Policy Research Working Paper 9283 The Evolution of Deep Trade Agreements Aaditya Mattoo Nadia Rocha Michele Ruta Development Research Group East Asia and the Pacific Region, Office of the Chief Economist Macroeconomics, Trade and Investment Global Practice June 2020 Policy Research Working Paper 9283 Abstract This paper presents new data on the content of preferential at facilitating flows of services, goods, and capital; (iii) deep- trade agreements. The data contain detailed information ening commitments have been accompanied by an increase on the 18 policy areas most frequently covered in prefer- in regulatory requirements, namely on enforcement; (iv) ential trade agreements, focusing on the stated objectives, developing countries tend to have fewer commitments in substantive commitments, and other aspects such as trans- preferential trade agreements, with larger gaps in areas such parency, procedures, and enforcement. Several new stylized as labor and the environment; and (v) preferential trade facts emerge: (i) preferential trade agreements have reduced agreements are more similar within blocs, but similarity can trade-weighted average tariff rates to less than 5 percent be significant even across blocs. The paper also discusses the for more than two-thirds of countries; (ii) the number of challenges of quantification of preferential trade agreements commitments in preferential trade agreements has increased “depth” and its effects and proposes a research agenda for over time, particularly since the 2000s and in areas aiming future work on trade agreements. This paper is a joint product of the Development Research Group; Office of the Chief Economist, East Asia and the Pacific Region; and the Macroeconomics, Trade and Investment Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at amattoo@worldbank.org; nrocha@worldbank.org; and mruta@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Evolution of Deep Trade Agreements1 Aaditya Mattoo, Nadia Rocha, Michele Ruta World Bank, Washington DC, United States 1. Introduction This paper takes a first look at new data on the content of all Preferential Trade Agreements (PTAs) that have been notified to the World Trade Organization (WTO), and highlights the emergence of Deep Trade Agreements (DTAs).2 The detailed description of the data and the methodology used to collect it are discussed in the Handbook of Deep Trade Agreements (Mattoo, Rocha and Ruta, 2020) -henceforth, “the Handbook”. DTAs are reciprocal agreements between countries that cover not just trade but additional policy areas, such as international flows of investment and labor, and the protection of intellectual property rights and the environment, among others. While these legal arrangements are still referred to as trade agreements, 1 This paper is based on the Introduction of the Handbook od Deep Trade Agreements (Mattoo, Rocha and Ruta, 2020). We would like to thank Petros Mavroidis for inviting us to present our work at Columbia Law School and Tom Prusa for his comments. We are highly indebted to Alvaro Espitia for providing background research for this paper and more broadly for his outstanding contribution to the production of the Handbook. Kazusa Yoshimura and Edith Laget worked on the quantification exercise using machine learning techniques in Section 4. 2 In the international economics and law literature, PTA is an umbrella term encompassing several types of reciprocal agreements between trading partners: Regional Trade Agreements (RTAs), Free Trade Agreements (FTAs), and Customs Unions (CUs). This definition differs from that of the World Trade Organization (WTO), which defines PTAs as agreements that grant unilateral (i.e., non-reciprocal) trade preferences such as the Generalized System of Preferences schemes, under which developed countries grant preferential tariffs to imports from developing countries. This study, following the definition from international economics and law, uses the term PTA to refer to all types of trade agreements, both within and across regions, and uses DTA to refer to PTAs that contain provisions aimed at deepening economic integration between trading partners. their goal is integration beyond trade or deep integration. DTAs aim at establishing five “economic integration” rights: free (or freer) movement of goods, services, capital, people, and ideas. DTAs also include enforcement provisions that limit the discretion of importing governments in these areas, as well as provisions that regulate the behavior of exporters. Preferential trade agreements have always been a feature of the world trading system but have become more prominent in recent years. The number of PTAs has increased from 50 in the early 1990s to roughly 300 in 2019. All WTO members are currently party to one, and often several, PTAs. While WTO rules still form the basis of most trade agreements, PTAs have in some sense run away with the trade agenda. Traditional trade policy areas, such as tariff reduction or services liberalization, are now more frequently negotiated in regional contexts rather than at the WTO, with PTAs often going beyond what countries have committed to at the WTO. The result is that PTAs have expanded their scope. While the average PTA in the 1950s covered 8 policy areas, in recent years they have averaged 17. In other words, there is some preliminary evidence that PTAs are becoming DTAs, both on the intensive margin (specific commitments within a policy area) and the extensive margin (number of policy areas covered). In this paper, we do not draw a sharp distinction between DTAs and other PTAs. Rather, the aim is to demonstrate the progressive deepening of PTAs. Deep trade agreements matter for economic development. The rules embedded in DTAs, along with the multilateral trade rules and other elements of international economic law, such as International Investment Agreements, influence how countries (and, hence, the people and firms that live and operate within them) transact, invest, work, and, ultimately, develop. Trade and investment regimes determine the extent of economic integration, competition rules affect economic efficiency, intellectual property rights matter for innovation, and environmental and labor rules contribute to social and environmental outcomes. It is, therefore, vital that rules and commitments in DTAs are informed by evidence and shaped more by development priorities than by international power dynamics or domestic politics. An impediment to this goal is that data and analysis on trade agreements have not captured the new dimensions of integration, which makes it difficult to identify the content and consequences of DTAs. The new data take a first step towards filling this important gap in our understanding of international economic law and policy. It presents detailed information on the content of the 18 policy areas most frequently covered in PTAs, focusing on the stated objectives, substantive commitments, and other aspects such as transparency, procedures and enforcement. In terms of the coverage of policy areas and the granularity of information within each area, this is the most comprehensive effort to date. The new data will inform experts and policy makers in their efforts to design, negotiate and take advantage of DTAs that promote development. This information will also enable researchers to develop indicators on the depth of trade agreements in different policy areas, assess the similarities between these arrangements, and benchmark countries’ DTAs relative to their partners. It will also help identify the rules that benefit only participants and those that have large spillover effects on non-participants or excluded countries. Finally, the new data and analysis in this study will allow researchers to identify areas where there is de facto convergence across different players, thus facilitating the adoption of commonly agreed multilateral rules. 2 This information will lay the groundwork for new research in international economics and other fields. A large body of economic literature has looked at the effects of PTAs on international trade flows and on welfare.3 However, this literature has two important limitations.4 On the theoretical side, the study of PTAs is mostly based on Vinerian5 logic, which focuses exclusively on tariffs, thus by construction excluding deep integration issues. On the empirical side, attempts to quantify the effects of PTAs suffer from a measurement error problem, as studies generally rely on dummies to identify trade agreements or distinguish between broad categories of agreements such as FTAs or CUs. The new data will help theorists to model DTAs and help empirical economists to properly identify their effects. Beyond economics, the data will inform research in other fields, primarily international law and international relations, on important issues such as the commonality or divergence of the rules set in PTAs and how they could evolve in the future. The new data is the result of collaboration among the World Bank, the International Trade Center (ITC), the Organization for Economic Development and Cooperation (OECD), the WTO, and experts from academic institutions. It builds on previous research by the World Bank and others. A first database on the content of deep trade agreements was published in 2017 with the goal of documenting how the policy areas covered by PTAs had increased over time (Hofmann et al. 2019). This data set allowed researchers to construct a first series of indicators which capture the scope of trade agreements; i.e., what policy areas they cover. We refer to this as the extensive margin of PTA depth. Based on this first data set, several research papers then looked, respectively, at the impact of deep trade agreements on trade, global value chains, foreign direct investment, and the effect of breaking up such agreements.6 The data have also been extensively employed for policy advice by the World Bank in several developing countries in Africa, Latin America, East Asia and the Balkans. The new data that we briefly review in this paper and that are analyzed in detail in the individual chapters of the Handbook offer insights into a different dimension of PTAs depth. They capture the detailed commitments to establish and preserve the rights to economic integration, and the procedures, institutions and enforcement mechanisms that countries set up to make deep integration work. The focus is therefore not on the extensive margin of integration (number of policy areas that are covered by the agreement), but on its intensive margin (the specific commitments within a policy area). While there are a number of individual studies that have documented the deepening of PTAs in specific areas, two major data collection projects—Dür et al. (2014) and Acharya (2016)—also aimed at documenting the specific commitments for a group of policy areas covered in PTAs. Both efforts have important merits. Dür et al. (2014) covered a large set of PTAs, including those that have been notified to the WTO but are no longer in force. Acharya (2016) provided a series of databases on the content of PTAs that go beyond specific policy areas and cover emerging issues such as e-commerce or the rules on dispute settlement in PTAs. Relative to these data collection projects, the new data set is more comprehensive, both in terms of the number of policy areas covered and in terms of the information on detailed disciplines in each area. 3 See, e.g., Freund and Ornelas 2010, Limao 2017. 4 See, e.g., Baldwin 2010, Mattoo et al. 2017. 5 Viner 1950. 6 Mattoo et al. 2017, Mulabdic et al. 2017, Laget et al. 2018, Laget et al. 2019. 3 The paper is organized as follows. Section 2 describes the scope and methodology underlying the research agenda on deep trade agreements. Section 3 highlights a novel set of stylized facts that can be inferred from a first look at the new data, while Section 4 offers some insights into future applications and areas for analysis. Concluding remarks follow. 2. Scope and methodology The number of policy areas covered by PTAs has increased in the last two decades. Until the late 1990s, when the number of PTAs started increasing, the majority of new agreements covered fewer than 10 policy areas. Since the 2000s, most new PTAs have covered between 10 and 20 policy areas, with some having even more than 20 (Figure O1). In a study of 28 trade agreements signed by the US and the EU, Horn et al. (2010) identify up to 52 policy areas that have been covered by at least one of the agreements. The inclusion of new policy areas in PTAs is not random. As shown in Mattoo et al. (2017), trade agreements covering few policy areas generally focus on traditional trade policy, such as tariff liberalization or customs (Table O1). Agreements with broader coverage (between 10 and 20 policy areas) tend to include trade-related regulatory issues, such as subsidies or technical barriers to trade. Finally, agreements with more than 20 provisions often include policy areas that are not directly related to trade, such as labor, environment and movement of people. Figure O1: Number of policy areas covered in PTAs, 1970-2017 25 300 Cumulative Number of Agreements 250 20 Number of Agreements 200 15 150 10 100 5 50 0 0 1990 2016 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2017 More than 20 Before 10 and 20 Less than 10 Cumulative Source: Authors’ calculations based on Hofmann et al. 2019. 4 Table O1: Share of policy areas for different PTAs Between 10 and Less than 10 More than 20 No. Provisions 20 Tariffs on manufacturing goods 97% 100% 100% Tariffs on agricultural goods 96% 100% 100% Export taxes 73% 81% 95% Customs 67% 95% 100% Competition policy 58% 73% 88% State aid 39% 69% 88% Anti-dumping 35% 88% 98% Countervailing measures 22% 77% 98% Statistics 20% 0% 23% TRIPS 18% 75% 98% STE 18% 69% 68% TBT 17% 73% 95% Movement of capital 15% 68% 93% GATS 14% 67% 98% SPS 12% 72% 98% Public procurement 12% 59% 80% IPR 6% 56% 75% Environmental laws 3% 14% 83% Labor market regulations 3% 13% 75% Investment 2% 58% 75% TRIMS 2% 42% 73% Visa and asylum 2% 37% 57% Industrial cooperation 2% 5% 33% Social matters 2% 5% 30% Agriculture 1% 10% 45% Energy 1% 8% 40% Data protection 1% 5% 20% Anticorruption 1% 5% 18% SME 1% 4% 25% Regional cooperation 1% 3% 15% Taxation 1% 2% 30% Approximation of legislation 1% 2% 25% Political dialogue 1% 1% 8% Research and technology 0% 6% 38% Public administration 0% 6% 5% Consumer protection 0% 5% 38% Mining 0% 5% 13% Education and training 0% 4% 33% Information society 0% 4% 15% Innovation policies 0% 4% 5% Illegal immigration 0% 3% 23% Illicit drugs 0% 3% 3% Economic policy dialogue 0% 2% 43% Cultural cooperation 0% 2% 38% Financial assistance 0% 2% 25% Audiovisual 0% 2% 18% Terrorism 0% 2% 8% Money laundering 0% 2% 3% Health 0% 1% 38% Human rights 0% 1% 3% Nuclear safety 0% 0% 15% Civil protection 0% 0% 5% Source: Mattoo et at. 2017. The policy areas studied in the Handbook are those that appear most frequently in trade agreements. They include (a) a set of 18 policy areas that are covered in 20 percent or more of trade agreements notified to the WTO (Figure O2): (b) tariffs on industrial and agricultural goods, which are covered by all trade agreements; (c) customs and export taxes, which are regulated in more than 80 percent of PTAs; (d) services and movements of capital, which are regulated in roughly half of the PTAs; and (e) environmental and labor issues, which are covered by around 20 percent of all trade agreements. Other issues that are sometimes (although infrequently) regulated in trade agreements, such as education, nuclear safety and human rights, are not included in the Handbook and could be the subject of future research. The focus on 5 individual areas helps us to identify specific policies that are the object of negotiation but may obscure cross-cutting issues—such as electronic commerce—that may be disciplined under multiple policy areas. Figure O2: Number of policy areas covered in PTAs, by policy 100% 90% 80% Share of agreements 70% 60% 50% 40% 30% 20% 10% 0% Antidumping and Countervailing… Sanitary and Phytosanitary… Energy Taxation Agriculture Investment Statistic Consumer Protection Competition Policy Movement of Capital Data Protection Trade Facilitation and Customs SMEs Innovation Policies Terrorism FTA Industrial Mining Audiovisual Subsidies Approximation of Legislation Public Procurement Environmental Laws Research and Technology Industrial Cooperation Cultural Cooperation Health Regional Cooperation Illicit drugs Political Dialogue Civil Protection Human Rights FTA Agriculture Export Taxes Intellectual Property Rights Social matters Education and Training Technical Barriers to Trade Services Visa and Asylum Anti-corruption Financial Assistance Illegal Immigration Public Administration Nuclear Safety Money laundering Labor Market Regulations Economic Policy Dialogue Information Society State Owned Enterprises Most common policy areas Source: Authors’ calculations based on Hofmann et al. 2019. The classification of policy areas used in Figure O2 deviates slightly from the one of Horn et al. (2010).7 Specifically, for the Handbook, we decided to include rules of origin, a policy area that was absent from the Horn et al. (2010) classification, and to treat as a single policy area: (a) trade remedies, which include anti-dumping and countervailing measures; (b) investment, which includes the areas covered under the WTO’s Trade Related Investment Measures, or TRIMs; and (c) intellectual property rights (IPR), which include the areas covered under the WTO’s Trade Related Intellectual Property Rights, or TRIPs. Trade agreements are generally assessed in terms of the market access they create. Given the complexity of policy areas that are covered by DTAs, the metric of market access—while still important—appears inadequate. In this paper, we propose to define deep trade agreements as international arrangements that aim to regulate three (partially overlapping) sets of policy areas (Figure O3). • First, the core policy areas included in DTAs aim to establish five economic integration rights: free (or freer) movement of goods, services, capital, people and ideas.8 The policy areas that directly impact these flows include: (a) tariffs and export taxes, which affect the movement of goods; (b) services, which regulate services trade flows; (c) investment and movement of capital, which affect the movement of capital; (d) visa and asylum, which regulate the movement of people; and (e) intellectual property rights, which influence the flows of ideas. 7 The Horn et al. (2010) classification was used to collect data on the extensive margin of PTA depth. 8 We use the words “aim to establish” rather than “establish” for two main reasons. First, DTAs may cover only a subset of integration rights. Second, provisions may not be justiciable. A contribution of the new data is to identify the extent to which integration rights are established in PTAs. 6 • Second, DTAs also include policy areas that aim to support these economic integration rights by limiting government discretion. Actions by importing governments that limit international flows can be taken at the border and behind the border and are often of a regulatory nature. The policy areas that fall in this category are: (a) customs; (b) rules of origin; (c) trade remedies; (d) public procurement; (e) technical barriers to trade (TBT); (f) sanitary and phytosanitary measures (SPS); (g) state-owned enterprises (SOEs); (h) subsidies; and (i) competition policy.9 • Third, DTAs cover policy areas that aim to enhance social or consumer welfare by regulating the behavior of exporters. Policy areas such as environment and labor impose obligations on exporters to further consumer or social interests in importing countries. Rules in areas such as competition, SOEs, and subsidies can have a dual aspect: in addition to regulating action that undermines economic integration rights, they can aim to address distortionary actions that lower economic efficiency thus hurting consumer or social welfare. Figure O3: A classification of policy areas in DTAs For each policy area, the experts followed a uniform approach to coding.10 The coding templates encompass several common headings such as objectives and definitions, institutional framework, enforcement mechanism, plus a series of discipline-specific questions. Under each heading, questions on specific provisions in the agreement are formulated so that they can be answered with Yes/No. For some 9 Some of these provisions apply only to cross-border trade in goods (e.g., customs, TBT and SPS). Others can also apply to cross-border trade in services (e.g., public procurement and competition policy). In some cases, services- related provisions are included separately in a services agreement. 10 One exception is preferential tariffs. Differently from the other policy areas, tariff commitments apply at the product level. The information for this area is therefore collected at the country-pair-product level. For rules of origin a sub-sample of agreements in Latin America and East Asia, the data set on regime-wide provisions is accompanied by a mapping of the rules of origin that apply at the product level. 