13 - The Trade and Labour Nexus

Summary

Preferential trade agreements (PTAs) of recent times have become much deeper and wider in the scope of issues and topics covered. We observe that PTAs now encompass a range of non-trade issues as well, such as sustainability and labour standards. This chapter examines the current trends in labour standards covered in PTAs, on both the design of provisions and the levels of success in implementation of these provisions at the domestic level. Upon evaluating past practices, this chapter recommends a higher level of ambition in labour chapters of trade agreements of the future. By relying on case studies, we prescribe that future labour chapters should i) expand the scope of labour rights, ii) emphasise involvement of local stakeholders for implementation and enforcement, and iii) reinvent the approach through which labour violations in trade are examined. The chapter concludes by exploring ways in which PTAs can act as a safeguard for labour rights amidst future developments, such as automation and migration.

13.1 Introduction

Since the 1990s, preferential trade agreements (PTAs) have proliferated, increasing international market access and reducing trade barriers. Stasis at the World Trade Organisation (WTO) following the collapse of the Doha Development Round in 2001 has only amplified this trend, contributing to a marked shift away from multilateralism towards regionalism. Preferential trade agreements are now a principal supranational forum through which countries engage in trade and, in recent years, have evolved to include a wide range of non-traditional trade issues.

This chapter focuses on labour provisions in PTAs. Trade and labour have an inherent tension due to the mixed effects that trade can potentially have on the domestic labour market. On the one hand, exports can create new job opportunities, increase wages, and improve working conditions. On the other hand, import competition can lead to job losses in competing industries and put downward pressure on wages and working conditions, particularly for low-skilled workers. Additionally, globalisation and increased competition from foreign workers can lead to a race to the bottom in terms of labour standards, as companies look to cut costs and increase profits. Ultimately, the impact of trade on labour depends on several factors, including the specific sectors and countries involved, labour market flexibility, the state of the global economy, and the presence or absence of strong and enforced domestic labour regulations. Given these complex dynamics, investigating the impact of trade agreements on labour standards and the optimal formulation of labour provisions in order to improve the status quo are important areas of enquiry that warrant attention.

This chapter is structured as follows. The first section starts with a general mapping of trends in labour provisions in PTAs on two levels: internationally and regionally. On the international level, it examines the role of international organisations such as the General Agreement on Tariffs and Trade (GATT), WTO, and, in particular, the International Labour Organization (ILO), in relation to labour provisions in PTAs. On the regional level, we examine how different regions approach the issue of trade and labour in their design of PTAs and how they differ in significant ways. After taking stock of the current state of play, the second section evaluates the successes or failures of past practices, focusing on four main points: i) enforcement mechanisms of labour provisions in PTAs, ii) a textual interpretation issue concerning the phrase ‘manner affecting trade’, iii) the possibility of the expansion of labour rights in PTAs to accommodate future trends in labour, and iv) the different approaches between developed and developing countries in their inclusion of labour provisions in PTAs. The chapter then concludes with recommendations for labour provisions in future PTAs, in both formulation and compliance.

13.2 Current State of Play

Advocates of free trade often claim that it allows individuals and workers to benefit from a system that offers a greater choice of goods for cheaper prices. This stems from the principle of comparative advantage famously posited by David Ricardo, where maximum efficiency can be obtained when countries of differing factor abundance use their comparative advantage to trade goods that utilise their abundant factors (Ricardo 1817). Since economies will be open to more competition, this creates a plethora of other benefits, such as a dynamic business climate, increased investments, knowledge transfer from foreign companies, and job creation. Indeed, several studies have shown that trade does in fact encourage economic growth (e.g. Edwards 1993; Krueger 1998; Frankel and Romer 1999). Regarding PTAs in particular, Baier et al. (2014) note that PTAs can have a positive effect on increasing trade flows between member countries but note that the overall impact on trade and economic growth varies depending on size, distance, and level of development of the countries involved.

However, recent research also points to the increasing costs of trade. Inequality within countries has been found to increase with the expansion of trade (Ghose 2001; Helpman 2016; Pavcnik 2017), and unsatisfactory labour conditions in developing countries have been linked to trade liberalisation via the well-known concept of the ‘race to the bottom’ enabled by international trade and foreign investment (Davies and Vadlamanatti 2013; Anner 2018; Wang 2018).

Given that trade is an issue that transcends borders and involves power asymmetries among private and public actors between trading partners, the debate concerning whether trade is ultimately beneficial or harmful raises interesting questions that must be addressed when considering future labour provisions. Which (if any) international institutions are the proper forum to protect labour rights and ensure a more equitable global economy? Should PTAs include labour provisions, or should those remain strictly under the purview of the ILO? Are labour provisions in PTAs consistent with other international rules? Additionally, what has worked and what has not worked in labour chapter templates and what are the potential avenues for improvement or expansion of labour provisions in PTAs?

13.2.1 Developments on the International Level (GATT, WTO, and ILO) and Their Relationship with Labour Provisions in PTAs

Preferential trade agreements are not the only treaties dealing with labour issues. The most prominent international organisations with an interest in labour issues are the WTO and the ILO. This section outlines the role of these two organisations and highlights the ways in which they are compatible in advancing workers’ rights in trade.

The United States (US) and the former European Economic Community/European Community sought to include labour standards in both the GATT and the WTO agreements. In 1953, for instance, the US State Department proposed an ‘unfair labour clause’ to the GATT, but the member countries could not reach a consensus on its definition (Cottier and Caplazi 1998). In 1994, the European Parliament proposed to broaden Article XX(e) of the GATT by including child labour as well as the infringement of the principle of freedom of association and collective bargaining (FACB) (Waer 1996).

However, these attempts failed mainly due to the opposition of India and other developing countries, who opposed the inclusion of labour clauses at the WTO on the grounds that they represent ‘protectionism in its new garb’ (Kolben 2006: 237; see also Hepple 2005: 130; Bhagwati 1995; Staiger 2003; McCrudden and Davies 2000). India’s government, business interests, and even labour and civil society organisations actively worked against including or expanding labour clauses, organising opposition among developing nations. Moreover, scholars have criticised the inclusion of non-trade issues (NTIs) at the WTO and in PTAs (and the agreements themselves, generally) as tools of neocolonialism and a means for enforcing foreign, Western norms on other parties to the agreements (Bhala 2007; Villalta Puig and Ohiocheoya 2011).

At the 1996 Singapore Ministerial Conference, WTO Members formally conferred on the ILO authority over labour standards in its Declaration: ‘We renew our commitment to the observance of internationally recognized core labour standards. The International Labour Organization (ILO) is the competent body to set and deal with these standards, and we affirm our support for its work in promoting them … We reject the use of labour standards for protectionist purposes, and agree that the comparative advantage of countries, particularly low-wage developing countries, must in no way be put into question…’. (WTO 1996).

Again in 1999, leading up to the third WTO Ministerial Conference in Seattle, both the US and the European Communities proposed a joint ILO/WTO working forum (WTO 1999a, 1999b). This forum would serve as a formal mechanism within the WTO to foster dialogue and research on labour issues and strengthen the ILO by granting it observer status at the WTO.

However, despite the proposed forum, labour unions and antiglobalisation protesters expressed concerns about the negative impacts of globalisation on workers’ rights, the environment, and human rights. The widespread protests were a turning point in the global debate on trade and globalisation and brought attention to growing public opposition to unfettered free trade, to the degree that the protests surrounding this conference were dubbed the ‘Battle in Seattle’. Since then, there has been little sustained effort to bring labour issues into the WTO agenda.

The ILO figures prominently in the relationship between trade and labour given that its internationally recognised labour standards are often integrated into trade agreements. Most labour provisions in PTAs refer to the 1998 ILO Declaration on Fundamental Principles and Rights at Work and its follow-up, the ‘1998 ILO Declaration’, or the eight fundamental conventions and the ILO Decent Work Agenda (Velut et al. 2022: 84). The 1998 ILO Declaration contains four fundamental principles regarding labour (ILO 1998):

1. 1. Freedom of association and the effective recognition of the right to collective bargaining.

2. 2. The elimination of all forms of forced or compulsory labour.

3. 3. The effective abolition of child labour.

4. 4. The elimination of discrimination in respect of employment and occupation.

There are various manners in which ILO commitments are referenced in PTAs. Sometimes, parties only reaffirm their ILO-related obligations entailing no further legal obligations or choose to incorporate ILO commitments directly into PTAs, either through ‘best endeavours’ clauses or by directly requiring parties to implement those standards in their domestic legislation (Agustí-Panareda et al. 2014). In principle, it is possible that a country binds itself by incorporating ILO conventions that it has not yet ratified into a PTA, and thus expands its obligations through the signing of the PTA. However, this has not happened so far, as countries have only included ILO treaties which they previously ratified.

While PTA labour provisions usually mirror the ILO-related obligations of a country, some PTAs require the parties to ‘make continued and sustained efforts towards ratifying and effectively implementing the fundamental ILO conventions’. Through this approach, countries with higher ambition on labour standards use PTAs to put additional pressure on a partner country to ratify the ILO conventions. This is especially common in North–South agreements. Moreover, as the ILO does not have a legally binding enforcement mechanism for non-compliance through their conventions, dispute settlement mechanisms in PTAs can help to close that gap. Allowing a party to take trade sanctions to offset the violation of a treaty provision has potential to be a stringent and efficient way to enforce ILO commitments (Melo Araujo 2018).

On the implementation front, the role of ILO is less pronounced (Peels and Fino 2015: 197). While the ILO is not so present as an actor in the domestic space, many agreements still specify ILO assistance in technical cooperation activities, monitoring, and capacity building (Agustí-Panareda et al. 2014: 20). For example, in the European Union (EU)–Caribbean Forum (CARIFORUM) and EU–Canada Comprehensive Economic and Trade Agreement (CETA), the ILO is named in a provision on advice for best practices and the use of effective policy tools in addressing labour rights challenges. In the US–Colombia agreement, the ILO is explicitly stated as a supporting pillar for national contact points in capacity-building work. The ILO is also involved in a consultative capacity for disputes in some instances. In the case of EU–Singapore, the PTA specifies that ‘parties may seek the views of [the ILO] in order to fully examine the matter’ in the case of labour disputes.

In addition to the link through the ILO or PTAs, labour standards and trade are connected via the Generalised System of Preferences (GSP). Based on what is commonly referred to as the enabling clause under WTO law, WTO Members can deviate from the ‘most-favoured-nation’ principle and accord more favourable treatment to developing countries. The US and the EU made use of this mechanism by establishing their respective domestic GSPs, where benefits such as preferential tariff treatment can be granted if a developing country meets certain criteria. However, preferences under GSP programmes may be suspended in case of violations of ILO core labour standards (Siroën 2013). Thus, these schemes provide for the unilateral trade suspension of benefits against developing countries that are independent from the suspension of concessions that may be used based on a PTA.

13.2.2 Variation of Labour Provisions PTAs by Region

As of 2022, 113 out of 354 PTAs included labour provisions, half of which were ratified over the last decade (ILO 2022). Although labour provisions tend to be concentrated in North–South trade agreements, there is an increasing trend to integrate labour provisions into trade agreements among developing countries, or South–South trade agreements.

