WPS6947 Policy Research Working Paper 6947 Institutional Arrangements for the Promotion of Regional Integration of Electricity Markets International Experience Musiliu O. Oseni Michael G. Pollitt The World Bank Development Research Group Environment and Energy Team June 2014 Policy Research Working Paper 6947 Abstract This paper focuses on the institutional arrangements second set focuses on three regional power pools in more needed for facilitating regional electricity cooperation. developed countries—one in the United States, the Single The paper begins by discussing the theory of Electricity Market in Ireland, and the South East Europe international trade cooperation in electricity, with a view market. These cases highlight the potential and difficulty to discussing what preconditions might be important in of having cross-jurisdictional power pools. In the light facilitating wide area trading across national borders. It of the theory and evidence presented, key lessons are then discusses two sets of case studies. The first set focuses drawn in the areas of preconditions for trading, necessary on three regional developing country power pools—the institutional arrangements, practicalities of timetabling, Southern African Power Pool, the West African Power reasons to be hopeful about future prospects, and Pool, and the Central American Power Market. The suggestions for future research. This paper is a product of the Environment and Energy Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at moo23@cam.ac.uk and mgp20@hermes.cam.ac.uk. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Institutional Arrangements for the Promotion of Regional Integration of Electricity Markets: International Experience Musiliu O. Oseni and Michael G. Pollitt Judge Business School, University of Cambridge JEL Classifications: F13, L94. Keywords: electricity trade; power pools; regional electricity markets. Sector: Energy, electricity. Institutional Arrangements for the Promotion of Regional Integration of Electricity Markets: International Experience Musiliu O. Oseni and Michael G. Pollitt 1 1. Introduction The problem of how to promote wide-area trade in electricity is a well-known one within individual countries. National electricity markets in advanced countries developed over time as initially local, vertically integrated distribution companies found that there were substantial cost and quality of service advantages to horizontal integration and interconnection between service territories. While some countries developed near monopoly generation utilities which made use of a national transmission system (e.g. England and Wales and France), other countries did develop (limited, but in some cases substantial) trading between continuing regionally vertically integrated utilities (e.g. Japan and the United States). The creation of a national or wide area electricity transmission system which is centrally dispatched has been key to the promotion of trade in electricity 2. Such a system physically allows energy from different power stations to be directed towards supplying given electrical loads from a common ‘power pool’. Within countries such electricity trade can be facilitated within a single jurisdiction by the creation of a separate transmission system operator or TSO (National Grid in England and Wales) which owns and operates the transmission system assets (as transmission owner or TO) and operates the system in real time (as system operator or SO). Many jurisdictions follow this model (see Chawla and Pollitt, 2013). The debate about the Third Energy Package in the European Union focused significantly on the role of the transmission system operator in promoting non-discriminatory access to the transmission grid – hence allowing different generating companies and new entrants to the generation market to effectively compete with each other to supply final customer demands. This package (2009/72/EC) favored the creation of ownership unbundled transmission system operators which only owned and operated national transmission networks. This arrangement was thought to minimize conflicts of interest between the system operator and individual generation companies within a competitive wholesale electricity market. Within a single country it is possible to reorganize the distribution of network asset ownership to better facilitate trade (e.g., the creation of RTE in France as a separate transmission company owned by EdF or the reorganization of the TSO TenneT in the Netherlands which acquired the higher voltage lines of distribution companies). 1 Judge Business School, University of Cambridge. The authors acknowledge the financial support of the World Bank and the support and advice of Mike Toman and Jevgenijs Steinbuks. All opinions expressed in the paper are those of the authors alone and should not be taken to represent those of the World Bank or any of its member countries. Financial support from the Australian Agency for International Development and from the Partnership for South Asia Regional Integration Programmatic Trust Fund is greatly acknowledged. 2 See for example Foreman-Peck and Waterson, 1985, who document the emergence of a national integrated transmission system in the England and Wales. 2 However between jurisdictions with different regulators this is very difficult and unlikely to happen easily 3. Instead what is needed is the creation of wide area coordination of cross border trade where the system operation is separated from the transmission asset ownership. This has been a significant issue in the US between states leading to the creation of Independent System Operators (ISOs) or Regional Transmission Organizations (RTOs). Under such a model transmission systems can continue to be owned by a number of separate and even integrated companies, while it is the role of the ISO/RTO to operate the system in real time and to ensure non-discriminatory access to the grid for individual generators. The ISO model is gaining considerable attention precisely because the ownership of transmission assets does not need to re-organized. The ISO is therefore a useful cross- jurisdictional model for facilitating trading of electricity. Chawla and Pollitt (2013) note the rapid rise in the number of countries exhibiting either of these models – TSOs or ISOs - as wholesale power markets have been introduced and vertically integrated generation monopolies have been horizontally and vertically unbundled. The above introduction suggests that there are both physical and governance barriers that need to be overcome in order to promote trade in electricity across a wider area than was previously the case. Clearly physical interconnection is necessary, because without it no electricity can flow across pre-existing electrical boundaries. Traditionally countries have been very reluctant to trade electricity across borders and hence have limited the construction of transmission lines across borders. This is actually unusual in energy. Globally exports of electricity are around 3% of total production, in contrast to c.64% for oil and c.31% for gas and c.16% for coal (with the average for all goods and services being c.31%) 4, suggesting that there may be substantial scope for increased trade in electricity across the world. Where trade does occur, it does so in the context of a governance arrangement that facilities the building and use of transmission capacity between areas. Both the TSO and ISO models are capable of providing the necessary governance to facilitate trading and indeed often run the associated markets which price wholesale power and ancillary services. Such governance arrangements have developed over time and often carefully represent the interests of generators (and their customers) across the interconnected area in a non-discriminatory way. TSOs and ISOs are two examples of institutions which embody non-discriminatory governance arrangements. In each case they received support from either national or state governments, to bring them into existence. This paper will focus on the institutional arrangements for facilitating electricity cooperation. We have in mind the application of the lessons in the paper to other regions, such as the South Asia Region (SAR), namely Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka. These countries are part of a free trade area – SAFTA (South Asian Free Trade Area, formed in 2006). The South Asia Region currently exhibits very little trade in electricity but exhibits significant potential for beneficial trade. We will begin by discussing the theory of international trade cooperation in electricity, with a view to discussing what preconditions might be important in facilitating wide area trading across national borders. Next we will discuss two sets of case studies. The first set of case studies focuses on three regional developing country power pools – the 3 Though TenneT did buy an adjacent transmission grid in Northern Germany. We also discuss asset ownership changes in Ireland later in this paper. 4 See IEA and UNCTAD statistics. 3 Southern African Power pool (SAPP), the West African Power pool (WAPP) and the Central American Power Market (SIEPAC). The second set focuses on three regional power pools in more developed countries – PJM in the United States, the Single Electricity Market (SEM) in Ireland and the South East Europe market (ECSEE). These cases highlight both the potential and difficulty of having cross- jurisdictional power pools. We will then go on to draw key lessons for the South Asian Region (SAR) by looking at what aspects of the theory and experience in the earlier sections might be relevant to the promotion of electricity trade across borders in this context. We conclude with some suggestions for future research. 2. The theory of cooperation and international trade applied to electricity In thinking about the institutional arrangements that might facilitate increased cross border in trade in electricity, it is useful to think about ideal electricity market design and institutions. Hogan (1995) suggests that a wholesale pool spot market and an independent system operator should go together. This is because short term generator dispatch and short term transmission system operation are ‘two sides of the same coin’ (Hogan, 1995, p.26). This suggests that power trading should be associated with an institution which is also responsible for the operation of the transmission system in real time. Hogan (1998) suggests that nodal pricing of the transmission system access is also desirable unless the networks are relatively simple. Thus the US Standard Market Design - which incorporates these ideas - may be the most sophisticated market design for wide area trading, but it may not be necessary for international trade in electricity. Other designs may work, but the institutional design of markets is undoubtedly important. Stoft (1996) correctly predicted (prior to the California electricity crisis of 2001) that the institutional conflict between the California ISO and the California Power Exchange might decrease system reliability and lead to inefficient dispatch! Efficient market design is also about the participation of the demand side in the wholesale electricity market. This is increasing in importance in many of the most sophisticated markets, such as PJM and New York (see Walawalkar et al., 2010). For many countries demand side response inside their own country might be much cheaper at the margin than expanding international imports of power. A key point about market design is the need for sufficient transmission capacity. Fuasch et al. (2013), in their examination of the EU, suggest that cost optimal trading within the EU would require to 76% more transmission capacity. It is important to note that transmission capacity is not just required at the border to facilitate cross border trading. Loop flows in the electricity system mean that the ability to export/import electricity across one transmission link is dependent on the absence of congestion on other transmission lines, which may be internal to one or other country. Without sufficient transmission capacity cross border trade is going to be limited. The degree of sophistication in markets may be limited when moving to cross border markets. Brunekreeft et al. (2005) note that locational marginal prices (LMPs) as recommended by Hogan and practiced in PJM may be desirable in the EU, but they are unlikely to be politically viable. This may explain why the EU has promoted market coupling between national markets and allowing some merchant interconnection, rather than LMPs. Perez-Arriaga and Olmos (2005) suggests that the 4 problem that LMPs try to solve in the US with 200+ control areas is much less in the EU with 17 to 27 control areas. Clearly having congestion constraints imposed internationally is difficult to sell to national politicians. International trade is always mutually beneficial under the assumption of costless adjustment of factors of production (and the other assumptions of the Heckscher-Ohlin model of international trade). However these assumptions are not clearly satisfied for the electricity sector. While one might assume that factors of production in the electricity sector can be moved to other sectors, it is not so obvious for electricity intensive industries. They may be dependent on cheap domestic power, which if it is exported may necessitate factor allocation adjustment in the wider economy. International trade may however alter the risk profile around electricity prices. This is a version of the ‘energy security’ problem. In theory if two countries begin trading electricity this will normally provide some insurance against large shocks to electricity prices. This will be the case where their domestic supply/demand risks are either independent or negatively correlated. However clearly there will be some imported price volatility and the possibility of a large supply/demand shock in one country inducing a large price effect in the other country, which it could have avoided under no- trade. Over time, there is the possibility that dependence on imports of electricity might develop and domestic production facilities might close. This could expose an importing country to a hold up problem if the other country refuses to export. However, in reality these would seem to be second order (and manageable) risks associated with increased trade dependence. It is worth pointing out that such energy security risk is two-sided, as the exporting country might become equally dependent on the export revenue from electricity sales. 