EU–China Comprehensive Agreement on Investment: An Appraisal of its Sustainable Development Section

I. A Distinctive Treaty in a Fast-Evolving Context

In the waning days of the indelible 2020, China and the European Union (EU) clinched an ‘in-principle’ deal on their Comprehensive Agreement on Investment (CAI). ¹ After seven years of negotiation, the treaty attracted widespread public scrutiny, partly due to the significant investment flows between China and the EU, although wider considerations are also at play. China’s 2001 accession to the World Trade Organization laid a major building block in the global order. Its economic rise and export growth have transformed China into the world’s factory, providing cheaper electronics, appliances and clothes for global consumers, but also prompting concerns about the environment and implications for jobs and working conditions in deindustrializing countries. This economic reconfiguration has created complex, often tense, relations between China and established powers such as the EU and the United States (US), to which multilateral arrangements have offered only partial responses. ²

While China has developed its own model of economic regulation, based on bundles of contracts, treaties and commercial arbitration, ³ negotiations have produced treaties such as the ‘phase one’ trade deal with the US, ⁴ and the CAI and an agreement on geographical indications with the EU. ⁵ Each of these agreements tackles specific trade or investment issues. At root, however, the deals grapple with coordinating fundamentally different development models, in a context of strategic rivalries, technological confrontation and global competition over regulatory approaches, and with close links between economic diplomacy and geopolitical considerations. As a major China–EU treaty, the CAI provides insights on how major powers are (re)negotiating their relations in this shifting international political economy.

From a legal standpoint, the CAI’s unique configuration attracted investment lawyers’ attention. The treaty deals exclusively with investment, rather than trade, although the two are related, as many foreign investments in China are oriented towards production for export. ⁶ However, while most investment treaties focus on investment protection and investor–state dispute settlement (ISDS), the CAI primarily covers issues such as market access terms, ⁷ prohibition of forced technology transfers, ⁸ non-discrimination by state-owned enterprises and regulatory bodies, ⁹ and transparency in economic governance, including subsidies. ¹⁰

These themes reflect key EU demands in the negotiations, informed by problems reported by EU businesses in China. ¹¹ Although the EU was already relatively open to foreign investment, China might secure a less fragmented response on its investments. ¹² Existing investment protection treaties between individual EU member states and China remain in force, ¹³ and the CAI requires the parties to seek to conclude a separate agreement on investment protection within two years. ¹⁴ Still, the CAI’s distinctive structure offers a glimpse of what investment treaties could look like if they responded to the problems businesses actually face, rather than assumptions about how legal protections would promote investment. ¹⁵

The CAI’s ‘sustainable development’ section has attracted particularly extensive scrutiny. Negotiators unveiled the agreement amid growing, disturbing reports of forced labour and the repression of Uyghurs in China’s Xinjiang Region, and heightened concerns about shrinking political space in Hong Kong. ¹⁶ A few months after the CAI’s release, tensions over these issues escalated in tit-for-tat sanctions and boycotts, souring relations between China and Europe and creating dilemmas for companies operating in both markets. ¹⁷

Also, public discourses in Europe have long considered that differentials in labour and environmental standards give China a competitive advantage, driving both exports and foreign investment, and some research has connected China’s rise with job market transformations and increased inequality in Europe. ¹⁸ While China has more recently made commitments on climate and the environment, ¹⁹ and met worker protests with both repression and improvements in labour conditions, ²⁰ concerns about labour rights and the environment are a recurring issue in China–EU relations.

The European Parliament has highlighted these concerns since the start of CAI negotiations, ²¹ calling for the treaty to include robust provisions, including on forced labour. ²² Parliamentary approval is essential for the EU to ratify the treaty, so assessing the effectiveness of the CAI’s labour and environmental clauses speaks not just to concerns about sustainable development but also to the prospect of the CAI coming into effect. Such assessment hinges on reviewing the sustainable development section of the CAI (section II of this piece), and outlining a preliminary appraisal of their import (section III).

