Bilateral free trade agreement a critical assement and WTO regulatory reform proposal

Folsom, © 2008 * At this writing, most-favored-nation MFN principles notwithstanding, every nation saveMongolia has entered into at least one bilateral or regional free trade agreement..

Legal Studies Research Paper Series

Research Paper No 08-070

September 2008

Ralph H Folsom

This paper can be downloaded without charge from the

Social Science Research Network Electronic Paper Collection:

Professor of Law, USD Law School Prof Folsom thanks the University of San Diego for a Summer

*

Research Grant in support of this publication.

BILATERAL FREE TRADE AGREEMENTS:

A CRITICAL ASSESSMENT AND WTO REGULATORY

REFORM PROPOSAL

Published by Prof Ralph H Folsom, © 2008 *

At this writing, most-favored-nation (MFN) principles notwithstanding, every nation saveMongolia has entered into at least one bilateral or regional free trade agreement The EuropeanUnion, for example, is so heavily engaged in bilateral deals that it has MFN trade relations withonly seven countries Hundreds of bilaterals have been negotiated since the early 1990s Theapparent failure of the Doha Round virtually guarantees their dominance of international tradelaw and policy

This article reviews the history of bilateral and regional free trade agreements

(“bilaterals”), failed attempts at their regulation under the General Agreement on Tariffs andTrade (GATT) and the World Trade Organization (WTO), and critically assesses the significance

of their proliferation for international trade law and policy Reform of WTO jurisdiction overbilaterals concentrated in the Dispute Settlement Understanding (DSU) is proposed Specifically,abandonment of all WTO regulation except for a Revised Transparency Mechanism (text

attached) operating in conjunction with expanded DSU opportunities is suggested

A word on nomenclature: This article utilizes the term “bilateral(s)” because it bestcharacterizes the free trade agreements that are presently sweeping the seas Generally in

GATT/WTO parlance, “regional trade agreements” (RTA) has been and continues to be

employed Those words, in my opinion, are not the best descriptors of what is happening when,for example, China and Chile, Japan and Mexico, the United States and Oman, and the EuropeanUnion and South Africa agree on free trade

Why Bilaterals?

A variety of factors help explain why bilaterals have become the leading edge of

international trade law and policy Difficulties encountered in the Uruguay, “Seattle” and DohaRounds of multilateral trade negotiations are certainly crucial GATT/WTO regulatory failuresregarding bilaterals have also fueled this reality Yet these “negatives” do not fully explain thefeeding frenzy of bilaterals

A range of attractions to bilaterals are also at work For example, bilaterals often extend

to subject matters beyond WTO competence Foreign investment law is a prime example, and1

many bilaterals serve as investment magnets Government procurement, optional at the WTOlevel, is often included in bilaterals Competition policy and labor and environmental matters2 3

absent from the WTO are sometimes covered in bilaterals In addition, bilaterals can reach4

beyond the scope of existing WTO agreements Services is one “WTO-plus” area where this isclearly true Intellectual property rights are also being “WTO-plussed” in bilateral free trade5

agreements Whether this amounts to competitive trade liberalization or competitive trade6

imperialism has been provocatively explored by Prof Bhala.7

Further, bilaterals are politically and economically selective In other words they avoidnot only global most-favored-nation principles, but also domestically “sensitive” areas of

national politics and economics For example, Singapore’s absence of farm exports helped make

it an ideal U.S and Japanese free trade partner The micro-sized economy of Chile contributed toits attraction as a free trade partner with Mexico, China, the European Union, the United Statesand others U.S free trade deals with Jordan, Bahrain and Oman fit economically in a similarfashion, not to mention national security objectives 8

Like it or not, the “spaghetti-bowl” maze of bilaterals is driven by powerful negative and9

positive forces It is not only the preferred trade medium of today, but very likely the future Already more than half of world trade is conducted under bilaterals While international trade10

lawyers may celebrate full employment, it bears remembering that bilaterals are discriminatory.

