1 Market Abuse: the things some companies do 29 CHAPTER 4 MARKET ABUSE: THE THINGS SOME COMPANIES DO This requires comprehensive structural reform, [because] even the most diligent competition enforcement cannot solve all the problems in these markets. Neelie Kroes, competition commissioner at September 2007 launch of the Commission's Third liberalization package. You might have thought the liberalization measures mentioned in the previous chapter would have resulted in less concentration and more integration. After all, doing away with national monopolies in the import and export of energy (in the early 1990s) should have been a blow to dominant incumbents and a spur to their domestic rivals, while the phasing in of full cross-border competition (by mid-2007) should have brought foreign rivals into national markets and increased EU-wide integration. This has not really come to pass. Some incumbents, so-called because they occupied the dominant position in their home markets when liberalization began, have lost market share. But frequently this has happened not as the result of market forces, but of government moves to cap and reduce an incumbent's market share action (as in the case of Italy's Enel) or to require an incumbent to auction off some of its gas or power to smaller competitors. Other incumbents have gained; Gaz de France already had a 90 percent share of the French gas market plus control of most storage and transmission before its government-orchestrated merger with Suez. Many have held their own. A few companies such as Germany's Eon have taken on a more pan-European character by making acquisitions in open markets like the UK and privatizations in East and Central Europe. But only in these latter markets has the concentration of dominant incumbents diminished at all. Brussels' competition directorate reviewed the energy industry's structure in its sector study. Liberalization, it found, had not changed the wholesale gas supply chain. 'The high level of concentration which existed in most national markets at liberalization largely remains." Incumbents dominate production, imports and trading on any gas hubs. But trading remains largely localized; only two incumbents trade across Europe to provide arbitrage and therefore price convergence. In electricity, the level of concentration also remains high. The early years of liberalization saw little new build — whether by incumbents or new entrants - except for some gas-fired plants in Italy, Spain and the UK.2 Generators' ability to influence price comes from non-storability of electricity and highly inelastic demand,3 and their desire to do so stems from the fact that the price offered to all is usually that of the most expensive. So it can be profitable to withhold capacity if the 'loss' on the electricity withheld is exceeded by the increase in profit on the remaining electricity sold. Obviously there are other factors in the setting of electricity prices — fuel costs and the impact of emission permit trading — but the concentrated market power of a few big generators also has a bearing on prices, feeding distrust of the industry. So what are the barriers to entry that have evidently shielded incumbents from competition? Vertical foreclosure This is the process whereby incumbents wittingly or unwittingly foreclose, or close in advance, the availability of crucial inputs or assets to potential rivals. This can take several forms of locking up energy in long term upstream supply contracts, or locking transmission networks up in long-term capacity contracts, or locking customers up in long term downstream distribution contracts. 1 DG Competition sector inquiry, 2007, p. 37. 2 Ibid, p. 134. 3 Ibid, p. 132. 30 Energy and Climate Change Table 3: How incumbents control most of the gas in their home countries Total imports Incumbents Total domestic Incumbent (2004 in bcm) % share of production % share of domestic imports (2004 in bcm) production - (2004) (2004) Austria 9 80-90 2 — Belgium 16 90-100 0 - Czech Republic 9 90-100 <1 - Denmark 0 - 10 80-90 France 49 90-100 1 - Britain 13 20-30 105 40-50 Germany 88 90-100 18 80-90 Hungary 11 90-100 3 90-100 Italy 67 60-70 13 80-90 Netherlands 18 50-60 73 90-100 Poland 10 90-100 5 90-100 Slovakia 7 90-100 <1 - Source: European Commission sector inquiry SEC (2006) 1724, p. 240 The issue of long term contracts (LTGs) is one that Brussels and the gas companies have argued over long and hard. Europe's gas companies, and their upstream suppliers in Russia and Algeria, regard these contracts as crucial to the planning and funding of long-distance pipelines. And the Commission does not disagree. Indeed in its latest attempt to get the gas companies to accept ownership unbundling, the Commission said, 'the key to conclude long-term supply agreements with upstream gas producers is not the ownership of the network but the existence of a strong customer