A Retrospective Evaluation of the GDF/Suez Merger: Effects on the Belgian Gas Hub.

• Author: Argentesi, Elena
• Position: Gaz de France

1. INTRODUCTION

   Competition policy enforcement activity in the EU energy sectors has increased significantly over the last twenty years, especially with regards to merger cases. Between 2000 and 2014, the European Commission (EC) has handled over 270 merger cases in the energy sectors, as opposed to just 20 in the five years prior to the year 2000 (European Commission, 2015). (1) In some of these decisions, the EC required merging parties to offer substantive remedies to, in principle, offset the anti-competitive effects of the merger (see e.g. Polemis, 2018).

   In reality, though, some of these remedies may have gone further than that. As argued by many commentators, they may have been used to promote market liberalisation and, in particular, to achieve effective unbundling of network and supply activities. The EC's view is that the EU liberalization directives in energy have not been wholly effective (Federico, 2011). (2) Member States have forced a number of compromises to defend their national energy champions, which limited or delayed the EU directives' effectiveness. Ensuring fair and transparent third-party access to transport facilities, and non-discrimination against non-vertically integrated firms, has been especially difficult (Harrison and Mordaunt, 2012).

   In response, the EC is said to have tried to "overcome significant obstacles to Europeanization endemic to the energy sector" (Eberlein, 2012) by using "windows of opportunities" created by large European cross border merger proposals (Pakalkaite, 2014). Others argue that, indeed, the EC is shaping the overall development of the EU energy markets through its merger policy decisions, and that the energy sector has been taken "out of the domain of (national) sector-specific regulation and put under the auspices of (EU) competition policy" (Hellwig, 2009).

   This paper provides a quantitative ex-post analysis of one of the most important mergers in the EU energy sector in recent decades, the Gaz de France's (GDF) acquisition of Suez in 2006. We analyse whether the merger and the associated remedies imposed by the EC restored, or indeed improved, initial market outcomes. We focus on the impact on the market for trading in Belgium's Zeebrugge (ZEE) hub. (3) The ZEE hub is a key part of the European gas market and an important target of the remedies imposed by the EC. Indeed, prior to the merger, the ZEE hub had suffered from limited infrastructure access and liquidity issues. Part of the remedies, which included ownership unbundling, aimed to free up access to the hub. If effective, the remedies should have allowed for a higher liquidity through more entry, higher traded volumes, and lower prices in the hub. This is not only important for the stakeholders of the wholesale hub gas market but also for end-users and consumers, as wholesale prices affect retail prices. (4)

   A series of Difference-in-Difference (DiD) analyses, which compare the evolution of prices at the ZEE hub relative to different counterfactual hubs before and after the merger, suggests that the remedies did more than successfully limit the merger's potential anti-competitive effects. Indeed, we find that not only did prices not increase, but they declined. This decline in prices supports the view that ownership unbundling improved access to the hub. We also provide evidence of the improvement of other outcomes at the ZEE hub in the post-merger period. In particular, we find evidence of increased entry, trading volumes and investment. Therefore, the remedies may have gone further than mitigating the potential anti-competitive effects of the merger. Given the absence of indicated efficiency gains by the merging parties, our analysis supports the view that the EC's merger policy actions may have been used as tools to improve market outcomes in the EU energy markets.

   Our paper provides, to the best of our knowledge, the first quantitative analysis of the ex-post effects of a merger and its associated remedies in the energy sectors. (5) Despite the importance of merger policy in energy markets, we are not aware of any retrospective study that provides an ex-post evaluation of the EC's merger decisions. (6) This may be because of complex market structures, or due to the specific technical features of energy markets. Notwithstanding these difficulties, retrospective studies are essential tools to assess past competition policy decisions and improve decision-making for future cases.

   There are some studies providing analyses of competition policy enforcement decisions in energy markets. However, most of the existing studies provide a qualitative, rather than a quantitative, analysis. Leveque (2006) argues that competition policy in the energy sector raises specific problems which require tailored solutions and call for a tougher approach to mergers. (7) Newbery (2007) compares the approach taken by the EU and the US regarding merger analysis in the energy sectors. He claims that, in contrast to the US, mergers between energy companies in Europe have been traditionally subject to rather relaxed standards (up until 2005). Pozzi (2004) shows that antitrust enforcement activity in a given year in the US electricity sector has a negative effect on industry profits in subsequent years.

