As discussed in the Executive Summary of this report, the term “crystalization of risk” refers to the timescale over which we realize that the risk is manifesting itself and how this view changes until ultimate understanding of quantum is reached and all liabilities are discharged. The “Reserving Risks” section in last year’s report, Ahead of the Curve: Understanding Emerging Risks looked at how information emerges in the presence of reserving cycles. The profit or loss in any particular financial year is made up of not only the profit or loss from the same accident year but also any recognized changes in the reserves on prior years.
A big movement in prior year reserves in any one financial year particularly from an unanticipated source of risk can cause a sharp drop in share price and an increase in regulatory scrutiny. In addition a small drip, drip of increasing reserves in consecutive financial years can also make shareholders, analysts and regulators nervous about where the deterioration will stop and question whether they can have any confidence in a company’s reserving process.
Such reserving increases can certainly erode the available surplus but should they also lead to a reexamination of the required capital, as they can be an additional item of information that would have changed the perspective at the time the capital was set? This is a somewhat philosophical question; a company can set capital aside initially to protect against extreme eventualities akin to a certain amount of water in a bucket, then someone trips and spills half the contents of the bucket due to something everyone thought was unlikely to happen. Does it make one think there should have been a bigger bucket when one sees what is left?
In a Solvency II world, capital (the solvency capital requirement (SCR)) is set as the change in a firm’s own funds (assets less liabilities) at a 99.5% confidence level over a 1-year time horizon. Setting this out as an equation, it looks like:
SCR = Change (A - L) @ VaR 99.5%
Expanding the L (in Solvency II, liabilities are actually the sum of the Best Estimate and the Risk Margin), the equation becomes:
SCR = Change (A - (BE +RM)) @VaR 99.5%
The Risk Margin is actually defined as an amount that a third party would require in order to accept the transfer of the liabilities over and above the best estimate. This is effectively the cost of the capital required to run-off the portfolio (i.e. the sum of the future SCRs calculated at each future point in time until the portfolio is fully run-off and discounted back at the risk-free rate).
So, roughly translated, the SCR is actually a reflection of how much surplus can change in a year, which in turn, is influenced by a firm’s view of risk at each point in time. For emerging risks that view can change fast, and presents issues as to how it should be reflected in the calculation of the risk margin and hence the SCR.
The discounting implicit in the Risk Margin calculation can be influential for emerging risks as well. What if a known or suspected risk is long-tailed but there is uncertainty in the way that it is likely to run-off? The model suggests that the ultimate 1-in-200-year loss is 100. However, it does not show either when that 100 could manifest itself or what inflationary influences may act on that 100 if it is presented in today’s values but may settle way into the future.