This white paper demonstrates that an investor is able to evaluate bond SCR by using only two variables, residual maturity and rating, rather than the nine initially tested. Moreover, the correlation between real credit spread and SCR is not high. This is due to the flat-rate treatment of spread risk under Solvency II (which assigns a single risk factor to each rating and does not account for internal variances in ratings). Given the additional marginal cost that could be considered excessive in proportion to the return generated, bonds rated BBB or lower could end up being neglected by investors, potentially resulting in significant implications for the financing needs of the economy.
We show that SCR - as defined by the standard formula - is, overall, an appropriate measure of risk. However, given their specificities, SCR does not fully reflect the risk associated with long-maturity investment grade bonds, high yield and unrated bonds. Due to the features of bond SCR (high correlation with volatility and historical VaR), bond management currently based on the return-VaR-volatility triple factor should evolve under Solvency II, more towards a management approach based solely on the bond return-SCR pair. Finally, an analysis of the efficiency of risk-taking measured by the bond return/SCR ratio shows that the standard formula favours low duration bonds, particularly high yield bonds. This management, within the constraints of SCR, could lead to a shortening of durations, given the calibration and current term structure of interest rates.