Select Committee on Environmental Audit Appendices to the Minutes of Evidence


Annex A

  This annex examines fundamental issues in the NETA design. The issues affect all players in the Balancing Mechanisms, but they have particularly disadvantageous effects on small and renewable generators.

BALANCING PRICES ARE NOT COST REFLECTIVE

  Contrary to the express intent stated in a number of Ofgem and DETR consultation documents, the System Buy Price and System Sell Price are not calculated in a manner that is designed to recover the costs of energy balancing from those parties that have caused the costs to be incurred. The calculations are complex and detailed, but it is possible to identify a number of fundamental problems with the design of the calculations.

1.  Half hourly balancing

  The design of the Balancing Mechanism requires that each party balances its contractual position over an integrated half-hourly period. There is no encouragement for a party to balance at a more disaggregated level and the traded markets also operate at the level of an integrated half hour. The Grid Code prevents portfolio players from deliberately offsetting imbalances within their own portfolios. Consequently, there is no incentive and little opportunity for a party to balance on a minute-by-minute basis. However, NGC must balance the system virtually instantaneously. To achieve this instantaneous balancing, NGC purchases services from both generators and the demand side. Some of these services are bought through long-term contracts, the costs of which may enter balancing prices through BSAD (information provided by NGC under the terms of the Balancing and Settlement Code), but frequently, balancing will be achieved by the acceptance of bids or offers in the Balancing Mechanisms. It is, therefore, perfectly possible for all parties to have a contractual and physical position that is in balance over an integrated half-hour, but for NGC to incur significant costs in balancing the system within that half hour.

  Such short-term balancing actions are more properly thought of as system balancing rather than energy balancing. Other elements of system balancing include actions taken by NGC to overcome transmission constraints. For the balancing prices to be reflective of the costs of energy balancing (costs incurred because parties are out of balance over the course of a half hour) would require that all system balancing actions are excluded from the calculation of the prices. Work undertaken on behalf of the Balancing and Settlement Code Panel indicates that in practice, the exclusion of system balancing actions from the Balancing Mechanism prices is impossible.

  Many of the higher System Buy Price spikes have been caused by the inclusion in the price calculation of system balancing actions, particularly offers accepted from pump storage plant. A number of modifications have been put forward to try to ameliorate this problem, but none could ever be entirely effective and all are necessarily crude in their effect. Essentially, it is impossible to identify the costs that are incurred solely because of imbalances as measured over the course of an integrated half hour.

2.  Individual imbalance versus system imbalance

  There is a very high probability that the imbalance positions of individual parties will be offsetting to some degree. Consequently, it is highly probable that summing the absolute values of the errors of individual parties will give a number far in excess of the actual level of imbalance on the system. Indeed, imbalance on the system may be either positive or negative in a particular half hour, but cannot be both, while some individual parties are likely to have positive balances and others are likely to have negative imbalances. Since all parties settle their imbalances at either System Buy Price of System Sell Price and these prices are calculated as the average cost per MWh of the appropriate subset of actions required to offset the aggregate energy imbalance on the system, it is most unlikely that the amount of money recovered from parties would equal the costs incurred by NGC. Indeed, these prices are not the mechanism by which NGC's costs are recovered, but are simply a means of disturbing funds between parties to the Balancing Mechanism. The redistributive mechanism means that a party in perfect balance in a half hour may either gain or lose from the redistribution.

3.  Dual Pricing

  The difficulties in constructing prices that are cost reflective, as note in one and two above, are exacerbated by the system of having two different prices for short and long positions. It is frequently the case that one of the prices is set based upon a very small volume of acceptances and that these acceptances are almost certainly undertaken for system balancing reasons rather than for energy balancing (see comments on the Balancing Reserve Level in Annex II). This has led to the creation of a large spread that fluctuates unpredictably between half hours. The size of the spread and the unpredictability of the imbalance prices, Particularly the System Buy Price, have actually discouraged parties from balancing since the cost minimising approach is to maintain a long position.

  A system of dual pricing, particularly where the balancing prices are not cost reflective, is very disadvantageous to small players. Small players inherently have greater difficulty in balancing because of the greater forecast error associated with small portfolios and also because of the difficulty in trading out small imbalances in the traded markets—see "Ability to balance" below.

  Because the spreads between System Buy and System Sell Prices are penal rather than cost reflective, they have serious impacts on any party with a greater than average level of imbalance. This will discourage the entry of more suppliers and more generators, but will be particularly harmful to generators with unpredictable outputs. NGC has estimated that if most of the Government's target, 10 per cent renewable generation by 2010, were to be met by wind energy and all of the wind turbines were held in a single portfolio, then the expected level of forecasting error for that portfolio would be 6 per cent. The reported spread for June and July was £30/MWh, so that a 6 per cent forecast error would be expected to cost the wind portfolio £1.80/MWh. Since the profit margin on wind generation is likely to be less than £0.5/MWh, the effect of the imbalance charges is to wipe out all the profit in the wind portfolio and create a significant loss. It is not an adequate response to this problem to state that NETA is designed to reflect the additional value attributed to reliable generation. The cost of being out of balance as currently calculated within NETA is penal rather than cost reflective and is itself evidently having a serious effect on those generators that the Government states that it wishes to encourage.

AGGREGATION OF IMBALANCES

  The Government has suggested in its Consultation Document that effective consolidation services are necessary in order to encourage small and renewable generators within NETA. This statement is correct, but as noted in "Balancing prices are not cost reflective" (3) above, it will not solve the problem for wind generators and actually illustrates one of the fundamental fallacies underlying the NETA design.

