1. INTRODUCTION AND PURPOSE 2. DEFINITIONS

Transcription

1 Solvency Assessment and Management: Steering Committee Position Paper 28 1 (v 6) Treatment of Expected Profits Included in Future Cash flows as a Capital Resource 1. INTRODUCTION AND PURPOSE An insurance company can potentially employ various sources of capital as own funds to cover the solvency capital requirement (SCR) of the insurance business. Typically this would comprise of the excess of assets over liabilities (capital provided by shareholders), and potentially also various forms of debt (capital provided by creditors). By recognising the expected profits included in future cash flows (EPIFC) as a part of a company s own funds, the policyholder s contractual commitments to the insurance company could be used as a source of capital. The purpose of this document is to identify the risks associated with recognising EPIFC as a source of own funds and whether these risks should be addressed by the tiering of own funds. 2. DEFINITIONS A detailed description of the EPIFC can be found in section OF 2.4 of the technical specifications of the latest draft version of the SA QIS 2 and is only briefly described below. Section OF 2.4 makes a distinction between a surrender value gap and a paid up value gap. For the purpose of this document reference to EPIFC includes both of these concepts. EPIFC results from the recognition of profits yet to be earned, emanating from the cash flows from existing (in-force) business that is expected to be received in the future. This means that the present value of future net cash-flows in future periods on existing (in-force) business is positive and where these cash-flows were projected and discounted on a best estimate basis. The calculation of EPIFC differs depending on whether or not there is a surrender or paid-up value payable: a) For policies where there is no surrender or paid-up value payable, the EPIFC is simply the value of any negative best estimate liabilities (BEL s) at the calculation date. So if the BEL is calculated to be negative, the EPIFC is the positive equivalent of the BEL. Where the BEL is calculated to be positive, the EPIFC is zero, i.e. there is no profit expected from these policies. b) For policies where there is a surrender value payable, the EPIFC is the surrender value-up gap, i.e. the difference between the amounts determined as the BEL and what would otherwise need to be held as the technical provisions in the event that ALL of these policies surrendered, if the latter is greater. Where the latter number is not greater, the result is zero. c) For policies that include a paid-up value, the EPIFC is the paid up value gap, i.e. the difference between the amounts determined as the BEL and what would otherwise need to be held as the technical 1 Discussion Document 28 (v 6) was approved as a Position Paper by the Steering Committee on 19 April Page 1 of 9

2 provisions in the event that ALL of these policies became paid up, if the latter is greater. Where the latter number is not greater, the result is zero. The EPIFC is based on instances where the surrender value or paid up value is greater than the best estimate liability. In practice the total liability or technical provision will include a risk margin. Where it is material the EPIFC can allow for the risk margin in the calculation. In the current pre-sam environment part (a) above is commonly known as negative reserves. The main difference is that BEL s are calculated under SAM on a best estimate basis whereas negative reserves are determined on a basis that includes valuation margins. To the extent that such negative reserves are recognised under the current legislation, rather than being set to zero, the positive equivalent of these reserves need to be shown as part of the TCAR. Also in the current pre-sam environment part (b) above is commonly known as a surrender strain, which also forms part of the TCAR. For the purpose of this document, additional capital requirements that may arise in the case of a windup is not relevant and do not contribute to EPIFC. 3. DETAILED DESCRIPTION OF THE PROBLEM Reserves (or technical provisions under SAM) are typically calculated as the present value, on a given valuation date of the expected future cash-flows that will be paid (negative cash-flows) less the present value of the cash-flows that will be received (positive cash-flows) for all future time periods, on existing obligations of an insurer, on a specified valuation basis. The purpose of calculating reserves is to determine the amount to be set aside as a liability on the balance sheet of the insurance company in order to ensure that the company will hold sufficient assets to pay claims and other outgo as and when they fall due. A reserve is a single number and does not reflect the impact of the timing of cash-flows in the future, and in particular the timing of negative cashflows versus positive cash-flows. E.g. the future negative net cash-flows may pre-date future positive net cash-flows. Therefore by setting up reserves there is no guarantee that all future cash-flow requirements will necessarily be met. Reserves can be either positive or negative. It is negative when the present value of the future positive cash-flows exceeds the present value of the future negative cash-flows. This typically happens on one of two scenarios: Scenario 1: Policies that carry sizable profit margins (i.e. higher premiums than what may otherwise be actuarially necessary to cover costs on the product given the underlying valuation basis), for the most part of the projected expected cash-flows of the policy. Scenario 2: Policies that include contractual premium increases rather than level premiums such that the expected claims costs and expenses in any given future time period remains less than the premiums payable under the contract. Traditional forms of life insurance typically did not include such increases and the increasing cost of future claims (typically due to increasing age) had to be pre-funded (reserved) from the level premiums received for the duration of the policy. However, newer generations of risk policies sold in the South African market have in recent years included contractual increases in premiums, allowing the premium to start from a lower base, but also allowing the premium in any given time period to be potentially larger than the expected claims costs and other expenses associated with the policy in that particular time period. 4. HISTORIC TREATMENT It is worthwhile at this stage to note that historically (pre-sam) reserves were calculated on a reserving basis, i.e. a best estimate basis plus valuation margins. These valuation margins would reduce the chances of reserves being negative and would otherwise also reduce the significance of negative Page 2 of 9