Figure 13.1 depicts the cumulative number of trade agreements that include labour provisions. We observe a very strong trend in the early 2000s of including labour provisions in North–South trade agreements. This indicates that countries in the North possibly use PTAs as a conditionality tool to grant market access in return for improvements in domestic labour standards within Southern counterparties. Interestingly, we find a smaller trend among South–South and North–North partner countries, which indicates that these governments may attach lesser importance to including provisions among partners with similar labour standards.

Figure 13.1 Labour provisions over time.

Source: Data taken from the ILO LP Hub (ILO 2022). Own classification and depiction

A similar picture is shown when we look at the variation of labour provisions in PTAs through a regional lens. Figure 13.2 depicts the number of PTAs with labour provisions by region. Most agreements of this sort have been concluded extra-regionally between European states and countries in the Asia-Pacific region. Sixteen agreements of this sort have been concluded within Europe, and another dozen within Asia-Pacific countries. To date, approximately ten agreements have been concluded that contain labour provisions in African countries.

Figure 13.2 Labour provisions by region.

Source: Data taken from the ILO LP Hub (ILO 2022). Own classification and depiction

As Figure 13.2 illustrates, labour provisions are especially present in North–South agreements since the labour standards of members of North–North and South–South treaties are often homogeneous. Due to the different levels of ambition, there exists variation in the scope, enforcement mechanisms, and compliance.

We categorise these various approaches into two broad schools of thought – enforcement or managerial (Tallberg 2002). The enforcement school is rooted in the political economy theory of collective action and assumes that states are rational actors. It emphasises that labour laws must be strictly enforced through penalties, fines, and other legal sanctions to ensure that states and companies comply with the law. This approach is based on the idea that some actors may not comply with the law unless there is a strong deterrent in place.

The managerial school, on the other hand, emphasises that actors, both states and companies, should be encouraged to voluntarily comply with labour laws and regulations through cooperation, mutual capacity building, and technical assistance. This approach relies on the idea that companies inherently want to comply with the law, but may need help understanding their obligations, especially if the language leaves room to be interpreted broadly.

Most parties adopt a mixed approach, which is a combination of both the enforcement and managerial. In this section, we take account of the different approaches taken by countries over time.

Since its first venture of including labour in its PTAs, the US has always adopted an enforcement approach in its PTA design. The US often requires its negotiating partners to make changes to domestic labour laws prior to the agreement’s entry into force.

A side agreement to the North American Free Trade Agreement (NAFTA), which entered into force in 1994, was the first time the US introduced binding labour provisions in connection with a trade agreement. The North American Agreement on Labour Cooperation (NAALC) contained clear obligations imposed on the contracting parties, such as an obligation to provide for high standards with reference to eleven labour principles, as well as obligations concerning the enforcement of domestic labour laws (Article 2 and Annex 1, NAALC). The agreement established domestic institutions called the National Administrative Offices (NAOs) to take on the role of acting as contact points between partners for information exchange and to oversee implementation.

The NAFTA also included a general dispute settlement mechanism that could be applied to labour issues. In the case of a labour law abuse, a committee of independent experts is convened if the issue is not resolved after rounds of consultations. A persistent pattern of failure to enforce the labour provisions, namely child labour, lack of minimum wage standards, and occupational safety and health risks could lead to the issue being escalated to an arbitral panel (Arestoff-Izzo et al. 2008). In the event the panel ruled against a country, the penalties could include a monetary assessment and temporary suspension of trade benefits outlined in the main text. In these ways, the US’ strong enforcement approach was a way to signal to partners the importance it ascribes to high labour standards and fair competition.

Following the NAFTA, the US signed an agreement with Cambodia in 1999 (US–Cambodia Textile and Apparel Trade Agreement [UCTA]) that limited labour standard provisions to the textile industry and conditioned market access on Cambodia demonstrating compliance with specific labour standards (Wells 2007). In 2001, the US–Jordan PTA’s labour clause reflected a commitment by both parties to not waiving or derogating from their labour laws in their pursuit of encouraging trade, as well as effective enforcement of domestic labour laws. Subsequent agreements led by the US also followed suit, such as bilateral PTAs with Chile, Morocco, Australia, Singapore, Bahrain, Oman, and Peru as well as multilateral agreements, most notably the Central American Free Trade Agreement (CAFTA).

Given that the US was the first mover on high-ambition labour provisions, the language in US labour chapters has spilled over into other agreements as well. The labour obligations in the Trans-Pacific Partnership (TPP) form part of the Canada–EU CETA concluded in 2016. They also appear in the recently concluded EU–Viet Nam Free Trade Agreement, in the sustainable development chapter (Claussen 2021).

Similar language can also be found in certain agreements signed by Asian countries. However, the main difference is that none of these agreements has a sanction mechanism in case of non-compliance. Hence, they embody the managerial approach as opposed to the enforcement tactic. Latin American countries do not include labour provisions in their PTAs – both Mercosur and the Pacific Alliance do not include any labour references in their PTAs.

The US has strengthened its approach towards enforcement of labour provisions in recent times. In 2020, the US along with Mexico and Canada reached a consensus on the United States–Mexico–Canada Agreement (USMCA), which is considered to be a modernised, high-standard reinvention of the now-defunct NAFTA. The revision was due to growing criticisms that the NAFTA no longer addressed the present-day issues, and the USMCA was designed to account for changes in the global economy and to address issues such as intellectual property protection, digital trade, labour rights, and environmental standards. Interest groups also raised concerns that the NAALC provided insufficient means for the parties to enforce its obligations, leading to concerns regarding a ‘race to the bottom’ regarding working conditions and standards.