They could render MFN the least favored status in world trade Such an outcome would beespecially harmful to the world’s poorest nations, those with whom few WTO partners seek abilateral agreement

Most-Favored-Nation Freer Trade Principles

When the United States and its allies emerged victorious after WWII, they faced a basicquestion What should be the international economic order in the post-war era? A return to tradeprotectionism, globally rampant in the 1930s, was unthinkable (unless you like economic

depression) But what about widespread trade-inhibiting national laws, for example the U.S.Smoot-Hawley Tariff Act of 1930?11

The answer came to be embodied in the General Agreement on Tariffs and Trade of 1947(GATT 1947) Ironically this agreement was not intended to be more than a temporary stopgap12

pending creation of the broadly conceived International Trade Organization (ITO) When13

ratification of the ITO failed in the U.S Congress, GATT 1947 became by default the anchor of14

post-war international trade law

GATT 1947 was never about free trade, merely freer trade This it achieved over decadesthrough tariff-reducing multilateral trade negotiations (known as “negotiating rounds”) and ever15

an expanding membership At its core, GATT 1947 and its successor GATT 1994 embrace the16

principle of general most-favored-nation (MFN) trading This principle is essentially one ofnondiscrimination, that is to say a rejection of discriminatory tariff and trade preferences Article

1 of the GATT (1947) and (1994) spells out the golden rules of MFN trading:

With respect to customs duties and charges of any kind imposed on or in

connection with importation or exportation or imposed on the international

transfer of payments for imports or exports, and with respect to the method of

levying such duties and charges, and with respect to all rules and formalities in

connection with importation and exportation, and with respect to all matters

referred to in paragraphs 2 and 4 of Article III, any advantage, favour, privilege or

immunity granted by any contracting party to any product originating in or

destined for any other country shall be accorded immediately and unconditionally

to the like product originating in or destined for the territories of all other

contracting parties (emphasis added)

GATT Article 24

The origins of Article 24 have long remained obscure Happily, Professor Kerry Chasehas recently illuminated those origins, emphasizing that a secret Canada-US free trade agreement(never realized) explains much of its content Primarily for this reason, Article 24 of the GATT17

(1947 and 1994) permits members to enter into “free trade area” and “customs union”

agreements of a fixed or interim character, provided tariff compensation is given to

nonparticipants in customs unions This permission constitutes a major exception to18

nondiscriminatory MFN trade Instead, such agreements embody trade preferences which

simultaneously create trade among their signatories and divert trade between those signatoriesand the rest of the world These effects are easily evident when customs unions are created19

because such agreements contain common external tariff and trade law provisions.

With free trade agreements, diversionary trade effects are usually not distinct because ofthe absence of a common tariff and trade wall against outsiders Trade diversion nonethelessoccurs “Rules of origin” in free trade area treaties keep third-party imports from seeking thelowest tariff or highest quota state and then exploiting trade advantages within a free trade area Under typically complex rules of origin, free trade areas are “free” only for goods substantiallyoriginating therein This causes member state goods to be preferred over goods from other states,the essence of trade diversion.21

Article 24 of the GATT attempts to manage the internal creating and external diverting effects of bilaterals Under its terms, free trade area and custom union proposals should

trade-be notified “promptly,” although most agreements have trade-been notified after their conclusion 22 23

Once notified, bilaterals should run the gauntlet of formal review and report procedures duringwhich recommendations are possible to bring them into conformity with Article 24 Prior to24

the WTO, GATT “working parties” conducted these procedures Since 1995, the WTO

Committee on Regional Trade Agreements (CRTA) reviews Article 24 notifications

Conformity recommendations might, for example, deal with Article 24 requirements forthe elimination of internal tariffs and other restrictive regulations of commerce on “substantiallyall” products originating in a customs union or free trade area Or they might concern Article25

24 requirements that tariffs and other regulations of commerce not be “higher or more restrictive”than before creation of the free trade area or customs union Whether the term “regulations of26

commerce” includes rules of origin is critically unclear The broad purpose of Article 24,27

acknowledged therein, is to facilitate trade among the GATT/WTO parties and not to raise trade