3 reserves (compared to a best estimate basis). Under SAM the technical provisions are determined on a true best estimate basis, which increase the chances of, and potential significance of the result of the calculation to be negative. Historically (pre-sam) a common practice has been, where negative reserves have been determined, to consider the zeroisation of these (i.e. setting negative values to zero), typically on a per-policy basis. Where negative reserves had been recognised (under the statutory valuation method), it had an impact on the CAR requirements. If Termination CAR is more than Ordinary CAR, CAR is equal to the amount of the negative reserves recognised plus an allowance for surrender strains This is admittedly a very prudent approach as the implicit assumption here is that sufficient capital is required to cover the risk of ALL policies lapsing, being selective against the company (i.e. only those policies with negative reserves lapse and all of these policies lapse). However, when following the principles of Solvency II and SAM, a key requirement is to determine the balance sheet on a market consistent basis. The technical provisions therefore also need to be determined on a market consistent basis. By zeroising such values, a potentially significant implicit margin is added to the liabilities, which may understate the net asset value of the business. By doing this the key Solvency II / SAM principle of market consistency is therefore not adhered to. Therefore, under SAM it is expected that the full EPIFC will be reflected on the balance sheet. 5. RISKS ASSOCIATED WITH RECOGNISING EPIFC AS PART OF THE ELIGIBLE OWN FUNDS The question therefore arises that if EPIFC is recognised under a SAM balance sheet, what risks will a company be exposed to and how should these be addressed. These risks are considered for two scenarios, namely: Scenario 1: Considering the risks where the future demographic and economic experience is worse than the best estimate assumptions used in the determination of the technical provisions. Scenario 2: Considering the risks where the future demographic and economic experience borne out is consistent with the best estimate assumptions used in the determination of the technical provisions. Risks under scenario 1: Lapse Risks: The typical risk associated with EPIFC is lapses, where these are worse than expected. However these risks relate to the extent to which, in this case lapses, may be different from the best estimate expectation and such risks should be adequately addressed in the determination of the SCR. Recognising this risk in the SCR identifies the need to hold the necessary capital. The detail of how the SCR lapse risk module works is not in the scope of this document. However it is important to highlight the basic mechanisms involved as specified in the SA QIS2. The impact of both higher than expected terminations (lapses, surrenders and certain policyholder options), as well as lower than expected terminations (i.e. the direction of the lapses) is tested on a homogeneous group basis at a 50% deviation from the base case. Secondly, a significant proportion, i.e. 45% for retail business (this increased from 30% in SA QIS1) and 70% for non-retail business (i.e. mainly management of group pension funds), of the surrender strain is shocked in the mass lapse risk calculation. This is a significant proportion of the historic TCAR. A combined mass lapse and level lapse shock is also considered. The condition with the worst result is included in the SCR. This means that the potential selective effect of worse than expected terminations is considered. The conclusion from the SCR calculation methodology is therefore that the risk of terminations, in the context of negative reserves and surrender strains, is considered in the mass lapse calculation. The calibration of the lapse risk shock in the SCR, should therefore address the lapse or surrender risks associated with EPIFC. Experience variation Risk: Page 3 of 9