The USMCA was designed with a prominent emphasis on labour standards, and those standards are found directly in the main text as opposed to being delineated in a side agreement. Moreover, the USMCA also includes novel mechanisms to enforce labour standards. The key achievements of the USMCA include commitments by Mexico to change domestic law to provide effective recognition of the right to collective bargaining, new provisions prohibiting imports of goods produced by forced labour, addressing violence against workers, and protection of migrant workers. The agreement also includes new trade rules of origin to push for higher wages by requiring that 40–45 per cent of auto content be made by workers earning at least USD 16 per hour. Finally, and possibly most notable, is the novel dispute settlement mechanism called the Rapid Response Mechanism (RRM) that ensures expedited enforcement of labour rights to ensure effective implementation (USTR 2016). The RRM will be expanded upon in a later part of this chapter.

Even though it relied almost entirely on legally non-enforceable labour provisions, Europe was also a pioneer with its inclusion of labour provisions in PTAs in the 1990s (Raess et al. 2018). With the expansion of the EU over time and trade linkages being formed outside of the region, many scholars have characterised the bloc’s approach to labour provisions as advancing a ‘values-based trade agenda’.

Manners (2002) conceptualised the EU as a normative power that works mainly through ideas and values using trade relations and instruments such as PTAs as a viable channel rather than military force. Normative power is not typically exerted with force, but instead through deliberation. The underlying rationale of this ‘norm-diffusion’ stems from a more social and human rights perspective, in which labour provisions can be used to improve labour conditions in countries that otherwise would lack the political will or implementing capacity. Hence, labour provisions are viewed as a means of ensuring respect for labour-related human rights that reflect values universally accepted by the international community (European Commission 2018). Labour rights in signatory countries to EU PTAs are impacted ex post rather than ex ante, which differs from the US. The EU also does not include sanctions as a remedy for failure to enforce labour standards. Hence, the EU approach has been characterised as more consultative and managerial.

However, this approach has led to criticism that the EU policy on labour in trade is not assertive enough (Nissen 2022). Coincidentally, in the wake of these criticisms, the EU has undergone institutional changes in its decision-making process. With the enactment of the Treaty of Lisbon, elected members in the EU Parliament (EP) now have an enhanced role in trade policy, and they advocate for heightened focus towards having obligations for investors and partners, such as adhering to environmental and labour standards to maintain public support for trade (Orbie et al. 2016).

Following this shift, EU PTAs since 2015 have included more comprehensive labour provisions within a broader sustainable development chapter that is made legally binding using the dispute settlement mechanism. It also confers a pronounced role for civil society in the monitoring process (Orbie et al. 2016), by establishing Domestic Advisory Groups (DAGs) and Civil Society Dialogues. In 2022, the European Commission released a communication after its annual trade policy review that proposes new legal instruments that will inter alia include trade sanctions for material breaches of trade and sustainable development (TSD) provisions in PTAs (European Commission 2022).

One would be hard-pressed to categorically assert which of these two approaches – enforcement or managerial – is better. Consulting the literature, Hafner-Burton (2005) argues that PTAs with hard human rights standards can rely on coercion (instead of reputational effects) to influence their treaty members towards establishing and adhering to high human rights standards. On the other hand, Spilker and Böhmelt (2013) counter this by illustrating that countries sign PTAs with hard human rights standards only if they are certain they can comply. As such, there might exist an inherent selection bias in which only countries with a good human rights record would be willing to add a hard human rights clause, and so paradoxically, the inclusion of hard human rights standards would primarily occur in those cases in which they are hardly necessary.

The Comprehensive and Progressive Agreement for a Trans-Pacific Partnership (CPTPP) is one good example of the enforcement approach and has been lauded as a significant improvement to the approach to labour rights in PTAs. The TPP labour chapter had specific and well-defined labour rights in the main text and side agreements, cooperative labour dialogue, and the participation of the public and any interested or involved groups, including organised labour and business. However, it also had notable shortcomings outlined by politicians (e.g. Warren 2016), trade unions (e.g. LAC 2015), and international non-governmental organisations (INGOs) (e.g. ITUC 2015). In future PTAs, providing relevant civil society groups (NGOs, INGOs, labour unions) with a larger role in negotiations could help prevent such shortcomings while also improving the acceptability of labour provisions in the agreements. A potential model of how to define what constitutes acceptable (and improved) labour rights above former PTAs is to begin with the rights outlined in the ILO’s core conventions as a minimum and build country-specific rights obligations up from there.

However, as with the inclusion of labour clauses at the WTO, countries have vocally opposed the inclusion of labour provisions in PTAs. Developing countries are most likely to oppose the inclusion of labour rights in PTAs on the grounds that they are 1) disguised protectionism, 2) that they would penalise countries for their stage-of-development, or 3) that they represent the imposition of Western norms on unwilling parties, representing neocolonial attitudes. At its core, these countries fear loss of their comparative advantage in labour-intensive industries and argue that improvements in labour rights will follow from economic growth and development. And yet, other voices from the developing world, like Sujata Gothoskar of the Workers Solidarity Centre, argue that such a view means that ‘every struggle by the workers for a better life may be argued as eroding the competitive advantage of [their] country’ (cited in Hensman 2000 and Fields 2003). Export-related economic growth is often associated with increasing inequality, growing informality, and structural barriers that can impede widespread improvements in labour standards (Paz 2014; Das et al. 2017). Labour provisions in trade agreements arguably can help rectify market failures, and adhering to core labour standards promotes a framework for inclusive growth (Martin and Maskus 2001; ILO 2016). Balancing the protection of workers’ rights within global value chains (GVCs) and export-oriented sectors with developing countries’ desire to pursue economic growth and their own developmental path requires including local, national, and international actors in the negotiation and enforcement of labour provisions in PTAs.