5 Trade theory has become increasingly concerned with considering departures from the assumptions of the basic Heckscher-Ohlin model. Markusen (1981) showed that if markets were initially monopolized a large country opening up to trade might suffer a loss of welfare due to the competition from another monopolist in the other country in a two country trade model. However Lahiri and Ono (1996) show that this result does not hold if new firms enter. Trade liberalization becomes beneficial again. The general result of Dixit and Norman (1986) emphasizes that trade can always be made beneficial as long as consumption and income taxes can be used to compensate losers within an economy. An important question in international trade theory is whether trade worsens the natural environment. This might be a concern for electricity trading where exploiting low cost resources might involve burning more coal in a low generation cost country. Antweiller et al. (2001) find that trade is generally good for the environment. This is because trade effects can be divided into three: an increased scale of activity; a composition effect on industrial structure; and an effect in improving technology. They find for a sample of sites in 43 countries over the period 1971-96, they find the scale effect worsens pollution by 0.25-0.5% for each 1% increase in GDP due to trade, the structure effect is neutral and the technology effect reduces pollution by 1.25-1.5%. This gives an overall positive effect on the environment due to trade. If wide area trading makes emissions control and 5 We discuss the dependence of Bhutan government revenue on electricity export income in the final section. This surely makes them a more reliable supplier of electricity than would otherwise by the case. 5 the spread of low emissions technology more likely, then it is likely to be good for the global environment. It is also important to be clear that where international trading is imposed on top of inefficient national arrangements is not always beneficial in theory. De Villemeur and Pineau (2012) model the integration of electricity markets in two jurisdictions, one selling at average cost, the other with a competitive market and prove that the overall welfare result is worse than under no interconnection. According to the theory of the second best we would expect that the price setting mechanism in each of the connecting national markets would need to be similar for trade not to worsen the initial distortions in the markets. Clearly if one jurisdiction were selling electricity at below cost and another at its true (higher cost) the result of the joining of the two markets would worsen the impact of the initial price distortion in the jurisdiction with prices below cost. This is likely to be a particular problem in some regions with a history of energy subsidies. Thus, international power pools are simply an example of the opening up to trade of a previously non-traded commodity. The initial situation in many developing countries is that there are monopoly electricity suppliers in each country each charging below cost in order to stimulate electricity intensive economic activity. If trade were to raise prices in one country but not in another that would affect the distribution of electricity intensive industry between the two countries. Of course this effect is tempered by the fact that commercial and industrial electricity consumers value the reliability of electricity supply as well as its price (Oseni and Pollitt, 2013) and hence if trade were to improve supply reliability and increase price at the same time in a particular country then it is possible that more electricity intensive industry would be attracted to that country. As noted above, distributional effects are important in considering whether to open up to trade. This is particularly true for electricity. First, low electricity price jurisdictions worry that electricity trading will raise prices for final consumers while increasing the profitability of electricity producers. If ownership of electricity production is concentrated while electricity consumption varies less than proportionally with income then electricity trading with higher price regions may be blocked (as is the case in some states of the US (by states with cheap coal) and between France (low price) and the UK (high price). Similarly in jurisdictions with cheap electricity for large electricity intensive industries, there may be a reluctance to export electricity at the expense of the trade dislocation caused by a potential reduction in electricity intensive manufacturing exports (e.g. Norway). Finon and Romano (2009) suggest that this could be dealt with by a windfall tax on hydro and nuclear producers (in the context of European countries), which could then be distributed to domestic consumers if necessary. These effects explain why high price jurisdictions are naturally keener on electricity trading with lower price jurisdictions than the other way round (see Joskow, 1997, on the United States). Neary (2007) predicts that trade liberalization in a particular sector will also lead to low cost firms in the country with comparative advantage in a sector taking over the similar high cost firms in the other country. This is likely to raise welfare in aggregate, but it raises profitability for producers in the low cost country at the expense of consumers generally. Neary predicts merger waves resulting from trade liberalization (subject to capital market liberalization). Indeed this is clearly what has been seen in the electricity sector in Europe. This may not be a problem for the distribution of 6 welfare internationally, as long as domestic shareholders can realize the benefit of the merger at the time of asset sale, however if there is capital market inefficiency this may not be the case. International trading of wholesale power is more valuable when the price of power fluctuates seasonally or across the day at individual locations due to weather fluctuations (Bahar and Sauvage, 2013). This means that trade will occur between countries even where their average pre-trade prices are the same. The rise of renewable energy in some countries adjacent to each other makes such trade more valuable. However it does put pressure on the physical characteristics of the system. Electricity is not a typical arbitraged product, because it has several dimensions of quality including voltage and frequency. What is needed is a market operation regime which adequately accounts for the need to maintain power quality in the face of energy price fluctuations. A good example of how important this is was the incident which occurred on November 4, 2006, when a problem in Northern Germany (a ship hitting a power line) caused blackouts in southern Europe (and other places). Market integration meant that one of the German system operators created an externality, which the Spanish grid operator failed to manage successfully. Where there are multiple countries seeking to reach an international trading arrangement there may be additional negotiation issues. One of the parties may block the agreement in an attempt to increase its share of the benefits. If it does so excessively or in a way that may set a precedent for other international agreements between the parties this may lead to the agreement not being reached. There are obvious ways to handle this using Shapley values which look at the value to the coalition of an additional member. However there is a real problem where side payments between regions must be made. If these are politically difficult to enforce then this may prevent agreement. Gately (1974) looks at the benefits from electricity cooperation between three Indian regions – Tamil Nadu, Andra Pradesh, Kerela-Mysore (KM). He finds that there are substantial benefits to all three regions cooperating in terms of reduced operation costs of their three power systems. However KM always has higher costs in any bilateral or trilateral agreement. The rise in KM’s costs are substantial (x3) as it exports its cheap hydro to other states, but costs in aggregate fall by 20%. Gatley shows that the order of joining the agreement may influence the value which an individual party can extract from agreement. He also notes that states may not just value the reduction in costs, but also care about the loss of jobs in the electricity sector as in state costs fall. The question for market arrangements put in place is therefore whether the market allows individual nations to capture a fair – both in the national and overall sense – of the benefits to the cooperation. This may be a particular problem where a transit state which hosts a transmission line does not actually import or export much electricity from it. Compensating this state fairly for its participating in the international agreement may be subject to these sorts of issues. Coordination of TSOs in terms of congestion management could bring significant benefits, if done under a single system operator. Kunz and Zerralin (2013) analyse congestion costs within Germany under the current regime of 4 separate TSOs versus a single German TSO. For a model calibrated to 2011, they find congestion costs of 30.36m Euros under full coordination versus 179.56m Euros under the current approach. They conclude that having four TSOs will become increasingly expensive relative to having a single TSO as the amount of intermittent generation continues to increase. 7 Kogut (1988) makes a general point about joint ventures which is relevant to power markets. He notes that the pooling of resources in a joint venture (in this case a power market, where the participants are countries and their firms) may not be just about transaction cost savings or strategic advantages. It may be about the benefits of tacit knowledge transfer. Agreeing to participate in a wide area power pool is a good way to learn about other markets and to benchmark against best practice for any individual firm, beyond a simple cost benefit comparison against the current national arrangements. Arguably, the most successful international power market in the world is Nord Pool (which includes Norway, Sweden, Finland and Denmark). Amundsen and Bergman (2007) discuss how prices across the area seem to be well integrated and vary according to underlying resource constraints. A particularly successful test of the Nord Pool system occurred in 2002-03 when a significant shortage of hydro capacity in Norway led to a severe price spike. Amundsen and Bergman (2006) note how the market coped well with this supply shock, maintaining political support. They suggest that the stakeholders involved understood that allowing market clearing prices to be high encouraged long term investment and that associated financial markets (which allowed hedging) also helped. This highlights how any international power pool needs to be able to withstand the inevitable stress tests that will come. In particular there needs to be confidence in the price determination process and a willingness to understand that a supply shock in one country will need to be supported by higher wholesale prices in connected markets. Thus temporarily high prices are an important price signal and represent a payment for mutual insurance. An interesting take on trading in electricity markets is the concept of fairness, which may be very important in extending international trade in electricity. Dickson and Kalipurakal (1994) suggest that the idea that competitive market determined prices are always ‘fair’ is only one of several potential concepts of fairness in market transactions. In particular the idea that if there is scarcity, scarcity prices should be charged to everyone may be one that is difficult to accept. Interestingly, Dickson and Kalipurakal (1994) find that the market traders in the US that they survey are generally happy with market determined prices (rather than the alternative of Dual Entitlement where prices go up if costs go up and do not go down if costs go down). This may not be true of electricity consumers (or their elected representatives) in general of course. When it comes to the benefits of wide area markets there is a lot of concern about the exercise of market power, particularly if competition policy enforcement is weaker internationally than it is domestically. In the European Union, which has pan-European competition policy and enforcement, this is not a big issue. However in other regions, such as SAR, if the market design gave rise to the exercise of market power this would be an issue. Market monitoring within wide area markets is an important activity to ensure that the market is behaving as intended. 6 Von der Fehr and Harbord (1998) look at different types of markets which may deviate from the perfectly competitive outcome: capacity constrained markets with Bertrand competition; collusive oligopolies in repeated games; supply function equilibria; and auction approaches. Clearly there is a worry about gaming, especially between countries within regions. Changing the nature of competition, by extending the market may change the current behavior of players within existing 6 See Goldman et al. (2004) who discuss five case studies where market monitoring was important. 8 markets. Similarly, Neuhoff and Newbery (2005) discuss how the move to integrated markets could initially lead to higher prices until enough competition emerges to definitely lower prices. 