ABSTRACT
This chapter contributes to the first line. We argue that the cause for the potential contradiction between the theoretical recommendation of quite sensitive portfolio weights and the practical finding of their comparatively high stability does not lie in properties of the Markowitz portfolio theory in itself. In contrast, this contradiction can be resolved, when return expectations are treated as endogenous in mean-variance optimization. Endogenous returns are then modeled on the basis of an inverted dividend discount model. Mean-variance optimization can then account for changes in return expectations that are either common knowledge (endogenous expectations) or insider knowledge (exogenous expectations). This should have a different impact on both the investor’s optimal portfolio and the market portfolio. In addition to these two extreme cases, there might be situations termed as mixed cases, with the investor’s expectations being only partially reflected by corresponding changes in dividend expectations of the capital market as a whole.
