Free trade v freedom of association? The EU/South Korea free trade agreement and the panel report on the EU challenge to South Korean labour law

 

 

Steve Peers, Professor of Law, University of Essex

Has international trade liberalisation, in the form of the World Trade Organisation and free trade agreements between its Members, aided economic growth generally, or rather increased gaps between the rich and poor? Should the pursuit of freer trade take into account concerns about labour standards (and parallel concerns about environmental protection) – and if so, how?

Concerns along these lines have become increasingly prevalent in debates over trade policy, leading in some countries to reluctance to negotiate new free trade deals or demands to renegotiate existing treaties. For its part, the EU has moved towards more assertively applying the provisions relating to trade, labour and the environment in its free trade agreements. 

The EU’s enforcement strategy has recently had its first concrete outcome, in the form of the first panel ruling concerning labour standards under an EU free trade agreement. This concerned freedom of association under the EU/South Korea free trade deal. Note that the EU is recently more active in enforcing other aspects of its bilateral FTAs too: see the recent panel ruling in a dispute brought by the EU against Ukraine (regarding a Ukrainian export ban on wood). Dispute settlement proceedings against South Africa (regarding poultry) and Algeria (regarding import restrictions) are also underway.

This blog post summarises the relevant law in the free trade agreement and its interpretation by the panel, then comments on the ruling in the context of the broader debate about the relationship between trade and labour standards. Finally, I discuss how the ruling might be relevant by analogy to disputes about trade and the environment, to the controversial new EU/China investment agreement and to EU/UK disputes under the new post-Brexit trade deal (see further my summary of that deal and discussion of dispute settlement and human rights, as well as Markus Gehring’s analysis of the environmental rules in the deal).

 

The trade and sustainable development provisions of the free trade agreement

The term ‘level playing field’ is commonly used to refer to these issues (and some other issues) in the recent UK/EU treaty, but the phrase ‘trade and sustainable development’ is used in the EU/S Korea treaty (and more commonly by trade lawyers generally). The EU/S Korea rules are similar to those in a number of other recent FTAs with the EU, and so the panel report may well be relevant by analogy to potential disputes between the EU and (some) other non-EU countries. They are also broadly similar to the recently agreed EU/China investment agreement (not yet in force). However, as discussed further below, the ‘Brexit deal’ provisions go further.

Also, it should be noted that ‘trade and labour’ rules are not unique to agreements with the EU. For instance, free trade agreements with the US contain some provisions on the issue. Indeed, as also discussed below, the panel report distinguishes itself from the ruling in a case between the US and Guatemala.

Chapter 13 of the FTA, the ‘trade and sustainable development’ Chapter, first defines its ‘context and objectives’: the parties ‘reaffirm their commitments to promoting the development of international trade in such a way as to contribute to the objective of sustainable development’ (Article 13.1.1). They ‘recognise that economic development, social development and environmental protection are interdependent and mutually reinforcing components of sustainable development’ (Article 13.1.2). Also, they ‘recognise that it is not their intention in this Chapter to harmonise the labour or environment standards of the Parties, but to strengthen their trade relations and cooperation in ways that promote sustainable development’.

The scope of the Chapter is then defined (Article 13.2.1):

Except as otherwise provided in this Chapter, this Chapter applies to measures adopted or maintained by the Parties affecting trade-related aspects of labour and environmental issues in the context of Articles 13.1.1 and 13.1.2.

Having linked trade with sustainable development, Article 13.2.2 then pushes back:

The Parties stress that environmental and labour standards should not be used for protectionist trade purposes. The Parties note that their comparative advantage should in no way be called into question.

The mixed message continues in Article 13.3, titled ‘right to regulate’, which first refers to ‘the right of each Party to establish its own levels of environmental and labour protection, and to adopt or modify accordingly its relevant laws and policies’, but then qualifies that immediately by stating that ‘each Party shall seek to ensure that those laws and policies provide for and encourage high levels of environmental and labour protection, consistent with the internationally recognised standards or agreements referred to in Articles 13.4 and 13.5, and shall strive to continue to improve those laws and policies.’

What are those international standards exactly? Article 13.4 defines the international labour standards:

1. The Parties recognise the value of international cooperation and agreements on employment and labour affairs as a response of the international community to economic, employment and social challenges and opportunities resulting from globalisation. They commit to consulting and cooperating as appropriate on trade-related labour and employment issues of mutual interest. 

2. The Parties reaffirm the commitment, under the 2006 Ministerial Declaration of the UN Economic and Social Council on Full Employment and Decent Work, to recognising full and productive employment and decent work for all as a key element of sustainable development for all countries and as a priority objective of international cooperation and to promoting the development of international trade in a way that is conducive to full and productive employment and decent work for all, including men, women and young people.

3. The Parties, in accordance with the obligations deriving from membership of the ILO and the ILO Declaration on Fundamental Principles and Rights at Work and its Follow-up, adopted by the International Labour Conference at its 86th Session in 1998, commit to respecting, promoting and realising, in their laws and practices, the principles concerning the fundamental rights, namely:

(a) freedom of association and the effective recognition of the right to collective bargaining;

(b) the elimination of all forms of forced or compulsory labour;

(c) the effective abolition of child labour; and

(d) the elimination of discrimination in respect of employment and occupation.

The Parties reaffirm the commitment to effectively implementing the ILO Conventions that Korea and the Member States of the European Union have ratified respectively. The Parties will make continued and sustained efforts towards ratifying the fundamental ILO Conventions as well as the other Conventions that are classified as ‘up-to-date’ by the ILO.

Article 13.5 then defines international environmental standards. I’ve discussed the application of this ruling to them by analogy separately below.

Article 13.6 is a general statement about the links between trade and sustainable development. Article 13.7 then sets out a rule about ‘upholding levels of protection’, ie reducing standards or failing to enforce domestic labour law.

