17 May 2010

Published in: Regulation - supervision

"UK Life" solvency ratio falls under QIS5

Milliman's scenario-testing using a notional life insurer reveals a likely decline in the standard formula solvency ratio, worsened by the fall-out from the financial crisis. Scott Mitchell and Fred Vosvenieks explain.

We created a notional life insurer, "UK Life", to give focus to our research into the impact of the draft fifth quantitative impact study (QIS 5) technical specifications on European life insurers' solvency positions under the standard formula. The specifications, published by the European Commission on 15 April, provide a clear indication of how the standard formula is likely to be set for the QIS5 exercise.

The following analysis is based on the 2007 and 2008 year-end positions of UK life insurers. We acknowledge that the QIS 5 exercise will be based on the 2009 year-end but, at the time of writing, complete data for 2009 was not yet available.

UK Life's opening balance sheet (as at 31 December 2007) represents an accumulation of the data available on the UK life insurance market at that date.

The study then considered the impact on UK Life's opening solvency position under QIS4 of the various changes likely to be implied by the new standard formula specification within the coming QIS5 exercise.

We performed this analysis taking into account both:

1. developments in the economic environment (over the period 31 December 2007 to 31 December 2008), and
2. development of certain components of the QIS standard formula for calculation of the solvency capital requirement (SCR) under Solvency II.

For (b) we considered the development of the parameters and methodology of the standard formula inferred from our reading of the draft QIS5 technical specification.

Given the nature of this analysis and some uncertainty over the detail of the final QIS5 technical specification, we note that the analysis is subject to several limitations, in particular:

• Most of the market data used in our analysis is based on publicly-available information (e.g. statutory accounts).
• The non-economic assumptions used are based on our knowledge of the market and a desire to reconcile our results as at 31 December 2007 with those published in respect of QIS4.
• The methodologies used have been set by our team of consultants. Other, possibly very different, approaches could also have been considered.
• The main objective of the study is to consider the impact relative to a known initial situation (i.e. the solvency position under QIS4 at 31 December 2007). Analysis of a figure taken in isolation could lead to incorrect interpretations.

Whilst we expect the main conclusions to remain valid, the results presented here should be therefore interpreted with the above points in mind.

Moving from QIS4 to QIS5

Recent development of key aspects of Solvency II has been driven by two "waves" of consultations and associated CEIOPS' final advice since last summer, and has culminated in publication of the draft QIS5 technical specifications:

• second wave
  • consultation papers: 2 July 2009
  • final advice: 10 November 2009
• third wave
  • consultation papers: 2 November 2009
  • final advice: 29 January 2010
• draft QIS5 technical specifications
  • QIS5 specifications: 15 April 2010

Our study focuses on certain key differences that we observe between QIS4 and the draft QIS5 standard formula, as illustrated in the following flowchart showing the steps in the sequence of analysis performed for UK Life:

The QIS4 solvency ratio of UK Life as at 31 December 2007 was 134%, in line with the observed QIS4 position published by CEIOPS and the FSA for the UK life market.

The following chart then illustrates the relative changes observed in the process of moving stepwise from this base QIS4 solvency position for UK Life to that as at 31 December 2008 under the draft QIS5 standard formula specification.

Under the QIS 4 parameterization the solvency ratio for 31 December 2008 declined from 134% to 125% as a result of the significant market falls observed over 2008. Incorporating the changes found in the draft QIS 5 technical specifications led to a further fall in solvency to 121%. This illustrates and confirms an overall tightening of the standard formula specification (although this has been scaled back in some respects from the proposals of CEIOPS' final advice).

Some key observations arising from the above analysis are:

• As expected, the global financial crisis has clearly impacted UK Life, as can be seen by the reduced solvency position one year after the QIS4 exercise as at 31 December 2007.
• The inclusion of an illiquidity premium in the valuation of best estimate liabilities has a significant impact on the Solvency II balance sheet. The large immediate annuity portfolios present within the UK market benefit considerably since their valuation can take credit for 100% of the illiquidity premium, but other business (except very short-term contracts) can also benefit from 50% of this premium.
• Despite a slight weakening in the stress to equity levels (after allowing for the dampener effect), the overall strengthening in the individual market sub-modules leads to a significant fall in the estimated solvency ratio. This is driven mainly by the introduction of the stresses to volatility for interest rates and equity and the change to the credit spread sub-module, which now also captures a change in the illiquidity premium.
• In our analysis, the benefit from the introduction of the illiquidity premium is then further diminished by the strengthening in the correlation coefficients used to aggregate the individual components of the market risk module.
• There were also changes to the sub-modules used to stress mortality and disability risk in the life underwriting component of the SCR, but, after allowing for diversification effects, these were not significant for our sample company.
• Similarly the SCR in respect of operational risk is somewhat more conservative under QIS5.
• While the classification of subordinated debt as tier 3 capital had no impact on our sample company's solvency ratio, the restriction to 15% of eligible own funds for the purposes of demonstrating solvency under the new regime may be significant for some companies.

QIS5 standard formula solvency ratios decline

Our study shows that the draft technical specification for QIS5 is likely to mean a decline in the QIS standard formula solvency ratios observed for UK life companies, because of an overall strengthening of the stresses applied to calculate the SCR outweighing the effect of uplifted risk-free rates.

The study also shows how this picture of declining QIS solvency has been exacerbated by the effects of the global financial crisis, with a fall from the year-end 2007 QIS4 levels.

Our analysis also appears to confirm that the end-2008 UK Life balance sheet, while looking worse under QIS5 than QIS4, seems to have been structured rather more robustly than that as at end-2007, which according to our calculations would have shown much worse solvency under QIS5 when compared to QIS4.

Our analysis has aimed to illustrate the possible effect on a hypothetical UK life company -- UK Life -- constructed to represent the life market as a whole and calibrated to the QIS4 results observed for the UK market as at the end of 2007. However it is important to note that this formulation is not necessarily typical of any given UK life company and the likely effect on a company's solvency ratio will vary depending on the specific circumstances of each life insurer.

Our research has been performed for a number of key European markets and an accompanying research report will soon be available that presents the background, results and analysis.

Scott Mitchell is a consultant in Milliman's Zurich office scott.mitchell@milliman.com
Fred Vosvenieks is a consultant in Milliman's London office fred.vosvenieks@milliman.com

Back to top