Labour Standards in International Trade Agreements: A Rule of Law Perspective

Abstract

Trade and sustainable development (TSD) chapters in international trade agreements include labour provisions with a view to prevent a “race to the bottom” and reaffirm the existing international labour commitments. When properly formulated and implemented, these provisions could have positive normative impacts on the international and domestic rule of law. This article provides a critical analysis of the evolution of labour standards in international trade agreements, particularly focusing on the European Union’s (EU) approach, in comparison with the approach in the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP). More specifically, it considers ways in which trade agreements contribute to the improvement of labour standards in domestic legal systems. While EU trade agreements embrace minimum standards under international labour law rather than higher standards that apply in the EU legal order, the article argues that there is at least a normative shift towards a more efficient implementation of labour rights through EU agreements, which could make a limited contribution to the rule of law.

I. Introduction

International trade and investment can have contradictory effects on human rights, and more particularly, labour rights. On the one hand, States could be tempted to reduce their domestic labour standards to attract foreign investment and trade; on the other, States’ efforts to improve domestic labour conditions can be received with scepticism and seen as disguised trade barriers. In particular, developing countries face social dumping as companies seek opportunities to outsource production abroad to reduce their labour costs, given that workers in the contracting companies established in these countries are often subject to precarious working conditions, paid lower wages and provided limited social protection. This trend is likely to increase pressures on the workers’ rights in developing countries.

Trade and sustainable development (TSD) chapters in international trade agreements primarily aim to prevent a “race to the bottom” that would lead to a significant degradation of domestic labour (and environmental) standards. These chapters thus aim to prevent the worst-case-scenario. However, they also seek to reaffirm State parties’ existing international commitments regarding labour standards, or to induce them to commit to basic International Labour Organisation (ILO) conventions.¹ Accordingly, TSD chapters have a potential to positively contribute to both domestic as well as the international rule of law by, on the one hand, encouraging the conclusion of these conventions and, on the other, promoting compliance, although their effectiveness in the scholarship remains debatable.²

In the EU, sustainable development in trade is both a political objective and a legal obligation under its founding treaties.³ The sustainable development agenda in EU trade agreements has become a standard for EU agreements post-Lisbon, since the agreement with South Korea in 2011.⁴ The lengthy consultation and revision process of the EU TSD chapters⁵ has led to recent strengthening of substantive provisions as well as stronger emphasis on their enforcement.⁶ The EU’s approach is thus shifting from traditionally “promotional” or “persuasive” towards “sanctions-based” model favoured by some other regional agreements, however, with significant differences. By providing greater clarity and strengthened enforcement mechanisms, these changes could also have positive normative impact on the rule of law.

This article critically explores the relationship between international trade agreements and the rule of law with respect to labour rights. In particular, it examines normative effects of labour provisions in TSD chapters on domestic labour laws and the manner in which they could enhance the international and domestic rule of law. Examining the new perspectives that the rule of law could have on the relationship between international trade law and labour protection law is not an easy task. The protection of labour rights and the promotion of international trade are operationally dissociated, with different legal frameworks and international organisations in charge of overseeing their implementation (primarily the WTO and the ILO respectively). Furthermore, the gradual incorporation of workers’ protection clauses into FTAs has been controversial from the outset, notably for fear of protectionism and that more protective labour standards would have the effect of lowering developing countries’ competitive advantage.⁷

Our analysis takes stock of recent trends in EU agreements, in particular new developments implemented in the text of the EU–New Zealand FTA concluded in 2022.⁸ This approach is compared to the approach implemented in the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) since New Zealand, as well as other significant EU FTA partners (e.g. Canada, Chile, Japan, Vietnam, Singapore), are also parties to the CPTPP. In addition, given that it covers 11 trading nations in the Pacific, the CPTPP is one of the largest regional trade agreements, having a significant impact on the Asia-Pacific supply chains.⁹ It is considered by its parties “a gold standard” for trade agreements.¹⁰ The CPTPP also belongs to the category of regional agreements, which implement “sanctions-based” model for breaches of TSD obligations.

The structure of the article is as follows. First, the article provides a theoretical framework, highlighting the key features of the rule of law and challenges of applying this principle in relation to labour law clauses in FTAs (Part II). Next, it considers the evolution of substantive labour commitments in trade agreements, followed by the analysis of the dispute settlement mechanisms (Part III). Finally, based on this analysis, the article discusses normative impacts of trade agreements on the international and domestic rule of law through consideration of practical implementation of labour obligations (Part IV). The article thus contributes to the existing scholarship on labour standards in international trade agreements by specifically examining their effects through the rule of law prism.

II. The rule of law and the challenges of its application in relation to labour clauses in free trade agreements

In order to understand the effects of the rule of law on labour protection clauses in FTAs, of importance is first to provide a theoretical framework, which will address the origin of the rule of law concept, as well as the status of this principle in different legal systems, at both domestic and international level.

Ontologically, the rule of law is the legal expression of political liberalism and the backbone of liberal democracy. As we highlighted in the introduction to this special issue,¹¹ democracy could be seen as a way of organising power (determining who governs?), while the rule of law as a way of limiting political power of the ruler through rules (how does the govern governs?). It is a concept that has its roots deep in European history. In the common law system of England, the emphasis of the rule of law is placed on the role of courts in protecting rights of private individuals (from arbitrary government power and in ensuring equality before the law). In the European continental systems, the rule of law was conceptualised from the nineteenth century onwards under the German neologism Rechtsstaat and the French concept of Etat de droit.