7 There is a question of whether existing independent power producers (IPPs) will individually participate in a power market or form a coalition (or merge) to get a higher price. Jia and Yokoyama, (2003) use Shapley values to investigate whether an IPP would be better of participating individually in a power market or via a coalition. Ferrero et al. (1997) however show theoretically that if power pools are big enough selling into them at marginal cost (and participating in a ‘grand coalition’) may be better than deviating and not participating in the power market. This modelling suggests that if the international power pool is big enough the gains for an individual country to participate may become bigger than refusing to participate (the EU Single Electricity Market project, may be a good example of this). Market power is not necessarily limited to incumbent generators in an international power market. National system operators may also exercise market power. System operators manage congestion on their networks and the transmission constraints behind them represent significant barriers to electricity trading, as suggested above (Kumar et al., 2005). However they are under national incentive schemes to minimize internal congestion within their control areas. Under international trading they may have incentives to shift transmission constraints to international interconnectors to reduce constraints within their own country, reducing the benefits of international competition. The Swedish transmission system operator was recently subject to anti-trust action by the European Commission for doing this within Nord Pool. 8 While power markets are good for short term competition and efficiency there is an issue with whether they induce optimal long term planning. Kagiannas et al. (2004), note that generation expansion planning is evaluated differently if done by several competing firms (or countries), than if done by one monopoly firm. They note that the scope for mistakes to be made in aggregate may be increased by increasing the number of firms in the market. In the context of the EU the impact of reducing market power, through more effective competition, is potentially very significant. Lise et al. (2006) show that for a simulation of 8 EU countries in 2000 moving from a situation of strategic competition to perfect competition would reduce profits substantially. In their analysis the profits of EdF and Electrabel (the incumbent generators in France and Belgium) would fall by one third. Hobbs et al. (2005) estimates the benefits of Netherland- Belgium market coupling to Belgian market using a transmission-constrained Cournot model. The study projects social surplus improvements on the order of 200m €/year, assuming market coupling does not encourage the largest producer in the region to switch from price-taking in Belgium to a Cournot strategy due to a perceived diminished threat of regulatory intervention. However benefits would be higher if transmission capacity was increased to allow the competitive baseline to be achieved. This implies the importance of optimal transmission capacity and of the monitoring of the competitive behavior of market players. 7 This was the initial experience within the England and Wales power pool where competition took 8-10 years to fully mature. 8 See http://ec.europa.eu/competition/elojade/isef/case_details.cfm?proc_code=1_39351, Accessed 31 March 2014. 9 3. Case studies of cross jurisdictional electricity trading In this section we examine six case studies of cross-jurisdictional electricity trading. We start with two African case studies from Southern Africa and from West Africa. We then look at Central American Power System. These developing country case studies indicate both the potential for wide area trading across a number of individually small national power markets. We then go on look at the evolution of power pools in three more developed jurisdictions. First, PJM, the largest and most developed power market in the US. Second, the Single Electricity Market (SEM) in Ireland which has integrated the power market in the island of Ireland, between Northern Ireland and the Republic of Ireland. Third, the South East Europe power market which covers a number of countries from Greece to Romania, and is part of the Single European Electricity market project of the European Union involving the creation of the Energy Community of South East Europe (ECSEE). In each case we look at the evidence on what happened to facilitate trade, the nature of the trading platform, what institutions were set up to support it, the governance of these institutions, the practical steps to implementation and the concrete evidence on the benefits of trade. The case studies are not all successful and some took many years (more than might be thought necessary) to reach fruition. African Integrated Power Markets Case A: Southern African Power Pool The Southern African Power Pool (SAPP), established in 1995, is currently the most advanced power pool in Africa (see Appendix 1 for details on its size relative to other international power pools). SAPP was a product of the efforts aimed at promoting energy development undertaken as part of the political goal of regional integration of the Southern African Development Community (SADC) established in 1992. One of the major political and economic forces behind the development of SAPP has been South Africa’s yearning to meet future energy demand increase by importing low-cost hydropower from its northern neighbors (Economic Consulting Associates (ECA), 2009). The original agreements of SAPP were drawn from the bilateral and multilateral agreements already existing (prior SAPP) among the member countries. SAPP’s guiding framework was based on intergovernmental memorandum of understanding (MOU) that authorized and guaranteed inter- utility MOU and operating agreements. Inter-utility memorandum of understanding (IUMOU) deals with the issue of ownership and rights among the participants. The Agreement between operating members (ABOM) defines the interaction between the utilities with respect to operating responsibilities. 9 Lastly, the Operating guidelines (OG) set out the arrangement for cost sharing and functional responsibilities for plant operations, maintenance and safety rules. The highest governing body of SAPP is the executive committee which comprises the chief executives of the various power utilities participating in SAPP. The committee receives and refers matters such as the requests for membership by non-SADC members and the major policy issues to 9 http://www.intechopen.com/download/get/type/pdfs/id/16031 10 the SADC energy ministers for onward consideration. The management committee which comprises officials from the member utilities oversees the running of the SAPP. Four sub-committees, including planning, operating, market and environmental sub-committees are responsible to the management committee. The SAPP member states and the interconnected grid are shown in Figure A1. SAPP comprises all 12 SADC member countries of which nine are operating members whose interconnected grid carries about 97% of the power produced by SAPP countries (Economic Consulting Associates (ECA), 2009). Coal dominates the fuel mix used for generating electricity and is predominantly abundant in the northern South Africa, western Botswana and eastern Zimbabwe. South Africa alone accounts for more than 80% of the total energy production in SAPP and hence coal contributes three-quarters of the total SAPP generation (Economic Consulting Associates (ECA), 2009). Hydro dominates the generation of electricity in the rest of the SADC countries, with power stations being located in the Zambezi Basin involving Zambia, Zimbabwe, Mozambique and Malawi, at Inga in the Democratic Republic of Congo, in central Angola, Northern Namibia and also in Tanzania. Overall, the generation mix in the SAPP currently stands at 74.3% coal, 20.1% hydro, 4% nuclear and 1.6% diesel and gas. Figure A1: The SAPP Interconnected Grid Source: (SAPP (Southern African Power Pool), 2013c) http://www.sapp.co.zw/docs/SD%20Bulletin,Sept,%202013.pdf The introduction of SAPP has led to rising investments in capacity building, and at best spurred the trading activities between member states. For instance, net capacity of 1,700 MW was added in 2007, 1,442 MW in 2008 and 1,061 MW in 2012, and the region is expected to have sufficient capacity by 2017. 10 Bilateral contracts dominate the trading arrangements in SAPP and it often accounts for between 90-95% of the energy traded. Bilateral agreements usually cover a period of 1- 5 years or more. These agreements guarantee the security of supply but are not flexible enough to accommodate varying demand profiles and prices (Musaba, 2009). In order to ensure a more 10 SAPP (Southern African Power Pool) (2013): Annual Review Reports for Various Years. www.sapp.co.zw 11 competitive electricity trading platform, a Day Ahead Market (DAM) was introduced in 2009. Between 2012 and 2013, however only about 6% of energy demand in the Day Ahead Market was traded. Transmission constraints remain a major issue within SAPP, with the African Development Bank funding new feasibility assessments. 11 The potential lost wheeling revenue that could have been generated in November 2013 alone was estimated at US$316,312 (SAPP (Southern African Power Pool), 2013b). In addition only around 20% of desired trades actually took place in 2012. The South African utility, Eskom, is the dominant supplier into SAPP, because of its extremely low (potentially ‘predatory’) prices. In 2008 it was charging just 1.7 US cents/kWh (Economic Consulting Associates (ECA), 2009) for its exports. This is well below the cost of new fossil and hydro power and is the reason that the potential regional projects originally planned to supply South Africa have so far not been accomplished. Case B: West African Power Pool (WAPP) Despite its great energy potential, the West African region remains among the world’s lowest electric power producers. The region accounted for only 9% of the total Africa’s installed capacity and about 7% of the total electricity generation in the continent in year 2000. Energy access stood at between 4% (Niger) and 70% (Benin) and it’s per capital electricity consumption (127 kWh) was 3 times less than Sub-Saharan Africa’s average (Zakharov, 2003). Established in October 2000, WAPP aims to promote energy trades between member countries through the integration of the national power systems in order to provide stable, reliable and affordable electricity supply to the citizenry. It comprises 14 of the 15 member countries of the regional economic community – Economic Community for West African States (ECOWAS). 12 WAPP member countries include Benin, Côte d'Ivoire, Burkina Faso, Ghana, Gambia, Guinea, Guinea Bissau, Liberia, Mali, Niger, Nigeria, Senegal, Sierra Leone and Togo. WAPP, which currently has 26 member utility companies, consists of public and private generation, transmission and distribution companies involved in the operation of electricity in West Africa. Article of Agreements, as filed with ECOWAS Secretariat upon the endorsement of the Meeting of Energy Ministers, set out the objectives of WAPP and its operating procedures among the member states. The General Assembly is the highest decision making body for the WAPP and comprises the representatives of all member states. It is responsible for the co-ordination of appropriate measures towards the implementation of the principles of the Articles of Agreement, facilitate the implementation of programs and projects, review and approve new membership applications, elect the members of the Executive Board, examine and adopt the financial regulations of the structures of governance of WAPP, among others. Next to the General Assembly is the Executive Board (EB) which, based on the overall policy objectives agreed upon by the General Assembly, has a decision making authority to develop and implement initiatives to achieve the mission of the organization. EB directs the activities of all WAPP committees; examines and recommends to the General Assembly, the entry, exit and re-entry of members to the WAPP Organization; authorizes all major contracts and [finance/debts] instruments; sets out the standards and policies of the WAPP Organization and penalties for non-compliance, and acts on appeals, among others. 11 For example, for the Zimbabwe-Zambia-Botswana–Namibia project (see www.sapp.co.zw). 12 http://www.ecowapp.org/ 12 The Organizational Committees of WAPP advise the Executive Board on all matters regarding collective policy formulation functions for developing, maintaining and updating common “rules of practice” on technical, planning, operational and environmental aspects of WAPP. The committees comprise the technical experts selected from WAPP members. The current Organizational Committees of WAPP are Engineering and Operating Committee, Strategic Planning Committee, and the Finance and Human Resources Committees. Despite its existence for over a decade, WAPP regional trading activities have been almost non- existent. Not until 2012 when WAPP commissioned Mercados (a consulting firm), there were no clearly specified trading platforms in WAPP. The few bilateral trading exchanges between member countries were outside the regional electric trading system, mostly based on separate and pre- existing agreements between the parties. Towards the establishment of a regional electricity market by 2015, however, Mercados Energy Market International was employed to design and develop the market models and rules for power exchanges between WAPP Member Utilities (WAPP (West African Power Pool), 2012). The final draft of the document has since been submitted (AF- MERCADOS EMI (Energy Market International), 2012). Based on the report, the pool’s main trading platform would be in the form of long-term contracts and allocation of excess production among members. As market develops, however, spot market energy trading such as a Day Ahead Market utilizing the remaining transmission capacity in the regional transmission system after allocating all the contracts will follow. Why did it take WAPP nearly a decade and half to come up with its trading frameworks unlike some other regional power markets? The main reason was inadequate existing installed capacity and poor infrastructures among the member nations. Unlike the Southern African Power Pool (SAPP), whose installed capacity at inception in 1995 was 48,461 MW (with about 38,000 MW in South Africa alone), WAPP had only 9,705MW capacity as at the time it was formally created. 