VIEWS: 3 PAGES: 22 POSTED ON: 9/17/2012 Public Domain
Generation Asset Divestment in the England and Wales Electricity Market: A Computational Approach to Analysing Market Power Christopher J. Day and Derek W. Bunn* April 1999 Decision Technology Centre London Business School Sussex Place, Regents Park London NW1 4SA, UK * Derek Bunn (dbunn@lbs.ac.uk) is Professor, and Christopher Day (cday@lbs.ac.uk), a research student, in Decision Sciences at London Business School, Sussex Place, Regents Park, London NW1 4SA, UK. Generation Asset Divestment in the England and Wales Electricity Market: A Computational Approach to Analysing Market Power Abstract The task of developing an adequate modelling approach to understanding strategic behaviour in competitive electricity markets is a major open research question. Contributions from economic theory based on equilibrium solutions give some comparative insights on market design, but provide an inadequate representation of the observed, dynamic evolution of these typically quite imperfect markets. In this paper, we develop an approach to modelling competition between electricity generating companies based on computational modelling and simulation. In this approach, each company, or agent, is modelled with specific objectives and bounded reasoning abilities. Using the price-setting mechanism found in the England and Wales pool, the interaction of the generating companies is simulated through time. For validation, we compare the computational approach with the theoretical equilibrium in continuous supply functions, for the scenario of symmetric firm sizes. As the computational approach is not constrained by the requirement of symmetry or continuity, we then demonstrate its ability to analyse more general circumstances. We apply the new approach to analysing the second round (1999) of capacity divestment proposals which the government and regulatory authorities in England and Wales required in order to improve the efficiency of the wholesale power market. In this context, we suggest that, for a second time, the level of market power may be underestimated and that although the proposed amount of divestment is substantial, it may still be insufficient to avoid the need for further regulatory controls in the short term. 2 1. Introduction With a well re-structured electricity industry, the aspirations of full de-regulation are to create wholesale markets which are perfectly competitive and deliver prices close to short-run marginal costs. Yet many of the re-structuring exercises of the past decade have been cautiously tempered with concerns about security of supply, stranded assets and various other externalities, the result of which has been the creation of power pools with market power on the generation side. The Electricity Pool of England and Wales is a prime example. Under such circumstances, continuous monitoring for the abuse of market power becomes necessary, and the persistent issues of price controls or further re-structuring add considerable regulatory risk to the industry. With the perception that generating companies in the England and Wales market had indeed been exerting excessive market power, the UK office of electricity regulation previously attempted to mitigate such abuses by requiring plant divestments (OFFER 1994). In accordance with this directive, 17% of the capacity owned by the two major price-setting generators (National Power and PowerGen) was divested in 1996. At that time, an official report on the market power of National Power and PowerGen concluded that with this amount of divestment and the prospect of further new entry, their ability “to affect the level of Pool prices, whether to keep them high or cause them to fluctuate, will be small” (MMC, 1996). However, this did not happen, and by 1998, those two generators were again accused of maintaining high prices (DTI, 1998), So much so, in fact, that the regulatory office was again saying that the “most effective route to increased competition in the short term would seem to be to transfer more of National Power’s and PowerGen’s coal-fired plant into the hands of competitors” (OFFER, 1988). Just how to decide upon the appropriate level of plant divestment is a formidable and unenviable regulatory task. Clearly, OFFER got it wrong the first time in 1996, but used very little formal analysis other than indices of market share and price-setting. For example, Littlechild (1996) referred to a reduction in the HHI index of market concentration from about 3,600 in 1991 to about 1,600 in 1996 as a measure of success in making the industry more competitive. Green (1996) addressed the divestment issue by applying the approach of supply function equilibria (Klemperer and Meyer, 1989), as an extension of previous analyses for the England and Wales pool (Green and Newbery, 1992; Green, 1996), and similarly to what has been done elsewhere 3 (eg Andersson and Bergman, 1995). In that study, Green (1996) broadly endorsed the level of divestment being undertaken at the time. However, the derivation of such supply function equilibria is analytically difficult, and the published studies have tended to make very strong assumptions to facilitate solutions. The basic framework is to assume that generating companies bid supply functions into the pool, representing the price at which they will make available a range of generation capacity, as happens in the Cal PX and the pools of England and Wales, Spain, Colombia and elsewhere. Demand being revealed later, as a forecast, or via demand-side bidding or in actuality (for ex post markets like Victoria) then clears the market. Analytical evaluation of the equilibrium solution either assumes that these supply functions are continuous (Green 1996), whereas the nature of generating units makes these functions increase with quite distinct steps in practice, or restricts the prospective analysis of various industry ownership structures to symmetric, equally sized firms (Rudkevitch et al, 1998), or both (Green and Newbery, 1992). It could therefore be argued that the underestimation of residual market power after the first round of divestment reflected the inadequacy of the modelling approaches available at the time. In this paper we present a new method for determining imperfectly competitive outcomes in electricity markets that utilises the approach of computational modelling and simulation. Rather than directly calculating the equilibrium solution, we formulate a computational model of profit maximising behaviour for each generating company and then simulate the interaction of the companies, observing the adaptation of supply functions during the repeated play of the daily game. In principle, this approach can take actual marginal cost data and deal with both discontinuous supply functions and asymmetric market concentrations. This rest of this paper proceeds as follows. First, we describe the computational framework we have developed for modelling supply function competition and discuss the properties of the dynamic behaviour resulting from simulating the strategic interaction. Then after specifying the demand and marginal cost assumptions of the model, we see how the results of this approach compare with those obtained by calculating a continuous supply function equilibrium, for those simplified circumstances to which the latter can be applied. This provides a calibration and validation of the approach. Then we apply the approach to analysing the second (1999) round of 4 plant divestment, thus demonstrating the applicability of this new approach to modelling competition between asymmetric generating companies with discontinuous supply functions. 2. Formulation of the Computational Model We develop a computational approach to modelling the interaction of generating companies who compete using supply functions in an electricity pool with rules similar to those found in the England and Wales market. Generating companies are modelled holding the conjecture that their competitors will submit the same supply functions as they did in the previous day. Given this conjecture, each company is modelled as a daily profit maximiser, optimising the supply function it submits to the market. Using data on observed demand and estimates of short-run marginal costs, we simulate the interactions between the generating companies and analyse the resulting supply functions. 16.0 14.5 No Capacity Price 13.0 in this bin Bin 11.5 10.0 Capacity in Capacity in Capacity in 1st bin 3rd bin 4th bin Figure 1: An illustration of the piece-wise linear supply function representation. We model the supply functions as having both discrete capacity and price ranges. In this market it is natural to have a discrete capacity range as a generator's capacity comprises of individual generating plants, hence cumulative available capacity increases in a discrete manner. 5 Discretizing the price range is not such an obvious transformation as the value a generator prices its plants at can vary almost continuously. We perform this discretisation as a modelling simplification to allow the subsequent application of an optimisation routine to the supply functions. Formally, for all K generating companies, let each generating plant be a standard size equal to PS and let the price range be discretised into N equal sized price ranges indexed from 1 to N, which we will refer to as price bins. If the price range of each bin is also a standard size and equal to BS, then the price range of bin i is (i-1)*BS to i*BS. Thus, we can define the lower price of bin i as lpi (i 1) * BS and the upper price as upi i * BS . In this representation we view generating capacity as being strategically allocated each day to individual price bins and define the capacity that company k has allocated to bin i as c k ,i (MW) which will be a multiple of PS, the standard size of each generating plant. With this representation, a piece-wise linear supply function can be defined for each company k by forming linear sections of the function between the lower price lpi and the upper price upi of each bin i using the capacity c k ,i that company k has allocated to the bins. Figure 1 gives an illustration of this piece-wise linear representation. Having defined the price range ( lpi to upi ) that each linear section of the supply function exists between, for each company k we can define the starting capacity sc k ,i of each bin i as i 1 sck ,i ck , j j 1 and the ending capacity ec k ,i of each bin i as i eck ,i ck , j j 1 We can now define the linear section of company k’s supply function (price pk as a function of quantity q k ) existing in the prices range lpi to upi and the cumulative capacity range sc k ,i to ec k ,i as 6 upi lpi up lpi pk * q k lpi i * sc k ,i (1) c k ,i c k ,i As we make the constraint that the price bins are the same uniform size across all companies, the process of generators submitting their supply functions to the market can be simplified to each company notifying the market operator of the number of generating plants (quantity) it wishes to allocate to the different price bins. With the submission of the bids by each