While traditionally the rule of law has been more commonly invoked by the common law courts, a spectacular development of the rule of law as a new value in the EU legal order stems from the case law of the Court of Justice of the European Union (CJEU), the European Court of Human Rights (ECtHR) and the constitutional courts of European States. The understanding of the rule of law in Europe thus requires an examination of a number of sources, including hard law (e.g. constitutional law), case law interpreting the hard law, as well as soft law instruments (recommendations of the Council of Europe and the European Commission).

A formal conception of the rule of law (focusing on separation of powers, legality, equality, legal certainty, justiciability, etc.) has long prevailed over its more substantive aspects (e.g. human rights, social justice, etc). As in the common law tradition, the concepts of Rechtstaat/Etat de droit aim to constrain actions of public authorities within a legal framework in order to reduce the risk of arbitrariness. This restrictive conception of the rule of law is similar to Joseph Raz’s rule of law, who sees it as one of the virtues to be used to judge a legal system.¹²

Most of the CJEU case law regarding the rule of law is based on the complementary requirements of legality and justiciability. Administration must be subject to judicial review, so that legality of its actions is fully guaranteed. In 2016, the European Commission for Democracy through Law of the Council of Europe (Venice Commission) provided a list of criteria for the rule of law, which includes legality, legal certainty, prevention of abuse of power, equality before the law and non-discrimination, and access to justice, thus also focusing on its formal conception. In effect, these principles are mostly formal and procedural in nature, as they require control over the executive action.

However, the relationship between formal and substantive perspectives on the rule of law is essential in labour law. A purely formal conception of the rule of law runs the risk of not guaranteeing the effective as well as just implementation and enforcement of protection systems for workers and their organisations. We will illustrate the interplay of the two perspectives in the context of the rule of law in the EU legal order. Given its specific features, the EU legal order also exemplifies the interaction between the international and domestic rule of law.

In EU law, procedural aspects of the rule of law are complemented with substantive aspects. This complementarity is the result of the evolution of the EU founding treaties. Having started as an economic integration project, the nature of the EU legal order has been reshaped significantly with the introduction of key values and principles. Since the entry into force of the Treaty of Lisbon in 2009, the rule of law has been enshrined in EU law through Article 2 TEU as a “value” of the EU, in the same way as are respect for human dignity, freedom, democracy, equality and respect for human rights. The constitutional nature of these fundamental values has been stressed in relation to the rule of law conditionally mechanism provided by Regulation 2020/2092.¹³ Moreover, for the European Commission, the constituent principles of the rule of law “are not purely formal and procedural requirements”.¹⁴ They are the vehicle for ensuring compliance with and respect for democracy and human rights. The rule of law is therefore a constitutional principle with both formal and substantive components.¹⁵ In addition, human rights have been constitutionalised through subsequent treaty reforms, culminating with the Treaty of Lisbon, which recognises their foundational nature.¹⁶

This articulation between the procedural and substantive aspects of the rule of law has been fostered by the CJEU. In effect, the Court does not refer to the rule of law as simply a formal and procedural requirement, but also highlights its substantive value by specifying that a “Union based on the rule of law” means that EU institutions are subject to judicial review of the compatibility of their acts not only with the Treaty but also “with the general principles of law which include fundamental rights.”¹⁷ This has been confirmed by the ECtHR, which gives to the rule of law a substantive nature by establishing that it is a concept inherent in all articles of the ECHR.¹⁸

Judicial protection is a key component of the rule of law, as well as labour law. Ubi jus, ibi remedium – where there is a right, there is a remedy. In other words, fundamental rights are effective only if they are justiciable. It follows that where EU law, or national law implementing EU law, grants rights to a person, it must also ensure that these rights are effectively protected. The principle of effective judicial protection of rights which individuals derive from Union law, referred to in the second subparagraph of Article 19(1) TEU, is the concrete expression of the rule of law as a “value” within the meaning of Article 2 TEU.¹⁹ As a “founding principle,”²⁰ effective judicial protection aims to ensure respect for Union law, which is inherent in an entity governed by the rule of law. It is a general principle of Union law deriving from the constitutional traditions common to the Member States.²¹ As the case law currently stands, the material scope of the principle of effective judicial protection encompasses all national rules and practices which could have a negative impact on the Member States’ obligation to provide effective remedies, including the independence and impartiality of their judicial systems. It encompasses access to justice, adequate adjudication procedures, as well as the compliance of the public authorities with the law – the core rationale of the rule of law.

In labour law, the principle of effective judicial protection is nothing new. Already in 1987, in Heylens, the CJEU held:

Since free access to employment is a fundamental right which the Treaty confers individually on each worker in the Community, the existence of a remedy of a judicial nature against any decision of a national authority refusing the benefit of that right is essential in order to secure for the individual effective protection for his right…. Effective judicial review, which must be able to cover the legality of the reasons for the contested decision, presupposes in general that the court to which the matter is referred may require the competent authority to notify its reasons.²²

As our analysis in the forthcoming section demonstrates, labour chapters of the EU FTAs do not provide the same level of protection; they only outline rather vague obligations upon State parties. Hence, there is a striking difference between the importance of fundamental rights, including workers’ rights, in the EU legal order, where rights can be invoked before the courts,²³ and EU FTAs that are based on the principle of reciprocity.

