انتقال قیمت در بازار برق انگلستان: آیا نتا مفید بوده است؟

This paper explores the relationship between domestic retail electricity prices in Great Britain and their determinants in the context of the New Electricity Trading Arrangements (NETA) introduced in 2001. We employ a consistent comparison of wholesale power price series before and after NETA, alongside a difference-in-differences analysis based on using Scotland as a control. Despite NETA's stated intention of reducing wholesale and thereby retail prices, we conclude that its net effect, alongside other developments, instead merely rearranged where money was made in the system.

Our paper explores the relationship between domestic retail electricity prices in Great Britain and their various determinants, focussing on the effect of the New Electricity Trading Arrangements (NETA), a milestone in restructuring the GB electricity market. In common with most advanced economies, the mantra “competition where possible, regulation where not” (Littlechild, 2005) has led to an evolution of processes to inject competition into electricity generation and supply through introducing, encouraging and then reforming markets in these areas; the recent series of reforms relating to NETA started in 2001. The fact that reforms come swiftly one upon another proves a significant intellectual challenge. Our basic question is: Did end users benefit? NETA was of course a reform of the wholesale market structure, aimed (and arguably successful) at reducing wholesale prices, but ultimately the test of this change in arrangements must be the impact on prices for final consumers. In our examination we exploit structural differences between the electricity and gas wholesale markets and in particular institutional differences between England and Wales and Scotland, which operated under a similar regime but without NETA until 2005. Our question has policy relevance much more broadly. The UK has tended to be in the vanguard of reforms to energy industries, an exemplar to others. Yet both here and in Europe more generally, doubts have emerged concerning the benefits of market liberalisation. Thus our analysis is not simply a statistical exploration with some economic interpretation — deregulatory strategies are being pursued worldwide in the electricity industry and there is a live policy interest that has at least a European dimension. Through NETA and BETTA the UK has chosen a particular deregulatory path that other countries may follow. The major unregulated determinant of retail prices is wholesale electricity costs, so it might seem obvious that when wholesale prices fall, so too will retail prices. Yet we show this is not so, because of the implications of the reforms for retail market competition. More recently (e.g. Business and Enterprise Committee, 2008) there has been significant concern that the drivers of retail price rises are somewhat opaque. Their (2008) investigation was prompted by “mounting public concern over gas and electricity prices” (p.5) and both it and the OFGEM (2008) “Supply probe” announced shortly afterwards appear to accept that there are barriers into entering at the retail level, with no supplier of significance outside the big six. Across Europe more generally, the European Commission sector inquiry in 2007 raised concerns about the presence of distorted competition in the energy sector, mostly as a result of high market concentration and instances of vertical foreclosure (European Commission, 2007). Similar issues were also raised by an OECD investigation into the German gas and electricity market (OECD, 2006). Since the early 1990s, the British electricity market has changed substantially. The first major steps, involving vertical disintegration and the introduction of competition in generation together with a Pool mechanism for coordinating generation and supply, have been well studied. Key papers on the British experience include Green and Newbery, 1992 and Wolfram, 1999 and, as Sweeting (2007) points out, competition in generation has been enhanced in the sense that the Herfindahl–Hirschman index in generation has been falling through divestiture, although he finds that the generators were nevertheless able to exercise considerable market power. The wholesale market arrangement instituted in 1990 employed a power pool, where a system marginal price was set for each half hour period of the day by the network operator, based upon matching predicted demand with bids from generators. This system received criticism, for example that it was subject to manipulation by generators, was insufficiently cost-reflective and that it did not contain demand revelation elements (OFGEM, 2002a). As a result, the replacement arrangements known as the New Electricity Trading Arrangements (NETA) were introduced in March 2001. These applied to England and Wales only, whilst Scotland then had a more vertically integrated structure (although its wholesale prices were designed to track those across the border).1 BETTA, a revision to incorporate the Scottish operators, was developed and became live in April 2005. The operating scheme under NETA and BETTA involves bilateral wholesale trades, hence a discriminating price auction, together with a small balancing market. It is fairly clear that, at least in the initial years, wholesale electricity prices fell around the time of the NETA reforms, although the underlying mechanism has been subject to contention (see Fabra and Toro, 2003).2 To our knowledge, the impact on the retail market has not been examined rigorously so far, although significant concerns have been expressed.3 One important feature of the GB market compared to other large economies is the extent to which the retail market is dominated by integrated