II. The CAI’s Sustainable Development Section: A Look at the Text

Sustainable development is a broad, contested concept, and the internationally agreed Sustainable Development Goals cover wide-ranging issues such as ending hunger and poverty, reducing inequality, and ensuring sustainable consumption. ²³ While the CAI innovates in setting standards of treatment for foreign investment, its sustainable development section largely follows the EU’s staple approach to negotiating trade agreements. ²⁴ In its preamble, the CAI refers to human rights, sustainable development, climate change, responsible business conduct, and ‘high levels of environmental and labour rights’ protection’. Two aspirational, scene-setting provisions provide a broad framing for linking investment to the ‘objective’ of sustainable development and for calling on parties to encourage businesses’ ‘voluntary uptake’ of responsible practices, such as those inspired by the Guiding Principles on Business and Human Rights. ²⁵

Like other recent EU trade agreements, however, the CAI is primarily concerned with two narrower issues: the environment and labour. The treaty reaffirms the parties’ ‘right to regulate’, ²⁶ requires parties to ‘strive’ to ensure laws provide for ‘high levels’ of environmental and labour protection, ²⁷ and recognizes that it is ‘inappropriate’ to promote investment by reducing environmental or labour rights protection. ²⁸ These latter two clauses contain statements of principle, but their wording makes them hard to enforce. Other CAI provisions require the parties not to waive or derogate from their own environmental or labour laws, or fail to effectively enforce these laws, as a way to encourage investment. ²⁹ These are more enforceable terms, but their effectiveness depends on national law: if this sets a low bar, investments could adhere to the law while still causing harm. ³⁰ The ‘right to regulate’ clauses make it clear that each party has the right ‘to establish its own levels of domestic labour and environmental protection’, ³¹ although parties affirm their aspiration (strive) to ensure ‘high levels’ of protection.

The CAI also mandates parties to comply with environmental and labour treaties they have ratified, including climate agreements and International Labour Organization (ILO) conventions, and to honour their commitments under the ILO Declaration on Fundamental Principles and Rights at Work. ³² These requirements reaffirm existing obligations but are significant because – unlike some of the underlying instruments – the CAI allows parties to bring compliance matters to international dispute settlement. The CAI also requires parties to ‘work towards’ ratifying all fundamental ILO conventions, and to ‘make continued and sustained efforts’ to ratify ILO Conventions No. 29 and 105 on forced labour. ³³ Worded in reciprocal language, these clauses reflect concerns that China has not ratified four of the eight fundamental ILO Conventions – those on forced labour, freedom of association and collective bargaining. ³⁴

In contrast to the indicative two-year timeline the CAI sets for investment protection negotiations, these ratification commitments establish ‘best-effort’ obligations without defining deadlines, resembling clauses in the EU–Korea and EU–Vietnam trade agreements. ³⁵ In 2018, under pressure from trade unions, ³⁶ the EU initiated a dispute settlement proceeding against Korea, partly claiming breach of ratification commitments. As the CAI was released, the international panel of experts issued its report on this dispute, accepting that the EU–Korea treaty establishes binding obligations to make efforts towards ratification and that international panels can review compliance, but dismissing this part of the EU’s complaint due to lack of specific timelines for ratification. ³⁷ This decision highlights the limited effectiveness of such ratification commitments. Vietnam ratified two fundamental ILO conventions after concluding its trade agreement with the EU, ³⁸ but only after the latter refused to ratify the trade agreement until Vietnam had ratified the labour conventions. ³⁹

The CAI’s ‘continued and sustained efforts’ clause only applies to the forced labour conventions; it does not apply to those on freedom of association and collective bargaining, although the more general ‘work towards ratification’ clause applies to all the fundamental conventions. This is a departure from the EU–Korea and EU–Vietnam trade agreements, which extend the ‘continued and sustained efforts’ commitment to all the fundamental conventions. Freedom of association and the right to collective bargaining are key pillars of international labour law, and feature in the ILO Declaration on Fundamental Principles and Rights at Work. Recognizing the great diversity of situations and developing countries’ resistance to what they see as rich countries’ ‘labour protectionism’, international labour law refrains from setting uniform conditions, such as wages and social protection, focusing instead on ground rules, and on rights and principles – including the rights to form or join a union and to negotiate collectively – that could help workers obtain more equitable, context-specific terms.

