1. Economics

The rise in preferential trade agreements between countries (in the form of GATT's Article XXIV-sanctioned free trade agreements, customs unions, and other arrangements involving developing countries, permitted by the GATT's Enabling Clause) stands as the dominant trend in the evolution of the international trade system in the recent two decades. Hundreds of preferential agreements have been negotiated during this period and nearly every member country of the World Trade Organization (WTO) belongs to at least one PTA.

Alongside this evolution of the world trade system towards preferential trade, the academic and policy literature on the economics of trade preferences has explored a variety of questions regarding the merits of preferential trade, the political and economic determinants of preferential agreements, and the interplay between the bilateral (preferential) and multilateral (non-preferential) approaches to achieving freer trade. On the one hand, it has sometimes been argued that the acceleration towards trade preferences reflects the deep frustration that countries felt with the ‘slow’ pace of the multilateral process and that PTAs are a faster and more efficient way of achieving trade liberalization. On the other hand, it has been recognized that preferences in trade may result in welfare losses, as imports may be sourced from inefficient partner countries rather than more efficient outsiders because of the lower tariffs faced by the former. This diversion of trade is also potentially costly to outsiders who are relatively handicapped in member country markets and may incur terms-of-trade losses in their exports. It has also been argued that preferential agreements are not to be seen as providing a simple monotonic path to multilateral free trade, warning that preferential agreements might create incentives within member countries against further multilateral liberalization.

For the year 2011, the World Trade Report (WTR) engages this question of preferential trade in depth. It starts by asking how much of merchandise trade is conducted on a preferential basis and points out that despite the proliferation of PTAs in recent years, the amount of liberalization that has been achieved through preferential agreements is actually quite limited. Specifically, only 16% of global merchandise trade (30% with intra-EU trade included) receives preferential treatment and less than 2% of world trade (4% when intra-EU trade is included) is eligible for preference margins greater than 10%. Furthermore, WTR 2011 reports that ‘sensitive sectors remain sensitive’ in preferential agreements. That is, approximately ‘66% of tariff lines with MFN tariff rates above 15% have not been reduced in PTAs’. The exclusion of ‘sensitive sectors’ from liberalization within preferential agreements also suggests that often the very same interest groups that object to multilateral or unilateral liberalizations will oppose liberalization in a preferential context. Taken together, the WTR analysis suggests that preferential agreements have not proven to be as effective in liberalizing trade as their proponents hoped.

Having noted the (only) modest liberalization in trade achieved through PTAs, WTR 2011 focuses on two (recently popular and interrelated) arguments to explain the proliferation of preferential agreements and to speculate on the consequences. The first argument is that the motivation for PTAs may have little to do with the lowering of trade barriers as such, and that PTAs should be understood instead as vehicles for undertaking ‘deeper’ forms of integration to achieve institutional harmonization with partner countries. The second argument is that the motivation for deep integration with partner countries comes from the increasingly fragmented pattern of global production in which the different stages of production are unbundled and spread across different countries. Underlying these arguments appears to be the loosely articulated view that PTAs formed with the intention of achieving deep integration in the context of globally fragmented trade are superior in their motivation and design, having more likely been driven by trade fundamentals rather than protectionist impulses and characterized by institutional arrangements and rules that are often ‘non-discriminatory’ in nature.

How much ‘deep’ integration has been achieved in practice? The WTR data on the ‘anatomy of preferential agreements’ suggest that institutional provisions that fall outside of the WTO mandates, but that are covered by PTAs (and are also deemed legally enforceable by the text of the PTA) are quite few in number. For instance, only 4% of the PTAs include legally enforceable provisions on anti-corruption measures, 12% include provisions on labor regulation, and 11% include provisions on environmental regulations. Regulations on the movement of capital and the protection of intellectual property rights are more significant at 40% and 34% respectively, but many possible provisions are simply not referenced in the vast majority of PTAs. This, in itself, permits some skepticism on how much deeper PTAs, on average, have gone beyond the possibilities offered by the WTO.

