^aAustrian Financial Market Authority, Praterstrasse 23, Vienna 1020, Austria. E-mail: patrick.darlap@fma.gv.at

^*The views expressed in this paper are those of the authors and do not necessarily reflect the opinion of the institution they represent.

Abstract

This article discusses group aspects that may be expected to be important determinants of regulatory reform in the insurance sector. Basically, we deal with the question of whether the risk level of a group is higher or lower than the simple sum of risks of all institutions individually. The application of simple portfolio theory to explain overall group risks may be myopic. A holistic approach to assess group (concentration) risks also has to cover risks that portfolio theory does not take into account, for example, reputational effects, contagion, etc. In fact, the diversification gains from a large group portfolio may well be outweighed by the risks encountered at group level. In order to assess group risks, we advocate the use of copula-based economic capital models. We believe that this procedure allows a more adequate judgement on group risks than traditional methods, which are generally based on rather restrictive assumptions. Data constraints, however, may impair an early implementation of such models. An early start of data collection for adequate modelling and further research on copulas have to be understood as prerequisites to advanced solvency determination in groups.