قیمت گذاری برق تحت "انتشار تجارت کربن": یک بنگاه مسلط با مدل حاشیه ای رقابتی

The aim of this paper is to analyze the impact of trading of CO2 emissions allowances on electricity pricing in the short run. We mainly refer to the European Emissions Trading Scheme (ETS) and are interested in understanding the role of electricity market structures. We carry out a simple analytical model useful to verify whether (and under which conditions) the impact of the ETS under market power could be lower (or higher) than that under perfect competition. We analyze a context where generators compete in a uniform, first price auction. Market power in the form of a dominant firm facing a competitive fringe model is assumed. The paper highlights that the marginal CO2 opportunity costs are fully included in energy prices when the electricity market is perfectly competitive. Under market power the impact of the ETS equals or exceeds that under the competitive scenario only when there is excess capacity and the share of most polluting plants in the market is low enough. Otherwise, the impact under market power is less than under perfect competition and significantly decreases in the degree of market concentration. This especially occurs when there is not high excess capacity and regardless of either the plant mix or the allowance price. In this case, moreover, the marginal pass-through rate is lower in the peak than in the off-peak hours and can be even nil if the degree of market concentration is high enough.

This paper studies the impact of “carbon emissions trading” (trading of CO2 emissions allowances) on electricity pricing. To provide a better understanding of the topic and stress its importance, we start by recalling the policy context from which the need of implementing a carbon emissions trading scheme (ETS) arises. It is widely recognized that the emissions of the so-called “anthropogenic” greenhouse gases (GHG), among which carbon dioxide (CO2) is the most important, contribute to global warming and, consequently, to climate change. This is the reason why, for many years, the international community has been engaged in finding the way of stabilizing the concentration of these gases in the atmosphere. For this purpose, in 1992, the United Nations Framework Convention on Climate Change (UNFCCC) has been signed and, in 1997, a large number of industrialized countries agreed to the Kyoto Protocol, coordinating on imposing emissions reduction targets. In order to minimize the social costs of achieving these targets, several tools of environmental policy are strongly recommended. Since emissions trading plays a crucial role among them, several countries and regions have already decided to implement an emissions trading scheme.1 This is the case for the European Union (EU) where an industry CO2 emissions trading scheme started in 2005.2 The European ETS covers several sectors of which the largest is power generation. Therefore, on the one hand, the effectiveness of the ETS largely depends on whether it will be able to induce power industry to significantly reduce its emissions. On the other hand, the ETS might have a significant impact on electricity prices and, consequently, on social welfare. This study focuses on this latter issue, attempting to provide a better understanding of how a CO2 price could impact on electricity pricing.3 The economic literature on emissions trading is wide enough and covers several fields.4 However, existing studies have been mainly concerned with design issues rather than with the impact on correlated (product) markets. Concerning the electricity sector, only recently specific research effort has been made to study the effects of the ETS on product prices but studies generally assume purely competitive frameworks which are far from the reality of electricity markets. These, in fact, are more or less concentrated markets where one or more firms are able to exercise market power. Thus the need of extending the study to imperfect competition arises and in particular the need of answering two important questions: (1) How does the impact of the ETS on electricity pricing depend on electricity market structures? (2) What role does market power in electricity markets play, in this respect5? Studies aimed at answering these questions do exist but they provide a very controversial framework. Sijm et al. (2005) and Wals and Rijkers (2003), by using a game theoretical simulation model based on the theory of Cournot competition and Conjecture Supply Functions,6 find that the electricity price in a competitive scenario increases more than under market power, on both percentage and absolute basis. They attribute this result to the assumption of linear demand function they adopt. Surprisingly, however, Lise (2005) achieves the opposite result (electricity price increases more under market power) even though the author uses the same model. Reinaud (2003), relying on price competition, and Newbery (2005), by assuming constant price elasticity, state that electricity prices are likely to increase more under market power. This controversial framework highlights that results significantly depend on the choice of competition models.7 In the economic literature on the electricity sector several approaches are generally used for modeling competition and several classifications are proposed.8 Examining recent developments in the literature on electricity spot markets, von der Fehr and Harbord (1998) distinguish three groups of approaches: the standard oligopoly models9; the “supply function” approach10; the “auction” approach.11 In the present work we will follow the suggestion of authors who argue in favor of the “auction” approach. In fact, several electricity spot markets have characteristics which make standard models not well-suited to their analysis. In particular in these markets pricing mechanism is a uniform, first price auction. Furthermore, since we have to isolate the effect of the environmental regulation, in the form of a typical cap and trade regulation (namely, a market of carbon emissions allowances), we do not account for the problem of capacity with holding, grid congestion and contract market which, in the opinion of some authors ( Borenstein et al., 1999), can be better investigated by using the standard oligopoly models. In addition, to simulate market power in electricity markets we use a dominant firm facing a competitive fringe model rather than the usual dupolistic-oligopolistic framework. This choice is due to several reasons, either methodological or practical. On the methodological side, the attraction of this characterization is that it avoids the implausible extreme of perfect competition and pure monopoly, at the same time escaping the difficulties of characterizing an oligopolistic equilibrium (Newbery, 1981). This does not mean, however, that it is only a useful benchmark. It is also useful on the practical side, as long as is suited to represent the reality of several electricity markets. We especially refer to those markets emerging from restructuring processes where the incumbent is obliged to sell a portion of his capacity to different firms and new independent producers meet the rise in power demand over time. The case of the Italian market is emblematic from this point of view.12 The article proceeds as follows. Section 2 summarizes the assumptions underlying the model and in particular those concerning power demand and supply, electricity market and allowance market regulations. Section 3 describes the impact of the ETS on power generation costs, a fundamental step for further evaluations. The competitive outcome is illustrated in Section 4. It provides a benchmark for the subsequent analysis. Section 5 simulates the impact of the ETS under market power in electricity market, by using a dominant firm facing a competitive fringe model. We will present various scenarios by altering the following factors: (1) the leader's share of the total capacity in the market; (2) the plant mix operated by either the dominant firm or the competitive fringe; (3) the allowance price (above or below the so-called “switching price”); (4) the available capacity in the market (whether there is excess capacity or not). In Section 6 a quantitative simulation is carried out by using plausible variable costs and emission rates. This simulation provides useful insights about the absolute value of the differential impact (the ETS impact under market power less that under perfect competition) and determines the pass-through rates in the peak and off-peak hours. Finally, Section 7 summarizes the main results of the article and suggests some reflections about their possible limits. Consistently with major economic remarks, the model confirms that the ETS causes a rise in power prices due to the pass through of the carbon opportunity cost. However, the main finding of the analysis is that the impact of the ETS significantly depends on electricity market structures (rather than on other factors). The carbon opportunity cost is fully included in energy prices when the electricity market is assumed to be perfectly competitive. Under market power prices increase less than under competition, except when significant excess capacity is combined with relatively low shares of the most polluting plants in the market. Before proceeding it is important to underline that, throughout the paper, we will focus on short-term issues, i.e. we will analyze the ETS impact on electricity pricing in the short run leaving the question of how the ETS can affect investment decisions for further research.