In this regard, we acknowledge the differences and thus also a difficulty of reconciling the concept of the rule of law at the international and domestic levels. On the one hand, the rule of law has only recently been recognised by the UN and the principles derived from it have no binding force. Moreover, international economic law as such does not proclaim the rule of law.²⁴ On the other hand, the concept of the rule of law is now at the core of the case law of the CJEU and the ECtHR. The importance afforded to the rule of law in the European context contrasts sharply with its uncertain status in international law.

Since the EU is called upon to promote its values and principles in its external action,²⁵ it must also ensure that the labour clauses in its FTAs are consistent by the rule of law in the field of labour law. However, to what extent is doable to enshrine the rule-of-law principles in the labour protection clauses in international trade agreements? Should these clauses confer workers’ rights, or should they be limited to obligations imposed on State parties? We consider these questions by examining the labour standards in EU FTAs, comparing them to the approach adopted in the CPTPP, with a view to understand whether the rule of law promoted by the EU is a game changer in this field.

III. Labour obligations and dispute settlement in trade agreements

The principle of legality is one of the core formal requirements of the rule of law. This principle requires that government acts must be based on law,²⁶ which must be clear and specific in order to avoid arbitrariness.²⁷

In the past, international rules have often been imprecise to leave sufficient political leeway to States in managing their relations. However, with the proliferation of international dispute settlement, the emphasis has been placed on more specific, clear, precise and detailed rules based on the best relevant technical evidence. The international rule of law thus presupposes that international law-making processes and rules should satisfy the requirements of clarity, publicity, certainty, transparency and fairness.²⁸ Accordingly, the principle of legality requires that the rights granted to workers should be enshrined in binding texts. The recourse to non-binding texts in social law generally has the effect of weakening the level of protection. However, soft law still has interpretative value. By way of illustration, although the 1989 Charter of Social Rights has no binding force upon social partners or Member States, the CJEU has nevertheless referred to this soft instrument to interpret European law.²⁹

The principle of legality ensures that rules can be implemented and enforced. It is, therefore, appropriate to start our discussion by considering substantive clauses articulating labour standards in trade agreements, in order to assess their ability to produce intended legal effects in accordance with the principle of legality. This will be followed by consideration of available dispute settlement mechanism to support the enforcement of substantive labour clauses in the final part of this section.

Labour standards in EU agreements could be classified into three types of obligations: non-regression clauses concerning domestic laws; provisions based on the minimum protection under the existing international commitments; and aspirational clauses. Neither of these types of clauses create new substantive obligations for the State parties: the emphasis is on upholding the achieved levels of protection or implementing the existing obligations with a view to avoid a race to the bottom.

The first type of clauses includes the obligation of parties not to weaken their levels of protection afforded in domestic law. These clauses are known as non-regression, non-derogation and non-enforcement clauses, and they aim to ensure the level-playing field between the parties. These obligations are conditioned upon the intended or actual effects on trade and investment. The obligations are formulated as a commitment not to derogate from domestic laws, or not to fail to effectively enforce domestic laws “in a manner affecting trade and investment between the parties.”³⁰ Depending on the agreement, the provisions are sometimes conditioned on the intention to “encourage trade and investment”,³¹ rather than actual effects, which arguably has a broader scope of application. In the EU–New Zealand FTA, in addition to a firm expression (“a Party shall not”) in non-derogation and non-enforcement clauses, a general non-regression commitment referring to the overall levels of protection afforded in domestic law has been reinforced³² in comparison to earlier agreements, where it is either non-existent,³³ or vaguely formulated.³⁴ In any event, for these clauses to be effectively enforced, the affected party would still need to demonstrate that regression is used as a tool for fostering domestic trade and investment.³⁵ We shall highlight further below how problematic this is.

The second set of obligations relates to pre-existing international commitments, whose aim is to promote sustainable international trade.³⁶ The EU approach requires the commitment of the parties to implement into their domestic legislation labour rights stemming from the eight core International Labour Organisation (ILO) conventions. This leads to four observations.

First, the ratification of the ILO conventions is not a precondition for the conclusion of an FTA between the EU and its partners. States, such as Japan, Singapore and South Korea, which have concluded FTAs with the EU, had not been parties to all ILO core conventions. In this respect, the approach followed at bilateral level by the EU departs from the one of the EU Generalised Scheme of Preferences (SGP), which requires developing countries to ratify the fundamental ILO conventions listed in its Annex VIII in order to benefit from the tariff preferences provided under the special incentive arrangement for sustainable development.³⁷ Accordingly, the FTAs compliance with the ILO core conventions are weaker compared to the arrangements under the GSP.³⁸ The European Parliament has thus taken the view that it would be preferable to require the EU’s trade partners to ratify all ILO core conventions before the FTA can be concluded.³⁹

Secondly, these core ILO conventions offer a minimum level of protection, such as: freedom of association and the effective recognition of the right to collective bargaining; the elimination of all forms of forced or compulsory labour; the effective abolition of child labour, and the elimination of discrimination in respect to employment and occupation.⁴⁰ While in earlier agreements parties “affirm” or “reaffirm” their commitment to “effectively implement” these international obligations, the language in the EU–NZ FTA has firmed to “shall.”⁴¹ In addition, the scope of obligations has broadened to include a more specific commitment to promote the ILO Decent Work Agenda, through parties’ laws and practices, as well as commitments in relation to occupational health and safety.⁴² There is also a new provision on trade and gender equality, with the commitment to effectively implement the relevant UN conventions and reiterating commitments regarding ILO conventions on gender equality and elimination of discrimination.⁴³