generator, a system supply function can be created as the aggregation of all bids and given demand, the market price (known as the system marginal price SMP ) can be calculated as the intersection of the demand function with the system supply function. This market is a uniform auction, with the same price being paid to all companies who generate in a given period t. Therefore, equation (1) can be inverted to attain the quantity an individual company is required to generate in a given period, such that if the system marginal price SMPt lies in the range ( lpi to upi ), then the quantity q k ,t company k will generate in period t will be given as up lpi c k ,i q k , t SMPt lpi i * sc k ,i (2) c k ,i upi lpi 2.1 Supply Function Optimization Using this representation of supply functions, we model each company optimising its next day profit, given its conjectures about its competitors by modifying its own supply function. In addition to the profit a company may make from electricity sales in the pool, we also model additional revenue earned from contracts that a generator may engage in with buyers of electricity. Known as two-way contracts for differences (CfDs), these purely financial instruments are used in the industry to hedge price volatility in the pool and take the form of a time period t, a strike price spt and volume x t . If during the time period, the market price ( SMPt ) is greater than the strike price, then the generator will pay the buyer the difference between the market price and the strike price multiplied by the volume. If the market price is lower than the strike price, then the buyer pays the difference between the strike price and the 7 market price multiplied by the volume under contract. Formally, we define the profit for company k in period t as k ,t q k ,t * SMPt Ck q k ,t spk ,t SMPt x k ,t (3) Where Ck is the total cost function of company k. As we have chosen to model the supply functions that are submitted by each generating company remaining constant across the 48 half hour periods of the day1, this results in the value of the total objective function for each company k being given as 48 k q k ,t * SMPt Ck q k ,t spk ,t SMPt x k ,t t 1 An iterative optimisation routine is used to calculate the best response for each generating company, given the supply function conjecture it holds about its opponents2. 2.2 Dynamic Behaviour and Bounded Rationality Simulating the interaction of generating companies using this computational model produces interesting dynamic behaviour. For the demand and marginal cost data that we have analysed, when each generating company is modelled fully optimising its supply function, we do not find convergence to a unique supply function. Instead, repeated cycling is observed in the supply functions in a manner similar to the Edgeworth cycle found in models of capacity constrained price competition. This should not be a surprising result. Although this model is one of supply function competition, we do not model the capacity commitment decision, rather we model pricing of individual plants that have already been committed. Thus, our model is one of price competition among many plants. 1 In the actual operation of the market, the generating companies must remain committed to a fixed set of prices for each plant for the entire day, although they submit a schedule of plant availabilities, allowing their supply function to differ across the day. 2 On every simulated day, a number of optimisation iterations are perform on each company’s supply function. On each iteration the routine performs an exhaustive search of the plants owned by the generator to find the one plant that when moved to a different price bin, will increase the value of the objective function the most. If there are N price bins then on each iteration a (N-1)N search is performed, with each plant being sequentially tried in all the other price bins to evaluate the expected change in the objective function. The move that produces the highest change is one that is performed when the search is complete, before moving to the next iteration. Each day, this optimization process is performed for every generating company before the new supply functions are submitted to the market. 8 From a game-theoretic point of view, the identification of Edgeworth cycles would indicate the lack of pure strategy Nash equilibrium, and this usually leads to the consideration of mixed strategy, Nash equilibria. Apart from being analytically complex and computationally problematic to calculate3, the idea that strategic interaction of this sort can be thought of in terms of mixed strategies is also intuitively unappealing. To quote Rubinstein (1991, p922) “One of the reasons that mixed strategies are popular in both game theory and economic theory, in spite of being so unintuitive, is that many models do not have an equilibrium with pure strategies. However, the nonexistence of a solution concept in pure strategy does not necessarily mean that we should look for stochastic explanations. … Expanding the model or changing the basic assumptions are alternatives which the modeller should consider at least as favorably as mixed strategies.” One plausible behavioural assumption is that of bounded reasoning on the part of the generating companies. Rather than assuming they will fully identify the optimal response, or formulate a mixed strategy, either explicitly or otherwise, we have assumed limited optimising behaviour, modelling each generating company seeking to incrementally improve its profitability by changing the price of only one or two of its plants each day. Thus, assuming companies conjecture that their opponents will not modify their supply function, this seems to be quite realistic as only small movements in the supply functions do actually occur from day to day. Thus, the fact that this type of process has been observed in practice over several years indicates that the players are not being so obviously mislead by their conjectures, and indeed this approach is consistent with the basic Cournot conjecture of best response. 