On first sight, the requirements of the new IFRS17 global accounting standard that will come into effect on January 1, 2021, appeared none too daunting for property and casualty (P&C) insurers. In general, such insurers appeared pretty relaxed, even nonchalant, since most of the more difficult challenges seemed to revolve around the treatment of life insurance contracts, particularly those with participation features.
Recently, however, P&C brows have become more furrowed. As companies have moved beyond setting up an IFRS17 gap analysis and implementation plan to scoping out the detailed requirements and the accompanying technical and process responses, some tricky issues have emerged. Assumptions that the standard would allow P&C insurers to continue with most of their existing accounting and reserving practices are beginning to look optimistic.
To use the analogy of the popular board game of snakes and ladders, many insurers initially saw a board comprising many ladders and few snakes. Yet it’s now becoming apparent that, as attention has switched to scoping out the detailed requirements and the accompanying technical and process responses, many of the ladders that companies thought they could use to speed their progress are nonexistent, or at the very least, more rickety than imagined. At the same time, the number of snakes hiding in the grass waiting to potentially send programs slithering off course seems to have multiplied.
Here are four examples that illustrate where some unexpected snakes may lie:
1. Premium allocation approach (PAA)
For reporting future short-term insurance contract income under IFRS17, the use of the PAA looked like a solid bet since it seemed comfortingly familiar to the existing concept of unearned premium reserves. With no need to go through explicit building-block calculations, including risk adjustment, discounting and the perceived horrors of the Contractual Service Margin, what was not to like?
The potentially venomous bite is in the detailed PAA wording. For contracts with a coverage period of more than one year, the eligibility requirements necessitate a comparison with the results under the general measurement model (GMM), and use terms such as “reasonably expects,” “not differ materially” and “no significant variability of fulfilment cash flows,” each of which requires the entity to exercise significant judgment.
More specifically, the PAA is based on actual premiums received, rather than on due or written premiums, whereas insurers’ current systems are generally unable to link unearned premium to cash. The need to match actual cash receipts to actuarial calculations could be a problem for many insurers, as could accounting for premiums due but not collected.
And last but not least, the requirement to recognize receipts on an instalment-by-instalment basis may represent a big change from current accounting in some jurisdictions. While the PAA is still the preferred route for most P&C insurers, the existence of these snakes means the decision may be far from automatic.
2. Onerous contracts and contract aggregation
At first glance, the definition of portfolios as contracts with similar risks managed together seems analogous to homogenous risk groups as defined in Solvency II. And in order to determine whether contracts are onerous (i.e., potentially loss-making), insurers using PAA are spared the burden of performing detailed quantitative modelling of cash flows (which would negate the benefits of the PAA), and can make the assessment qualitatively, based on facts and circumstances. A couple of ready-made ladders then – seemingly.
Ignoring the fact that the IASB has assumed that onerous contracts are “likely to be relatively infrequent” and “the result of an intentional pricing strategy” – which is hard to reconcile with the prevalence of reported combined ratios over 100% – there are other issues to contend with on closer inspection, particularly around contract aggregation. These include questions about whether companies have to identify onerous contracts on an individual basis and, if not, how far down they need to dig to determine a set of similar contracts; how to treat differential pricing of new business compared to renewals; how to reconcile the potentially different views of pricing and reserving actuaries; and the level to which companies may need to unbundle different risks in a single contract.
3. Outstanding claims
Here again, there appears nothing very taxing about the approach required for IFRS17. Outstanding claims calculations are based on best-estimate cash flows, adjusted for the company’s perspective of risk. While the need for discounting of the cash flows is, in many cases, new for statutory accounting purposes, insurers in Europe should be able to leverage their experience from Solvency II.
The biggest potential snake lurking here is the need to produce numbers by group (as determined at initial recognition and based on the requirement for annual cohorts) and what this means for traditional accident year reserving processes. Does this inevitably mean that a calculation of outstanding claims on an underwriting year basis will be required? Allied to that, there is the need to track and apply multiple discount rates where the entity chooses to disaggregate insurance finance income and expense, as well as the different starting point for determining the discount rate under the PAA and the GMM.
If all of this wasn’t sobering enough, it’s when we come to the treatment of reinsurance contracts that things become even more unpleasant. Indeed, it’s actually difficult to find any ladders at all but rather a veritable pit of snakes.
The standard requires companies to classify, measure and present reinsurance contracts explicitly, independent of the underlying business. One consequence is that businesses might have to recognize the loss on an onerous contract immediately while any potential offsetting gains from reinsurance cover are recognized over the period of the contract. Further mismatches may arise due to the selection of measurement approach, inconsistent (dis)aggregation requirements and non-coterminous coverage periods. And, as if that was not enough, IFRS 17 requires the risk of default by reinsurers to be assessed on an expected loss rather than incurred loss basis and will significantly increase the complexity of dealing with the elimination of intra-group reinsurance on consolidation.
This selection of potentially tricky issues for P&C insurers demonstrates the need to delve into the details of IFRS17 as soon as possible, even within the relative comfort of the simplifications provided by the PAA. What’s clear is that any complacency on the part of P&C insurers is misplaced: yes, there are many ladders but the snakes are often numerous and less immediately visible. Avoiding being bitten requires keen eyesight, careful analysis of the paths offered and appreciation of the risks as well as the opportunities.