7 policy areas, additional information is provided at the provision level, including (a) the relationship between the coverage of the disciplines on and the corresponding regulation in the WTO; (b) the level of enforceability of each provision;11 (c) whether the specific commitment can be applied discriminatorily or whether it is de facto non-discriminatory. Finally, when applicable, for example in services and government procurement, the coders included information at the sectoral level on exclusion of certain sectors from an agreement, or the applicability of an agreement to a specific industry. The analysis covers the realm of PTAs that are in force and notified to the WTO as of end-2017. The basis of the coding analysis is the legal text of the trade agreements and the relevant annexes that accompany the agreement (and have been notified to the WTO). This approach comes with two main limitations that should be clear to the user of the database. First, the focus on the legal text of the agreement implies that secondary law (the body of law that derives from the principles and objectives of the treaties) has not been coded. This is a concern particularly when assessing the depth of integration of the EU, since in most policy areas covered in the Handbook, EU institutions have used secondary law such as regulations, directives, and other legal instruments to pursue integration.12 Second, the focus on the legal text also excludes from consideration issues of implementation of the trade agreement into national laws and regulations or subsequent annexes that the parties might agree on which are not reported to the WTO. These are important areas for future research. Despite the similarity in the coding approach, policy areas differ widely from each other. First, some policy areas are inherently more complex than others and their description requires a larger number of questions to reflect the more detailed provisions. IPR has the highest number of provisions (120), while labor has the lowest (18). Second, some policy areas focus primarily on substantive provisions: specific commitments on integration, such as market access commitments, and specific obligations such as harmonization of standards. Others tend to have a larger number of procedural provisions, such as transparency provisions and procedural requirements. Table O2 provides an overview, showing the heterogeneity across policy areas in these different dimensions and identifying the set of “substantive” provisions as those that require specific integration/liberalization commitments and obligations. Table O2: Number of substantive and other provisions per policy area in all PTAs notified to the WTO Intellectual Property Rights Sanitary and Phytosanitary Technical Barriers to Trade Labor Market Regulations State Owned Enterprises Discipline Trade Facilitation and Countervailing Duties Movement of Capital Environmental Laws Public Procurement Competition Policy Anti-dumping and Visa and Asylum Total provisions Rules of Origin Export Taxes Investment Measures Subsidies Customs Category Services Objectives 1 1 8 1 2 2 2 2 2 2 23 Scope and definitions 1 16 11 7 2 17 2 4 2 1 10 8 10 91 11 The legal enforceability of the PTA provisions is coded according to the language used in the text of the agreements. It is assumed that commitments expressed with clear, specific, and imperative legal language can more successfully be invoked by a complainant in a dispute settlement proceeding, and therefore are more likely to be legally enforceable. In contrast, unclearly formulated legal language might be related with policy areas that are covered but that might not be legally enforceable. 12 Note that the figures and tables in this paper refer to the EU as a single entity (i.e. the European Union agreement and enlargements are excluded) and report data for EU PTAs with third countries where this concern does not apply. 8 Transparency 4 9 3 13 3 8 6 3 10 7 4 5 1 76 Substantive commitments 17 19 13 59 3 6 2 19 20 4 3 8 11 27 20 12 243 Liberalization/Integration 14 8 11 19 4 1 3 4 3 13 80 Conditions/Obligations 3 11 2 40 3 2 1 16 16 1 3 8 11 27 7 12 163 Procedural requirements 17 8 12 3 28 10 3 28 2 2 17 130 Enforcement mechanism 1 3 8 22 1 2 4 5 7 5 4 1 63 Sectoral coverage 2 1 2 5 33 9 8 60 Specific coverage 2 1 13 9 8 2 8 1 44 Exceptions 5 6 2 35 4 4 3 1 3 63 Safeguards 1 10 31 1 43 Special and differentiated treatment 7 2 2 11 Institutional framework 1 1 2 2 2 6 2 11 2 1 2 32 Cooperation 2 3 1 8 3 3 1 1 1 5 4 1 33 Miscellaneous 9 6 5 2 2 1 25 Total provisions 46 64 57 95 120 30 52 51 34 59 100 36 54 35 48 38 18 937 Source: Authors’ calculations based on Mattoo et al. (2020). We also make an effort to identify the set of provisions within each policy area that are essential to achieve the objectives of the agreement. The provisions we refer to as “essential” comprise the set of substantive provisions plus the disciplines among procedures, transparency, enforcement or objectives, which are viewed as indispensable and complementary to achieving the substantive commitments. Non-essential provisions are referred to as “corollary.” A caveat is that this exercise is based on the experts’ knowledge and, hence, is subjective. However, this approach has the advantage of limiting the dimensionality of the data in an informed way.13 3. Stylized facts A number of new stylized facts emerge from a preliminary analysis of the data. Each of the chapters in the Handbook provides a first look at the data by policy area. In this paper, we present a bird’s-eye view of the entire data set put together by the experts. Given the differences among policy areas and among provisions within each policy area, this approach presents many quantification challenges, which are discussed below. In this section, we rely on simple counts of the provisions and on coverage ratios14 to investigate the evolution of the content of deep trade agreements. The underlying assumption in this approach is that deeper trade agreements imply a larger number of provisions. As shown in Espitia et al. (2019), liberalization in PTAs has reduced trade-weighted average tariff rates to less than 5 percent for more than two-thirds of countries (Figure O4). While there are still pockets of high protection in some countries, most notably lower-income economies, PTAs have been broadly successful in committing national governments to maintaining low tariffs. Trade-weighted applied tariffs are, on average, 2.3 percentage points lower than average most-favored nation (MFN) rates, with gaps of greater than 6 percentage points for countries like Tunisia, Morocco, Bosnia and Herzegovina, Namibia and Lao PDR. So, while from an efficiency perspective, preferential tariff liberalization is inferior to non- preferential liberalization, the commitments countries have taken in the network of preferential trade 13 A statistical approach on how to assess the importance of specific provisions included in the different policy areas in explaining trade outcomes is presented in section 4. 14 The coverage ratio is defined as the share of provisions for a policy area contained in a given agreement relative to the maximum number of provisions in that policy area or agreement. 9 agreements may provide a safety net at a time when trade tensions are escalating and some countries are disregarding their multilateral commitments. Figure O4: Tariffs in PTAs and MFN tariffs Source: Espitia et al. 2019. The number of commitments that governments have taken in trade agreements, particularly since the early 2000s, has increased over time. Figure O5 shows how the coverage ratio has changed over time for the 17 policy areas analyzed in the Handbook (all but tariffs) in aggregate. With only few exceptions, the majority of new PTAs signed after 2000 have a coverage ratio higher than 25 percent. This stands in sharp contrast to the trade agreements signed in the 1980s and 1990s, when coverage ratios were below 15 percent and, in many cases, even below 5 percent. The reduction in tariffs accomplished through preferential trade liberalization, together with the increased depth of agreements over time, suggests that countries that are willing to cut tariffs reciprocally may also be willing to accept deeper mutual commitments in other areas. Figure O5: Number of agreements over time vs average coverage ratio 25 300 20 250 Number of Agreements 200 Cumulative 15 150 10 100 5 50 0 0 1968 1960 1962 1964 1966 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 More than 25% Between 15 and 25% Between 5 and 15% Less than 5% Cumulative 10 Note: Coverage ratio refers to the share of provisions contained in a given agreement relative to the maximum number of provisions. European Union agreement and enlargements excluded. While the overall number of provisions is suggestive, it can hide important elements of the evolution of deep trade agreements. First, as discussed above, some provisions imply substantive commitments while others concern broad objectives, definitions or procedural matters. Second, deep trade agreements, as defined in the Handbook, not only concern themselves with market access in goods, but also aim to establish freedom of mobility for services, capital, ideas and people, as well as regulating policy areas that have an impact on consumer and/or social welfare, such as labor and the environment. To gain a better understanding of how the commitments in PTAs have changed over time, we look at the evolution of coverage ratios by policy area. Figure O6 shows that the coverage of essential disciplines in PTAs has increased over time across all policy areas. This is most clearly the case for the policy areas aimed at facilitating the flows of goods (customs and trade facilitation), capital (investment and movement of capital) and services. IPR and movement of people (visa and asylum) also saw a steady by less remarkable increase in essential commitments over time. Along with economic integration rights, PTAs increasingly include essential commitments in policy areas that support these rights or impose obligations on exporters. The ones that appear to stand out are subsidies, competition and SOEs, areas that are either excluded from the WTO or for which reform of multilateral rules is considered difficult. Interestingly, while essential commitments in labor have largely increased in recent years, this happened to a lesser extent for provisions on the environment. 11 Figure O6: Coverage ratios by policy area, over time Note: Coverage ratio by policy area refers to the share of provisions for a policy area contained in a given agreement relative to the maximum number of provisions in that policy area. Years refer to entry into force date. European Union agreement and enlargements excluded. The presumption is that the increase in the essential disciplines in deep PTAs has been driven by countries taking on more substantive commitments over time. Indeed, Figure O7 shows that this is the case, but it also uncovers interesting insights about the evolution of non-substantive commitments. We focus on the three (numerically) most relevant non-substantive provisions: procedural rules, transparency and enforcement provisions. The deepening of substantive commitments has been accompanied by an increase in the number of corollary provisions, suggesting that achieving deeper commitments may require more procedural rules for implementation, transparency, and enforcement. A second insight is that, while these disciplines are all necessary to render substantive commitments in trade agreements effective, they have evolved differently in recent years. Starting in the early 2000s, the relevance of enforcement provisions in DTAs has increased disproportionally relative to procedural and transparency provisions. The growing enforcement capacity of DTAs may help explain the success of these institutional arrangements as tools for deep integration. 12 Figure O7: Substantive provisions and a breakdown of non-substantive provisions in PTAs, over time 42 Average coverage by category (%) 35 27 27 22 19 20 15 16 14 13 13 12 10 10 8 8 6 6 4 Before 1995 1995-1999 2000-2004 2005-2009 2010-2017 Transparency Procedural requirements Enforcement mechanism Substantive Note: Coverage ratio refers to the share of provisions for a policy area contained in a given agreement relative to the maximum number of provisions in that policy area. Years refer to entry into force date. European Union agreement and enlargements excluded. When we break down the trade agreements by level of development of the signatories, we observe two facts. First, the deepest PTAs are those involving developed economies, followed by PTAs between developed and developing economies. PTAs between developing countries are the shallowest. Indeed, there is a sizeable gap between average coverage ratios for the latter group of PTAs relative to the first two (Figure O8). This could reflect a focus of negotiations on tariffs and traditional trade barriers, which are still high for several low-income economies. Second, in terms of composition, PTAs between developed countries and those between developed and developing economies include similar shares of provisions establishing economic integration rights, supporting these rights and aiming to regulate exporters (Figure O9). PTAs between developing countries are shallower across the board, with a stronger gap in areas such as environment and labor that aim at improving social welfare. Figure O8: Inclusion of substantive commitments in PTAs, by level of development 27.7 23.2 Average coverage ratio (%) 18.8 18.9 19.5 13.2 13.9 13.7 10.1 10.9 8.4 8.2 9.1 5.8 4.8 Before 1995 1995-1999 2000-2004 2005-2009 2010-2017 Developing - Developing Developing - Developed Developed - Developed Note: Coverage ratio refers to the share of provisions for a policy area contained in a given agreement relative to the maximum number of provisions in that policy area. Years refer to entry into force date. European Union agreement and enlargements excluded. 13 Figure O9: Coverage ratio by type of policy and level of development 26.0 23.5 22.4 Average coverage ratio (%) 22.1 20.8 20.7 11.9 9.5 4.9 Establishing Integration Rights Protecting Integration Rights Expanding Consumer Rights Developing - Developing Developing - Developed Developed - Developed Note: Coverage ratio refers to the share of provisions for a policy area contained in a given agreement relative to the maximum number of provisions in that policy area. Years refer to entry into force date. European Union agreement and enlargements excluded. We next analyze the depth of trade agreements by country. Here, we focus on the substantive commitments.15 As several countries have multiple agreements with different levels of depth, we present the average number of substantive commitments per country in panel a of Figure O10 and the maximum number in panel b of Figure O10. The main takeaway is that developing countries in Sub-Saharan Africa, Middle East and North Africa, South America, South Asia and, to a lesser extent, East Asia tend to have fewer substantive commitments in trade agreements relative to advanced economies. The few exceptions include countries in South America that are signatories of the Pacific Alliance and other developing economies that have signed deep trade agreements with an advanced trade partner, such as Mongolia with Japan and Caribbean countries with the EU. In terms of depth as measured here, North America and Europe are the most integrated regions, through NAFTA and its successor agreement, and through the agreements the EU has signed with neighboring countries. East Asia is a region with a mixed profile: the network of ASEAN agreements includes most countries but tends to have fewer substantive commitments relative to North America and Europe, except for some countries such as Vietnam, which have signed onto the Comprehensive Agreement for the Trans-Pacific Partnership (with a coverage ratio of 61 percent). Figure O10: Substantive provisions in PTAs by country 15 Appendix Tables O1 and O2 provide other indicators by PTA and by country. 14 Panel a. Average number of provisions Panel b: Maximum number of provisions 15 With the increasing depth and complexity of trade agreements, both similarities and dissimilarities between PTAs could potentially increase. Older agreements that covered only preferential tariff liberalization and other aspects of market access tended to be very similar. As PTAs now cover more ground, there can be provisions that are included in two agreements, making them more similar, or there can be provisions that are covered by one PTA but not by another, making them more dissimilar. To capture this information, we construct a similarity index for DTAs, calculated as the ratio between the number of provisions for which two agreements have a “yes” (a measure of similarity) and the total number of provisions covered by the agreements, independently of whether they have the same answer or not. The closer the similarity index is to one (or zero), the more (or less) similar are the two DTAs; i.e., include the same type(s) of provisions. Figure O11 plots the degree of similarity for the PTAs signed by the three major trading blocs: the European Union, United States, and Japan. Each color represents a DTA signed by a third country with the US (dark blue), EU (light blue) or Japan (red). The size of the bubbles represents the depth of the agreements, measured as the number of provisions covered. Each agreement is connected to the one which is most similar within a trading bloc. The figure also links the three trading blocs, by connecting the pair of agreements that are the most similar between two blocs. As expected, within each bloc, DTAs are highly similar: up to 0.89 for the US (US-Peru; US-Colombia), up to 0.80 for the EU (EU-Republic of Moldova; EU-Ukraine), and up to 0.75 for Japan (Japan-Indonesia; Japan-Mongolia). This fact often reflects a “template effect,” where the EU, US and Japan tend to negotiate based on a template offered to third countries. Interestingly, the similarity of DTAs is relatively high even across blocs, although lower than within blocs. For example, the EU-Korea agreement shares more than 50 percent of the provisions with the Japan-Switzerland agreement (similarity index of 0.54) 16 and with the US-Peru agreement (similarity index of 0.51). These results indicate that concerns about the fragmentation of the global trade system have some foundation (i.e., they do not share almost half of provisions), but also point to substantial similarities—based on which multilateral rules can be agreed upon. Figure O11: Similarity of agreements Note: The size of the bubbles represents the depth of a trade agreement, as captured by the number of provisions included in the agreement. Each edge connects an agreement with one that is most similar. Light blue bubbles represent EU agreements with non-EU countries, dark blue represent US agreements, and red represent Japan agreements. 4. The challenge of quantifying the effects of DTAs Quantification of the effects of DTAs poses a serious challenge. DTAs cover heterogeneous areas: tariffs, contingent protection, export taxes, customs procedures, technical barriers in goods; a wide range of restrictions across modes in services; investment measures, subsidies, procurement, state enterprises, competition policy affecting both trade and investment in goods and services, visas and asylum, and a range of regulatory requirements affecting labor mobility; and a variety of policies affecting the protection of intellectual policy rights and the environment. How can the diversity of policies be quantified and aggregated within separate areas? How can we aggregate across the different areas? We briefly discuss here two approaches to quantification—directly constructed indices and indirectly estimated measures— and some analytical issues going forward. Directly constructed indices The count variables and coverage ratios presented in the previous section are the simplest directly constructed indices of depth. They provide an immediate view of how commitments in PTAs have changed over time, across countries and for subsets of provisions. Still, aggregate indicators based on some form 17 of counting disregard the fact that DTAs cover multiple policy areas and sectors and that the “value” of each provision is unlikely to be the same even within the same policy area. In some cases, it may be possible to construct a hierarchy of measures. For example, in the areas of services and government procurement, provisions could be divided into three tiers. Tier 1 would comprise provisions ensuring market access and national treatment at entry. Tier 2 would comprise provisions on post-entry operation; e.g., preferences or offsets. Tier 3 would comprise procedural rules limiting discretion in licenses and awards. The construction of an index could then be lexicographic, in that we would consider first only differences between countries or sectors in Tier 1 and move to subsequent tiers only to break ties. Such an approach is ideally suited to the construction of an ordinal rather than cardinal (i.e., qualitative rather than quantitative) measure. A pragmatic approach to overcoming some of the constraints to constructing representative indices is to rely on experts’ judgment. This is the method adopted in the Handbook. The individual chapters offer a disaggregated set of stylized facts for each policy area using count variables, coverage ratios and the individual assessment of the authors of the key provisions in each policy area. We have already discussed the distinction between substantive and essential provisions. Some chapters go even beyond these categories. For instance, the chapter on SOEs (Rubini and Wang, 2019) identifies four commitments concerning issues of ownership, discrimination, subsidization and anti-competitive behavior as key. The chapter on technical barriers to trade (Espitia et al. 2019) identifies a subset of seven commitments which are key to achieving deep integration in the area of technical regulations. This type of information can be used in the estimation exercises we discuss below, as it allows the researcher to address problems associated with large numbers of possible variables at hand, such as multicollinearity (i.e., the high correlation between the different provisions within and across policy areas). Indirectly estimated measures These measures are obtained by estimating the impact of the provisions on a variable of interest. For example, we could infer the value of individual provisions by estimating their impact on bilateral trade, controlling for other influences. In principle, each binary element in the relevant DTA areas could be included in a country-product import regression as a right-hand variable while controlling for applied policies, including tariffs and non-tariff measures. Similar methods have been used to estimate the Overall Trade Restrictiveness Index.16 However, even for trade in goods we have limited degrees of freedom, and in other areas (such as services), we do not have sufficiently fine outcome data. In these areas, it may be necessary to take a hybrid approach, based on first constructing more aggregated indices. Another approach to quantify the effects of DTAs and build indicators of depth is to use new statistical methods. As a first example, we employ machine learning techniques to detect the influential variables/provisions in DTAs for trade.17 Machine learning is a generic term referring to a wide variety of algorithms which detect a certain pattern from a large data set, often referred to as “Big Data,” and make predictions based on that pattern. In this case, we use a method called Random Forest (RF) to calculate 16 Kee, Nicita and Olarreaga, 2009. 17 This exercise has been carried out in collaboration with Kazusa Yoshimura and Edith Laget. Parallel work by Breinlich et al. (2020) also uses machine learning techniques to precisely quantify the impact of individual provisions in trade agreements on trade flows. 18 the importance of each variable/provision for international trade flows.18 Specifically, we run as a first step a structural gravity model with the standard set of fixed effects and then use the residuals as the left- hand variable in the RF. Figure O12 shows the boxplot of scores calculated by the RF of variables/provisions in PTAs belonging to the 17 (non-tariff) policy areas analyzed in the Handbook.19 The areas are colored according to their categorization into the three main groups illustrated in Figure O3; red indicates policies that establish economic integration rights, blue is assigned to those supporting these rights, and green to those that promote welfare. Each box shows the range of the first (25 percent) and third (75 percent) quartiles, and the black line in the box shows the median of the scores. The vertical lines extending from the box indicate the variability outside the above quartiles, and the dots outside the lines are regarded as outliers. Boxplots are ordered according to the magnitude of the median. Focusing on the entire set of PTAs, we find that provisions such as investment, subsidies, and services, and to a lesser extent, rules of origin and movement of capital have a median score above the overall score average, suggesting that these policy areas are good predictors of bilateral trade, after controlling for the usual gravity determinants of trade flows. Provisions in policy areas such as SPS, environmental laws, and visa and asylum are located at the other extreme of the distribution of median scores, suggesting a more limited role in predicting bilateral trade flows. The size of the boxes and the vertical lines also indicate that there are policy areas such as movement of capital and IPR for which the contribution to trade is more or less uniform across provisions. For other policy areas such as competition policy and SOEs, there is more heterogeneity within provisions in terms of their contribution to trade. 18 RF is a frequently used machine learning algorithm that predicts a Y variable by combining the results from hundreds of regression/classification trees. It has the merit of not imposing a linear relationship between the Y and X variables, which is an advantage when analyzing the impact of a highly heterogenous set of variables, such as the provisions in PTAs. 19 A score should not be interpreted as a coefficient in a regression analysis. It measures how much the accuracy of the prediction for Y gets worse if the particular X variable is randomly permuted. 19 Figure O12: Boxplot of scores calculated by the RF of variables/provisions in PTAs 20 Quantification challenges: Some analytical issues going forward Looking ahead, there is a need for stronger analytical underpinnings for any quantification exercise. Ideally, the “value” of a commitment must be evaluated in light of the objective that the provision or the deep trade agreement is trying to achieve. In other words, depth indicators could use different weights depending on whether the outcome variable is market access, welfare or another metric. For trade policy, market access may seem to be the most obvious metric, but for intellectual property rights, welfare may be the more relevant. In still other areas, such as competition policy, both might be relevant: the market access measure would include only provisions restricting barriers to foreign entry and operation while the welfare measure would include provisions requiring action against anti-competitive behavior affecting consumers. One indicator cannot provide a measure of both the trade distortions a country imposes on its trading partners (market access) and the trade distortions a country imposes on itself (welfare). For a market access-based measure in the goods context, the relevant question could be: what is the uniform tariff that if imposed on home imports instead of the existing structure of protection would leave aggregate imports at their current level? And for a welfare-based measure: what is the uniform tariff that if applied to imports instead of the current structure of protection would leave home welfare at its current level? The relationship between the two measures is likely to vary across policy areas: positive correlation for tariffs; perhaps negative for environmental standards; and ambiguous for intellectual property rights. A further issue relates to whether we should be interested in what legal commitments do to the level of a policy or to its variance. Provisions such as the elimination of tariffs, or of a national treatment rule in services or government procurement, fix the level of protection at zero. Provisions which legally bind policy (e.g., the permissible levels of fees, subsidies or preferences) truncate the distribution of possible policy outcomes by reducing the variance and hence the expected level of protection. Provisions which reduce discretion, such as rules on customs valuation, licensing or procurement procedures, narrow the distribution of possible policy outcomes. Finally, we also need to consider whether we should assess agreements per se or agreements relative to applied policies. If we have the relevant data, the mean and variance shift would ideally be assessed relative to the prevailing policy (and not just the law or policy on paper but how it is implemented). For example, a legal binding tariff at 10 percent might have a different value depending on whether the existing tariff was 5, 10 or 20 percent. The creation of new databases on applied policies in goods and services trade may facilitate such analysis. 5. Conclusions The World Development Report 2009 made the case that “thicker” borders between countries hurt economic growth, especially in developing countries. Policies that directly or indirectly restrain the international mobility of goods, services, capital, people and ideas limit, among other things, the scale of the market, which is vital for development.20 Deep trade agreements aim at establishing the rights of economic integration, protecting these rights from importing governments’ actions that could undo them, and regulating actions of exporters that can have negative welfare effects. These agreements have 20 World Bank 2009. 21 developed over time into a key institutional mechanism for countries to overcome the constraints to economic development created by the thick borders that fragment markets. Of course, deep integration is not an end in itself. First, countries at different levels of development may have different institutional needs, and trade agreements still need to strike the right balance between rules in PTAs and the needed discretion at the national level to pursue desirable social objectives. Second, while many deep provisions may be de facto non-discriminatory and apply to members and non-members alike, there is still a tension between the proliferation of regional approaches and multilateral rules enshrined in the WTO. Therefore, from the perspective of both economic development and global governance, the efficient set of rules in DTAs is an empirical question. The wealth of information on the content of the policy areas commonly included in PTAs could provide new impetus to the analysis on the determinants and impact of deep trade agreements. Such analysis would also provide the necessary tools to further understand the opportunities and challenges that countries face in terms of negotiation and implementation of deep trade agreements. We suggest three areas of work going forward. A first step is to improve the measurement of the depth of trade agreements and quantification of its effects. Beyond simple count variables and coverage ratios, more work will be needed to develop new analytic methods to overcome the challenges discussed in the previous section. As shown, machine learning techniques may provide a useful innovative approach. Second, the detailed information at the level of individual policy areas could inform a series of studies to assess how specific provisions impact trade and other relevant economic variables. As trade policy experts well understand, the devil is often in the details. Finally, the new data and analysis could provide essential information to policy makers on priorities for the negotiation and implementation of trade agreements: finding what potential partners include in their trade deals, identifying best practices in DTAs and areas where practices diverge or overlap across different players, and assessing gaps between international commitments and domestic legislation. 22 References Acharya, R. (ed.) 2016. Regional Trade Agreements and the Multilateral Trading System. Cambridge, UK: Cambridge University Press. Baldwin, R. 2010. 21st century regionalism: filling the gap between 21st century trade and 20th century trade rules. Geneva Graduate Institute, Working Paper No. 2010-31. Breinlich, H., V. Corradi, N. Rocha, M. Ruta, J. Santos Silva, T. Zylkin. 2020. Machine learning in international trade research: Evaluating the impact of trade agreements. Mimeo, University of Surrey and World Bank. Dür, A., L. Baccini and M. Elsig. 2014. The design of international trade agreements: introducing a new dataset. The Review of International Organizations 9(3), 353-375. Espitia, A., A. Mattoo, M. Mimouni, X. Pichot and N. Rocha. 2019. Preferential tariffs. Chapter 1 in A. Mattoo, N. Rocha and M. Ruta (eds.). Handbook of Deep Trade Agreements. World Bank. Espitia, A., S. Pardo, R. Piermartini, N. Rocha. 