تاثیر ساختار بازار برق بر هزینه CO2 برای قیمت های برق تحت رقابت کمی - رویکرد نظری
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|19838||2012||10 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Energy Economics, Volume 34, Issue 4, July 2012, Pages 1143–1152
We present a theoretical analysis of the impact of power market structure on the pass-through rate (PTR) of CO2 emissions trading (ET) costs on electricity prices. Market structure refers in particular to the number of firms active in the market and the intensity of oligopolistic competition as measured by the conjectural variation, as well as to the functional form of the power demand and supply curves. In addition, we analyse briefly the impact of other power market-related factors on the PTR of carbon costs to electricity prices. These include in particular the impact of ET-induced changes in the merit order of power generation technologies and the impact of pursuing other market strategies besides maximising generator profit, such as maximising market shares or sales revenues of power companies. Each of these factors can have a significant impact on the rate of passing-through carbon costs to electricity prices.
During the first phase of the EU Emissions Trading Scheme (ETS; 2005–2007), the impact of the scheme on electricity prices became a major political and academic issue (Sijm et al., 2005 and Sijm et al., 2008). In general, the impact of carbon trading on power prices depends first of all on the price of a CO2 emission allowance and the carbon intensity of the power sector, especially of the generation technologies setting the electricity price at different levels of power demand. These two factors, i.e., allowance price times carbon intensity, determine the (marginal) carbon costs of power generation. In addition, however, the impact of emissions trading on electricity prices depends also on the extent to which the CO2 allowance costs of power generation are passed through to these prices. This so-called ‘pass-through rate’ (PTR) is determined largely by the structure of the power market. 1 By structure, we refer in particular to the interaction of the following three elements: 1. The number of firms active in the market (N), indicating the level of market concentration or market competitiveness. Depending on this number of firms, the market structure is called either monopolistic (N = 1), duopolistic (N = 2), oligopolistic (N = small) or competitive (N = large). 2. The shape of the demand curve, notably whether the (inverse) demand curve is linear or iso-elastic.2 3. The shape of the supply (or marginal cost) curve, in particular whether the marginal costs are constant – i.e., a flat, horizontal line of perfectly elastic supply – or variable, i.e., sloping upward in either a linear or iso-elastic way.3 The main purpose of this paper is to analyse the impact of power market structure on the pass-through rate of CO2 emissions trading costs to electricity prices from a theoretical point of view, including graphical illustrations and mathematical proofs. It builds on Section 3 – including the Appendix – of an article by the authors (Chen et al., 2008), making the following additional contributions: • The results herein are more general in that we distinguish between cases of constant versus non-constant marginal costs and discuss the implications of this distinction for the derivation of the PTR under these cases. • The implications of ETS-induced changes in the so-called merit order of power generation technologies for the PTR of carbon costs to electricity prices are analysed. These changes account for much of the ETS-caused decreases in emissions, but a general analysis of their implications for the PTR of CO2 costs to power prices has not previously been presented. • The effects of oligopolistic competition, considering a full range of conjectural variations, including special cases of Bertrand (perfect competition), Allaz–Vila competition, Nash–Cournot competition, and perfect collusion. • We discuss briefly the implications of other, market-related factors for the pass-through of emissions trading costs to power prices. These factors include in particular the incidence of (i) market regulation, (ii) market imperfections, and (iii) other market strategies besides maximising profits, such as maximising market shares or sales revenues. In the literature on the electricity sector, several approaches are generally used for modelling competition such as Cournot–Nash models, the ‘supply function’ approach and the ‘auction’ approach.4 In this paper, we predominantly apply the so-called ‘conjectural variations’ approach, of which the Cournot–Nash model is a special case, to analyse the impact of different market structures on the pass-through of CO2 emissions trading costs to electricity prices. The basic assumption of this approach is that quantity, i.e., output generated, is the decision variable of rival electricity producers. Whereas the Cournot model is based on the conjecture that rivals will not react to a production change by changing their output, the conjectural variation models are flexible with regard to the conjectures on the response of competitors. By parametrically changing the assumed supply response, different degrees of competitive intensity can be modelled, ranging from pure (Bertrand) competition (infinitely large positive quantity response by rivals to price increases), to oligopolistic Cournot competition (no response), and even collusion (which can be simulated by a negative quantity response to price increases). An intermediate case is Allaz-Vila competition which, under some assumptions, implies that a unit change in output by one firm is believed by that firm to stimulate a 0.5 change in output in the other direction by rival firms (Murphy and Smeers, 2010). Positively sloped conjectured supply functions (CSFs) also represent different degrees of competitive intensity between the Cournot and Bertrand cases (Day et al., 2002 and Hansen, 2010). Recently, Gulli et al. have used an alternative approach – the so-called ‘auction’ approach – to analyse the impact of market structures on CO2 cost pass-through to electricity prices (see Bonacina and Gulli, 2007, Chernyavs'ka and Gulli, 2008 and Gulli, 2008). More specifically, by using a dominant firm facing a competitive fringe model, they analyse the short-run impact of CO2 emissions trading on wholesale electricity spot markets where the pricing mechanism is a uniform, first-price auction. Their main finding is that the impact of carbon trading on power prices significantly depends on the structure of the electricity market. Under perfect competition, electricity prices fully internalise the carbon opportunity costs. Under market power, however, the extent to which these costs is passed through into electricity prices depends on several factors, including (i) the degree of market concentration, (ii) the plant mix operated by either the dominant firm or the competitive fringe, (iii) the carbon price, and (iv) the available capacity in the market, i.e., whether there is excess capacity or not.5 Our results confirm (i) for the case of Cournot and, more generally, conjectural variation competition. The remainder of the paper is structured as follows. 2 and 5 discuss the PTR of carbon costs to electricity prices under different power market structures, in particular under different levels of market competitiveness (competitive, Cournot, and monopoly) and different combinations of shapes for power demand and supply (marginal cost) curves. Section 6 generalizes some of these results for oligopolies using conjectural variation competition, while Section 7 discusses two bounding cases of linear demand and supply under competitive markets. Subsequently, Section 8 analyses the implications of ETS-induced changes in the merit order of power generation technologies for the PTR of carbon costs to prices. Next, Section 9 discusses the implications of other, market-related factors for the pass-through of emissions trading costs to power prices, such as the effect of company strategies other than profit maximization. Finally, Section 10 summarizes our major findings.
نتیجه گیری انگلیسی
A major factor affecting the impact of emissions trading on electricity prices is the structure of the power market. This structure refers primarily to the interaction of three elements: • The number of firms active in the market (N), indicating the level of market competitiveness or market concentration. • The shape of the demand curve, notably whether this curve is linear or iso-elastic. • The shape of the supply curve, particularly whether the marginal costs before emissions trading are constant – i.e., a flat, horizontal line – or variable, i.e., sloping upward in either a linear or iso-elastic way. Table 1 gives an overview of the cost pass-through formulas for different market structures, assuming profit maximisation among Cournot producers. The table makes a distinction between two definitions of the pass-through rate (PTR), i.e., PTR1 = dP/dMC (where dP is the change in price and dMC the total change in marginal costs, including carbon costs) and PTR2 = dP/dCC (where dCC refers to the change in carbon costs only). If the supply function is perfectly elastic (i.e., marginal costs are constant) PTR1 is similar to PTR2. However, if the marginal costs are variable (i.e., sloping upwards linearly or iso-elastically), the two rates are no longer similar, with PTR1 > PTR2. Table 1. Overview of cost pass-through formulas for different market structures, assuming profit maximisation among Cournot producers, and different definitions of the pass-through rate. Demand function Perfectly elastic Perfectly inelastic Linear Iso-elastic Definition of PTR/Supply function First definition:View the MathML sourcePTR1=dPdMC All supply functions 0 1.0 View the MathML sourceNN+1 View the MathML sourceNεNε−1 Second definition:View the MathML sourcePTR2=dPdCC Perfectly elastic N.A. 1.0 View the MathML sourceNN+1 View the MathML sourceNεNε−1 Linear 0 1.0 View the MathML source11+1/N+u/v View the MathML source11−1Nε+uP−ε−1 Iso-elastic 0 1.0 View the MathML source11+1/N+εb View the MathML source11−1Nε1+bε Note: PTR is pass-through rate, dP is the change in price, dMC is the change in marginal costs, dCC is the change in carbon costs, N is the number of firms active in the market, 1/b is the price elasticity of supply (b > 0), −ε is the price elasticity of demand (ε > 0), v > 0 is the absolute value of the slope of the inverse, linear demand function, and u > 0 is the slope of the inverse, linear supply function. Table options Based on Table 1, our findings regarding the impact of market structure on cost pass-through include: • If demand is perfectly elastic, i.e., the price is given, then the PTR is zero. This outcome applies also to cases of outside competition – when prices are set by competitors outside the ETS – or price regulation, in particular when the cost pass-through of freely allocated allowances is not accepted. • If demand is perfectly inelastic, i.e., demand is fixed and unresponsive to price changes, then the PTR is always 100% (in the case of competitive markets), regardless of the shape of the supply function, assuming no change in the merit order of generating plants. • If supply is perfectly elastic, i.e., marginal costs are constant, the PTR depends on the shape of the demand curve and the number of firms active in the market (N). If demand is linear, the PTR is significantly lower than 100% when N is small (for instance, it is 50% in the case of monopoly, i.e., N = 1) but increases when markets become more competitive (it approaches 100% in the case of perfect competition, when N = ∞). If demand is iso-elastic, however, the PTR may be substantially higher than 100% when N is small (and demand is less elastic), but decreases towards 100% when markets become more competitive (or demand becomes more price-responsive). Therefore, if supply is perfectly elastic, the PTR always tends towards 100% when the number of firms becomes large and, hence, markets approach the case of full competition, regardless of the shape of the demand function. • If supply is not perfectly elastic, i.e., marginal costs are variable, the PTR should be carefully defined, distinguishing between PTR1 = dP/dMC and PTR2 = dP/dCC. When using the first definition, the pass-through rate (i.e., PTR1) under variable marginal costs is similar to the PTR under constant marginal costs (as discussed above). However, when applying the second definition, the pass-through rate (i.e., PTR2) under variable marginal costs is always lower than the PTR under constant marginal costs. Moreover, the PTR2 under variable costs decreases when supply becomes less elastic or demand becomes more elastic. The distinction between the two definitions of the pass-through rate is also relevant in the case of ETS-induced changes in the merit order of the power supply curve (i.e., changes in the ranking of generation technologies according to their marginal costs, including carbon costs). For instance, if the PTR is defined as dP/dMC (where dMC refers to the difference between the marginal costs of the price-setting production technology after and before emissions trading), its value is and remains 100% in competitive markets, regardless of whether the merit order changes or not. However, if the PTR is defined as dP/dCC (where dCC refers to the carbon costs of the production unit that becomes marginal after emissions trading), the PTR can deviate substantially from 100% (either > 1.0, or < 1.0) if the merit order changes, even under competitive markets with perfectly inelastic demand and perfectly elastic supply, depending on the carbon intensity of the marginal generation technology after emissions trading. In addition, there are additional factors related to the power market that influence the pass-through of carbon costs to power prices, including: • Market strategy. Besides profit maximisation (as assumed above), firms may pursue other objectives such as maximising market shares or sales revenues. These differences in market strategy affect the PTR, regardless of whether carbon costs are actual cash outlays or opportunity costs. • Market regulation. However, in the case of market regulation (or ‘regulatory threat’) public authorities (or firms) may treat the actual, real costs of purchased allowances differently than the opportunity costs of freely obtained allowances, resulting in different levels of cost pass-through to power prices. • Market imperfections. The pass-through of carbon costs to power prices may be affected by the incidence of market imperfections such as (i) risks, uncertainties or lack of information, and (ii) other production constraints, including ‘must run’ limits, highly non-convex operating cost functions (such as high start-up costs), lack of flexible fuel markets, and time lags.