17 - Regulatory Convergence in Preferential Trade Agreements

Summary

This chapter traces the evolution of the recent inclusion in preferential trade agreements (PTAs) of chapters dealing with the process of procedural or substantive regulatory convergence, under different denominations such as ‘regulatory cooperation’, ‘good regulatory practices’, ‘regulatory improvement’, and ‘regulatory coherence’, among others. After mapping this discipline in existing PTAs, we categorise its development in three different approaches: substantive regulatory harmonisation (or ‘substantive coherence’ – the same rule), procedural coherence (the same process), and regulatory cooperation (or ‘regulatory compatibility’ – equivalent rule / equivalent process). The chapter concludes with the possible ways forward for regulatory convergence, considering the use of dispute settlement, regional and sectoral harmonisation, as well as an increased role of private and mixed actors.

17.1 Introduction

In recent years, a new type of chapter has increasingly been included in preferential trade agreements (PTAs), under different monikers, like ‘regulatory cooperation’, ‘regulatory improvement’, ‘regulatory coherence’, and more commonly, ‘good regulatory practices’. They represent a significant new development in trade governance.

Scholars generally use these concepts interchangeably with no agreed typology. Following the framework established in previous works (Polanco Lazo 2013; Polanco Lazo and Sauvé 2018), this chapter considers that these mechanisms should be viewed as forming part of the overarching notion of ‘regulatory convergence’, since they all aim to reduce unnecessary regulatory incompatibilities between countries.

Such a development raises important questions to which this chapter seeks answers: What are the main similarities and differences between these concepts? Do the different denominations matter? What are the guiding principles and objectives of such provisions? What are the challenges to implementing them? What are the possible future developments?

The chapter is organised as follows. Section 17.2 explores the foundations of the notion of regulatory convergence. Section 17.3 looks at its origins, while Sections 17.4 and 17.5 map and compare its different denominations, content, and meaning in early and recent PTAs. Section 17.6 examines the possible future developments of regulatory convergence in PTAs. The final section provides some brief conclusions.

17.2 Regulatory Divergence and Regulatory Convergence

The law and political science literature has described several reasons why regulatory divergence might occur. First, the legal framework and social norms that countries have inherited create a strong path dependency that intrinsically limits the cross-country convergence of legal systems. Countries usually choose rules primarily consistent with their legal traditions (Glaeser and Shleifer 2002; La Porta et al. 2008). Second, regulatory divergence may result from different public policy choices, underlying objectives, and local preferences. Even if countries have the same choices, objectives, and preferences, regulators often adopt different approaches to reach them (e.g. more or less stringent environmental or health regulations), typically resulting in distinct compliance requirements (Basedow and Kauffmann 2016). Third, regulatory differences may also reflect information asymmetries between states (e.g. different access to relevant information) and network effects (e.g. sharing of information across regulators of different countries) (Chirico and Larouche 2013).

Regulatory divergence could be considered a trade barrier if it significantly increases trade costs that hinder cross-border exchanges. Non-tariff barriers have become a significant hurdle for businesses seeking enlarged access to other markets (Basedow and Kauffmann 2016; Chirico and Larouche 2013; Sheargold and Mitchell 2016). For example, they can lead to costly duplication in product development, manufacturing, and testing, affecting internationally active enterprises and especially small or medium-sized enterprises (SMEs), for which such costs can be a decisive factor in exporting (Malmström 2015).

By contrast, the literature highlights some factors that lead to regulatory convergence, based on the repeated interactions at the international level of policymakers, traders, and academics. Other causes of this phenomenon include the perceived need or pressure to comply with a dominant culture of regulatory hegemons, for example by using ‘legal transplants’ (Watson 1993). International law facilitates the dissemination of shared standards and underlying values or ideologies, and the adoption and implementation of uniform laws (Street 2013). Functional convergence among regulators may sometimes occur even as formal regulations diverge (Crettez et al. 2014).

Regulatory convergence and divergence are incomplete and dynamic processes for several reasons. Equal rules may mean different things in different legal systems or have distinct interpretations or applications, even in systems that share a legal culture. Tribunals typically accept a degree of divergence when applying laws (Chirico and Larouche 2013). The pace, form, and degree of convergence also differ over time, across sectors, and are affected by the costs and benefits of change, bureaucracy, and the cultural or social importance of the process (Crettez et al. 2014). Regulatory convergence involves actors with different ‘regulatory capacity’ and degrees of autonomy, and it requires coordination across ministries and regulators within a state (Young 2015). Furthermore, regulatory differences may evolve, fostering convergence or nurturing further divergence (Hoekman 2015).

Regulatory convergence can include substantive or procedural aspects that are aimed at two different types of regulatory outcomes. In the first type, regulatory convergence seeks to achieve substantive regulatory convergence (same or equivalent regulations). The second type considers the convergence of the processes by which regulations are developed, adopted, publicised, and implemented (same or equivalent regulatory procedures), or procedural regulatory convergence. In other words, those that address problems related to the quality of the regulations and their effects are mainly substantive, and those that address issues of design and implementation of regulations are primarily procedural (Sheargold and Mitchell 2016). With different denominations, both approaches are present in the PTAs examined in this chapter.

Regulatory convergence can be achieved through top-down or bottom-up processes (Chirico and Larouche 2013), like harmonisation at the international level (top-down) or exceptions in domestic law for regulatory measures of other countries (bottom-up). Preferential trade agreements consider several mechanisms for these purposes.

There are different typologies of these mechanisms. The Organisation for Economic Co-operation and Development (OECD) has identified eleven forms of ‘international regulatory cooperation’ (IRC), from the least to the most legally binding, including dialogues/ad hoc exchange of information, soft law, recognition/incorporation of international standards, unilateral convergence through good regulatory practices (GRPs), trans-governmental networks of regulators, mutual recognition agreements (MRAs), PTAs with regulatory provisions, joint standard-setting through intergovernmental organisations, formal regulatory cooperation partnerships, specific regulatory agreements between countries, and integration/harmonisation through supranational or joint institutions (Malyshev and Kauffmann 2015; OECD 2013). Good regulatory practices encompass regulatory impact assessments (RIAs), stakeholder engagement, and ex post evaluation (Basedow and Kauffmann 2016).

The distinctions among all these mechanisms are not always clear. Still, one can consider that some aim mainly for substantive convergence (e.g. MRAs, recognition of international standards, joint standard-setting) and others for procedural convergence (e.g. GRPs, exchange of information).

17.3 The Roots of Regulatory Convergence

The roots of regulatory convergence are found in the evolution of the United States (US) administrative law after World War II. The philosophy that has governed American administrative law since then is mainly focused on what we have called procedural convergence, including, among others, transparency and public consultation, RIAs, inter-agency cooperation or coordination and compatibility, and judicial or administrative review. Other developed countries like Australia and New Zealand have also highlighted elements of regulatory convergence in their domestic settings (Lin and Liu 2018).