13 This low capacity and lack of adequate transmission networks made regional trading difficult as member nations could not even meet their domestic energy needs, and because majority of them were yet to be interconnected. The slow progress in the regional trading in WAPP relative to the development in SAPP underscores the importance of existing infrastructural capacity in the development and progress of market integration. Recognizing the need for quick intervention in infrastructure developments, WAPP declared a state of emergency for rapid power infrastructures in its master plan. Building of power plants and transmission networks, and perfecting other technicalities required for loads scheduling and power trading were prioritized in WAPP project financing, with few investments on information, transparency and regulation (Pineau, 2008). WAPP master plan was designed by an American consulting firm, Nexant, and presented to the funding agency of the project, USAID, and to the ECOWAS Secretariat (Nexant, 2004b). In the master plan, WAPP countries were divided into two zones, and 14 major priority interconnections were identified (Nexant, 2004a). WAPP Zone A comprises Benin, Burkina Faso, Côte d’Ivoire, Ghana, Niger, Nigeria and Togo. Countries in Zone B include Gambia, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Senegal and Sierra Leone. Figure B1 illustrates the major interconnection projects in WAPP as of 2013. 13 http://www.eia.gov/ 13 Figure B1: WAPP Major Interconnection Projects (Source: WAPP website) 14 The majority of the priority interconnection projects – which have been supported with funding from the World Bank - have since been commissioned while some are nearing completion. With the completion of many of the projects and the near completion state of the others, WAPP is set to have an operative regional power exchange by 2015. The roadmap for regional energy trading sets out three different development phases, consisting of a transition from bilateral market arrangements between neighboring countries to an open and competitive regional market system. However, how the 2015 target for the operation of the regional market will be achieved still remains unclear considering the high level of unreliability still facing most of its member countries’ power sectors. Case C: The Central American Power Market (SIEPAC) Although Central American countries have witnessed several electricity market reforms, individual electricity markets still differ considerably, ranging from vertically integrated to fully unbundled wholesale energy markets. Also there are significant differences in the quality of service, and the production and delivery efficiency (Reinstein, Mateos, Brugman, Berman, & Johnson, 2011). In addition, the power system in each country has faced huge challenges of meeting growing demand and high costs of production due to their relatively small size and fragmentation. Apart from poor and unreliable energy services often experienced by the countries, the region has struggled to achieve cost-competitive energy services (Martin & Posadas, 2012). Thus, establishing a regional electricity market was considered a means for creating a larger market that would enhance efficiency and promote competition among power producers. 14 http://www.ecowapp.org/?page_id=168 See also Pineau (2008). 14 The Central American Power Market is a product of the cooperation between six Central American countries to form a reliable, efficient and affordable energy market. The focus has been on the interconnection of the sub-region electric grid for the purpose of reaping economies of scale in larger generation as opposed to small production constrained by national boundaries. An important interconnection project – the Central American Electrical Interconnection System or SIEPAC 15 in Spanish – involving 250 KV single circuit transmission line approximately 1,800 km long was launched following an initial feasibility study in 1987. 16 A treaty was signed in 1996, but the line construction was not started until 2006 and was finally completed in 2013. The line connects the member countries, linking Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua and Panama for easy energy exchanges (Figure C1). The project has received significant financial support from multi- lateral lenders, with the Inter-American Development Bank providing 59% of the initial funding. Figure C1: SIEPAC Project Map and Basic Characteristics Source: Tarbuck (2013) 17 The region’s electricity market – the Central American Regional Electricity Market (MER) – was created after the governments of the six countries have signed and ratified the treaty for its establishment in 1998. To support the operations of the market, the regional regulatory commission CRIE (Comisión Regional de Interconexión Eléctrica), the regional system operator EOR (Ente Operador Regional), and the company that owns the grid, EPR (Empresa Propietaria de la Red), were also created. The MER was established as the seventh market, superimposed on the other six national markets, with regional regulation which enabled the Regional Operating Agency (EOR) perform international transactions of electricity in the region. 18 15 SIEPAC is the acronym for the Sistema de Interconexión Eléctrica para los Países de América Central (Central American Electrical Interconnection System). 16 See Economic Consulting Associates (2010). 17 Emily Tarbuck (2013), ‘Panama: Regional Support Provides Relief from Energy Crisis’, The Argentina Independent, May 17 2013, Available at: http://www.argentinaindependent.com/currentaffairs/newsfromlatinamerica/nama-regional-support- provides-relief-from-energy-crisis/ 18 http://www.eprsiepac.com/operacion_mer_siepac_transmision_costa_rica.htm 15 Apart from the Consejo de Electrificación de América Central (the Council for Central American Electrification), established in 1985, which provides advice at both national and regional level, MER has three distinct institutions set up to support its operation. CRIE (Comisión Regional de Interconexión Eléctrica), established in 2002, serves as the regulator for the regional wholesale electricity market. It consists of six members, comprising one commissioner drawn from the electricity regulatory agency of each member states. The selection of representatives from the regulators of various countries was designed to minimize possible inconsistency between national and regional regulatory techniques and encourage uniformity in technical and operating procedures. The presidency of CRIE is rotated among its members. The regional system operator EOR (Ente Operador Regional) is responsible for the coordination of the market operation, including the Day Ahead Market, real-time dispatch, financial settlements, and dissemination of information through designated national system operators. It is also responsible for the formulation of plan for the regional generation and transmission networks expansion. Lastly, EPR (Empresa Propietaria de la Red) owns and operates the interconnectors connecting the six countries in the region. The interconnectors are 75% publicly owned by the utilities and transmission companies in the six member states, and 25% private sources, including the Spanish company, ENDESA. The volumes of energy traded in the Central America regional power market have varied since cross- border exchange began and have always been less than 5% of the region’s electricity consumption (CASTALIA, 2009). Power exchanges are currently dominated by bilateral contracts between parties. The bilateral contracts consist of medium to long-term agreements between parties to inject and withdraw energy at specific nodes of the regional transmission network based on the guiding rules and regulation (Prada, et al., 2004, Tinoco, 2009). Between 2003 and 2007, these contractual trading arrangements accounted for more than 80 percent of cross-border power exchange in the region (CASTALIA, 2009). It accounted for 94% and 98% of the regional transactions in 2010 and 2011 respectively. The contribution of each of the countries differs considerably. In September 2011 for instance, while 69% of the cross-border injection was contributed by El Salvador followed by 29% from Guatemala, Honduras topped the ranks of importers accounting for 44% of the regional bilateral trade withdrawals in the same month. Panama, El Salvador, and Nicaragua imported 36%, 16% and 4% respectively during that month. 19 As highlighted by (CASTALIA, 2009), the Central America regional market currently operates three types of spot-market transactions, namely financial contracts and two different forms of physical flexible contracts. Financial contracts are based on net settlement and have no impact on the dispatch other than through bids and offers to the opportunity market. Physical flexible contracts are bids for transmission services between two nodes and a maximum price that the bidder is willing to pay. Lastly, the second physical flexible contracts allow buyers to substitute their own planned generation with spot market purchases. Despite operating three different spot-market transactions, however, spot trading in the region’s market has been very limited, averaged 18% of the region’s cross-border power exchange between 2003 and 2007 (CASTALIA, 2009). It peaked only at 26% in 2007 before maintaining a downward trend accounting for just 5.8% and 1.6% of the regional trading volumes in 2010 and 2011, respectively. 19 http://www.enteoperador.org/InformesEstadisticosMensuales.jsp 16 Case D: PJM Interconnection 20 In the United States there are now four significant interstate electricity trading areas run by Independent System Operators or Regional Transmission Organizations (MISO, SPP, ISO-NE and PJM). These entities are of interest because they do cross traditional ownership boundaries and because they have evolved organically and often in competition with each other at their boundaries. PJM is the oldest, largest and arguably most developed. PJM’s three initial utilities began cooperating in 1927 to form the world’s first power pool in order to save costs. It expanded over time becoming an independent organization in 1997.21 PJM has been expanding over time in terms of coverage and in terms of the scope of its activities. In 2012 PJM covered around 61m people, 62,556 miles of transmission lines, 183 GW of generation capacity across 13 US states plus the District of Columbia. The recent history of transmission arrangements in the US is well discussed in Joskow (2005), who traces FERC’s attempts to liberalize the electricity sector by promoting ISOs (in FERC Order 888) and then RTOs (in FERC Order 2000) and specifically the ISO-RTO model of PJM (in Standard Market Design proposals: SMD 2002). ISOs have been described as the ‘soul of the grid’ 22 and as the ‘air traffic controllers’ 23 of the electricity system. 24 PJM, like the others, undertakes a number of key functions. These include: real time physical system operation; ensuring short term reliability; tariff administration and design; congestion management; and procurement of ancillary services. It also runs the associated Day Ahead Market and real time energy, capacity and transmission rights markets. An interesting feature of the governance structure is that several of the US ISOs, including PJM, have fully independent boards where individuals should have no ongoing relationship with market participants in the sector 25. PJM is run as a not-for-profit organization, owned by its member companies. The Board of Managers is separate from its Members Committee. Board of managers is the top management level of the PJM. It comprises 10 members whose responsibility is to ensure that PJM fulfils its business and legal and regulatory requirements. In order to ensure impartiality, members’ actions are strictly subject to the Code of Conduct guiding the operation of the board. Next to the PJM Board is the Member Committee which reviews and decides upon any major changes and proposals made by committees and user groups. This committee serves as the link between the committees and user groups and the board of managers. It advises PJM and makes recommendation on all matters relating to secure and dependable operation of PJM, development and operation of a standard, competitive and non-discriminatory power market, and ensuring there is no undue influence over PJM’s operations by any market participants. 20 This section draws substantially on Pollitt (2012a). 21 th Source: PJM website, http://www.pjm.com/about-pjm/who-we-are/pjm-heritage.aspx, Accessed 28 January 2011. 22 See O’Donnell (2003). 23 Source: SPP website, Available at http://www.spp.org/section.asp?pageID=1 Accessed 28 January 2011. 24 A good general introduction to ISOs is contained in O’Neill et al., (2006). See: http://www.nyiso.com/public/about_nyiso/nyisoataglance/board/index.jsp ; 25 https://www.midwestiso.org/Library/Repository/Meeting%20Material/Stakeholder/BOD/Corporate%20Gover nance%20and%20Strategic%20Planning%20Committee/2011/20110615/20110615%20BOD%20Corporate%20 Governance%20and%20Strategic%20Planning%20Committee%20Item%2002%20%20Principles%20of%20CG.p df; http://www.pjm-miso.com/about/board-managers.html. Accessed August 22nd 2011. 