and quantitative management information will need to be developed to support this. Regulators will also we presume take a keen interest in this aspect. Put starkly, there is a spectrum of choice between a low risk but small DTA and a higher risk big one. Key questions for Boards include:- What are the implications for reported numbers and business plans should a PRA challenge of DTAs be successful? If you are very prudent are you satisfied you are not at a competitive disadvantage? If there is enforced prudence elsewhere in the base balance sheet for example contract boundaries - can that be used to support a more favourable tax position, either directly by assuming that the prudence unwinds and so can be used to support a DTA, or indirectly by allowing a view to be taken in the round. Don t leave it all to the tax function The judgements and major points of principle are not purely tax technical matters - there is little guidance on how to determine future profit or future taxable profits. There is a risk that such matters fall between the stools and that the tax function and the actuarial/ modelling function each assume the other has primary responsibility. This is a classic multi-disciplinary situation. It is particularly important that this is considered when scoping validation. It is not sufficient for an independent tax technician to review the tax logic. The PRA challenge Well, I never heard it before, and it sounds like inconsistent nonsense If you wish to increase own funds or decrease the SCR through use of DTAs in a robust manner, then you need to be sure you have a clear message and be confident in articulating it to the PRA. The PRA will be instinctively cautious when it comes to DTA recognition as DTAs are not part of Peak I in Solvency I at present, and so represent something the regulator is unfamiliar with. We understand that pre-consultation drafts of Solvency II Level 3 materials say relatively little in this area. There is a suggestion that DTA recognition should be limited to what is recoverable within the normal business planning cycle (which is suggested to be 3-5 years). This requirement is not part of IAS 12 and some companies have we understand made representation to that effect. Nonetheless, it would be wise to quantify the risk of DTAs being restricted in this way. That point aside, there is little objective guidance. The PRA perhaps will have three ways of challenging a deferred tax position: (i) technical objection to the principles used, (ii) time horizon for recoverability and (iii) quality of the supporting evidence. It will be necessary to develop and document the approach and evidence in each of these areas in order to be prepared to respond to challenges on them. I seem to have an OK result just now, can I do nothing? Even if an under-developed and underdocumented deferred tax methodology gives you a reasonable result for now, the disadvantages will become more acute as transitional DTLs run off over the next 10 years. This is illustrated in example 1b. Is I-E tax relevant: Curiouser and curiouser! The post-2012 tax regime is good news for modelling actuaries as the change to protection business, introduction of a commercial basis of apportionment and the ending of interactions between gross roll up business profits and the I-E all significantly simplify modelling. Nonetheless, I-E tax and BLAGAB continue to present some challenges. It is important to consider whether the I-E tax logic in BEL models will be satisfactory under stress. It may be necessary to introduce restrictions so that credit is not inadvertently taken for net investment losses or excess E (see example 2). In addition to considering the modelled I-E tax cashflows, the effect that tax adjustments have on unit liabilities and asset shares should be considered. Example 2: Tax in BEL models A life office is projected to have income of 100 and expenses of 90 each year and so is marginally XSI. BEL models assume a risk free yield and calculate I-E tax as 20% of income + gains less expenses (tax = 20% = 2). Under a particular stress, the company generates substantial XSE of 40 p.a. (due to expense inflation, lapses of unit linked bonds and/or a resilient commission tail from pre-2013 protection business). The model uses net I and net E assumptions and does not distinguish between positive and negative results and so a tax credit of 20% = 8 is assumed each year. Comment The stressed BEL model run will include implicit deferred tax assets ( 8 p.a.) and therefore the stressed BEL numbers will be understated. This affects the calibration of the internal model and understates the SCR. It may be difficult to extract data from the BEL models and to quantify the effect. Were the XSE to arise immediately (as opposed to being projected to arise in future) then any DTA would be within the scope of IAS12 and the use of more likely than not real world returns as opposed to risk free returns to support the DTA could be explored. There is however no clear authority for the extrapolation of IAS12 principles into the calculation of BEL. 4 Actuate newsletter / April 2013