basis',4 presumably enshrined in a long term contract. But Brussels tries to insist that the LTCs do not lock up too high a percentage or for too long, though without so far really defining what that means. The tendency or temptation for companies to use any transmission networks they may own to 'disadvantage' their rivals is what the whole unbundling saga is about. As if to dramatize 4 Commission explanatory memorandum of Third legislative package, September 2007. Market Abuse: the things some companies do 31 the problem, the competition directorate launched its Eni and RWE anti-trust investigations in summer 2007. It announced an investigation into Eni's 'alleged capacity hoarding and strategic under-investment in the gas transmission system' in Italy, so shutting competitors out of some of the country's markets. A year later, in May 2008, the Commission announced a similar infringement inquiry into Gaz de France's 'behaviour that might prevent or reduce competition on downstream supply markets for natural gas in France through, in particular, a combination of long-term reservation of transport capacity and a network of import agreements, as well as through under-investment in import infrastructure capacity'. The investigation that the Commission opened against RWE in May 2007 appears to have had a swifter, and from the Commission's viewpoint rather satisfactory, denouement. The Commission said it was investigating RWE for possibly 'abusing its dominant position' in the North Rhine Westphalian regional gas market 'by raising rivals' costs and preventing new entrants from getting access to capacity on gas transport infrastructure in Germany'. It said it suspected RWE of charging high prices for access to its gas network, inflating its network operating costs, maintaining artificial fragmentation of the network, and failing to free up pipeline capacity so that customers could switch to rival suppliers. A year later, in May 2008, RWE said it was ready to sell its entire German gas grid of some 4,000 km, with the aim of getting Brussels anti-trust investigators off its back. The German company said its offer was not an acknowledgement of guilt. Eon had maintained its innocence in identical terms a couple of months earlier, in a similar deal with the Commission. In February 2008, Eon said it would sell off its high-voltage electricity grid in Germany in return for Brussels dropping an investigation into Eon for suspected manipulation of the wholesale and balancing markets, in which Eon's control of the grid may have played a part. Yet many energy operators and specialists regard the vertical integration model - inherited from the old national monopolies — as a legitimate business model, especially to offset other risks in liberalized markets. In electricity Malcolm Keay has noted that 32 Energy and Climate Change vertical integration — though not necessarily including ownership of networks - is considered a useful form of risk management. He has made a convincing case that 'the pressure to hedge risk by integrating downstream is reinforced by the special nature of electricity markets. Because it cannot be stored, an electricity sale, once missed, is lost forever. And because hedging is difficult, direct physical ownership of generating assets, combined with access to retail customers, is the easiest way of securing a match between supply and demand.' He notes that it was stand-alone 'merchant companies which suffered most in the UK and USA in the early 2000s, when many faced or suffered bankruptcy'.3 On a more mundane level, many ordinary energy customers may appreciate the organizational simplicity of vertical integration (even if that is not what they would call it). The argument has surfaced in an interesting way in France, whose government has been telling Brussels that all that was needed was rigorous enforcement of existing unbundling legislation (putting transmission networks into separate subsidiaries) rather than any new reform. As part of this pitch to enforce maximum separation within the existing EU law, the French regulator, the CRE (Commission de Regulation de l'Energie), has complained about the striking similarity of the marketing logos of EdF on the one hand, and of RTE (Reseau de Transport d'Electricite), EdF's 100-percent owned but separately managed transmission subsidiary. 'Look, the network is part of our brand, whether the CRE likes it or not', says an EdF executive. 'People know the network because if anything goes wrong it is the network people that come and fix it, and customers don't want someone strange turning up looking totally unrelated with EdF when they have signed their contract with EdF. Industry may know RTE, but householders know only EdF.'6 In gas, the vertically integrated incumbents defend their raison d'etre even more strongly, arguing that Europe needs strong players to negotiate with the world's Gazproms and to 5 Malcolm Keay, The Dynamics of Power, Oxford Institute for Energy Studies, Oxford, 2006, pp. 48-9. 