The mixed results of labour conditionalities in PTAs have also fuelled the debate further. In a study comparing the effects of human rights agreements (HRA) and PTAs, Hafner-Burton (2005) found that when benefits of trade are conditional to human rights standards through legally binding and enforceable provisions, change in behaviour is more apparent. On the flip side, PTAs that use ‘soft’ standards that are only vaguely tied to market access reflect no difference in the actions of parties with regard to increasing human rights standards. On labour clauses specifically, Carrère et al. (2022) found that including such clauses in PTAs has on average no impact on bilateral trade flows. However, in North–South agreements, the impact of labour clauses is higher when there is more cooperation among parties. Carrère et al also found that strong enforcement mechanisms only reflect marginal improvements on compliance. Nonetheless, with the rise of PTAs, countries that prioritise countries that prioritise extensive labour rights have engaged on a bilateral basis to push for the inclusion of labour provisions in trade agreements.

In examining these patterns in past PTAs empirically, we can evaluate the instances where some approaches worked better than others. First, the coverage and level of ambition with regard to labour rights in PTAs exhibit a lack of consistency. Second, low legalisation – in the precision, obligations, and enforcement – of labour provisions presents as a problem in adhering to these standards in actualisation. As such, future PTAs should move towards more effective legalisation of labour provisions. Finally, on the enforcement of these labour provisions, the numbers are – at first glance – very disappointing: up until 2020, only two cases had been solved under the dispute settlement mechanisms of these PTAs, which implies that the impact of having labour provisions enforced under PTA dispute settlement is limited. Bearing in mind that enforcement does not exclusively include the formal settlement of disputes, the enforcement of labour conditions could improve if dispute settlement procedures were more accessible, and potentially more inclusive towards civil society groups and labour organisations. In fact, some provisions could have been enforced on a political level by these actors.

The lack of a stringent enforcement mechanism at the ILO was one of the main motives for including labour provisions in PTAs. Thus, PTAs and dispute settlement mechanisms might be an avenue to ensure labour provisions can be properly enforced.

13.3 Future Trends and Recommendations

Having mapped past practices and institutional interaction, the next section puts forward key recommendations on the design, enforcement, and compliance mechanisms of labour provisions in future PTAs.

13.3.1 Expansion of Labour Rights

Labour rights are often divided into two categories: collective and individual (sometimes called substantial) rights. The former refers to FACB rights, that is, the right of workers to act collectively and negotiate as a collective in their own best interest. The latter category of rights is less well-defined but overlaps to a large degree with the components of the ILO and UN’s concept of decent work, that is, formal work contracts, humane work hours, adequate earnings, occupational safety and health, non-remuneration benefits, and social security (ILO 2009). While unions may be viewed as a threat in developing countries for political (regime stability) and economic reasons (disruption of production and higher negotiated wages, etc.), allowing workers the right to organise does not imply higher labour costs, per se. Employers in developing countries can (and often do) refuse to negotiate in good faith with workers’ legitimate representatives and will even find ways to circumvent laws pertaining to FACB rights by, for example, installing ‘yellow’ or employer-friendly unions. Collective labour rights, therefore, represent a relatively inexpensive way of signalling labour upgrading to (potential or current) trading partners without losing price advantage as an exporter (Wang 2020; Messerschmidt and Janz 2023). Individual rights, on the other hand, can be very costly to implement. Upgrading workplace safety, reducing excessive hours or paying overtime, providing non-wage benefits, providing formal employment contracts, and paying into pension schemes all are associated with higher labour and therefore production costs which is perceived as a loss in competitiveness in export markets, especially for countries whose comparative advantage is in low-skill production.

For these reasons, labour provisions in trade agreements have primarily focused on enforcement of domestic laws (Raess and Sari 2018). Doing so avoids the projection of exogenous standards (i.e. from the Global North) on developing (or otherwise low-standard) countries and pre-empts opposition based on claims that trade-related labour provisions are little more than disguised protectionism. However, while substantive commitments to enforcing domestic laws are widespread, they are largely ineffective in improving the working conditions on the ground. Partner countries are unlikely to trigger dispute settlement regarding violations, and enforcement within a violating country is more likely to occur as a means of signalling suitability as a potential future partner in PTAs, rather than to stop abuses (Kolben 2013; Van Roozendaal 2015; Vogt 2015).

When labour provisions do mention protection of specific rights, they tend to focus on two things, both derived from the ILO 1998 Declaration: FACB rights and prohibitions on the worst forms of labour abuses, namely, child and forced (i.e. slave) labour. After this, several individual rights are specifically referenced in a steadily declining percentage of agreements as substantive commitments. Elimination of discrimination and equal remuneration are included in ~40% of PTAs, followed by health and safety (~30%), working conditions and wages (in ~20%), and decent work and working time, which can be found in only about 15% of agreements. Freedom of association and collective bargaining labour rights are frequently included in trade agreements because they are considered the cheapest option (i.e. they do not incur increased labour costs), second only to references to upholding domestic laws, and because proponents of their inclusion argue that FACB rights can lead to improvements in working conditions and individual labour rights. However, while signatory countries often improve their collective labour laws, they do little to improve de facto enforcement, leading to no improvement or sometimes even worsening conditions (Raess et al. 2018).

