Sector-specific optimum asset allocation — An example for non-life insurers

Abstract

Quantitative asset allocation models often fail to take account of the many explicit and implicit constraints that asset managers are subject to. When optimising the portfolio of a given investor, this problem can be circumvented by looking at the allocation of a peer group as a reference. The key advantage of such an approach is that the allocation of the peer group reflects the allocation constraints of the industry the investor is acting in. Given the allocation of the peer group, the investor’s allocation can still differ from that of the industry: the more the investor’s risk appetite differs from that of the peer group, the more his portfolio will diverge from the allocation of his competitors. Another way for the investor to deviate from the peer group allocation is to formulate tactical forecasts regarding asset performances with corresponding confidence levels. This is made possible by Black/Litterman’s model, on which our approach is based. In this paper, we describe our model using the example of a German non-life insurer whose peer group consists of its own competitors. But our approach could as well be used by a fund manager in the UK, a saving bank treasurer in the US or more generally by any investor knowing the allocation of its competitors.