The critical role of fit-for-purpose financial modelling and analytics in the Oil & Gas industry
These are challenging times for the Oil & Gas industry, with the dramatic falls in oil prices triggering cutbacks in production levels, headcount reductions, renegotiations with oil services companies, lower planned capex and postponed projects.
There is a very real risk of an economic triple-whammy: as the falling oil price reduces income, incremental investment may no longer be economic, with the risk that field life diminishes and decommissioning is accelerated. Although there are a series of levers Oil & Gas business leaders can seek to pull, a sub-optimal in-house financial model and associated analytics will not allow the benefits and associated costs of such actions – nor different future scenarios around oil prices, capex, opex and financing arrangements – to be fully explored and assessed.
At the same time, lenders to highly leveraged Oil & Gas businesses are increasingly reassessing the debt servicing capacity of key assets, as well as that of the whole group. In many cases, existing in-house financial and forecasting models and the current suite of analytics are coming up short in their ability to provide a suitably robust and flexible platform for such assessments, with recurring themes including the following:
- many existing models base future oil price assumptions on forward curves sourced from, at least until more recently, more buoyant times for the industry. Oil price assumption modules consequently may not have the functionality to independently generate potential future oil price shocks, or generate scenarios that reflect the risks of changing macroeconomic conditions and regional/global supply and demand;
- many models do not break opex and capex into each component part. Instead, the total costings from the original well plans (based on $100 or similar oil) have been input to the financial models and analytics dashboards. This severely constrains the ability to consider the impacts of alternative capex and opex scenarios at a more granular level (for example, the early termination or renegotiation of the lease for specialist equipment or a rig), thereby limiting the value of existing analytics suites;
- 1P and 2P reserves may not have been revalued, or the functionality around production profiles is not sufficiently sophisticated or dynamic to be able to respond, without manual overrides, to revaluations of the Oil in Place which can be economically recovered;
- some other key business drivers are not modelled explicitly or dynamically, and so approximations or proxies need to be used in attempting to assess the business impacts of specific options and scenarios. In some other models, the hard-coding of values for certain key business drivers, rather than being sourced from a central repository of assumptions and inputs, makes it a much more time-consuming task to rerun the model under different scenarios;
- the models are often purely deterministic in nature, such that uncertainty around future outcomes can only be assessed by considering stresses around individual business performance drivers one at a time, or as a combination of these stresses in the form of a specific scenario. Stochastic functionality, which instead runs thousands of simulations for the business as a whole, facilitates an understanding of the “corridor of reasonable likelihood” of future outcomes by, along with quantifying the risk of more extreme outcomes (as illustrated in the graphic). Such analytics enable refinancing decisions to be taken with greater confidence. It also allows the risks and financial impacts of sharp movements in oil prices to be modelled more explicitly, and appropriate hedging strategies to be deployed;
- in a similar vein, the potential for geopolitical events, adverse foreign exchange exposures, resource nationalisation and sudden tax hikes have not necessarily been anticipated within existing financial models;
- the model structure was poorly conceived at the outset, and its subsequent expansion to include new projects or functionality has been haphazard and inconsistent;
- model documentation is limited or non-existent, creating key person risk;
- the model has never been validated independently, raising concerns about its integrity and the reasonableness of the methodologies and assumptions employed.
In light of all of the above issues, many financial models and analytics suites may no longer be fit for purpose. But the decision to start afresh, and design and build a completely new model and analytics dashboards, needs to be weighed up carefully. In some cases, the required enhancements to functionality can be retro-fitted to the existing model relatively easily. In some others, the constraints of the existing model framework or the amount of work required to address the core weaknesses strongly indicate the need to start with a blank piece of paper. Either way, it is critical that the design phase involves close consultation with key stakeholders, to ensure that all of the required functionality and capabilities will be captured and delivered.
A well thought through financial model and analytics suite – whether or not it offers stochastic functionality – should, in our view:
- model explicitly the business’s internal and external performance and cost drivers at an appropriately granular level for each asset, including production at each well, opex (for example, rig lease costs, contractor costs, transportation costs) and capex (such as exploration costs, appraisal well drilling costs, field development expenditure). This then enables management to better understand, quantify and mitigate the key threats to that performance, as part of strategic decision making;
- facilitate the production of a comprehensive suite of analytics, including production forecasts, P&L accounts, operating cash flows, net cash flows, balance sheets, and all the required Key Performance Indicators and Key Risk Indicators for individual projects, business units or the entire business, in a consistent manner, from a bottom up perspective. This will require the explicit modelling of each asset’s Production Sharing Agreements, Cost Recovery Pools, royalties and various tax liabilities, for example;
- dynamically link valuations of 1P and 2P reserves to future oil price scenarios, potentially parameterised through in-depth discussions with sub-surface engineers (see the diagram below);
- model dependencies between key business drivers and risks in an explicit manner, to allow risk analytics to be brought more to the fore;
- readily facilitate changes to assumptions, parameters and other inputs for a wide variety of key business performance drivers, which are captured in a central module, and which are all well documented in terms of their provenance and rationale. This will enable alternative business plans and scenarios to be explored in a robust manner, and the effectiveness of potential risk mitigating hedges to be tested;
- be sufficiently flexible that new assets and projects can be incorporated consistently and seamlessly, to allow rapid scalability of the analytics suite, while recognising that such assets and projects will have unique features;
- be subject to a rigorous governance framework that addresses controls around model change, data quality and management sign-off of key assumptions, as well as ensuring that management understands the model’s operation and limitations.
A fit-for-purpose financial and forecasting model is still only a tool, and should not be relied upon blindly. However, a high level of model functionality and a well-run model parameterization process (along with regular recalibrations) will mean that much of the corporate expertise and knowledge has been captured effectively within the model and analytics dashboards, engendering greater buy-in across the business. This in turn will translate into greater use and acceptance of the model and associated analytics suite as a critical input to business strategy and decision making.
If you would like to discuss these issues, or the impact of emerging technology or data and analytics on your industry, then contact our Data & Analytics team.