17 Besides the PJM Board of Managers and the Member Committee, there are two other broad committees and groups that are responsible for the operation of the PJM. These include the PJM Interconnection committees and groups which are responsible for formulating, developing and refining PJM’s rules and regulations, policies and procedures. The second group is the market committees. The market committees ensure the administration of an open grid and transparent energy market. Figure D1: Map of PJM showing states and transmission ownership areas covered Source: PJM 26 The estimates of the gains from the creation of ISO areas (and particularly PJM) are impressive. FERC (2004) estimates the benefits of improved market integration facilitated by ISOs in the US is of the order $3.8-5.4bn per year. This attributes the benefits of increased generator competition to ISOs. This study also suggests that better use of transmission assets and better sharing of reserves might be worth around $0.8bn per year (by 2010). Douglas (2006) estimates that the regional dispatch of coal fired power plants, by ISOs, has led to average variable cost savings of 1.5-3%. Mansur and White (2009) find significant gains from increased trade in their analysis of one particular expansion of PJM in 2004 when 19 Mid-west based utilities joined the PJM area. These gains were estimated at $163m in the first year, against an initial investment cost of $40m. Case studies E and F: Drawn from the EU Single Electricity Market The European Union has had a significant regional push towards the creation of a single electricity market across the whole the European Union. This has involved a series of Electricity Directives aimed at setting common set of rules that national governments must pass into their national legislation. The 1996, 2003 and 2009 directives gradually liberalized national electricity markets and facilitated increased cross border trading directly through specific rules encouraging non- discriminatory access to interconnector capacity and indirectly through creating consistent national market rules which reduced the entry barriers into national markets. The Single European Electricity market is part of the European Single Market project (from 1986) which aimed to encourage the creation of genuine pan European competition in all sectors by removing all non-tariff barriers to trade. The Single Market project encouraged mergers between national companies and cross-entry 26 PJM Transmission Zones map (pjm-zones.pdf), available at: http://www.pjm.com/documents/maps.aspx. 18 into markets. In traditionally heavily nationally regulated sectors, such as telecoms, electricity, gas and transport, the impact of the extension of the single market has been particularly noticeable. Progress has been considerable in terms of forcing convergence of national market rules (Jamasb and Pollitt, 2005 and Pollitt, 2009) but rather slower in causing wholesale electricity price convergence across the EU. 27 As early as 1988 the European Commission, which oversees competition in the EU identified the possibility of the extending the single market from traditionally heavily traded sectors into electricity and gas. A key feature of the process that emerged was the encouragement of regional sub markets between neighboring countries where the benefits of cross border trading were clear and where interconnection capacity already existed (see Jamasb and Pollitt, 2005; Pollitt, 2009). The European Commission welcomed this process as it meant that the harmonization of cross border trading platforms and rules could take place organically, in line with the EU’s subsidiarity principle, at a pace that national governments were comfortable with. There have been notable successes with emergence of a Northwest European trading area initially involving France, Belgium and The Netherlands, being extended to include Germany and now UK and Nord Pool. These projects have involved coordinated allocation of transmission capacity and coupling of day ahead power markets. In the next two cases we discuss two manifestations of regional markets within the EU (and its near neighbors) which are particularly relevant to the South Asian Region countries. These are the creation of the Single Electricity Market in Ireland (SEM) and the Energy Community of South East Europe (ECSEE). The Irish market is interesting because of governance structure, while the ECSEE market is interesting because of the diverse nature of countries involved. Case E: The Single Electricity Market (SEM) in Ireland The island of Ireland was divided into two distinct electricity markets until the creation of the SEM. In Northern Ireland there was a small market with approximately 0.7m customers and a peak demand of around 2000 MW, interconnected with both Scotland and with the Republic of Ireland. In the Republic of Ireland peak demand is around 5000 MW and there are around 1.8m customers. The system has recently been connected to Wales in Great Britain. Prior to 1975 there was an operational North-South interconnector, rated at 300 MW. Following the escalation of the Troubles in Northern Ireland this was targeted by the IRA and remained non- operational until 1995 and the acceleration of the Peace Process within Northern Ireland. However the benefits of interconnection between the two systems were well understood, in terms of the reduced need for spinning reserve capacity in the island of Ireland, the non-coincident peaks on the two systems, leading to differential prices and the different generation mixes within the two systems which led to marginal plants alternating between North and South. The emergence of wind as a major source of generation across the island of Ireland has also significantly increased the value of interconnection. Wind now accounts for 20-25% of energy generation in Northern Ireland and 40% in the Republic of Ireland, with the share being much higher than this at certain times. 27 Zachmann (2008) looked at the convergence of wholesale prices across the EU over the period 2002-06. He found that while there was some evidence of convergence, there was less than full convergence even allowing for transmission constraints between countries. He concluded that market integration was only ‘partially successful’. 19 In 2004 the two utility regulators laid out principles for the establishment of a single wholesale electricity market in Ireland which was approved in 2005. The market is claimed to be a world first: the first market between sovereign countries to operate with multiple currencies (Northern Ireland is in the GBP area, while the Republic of Ireland is in the Euro area). The SEMO sought to deliver a larger market, transparency, a stable investment environment, greater security of supply and greater efficiency. 28 The SEM was created on 1 November 2007. The SEM is overseen by the SEM committee (SEMC) which has representatives from UREGNI, CER and an independent member. It involves marginal cost based bids from all generators in the island. These bids are then aggregated and matched to electricity demand to give a single half hourly trading price. There is also a capacity payment. This is fixed in total each year as the total required MW forecast by the transmission system owner (which is now EirGrid for the whole of Ireland) times the cost of a MW on new peak capacity. This amount is then divided up among generators over the year on the basis of their half hourly availability. 29 In the event of their being transmission constraints which prevent the winning bidders being dispatched in the market constraint payments are made to those forced to generate because constraints force losing bidders to generate. The use of marginal cost based bidding is significant because research prior to the opening of the market had demonstrated price based bidding would have given rise to the ability for some firms to exercise significant market power in the SEM (McCarthy, 2005). Short run marginal cost bids with a capacity payment were shown to reduce wholesale prices significantly relative to average cost bidding or medium run cost bidding (FitzGerald et al., 2005, p.73). Overall the SEM has been judged to working efficiently (O’Mahony and Denny, 2013). One important feature of SEM is that it is designed in a way that strategically mitigates participants’ abuse of market power. The pool imposes Directed Contracts on generators with significant market power, and enforces a license condition on generators to strictly adhere to a bidding code of practice. Generators viewed as possessing substantial market share are compelled to enter into forward contracts with suppliers for a specified volume at a pre-determined price. This is done to prevent those generators from withholding capacity for the purpose of influencing the market price (Conlon, 2009). In addition, the pool has a Market Monitoring Unit (MMU) which constantly monitors participants’ bidding behavior in order to ensure participants’ compliances with the market rules. Nera Economic Consulting (2006) conducted a cost-benefit Analysis of SEM over a period of 10 years covering 2005-2016. The estimates also show that the majority (NPV €127Mn out of €155m) of the social net benefit of SEM implementation goes to consumers, with a fairly even split between NI and ROI consumers. The SEM is increasingly integrated with the market in Great Britain. The East-West interconnector between Dublin and North Wales opened in 2012, having received a grant from the European Commission and project finance from the European Investment Bank. It is estimated to 28 See ‘SEMO Success Story’ available at http://www.ventyx.com 29 See ‘The SEM Capacity Payment Mechanism and the impact on trade between Ireland and GB’, Presentation by John Lawlor, 19 June 2012, available at www.elforsk.se/.../Market%20Design/seminars/CapacityMarkets/9_ESB.pdf 20 bring annual benefits of 66m Euros at an NPV cost of 395m Euros, making it extremely worthwhile (see de Nooij, 2011). Figure E1: Map of all Ireland Electricity System Source: Eirgrid 30 Case F. The Energy Community of South East Europe (ECSEE) 31 In 2005 nine South East Europe (SEE) countries - Albania, Bosnia-Herzegovina, Bulgaria, Croatia, UNMIK (Kosovo), FYROM (Macedonia), Montenegro, Romania and Serbia 32 - signed a Memorandum of Understanding (acquis communitaire) with the EU to form the Energy Community of South East Europe, ECSEE. The MOU can be seen as an effort to get back to the roots of the EU in the European Coal and Steel Community, ECSE (see Dietz et al., 2007; 2009)). In May 2007 the ownership of the EU transferred ‘ownership’ of ECSEE to its member countries. There has been some significant progress towards electricity market integration (see Vailati, 2009) in ECSEE. An inter-TSO compensation mechanism has been implemented for cross border power transfers. The TSOs conducted trials of a coordination mechanism in 2008-10 and have now moved on setting up an office to coordinate auctions of transmission capacity (the CAO). This is notable 30 All Island Transmission Map (All-Island TransmissionMap.pdf). Available at: http://www.eirgrid.com/transmission/transmissionsysteminformation/ 31 The section draws on Pollitt (2009). 32 Of these countries, Bulgaria, Croatia, Romania, Greece and Slovenia are members of the European Union and are subject to EU Electricity directives on the creation of the Single European Electricity market. The rest are actual or potential candidate countries for EU membership which would imply that compliance with the EU Electricity directives would a necessary part of any formal accession process. 21 given that Kristiansen (2007) observed that the previous transmission capacity allocation mechanisms existing across the ECSEE region were different and often bilateral between countries and that these were barriers to trade across the ECSEE. However so far there has been no impact on investment in extra transmission capacity. There is long history of regional electricity trading in general between ECSEE countries, though several were involved in the Balkan wars following the breakup of Yugoslavia. These wars redefined the country boundaries and caused substantial damage to the electricity network which existed in Yugoslavia prior to its break up. The Yugoslavian region was a significant transit route between the larger markets to the north, especially Bulgaria and Romania, connecting Greece with the rest of Europe. However the problems in delivering electricity integration benefits to society are not to be underestimated. This may be particularly true of the smaller ECSEE countries. For small countries, where the scope for national competition may be limited and managerial expertise is scarce, the benefits of a full reform package may be small in relation to the costs. This should make interconnection more attractive but it does imply a potentially problematic loss of sovereignty in a region where this is an important issue. Thus, without adequate interconnection wholesale competition may therefore be an issue within some ECSEE countries (e.g. Albania, Bosnia- Herzegovina, UNMIK, FYROM and Montenegro). The internal decision making in ECSEE rests with a Ministerial Council (MC) and a Permanent High Level Group (PHLG). The MC provides general policy guidelines and takes measures to ensure that the objectives of ECSEE are met. The body also delegates, when necessary, the Permanent High Level Group, the Regulatory Board or the Secretariat to carry out some functions in order to achieve the policy objectives of the Energy Community. The presidency of the MC is rotated annually between contracting parties. The PHLG which is also chaired by a president is responsible for preparing the work of the MC, ratifying technical assistance requests made by international donor organizations, international financial institutions and bilateral donors, and reporting to the Ministerial Council on progress made toward achieving the ECSEE’s objectives, among others. The Regulatory Board provides advice on the details of statutory, technical and regulatory rules to the Ministerial Council or the Permanent High Level Group, and issues Recommendations on cross- border disputes involving two or more Regulators upon request by any of the parties involved in dispute. The President of the Board is elected for a term determined by the Regulatory Board while European Commission acts as Vice-President. The Secretariat serves as the administrative hub of ECSEE and is headed by a Director appointed by a Procedural Act of the Ministerial Council. The Energy Community has developed a Coordinated Auction Office (CAO) for an effective regional congestion management system. The South East European Coordinated Auction Office (SEE CAO) is to be responsible for the harmonization of congestion management and optimization of cross- border capacity allocation. The lack of a regionally coordinated capacity allocation and congestion management system creates a barrier for cross-border electricity trade and as a consequence for the establishment of regional electricity market in South East Europe. The Office is expected to tackle this problem by acting as a central point for cross-border capacity allocation in the region. The application of one single set of auction rules and one single allocation platform for the whole region by the CAO is expected to result in the full harmonization of the system. The CAO was expected to 22 perform the first round of yearly auctions for 2014 annual capacities (Poyry Management Consulting and Nord Pool Consulting, 2011). 