Skipping over a number of general provisions, the Chapter ends with rules on dispute settlement. Article 13.14 first calls for consultation between the parties on a ‘matter…arising under’ the Chapter. If these do not lead to a settlement within 90 days, a party can call for a panel of experts to be set up (Article 13.15). The panel is to make recommendations, and the parties ‘shall make their best efforts to accommodate advice or recommendations of the Panel of Experts on the implementation of this Chapter.’ This replaces the main dispute settlement system of the FTA, which is switched off for any ‘matter arising under’ this chapter (Article 13.16).

The panel report

The EU’s argument was based on two separate parts of Article 13.4.3 of the FTA, which was set out in full above. First, South Korea’s domestic labour law had not fully complied with ILO obligations as regards freedom of association. Second, South Korea had not made sufficient efforts towards ratifying core ILO treaties. 

Rules of interpretation

The report starts out by asserting that the level playing field provisions must be interpreted in accordance with the normal rules of interpretation of international law, in Articles 31 and 32 of the Vienna Convention of the Law of Treaties (VCLT). This followed from the obligation in the FTA for dispute settlement arbitrators to apply those rules; and the panel of experts was acting in lieu of arbitration.

Jurisdiction

Next, the panel turned to its jurisdiction to hear the case. Jurisdiction can be a dry technical issue, but here the dispute went to the core of the trade/labour nexus.  South Korea argued that the dispute was not ‘a matter arising under’ the Chapter, so could not be subject to dispute settlement at all, because Article 13.2.1 limited the Chapter to measures by a party ‘affecting trade-related aspects of labour and environmental issues’ (emphasis added). There was no direct impact on trade here, therefore no matter could arise under the Chapter.

If this argument had been accepted, it would have ended not only this particular dispute but also (if the ruling had been taken as a precedent) potentially many other disputes under this and other EU FTAs (as well as the EU/China investment treaty).  However, the panel did not accept it. It pointed instead to the words at the start of Article 13.2.1: ‘Except as otherwise provided in this Chapter…’, which indicated that there were exceptions to the rule that the Chapter only related to trade-related aspects. Article 13.4.3 was one of those exceptions. This was confirmed by the broad references to ILO obligations, the ‘fundamental rights’ aspects of those obligations, the reference to the abolition of ‘all’ compulsory labour, and the impossibility of ratifying ILO Conventions only as regards a limited category of workers. It was further confirmed by a contrario comparison with Article 13.4.1 and 13.4.2, which refer to trade (as to Articles 13.7 to 13.9), and by interpreting the scope of the Chapter in light of the broad context and objectives set out in Article 13.1 and the preamble to the FTA.

South Korea also argued that the EU aimed to harmonise the labour law of the parties, in breach of Article 13.2.2. The panel did not accept this either. Harmonisation meant ‘alignment of actual standards such as minimum rates of pay, maximum hours of work, or access to job security procedures’ as distinct from ‘fundamental principles and rights and core labour standards’, which ‘do not require harmonisation of domestic labour laws or outcomes’. The proof of this was that ‘many of the member States which have ratified the relevant Conventions both comply with their international obligations and maintain disparate systems of industrial relations, with very different substantive outcomes in terms of levels of economic development’. The panel also noted that the domestic right to regulate in Article 13.3 was subject to the obligation to uphold core standards.

Likewise, the panel rejected the argument that the EU aimed at protectionism, referring to ILO research that asserts no link between competitiveness and raising labour standards. With respect, a better argument here might be that the EU was not arguing for a protectionist measure (ie trade retaliation), if only because of the limits on remedies set out in the FTA itself. If there is no link between lower labour standards and trade, why does this Chapter often make that link in the first place? The panel’s interpretation is a hostage to fortune because it might make it harder for the EU to bring a claim in a different case relating to reduction or non-enforcement of labour standards, where it would have to show an effect on trade or investment to win the case.

On that point, next, the panel rejected any comparison with the ruling in the US/Guatemala dispute, because the rule in the FTA in question was a requirement that 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…’. While this was nearly identical to Article 13.7.1 of the EU/S Korea FTA (‘(a) Party shall not fail to effectively enforce its environmental and labour laws, through a sustained or recurring course of action or inaction, in a manner affecting trade or investment between the Parties’ – though note the extension to include environmental law, and an effect on investment), the EU was not alleging a breach of Article 13.7, ie a failure to enforce South Korea’s labour law. It was  instead alleging a breach of the separate obligation for that domestic labour law to meet minimum standards. More broadly, the context of the two treaties is different: the US/Central America/Dominican Republic treaty ‘does not have the same contextual setting of sustainable development as the EU-Korea FTA, nor does it refer to the range of multilateral and international agreements and declarations which the Parties have included in the EU-Korea FTA’.

Substance: domestic labour law

The panel then turned to the EU’s arguments about South Korean labour law, which were fourfold: a narrow definition of ‘workers’ who can join trade unions (excluding self-employed or dismissed or unemployed workers); the definition of ‘trade union’ as excluding bodies who admit persons excluded from the category of ‘worker’ as members; the requirement that only members can be trade union officials; and the discretionary registration of trade unions.

It started with looking at the different elements of Article 13.4.3. As regards the requirement to act ‘in accordance with the obligations deriving from membership of the ILO’, the panel found that it included the obligation to ensure freedom of association, even where (like South Korea) a country had not signed up to specific treaties on that issue. Moreover, the rulings of the ILO’s Committee on Freedom of Association could be taken into account.

As for ‘the ILO Declaration on Fundamental Principles and Rights at Work and its Follow-up’, the panel decided that the EU was not claiming that it was legally binding as such. What was legally binding between the parties was Article 13.4.3, taken as a whole.

Next, the panel considered the legal strength of the term ‘commit to’. Here the panel rejected South Korea’s argument that this wording was purely aspirational, ruling instead that ‘this represents a legally binding obligation of commitment to respecting, promoting and realising the obligations arising from membership of the ILO and the 1998 ILO Declaration in relation to the principles concerning the fundamental rights’.