3. Model Specification 3 Although the calculation of mixed strategies is a feasible option for the simplified strategic setting of companies competing with just one price (Kreps and Scheinkman 1983), in electricity markets where there may be in excess of 100 generating units such analytical calculations are at best complex. Another approach to the problem of calculating mixed strategy equilibria would be to tabulate the payoff matrix of each company and then numerically determine the mixing probabilities, where each non-decreasing supply function would be an individual strategy. This approach is infeasible as even for a company with a small number of plants, the number of non-decreasing supply functions it could formulate is vast. For example, a company with 25 generating units and 60 different possible prices per plant has ((6025 60) / 2) 60 1.42 *1044 different non-decreasing supply functions that it could bid into the market. Even with removal of dominated strategies, this problem is likely to be too large to be solved numerically. 9 In the analysis that follows, we will be concerned with modelling competition in the England and Wales pool between the three largest fossil fuel generating companies, namely National Power, PowerGen and Eastern Group4. Between them these companies controlled 63% of the all generating plant in this market (NGC 1998), but since the market price was set by the marginal plant in each demand period, and these companies owned most of the marginal plant, they set the market price in excess of 80% (OFFER 1998). 50 40 30 Price (£/MW/hr) 20 10 0 0 5000 10000 15000 20000 25000 30000 35000 40000 Cumulative Capacity (MW) Figure 2: Aggregate short-run marginal cost function for National Power, PowerGen, and Eastern Group 3.1 Estimates of Short-Run Marginal Costs We have estimated the short-run marginal cost of generation for the plants owned by National Power, PowerGen and Eastern Group using data provided by these companies in environmental reports that generating companies in the UK are required to produce annually, by the UK government. In these reports, there is a requirement to provide data on the amount of fuel that individual plants used, plus data on the amount of electricity that was produced. Using this data it possible to calculate thermal efficiencies for each plant, which combined with fuel cost data 4 When required by the regulator in 1994 to divest themselves of a combined total of 6GW of plant (OFFER 1994), both National Power and PowerGen divested their plants to a third company which was Eastern Group. 10 can be used to give an estimate of short-run marginal costs5. Figure 2 shows the aggregation of National Power, PowerGen and Eastern Group’s marginal cost functions (as estimated for 1998) into a system marginal cost function. 3.2 Demand In the following analysis of the England and Wales pool, demand elasticity is represented by locally linear functions. Although there have been several attempts to estimate the demand response to changing prices of electricity (Taylor, 1975; Branch, 1993), rather than relying on one of these estimates, we analysed several scenarios for the slope of the demand function, ranging from -10 MW/£ to -100 MW/£ to determine the influence of demand response that may exist over the short and longer-term time horizons on the generators’ability to exercise or sustain market power. We define 48 demand functions that exist in one day by subtracting, from the forecast of demand given by the market operator6, the demand that would be met by companies in this market other than the three main prices setting generators. The plants that fall into this category are nuclear and many of the combined cycle gas turbines (CCGTs) owned by independent power producers (IPPs) that have entered the market since privatisation7, plus imports via interconnectors from Scotland and France and demand that is met by two pumped storage hydro dams. Thus, after adjusting demand in this manner, we are left with the residual demand for each period of the day to be met by National Power, PowerGen and Eastern Group. 5 We assumed delivered fuel prices of £0.13/therm for gas, £1.30/GJ for coal and £1.85/GJ for heavy fuel oil. 6 In the operation of the England and Wales market, demand is forecast for the following day by the market operator rather than requiring the submission of demand functions, and it is this forecast of demand that is used to set prices. Thus, on the day in which the market operates, there can be no demand response to the level of prices. We have chosen to model demand functions in an attempt to capture the demand response that may be observed over differing time lengths. 7 Nuclear plants adopt this strategy as they do not wish to be scheduled on and off as demand varies on the day since they cannot rapidly change their output nor restart quickly once they have been desynchronized from the gird. IPPs bid in such a manner as many have entered the market by signing take-or-pay fuel contracts, plus off-take contracts for their output, which results in them wishing to burn the gas they have contracted for, creating an incentive for them to attain high load factors. 