IFRS 17 And Navigating Financial Condition Testing

Insurers that report on an International Financial Reporting Standards (IFRS) basis are required to apply IFRS 17 Insurance Contracts for annual reporting periods starting on or after January 1, 2023. The implementation of IFRS 17 demands a different approach to financial condition testing (FCT), a risk management tool insurers use to assess their forecasted capital and solvency position, typically over a three- to five-year horizon. FCT involves stress testing an insurer's financial position under various scenarios, which allows management to understand the capital implications of their business plan and provide awareness of significant risk exposures.

Here, we delve into the impact of IFRS 17 on FCT, how it compares to FCT performed on an IFRS 4 basis, and explore how insurers can adjust to these changes.

This article assumes some familiarity with Canadian actuarial practice, and focuses primarily on Canadian insurance business measured under the General Measurement model (GMM). But many of the principles discussed could be applied to other jurisdictions and types of insurance business.

Design and presentation of adverse scenarios

The design of the FCT base scenario should be largely unchanged under IFRS 17. An insurer’s view of expected future insurance claims and market conditions would not normally be affected by the introduction of a new accounting standard. An exception would be if an insurer affected changes to its strategic priorities or its risk management practices as a result of the transition to IFRS 17, in which case these changes would be reflected in the FCT’s base scenario.

For adverse scenarios it is important to note that the severity of an adverse scenario should be based on its inherent loss distribution, not on how a given accounting or capital regime presents those results in time. What constitutes a 90^th percentile adverse mortality outcome or a 95^th percentile adverse interest rate movement, for example, is based solely on the perceived distributions of those risk factors. The role of FCT is to assess the impact of those adverse risk factor outcomes through the lens of the prevailing accounting and regulatory capital regimes.

The change in regime to IFRS 17 changes the lens through which the adverse scenarios are viewed. Financial shocks (for example, adverse interest rate or equity changes) are likely to have a greater impact on the insurer’s income and capital ratios under IFRS 17 than under IFRS 4. Adverse non-financial shocks will in many cases be absorbed by contractual service margin (CSM) and therefore have little impact on short-term IFRS 17 income; however, given that CSM counts as available capital under the Canadian capital regime, the impact on CSM will still have an immediate adverse effect on capital ratios.

The change in lens highlights at least two important considerations:

CSM as a key metric under IFRS 17. Under IFRS 4, FCT reporting typically focused on the impacts to income and capital ratios, and possibly to capital transfers. While still relevant, under IFRS 17 it is also important to consider the impact to the CSM in order to understand the full story of an adverse scenario.
Changing sensitivity of an insurer’s key metrics to each risk type. For example, the valuation of insurance liabilities is more independent from the valuation of assets under IFRS 17 than it was under IFRS 4 and, as such, market risks could gain importance in FCT analysis. These kinds of changes can affect which risks are deemed most important to analyze and include in the FCT reporting. An insurer’s IFRS 17 policy decisions will also play a part in relative sensitivities, for example, whether products are variable fee approach (VFA) vs. GMM, whether the other comprehensive income (OCI) option is elected for insurance finance expense; these choices will affect income volatility under adverse economic conditions and as such may affect the choice of adverse scenarios included in the FCT report.

Overall the design of an adverse scenario exploring a given risk is not expected to change as a result of the transition to IFRS 17, but it is expected that the risks and scenarios prioritized for inclusion in the FCT report will change due to changing sensitivities, and that there will be an update to the metrics used to present the results.

Putting new business profitability in focus

Insurers should consider how they would adjust their new business plan (such as new business volumes and mix) under the new accounting regime. For example, some insurers will see higher new business strain due to profit deferral under IFRS 17 from profitable contracts, which may impact either pricing or product design strategy. It would be important to discuss these considerations with the pricing and product management teams during the FCT process.

Modeling of new business is now more complicated under IFRS 17. New policies need to be allocated to different IFRS 17 profitability groups according to the insurer's grouping policy, on a seriatim basis or using a heat map based on risk characteristics. Some insurers may consider simplifying their new business modeling by offsetting new business CSM and loss component (LC) depending on the materiality of new business CSM/LC at initial recognition. Others may consider approximations such as estimating in advance what proportion of new business will be onerous vs. non-onerous instead of modelling this dynamically. Regardless, it is important to revisit how the grouping logic is applied under each adverse scenario where profitability could be impacted.

Typically, new business projections for onerous groups should be similar under IFRS 17 as under IFRS 4 as the losses get recognized through income right away while non onerous groups would experience different net income impacts due to the establishment of CSM under IFRS 17, which defers profit over the life of the contract.

Other IFRS 17 considerations for new business could include:

Reinsurance held contracts. While less material from an income perspective (as reinsurance held contracts can have a negative CSM), IFRS 17 does require new reinsurance held contracts to also be split between loss-making and profit-making. Reinsurance contracts must also be modelled separately from direct, which is different from IFRS 4
Contract boundaries. What proportion of “new business” assumed would belong in existing groups (in cases such as whole life conversions and riders added on to existing products) vs in ‘new’ groups (such as true new contracts)

The roll-forward of liabilities gets more granular

The roll-forward of the liabilities is one of the more complex requirements of an FCT on an IFRS 17 basis, and a big change compared to doing the same on an IFRS 4 basis. Under IFRS 4, the liability at each projected year-end had the complexity of being dependent on the insurer’s asset holdings and its evolving mismatch relative to the liabilities, but had the simplicity of being modeled in aggregate and often on a net of reinsurance basis. Under IFRS 17, the roll-forward of the liabilities, while now detached from the insurer’s asset portfolio, needs to be much more granular: fulfilment cash flows separate from CSM, direct separate from reinsurance, financial assumption changes separate from non-financial assumption changes, tracking experience gains/losses, and all at the IFRS 17 group level.

Depending on the actuarial modeling platform used, it is possible to get some or even all of the required breakdowns for the liability roll-forward; this significantly reduces the burden on external tools to calculate these items explicitly. However, most modeling platforms still do not have a CSM projection capability which therefore likely needs to be completed in a separate tool. The complexity of these tools depends on insurer requirements such as, “Is CSM/loss component/loss recovery component on reinsurance contracts material?”, “Do they have both direct and reinsurance?”, “Is the business measured under GMM, VFA, or premium allocation approach (PAA)?”, and so forth.

There are also quite a few notional items to be rolled-forward under IFRS 17. This potentially includes the loss component (LC), the loss recovery component on reinsurance contracts held (LRECC), and unamortized acquisition expenses, summarized below:

The LC roll-forward is a presentation item on the income statement and may be immaterial but can become more important under adverse scenarios where experience changes can result in groups switching between LC and CSM. LC is important if an insurer specifically looks at insurance revenue and insurance service expense results separately; however, if an insurer does not expect to write a significant amount of onerous business, this roll-forward can potentially be simplified or skipped entirely
The LRECC does have a direct impact on the reinsurance held CSM and therefore income; similar to LC roll-forward, this LRECC roll-forward can potentially be simplified or skipped entirely if immaterial
Similarly, the recovery and amortization of acquisition expenses on new business does not affect the bottom line for insurers but does affect individual line items on the income statement

Adverse scenarios add another layer of complexity to the liability roll-forward. Impacts to the CSM from experience changes (either due to survivorship or non-financial assumption changes) need to be costed at the locked-in rates for GMM business under IFRS 17. The impact of actual vs. expected premiums, premium related cashflows, and investment components also needs to be tracked to pass through to the CSM. Other items to be considered under adverse scenarios are:

The possibility of the CSM being wiped out and turning into a LC
Measuring the impact of adverse survivorship to pass-through CSM (this can be particularly difficult as the "expected” will re-base each reporting period as actual experience unfolds)
Accommodating assumption changes at various times within the projection period

Taxes are certain, but future tax provisions do change

The Canada Revenue Agency (CRA) adapted the income tax rules for insurers to account for the change to IFRS 17. The two main elements to consider for FCT purposes are:

Non-deductibility of CSM: while for many insurers the maximum tax actuarial reserve was largely the same as the IFRS 4 liability, under IFRS 17, 90% of most CSM is non-deductible, creating a temporary difference between the tax and statement reserves and an associated Future Tax Asset (FTA)
Transition rules: this rule phases in the change in the maximum tax actuarial reserve associated with the change from IFRS 4 to IFRS 17 over five years. Insurers will have either an associated FTA or Future Tax Liability (FTL) depending on whether the IFRS 17 liabilities are higher or lower than the IFRS 4 liabilities at the start of the transition.

For many insurers, the FTA will have increased due to the non-deductibility of CSM. This may result in additional capital implications, given that there are limits on the amount of FTA that can be included in available capital and any amount included attracts a 25% credit risk requirement. Where this is material, insurers should explore the implications on current or future capital needs in the FCT scenarios.

Impacts of IFRS 17 on financial condition testing

Overall, IFRS 17 changes the lens with which we view insurer results. This change in lens alters insurers’ sensitivities to risks compared to what they were under IFRS 4. This in turn affects which risk factors command the greatest attention, which adverse scenarios are included in FCT reporting, and how FCT results are presented and explained.

Technically, performing FCT is more challenging under IFRS 17 than under IFRS 4, due in large part to the more granular nature of IFRS 17 liability calculations. The CSM further adds a distinct component that affects the timing of income and taxes, plays into solvency ratios and becomes a new key performance metric. Therefore, the existing tools used in FCT analyses need to be modified and augmented to adapt to the new requirements and new realities. While it is important to consider materiality and to recognize that approximations can be appropriate, we caution that the effort required to update these tools to produce reliable IFRS 17 projections can be significant.