The negotiations for an EU-China Comprehensive Agreement on Investment (hereafter: CAI) were completed, after seven years and 35 rounds, on December 30, 2020. The main text was published by the European Commission on January 22, and important annexes (which actually list sectors open for investment and the reservations made by each party) were published on March 12.

This is not the end of the process, which requires actual signature by China and the European Union, and ratification by the European Parliament. An investment protection chapter also remains to be agreed: according to the present text, “the two Parties will endeavor to complete such negotiations within two years of the signature of this agreement.” This is not a binding deadline. However, there are precedents of investment agreements entering provisionally into force before the full completion of the process.

This dry summary hardly captures the debates among Europeans that has followed the conclusion of the CAI in the final days of 2020. To a much lesser extent, there have been also different judgments on the agreement among China’s Europe experts and watchers.

The interrogations can be split in three categories. The first is what has actually been conceded by either side in terms of openings for investment, and under what terms. The goal of a level-playing field for Europe, and that of continued access to a European market that is already open to external investment are the main issues. The second concerns the guarantees for implementation of the various concessions made under CAI, including dispute resolution mechanisms, which differ between general cases and obligations undertaken under clauses for sustainable development. This second category intersects a third set of interrogations which concern both values and geopolitics: how binding are commitments made under one form or another by China towards labor and environmental standards, and particularly the issue of forced labor, which has dominated the public debate since the negotiations were concluded? Has the European Union secured on these issues a credible commitment? And, hanging over this interrogation, is it right to sign an investment agreement when tensions of a political nature are rising? Should there have been more coordination with the USA to secure better terms, whereas December 2020 was a transition period in Washington after a bitter presidential campaign? Analyses from China of CAI often underlined the differences between the EU and the USA, lending some credence to those who argued that China’s goal was in part political—to divide both parties. Both China and the European Union have underlined that a deadline for the negotiations had in fact been set in 2019 for the end of 2020, and that the date for the agreement reflected the will to meet this deadline.

For a time, the debate seemed to diminish in intensity. The Commission, which had initially declared CAI a victory for European “values and interests,” now presented CAI as just one aspect of a series of policies regarding China, with several defensive policies being in the pipeline: after investment screening and a reform of the terms for anti-dumping, a move towards an instrument for foreign subsidies, an anti-coercion instrument are also under consideration. A Global Human Rights Sanctions Regime, and its guidelines, had also been adopted by the Council in December 2020: the adoption of restrictive measures (sanctions) on individuals.

But the debate has rebounded. The EU’s adoption of targeted sanctions against four Chinese officials over their role in Xinjiang, and the wider counter-sanctions edicted by China, which among other target MEPs from the four most important group of the European Parliament, are throwing open more interrogations as to the ratification of CAI. Once again, the political debate and its context matter more than the actual content of CAI.

The CAI is based on three pillars with one important omission: market access (pre-establishment sectoral openings and conditions for foreign employees), level playing field (this includes disciplines notably on technology transfers, status of SOEs transparency for subsidies), sustainable development (corporate social responsibility, labor and environment norms). Missing so far is the investment protection chapter, which is supposed to be completed within one year after the signing of CAI.

The first pillar or market access component lists manufacturing sectors that are open to European investment—a positive list which includes sectors ranging for example from agriculture processing, wearing apparel, and paper. It is largely a confirmation of earlier openings by China, either under the aegis of WTO, or through the successive issuance of several “negative lists” for foreign investment: these have been published since 2014 for Special Investment Zones (SEZs) and turned into national negative lists, the last of which was published in January 2020. The new wins for European firms are much more limited: bioresearch (except stem cells and genetics), health clinics in 8 Chinese cities and Hainan, electric cars assembly, cloud services, and some telecom services. Even these concessions are often bounded by considerations of planning from China’s government. In the other direction, the main Chinese win is a right to invest in the production of renewable energy to a 5% percentage. But here too, temporary ceilings are set much lower and seem to reflect an expectation of reciprocity. On employment of foreign manpower, the concessions may be considered a win for China. Residency and work permits are granted for up to 3 years to senior managers and “specialists” working locally for a foreign investor. Although visa and work permit hassles are an important issue for European and other foreign investors in China, reciprocity given to Chinese employees—even if their contract has to conform to local rules—is notable.

But the agreement is also a reassurance from Europe that investment remains open. And that is clearly a win for China, which practices some economic decoupling under the guise of the “dual circulation” economy, but fears more restrictions by its partners. However, the reassurance does not include sectors covered by investment screening for security reasons, a reservation amply matched by China’s 2019 Foreign Investment Law, which defines in broad and general terms the sectors for which transactions can be blocked. More generally, both sides have reserved the right to create new legislation that would affect foreign investment, provided they are “in good faith” and not a cover for protective measures.

The last means, of course, that neither side has lost policy space, and that was a goal of the European Union, which does not create informal barriers beyond law. Overall, the gains for European investment market access appear minimal, but neither can the result be described as negative. They concentrate in large measure in services because investment in manufacturing and non-services remains far less covered than trade by WTO related agreements. The agreement is also limited to local entry and operation—cross-border investment and services are not part of the agreement.

The CAI’s second ambition concerns a level playing field in many areas. This comes under various headings. In the Commission’s own summary, “in addition to rules against the forced transfer of technologies, CAI will also be the first agreement to deliver on obligations for the behavior of state-owned enterprises, comprehensive rules for transparency on subsidies and commitments related to sustainable development.” Indeed, the EU has aimed in recent years at crafting ambitious “new generation” and WTO-plus agreements, often including similar approaches to the issue of state support and subsidies, to intellectual property rights and sustainable development, encompassing both environment and labor ethics, if not actual standards. The EU negotiators have worked from a common mold and applied this to China’s case. This is consistent with the claim to have brought China, for the first time, to sign its “most ambitious agreement ever concluded with a third country.”

