Opinion 2/15: Maybe it is time for the EU to conclude separate trade and investment agreements

By Szilárd Gáspár-Szilágyi

Opinion 2/15 is already causing quite a stir in legal academia. While some take an EU law perspective, others look at it from the perspective of investment law or public international law. In this short post I will not focus on purely legal issues. Instead, I will look at the Opinion’s effects on the EU’s investment policy and propose a change in the Commission’s approach to the negotiation of international economic agreements.

The Current Approach and Its Drawbacks

 The EU is not new to negotiating preferential trade agreements (PTAs). However, negotiating free trade agreements that include investment chapters (FTIAs), resembling bilateral investment treaties (BITs), is a new and challenging experience for the EU. In its relations with Canada, Vietnam, Singapore, the USA, and Japan the EU has chosen an ‘all-in’ approach, seeking the negotiation and conclusion of comprehensive and lengthy trade agreements which are bolstered with extensive provisions on sustainable development, labour standards, the protection of intellectual property rights, and investment protection.

Combining trade and investment under one roof is not a novel phenomenon. In a previous co‑authored paper, Maxim Usynin and I have analyzed over 150 PTAs world-wide and we found that ever since the dawn of NAFTA, countries and REIOs such as Japan, Canada, US, Australia, and ASEAN have included investment chapters in most of their PTAs, while others, such as India, China, CARICOM, Chile and the EU are increasingly doing so. The reasons for this approach can be plentiful: states might want to export their norms, stronger parties might want to impose pre-existing templates on weaker parties, states might want to replace existing international economic agreements, or it might be more cost effective to conclude one set of negotiations, covering a vast array of fields, instead of having a sector-based approach.

Choosing an ‘all-in’ approach is not in itself problematic, provided that the issues and fields discussed in such agreements are not contentious internationally or domestically. Problems do occur, however, when a contentious issue is included in one of these agreements. According to Putnam’ s seminal article, a country entering into international negotiations takes part in a two-level game because it needs to simultaneously satisfy the international partner(s) and domestic constituencies. If a certain issue covered by the negotiations, such as ISDS, is highly contentious domestically, then the negotiation process might become more time and resource consuming or might even end up in a deadlock. If ISDS is not a contentious issue then the domestic ‘win-set’ for both parties is large and mostly overlapping. Thus, it is highly likely that in such a case the contracting parties will agree on the inclusion of ISDS in the PTA and conclude the agreement in a timely and resource efficient fashion. For example, the China-New Zealand FTA (includes ISDS) was negotiated in only 3 years. However, if ISDS is a contentious domestic issue in State A, but not in State B, then the inclusion of ISDS in the agreement is less certain. The negotiation outcomes in this scenario will vary according to how the perception of the contentious issue in State A changes over time and according to whether State B conditions the existence of the agreement on the inclusion of the contentious issue. In this latter case State B might be willing to change its stance if certain concessions are given by the other party. In case the inclusion of ISDS might compromise the conclusion of the trade agreement or might lead to protracted and costly negotiations, states could choose to have a sequential approach to their economic relationship, as observed in late Chilean FTAs. The parties can include ‘anchor’ clauses on future talks or consultations on investment protection (e.g. Chile-Turkey FTA, Article 61).

In case of the EU one can talk about a multi-level game, in which the Commission – as the EU negotiator – needs to ensure that new international economic agreements satisfy the third-state contracting party, on the one hand, as well as the Member States, their constituencies, and the EU level institutions, on the other. Thus far the Commission’s attempt to satisfy all the different levels is not entirely successful. On the international level contracting parties such as Canada and Singapore are clearly frustrated. In the case of Canada, the 2014 version of the treaty text had to be revised so as to include the EU’s new Investment Court System (ICS) and the negotiations ended up taking seven years. Furthermore, the Belgian and possibly the Slovenian governments are about to ask for a CJEU Opinion on the compatibility of CETA’s ISDS mechanism with EU law. Singapore had to wait two years in order for Opinion 2/15 to be handed down and now it faces a renegotiation of the agreement so as to include the ICS. Furthermore, a Japanese official has recently declared that they would favour a classical type of ISDS mechanism in their FTIA with the EU, instead of the ICS. Even more problems will result from including the ICS in the negotiations with more powerful actors, such as China and the USA (currently on hold). Domestically things do not look brighter. Civil society and NGOs have been protesting against including ISDS in EU FTIAs, followed by groups of academics and regional parliaments.

All in all it is fair to conclude that the inclusion of ISDS and investment protection in EU trade agreements is causing enormous headaches for the Commission, it is tarnishing the EU’s image as a reliable treaty partner and it causes domestic discontent. In light of these, some authors have argued for a removal of ISDS from these agreements.

 Proposal: Split the FTIAs into Separate Trade and Investment Agreements

 The current hurdles can be alleviated if the Commission decides to change its ‘all-in’ policy for concluding trade and investment agreements, and instead follows a sector-based and sequential approach, concluding two separate agreements: one on trade and one on investment.