11 We define the locally linear demand functions by assuming a particular slope and calculate the intercept by assuming the functions cross the system marginal cost function (the aggregation of the three generators plants) at the observed demand level8. 4. A Comparison with Continuous Supply Function Equilibrium In this section we compare the supply functions obtained from the computational model with the equilibrium in continuous supply functions calculated using the approach developed by Klemperer and Meyer (1989). As Klemperer and Meyer found, there exists a continuum of equilibria for the case of finite demand, with the range being defined by the location of the supply function in the maximum demand period. The highest equilibrium exists where the supply function passes through the Cournot outcome and lowest through the perfectly competitive outcome in this period. We follow Green and Newbery (1992) and analyse the highest possible equilibrium, as this gives an indication of the possible markup above marginal costs that may be obtained by profit maximising firms without the discounting of future profits. In practice, the calculation of the continuous supply function equilibrium has to be solved numerically9 and suffers from instability problems for asymmetric firm sizes. Thus, we have chosen to compare the two using a stylised market concentration of 3 symmetric companies, created by dividing the plants of the three main price setting generators. An additional constraint in the calculation of supply function equilibrium is the requirement of the marginal cost function to be continuous. Thus, for purpose of comparison in this section we have approximated the function as linear for both approaches. We have assumed that plants will not be available at their maximum registered capacities throughout the year, due to forced and unforced outages, and in the specific winter day that we have analysed, we assume 90% availability. Although the issue of strategic capacity with- holding as an instrument of market power in electric power pools has been addressed by other 8 We could have defined the functions passing through the demand and price points forecast by OFFER or NGC, although these prices consist of payments made in addition to those that we model such as for start-up and no-load, thus do not equate well with our model. 9 Unless specific constraints are made on the functional form of the supply function and marginal cost function, such as linearity (Green 1996). 12 authors (Wolak and Patrick, 1997; Rudkevich et al. 1998), we have just concentrated on strategic price bidding in this study10. 70 60 Avg. Agent-Based Supply Function Supply Function Equilibrium Highest Demand Function Modeled Marginal Costs 50 40 Price (£/MW/hr) 30 20 10 0 0 5000 10000 15000 20000 25000 30000 Cumulative Capacity (MW) Figure 3: System supply functions from the equilibrium approach and the computational approach with 0% contract cover across all companies, plus the short-run marginal cost function and linear demand function for the highest demand period with a slope of -125 (MW/£). For the computational approach we use a uniform plant size of 100MW and a price-bin size of £1.5/hr with 60 individual bins, giving a possible price range of £0-£90/hr. Although this may be seen to artificially constrain the price, as in the actual market the price is capped by the regulator at over £2000/hr, in both models the Cournot price defines a lower maximum, where the supply functions pass through the highest demand function. In both approaches, we use the same demand functions, which have a slope of -125 MW/£, and are defined using demand data given by the market operator for the 1st of December 199711. For the computational approach, we simulated 600 days of interaction. We discarded the first 100 days to allow the supply functions to move away from their initial conditions, which were linear functions starting at the companies marginal cost at negligible output to the Cournot price at the 10 Following the first pool price enquiry (OFFER 1991), a number of measures were introduced to reduce the potential for strategic capacity with-holding, eg availability monitoring, redeclaration restrictions and an 8-day delay rule for computing the loss-of- load probability. 11 The demand for this day is given by the National Grid Company, who act as both market and system operator, as typical demand for a winter day of that year (NGC 1998). 13 Cournot output12. The remaining 500 supply functions were averaged for comparison with the numerically calculated equilibrium. 50 45 40 Avg. Agent-Based Supply Function Supply Function Equilibrium Highest Demand Function 35 Modeled Marginal Costs 30 Price 25 (£/MW/hr) 20 15 10 5 0 0 5000 10000 15000 20000 25000 30000 Cumulative Capacity (MW) Figure 4: System supply functions from the equilibrium approach and the computational approach with 50% contract cover across all companies, plus the short-run marginal cost function and linear demand function for the highest demand period with a slope of -125 (MW/£). The results from the two approaches are shown in Figures 3 and 4, along with the linear system marginal cost function and the demand function in the maximum demand period for the case of 0% aand 50% of contract cover13. These figures show that the system supply functions that result from the two very different approaches are very similar, although the computational approach produces a supply function with plants being priced at lower levels than does the equilibrium approach. The results of the approaches differ slightly for two reasons. First, as the computational approach uses discrete supply functions, there exists an incentive not present when using continuous functions, for companies to compete, with their individual discrete plants, by undercutting the price at which rivals price their plants. Figure 1 demonstrates how discontinuities of this form may appear in the piece-wise linear supply functions. Second, as we 12 From the experiments that we have performed, we do not find the initial location of the supply functions affects the long term behavior. 