5 A further point is to consider the immediate tax impacts of asset shocks. If these are input into BEL models as an opening adjustment these models may not calculate tax on the losses. Also there will be assets impacted by the shock which will be outside the scope of the BEL models. Simply assuming that I-E tax credits are not relevant to the SCR is an oversimplification (see example 3). There is considerable latitude when it comes to forecasting future taxable profits to support DTAs. Example 3: Asset shocks in a WPF A poorly capitalised BLAGAB WPF with a low equity backing ratio has 1bn of assets and 1bn of asset share liabilities. A 40% asset shock is applied. For simplicity, asset shares are reduced by 400m to 600m at time 0 in the BEL models (and no I-E tax credit is considered). Asset shares grow at a net of tax yield in the BEL model. Post stress, guarantees become material and so the liability is 800m (as opposed to the 600m asset share). The capital requirement is therefore 200m. Comment There is a potential I-E tax credit of 20% = 80m. If all of this could be supported by a projection of future income or gains, the capital required could be reduced from 200m to 120m. Even partial credit could be of considerable value. As this investment loss is immediate, any value would be in the form of a deferred tax asset (and so within the scope of IAS 12). Using real world returns to support the asset could enhance the value of the DTA considerably. Conclusion It would be so nice if something made sense for a change. Tax assets are potentially an important part of an economic or SII balance sheet and can represent a material reduction to the economic required capital. The modelling of other key risks has been the subject of considerable theoretical research and practical development, and recognized practices have been emerging across the industry over recent years. However, the same attention has generally not been given to tax generally and to deferred tax in particular. Core to any methodology for recognizing deferred tax assets are the acceptance and recognition that net deferred tax assets generally require value which is not recognized on the economic balance sheet for their support. There is considerable latitude when it comes to forecasting future taxable profits to support DTAs. This opens up number of methodology choices and judgments. There is not necessarily a single answer to the questions arising in connection with deferred tax. Tax assumptions and results need to be understood, probed, challenged and managed. Companies which are overly prudent in their tax treatment for their economic or SII balance sheet may well be putting themselves at a commercial disadvantage. Companies which are overly aggressive will expose themselves to regulatory challenge and potential future balance sheet volatility if assumptions are not borne out. Where a company should be within this spectrum needs to be as a result of a deliberate and well-governed choice after weighing up all the pros and cons of the alternative approaches. April 2013 / Actuate newsletter 5

6 Solvency II - A European View Michael Rallings & John Jenkins Michael Rallings John Jenkins The adaptation and implementation of Solvency II (SII) has dominated the insurance industry over the last few years. And even though delays and uncertainties in recent months have left a question mark hanging over SII, and have resulted in a reduced overall focus on it by many companies, EIOPA still believes SII to be a fundamentally important regime in setting a harmonised European wide capital and risk management framework. Further, to assume that Solvency I will continue as it stands is probably not appropriate, as there exists an improved framework directive approved by the EU. Therefore, even if Solvency II continues to be delayed, one would expect a range of national solutions topping up the existing legislation to emerge. For example in the UK, the FSA has already taken a first step in this direction. Julian Adams, Director, Insurance and Deputy Head, Prudential Business Unit, FSA, said in a speech that FSA would consider merging some of the SII Pillar II and III work into the existing regulatory framework in what is being called an ICAS Plus. This suggests that some local regulators may still go ahead with an effective early introduction of Solvency II or a similar regime (ICAS plus) on a national basis. This short article presents a quick snap shot across Belgium, Germany, Netherlands and United Kingdom of how firms have progressed with the implementation of internal model or standard formula approach. It also highlights some challenges and views across technical provisions, risk margin and wider issues related to SII. The following table details the findings. These finding were also discussed the Actuarial Profession s life convention in Belgium on 5 November We are grateful to our KPMG colleagues in the countries referred to above for their input in compiling this analysis. 6 Actuate newsletter / April 2013

7 Area Question Response Standard formula vs Internal model What proportion of companies are using a standard formula approach as opposed to an internal model? Is there any pattern to this and has there been any switching from one to the other? Most large firms across all four regions opted for an internal or partial internal model, with medium to smaller sized firms using a standard formula approach. However, more recently some of the internal model firms have switched back to a standard formula approach as internal model provided limited benefit. Do companies using the standard formula think that it is conceptually correct and fit for purpose? Risks covered Methodologies Risk aggregation Is it practicable to evaluate? Are they getting a sensible answer? In Belgium and UK the consensus was that the standard formula approach was generally fit for purpose. However, the coverage of risk could be extended to include asset volatility, inflation and pension scheme risk. In Germany and Netherlands firms have not yet made an assessment as to whether standard formula is appropriate. This suggests that firms might have adopted standard formula as a default approach. Are companies being challenged by the regulator as to whether the standard formula is appropriate for them or not? UK firms are being asked to consider standard formula as a conscious choice and justify that it is appropriate to their risk profile. In Belgium the focus in mainly on larger firms to develop an appropriate internal model. Whereas, the Dutch regulator considers standard formula to sufficiently calculate SCR and has presented limited challenge to standard formula companies. In Germany the regulator has challenged the use of a standard tool created by the German Insurance Association (GDV) to calculate the SCR but has not challenged the companies on whether the standard formula is appropriate or not. April 2013 / Actuate newsletter 7

