In its series of Consultation Papers on Level 2 implementation Measures for Solvency II, the Committee of European Insurance and Occupational Pensions Supervisors (CEIOPS) drafted, in Consultation Paper 71, a new proposal for the calibration of non-life underwriting risk. Additionally, CEIOPS published its final and third set of advice to the European Commission (EC) at the end of January 2010. It is noteworthy that non-life underwriting risk is not covered in this final advice. Thus, the calibration changes suggested in CP 71, which would lead to an average of a 35 percent increase in the SCR for non-life underwriting, are still valid. It should be noted that the proposals in the CP are subject to a consultation process resulting in final recommendations by the end of March 2010 and therefore may not be final. The purpose of this briefing is to outline the rationale provided by CEIOPS behind the proposed increase in the SCR in respect of non-life underwriting risk.
In view of the criticism received in reaction to Quantitative Impact Study (QIS) 4, CEIOPS performed a thorough analysis of the underwriting risk factors. It analyzed data from six member states, which constitute a broader sample than the one used for the QIS 4 calibration. While probably not sufficient for a rigorous statistical analysis, this sample includes a broad representation, with participants from two large states (Germany and UK), two medium states (Poland and Portugal) and two small states (Luxemburg and Slovenia).
The data items collected for the analysis were standard net premium and claim data for each company at the line of business and accident year levels. Based on this data set, CEIOPS calculated the standard deviations for each line of business by means of six methods for reserve risk. For premium risk four methods for the standard deviation calculation were used with an additional analysis of the loss ratio time series which participants submitted with their QIS 4 calculations. The methods used are in accordance with QIS 3, QIS 4 and corresponding variations.
The results of this analysis vary substantially depending on the method, the size of the company and the quality of the data submitted. CEIOPS interpreted the results and made a final factor selection allowing some degree of conservatism. The interpretation also reflected the fact that the smaller companies in the sample (with less weight in the analysis) tend to not benefit from the same diversification credit as the larger ones. Interpretation of the results is not straightforward and the final factor selection does not demonstrate a clear link to the outcome of the analysis.
Comments received by CEIOPS
Participants expressed thanks that CEIOPS published its analysis explaining the rationale behind the increase. Participants agree that these kinds of communications should be continued.
The general view is that the already conservative QIS 4 factors have been made even more conservative. Most participants commenting on Consultation Paper 71 believe that the methods are not always consistent with Solvency II principles. The approach used by CEIOPS in the factor final selection seems quite arbitrary and, at a minimum, lacks transparency. Commenters observe that the final factors are higher than the levels supported by the analysis. There is a broad consensus that the gap between the results of the analysis and the final selections is not justified.
Some comments indicate that the analyzed data are not consistent in the first place:
- Data quality is arguable, in that the data are not on an economic basis as stipulated by Solvency II, but rather on a local statutory basis.
- Scarcity and lack of geographical diversification is not representative of the EU market.
- Catastrophic events have potentially been double counted.
- Consideration of reinsurance changes over time and underwriting cycles have been excluded.
Performing the analysis on a more consistent data basis in which the underwriting cycle and catastrophic events are removed from the data would substantially reduce the volatility parameters, whichever method is used to derive them.
Some participants reiterated that a more adequate view of the risk landscape would be achieved through estimation of the gross volatility and calculation of the net volatility by incorporation of the actual reinsurance structure. Similarly, some regret that CEIOPS did not reintroduce the allowance for the expected profit, dropped after QIS 2.
The high volatility parameters should be an incentive for undertakings to apply for a (partial) internal model. Others call for a broader flexibility in the use of undertaking-specific parameters or at the very least, different parameters depending on the size of the company. The average factor will usually be too high for large companies and too low for small companies.
In addition to the calibration of the premium and reserve risk parameters, in CP 71 proposals CEIOPS also made several changes to the calculations of the capital charge for catastrophe risk. Companies may quantify their individual catastrophe risk charge as part of the standard formula or as part of an internal model. In the standard formula, CEIOPS proposes either a standardised scenario approach similar to QIS 4, but with revised scenarios harmonised among countries, or a factor method where the gross catastrophe risk is split by peril and line of business affected. A more detailed analysis of the cat risk proposal can be found in “Higher Pressure on Cat Risk under Solvency II”
CP 71 only deals with the calibration of the underwriting risk module. Adjustments in the correlations among lines of business are addressed in CP 74. Results demonstrate a 7 percent increase in SCR. This is a dramatically smaller figure than the result of the new calibration of underwriting parameters, but obviously comes in addition to the other increases.
In light of the many - mainly critical - comments received, CEIOPS may reconsider the calibration within the final advice it provides to the European Commission. If implemented unchanged, the effect of the increased capital requirement for underwriting risk coupled with similar increases for other risk classes may trigger a wave of capital raising and/or the increased use of internal models.
This internal briefing was prepared by Guy Carpenter’s Financial Intelligence Team (FIT). Questions regarding this briefing may be directed to any of the FIT group members, listed below.
Susan Witcraft, Managing Director, Minneapolis +1 952.832.2143
Frank Achtert, Managing Director, Munich +49 89.28.66.03.361
Eddy Vanbeneden, Managing Director, Brussels +32 2.674.98.11
David Flandro, Senior Vice President, London +44 (0)20 7357 3267
Benoît Butel, Vice President +33 188.8.131.52.26
Sebastien Portmann, Vice President, Zurich +41 44.285.9322
Florent Scarabin, Vice President, Stockholm +46 8 505.73549
The information contained in this internal briefing is proprietary and confidential and is intended for the use of employees of Guy Carpenter & Company, LLC and its affiliates (collectively, “Guy Carpenter”). If you are not an employee of Guy Carpenter, you are hereby notified that any disclosure, copying, distribution or the taking of any action in reliance on the contents of this internal briefing is strictly prohibited. If you have received this internal briefing in error, please immediately notify Guy Carpenter to arrange for return of the original documents to us.
The data contained herein is provided for information purposes only. Guy Carpenter makes no representations or warranties as to the accuracy of such information nor is it obligated to update such information. The information is not intended to be taken as advice with respect to any individual situation and cannot be relied upon as such.
Statements concerning, tax, accounting, legal or regulatory matters should be understood to be general observations based solely on our experience as reinsurance brokers and risk consultants, and may not be relied upon as tax, accounting, legal or regulatory advice which we are not authorized to provide. To the extent that you discuss such statements with your clients, be sure to advise your clients that all such matters should be reviewed with their own qualified advisors in these areas.
The trademarks and service marks contained herein are the property of their respective owners.