4. Key lessons from the Case Studies This section highlights the lessons learned from the various trading arrangements discussed in this paper. We organize these under a number of key headings. Pre-conditions for trade: Trade Agreements/Transmission Availability Both expanded bilateral power trading and more formal power pools require a broader pre- commitment to free trade to be successful. Electricity is just one commodity that could be freely traded across borders. It is not clear that it can be traded easily without a prior commitment to the creation of a free trade area. The existence of a regional trade can aid regional electricity trading in several ways: the existence of a regional trade reduces/removes possible trade barriers to regional power market and reduces planning time as most of the rules and regulations necessary for regional trading would have been established. The need for specific trade agreements to support electricity trade is heightened by the fact that WTO rules do not adequately address trade in electricity, partly because it combines goods (production) and services (transmission), and involves other policy objectives to do with the environment and energy security. 33 Free trade arrangements between countries leads to the establishment of the trust required to promote the development of a regional power pool. Five of our case studies are based within established free trade areas; the sixth – ECSEE straddles the EU and its near neighbors and is the least institutionally developed. All of them are clearly being promoted within a wider trade development context. Thus, SAPP is an offshoot of the Southern Africa Development Community (SADC) while WAPP is a subsidiary of the Economic Community of West African States (ECOWAS). The existence of regional trade might be the reason why bilateral power trades always continue to thrive even before an advanced power pool is achieved. Our case studies are consistent with the hypothesis that greater trade openness leads to more cross- border trade in electricity, even keeping the potential gains from trade constant. The existence of a common currency is not a pre-condition for an effective functioning electricity market. Although a common monetary unit does help facilitate trade because it signals deep trade integration, this is not necessary and fully functioning cross border electricity markets can exist in the presence of multiple currencies – the Single Electricity Market (SEM) in Ireland is a good example here. Similarly, SAPP members do not have a common currency. Similarly, it should be pointed out that institutions of competition policy enforcement across a free trade area are not necessary for power pools to develop and function effectively (they do not exist 33 For a discussion see: http://www.wto.org/english/res_e/publications_e/wtr10_forum_e/wtr10_7may10_e.htm 23 for the SAPP or SIEPAC countries). However, such institutions are a sign of deep trade integration – and exist in the US and in the EU – and do act as regulators of mergers and market power in cross jurisdictional power pools (such as PJM, Nord Pool and SEM). Fuel costs may be reflected more quickly in power prices in an international power market, than in a managed national market. This is economically efficient but does expose wholesale electricity price dependent customers to significant variability in prices. PJM is a good example of this. PJM note that average real time prices rose 24% in 2010 and fell 25% in 2012. 34 Modelling would expose whether such volatility is likely to be higher or lower with an integrated market. Such market risks may have implications for the willingness of jurisdictions to be involved in cross-border trading of electricity or their need for institutional arrangements which help mitigate risks. It almost goes without saying that for trade to occur there must be a no trade price differential between the potential parties to the trade. This implies that there must be a relatively low electricity price country, where the unit value of electricity exports to the relatively high price country is greater than the willingness to pay for a unit electricity consumption within the country. In many countries of WAPP this is not the case, in the sense that the price an importing country would be willing to pay for a kWh is less than the willingness to pay of demand for a kWh within the potential exporting country. This could be case for a country with a lot of un-served load and underdeveloped hydro capacity. In theory this country should be exporting, but only after satisfying its latent domestic demand. This fact may explain the slowness of the development of international interconnection from countries with favorable export potential. Adequate transmission capacity is essential for power trading to occur. Thus, agreements for expanding transmission capacity are integral to the development of an international power pool. A generalized commitment to free trade is not enough to promote the development of a fully functioning cross-jurisdictional power pool, with a single market price. This would seem to require physical investment in sufficient cross-border transmission capacity. SEM only came about after the re-energizing of the North-South interconnector. The development of WAPP beyond bilateral trading has been prevented by a lack of transmission capacity. Lack of transmission capacity has not prevented the emergence of spot markets in SIEPAC or SAPP but has severely limited their significance and explains the prolonged dominance of bilateral trading. By contrast in developed regional power pools such as PJM or Nord Pool there has been a significant amount of transmission capacity which has supported the development of significant power pools. While our case studies suggest power markets can come before the development of interconnection (e.g. in the case SIEPAC), they can only be significant in the presence of sufficient interconnection. The development of sufficient transmission capacity is a problem within all power systems. This is because there almost always practical and political problems in allocating the gains from transmission capacity expansion. This explains why in national power systems transmission expansion costs were historically socialized across the whole of society (e.g. within the CEGB in England and Wales, or within US vertically integrated utilities) rather than fully allocated to specific sets of customers and generators, via locational charging. Thus it is easy to understand the difficulty in getting agreements to allocate the costs of interconnecting transmission between jurisdictions. 34 ‘PJM The Cost of Electricity’, PJM website. 24 However ‘cross border’ investments in transmission capacity did occur to a limited extent historically where there were sufficiently large bilateral gains relative to the cost of transmission expansion (i.e. large price differentials which can be arbitraged by relatively short distance wires). This explains the existence of some transmission capacity between jurisdictions within all of our case studies, prior to their most recent institutional incarnation. However the expansion of this initial transmission capacity - to the extent that it has occurred – has been supported by feasibility studies which have attracted multilateral agency finance. SIEPAC, WAPP, SAPP have all been financed by international development agencies (IADB, World Bank and AfDB); while the EU has financially supported interconnection projects in the SEM and ECSEE. Institutional Arrangements The role of strong, efficient and independent institutions in ensuring an effective functioning integrated power market cannot be over-emphasized. An integrated power pool needs an efficient operator who can oversee and sanction the activities of participants in order to prevent predatory pricing, non-disclosure of capacity, and other forms of unruly behavior. Cross-border institutional arrangements are essential, but can take a variety of forms. WAPP and SAPP are international arrangements for electricity exchanges. PJM is an actual cross-jurisdictional ISO. The ECSEE has evolved a joint institution for allocating cross border transmission capacity, separate from the national system operators. While the strength of the institutional arrangements governing cross-border trading is undoubtedly limited if it is cross border, it clearly is possible to build strong cross-border power markets (as demonstrated by SEM and SIEPAC). The SEM has involved the merger of the system operators in both participating jurisdictions (though they still continue to exist separately). This is helped where the parties to the transaction are private entities and can submit themselves to be bound by a common legal system. Indeed the problem is not the lack of a common legal system to which all the parties can be bound, but the unwillingness of sovereign countries to bind themselves to a common enforcement mechanism. Such institutions do involve some additional cost (as Mansur and White, 2009, found for PJM) but these are likely to be small in relation to the benefits. Getting the appropriate combination of regulation and market design for power pools is important. While a new cross border regulatory agency is not necessary, some regulatory oversight is beneficial. SAPP could have benefited from some regulation of Eskom’s potentially predatory pricing behavior. The SIEPAC, PJM, SEM and ECSEE are subject to some form of external regulatory oversight. PJM is clearly the most developed as it is subject to oversight from the US national energy regulator, FERC. Cross-border electricity markets within the EU are subject to the jurisdiction of the European Commission, which regulates EU wide competition. Market design may be important for reducing the need for cross-border regulatory enforcement action. For instance, in small markets such as Ireland price based bidding would have involved large distributional transfers and potentially significant welfare losses. Hence the introduction of cost based bidding with a capacity market in the SEM. The price determination process should both be based on underlying economic cost and take account of the potential for market power and the existing pricing inefficiencies within the trade partners. Clearly disputes about price determination are potentially more difficult to resolve in a cross-border market than in a national market and hence should be avoided. 25 Countries may have rational reasons to wish to mitigate the price risks associated with electricity market integration. Small countries faced with increasing their exposure to foreign sources of electricity price volatility may be unwilling to have much exposure to short term international market prices and prefer most trades to occur under long term contracts at fixed prices. This may explain the lack of trading in the DAM market of SAPP and the preference for bilateral contracting in SAPP and SIEPAC. Clearly as long as the price determination process for long term contracts is reasonably efficient then this may not be a problem and the market should decide the mix of contract terms under which electricity is exchanged. The use of day-ahead markets and/or real time markets facilitates more trade and greater market efficiency. Day-Ahead/real time trading leads to more competition than in bilateral arrangements and therefore results in more efficient utilization of resources. The bidding mechanism in day ahead/real time market tends to make suppliers to be more efficient in order to keep their marginal costs as low as possible since they can be bid out of trade/market if they bid too high price. A day ahead/real time market also is more flexible and does more to facilitate trade than bilateral arrangements alone. This seems to be in line with the experience of SIEPAC (where most trade is bilateral) vs the SEM in Ireland (where there is a compulsory day ahead pool). Timeline/Process of how to get there The creation of cross border trading capacity often has the effect of inducing further market integration. Trade success demonstrates the concept of trade being mutually beneficial. Several of our examples show that initial trading arrangements have expanded over time to include more jurisdictions, building on the initial cooperating group. This is true of PJM (extending from three utilities) and SEM (which is extending its transmission capacity and increasingly integrating with the GB market). There can and should be a timetable for reform and development. Developing and keeping to a timetable and is essential for the rapid development of integrated power market. It is necessary that agreeing parties design a timetable with clear objectives and the procedures or processes required to get there. This would allow them to keep tracks of their progress as well as having a definite focus. In each of our jurisdictions a timeline of development can be identified. The setting of a clear timetable in the case of the SEM and ECSEE seems to have been successful in reaching the next stage of development in the market. There can be an important role for international organizations to facilitate the creation of power pools. The creation of a regional power pool requires substantial investments in building and updating generation capacity, transmission networks, and human development. All five of the international power markets we look at have received significant external development support and financing (from the AfDB, IADB and the EU). The support levels are significant (59% of the funding for SIEPAC came from the IADB). Thus, the evidence is that the support of international organizations or foreign capital (as bilateral aid or development finance) is required. 