Transmittal Memorandum - International Actuarial Association
ISAP [5]
(Pro
Insurer Enterprise Risk Models
NOTE: Defined terms and references to ISAP 1 in this proposed final ISAP are marked in blue coloured text with a dotted underline. The defined terms in the approved final ISAP will have hyperlinks to the relevant definition in the glossary, and references to ISAP 1 will have hyperlinks to ISAP 1. Please note that the hyperlinks have not been created in this proposed final version.
Approved by the Council
[Day Month Year]
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
TABLE OF CONTENTS
i
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
Preface
This International Standard of Actuarial Practice (ISAP) is a model for actuarial standardsetting bodies to consider.
The International Actuarial Association ( IAA ) encourages relevant actuarial standard-setting bodies to maintain a standard or set of standards that is substantially consistent with this ISAP to the extent that the content of this ISAP is appropriate for actuaries in their jurisdiction. This can be achieved in many ways, including:
Adopting this ISAP as a standard with only the modifications in the Drafting Notes;
Customizing this ISAP by revising the text of the ISAP to the extent deemed appropriate by the standard-setting body while ensuring that the resulting standard or set of standards is substantially consistent with this ISAP;
Endorsing this ISAP by declaring that this ISAP is appropriate for use in certain clearly defined circumstances;
Modifying existing standards to obtain substantial consistency with this ISAP; or
Confirming that existing standards are already substantially consistent with this ISAP.
A standard or set of standards that is promulgated by a standard-setting body may be considered to be substantially consistent with this ISAP if:
There are no material gaps in the standard(s) in respect of the principles set out in this
ISAP; and
The standard or set of standards does not contradict this ISAP.
If an actuarial standard-setting body wishes to adopt or endorse this ISAP, it is essential to ensure that existing standards are substantially consistent with ISAP 1 as this ISAP relies upon ISAP 1 in many respects. Likewise, any customization of this ISAP, or modification of existing standards to obtain substantial consistency with this ISAP, should recognize the important fact that this ISAP relies upon ISAP 1 in many respects.
If this ISAP is translated for the purposes of adoption, the adopting body should select three verbs that embody the concepts of “must”, “should”, and “may”, as described in paragraph 1.6. Language of ISAP 1
, even if such verbs are not the literal translation of “must”, “should”, and “may”.
This ISAP is binding upon an actuary only if so directed by the actuary
’s standard-setting body or if the actuary states that some or all of the work has been performed in compliance with this ISAP (e.g., if the actuary is directed by the principal to comply with this ISAP).
This ISAP was adopted by the IAA Council in [month year].
[Drafting Notes: when an actuarial standard-setting organization adopts this standard it should:
1.
Replace “ISAP” throughout the document with the local standard name, if applicable;
2.
Modify references to ISAP 1 in paragraphs ………………and ……..to point to the local standard(s) that are substantially consistent with ISAP 1 , rather than referring to ISAP
1 directly, if appropriate;
3.
Choose the appropriate phrase and date in paragraph 1.7;
4.
Review this standard for, and resolve, any conflicts with the local law and code of professional conduct; and ii
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
5.
Delete this preface (including these drafting notes) and the footnote associated with
iii
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
Introduction
This International Standard of Actuarial Practice (ISAP) provides guidance to actuaries when performing actuarial services involving the use of models which assess solvency and produce risk metrics for Enterprise Risk Management (ERM) programs of insurance entities.
Actuaries play a principal role in assuring financial soundness of insurers. Their tools include ERM and the use of enterprise risk models for assessment of capital. Enterprise risk models are those models that are developed for insurers to comprehensively and consistently evaluate risks.
Examples include “Economic Capital Models” and “Internal models” as defined under Solvency II and by the International Association of Insurance Supervisors (IAIS). Specifically, the central importance of enterprise risk models to insurance business management is clearly demonstrated in two of the core principles published by the IAIS for assessment and supervision of insurance firms, ICP 16 - Enterprise Risk Management for Solvency Purposes and ICP 17 - Capital
Adequacy.
Increasingly Boards and senior managements of insurers rely on enterprise risk modeling for both regulatory and management decision-making purposes. As a result, insurance firms, their stakeholders and other interested parties have a strong interest in the reliable operation and transparent governance of the use of enterprise risk models. As employees or advisors, actuaries play an important role in advising firms and others on the development or selection of the appropriate models and the related testing techniques.
