Use Test: Challenges and Opportunities

Christopher Chappell, Elliot Varnell, and Coomaren Vencatasawmy

The use test is arguably the most important yet most challenging of the requirements that European supervisors have developed for the Solvency II internal model. The 2007–08 financial crisis highlighted the fallibility of internal models, and much has been written about the way risks were wrongly assessed and how early warnings were ignored because of internal models. There has also been a lot of discussion about the use of Gaussian copulas, and how these might have contributed to the financial crisis (Salmon, 2009). It should be recognised, however, that all models are wrong. We are very far from internal models that can closely replicate all real-life conditions, especially emerging ones. Therefore, saying that models caused the crisis is a moot point. The crisis was caused just as much by decision-makers and by how outputs from the models were used. This is why the use test is so crucial.

Internal models’ failings have been mostly about how the models have been used within the decision process of financial institutions. Most of the issues related to failures to understand the weaknesses of the models used to assess risks and how the models were used in decision-making. As