Carbon emissions trading may cause a rise in electricity prices due to the pass-through of the carbon opportunity cost. The analysis described in this paper highlights that the extent of this effect significantly depends on electricity market structures. Electricity prices fully internalize the marginal opportunity cost of CO 2 emissions allowances when electricity market is assumed to be perfectly competitive. Under market power the impact on the volume-weighted average price can be lower or higher than that under competition depending on several factors: (1) the degree of market concentration, (2) the plant mix operated by either the dominant firm or the competitive fringe, (3) the price of the CO 2 emissions allowances; (4) theavailable capacity in the market (whether there is excess capacity or not). In particular, the model points out that under market power the ETS impact is higher than under perfect competition only when there is excess capacity and the share of the most polluting plants in the market is low enough. Otherwise, and especially without excess capacity, the impact under market power is less than under competition and significantly decreases in the degree of market concentration. In addition, looking at the time distribution of impacts, it can be observed that, under market power, the marginal pass-through rate in the peak hours can be less 100% and lower than that in the off-peak hours. This especially occurs when there is not high excess capacity and regardless of either the plant mix or the allowance price. In this case the marginal pass-through rate is even nil if the degree of market concentration is high enough.The model set out here is simple. Reality, in fact, will be more complex. For instance, power technology mix is generally more heterogeneous than assumed here, where only two kinds of technologies have been taken into account (excluding the backstop technology).Furthermore, one may remark that the competition model is not close to reality of power markets. This objection, however, is not entirely convincing. In fact, simulating a competition context in which market power emerges from the strategic behavior of a dominant firm is,in our opinion, a good representation of several markets and especially of those emerging from restructuring processes where the incumbent is obliged to sell a portion of his capacity to different firms and new independent power producers meet the rise in the power demand over time. The case of the Italian market is emblematic from this point of view. However, we cannot ignore that assuming the other firms (other than the leader) behave as a competitive fringe might be unrealistic. This assumption might be acceptable if we are looking at a market just emerging from a restructuring process but less acceptable if we look at the longer term. In this case, effects of concentra- tion and tacit coordination within the fringe (and possibly among the fringe and the leader) are more likely so that simulating a context in which two or more strategicfirms face a competitive fringe could be closer to the reality. Nevertheless, what is assumed in this paper remains, in our opinion, a good benchmark and an useful starting point for further research. However, other kinds of strategic behavior as well as electricity market regulations (i.e. discriminatory auction, contract market, etc.) should be investigated. In addition, appropriate empirical analyses need in order to check what simulating models suggest.