It then turned to those specific terms. ‘Respecting’ meant a ‘negative obligation not to injure, harm, insult, interfere with or interrupt freedom of association’. ‘Promoting’ meant a ‘positive obligation on States, which in human rights statements on the content of the freedom of association means that States should ensure third parties do not disrupt workers engaging in their right to freedom of association’. Also, ‘States should create a ‘climate’ in which the civil rights of workers and employers allow them to freely exercise their rights to freedom of association’. As for ‘realising’, this meant ‘a binding requirement involving a commitment to realising the principles concerning the fundamental right of freedom of association’ as distinct from ‘a binding requirement that a Party comply with the terms of Conventions’ as such – which was the subject of a separate complaint, considered below.

Finally, ‘the principles concerning the fundamental rights’ did not mean that there was no legally binding obligation, and the panel was unconvinced by the South Korean argument that these principles were not sufficiently clear.  It did, however, hint that the freedom of association principles were clearer than the others referred to (forced labour, child labour and discrimination), raising the prospect that a case concerning those principles might face an extra hurdle to be successful.

Applying these principles to the EU’s specific complaints, first of all South Korea had wrongly excluded self-employed or dismissed or unemployed workers from joining trade unions. On the first point it noted that this issue arose under some EU Member States’ law too – noting that this might be an issue for discussion between the parties going forward. On dismissed workers, it noted in particular that there were insufficient safeguards against being dismissed for trade union-related activity.

Next, the panel concluded that South Korea had wrongly defined ‘trade unions’ as excluding bodies who admit persons excluded from the category of ‘worker’ as members, noting in particular that some trade unions had been deregistered because their members had been dismissed. Needless to say, this gives employers considerable power over not only individual employees but also entire trade unions by threatening dismissal of a union member. It also ruled against the national law requirement that only union members can be trade union officials, on the basis that ti was up to union members to choose who to represent them.

On the other hand, the panel ruled against the EU complaint regarding the discretionary registration of trade unions, holding that the evidence was contradictory and it was not sufficiently certain that the complaint was well-founded.

Substance: ratification of ILO Conventions

As a reminder, treaty provides that South Korea ‘will make continued and sustained efforts towards ratifying the fundamental ILO Conventions as well as the other Conventions that are classified as “up-to-date” by the ILO’. The EU argument related only to the ‘fundamental’ Conventions.

The panel started out by noting that of the eight ‘fundamental’ ILO Conventions, South Korea had yet to ratify four: two on forced labour and two on trade unions. The first issue here was the legal strength of the word ‘will’: the panel rejected South Korea’s argument that it was not really binding, holding that it ‘establishes a binding legal obligation’.

But a legal obligation to do what exactly? Here the panel rejected both South Korea’s argument that the status quo was sufficient, and the EU’s argument that progress towards ratification must continue ‘without interruption’. There was no concrete requirement or target date, and the parties had ‘leeway’ in achieving the objectives. This was an obligation of effort, not result.

Was there sufficient effort then? The panel ruled that it was sufficient that the South Korean government had tabled bills before Parliament in 2019 to ratify three of the treaties concerned. As for the other treaty, concerning prison labour, the panel accepted South Korea’s argument that changes to penal law take time. (One might note that they take very little time when a State responds to a terrorist bomb or other outrage). Overall, while expressing mild disappointment, the panel did not think that South Korea had breached its obligation to make effort towards ratifying the Conventions.

What happens when and if these treaties are ratified? That’s a hypothetical issue, and the panel was not called upon to address it. But it should be recalled that Article 13.4.3 also provides for a ‘commitment to effectively implementing the ILO Conventions that Korea and the Member States of the European Union have ratified respectively’. This would apply already, of course, to ILO Conventions which both sides have ratified. Logically much of the panel’s analysis would apply by analogy to this provision: it falls within its jurisdiction; there is no requirement to show a link to trade or investment; and the word ‘commitment’ entails a legally binding obligation. But here the wording is stronger than the obligation to make ‘efforts towards ratifying’ treaties: ‘effectively implementing’ them arguably suggests rather an obligation of result, going beyond halting and partial ‘efforts’ to do so.   

 

Comments

The EU has won some significant victories here, as regards: jurisdiction; the legal effect of some vague quasi-soft law terms; the absence of a requirement for a trade or investment impact; the references to some ILO Conventions; and the use of ILO ‘soft law’. All of this sends a signal to other non-EU countries, should the EU seek to invoke the relevant provisions (where they exist) in other FTAs or the EU/China investment treaty. Of course, the obligations also apply to the EU – as the panel points out at several places – although labour unions and environmental groups would likely be happy to see the EU held to account as well.

The EU has been successful also on some key points of criticism of domestic labour law. The panel’s close attention to the effect of labour law rules in practice – ie as a means to decertify trade unions simply by firing a member – is significant. On the point where the EU failed to table enough evidence to convince the panel, that is a pointer for how such disputes might be litigated in future.

The EU will likely be disappointed by the panel’s approach to ILO Conventions, where it held that they created a legal obligation but applied a relatively low threshold to assessing South Korea’s compliance with the obligation to move towards ratifying them. As noted above, some aspects of its reasoning (as regards the prison labour Convention, and the ‘disguised protectionism’ point) are particularly unconvincing, with great respect. This sends a converse message to non-EU countries: that a delay of nearly a decade in ratifying such Conventions, including extra tardiness in ratifying one important Convention, is acceptable. However, it should be noted that as discussed above, the obligations to implement ILO Conventions after ratification, while not discussed by this panel report, are arguably stronger.

Of course, the EU’s partial success in its action is inevitably limited by the lack of any further remedies to enforce the panel ruling. It might be possible that the process has some effect on domestic political opinion in the other party, perhaps helping to persuade the government to move faster on the relevant issues. However, considerations like these are only relevant where there is a form of democracy in the other party – so they are hardly relevant in the context of the EU/China investment agreement.  

The context of the panel ruling includes the gradual development of the EU’s own trade policy. A revision of internal EU law on trade remedies (which does not apply to investment treaties) is about to be formally adopted, and will be accompanied by a statement from the Commission, including the following commitment to enforcement of sustainable development provisions in the EU’s FTAs:

In deploying the enhanced enforcement system [regarding alleged breaches of trade agreements], the Commission will pay equal attention to alleged breaches of the trade and sustainable development provisions of EU trade agreements as to alleged breaches of market access systems. The processing of alleged breaches of trade and sustainable provisions will be fully integrated into the system. The Commission will prioritise those cases which are particularly serious in terms of their effect on workers or the environment in a trade context, which have systemic importance and which are legally sound.  