   There are also few studies that analyse the performance of wholesale gas markets. Most of the existing analyses focus on assessing the degree of integration between hubs, which is, in itself, an important goal of the EU energy strategy. Heather (2012) and Petrovich (2013), for example, examine the integration of gas hubs using price correlations. Ruperez Micola and Bunn (2007) test for the existence of market power on the arbitrages performed in the "Interconnector" pipeline linking the gas hubs in Belgium and in the UK. Massol and Banal-Estanol (2018) develop a methodology to examine the overall efficiency of the arbitrages performed between gas hubs and use the same Interconnector pipeline as a case study.

   The structure of this paper is as follows. In section 2, we review the EC's GDF and Suez merger decision within the context of the ZEE hub. We then present the data and the empirical analysis of the impact on the market for trading on the ZEE hub in section 3. Section 4 discusses our main results on the effect of the merger on prices at the ZEE hub. In section 5 we discuss limitations and several robustness checks. In section 6 we present additional evidence on the evolution of non-price variables such as traded volumes, entry and investment. Section 7 concludes and discusses policy implications following on from the empirical analysis.

2. THE EC'S MERGER DECISION WITHIN THE CONTEXT OF THE ZEE HUB

   This section presents the main features of the GDF/Suez merger decision within the context of the Belgian wholesale gas market and, in particular, of the ZEE hub (for further institutional details, see European Commission, 2006). We also identify the dates of the key merger events, which are crucial for our empirical approach.

   The Belgian gas network is an integrated network used for domestic transmission, as well as for international transit. The ZEE hub was, and remains, crucial for this transit role; the pipelines that go through the ZEE hub are being used almost exclusively for transit (Heather, 2012). (8) The role of the hub became more prominent leading up to 2006 with the gradual liberalization of gas markets. As explained in detail by Miriello and Polo (2015), the ZEE hub, as well as the other European hubs, were initially developed to cope with local network balancing needs. However, they turned out to become a source of gas procurement as the previously monopolized industry structure gradually became more fragmented.

   At that time, GDF and Suez groups were the two most important players in the Belgian gas sector. Suez, the largest company in the market, had a majority stake in Distrigas (which was dominant in gas wholesale and supply) and in Fluxys (which controlled gas infrastructure, transit, storage and transport). (9,10) Through these companies, Suez also controlled GIE Finpipe (which owned the rTR/vTn pipelines on which the ZEE hub is physically located), Distrigas & Co (which managed the capacity rights on these rTR/vTn pipelines) and Huberator (which operated the ZEE hub). Thus, while the first two gas directives of the European Parliament (98/30/EC and 2003/55/EC) had led to "legal unbundling", in terms of ownership, Suez still controlled the access to the ZEE hub. GDF, on the other hand, was the second largest competitor in Belgian gas wholesale and supply markets.

   In May 2006, the European Commission received prior notification of a concentration between the GDF group and the Suez group via an exchange of shares. The Commission's investigation found that, given the horizontal and vertical overlaps between the two companies' activities, the proposed transaction raised significant competition concerns across all levels of the Belgian gas market. Of particular concern was the "market for trading at the ZEE hub", which was considered a relevant product market on its own, separated from the various gas-supply markets in Belgium (European Commission, 2006).

   The Commission was concerned that the merger might further impede access to the ZEE hub, which was already suffering access and liquidity issues. Actual or potential competitors of Distrigas wishing to access the hub had to obtain capacity rights through an entry/exit agreement with Distrigas & Co, which was the capacity rights seller of the transit network. According to the Commission, these negotiations, which took place on a bilateral non-transparent basis, posed a problem in terms of access. For example, there were issues regarding access to information, because Distrigas could obtain details of the positions of competitors. This may have undermined confidence and discouraged market entry.

   In response to the EC's concerns, GDF and Suez...