However, the idea of adopting measures to address regulatory divergence has also been present for a long time in the global arena. Examples include international organisations and forums like the World Trade Organization (WTO), the OECD, the Asia-Pacific Economic Cooperation (APEC), and transatlantic cooperation between the European Union (EU) and the US.

17.3.1 WTO

Since the establishment of the General Agreement on Tariffs and Trade (GATT), behind-the-border regulatory measures have raised concerns about diluting or nullifying the value of tariff bindings by discriminating against foreign products and thus affecting international trade (Sykes 1999). For that reason, drafters of the GATT included general substantive rules on ‘national treatment’ and ‘most-favoured-nation treatment’ (MFN) in GATT Articles III and I, respectively.¹ Similar provisions concerning services are enshrined in Articles XVII and II of the General Agreement on Trade in Services (GATS).²

But, regulatory discrimination often involves barriers that are not as easy to detect as tariffs or quotas applied at the border. Areas such as food inspection, product labelling, and safety guidelines are governed by internal health, safety, and technical regulations, all of which can be designed or implemented to raise costs for foreign firms in comparison with domestic ones, giving the latter an advantage in the home market (Fishbein and Trebilcock 2007). Over time, countries have recognised that implementing GRPs minimises unnecessary trade barriers (Bollyky 2012). In the WTO, the agreements addressing such measures are the Agreement on Technical Barriers to Trade³ (TBT), concerning technical regulations, standards, and procedures for testing and certification, and the Agreement on Sanitary and Phytosanitary Measures⁴ (SPS), regarding food safety and animal health. Both deal with substantive and regulatory convergence using tools such as transparency obligations, recognition of equivalence, MRAs, and treaty bodies to oversee treaty implementation.

But these agreements also further promote substantive regulatory convergence by encouraging members to use international standards and opting for least trade-restrictive ways to satisfy public policy objectives or minimise their adverse effects. Neither the TBT nor the SPS Agreements explicitly compel members to use international standards. However, if they adopt measures that depart from them, they shift the burden of proof regarding the necessity of a measure to the country adopting any regulation that departs from the international standard.⁵

17.3.2 OECD

The notion of ‘regulatory cooperation’ emerged at the OECD in the mid-1990s. A study noted that building more integrated systems for rulemaking and implementation required institutional and procedural cooperation frameworks within national governments, sub-national governments, and the wider public, subject to the constraints of democratic values, such as accountability and openness (Jacobs 1994).

Members of the OECD were the first to consider concerted actions directed at improving the quality of their regulatory systems, particularly by using procedural convergence tools, like RIAs, which are defined as ‘a process of systematically identifying and assessing the expected effects of regulatory proposals, using a consistent analytical method’ (OECD 2008). A study noted that from a couple of countries using this tool in 1980, by 1996, more than half of OECD Members had adopted RIAs (Jacobs 1997). By 2009, RIAs were/became mainstream among OECD Members to improve the quality of regulatory decision-making and inform whether and how to regulate to achieve public policy goals (OECD 2009). In 2012, the OECD Council on Regulatory Policy and Governance issued a recommendation that explicitly promotes the integration of RIAs into the early stages of policy formulation (OECD 2012).

Regulatory impact assessments are comparative and ex ante in character. They determine the underlying regulatory objectives sought and identify all the policy interventions capable of achieving them, using risk analysis or, more often, cost–benefit analysis. The latter can be used as a mechanism of regulatory convergence on its own, without the comparative element of RIAs. Key aspects of RIAs include defining a regulatory problem, identifying different regulatory options (including no policy change), collecting data (from public consultations, interviews, questionnaires, surveys, focus groups, etc.), assessing alternative options, identifying preferred regulatory options, and communicating RIAs results (usually through websites) (Lemoine 2018).

Stakeholders are vital drivers of RIAs. They include groups concerned with the costs and impact of regulations on their business activity, with the promotion of specific values (like transparency and accountability), and with the effect of regulations, such as job creation or the environment (Jacobs 1997). To avoid ‘capture’ by stakeholders, effective and transparent RIA processes are often conducted or overseen by a specialised government body that can provide regulators with high-quality, trusted, and impartial advice (Lemoine 2018).

17.3.3 APEC

In the 2000s, APEC economies⁶ developed a work programme on regulatory reform as part of a joint initiative with the OECD, establishing the APEC–OECD Integrated Checklist on Regulatory Reform (the ‘Checklist’) (APEC and OECD 2005).

The Checklist is a voluntary tool that contains four pillars: regulatory policy, competition policy, market liberalisation policy, and regulatory reform. Concerning the latter, the Checklist comprises mechanisms that APEC economies may use to evaluate their process of development and implementation of regulatory policy, with flexibility in the methods applied, recognising the diversity of economic, social, and political environments and values of APEC economies (Polanco Lazo and Sauvé 2018).

The focus here is also primarily on procedural regulatory convergence, mainly through RIAs, allowing policy comparisons. However, as RIAs alone cannot build a regulatory policy, the Checklist also highlights the need to focus on an integrated system, involving all state sectors acting in coordination and mutual support. It is often difficult for regulators to reform due to close identification with outdated regulatory regimes, countervailing pressures from regulated sectors or different parts of society, as well as personal or bureaucratic interests. For that purpose, the Checklist proposes the creation of a central coordinating body and establishing a coordination mechanism. Additionally, it fosters the use of transparency measures. National and foreign stakeholders are considered in the creation of regulations, through appropriate and publicised procedures of public consultation, not only limited to affected businesses but also including other interested parties, such as trade unions, interest groups (e.g. consumer or environmental organisations), and other pertinent levels of government. Finally, the Checklist suggests a procedure for evaluating a proposed regulatory action’s compliance with international obligations, like commitments under the WTO, PTAs, and international investment agreements (IIAs) (APEC and OECD 2005).

The Asia-Pacific Economic Cooperation developed a baseline study which reviewed the application of selected GRPs across the twenty-one APEC economies (APEC 2011). By 2016, these countries had made visible progress in three practices: internal government coordination of rulemaking activity, public consultation mechanisms, and RIAs – with the latter becoming ‘the norm’ in the APEC region. However, as the implementation of the Checklist is voluntary, notable differences in regulatory processes remain (Jacobs 2017). Additionally, some have warned that a core tenet of RIAs – the need for ‘clear evidence’ that regulation is needed – could be dangerous/questionable if it is subordinated to free trade and investment. For them, reducing red tape has served primarily business and not society as a whole (Trade Union Advisory Committee 2016).

