Quantitative Methods for Electricity Trading and Risk Management

Traditionally, economics sectors of public utility such as power and gas have been developed and run under strict monopolistic regimes. We may find examples of state monopoly, especially in south-east Europe, regional or municipal monopoly in central and north Europe or somewhere in the United States, and, finally, natural monopoly (privately owned). There have been two main reasons which justify the success of monopolistic regimes. The first is the Keynesian economics beliefs which influenced a lot of the economics policies of many industrialized countries just after the Second War World. The Keynesian economics vision suggests a significant intervention of the state in economic affairs, and the period we are talking about was characterized by the greatest development of utility sectors all over the world. The second important reason is related to the huge initial investment necessary to start up any industrial initiatives in the utility sector itself. Only big industrial subjects or national governments have the resources necessary to face such investments. From the beginning of the 1990s a growing number of countries worldwide, including most of the developed ones such as the USA and the UK, have chosen a substantial restructuring of their utility sectors, starting from their electric power sectors, massively oriented towards liberalization of the whole sector.

Classical economic theory teaches us that in a free and competitive market the price of any traded good or service is completely determined by the interaction between aggregated demand and aggregated supply. Hence, at least in theory, the price of electricity (the commodity) and ancillary service related to it is determined, in a liberalized market, by this kind of dynamic equilibrium. Because of that, before considering electricity price behavior it is fundamental to understand and analyze the shape and dynamics of aggregated demand and supply.

In the previous chapter we saw that electricity price dynamics are influenced by many economic and physical factors most of which are strongly related to the specific system and market structure of the country or geographic area. The natural conclusion is that the dynamics of electricity prices for a certain area cannot be fully generalized. However, there are common features which characterize the time evolution of electricity prices over heterogeneous areas, which, mostly, can be analyzed and studied with common tools. The determination of the principal common features of electricity price dynamics and the specification of the most advanced methodologies for their detection is the main focus of this chapter. The topics discussed will provide the essential basis for the price modeling methodologies to be discussed in the next part of the book.

During the first part of the book we have seen that electricity price dynamics, in organized markets, display peculiarities which partly have to be attributed to the characteristics of the particular market we are looking at, as well as more general features of markets.

In the previous chapter we noted that the main scope of an econometric model is that of describing in the best possible way the interrelation exist­ing among different economic variables. Applying this idea to the study of electricity price dynamics we have the possibility to investigate the rela­tion that links electricity prices (dependent variable of the model) to other economic drivers such as, for example, electricity consumption. This rela­tionship can be expressed in mathematical terms by a simple and generic functional representation of the following type: where Xi represents the value of the i-th generic explanatory variable.

The use of continuous time probabilistic modeling tools for financial applications can be dated back from the beginning of the 1960s, with Samuelson and Mandelbroat, or even at the beginning of the twentieth century with Bachelier. However, we can observe its massive diffusion after the publication of the famous Black–Scholes and Merton articles (1973) on option pricing and the subsequent growing importance of derivative pricing problems in financial economics.

Liberalized electricity markets are characterized by high volatility levels, which mean high risk for both electricity producers and consumers. This high-risk level is not always compatible with agents’ risk attitudes, and hence derivative instruments represent a necessary tool to reconcile agents’ economic exigencies with electricity markets’ natural characteristics.

The Black–Scholes–Merton theoretical framework represents the benchmark for derivative pricing in standard financial markets. However, in the electricity field some of the basic assumptions of this famous approach are violated, and consequently the traditional derivative pricing approach has to be revised in order to work properly for electricity derivatives. In this chapter we will briefly review the traditional pricing approach focusing attention on those assumptions which cannot be considered realistic for electricity derivatives.

In the previous chapter we have seen that, for various reasons, no closed-form pricing formulas are available for the pricing of electricity derivatives, at least if we concentrate on realistic pricing approaches. Not even the simplest plain-vanilla instruments can be priced realistically by means of a traditional Black–Scholes-type model, because of the particular features which characterize the stochastic dynamics of electricity spot and forward prices. Hence, numerical procedures have to be used in order to determine electricity derivative prices and analytical risk measures, such as Greeks. In this chapter we will discuss general numerical procedures for derivative pricing such as Monte Carlo simulations and Lattice methods, presenting also some practical pricing applications related to derivative instruments introduced in previous chapters.

Optimization problems have always been an important issue in any industry, particularly in the power generation industrial sector. Before the liberalization process of the electricity markets started, the electricity market was monopolistic; the government authority fixed power tariffs and the only uncertainty faced was exogenous demand and mechanical failures giving operational and demand risk top priority. Under a regulated regime, optimization meant cost-minimization. Electricity producers were forced to meet demand by forecasting future demand preferences along with efficiently managing their generation assets, reaching demand targets at the lowest cost. Liberalization, however, shed new light on the optimization argument. The presence of the free market brought equilibrium in the supply and demand of electricity along with price volatility and uncertainty.

In the previous chapter, we saw that a thermal power generation asset can be thought of as a highly structured financial derivative whose optimal exercise depends upon market price realization. The asset manager has to ‘exercise’ the asset flexibilities in order to maximize the expected return, respecting operational constraints. Of course, for the solution of this kind of problem the dynamic hypothesis we decide to make for the market price variables is fundamental. We can decide to work within a deterministic environment (optimization of price forecasts) in which case the complexity of the optimization problem defined in the previous chapter is drastically reduced, but obviously also the solution is not then extremely accurate.

In competitive markets such as the liberalized electricity markets, market risk is an extremely important variable both in day-by-day business activity and in the strategic decision-making process. For this reason its correct definition and assessment is a fundamental issue for companies which operate in those markets.

Market risk is a complex subject with multiple dimensions and implications for electricity business activity. Analytical risk measures such as traditional Greek measures or high-order and cross-sensitivities allow us to control in detail market risk, but sometimes these analytical risk measures are too technical to be understood by non-technical staff or by management. Hence, it is necessary to make a synthesis of the information contained in analytical measures into a more intelligible form. The natural way of creating such a type of risk measure is of course that of assessing the impact of risky events in monetary terms, because non-technical people are also capable of understanding the meaning of a potential monetary (or economic) loss.

Deregulated electricity markets are highly dynamic markets and consequently static and backward-looking management tools should be supplemented by new instruments which help management to understand and react promptly to unexpected market changes. Risk is not only a random noise that affects a firm’s economic performance, it is also a strategic opportunity which should be appropriately evaluated.