The impact of IFRS 17 on insurers’ operations will be anything but standard, says Alex Foreman-Peck FCCA

Alex Foreman-Peck FCCA is a partner at AFP Advisory.


Get verifiable CPD units by answering questions on this article

IFRS 17, Insurance contracts, issued in May 2016, had the objective of aligning accounting treatment across the industry to increase the understandability and visibility of insurance companies’ performance.

The International Accounting Standards Board’s recent consultation proposing amendments to the standard in the light of stakeholder concerns has meant that the implementation date is likely to be pushed back to January 2022, giving insurers more time to prepare for the changes.

The extent of these changes, including those relating to internal reporting, will vary from insurer to insurer depending on the risks they underwrite and their existing operating model (the interaction of data, systems, processes and people).

The practical impact is more confined to the finance function than first thought

When the standard was near finalisation, there was much speculation about its potential impact. Let’s explore four of the hypotheses.

True or false?

The first hypothesis was that IFRS 17 would require fundamental changes to data capture and analysis in upstream systems. This has proved to be false.

All business functions produce an output. In the case of finance, this is often in the form of numerical reports with interpretation. This requires a process of taking raw data and turning it into information, to be interpreted by skilled people. IFRS 17 states that insurance contract data must be segmented by portfolio, annual cohort and profitability. While the administration systems that feed the finance system often record year of sale, the grouping required might result in extra data having to be captured up front, and these administration systems may require costly modification or even re-implementation.

The standard dictates that contracts should be grouped according to risk type and how the organisation manages that risk when deciding what is a portfolio. However, this grouping tends to occur already, as it is needed for reserving and reporting internally downstream. Sufficient data capture is typically already available in the underwriting/claims systems.

The second hypothesis was that general insurers would not be impacted due to the simplified premium allocation approach. This has also proved to be false. Although the liability for remaining coverage remains substantially the same as under current standard IFRS 4, the fundamentals of calculation for the liability for incurred claims often differs.

New inputs are required that are themselves outputs of calculation –specifically, the discount rate and the risk adjustment to expected cashflows. Where these or equivalents are already used, they are often different from those required by IFRS 17. Consequently, they will require new processes to determine them and, potentially, system support to do so. The discount rate may require new yield curves to be sourced. The illiquidity premium must be calculated and applied at regular intervals.

The risk adjustment has many more options for determination, and these are likely to be different from existing regulatory and statutory equivalents.

The third hypothesis is that, as the standard integrates accounting and actuarial disciplines, processes within finance will need to be re-engineered. Again, false. Many finance functions are divided between the accountants who deal with the accounting for premium and outstanding claims, and the actuaries who calculate the liability for incurred claims for short-term business and/or the long-term liability for business written on a periodic basis. Typically, the function performs a ‘light touch’ calculation at quarterly or half-yearly intervals, with a full report at the year end.

With IFRS 17, the expected cashflows become far more integrated, with those flows realised each period in the income and expenditure statement. Every reporting period requires the unwinding of a discount rate and amortisation of the deferred profit earned for that period. Most of these movements are formulaic and can be accommodated through automated, recurring, periodic journal posting into the general ledger. The month-end process is therefore unlikely to change radically.

Reassessment and potential remeasurement of the capitalised cashflows on the balance sheet, adjusting for actual versus expected cashflows, can continue to be updated quarterly or less frequently. So, while the nature of calculations may change, the activities and their sequence are unlikely to need to.

The final hypothesis is that there will be fundamental change to systems downstream. This has proved true. All insurers will require changes and investment in their systems. This begins at the calculation of cashflows for contract groups, the determination of risk adjustment to those cashflows and consequential deferred profit, the contractual service margin (CSM). Thereafter each component listed above should be recorded at contract inception and subsequently. These balances and movements must be converted to double-entry for posting to the new nominal accounts in the general ledger.

Organisations face a number of choices of where and how the CSM calculation and tracking occurs. Usually the first port of call is the actuarial liability tool, with modifications or upgrades to calculate the building blocks that result in the CSM.

Some providers are extending their functionality in this area, and this can be attractive to insurers looking to upgrade this system anyway. Those not already on the cloud may have problems with storage capacity and may wish to use an existing data warehouse for storage and retrieval. A data warehouse is just that though, and not a calculation engine of the type needed by IFRS 17.

Rather than alter the actuarial tool too much, middleware packages may be purchased to take in the cashflows and do the necessary calculations. This option is more attractive to larger insurers, which may have many actuarial tools in operation and may wish to amalgamate data before posting to the general ledger. These software tools often generate the double entry needed for the general ledger. Again, storage may be a problem.

Surprisingly few finance or enterprise resource planning (ERP) systems have a subledger to provide a more integrated solution than might be available on middleware. The existing IT strategy and architecture may determine which solution type is best.

The path to IFRS 17 will be a long one. The practical impact is more confined to the finance function than first thought and changes measurement rather than the fundamentals of what accountants and actuaries do. Any solution will be IT-dependent.

While the destination has become more certain, the path there continues to provide surprises.