Regardless of whether they have ratified the relevant conventions, all ILO member states must comply with the Declaration on Fundamental Principles and Rights at Work, which the CAI requires parties to adhere to. The EU–Korea panel found that, despite not having ratified the relevant ILO conventions, Korea’s restrictions on freedom of association were inconsistent with fundamental principles the trade agreement mandated parties to respect, promote and realize. ⁴⁰ However, the compliance monitoring arrangements envisaged in labour conventions only apply if those conventions are ratified. Even if China did so, questions would remain about the difference this would make in practice, without independent courts, media or social movements to scrutinize compliance. ⁴¹ Research suggests that ratifying labour conventions could even have a negative effect on respect for labour rights, by virtue signalling and therefore reducing public pressure for domestic reform. ⁴²

The CAI establishes a Working Group on Sustainable Development to facilitate and monitor implementation, ⁴³ and dispute settlement arrangements that allow parties to bring compliance issues to an independent body. However, consistently with EU treaty practice, sustainable development issues have a tailored dispute settlement procedure that involves consultations between parties, with referral to a panel of experts in the absence of an amicable solution – unlike other disputes, which are settled through state-to-state arbitration. ⁴⁴ Independent third parties can make written (amicus curiae) submissions to the panels; ⁴⁵ indeed, the EU–Korea expert panel cited such submissions from trade unions. ⁴⁶ While state-to-state arbitration leads to a binding decision, with sanctions for non-compliance, expert panels issue non-binding reports and recommendations with no sanction mechanisms.

III. Preliminary Appraisal

Overall, the CAI reflects compromises between competing considerations. Some compromises – such as balancing each party’s right to define its own labour and environmental policies with the commitment to strive for high levels of protection – align with drafting approaches found in other EU treaties. Others – such as the reduced emphasis on trade union rights – seem tailored to the CAI. Commitments on labour rights can challenge structural features of China’s economic and political system. Observers have suggested, for example, that historical experience in other countries, where independent trade unions undermined one-party rule, may have made the Chinese authorities suspicious of independent labour activism. ⁴⁷ These political considerations might explain features of the CAI, and it seems unlikely that the CAI’s generic commitment to ‘work towards’ ratification can overcome such foundational concerns.

That China agreed to adopt an adjusted version of the EU’s sustainable development model is significant. However, if it wanted to ratify the ILO’s fundamental conventions, it could presumably do so, and incremental formulations such as ‘continued and sustained efforts’ towards ratification seem to interrogate political will more than technical obstacles. Also significant is that compliance is subject to international review, creating space to scrutinize environmental issues and labour rights abuses, including against international commitments. Although a finding of non-compliance could be politically embarrassing, the tailored dispute settlement arrangement lacks legal teeth, and one wonders whether China accepted these commitments because it considered them substantively and procedurally unenforceable.

In Europe, public debates around the CAI have emphasized tensions between ethical principles and economic realities. ⁴⁸ Pragmatism seems to underpin the CAI’s approach to easing those tensions. Treaties are the product of negotiation, and it is possible that European negotiators ‘could never have hoped for more’. ⁴⁹ Beyond these transactional dimensions, however, there are questions about the EU’s own policy approach, including its preference for non-binding dispute settlement in sustainable development matters, and about the wider intersection of economic treaties, human rights and sustainable development. Certain features of the CAI – such as its setting timelines for investment protection negotiations but not for ratifying fundamental labour conventions, and the varying effectiveness of its dispute settlement arrangements – reproduce, and potentially compound, imbalances between social, environmental and economic considerations in the legal architecture of the global economy. Except for an aspirational provision, the treaty is silent on businesses’ responsibility to address human rights and sustainable development issues in their operations and supply chains.

Debates around the CAI hinge partly on diverging views about the effectiveness of law in addressing deep-seated problems. It is perhaps unrealistic to expect a treaty – no matter how tightly formulated – to deliver such far-reaching changes; ⁵⁰ although it should be possible to scrutinize that treaty’s contribution within wider strategies, both domestic and international, including other China–EU arrangements and multilateral processes. The CAI projects remarkable confidence in the power of international law, with public announcements staking much on the effects of ratifying an investment agreement and, in turn, labour conventions. ⁵¹ For sceptics, this approach masks choices favouring commercial over social and environmental interests, and risks compounding problems rather than addressing them. In certain respects, the CAI displays little trust in legal processes, effectively leaving disputes over labour and the environment to analysis-backed conciliation, and illustrating how the ‘power of law’ can be deployed unevenly for different policy objectives.

Questions of effectiveness are an empirical matter, and socio-legal methods can help measure outcomes. ⁵² Whether the European Parliament will ratify the CAI, and whether the treaty will ever come into force, remains to be seen. As more economic treaties integrate comparable provisions on labour and the environment, however, there is a need for rigorous research to interrogate these arrangements, their effectiveness and possible ways to enhance them.