17.3.4 Transatlantic Cooperation

Historically, various initiatives have addressed regulatory divergence between the EU and the US.⁷ Still, it was not until 2005, with the establishment of the EU–US High-Level Regulatory Cooperation Forum (HLRCF), that both parties decided to create a body to discuss these issues.

In 2007, the EU and the US renamed the HLRCF as the Transatlantic Economic Council (TEC), with the formal task of avoiding increased cross-border trade costs stemming from duplicative regulations and to ‘oversee and accelerate government-to-government cooperation in furthering economic integration’ (Posner and Wolff 2011). The TEC addressed the ‘regulatory cooperation’ issue by establishing a framework for fostering cooperation and reducing regulatory burdens.⁸ To meet these objectives, the EU and the US established two kinds of cooperative efforts on substantive and procedural convergence. On the latter, the TEC promoted the exchange of experiences and information among regulators and developed a methodological framework allowing comparisons between RIAs, risk assessments, and cost–benefit analyses concerning regulations.⁹ On substance, the TEC advanced regulatory convergence in specific sectors (cosmetics, medical devices, pharmaceuticals, automobiles, electrical equipment, and nano-materials)(Woolcock et al. 2015).

The TEC did not succeed in promoting regulatory convergence between the US and the EU and stumbled ‘over efforts to resolve disputes involving sales of poultry, cosmetics, and electrical equipment’ (Ahearn 2009). Obstacles to bilateral regulatory cooperation remained, particularly in food safety, the environment, and security, where both parties have maintained diverging risk perceptions and regulatory philosophies (Mildner and Ziegler 2009). After five meetings, the TEC stopped functioning at the ministerial level in 2010, although cooperation was still being pursued at a technical level (European Commission 2015).

The EU and the US reignited transatlantic cooperation after a summit on 15 June 2021 in Brussels, establishing a Trade and Technology Council (TTC) to serve as a forum to coordinate approaches to key global trade, economic, and technology issues and to deepen transatlantic trade and economic relations (European Commission 2021). The EU and the US have established ten TTC Working Groups, respectively, focused on tech standards, climate and green tech, secure supply chains, information and communications technology and services (ICTS) security and competitiveness, data governance and tech platform regulation, misuse of technology threatening security and human rights, export controls, investment screening, promoting SME access to and use of digital technologies, and global trade challenges (US Department of Commerce 2023).

Up to now, the TTC focus largely seems to be on procedural convergence, as manifested in the initial functioning of its Working Groups on activities of cooperation and promotion. But a more substantive convergence is intended in at least two areas: joint standards in post-quantum encryption and cybersecurity of the internet of things, and extending MRAs on marine equipment and vaccines (European Commission 2022). It is too early to assess if the TEC will deliver. Swift regulatory alignment between the US and the EU is still unlikely, and initial effects have been mainly geopolitical (Ringhof 2022).

17.4 Regulatory Convergence in Early PTAs

Individual chapters on regulatory convergence are not found in most PTAs, but disciplines relating to some of its elements, like transparency, were common even before the notion of regulatory convergence appeared (Lin and Liu 2018).¹⁰

With different degrees of success, some regional PTAs were early movers, as questions about regulatory convergence matter for countries that trade a lot and have borders. In 1983, the Australia−New Zealand Closer Economic Relations Trade Agreements (ANZCERTA)¹¹ established joint institutions, including common regulatory agencies, to develop ‘joint food standards, harmonised approaches to certification of quality management systems, and mutual recognition of product standards and occupational qualifications’ (Steger 2012). Both countries established an MRA on product standards and professional qualifications in 1997,¹² and a common food standards code in 2003 issued by a joint agency, the Food Standards Australia New Zealand (FSANZ).¹³

Canada, Mexico, and the US concluded several ad hoc or informal cooperative arrangements outside the North American Free Trade Agreement (NAFTA), like the Security and Prosperity Partnership (SPP) of North America, the Mexico−US High-Level RCC (HLRCC), and the Canada−US Regulatory Cooperation Council (RCC). Besides some substantive regulatory convergence from the latter, in health products, energy efficiency, and motor vehicle safety rules, these initiatives have been largely unsuccessful (Manak 2022).

The Association of Southeast Asian Nations (ASEAN)¹⁴ aims to align domestic norms with international standards and develop MRAs on conformity assessment and professional services (Steger 2012). So far, however, the extent of convergence has been limited. The Association of Southeast Asian Nations has a Consultative Committee on Standards and Quality (ACCSQ) with the goal of ‘One Standard, One Test, Accepted Everywhere’ (ASEAN 2014b), and a Framework Agreement on Mutual Recognition Arrangements (ASEAN 2014a). Using these tools, ASEAN Members have fostered substantive regulatory convergence through MRAs on medical services, architecture, accountancy, and tourism services, among other areas (Sheargold and Mitchell 2016).

Members of the Southern Common Market (‘Mercado Común del Sur’ – Mercosur)¹⁵ established in 1992 a standardisation committee to elaborate voluntary standards, promote cooperation to facilitate harmonisation, align Members’ regulations on international standards, promote training in standardisation and quality control, and encourage the development of certification systems and MRAs.¹⁶ However, the Protocol of Ouro Preto in December 1994¹⁷ suppressed the standardisation committee without providing reasons behind this decision.

The 2010 EU–South Korea Free Trade Agreement (FTA),¹⁸ included a section on ‘regulatory framework’, in Chapter 7 on trade in services, investment, establishment, and electronic commerce, containing some tools of substantive regulatory convergence, like mutual recognition or harmonisation of minimum common standards, avoiding needless or unjustified differences in rulemaking. It also considers mechanisms of procedural regulatory convergence, such as dialogues and committees, exchange of information, experiences, and data, as well as through deepened scientific and technical cooperation (Laurenza and Mathis 2013). Under the guise of ‘regulatory cooperation’, parties adopted international standards in pharmaceutical goods, medical products,¹⁹ and the automotive industry.²⁰

17.5 Regulatory Convergence in Recent PTAs

Regulatory convergence in PTAs has evolved from a relatively voluntary principle (e.g. OECD and APEC) to a more legalised treaty obligation (with some elements in early PTAs), to having dedicated chapters in this discipline, as explained in this section.²¹

17.5.1 Transatlantic Trade and Investment Partnership

In June 2013, the EU Member States mandated the European Commission (EC) to negotiate a Transatlantic Trade and Investment Partnership (TTIP) with the US. One of its objectives was to increase trade and investment through greater regulatory compatibility, setting the path for global standards.

