Incorporating technical uncertainty in real option valuation of oil projects

Abstract

Real options are a way of valuing projects such as oil fields, which involve irreversible investment decisions subject to uncertainty. Whereas discounted cash flow (DCF) analysis is based on fixed estimates of costs and revenues, and a predetermined development scenario, real options focus on project flexibility (e.g., being able to defer starting a project, or conversely, to accelerate its development, etc.). Early applications concentrated on flexibility to overcome uncertainty on financial parameters and tended to ignore uncertainty on technical parameters (such as recoverable reserves). In some cases, companies can reduce the latter by acquiring additional information and sequentially updating parameter values and then the estimate of the project's value. This paper addresses the question of how to evaluate the option to acquire more information. The natural way to incorporate the new data is by Bayesian analysis. Choosing a multivariate normal framework for this analysis considerably simplifies the computations but this distribution has symmetric upper and lower tail dependence. To overcome this limitation, we have developed a novel form of Bayesian updating based on Archimedean copulas. We present a case study on production enhancement that shows how to combine Bayesian updating and real options. In this study, the oil company has the option to gather information from a production logging tool (PLT) before carrying out a workover.