Allocation of Systemic Risk

We critically review the translation of the core originating from game theory to the allocation of systemic risk. Whereas the core is commonly accepted for the portfolio allocation we will see that in a systemic context it might result in unfairnesses for certain members of the system. We observe that due to the presence of possible feedback mechanisms between the single entities, apart from the height of the induced losses, also the ability to transfer these losses has to be considered for the risk allocation problem. Since this new source of risk is only fully assessable in the complete system and not in the subsystems, it can not be captured by a core allocation. Thus, instead of considering upper bounds, as it is done for allocations in the core, we reverse the definition to lower bounds.