Early Regulatory Failure

The inadequacy of Article 24 as a regulatory mechanism quickly became apparent Thiswas perhaps most noticeable regarding the politically and economically significant customsunion treaties of western Europe With the European Coal and Steel Community (1951) onlytwo products were involved Clearly no case could be made for its compliance with the

requirement of elimination of internal trade barriers on “substantially all” products Hence theGATT members, passing over Article 24's own waiver proviso for proposals leading to a

customs union or a free trade area “in the sense of Article 24,” reverted to Article 25 That29

article allows a two-thirds vote by the contracting parties to waive any GATT obligation

In reviewing the European Economic Community Treaty (1957), many “violations” of theletter and spirit of Article 24 were cited For example, the derivation of the EEC common

external tariff by arithmetically averaging existing national tariffs was challenged as more

restrictive of trade than previous arrangements Such averaging on a given product fails to takeaccount of differing national import volumes If a product was faced originally with a lower thanaverage national tariff and a larger than average national demand, the new average tariff is clearlymore “restrictive” of imports than before Averaging in high tariffs of countries of low demandquite plausibly created more restrictions on third-party trade If so, the letter and spirit of Article

while GATT officials viewed them as rather open efforts at purely preferential tariff status Similar problems arose later in GATT review of the multitude of “interim” EEC free trade areatreaties with Mediterranean nations In 1975, the openly preferential and discriminatory Lomé32

Convention negotiated between the European Community and forty-six African, Pacific andCaribbean nations (including many former colonies) challenged the interim character of

Community “free trade areas” with developing states Once again it was the GATT and not the33

European Community that gave way

Despite these and other arguments, the 1957 EEC Treaty passed through GATT study andreview committees without final resolution of its status under Article 24 Postponement of theseissues became permanent GATT attempts through the lawyer-like conditions of Article 24 tomaximize trade creation and minimize trade diversion must be seen in the context of Europeanintegration as generally inadequate Treaty terms became negotiable demands that were notaccepted This regulatory failure is representative of 60 years of GATT/WTO practice Asseveral notable commentators have observed, Article 24 standards for bilaterals are deceptivelyambiguous, make little economic sense, and in application have proved a dismal experience

“if not a fiasco.”34

Developing Nation Bilaterals

Developing nations in Africa, the Caribbean, Central America, South America and

Southeast Asia (among others) had free trade and customs union agreements in place as early asthe 1960s In 1979, under what is commonly called the Enabling Clause, the GATT parties35 36

decided to permit developing nations to enter into differential and more favorable bilateral,

regional or global arrangements among themselves to reduce or eliminate tariffs and nontariff

barriers applicable to trade in goods Like Article 24, the Enabling Clause constitutes an

exception to MFN trade principles It has generally been construed to authorize third world freetrade area and customs union agreements Whether the Enabling Clause was intended to takesuch agreements out of Article 24 and its requirements, or be construed in conjunction therewith,

is unclear However, the creation of alternative notification and review procedures for EnablingClause arrangements suggests Article 24 is inapplicable

Notification to GATT of Enabling Clause arrangements is mandatory Since 1995, the37

WTO Committee on Trade and Development (CTD) is the forum where such notifications arereviewed, but in practice not examined in depth Enabling Clause arrangements should be38

designed to promote the trade of developing countries and not raise external trade barriers orundue trade difficulties Consultations with individual GATT members experiencing such39

difficulties must be undertaken, and these consultations may be expanded to all GATT members

if requested Unlike GATT Article 24, neither compensation to nonparticipants nor formal40

reporting on the consistency with the Enabling Clause of developing nation arrangements isanticipated Despite these relaxed procedures, the Enabling Clause of 1979 did not trigger a rush

of third world bilaterals More recently, the ASEAN-China (2004), India-Sri Lanka (2002),41

and “revived” Economic Community of West African States (ECOWAS 2005) agreementsillustrate notified but unexamined preferential arrangements sheltered by the Enabling Clause.42