6 Author interview, July 2007. Market Abuse: the things some companies do 33 make big long term financial bets on expensive and extensive pipelines. But the Commission has accepted none of these arguments. The Directorate General for Competition claimed in its report that its public consultation on its sector inquiry findings 'has not revealed any significant synergy effects linked to vertical integration',7 and that experience had shown that all parts of unbundled businesses continued to thrive after separation. Sometimes, downstream contracts can damage competition as much as upstream ones because incumbents can write them to effectively lock customers in and so exclude would-be new suppliers getting a foothold. The German cartel office took a stand on this in 2005 in the case of Eon-Ruhrgas' downstream contracts. The agency had earlier tried to block Eon's takeover of Ruhrgas (see next chapter), but had been overruled by the German government. But in 2005 it got some of its own back by preventing Eon-Ruhrgas from writing new contracts covering more than four years for more than 50 percent of a customer's annual demand, or covering for more than two years more than 80 percent of a customer's demand. In 2007 the Commission reached a similar deal with Distrigaz in Belgium by getting it to agree to make some 70 percent of its gas supply to industrial and wholesale customers 'contestable' by other suppliers. Shortly afterwards, the Commission opened an investigation of Suez's Electrabel division in Belgium and EdF in France for effectively locking up their customers by making it hard for them to switch suppliers. Commission officials indicated they were looking for a 'Distrigaz' type solution. Market segmentation This phenomenon exists, because incumbents rarely enter other national markets as competitors. Incumbents tend to sit on existing pipeline through-put capacity via long-term, pre-liberalization capacity agreements not subject to ordinary TPA rules. And when pipelines are expanded it is generally to serve incumbents, not new entrants. In electricity, the bottlenecks 7 DG Competition sector inquiry, p. 14. 34 Energy and Climate Change are less contractual and more the physical lack of sufficient interconnection between national markets. Historically, power companies only created links between each other so as to be able to help each other out in emergencies. As a goad to more action, EU leaders, at their Barcelona summit of 2002, said that each state should aim at having import or interconnector capacity amounting to 10 percent of its total generation capacity. But several countries — islands or peninsulas — do not have this: the UK, Ireland, Spain, Portugal and Italy. However, having more than a 10 percent ratio of connection that does not guarantee that countries are free from congestion or subject to competition from outside.8 Often national politicians and regulators are chiefly or entirely to blame for this segmentation. But sometimes companies do their own segmenting. Table 4: Some Weak Links in Europe's Power Chain - Average hourly net import capacity (NTC) relative to electricity generating capacity 2004 Country percent Country percent UK 2 Czech Republic 23 Italy 6 Austria 24 Spain 6 Belgium 25 Ireland 6 Sweden 29 Portugal 9 Hungary 38 Poland 10 Slovakia 39 Greece 12 Denmark 50 Finland 14 Estonia 66 France 14 Slovenia 68 Germany 16 Luxembourg 90 Netherlands 17 Source: European Commission sector inquiry SEC (2006) 1724, p. 175 This is what the Commission suspected when in July 2007 it announced an inquiry into possible collusion by Eon and Gaz de France. 'The possible infringement, which will be further investigated, takes the form of a suspected agreement and/or concerted practice between Eon and Gaz de France whereby they agreed not to sell gas in each other's home market. This 8 Ibid, p. 175 Market Abuse: the things some companies do 35 agreement and/or concerted practice may concern in particular supplies of natural gas transported over the MEGAL pipeline, which is jointly owned by Eon and Gaz de France and transports gas across Southern Germany between the German-Czech and German-Austrian borders on the one side and the French-German border on the other side.'9 A year later, in June 2008, the Commission sent both companies a 'statement of objections', taking its investigation of their possible market sharing behaviour to a more formal stage. However, the companies still insisted that the Commission, which had conducted dawn raids on their respective premises back in 2006, had just dug up an old and out of date story. The companies claimed that the gas transport agreements that Brussels was so fussed about went back to the construction of the MEGAL pipeline in 1975, and had been terminated in 2004. Whatever the