Opinion 2/15 also points into this direction. If one looks at the Annex to this post, which illustrates how the Advocate General and the CJEU decided on the issues of competence over the various chapters, one can see a clear split between the various areas relating to trade and those relating to investment. It is as if the CJEU is telling the Commission to split the agreements into two separate ones.


First, a far reaching trade agreement should be concluded by the EU alone. All the traditional trade related areas, such as market access for goods, trade remedies, various barriers to trade, customs and tariffs, fall under exclusive EU competence. Furthermore, services – including all five means of transport services -, public procurement, intellectual property, sustainable development, and competition are also covered by the EU’s external competences, as well as those parts of the EUSFTA’s chapters on dispute settlement between the parties and transparency that do not relate to areas of shared competence. Given that all the areas that such an agreement would cover fall under exclusive EU competence, the Commission would have a very solid argument against the political choice of its Member States to participate in the conclusion and ratification of such trade agreements.

Concluding a trade agreement without an investment chapter by the EU alone would have several advantages.  First, during negotiations the win-set of the EU and the other contracting party would be higher because the contentious ISDS would not be included. This would result in swifter negotiations, a higher chance for the agreement to be concluded, and enhance the EU’s reputation as a credible partner. Furthermore, it would also not lead to far-reaching domestic opposition and backlash. For example, in the case of CETA, Vietnam and EU-Singapore the inclusion of investment chapters has delayed negotiations by several years, has caused domestic discontent and tarnished the EU’s international image. Second, having only the EU conclude such a trade agreement would mean that the ratification process would be a lot faster, thus lowering the chances of a prolonged provisional application of the agreement. This would create more legal certainty, since provisional application often does not include the whole agreement and it might create further complications in case of potential disputes brought under the agreement’s dispute settlement provisions..

Second, a separate investment agreement can be concluded later, in the form of a mixed agreement. If we take away the afore-mentioned exclusive EU policy fields what we are left with is investment protection and ISDS, parts of which still fall under shared competences, and those parts of the objectives of the agreement and the final chapters that relate to areas of shared competence. The CJEU made it fairly evident that the Member States would also need to be parties to an agreement that includes such provisions. An ‘anchor clause’ could be included in the trade agreement that the parties shall also negotiate a separate BIT. If the investment agreement fails, one would still be left with the trade agreement. This is preferred to the current situation in which the Common Commercial Policy is almost blocked because of investment protection and ISDS.

Having a sequential and sectorial approach is not uncommon; it is often a result of domestic opposition to certain policy fields or other political concerns. For example, in recent years China and Taiwan decided to re-launch economic cooperation between each other. However, given half a century of hostile relations they decided to follow a sequential approach. They first concluded a PTA called the China-Taiwan Economic Cooperation Framework Agreement (2010) which was later followed by the Cross-Straits Bilateral Investment Protection and Promotion Agreement (2012). Furthermore, already in the EU’s investment policy the negotiation of an EU-China BIT, separate from an FTA, is a sign that such an approach could be followed.

 Possible Drawbacks of the New Approach

 One could argue that negotiating several issues at once is less time and resource consuming than negotiating such issues separately, and it can lead to ‘issue linkage’. However, ‘all-in’ type of negotiations work well when the issues discussed do not cause much backlash. This is unfortunately not the case with the inclusion of investment chapters and ISDS into EU FTIAs. Instead of time- and resource-friendly negotiations, we ended up with very lengthy ones. Therefore, this argument does not hold its place given the current circumstances.

There could also be a risk that the second agreement, the investment one, would not materialise. Still, given the EU’s negotiating power and clout it is fairly likely that a second agreement would materialise, without the risk of dragging the trade components with it in case there is no agreement. One can of course argue that the trade components of such far reaching PTAs can be provisionally applied. Still, this presupposes that an agreement can be reached in the first place. While CETA is currently being ratified, the agreement might in the end not come into force if the CJEU decides that CETA’s ISDS is not compatible with EU law. Furthermore, as mentioned, provisional application as opposed to entry into force raises a set of legal issues with regard to the parts of the agreement that can be provisionally applied and application of dispute settlement procedures.

Furthermore, one could also argue that a split approach would delay the replacement of existing Member State BITs even more. While it is true that four of the FTIAs (Singapore, CETA, Vietnam, TTIP) would replace 46 existing MS BITs, this will only happen if these agreements get ratified in the first place. Furthermore, under Regulation 1219/2012 Member States continue concluding new, post-Lisbon BITs. 43 such agreements exist thus far, 18 of which have entered into force with many more awaiting approval. If the EU is really committed to include the ICS in all the new investment agreements of the EU or its Member States, then the Commission could simply authorize these new Member State agreements on the condition that they include the ICS. An amendment of Article 9.1 of the Regulation might not be needed.

Annex  – Competences over EUSFTA (click image to enlarge)