According to the EU, the TTIP should include cross-cutting disciplines on ‘regulatory coherence’, featuring both substantive and procedural convergence. Some were not wholly new, like GRPs on transparency, early stakeholder consultations, periodic reviews of regulations, and RIAs. Novel mechanisms included focal points to facilitate information exchange between regulators or, if specific common interests were identified, recognition of equivalence or harmonisation of regulatory acts. A Regulatory Cooperation Body (RCB) would develop horizontal provisions on regulatory cooperation, complemented by commitments in nine specific sectors: automotive, chemicals, cosmetics, pharmaceuticals, information and communication technology, engineering, financial services, medical devices, and textiles. The Regulatory Cooperation Body was not supposed to have regulatory competence per se, and both parties would continue to regulate following their domestic regulatory frameworks, procedures, and principles (European Commission 2013, 2015).

The US position concerning regulatory convergence in the TTIP was never publicly available. It was reported that a July 2014 proposal sought only to cover federal rulemaking on the US side but both community-level and Member States’ legislation and regulations on the EU side. A negotiating draft leaked in 2016 included both procedural and substantive convergence. On the latter, it proposed a bilateral cooperation mechanism, information and regulatory exchanges, the promotion of regulatory compatibility and international cooperation, as well as the establishment of an RCB. The procedural part featured a section on GRPs, including RIAs, decision-making based on evidence, internal coordination of regulatory development, prior notification procedures, stakeholder consultations, transparency obligations, and the possibility of reviewing regulations. (Greenpeace Nederland 2016).

Transatlantic Trade and Investment Partnership negotiations were halted by the US during the Trump Administration, and on 15 April 2019, the EC declared them ‘obsolete and no longer relevant’ (Council of the European Union 2019).

17.5.2 PTAs with Regulatory Convergence Chapters

Several recent trade agreements include standalone chapters on regulatory convergence. According to the OECD, PTAs increasingly incorporate horizontal chapters on GRPs, IRCs, or both (Kauffmann and Saffirio 2021). We have found a total of twenty PTAs, including regulatory convergence chapters, which are detailed in Table 17.1 below.²²

Table 17.1 PTAs with regulatory convergence chapters

	Name	Date of signature	Date of entry into force	Name	Chapter / Title
1	Pacific Alliance Additional Protocol (PAAP)	03.07.15	01.04.20	Regulatory Improvement	Ch. 15bis
2	Trans-Pacific Partnership (TPP)/ Comprehensive and Progressive Trans-Pacific Partnership (CPTPP)	04.02.2016 / 08.03.2018	30.12.2018	Regulatory Coherence	Ch. 25
3	Chile–Uruguay FTA (2016)	04.10.16	13.12.18	Regulatory Coherence	Ch. 15
4	Canada–EU Comprehensive Economic and Trade Agreement (CETA)	30.10.16	21.09.2017 (provisionally applied)	Regulatory Cooperation	Ch. 21
5	Australia–Peru FTA	12.02.18	11.02.20	Regulatory Coherence	Ch. 24
6	EU–Japan Economic Partnership Agreement (EPA)	17.07.18	01.02.19	Good Regulatory Practices and Regulatory Cooperation	Ch. 18
7	Brazil–Chile FTA	21.11.18	25.01.22	Good Regulatory Practices	Ch. 3
8	US–Mexico–Canada Agreement (USMCA)	30.11.18	01.07.20	Good Regulatory Practices	Ch. 28
9	Protocol to amend the Agreement between New Zealand and Singapore Comprehensive Economic Partnership Agreement (CEPA)	17.05.19	01.01.20	Regulatory Cooperation	Ch. 13
10	Japan–United Kingdom (UK) CEPA	23.10.20	01.01.21	Good Regulatory Practices and Regulatory Cooperation	Ch. 18
11	EU–UK and Cooperation Agreement	30.12.20	01.05.21	Good Regulatory Practices and Regulatory Cooperation	Part. II; Tit. X
12	Chile–Paraguay FTA	12.01.21	14.02.24	Good Regulatory Practices	Ch. 3
13	Iceland–Liechtenstein–Norway–UK FTA	08.07.21		Good Regulatory Practices and Regulatory Cooperation	Ch. 11
14	Australia–UK FTA	16.12.21		Good Regulatory Practices	Ch. 26
15	Pacific Alliance–Singapore FTA	26.01.22		Good Regulatory Practices	Ch. 20
16	New Zealand–UK FTA	28.02.22		Good Regulatory Practices	Ch. 21
17	Taiwan–US Trade Initiative	01.06.23		Good Regulatory Practices	Ch. 3
18	EU–New Zealand FTA	09.07.23	01.05.24	Good Regulatory Practices	Ch. 22
19	Canada–Ukraine Modernized Free Trade Agreement	22.09.2023		Good Regulatory Practices	Ch. 26
20	Chile–EU Interim Trade Agreement	13.12.2023		Good Regulatory Practices	Ch. 29

When negotiations of the Comprehensive Economic and Trade Agreement (CETA) between the EU and Canada concluded in August 2014, they included for the first time a chapter on ‘regulatory cooperation’ (Chapter 21). CETA focuses mainly on substantive convergence, including improving regulatory quality by avoiding or reducing unnecessary differences, identifying alternatives (like the recognition of equivalence), and minimising administrative costs.²³ It also has features of procedural convergence like RIAs, data collection and analysis, sharing regulatory proposals, joint high-level dialogue, and specific sectoral cooperation initiatives dealing with consumer safety, among others (Steger 2014).

The next milestone in incorporating regulatory convergence chapters in PTAs can be observed in the Pacific Alliance (PA). The PA Members (Chile, Colombia, Mexico, and Peru) aimed to deepen intra-regional trade, investment, and mobility, focusing on improving regulatory processes and their transparency (Ministerio de Comercio, Industria y Turismo de Colombia 2016). On 3 July 2015, Members signed a Protocol Amending the First Additional Protocol to the Framework Agreement of the Pacific Alliance (PAAP, in force since 1 May 2016), which includes (in its Annex 4) a new chapter on ‘Regulatory Improvement’.²⁴ This chapter primarily focuses on procedural convergence, with Members affirming their desire to improve their regulatory performance through establishing and systematically implementing tools such as transparency and public consultation, review, and ex ante and ex post measurement of the impact of regulations, and the simplification of procedures.

On 4 February 2016, the TPP was signed by twelve countries. Although the chapter on ‘Regulatory Coherence’ was concluded after CETA and the PAAP, its negotiations predate both agreements. It primarily focuses on procedural convergence, improving regulatory practices, conducting regulatory processes more trade-facilitatively, eliminating redundancies in testing and certification, and increasing transparency and stakeholder engagement. It also aimed to make the regulatory systems of member countries more compatible, removing unnecessary barriers, reducing regional divergences in standards, and promoting cooperation on specific regulatory issues (Bollyky 2012). In the wake of the January 2017 decision of the US to withdraw from the agreement, a revised version of the treaty, now renamed CPTPP, was signed on 8 March 2018, keeping the same chapter on regulatory coherence as in the original TPP.

