Fourth revision of the proposed amendment of the Gas Directive

Further muddying of the waters?

February 2019

Executive summary

1. The revised Romanian proposal of the Gas Directive amendment published 4 February 2019 addresses five issues:

a. The conflict of law with UNCLOS

b. Exclusion of upstream pipelines

c. Derogation of existing pipelines

d. Authority of Member States to negotiate with third countries

e. The applicability of Network Codes

2. The conflict of law with UNCLOS is resolved in theory by limiting applicability to territorial waters, but creates practical difficulties which prevent the amendment from having any real impact, other than creating uncertainty, bureaucracy and cost.

3. Clarification of the status of upstream pipelines as excluded is good, but in practice means that exporters from countries such as Norway are treated differently than exporters falling under the Directive (discrimination).

4. The reasons specified as acceptable examples for granting derogation are either too limited (recovery of investment made) or arbitrary (security of supply). It is also unclear what is meant by “objective reasons duly justified”.

5. The specifications for how and when to grant Member States authority to negotiate with and conclude agreement with third countries in effect gives more power to the European Commission.

6. The unenforceability of the Network Codes as a result of the definition of pipelines connecting with third countries as “interconnectors” remains unresolved.

7. All other concerns that we have raised in previous reports also remain unresolved – the amendment still creates market distortion and risks turning away otherwise competitive gas supplies from Europe to other markets.

8. The revised amendment still creates more problems than it solves, and for this reason the amendment should be abandoned. Any remaining issues regarding how to improve further the functioning of the internal gas market can instead be dealt with in the imminent
overhaul of the entire Gas Directive in 2020.


On February 4, 2019, the General Secretariat of the European Council, currently under Romanian presidency, recommended a fourth revision of the text1 for the proposed amendment of the Gas Directive, a proposal that has been under discussion since late 2017.

Passing the amendment has become an issue of urgency, for at least two reasons: (1) the Gas Directive will become subject to a major overhaul in 2020, at which time any remaining issues preventing the completion of the internal natural gas market as intended can be addressed, and (2) after the elections to the European Parliament in May 2019, the newly elected Commission could potentially take a different view than the current one to the proposed legislation.

Many Member States are still concerned about the potential effects of the amendment, which, as we have pointed out in previous reports2, has a number of weaknesses. These include, for example:

  • The shift of competence from Member States to the European Commission regarding any required agreements with third countries over existing and new infrastructure to and from Member States.
  • The conflict of law with the United Nations Convention on the Law of the Sea (UNCLOS) and WTO trading rules.
  • The requirement of compatible legislation for implementation of the Network Codes on both sides of interconnection points with pipelines from third countries.

The fourth revision seems to address these issues, but in reality, only makes things worse. In this article we will explain why.


In late 2017, the European Commission published a proposal for amending the Gas Directive3. The purpose, as stated, was to complete the Gas Directive and internal gas market by extending the provisions and principles of the third energy package – third-party access, unbundling, transparency and tariffication – to import pipelines from third countries. The amendment covers all pipelines from third countries, onshore as well as offshore. At the time, the accompanying working papers and Q&A fact sheet seemed to indicate that the effect on onshore pipelines would be limited4, but this has remained unclear. We have described, in one white paper5 and two complementary viewpoints6, why we believe this proposal to be unnecessary and ineffective in view of the stated objectives, and most likely harmful to the functioning of the internal gas market, security of supply and welfare of European consumers.

Member States too, it seems, have had some misgivings about the proposed amendment, which has delayed its speedy implementation. This is understandable, especially since the proposal has the power to affect relations between Member States and third countries; and Member States’ influence over such relations. The potential effects are far-reaching, complex and somewhat opaque.

In our previous assessments, we concluded that the amendment suffered from a number of weaknesses and undesired potential effects:

  • It potentially moves the negotiating mandate concerning cross-border infrastructure with third countries from Member States to the European Commission.
  • It is in conflict with the United Nations Convention on the Law of the Sea (UNCLOS7 ).
  • In It is potentially in conflict with WTO trading rules (as it may discriminate against certain types of infrastructure and sources of origin).
  • It does not deliver on the stated objectives of increasing competition, internal gas market functioning or security of supply.
  • The lack of specified terms for granting exemptions and derogations and potential differences in conditions risks creating uncertainty and market distortion.
  • The fact that LNG and pipeline gas are treated differently risks otherwise commercially competitive gas supplies being turned away from the European market, which could lead to higher prices.
  • It is not at all clear what powers the European Union has to enforce compliance with this legislation, since nonagreement with third countries could lead to a halt in or redirection of supplies, which could harm European gas consumers.
  • The unclear effects of having to renegotiate existing agreements with third-country governments, whose priorities may have shifted since the infrastructure was built, adds further complexity and the risk of ending up with less favorable terms for European gas consumers.

