Bond portfolio management under Solvency II regulation

We develop a novel approach to the bond portfolio optimization for insurance companies that are subject to the new Solvency II regulation. The regulatory efficient portfolios are determined using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The characteristics of the estimated efficient portfolios are examined in different market regimes. Our findings suggest low cardinality of all estimated efficient portfolios despite explicit regulatory penalties for highly concentrated portfolios. The efficient portfolios are dominated by short term and BBB rated bonds. The lack of diversification and over-exposure to bonds with higher credit risk in different market regimes represents a weakness of the Solvency II regulation with unintended consequences for management of insurance companies.