Thirdly, the focus in the EU FTAs on the ratification of core ILO conventions has had the effect of setting aside other areas of labour protection, such as collective redundancies,⁴⁴ carers’ leave⁴⁵ and parental leave,⁴⁶ flexible working arrangements, equal opportunities,⁴⁷ occupational health and safety,⁴⁸ issues which are all subject to harmonisation rules in the EU. In addition, areas such as social benefits, protection against unemployment, hygiene and medical examinations, that have not been subject to EU harmonisation, are not covered by TSD chapters. However, there are some recent exceptions. For example, the EU–New Zealand FTA provides for the application of the ILO Decent Work Agenda, in particular in relation to wages and earnings, working hours, social protection.⁴⁹ Is it possible to go further than the minimal ILO standards in labour clauses in FTAs? In this respect, the EU social harmonisation process is a case in point. There has been constant tension among Member States regarding the harmonisation of social standards, given their diverging national approaches. For instance, in the civil law family, the State has always played a central role in industrial relations. The constitutions of countries, such as Belgium, France, Germany, Greece, Italy, guarantee a set of fundamental social rights and freedoms. By contrast, in the Nordic countries, as wages and working times are usually agreed by the social partners in concluding collective agreements, the State assumes a relatively limited role in industrial relations. As a result, the State intervenes only when asked to do so by the social partners.⁵⁰ Thus, if States belonging to a regional organisation such as the EU, that tends to harmonise market competition, find it difficult to identify a common denominator with respect to social matters, it could be very difficult to determine common standards in terms of wages or working hours between the EU and States such as South Korea or Mexico.

Fourthly, the rather soft and imprecise language of these commitments raises a question about their legal effectiveness. The dispute settlement practice has confirmed that these provisions create legally binding obligations under international law given the pre-existing source of these obligation in the ILO conventions.⁵¹ However, they are obligations of means, not results, and in the absence of more specific targets, milestones or schedule detailing the form or content of the obligation, are likely to be legally ineffective.⁵² The firming of commitments regarding pre-existing international obligations is thus likely to have insignificant effects on the compliance, unless complemented by more specific content of the commitment, also demonstrating the importance of norms’ clarity for the effective rule of law.

The third type of clauses in EU trade agreements are clauses about promotion of higher levels of protection. These are usually expressed in terms of parties’ commitment to “endeavour” or to “seek” to ensure that existing laws and policies “provide for and encourage high levels of protection,” while “striving” to continue improving them towards that goal.⁵³ The position of these provisions in the same article that guarantees parties’ right to regulate demonstrates the emphasis on sovereignty and thus aspirational nature of these clauses. In this sense, it has been argued that the “export” of EU’s values and higher levels of protection has been only rhetorical, with EU’s focus instead placed on the compliance with minimum standards under international law.⁵⁴ This is even more obvious with respect to the aspirational clauses in the EU–NZ FTA,⁵⁵ which are placed in the same article with non-regression/non-derogation/non-enforcement provisions and a more specified right to regulate, that explicitly refers to “consistency” of parties’ commitments with international standards.⁵⁶ However, aspirational clauses, even if they are unlikely to lead to higher levels of protection, could be used as an interpretative tool to reinforce the other two types of clauses.

In comparison to EU agreements, labour obligations in the CPTPP are narrower in their scope,⁵⁷ with the key objective of maintaining the level-playing field.⁵⁸ In terms of pre-existing international obligations, the emphasis is placed on the commitments in the ILO Declaration on Fundamental Principles and Rights at Work and its Follow-up (1998), rather than ILO conventions.⁵⁹

The non-regression clause refers to any waiver or derogation in domestic laws, which would be “inconsistent” with the minimum international standards implemented in domestic laws, or whose implementation would “weaken or reduce” adherence to these standards. The reference is also made to “acceptable conditions of work with respect to minimum wages, hours of work, and occupational safety and health,” whose levels are determined by each party, rather than international standards.⁶⁰ However, the application of these provisions is conditioned on their effects on trade, requiring that weaver or derogation should be done “in manner affecting trade and investment between the Parties.” Accordingly, violations of the TSD chapter can only be established if they have an actual impact on bilateral trade and investment,⁶¹ which is narrower in comparison to the EU approach.

The formulation in the CPTPP excludes the non-regression principle in cases where breaches of labour legislation could be sufficient to hold the party liable but do not have effects on trade and investment. Another limitation is the threshold for triggering the non-regression provision. The requirement that the failure to enforce laws in a “manner affecting trade” is highly ambiguous. This threshold was tested is the US’ claim against Guatemala under the Central American Free Trade Agreement (CAFTA).⁶² The US argued in favour of an expansive interpretation of the term, drawing on the WTO case law regarding the term “affecting” as used in Article III:4 of the General Agreement on Tariffs and Trade (GATT). Accordingly, the requirement encompasses any course of action or inaction by a Party that has a bearing on, influences, or changes – i.e. affects – cross-border economic activity.⁶³ Guatemala argued that such a requirement entailed a high threshold requiring robust evidence that the challenged conduct has influenced trade between the parties. However, the panel in this case did not provide a clear trade effects legal test, requiring only “some competitive advantage” for Guatemalan employers engaged in trade with the US.⁶⁴ This seems to be a rather low threshold, which the US nevertheless could not meet, as it could not demonstrate that Guatemala’s failure to effectively protect the right of association occurred in a “manner affecting trade” between the parties. The case thus also demonstrates the difficulty of enforcing non-regression clauses conditioned on the effects on trade or investment.