In the same context, the panel ruling is comparable to the 2017 CJEU judgment on the EU’s legal competence to agree the EU/Singapore free trade agreement, which decided (at paras 139 to 167) that the ‘sustainable development’ provisions fell within the scope of the EU’s common commercial (ie trade) policy. Having said that, it is striking that the CJEU took a different view from the panel as to whether lower labour standards might have an effect on trade, ruling that the sustainable development provisions:

are such as to have direct and immediate effects on trade between the European Union and the Republic of Singapore since they reduce the risk of major disparities between the costs of producing goods and supplying services in the European Union, on the one hand, and Singapore, on the other, and thus contribute to the participation of EU entrepreneurs and entrepreneurs of the Republic of Singapore in free trade on an equal footing (para 159).

There’s a context beyond the EU as well: while the new US President will not be tweeting angrily in the middle of the night about trade issues, the Biden administration is nevertheless lukewarm about further trade liberalisation. Whether this means ruling out trade deals entirely, or insisting on more account being taken of labour and environmental standards in such deals, remains to be seen.  

More broadly, concern in wealthier countries about opening up (or retaining) free trade without more enforceable labour and environmental standards may have reached a tipping point. One might draw comparisons with the EU’s own moves to adopt more labour and environmental law in the context of completing the internal market. Free trade’s hardcore advocates have long resisted making a strong link between trade and labour or environmental standards. They may now face a choice not necessarily between socialism or barbarism, but at least between legalism or nativism.

 

Application to environmental law

As noted above, some parts of the panel ruling are relevant to environmental disputes by analogy – in particular to the equivalent provisions on multilateral environmental treaties in Article 13.5:

2. The Parties reaffirm their commitments to the effective implementation in their laws and practices of the multilateral environmental agreements to which they are party.

3. The Parties reaffirm their commitment to reaching the ultimate objective of the United Nations Framework Convention on Climate Change and its Kyoto Protocol. They commit to cooperating on the development of the future international climate change framework in accordance with the Bali Action Plan.

First of all, interpretation in accordance with the VCLT is equally relevant to environmental issues. The panel’s analysis of jurisdiction applies also to environmental disputes by analogy:  Articles 13.5.2 and 13.5.3 are equally exceptions to the rule that the Chapter only covers trade-related aspects of the environment, given that there is no reference to trade only here.

As with the labour provisions, one can interpret the scope of those rules a contrario by comparison with Article 13.5.1 (which does mention trade) and Articles 13.7 to 13.9, and in light of the broad context of Article 13.1 and the preamble. The indivisibility of ILO Conventions applies equally to environmental treaties. The limited scope of the ban on harmonisation of law, the distinction between the right to regulate and the obligation to uphold core standards, and the absence of a requirement to show an effect on trade applies equally to the clause on multilateral environmental treaties.

By analogy with the panel ruling, the words ‘commit’ and ‘commitment’ as regards environmental treaties – appearing three times here – denote a legally binding obligation. A commitment to cooperation is arguably too vague to define in concrete terms; but on the other hand, a ‘commitment to reaching the ultimate objective’ of a treaty is a stronger obligation than ‘continued and sustained efforts towards ratifying’ the ILO Conventions at issue in part of the labour dispute.

The strongest obligation here may be the obligation ‘to the effective implementation in their laws and practices of the multilateral environmental agreements to which they are party’. This closely parallels the commitment in the labour provisions ‘to effectively implementing the ILO Conventions that Korea and the Member States of the European Union have ratified respectively’ – discussed further above.

Just as the panel report may be relevant by analogy to the labour provisions in other EU FTAs and the EU/China investment treaty, this extension of its reasoning could be relevant to the environmental provisions of those FTAs and the EU/China treaty too.

Application to the EU/China investment treaty

Comparing the EU/S Korea FTA with the EU/China investment treaty, the latter (understandably) refers only to investment aspects throughout, rather than trade also. However, it does not limit its scope in the same way as the EU/S Korea treaty, so South Korea’s jurisdictional objections (which were unsuccessful anyway) would not be applicable. The two treaties have a similar clause on their context, and comparable provisions on environmental treaties.

As regards labour standards, the non-regression and non-enforcement clauses are similar to the EU/S Korea treaty. The provision on ILO standards and domestic law, and ratification of ILO Conventions, provides:

1. Each Party, in accordance with its obligations assumed as a member of the International Labor Organization (“ILO”), and its commitments under the ILO Declaration on Fundamental Principles and Rights at Work and its Follow-up, shall respect, promote and realize, in good faith and in accordance with the ILO Constitution, the principles concerning the fundamental rights which are the subject of the fundamental ILO Conventions.

2. Each Party is, in accordance with the commitments of the members of the ILO and the 2019 ILO Centenary Declaration for the Future of Work, committed to effectively implement the ILO Conventions it has ratified and work towards the ratification of the ILO fundamental Conventions. In particular, in this regard, each Party shall make continued and sustained efforts on its own initiative to pursue ratification of the fundamental ILO Conventions No 29 and 105, if it has not yet ratified them. The Parties will also consider the ratification of the other Conventions that are classified as "up to date" by the ILO.

It is notable that this wording leaves out the list of fundamental rights in question, thereby omitting to mention forced labour explicitly. However, it is nevertheless covered by the general reference to ILO principles, in light of the panel interpretation.

The wording is basically similar to the clause in the EU/S Korea treaty which the panel interpreted, and so should logically be interpreted the same way, particularly in light of the use of the stronger word ‘shall’ at two points. Remember that the panel interpreted the word ‘committed’ (here referring to effective application of ILO Conventions once ratified) as legally binding.