26 In SAPP the gap between bids and actual traded volumes in DAM are always due to transmission constraints. Similarly, the slow development of trading in the WAPP are caused by lack of adequate funding required to embark on massive infrastructural developments necessary for effective trading arrangements, although they do receive foreign support. Things probably would have been better if an international organization had been fully in charge of the creation and development of WAPP for a possible hand-over to the member countries at a later date. Clearly sovereign countries may be reluctant to do this, but the fact is that the SIEPAC and ECSEE countries have been willing to do this. The creation of the ECSEE is a particularly good example of such an initial agreement between an international organization and the countries, which has now been given to the countries. The viability (or otherwise) of an international power pool should be assessed in advance by a careful cost-benefit analysis. Clearly any major policy change should be subject to an impact assessment. This is particularly true of electricity markets which lend themselves to modelling and can provide a quantification of likely benefits and identify the need for any side payments to countries who facilitate trade but do not directly benefit (this was a particular issue in the SIEPAC case). Also, fuel costs may be reflected more quickly in power prices in an international power market, than in a managed national market. This can expose wholesale electricity price dependent customers to significant variability in prices. Modelling would expose whether such volatility is likely to be higher or lower with an integrated market. Power pools can (and should) start with a small number of countries and grow over time. The US and EU experience also suggests that regional power markets – involving just two or three parties - might be a good place to start on the road to full market integration. SEM involves two countries, PJM started with three utilities. Nord Pool became an international power market by expanding from Norway to include Sweden in 1996 (it now covers trades between nine countries). Indeed, one of the observations from our case studies is that the most integrated markets are those that have grown organically rather than those that started with a large number of jurisdictions. The slow growth of WAPP may be a function of the large number of participating countries – 14 - at the beginning. Starting small means that large gains from trading can be demonstrated and new parties willingly opt in to an existing working arrangement. This would seem to offer more chance of steady deep progress, rather than prolonged initial development periods. Reasons to be hopeful about the prospects for international power trading in the other regions, such as SAR Trust building around electricity trading is possible even between countries with a history of conflict. Five of our six case studies are drawn from troubled regions. The nature of the trouble was not necessarily at the border but sometimes internal conflict (and hence a potential source of supply risk for international partners). However in the case of SAPP and ECSEE there have been cross-border conflicts in the past. Often electricity trading by reinforcing mutual interdependence can be a significant positive outcome for the conflict resolution process (e.g. SEM and hopefully ECSEE). 27 The potential gains within the South Asian Region (SAR) from cross-border electricity trade are large. ESMAP (2008) discusses the nature of the gains from increased energy trading in the SAR. This report discusses the bilateral (and multilateral) electricity trades that might be mutually beneficial within the region. Of these, six are relevant to our initial list of eight SAR countries: Pakistan-India and Pakistan-Afghanistan; Bhutan-India; Nepal-India; Bangladesh-India; India-Sri Lanka; and Bangladesh, Bhutan, Nepal and India multilateral trade. The ESMAP report discusses the absence of electricity interconnection between Pakistan and India and the difficulties of interconnecting Pakistan with Afghanistan, suggesting that the gains from trade are modest. Bhutan-India is the one electricity trade success story in the region, with Bhutan deriving significant government revenue from the export of hydro based electricity to India. In 2010 Bhutan exported 5.4 TWh (or 75% of its production) 35 to India. Nepal-India has the biggest unexploited trade potential. Nepal has 43 GW of identified economic hydro capacity, and an installed capacity of just 721 MW in 2010. Much of any hydro capacity could be exported to India, but current total exports were 1% of its current production (or 0.03 TWh) in 2010. Bangladesh-India electricity trade is also at a very low level in spite of good prospects for the export of gas based generation from Bangladesh to India. In 2010 Bangladesh imported a total 0.5 TWh of electricity (with zero exports). In spite of considerable under-utilized hydro resources in Sri Lanka and a mere 30 km of sea to India, there remains no interconnection between the two countries. In terms of a regional electricity market, there would seem to be a lot of potential for a joint Bangladesh, Bhutan, Nepal and India market, aimed at exploiting the considerable hydro potential of Bhutan and Nepal. Studies carried out by USAID under its SARI-E program 36 identified relative low cost transmission investments ($9m to $52m at the time) which would significantly increase cross border transmission capacity in the north east border region of India and its three neighbors. Chattopadhyay (2013) quantitatively examines the high benefit to cost ratios across three potential cross border links within the SAR: India-Sri Lanka, India-Bhutan and India-Nepal. The major sources of benefit are avoiding unserved energy, operational cost benefits and capacity benefits, when measured using an investment planning and optimization model. These benefits are $1.8bn p.a., against one-off transmission line costs of $700m. Chattopadhyay does not discuss the distributional impacts and why the links are not already built if they are really so beneficial. Clearly if such large potential gains exist it would be worth understanding what barriers stand in the way of electricity trade across the SAR and how they can be overcome. The measured benefits in the SEM and PJM cases were significantly less attractive than this (proportionally), but are in the process of being realised, in the light of the prescriptions for successful power pool development outlined above. Taken together ESMAP (2008) and Chattopadhyay (2013) identify considerable potential for regional electricity trading, much of it unexploited, and only at the feasibility study stage. 35 Source of recent electricity production, import and export figures: CIA The World Factbook. Available at https://www.cia.gov/library/publications/the-world-factbook/index.html. 36 See: http://www.sari-energy.org/ 28 Suggestions for further research There is much we do not know about the promotion of international electricity trade. While the empirical literature has extensively examined the impact of electricity liberalization on electricity market outcomes (see Pollitt, 2012b, for a review of the literature), there has been relatively little examination of either how international electricity trade can be promoted or the impact of trade. There has been some work on the impact of market extension within the US, but outside the US the work is almost entirely at the level of impact studies of the potential for trade. What is required is work looking at actual market evolution and impact. A number of interesting areas for future econometric work suggest themselves. First, what explains the extent of involvement in international trade? We have described the evolution of six power trading arrangements that took years to develop. Can we explain the extent of trade with reference to opportunity, national institutional variables, membership of a free trade area, institutions of the international electricity market, international financial support etc.? Taking a sample of countries over time, this analysis would draw on the New Institutional Economics literature (in line with Erdogdu’s (2013) analysis of national progress with electricity reform). Nord Pool evolved steadily (and reasonably quickly) from its starting point in Norway. WAPP has taken decades to get going, while SIEPAC has finally completed its major transmission line (more than 25 years after its initial feasibility study). Distinguishing the relative impact of trade potential and institutional arrangements would seem to be worthwhile. Second, what price or quality of service impact does international trade in electricity actually have? These impacts could be on national electricity price levels or volatility, power shortages etc. Trade should have an impact in bringing about price convergence (following Mansur and White, 2009, on PJM extension). It might also mitigate domestic power shocks (such as hydro shortages, or a major power plant outage). It might relieve shortages of electricity (or allow domestic market extension to more customers). Econometric analysis of national wholesale or industrial prices could be linked within trade involvement variables to see the extent to which international power trading has demonstrable national impacts. Clearly, small amounts of power trading might be enough to cause significant price convergence between neighboring countries. Such analysis could focus on particular institutional developments within power pools. For example, SAPP has existed for many years and has shown significant institutional development. One could look at the impact of the move from bilateral trading only to a Day Ahead Market. Third, what is the productive efficiency impact of electricity trade? The merit order within a trading region is significantly altered when the market moves from being national to being international. Following, Douglas (2006, on US coal fired power plants), the efficiency of power plant utilization after the introduction of a power pool can be examined (e.g. within SAPP or SIEPAC). It would also be possible to examine the extent to which international trade impacted the efficiency of individual power plants, in a way similar to the impact of privatization or liberalization more generally. Alternatively, the impact of involvement in a power pool on the promotion of electricity productivity could be examined, at the plant or utility level. For instance, Estache et al. (2008) and Jaunky (2013) look at utility level productivity evolution within SAPP. 29 Finally, it is worth noting that it would be good to see more careful cost-benefit analyses of the actual outturn performance of power pools in order to see if the potential initially identified is being realized. SIEPAC and SEM were both undertaken following positive cost benefit analyses: it would be highly informative to undertake actual cost benefit analyses (following Newbery and Pollitt, 1997) of their realization in order to quantify the outturn impacts of these developments. REFERENCES AF-MERCADOS EMI (Energy Market International). (2012). Development of WAPP Market Design and Market Rules. Final Report Prepared for WAPP. AF-MERCADOS EMI (Energy Market International). Retrieved from http://www.erera.arrec.org/getattachment/News-Events/News-archive/Proposed- Regional-Market-Design-and-Market-Rules/WAPP-Market-Rules-last-Final.pdf.aspx Amundsen, E.S. and Bergman, L. (2006). ‘Why has the Nordic electricity market worked so well?’, Utilities Policy, 14(3), 148–157. Amundsen, E.S. and Bergman, L. (2007). ‘Integration of multiple national markets for electricity: The case of Norway and Sweden’, Energy Policy, 35(6), 3383–3394. Antweiler, W., Copeland, B.R. and Taylor, M.S. (2001), ‘Is Free Trade Good for the Environment?’, American Economic Review, 91(4), 877-908. Bahar, H. and Sauvage, J. (2013). ‘Cross-Border Trade in Electricity and the Development of Renewables-Based Electric Power: Lessons from Europe’, OECD Trade and Environment Working Papers, 2013/02, OECD Publishing. Brunekreeft, G., Neuhoff, K. and Newbery, D. (2005). ‘Electricity transmission: An overview of the current debate’, Utilities Policy, 13(2), 73-93. CASTALIA. (2009). International Experience with Cross-border Power Trading. Report to the Regional Electricity Regulators' Association (RERA) and the World Bank. Retrieved from http://documents.worldbank.org/curated/en/2009/09/16453555/international-experience-cross- border-power-trading Chattopadhyay, D. (2013). ‘Cross-border Power Trading in South Asia: Modelling Analysis to Assess Economic Benefits’, Power and Energy Society General Meeting (PES), 2013 IEEE 21-25 July 2013, Vancouver, BC. Chawla, M. and Pollitt, M. (2013). ‘Energy-efficiency and environmental policies & income supplements in the UK: Their evolution and distributional impact in relation to domestic energy bills’, Economics of Energy and Environmental Policy, 2(1), 19-38. Conlon, P. (2009). The Integration of Electricity Markets in Ireland under the ISO Model. ESB International. Retrieved from http://www.esbi.ie/news/pdf/White-Paper-Integration-Electricity- Markets.pdf de Nooij, M. (2011). ‘Social cost-benefit analysis of electricity interconnector investment: A critical appraisal’ Energy Policy, 39(6), 3096–3105. 30 de Villemeur, E. B., & Pineau, P.-O. (2012). ‘Regulation and electricity market integration: When trade introduces inefficiencies’, Energy Economics, 34(2), 529-535. Deitz, L., Stirton, L., and Wright, K. (2007). The Energy Community of South East Europe: Challenges of, and Obstacles to, Europeanisation. Centre for Competition Policy, CCP Working Paper 08-04. https://www.uea.ac.uk/polopoly_fs/1.104657!ccp08-4.pdf Deitz, L., Stirton, L., and Wright, K. (2009). ‘South East Europe’s electricity sector: Attractions, obstacles and challenges of Europeanisation’, Utility Policy, 17, 4-12. Dickson, P.R. and Kalapurakal, R. (1994). ‘The use and perceived fairness of price-setting rules in the bulk electricity market’, Journal of Economic Psychology, 15(3), 427-448. Douglas, S. (2006). ‘Measuring Gains from Regional Dispatch: Coal-Fired Power Plant Utilization and Market Reforms’, Energy Journal, 27 (1), 119-138. Dixit, A. and Norman, V. (1986). ‘Gains from trade without lump-sum compensation’, Journal of International Economics, 21(1-2), 111-122. Economic Consulting Associates (ECA). (2009). The Potential of Regional Power Sector Integration: South African Power Pool (SAPP) Transmission & Trading Case Study. A Report Submitted to ESMAP. Retrieved from http://www.esmap.org/sites/esmap.org/files/BN004-10_REISP- CD_South%20African%20Power%20Pool-Transmission%20&%20Trading.pdf Economic Consulting Associates (ECA). (2010), The Potential of Regional Power Sector Integration: Central American Electric Interconnection System (SIEPAC) Transmission & Trading Case Study. A Report Submitted to ESMAP. Retrieved from http://www.esmap.org/sites/esmap.org/files/BN004- 10_REISP-CD_Central%20American%20Electric%20Interconnection%20System- Transmisison%20&%20Trading.pdf Erdogdu, E. (2013). ‘A cross-country analysis of electricity market reforms: Potential contribution of New Institutional Economics’, Energy Economics, 39(C): 239-251. ESMAP (2008). Potential and Prospects for Regional Energy Trade in the South Asia Region. Washington, D.C.: ESMAP, World Bank Group. Estache, A., Tovar, B., and Trujillo, L. (2008). ‘How efficient are African electricity companies? Evidence from the Southern African countries’, Energy Policy, 36(6), 1969–1979. FERC (2004). Staff Report on Cost Ranges for the Development and Operation of a Day One Regional Transmission Organization Docket No. PL04-16-000, FERC: Washington, D.C., available at: http://www.ferc.gov/EventCalendar/Files/20041006145934-rto-cost-report.pdf, last accessed 29 January 2011. Ferrero, R.W., Shahidehpour, S.M. and Ramesh, V.C. (1997). ‘Transaction Analysis in Deregulated Power Systems using Game Theory’, IEEE Transactions on Power Systems, 12(3), 1340-1340. Finon, D. and Romano, E. (2009). ‘Electricity market integration: Redistribution effect versus resource reallocation’, Energy Policy, 37, 2977–2985. 31 FitzGerald, J., Keeney, M., McCarthy, N., O’Malley, E. and Scott, S. (2005). Aspects of Irish Energy Policy, Economic and Social Research Institute Policy Research Series WP 57. Foreman-Peck, J. and Waterson, M. (1985). ‘The Comparative Efficiency of Public and Private Enterprise in Britain: Electricity Generation between the World Wars’, Economic Journal, 95(380a), 83-95. Gately, D. (1974). ‘Sharing the Gains from Regional Cooperation: A Game Theoretic Application to Planning Investment in Electric Power’, International Economic Review, 15(1), 195-208. Goldman, C., Lesieutre, B.C. and Bartholomew, E. (2004). ‘A review of market monitoring activities at U.S. independent system operators’, Lawrence Berkeley National Laboratory http://escholarship.org/uc/item/87m978zf Hobbs, B.F.; Fieke A.M.R.; Maroeska G. Boot (2005). ‘The More Cooperation, The More Competition? A Cournot Analysis of the Benefits of Electric Market Coupling’, The Energy Journal, 26(4), 69-97. Hogan, W. (1995). ‘A Wholesale Pool Spot Market Must Be Administered by the Independent System Operator: Avoiding the Separation Fallacy’, The Electricity Journal, 8(10), 26–37. Hogan, W. (1998). Independent system operator: Pricing and flexibility in a Competitive Electricity Market, Center for Business and Government, Harvard University, http://www.hks.harvard.edu/fs/whogan/iso98.pdf. Jamasb, T., and Pollitt, M. (2005). ‘Electricity Market Reform in the European Union: Review of Progress toward Liberalization & Integration’, The Energy Journal, European Energy Liberalisation Special Issue, 11-40. Jaunky, V. C. (2013). ‘Divergence in technical efficiency of electric utilities: Evidence from the SAPP’, Energy Policy, 62, 419–430. Jia, N.X. and Yokoyama, R. (2003). ‘Profit allocation of independent power producers based on cooperative Game theory’, International Journal of Electrical Power & Energy Systems, 25(8), 633- 641. Joskow, P.L. (1997). ‘Restructuring, Competition and Regulatory Reform in the U.S. Electricity Sector’, Journal of Economic Perspectives, American Economic Association, 11(3), 119-138, Summer. Joskow, P.L. (2005). ‘Transmission policy in the United States’, Utilities Policy, 13(2), 95–115. Kagiannas, A.G., Askounis, D. and Psarras, J. (2004). ‘Power generation planning: a survey from monopoly to competition’, International Journal of Electrical Power & Energy Systems, 26(6), 413- 421. Kogut, B. (1988). ‘Joint ventures: Theoretical and empirical perspectives’, Strategic Management Journal, 9(4), 319-332. Kristiansen, T. (2007). ‘Cross-border transmission capacity allocation mechanisms in South East Europe’, Energy Policy, 35(9), 4611–4622. 32 Kumar, A., Srivastava, S.C. and Singh, S.N. (2005). ‘Congestion management in competitive power market: A bibliographical survey’, Electric Power Systems Research, 76(1-3), 153–164. Kunz, F. and Zerrahn, A. (2013). ‘The Benefit of Coordinating Congestion Management in Germany’, 10th International Conference on the European Energy Market (EEM), 27-31 May 2013, 1-8. Lahiri, S. and Ono, Y. (1996). ‘The Role of Free Entry in an Oligopolistic Heckscher-Ohlin Model’, International Economic Review, 36(3), 609-24, August. Lise, W., Linderhof, V., Kuik, O., Kemfert, C., Östling, R. and Heinzowe, T. (2006). ‘A game theoretic model of the Northwestern European electricity market—market power and the environment’, Energy Policy, 34(15), 2123–2136. Mansur, E.T. and White, M.W. (2009). ‘Market Organization and Efficiency in Electricity Markets’, Mimeo. Markusen, J.R. (1981). ‘Trade and the gains from trade with imperfect competition’, Journal of International Economics, 11(4), 531-551, November. Martin, J. M., and Posadas, J. C. (2012). ‘Central America’s Electric Sector: The Path to Interconnection and a Regional Market’, Journal of Energy Security. Retrieved from http://www.ensec.org/index.php?option=com_content&view=article&id=375:central-americas- electric-sector-the-path-to-interconnection-and-a-regional-market&catid=128:issue- content&Itemid=402 McCarthy, N. (2005). ‘Market Size, Market Structure & Market Power in the Irish Electricity Industry’, Working Paper No. 168, August, Dublin: The Economic and Social Research Institute. Musaba, L. (2009). Energy Trading in the SAPP. SAPP Coordination Centre, Zimbabwe. Retrieved from http://www.sapp.co.zw/ Neary, J.P. (2007). ‘Cross-Border Mergers as Instruments of Comparative Advantage’, Review of Economic Studies, 74, 1229–1257. NERA Economic Consulting. (2006). A Cost-Benefit Study of the Single Electricity Market: A Final Report for NIAER & CER. Retrieved from http://www.detini.gov.uk/a_cost_benefit_study_of_the_single_electricity_market_- _a_final_report_for_niaer_and_cer__december_2006_.pdf Neuhoff, K. and Newbery, N. (2005). ‘Evolution of electricity markets: Does sequencing matter?’, Utilities Policy, 13, 163-173. Newbery, D. M. and Pollitt, M. G. (1997). The Restructuring and Privatisation of Britain’s CEGB-Was It Worth It? The Journal of Industrial Economics, 45(3), 269–303. Nexant. (2004a). West Africa Regional Transmission Stability Study — Project Summary Report. Final report prepared for USAID and ECOWAS Secretariat, Nexant, Washington, DC. 33 Nexant. (2004b). West Africa Regional Transmission Stability Study — Vol. 2: Master Plan. Final report prepared for USAID and ECOWAS Secretariat, Nexant, Washington, DC. O'Donnell, A. (2003). Soul of the Grid: A cultural biography of the California Independent System Operator. Lincoln, US, iUniverse. O'Mahoney, A. and Denny, E. (2013). ‘Electricity prices and generator behaviour in gross pool electricity markets’, Energy Policy, 63, 628–637. O’Neill, R., Helman, U., Hobbs, B.F. and Baldick, R. (2006). ‘Independent System Operators in the USA: History, Lessons Learned, and Prospects’, in F.P.Sioshansi and W.Pfaffenberger (eds.). Electricity Market Reform: An International Perspective, Oxford: Elsevier. Oseni, M. and Pollitt, M. (2013). The Economic Costs of Unsupplied Electricity: Evidence from Backup Generation among African Firms’, EPRG Working Paper, No.1351. Pérez-Arriaga, I.J. and Olmos, L. (2005). ‘A plausible congestion management scheme for the internal electricity market of the European Union’, Utilities Policy, 13 (2), 117-134. Pineau, P.-O. (2008). ‘Electricity sector integration in West Africa’, Energy Policy, 36(1), 210-213. Pollitt, M. (2009). ‘Evaluating the evidence on electricity reform: Lessons for the South East Europe (SEE) market’, Utilities Policy, 17(1), 13-23. Pollitt, M. (2012a). ‘Lessons from the History of Independent System Operators in the Energy Sector’, Energy Policy, 47(August), 32-48. Pollitt, M. (2012b), ‘The Role of Policy in Energy Transitions: Lessons from the Energy Liberalisation Era’, Energy Policy, 50, November: 128-137. Poyry Management Consulting and Nord Pool Consulting. (2011). South East Europe Wholesale Market Opening: Final report – updated with Ukraine and Moldova. Retrieved from http://www.energy-community.org/pls/portal/docs/1304177.PDF Prada, J. F., Bowman, D., Petrov, K., Calderón, E., Ríos, R., & Corredor, P. (2004). The Regional Electricity Market of Central America. Retrieved from http://www.cigre.org Reinstein, D., Mateos, A., Brugman, A., Berman, L., & Johnson, T. (2011). Regional Power Integration: Structural and Regulatory Challenges. Report No: 58934 - LAC. Energy Sector Management Assistance Program (ESMAP). SAPP (Southern African Power Pool). (2013a). SAPP Annual Review Reports for Various Years – Statistics and Planning Data. Retrieved from http://www.sapp.co.zw SAPP (Southern African Power Pool). (2013b). SAPP Market Monthly Performance Report, November 2013. Retrieved from http://www.sapp.co.zw/mreports.html SAPP (Southern African Power Pool). (2013c). SAPP Sustainability Bulletin. SAPP Bulletin, 21(21). Retrieved from http://www.sapp.co.zw 34 Stoft, S. (1996). ‘California's ISO: why not clear the market?’, The Electricity Journal, 9(10), 38-43. Tinoco, M. (2009). ‘Transmission tariffs and planning in the central American regional electricity market’, 2009 IEEE Power & Energy Society General Meeting, (pp. 1-7). Retrieved from http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5275744 Vailati, R. (2009). ‘Electricity transmission in the energy community of South East Europe’, Utilities Policy, 17, 34-42. von der Fehr, N.-H. and Harbord, D. (1998). ‘Competition in Electricity Spot Markets. Economic Theory and International Experience’, Memorandum 05/1998, Oslo University, Department of Economics. Walawalkar, R., Fernands, S., Thakur, N. and Chevva, K.R. (2010). ‘Evolution and current status of demand response (DR) in electricity markets: Insights from PJM and NYISO’, Energy 35(4), 1553– 1560 WAPP (West African Power Pool). (2012). Annual Activity Report of the WAPP Secretary General., (pp. 1-48). Retrieved from http://www.ecowapp.org/?page_id=10 Zachmann, G. (2008). ‘Electricity wholesale market prices in Europe: Convergence?’, Energy Economics, 30, 1659-1671. Zakharov, I. (2003). West Africa Power Pool (WAPP) Just a Matter of Time? USAID Project Managed by PA Government Services, Inc. Retrieve from lasen.epfl.ch/webdav/site/lasen/shared/import/migration/Zakharov.pdf 35 APPENDIX 1: Cross-Border Trading in Selected Countries by Power Pool The table below shows the proportion of cross-border electricity trades to interconnector capacity (transmission capacity) and total electricity consumption in some selected countries across various Power Pools. Data on annual electricity consumption and trades were converted from kWh to MW for them to be in the same unit as interconnector capacity. Conversion was done by dividing the annual data on each variable in kWh by 1000*24*365. The results as reported in the table indicate that a lot of trades take place in Nord Pool and in SAPP than in the other power pools. For instance, while the interconnection capacity utilisation are respectively 39% and 50% for Nord Pool and SAPP, WAPP and SIEPAC records just approximately 9% and 4% respectively. Similarly, the proportions of cross-border trades in consumption are 28% and 21% for Nord Pool and SAPP respectively, whereas WAPP and SIEPAC trade only 5% and 2% of their consumption. It is worth noticing that while SAPP appeared to have utilised its transmission capacity (50%) more than the Nord Pool (39%), Nord Pool has traded more of its consumption (28%) than the SAPP (21%). The performance of SAPP does stand out relative to other developing countries’ power pools. The reason for SAPP to have achieved so much might be because of the (excess) existing capacity in South Africa which makes trading possible. Unlike the WAPP whose installed capacity stood at only 9,705MW as at the time it was formally created, Southern African Power Pool (SAPP) installed capacity at inception in 1995 was 48,461 MW, with about 38,000 MW in South Africa alone. Similarly, SAPP had been created many years before other pools in developing countries - e.g. WAPP was established in 2000/2001. Those years of existence might have been used to develop the pool – development and upgrading of transmission networks, etc, in order to promote trading within the pool. 36 Table A1: Share (%) of Trade in Electricity Consumption and Transmission Capacity in Selected Countries by Pool Selected Transmission Installed Electricity Exports Imports Crossborder Trade % of Trade Country cap. (MW) cap consp (MW) (MW) trade (MW) transmision % of (MW) (MW) cap. consp Nord Pool Norway 5395 30180 14703.2 1,974 390 2,363 43.8 16.1 Sweden 11825 36510 15353.9 1,679 1,455 3,135 26.5 20.4 Finland 3310 16680 9960.0 381 1,839 2,220 67.1 22.3 Denmark 6405 13710 3915.5 1,297 1,463 2,760 43.1 70.5 Estonia 1941 2751 877.4 264 156 420 21.6 47.9 Latvia 2084 2166 778.8 242 530 772 37.1 99.2 Lithuania 2714 3820 1097.3 754 645 1,399 51.5 127.5 Total 33674 105817 46686.1 6,591 6,479 13,070 38.8 28.0 West African Power Pool (WAPP) Ghana 627 1985 650.9 28.4 49.7 78.1 12.5 12.0 Nigeria 855 5900 2192.9 0.0 0.0 0.0 0.0 0.0 Senegal 100 638 158.0 0.0 0.0 0.0 0.0 0.0 Cote d'Ivoire 327 1222 368.8 88.1 0.0 88.1 27.0 23.9 Total 1909 9745 3370.7 116.6 49.7 166.2 8.7 4.9 Central American Power Market (SIEPAC) Guatemala 300 2745 812.2 15.1 0.9 16.0 5.3 2.0 El Salvador 300 1491 533.8 0.8 4.3 5.1 1.7 1.0 Honduras 300 1701 536.1 0.0 1.3 1.3 0.4 0.3 Nicaragua 300 1108 293.3 0.0 7.3 7.3 2.4 2.5 Costa Rica 300 2800 920.5 8.8 23.2 32.0 10.7 3.5 Panama 300 1976 590.2 14.3 1.0 15.3 5.1 2.6 Total 1800 11821 3686.1 38.9 38.1 77.0 4.3 2.1 Southern African Power Pool (SAPP) DRC 260 2437 684.6 218.7 0.7 219.4 84.4 32.0 Zambia 1400 1679 1008.9 30.6 25.3 55.9 4.0 5.5 Mozambique 5800 2428 1159.8 1,349.3 945.0 2,294.3 39.6 197.8 Botswana 1450 132 302.3 0.0 249.0 249.0 17.2 82.4 South Africa 2050 44260 24554.8 1,616.4 1,206.6 2,823.1 137.7 11.5 Lesotho 230 76 35.1 0.0 28.2 28.2 12.3 80.5 Namibia 750 508 415.0 10.4 287.6 298.0 39.7 71.8 Total 11940 51520 28160.4 3,225.5 2,742.4 5,967.8 50.0 21.2 Data Sources: CIA: The World Fact Book https://www.cia.gov/library/publications/the-world- factbook/index.html International Renewable Energy Agency, IRENA (2013) WEST AFRICAN POWER POOL: Planning and Prospects for Renewable Energy. http://www.irena.org/DocumentDownloads/Publications/WAPP.pdf International Renewable Energy Agency, IRENA (2013) SOUTHERN AFRICAN POWER POOL: Planning and Prospects for Renewable Energy. http://www.irena.org/DocumentDownloads/Publications/SAPP.pdf 37