4 A second risk under scenario 1 above is adverse operating experience where these result in a call on available capital. This again is addressed in the SCR calibration. Recognising EPIFC may mean that the funds a company has access to are less liquid, or the company has fewer funds that are immediately available on a going concern basis to pay for this variation in experience. The illiquidity of capital is considered in more detail later in the document. Risks under scenario 2: Liquidity Risk: Where experience is not worse than expected, there remains a liquidity risk associated with EPIFC. Since, as explained above, a fundamental drawback of reserves is that it is a single figure that does not reflect the timing of future cash-flows, there is no guarantee that the present value of future cash-flows (positive and negative) would necessarily be sufficient to cover future negative cashflows as and when they fall due. This problem is however not limited to the recognition of EPIFC as a source of own funds, as is illustrated in the two simple examples shown below. Example A Net cashflow Best estimate technical provision (-NPV of Net cashflow) Cash Required as at start of the projection period Example B Net cashflow Best estimate technical provision (-NPV of Net cashflow) Cash Required as at start of the projection period 1-50 ( ) Valn int (p.a.) % Valn int (p.m.) % The reserve in each of the examples above is the negative present value of the future cash-flows at an annual interest rate of 7%. However the cash required as at the start of the projection period is a calculation such that sufficient cash is held as and when negative cash-flows occur, without taking credit of any future positive cash-flows that are received subsequent to the negative cash-flows. The two examples above are admittedly extreme and rare, but the principle illustrated is that the technical provision (reserve) calculated may not be appropriate and does not necessarily address the risk of liquidity even if this figure is positive. Therefore the risk of liquidity is not limited to the recognition of EPIFC and may also apply where reserves are positive. Secondly, the risk of liquidity applies equally to other sources of funds (other than EPIFC) such as strategic investments which may also not be easily and readily converted to cash when significant negative cash-flow requirements arise. This discussion document does not suggest that the risk of liquidity should be ignored, but that all sources of funds should be treated in a consistent manner when considering and addressing the risk of illiquidity of funds. Page 4 of 9

5 6. TREATMENT OF LIQUIDITY RISK IN SOLVENCY II AND SAM The risk of liquidity did not receive a great deal of written consideration in the EIOPA publications. However the following should be noted: 6.1 The Solvency II directive Article 44 of the Solvency II directive (Risk Management) refers to the risk management system which includes risks that are not, or not fully, included in the calculation of the SCR. Explicit mention is made of asset liability management (ALM) and liquidity and concentration risk management. Article 45 of the Solvency II directive (Own Risk and Solvency Assessment), refers to the need to assess the significance with which the risk profile of the undertaking concerned deviates from the assumptions underlying the Solvency Capital Requirement (SCR). In other words if the risk of liquidity is identified for the company this should be addressed as part of the ORSA given that it does not specifically form part of the assumptions underlying the SCR. Article 132 of the Solvency II directive (Investments the Prudent Person Principle), requires the investments to be made such that all assets, in particular those covering the Minimum Capital Requirement and the Solvency Capital Requirement, shall be invested in such a manner as to ensure the security, quality, liquidity and profitability of the portfolio as a whole. All three the above articles make it clear that the risk or liquidity, if it does exists in the business, need to be addressed appropriately, in the context of Risk Management, the ORSA and the Investment decisions made by the company. 6.2 The Level II advice The Level II advice (former CP 33 under Risk Management) explains in paragraph that the ALM policy shall provide for: - A structuring of the assets that ensures the undertaking holds sufficient cash and diversified marketable securities of an appropriate nature, term and liquidity to meet its obligations, including obligations to pay bonuses to policyholders, as they fall due; - A plan to deal with unexpected cash outflows, or changes in expected cash in- and outflows 6.3 Draft Solvency II regulations Article 252 (SG4) of the draft Solvency II regulations ( SEG consolidation commission draft regulation ) explains in detail what is understood by illiquidity risk management including having to plan how to deal with changes in expected cash in-flows and out-flows. The conclusion from the above is that although the risk of liquidity of own funds held to cover the SCR and assets held to match liability cash-flows is not specifically addressed in the calculation of the SCR (as the case is with the risk of lapses), the risk is addressed through the Pillar II mechanisms which should then also translate into how the own funds of the business, and more generally the assets on the balance sheet are selected by management. Under SAM the FSB would be expected to review a company s ORSA report as part of its supervisory function under Pillar II. In this report it would be expected that the risk of liquidity (if applicable) is identified and addressed to the extent that this risk is not covered in the calculation of the SCR. Page 5 of 9

