Application of the Efficient Frontier Concept to Resource Play Budget Evaluation and Portfolio Optimization

Abstract

The oil and gas industry contains significant inherent risks that greatly affect a company's ability to achieve predicted performance. In an attempt to better manage risk and uncertainty during the decision making process, the efficient frontier concept has been proposed as a method for portfolio optimization. This paper analyzes the application of the efficient frontier concept to a yearly budget allocation process considering unconventional resource play opportunities. Economic and optimization models were developed with assumptions specific to resource plays. The models show that the expected risk, uncertainty, and return on investment change according to the level of investment, project selection, the number of planned wells, and to the grouping of wells for budgeting purposes. The budget dependency of the risk and return cause an oddly shaped efficient frontier resulting in mean-variance efficient portfolios that may not be optimal when based on a set level of risk. In addition, the results show that conventional rank and cut methods for portfolio selection may suffice for resource play budget allocation. However, even though the efficient frontier does not often offer definitive solutions for portfolio optimization, illustrating the interactions among portfolio properties and providing insight into inefficient budget allocations could make the efficient frontier optimization process invaluable. Introduction In Markowitz's seminal article of 1952, he introduced the investment community to the idea of the Efficient Frontier (EF) and mean-variance efficient portfolios. An efficient portfolio maximizes expected value for a given level of risk (or uncertainty) and a set capital investment (Figure 1). Many articles have analyzed the possibility of using EF concepts for oil and gas development decisions (Adekunle 2006; Edwards and Hewett 1993; Erdogan et al. 2005; Faya et al. 2007; Haskett et al. 2004; Merritt 2000; Orman and Duggan 1999; Schuyler 2003; Simpson 2002) and a review of these articles is suggested for more elaborate elucidation on the topic. The goal of the project was to explore the potential for the application of the EF for budget allocation among resource play opportunities. Can the EF provide insight for the decision making process? Ideally, a decision framework could provide a concise answer to a problem. However, most studies suggest that the EF should only be used to gain clarity on the interactions among a large number of correlated and uncorrelated investments, and to illustrate which portfolios are not efficient and thus should be avoided (Adekunle 2006; Bratvold 2003; Edwards and Hewett 1993; Erdogan et al. 2005; Faya et al. 2007; Simpson 2002). In contrast, the more conventional rank and cut method based on an economic metric for budget allocation may not be perceived as the best method for risk reduction, but it does provide a specific answer, and it may in fact be an acceptable risk-mitigating method for resource play budget allocation.

Creating Efficient Portfolios That Match Competing Corporate Strategies

“…When the assumption that no correlation exists between assets in a portfolio is invalid, this may result in overestimating portfolio returns and underestimating portfolio risks (Tonnsen 2008;Begg and Bratvold 2008;Adekunle 2006;Brashear et al 1999;Campbell et al 2003). Although dependencies will complicate the portfolio-generation process, the correlation between different assets can be modeled by introducing the relevant covariances in the simulation (Walls 2004;Begg and Bratvold 2008).…”

“…Several authors advocated the application of efficient frontier (EF) and the identification of mean variance developed by Markowitz (1952) in the process of portfolio optimization (Tonnsen 2008). The literature on portfolio optimization in an exploration-andproduction (E&P) context covers the general application of EF (Orman and Duggan 1999;Merritt 2000), the practical applications of EF (Pinkney 2003), the integration of EF with real option valuation (Lima and Suslick 2002;Lima et al 2008), and the impact of decision-maker risk attitude (Walls 2004).…”

Creating Efficient Portfolios that Match Competing Corporate Strategies

“…When the assumption that no correlation exists between assets in a portfolio is invalid, this may result in overestimating portfolio returns and underestimating their risks (Tonnsen, 2008;Begg and Bratvold, 2008;Al-Harthy and Khurana, 2006;Brashear et al, 1999;Campbell et al, 2003). Although dependencies will complicate the portfolio generation process, the correlation between different assets can be modeled by introducing the relevant covariances in the simulation (Walls, 2004;Begg and Bratvold, 2008).…”

“…Several authors advocated the application of Efficient Frontier (EF) and the identification of mean-variance developed by Markowitz (1952) in the process of portfolio optimization (Tonnsen, 2008). The literature on portfolio optimization in an Exploration and Production (E&P) context covers the general application of EF (Orman and Duggan, 1999;Merritt, 2000), the practical applications of EF (Pinkey, 2003), the integration of EF with real option valuation (Lima, 2002;Lima 2008), and the impact of decision makers' risk attitude (Walls 2004).…”

Creating Portfolio Insight by a Pratical Multi-Method Optimisation Approach

“…The M-V method, rather than search for a single optimal portfolio, aims to establish the highest feasible portfolio value for a range of portfolio variances. In the M-V approach, the variance is used as a measure of risk, and the analysis allows for the identification of an optimal portfolio given the risk attitude of an investor (Tonnsen, 2008). M-V applied to portfolio optimization problems in E&P covers general applications (Orman and Duggan, 1999;Merritt, 2000;Pinkney, 2003), integration with real option valuation (Lima, 2002(Lima, , 2008 and the impact of decision makers' risk attitude (Walls, 2004).…”