13 In this industry, contract cover normally refers to the percentage of attained generation that was contracted, with this volume being agreed before the market operates. Similarly, for the pool models we compare a contract volume has to be determined in advance of developing the supply functions. In both approaches used here, we have taken the percentage of contract cover to be the percentage of generation that would be obtained if all companies bid at the perfectly competitive level. With this quantity equating to the contract volume x k ,t in equation 3. 14 use a specific demand day with only 48 demand periods that have significant differences between demand levels, the supply functions that result from the computational approach, unlike those from the supply function equilibrium, will not be optimal at every possible observation of demand. Thus, the computational supply functions may have significant discontinuities as a result of them becoming profiled toward the levels of realised demand during the simulation. These discontinuities are still visible in figures 3 and 4 even after the averaging process that we perform. Nevertheless, with these two practical points in mind, the calibration of the computationally-derived average supply function with that of the supply function equilibrium is reassuringly close, and give grounds for confidence in applying it to real circumstances too complex for the analytical approach. 5. Analysis of the 1999 Plant Divestment Issue in England and Wales The 1999 divestment requirement is the second such attempt to reduce market power in England and Wales, with OFFER previously requiring National Power and PowerGen to divest 4GW and 2GW of capacity respectively in 1996 (OFFER 1994), after a series of reviews into the high level of pool prices (OFFER 1991; OFFER 1992; OFFER 1993). OFFER has also indicated that the 1999 divestments should not be made to just one company, unlike 1996 when both companies leased their plants to Eastern Group14. The belief then was that three main price- setting generators would be significantly better than two, but as it turned out, this was a mistake and prices did not go down (OFFER 1998b). For the 1999 divestment, both PowerGen and National Power have agreed to divest themselves of 4GW of coal-fired generating capacity and have also agreed to forego the variable earn-out payments they receive from Eastern Group (PowerUK, 1999). We have chosen to analyse two possible divestment, plus an alternate option of not requiring plant divestments but abolishing the earn-out payments by Eastern . The two divestment options we have analysed are that of National Power and PowerGen both selling either 25% or 50% of 14 These deals with Eastern Group actually took the form of a long term leases with fixed payments plus a variable payment, known as an earn-out payment. These per megawatt hour payments that Eastern Group made to National Power and PowerGen effectively increased the marginal cost of the divested plants that Eastern Group operated by £6/MWhr. 15 existing capacity coal-fired capacity15 to two separate companies, creating five main price setting generators. We have modelled plant divestment by allocating either 25% or 50% of each of the existing companies coal plants to the newly created companies. Following the same method employed in the comparison (section 4), in this analysis we again solve for the Cournot outcome to define the point where the supply functions cross the demand function in the highest demand period16. We analyse the final 500 supply functions after simulating the interaction of the companies for 600 days. The metric with which we have chosen to compare the effectiveness of the divestment options is the average percentage bid above marginal cost in each period where price was set17. This measure has been used in addition to the Hirschmann-Herfindahl index (HHI) by Rudkevich et al. (1998) in their analysis of in Pennsylvania. Using residual demand functions calculated following the approach described in section 3.2, we simulated the interaction of National Power, PowerGen and Eastern Group, plus the two companies we assume to be created after the divestment process. The 48 demand functions are constant throughout the simulations. To account for the change in demand levels across the seasons of the year and the associated change in the availability of generating capacity, we have simulated the interaction of firms in three demand seasons. Using demand data given by the National Grid Company for a typical winter and summer day in 1997, we have averaged the two creating demand for a typical spring/autumn day. As plants are not available to generate continuously throughout the year due to planned and unplanned outages, we have modelled this by reducing the available capacity of all plants, including the baseload capacity, by 10% , 20% and 30% in the winter, spring/autumn and summer seasons respectively. 