Proposed changes and practical impact

In February 2019, a revised text of the proposed amendment was published8 , presumably to address the concerns raised by Member States. Given that the revision intends to assuage the concerns raised, it addresses five key issues:
  1. The conflict of law with UNCLOS
  2. Exclusion of upstream pipelines
  3. Derogations for existing pipelines
  4. The authority of Member States to renegotiate or enter into new agreements with third countries
  5. Interconnector status and the applicability of the Network Codes

1. Conflict of law with UNCLOS

The amendment seeks to impose EU gas market regulations on pipelines from third countries, including those entering the European Union from across the sea. This includes pipelines from, for example, Algeria, Libya and Russia, and in future between post Brexit UK and the EU. However, the ability of the European Union, Member States and third countries to regulate pipelines outside of territorial waters in their Exclusive Economic Zones is limited by international law, for example, the UN Convention on the Law of the Sea (UNCLOS9 ). Hence, to avoid a conflict of laws, the revised text limits the application of the amendment to the "territorial sea." (See Recital 510 and the revised definition of interconnector, which says, "Up to the border of Union territory."11)

While this, at first glance, seems like a reasonable concession, it has some practical flaws. Interconnection points (metering stations, valves, etc.) with offshore pipelines are typically placed at or in the vicinity of a landfall, and not at the border of the 12 nautical mile territorial waters zone at the bottom of the sea. How will the point where legislation changes be determined, and flows and qualities be measured, on a seamless stretch of pipe? Will they be back-calculated from the landfall station? 8 See footnote 1 9 See footnote 7 10 Page 6 11 Page 8 Article 1 12 Page 5, Recital 4a Or will the interconnection point be agreed to be at landfall for practical purposes? If so, what practical difference does this amendment make, other than to potentially shift the costs of the 12 nautical miles of gas transport from one system to another? It is difficult to see how this could possibly enhance competition or improve cost efficiency or security of supply.

2. Exclusion of upstream pipelines

Very sensibly, the revised text now specifies that pipelines connected to processing terminals or production systems12 are classified as upstream pipelines and, as such, not subject to any new regulation under the Gas Directive, other than that currently applying. This clarification is laudable.

However, it also means that some export pipelines (e.g., those from Norway) from third countries will be treated differently to others, which seems discriminatory.

On the other hand, it could provide, at least in part, a solution for projects such as the EastMed pipeline, which we described in an earlier paper13. That pipeline will connect gas fields in the Levantine basin with markets in Cyprus, Greece, Italy and Bulgaria. It will thus consist of several sections, some of which may well fall under the amended Gas Directive. To what extent relief can be obtained for the entire pipeline remains to be clarified.

3. Derogations for existing pipelines

The revised text acknowledges the right of Member States to grant derogations from the amended Gas Directive14 for pipelines from third countries that were completed before the amendment came into force. This also applies to stretches of pipeline between the border of a Member State and the first interconnection point on its territory (which may be located some distance away from the border). It is clear from the text15 that Member States may only grant derogation with the approval to do so from the European Commission, since it is specified that "objective reasons duly justified" have to be present, although it is unclear how "objective reasons duly justified" are defined, and who decides what they might be. Such reasons are specified as (but presumably not limited to):

(1) Enabling recovery of investments made.

(2) Safeguarding security of supply.

Examining these two in detail gives rise to some provoking thoughts.

Recovery of investments made

Many pipelines entering the European Union were completed a long time ago, and investments should already (reasonably) have been recovered. This reason, thus, will hardly be relevant for consideration, other than in a few cases (e.g., Medgaz and Nord Stream).

