By John Hoskin, Associate at Barnett Waddingham
These assumptions may have made it past the board, auditor and regulator, but do they reflect reality and how will they stand up to the effect of future experience differing from that assumed? In particular, are the processes in place sufficiently robust and objective to analyse historical data, take into account expectations and manage the emergence of new data?
We know that Solvency II technical provisions should be calculated using best-estimate assumptions with the prudence required to reflect the uncertainty in the business allowed for in the risk margin calculation. However, we also know that finding the true best estimate of assumptions is a much harder job than it sounds. In this blog we sketch some of the complexities in setting and maintaining best estimates, and examine how independent review can add value.
Finding the best…
Most actuaries are experienced in analysing data sets and fitting suitable models to that information. However the need to be able to demonstrate ‘best estimate’ raises some interesting challenges in practice, including:
-
What data sets to use – the most recent data may be more relevant, but could be incomplete. Including historical data may make for a smoother experience, but is such information still relevant?
-
How best to allow for outliers – is that spike in experience an anomaly or part of the experience?
-
When should we update our assumptions – is the most recent experience just noise, or do we need to adopt a revised set of assumptions?
-
How should the business be split into homogeneous risk groups?
-
How can we demonstrate to other stakeholders that our processes are robust and objective?
These and other complexities bring judgement into the calculation and when it comes to judgements, old habits die hard. Under previous regimes, actuaries have not been required to separate prudence from the underlying reserving assumption, and many valuations inherently allowed for some margins based on the uncertainty within the calculation.
These are the types of thoughts that have been common place within insurers for many years. This tendency to include prudence as a safety net may have resulted in actuaries producing less in-depth analysis into understanding the trends or changes in experience than is perhaps suggested as required under Solvency II by Article 272 of Commission Delegated Regulation (EU) 2015/35. It is therefore essential that management understand if and how the mind-set and the processes used by the actuaries advising on the assumptions have developed with the new regulations.
So how can you ensure a robust approach to setting material best estimate assumptions? How can you remove any anchoring effect or unconscious bias from previous prudent assumption setting techniques?
A blend of technical expertise in analysing data alongside a clear and thorough framework for setting assumptions is a good start for helping a firm’s management to gain comfort in the process and a better understanding of changes as they develop. But is this enough?
Proving it is the best…
In our opinion there are three key requirements to ensuring a more reliable result:
Independent review
Looking at the data analysis and assumption setting process with ‘fresh eyes’ allows a reviewer to avoid being led in the same direction as those that produced the result. An independent review should support the derivation of assumptions if they can be considered best estimate, or else identify weaknesses.
Detailed technical knowledge of the underlying risk/assumption
Independence alone is not enough. Neither is it appropriate for the reviewer to simply consider the assumption paper to gain understanding of the risk. To validate the assumption the independent reviewer should have expert understanding of the risk in general. Given certain information about the firm, the reviewer should ask themselves and, more importantly, be able to answer questions such as “What do I expect?”, “Why might the firm’s experience differ from industry benchmarks?” and “What developments have occurred since the last review?”.
Experience of the industry and different modelling techniques
By understanding different approaches and their associated pitfalls, the reviewer should form a clear view as to the robustness of the analysis. Experience of setting and reviewing the assumptions for different businesses and access to industry benchmark data, will help to reveal areas of added prudence or other weaknesses in the approach. In addition, knowledge and use of modern data analytic techniques, which have moved apace in recent years, can add valuable insight.
The clock is ticking
2016 is the year when the Solvency II requirements kick in to full flow. With draft narrative reports, quarterly quantitative reporting templates many more challenges to face (even before we get to year-end), there won’t be many quiet days in the near future.
With so much to do it would be easy to deprioritise reviewing the best estimate assumptions. Why rush? Leaving this for a quieter time will come with a cost.
The most obvious cost will be to the firm’s Solvency II cover ratio, which may be deflated due to the additional capital held against the unnecessary prudence.
A delay may also dent future market and customer confidence. A great deal of information will be made available to both the regulator and the public. Following disclosure of the best estimate liabilities, firms may be encouraged or have a desire to maintain a stable view of their business. Material changes in any assumptions will raise questions. For listed companies, this could impact the market’s confidence in the model and therefore future cashflow projections. All firms may experience some fall-out with regards to sales and retention if policyholders and/or their advisors view any change in assumptions as indicating a weakening in security.
With many companies promising that dividends will be in line with Solvency II results, every bit of capital needs to be used efficiently. In the competitive market, inefficient insurers will soon see the price they pay for not making the most of their money.
|