The Devil Is in the Details. Five Points on the EU-China Comprehensive Agreement on Investment (CAI)

On the 30th of December 2020, the European Commission announced its agreement in principle on the EU-China Comprehensive Agreement on Investment (CAI), after nearly seven years of negotiations.
The CAI is a very ambiguous agreement, or rather, it is an agreement with some details that need to be clarified. For example, some days ago the European Commission published the annexes containing the parties’ market access commitments (i.e., the sectors in which the EU and China agreed to limit quantitative restrictions or joint venture requirements) and their lists of present and future non-conforming measures (e.g., on national treatment). These annexes are helpful to understand the agreement a little better, although many details require further clarification.
Commentators in the blogosphere have already analysed the text of the agreement, and in this post, I am not going to offer an in-depth analysis of the CAI. Instead, I will make five general comments.
My first point regards the nature of the agreement. The CAI is not a revolutionary text (with some exceptions, for example, in the field of technology transfer). For the most part, it seems to reiterate and consolidate some of the parties’ existing commitments under the World Trade Organization (WTO). It also overlaps with other strategies undertaken by China at an international level, such as the Belt and Road Initiative (BRI). Think, for instance, of the transport sector, which is recalled by the CAI and is crucial in the BRI. In this sense, the CAI should be read in conjunction with other initiatives undertaken by China since it is part of a strategy aimed at reshaping the international legal order. It is probable the agreement was announced to pursue what has been called the strategic autonomy of the EU (whatever this formula really means), and it can be argued that it is a response to the so-called “phase one” trade agreement conceived under the Trump presidency and signed  in January 2020.
My second point looks at Art. 14 of Section VI: “Nothing in this Agreement shall be construed as conferring rights or imposing obligations that may be directly invoked before the Parties’ courts or tribunals”.
This resembles the text of other EU agreements, for instance, the Comprehensive Economic and Trade Agreement (CETA) with Canada, which denies direct effect and limits private party enforcement before courts. Elaine Fahey criticised the CETA arguing that “this state of affairs creates a highly undesirable enforcement gap“. I agree with this view and this is a criticism somewhat independent of the ambiguity of Opinion 1/17, in which the Court of Justice of the EU said that the CETA (rectius, its Section F of Chapter Eight) is compatible with the EU Treaties.
The third point has to do with the elephant in the room, namely human rights. One could argue that an agreement like this could end up legitimising China’s human rights record, in spite of the EU’s commitment to respect the values of Art. 2 of the Treaty on European Union (TEU). These are strong value-based concerns regarding the CAI, but what can we say about that? There are obviously differences between China and the EU that cannot be reconciled, despite the “common” terms often used in international fora. Think, for instance, of the words used by H.E. Wang Yi, the Foreign Minister of the People’s Republic of China, at the High-level Segment of the 46th Session of the United Nations Human Rights Council last February, when he referred to a “human rights philosophy” that barely relates to the Western understanding of the concept of human rights as the recent exchange on sanctions on Xinjiang abuses confirms. Similar differences can be found with concepts such as the rule of law, which also entered the text of the Chinese constitution (Art. 5). Clearly, the Chinese concept of the rule of law is different from the European one, which is, in turn, distinguishable according to the European constitutional traditions taken into account (the German, British, French or Italian ones, for instance). I use here the linguistic concept of “false friends” to refer to similar words which actually mean completely different things. As a counter-argument, it was said that through this agreement, the EU could induce China to guarantee higher protection standards by “adding pressure on China to comply with International Labour Organisation (ILO) Conventions.” In this respect, it is worth recalling the European Commission’s recent activism, which, before a panel of experts established under the EU-Korea FTA, successfully challenged certain Korean measures restricting workers’ rights to join a union. Should we aspect more of this activism under the CAI? However, while the EU thinks it can convince China to converge on its own values, China thinks the same, and this could lead to a never-ending tug-of-war.
My fourth point relates to the international strategy pursued by China. It has been argued that the CAI “showcases its increasing desire to shape multilateralism, rather than merely be shaped by it.” This agreement is part of a trend in which China would be moving from a paradigm of “selective engagement” (or “selective adaptation“) to a paradigm of “selective reshaping.” What is the difference between these two concepts? Heng Wang writes that “Selective adaptation is concerned with the ‘downloading’ of external norms, in the form of ‘rules, structures, processes, and practices.’ Selective reshaping is the ‘uploading’ of China-led institutions and China-preferred rules at the extra-regional level.”
Henry Gao, in this sense, writes of a shift from a situation in which China was a rule-taker to a context in which China can be described as a rule-maker, or even perhaps a rule shaker. In so doing, China would also contribute to questioning the existing international economic legal order, which is frequently depicted as characterised by American dominance and liberal internationalism.
But how can China develop this strategy? This leads to my fifth point.
The CAI does not exclude the possibility of Memoranda of Understanding (MoU). Indeed, there are many MoU in this area. For instance, the Switzerland – China free trade agreement refers to an MoU that states cooperative activities may take place through dialogue, joint studies, and capacity building. MoU are soft law instruments, and they are frequently employed by China since they guarantee flexibility, a key ingredient of its approach to the international community; think of its use within the BRI. There are also existing MoU between the EU and China, for instance, the MoU on “establishing a Connectivity Platform between the EU and China”. In theory, since many details in the CAI need to be clarified, this agreement could cover the contents of these MoU, but something will remain out of that and this confirms the margin for flexibility. After all, soft law concerns “measures which are not legally binding but which nevertheless have practical and even legal effects” as Francis Snyder puts it.
As scholars have argued, soft law is a very general term that has been used to refer to a variety of processes. Frequently, soft law is identified as opposed to hard law since these forms of law would respond to antithetical dynamics: “rigid versus flexible approach to implementation”; “presence versus absence of sanctions”; “material versus procedural regulation”.
However, soft law and hard law should be seen as complementary (Shaffer, Pollack 2010). Among the advantages of soft law, one can mention “lower contracting costs”, “lower sovereignty costs”, “simplicity and speed”, and “participation and incrementalism”. In international law, soft law is frequently employed since ‘soft legalisation is much easier to achieve. A recent proliferation of soft law instruments can be found in the creation, enforcement, oversight, and regulation of: counterterrorism measures, financial regulation, and international taxation law.
Of course, concerns over transparency and ambiguity remain, but my point is that soft law is not bad or weak per se. Indeed, soft law is, above all, a multi-functional device or, better, it refers to a jungle of measures and acts that may have legal relevance depending on the context in which they are employed. In a strong integrated system like that of the EU, soft law has acquired an interesting role even before courts. On the contrary, in the Canadian case, soft law has been employed to galvanise political mediation and consensus, avoiding the involvement of judges.
Comparative law shows that soft law can have an added value in favouring the coexistence of different cultures and thus pluralism. In the Chinese approach to international law, soft law is frequently coupled with flexibility, and I also think that in light of the CAI’s ambiguity, it can continue to play a prominent role.