Expanding the types of labour rights in PTAs has the potential to lead to improvements in on-the-ground conditions that have remained elusive thus far, particularly when paired with the other advancements that we recommend. Inhumane working hours, unsafe working conditions, and informal employment are endemic in developing countries and in global supply chains, in particular (Beliner et al. 2015). So far, state-centred approaches in labour provisions have proven ineffective in alleviating these violations (Vogt 2015; Brown 2016). In tandem with targeted sanctions and empowerment of local civil society, more clearly defined and expanded labour rights would make improved working conditions a more attainable and enforceable goal (Brown 2016). We can also envision an expansion of protected rights to those of migrant workers, a group whose relevance continues to grow, as many regions in the world grapple with climate change and conflict-exacerbated migration flows. As an example, the draft TPP, which made many progressive steps regarding protection of labour rights, included ‘Labor Consistency Plans’ (LCPs) for some potential signatories. One of these was an LCP for Malaysia, which called on the country to stop withholding migrant workers’ passports and provide them with formal contracts (Dong 2016). Although the US ultimately withdrew from the TPP and it has since been replaced by the CPTPP, many of the advances made during the negotiations are being held up as an indication of shifting political will among parties as well as examples of what is possible when negotiating far-reaching trade agreements. Of course, defining the acceptable floor for labour rights in such a way as to avoid the perception (or reality) of disguised protectionism or projection of standards from one country to another will require involvement of not only state representatives, but relevant societal actors from each party. By involving labour representatives or other civil society actors, inclusion and enforcement of labour rights in PTAs can avoid sanctioning countries simply due to their stage-of-development (Fields 2003).

13.3.2 The ‘Manner Affecting Trade’ Issue

Another key issue observed in past dispute cases regarding labour rights violation is the ‘manner affecting trade’ term. According to Article 16.2.1(a) of the Dominican Republic–Central America FTA (CAFTA-DR), ‘A Party shall not fail to effectively enforce its labour laws, through a sustained or recurring course of action or inaction, in a manner affecting trade between the parties, after the date of entry into force of this Agreement.’ In order to establish a labour violation, two conditions must be met. First, the claimant must prove a persistent pattern of violations of labour laws. Second, this lack of enforcement must have affected trade between the Members of the PTA (Paiement 2018). This term is a key feature from the May 10 template from the US that appears in several agreements (Cimino-Isaacs 2020). To demonstrate the complications brought upon by this term, this section unpacks the dispute between the US and Guatemala over Article 16.2.1(a) of the CAFTA-DR agreement.

The CAFTA-DR includes a state-to-state mechanism that allows for a complaint of labour violations to be brought for dispute under the arbitration mechanisms of the agreement. In 2008, the US government raised a dispute under this mechanism against Guatemala, claiming that they violated commitments to enforce labour laws outlined in the CAFTA-DR PTA. This was after the American Federation of Labour and Congress of Industrial Organizations (AFL-CIO) and six Guatemalan labour unions filed complaints. Formal labour consultations between the two nations failed to yield results, and the United States Trade Representative (USTR) filed for arbitration under the CAFTA-DR dispute resolution chapter in 2011. Nine years after the complaint was submitted, the panel put forth its decision in 2017, finding that the US had not shown that Guatemala breached the agreement in relation to labour rights.

During the proceedings, both countries came forth to explain their interpretation of the language in the agreement. The US submitted that ‘in a manner affecting trade’ means that a violation ‘has a bearing on, influences or changes cross-border economic activity, including influencing conditions of competition within and among the CAFTA-DR Parties’. Conversely, Guatemala submitted that Article 16.2.1(a) requires an ‘unambiguous showing that the challenged conduct has an effect on trade between the Parties’.

However, the panel had differing interpretations. The panel interpreted the notion of ‘sustained or recurring course of action or inaction’ in Article 16.2.1 (a) of CAFTA-DR as either ‘(i) a repeated behaviour which displays sufficient similarity’ or ‘(ii) prolonged behaviour in which there is sufficient consistency in sustained acts or omissions as to constitute a line of connected behaviour by a labour law enforcement institution, rather than isolated or disconnected instances of action or inaction’. The panel acknowledged that ‘in a manner affecting trade between the Parties’ only applies if it provides a competitive edge to an employer or employers involved in trade between the parties. They explained further how this can be proven – by first demonstrating that the business is engaged in markets or competes with imports of CAFTA-DR Parties and identifying that a failure to enforce labour obligations is sufficient to cause ‘some competitive advantage’ on said enterprise.

The panel’s interpretation faced significant criticism. They conclusively determined that Guatemala did not uphold its labour laws according to Article 16.2.1(a) of the CAFTA-DR, as evident from violations affecting 74 workers at eight different locations. Though this appeared to be a straightforward win for the US, two key criteria remained: the persistence of the violations and their impact on trade. The panel firmly concluded that even though Guatemala consistently neglected its labour laws, this neglect did not impact trade. Additionally, in the single instance where trade was affected, the violation was neither sustained nor recurring. Essentially, according to the panel, one of the conditions of Article 16.2.1(a) was always unmet, leading them to rule in favour of Guatemala and against the US.

Given that this was the first labour enforcement case brought to dispute settlement under a PTA, the expectations on whether this mechanism can be evoked to protect workers’ rights were high. The fact that the panel did not find a breach was met with disappointment from stakeholders, including from the Office of US Trade Representative (USTR 2014). The results raised questions of whether labour chapters act as safeguards for workers in the way they are currently formulated (Claussen 2020).

