At a recent meeting of the London Market Actuaries’ Group (LMAG) we attended, James Orr, Chief Actuary – General Insurance at the PRA, with his colleagues Cassandra Archer and Elena Papastylianou, delivered a PRA Update presentation on the recent “Dear Chief Actuary” and “Dear CEO” letters*.
We think the contents of these letters should be in the forefront of the minds of those involved in the year-end reserving process, as well as Boards and Reserve committees, as they contain important findings from recent reserving reviews.
Three Key Issues & Elephant Traps
- Avoid bias in reserve estimates – the PRA have observed a heavy focus on favourable development potential in claims development, and optimism in the selection of Initial Expected Loss Ratios.
The Elephant Trap: The Bornhuetter-Ferguson method is a staple of actuarial reserving, particularly where data is sparse and forecasting is difficult. Don’t be fooled by mechanical application of this method, as it requires important expert judgements as inputs. Over-reliance on this method (and worse still, the IELR method) exacerbates under-pricing mistakes and delays the recognition of problems.
- Beware of weakening case estimates – the PRA have observed weakening of case estimates, judged by trends in paid to incurred ratios, particularly in certain casualty lines.
The Elephant Trap: A recurrent feature of soft markets is that case estimates are not as strong as they used to be, and that firms miss the indicators. It is therefore key for firms to ensure that case reserve strength is robustly assessed and monitored, and assertions on faster claims development (e.g. due to process changes in the claims function) need to be sufficiently challenged. Also beware of fast-close actuarial reserving exercises, where fourth quarter data is often not reviewed for year-end reserving to the same extent.
- Keep an eye on inflation – the PRA has seen insufficient consideration being given to the possible impact of inflation on long-tailed classes.
The Elephant Trap: Standard actuarial techniques, particularly the Chain-Ladder and other triangulation methods, implicitly assume a constant level of inflation. This will be rendered invalid if inflation is increasing as many claims teams are reporting. Firms must make sure that they have understood these dynamics and have an approach for incorporating evolving inflationary effects in their projections.
* Links to these letters are below: