This blog series is a practical guide on solvent exit planning for practitioners and those charged with governance. Part 6 explores assumptions that require simplification.
Specifying, analysing and modelling a solvent exit is a complex undertaking requiring many assumptions. Furthermore, where the starting position for a solvent exit is significantly removed from the current financial or strategic position of the insurer, the exercise can feel hypothetical, making it harder to be accurate over certain assumptions. Therefore, for all insurers, there will be some assumptions that can be specified more precisely and some that require simplification.
The following are some common assumptions that insurers may simplify as part of solvent exit analysis (SEA):
Simplification:
Most insurers would experience a significant period where the business was under-performing before the point of non-viability is reached and the prospect of exit becomes likely. During this period, the underlying composition of the business (for example product volumes and SCR) may change and/or management may take actions to try to recover the business.
A common simplification is to model the impact of this period as an instantaneous shock at T=0. The subsequent run-off projection would then start from this point.
Impact / risks:
The Prudential Regulation Authority (PRA) has been clear that the purpose of SEA is not to create a detailed modelling exercise but to encourage insurers to analyse their business models, understand their vulnerabilities and identify the key barriers and risks to being able to exit in an orderly manner. Therefore, in most cases this simplification is likely to be appropriate, although firms need to think carefully about how actions taken ahead of T=0 would manifest throughout the run-off period.
Simplification:
As part of identifying the levels of non-financial resources required to enact a solvent exit, insurers must consider the key activities in run-off, the staffing levels needed to continue to support such activities and any key staff required (such as management, niche expertise, and retention of corporate knowledge).
Rather than do an analysis at the level of individual staff, many insurers will use simplifications through consideration of percentage reductions in headcount over the run-off or percentage of key staff that may need to be retained. They may also choose to linearly run down staffing levels or align the run-down profile with policy count.
It may also be simpler to assume that redundancies happen at specific dates (for example, year-end) rather than try to model inter-year run-off of staffing levels.
Impact / risks:
It is hard to be precise about staff levels required during run-off and the cost of doing a detailed analysis may not be proportionate to the benefits. In most cases it will therefore be reasonable to use higher level estimation as a basis for cost analysis.
Notwithstanding this, detailed consideration and documentation of the activities required during exit will be required to demonstrate an understanding of the requirements in run-off.
Where there are particularly important individuals for example,. that provide maintenance to specific IT systems, a more granular identification may be necessary.
Simplification:
Insurers typically have a range of arrangements with suppliers, outsourcers and other third parties (for example reinsurers) that require consideration in exit to ensure that:
While the most comprehensive approach is to perform a detailed review of all contracts, this is likely to be a costly and time-consuming exercise even for smaller insurers.
Therefore, insurers may choose to focus on a small list of the most material contracts or those known to have complex features. A simplifying assumption may be to extrapolate costs from this subset to the full supplier list or even assume that less material contracts can be terminated without penalty.
Impact / risks:
Any simplification risks missing complexity in the contractual position. However, for less material contracts this is likely to have limited impact on the overall financial trajectory of the run-off (even where longer contracts need to be broken early).
For most insurers, reviewing a sample of contracts including those above a materiality threshold is likely to be considered proportionate – especially if this covers all material outsourcers and those central to best estimate liability assumptions (for example claims handling costs).
It is common in the SEA to document overall barriers and risks to being able to enact a solvent exit. Therefore it would be prudent for those making a simplification to document the risk of having imperfect information about the contractual position and assessing the likelihood of this impacting other key assumptions (such as the ability to transform the operating model during run-off).
Simplification:
As part of the financial resource assessment, insurers need to identify and consider the impact of any additional costs in exit (for example retention payments for key staff and redundancy payments). In practice these costs would be spread across the run-off period. However, a simplifying assumption could be to assume that they all happen together upon closure to new business.
Impact / risks:
While this is not realistic from a real-world perspective, it has the benefit of making the modelling simpler.
The risk of such an assumption is the creation of a solvency or liquidity problem that in practice would not exist if costs were spread across the run-off.
To manage this, insurers should ensure that enough time is set aside for iterating modelling if the assumptions need to be changed once initial results are produced.
The above is just a selection of examples of simplifying assumptions and there may be many more. In general, firms may choose to use simplifying assumptions where these represent conservatism but should be more careful where the potential impacts could have a material financial implication and/or make solvent exit less certain.
Where possible an insurer should review all of its assumptions and the potential for the outcome to be different to those modelled, clearly documenting the rationale for its choices and the associated sensitivity of the overall solvent exit to the choices made.
While some will see the requirements to perform solvent exit analysis as an additional compliance burden, reports from those that undertook such analysis early (either at the request of the PRA or as part of broader resolution planning) suggest that there can be significant value gained from the underlying analysis.
Where analysis is new or likely to provide valuable insight into the business model, business dynamics, cost base, or complexity of operations, insurers may choose to dive deeper whereas they may not if the analysis has been previously performed or is already well understood.
Some examples of where firms have generated lasting business value include:
Approaching the analysis with an open mind will ensure that insurers are best placed to maximise value from the exercise.
The PRA has been clear that firms should leverage existing analysis where possible and ensure that SEA fits within a harmonised suite of analysis including ORSA reporting, stress and scenario testing, recovery planning and resolution considerations.
Since the 2008 financial crisis there has been significant effort from insurers to increase financial resilience and develop deep understanding of how the business performs under stress. This body of work is a great starting point for developing the point of non-viability (for example with reference to existing stress testing and reverse stress testing) and for thinking through the actions available and associated impacts both before and after closing to new business (for example with reference to existing recovery planning actions).
Where firms already have a resolution plan, some of the analysis needed for SEA may already have been completed or could be efficiently tailored. However, consideration should be given to how solvent exit differs from the assumptions underpinning it (for example, the circumstances that may lead to a management led exit versus a disorderly Bank of England/PRA led exit).
One efficient way to leverage existing information is to cross-refer to said information within the SEA document as opposed to reperforming or regurgitating the analysis. This is particularly pertinent for contextual information that is important for understanding the business model, operating model and intragroup connectivity which underpins SEA analysis but likely already exists.
For smaller insurers, the SEA may become part of an existing document such as the ORSA or recovery plan. For larger insurers, it is more likely that the SEA will be a standalone document unless an existing resolution plan already exists (in which case there is a choice as to whether this remains two documents or whether SEA and resolution planning become part of the same set of analysis).
When the ORSA was introduced during Solvency II, the concept of proportionality was central to firms’ development (and the subsequent evolution) of analysis over time. Early ORSA reports tended to be long documents with comprehensive analysis.
However, over time they have shortened since boards and executive teams have become comfortable with the type of analysis included. The working group expects this will also be the case for SEA, with a period of longer, more detailed documents that become shorter and more focused over time.
Proportionality does not allow practitioners to simply ignore parts of the regulation but allows for simpler approaches for less material and less complex analyses. When practitioners are thinking about applying proportionality, they should have regard for the considerations set out above and be prepared to flex the approach if the analysis highlights a risk or complexity that was previously unforeseen or expected to be less material.
Beyond anything practitioners should be prepared to explain why the approach taken is appropriate and those charged with governance should ensure appropriate robust challenge of this rationale and flag any areas where they feel that a more detailed analysis is required.
