IFRS 17 technology solutions the CSM calculation
IFRS 17 and the Contractual Service Margin This briefing document focuses on how to incorporate the Contractual Service Margin (CSM) calculations required under IFRS 17 into existing models and processes. In May 2017 the IASB published IFRS 17 Insurance Contracts, the first global accounting standard for insurance contracts. Global adoption will likely be required by many listed insurers and some mutuals. With an effective date of 1 January 2021, insurers have two and a half years left to implement IFRS 17. Central to IFRS 17 is the General Measurement Model (GMM), which insurance firms will apply to determine the liabilities for their long-term insurance contracts and short-term contracts in the post claims incurred period. This model has also been referred to as the building-block approach as reference to the various components of value as shown in Figure 1. Figure 1. The CSM set to negate day one accounting profit Total Insurance Contract Liability Expected Cash Flows Discounting Contractual Service Margin (CSM) Risk adjustment 1 willistowerswatson.com
The CSM challenge While IFRS 17 is an accounting standard, much of the GMM is closer to a typical actuarial model, such as Solvency II, rather than a traditional accounting model. The building blocks of both IFRS 17 and Solvency II are based on a set of risk-adjusted discounted best estimate cash flows albeit the allowance for risk and discounting between these two measures may differ. However, there is a more fundamental difference: IFRS 17 assures nil profit as of day one on the sale of a contract, by creating a CSM. This is then run off over the lifetime of the contract, effectively smoothing profit release as the service is provided. There is no equivalent of the CSM under Solvency II. While traditionally actuarial and accounting functions have been relatively distinct in people and systems, the move to IFRS 17 challenges this separation in particular, asking the question where should the CSM calculation take place? This is because the CSM calculation uses both forward looking information that typically comes from an actuarial model and backward looking actual experience that typically comes from an accounting system. The decision of where the CSM calculation should go will depend upon a firm s specific circumstances, models and processes. However, we believe it is important that the solution meets the following criteria: Flexibility: As well as the requirement to publish IFRS 17-compliant results, Firms will want to regularly stress/scenario test these results for various purposes including business planning projections and helping make key decisions. Control and auditability: the CSM calculation adds significant data and process requirements so having good controls will be critical to ensure that results are reliable and meet audit requirements. Speed and efficiency: for many insurers, people and systems resources are already stretched so the additional requirements of IFRS 17 within the financial reporting time frames are likely to drive the need for efficiency gains across the reporting function. Developing a solution As IFRS 17 is a principles-based standard there is scope to interpret how each of the principles is implemented. As such, off-the-shelf solutions will not provide an optimised solution and some customisation will always be required. We have therefore developed an example CSM solution (see later) that is likely to provide much of the functionality and processes needed. The existing actuarial systems currently used for Solvency II purposes will have all the attributes needed to project and discount future cash flows using cohort-specific discount rates. A common and obvious starting point therefore will be the use and incorporation of these existing actuarial systems and processes into the IFRS 17 solution. This has the added benefit of automatically providing consistency between the underlying cash flows for both metrics and reducing time and effort attempting to reconcile differences. Assuming the use of existing actuarial systems, key requirements to produce the IFRS 17 balance sheet and income statement include: 1. A net present value calculation using cohort-specific discount rates, to determine the CSMs at the point of sale 2. A process to allocate new policies to a profitability group 3. A roll forward and analysis of movement of the CSM from one period to the next on historic assumption sets (including from point of sale to the next valuation date). Items 1 and 2 do not require any accounting inputs, but require full policy by policy cash flows and discounting using prescribed discount rate curves, which of course existing actuarial systems are very well placed to perform efficiently. Carrying out these calculations within the actuarial systems would therefore remove any need to write out large volumes of data to a different system, reducing risk, cost and time. Item 3 requires actuals from the accounting ledger in much the same way these are currently required to produce the analysis of change for Solvency II, albeit in some cases the data may be required at a more granular level. Once the CSM profit test calculations have been created for Items 1 and 2, subsequent steps just require looping of these calculations for different cash flow and assumption sets. It would therefore make sense to have all CSM calculations together in one place. As items 1 and 2 are likely to generate the vast majority of the data it is likely to make sense for the CSM calculations to form part of the actuarial systems. It will be important that the additional data and processing are handled with sufficient validations, controls and governance to gain audit approval. For some insurers, the current actuarial systems environment can provide this while for others it may lead to some additional investment in a workflow solution, such as Unify. However, such investment could benefit not only the IFRS 17 production, but also the Solvency II and other metrics currently produced from the same environment. IFRS 17 technology solutions the CSM calculation 2
An example CSM calculation in Willis Towers Watson Software We have produced an example of a CSM calculation for an insurance contract under the GMM using our in-house software in order to demonstrate how using the actuarial systems would work in practice. The CSM calculation requires interpretation of the standard so the first step in producing a sample calculation in our software was to revert to the standard and make a number of judgements. The key areas of judgement we considered are set out in Figure 2. (Contracts under the Variable Fee Approach are discussed below; the CSM for reinsurance contracts operates a little differently with no onerous contracts test.) Step 1: Net present value calculation The point of sale CSM is effectively defined as the maximum of net present value of the cash flows (including expected premiums and allowing for the risk adjustment) at the appropriate discount rate on a policy by policy basis and nil (i.e. the CSM cannot be negative). Insurers actuarial systems Figure 2. The CSM calculation key areas of judgement Key areas of judgement Onerous contracts test at point of sale Things to consider/what the impact is? All judgements affecting the other building blocks (projection and discounting of cash flows and the risk adjustment) will also affect the onerous contracts test Identifying a contract group flag which identifies whether contracts are onerous or have a possibility of becoming onerous later or not Point of sale interest rates used in the roll forward Choice of coverage unit Actual experience information Ordering of processing the income statement Depending on the products written and the volatility of the profit, how the groups are set could affect the pattern of profit earned Firms will need to consider how many interest rates to apply, taking into account the type of business, how sensitive the business is to interest rates and the availability of interest rate curves at different dates The coverage unit is the measure used to amortise the CSM over time hence affecting the profit recognition. Decisions are needed on which drivers are applied for which business, and whether or not to allow for discounting The level of granularity required for bringing through the actual experience and how this is brought in needs consideration The ordering of the analysis of change can have an impact on the measure of CSM at interim steps, but will not affect the valuation date CSM allocation of profit and loss which is exacerbated with the no-negative rule 3 willistowerswatson.com
will already efficiently project policy cash flows on a set of assumptions and discounting using prescribed discount rate curves. Adapting these systems to perform the projections and discounting required under IFRS 17 is therefore likely to be relatively light touch. The additional functionality needed by most insurers in this step is likely to be data preparation and storage, including policy data, discount rate curves and the granularity of assumptions required. We have implemented this step in our actuarial cashflow modelling system RiskAgility FM for a sample set of policies. Our calculation uses the Solvency II best estimate cash flow functionality but adds in a risk adjustment projection on a cost of capital approach and IFRS 17 discount rates by cohort for a block of new business. The development is easily re-useable within company specific systems but will require some customisation. For example, different firms will have different views regarding the discount rate and risk adjustment Step 2: Allocating contracts to profitability groups Following the requirements of IFRS 17, contracts are allocated to a profitability group depending on the level of the CSM at outset. IFRS 17 requires a high level of granularity with policies needing, for CSM purposes, to be grouped by policy type, entry year and at least the following three categories: onerous; likely to become onerous in the future; and unlikely to ever become onerous. The allocation remains unchanged for the lifetime of the contract. Each group will use the same assumption set and the CSM calculations are typically carried out at this level. We have implemented this part of the calculation in our data validation tool DataValidator, which ranks new policies by CSM and assigns a cohort flag to each policy record depending on pre-defined grouping rules. This will form part of the new business reporting process and is therefore only relevant to insurers which are open to new business, except for the one-off retrospective piece of work required for transition. Step 3: CSM roll forward and analysis of change At the point of sale of a new business contract, the CSM is intended to measure the profit an insurer expects to earn over the lifetime of a contract or a group of contracts. As the contract (or group of contracts) matures, the CSM is rolled forward and serves as a buffer against some elements of actual experience. Effectively, for the GMM some non-economic effects can be offset against the CSM reducing volatility, whereas most economic effects cannot. For the Variable Fee Approach, applicable under certain conditions to some with profits and unit linked business, the CSM can also be used as a buffer for some investment variance. See Figure 3. Figure 3. Subsequent measurement of the CSM differs between the approaches General measurement model Variable Fee Approach for direct participating business Accretion of interest Investment variance Cash-flow changes related to future services Cash-flow changes related to future services CSM Beginning of period Allocation to current period in income statement CSM Beginning of period Allocation to current period in income statement CSM End of period For direct participating businesses the CSM also buffers changes in the entity s share of the underlying item CSM End of period IFRS 17 technology solutions the CSM calculation 4
In the modelling this requires an additional CSM calculation routine to be run on top of each of the analysis of change runs in the main actuarial cashflow model. This effectively requires a loop of the same calculation to be run under multiple different assumption sets. This in turn produces the IFRS 17 components of the income statement and balance sheet. Specification of the output is required for the accounting ledger and for review and audit. We have implemented this part of the process within our workflow system Unify to show the power of automating the calculation looping within tight financial reporting timeframes and demonstrate why we think automation is so important for good process control and governance. Figure 4 shows a sample CSM analysis of change and the balance sheet and income statement output as created by our proof of example model. Figure 4. Balance sheet and presentation of the income statement Unify takes the output of the IFRS 17 model via DataValidator to a Reporting Front end IFRS 17 income statement Year 2017 2018 Life insurance Month 12 12 Insurance service result 790 1,361 Insurance revenue 1,203 2,217 Expected claims death 68 131 Expected maintenance expenses 97 168 Allocation of recovery of insurance acquisition cash flows 313 557 Release of contractual service margin to P&L 617 1,151 Release of risk adjustment to P&L 108 211 Insurance service expense 413 856 Incurred claims death (excl. investment component) - 131 Incurred insurance service expenses 100 168 Allocation of insurance acquisition cash flows 313 557 Net finance result (3,079) - Investment income (322) (595) Insurance finance expense (or income) 2,757 (595) Interest expense (322) (595) Charge in discount rates 3,079 - Profit or loss for the year (2,289) 1,361 CSM roll forward 2017 Opening CSM - New business CSM 14,591 Accrete interest (13) Experience adjustments Change in CF (27,639) Inv. comp/premiums exp. adjustment 28,741 Change in RA (240) Change in assumptions Change in CF 244 Change in RA - CSM pre allocation 15,684 Allocation to the period 617 Closing CSM 15,067 CSM roll forward under the General Measurement Model No deaths or surrenders in first half year Change in future expense assumptions Balance sheet Life insurance Best estimate liability (incl. O&G) (17,537) 356,369 305,355 Discounted future cash flows (det.) (17,537) 356,369 305,355 Options and guarantees - - - Risk adjustment 2,946 3,100 2,884 Contractual service margin 14,591 15,067 13,987 Total insurance liabilities - 374,536 322,136 5 willistowerswatson.com
Conclusion We believe that an IFRS 17 implementation solution which has the CSM calculation in the actuarial systems is likely to be the most cost-effective and process-efficient approach. This helps ensure that the current assumptions and functionality that exists does not need to be re-invented and consistency is ensured. Insurers will need suitable controls and governance in place around their actuarial systems to deal with the additional data, storage and calculation looping required under IFRS 17. Any investment in process and data controls, in automation, or in workflow solution, such as Unify, can be used to benefit all metrics produced from the actuarial systems. Further information For further information please contact our IFRS and modelling experts below, or your Willis Towers Watson contact. Neil Chapman Senior Director +44 207 170 2460 neil.champman@wilistowerswatson.com Lucia Lumsdon Director +44 207 170 2131 lucia.lumsdon@willistowerswatson.com Matthew Ford Senior Director +44 207 170 2367 matthew.ford@willistowerswatson.com John Morley Senior Director +44 207 170 2436 john.morley@willistowerswatson.com Kamran Foroughi Senior Director +44 207 170 2743 kamran.foroughi@willistowerswatson.com David Pond Director +44 207 170 2521 david.pond@willistowerswatson.com IFRS 17 technology solutions the CSM calculation 6
About Willis Towers Watson Willis Towers Watson (NASDAQ: WLTW) is a leading global advisory, broking and solutions company that helps clients around the world turn risk into a path for growth. With roots dating to 1828, Willis Towers Watson has over 40,000 employees serving more than 140 countries. We design and deliver solutions that manage risk, optimise benefits, cultivate talent, and expand the power of capital to protect and strengthen institutions and individuals. Our unique perspective allows us to see the critical intersections between talent, assets and ideas the dynamic formula that drives business performance. Together, we unlock potential. Learn more at willistowerswatson.com. Willis Towers Watson 51 Lime Street London EC3M 7DQ Towers Watson Limited (trading as Willis Towers Watson) is authorised and regulated by the Financial Conduct Authority in the UK. The information in this publication is of general interest and guidance. Action should not be taken on the basis of any article without seeking specific advice. To unsubscribe, email eu.unsubscribe@willistowerswatson.com with the publication name as the subject and include your name, title and company address. Copyright 2018 Willis Towers Watson. All rights reserved. WTW-HP-2018-0048 willistowerswatson.com willistowerswatson.com/social-media