Valuating risk from sales contract offer maturity in electricity market

Abstract

The electricity retail business has always been exposed to a significant risk due to a very volatile nature of the wholesale electricity market. Recently, this trend has been accelerating, as electricity-forward price can rise or fall during one day by an amount comparable to the retailer’s margin. As a result, the risk that arises from sales contract offer maturity can significantly reduce or even neutralize the retailer’s profit. To develop a mitigation strategy for this risk, a retailer clearly needs a model for a detailed risk analysis. Many attempts have been made to analyze several risk issues of an electricity retail company, but so far none has explored and evaluated the sales contract offer maturity risk. This paper presents a fundamental model for evaluating this risk and a methodology for its use, and compares its analytical performance with traditional techniques obtained from the option-pricing theory. A realistic example illustrates the use and the benefits of the proposed model in risk analysis of the electricity retail business.

Introduction

Following deregulation and liberalization of the electricity market, a new era has begun for the electricity companies participating in the market. While the aim of deregulation is to create conditions for electricity prices to gravitate towards their marginal values by introducing competition, a liberalized market with its regulated third party access to the grid creates an opportunity for new participants to enter the market and a chance for consumers to choose from among a number of different electricity retailers. However, while in a vertically-integrated monopoly dominated market, the cost of risk is simply passed on to consumers, in the liberalized market, exposure to market risk is a much more significant issue for the players, especially in light of highly volatile market conditions that currently exist. Consequently, the market participants are forced to start measuring their risk exposure and manage it in accordance with the individual company’s risk policy.

This paper focuses on retail electricity companies that purchase electricity on the wholesale market and sell it on the retail market, supplying it to the individual consumers. Since most of the consumers do not possess the means, knowledge or sophistication to be able to participate in a volatile wholesale electricity market on their own, the retail companies assume the market risk on behalf of the consumers and charge a premium for this service. It is generally agreed that price risk and volume risk are the most important risks faced by a retailer on the electricity market [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11]. Namely, the electricity can not be easily and economically stored, so it has to be produced and consumed at the same instant. Hence an important price risk component is the risk that arises from the maturity of the sales contract offer (i.e. the offer). Theoretically, a retailer could purchase electricity on a wholesale market and sell it to the consumer instantly; in reality, he often first has to find a buyer in the retail market in order to buy the electricity on the wholesale market. A retailer therefore offers to a consumer a fixed-price sales contract to buy electricity. If the fixed price offer is valid for a few days, this exposes the retailer to unfavorable wholesale price movements during the time until the offer expires at its maturity. Since wholesale electricity price can easily rise or fall during a single trading day by an amount which is comparable to the retailer’s margin, the risk resulting from the offer maturity can reduce or even neutralize the retailer’s profit even before the energy delivery period has started. The aim of this paper is to analyze and evaluate the risk arising from the offer maturity.

The existing models deal with consumer pricing considering demand and electricity price uncertainty. The retailer has to assess the risk that arises from demand uncertainty and incorporate it into the contract sales price. Keppo and Räsänen [2] worked on the consumer pricing based on the assumption that electricity consumption and electricity price are both stochastic processes. Determination of the optimal price and quantity in the sales contract by setting the optimal strategy using a stochastic optimization model for both the supply and purchase sides was reported in [3], [4]. The proposed stochastic optimization model can help the retailer to determine both the optimal price and the optimal quantity. Demand-forecast uncertainty should also be properly taken into account when setting the annual risk management strategy as seen in [5], [6].