upshot of this case, it is another instance of alleged market segmenting involving Germany This is not surprising. Segmenting is literally built in to the German system. For it has by far the most complex energy network in terms of ownership. In electricity, it has four transmission system operators (TSOs) with ultra high voltage, 40 TSOs with high voltage, and a remaining 800 networks with medium to low voltage. Its gas network is no less of a nightmare, though the two dominant suppliers, Eon and RWE, are making a big effort to simplify it. Germany has nineteen gas balancing zones (compared to one, two or maximum three in other states). This is awkward for new entrants: the smaller the balancing zone the greater the risk of imbalance in volume and pressure that a new supplier can cause and that the new supplier has to compensate for. The subtlest barrier, though one of the most effective, is the lack of transparent market information. New entrants find themselves at a particular disadvantage if they do not know as much, or as rapidly, as the incumbents about such vital data as network availability, infrastructure outages, or congestion in electricity interconnectors and gas transit pipelines. And in the normal course of events new entrants will not be as much 'in the know' as incumbents who, after all, control much of these 9 Commission press Memo/07/316, 30 July 2007. 36 Energy and Climate Change Market Abuse: the things some companies do 37 networks and whose operations weigh most on the market. If the reason for the prices swinging around is a big utility taking a nuclear plant off line or an incumbent gas company increasing its imports, it helps to be that utility or that gas incumbent knowing this in advance. Therefore, to neutralize this advantage, there need to be means of ensuring this information is disseminated to all, and at the same time. In terms of pursuing some of these market abuses, we have seen that the Commission has been ready to launch investigations and take, if necessary, court action. But preventing these abuses arising in the first place is not so easy. They stem from the concentrated nature of national energy markets, and that in turn is a legacy of the old national energy monopolies. What can Brussels do about this? Not much. In 2004, it did get the power to impose 'structural remedies', in plain language, to break up a company or group, 'where there is substantial risk of a lasting and repeated infringement that derives from the very structure of the undertaking'. In practice, the Commission is most unlikely to dare to use this power. Fining a company on proof of market abuse is something that Brussels has long done. To go on to break it up, in cold blood, so to speak, because there is a risk the company will keep repeating the abuse, is probably a step too far. So the Commission will continue to rely on its merger control authority for any shaping of the energy sector. But this is a passive power. Brussels cannot use it proactively. It has to wait for companies to want to merge before intervening. Liberalization has certainly stimulated the merger market. Companies when forced to compete often choose to join rivals they can't beat. During the first phase of liberalization — from 1998 (date of the first gas directive) to 2003 (date of the Second directives) - the Commission was notified of 135 mergers and acquisitions in the electricity and gas sectors, of which two-thirds were of national dimension and one-third were cross-border.10 In general, the Commission has taken a deeper interest in energy mergers than in other mergers. Of all mergers notified to Brussels for approval since 1994, the Commission has raised queries, conducted investigations, set conditions and occasionally issued vetoes in 18 percent of energy mergers, compared to a 12 percent average for mergers in all sectors.11 In summary, then, it is evident that anti-competitive practices persist in Europe's energy markets. There are special economic and political features, such as the legacy of state monopolies, which have contributed to these practices. But there are also special technical and commercial features, such as synergies in vertical integrated utilities, which can mitigate these practices. Just as the issues are not always clear cut, so the remedial instruments are not always ideal. The Commission has used anti-trust investigations to the full. And it has often used its control over cross-border mergers to impose concentration-reducing measures on dominant incumbent companies such as market caps, gas release programmes or electricity auctions. But sometimes, as we shall see in the next chapter, it is governments that stitch up national mergers that are beyond the Commission's reach. 10 Peter Cameron, Competition in Energy Markets, Oxford University Press, Oxford, 2007, p. 368. 11 Commission officials at Casteel-Claeys seminar, January 2007.