IBM’s view on Solvency II

Regulators are not simply asking insurers to come up with a new number for the Solvency Capital Requirement (SCR). During accreditation they will look to get answers to a number of questions: Is the SCR an unbiased estimate of the risk? Are the data and methodology sound and sufficiently reliable? Is the actuarial model actually used within risk management? Is risk management objective, with a proper separation of roles? Are the models and processes supported by the right level of documents? The calculation of the SCR will drive larger insurance firms to the adoption of an internal risk model, because the standardised approach will be conservative so that firms using it will stand at a competitive disadvantage. Also, the first companies to introduce internal risk models will benefit most from better capabilities in risk based pricing and pricevolume optimisation. Solvency will trigger a reorientation of controlling, as it enables steering based on a risk and value based management approach. Although not all regulations are finalised,insurers cannot afford to wait because assuming (1) compliancy in 2012 (in line with the current timelines), (2) needing two+ years running for an internal model to produce significant results, before 2009 the internal risk model and data integration needs to be implemented for all units within the company. And one should not underestimate the time required to achieve this: Many banks needed for years after the start of the implementation to reach Basel II compliance. Add up the expected shortage of skills: Putting together a team for Solvency II requires a broad range of specialist skills – in actuarial modeling, project management and risk management. Insurers will all be approaching this together – at critical stages of the Basel implementation banks found that the market for specialists completely dried up, causing additional delays for the banks that had started later. Many Basel implementers have commented that if they were to do Basel again they would invest in more technical specialists who understand IT, to bridge the gap between complex highly technical and statistical specifications and pragmatic IT implementation. The chart below shows an example of the breakdown structure of costs associated with Basel II implementation, illustrating that most of the costs are related to data integration. We expect a very similar situation for Solvency II, as the breadth and depth of data requirements will be at least equally challenging. Additional data will be needed for the risk capital calculation. For example, the SCR is based on the calculation of the market value of liabilities which is significantly different from the calculation of statutory reserves and requires different data sets. Also, this data will come from varied sources.

Data integrity will become ever more important as the high level of confidence chosen for Solvency II risk capital calculations implies a high sensitivity to outliers. In particular, an insurance company striving for an internal model would require a risk engine which supports extensive sensitivity analysis.

Note: FO/BO stands for Front-Office and Back-Office.

All of the well-established principles of business intelligence, data warehousing and data governance completely apply to Solvency II and a well-known success factor of such initiatives is the availability of a robust, enterprisewide data model reflecting the common processes and product types rather than the existing (legacy driven) data storage implementation. While such models can certainly be developed in-house, it makes economical sense to leverage industry efforts instead of reinventing the wheel since a welldesigned industry model can cover as much as 85% of any enterprise’s data. Such models also have the added advantage of taking a broader view of requirements that can be designed at the level of an enterprise. Typically an enterprise data warehouse will be supported by a dedicated interface which both controls the quality of data and consolidates data from several sources (see schematic chart below). The data warehouse will support in a consistent manner the requirements of different business users and business applications, including development of the internal risk model and risk capital calculation, Value Based Management (VBM), Investment Planning and Balance Sheet preparation. The data warehouse (“one version of the truth”) will feed the risk engine with consistent data sets. The risk engine (eg, VIPitech, EMB IGLOO, DFA Advice) will then support the development of the internal risk model, as well as simulation, reinsurance optimization, asset allocation, and so on. High data quality is characterised by a unique data format and standardized data transfer procedures. Based on the inventory and the grouping of data of the legacy systems on the one hand, and the corporate data model on the other hand, the creation of an unified interface between the legacy systems and the corporate model would be a crucial task. As the physical data systems of the company evolve, this interface would be continuously updated. It would therefore need to be clearly documented and to include quality criteria and consistency checks. The unified interface could apply to the corporate business unit and to all individual business units.

IBM´s reference datawarehouse IIW V8.0 can jumpstart a Solvency II implementation because it contains Business Solution Templates that represent in a structured way the Solvency II (and IAS and IFRS4) reporting requirements. It also contains Business Solution Templates covering the Key Risk Indicators (KRIs) (eg, Economic Capital, RAROC, European Embedded Value) and risk analysis areas (eg, insurance risk, financial risk, credit risk, liquidity risk, operational risk). Moreover IIW contains over 2000 well defined business terms and an insurance specific reference data model covering up to 80-90% of insurer´s business requirements. Amongst other due to the IIW metadata management (complete and accurate data definitions and traceability between business level and IT level artifacts) and tooling (Multi-Model Mapper) all IIW models can be easily customized to accommodate specific business requirements. IBM Solvency II Practice has been very active in the last two years, particularly in the German insurance market, with the implementation of Regulatory Compliance (eg, Solvency II, IFRS 4) and Enterprise Risk Management (ERM) projects. Also, IBM has developed business partnerships with specialized Actuarial & Risk Management firms with deep expertise in the development of insurance risk models and Solvency II compliance.