Hence, the main recommendation of this section regarding ‘manner affecting trade’ is that this term should be rephrased with unambiguous language, encompassing strong legal obligations and sharper precision (Abbott et al. 2000). We suggest this change on two grounds. First, the term ‘manner affecting trade’ should be phrased in a clearer way that does not allow for narrow interpretations by arbitration panels that primarily adopt a perspective of trade considerations as opposed to worker’s rights, and instead be inclusive to the voices and submissions of civil society. Second, labour rights violations should not only be considered when they are predicated upon whether the violation affects trade. The central goal of including labour provisions should be to protect workers’ rights in traded sectors, not to protect workers’ rights only if labour law violations distort competition in ways that negatively affect trade. Moreover, the combined conditionality of ‘sustained or recurring course of action’ sets a high requirement that worker groups and unions might not be able to prove or reach. Overall, these recommendations can prevent possibilities of interpretations that have adverse effects on the welfare of the workers and instead prioritise labour rights-based considerations.

13.3.3 Enforcement Mechanisms

A main criticism of labour provisions in PTAs is that even if they are included in the agreement, enforcement is weak. Using the USMCA’s enforcement mechanisms as the current gold standard, we make three broad recommendations on how enforcement can be strengthened in future PTAs to fully unlock the potential of these provisions that have been painstakingly negotiated and advocated for.

13.3.3.1 USMCA’s Rapid Response Mechanism

When contemplating novel enforcement mechanisms in PTAs for the future, parties ought to look at the USMCA, since its enforcement mechanisms have been praised as a ‘significant innovation’ (Corvaglia 2021: 664) and as ‘forging a path toward a third generation’ (Polaski et al. 2022: 149) of enforcement possibilities for labour provisions. While the USMCA will serve as the baseline, we will provide additional ideas and concepts regarding the enforcement of labour provisions in PTAs.

The enforcement regime of the USMCA’s labour chapter consists of three main pillars: first, there is the pre-ratification conditionality regarding worker representation in collective bargaining in Mexico (Annex 23-A USMCA). It focuses on domestic provision of collective bargaining rights in Mexico and the establishment of independent entities and courts for the registration of unions and the adjudication of labour disputes. To ensure compliance with these provisions, the entry into force of the entire PTA was made dependent on the implementation of such legislation by Mexico (Annex 23.A (3.) USMCA). The US has regularly made use of its bargaining leverage to push for labour reforms by inserting such preconditions into its PTAs. The second pillar addresses the effectiveness of collective bargaining rights in Mexico after the ratification of the PTA by establishing an innovative adjudicatory mechanism called ‘Facility-Specific Rapid Response Labour Mechanism’ (hereinafter: ‘RRM’). The mechanism applies whenever a party believes that ‘workers at a Covered Facility are being denied the right of free association and collective bargaining’ (Article 31-A.2 USMCA). Thus, while the parties to these disputes are still the respective countries, it is specific to a particular denial of rights at a specific facility. After an on-site review by the respondent party of whether a denial of rights exists, the complainant party may request the establishment of a panel if an agreement on the matter cannot be reached. The short duration of the procedure is where a first advantage of the RRM lies: it can be expected to last only 120 days from the initial request for a review to the implementation of a remedy (Claussen 2021: 356). In all five occasions in which the RRM has been used so far, an agreement was found after around two months without the need to establish a panel. The second major innovation can be seen in the facility specificity of not only the investigation of the denial of rights, but also the remedies provided by Article 31-A.10 (2.) USMCA. These remedies include the suspension of preferential tariff treatment or the imposition of penalties specifically against the goods from a particular factory found guilty of the denial of rights. Third, in addition to the settlement of disputes concerning the denial of rights via the RRM, the USMCA also features a traditional dispute settlement mechanism under which all labour disputes not concerning the denial of collective bargaining rights are adjudicated (Article 31.2 USMCA).

Clearly, this approach also has a few drawbacks of its own. First, there is the issue of extraterritorial application and enforcement of domestic law, which results in the loss of sovereignty for the contracting party; we address this issue in our recommendation for the dispute settlement mechanism below. Second, it is evident that the further apart working conditions and domestic labour laws are for the two (or more) countries under a PTA, the more difficult it will be to implement such a system: for example, the interpretation of collective labour rights is a concept very specific to Western countries. The implementation of it might work well between the US and Mexico. In a PTA between two countries of the Global South, this might not be as easy. However, we consider the design of the enforcement mechanisms in the USMCA a step in the right direction, especially given that the RRM has been a success, with five cases already having been resolved in the two and a half years since its entry into force. Inspired by this, we suggest that enforcement in future PTAs should focus on the three aspects explained in more detail below.

13.3.3.2 Targeted Sanctions

One of the main innovations of the RRM is the implementation of sanctions targeted at specific worksites. We consider this facility specificity a step in the right direction and a necessity for future PTAs for two different reasons. First, it increases accountability on a company level. Companies that do not comply with labour rights will see their goods being charged additionally and hence sustain a disadvantage with respect to their law-abiding competitors. This should serve as a deterrent against violations. Second, since preferential tariff rates keep being applied on goods from other facilities, trade flows are not hampered the same way they would be if these higher tariff rates were applied to an entire sector or even additional sectors as well. With the negative effect on the most vulnerable stakeholders being the main point of criticism against the traditional sanction mechanisms (Orbie 2021: 200), targeted sanctions would be much more effective. However, there are certain aspects that need to be considered when designing targeted sanctions for future PTAs. First, such enforcement schemes would have to make sure that companies in the territory of all parties face enforcement in case of a denial of rights at their facility. Under the USMCA, limiting the scope of possible action under the RRM concerning facilities in the US (LeClercq 2021) has led to a very one-sided application so far. Second, due process rights for affected companies must be guaranteed. This is not an easy undertaking, since a lot of cases are solved on a political level before the establishment of an independent panel. However, not only panel decisions have to be open for judicial review, but there is also a need to give companies the possibility to challenge the legality of the actions of state actors in the context of their accusations.