Security of supply

To use this reason for granting derogation, a Member State would have to argue that without it, security of supply would be at risk. This is equivalent to saying that without derogation, deliveries through the pipeline in question would stop. While in theory, this would make it possible for an exporter to hold the Member State (or EU) at ransom to preserve the status quo, it is hard to see, in reality, that an exporting nation would not, rather, enter into a compromise agreement before breaking off deliveries. Security of supply is thus a reason that could or could not be deemed as a valid argument for granting derogation, depending on the current preferences of the European Commission. It could be applicable in all or no cases.

There is also an interesting conflict of objectives inherent in using security of supply as a reason for granting derogation from an amended Directive. It seems odd that security of supply has been used both as a justification for amending the Directive to capture import pipelines from third countries, and as a reason the amended Directive may not be applied.

Other concerns regarding derogations

The text concerning derogations goes on to say16 that these may only be granted if they would not significantly impede competition, the functioning of the gas market or the security of supply in the European Union. It is difficult to see how this could be the case for pipelines which have been operating for some time under current legislation, while all these parameters have steadily improved17. For example, for the Commission to deny a derogation because it would reduce competition, we would have to assume that the derogation would reduce the number of suppliers able to supply the European market. Since the purpose of derogation is to provide relief against the provisions of the Gas Directive for pipelines that have been able to compete prior to the amendment, this does not appear logical, unless the purpose is to increase the competitiveness of other, higher priced sources (which is equivalent to discrimination). The same type of reasoning can be applied to gas market functioning and security of supply.

Unless the European Commission significantly twists common logic regarding what factors could be expected to improve or, conversely, worsen competition and security of supply, it is hard to see the value of these limitations. For the same reasons, as we have said before, it is also hard to see why the amendment is needed in the first place.

Derogations from the Gas Directive for existing pipelines shall be limited in time for "up to 20 years based on objective justification, renewable if justified"18. Effectively, they could thus become valid indefinitely, in contrast with conditions for new pipelines, which will have to apply the Gas Directive from day one, unless they receive time-limited exemptions under Article 36 of the Gas Directive19. This creates an unacceptable two-class system.

4. The authority of Member States to renegotiate or enter into new agreements with third countries

The revised text of the proposal introduces a new Article 49aa20, which details the empowerment procedures envisaged for granting Member States the right to enter into negotiations and conclude agreements with third countries regarding issues that fall within the scope of the amended Gas Directive. It makes it clear that the European Commission has the power to grant such authority (and thus, that Member States are not able to take such initiatives unilaterally). It states that the Commission shall grant this authority, unless such negotiations would:

  • Be in conflict with EU law.
  • Be detrimental to competition, gas market functioning or security of supply.
  • Undermine the objectives of other negotiations between the European Union and the third country.
  • Be discriminatory.

Again, at first glance, this appears to make sense. However, if we consider the points in detail, concern arises.

Negotiations with third countries resulting from the implementation of the amended Gas Directive will concern derogations, renegotiation of pre-existing agreements, or new agreements to implement the provisions of the Directive. The very purpose of such negotiations will be to come to terms with the fact that EU law may not be compatible with legislation in the third country. If negotiations thus cannot be in conflict with EU law, it severely restricts negotiators' maneuvering room to find workable compromises that satisfy both parties.

For such negotiations to be detrimental to competition, gas market functioning and security of supply, we would have to assume their aim would be fewer suppliers (than would be available if the Gas Directive applied in full). This does not seem realistic, or even logical. Furthermore the criteria for determining a pipeline's impact on competition, gas market functioning and security of supply are not made clear.

The third bullet, concerning undermining other negotiations with third countries, is difficult to comment on since it is purely hypothetical. Presumably, it would not be in the interest of Member States to jeopardize common objectives. It may, however, have some relevance as a safeguard, since inclinations of Member States to act in their own self-interest cannot be ruled out.

On the other hand, since this condition is so vaguely formulated, not even limiting the type of intergovernmental negotiations concerned to the gas market or the energy sector, it opens up the risk of politicization. The European Commission could thus give unrelated issues preference over otherwise legitimate concerns regarding safe and secure operation of import pipelines from third countries.

For negotiations to be discriminatory, we would have to assume that they would aim to provide more favorable conditions to the third countries concerned than to others laboring under the regulations of the Gas Directive. Since that is the very point of granting, for example, a derogation, this provision seems oddly misplaced.

The Commission reserves the right to provide guidance and inclusion of specific clauses to ensure compatibility with EU legislation. This contradicts the purpose of such negotiations and introduces uncertainty, as it is too vaguely formulated. It also transfers more power from Member States to the European Commission.