15 The 4GW of coal plant that each company has actually agreed to divest equates to 43% and 34% of PowerGen and National Powers total coal-fired plants respectively and thus both are in the range of the two scenarios we have analysed. 16 Although we have no theoretical justification for using the Cournot outcome in the maximum demand period, we choose this point to be consistent with the theory of continuous supply function equilibria applied to analyzing this market by Green and Newbery (1992) with which we have shown a close similarity of results in the previous section. 17 By analyzing 500 days with 48 demand periods per day the quoted figures for average percentage bid above marginal cost are an average of 24000 recorded prices. 16 Avg. Percentage Bid Above Marginal Cost 70% Existing 60% Existing (No Earn-Out) 25% Divested (No Earn-Out) 50% 50% Divested (No Earn-Out) 40% 30% 20% 10% 0% 75% 80% 85% 90% 95% Percentage of Contract Cover Figure 6: Average percentage bid above short-run marginal cost using demand functions with slopes of -50 MW/£. To give a clearer understanding of the influence that contract volumes and the slope of the demand functions have, we have analysed several scenarios of these variables for each divestment option. Historically, we know that generators have been close to 90% contracted (MMC 1996) and that the majority of these contracts have taken the form of contracts for differences (CfDs), so the range of 75% to 95% of contract cover that we have analysed should include existing contract levels. We have analysed slopes of the demand functions in the range of -10 MW/£ to -100 MW/£, with the lower value being indicated to us by the demand forecasting unit of the National Grid Company as representative of the daily demand response. Conversely, the higher value can be thought of as a longer term demand response. The first result that is apparent from figures 6 and 7 is that as the percentage of contract cover tends toward 100%, the percentage bid above cost approaches 0, which is a confirmation of the result of Green (1999). Also, the abolition of the earn-out payments on its own appears to have a significant influence on reducing the ability of the existing companies to raise prices, and raises further evidence to question the general rationale of leasing as a divestment mechanism. 17 Avg. Percentage Bid Above Marginal Cost 35% Existing 30% Existing (No Earn-Out) 25% Divested (No Earn-Out) 25% 50% Divested (No Earn-Out) 20% 15% 10% 5% 0% 75% 80% 85% 90% 95% Percentage of Contract Cover Figure 7: Average percentage bid above short-run marginal cost using demand functions with slopes of -100 MW/£. However, the most striking result in these figures is that the main reduction in markup comes from the creation of the 5 companies, with the difference between 25% or 50% divestment being significant, but relatively less so. We may think that this is further vindication of the “5 company rule” that has often been proposed as the basis for efficient competition in wholesale electricity markets (eg Green and Newbery, 1992). However, more details of the numbers behind these figures, and the variants for lower elasticities can be seen in Tables 1.1-1.4. For the lower (short- term) elasticity gradient of 10MW per £/MWh (which is close to the heuristic used by NGC for daily operational forecasting), we see that a substantial difference does exist between the scenarios for 25% and 50% divestment, and indeed for the 90% level of contracting which is generally assumed to have been taking place in the England and Wales Pool, we still see prices at sunstantially above marginal cost. The conclusion of this, therefore, is that with the low price elasticity which has been estimated in the short -term in the short term, the proposed 1999 divestment of about 40% of National Power and Powergen’s plant will still leave market power with generators to the extent that profit maximising behaviour on their part could result in prices in the short-term at more than 20% above short-run marginal costs. In the longer term, the higher elasticity assumption, plus the attractiveness of new entry, would erode the sustainability of this profit margin. In looking at Table 1.1 for the existing market structure, and the low price 18 elasticity assumption, the high level of profits are model-based assessments of what could be achieved by the generators profit-maximising in the absence of regulatory constraint. The history of the pool of England and Wales from 1990 to 1998, however, has been on of persistent tight regulatory control and threat, to preclude the full attainment of these rents. E Contract Cover 75% 80% 85% 90% 95% Demand -10 * * * 97.53% 53.99% Slope -50 64.78% 48.11% 36.06% 24.87% 9.65% -100 34.26% 27.46% 20.04% 12.43% 4.27% Table 1.1: Average percentage bid above short-run marginal cost for the existing market structure. Contract Cover 75% 80% 85% 90% 95% Demand -10 * * * 78.62% 34.55% Slope -50 55.73% 41.72% 29.11% 16.74% 6.33% -100 29.66% 22.58% 15.19% 8.99% 2.82% Table 1.2: Average percentage bid above short-run marginal cost for the scenario of abolishing earn-out payments Contract Cover 75% 80% 85% 90% 95% Demand -10 * * * 32.91% 7.35% Slope -50 27.97% 20.33% 11.47% 7.31% 2.58% -100 13.46% 9.62% 6.27% 3.82% 2.19% Table 1.3: Average percentage bid above short-run marginal cost for the scenario of National Power and PowerGen divesting 25% if their coal plants. Contract Cover 75% 80% 85% 90% 95% Demand -10 * * 19.57% 5.19% Slope -50 18.18% 12.22% 7.15% 4.21% 2.30% -100 9.04% 6.55% 3.65% 3.08% 2.00% Table 1.4: Average percentage bid above short-run marginal cost for the scenario of National Power and PowerGen divesting 50% if their coal plants. 