By Tim Edwards, Director, Insurance and Investment Management, PwC
So, how should proportionality apply in practice? A smart answer is that it comes down to the “nature, scale and complexity” of a risk or issue, and thus to the amount of effort that is required to demonstrate that the various requirements are met as a result.
1. Scale – this is the easiest one. The bigger something is, so the bigger the likely impact if it goes wrong. That can apply to a company overall, or to a line of business or a type of risk within a company. You are more likely to devote time and effort to managing and modelling major lines than smaller lines, and the amount of data to be worked with is of course bigger too.
2. Nature – this relates to the inherent riskiness of the business or the firm. A risk or line of business might be big but have very low volatility, and so is unlikely to have a significant impact on the viability of the firm, even in the event of a problem. You are likely to need to – and choose to - spend more time managing and modelling high risk business that could impact the firm.
3. Complexity – if a risk or issue is easy to understand, or has tried-and-tested management, mitigation and/or modelling techniques, then it has relatively low complexity. If, however, it is unusual, or hard to model, or difficult to describe, then comparatively more effort will be required to quantify and manage it.
Put simply, if a line of business or type of risk is big, volatile and hard to model, then of course you will spend a lot more time on it than if it were small, low risk and easy to quantify or manage.
How, though, do you know when “enough is enough”?
Every firm’s resources are finite – so it is a matter of applying resources in the most appropriate way. Hard decisions may need to be made – just because your technical experts find something fascinating, it does not justify them spending a disproportionate amount of time on it. To help you to prioritise, I suggest you pose a few questions:
• Which are your major risks or issues?
• Which justify most discussion time at board level?
• Which have been agreed as the main concerns when you last met the PRA?
• Which are the main risks in your risk register?
• Which lines of business generate most premium income, technical provisions or profit?
• Are there any modelling approaches that are a bit untested, or harder to justify?
• Are there sensitivities to key assumptions or judgements?
• Is there any aspect in your business that has caused recent adverse headlines elsewhere, maybe to another firm, such that you’ll need to take care to demonstrate you are not or will not be impacted?
But to show that proportionality is not a one-way street, you could equally turn each question around. I won’t, for the sake of space – but for example “Are there any modelling approaches that are well understood, tried, tested and proven?” or “Which lines of business generate least premium income (etc.)”.
And then - do you really need to use five people for ten days each to model something that could be done adequately by one person over three days? Remember that the more effort you put into the modelling, so the more documentation you will be producing and the more validation effort will be needed.
Proportionality – it’s important to know when enough is enough. But also, of course, to be able to justify it.
|
