A practical methodology to optimise marginal mineral deposits using switching real options

Currently depressed commodity prices have rendered many mining projects marginal irrespective of their geological merit. Tight capital markets discourage investment in their development because of their unappealing deterministic NPVs, which in the majority of cases reflect conceptual designs focused on achieving primarily economies of scale often at the expenses of operating flexibility. Given that project profits and cash flows are highly sensitive to movements in volatile commodity prices, circumstances now call for a re-direction of emphasis towards creating managerial flexibility to facilitate and minimize the cost of temporarily placing projects in care and maintenance and re-opening them in response to increases in prices. This flexibility, that is to say the option to alternatively switch the project between an open and closed state, can be created through an appropriate combination of mine design, commercial procurement arrangements and mode of operations that enables managers to anticipate and take advantage of future hikes in prices, while minimizing the negative effect of downturns. This paper presents a practical example of how to estimate the real option value (ROV) of this type of switching option, which is generally not captured by the deterministic DCF/NPV of projects. To facilitate the numerical presentation, initially the binomial lattice method is applied only to the first 2 years of a realistic DCF model of a gold mine, with an expected life of 5 years and a negative deterministic NPV. The model is limited to assessing the ROV created by introducing switching flexibility as a result of the volatility of the gold price in isolation. A consistent ROV is then obtained using as an alternative the unrelated decision tree methodology. This result is considered important as using decision trees for this type of analyses in cases where more than one source of uncertainty is involved (e.g. that of grades, costs, and exchange rates) does not require, as in the case of binomial lattices, estimating the volatility of a project cash flow. This process, which may create computational ambiguity and possible bias, can be avoided in decision trees as each source of uncertainty is represented by an individual event node. Finally the ROV of the project, including the switching option, is calculated over its whole 5-year life to provide some indication of the amount that could justifiably be invested up-front to create the necessary switching flexibility.