Furthermore, as the WTR recognizes, it is less than clear that the achievement of ‘deep integration’ at a regional level is a desirable goal from either a regional or multilateral perspective. Common policies may benefit countries with common policy preferences, but may be costly if there are wide differences in the preferences of member countries (as is often the case with provisions involving environmental and labor standards, for instance). In the context of North–South agreements, there is an additional concern that the greater resources and organizational ability of government and sectoral lobbies in the North will shift policy in a direction that is closer to their own interests and away from the interests of the South. Furthermore, the establishment of policies and standards at a regional level may inhibit multilateral liberalization if the multilateral standards vary from regional ones and there are costs to switching standards. While some features of ‘deep’ PTAs such as competition policy may be non-discriminatory in general, the possibility of negative spillovers on non-member countries cannot be ignored.

Does fragmentation of global production trade provide a new basis for regionalization of trade? While this argument is gaining currency in some quarters, it would seem that production fragmentation should provide greater incentives instead for broader multilateral liberalization. After all, the most efficient producers of any given intermediate good need not lie within the jurisdictional boundaries of any specific preferential agreement, and the identity and location of the efficient producers of intermediates may be expected to vary faster than any country's ability to sign new preferential agreements. As a practical matter, if PTAs were designed to support fragmented production networks, we might expect to see greater geographic concentration of trade over time as many production networks are regional in nature. However, as the WTR notes, the share of intra-regional trade in Europe has remained roughly constant at around 73% between 1990 and 2009. While Asia's intra-regional trade seems to have risen from 42% to 52% during the same period, North America's intra-regional trade shares rose from 41% in 1990 to 56% in 2000 and fell back to 48% in 2009. Thus, no clear pattern linking trade preferences and production fragmentation can be discerned in the data.

Alongside these recently emerging hypotheses regarding the determinants and the ‘true’ nature and motivation of preferential agreements and their putative impact on the world trade system, WTR 2011 uses new data and novel analysis of older data to assess several ongoing debates relating to the prevalence and merits of preferential trade. It also contains a number of very interesting case studies and analytical discussions that inform various economic, political, and legal aspects of this complex question. This important and multidimensional contribution is essential reading for the expert and the novice alike.

2. Political science

Since 2003, the World Trade Organization (WTO) has issued an annual World Trade Report that addresses a topic bearing on trade policy and the multilateral trading system. The World Trade Report 2011 focuses on preferential trade agreements (PTAs), institutions that include free trade agreements, customs unions, common markets, and economic unions. A WTO report on PTAs is long overdue. More than 300 preferential arrangements currently dot the international political economy and every country except Mongolia belongs to at least one. As PTAs have proliferated, so has the amount of trade guided by them. Roughly half of all merchandise trade is conducted between PTA members and there is little reason to expect that this figure will decline anytime soon.

For well over half a century, economists have debated the welfare implications of PTAs, focusing largely on whether PTAs are trade creating or trade diverting and whether they bolster or undermine the multilateral trading regime. The World Trade Report 2011 argues that these debates are outdated. So too, it claims, are analyses of preferential arrangements that center primarily on tariffs and other cross-border trade barriers. The reason is that very little trade currently receives preferential treatment within PTAs. Over half of world trade is subject to a most favored nation (MFN) tariff rate of zero so no preferences can be granted, and preference regimes often contain exemptions whereby trade in a given good is subject to the MFN rate.

Instead of focusing on these issues, the report argues, attention should be redirected at the rise of ‘deep’ PTAs. The concept of depth is rather murky, but the report follows a number of studies in characterizing shallow integration as that which shrinks or eliminates ‘border measures’, such as tariffs. Conversely, deep integration involves ‘inside the border’ measures, such as investment provisions, harmonization of regulations across countries, intellectual property protection, competition policy, and the like. In certain cases, deep integration also stems from the delegation of authority over economic policy to a supranational body. Central to the rise of these deep arrangements, the report claims, are international production networks, which break down the production process across various countries in order to reduce costs. As global production networks have become increasingly pervasive, they have also become of rising importance to international trade since these networks involve the movement of goods and services across national borders at various stages of the production process.

