Overcoming Complexity in Measuring an Energy Utility Enterprise’s Financial Risk Pt. 2
The Solution: An Alternative Framework
At the end of my previous blog, I posed a question as to whether there is a reasonable way to meaningfully measure enterprise-wide financial risk for an energy utility while overcoming the incompleteness and complexity associated with current methods.
The good news is there is a way to generate a utility’s enterprise-wide financial risk that is commercially viable and operationally meaningful — one that is practical and relatively simple, steeped in the use of Regional Transmission Organization (RTO)-generated market data combined with proven statistical tricks. But in order to do so, we need to start from the basics by effectively understanding how energy utilities make money in these deregulated markets while using a well-known concept of the equity markets.
For any operating day, market participant utility financials from asset operations measured by revenue, cost, cash flow, and earnings are impacted by data on dispatched MWh quantity, MWh load served, cleared LMP prices (both Next Day and Real Time), fuel costs, and RTO performance penalties, among other factors. From the perspective of a high-level financial risk assessment, the process by which the above-mentioned input data are determined should be of less importance than ensuring that actual implied activity levels are accounted for in the asset component of the analysis. For now, let’s keep this important tidbit in mind and come back to it later. Because there is uncertainty surrounding these key input data, it makes sense to conduct the financial and corresponding risk assessment in probabilistic terms as opposed to deterministic terms. Deterministic assessment always implies input data are known with certainty, which simply is not the case in this instance.
Against this backdrop, I suggest that one can effectively employ a simpler alternative modeling framework as follows. For the asset component of the analysis, use RTO market data in conjunction with a stochastic financial statement model, juxtaposed against scenarios based on the organization’s market view over the analysis or budget period. Subsequently, extract underlying probability distribution properties of the asset operational data to power a simulation program of the company’s business. The company business model should simultaneously cover the asset operations, trading operations, and structured products to the fullest extent possible. Effectively, this will be a Monte Carlo simulation program that utilizes the concepts and methods of financial statement modeling under uncertainty while capturing the key characteristics of a utility’s interaction with RTO market operations. A key feature will be the inclusion of stochastic representations of the P&L profiles of the utility’s other businesses, specifically trading and structured products.
In employing stochastics, 1) RTO variables and other financial performance drivers can be modeled using probability density functions (pdfs), and 2) power and gas forward prices can be modeled using the industry-acclaimed Pilipovic two-factor price mean reverting model with a seasonality function.
Implementation is facilitated if necessary steps are taken to routinely gather the daily RTO market data on an ongoing basis. Alternatively, the data can be obtained from software vendors that are connected to the RTOs. The parameters of the two-factor price mean reverting model with a seasonality function can be calibrated using actual market quotes that power a statistical non-linear regression process. RTO historical data can be used to estimate the pdf location parameters such as the mean and standard deviation. The justification for and significance of this is that the RTO cleared data already reflect the effects of underlying economic dispatch principles, generator physical constraints, transmission network congestion, forced outage rates, and inclement weather variations. To reiterate, how this data is generated (i.e., the source) is inconsequential to this brand of analysis. What matters is ensuring that the underlying variability of the data is adequately captured to align with reality for the purposes of enterprise financial and risk modeling. This is analogous to how stock prices inherently reflect the collective views of all players (buyers and sellers alike) in the equity market without regard to the source of the individual views.
With this logic, one can capture the probability distribution properties of the revenue and cost drivers of the utility’s asset operations simply by taking advantage of the operational intelligence embedded in the historical RTO cleared data. For a high level assessment, this alternative approach obviates the need for the more complex and time-consuming stochastic dynamic programming route. And, in this context, we are still able to accurately handicap the range of expected levels for the financial drivers represented by the RTO variables. This ensures that, even without using the burdensome complex approach, the financial and risk analysis results are aligned with reality on the ground. By extension, the same logic could be applied to the non-RTO input data covering trading and structured products for purposes of the Monte Carlo simulation. While not specifically mentioned, it is understood that appropriate input correlation structures must be provided to control the simulation program so that meaningful and reasonable instances of the randomly sampled input data are generated.
In the end, through this simpler market-driven method, an energy utility or market participant can accurately and quickly quantify its enterprise financial risk over a specified analysis period in a less costly fashion. Given stated financial targets for revenue, cost, cash flow, and earnings, the corresponding at-risk measures can be accurately derived at various probability levels. The resulting intelligence data can then provide the foundation for enhanced executive decision-making concerning ongoing business operations.
The view expressed above is not meant to discount the significance of energy utilities employing complex dynamic programming methods to analyze the various operational states of their assets for resource planning. On the contrary, asset operators on the ground need that stuff. That being said, this blog conveys a different, albeit relevant mindset. Its primary purpose is to show how a high-level consolidated enterprise risk analysis for an energy utility can be accurately performed, using already-available market data in conjunction with non-RTO business data, to quickly generate ‘at-risk’ metrics that provide real-time actionable insights for executive management, boards, and decision makers. What’s more, this is accomplished without employing time-consuming and complex asset programming algorithms. The advantages to having access to this type of analytical tool, supported by sound financial analysis for resource planning, consolidated risk mitigation, and rapid response to market events, are beyond doubt.
For a utility to pull this off successfully within a commodity trading and risk management (CTRM) production system or an enterprise resource planning production system, requires a risk solution that demonstrates strong economic fundamentals, enables easy and varied data capture on a single platform, allows flexible analysis configuration, and provides good visualization of the key solution outputs. The OATI enterprise risk solution meets these requirements and much more.
Stay tuned for more trading and risk management discussions from OATI.
If you missed Geoff’s first blog and would like to read it, click here.
Mr. Geoff Anato-Mensah has over 35 years of corporate business experience, with the last fifteen focused on energy risk management. Mr. Anato-Mensah is currently Managing Director of Risk at Open Access Technology International, Inc. (OATI). He has directed, managed, and implemented financial valuation and risk management programs for energy companies. He has strong quantitative skills in asset valuation, credit risk modeling, market risk modeling, derivatives pricing, options trading, strategic planning, and risk policy development. He has been certified as an Energy Risk Professional (ERP) by the Global Association of Risk Professionals (GARP) following successful completion of a rigorous energy risk management program.
In his current role, Mr. Anato-Mensah is leading the vision for the OATI risk solution for energy Market Participants. In previous managerial risk roles, he developed and directed risk management programs covering risk identification, risk assessment, risk limit setting, risk monitoring, risk mitigation, and risk reporting to company stakeholders.