In both the PAAP and the CPTPP, emphasis is laid on the use of international GRPs in planning, drafting, promulgating, implementing, and reviewing regulatory measures, considering the input from interested stakeholders in their development.²⁵ The USMCA – an agreement that partially follows and expands this model – also encourages regulatory compatibility between the contracting parties.²⁶

Subsequent agreements have primarily followed the template of the CETA and CPTPP/PAAP. A more detailed comparison of both models follows in the next section.

17.5.3 Comparison of PTAs Models of Regulatory Convergence

This section offers a comparative analysis of provisions in the PTAs that currently include regulatory convergence chapters. With some variations, we can detect two main models among these treaties: the CETA model (followed mainly in subsequent EU and UK agreements), and the CPTPP/PAAP model (followed by some countries of those trading blocs).

17.5.3.1 Definition

The chapter on regulatory cooperation in the CETA does not include a definition of such a discipline. The same happens in the EU–Japan EPA, the Japan–UK CEPA, the EU–UK TCA, Iceland–Liechtenstein–Norway–UK FTA, Australia–UK FTA, and New Zealand–UK FTA. Definitions help to identify the real focus of regulatory convergence, whether substantive, procedural, or both.

The definition of ‘regulatory coherence’ in the CPTPP is mainly about procedural convergence. It refers to the ‘use of good regulatory practices in the process of planning, designing, issuing, implementing, and reviewing regulatory measures to facilitate achievement of domestic policy objectives, and in efforts across governments to enhance regulatory cooperation to further those objectives and promote international trade and investment, economic growth, and employment’.²⁷ Basically the same definition is included in the Chile–Uruguay FTA, Australia–Peru FTA, and PA–Singapore FTA (although the latter uses ‘best practices’ to define ‘good regulatory practices’).²⁸ The PAAP definition of ‘regulatory improvement’ is essentially identical to ‘regulatory coherence’ in the CPTPP, with one telling difference. While the CPTPP refers to the use of ‘good regulatory practices’, the PAAP refers to ‘good international regulatory practices’.²⁹ This would appear to acknowledge that such good practices are most likely to be ‘imported’ and not generated by the PA Members.

A different definition, but also concerning procedural convergence, is found in the Chilean FTAs with Brazil and Paraguay, which define ‘good regulatory practices’ as the use of tools in the process of planning, drafting, adoption, implementation, review, and follow-up of regulatory measures.³⁰

Broader definitions, including substantive and procedural convergence elements, are found in three PTAs following the CPTPP/PAAP model. The United States–Mexico–Canada Agreement and the Taiwan–US PTA define ‘regulatory cooperation’ as ‘an effort between two or more Parties to prevent, reduce, or eliminate unnecessary regulatory differences to facilitate trade and promote economic growth, while maintaining or enhancing standards of public health and safety and environmental protection’.³¹ The updated New Zealand–Singapore CEPA defines it as ‘efforts between the Parties to enhance regulatory cooperation to further domestic policy objectives, improve the effectiveness of domestic regulation in the face of increased cross-border activity, and promote international trade and investment, economic growth, and employment’.³²

17.5.3.2 Scope of Application

The regulatory cooperation chapter in the CETA considers the development, review, and methodological aspects of regulatory measures covered by WTO agreements: TBT, SPS, GATT, and GATS, all deemed incorporated in the treaty.³³ The parties may expand this scope of application by discussing regulatory policy issues of mutual interest, via consultations or in a treaty body (the Regulatory Cooperation Forum – RCF).

Other treaties following the CETA model have a broader scope. Regulatory convergence applies to all measures issued by the parties’ regulatory authorities regarding any matter covered by the agreement, or when they are relevant to regulatory cooperation activities, such as guidelines, policy documents, or recommendations.³⁴ Yet, some PTAs consider exclusions to this rule (e.g. taxation, public procurement, and public financial assistance).³⁵

In contrast, regulatory convergence in PTAs following the CPTPP/PAAP model applies only to ‘covered measures’. These are measures of general application, with which compliance is mandatory, adopted, issued, or maintained by a regulatory authority. In determining the covered measures, each party should aim to achieve ‘significant’ coverage, promptly after the date of entry into force of the agreement: no later than one year (CPTPP, Brazil–Chile FTA), two years (PA–Singapore FTA), or three years (PAAP, Chile–Uruguay FTA) and make it publicly available.³⁶

Some of these treaties briefly depart from the CPTPP/PAAP model, and have a larger scope, applying to all regulatory practices, not requiring to achieve significant coverage.³⁷ However, some of these PTAs include exceptions to this rule. The United States–Mexico–Canada Agreement excludes measures concerning military or defence, agency organisation, taxation (for Canada and Mexico), and financial services (for Mexico and the US).³⁸ Financial and military regulations are excluded from the Taiwan–US PTA.³⁹

17.5.3.3 Guiding Principles and Objectives

Both the CETA and CPTPP/PAAP models of regulatory convergence share general principles, and more detailed objectives like the right to regulate and inter-state cooperation.

17.5.3.3.1 General Principles

Some general principles are recurrent in PTAs with regulatory convergence chapters. Two are not entirely novel or specific to this discipline, namely facilitating increasing trade and investment between the parties,⁴⁰ and reducing or removing potential regulatory barriers,⁴¹ which are also generally found in other PTAs.

But PTAs with regulatory convergence chapters also include principles that are more specific to the domain, like having an effective, transparent, and predictable regulatory environment,⁴² taking into account input from interested persons in developing regulatory measures,⁴³ and reducing or removing unnecessarily burdensome, duplicative, or divergent regulatory requirements.⁴⁴

Other principles mentioned in a few PTAs include facilitating innovation, adopting new technologies,⁴⁵ and international models, norms, and rules.⁴⁶

17.5.3.3.2 Right to Regulate

All regulatory convergence mechanisms can be seen as diminishing the regulatory autonomy of states, limiting their choices in how to pursue public policy objectives (Sheargold and Mitchell 2016). Probably for that reason, most PTAs address the relation of this discipline with the state’s right to regulate either directly or indirectly, balancing more depth with flexibility.