Member States concerned about the potential impact of the Gas Directive amendment on existing infrastructure entering their territories and any related arrangements would be wise to question these provisions.

5. Interconnector status and the applicability of Network Codes

The new drafting attempts to answer concerns about existing technical agreements between transmission operators in Member States and third countries. Article 48a says that such agreements can be maintained or concluded "insofar as these agreements are compatible with Union law and relevant decisions of the national regulatory authorities of the Member States concerned". However the Network Codes are EU law and they cover exactly the type of issues included in technical agreements between operators, for example the Network Code on interoperability and data exchange. Thus there is still a problem if there is a conflict between such technical agreements and the Network Codes.

The problem arises from the fact that pipelines from third countries in the original amendment are redefined as interconnectors21. Previously, interconnectors were defined as pipelines between Member States, connecting one transmission system with others, and often operable in both directions. Interconnectors serve the purpose of facilitating gas flows between markets, not just exports from one country to another.

Under the proposed amendment to the Gas Directive, export pipelines to and from third countries would also be classified as "interconnectors". This brings about some interesting consequences. It means that certain mandatory Network Codes (e.g., for capacity allocation or tariffs) would have to be applied at interconnection points between EU pipelines and those from third countries, both on land and that enter from the sea. The definition of an interconnection point is "a physical or virtual point connecting adjacent entry-exit systems or connecting an entry-exit system with an interconnector"22. And, of course, the amendment now classifies an interconnector as "a transmission line between a Member State and a Third Country"23. This raises difficulties, especially for onshore pipelines, since it would require EU rules, such as those for capacity allocation, or interoperability and data exchange, to apply on both sides of an interconnection point, in order for the procedures of the relevant Network Code to work, and for EU Member States to be in compliance with EU law. Hence, EU rules would be applied to pipelines on foreign territory, subject to different, potentially incompatible legislation. This creates a conflict of laws, and would be unenforceable.

For this reason, presumably, onshore pipelines were at first assumed (in practice) not to be affected by the Gas Directive Amendment24. Formulations on this point have, however, not been entirely clear. As well as the new Article 48a, the revised text recites that the Network Codes (except the Balancing Code, which is purely internal) shall apply to entry points to and exit points from the European system, subject to the decision of the National Regulatory Authority25. This appears to be an attempt to assuage concerns that a conflict of laws would be created for onshore pipelines, or that existing technical agreements would need to be changed. But it ignores the definition of "interconnection point", which, in turn, is dependent on the definition of "interconnector".

Thus, there is a conflict between the revised amendment, which states that that existing technical agreements can be maintained, and that Network Codes only apply to entry and exit points to and from third countries (if the local regulator has so decided), and the implications of the amended definition of "interconnector". This, in turn, creates uncertainty not only for offshore pipelines, but for those crossing land borders as well.

How does the revision impact the conclusions in the original report?

Unfortunately, we cannot see that the revised text solves any of the concerns raised in our previous reports, other than removing the conflict of law with UNCLOS26 by limiting the applicability of the revised amendment to territorial waters.

At best, it limits the applicability of the Directive to the 12 nautical mile zone for offshore pipelines, and to the entry points to the European system for onshore pipelines. This creates a potential conflict of laws with third countries. The potential for infinite derogations ends up in no discernible difference for existing pipelines from conditions before the amendment, but potentially full applicability of the Gas Directive to any pipelines completed after the amendment comes into force. This, thus, still has the effect of potentially discriminating against, or distorting competition with, new infrastructure.

At worst, the European Commission could deny derogation/ exemption to all existing and new pipelines, and authority to conclude agreements with third countries to any Member State, thus potentially putting all current and future pipeline imports at risk. While this seems far-fetched, it illustrates the degrees of freedom that will be granted to the European Commission should this amendment be adopted in its present form. Any outcome between these two extremes is conceivable.

In our view, the proposed amendment in its current form should not be considered for adoption. It clearly creates more problems than it solves, and is too ambiguous and vague to be evaluated in full. Since it is also unnecessary from a marketfunctioning point of view, we question the wisdom of retaining it in discussion. It would be better to discuss any potential changes as part of the imminent 2020 review of the Gas Directive.



This report was written by:

Annette Berkhahn

Senior Advisor