As with the EU/S Korea FTA, the weak message on ratification of ILO Conventions (in this case, two forced labour Conventions) is arguably compensated for by the stronger wording on the domestic implementation of ILO principles. But this is subject to the important caveat above – that the recent panel ruling suggests that freedom of association principles might be easier to identify than the other ILO principles being referred to. The particular risk here is that the crucial issue of forced labour could not be effectively addressed. Whether this is sufficient will likely be subject to much debate in the near future. 

Finally, the dispute settlement system is essentially the same as for the EU/S Korea FTA, although the parties merely ‘discuss measures’ following any panel ruling.

Application to the EU/UK treaty

Comparing the EU/S Korea FTA with the EU/UK deal (chapter 8 of the ‘level playing field’ rules), the list of international measures includes some later measures in the latter treaty, but the references to integrating sustainable development into the EU/UK relationship are shorter. Conversely, the EU/UK treaty makes more references to transparency. The wording relating to ILO Conventions is only slightly different, bringing together the rules on promoting and effectively implementing such Conventions: ‘each Party commits to respecting, promoting and effectively implementing the internationally recognised core labour standards, as defined in the fundamental ILO Conventions’.

But there are additional commitments to: implementing the parts of the Council of Europe Social Charter that each party has signed up to; promoting the ILO Decent Work Agenda; protecting the social dialogue; and cooperating in multilateral fora on ‘trade-related’ labour issues.  There is a much longer list of environmental obligations as compared to the EU/S Korea FTA.  

Crucially, there is no general clause limiting the scope to ‘trade-related’ matters; such wording appears only in a handful of provisions of chapter 8. There is no rule against harmonisation or protectionism either.  So the various jurisdictional objections raised by South Korea could not so easily be made; and in any event, this panel report’s rejection of such arguments might well serve as a precedent. The panel’s findings that terms like ‘will’ and ‘commitment’ contain are legally stronger than one might think could also be relevant by analogy.

However, one similarity between the UK/EU and UK/S Korea treaties is crucial: the relevant provisions are both subject to a relatively limited form of dispute settlement, consisting of consultation followed by expert panels. No trade remedies can result, even if the panel finds a breach of the treaty. On the weak legal effect of a panel report, the EU/UK treaty is even blunter than the ‘best efforts’ clause in the EU/Korea treaty: ‘the Parties share the understanding that if the Panel makes recommendations in its report, the responding Party does not need to follow these recommendations in ensuring conformity with the Agreement.’

Conversely, though, the references to upholding domestic legal standards – ie the cases similar to US/Guatemala, not EU/S Korea – are removed from the ‘fluffy’ sustainable development chapter, being placed instead in separate chapters in the UK/EU treaty, where trade retaliation can apply.  It is worth comparing the substantive test in full. First of all, the EU/UK treaty provides:

A Party shall not weaken or reduce, in a manner affecting trade or investment between the Parties, its labour and social levels of protection below the levels in place at the end of the transition period, including by failing to effectively enforce its law and standards.

(There’s an identical provision for environmental law). The EU/S Korea treaty provides:

1. A Party shall not fail to effectively enforce its environmental and labour laws, through a sustained or recurring course of action or inaction, in a manner affecting trade or investment between the Parties.

2. A Party shall not weaken or reduce the environmental or labour protections afforded in its laws to encourage trade or investment, by waiving or otherwise derogating from, or offering to waive or otherwise derogate from, its laws, regulations or standards, in a manner affecting trade or investment between the Parties.

Note that there are two rules in the EU/S Korea treaty – non-enforcement and non-regression – which are in effect merged in the EU/UK treaty. In the EU/S Korea treaty, non-enforcement is subject to two different thresholds: not only ‘in a manner affecting trade or investment’ but also ‘sustained or recurring course of action or inaction’. As for non-regression, it is also subject to the ‘manner affecting trade or investment’ test, but also a requirement that the reduction in standards must take the form of a waiver of a derogation of its laws ‘to encourage trade or investment’.

By comparison, while the EU/UK treaty also contains a threshold of ‘in a manner affecting trade or investment between the Parties’, it does not contain the requirement that any non-enforcement must take the form of a ‘sustained or recurring course of action or inaction’. Nor does it require that a reduction in standards take the form of a waiver of a derogation of its laws ‘to encourage trade or investment’. Any reduction of standards will do, not just a waiver or derogation of them, although unlike under the EU/S Korea treaty, the time period to judge the reduction of standards is fixed at a particular date: the end of the transition period. On the face of it, there is no scope for a dynamic interpretation, ie a reduction of standards in 2025 from a higher level which was legislated in 2022 would not fall within the scope of the EU/UK clause, unless that drop in standards went so far as to drop also below the level at the end of 2020.

However, a development like that might fall within the scope of the ‘rebalancing’ clause in the EU/UK treaty, which concerns future developments, and is also subject to potential trade retaliation.  Here there is a different threshold: the rule applies where ‘material impacts on trade or investment between the Parties are arising as a result of significant divergences between the Parties’ as regards labour, environment or subsidy laws.

Finally, just for fun: it’s a moot point now, but can we make any analogies between the EU/S Korea panel’s interpretation of the obligation to move toward ratification of the ILO treaties and the provision in the withdrawal agreement about negotiating a future relationship treaty? Here’s that text:

The Union and the United Kingdom shall use their best endeavours, in good faith and in full respect of their respective legal orders, to take the necessary steps to negotiate expeditiously the agreements governing their future relationship referred to in the Political Declaration of 17 October 2019 and to conduct the relevant procedures for the ratification or conclusion of those agreements, with a view to ensuring that those agreements apply, to the extent possible, as from the end of the transition period.

The wording is far from identical. But in general, remember that the panel report found an obligation of effort, not result, which was moreover not very strict. So the panel report is far from helpful to those who claimed that the EU was in breach of the clause for being ‘mean’ to the UK by not agreeing with the UK’s negotiation position. (Nor would it help anyone making an argument the other way around).  And suffice it to say that while the panel made extensive reference to primary and secondary literature, it did not find that any legal obligations derived from power point slides or tweets from EU officials.