The CETA explicitly upholds the right to regulate in its investment, labour, and environment chapters.⁴⁷ Other PTAs that follow the same model recognise that right in their chapters on trade and sustainable development,⁴⁸ labour,⁴⁹ or animal welfare.⁵⁰ One recognises it in services, investment, digital trade, competition, and sustainable development.⁵¹ However, most of them do not explicitly recognise the right to regulate in pursuit of public policy objectives according to each party’s legal system, in their regulatory cooperation chapter, except for three UK PTAs, where parties are free to adopt regulatory measures to address their regulatory priorities to ensure the levels of protection they consider appropriate to achieve their public policy objective.⁵² Likewise, some recognise the freedom of each party to determine its approach to GRPs in a manner consistent with its own legal framework, practice, procedures, and fundamental principles.⁵³

Without explicitly mentioning the ‘right to regulate’, the CPTPP/PAAP model reaffirms the parties’ sovereign right to establish regulations, identify their regulatory priorities, and implement measures to address them, at the scope and the levels they deem appropriate.⁵⁴

Compared to the CPTPP/PAAP model, the CETA model clearly delineates its regulatory objectives. The CETA mentions the parties’ commitment to ensuring high levels of protection for human, animal, and plant life, health, and the environment.⁵⁵ Similar PTAs explicitly include health, safety, and environmental goals as public policy objectives.⁵⁶ Some add occupational health and safety, labour conditions, climate change, consumer protection, social protection and social security, personal data and cybersecurity, cultural diversity, financial stability, and energy security.⁵⁷ The EU–UK TCA also adds money laundering.⁵⁸ In contrast, among the treaties following the CPTPP/PAAP model, only the USMCA and the Taiwan–US PTA explicitly mention public policy objectives like health, safety, and environmental or sustainable goals.⁵⁹

17.5.3.3.3 Inter-state Cooperation

Both models promote cooperation activities between state parties to develop regulatory reform policies.⁶⁰ In some, these activities also include capacity building and strengthening in this field.⁶¹ In most cases, the regulatory cooperation activities are voluntary or include hortatory language, although some agreements have binding commitments to cooperate.⁶²

Several agreements include concrete cooperation activities, such as information exchange, dialogues, meetings between the parties, their regulatory authorities, and stakeholders; training programmes, seminars, and other technical assistance initiatives.⁶³ Certain PTAs add activities of data exchange, information and practices, including management, RIAs, cost–benefit analysis, ex post review, and public consultations.⁶⁴

Most of these PTAs foresee cooperation exclusively between parties. The CETA considers it between ‘relevant trading partners’, and is open to others ‘only if practicable and mutually beneficial’.⁶⁵ The CPTPP acknowledges cooperation between treaty partners, without any express limitation or exclusion of third parties.⁶⁶ However, some are explicitly open to IRC with non-parties,⁶⁷ or in international forums.⁶⁸ According to the PAAP, parties can consider existing measures of Member States or regional forums.⁶⁹ Likewise, the Brazil–Chile FTA encourages parties, when developing regulatory measures, to consider international and foreign references, to the extent appropriate and consistent with domestic law.⁷⁰

17.5.3.4 Mechanisms of Regulatory Convergence

Preferential trade agreements with horizontal chapters on regulatory convergence promote several GRPs that we can classify as internal and external mechanisms. Internal mechanisms aim to build convergence starting from each state (‘bottom-up’), which are then coordinated, implemented, and reviewed with other contracting parties. External mechanisms aim first to achieve convergence between contracting parties, a process that is later internalised (‘top-down’) (Polanco Lazo and Sauvé 2018).

1. a) Internal mechanisms refer to bottom-up processes which are generally binding, such as RIAs⁷¹ (carry out and make public an analysis of regulatory measures under preparation, considering the need of them, and any feasible and appropriate alternatives, including the option of not regulating);⁷² transparency (including publication, free and searchable access to regulatory measures online through a ‘regulatory register’ or a dedicated website),⁷³ stakeholder participation (usually through public consultation),⁷⁴ and establishing internal coordination and review processes and mechanisms.⁷⁵ In addition, some PTAs recognise the important role of existing central regulatory coordination bodies (CRCBs) in the contracting parties,⁷⁶ or encourage its establishment.⁷⁷ Other less common internal mechanisms are the periodic or retrospective review or evaluation of regulatory measures,⁷⁸ determining whether they should be modified, streamlined, expanded, or repealed,⁷⁹ to reduce regulatory burden;⁸⁰ the use of plain language to ensure that new regulatory measures are clear, concise, well-organised, and easy to understand;⁸¹ and early information on planned regulatory measures, at least on an annual basis, with a brief description of its scope and objectives, the estimated time for its adoption, and opportunities for public consultation.⁸²

2. b) External mechanisms follow a top-down process, such as establishing treaty bodies,⁸³ contact points,⁸⁴ or specific cooperation activities between parties to the agreement, like exchanges of information on regulatory practices⁸⁵ in case of planned regulatory measures.⁸⁶ The New Zealand–UK FTA further details informal cooperation (including dialogues or meetings between policy officials), formal cooperation (including MRAs, equivalence, or harmonisation), and engaging with interested persons, including businesses and consumers.⁸⁷

As we can see, PTAs largely favour procedural convergence. One of the few internal measures aimed at substantive convergence found in both models calls for a general ‘consideration’ of other treaty parties’ regulatory measures. But such soft commitments leave plenty of room for divergence. In CETA and like agreements, parties undertake, when appropriate, to consider the regulatory measures or initiatives of the other party on the same or related topics. However, that does not prevent parties from adopting different approaches, including institutional and legislative set-ups, circumstances, values, or priorities.⁸⁸ In other PTAs following this model, when developing new or revising existing regulatory measures, parties shall consider, to the ‘extent feasible’, any regulatory approach by the other party, on the same or a related matter.⁸⁹

In the CPTPP/PAAP model, each party should encourage its relevant regulatory agencies to consider regulatory measures in other parties, as well as relevant developments in international, regional, and other forums when planning covered regulatory measures, but only to the extent appropriate and consistent with its legal system.⁹⁰

As we can see, most of the internal mechanisms of regulatory convergence included in existing PTAs are binding and aim to achieve procedural rather than substantive convergence. Provisions on substantive convergence are increasingly found, but their ‘bindingness’ is very small. We are still at the beginning when thinking about substantial commitments.

17.5.3.5 Implementation

For purposes of transparency and to serve as a basis for cooperation and capacity-building activities, several PTAs following the CPTPP/PAAP model undertake to issue a regular report on the implementation of their regulatory convergence chapters. In these agreements, each party shall submit a ‘notification of implementation’ to the free trade commission (FTC) through national contact points, describing the steps taken since the entry into force of the treaty and those intended to take in implementing the regulatory convergence chapter. The FTC will examine these notifications and ask questions or discuss specific aspects of the notification of that party.⁹¹ In contrast, CETA and other PTAs following that model do not have explicit commitments concerning the implementation of regulatory convergence chapters.