players. The number of active residential power suppliers has now effectively been reduced to six, who together control over 99% of domestic consumer supply.4 Each has significant generation capacity and each competes in all 14 regions of the electricity industry.5 The drive by retailers to vertically integrate back into generation started in 1998 and has continued.6 As illustrated in Fig. 1, after an initial decline in the big six's share in generation capacity at market opening, by the end of the period they had regained their original total share to the detriment of independent generators. Out of the big six however some companies managed to increase their individual share, such as EDF, SSE and Centrica, while others, such as E.ON and npower, have registered a decline. At the same time the big six were also involved in merger activity at the retail level through the acquisition of former regional suppliers and independent companies. Full-size image (40 K) Fig. 1. Cumulative electricity generation capacity by company, England and Wales (MW). Source: OFGEM. Total includes independents. Figure options Vertical integration, by itself, is likely to reduce wholesale, and possibly retail, prices. In addition to the vertically integrated firms, there are non-integrated generators, including Drax and International Power. Certain supply companies have interests in electricity distribution as well, but these activities are subject to separation and regulatory arrangements and can be viewed as separate. Though an important topic to study, the experiment of introducing NETA cannot be considered as “clean”. Since one of the main tasks of suppliers in the market is, in effect, to absorb risk, the drive towards vertical integration across generation and supply and a simultaneous squeeze of independent supply companies may well in part have resulted from the NETA arrangements, or indeed the knowledge that NETA was to happen. Moreover, the process of deregulating competition in supply was still under way. We tackle this range of effects through a form of difference-in-differences analysis looking before and after NETA's introduction. The controls are (i) the situation in Scotland, where unlike England and Wales, an unchanging wholesale regime was in operation until 2005, spanning the introduction of NETA,7 and (ii) the wholesale market in gas, which had an unchanging regulatory structure over our entire period. In addition, at the retail level we are able to distinguish between effects on customers of non-incumbents (an unregulated market) versus those of incumbents. Our analysis differs from the existing academic literature in many respects. First, the period — our focus is the period of transition between Pool and NETA during which competition in supply was operative. Second, unlike others including Bushnell et al. (2008), our focus is firmly at the retail level. Wholesale prices might be an appropriate focus if the supply activity was regulated, but in the UK it is not.8 Since 1999 retail consumers have all been free to choose their supplier (and since 2002 all prices have been unregulated) and large numbers have switched. Was NETA the hoped-for success? We examine this by making comparison between the (relevant dates within the) two years 1999–2001 in which there was competition in supply but there was a wholesale pool and the period of four years 2001–2005 in which NETA operated, but the picture in Scotland remained unchanged. Extending this to the period after BETTA was introduced would be more problematic, although terminating in 2005 misses out very significant price hikes. There is then no clear endpoint, and the empirical pattern of retail and wholesale prices suggests that this choice may be crucial.9 Also over time, but particularly since 2005, the range of retail tariff arrangements has extended, so that standard retail data series are less representative, in particular, of prices the marginal switching consumer will face. To preview findings, we confirm that NETA succeeded in its primary aim, of reducing margins at the wholesale level, i.e. generation margins. However this did not lead to lower retail prices. Indeed, whilst wholesale margins fell significantly, retail margins rose, we suggest as a side-effect of the introduction of NETA and the forces this unleashed. It seems most likely that both vertical integration and NETA had an impact upon the outcome. Moreover, at least within our period, removing retail tariffs had the effect of allowing prices to rise. The net result is that market restructuring seems not to have reduced retail prices compared with underlying costs, if anything the opposite. As we say in the conclusion, the longer-term policy implications are potentially very significant, although they may be less relevant for other European countries which have retained a pool. Our conclusions about the limited benefits to final consumers are not dissimilar to those reached by Newbery and Pollitt's (1997) cost benefit analysis of the UK electricity privatisation program in the early 1990s. Our analysis also confirms the reasons for concerns which emerged more recently about the lack of competitive market conditions and inability to achieve the expected welfare benefits of the energy market liberalisation and which led in 2008 to the UK regulator's Energy Supply Probe (Ofgem, 2008). In this paper, after describing the data sources in Section 2, we consider an appropriate theoretical framework for the analysis (Section 3), outline our empirical strategy (Section 4), analyse margins at various levels using various cuts of the data (Section 5) and then conclude (Section 6). A few more technical aspects of the relationships between various wholesale costs are relegated to the Appendices.