Regulatory Reform: Interpreting Article 24

The early regulatory failure of Article 24 and the limited requirements of the EnablingClause arguably created an incentive to reach free trade area and customs union agreements as ameans to avoid MFN trade principles In the 1960s, 1970s and into the 1980s, a goodly number

of bilaterals were established, especially in the developing world Yet there was no avalanche

of agreements, in part because of a steady stream of MFN successes in GATT negotiating

rounds The turning point came when major delays and perceptions of possible failure in the44

Uruguay Round (1986-1994) accelerated the creation of bilaterals, most visibly the Canada-U.S.FTA of 1989 and NAFTA (1994) The emergence, also, of export-driven (not import

substituting) developing economies, such as Mexico and Chile, also contributed to this45

acceleration

The Uruguay Round, which created the World Trade Organization, presented an

opportunity to come to grips with the regulatory failure of Article 24 and the implications of theEnabling Clause Agreement was reached in 1994 on an “Understanding on the Interpretation ofArticle 24,” which presently binds the roughly 150 member nations of the WTO This46

Interpretation reaffirms that free trade area and customs union agreements must satisfy the

provisions of Article 24, clarifies the manner in which before and after evaluations of common47

external tariffs are to be undertaken, limits in most cases interim agreements to 10 years, and48 49

details Article 24 notification, report and recommendation duties and processes Most50

importantly, the 1994 Understanding on Interpretation expressly permits invocation of standardWTO dispute settlement procedures (DSU) regarding any Article 24 matters That said, the51

1994 Understanding did not come to grips with the systemic ambiguities that led to Article 24'searly and ongoing regulatory failure.52

Integrated Services’ Agreements

DSU procedures may also be invoked since 1995 regarding “economic integrationagreements” (EIAs) covering services under Article 5 of the General Agreement on Trade in

Services (GATS) Such agreements, which can be staged, must have “substantial sectoral

coverage,” eliminate “substantially” all discrimination in sectors subject to multilateral53

commitments, and not raise the “overall” level of barriers to trade in GATS services compared54

to before the EIA EIAs involving developing nations are to be accorded “flexibility Like55 56

GATT Article 24 customs unions, there is an Article 5 duty to compensate EIA nonparticipants.57

Review of GATS Article 5 notifications is undertaken, when requested by the WTOCouncil for Trade in Services, by the Committee on Regional Trade Agreements Thus, whereasCRTA examinations of GATT Article 24 agreements are required, such examinations are

optional under GATS Nevertheless, numerous Article 5 examinations have been conducted,including notably the services components of NAFTA, the EEC Treaty (1957) and EU

Enlargement (2004), Japan’s FTAs with Singapore, Mexico and Malaysia, China’s FTAs withHong Kong and Macau, and various U.S bilaterals None of these examinations have resulted58

in a final report on consistency with GATS Article 5

The Impact of the Seattle and Doha Rounds

The failure to launch a new round of WTO negotiations in Seattle (1999), followed bydelays and perceptions of possible failure in the Doha Round that commenced in 2001, hascontributed to a veritable feeding frenzy of bilaterals Well over 100 new agreements have been

notified to the WTO, and a large additional number are believed not to have been notified In59 60

general, most of the notified agreements are bilateral, not regional in character Meanwhile, theWTO Regional Trade Agreements Committee, working by consensus, has been unable since

1995 to complete even one assessment of a bilateral agreement’s conformity to GATT Article 24

or GATS Article 5 The same is true for WTO Committee on Trade and Development61

“review” of Enabling Clause arrangements.