In practice, this does not make much of a difference, as there is very little information about the implementation of these agreements in general. The CPTPP Committee on Regulatory Coherence just met for the first time in May 2021.⁹² The EU–Japan Committee on Regulatory Cooperation has met three times in the past two years, with no concrete commitments.⁹³

17.6 Possible Future Developments

17.6.1 New Mechanisms of Internal Regulatory Convergence

Recent PTAs expand the toolkit of regulatory convergence, particularly those we have identified as ‘internal’ mechanisms. Among them, we find:

1. a) Descriptions of regulatory processes and mechanisms, including RIAs and other relevant GRPs.⁹⁴ Some add that these descriptions shall be freely and publicly available online.⁹⁵

2. b) Provide any person with an opportunity to submit comments for improvements of regulatory measures in force, including suggestions for simplifying or reducing unnecessary burdens.⁹⁶

Among the most recent PTAs, the USMCA and the Taiwan–US PTA include more of these new mechanisms. In addition to those referred to above, both treaties recognise the need for regulations to be based on reliable information and of high quality (including scientific, technical, economic, and other information relevant to the regulation). To that end, parties should adopt or maintain publicly available guidance or mechanisms that encourage regulatory authorities when developing a regulation to seek such information.⁹⁷ Likewise, they recognise that regulatory authorities may seek advice and recommendations to prepare or implement regulations from expert advisory groups, including those outside the government.⁹⁸

Future PTAs dealing with regulatory convergence will probably expand and refine such mechanisms, based on the experience gained from their implementation. However, increasing internal regulatory convergence is likely to require more than that. One possible way forward would be actively promoting central regulatory coordinating bodies in each country, one of the few internal mechanisms of regulatory convergence found in PTAs that are not binding. The mandatory establishment of such bodies could be coupled with assistance (both technical and financial) from states with prior experience in this regard (like OIRA in the US,⁹⁹ or the Treasury Board in Canada),¹⁰⁰ to help this development in countries lacking such institutions, particularly developing countries. There are already some signs in that direction. A few Latin American countries have established CRCBs, including some of those that have negotiated PTAs with regulatory convergence chapters, like Chile, Mexico, Peru, and Colombia (as well as El Salvador).¹⁰¹

17.6.2 More Substantive Regulatory Convergence?

As we have seen, up to now, PTAs have primarily focused on procedural convergence. Still, another possible future development could be the increase of substantive regulatory convergence in general and specific sectors.

Some recent PTAs include glimpses in that direction. The EU–Japan EPA aims to promote common principles, guidelines, codes of conduct, and implementing tools.¹⁰² In 2021, the PA agreed on common rules for eliminating technical barriers to food supplements, cosmetic products, household cleaning products, and medical devices, mostly on certificates and labelling.¹⁰³ Although the current CETA RCF work plan is mainly about information-sharing activities, since 2017, it has considered initiatives to align regulations between Canada and the EU in consumer product safety, pharmaceutical inspections, and animal welfare (started in 2021).¹⁰⁴

Furthermore, outside PTAs, substantive regulatory convergence has increased in specific sectors, particularly at a regional level. For example, the Pan American Network for Drug Regulatory Harmonization (PANDRH), created in 1999, focuses on harmonising regulations for small molecule drugs in the Americas. The African Medicines Regulatory Harmonisation Initiative (‘Zazibona Process’) was launched in 2009 and involves cooperation between Zambia, Zimbabwe, Botswana, and Namibia, among others (Alvaro et al. 2019).

Among the different challenges posed by the COVID-19 pandemic, we found the recognition of vaccination certificates from other countries. Still, some convergence was partially achieved concerning WHO-approved vaccines – even though global vaccination coverage on this and other vaccines is lacking.¹⁰⁵ Harmonisation of regulatory guidelines of clinical trials is much needed (e.g. consent, handling of data, the equivalence of conclusions).¹⁰⁶

Avoidance or reduction of regulatory divergence during the COVID-19 crisis was possible through cooperation and coordination across regulatory agencies, not only on vaccines and related issues, but also on other issues like food and refugees.¹⁰⁷ Inter-agency cooperation or coordination is a model for avoiding divergence in the first place and can be used to achieve more substantive regulatory convergence in the future (although this is likely to happen only for specific topics).

17.6.3 Increased Role of Private Actors

The distinction between public and private regulations is not always clear, as there are different sources of regulations following the ‘governance triangle’ that includes state-led, private-led, and collaborative regulations (Abbott and Snidal 2010).

Future developments in regulatory convergence in PTAs may include an increasing role of private actors in developing regulations. As mentioned, one of the most common general principles found in agreements with regulatory convergence chapters is precisely to consider inputs from interested persons in developing regulatory measures.

As Vallejo points out, some private actors already have regulatory authority. However, such recognition is subject to formal and substantive conditions of validity that are not equivalent to those applicable to public agents (Vallejo Garretón 2021). Examples of these entities include the International Financing Reporting Standards (IFRS) Foundation,¹⁰⁸ which has become the focal point of transnational accounting standard-setting being endorsed by the WTO, the International Organization of Securities Commissions (IOSCO), the World Bank, the G7, and the Basel Committee (Vallejo Garretón 2020). Likewise, since its creation in 1990, the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) has brought together regulatory authorities and the pharmaceutical industry to discuss scientific and technical aspects of pharmaceuticals and develop guidelines to achieve greater harmonisation worldwide.¹⁰⁹

However, an increasing role of these stakeholders may increase the risk of ‘regulatory capture’, without a clear delineation of how the public consultation processes should be conducted, and whether they should be opened to include other stakeholders and governments than the one planning to regulate. Some fear that as ‘interested persons’ or ‘stakeholders’ are not always clearly defined, that may lead to agenda capture by major corporations and business lobby groups (Kelsey 2011). However, it could also be the case that the influence of private standards could decrease trade costs and facilitate convergence, particularly in sectors where private standards are commonplace, as in the automotive industry (Costa and Jacoby 2014).

17.6.4 Dispute Settlement

A significant limitation of most treaties following the CPTPP/PAAP model is that they are not subject to dispute settlement provisions. This means that instances of non-compliance with the chapter’s obligations are not directly enforceable by the Member States.¹¹⁰

There is no explicit exclusion from the dispute settlement mechanism in CETA. However, the limitations inherent in the CETA approach make conflict unlikely, as the parties commitments are largely voluntary and do not limit their ability to carry out their regulatory, legislative, and policy measures.¹¹¹ Activities like conducting, comparing, and sharing post-implementation reviews are also only facultative. All other treaties following the CETA model have explicitly excluded their regulatory convergence chapters from the treaties’ dispute settlement mechanism.¹¹²

A further significant limitation of the CPTPP/PAAP model is that in the event of any inconsistency between their respective chapters, and any other chapter of the agreement, the latter will prevail.¹¹³ The CETA has no similar provisions, and neither treaties that follow its model except for the UK FTAs with Japan, Iceland–Liechtenstein–Norway, Australia, and New Zealand.¹¹⁴

However, current developments in both models open the possibility of dispute settlement concerning regulatory convergence commitments. A couple of recent agreements that largely follow the CETA model – the EU–Japan EPA and the Japan–UK CEPA – include a mechanism through which a party may submit to the other party a request to consider its concerns about a planned or existing regulatory measure. The requesting party shall identify the regulatory measure at issue, describe its concerns, and, where relevant, submit questions. The responding party shall, as soon as possible but, unless justified, no later than 60 days after the receipt of the request, provide written comments as regards the concerns raised by the requesting party. Even though there is no possibility of accessing the general dispute settlement mechanism, the requesting party may request consultations with the responding party, to address the concerns of the requesting party. This might include proposals for adjusting the regulatory measure or adopting a less trade- or investment-restrictive one. However, they do not require the responding party to achieve a particular regulatory outcome and shall not delay the adoption of a regulatory measure.¹¹⁵ It is also noteworthy that in both agreements, no provision makes other chapters prevail in case of inconsistency with the regulatory convergence chapter.

