How Should Mining Companies Select the Optimal Portfolio of Production Projects Considering the Risk of their NPV

Consider the case of a mining company with N investment opportunities in projects of different commodities. The problem is: since the company has limitation of capital, equipments, technical staffs, etc, what is the best portfolio of production projects in order to maximize expectations from shareholders? In the no-longer past, the traditional method for the selection of portfolio of projects was the rank-an-cut method. But, this does not properly account for the risk to which the company is exposed. The Markowitz?s Mean-Variance is an advance because the risk of the NPV can be estimated at the level of the company or even division of the corporation. The final product of the Markowitz?s model is the efficient frontier, in which, the manager can clearly understand the maximum return from existing assets and the associated risk level - but this model does not recommend which one is the best portfolio. In this paper we present a model to select the best portfolio considering not only economic indicators, but also requirements of production, constraint in investments, etc. We show that if investors are diversified, managers should select projects with highest return, regardless of risk. However, in practice, if bonuses of managers are tied to their performance, the choice for portfolio with lower return and risk can happen. Numerical application considering typical portfolio of large mining company is used as example.