The rule of law requires not only that laws are drafted in clear and precise terms in accordance with the principle of legality, but also that they are effectively enforced through the available dispute settlement mechanisms. Unlike in the field of investment protection where justiciable rights are conferred upon foreign investors,⁶⁵ breaches of trade agreements are settled through dispute resolution mechanisms taking place at a State-to-State level. Given their imprecise and conditional formulation, disputes about breaches of TSD obligations are rarely resolved through legal mechanisms, and trade diplomacy plays a much stronger role. However, in addition to strengthening substantive labour obligations (and more generally, broadening its TSD agenda), the second aspect of the EU’s TSD ambition relates to the strengthening of the enforcement of TSD commitments. In the EU–New Zealand FTA, the EU switched to a sanctions-based model, which is nevertheless, distinct from the approach implemented in the CPTPP.⁶⁶

The EU’s pre-review TSD chapters have relied on the engagement with third parties for the purposes of implementation. This implementation is monitored by a Board or Committee on Trade and Sustainable Development, and domestic stakeholders may submit their observations.⁶⁷ In case of a suspected TSD breach, a concerned FTA party can initiate State-to-State consultations. If that fails, and rather as a last resort, a specific TSD legal dispute settlement mechanism is provided, but it lacks financial sanctions for non-compliance. A panel of three independent experts is convened to determine the breach and recommend solutions. However, the implementation of panel’s recommendations is reliant on political pressure as there are no further legal means to induce compliance.

Following the TSD chapters review, TSD disputes are to be subject to the FTA’s horizontal dispute settlement mechanism, and this approach has been implemented in the EU–New Zealand FTA. This process follows the consultations and panel procedures, which are laid down in more detail than the specific TSD mechanism in earlier TSD chapters.⁶⁸ The findings and recommendations of the panel are legally binding, and the defaulting party has an obligation to “take any measure necessary to comply promptly” with them.⁶⁹ The compliance with the panel’s findings is monitored through the compliance stage of the dispute settlement mechanism.⁷⁰ In cases of non-compliance, the aggrieved party has recourse to general temporary remedies (compensation and trade retaliation) but only for breaches of the commitment on multilateral core labour standards and ILO conventions.⁷¹

In comparison, the dispute settlement mechanism in the CPTPP comprises of two stages: labour consultations between State parties, and if that fails, the general dispute settlement mechanism outlined in chapter 28, involving State-to-State arbitration. This approach to dispute settlement is thus similar to the EU’s approach in EU–New Zealand FTA, but it is nevertheless different with respect to its scope. While the CPTPP dispute settlement covers the entire labour chapter, it is de facto limited only to those violations that have an actual impact on bilateral trade and investment given the formulation of its substantive labour obligations. At the same time, the CPTPP has a broader scope for trade sanctions as they cover the entire labour chapter.

In international law, dispute settlement mechanisms between States do not always produce satisfactory results regarding State compliance with their international obligations.⁷² There have been only two cases under international trade agreements involving labour commitments.⁷³ While these cases have had some effects on the enforcement of labour standards in trade agreements, their contribution primarily concerns interpretation of labour standards and clarifying their legal value in trade agreements, as discussed above. The effects of labour standards on the rule of law are thus better considered through their normative dimension. Before we turn to that discussion, the following Table 1 summarises the preceding discussion on TSD chapters and compares the different approaches in EU FTAs and the CPTPP.

Table 1. Comparison of labour standards in select international trade agreements

IV. Normative effects of labour standards in international trade agreements on the rule of law

Having discussed the characteristics of labour provisions in international trade agreements in light of the rule of law’s principle of legality, we now turn to consideration of different normative effects, stemming from these labour standards, on both international and domestic rule of law. In doing so, we draw on the theoretical framework outlined in Part II, and we provide food for thought on how the rule of law with respect to labour law could be further strengthened through international trade agreements.

On the one hand, TSD chapters have a potential to contribute to the international rule of law by ensuring compliance with pre-existing international legal obligations of States, thus bridging the gap between different international regimes (labour and trade). In this sense, they could be seen as contributing to the international rule of law by ensuring greater enforceability through the threat of dispute settlement and trade sanctions, which is otherwise unavailable under the international labour instruments. At the same time, given that different trade agreements differently conceive labour obligations, it may be more challenging to assess State’s compliance with legal obligations in specific cases. Diverging interpretations of ILO obligations by different dispute settlement bodies set up under trade agreements could thus lead to further contradictions and more fragmentation of international law. This could ultimately undermine predictability and certainty in legal relations between States, and thus also the international rule of law. However, this is less of a problem for the EU type of trade agreements, which refer to core ILO conventions, rather than the ILO Declaration (as it is the case in the CPTPP), which lacks more clarity in defining the rules of conduct.⁷⁴ In addition, EU FTAs envisage the utilisation of information from the ILO in trade disputes concerning labour standards, which could, if properly implemented, contribute to more coherence between international trade and labour regimes.⁷⁵

On the other hand, the examination of the rule of law is probably more meaningful at the level of national implementation of legal standards enshrined in the FTAs. The rule of law implies compliance with a series of legal principles, including equality and non-discrimination, legality, legal certainty and separation of powers. As these general principles have been developed mainly by the supreme courts of the civil family countries since the 1960s, they were not designed to govern trade relations between States. Compliance with labour law standards requires the application of these principles. Accordingly, legal arrangements aiming at protecting workers must be drafted in a clear and precise manner, without ambiguity, by the legislator and implemented by the executive agencies.⁷⁶ Further, in accordance with the principle of non-discrimination, labour rules must be applied in the same way to undertakings in similar situations. In addition, categories of workers may not be discriminated against. The rule of law also requires that standards restricting fundamental rights – freedom of association, the right to strike, etc – can be challenged before independent and impartial courts. The right to take legal action, whether for workers, trade unions or employers, is mainly a domestic issue.