The central claim of the World Trade Report 2011 is that global production networks foster deep PTAs and that such PTAs, in turn, reinforce these networks and the international trade flows that these networks generate. Furthermore, deep PTAs are usually trade creating; they also tend to reinforce and further strengthen rules that are embodied in existing WTO agreements. The clear implication is that deep integration is welfare enhancing. Equally, the growth of deep integration requires scholars and policy makers to reconsider what is meant by preferential trade and to reassess whether existing theories explain emerging patterns of PTAs.

There is much to recommend about this report. It summarizes a great deal of research on various aspects of PTAs, conducts a set of interesting and important analyses, and includes a substantial amount of valuable data. Furthermore, the report is surely correct that the content of various PTAs has changed and evolved in recent years. Its emphasis on the depth of integration in PTAs is useful and strongly suggests that more attention should be focused on the design of these arrangements (e.g., Downs et al., Reference Downs, Rocke and Barsoom1998; McCall Smith, Reference McCall Smith2000; Koremenos et al., Reference Koremenos, Lipson and Snidal2004; Busch, Reference Busch2007).

However, as the report acknowledges, the concept of deep integration is both blunt and rather imprecise. It notes, for example, that ‘the concept generally refers to any agreement that goes beyond shallow integration’ (p. 113), hardly a satisfying definition. Partly as a consequence of this drawback, it is difficult to assess how widespread deep integration has become. The report strongly suggests that such integration is becoming more pervasive; but without better definitions and measures of deep integration, it is hard to determine the extent to which this is actually the case. In the same vein, it is important to develop operational measures that distinguish different degrees of deep integration. The report argues that PTAs marked by any inside the border policy are deep, but it seems to draw no distinction between a PTA that includes a single policy of this sort and one that includes many.

One of the report's key claims is that ‘traditional theories’ of PTAs are designed to explain shallow integration agreements, but fail to explain both the causes and effects of deep integration agreements (p. 114). As a result, the report can be read as a call for further theorizing on deep integration and such research will undoubtedly prove useful. However, the report gives short shrift to a set of existing political theories that may also shed light on this issue. Some of these theories are reviewed briefly in the report (pp. 95–97), but they are not linked to the depth of integration.

First, neo-functionalists have long argued that integration occurs when economic transactions become dense enough among a group of countries that organized economic interests benefitting from these transactions pressure governments to manage economic interdependence by centralizing policies and creating common institutions (e.g., Mitrany, Reference Mitrany1943; Haas, Reference Haas1958, Reference Haas1964). Neo-functionalists stress that any initial decision to integrate because of such dense transaction networks produces, and unintentionally creates, both economic and political spillovers that ultimately deepen regional integration. Cooperation between countries on certain economic issues is likely to trigger cooperation in other related areas (Lindberg, Reference Lindberg1963).

Second, the number of domestic ‘veto players’ (that is, actors with the capability to block policy change) in member states may affect the depth of integration in PTAs. As the number of veto players rises, domestic ratification of a deep PTA becomes more difficult (Mansfield and Milner, forthcoming). The magnitude of domestic change needed to comply with a trade agreement and the associated political costs borne by leaders for entering it depend on the extent of integration that the agreement aims to achieve. Arrangements that envision more extensive integration are likely to cover more goods and services and therefore to affect more sectors and a larger segment of society. Deeper arrangements tend to reduce the decision-making power of certain veto players (such as domestic legislatures), increase the adjustment costs and the portion of society affected, and attenuate the ability of domestic groups to lobby the government. The result is mounting opposition to trade agreements among the increasing number of segments of society that will anticipate being adversely affected. The increasing number of veto players that represent those affected groups will likely resist these expected consequences of deeper integration and thereby impede ratification.