  If consolidation were to occur, then the sum of imbalance payments would decrease (this is the stated aim of consolidation), but the actual costs incurred by NGC in balancing the system would be entirely unaffected. This illustrates simply that the amount of money that is recovered through the imbalance charges reflects the industry's structure rather than the cost of balancing the system. Since it is a fundamental stated principle of the NETA design that it is intended to target costs to those that have created them, it is paradoxical that the level of such charges can be affected simply by how effective the parties are in aggregating individual small generators into a single Balancing Mechanism account. The acknowledged need for consolidation services fundamentally undermines the pretence that the NETA system targets costs at those that have created them.

  The failure of logic underlying the design affects all the parties in NETA, but the effects are particularly noticeable for small players and renewable generators. Indeed, large integrated portfolio players may actually enhance their profitability at the expense of small players through the redistribution of the Balancing Mechanism charges.

ABILITY TO BALANCE

  The intent of the NETA system is that each party would balance their own contractual positions. An economic market place would encourage parties to balance to the extent that it is economically justified. However, the practical effect of NETA, and indeed, on occasions the stated intent set out in Ofgem's documents, is to force parties to balance. This appears to be based upon the belief that individual parties have better information and are more able to balance the system effectively than the central actions of the system operator. While it is desirable to limit the responsibilities and actions of the grid operator monopoly as far as is practicable and economic, the concept of forcing parties to balance is fundamentally flawed for three reasons.

  First, although individual generators have better information about their own plant than NGC, they do not have information about other generating stations. This leads to inefficient decisions being taken where a generator will balance its position using its own equipment when a more efficient solution would have been to effect that balancing on the equipment of another generator.

  Secondly, simple probability theory would suggest that the individual demand forecasts for the vast majority of suppliers will have higher forecast errors than would be associated with a central forecast made with the same degree of expertise. This is particularly true in the near term, since NGC has access to operational metering in real time. It is neither economic nor practicable for each individual supplier to have available and act upon the same operational information available to NGC.

  A third element that particularly impacts upon small players is the significant administrative costs and practical difficulties of trading out of imbalance. For example, a 10MW generator with a maximum expected contractual imbalance in a half hour of 1MW would find it inordinately expensive to maintain a trading capability in order to trade out that imbalance. It has been independently estimated that maintaining a 24-hour, seven day a week trading capability would cost a party around £750,000 per annum. This implies an average cost of approximately £85 per hour. With an average spread between imbalance prices of £30/MWh, the expected average level of imbalance would need to be at least 3MWh (6MW or 60 per cent) before having the trading capability could be justified economically.

  It should be clear from the example above that asking small players to balance precisely is an economic nonsense, but the situation is made worse by the lack of liquidity in the traded markets for small volume trades. There is an irreducible overhead associated with each trade that discourages a large number of small trades and leads large players to favour undertaking a small number of large transactions. Further, it is unlikely that large players would be active in the market trying to trade out of small levels of imbalance, since such small imbalances would be well within the margin of error on their forecasts.

NETA ADMINISTRATIVE COSTS

  The NETA mechanism is considerably more complex that the Electricity Pool and requires a great deal more involvement from all parties. This necessarily forces small players to incur proportionately very large administrative costs and hence, to disadvantage them in comparison to large players. In the Report to the Dti on the Review of the Initial Impact of NETA on Small Generators (August 2001) Ofgem notes that small generators have incurred average total cost increases of 16 per cent and that fuel costs had risen by 14 per cent. These cost differences were calculated by comparing the first two months of NETA operation to the equivalent period in 2000. Ofgem draws no conclusions from the fact that total costs have risen faster than fuel costs. However, small generators have two areas of costs that predominate: fuel costs and the costs of servicing debt. Since it has been noted that fuel costs have risen by 14 per cent and since it is likely that the costs of servicing debt will have fallen over the period in question, we must conclude that other small cost categories have risen very fast in order that the average increase may be greater than the increase in fuel costs. Although one may only speculate, a likely cause of these increased costs is the massively increased administrative burden brought about by the introduction of NETA.

MARKET POWER

  Since the underlying principle of NETA is to force parties to contract to balance their physical positions, and since there is an acknowledged need for consolidation services to reduce the imbalance exposure and administrative costs of small players, then there should be an expectation that small players will be forced to enter into contracts with large players. Consequently, parties with significantly different degrees of market powers are being pitted against each other in the commercial arena. The small parties can be expected to be seriously disadvantaged in this context and this presumption appears to be borne out by practical experience.

  The Electricity Pool had a single price available to all so that small players were able to trade on the same terms as large players and were not forced to enter into contracts on disadvantageous terms. The large and penal price spread within NETA means that this useful facility is no longer available to small players. Therefore, for this reason and because of the large and penal imbalance charges, NETA enhances the competitive position of large players and is encouraging a further concentration of generation and supply in an increasingly small number of companies.

CONCLUSIONS ON NETA DESIGN ISSUES

  The electricity production and delivery system is subject to significant economies of scale and has an innate tendency towards monopoly. Indeed, prior to privatisation many thought that the provision of electricity was a natural monopoly. The design of the Electricity Pool was explicitly targeted at counterbalancing this natural tendency towards monopoly. The design of NETA exacerbates these monopolistic tendencies by imposing penal imbalance charges on the small players and by imposing very high administrative costs.

  Modifications to the NETA trading system have been proposed to ease the burden of small players. However, each of these modifications attempts to deal with the symptoms rather than the cause of the problem and this in itself creates other problems. The problems that small players face in NETA will not be addressed by tinkering around the edges but only by fundamental redesign.



 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 22 July 2002