6. Summary and Conclusions 19 This paper has sought to demonstrate the practical value of a new computational approach to deriving outcomes for imperfect competition in power pools where generating companies bid supply functions. The approach is consistent with the previously applied analytical method for the simplified circumstances in which the latter can be applied, but has the potential to model much more realistic situations with respect to the discontinuous supply functions, non-linear elasticity, and asymmetric ownership structures which we see in the industry Using this approach, we find that, even with the 1999 divestment proposals, the England and Wales electricity market may continue to experience high prices and/or require a continuation of the tight regulatory level of price monitoring and control. The first round of divestment (17% in 1996), failed to have any impact on prices and it looks again as if the second round (40% in 1999) may still leave the opportunity for profit maximising behaviour in an imperfect market to result in prices substantially (about 20%) above short-run marginal costs. One of the lessons here is that the price-setting mechanism in markets which have very small short-term price elasticity can provide market power at levels of concentration which are usually seen as quite acceptable in other industries. For example, the 1999 divestment proposals would reduce the HHI index of concentration (on generation capacity) from about 1,500 to 1,200; both of which are below the 1,800 benchmark generally adopted by regulatory bodies. Encouraging forward contracting can appear to mitigate this, but as it is perceived risk that encourages contracting, and risk premia that inflate contract prices, then generators’ rents may just be transferred to the forward markets. The second option of encouraging greater elasticity is also appealing, but demand-side participation in wholesale electricity trading only seems to become significant at high prices. All of which seems to suggest that, whilst the second round of divestments are substantial, and may in the long term be sufficient, they will not fulfil the ambition of OFFER (1998) to ensure a competitive market in the short term, and that a further period of regulatory price management will be required. 20 References Andersson, B., and Bergman, L., (1995) “Market Structure and the Price of Electricity: An Ex Ante Analysis of the Deregulated Swedish Electricity Market” The Energy Journal 16(2): 97-130 Branch, E. R. (1993) “Short Run Income Elasticity of Demand for Residential Electricity Using Consumer Expenditure Survey Data.” The Energy Journal 14(3): 111-121. DTI (1998), “Conclusions of the Review of Energy Sources for Power Generation.” Department of Trade and Industry. HMSO Cm4071. Green, R. (1996). “Increasing Competition in the British Electricity Spot Market.” The Journal of Industrial Economics 44(2): 205-216. Green, R. (1999) “The Electricity Contract Market in England and Wales.” Journal of Industrial Economics. Green, R. J. and D. M. Newbery (1992). “Competition in the British Electricity Spot Market.” Journal of Political Economy 100(5): 929-953. Klemperer, P. D. and M. A. Meyer (1989). “Supply Function Equilibria in Oligopoly Under Uncertainty.” Econometrica 57 (November): 1243-1277. Kreps, D. and J. Scheinkman. “Quantity Precommitment and Bertrand Competition Yield Cournot Outcomes.” Bell Journal of Economics. 14:326-337, 1983. Littlechild, S. (1996), Regulation of the Electricity Market, Speech to the IEA Third Annual Electricity Conference, available from the library at OFFER, Hagley House, Birmingham B16 8QG MMC (1996) National Power and Southern Electric, Monopolies and Mergers Commission, HMSO, PO Box 276, London SW8 5DT NGC (1998). 1998 Seven Year Statement. National Grid Company, National Grid House, Kirby Corner Road, Coventry, CV4 8JY, England. OFFER (1991). Pool Price Enquiry. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. OFFER (1992). Review of Pool Prices. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. OFFER (1993). Pool Price Statement. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. OFFER (1994). Decision on a Monoplies and Mergers Commsion Reference. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. 21 OFFER (1998a). Review of Trading Arrangements - Background Papers 1. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. OFFER (1998b). Report on Pool Price Increases in Winter1997/98. Office of Electricity Regulation, Hagley House, Hagley Road, Edgbaston, Birmingham, B16 8QG. Rubinstein, A. (1991). “Comments on the Interpretation of Game Theory.” Econometrica. 59(4): 909- 924. Rudkevich A., M. Duckworth and R. Rosen. (1998). “Modeling Electricity Pricing in a Deregulated Generation Industry: The Potential for oligopoly Pricing in a Poolco.” The Energy Journal. 19(3): 19-48. Taylor, L. D. (1975). “The Demand for Electricity: A Survey.” Bell Journal of Economics 6(Spring): 74- 110. Wolak, F. A. and R. H. Patrick (1997a). “The Impact of Market Rules and Market Structure on the Price Determination Process in the England and Wales Electricity Market.” University of California Energy Institute. Working Paper No. PWP-047. Wolak, F. A. and R. H. Patrick (1997b). “Estimating the Customer-Level Demand for Electricity Under Real-Time Market Prices” Mimeo, Dept. Economics Stanford University. 22