Scholars have questioned the utility of TSD standards in trade agreements, emphasising their legal weakness, ineffectiveness, and domestic politics as their core rationale.⁷⁷ However, as non-compliance with the ILO standards does not lead to sanctions under the existing ILO mechanisms, the question is whether the availability of the enforcement mechanism under trade agreements could serve as a leverage in achieving compliance of domestic labour laws with international standards, and, at least in theory, enhance the domestic rule of law. A comparison could be made between labour and environmental clauses in FTAs. They are usually similar (non-regression, implementation of the existing international obligations, aspirational clauses) and encapsulate the same objectives. However, as far as compliance with international obligations is concerned, it seems much easier to demonstrate a breach of an ILO Convention than a breach of an international environmental agreement such as the Convention on Biological Diversity or the Basel Convention on transfrontier movement of waste. As environmental provisions are often rather vague, they give the State parties a very wide margin of discretion that is not likely to be subject to judicial review at domestic level. Thus, when they refer to ILO agreements, clauses concerning the protection of workers can be more easily invoked than environmental clauses.

A case in point in this regard is the EU–Korea dispute under the 2011 EU-Korea FTA, which was the first and only dispute settlement procedure for a breach of TSD obligations in EU FTAs. The dispute concerned Korea’s insufficient action in implementing into domestic legislation labour rights stemming from ILO conventions,⁷⁸ and it took place prior to the EU TSD chapters review. The EU succeeded in three claims demonstrating that Korea’s measures in its domestic law, particularly those relating to the principle of freedom of association, are inconsistent with Korea’s ILO labour obligations under the FTA.⁷⁹ However, with respect to Korea’s commitment to make “continued and sustained efforts towards ratifying the fundamental ILO Conventions,” the Panel assessed Korea’s efforts as “less-than-optimal” but nevertheless concluded that these efforts, especially those taken since 2017, did not fall short of Korea’s obligations under the FTA.⁸⁰ The Panel ruled in favour of Korea on this point, demonstrating the legal ineffectiveness of this FTA obligation in the absence of any schedule detailing the form or content of the obligation.⁸¹

A further question is whether these international standards oblige trade partners to level the playing field between themselves? Or does the focus remain on aligning domestic laws with the minimum level of protection enshrined in the ILO conventions, regardless of the higher standards that might be enshrined domestically in one of the parties? The above analysis demonstrates that worker protection clauses in TSD chapters merely require a minimum available under international labour law. The fact that higher standards might be applied domestically in the territory of one party does not oblige the parties to enshrine these higher standards in their trade agreements, which would more genuinely contribute to levelling the playing field. This is particularly illustrated in the case of the EU, whose legal regime enshrines higher standards compared to those included in the ILO conventions.

That being said, certain minimum rules imposed by the ILO conventions correspond to fundamental rights recognised by the European courts. For example, alignment with ILO Convention 98 has been required by the new generation of FTAs concluded between the EU and third States (e.g. Singapore, Vietnam, etc.). Article 4 of that Convention regarding collective bargaining requires that the law promotes “the full development and utilisation of machinery for voluntary negotiation” between workers’ organisations and employer groups to ensure the regulation of employment “by means of collective agreements.” In EU law, Article 28 of the Charter of Fundamental Rights (CFR) aims to promote social dialogue through collective bargaining. In the case law of the ECtHR⁸² and the CJEU,⁸³ the right to collective bargaining has been recognised as a fundamental right. Accordingly, the protection of the fundamental right to bargain collectively must “take full account, in particular, of national laws and practices.”⁸⁴ By the same token, the right to strike has been recognised by the ILO’s supervisory bodies “as an intrinsic corollary of the right to organize,” protected by the Convention No. 87.⁸⁵ Last, the ban on child labour provided for in Article 32 of the CFR mirrors the 1999 ILO Convention on the Worst Forms of Child Labour. It is thus unsurprising that FTAs concluded by the EU impose on the other parties the obligation to ratify and implement Convention 98.

However, beyond these international conventions, of which there are only a few, there is a risk of discrepancy between the level of protection achieved domestically in the EU and the level of protection required under EU FTAs. This raises the question of the appropriateness of these labour clauses to prevent social dumping, let alone a race to the bottom. The importance of social rights proclaimed in the EU legal order is not sufficiently reflected in the FTAs concluded by the EU. We will illustrate this in light of several provisions of EU Treaty law that entered into force with the Treaty of Lisbon in 2009. These provisions have affected the EU framework of fundamental freedoms.⁸⁶

First, Article 9 TFEU lays down a “cross-cutting” social protection clause obliging the institutions “to take into account requirements linked to the promotion of a high level of employment, the guarantee of adequate social protection, the fight against social exclusion, and a high level of education, training and protection of human health.” Secondly, Article 3(3) TEU states that the construction of the internal market is to be realised by means of policies based on “a highly competitive social market economy, aiming at full employment and social progress.” Furthermore, adopted in 1961 by the Council of Europe, the European Social Charter, revised in 1996, brings together the fundamental social rights to which employees are entitled. This Charter was a major source of inspiration in the drafting of the EU Charter of Fundamental Social Rights and the social provisions of the CFR (Articles 27, 28, 30, 31, 32).