This ISAP is intended to:
Facilitate convergence in standards of actuarial practice in connection with Insurer
Enterprise Risk Models within and across jurisdictions;
Increase public confidence in actuaries
’ services for ERM purposes; and
Demonstrate the IAA
’s commitment to support the work of the IASB in achieving high quality, transparent, and comparable, financial reporting internationally, as envisaged by the Memorandum of Understanding between the IAA and the IASB. iv
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
Section 1.
General
1.1.
Purpose - This ISAP provides guidance to actuaries when performing actuarial services involving the development and use of models which assess solvency and produce risk metrics for ERM programs of insurance entities. It is expected to:
Facilitate widely accepted convergence of principle based ERM actuarial standards within and across jurisdictions;
Demonstrate the IAA
’s commitment to support the work of the profession in achieving acceptance of actuaries as valuable professionals in the ERM area.
Help to increase public confidence in the ERM work provided by intended users confidence that: actuaries b y giving
Actuarial services are carried out professionally and with due care;
The results are relevant to their needs, are presented clearly and understandably, and are complete; and
The assumptions and methodology (including, but not limited to, models and modelling techniques) used are disclosed appropriately in the actuary
’s report .
1.2.
Scope
– This standard will apply to actuaries when performing actuarial services involving the development and use of enterprise risk models, including stress and scenario tests, to assess solvency and produce risk metrics for ERM programs of insurance entities.
1.3.
Compliance
– There are situations where an actuary may deviate from the guidance of this
ISAP but still comply with the ISAP:
1.3.1.
Law may impose obligations upon an actuary . Compliance with requirements of law that conflict with this ISAP is not a deviation from the ISAP.
1.3.2.
The actuarial code of professional conduct applicable to the work may conflict with this ISAP. Compliance with requirements of the code that conflict with this ISAP is not a deviation from the ISAP.
1.3.3.
The actuary may depart from the guidance in this ISAP while still complying with the ISAP if the actuary provides, in the report , an appropriate statement with respect to the nature, rationale, and effect of any such departure.
Paragraphs 2.6, 2.8, and 2.9 of ISAP 1, as modified by paragraph 2.4 of this ISAP, cover the situation where the actuary is directed to use certain assumptions or methodology. The actuary who complies with these paragraphs is not deviating from this ISAP.
1.4.
Relationship to ISAP 1
– Where possible, this ISAP does not repeat guidance already provided in ISAP 1 . Any actuary who asserts compliance with this ISAP (as a model standard) must also comply with ISAP 1 . References in ISAP 1 to “this ISAP” should be interpreted as applying equally to this ISAP [5], where appropriate.
1.5.
Defined Terms – This ISAP uses various terms whose specific meanings are defined in the
Glossary. These terms are highlighted in the text with a dashed underscore and in blue, which is a hyperlink to the definition (e.g., actuary ).
1.6.
Cross References
–
1
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
1.7.
Effective Date – This ISAP is effective for { actuarial services performed/ actuarial services commenced/ actuarial services performed with respect to an IFRS financial statement for a reporting period ending} 1 on or after [Date].
1
[Phrase to be selected and date to be inserted by standard-setter adopting or endorsing this
ISAP.] .
2
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
Section 2.
Appropriate Practices
2.1.
Knowledge of Risk and Uncertainty – To be confident in performing the actuarial services , the actuary should have or obtain sufficient knowledge, understanding and experience of the nature of risk and uncertainty in relation to the subject of the work . In this context risk is related to the unknown outcomes of a known probability distribution and uncertainty is related to the unknown outcome of an unknown probability distribution.
2.1.1
In performing services related to risk evaluation, the actuary should consider, or may rely on others who have appropriately considered, the following: a.
Information about the financial strength, risk profile, and risk environment of the organization that is appropriate to the assignment; b.
Information about the organization’s own risk management system as appropriate to the assignment; c.
Relationship between the organization’s financial strength, risk profile, and risk environment as identified in 2.1.1.a. above, and the organization’s risk management system as identified in 2.1.1. b. above. If in the actuary
’s professional judgment, as appropriate to the assignment, a significant inconsistency exists, then that inconsistency should be reflected in the risk evaluation; and d.
