Market Power and Long-term Gas Contracts: The Case of Gazprom in Central and Eastern European Gas Markets.

AuthorChyong, Chi Kong

    Healthy energy markets hinge on effective competition rules and, critically, on how those rules are implemented. From the early 2000s onwards, the Directorate-General for Competition (DG COMP) of the European Commission (EC) has emphasised competitiveness in European gas and electricity markets. Its antitrust investigations have spanned merger cases, like the 2006 GDF-Suez merger and the 2008 EdF-British Energy merger, as well as investigations into long-term contracts (LTCs) for gas supply with major exporting countries, namely Russia, Norway and Algeria (see Note 1 in online Supplementary Information (SI) for a detailed history of LTCs).

    In 2009, the EU adopted the Third Energy Package, which sought to implement new policies to open markets and accelerate investment such as ownership unbundling and independent system operators. Under this new regime, Lithuania filed a complaint against the Russian state-owned gas exporter, Gazprom, for using its monopoly position to charge high gas prices in the Lithuanian market through specific clauses in its LTCs. The EC subsequently carried out 'dawn raids' on Gazprom offices in ten Central and Eastern Europe (CEE) Member States (MS) in September 2011.

    After finding evidence of anticompetitive behaviour on the part of Gazprom, in September 2012, DG COMP announced it had opened proceedings into contracts between Gazprom and eight CEE MS: Bulgaria (BG), Czech Republic (CZ), Estonia (EE), Hungary (HU), Latvia (LV), Lithuania (LT), Poland (PL) and Slovakia (SK). In April 2015, the Commission adopted a 'statement of objections' (SO), describing its preliminary assessment of the case (EC, 2015). Gazprom was accused of:

    a. Imposing LTC restrictions on customers in the eight CEE MS, restricting exports and usage of gas to certain territories, and refusing to change delivery points for their gas imports;

    b. Pursuing unfair pricing policy in five MS - Bulgaria, Estonia, Latvia, Lithuania and Poland (henceforth MS5); and

    c. Obtaining unrelated commitments from its contractual counterparties concerning gas transport infrastructure (in Poland, the Yamal-Europe pipeline and in Bulgaria, the South Stream pipeline).

    In February 2017, Gazprom proposed commitments to address the EC's objections. In response, the EC subjected these proposed commitments to a market test, following which Gazprom revised its proposed commitments in March 2018. Although the Polish state oil and gas company PGNiG filed a complaint with the EC against Gazprom following the initial proposal, it was rejected on grounds that the commitments already addressed the allegations made in the complaint, which mirror the EC's own concerns. The revised commitments, effective until mid-2026, were made legally binding on Gazprom in an Article 9 'commitments decision' passed on 24 May 2018 (EC, 2018). PGNiG subsequently appealed to the European Court of Justice on 16 October 2018 (Euractiv, 2018). In February 2019, the appeal was joined by the Polish and Lithuanian governments (Polandin, 2019), as well as the Bulgarian gas company Overgas (PGNiG, 2019).

    We develop a bespoke modelling framework to offer new insights into European antitrust investigations in upstream gas markets over the past two decades and explore possible consequences. Our research contributes to the energy economics literature in several ways. Although there have been several studies on the state of EU gas markets focusing on Russian exports (Aune et al, 2017), supply disturbances (Bartelet and Mulder, 2020), projects of common interest (PCIs) for gas infrastructure (Kiss et al 2016), and optimal deployment of gas (Fodstad et al 2016), there have been no publicly-available studies analysing the future of gas markets in this new landscape.

    We examine the impacts of Gazprom's 2018 settlement of the case on future gas markets in Europe using a large-scale computational model. The model builds on Chyong and Hobbs (2014) but incorporates a number of significant extensions. The model now explicitly represents Former Soviet Union (FSU) and individual CEE states in a global gas market and captures the significant changes in European gas infrastructure including those anticipated for the 2020s. Further, the model adds an important feature of the gas market--swap transactions proposed by Gazprom as part of its commitments; swap transactions are related to well-established 'backhaul' transactions in the gas industry, which were explicitly modelled by Kiss et al. (2016). On the policy side, the analysis adds to existing ex ante modelling studies by investigating the impacts of implementing Gazprom's commitments on gas markets in CEE and North-Western Europe (1) (NWE) along three dimensions:

    i. Delivery points. Under what circumstances, if any, does the possibility of changing delivery points (or 'swap deals') for Russian gas within CEE markets improve the welfare and market efficiency of MS5?

    ii. Product market definition. How effective are swap deals in constraining the potential market power of Gazprom in MS5 vis-a-vis diversifying gas infrastructure?

    iii. Geographic market definition. What is the possible impact of swap deals on the wholesale prices of other markets and how geographically 'wide' would those impacts be? This is important, as any changes to the service charges for swap deals could affect market prices beyond those markets directly affected by swap deals.