Third, domestic institutions may influence the depth of integration. One possibility is that democracies tend to enter deeper PTAs than other regime types (Mansfield and Milner, forthcoming). Since leaders cannot credibly commit to ignore special interest pleading for trade protection, voters may hold heads of state responsible for adverse economic conditions stemming from protectionism even if these economic conditions were not caused by policies driven by the demands of special interests. Leaders can help address this problem by entering a PTA. Since this problem is more severe in more competitive electoral systems, democratic chief executives are especially likely to join preferential arrangements. PTAs can ameliorate these credible commitment problems, and deeper integration agreements can provide greater credibility for leaders. Democratic leaders thus have greater incentives to sign agreements that aim to achieve fairly extensive integration, while autocrats do not.

Finally, political–military relations may affect the depth of integration among PTA members. Deeper integration is expected to yield greater economic benefits for the PTA as a whole, but also to have more profound distributional consequences among the participating countries. Since the distribution of these economic benefits can influence political–military power relations, political antagonists have few reasons to form a PTA and fewer still to form a deep arrangement. Those members that derive fewer economic benefits from the PTA are likely to experience a reduction in their national security relative to those members that gain more. In PTAs composed of allies, by contrast, the economic gains from deep liberalization bolster the alliance's overall political–military capacity. This, in turn, promotes the security of all members and attenuates concerns about the political–military implications of the distribution of economic gains from the PTA (Mansfield, Reference Mansfield, Anderson and Blackhurst1993; Gowa, Reference Gowa1994).

In this regard, it is no coincidence that the World Trade Report 2011 observes that ASEAN has generated deep integration, whereas various African PTAs have not. After all, ASEAN was intended to dampen political–military tensions among states in the region and promote political cooperation. From a political standpoint, it is entirely understandable that this PTA led to deeper integration than those in Africa, which has been riddled with political hostilities and adversarial relations, even among PTA members.

These critiques should not be interpreted as implying that the World Trade Report 2011 disregards the politics of preferential trade. In fact, at various points, the authors note that political factors are central to an understanding of PTAs. It is obvious that no study can cover everything and this report offers a fresh and insightful analysis of many key features of the economics of preferential trade. But it also raises various issues where a fuller understanding of the politics and the political economy of PTAs is likely to prove valuable.

3. Law

This assignment, happily accepted, was to consider international law aspects of preferential trade agreements (PTAs) in light of the WTO's 2011 World Trade Report.Footnote ¹ After reading the relevant Report sections however, it became somewhat clear that a general public international law treatment of the subject, while interesting, would be a bit ‘over-academic’ for the task and not well related to the way the Report is structured and the primary points being made that have legal implications. We understand anyway that the great bulk of applicable international law to trade agreements (WTO and PTAs) is a matter of treaty law and the customary rules of interpretation. Thus, the main international law focus remains on interpretation of the WTO provisions and how they relate to the law created among members to PTAs.

With that small caveat in mind, the WTO Report offers an opportunity to visit some important legal aspects of the relationship between the WTO and PTAs that the authors raise for attention. This material is mainly found in Part E titled, ‘The multilateral trading system and PTAs’. This is not the only place in the Report where legal aspects are considered, but on the whole this Part, dealing in turn with sections on the systemic effects of PTAs, deep PTA provisions, the historical perspective of regionalism in the WTO, and the relationship between PTAs and the WTO, contains the material for discussion here.Footnote ²