Intended purpose and uses of the model.
2.2.
Data Quality Considerations –
2.3.
Assessing Consistency with the Valuation Framework (Bases)
–
Insurer enterprise risk models often rely on the stochastic nature of a portfolio of assets and liabilities, the entire balance sheet or off balance sheet items.
2.3.1.
For risk modelling purposes special valuation frameworks may be used to place a value on an uncertain item that is different from the value stemming from any accounting or regulatory type valuation frameworks.
In such circumstances the actuary should consider to what extent the valuation framework used in the risk model should be the same or consistent with those used for other valuation frameworks. Such other valuation frameworks may include accounting type valuation or other risk type valuation (such as used in regulatory risk modelling).
2.3.2.
When making a choice for evaluation consistency the actuary should consider the following aspects where appropriate: a.
Relevant accounting valuation framework; b.
Relevant regulatory valuation framework; and c.
Relevant risk valuation framework used for other purpose than the particular purpose of the work .
2.3.3.
The actuary should assess the impact of using a different valuation framework for risk modelling from other relevant purposes.
2.3.4.
The actuary should disclose in the report : a.
Whether a consistency check has been conducted between the valuation framework used for the particular purpose of the work and the valuation
3
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year] framework used for other relevant purposes and the main reasoning of the actuary
’s choice; b.
Main results of the actuary ’s evaluation process; c.
The impact of using a different valuation framework where appropriate; and d.
A reasonable reconciliation between the valuation frameworks considered where appropriate.
2.4.
Assumption Setting – The actuary should take into account the requirements of ISAP 1 regarding the setting of assumptions.
2.4.1.
The actuary should give careful consideration and apply professional judgement in the choice of assumptions for the model.
Assumptions can be classified into the following four categories:
Observable from market data
– Assumptions might be based on observable market data, if available. For example, assets and liability products with deep and liquid markets will have reliable market prices readily available. The actuary should consider the appropriateness of the assumption taking into account trends and likely future experience.
Best estimates
– The calculation of best estimate assumptions should incorporate company experience, industry experience, changes in the environment and observed trends. The actuary should consider whether it is appropriate to include a risk margin or PAD (provision for adverse deviation) to reflect the uncertainty in the determination of the best estimate.
Management action
– Management actions can impact many aspects of the business, including compensation, expenses, reinsurance, investment and hedging strategies. In setting assumptions based on assumed management action, the actuary should consider any contractual requirements, policy language, approval process, timing, and past experience.
Assumptions outside of management control
– Examples of assumptions outside management control include tax rates, regulatory requirements and reserving requirements. The actuary’s assumptions should normally reflect the actual situation as of the valuation date, as well as any known future changes.
2.4.2.
The actuary should exercise professional judgment in the selection of assumptions, taking into account factors such as historical data, market prices, views of other experts, fit of the assumed distribution to available data, the ability of the assumed distribution to reflect possible extreme values, sensitivity of results to changes in assumptions, internal consistency of the assumptions, and consistency in the application of assumptions. In selecting the assumptions, the actuary should consider the time horizon to which the model will be applied.
2.4.3.
The actuary should ensure that the assumptions are realistic, applicable and relevant for the situation, objective, used consistently through the models, accounting for future developments, and consistent with the assumptions used for other purposes.
2.4.4.
The actuary should update experience studies regularly to verify whether the assumptions are still appropriate, including back-testing comparisons with actual
4
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year] experience. When substantial differences are identified the actuary should implement remedial action.
2.4.5.
The actuary should calibrate the model’s underlying assumptions and parameters such that the resulting model output aligns with observable or known values, including industry or regulatory guidance. This re-calibration exercise should be repeated periodically to allow for recent, relevant and credible experience.
2.4.6.
The actuary should document the justification for, and derivation of, the assumptions such that they can be independently verified by another actuary qualified in the same practice area.
2.5.
Scenario and Stress Testing – Models normally assume that the external economic and internal business environment is stationary or develops along a smooth curve on a forward looking basis. Stress and Scenario testing (SST) is used to understand what happens if the environment experiences sudden discontinuities. A stress causes a discontinuity in one variable whereas a scenario consists of stresses related to each other in a consistent manner.
2.5.1.
