19 February 2014

Risk.net: UK insurers' internal model validation processes 'hugely inefficient'

UK insurers are wasting time and resources with inefficient model validation processes that risk underestimating the level of risk inherent in their Solvency II internal models, experts say.

A slew of recent surveys have highlighted the significant challenges insurers are facing in creating a validation framework that is both cost-effective and suitably tailored to their model risk.

A survey of Lloyd's of London insurers by actuarial consultancy Lane, Clark and Peacock, published last month, found that half the companies surveyed said modeling the dependencies in their validation process to be a challenge. In addition, one-third (32%) of companies said setting and managing pass/fail criteria was challenging, while 29% highlighted validating expert judgment as a key area of difficulty, the survey also revealed.

Consultants say the results demonstrate that insurers have yet to develop streamlined validation processes, despite an abundance of guidance provided by the Prudential Regulation Authority (PRA).

Paradoxically, this in turn creates the risk that the model being validated is not subject to the appropriate level of scrutiny and senior oversight, he adds.

Model validation is one of six criteria an insurer must fulfill to pass the PRA's internal model approval process (Imap). The European Insurance and Occupational Pensions Authority (Eiopa) describes validation as "a set of tools and processes used by the undertaking to gain confidence over the results, design, workings and other processes within the internal model".

Once the PRA is given powers to approve models at the date of Solvency II's transposition into national law (currently scheduled for March 31, 2015), all UK firms hoping to use an internal model will need to submit a formal application to the PRA for a decision ahead of Solvency II's implementation date.

If a firm's model does not meet the requirements when the approval decision is made, it will go to the back of the queue in order to provide time to remedy existing issues.

Insurers report that an additional problem is effectively validating third-party models, such as economic scenario generators and catastrophe models bought from external providers.

Full article (Risk.net subscription required)

© Risk.net