It has been suggested that this record can be explained by the ambiguous relationshipbetween Committee reports and WTO dispute settlement proceedings For example, can such62

reports be used in evidence in WTO dispute proceedings? Can fact-finding by WTO Secretariat and information gathered for WTO regulatory purposes be similarly used? This “dispute

settlement awareness” makes WTO members reluctant to provide information or agree on

conclusions that could later be used or interpreted in DSU proceedings.63

The Transparency Mechanism of 2006

The second post-Uruguay Round of regulatory failure put bilaterals on the Doha

negotiating agenda Special emphasis has been placed on “transparency” issues, i.e., notificationand reporting duties Surprisingly, in June 2006, agreement was reached on a “TransparencyMechanism for Regional Trade Agreements.” This agreement has been provisionally64

implemented, although it is less than clear that it will bring about all required notifications andstill retains the ex post facto nature of WTO regulatory review Whether this Mechanism willsolve the present regulatory gridlock is seriously problematic

The Transparency Mechanism envisions multiple electronic filings and detailed datasubmissions by parties to bilaterals, and substantial posting of such information on the WTO65

website Announcement of negotiations should be “early,” notification of agreements “as66 67

early as possible” but no later than after ratification and before application of preferential

treatment between the parties In addition, the separate 1994 WTO Agreement on Rules of68

Origin specifies transparency requirements for bilaterals, including a duty to notify the WTOSecretariat of preferential rules of origin.69

Review by the WTO of bilaterals should “normally” be concluded within a year of

notification The WTO Secretariat will prepare a “factual presentation” primarily based on70

information submitted by the parties, but if necessary from other sources The Secretariat is71

admonished to “refrain from any value judgment, which appears true of the presentations

reported to date on the WTO website.” Most importantly, its factual presentation may not be72

used “as a basis for dispute settlement procedures” or “to create new rights and obligations formembers.” A general notice and comment period is prescribed Subsequently, the CRTA and73 74

CTD is supposed, as a rule, to devote a single formal meeting to consider each notified

agreement 75

A Reform Proposal

Bilaterals presently fall under two WTO jurisdictions: (1) The disparate regulatory

regimes of GATT Article 24, GATS Article 5 and the Enabling Clause, and (2) the WTO DisputeSettlement Understanding (DSU) The regulatory regimes have a 60-year history of failure Atmost they have contributed to information sharing and peer pressure regarding bilaterals, andeven that accomplishment has a spotty record

The DSU, this author believes, has broadly achieved a reasonable degree of success inobtaining member state compliance with WTO agreements, and has already developed some76

expertise in proceedings where bilaterals have been at issue Specifically, the Appellate Body77

has suggested that parties raising Article 24 defenses in DSU proceedings must be prepared tohave their bilateral agreements examined for compliance with WTO rules Moreover, the78

professionalism of WTO panels and its Appellate Body has been commendable, the DSU doesnot suffer from gridlock, and it relies on member state interests (not WTO consensus) for its

driving force Indeed, because GATT Article 24 and GATS Article 5 often trigger compensationrights for nonparticipants, member states are already quite interested in bilaterals and have reason

to challenge their WTO conformity

Prof Picker has thoughtfully proposed altering the relationship between the WTO andbilaterals by having the CRTA identify bilateral derogations to WTO rules that cause or willlikely cause “institutional harm,” by which he principally means the creation of conflicts with ordiversion of resources from the WTO Positing a rule of WTO supremacy, he would use DSU79

procedures to finally and conclusively resolve disputes regarding bilaterals, excepting only

selected waivers granted by the WTO membership on a two-thirds vote While I welcome thecall for greater use of the DSU, I am much less sanguine about continuing attempts to regulatebilaterals

Unlike Prof Picker, I therefore propose abandoning the WTO regulatory regimes

applicable to bilaterals, save only notification and data reporting duties established in the 2006Transparency Mechanism Any failure to fulfill these duties should be sanctioned by creating arebuttable presumption of WTO nonconformity The opportunity for interested member states topursue DSU proceedings regarding bilaterals should be as clearly delineated for GATS Article 5and the Enabling Clause as it is in the 1994 Understanding on Interpretation of GATT Article 24

In all cases, the right to pursue such remedies should not depend upon proper notification of the

bilateral agreement in question A revised version of the Transparency Mechanism reflectingthese approaches is appended

This proposal, I submit, recognizes the realities of GATT/WTO practice It avoids thediversionary impact of the regulatory regimes, accepts their failure, and by concentrating