Barnard & Peers: chapter 9; chapter 24; chapter 27; chapter 20; chapter 22

Ending the exploitation of seasonal workers: EU law picks the low-hanging fruit

 

Steve Peers

For a long time, it proved impossible for the EU to agree on legislation on migrant workers coming from non-EU countries. Eventually, the Member States were able to agree on some laws that mostly concerned higher-income migrants: the Blue Card Directive (on its implementation, see here) and the Directive on intra-corporate transferees (see discussion here). The EU has also adopted some general rules on the overall framework for admission of labour migrants (the so-called single permit Directive).

But for the first time last year, the EU also adopted rules on a less well-paid group of migrant workers: seasonal workers. This group of workers is potentially particularly vulnerable to exploitation and abuse. Does the recent Directive go far enough to protect them from these risks?

Content of the Directive

Member States have to apply this Directive by 30 September 2016, and the UK, Ireland and Denmark opted out of it. It is limited in scope to those who normally reside outside the territory of the EU, and who apply to be admitted as seasonal workers, or who have already been admitted under the terms of the Directive. Also, it applies to those admitted for less than three months as well as those admitted for a longer period. For the former group, the Directive specifies that the EU’s borders and visas legislation continues to apply, and makes a number of cross-references to those measures. Furthermore, the Directive does not apply to those workers who are usually employed in other Member States, and who are ‘posted’ by their employers to work in a second Member State, to non-EU family members of EU citizens, and to non-EU citizens covered by an agreement which extends free movement rights (the EEA or EU/Swiss treaties).

A ‘seasonal worker’ is a worker who normally resides outside the EU, and who lives temporarily in the EU to ‘carry out an activity dependent on the passing of the seasons’, pursuant to a fixed-term contract concluded directly with an employer established in a Member State. The concept of a seasonal activity is in turn defined as an ‘activity that is tied to a certain time of the year by a recurring event or pattern of events linked to seasonal conditions during which required labour levels are significantly above those necessary for usually ongoing operations.’ Member States have to define what the relevant sectors are; the preamble refers to tourism, agriculture and horticulture as areas where seasonal work is usually needed.

Member States are free to set higher standards for certain issues (procedural safeguards, accommodation, workers’ rights and facilitation of complaints), but otherwise the Directive has set fully harmonised rules. So Member States can’t alter the substantive grounds for admission or the rules on duration of stay and re-entry.

The key criteria for admission are fully mandatory. Member States have to ensure that an application to enter as a seasonal worker is accompanied by: a valid work contract or binding job offer, setting out all of the details of the job; a valid travel document (possibly valid for the entire duration of the seasonal work); evidence of having, or having applied for, sickness insurance (unless such coverage comes with the work contract); and evidence of having accommodation, as defined in the Directive (see below). Member States have to check that the seasonal worker has sufficient resources not to have to use the social assistance system, cannot admit persons considered to pose a threat to public policy, public security or public health, and must check that the applicant does not pose a risk of illegal immigration and intends to leave the Member States’ territory when the authorization for seasonal work expires.

Applications have to be rejected whenever these conditions are not met, or where the documents presented with an application are ‘fraudulently acquired, or falsified, or tampered with.’ Member States also have to reject applications, ‘if appropriate’, where there has been a prior sanction against the employer for ‘undeclared work and/or illegal employment’, the employer is being wound up or has no economic activity, or the employer has been sanctioned for breach of the Directive.

Otherwise the grounds for refusal of an application are optional: a labour market preference test for home State citizens, other EU citizens or third-country nationals lawfully residing and forming part of the labour market; the application of Member States’ rules on volumes of admission of third-country nationals; or breaches of employment law by the employer, the use of seasonal work to replace a full-time job, or a prior breach of immigration law by the would-be worker. There are similar provisions on withdrawal of the authorisation to work as a seasonal worker, although it should be noted that Member States can withdraw authorization if the worker applies for international protection.

As for the admission procedure, Member States have to make information available on the conditions of entry and residence and rights, as well as the admission process. It’s up to Member States to decide whether the applicant or the employer makes the application, and the application process takes the form of a single application procedure for a combined work/residence status. Those applicants who fulfil the admission criteria and who do not fall foul of the grounds for refusal must be granted a permit or visa, in the format of the EU standard visa or residence permit.

There’s a total maximum limit of between five and nine months per calendar year of residence for a seasonal worker; they must then return to a third country. Since the Directive only regulates admission and stay of seasonal workers, it should follow that Member States still retain discretion to permit the worker to stay for longer on some other ground.

Within the maximum time limit, seasonal workers will be able, on one occasion, to change employers or to obtain an extension of their stay with their employer, if they still meet the criteria for admission, although the grounds for refusal will still apply. The preamble makes clear that this possibility is intended to avoid abuse, since the worker will not be tied to a single employer. Member States will have an option to allow further extensions or changes of employer. But again they can punish any worker who applies for international protection, by refusing to extend that worker’s stay.

Next, the Directive facilitates the re-entry of seasonal workers who were admitted at least once within the previous five years, if they complied with immigration law during their stay. This could include a simplified application process, an accelerated procedure, priority for previous seasonal workers, or the issue of several seasonal worker permits at the same time. The idea is to give an incentive to workers to comply with immigration law.

Member States have to impose sanctions against employers who have breached their obligations under the Directive, including a possible ban on employing seasonal workers. If seasonal workers’ permit to work is withdrawn because of the employer’s illegal behaviour, the employer must compensate the employees for all the work they have done or would have done. There are specific rules on the liability of sub-contractors.

Moving on to procedural safeguards, the Directive provides for: a notified decision in writing within ninety days of the application; special rules on the renewal of authorization; a chance to provide additional necessary information within a reasonable deadline; and a requirement that a rejection (or withdrawal or non-renewal of a permit) be issued in writing and open to a legal challenge, with information on the reasons for the decision, the redress available, and the relevant time-limits. Member States may charge fees for applications, if they are not disproportionate or excessive, and may require employers to pay the costs of workers’ travel and sickness insurance. Workers’ accommodation must ensure an ‘adequate’ standard of living, rents cannot be excessive, a contract for housing must be issued, and employers must ensure that accommodation meets health and safety standards.