The risk resulting from electricity-price volatility has been the subject of several studies. Vehviläinen and Keppo [7] introduced the idea of managing the financial risk of a power portfolio by using a model employing Monte Carlo simulation method. The result is a risk management method that can be applied in daily electricity trading. Buying electricity on the volatile and potentially unpredictable wholesale market and selling it on the retail market through fixed-price contracts puts the retailer in a high-risk situation. This is why managing the procurement costs and the associated risk is one of the vital issues for any electricity retailer. The answers to the key questions of what products to buy and how to buy them are very important in any procurement strategy [9], [10]. One of the possible strategies proposed by Woo et al. [11] is to derive a criterion that summarizes the optimal tradeoffs between the procurement risk and the expected cost by solving a mathematical model. All such models should include proper values for the technical and economic boundary conditions in the optimization. The studies mentioned above deal with the retailer’s aggregated consumption and the associated price uncertainties when participating on the market in order to define an optimal procurement strategy. However, none of the studies include the sales contract offer maturity risk.

In this paper, a novel methodology is presented that deals with the risk premium determination by considering the risk resulting from the offer maturity, in addition to the above-mentioned uncertainties. In the retail market, it is customary that the offer matures in a period ranging from a few days up to 1 month, or even more. Since the retail business is different from the wholesale business and consumers are not very familiar with the wholesale market conditions, a contract offer with a “zero-day” maturity is not acceptable to most of them. When considering the consumers’ needs, the retailer must be able to produce the offers with different maturity times. In the meantime, the consumer analyzes the offered contract and when the contract offer matures he decides whether or not to accept the offer. Of course, the consumer can always reject the offer without any penalty. Thus, the retailer has to two possibilities as to when to purchase the energy for the consumer: either on the day of issue of the offer, taking the risk that the consumer will reject the offer, or on the maturity date of the contract, facing the risk of unpredictable electricity price movements on the wholesale market in the mean time. Often, the offer is accepted when the market price increases, which is bad for the retailer as he needs to purchase electricity at a higher wholesale market price. Consequently, unfavorable price movements on the wholesale market are reflected in a higher percentage of accepted offers, resulting in the retailer being exposed to higher risk.

In the paper, a Fundamental Retail Market Model (FRMM) for the retailer is derived, and a methodology based on Monte Carlo simulation is proposed to use FRMM for risk analysis. The methodology is used to determine the risk premium for the sales contract that a retailer offers to a consumer, with the focus on the risk exposure due to different offer maturities. Monte Carlo simulation is a commonly used method for pricing financial options, [12]. Since in the paper the offer is treated as an option, Monte Carlo simulation technique was chosen as the most appropriate. The simulation results of the proposed risk analysis methodology could also be used as an input in an optimal portfolio selection process. Such a process would entail definition of the optimization criterion and would aim to maximize expected profit at a given risk level as reported in [7], [13]. While the optimal portfolio selection including the associated procurement strategy and the analysis of uncertainties on the consumer side is beyond the scope of this paper, it is one of the goals in our ongoing research.

The retailer is in a position of a seller of a call option on electricity supply, and the consumer is the buyer of the option, which gives the consumer the right (but not the obligation) to exercise the option – buy the energy at the offer maturity. The premium for this option is fixed and determined at the offer time. Since the retail consumer as the buyer of the call option usually isn’t sophisticated enough to be able to price the option; instead, the consumer gathers several offers from different retailers and chooses one of them. The retailer has to price the sales contract he is selling – determine the level of risk exposure and set the risk premium based on his target risk-profit ratio.

The proposed methodology for determination of risk premium R_p regarding the maturity of the offer includes the following steps. First, the dependence between profit volatility and the maturity of the contract offer can be observed by finding the possibility of decoupling risk issues related to different stochastic model variables. The risk exposure is expressed as the Conditional Value at Risk (CVaR), providing an immediate solution for defining a risk premium to cover the risk resulting from the maturity of the offer, [14], [15]. In the paper, the risk exposure in the form of CVaR and the expected profit from supplying energy to the consumer is obtained using the proposed retailer’s model. Since the offer can be modeled as a call option, the model results were compared to results obtained by the widely known Black-Scholes option-pricing formula, as described in [16].

The rest of the paper is organized as follows: Section 2 discusses the retailer’s model; Section 3 presents the numerical simulation results; and the results are discussed in the concluding Section 4.