MODELLING IMPROVED WORKING INTEREST IN OIL AND GAS INVESTMENTS

Abstract

Risk pervades all aspects of the oil and gas Industry. Of all the risks, technical risk in the exploration phase is the most challenging, in which the chance of drilling a dry hole is quite significant. Investors usually increase their odds of success by investing in a fraction of a risky prospect, the Working Interest, (WI). The commonly used analytical expressions for estimating working interest suffer from the “Paradox of Aversion to Incremental Reward” (PAIR) – decreasing WI recommendation when returns are better than expected, contrary to what investors actually do in practice. This study was designed to develop a Risk Adjusted Value (RAV) model that corrects for PAIR and predict realistic estimates of WI in oil and gas assets. Analytical models were developed using a 2-outcome risky prospect, with specified parameters of chance factor (Ps), success value (V), and failure cost (C). Relationships were derived for Expected Value (EV), RAV for specified levels of Risk Tolerance, (RT) and WI. Two hybrid Expected Utility/Expected Value (EU/EV) models were then constructed in order to correct for PAIR – one combining exponential utility function with EV, and the other hyperbolic utility function with EV. The relative impact of the significant variables on RAV was investigated through sensitivity analysis and Monte Carlo simulation. Analysis was then extended to two risky ventures. Optimizations of portfolio of selected risky ventures reflecting the risk characteristics in different phases of upstream oil and gas business were performed for unlimited and limited capital allocation conditions. For particular values of C, V and Ps, WI varied linearly with RT for the two hybrid models in the expected utility maximization region. Also, the higher the RT value, the higher the recommended WI lies in the EV maximization region. Cost has the highest impact on asset’s RAV regardless of the preference function employed. Relative impacts of V and Ps depended on the individual project under consideration. The Hyperbolic/EV model consistently recommended higher WI than the Exponential/EV model. Optimization of the portfolio of selected ventures with RAV as the objective function resulted in 257% higher investment recommendation by the hyperbolic model than the exponential model for the unlimited capital situation. For limited capital, the risk premium for the exponential model was 611% of that required for the hyperbolic model which is an indication of the latter models’ risk tolerance. A method for the determination of RAV and WI was developed which corrected for PAIR. The method can be employed for risky asset valuations, especially in budget constrained environments where different investment options compete for limited capital.

Key Words: Working Interest, Preference functions, Aversion to Reward, Risk Adjusted Value