6 7 TREATMENT OF EXPECTED FUTURE PROFITS IN OTHER JURISDICTIONS 7.1 Europe The EIOPA Level II text (former CP 46) calculates Expected Profits Included in Future Premiums (EPIFP) and suggests that EPIFP and the wind up gap should be classified as Tier III own funds and be subject to the eligibility of Tier III funds. The main difference between EPIFP and EPIFC is that EPIFP ignores single premium contracts where the premium has already been received as well as multi premium contracts where all the premiums have already been received. According to the Level I text (the Solvency II Directive) only one third of Tier III should be eligible. This proportion will not be specified in the primary SAM legislation of South Africa but rather in the subordinate legislation. It is therefore not cast in stone as the case is for the European Solvency II Level I text. Subsequent to the publication of the Solvency II Directive, the Solvency II QIS 5 specified that only 15% of the SCR can be covered by Tier III capital. By contrast to the initial level II text (CP 46 mentioned above), the draft Solvency II regulations ( SEG consolidation commission draft regulation ) states in paragraph (23) [IM7] that: The amount of the excess of assets over liabilities that is included in Tier 1 should not be adjusted for amounts that relate to expected profits that result from the inclusion in the calculation of technical provisions of premiums on existing contracts that will be received in the future. The conclusion therefore from EIOPA is that contrary to their initial thinking EPIFP now remain in Tier I. 7.2 Canada The Office of the Superintendent of Financial Institutions Canada (OSFI) published a guideline on the Minimum Continuing Capital and Surplus Requirements (MCCSR) for Life Insurance Companies (dated December 2010). This guideline considers three primary considerations for defining the capital of a company for purposes of measuring capital adequacy into two tiers, namely Tier 1 ("core capital"), the highest quality capital elements and Tier 2 ("supplementary capital"), Tier 2 is then further split into Tier 2A, Tier 2B, and Tier 2C and requires negative reserves and cash surrender value deficiencies to form part of Tier 2C capital, which would receive the lowest qualification. Source: Australia The Australian Prudential Regulation Authority (APRA) published a discussion paper entitled Review of capital standards for general insurers and life insurers (dated 13 May 2010). This document outlines that policy liabilities would be adjusted to be the best estimate value of the liabilities to policyholders (BEL), with a minimum being the amount payable on voluntary termination of policies or current termination value (CTV). The minimum would be applied for groups of policies, not for each individual policy. This effectively means that the intention of APRA is that there will be no recognition of EPIFP in the determination of the eligible own funds as defined in this document. However this happens on a group level rather than on an individual policy level, so there would be some recognition within groups of policies. It is therefore not as stringent as what it would have been had these profits been ignored on a per policy basis. Source: Page 6 of 9

7 7.4 International Association of Insurance Supervisors (IAIS) The IAIS insurance core principles do not elaborate on the treatment of future profits. There is in summary no conflict with the principles under IAIS in terms of how EPIFC is proposed to be treated in the SAM regulations. 7.5 Comments on Other Jurisdictions In the latest publications, the preferred route in Europe is for EPIFP to be classified as Tier 1. This is justified based on the fact that the remaining risk not addressed in the SCR (liquidity) is addressed via Pillar II. However OSFI and APRA have thus far opted for very limited allowances of these forms of capital in the determination of eligible own funds. Having said that, it should be added that the OSFI and APRA regulations were not studied in its totality and therefore the justification of not recognising EPIFP as eligible capital should be interpreted in the context of the entire OSFI and APRA regulatory frameworks respectively as was done in the case of Solvency II. Secondly the OSFI and APRA frameworks referred to above are also still in development and may still change based on the consultative process that is yet to follow in these two respective regulatory environments. 8 RESULTS FROM QIS1 The submission of QIS included the quantification of Expected Profits In Future Premiums (EPIFP). The main difference between EPIFP and EPIFC is that EPIFP ignores single premium contracts where the premium has already been received as well as multi premium contracts where all the premiums have already been received. We analysed the results from QIS1 to determine the size of EPIFP in relation to the total own funds or available capital. We considered long term direct insurance companies only (including linked companies), representing 35 of the 95 companies that submitted QIS1 returns. We used the EPIFP assumed to lapse figure to derive the negative liabilities on premium paying business and the EPIFP paid up gap to derive the surrender strain. The negative liabilities were estimated as the EPIFP (assumed to lapse) plus negative liabilities derived from the difference between the abovementioned combined EPIFP figures and the technical provisions to identify negative liabilities not captured in the EPIFP calculation. The investigation is subject to the quality and accuracy of the QIS returns. The EPIFP ratios of 4 submissions were nonsensical and ignored for this investigation. The FSB provided us with ratios rather than absolute figures of the QIS1 returns, to ensure confidentiality of the participants. This limited the scope of the analysis. However, it was possible to draw some conclusions on the impact of EPIFC on the own funds (or available capital) and solvency cover of the participants. The graph below focuses on the 24 companies where we ve seen the biggest change in the SCR cover and shows the following: - Adjustment to OF as a percentage of QIS1 OF (2 nd axis). This shows the impact of excluding negative liabilities from own funds (e.g. by a tiering process) as the reduction in the own funds as a percentage of the QIS1 own funds. The elimination of negative liabilities is limited to the difference between the negative liability and the mass lapse, as the risk Page 7 of 9

8 associated with negative liabilities is partly allowed for in the mass lapse shock of the SCR. (Similarly, it could be argued that a portion of all other SCR modules should also offset this adjustment since the other modules also entail shocking these negative liabilities. Furthermore, the diversification allowed for in the SCR through the correlation matrices should also have been allowed for. However, this has not been possible with the available data and the analysis sufficed with incorporating only the Mass Lapse module which is expected to have the biggest impact on the SCR) - SCR Cover (QIS1) is a bar graph showing the SCR cover, i.e. own funds (available capital) covering the SCR. - SCR Cover (OF excl Neg Liab) shows the impact of the reduced own funds, as described above, on the SCR cover. A negative SCR cover is capped at zero. - SCR Cover (OF excl EPIFP) shows the impact on the SCR cover of removing the total EPIFC from own funds, including the paid up gap. The reduction in the own funds is again limited to the difference between the EPIFC and the mass lapse shock. A negative SCR cover is again capped at zero Change in OF & SCR cover by excluding Negative Liabilities in OF 140% % 100% 80% 60% 40% SCR cover (QIS1) SCR cover (OF excl Neg Liab) SCR cover (OF excl EPIFP) Adjustment to OF as a percentage of QIS1 OF (2nd axis) % % The QIS1 results show that 5 companies did not have enough available capital to cover their SCRs. Excluding the negative liabilities (adjusted to allow for the SCR mass lapse as explained above) results in 14 companies not having enough available capital to cover their SCRs. After removing the total EPIFP, 15 companies did not cover their SCRs. The above shows that for about 14 companies, a significant portion of their own funds is represented by negative reserves. For 2 companies the negative liabilities, in excess of the mass lapse shock, are more than their own funds, and for 13 companies negative liabilities are more than 40% of their own funds. We recognise that EPIFP as defined in QIS 1 excludes business where no future premiums are payable and it does not take into account the impact of EPIFP on the risk margin. The broader definition of EPIFC in QIS2 takes this into consideration and explicitly requests the quantification of these components to enhance future analysis. Page 8 of 9

9 9 SUMMARY The risks associated with the recognition of EPIFC as part of the company s eligible own funds are as follows: a) Policy terminations (such as lapses and surrenders, as well as claims that result in policy terminations) being worse than expected; b) Experience variance other than terminations (being worse than expected). c) Liquidity constraints generally due to the timing of cash-flows. Comments on the above risks: a) & b) The risk associated with surrender strains and lapses (and claims resulting in policy terminations) being worse than expected, as well as strains arising from other stressed scenarios, while holding negative technical provisions should be addressed in the calibration of the SCR. It is therefore not required to tier EPIFC in the Own funds. c) It is important to recognise that holding the technical provisions which are determined in the prescribed manner, even if these are positive, does not necessarily protect a company against the risk of illiquidity. The liquidity constraints should be considered for all sources of funds and is not limited to EPIFC. The current requirement under Solvency II and SAM is that the risk of illiquidity should be tested by a company as part of the risk management of the business under Pillar II and as addressed in Article 44 of the Solvency II directive. In addition it would be expected that the ORSA (as stated in Article 45) reports on this risk and how it is addressed as part of the risk management process of the company, where this is not addressed in the calculation of the SCR. In this regard, the treatment of liquidity risk associated with EPIFC and illiquid assets in general, is addressed in Discussion Document 95, which is produced by the capital requirements task group. Tiering EPIFC would be an inconsistent treatment of own funds when compared to certain assets (such as strategic investments) that may be equally illiquid. Given the calibration of the SCR, such an approach would not be market consistent and therefore inconsistent with one of the basic tenets of Solvency II and SAM. 10 RECOMMENDATIONS Based on the above it is therefore recommended that EPIFC be classified as part of Tier 1. Consideration may however be given to how the risk of illiquidity as currently addressed in Pillar II of Solvency II / SAM and how liquidity risk management will be regulated by the FSB under SAM. The disclosure and reporting requirements as a result of holding illiquid assets is discussed in more detail in Discussion Document 95. It is also recommended that all sources of own funds be treated consistently in how the risk of liquidity is addressed. Even without the tiering of EPIFC, we recommend that EPIFC still be quantified to monitor and assess the extent of exposure to liquidity risk. Page 9 of 9