When constructing a scenario for purposes of SST, the actuary should ensure that the scenario is adverse but plausible by obtaining input from appropriate sources, for example: a.
Management of the company being modelled; b.
Knowledgeable persons at the company; c.
The company’s business plan regarding how the company will function during a catastrophic event; d.
External industry experts; e.
Regulatory requirements; f.
Economists; and g.
Subject matter experts.
2.5.2.
The actuary should consider the following, if appropriate: a.
Some assumptions, including those regarding risk distributions, may differ from their baseline values because of the defined stress; b.
In an extreme event, management may delay decisions or make quick decisions that are inconsistent with business plans or business practice; c.
The actions and reactions of various stakeholders, including policyholders, and markets may differ from those during “normal” times; d.
Regulatory and legislative authorities may require additional capital; e.
Free transfer of assets may be restricted by either operational or legal limitations; [from ICP 16 of IAIS paragraph 16.14.14] f.
Secondary effects under the scenario might occur in a later time period than the stress itself; and g.
Using different degrees of adversity may affect the comparability of stress tests. [from ASOP 46 paragraph 3.4]
5
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
2.5.3.
Reverse stress testing is the process used to back-solve the required stress and / or scenario events that will give rise to a specific adverse business outcome. When applying reverse stress testing the actuary should derive the stress test in a way that gives rise to the most likely combination of risk factor events. The Actuary should consider the merits of performing reverse stress testing. These reverse stress tests may be more extreme than plausible scenarios.
2.5.4.
In relation to the SST the actuary should disclose: a.
The significant assumptions used in the SST, including the actions assumed to be taken by management; b.
Any known limitations of the SST and an assessment of the potential impact of these limitations on results; c.
The time frame; and d.
The basis of measuring loss.
2.6.
Model Selection – The actuary should consider the appropriateness of the method used to model each risk. Some risks are more appropriately modeled stochastically while others may be more appropriately modeled deterministically or by stress testing.
2.6.1.
Considerations Related to Risk Evaluation Models – In developing, reviewing, or maintaining models used in risk evaluation, the actuary should consider, or may rely on others who have considered, whether the models are fit for the purpose. In making that determination, the actuary should, where appropriate, review the following: a.
Degree to which the models need to be reproducible and adaptable to new risks; b.
Sophistication of the models in proportion to the materiality of the risks they cover; c.
Practical considerations for the models, including usability, reliability, timeliness, process effectiveness, technological capabilities, and cost efficiency; d.
Inherent statistical and theoretical limitations of the models; e.
Quality, accuracy, appropriateness, timeliness, and completeness of data underlying the models; f.
Appropriateness of the methodologies used for model verification and validation, calibration, and sensitivity testing; g.
Appropriateness of the methodologies used for modeling dependencies among risks; h.
Appropriateness of the cash flow and discounting methodologies used in the models; and i.
Appropriateness of accounting and valuation frameworks used in the model and the consistency of their application throughout the model.
2.6.2.
Considerations Relating to an Economic Capital Model – In performing actuarial tasks relating to designing, developing, and reviewing an economic capital model, the actuary should consider the following, if appropriate to the assignment:
6
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year] a.
Appropriateness of the selected time frame, basis of measuring loss (for example, solvency, regulatory standards, earnings loss, reputation damage), and risk metric underlying the organization’s definition of economic capital relative to how it is used to support strategic decisions; and b.
Degree to which the economic capital model reflects the significant risks of the organization and the interdependencies of those risks in a consistent and comprehensive manner.
2.7.
Model Validation –
2.7.1.
To validate the model, the actuary should consider whether: a.
The structure of, and the assumptions underlying, the model itself are reasonably suited to the risks being modeled; b.
The logic by which the intended results are produced by the model is appropriate; and c.
Dependencies between risks are appropriately treated.
2.7.2.
When the main purpose of the model is the determination of appropriate levels of capital to be held for remote events, then the manner in which the model extrapolates from the observed data to the tails of the distribution of gains and losses needs to be considered and disclosed.
The actuary may choose to verify an appropriate implementation by reviewing the major steps in the process and by reviewing the results of the calculations. The actuary may also review the stability of results of the model in actual operation as well as in tests where both minor and major changes are made to various input parameters and the magnitude of changes in output values are reviewed.
2.7.3.
The actuary should test the model before use and also consider whether there are adequate controls to assure that the inputs to the model are as expected and that the output of the model moves appropriately from step to step in the processing.
2.7.4.
The actuary should disclose whether and how the modeled future economic conditions have been reviewed and tested for reasonableness. Items such as the sensitivity of the results to significant changes in the assumptions, time frame, basis of measuring loss, and risk metric may be disclosed.
2.8.
Review of Results – Opening sentence?
2.8.1
In applying the ISAP 1 paragraph 2.10.2 Reasonableness checks, the actuary should consider the following items when reviewing the results of the work : a.
Intended usage of the results: b.
Reasonability of the results in the light of: i.
Relevant risk appetite; i.
Current risk profile and its changes in time; ii.
Applied risk measures and the valuation framework; and iii.
Data used. c.
Consistency between the results and the inputs;
7
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year] d.
If the model includes many stochastic runs then an assessment of the appropriateness of the distribution of the outcomes from the various runs and the consistency between that distribution and the overall results of the model; e.
Effect of any simplification used; f.
Robustness of the results, in particular considering the potential effect of model risk; g.
Extent to which the results can be considered reliable, especially identification of the cases when the results cannot be considered reliable for the purpose of the actuarial service (limitation of the model).
2.8.2
The actuary should consider specifically the effect of any aggregation techniques used by comparing pre- and post-aggregation results for reasonableness.
2.8.3
The actuary should consider obtaining validation of the results of the model by means of an independent peer review and/or by obtaining input from independent parties
(internal and/or external) that understand the business and its goals and strategies.
2.9.
Frequency of Calculations – The fundamental purpose of the enterprise risk model is to provide management and other stakeholders, such as the regulator, with timeous information to enable the organisation to address the various risks to which it is exposed. The actuary should therefore monitor the results of the model with a frequency that is appropriate to the risk in question, noting that for some risks it might be sufficient to monitor on a monthly basis whereas other risks might be monitored as close to real-time as possible. Monitoring should be sufficiently frequent to allow decisions to be made and for action to be taken on an informed basis.
2.10.
Update Process – The effectiveness of the risk model depends on the validity of the underlying assumptions and methodology. These need to be continuously reviewed and updated for any actual and anticipated changes in the environment in which the insurer operates. To this end the actuary should consider the following, and make the appropriate changes to the model: a.
b.
Changes to the ERM practices of the insurer;
Changes to the business practices of the insurer, for example: i.
Changes in sales and marketing practices that might result in a change in the make-up of the customer base; ii.
Introduction of new products or changes to existing products; iii.
Changes in advertising or sales process communications that might result in significantly different policyholder expectations; iv.
Changes in customer services practices that might result in better or worse renewal or retention experience going forward; v.
Changes in claims payment practices that might affect the amount of future claims or cause better or worse renewal or retention experience going forward; and vi.
Changes in practices regarding premiums or indeterminate policy charges that might result in a change in the claims ratio or in a different make-up of
8
e.
f.
g.
h.
c.
d.
i.
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year] customer base and/or better or worse renewal or retention experience going forward;
Similar changes in business practices of competitors, if known, and their potential impact on the insurer;
Changes in the financial markets for investments of the insurer that make some or all financial instruments more or less risky going forward;
Changes in the environment regarding physical risks drivers for applicable coverages that are evidencing secular trends or regime shifts;
Changes in the environment regarding casualty risks for applicable coverages that are evidencing trends or regime shifts;
Changes, or anticipated changes, in insurance legislation;
Changes that are likely to be introduced as a result of legislators placing greater emphasis on Treating Customers Fairly ("TCF");
Changes to the usage of the enterprise risk model and the different information and timing of output that might be needed to support the new usage.
9
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
Section 3.
Communication
3.1.
Disclosures in the Report – In addition to complying with ISAP 1 Section 3.
Communication, the actuary should disclose in the report : a.
Any material deviation from the guidance in this ISAP (1.3.);
b.
Any reliance on the principal
’s representations regarding constructive obligations
( Ошибка! Источник ссылки не найден.
.); and c.
Any information regarding a change in the process for selecting assumptions that is requested to be disclosed (
Ошибка! Источник ссылки не найден.
.).
10
Exposure Draft ISAP [5] Insurer Enterprise Risk Models [Month Year]
APPENDIX
11
