The Application of Quantitative Risk Analysis Techniques in the Oil and Gas Industry

Investment projects in the oil and gas industry involve great technical challenges, considerable risks and massive financial resources. Given the inherent obstacles required to make any project successful, this industry has served as a proving ground for cutting edge project valuation methodologies and tools, such as quantitative risk analysis. As oil and gas organizations attempt to maximize the value of each project and optimize their portfolio of investment opportunities, it is imperative that all risks are properly identified and quantified, to maximize value and increase the effectiveness of mitigation strategies.

Discounted Cash Flow, Sensitivity and Scenario Analyses

The Discounted Cash Flow (DCF) method is the most commonly used tool for decision analysis in project valuation. It is based on the estimation of the net present value (NPV) of the cash flow of the project under consideration, discounted at the company’s hurdle rate. Easy to implement and widely used in the oil and gas industry and management schools, DCF acknowledges the time value of money, provides a common language for valuation, establishes clear decision criteria and allows comparison of projects based on metrics. However, DCF fails to address a number of key factors:

  • The uncertainty of future cash flow (assumptions about cash flow components are all static)
  •  The recognition of explicit project risks (risks are all assumed to be accounted for by the discount rate)
  • The valuation of risk mitigating strategies

Sensitivity analysis and scenario analysis attempt to address these issues by acknowledging uncertainty over the project’s inputs and evaluating their impact on the project, thus generating ranges for the project’s metrics in the form of spider graphs, sensitivity tornadoes, and other outputs. Unfortunately, neither sensitivity analysis nor scenario analysis appropriately provides probabilities for the ranges of possible outcomes of the project. For project managers and other decision makers, there is a clear need to fill the gaps left by these processes.

Quantitative Risk Analysis, Probability Distributions, and Monte Carlo Simulation

Quantitative risk analysis goes further than the traditional DCF method, moving the analyses beyond the static world. With quantitative risk analysis, uncertainty in the project’s future cash flow is explicitly acknowledged, risks are objectively assessed based on probabilities and impacts and relevant inputs are modeled as probability distributions. Static models used in traditional DCF, sensitivity and scenario analyses use point estimates as inputs, whereas stochastic models used in quantitative risk analysis utilize probability distributions as inputs.

Quantitative risk analysis introduces a new layer of complexity – and accuracy – to the valuation problem by acknowledging that there are multiple possibilities for each of the inputs’ values. The analyst must describe each input, not only through a number but through a group of characteristics that effectively represent each input’s possible values, that is, its probability distribution. Typically, probability distributions are described by parameters like most likely value, expected value, minimum and maximum values, standard deviation, percentiles or shape parameters. At this point, historical data, expert opinion and time series are the best sources for estimating the probability distributions’ parameters.

Quantitative risk analysis also combines all of the computed inputs’ distributions in a meaningful way inside the valuation model, including their correlated interdependencies to properly assess the project’s possible outcomes. This is where Monte Carlo Simulation (MCS) becomes a vital tool for properly forecasting risk.  MCS is a very powerful, straightforward and robust numerical method for sampling random numbers. In this case, each of the inputs’ possible values from its probability distributions is considered. MCS allows the evaluation of multiple probability distributions (including correlations) in practically any kind of problem, regardless of size or complexity. Static models typically require only minor changes, if any, to be suitable for MCS.

 Examples of Risk Analysis in the Oil and Gas Industry

With quantitative risk analysis, uncertainty in relevant inputs is explicitly modeled as risks, using probability distributions and stochastic processes. Determining which inputs are relevant relies on the nature of the project being evaluated and the business environment surrounding it. Typical risks observed and modeled in the oil and gas industry include exploratory success chance, oil, gas and water production curves, capital expenditures (CAPEX) and operational expenditures (OPEX). Quantitative risk analysis also addresses commodity prices and demands, which are more challenging to model, as they often rely on elaborate stochastic processes and political/regulatory risks, where data to estimate the model is always an issue.

In the oil and gas industry, quantitative risk analysis is usually undertaken at different stages of a project, shifting the focus to the specific tasks at hand:

  • Integrated project risk analysis forecasts the risks surrounding the oilfield projects and considers the probability distribution of the project’s NPV as its main output
  • Cost risk analysis focuses on the cost structure of the project, explores the deeper details of cost inputs and provides the probability distribution of the CAPEX as its main output
  • Schedule risk analysis focuses on the time required to complete each task, and its main outputs are the probability distribution of the project’s first oil and its possible critical paths

Some risks have a compound effect on schedule and cost that need to be acknowledged, like the drilling time in an offshore oil project, for example. Larger, more complex models can even combine the three analyses simultaneously.

Benefits of Quantitative Risk Analysis for Decision Making

Quantitative risk analysis acknowledges uncertainty project outcomes, offering new metrics to project valuation in terms of ranges and probabilities, including the probabilities of achieving target values, correlation tornadoes and statistical contingencies. Quantitative risk analysis also allows identifying the key risk drivers underlying a project, which is not possible using traditional DCF, sensitivity or scenario analysis. Clearly identifying a project’s key risk drivers is crucial to effectively designing and evaluating risk-mitigating strategies according to their cost and benefits. Usually, the benefits of a risk-mitigating strategy are not better expected values, but better risk profiles, in the form of a lower standard deviation of the project’s NPV distribution, a higher probability of positive NPVs, reduced downside and so forth. Ultimately, decisions are based on the probability distributions of the project outcomes, their expected and extreme values, available mitigation strategies and remaining risks.

Conclusion and Further Uses of Quantitative Risk Analysis

Quantitative risk analysis is the natural evolution of the traditional DCF method, moving the analyses beyond the static world. Investment projects in the oil and gas industry, where risk is plentiful, offer an abundance of examples of successful application of risk analysis. This method allows the identification of the key risk drivers underlying a project and the design and evaluation of risk mitigating strategies. Ultimately, decisions are based on the probability distributions of the project outcomes, their expected and extreme values, available mitigation strategies and remaining risks.

A good quantitative risk analysis model is also the starting point of other more advanced analyses common in the oil and gas industry, like value of information (widely used in exploration and early production projects), optimization under uncertainty (used in refining projects, portfolio allocation and corporate strategic planning), and real options (complex production projects, expansion, abandonment and acquisitions). The challenges involved in any project are great, and utilizing the analysis that forecasts not only risk, but risk’s probability, will enhance the ultimate success of any venture.