The theme of the Report centres on the rising prominence of ‘deep integration’ provisions in PTAs as compared to ‘shallow agreements that mostly deal with border measures’.Footnote ³ This ‘deep integration’ term is used to refer to trade agreement provisions concerning rules on domestic policies, the bulk of which are in the character of domestic laws and regulations. The Report draws this picture through an extensive documentation of this deep integration evolution to increasingly binding forms and then goes on to consider the implications presented for the WTO system. For the legal aspects, both substantive and institutional elements are considered and eventually drawn together with the other disciplinary threads for a closing discussion on the ‘coherence’ between PTAs and the WTO and how it can be managed or improved. The substantive aspects of deep integration provisions consider the application of general WTO rules on discrimination. The institutional side considers the relationship between the different dispute settlement systems in play between PTAs and the WTO. The ‘documentation’ referred to above is an impressive Secretariat study analyzing dozens of PTAs for their deep integration provisions which are also sorted into WTO plus (WTO+) and WTO extra (WTO-X). These are then sorted as to their ‘binding’ or ‘nonbinding’ character.Footnote ⁴ While beyond the discussion here, the sections detailing ‘PTA commitments in selected policy areas’ also provides some nice terrain for legal researchers seeking credible generalizations about the types of provisions being found in PTAs for services and investment, technical barriers to trade, and competition policy.

This study of the provisions allows the theme to go forward in Part E that many (or most) of the deep integration areas being treated in PTAs are often non-discriminatory as to WTO third parties. Cooperation in areas that are inherently concerned with the formulation and application of domestic regulatory regimes do not easily lend themselves to discriminatory application. A domestic regime tends by its nature to apply to all. This section provides examples for a number of deep integration areas including services, where ‘barriers to trade … are generally behind-the-border’; mutual recognition agreements for product testing, which do not discriminate on the origin of the product being tested; and domestic competition policy regimes, which have the potential to improve the opportunities for foreign enterprises generally in the domestic market relative to domestic competitors.Footnote ⁵ Similarly, the Report refers to transparency provisions in PTAs that tend to be supportive of existing provisions in the WTO (e.g., The TBT Agreement) and therefore tend to enhance the effectiveness of transparency overall.

The WTO non-discrimination issue as to deep integration is clear enough on the surface. If a provision enhancing a domestic regulatory subject area is not by its nature able to be formulated or applied in a manner that can distinguish goods or services on the basis of their origin, well, then, we do not have a WTO discrimination issue. For an old ‘MFN detective’ searching the back alleys of PTA regulatory provisions for hidden discrimination, there is some reluctance to accept the Report's findings on this as a general proposition. However, the Report on this point is also not conclusory. As mentioned, the documentation of the Part D study is extensively made across a broad range of PTAs and provisions. If a constructive generalization cannot be formed based on this kind of data and analysis, then such a generalization is probably not possible on any. That said, one also notes the Report's own caveats along the way, for example in the characterization of ‘often’ non-discriminatory rather than ‘always’. There are also recognized points of analysis that cannot be well completed in regard to discrimination. As noted, it is difficult to consider the de facto application of a domestic regime or to have consistent data on how service suppliers are actually treated in the market.Footnote ⁶

What also lingers are some potentialities in PTA regulatory provisions as they could be expected to continue to evolve and become combined with improving PTA institutional structures. One considers whether similar non-discrimination generalizations will be able to be made for mutual recognition agreements that go beyond conformity assessment to recognizing foreign standards as equivalent, or for more advance transparency notifications that provide a strong and selective right of comment (and redress) at early consideration stages. While these types of bugs are admittedly concerned about what might occur rather than what is evident today, we have already moved quite some ways from the early PTA days of very non-binding cooperation endeavours and very weak or non-existent dispute settlement recourse.

A further consideration for the Report is that the legal concepts underlying non- discrimination – defining its scope in the first place – are not given that intensive a legal treatment. This is also the case for the interpretation issues regarding the scope of Article XXIV and GATS V provisions as they may or may not apply to grant a non-discrimination exception for a deep integration subject area found otherwise to be discriminatory. Understandably, this kind of treatment is beyond the Report's mission, but if we are going to say that deep provisions are inherently non-discriminatory, then it can also be noted that the scope of those WTO general obligations are quite broad and capable of capturing, for trade in goods, any internal regulation that affects the internal sale of a product; and for trade in services, ‘any measure covered by this Agreement’ which means measures affecting trade in services.Footnote ⁷ This makes for a wide net, and certainly wide enough to capture the full range of regulatory possibilities in a PTA.

While there are no panel interpretations for Article XXIV:8 ‘other restrictive regulations of commerce’ and one does not easily document its scope of application, we can also acknowledge that this has been at issue in the academic literature for some time now. In particular, the term may or may not relate to PTA regulatory cooperation activities undertaken under the coverage of another Annex 1A Agreement, or for that matter, undertaken in another WTO Agreement, for example in the TRIPS. In services provider recognition, it is also an established question whether or not a bilateral or regional recognition agreement can be included within a GATS V Economic Integration Agreement without invoking the MFN provisions in the GATS ‘Recognition’ Article VII. It is not incumbent on the Report to attempt to analyze and answer these questions, but some recognition of the peculiarities of dealing with regulatory discrimination issues and their available regional exceptions would be helpful.

The Report turns to the institutional side on the subject of dispute settlement and the relationship between the WTO and PTA dispute settlement systems. This is an important legal element in the WTO coherence discussion, and as well in general public international law where a body of literature has developed on dispute settlement fragmentation.Footnote ⁸ The WTO cases themselves have been at the centre of much of this interest both in reference to other treaties and arbitral systems with arguably overlapping jurisdictions and for the closely related substantive issue of determining whether applicable law (outside the scope of the WTO covered agreements) may or may not be applied in a WTO dispute, here as between PTA members in a WTO case. Another facet of that question is the possible application of WTO provisions between PTA members in their own arbitral systems. A third angle is the presentation of concurrent or consecutive claims in both regional and WTO systems on the same subject matter. The Report commences by noting the different types of PTA systems in use and then divides its treatment into a description of how the different dispute systems may overlap. This is followed by a discussion of the concerns that have been raised over the coexistence of dispute systems together with some of the recommendations that have been made to reduce the potential for conflicting rulings.Footnote ⁹ A separate section summarizes the WTO cases where PTA dispute settlement issues have risen, including Mexico–Taxes on Soft Drinks,Footnote ¹⁰ which is understood to be the main Appellate Body Report on the subject. This case also illustrates the point of one of the recommendations for avoiding conflicts, this suggesting the need for more clearly drafted or more strict ‘fork in the road’ clauses in the PTAs themselves.Footnote ¹¹

As desirable as that seems, it is also a somewhat difficult remedy in the absence of PTA drafters' adherence to the notion that the WTO has an exclusive zone of jurisdiction for claims under the covered agreement in the first place. As noted in the Report, this seems to be the clear text of the WTO Dispute Settlement Understanding and the tenor of the Soft Drinks reports.Footnote ¹² The problem remains however that an identical subject matter claim to that of a covered agreement may be brought in a PTA dispute according to its own Article provision. That a PTA member may lose first in that forum and then bring the same claim anew in the WTO would also appear to raise the possibility of asserting some general public international law principles to avoid re-arbitrating an identical dispute. This is noted in the Report by its references to comity, res judicata and lis alibi pendens.Footnote ¹³ Although the Appellate Body in Soft Drinks did not foreclose the possibility of being presented with a legal impediment to exercising a WTO panel's jurisdiction in a future case, the more predominant tone of the case nevertheless seems to stand with the idea that a WTO panel given a claim made under a WTO covered agreement cannot (or will not) refuse its responsibility to determine the rights and obligations of the parties.Footnote ¹⁴

While this approach upholds the sanctity of the availability of the DSU, it may also appear somewhat isolating and contrary to the desire of some to see less fragmented systems of international dispute settlement. However, it also appears to avoid (for now) an underlying paradox arguably presented by the Soft Drinks case in the Appellate Body's refusal to consider a NAFTA provision as a source of possible applicable law to be applied in the WTO case.Footnote ¹⁵ This would seem to lead inevitably to the question of how a WTO panel would give effect to a PTA ‘fork in the road’ clause – being raised by the respondent to argue the absence of the WTO panel's jurisdiction – without applying it and necessarily interpreting it as a source of applicable law between the parties? While the Report does a concise job of setting out the issues of coexisting dispute settlement systems, the WTO legal context itself for that discussion may be a bit off the mark if the Appellate Body has already determined that a regional exclusion clause cutting off WTO recourse is not a legal issue in the WTO at all, but rather a matter of PTA law and therefore ‘their problem’. The same possibility is present for the risk of multiple claims and varying outcomes, as in, ‘their problem’ and not the WTO's.

Perhaps the final section in this part presenting WTO dispute settlement data on cases between PTA members also validates the ‘what, us worry?’ conclusion as above. This summation indicates that the share of WTO disputes between PTA partners advancing to the panel stage is close to the overall average of the disputes between non-PTA parties. Apparently, PTA members are not bailing out of the WTO system after all, at least not so far. One might note that this may change over time in systems that move from ad hoc panels to standing arbitral bodies to develop a more coherent body of regional law.

The Brazil–Tyres case is also pertinent on the question of conflicting dispute rulings and noted here for its absence of treatment in the Report section on dispute settlement but treated later in a section on the historical perspective under ‘WTO developments’ along with the landmark Turkey–Textiles case.Footnote ¹⁶ This seems to be a missed opportunity where the treatment of Brazil–Tyres tends to focus on the Article XX ruling somewhat narrowly as applied to Brazil's discriminatory prohibition of imports in favour of other Mercosur members. That Brazil could have avoided a conflict in rulings by also seeking the use of the comparable public health exception under the Mercosur treaty before that court seems to miss the implication presented by the ruling both for the Report's dispute settlement discussion as well as for the later and concluding coherence discussion.

For this, consider if Brazil had raised such a defence in its regional case but the Mercosur court ruled it an invalid invocation of the human health exception according to the dictate of the Mercosur treaty's own free movement provisions; that is, its own regional law. Such a ruling could also not find validation as a form of justifiable discrimination in the GATT Article XX case. Brazil–Tyres is clear that the grounds of allowable discrimination under the Article XX Chapeau must relate to the objective that was found to justify the measure from the listed Article XX exceptions.Footnote ¹⁷ How could any regional court ruling based on free movement provisions in the other treaty possibly be connected to an invoked Article XX exception? At least for the purpose of GATT and GATS general exceptions, this ruling on conflicts seems fairly obvious. The WTO general exception chapeau law prevails over all regional law to the extent of any ruling that results in discrimination.

This starts to inform a hierarchy discussion that could make a meaningful contribution to the Report's closing of the chapter about enhancing coherence between the WTO and PTAs. Both the Brazil–Tyres and the Turkey–Textiles cases go a long way to set the nature of the hierarchy that arises when a WTO violation is defended by the MFN exception possible in a WTO regional trade agreement. We know already that a panel's responsibility is to assess that agreement for its compatibility with the applicable WTO provisions. For free-trade areas and customs unions, this means its assessment according to paragraphs 5 and 8 of Article XXIV. While the Brazil–Tyres panel begged off doing that assessment on the basis of judicial economy, and the Appellate Body did not need to make the assessment as it found the measure did not comply with Article XX, its final paragraph on this seems to offer future guidance to panels to make findings for the DSB ‘in order to insure effective resolution of disputes to the benefit of all Members.’Footnote ¹⁸ The Article XXIV assessment should have been done by the panel.

As the Report takes up the ‘coherence’ discussion, it considers whether clarification of the WTO rules might occur via WTO dispute settlement. It rightfully notes that there are few cases and Members have been reluctant to bring disputes to clarify the rules, and that this does not ‘seem to be a promising starting point’ for enhancing coherence.Footnote ¹⁹ But this is not to say that cases have not occurred – as they have – and those rulings and interpretations should form the legal context of how we look at coherence between the WTO and PTA legal regimes. To put it another way, the legal context for the discussion of coherence deserves to be present and accounted for. The Appellate Body has already clarified the WTO rules for regional trade agreements on important points, the first being that PTAs in the WTO are ‘conditional exceptions’ and when raised as a defence, the PTA member has the burden to validate the regional agreement. This means that the complainant can and will attack the regional agreement, just as the EC did before the Brazil–Tyres panel on a number of points relating to ‘substantially all trade’ coverage and the conformity of Mercosur's external common tariff. Together with that case's ruling on the application of the GATT Article XX Chapeau, we see the outlines of a lawful hierarchical relationship between the WTO and PTAs, as a matter of WTO law. That is a form of coherence.

The Report also gives little credence to the ongoing negotiations to clarify the GATT provisions in Article XXIV, concluding that, ‘it seems unlikely that … clarifying and strengthening existing rules – would be viable’.Footnote ²⁰ Besides the point that many of the ‘substantially all trade’ submissions made by Members in the Rules negotiations deserve better attention than that, one wonders if these conclusions are what the WTO Membership ‘wants’ to hear as contrasted with what they perhaps ‘need’ to hear, as in ‘dispute settlement does not do anything to affect our PTAs and we are not called upon either to do much on the negotiation side of it, so we are pretty happy with the situation as it stands overall’. If so, this is also a form of continuing group disillusionment. Dispute settlement poses a true legal hazard for a number of PTAs and the clarification of the rules (either within the Doha mandate or otherwise) is outstandingly in the general interest of all the Members in light of the alternative of having a panel ultimately configure the definition of ‘substantially all trade’ for them. This has been known since Turkey–Textiles and Brazil–Tyres served as one more step to a panel assessment of a regional trade agreement. This is perhaps why this reviewer prefers to see the negotiations on Article XXIV clarification as a glass half full rather than one half empty. The submissions have gone further than those in the Uruguay Round and the issues to be settled are recognized and fairly clear. The outlines of ‘substantially all trade’ are there. It is even possible that a form of PTA practice has tacitly evolved that services an attempt by most PTAs to meet the coverage requirements more or less as proposed. Admittedly, this is not the case for the term ‘other restrictive regulations of commerce’, the area which touches the deep integration issues that rise to the fore in the Report, but also further down the list of clarifications in the negotiations. That one will also need to happen, but it is a complex expression and its relation to domestic regulation should take considerable thought and time. ‘Substantially all trade’ on the other hand is a cake three-quarters baked. It is almost ready to go.

This idea of ‘practice’ is worth visiting. The transparency instrument for notification of PTAs contributes to the possibility of the development of a common and acceptable practice and it also informs the Report's proposal for clarifying the rules over time by a softer law instrument, a code of good practice for PTAs. The rationale for this soft law approach would be to ‘allow WTO Members to better understand their respective priorities and interests’, and eventually, ‘unblocking progress towards legal interpretations of particular provisions that would insure coherence’, as would completion of the rules negotiations on this subject in the WTO.Footnote ²¹ One senses here that a soft law approach is a not so implicit admission that the rules negotiations appear hopeless. One can hope not, but, either way, the two tracks do not seem mutually exclusive and the soft law proposal would at least not seem to cause any further harm.

What would also be helpful for advancing both of these fronts is a more clear recognition from the outset that WTO Members in a PTA do not live in an isolated legal solar system. Their PTA systems remains within a larger WTO legal galaxy and are continuingly bound by those rules, whether clarified or not. That some deep integration subjects fall outside of that WTO legal framework is understandably the case, just as there are many areas of deep integration that relate to the existing WTO regimes. How they relate is the central legal question that informs coherence for deep integration PTAs. The Report can be applauded for the focus on these regulatory areas and hopefully it will help move the discussion forward, just as a discussion about coherence that sets the legal framework at the outset would contribute well to that goal.