As for the rights of seasonal workers, first of all they have the right to enter and stay on the territory of the relevant Member State, free access to the territory of that Member State, and the right to carry out the economic activity which they have been authorized to take up.  Furthermore, they have the right to equal treatment with nationals as regards terms of employment (including working conditions), freedom of association, back payments, social security, the transfer of pensions, access to goods and services available to the public (except housing), employment advice (on seasonal work), education, and recognition of diplomas, and tax benefits. However, equal treatment can be restricted as regards family benefits, unemployment benefits, education and tax benefits, and Member States are still free to withdraw or to refuse to renew the permit in accordance with the Directive. Finally, Member States must ensure monitoring, assessment, and inspections, and facilitate complaints workers or by third parties supporting or acting on their behalf.

Comments

According to its preamble, the intention of this Directive is to regulate the admission of seasonal workers with a view to enhancing the EU’s economic competitiveness, optimizing the link between migration and development, while guaranteeing decent working and living conditions for the workers, alongside incentives and safeguards to prevent overstaying or permanent stay. In principle it has achieved some of these goals, in particular by including a number of provisions to ensure equal treatment and decent accommodation for seasonal workers, to punish employers who mistreat workers or who breach immigration law, and to guarantee that the rules in question are enforced.

In fact the Directive was significantly improved on these points during the legislative process, in particular as regards monitoring and punishment of dodgy employers, accommodation standards, equal treatment (which was significantly extended in scope), employees’ costs, and remedies against employers (compare the final Directive to the original proposal). Doubtless this was largely due to the hard work of NGOs which raised these issues (see their joint statement here). Perhaps the EU should use this Directive as a template to try and address the exploitation of other vulnerable groups of migrant workers – for instance domestic workers, who are at particular risk of being enslaved or trafficked.

Having said that, there are some limits to what Member States were willing to agree. There are exceptions from the equal treatment rule, and some of the provisions on dodgy employers, as well as the ban on passing costs along to the workers, are optional, not mandatory. (See the comments on the final Directive by a group of NGOs here). The right to change employers is subject to conditions, and Member States might decided to allow only one such change. More broadly, while the provisions on enforcement are stronger than what Member States are usually willing to agree to in EU laws about migrant labour (or indeed EU employment law), it remains to be seen how much resources Member States are actually willing to expend on enforcement in practice.

Furthermore, since the Directive is limited in scope to those who are not yet on the territory, it can do nothing to alleviate the position of those who are present without authorization but who cannot be returned (ie who are in limbo) and it gives Member States express carte blanche to deprive asylum-seekers of even the modest income which they were previously earning as seasonal workers. Overall, while the Directive will hopefully have some effect achieving its objectives, it may be a classic example of what academics call ‘picking the low-hanging fruit’ – focusing on the easier issues and avoiding the harder ones.

*This post is based on my ongoing research for the 4^th edition of EU Justice and Home Affairs Law (forthcoming, OUP)

Photo credit: Globalpost.com

 

Barnard & Peers: chapter 26

Can EU law make an effective contribution to the prohibition of forced labour by the ILO?

Steve Peers

While many employees jokingly refer to themselves as ‘wage slaves’, millions of people worldwide are actually forced to supply their labour – a modern form of slavery. Alongside sexual exploitation, labour exploitation constitutes a particular form of trafficking in persons.

In recent years, the chief international actors combatting trafficking in persons have been the United Nations (adopting a Protocol to the Convention on Organised Crime), and (within Europe), the Council of Europe, adopting a Convention on the topic. However, there is another long-standing actor in the field: the International Labour Organisation (ILO), which has particular expertise as regards labour exploitation. Due to concerns that its existing treaties, notably Convention 29 of 1930, are not being fully applied (see the 2013 experts’ report for details), the ILO is considering the adoption of a Protocol and/or a Recommendation relating to that Convention in the near future.

The draft Protocol consists of only six short Articles. It starts with a general obligation for States to prevent and eliminate forced labour, and then sets out a more concrete obligation to adopt action plans to this end. States would have to take ‘specific action against trafficking in persons’ as regards labour or sexual exploitation. They would also have to take steps to educate the public, to broaden the coverage of labour law and strengthen inspection services, and to protect workers who use placement services, particularly migrant workers, against abuse.

Next, States would have to secure the identification, recovery, release and rehabilitation of victims of forced labour. They would also have to secure access to remedies for victims, including compensation. Victims would also have to be exempted from liability as regards crimes which they were forced to commit. Finally, there would be a general obligation for States to cooperate with each other, and the Protocol would specify that States would enforce its rules by means of national law.

As for the draft Recommendation, it sets out further details as regards the substance of the draft Protocol. It includes provisions on: gathering statistics relating to forced labour; skills training for vulnerable groups; programmes to combat related discrimination; the promotion of freedom of association for at-risk groups; setting out terms and conditions of work in a contract; basic social security guarantees; pre-departure orientation for migrants; coherent labour and immigration policies, which take account of the risks posed to irregular migrants; and efforts to reduce the trade in and demand for goods and services produced by forced labour.

The provisions on victims’ protection would be conditional on victims’ informed consent, but conversely could not be conditional on their willingness to participate in criminal proceedings. There would be more details on the specific forms of protection which should be granted to victims (protection from retaliation, housing, health care, privacy, and social assistance). For migrants, there should be a reflection and recovery period, the provision of residence permits ‘as appropriate’ and the facilitation of ‘preferably’ voluntary repatriation.

As regards access to justice, there should be rules allowing representatives to act on behalf of victims, a right for victims to get compensation from perpetrators, (state) compensation schemes, information for victims, and access to court. Enforcement rules should include providing for penalties, including confiscation of profits alongside penal sanctions and liability for legal persons.