2019. Technical Barriers to Trade. Chapter 12 in A. Mattoo, N. Rocha and M. Ruta (eds.). Handbook of Deep Trade Agreements. World Bank. Freund, C. and Ornelas, E. 2010. Regional trade agreements. Annual Review of Economics 2, 139-166. Hofmann, C., A. Osnago and M. Ruta. 2019. The Content of Preferential Trade Agreements. World Trade Review, 18(3), 365-398. Horn, H., P.C. Mavroidis and A. Sapir. 2010. Beyond the WTO? An anatomy of EU and US preferential trade agreements. World Economy 33: 1565-1588. Kee, Hiau Looi, Alessandro Nicita and Marcelo Olarreaga. 2009. Estimating trade restrictiveness indices. Economic Journal 119: 534, pp. 172-199. Laget, E., A. Osnago, N. Rocha and M. Ruta. 2018. Deep trade agreements and global value chains. Policy Research Working Paper 8491. World Bank. Laget, E., N. Rocha and G. Varela. 2019. Deep trade agreements and foreign direct investments. Mimeo, World Bank. Limao, N. 2016. Preferential trade agreements. In K. Bagwell and R. Staiger (eds.), Handbook of Commercial Policy Vol. 1, pp. 279-367. Mattoo, A., A. Mulabdic and M. Ruta. 2017. Trade creation and trade diversion in deep agreements. Policy Research Working Paper 8206. World Bank. Paper funded by the World Bank’s Strategic Research Program. Mattoo, A., N. Rocha and M. Ruta (eds.). 2020. Handbook of Deep Trade Agreements. Washington DC. World Bank. Mulabdic, A., A. Osnago and M. Ruta. 2017. Deep integration and UK-EU trade relations. Policy Research Working Paper 7947. World Bank. 23 Rubini, L. and T. Wang. 2019. State-owned enterprises. Chapter 15 in A. Mattoo, N. Rocha and M. Ruta (eds.), Handbook of Deep Trade Agreements. World Bank. Viner, J. (1950). The Customs Union Issue. New York: Carnegie Endowment for International Peace. World Bank. 2009. World Development Report 2009: Reshaping Economic Geography. Washington DC. World Bank. World Bank. 2019. World Development Report 2020: Trading for Development in the Age of Global Value Chains. Washington DC. World Bank. World Trade Organization. 2011. World Trade Report 2011: The WTO and Preferential Trade Agreements: From Co-Existence to Coherence. Geneva. 24 Appendix Table O1: Number of provisions included and coverage ratio – by agreement Entry into Provisions Coverage Ratio Force Agreement Substantive (Year) Overall Substantive Overall (%) (%) 1958 EC Treaty 121 24 13.5 10.8 European Free Trade Association 1960 224 38 24.9 17.0 (EFTA) Central American Common Market 1961 155 44 17.2 19.7 (CACM) EU - Overseas Countries and 1971 57 11 6.3 4.9 Territories (OCT) EC-Iceland 57 11 6.3 4.9 EC-Norway 57 9 6.3 4.0 EC-Switzerland Liechtenst0 58 9 6.5 4.0 1973 EC (9) Enlargement 127 31 14.1 13.9 Caribbean Community and 152 36 16.9 16.1 Community Market (CARICOM) 1976 APTA 68 10 7.6 4.5 Australia - Papua New Guinea 19 3 2.1 1.3 1977 (PATCRA) EC-Syria 30 2 3.3 0.9 South Pacific Regional Trade and Economic Cooperation Agreement 21 2 2.3 0.9 1981 (SPARTECA) LAIA 32 7 3.6 3.1 EC (10) Enlargement 114 30 12.7 13.5 1983 Australia-New Zealand (ANZCERTA) 108 35 12.0 15.7 1985 US-Israel 90 7 10.0 3.1 1986 EC Enlargement (12) 118 32 13.1 14.3 1987 Panama - Dominican Republic 20 1 2.2 0.4 1988 CAN 116 25 12.9 11.2 Global System of Trade Preferences 1989 7 0 0.8 0.0 among Developing Countries (GSTP) Lao-Thailand 9 1 1.0 0.4 1991 EU - Andorra 30 6 3.3 2.7 MERCOSUR 161 32 17.9 14.3 Economic Cooperation Organization 46 8 5.1 3.6 (ECO) 1992 Turkey-EFTA 120 30 13.3 13.5 ASEAN free trade area 133 18 14.8 8.1 Russian Federation - Uzbekistan 34 13 3.8 5.8 Russian Federation - Tajikistan 36 8 4.0 3.6 Russian Federation - Turkmenistan 40 13 4.4 5.8 1993 Russian Federation - Azerbaijan 41 15 4.6 6.7 Faroe Islands - Norway 65 17 7.2 7.6 ECOWAS 99 18 11.0 8.1 25 EFTA-Israel 101 22 11.2 9.9 MSG 26 3 2.9 1.3 Georgia - Russian Federation 42 10 4.7 4.5 CIS 109 23 12.1 10.3 1994 COMESA 158 40 17.6 17.9 EEA 213 42 23.7 18.8 NAFTA 360 68 40.0 30.5 South Asian Preferential Trade 11 2 1.2 0.9 Agreement (SAPTA) Faroe Islands - Switzerland 21 6 2.3 2.7 Kyrgyz Republic - Armenia 37 11 4.1 4.9 1995 Ukraine-Turkmenistan 38 8 4.2 3.6 Kyrgyz Republic - Kazakhstan 39 7 4.3 3.1 Armenia - Moldova 39 10 4.3 4.5 EC Enlargement (15) 111 31 12.3 13.9 Colombia - Mexico 254 61 28.3 27.4 Ukraine - Azerbaijan 34 6 3.8 2.7 Armenia - Turkmenistan 36 10 4.0 4.5 Ukraine - Uzbekistan 37 5 4.1 2.2 Georgia - Azerbaijan 38 11 4.2 4.9 1996 Georgia - Ukraine 40 12 4.4 5.4 Armenia - Ukraine 40 7 4.4 3.1 Kyrgyz Republic - Moldova 42 10 4.7 4.5 EC-Turkey 110 40 12.2 17.9 Russian Federation - Belarus - 12 1 1.3 0.4 Kazakhstan EAEC 52 8 5.8 3.6 EC-Faroe Islands 59 8 6.6 3.6 1997 Turkey - Israel 74 15 8.2 6.7 EC-Palestinian Authority 107 21 11.9 9.4 Canada - Israel 141 21 15.7 9.4 Canada - Chile 290 62 32.3 27.8 PAFTA 11 0 1.2 0.0 Kyrgyz Republic - Uzbekistan 37 7 4.1 3.1 Georgia - Armenia 37 10 4.1 4.5 1998 Ukraine-Kazakhstan 37 6 4.1 2.7 Kyrgyz Republic - Ukraine 39 10 4.3 4.5 EC-Tunisia 126 35 14.0 15.7 Georgia - Kazakhstan 40 9 4.4 4.0 Economic and Monetary Community 57 14 6.3 6.3 of Central Africa (CEMAC) 1999 EFTA - Palestinian Authority 100 24 11.1 10.8 EFTA - Morocco 130 31 14.5 13.9 Chile - Mexico 265 69 29.5 30.9 26 Georgia - Turkmenistan 40 10 4.4 4.5 West African Economic and Monetary 61 12 6.8 5.4 Union (WAEMU) Turkey - Former Yugoslav Republic of 73 14 8.1 6.3 Macedonia Southern African Development 84 22 9.3 9.9 Community 2000 East African Community (EAC) 103 32 11.5 14.3 EC-Mexico 121 20 13.5 9.0 EC-Morocco 122 27 13.6 12.1 EC-Israel 125 31 13.9 13.9 EC-South Africa 128 33 14.2 14.8 Israel - Mexico 139 18 15.5 8.1 Armenia - Kazakhstan 38 11 4.2 4.9 India-Sri Lanka 39 8 4.3 3.6 Ukraine - Former Yugoslav Republic of 65 19 7.2 8.5 Macedonia 2001 US-Jordan 80 22 8.9 9.9 EC-FYR Macedonia 171 46 19.0 20.6 New Zealand - Singapore 179 34 19.9 15.2 Dominican Republic - Central America 189 44 21.0 19.7 EFTA - Mexico 233 37 25.9 16.6 Asia Pacific Trade Agreement (APTA) - 29 3 3.2 1.3 Accession of China EU-San Marino 32 11 3.6 4.9 Ukraine Tajikistan 39 8 4.3 3.6 EFTA - Former Yugoslav Republic of 121 26 13.5 11.7 Macedonia EFTA - Jordan 122 25 13.6 11.2 2002 Canada - Costa Rica 155 35 17.2 15.7 EC-Jordan 162 46 18.0 20.6 Chile - Costa Rica (Chile - Central 230 49 25.6 22.0 America) Chile - El Salvador (Chile - Central 230 49 25.6 22.0 America) Japan-Singapore 237 29 26.4 13.0 GCC 32 3 3.6 1.3 Pacific Island Countries Trade 37 1 4.1 0.4 Agreement (PICTA) India - Afghanistan 37 4 4.1 1.8 China - Macao, China 60 14 6.7 6.3 2003 Turkey - Bosnia and Herzegovina 77 16 8.6 7.2 EC-Lebanon 93 32 10.3 14.3 EFTA - Singapore 264 52 29.4 23.3 Panama - El Salvador (Panama - 271 59 30.1 26.5 Central America) 27 EC-Chile 301 54 33.5 24.2 Australia-Singapore 303 76 33.7 34.1 SACU 25 7 2.8 3.1 CEZ 45 13 5.0 5.8 China-Hong Kong 64 12 7.1 5.4 EC Enlargement (25) 103 29 11.5 13.0 EC-Egypt 132 32 14.7 14.3 2004 EFTA - Chile 232 38 25.8 17.0 Mexico - Uruguay 233 59 25.9 26.5 Panama - Taipei, China 271 73 30.1 32.7 US-Singapore 318 56 35.4 25.1 Chile-Korea 330 55 36.7 24.7 US-Chile 348 64 38.7 28.7 Pakistan - Sri Lanka 43 7 4.8 3.1 Ukraine - Moldova 65 14 7.2 6.3 Turkey - Tunisia 87 18 9.7 8.1 Turkey - Palestinian Authority 104 15 11.6 6.7 Jordan - Singapore 120 17 13.3 7.6 EC-Algeria 132 37 14.7 16.6 2005 EFTA - Tunisia 145 30 16.1 13.5 Thailand - New Zealand 178 28 19.8 12.6 China-ASEAN 184 36 20.5 16.1 Australia-Thailand 197 33 21.9 14.8 India-Singapore 197 37 21.9 16.6 Japan-Mexico 252 35 28.0 15.7 US-Australia 324 61 36.0 27.4 India - Bhutan 8 1 0.9 0.4 Ukraine-Belarus 36 8 4.0 3.6 SAFTA 39 2 4.3 0.9 Russian Federation - Serbia 39 7 4.3 3.1 Iceland - Faroe Islands 66 18 7.3 8.1 Turkey - Morocco 85 18 9.5 8.1 Chile-China 130 22 14.5 9.9 EC-Albania 158 39 17.6 17.5 2006 Guatemala – Taipei, China 227 52 25.3 23.3 US-Bahrain 237 38 26.4 17.0 Japan-Malaysia 245 47 27.3 21.1 Trans-Pacific Strategic Economic 260 52 28.9 23.3 Partnership US-Morocco 273 52 30.4 23.3 Panama - Singapore 274 52 30.5 23.3 EFTA-Korea 296 47 32.9 21.1 CAFTA-DR 300 53 33.4 23.8 28 Korea, Republic of-Singapore 315 54 35.0 24.2 Agadir Agreement 22 0 2.4 0.0 East African Community (EAC) - 40 14 4.4 6.3 Accession of Burundi Mauritius and Pakistan 42 7 4.7 3.1 Chile-India 70 11 7.8 4.9 Egypt - Turkey 84 18 9.3 8.1 EC Enlargement (27) 102 29 11.3 13.0 2007 Turkey - Syria 109 17 12.1 7.6 China-Pakistan 120 10 13.3 4.5 EFTA - Lebanon 142 36 15.8 16.1 EFTA - Egypt 150 37 16.7 16.6 CEFTA 153 55 17.0 24.7 Chile-Japan 244 39 27.1 17.5 Japan-Thailand 267 55 29.7 24.7 Japan-ASEAN 67 8 7.5 3.6 Turkey - Albania 73 14 8.1 6.3 Turkey - Georgia 78 15 8.7 6.7 EFTA - SACU 108 21 12.0 9.4 Panama - Chile 145 27 16.1 12.1 EC-Montenegro 194 63 21.6 28.3 Brunei Darussalam - Japan 197 34 21.9 15.2 EC-Bosnia Herzegovina 204 57 22.7 25.6 Pakistan - Malaysia 212 37 23.6 16.6 2008 Japan-Indonesia 222 38 24.7 17.0 Chile - Honduras (Chile - Central 226 49 25.1 22.0 America) El Salvador - Honduras - Taipei, China 233 54 25.9 24.2 Panama - Costa Rica (Panama - 241 49 26.8 22.0 Central America) Japan-Philippines 251 60 27.9 26.9 China-New Zealand 256 50 28.5 22.4 EC-CARIFORUM 334 74 37.2 33.2 Nicaragua - Taipei, China 334 95 37.2 42.6 India - Nepal 19 0 2.1 0.0 MERCOSUR-India 52 7 5.8 3.1 EU-Côte d'Ivoire 78 11 8.7 4.9 EU - Papua New Guinea/Fiji 96 14 10.7 6.3 EU-Cameroon 108 18 12.0 8.1 2009 Canada-EFTA 158 23 17.6 10.3 China-Singapore 214 39 23.8 17.5 Colombia - Northern Triangle (El 223 41 24.8 18.4 Salvador, Guatemala, Honduras) Panama - Nicaragua (Panama - 231 45 25.7 20.2 Central America) 29 Panama - Guatemala (Panama - 237 49 26.4 22.0 Central America Panama - Honduras (Panama - Central 239 49 26.6 22.0 America) Japan-Viet Nam 250 38 27.8 17.0 Peru - Chile 258 64 28.7 28.7 Chile - Colombia 266 60 29.6 26.9 US-Oman 286 59 31.8 26.5 Japan-Switzerland 292 47 32.5 21.1 Peru - Singapore 306 57 34.0 25.6 Chile-Australia 332 51 36.9 22.9 US-Peru 355 74 39.5 33.2 Canada-Peru 375 79 41.7 35.4 Turkey - Montenegro 71 15 7.9 6.7 Turkey - Serbia 71 14 7.9 6.3 EFTA - Serbia 162 30 18.0 13.5 EFTA - Albania 173 29 19.2 13.0 ASEAN-India 175 29 19.5 13.0 ASEAN-Korea 194 31 21.6 13.9 2010 EU-Serbia 209 56 23.2 25.1 Chile - Guatemala (Chile - Central 228 49 25.4 22.0 America) New Zealand - Malaysia 245 44 27.3 19.7 Korea, Republic of-India 252 45 28.0 20.2 China-Peru 260 44 28.9 19.7 ASEAN-Australia-New Zealand 260 59 28.9 26.5 South Asian FTA (SAFTA) - Accession 30 0 3.3 0.0 of Afghanistan Turkey - Jordan 82 16 9.1 7.2 Turkey - Chile 93 19 10.3 8.5 China - Costa Rica 176 23 19.6 10.3 Hong Kong, China - New Zealand 211 42 23.5 18.8 2011 India-Malaysia 213 32 23.7 14.3 India-Japan 234 40 26.0 17.9 EFTA - Peru 291 61 32.4 27.4 EFTA - Colombia 330 63 36.7 28.3 EU - Korea, Republic of 392 81 43.6 36.3 Canada - Colombia 401 79 44.6 35.4 Peru - Korea, Republic of 403 77 44.8 34.5 Treaty on a Free Trade Area between members of the Commonwealth of 37 5 4.1 2.2 Independent States (CIS) 2012 El Salvador-Cuba 51 10 5.7 4.5 EU - Eastern and Southern Africa 78 16 8.7 7.2 States Interim EPA 30 Chile - Malaysia 91 12 10.1 