8 Area Question Response Standard formula vs Internal model (continued) What are the key challenges companies face in applying the standard formula? The key challenges faced by standard formula companies in Belgium and Netherlands include data and IT (run times). However, for German companies the key issue was to understand the tool created by GDV (which is the German equivalent of the UK ABI). UK entities generally require further guidance in respect to certain areas such as the staff pension scheme. How are companies finding interaction with their regulator in relation to internal models and the approval process? Overall across all four regions firms believe that the communication from the regulators could be clearer. However, having said that in Germany the interactions between firms and regulator are perceived constructive, even though the regulator is not in the process of approving internal models. Similarly in UK the FSA is thought to be open with the firms and feedback is becoming more detailed and specific. The Dutch regulator is also believed to have pragmatic approach. However, in Belgium the regulator, National Bank of Belgium (NBB) is perceived to be too strongly focused on technical details while not paying enough attention to the qualitative side. What level of detail are companies required to submit to the regulator to obtain approval? The documentation requirements all regions are observed to be quite demanding and challenging. What progress has been made at present towards internal model approval? The firms in UK are progressing relatively well. However, this is expected to be because a good base (ICAS regime) is already in place. What are the main challenges being faced? The key challenges faced across the different regions include, data, documentation, IT (run times), validation, justifying expert judgement, satisfying use test and transition to BAU. 8 Actuate newsletter / April 2013

9 Area Question Response Standard formula vs Internal model (continued) Has the internal model approval process been significantly more or less onerous than expected? The internal model approval process has been significantly more onerous than expected. Have the questions posed so far by the regulator been of a more quantitative nature or a more qualitative nature? Across UK, Germany and Netherlands the questions have mainly been qualitative. However, in Belgium the regulator has concentrated more on quantitative aspects. On a scale of 1 10, how far along are companies in the approval/ completion process? Even though some view to be much further along in the internal model approval process, overall its believed that firms have progressed approximately 50% - 60% with the approval process. Have companies done dry runs? Generally, to what extent and how detailed have these been? Most of the companies in Germany, which intend to use internal models, have integrated the models in their risk management framework already and use them regularly. Similar in Netherlands the regulator requires a dry run every year toward implementation of Solvency II. Therefore, firms are getting more and more familiar with the required calculations. In UK generally full dry runs have not been performed and the processes are not at a stage of meeting external financial reporting standards. However, firms are updating their numbers at year-end. In Belgium on standard formula firms have performed dry runs. Technical Provisions What are the most controversial areas of the technical provision calculation? The most controversial areas recognised by the underlying regions included the valuation of long term guarantee business (including risk-free yield curve and match adjustment) as well as contact boundaries. April 2013 / Actuate newsletter 9

10 Area Question Response Technical Provisions How do technical provisions under SII compare with those under Solvency I? The technical provisions under SII compared to those under Solvency I are expected to vary between products. In Germany, where underlying interest rates are guaranteed entities might experience an increase in technical provisions if the risk free yield is lower than the guaranteed interest rates. In UK depending on the matching adjustment and contract boundaries, the technical provisions for annuity and unit linked business may be more onerous. Whereas, the technical provisions for term assurance business would less onerous. Risk Margin How challenging are companies finding the risk margin calculation in practice? Project of SCR has been an issue for the UK firms which have developed a range of methods to address this issue. Other regions have not faced any key issues with respect to risk margin calculations. Overall Balance Sheet What are the main areas of difficulty in the overall balance sheet calculation? In UK some firms had to revise their models to produce results on realistic cash flow basis. However, in continental Europe firms have not come across any particular issues. Other Are companies still enthusiastic about SII? What do they think about the delays to implementation? In some regions the delays have been well received as it allows them to develop and implement the underlying SII requirements. However, in other regions such as UK and Netherlands the firms are less enthusiastic about SII due to the uncertainty and would prefer the SII implement as soon as possible. Have companies transitioned SII to a business as usual environment yet or are they still in a project environment? Besides UK, SII in the other regions is still in a project environment. In UK firms had been planning to move SII into BAU. 10 Actuate newsletter / April 2013

11 April 2013 / Actuate newsletter 11

12 Contact us If you would like more information on any of the articles discussed in this newsletter, or if you would like more information or assistance with regard to our actuarial practices, please contact: Isobel Gaston Executive Assistant t: +44 (0) e: KPMG in the UK Life Tim Roff t: +44 (0) e: John Jenkins t: +44 (0) e: Nick Dexter t: +44 (0) e: Ferdia Byrne t: +44 (0) e: Trevor Jones t: +44 (0) e: Richard Care t: +44 (0) e: Gavin Palmer t: +44 (0) e: Simon Perry t: +44 (0) e: Gerard Callaghan Director t: +44 (0) e: Michael Rallings Director t: +44 (0) e: Gina Craske Director t: +44 (0) e: James Cruttenden Director t: +44 (0) e: The information contained herein is of a general nature and is not intended to address the circumstances of any particular individual or entity. Although we endeavour to provide accurate and timely information, there can be no guarantee that such information is accurate as of the date it is received or that it will continue to be accurate in the future. No one should act on such information without appropriate professional advice after a thorough examination of the particular situation KPMG LLP, a UK limited liability partnership, is a subsidiary of KPMG Europe LLP and a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative, a Swiss entity. All rights reserved. Printed in the United Kingdom. The KPMG name, logo and cutting through complexity are registered trademarks or trademarks of KPMG International. RR Donnelley RRD April 2013 Printed on recycled material.