Most importantly, there is an explicit recourse to dispute settlement in the USMCA. Recognising that a mutually acceptable solution can often be found outside recourse to dispute settlement, a ‘sustained or recurring course of action or inaction’ that is inconsistent with a provision of the chapter on good regulatory practices is subject to dispute settlement. No other chapters of the agreement take precedence in case of inconsistency.¹¹⁶ Therefore, it seems likely that future PTAs may include dispute settlement provisions – either general or special – concerning their regulatory convergence chapter. This could make a difference concerning the implementation of internal mechanisms, which are largely binding.

Another concern that might appear – although it seems highly unlikely in the current development of investment law – is that the breach of regulatory convergence commitments is used as a basis to start investor–state dispute settlement (ISDS), if the PTA has an investment chapter that considers such possibility. Several ISDS arbitral tribunals have interpreted the obligation to provide a ‘predictable and transparent legal framework’ as a source of investors’ legitimate expectations as an element of the fair and equitable treatment (FET) or indirect expropriation standards, although the contours of these obligations are controversial (Henckels 2016). These tribunals have interpreted transparency obligations as requiring states to be forthcoming with information about intended changes in policy and regulations that may significantly affect investments so that the investor can adequately plan their investment and, if needed, engage the host state in dialogue about protecting their legitimate expectations.¹¹⁷ Although the purpose of measures subject to RIAs may not be to expropriate or treat a foreign investor unfairly, it could be considered to do so if the assessment does not appropriately balance policy goals with science- or evidence-based information (Sheargold and Mitchell 2016).¹¹⁸

It is also important to note that regulatory convergence does not work in the same way as FET, or indirect expropriation provisions. While the latter usually requires an ex post establishment infringement of investment treaty rules, regulatory convergence usually implies an ex ante analysis of a potentially challenged measure even before its legal existence, without necessarily examining the measure’s effects concerning foreign investors. Thus, even if a party undertakes an RIA and adopts a measure that has net benefits, such a measure could still be deemed inconsistent with investment treaty obligations if it disproportionately affects foreign entities (Polanco Lazo 2013).

17.7 Conclusion

Under different names, the introduction of notions of regulatory cooperation, improvement, coherence, or, more commonly, GRPs have become a new frontier in PTA governance, primarily built on the roots of domestic and international rulemaking principles.

The distinction between these typologies is not really relevant, as all these concepts aim to increase regulatory convergence, whether substantive or procedural, through different internal (‘bottom-up’) and external (‘top-down’) tools. While early PTAs only included some provisions on regulatory convergence, recent agreements consider specially dedicated chapters on this discipline, which are applied horizontally.

Most PTAs having dedicated chapters to this new discipline focus on different internal mechanisms of procedural convergence, which are largely binding. Starting with RIAs, these tools have grown over the years, including commitments to transparency, inter-agency coordination, stakeholder participation, periodic or retrospective review or evaluation of regulatory measures, use of plain language, and early information on planned regulatory measures. Future design for these areas could include more detailed and transparent descriptions of RIAs, increased opportunities to suggest improvements of regulations, more access to scientific, technical, economic, and other information used as a basis to develop the regulation, and mandatory establishment of CRCBs.

The most common mechanisms envisaged to achieve substantive convergence in PTAs are establishing treaty bodies (up to now with somewhat limited or indirect regulatory powers) and a host of general and specific cooperation activities. But it still remains a voluntary undertaking under which neither party is obliged to enter into particular regulatory cooperation activities, nor where either party may refuse to cooperate or withdraw from ongoing cooperation initiatives.

The scope of these chapters has also grown over time, extending from some regulations to almost all regulations. These provisions are also increasingly less of a ‘second-class’ commitment – meaning that PTAs are less likely to include a provision where other chapters prevail in case of inconsistency with the regulatory convergence chapter. Yet, many provisions on regulatory convergence in PTAs are aspirational or ‘best efforts’.

Up to now, the main drivers of this new discipline are the US, EU, the UK, and countries from the Asia-Pacific rim like Australia, Chile, Japan, New Zealand, and Singapore. China is conspicuously absent and probably hesitates to embrace international commitments on regulatory convergence for political reasons (Lin and Liu 2018).

The notion of regulatory convergence raises questions of influence and legitimacy. In a world of pronounced regulatory asymmetries, will the regulations of the most powerful actors not tend to prevail? Will this increase the tendency towards harmonising or adopting dominant regulatory standards issued by the small group of rulemaking nations with greater negotiating power (‘regulatory hegemons’)? (Kelsey 2011). Does it run the risk of resulting in the dissemination of ‘the regulatory gospel’ of ‘regulation-exporting’ states? (Slaughter 2005).

Developing countries may prefer to ascribe to substantive or regulatory convergence for other reasons, like participation in their development, as it has happened at the regional level with ASEAN and the PA. On the other hand, regulatory convergence by ‘developed economies’ may not lead to emerging markets automatically upgrading to higher standards, either because they were not part of the negotiation that created the standard or because they are simply unable to meet and implement them, in the absence of capacity-building provision for North–South agreements.

If the introduction of regulatory convergence disciplines in PTAs is successful, it is likely to inform the evolution of trade global norms and play a significant role in a wide range of subject areas subject to PTA disciplines, including TBTs, intellectual property rights, financial services, investment, competition policy, consumer rights, and the recognition of professional service qualifications (Stoler et al. 2014)

But the mere existence of regulatory divergence need not always be problematic. Alternative legal solutions adopted by countries might be neutral from an efficiency viewpoint (Mattei et al. 2000). Legitimate cross-country differences in collective preferences, attitudes to risk and uncertainty, legal traditions, etc., can also explain such differences (Chirico and Larouche 2013). Fostering deeper forms of regulatory convergence could generate high costs, particularly if enacted in asymmetrical North−South relations and agreements. Finally, addressing the competition-impairing effects of regulatory differences can be administratively and politically more challenging than eliminating traditional impediments to trade and investment.