Among improvements, three areas will likely stand out. One is the so-called binding and ratcheting clause. The first term prevents China from reversing concessions. The second obligates it to mutualize later concessions to another partner: this is in effect the Most Favored Nation (MFN) rule. It is worth noting that it does not prevent China (or the EU) from establishing new regulations, and it applies only to services, to the exclusion of manufacturing, and to that of norms (environment, labor) that fall under sustainable development. Therefore, this is the literal implementation of WTO rules—including with their very extensive limitations. There seems to be no improvement on the US-China Phase One trade deal regarding forced technology transfer.

Another improvement is that for national treatment, which under WTO rules is mandatory for goods but not for services. This covers the behavior of state enterprises, which are not allowed, except in their public role, to discriminate in their purchase or sales of services. There will likely be also an extension of the criteria for state ownership to companies where a state entity holds a significant minority stake, and to companies formed at the local government level. A wide definition of SOEs, together with the obligation to provide information on SOEs, would close a loophole. It is also important because of the hybrid nature of China’s economy, and the move towards even more control of private companies since 2017.

To the leveling of the playing field described above, there are important exceptions. Government services (outside the remiss of GATS in any case) and procurement, news organizations, social services, education, and health are not included. The European negotiators have not acceded to Chinese requests to exempt infrastructure investment in Europe from screening, as well as most of the energy sector.

Overall, the concessions gained by the European Union over terms of investment appear more significant in principle than those related to sectoral openings. They are not matched by a very large European offer. Investment screening and future legislation, including anti-subsidy competition rules applying to non-European investors, have not been given up. The mutual liberalization and opening is much larger in the case of the EU-Japan Free Trade Agreement (FTA). We shall see that these commitments, besides their value in principle, are worth only the process to enforce them, and in particular to lodge complaints and resolve disputes.

It is the terms under sustainable development which have been often criticized by the European media and civil society. They include provisions on labor, climate, and corporate social responsibility (CSR). There is a commitment not to lower norms and rules in order to attract investment or to prevent investments from happening. One needs to acknowledge that within CAI, China has made new international commitments, notably that “each Party shall make continued and sustained efforts on its own initiative to pursue ratification of the fundamental ILO conventions No. 29 and 105, if it has not already ratified them”: this includes the issue of forced labor. But the mention of “on its own initiative” and the absence of a deadline seem to weaken the commitment. The CAI has the merit to open the conversation officially.

The other area that has stirred controversy is that of dispute arbitration and enforcement. Critics have focused on the process that applies to issues coming under sustainable development. Here, the process for dispute resolution is much weaker than the provisions set in other areas. The agreement cites general declarations or agendas of the UN, ILO, or OECD, rather than specific rules or conventions. Even with partners such as Japan or Korea—full rule of law democracies—a commitment to “work towards,” “make efforts” to ratify any ILO conventions is not binding. And for a good reason: “new generation” trade and investment treaties involving labor, environment, and sustainable development for the most part are not subject to WTO enforcement, including any legal recourse. As of now, trade norms still do not mix with legal norms. At the most, the last are subject to a process of bilateral arbitration, usually without sanction unless precisely specified. In this case, there are hearings for arbitration, meant to be transparent when no confidential data are involved. But there is no appeal, no sanction, and no enforcement mechanism. Unavoidably, this and the Xinjiang issues—which in fact overlap—will be an object of debate at the European Parliament. In defense of CAI, the Commission argues that in the case of the EU-Korea Free Trade Agreement, the EU is indeed using a similar mechanism to argue that Korea has not fulfilled its commitments regarding labor. Viewed in a geopolitical rather than purely economic perspective, the high-level dialogue set up by CAI between the EU and China will have to obtain a resolution of this issue.

For investment issues not falling under sustainable development, a complex but standard mechanism applies, governing dispute resolution, arbitration, recourse to WTO, and remedies. It is focused around state-to-state dispute resolution, in a process that is very close to that of WTO. There are arbitration panels of experts (presided by a third party), deadlines for each phase, a complaint process. At the end of the road, if decisions are not followed with remedies by the target of the complaint, the complainant may retaliate by suspending other concessions under the CAI in proportion to the damage incurred. Importantly, this last phase does not include the issue of state subsidies, for which only transparency is mandated. The process—which is present in other treaties involving investment that the EU has signed—does not exclude a direct recourse to the WTO, although the two paths are mutually exclusive: one must choose at the beginning of the process.

To complement what is a hazardous route for arbitration over service investment disputes, and possibly to improve the resolution process for disputes related to sustainable development issues, the EU negotiators have secured a political road: working group meetings at the Commissioner for Trade and Minister of Commerce level twice a year, and a high-level dialogue involving once a year a Vice-president of the Commission and a Chinese vice-premier. Other European FTAs generally include only a political review mechanism at the level of commerce minister or trade commissioner.

For its part, although China has hailed the agreement as a breakthrough and a sign of Europe’s will to chart its own course towards China, other experts have either debated that the negotiation may have granted too much to Europe, and mentioned that this was a generous gesture at a time when Europe is hit by a pandemic and the accompanying economic difficulties. That estimate neglects that the whole point of negotiating an investment agreement was for European to redress a situation where rules and restrictions are much more strict in China than in the European Union: the starting points were that of a developing and statist economy on one side, a set of developed market economies on the other side.

But there may be a kernel of truth in this estimate by some Chinese experts. Whatever the criticism in Europe on the shortcomings of the deal, it remains easy for China to renounce its benefits and commitments. CAI is now in danger from both ends: it is hard to envisage a European Parliament ratifying the agreement while some of its members are sanctioned by China, and easy to imagine that China can sustain a test of will in today’s charged political atmosphere.