While EU FTAs insist on compliance with the basic ILO conventions, there are gaps between these treaties and the social provisions of the CFR. For example, CFR Article 30, which provides for the protection of workers against unjustified dismissal, is close to ILO Convention 158, which is not included in the labour clauses of the TSD chapters. By the same token, the workers’ right to working conditions which respect (their) health, safety and dignity enshrined in CFR Article 31 has no equivalent in ILO law. We might, therefore, conclude that compliance with ILO conventions by trading partners is essential in terms of respect of EU fundamental social rights (collective bargaining, collective action, prohibition of child labour), but that it is not sufficient to reflect all EU’s social aspirations. Certainly, the legal relationships that the EU develops with third countries must not be a carbon copy of EU domestic law. Nevertheless, Article 21(1)(a) TEU requires the EU institutions to work for a high degree of cooperation in all fields of international relations, in order to safeguard its values. Taking account of values, including social values, is a legal imperative, not merely a political aspiration.⁸⁷

It could be also possible to strengthen the domestic rule of law by placing stronger obligations on the parties to provide effective labour inspection systems. So far, only a few FTAs impose such obligations. Where this is the case,⁸⁸ additional legal issues could come into play from the EU’s point of view. On the one hand, the conclusion and the implementation of the FTA is subject to an exclusive EU competence on which the Member States have no say.⁸⁹ On the other, labour inspection systems have not been subject to harmonisation measures in EU secondary law. The EU–New Zealand FTA thus obliges the 27 Member States to maintain a labour inspection system, even though they did not participate in the negotiation of the treaty, which could ultimately undermine its effective implementation at the domestic level, and thus also the rule of law.

We have discussed the limitations of the dispute settlement schemes provided for by the FTAs. There are further factors that may explain the failure of FTA obligations to protect workers.

First, limited resources prevent the strict oversight of the implementation of these agreements. For monitoring and cooperation purposes, EU TSD chapters rely on specialised TSD committees and civil society dialogue through domestic advisory groups (DAGs), comprising of business, environment and labour groups, and Civil Society Forums.⁹⁰ The roles of these bodies, as well as their concrete contributions towards the implementation of TSD objectives, are rather undefined.⁹¹ Studies demonstrate their limited political influence and policy impact,⁹² which has ultimately led to attempts to strengthen their role. Starting with the EU–UK Trade and Cooperation Agreement, the mandate of DAGs and Civil Society Forums has been widened to cover the implementation of the entire FTA,⁹³ with further efforts to improve their transparency. However, for trade unions, the considerable efforts to prepare and file a comprehensive complaint in order to prompt consultation and dispute settlement explain why the investigation of their complaints may take several years, as demonstrated by the EU–Korea dispute.

Similar to the EU approach, the CPTPP also provides for mechanisms of engagement at the inter-governmental (through the Labour Council⁹⁴) and the domestic level. Public engagement is envisaged through less structured set-up of contact points, which are to act as a channel of communication with the public, and to assist the inter-governmental Council.⁹⁵ In addition, each party should maintain a national labour consultative or advisory body (or a similar mechanism), consisting of business and labour organisations “to provide [their] views on matters regarding [the CPTPP labour] chapter.”⁹⁶ These, primarily political processes, are the only available avenue for the domestic stakeholders to raise their concerns, as any direct claims against CPTPP States are explicitly excluded.⁹⁷

Recent developments in the EU have aimed at improving the monitoring of the implementation of labour clauses by providing greater opportunities for private actors to report breaches. As part of the efforts to strengthen enforcement, a new institution of the Chief Trade Enforcement Officer (CTEO) was created in 2020. At the same time, the Single Entry Point for trade-related complaints has also been extended to complaints about non-compliance with TSD commitments and is now available to a broader range of stakeholders (industry, but also trade unions, NGOs, citizens, Member States). The aim is to enable a privately induced and centralised complaints mechanism for breaches of TSD commitments in EU FTAs, which can then serve as a trigger for an EU enforcement action, either through the dispute settlement of the FTA in question, or unilateral measures, while still leaving sufficient discretion to the Commission in prioritising actions.⁹⁸

Secondly, the complaints mechanisms, even if more open to private actors, still primarily focus on government inaction and not on breaches committed by companies.⁹⁹ The first complaint submitted to the EU’s CTEO through the Single Entry Point concerned the alleged breaches of a number of labour obligations in the mining sector in Columbia and Peru, under EU trade agreement with these countries, and it was submitted by a Dutch complainant, on behalf of trade union organisations from Columbia and Peru.¹⁰⁰ For private actors’ complaint to succeed, the violation of labour rights in the third country must be “systemic,” either regarding “systemic misapplication of legislation” or “systemic failure to apply a law or regulation,” which excludes any “isolated cases of non-compliance”.¹⁰¹ It is thus likely that the new mechanism will lead to more diplomacy.¹⁰² In the absence of trade sanctions envisaged in the FTA itself, the legality of any unilaterally imposed measures (e.g. rebalancing duties, suspending tariff concessions) would be debatable under international law, thus also questioning the practical effectiveness of the new monitoring mechanisms.¹⁰³