    EU competition cases play a critical role in the evolution of European gas (and electricity) markets (SI Note 2) and yet there have been no detailed analyses of any major competition decision in the public domain. We explore one of the most important recent decisions, the 2012-18 Gazprom antitrust case, using a global gas market simulation model. The rest of this paper is organised as follows: [section]2 provides an overview of relevant gas market modelling studies; [section]3 describes the methodology used to assess and quantify the impacts of Gazprom's commitments on the European market; [section]4 discusses the simulation results; and [section]5 concludes.


    Energy market models have been used by firms and governments to study both ex post and ex ante impacts of energy policy and regulatory decisions, or to examine macroeconomic development pathways and competition analysis (Herbst et al. 2012; Davies 2010). Competition authorities use ex post statistical analysis to prioritise policy interventions based on past effectiveness and to help justify their decisions (Ilzkovitz and Dierx (2015) provide a detailed review). Chauve and God-fried (2007) present DG COMP's ex post electricity market simulation and withholding analysis (2) as part of the 2005 Sector Inquiry and find a positive correlation between market concentration and wholesale electricity price mark-ups. Further, Argentesi et al. (2017) provide the only independent quantitative ex post assessment of a DG COMP case, using a Difference-in-Differences method to assess actual market conditions and the counterfactual after the 2006 GdF-Suez merger. They conclude that the unbundling required in the settlement increased competition in the Belgian gas market and reduced prices.

    The ex ante quantitative analyses used by DG COMP to inform its antitrust and merger decisions are not publicly available. Budzinski and Ruhmer (2009) provide a review of the merger simulation models (MSM) employed in competition policy, noting that MSM is still a novel instrument for ex ante antitrust analysis and hence remains underexploited. The authors classify MSMs and summarise their usage in mergers cases, concluding that MSMs must be combined with traditional competition policy instruments. Further, Ilzkovitz and Dierx (2015) mention that a simple version of the Proportionally Calibrated Almost Ideal Demand System (PCAIDS (3)) model (Coloma 2006) has been used for pre-merger impact evaluations.

    In 2003, the Dutch competition authority commissioned ECN and Frontier Economics to develop two electricity market simulation models to assess ex ante the merger between two leading utilities Nuon and Reliant (Budzinski and Ruhmer, 2009). The competition authority's decision was subsequently challenged by Nuon and NERA (2005) was asked to reconstruct the ECN/Frontier Economics models. Other examples of energy MSMs employed for ex ante quantitative analysis to support competition policy are rare, as noted by Budzinski and Ruhmer (2009). Wolak (2011) uses a Cournot competition model to study potential post-merger effects in the Maryland electricity market using a withholding analysis and finds potential post-merger price increases.

    Budzinski and Ruhmer (2009) classify MSMs based on two key assumptions: the form of competition and the functional form of systems of demand used to develop MSMs. Bertrand models, Cournot and supply function models, and auction models are three forms of competition most often used to develop MSMs. Froeb and Werden (2008) noted there is a widespread consensus that Ber-trand models are appropriate for oligopolies in heterogenous product markets while Cournot models are more appropriate for oligopolies in homogenous product markets (e.g, commodities like crude oil, gas, electricity etc.) (Budzinski and Ruhmer, 2009).

    Indeed, academic gas market models represent oligopolistic market structure by Cournot model (e.g., Abada et al., 2013; Chyong and Hobbs, 2014; Egging et al., 2009; Gabriel et al., 2012, 2013; Growitsch et al., 2014; Holz et al., 2008; Holz et al., 2017; Huppmann and Egging, 2014; Lise and Hobbs, 2009; Siddiqui and Gabriel, 2017). These models have been extensively used for policy assessments and energy security analyses but large-scale, Cournot-based gas market simulation models are rarely applied to competition policy assessments. Only a handful of studies quantified the impacts of flexibility in global LNG markets using various methods (e.g., real options, econometrics and simulation...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT