Vigilant measures of risk and the demand for contingent claims (Q2347093)

The authors study a class of utility maximization problems with a risk measure constraint \(\rho(Y)\leq M\) on the claim \(Y=I\circ X\) contingent on the underlying random variable \(X\) where the risk measure \(\rho\) satisfies a consistency property called vigilance and a continuity property named weak dominated convergence property (weak DC-property). The main result of the paper is that supermodularity of the utility function \(U:R^2\rightarrow R\) is sufficient for the optimal choise of the contingent claim \(Y=I\circ X\) to be a non-decreasing function of the random variable \(X\). As an illustration, the authors consider a problem of optimal insurance design where the premium principle satisfies the vigilance property, including premium principles which are not law-invariant. The existence of optimal indemnity schedules is proven as well as the fact that they are nondecreasing functions of the insurable loss.