Finally, the proper application of labour law in light of the rule of law implies that both employers and employees should be able to bring their disputes before independent and impartial courts. Until now, labour clauses in FTAs have made no provisions for such requirements beyond rather vague non-enforcement clauses, whereas EU law and European human rights law pay heed to these requirements. Could the rule of law require that a new generation of FTAs imposes obligations on States to guarantee that any measures restricting fundamental labour rights guaranteed in FTAs (e.g. freedom of association, the right to strike, etc.) could be challenged before independent and impartial courts, established by law? This issue raises a fundamental question that has not been discussed in the legal literature related to trade law since the right to take legal action, whether for workers, trade unions or employers, is mainly a domestic issue. We offer some food for thought based on the standards of independence and impartiality discussed in EU law and ECHR.

The overlapping requirements of independence, impartiality and prior establishment by law are intrinsically linked. Failure to comply with one of them is likely to result in failure to comply with the other requirements.¹⁰⁴

First, independence of courts and tribunals (Article 19(1) TEU read in the light of Article 47 of the CFR) is essential for guaranteeing effective judicial protection.¹⁰⁵ Courts must disregard any national provision or case law that would call these requirements into question.¹⁰⁶ It is important to remove any legitimate doubt in the minds of litigants as to the imperviousness of the court in question to external factors and its neutrality in relation to the conflicting interests.¹⁰⁷ Judicial bodies are required to carry out adequately the specific function that the State has entrusted to them, in accordance with the principle of the separation of powers,¹⁰⁸ which is also one of the component of the rule of law.¹⁰⁹

In this regard, the ECtHR highlights four elements of judicial independence that are either external or internal: the appointment of its members, the term of office, the existence of guarantees against outside pressures – including in budgetary matters; and whether the judiciary appears independent and impartial.¹¹⁰ Similarly, the notion of independence in the CJEU’s case law encompasses an external as well as an internal dimension. The rationale of these two dimensions is related to the public perception of the independence of courts. The aim is “to dispel any reasonable doubt in the minds of individuals as to the imperviousness of that body to external factors and its neutrality with respect to the interests before it.”¹¹¹ In that way, “justice must not only be done, it must also be seen to be done.”¹¹²

The political will to appoint judges with a view to representing the different opinions that run through society can be salient. For example, many countries provide for labour courts to be composed of equal numbers of workers’ and employers’ representatives.¹¹³ The question arises as to whether such a political balance is consistent with the requirements of independence and impartiality that flow from the rule of law. We believe that this balance does not pose a problem as long as labour judges are appointed solely on the basis of their professional competences and not their political affiliations. The national judicial system must therefore ensure that candidates have sufficient skills and competences to take on such duties. This could require the intervention of an independent commission that would check whether the professional skills comply with the degree of independence required.

Would it be appropriate, when modernising labour protection clauses in the future, to include more precise obligations regarding the independence and impartiality of the competent and specialised courts, which would go beyond ILO international obligations? One could argue that the organisation of the judiciary is a matter of national sovereignty. However, assistance, cooperation and compliance with these requirements by State parties to a free trade agreement could strengthen the rule of law in labour law in the long term.

V. Conclusion

International trade agreements include a number of labour standards clauses, the main purpose of which is to limit the risk of social dumping, in particular by obliging States to improve and effectively apply their labour laws.

The literature highlights that commitments on labour rights in EU trade agreements have had little positive impact and fell short of ensuring compliance with ILO fundamental labour rights in law or in practice.¹¹⁴ By including minimum standards under international labour law rather than higher standards in the EU legal framework, EU agreements have also missed an opportunity to level the playing field in a way that would, at least at a normative level, contribute more significantly to the improvement of labour rights in third countries.¹¹⁵ At the same time, the political reality is such that more often than not, EU trading partners will be hesitant to agree to the strengthening of even the minimum standards, and the EU will not risk the conclusion of an FTA for the sake of labour standards.¹¹⁶

The EU-Korea case demonstrates the use of legal dispute settlement as the last resort, in line with the EU approach pre-TSD review, which focused on the promotion of sustainable development values through political dialogue and regulatory cooperation, rather than more adversarial methods. The enforcement of TSD obligations in earlier FTAs ultimately depends on the diplomatic pressure and voluntary compliance by third countries, which in the case of Korea proved effective.¹¹⁷ Post-TSD chapters review, it can be expected that legal dispute settlement will become more common, while the use of trade sanctions will be the last resort. The EU approach is also increasingly focused on private actors’ complaints and unilateral measures¹¹⁸ in ensuring greater effectiveness of its trade agreement mechanisms.

Ultimately, the rule of law requires, both at international level and in the implementation of labour clauses at domestic level, that respect for the fundamental rights granted to employers, workers and their organisations be clearly specified in accordance with the principles of legality and legal certainty. In addition, State parties to FTAs should ensure that not only the enforcement and control mechanisms are effective, but also that interested parties can easily initiate proceedings before their domestic courts. Although the rule of law is applied differently in international economic law than in national law, of importance for both is that rights and obligations are clearly and precisely framed to ensure their effective and just implementation. Accordingly, the enforcement of labour rights through international trade treaties could have beneficial effects on the rule of law, insofar as it enhances the common playing field.