5.4 Canada - Jordan 120 29 13.3 13.0 EFTA - Montenegro 206 52 22.9 23.3 Chile - Nicaragua (Chile - Central 227 48 25.3 21.5 America) Peru - Mexico 236 45 26.3 20.2 Mexico - Central America 244 56 27.1 25.1 EFTA - Hong Kong, China 295 54 32.8 24.2 EFTA - Ukraine 302 54 33.6 24.2 Panama - Peru 311 65 34.6 29.1 Japan - Peru 324 51 36.0 22.9 US - Panama 328 71 36.5 31.8 Korea, Republic of - US 344 68 38.3 30.5 US - Colombia 355 76 39.5 34.1 Turkey - Mauritius 55 7 6.1 3.1 EU (28) Enlargement 102 31 11.3 13.9 Ukraine - Montenegro 133 22 14.8 9.9 Korea, Republic of - Turkey 140 42 15.6 18.8 Gulf Cooperation Council (GCC) - 184 27 20.5 12.1 Singapore 2013 Malaysia - Australia 291 65 32.4 29.1 Costa Rica - Singapore 294 57 32.7 25.6 Costa Rica - Peru 301 64 33.5 28.7 New Zealand - Taipei, China 340 70 37.8 31.4 Canada - Panama 345 74 38.4 33.2 EU - Central America 395 88 43.9 39.5 EU - Colombia and Peru 399 89 44.4 39.9 Chile - Viet nam 75 5 8.3 2.2 Iceland - China 180 32 20.0 14.3 Hong Kong, China - Chile 189 27 21.0 12.1 Switzerland - China 210 40 23.4 17.9 Singapore - Taipei, China i 274 40 30.5 17.9 EFTA - Central America (Costa Rica 2014 356 74 39.6 33.2 and Panama) Canada - Honduras 357 74 39.7 33.2 EU - Georgia 376 96 41.8 43.0 EU - Republic of Moldova 402 94 44.7 42.2 Korea, Republic of - Australia 415 91 46.2 40.8 EU Ukraine 448 111 49.8 49.8 Southern African Development Community (SADC) - Accession of 40 10 4.4 4.5 2015 Seychelles Eurasian Economic Union (EAEU) - 114 38 12.7 17.0 Accession of Armenia 31 Eurasian Economic Union (EAEU) - 117 40 13.0 17.9 Accession of Kyrgyz Republic Mexico - Panama 169 24 18.8 10.8 Korea, Republic of - Viet Nam 186 18 20.7 8.1 Australia - China 202 39 22.5 17.5 EFTA - Bosnia and Herzegovina 206 48 22.9 21.5 Korea, Republic of - New Zealand 240 37 26.7 16.6 Eurasian Economic Union (EAEU) 243 63 27.0 28.3 China - Korea, Republic of 262 56 29.1 25.1 Japan - Australia 342 63 38.0 28.3 Canada - Rep. of Korea 415 93 46.2 41.7 Pacific Alliance 110 27 12.2 12.1 Costa Rica - Colombia 172 28 19.1 12.6 2016 Japan - Mongolia 210 42 23.4 18.8 Korea, Republic of - Colombia 267 49 29.7 22.0 2017 Trans-Pacific Partnership 486 136 54.1 61.0 32 Appendix Table O2: Average number of provisions included and coverage ratio – by country Provisions Coverage Ratio Number of Country Substantive Agreements Overall Substantive Overall (%) (%) Afghanistan 2 33.5 2.0 3.7 0.9 Albania 4 139.3 34.3 15.5 15.4 Algeria 2 71.5 18.5 8.0 8.3 Andorra 1 30.0 6.0 3.3 2.7 Angola 2 62.0 16.0 6.9 7.2 Antigua and Barbuda 1 152.0 36.0 16.9 16.1 Argentina 4 63.0 11.5 7.0 5.2 Armenia 9 77.9 22.2 8.7 10.0 Aruba 1 57.0 11.0 6.3 4.9 Australia 13 253.8 54.9 28.2 24.6 Austria 37 170.1 39.7 18.9 17.8 Azerbaijan 5 51.8 12.0 5.8 5.4 Bahamas, The 2 243.0 55.0 27.0 24.7 Bahrain 4 116.0 17.0 12.9 7.6 Bangladesh 5 35.4 3.4 3.9 1.5 Barbados 1 334.0 74.0 37.2 33.2 Belarus 7 88.4 24.4 9.8 11.0 Belgium 37 170.1 39.7 18.9 17.8 Belize 2 243.0 55.0 27.0 24.7 Benin 2 80.0 15.0 8.9 6.7 Bermuda 1 57.0 11.0 6.3 4.9 Bhutan 4 22.0 1.3 2.4 0.6 Bolivia 2 74.0 16.0 8.2 7.2 Bosnia and Herzegovina 4 160.0 44.0 17.8 19.7 Botswana 4 64.3 15.0 7.1 6.7 Brazil 3 81.7 15.3 9.1 6.9 Brunei Darussalam 9 217.3 44.8 24.2 20.1 Bulgaria 37 165.4 39.7 18.9 17.8 Burkina Faso 2 80.0 15.0 8.9 6.7 Burundi 3 100.3 28.7 11.2 12.9 Cabo Verde 1 99.0 18.0 11.0 8.1 Cambodia 6 168.8 30.2 18.8 13.5 Cameroon 2 82.5 16.0 9.2 7.2 Canada 12 300.3 64.4 33.4 28.9 Cayman Islands 1 57.0 11.0 6.3 4.9 Central African Republic 1 57.0 14.0 6.3 6.3 Chad 1 57.0 14.0 6.3 6.3 Chile 26 218.8 44.0 24.3 19.7 China 14 167.6 30.0 18.6 13.5 33 Colombia 12 243.8 50.4 27.1 22.6 Comoros 1 158.0 40.0 17.6 17.9 Congo, Dem. Rep. 3 94.0 24.0 10.5 10.8 Congo, Rep. 1 57.0 14.0 6.3 6.3 Costa Rica 13 246.8 51.1 27.4 22.9 Côte d'Ivoire 3 79.3 13.7 8.8 6.1 Croatia 37 170.1 39.7 18.9 17.8 Cuba 2 41.5 8.5 4.6 3.8 Cyprus 37 170.1 39.7 18.9 17.8 Czech Republic 37 165.4 39.7 18.9 17.8 Denmark 37 170.1 39.7 18.9 17.8 Djibouti 1 158.0 40.0 17.6 17.9 Dominica 2 243.0 55.0 27.0 24.7 Dominican Republic 3 181.0 39.7 20.1 17.8 Ecuador 2 74.0 16.0 8.2 7.2 Egypt, Arab Rep. 6 92.8 21.2 10.3 9.5 El Salvador 10 229.1 49.8 25.5 22.3 Equatorial Guinea 1 57.0 14.0 6.3 6.3 Eritrea 1 158.0 40.0 17.6 17.9 Estonia 37 165.4 39.7 18.9 17.8 Ethiopia 1 158.0 40.0 17.6 17.9 Faroe Islands 4 52.8 12.3 5.9 5.5 Fiji 4 45.0 5.0 5.0 2.2 Finland 37 165.4 39.7 18.9 17.8 France 37 170.1 39.7 18.9 17.8 French Polynesia 1 57.0 11.0 6.3 4.9 Gabon 1 57.0 14.0 6.3 6.3 Gambia, The 1 99.0 18.0 11.0 8.1 Georgia 8 86.4 21.6 9.6 9.7 Germany 37 170.1 39.7 18.9 17.8 Ghana 1 99.0 18.0 11.0 8.1 Greece 37 170.1 39.7 18.9 17.8 Greenland 1 57.0 11.0 6.3 4.9 Grenada 2 243.0 55.0 27.0 24.7 Guatemala 9 244.2 52.9 27.2 23.7 Guinea 1 99.0 18.0 11.0 8.1 Guinea-Bissau 2 80.0 15.0 8.9 6.7 Guyana 2 243.0 55.0 27.0 24.7 Haiti 1 152.0 36.0 16.9 16.1 Honduras 10 256.1 55.2 28.5 24.8 Hong Kong SAR, China 4 189.8 33.8 21.1 15.1 Hungary 37 165.4 39.7 18.9 17.8 Iceland 29 189.1 37.4 21.0 16.8 34 India 16 92.1 14.4 10.2 6.5 Indonesia 7 176.4 31.3 19.6 14.0 Iran, Islamic Rep. 1 46.0 8.0 5.1 3.6 Iraq 1 11.0 0.0 1.2 0.0 Ireland 37 165.4 39.7 18.9 17.8 Israel 6 111.7 19.0 12.4 8.5 Italy 37 170.1 39.7 18.9 17.8 Jamaica 2 243.0 55.0 27.0 24.7 Japan 16 257.5 47.6 28.6 21.4 Jordan 8 89.9 19.4 10.0 8.7 Kazakhstan 10 73.7 19.6 8.2 8.8 Kenya 3 100.3 28.7 11.2 12.9 Kiribati 2 29.0 1.5 3.2 0.7 Korea, Rep. 17 267.5 50.4 29.8 22.6 Kuwait 3 75.7 10.0 8.4 4.5 Kyrgyz Republic 9 80.0 21.6 8.9 9.7 Lao PDR 9 124.3 21.7 13.8 9.7 Latvia 37 170.1 39.7 18.9 17.8 Lebanon 3 82.0 22.7 9.1 10.2 Lesotho 4 64.3 15.0 7.1 6.7 Liberia 1 99.0 18.0 11.0 8.1 Libya 2 84.5 20.0 9.4 9.0 Liechtenstein 27 194.0 38.3 21.6 17.2 Lithuania 37 165.4 39.7 18.9 17.8 Luxembourg 37 170.1 39.7 18.9 17.8 Macao SAR, China 1 60.0 14.0 6.7 6.3 North Macedonia 5 116.6 32.0 13.0 14.3 Madagascar 4 90.0 22.0 10.0 9.9 Malawi 3 94.0 24.0 10.5 10.8 Malaysia 13 215.1 42.6 23.9 19.1 Maldives 3 26.7 1.3 3.0 0.6 Mali 2 80.0 15.0 8.9 6.7 Malta 37 165.4 39.7 18.9 17.8 Marshall Islands 1 21.0 2.0 2.3 0.9 Mauritius 6 76.2 17.0 8.5 7.6 Mexico 14 223.9 47.3 24.9 21.2 Micronesia, Fed. Sts. 2 29.0 1.5 3.2 0.7 Moldova 5 140.2 36.6 15.6 16.4 Mongolia 1 210.0 42.0 23.4 18.8 Montenegro 5 151.4 41.4 16.8 18.6 Morocco 6 107.2 21.3 11.9 9.6 Mozambique 2 62.0 16.0 6.9 7.2 Myanmar 6 168.8 30.2 18.8 13.5 35 Namibia 4 64.3 15.0 7.1 6.7 Nauru 2 29.0 1.5 3.2 0.7 Nepal 4 24.8 1.0 2.8 0.4 Netherlands 37 170.1 39.7 18.9 17.8 New Caledonia 1 57.0 11.0 6.3 4.9 New Zealand 12 232.0 49.1 25.8 22.0 Nicaragua 8 259.4 59.1 28.9 26.5 Niger 2 80.0 15.0 8.9 6.7 Nigeria 1 99.0 18.0 11.0 8.1 Norway 28 189.4 37.5 21.1 16.8 Oman 4 128.3 22.3 14.3 10.0 Pakistan 9 61.1 8.1 6.8 3.6 Panama 14 245.6 50.9 27.3 22.8 Papua New Guinea 5 39.8 4.6 4.4 2.1 Paraguay 3 81.7 15.3 9.1 6.9 Peru 16 285.2 60.3 31.7 27.0 Philippines 7 180.6 34.4 20.1 15.4 Poland 37 165.4 39.7 18.9 17.8 Portugal 37 170.1 39.7 18.9 17.8 Qatar 3 75.7 10.0 8.4 4.5 Romania 37 170.1 39.7 18.9 17.8 Russian Federation 12 67.9 19.1 7.6 8.6 Rwanda 3 100.3 28.7 11.2 12.9 Samoa 2 29.0 1.5 3.2 0.7 San Marino 2 240.0 61.0 26.7 27.4 Saudi Arabia 3 75.7 10.0 8.4 4.5 Senegal 2 80.0 15.0 8.9 6.7 Serbia 5 126.8 32.4 14.1 14.5 Seychelles 4 90.0 22.0 10.0 9.9 Sierra Leone 1 99.0 18.0 11.0 8.1 Singapore 22 238.1 45.3 26.5 20.3 Slovak Republic 37 170.1 39.7 18.9 17.8 Slovenia 37 165.4 39.7 18.9 17.8 Solomon Islands 3 28.0 2.0 3.1 0.9 South Africa 5 77.0 18.6 8.6 8.3 South Sudan 1 158.0 40.0 17.6 17.9 Spain 37 165.4 39.7 18.9 17.8 Sri Lanka 7 37.0 4.6 4.1 2.0 St. Kitts and Nevis 2 243.0 55.0 27.0 24.7 St. Lucia 2 243.0 55.0 27.0 24.7 St. Vincent and the Grenadines 2 243.0 55.0 27.0 24.7 Sudan 1 11.0 0.0 1.2 0.0 Suriname 2 243.0 55.0 27.0 24.7 36 Eswatini 5 83.0 20.0 9.2 9.0 Sweden 37 165.4 39.7 18.9 17.8 Switzerland 30 192.0 37.5 21.4 16.8 Syrian Arab Republic 3 50.0 6.3 5.6 2.8 Taiwan, China 6 279.8 64.0 31.1 28.7 Tajikistan 3 42.3 8.0 4.7 3.6 Tanzania 4 66.8 19.5 7.4 8.7 Thailand 10 166.4 29.8 18.5 13.4 Togo 2 80.0 15.0 8.9 6.7 Tonga 2 29.0 1.5 3.2 0.7 Trinidad and Tobago 2 243.0 55.0 27.0 24.7 Tunisia 5 78.2 16.6 8.7 7.4 Turkey 19 85.9 18.5 9.6 8.3 Turkmenistan 6 50.0 11.5 5.6 5.2 Turks and Caicos Islands 1 57.0 11.0 6.3 4.9 Tuvalu 2 29.0 1.5 3.2 0.7 Uganda 3 100.3 28.7 11.2 12.9 Ukraine 14 67.9 13.7 7.5 6.1 United Arab Emirates 3 75.7 10.0 8.4 4.5 United Kingdom 37 165.4 39.7 18.9 17.8 United States 14 285.6 54.9 31.8 24.6 Uruguay 4 119.5 26.3 13.3 11.8 Uzbekistan 5 50.8 10.6 5.7 4.8 Vanuatu 3 28.0 2.0 3.1 0.9 Venezuela, RB 3 81.7 15.3 9.1 6.9 Vietnam 10 201.0 37.8 22.4 17.0 West Bank and Gaza 3 103.7 20.0 11.5 9.0 Yemen, Rep. 1 11.0 0.0 1.2 0.0 Zambia 3 94.0 24.0 10.5 10.8 Zimbabwe 4 90.0 22.0 10.0 9.9 37 Appendix Table O3: Maximum number of provisions included and coverage ratio – by country Provisions Substantive Country Agreement Provisions Coverage Ratio (%) Agreement Provisions Coverage Ratio (%) India – India - Afghanistan 37 4.1 4 1.8 Afghanistan Afghanistan EFTA – Albania 173 19.2 CEFTA 55 24.7 Albania Algeria EC-Algeria 132 14.7 EC-Algeria 37 16.6 Andorra EU – Andorra 30 3.3 EU - Andorra 6 2.7 Southern Southern African African Angola 84 9.3 22 9.9 Development Development Community Community Antigua and Barbuda CARICOM 152 16.9 CARICOM 36 16.1 Argentina MERCOSUR 161 17.9 MERCOSUR 32 14.3 Armenia EAEU 243 27.0 EAEU 63 28.3 EU - Overseas EU - Overseas Countries and Countries and Aruba 57 6.3 11 4.9 Territories Territories (OCT) (OCT) Trans-Pacific Trans-Pacific Australia 486 54.1 136 61.0 Partnership Partnership Austria EU Ukraine 448 49.8 EU Ukraine 111 49.8 Azerbaijan CIS 109 12.1 CIS 23 10.3 EC- EC- Bahamas, The 334 37.2 74 33.2 CARIFORUM CARIFORUM Bahrain US-Bahrain 237 26.4 US-Bahrain 38 17.0 Bangladesh APTA 68 7.6 APTA 10 4.5 EC- EC- Barbados 334 37.2 74 33.2 CARIFORUM CARIFORUM Eurasian Eurasian Belarus Economic 243 27.0 Economic 63 28.3 Union (EAEU) Union (EAEU) Belgium EU Ukraine 448 49.8 EU Ukraine 111 49.8 EC- EC- Belize 334 37.2 74 33.2 CARIFORUM CARIFORUM Benin ECOWAS 99 11.0 ECOWAS 18 8.1 EU - Overseas EU - Overseas Countries and Countries and Bermuda 57 6.3 11 4.9 Territories Territories (OCT) (OCT) Bhutan SAFTA 39 4.3 SAFTA 2 0.9 Bolivia CAN 116 12.9 CAN 25 11.2 EFTA - Bosnia EC-Bosnia Bosnia and Herzegovina and 206 22.9 57 25.6 Herzegovina Herzegovina Southern African Botswana EFTA – SACU 108 12.0 22 9.9 Development Community Brazil MERCOSUR 161 17.9 MERCOSUR 32 14.3 Trans-Pacific Trans-Pacific Brunei Darussalam 486 54.1 136 61.0 Partnership Partnership Bulgaria EU Ukraine 448 49.8 EU Ukraine 111 49.8 Burkina Faso ECOWAS 99 11.0 ECOWAS 18 8.1 38 Burundi COMESA 158 17.6 COMESA 40 17.9 Cabo Verde ECOWAS 99 11.0 ECOWAS 18 8.1 ASEAN- ASEAN- Cambodia Australia-New 260 28.9 Australia-New 59 26.5 Zealand Zealand Cameroon EC-Cameroon 108 12.0 EC-Cameroon 18 8.1 Trans-Pacific Trans-Pacific Canada 486 54.1 136 61.0 Partnership Partnership EU - Overseas EU - Overseas Countries and Countries and Cayman Islands 57 6.3 11 4.9 Territories Territories (OCT) (OCT) Economic and Economic and Monetary Monetary Central African Republic Community of 57 6.3 Community of 14 6.3 Central Africa Central Africa (CEMAC) (CEMAC) Economic and Economic and Monetary Monetary Chad Community of 57 6.3 Community of 14 6.3 Central Africa Central Africa (CEMAC) (CEMAC) Trans-Pacific Trans-Pacific Chile 486 54.1 136 61.0 Partnership Partnership China - Korea, China - Korea, China 262 29.1 56 25.1 Republic of Republic of Canada – EU - Colombia Colombia 401 44.6 89 39.9 Colombia and Peru Comoros COMESA 158 17.6 COMESA 40 17.9 Congo, Dem. Rep. COMESA 158 17.6 COMESA 40 17.9 Economic and Economic and Monetary Monetary Congo, Rep. Community of 57 6.3 Community of 14 6.3 Central Africa Central Africa (CEMAC) (CEMAC) EU - Central EU - Central Costa Rica 395 43.9 88 39.5 America America Côte d'Ivoire ECOWAS 99 11.0 ECOWAS 18 8.1 Croatia EU Ukraine 448 49.8 EU Ukraine 111 49.8 El Salvador- El Salvador- Cuba 51 5.7 10 4.5 Cuba Cuba Cyprus EU Ukraine 448 49.8 EU Ukraine 111 49.8 Czech Republic EU Ukraine 448 49.8 EU Ukraine 111 49.8 Denmark EU Ukraine 448 49.8 EU Ukraine 111 49.8 Djibouti COMESA 158 17.6 COMESA 40 17.9 EC- EC- Dominica 334 37.2 74 33.2 CARIFORUM CARIFORUM EC- EC- Dominican Republic 334 37.2 74 33.2 CARIFORUM CARIFORUM Ecuador CAN 116 12.9 CAN 25 11.2 Egypt, Arab Rep. COMESA 158 17.6 COMESA 40 17.9 EU - Central EU - Central El Salvador 395 43.9 88 39.5 America America Economic and Economic and Monetary Monetary Equatorial Guinea Community of 57 6.3 Community of 14 6.3 Central Africa Central Africa (CEMAC) (CEMAC) 39 Eritrea COMESA 158 17.6 COMESA 40 17.9 Estonia EU Ukraine 448 49.8 EU Ukraine 111 49.8 Ethiopia COMESA 158 17.6 COMESA 40 17.9 Iceland - Iceland - Faroe Islands 66 7.3 18 8.1 Faroe Islands Faroe Islands EU - Papua EU - Papua Fiji New 96 10.7 New 14 6.3 Guinea/Fiji Guinea/Fiji Finland EU Ukraine 448 49.8 EU Ukraine 111 49.8 France EU Ukraine 448 49.8 EU Ukraine 111 49.8 EU - Overseas EU - Overseas Countries and Countries and French Polynesia 57 6.3 11 4.9 Territories Territories (OCT) (OCT) Economic and Economic and Monetary Monetary Gabon Community of 57 6.3 Community of 14 6.3 Central Africa Central Africa (CEMAC) (CEMAC) Gambia, The ECOWAS 99 11.0 ECOWAS 18 8.1 Georgia EU – Georgia 376 41.8 EU - Georgia 96 43.0 Germany EU Ukraine 448 49.8 EU Ukraine 111 49.8 Ghana ECOWAS 99 11.0 ECOWAS 18 8.1 Greece EU Ukraine 448 49.8 EU Ukraine 111 49.8 EU - Overseas EU - Overseas Countries and Countries and Greenland 57 6.3 11 4.9 Territories Territories (OCT) (OCT) EC- EC- Grenada 334 37.2 74 33.2 CARIFORUM CARIFORUM EU - Central EU - Central Guatemala 395 43.9 88 39.5 America America Guinea ECOWAS 99 11.0 ECOWAS 18 8.1 Guinea-Bissau ECOWAS 99 11.0 ECOWAS 18 8.1 EC- EC- Guyana 334 37.2 74 33.2 CARIFORUM CARIFORUM Caribbean Caribbean Community Community and and Haiti 152 16.9 36 16.1 Community Community Market Market (CARICOM) (CARICOM) EU - Central EU - Central Honduras 395 43.9 88 39.5 America America EFTA - Hong EFTA - Hong Hong Kong SAR, China 295 32.8 54 24.2 Kong, China Kong, China Hungary EU Ukraine 448 49.8 EU Ukraine 111 49.8 EFTA - Central EFTA - Central America America Iceland 356 39.6 74 33.2 (Costa Rica (Costa Rica and Panama) and Panama) Korea, Korea, India Republic of- 252 28.0 Republic of- 45 20.2 India India ASEAN- ASEAN- Indonesia Australia-New 260 28.9 Australia-New 59 26.5 Zealand Zealand 40 Economic Economic Cooperation Cooperation Iran, Islamic Rep. 46 5.1 8 3.6 Organization Organization (ECO) (ECO) Iraq PAFTA 11 1.2 PAFTA 0 0.0 Ireland EU Ukraine 448 49.8 EU Ukraine 111 49.8 Canada – Israel 141 15.7 EC-Israel 31 13.9 Israel Italy EU Ukraine 448 49.8 EU Ukraine 111 49.8 EC- EC- Jamaica 334 37.2 74 33.2 CARIFORUM CARIFORUM Trans-Pacific Trans-Pacific Japan 486 54.1 136 61.0 Partnership Partnership Jordan EC-Jordan 162 18.0 EC-Jordan 46 20.6 Eurasian Eurasian Kazakhstan Economic 243 27.0 Economic 63 28.3 Union (EAEU) Union (EAEU) Kenya COMESA 158 17.6 COMESA 40 17.9 South Pacific Pacific Island Regional Countries Trade and Kiribati Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) Canada - Rep. Canada - Rep. Korea, Rep. 415 46.2 93 41.7 of Korea of Korea Gulf Gulf Cooperation Cooperation Kuwait 184 20.5 27 12.1 Council (GCC) Council (GCC) – Singapore - Singapore Eurasian Eurasian Kyrgyz Republic Economic 243 27.0 Economic 63 28.3 Union (EAEU) Union (EAEU) ASEAN- ASEAN- Lao PDR Australia-New 260 28.9 Australia-New 59 26.5 Zealand Zealand Latvia EU Ukraine 448 49.8 EU Ukraine 111 49.8 EFTA – EFTA - Lebanon 142 15.8 36 16.1 Lebanon Lebanon Southern African Lesotho EFTA - SACU 108 12.0 22 9.9 Development Community Liberia ECOWAS 99 11.0 ECOWAS 18 8.1 Libya COMESA 158 17.6 COMESA 40 17.9 EFTA - Central EFTA - Central America America Liechtenstein 356 39.6 74 33.2 (Costa Rica (Costa Rica and Panama) and Panama) Lithuania EU Ukraine 448 49.8 EU Ukraine 111 49.8 Luxembourg EU Ukraine 448 49.8 EU Ukraine 111 49.8 China - China - Macao SAR, China 60 6.7 14 6.3 Macao, China Macao, China EC-FYR North Macedonia 171 19.0 CEFTA 55 24.7 Macedonia Madagascar COMESA 158 17.6 COMESA 40 17.9 Malawi COMESA 158 17.6 COMESA 40 17.9 41 Trans-Pacific Trans-Pacific Malaysia 486 54.1 136 61.0 Partnership Partnership Maldives SAFTA 39 4.3 SAFTA 2 0.9 Mali ECOWAS 99 11.0 ECOWAS 18 8.1 Malta EU Ukraine 448 49.8 EU Ukraine 111 49.8 South Pacific South Pacific Regional Regional Trade and Trade and Marshall Islands Economic 21 2.3 Economic 2 0.9 Cooperation Cooperation Agreement Agreement (SPARTECA) (SPARTECA) Mauritius COMESA 158 17.6 COMESA 40 17.9 Trans-Pacific Trans-Pacific Mexico 486 54.1 136 61.0 Partnership Partnership South Pacific Pacific Island Regional Countries Trade and Micronesia, Fed. Sts. Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) EU - Republic EU - Republic Moldova 402 44.7 94 42.2 of Moldova of Moldova Japan – Japan - Mongolia 210 23.4 42 18.8 Mongolia Mongolia EFTA - EC- Montenegro 206 22.9 63 28.3 Montenegro Montenegro Morocco US-Morocco 273 30.4 US-Morocco 52 23.3 Southern Southern African African Mozambique 84 9.3 22 9.9 Development Development Community Community ASEAN- ASEAN- Myanmar Australia-New 260 28.9 Australia-New 59 26.5 Zealand Zealand Southern African Namibia EFTA - SACU 108 12.0 22 9.9 Development Community South Pacific Pacific Island Regional Countries Trade and Nauru Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) Nepal SAFTA 39 4.3 SAFTA 2 0.9 Netherlands EU Ukraine 448 49.8 EU Ukraine 111 49.8 EU - Overseas EU - Overseas Countries and Countries and New Caledonia 57 6.3 11 4.9 Territories Territories (OCT) (OCT) Trans-Pacific Trans-Pacific New Zealand 486 54.1 136 61.0 Partnership Partnership EU - Central Nicaragua - Nicaragua 395 43.9 95 42.6 America Taipei, China Niger ECOWAS 99 11.0 ECOWAS 18 8.1 Nigeria ECOWAS 99 11.0 ECOWAS 18 8.1 42 EFTA - Central EFTA - Central America America Norway 356 39.6 74 33.2 (Costa Rica (Costa Rica and Panama) and Panama) Oman US-Oman 286 31.8 US-Oman 59 26.5 Pakistan – Pakistan - Pakistan 212 23.6 37 16.6 Malaysia Malaysia EFTA - Central America Canada - Panama 356 39.6 74 33.2 (Costa Rica Panama and Panama) EU - Papua EU - Papua Papua New Guinea New 96 10.7 New 14 6.3 Guinea/Fiji Guinea/Fiji Paraguay MERCOSUR 161 17.9 MERCOSUR 32 14.3 Trans-Pacific Trans-Pacific Peru 486 54.1 136 61.0 Partnership Partnership ASEAN- Japan- Philippines Australia-New 260 28.9 60 26.9 Philippines Zealand Poland EU Ukraine 448 49.8 EU Ukraine 111 49.8 Portugal EU Ukraine 448 49.8 EU Ukraine 111 49.8 Gulf Gulf Cooperation Cooperation Qatar 184 20.5 27 12.1 Council (GCC) Council (GCC) - Singapore - Singapore Romania EU Ukraine 448 49.8 EU Ukraine 111 49.8 Eurasian Eurasian Russian Federation Economic 243 27.0 Economic 63 28.3 Union (EAEU) Union (EAEU) Rwanda COMESA 158 17.6 COMESA 40 17.9 South Pacific Pacific Island Regional Countries Trade and Samoa Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) San Marino EU Ukraine 448 49.8 EU Ukraine 111 49.8 Gulf Gulf Cooperation Cooperation Saudi Arabia 184 20.5 27 12.1 Council (GCC) Council (GCC) - Singapore - Singapore Senegal ECOWAS 99 11.0 ECOWAS 18 8.1 Serbia EU-Serbia 209 23.2 EU-Serbia 56 25.1 Seychelles COMESA 158 17.6 COMESA 40 17.9 Sierra Leone ECOWAS 99 11.0 ECOWAS 18 8.1 Trans-Pacific Trans-Pacific Singapore 486 54.1 136 61.0 Partnership Partnership Slovak Republic EU Ukraine 448 49.8 EU Ukraine 111 49.8 Slovenia EU Ukraine 448 49.8 EU Ukraine 111 49.8 Pacific Island Countries Solomon Islands Trade 37 4.1 MSG 3 1.3 Agreement (PICTA) EC-South EC-South South Africa 128 14.2 33 14.8 Africa Africa 43 South Sudan COMESA 158 17.6 COMESA 40 17.9 Spain EU Ukraine 448 49.8 EU Ukraine 111 49.8 Sri Lanka APTA 68 7.6 APTA 10 4.5 EC- EC- St. Kitts and Nevis 334 37.2 74 33.2 CARIFORUM CARIFORUM EC- EC- St. Lucia 334 37.2 74 33.2 CARIFORUM CARIFORUM EC- EC- St. Vincent and the Grenadines 334 37.2 74 33.2 CARIFORUM CARIFORUM Sudan PAFTA 11 1.2 PAFTA 0 0.0 EC- EC- Suriname 334 37.2 74 33.2 CARIFORUM CARIFORUM Eswatini COMESA 158 17.6 COMESA 40 17.9 Sweden EU Ukraine 448 49.8 EU Ukraine 111 49.8 EFTA - Central EFTA - Central America America Switzerland 356 39.6 74 33.2 (Costa Rica (Costa Rica and Panama) and Panama) Syrian Arab Republic Turkey - Syria 109 12.1 Turkey - Syria 17 7.6 New Zealand - Nicaragua - Taiwan, China 340 37.8 95 42.6 Taipei, China Taipei, China Tajikistan EAEC 52 5.8 EAEC 8 3.6 East African East African Tanzania Community 103 11.5 Community 32 14.3 (EAC) (EAC) ASEAN- Japan- Thailand 267 29.7 Australia-New 59 26.5 Thailand Zealand Togo ECOWAS 99 11.0 ECOWAS 18 8.1 South Pacific Pacific Island Regional Countries Trade and Tonga Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) EC- EC- Trinidad and Tobago 334 37.2 74 33.2 CARIFORUM CARIFORUM Tunisia EFTA - Tunisia 145 16.1 EC-Tunisia 35 15.7 Korea, Korea, Turkey Republic of – 140 15.6 Republic of - 42 18.8 Turkey Turkey Turkmenistan CIS 109 12.1 CIS 23 10.3 EU - Overseas EU - Overseas Countries and Countries and Turks and Caicos Islands 57 6.3 11 4.9 Territories Territories (OCT) (OCT) South Pacific Pacific Island Regional Countries Trade and Tuvalu Trade 37 4.1 Economic 2 0.9 Agreement Cooperation (PICTA) Agreement (SPARTECA) Uganda COMESA 158 17.6 COMESA 40 17.9 EFTA – EFTA - Ukraine 302 33.6 54 24.2 Ukraine Ukraine Gulf Gulf United Arab Emirates 184 20.5 27 12.1 Cooperation Cooperation 44 Council (GCC) Council (GCC) - Singapore – Singapore United Kingdom EU Ukraine 448 49.8 EU Ukraine 111 49.8 United States NAFTA 360 40.0 US - Colombia 76 34.1 Mexico – Mexico - Uruguay 233 25.9 59 26.5 Uruguay Uruguay Uzbekistan CIS 109 12.1 CIS 23 10.3 Pacific Island Countries Vanuatu Trade 37 4.1 MSG 3 1.3 Agreement (PICTA) Venezuela, RB MERCOSUR 161 17.9 MERCOSUR 32 14.3 Trans-Pacific Trans-Pacific Vietnam 486 54.1 136 61.0 Partnership Partnership EFTA - EC-Palestinian West Bank and Gaza 107 11.9 Palestinian 24 10.8 Authority Authority Yemen, Rep. PAFTA 11 1.2 PAFTA 0 0.0 Zambia COMESA 158 17.6 COMESA 40 17.9 Zimbabwe COMESA 158 17.6 COMESA 40 17.9 45