Insurers are forging ahead with the development of their internal model change policies, despite uncertainty about what alterations to the model will trigger supervisory intervention.
A key consideration for firms in the development of their internal systems is the model change policy, which explains the procedure by which they are allowed to alter their models to reflect improved calculation techniques. Insurers are required to identify what counts as a major change, and what counts as a minor change in their applications for internal model approval.
Major changes, such as to the key calculation kernel of the model, will oblige firms to submit the change for approval from supervisors before the updated model can be used for computing the solvency capital requirement (SCR). Minor changes will only need to be reported to supervisors and will not need approval.
But existing guidance on what can be categorised as major or minor change is sketchy at best, say insurers, raising the possibility that supervisors will be required to approve many changes that initially were not expected to be material. Changes to model parameters, assumptions and even usage could all potentially trigger supervisory review.
Supervisors say that with the finalisation of Solvency II still some time away, it is difficult to judge how the categorisation of major and minor changes in a firm's internal model policy will be assessed.
Despite the uncertainty, some firms are already taking steps to design change policies for all contingencies. Firms, say consultants, are experimenting with a number of ways of defining major changes. One method involves setting a quantitative measure at a certain percentage of the SCR calculated by the model. If this changes as a result of altering the model, then this would trigger an internal appraisal of the model by the firm's risk committee to determine whether a major change has occurred. A second method uses a qualitative trigger, whereby changes are analysed to check whether they materially influence the use of the model and are then reported upstream.
Yannis Pitaras, head of prudential regulation at Insurance Europe in Brussels, says: "Model governance and model approval processes need to be kept manageable, because internal models should be encouraged and not made so burdensome that supervisors cannot cope with the process or companies decide it is not worth the cost".
Full article (Risk.net subscription required)
© Risk.net
Key
Hover over the blue highlighted
text to view the acronym meaning
Hover
over these icons for more information
Comments:
No Comments for this Article