13.3.3.3 Private Arbitration

A concept that has already been proposed by other scholars but has faced criticism due to political feasibility concerns (Claussen 2021: 366) is the establishment of mixed arbitration for dispute settlement concerning labour rights issues. The idea would be to have an adjudicative body comparable to an investor–state arbitration tribunal where trade unions, NGOs, or even individuals could make claims regarding the violation of labour rights in the PTA. Such an advancement would be paramount for the enforcement potential of labour chapters, as various scholars have attributed their underperformance to the lack of non-governmental actors involved (Gött 2020: 151; Dombois et al. 2004: 267, Kolben 2017). An independent panel made up of labour experts could review the claim and issue a decision, including its reasoning. With more cases being brought forward by trade unions and NGOs, panels could – taking into consideration the input of the ILO and its experts – develop lines of jurisprudence and thus breathe life into the often rather vaguely phrased labour provisions in PTAs. Admittedly, this would mark a huge shift in the approach taken towards enforcing PTA provisions and countries are wary of giving away even the slightest degree of sovereignty when it comes to PTAs. This could be countered by leaving the enforcement action in the hands of the parties. If a company violating labour rights were to receive a monetary fine instead of the suspension of tariff benefits, the party where said company is located would be competent to enforce this fine. This approach would include all the above-mentioned benefits while leaving the ultimate say to the parties of the PTA. Moreover, it could foster the legitimacy of PTAs and increase public approval, which in times of increased scepticism towards free trade and calls for deglobalisation could become a good hard to come by. Another possible solution to assuage fears over loss of sovereignty is to implement the so-called pyramid model proposed by Kolben (2007: 248). In such a model, an overall council organises numerous local-level councils built of stakeholders who can shape how the standards in a labour provision are implemented and can monitor compliance, making information available regarding plants’ compliance (or lack thereof). This model would function similarly to Better Factories in Cambodia, which has received positive feedback regarding its effectiveness at improving working conditions as a result of included and empowered local actors.

13.3.3.4 Certification System

An idea we consider worth exploring is the inclusion of a certification system for factories and companies that adhere to certain labour standards. For that matter, we take inspiration from the European Free Trade Association (EFTA)–Indonesia Comprehensive Economic Partnership Agreement (CEPA): in its Annex V, Switzerland limits preferential tariff treatment for palm oil conditional on meeting the sustainability objectives as set out in Article 8.10 in the main text of the PTA, which requires parties inter alia to ‘ensure that [palm oil] traded between the Parties [is] produced in accordance with the sustainability objectives referred to in subparagraph (a)’. Switzerland implemented this provision by enacting the so-called palm oil regulation, pursuant to which only palm oil certified under an approved certification system (Article 3 of the palm oil regulation) enjoys preferential tariff treatment. If this is possible for certified palm oil, why would it not be possible for goods produced at a certified worksite? Admittedly, the search for an effective common certification standard might pose its challenges. However, given such certification systems might become a standard in future PTAs, even the involvement of the ILO together with existing private initiatives is conceivable. As for the compatibility of such preferential tariff schemes in PTAs with WTO law, research suggests that it is not whether tariff differentiation based exclusively on production and process methods (PPMs) violated WTO law per se, but rather how such a scheme is designed (Bürgi Bonanomi 2021: 373).

13.4 Conclusions

To conclude, there is no ‘one size fits all’ in terms of labour provisions. Countries must consider domestic capacities of their partner countries and widen the engagement of stakeholders when conceiving a labour chapter in a PTA. There currently exist some templates of labour chapters, which we loosely classify into managerial and enforcement approaches. The first part of this chapter has outlined the successes and failures of both these models, and the latter part includes some key recommendations on the design, enforcement, and compliance of labour provisions for the future.

On the design of labour provisions in future PTAs, we generally recommend continuing the current practice of using internationally recognised standards by the ILO. However, keeping in mind the domestic capacity gap of certain developing countries, we recommend that standards be determined on a case-by-case basis, which encourages flexibility in adopting some domestic standards as well, to account for the particularities of certain countries. This is also a good way to avoid sentiments of labour provisions being viewed as neocolonial by developing countries, especially if done in a consultative manner between negotiating parties, with stakeholder representation including unions, activists, and members of the civil society. Considerations regarding domestic capacity and stage-of-development must be balanced with the ultimate goal of mediating labour abuses and accounting for the preferences of labour and civil society groups. Expanding the types of labour rights incorporated in PTAs can help ensure that trade-related economic development is sustainable and equitable and that labour provisions are more effective.

Additionally, we recommend reforming the language regarding ‘manner affecting trade’, which is prevalent, especially in US-led labour chapters. There is a need to adapt new templates and language that embody higher levels of legalisation in the case of disputes. This will confine room for interpretation and make disputes fairly straightforward to adjudicate. Furthermore, future labour provisions in PTAs need to eschew this embeddedness of labour violations being considered only when they affect a traded sector. To maintain high standards in labour, worker’s rights should be a standalone issue, regardless of whether violations substantially impact trade flows or not.

As for enforcement mechanisms, we think that the RRM in the USMCA displays a substantial step forward and will be copied into future PTAs. However, there needs to be higher ambition on the enforcement front. Whereas trade sanctions targeted at specific facilities limit the disturbance of trade flows and increase accountability for firms, we also pointed out several flaws of the RRM that will need to be addressed in future PTA negotiations. The inclusion of private arbitration mechanisms providing for a direct access for civil society organisations to dispute settlement, as well as the implementation of a certification system could help enhance the effectiveness of future labour chapters.