EU law aspects

But how does this relate to EU law? The EU has not comprehensively regulated forced labour as such, but a number of separate EU measures do touch on aspects of the issue. In particular, an EU Directive adopted in 2011 regulates the criminal law aspects of trafficking in persons, while a 2004 Directive regulates the immigration status of trafficking victims. The 2012 Crime Victims’ Directive concerns the status of victims of crime more generally. There is also EU employment law dealing with the issue.

The normal rule relating to EU external competence is that the adoption of internal EU legislation on an issue brings with it external EU competence relating to that issue. That competence becomes exclusive where the EU has fully harmonised the issue in question (see Article 3(2) TFEU). On that basis, the EU has concluded the relevant Protocol to the UN Convention, although it has not signed the Council of Europe Convention on trafficking in persons.

However, there is a complication as regards the planned ILO measures, or indeed any measures emanating from the ILO. Unlike many of the treaties drawn up within the framework of the United Nations or the Council of Europe, only States can be party to ILO legal instruments. This also means that the EU as such cannot fully take part in ILO discussions, even when they concern matters within the scope of EU external competence.

Traditionally, this complication has been addressed by adopting EU measures authorising the Member States to act on the EU’s behalf in the ILO framework, and then authorising Member States to sign and ratify the relevant Conventions which result. For instance, the EU has recently authorised Member States to ratify the ILO Convention on Domestic Workers.

Similarly, the Commission has recently proposed a Council Decision harmonising Member States’ positions as regards the planned ILO Recommendation on forced labour, as well as a separate proposal as regards the possible Protocol (the latter proposal has not been published).

But this process, which is also applicable in other cases where the Member States act on the EU’s behalf in international fora, is not always smooth. For instance: the Commission successfully sued Greece for infringing EU external competence as regards maritime security, within the framework of the International Maritime Organisation; the CJEU has been asked to decide whether the EU has to authorise Member States to admit new countries as parties to the Hague Convention on Child Abduction; and the Court also had to settle a dispute relating to competence relating to earlier ILO treaties on the use of chemicals at work.

More fundamentally, in a pending case, Germany is suing to annul a Council Decision coordinating Member States’ positions in an international wine organisation on behalf of the EU, arguing that the EU’s powers to establish positions in international bodies can only apply where the EU itself is a party to the relevant organisation and/or treaty, not where the Member States are acting on the EU’s behalf. Germany also argues that the EU’s powers can only apply where the organisation concerned will establish legally binding rules; this could arguably be relevant as regards the draft ILO Recommendation.

If the German case is successful (an Advocate-General’s opinion is due in May), then the new proposal obviously cannot be adopted (or would be invalid, if it had been adopted in the meantime). But even if the EU is legally entitled to coordinate Member States’ positions in international conferences, including as regards recommendations, there are several complications resulting from this process.

First of all, the Commission proposal does not mention the difficulty arising from the various opt-outs from EU law which some Member States enjoy. The immigration and criminal law measures referred to do not bind all Member States, while the employment law measures do. So arguably two different Council Decisions might be necessary, in order to address this problem.

Secondly, there might be arguments about the precise extent of the external competence of the EU as compared to the proposed ILO measures. In this case, there are some detailed problems with the Commission proposal, although it is too early to say whether the Council might amend or refuse to adopt the proposal. (For more details of these problems, see the annex below.)

Thirdly, if the EU itself is not party to an international treaty, then the CJEU takes less account of that treaty. Although it has made brief reference to ILO treaties which EU legislation implements (for instance, the Schultz-Hoff judgment on the EU’s working time Directive), this can be compared to (for instance) its more frequent references to the UN Convention on the Rights of Disabled Persons (which the EU has ratified): there have been two such references in the past year (the Ring case and the Z case).

Fourthly, it sometimes appears as if the institutional arguments over the exact extent of the EU’s external competence take precedence over the substance of the treaty concerned. In this case, the substance of the planned measures on forced labour are extremely important, given that millions worldwide are subject to this severe breach of human rights. Moreover, the EU could usefully take this opportunity to reflect on whether its own legislative framework is sufficient to address this problem.

Unfortunately, since the EU would not be a party to the planned Recommendation or Protocol, it would not be required to carry out such a reflection. And conversely, even if Member States become parties, there is a limit to the effectiveness of any such reflection, given that some of the measures concerned can only be taken by the EU.

While the division of powers between the EU and its Member States in external relations is an inevitable consequence of the rules on EU competence, there is a growing need for the EU and its Member States to find an effective mechanism to ensure that their potentially important contribution to achieving international objectives, such as (but not only) the elimination of forced labour, are not frustrated by the EU’s internal disputes.

The EU has found some creative solutions to these problems, such as the use of ‘mixed agreements’ (treaties which the EU and its Member States are both parties to), and the adoption of EU legislation regulating Member States’ exercise of their external competence in areas closely related to EU law (such as air services agreements). It is long past time for it to adopt measures which effectively coordinate the Member States’ and the EU’s exercise of their competence as regards the legal framework for international human rights protection.

Annex

Links between EU law and proposed ILO Protocol

Article 2(c): Directive on temporary agency work
Articles 3 and 4: Crime victims Directive; Trafficking in persons Directive

Links between EU law and proposed ILO Recommendation

Clause 2(2): Directive on data protection

Clause 3(c): EU anti-discrimination legislation Clause 3(e): EU Directive on contracts of employment

Clause 3(j): EU trade policy; EU legislation on corporate social responsibility

Clauses 4, 5, 7-10: Crime Victims Directive; Trafficking in Persons Directive; Trafficking victims’ residence permits Directive

Clause 6: Temporary agency workers’ Directive Clause 11(b) and (c): Trafficking in persons Directive

Legal Base problems:

The Commission proposes legal bases relating to criminal law and employment law to adopt the Council Decision. But the preamble also refers to EU immigration legislation and rules on free movement of EU citizens, so the relevant legal bases (Articles 45 and 79 TFEU) logically have to be cited, particularly the immigration legal base given the very close link with the EU legislation concerned. Since the immigration and criminal legislation does not apply to all Member States, arguably there will need to be separate Council decisions to take account of this.

Barnard & Peers: chapter 9, chapter 20, chapter 24, chapter 25, chapter 26