return to Articles & News  

Assessing the WTO General Agreement on Trade in Services
And Improving the GATS Architecture

Geza Feketekuty[1]

           One of the major achievements of the Uruguay Round of Multilateral Trade Negotiations was the negotiation of the General Agreement on Trade in Services (GATS), the first comprehensive framework for the global liberalization of trade in services.  In addition to the legal framework, countries negotiated national commitments on access to their markets in individual sectors and sub-sectors, and on the treatment of foreign services and service suppliers within their national markets. The major thrust of this first round of negotiations on market access and national treatment was to commit governments to preserve the degree of access provided by current regulations.  The hard part of reducing the barriers built into existing government regulations was left largely to future rounds of negotiations.

            The negotiators were very much aware of the need for further negotiations, and included a provision for the resumption of negotiations in the year 2000. [2] The agreement calls for successive rounds of negotiations for the liberalization of trade barriers, of which the round beginning in the year 2000 will be the first.  It also calls for the resumption of negotiations on GATS rules, particularly with respect to areas such as subsidies, safeguards, government procurement and regulations “relating to qualification requirements and procedures, technical standards and licensing requirements” [3] It tasks the Council for Trade in Services with the development of negotiating guidelines and procedures for each round of negotiations.[4]  This paper is designed to provide an assessment of the existing provisions and their application, as input into the development of these negotiating guidelines and procedures.

 In framing the negotiating guidelines and procedures, the Council should go beyond the built-in agenda and expand the terms of reference for the negotiations in several respects.  First, the negotiators should be directed to develop enhanced procedures for the negotiation of national commitments, in order to ensure a substantial liberalization of barriers to trade in services.  Second, the negotiators should be asked to fine-tune the legal framework and deepen the provisions of the GATS, in order to enhance its clarity and effectiveness.  Third, the WTO Council should establish an institutional structure for the negotiation and administration of horizontal agreements that cover trade in both goods and services, since such agreements would not fall either under the Council for Trade in Goods or the Council for Trade in Services.  In the future, it is likely to make more sense to negotiate agreements on issues of common concern to both trade in goods and trade in services on a unified basis, and to administer such agreements within a unified institutional structure. 

An Initial Assessment of the Agreement

An assessment of the overall effectiveness of an international trade agreement such as the GATS should be based on at least four criteria:

(1)   the effectiveness of the negotiating procedures in achieving a reduction of trade barriers;

(2)   the effectiveness of the disciplines built into the agreement in restraining the introduction of new trade barriers;

(3)   the effectiveness of the legal framework and the dispute settlement procedures in facilitating the settlement of disputes arising over the interpretation of the commitments; and

(4)   the effectiveness of the agreement in establishing a transparent and stable environment for enterprises engaged in international trade in services.

Five years have passed since the adoption of the GATS.  This is not a long enough period of time to form any conclusive judgments. The first round of negotiations on market access and national treatment commitments was more designed to establish an initial set of national commitments that would preserve the existing degree of market access, than to achieve a substantial reduction in barriers to trade.  The continuing rapid growth in demand for services has meant that the provisions of the agreement have not been tested for their ability to withstand political and economic pressures for protection.  Empirical evidence from other trade agreements such as the GATT and the EEC suggests that such agreements often have to mature for a decade or two before either enterprises or governments actively use the agreement in managing their affairs.

Notwithstanding the shortage of data in making an assessment of the operational effectiveness of the GATS, it is possible to examine the available data for a preliminary assessment.  Since the adoption of the agreement, there have been two multilateral negotiations under the GATS umbrella on market access and national treatment commitments – one on telecommunication services and the other on financial services. 

The negotiations on basic telecommunication services were a considerable success.  The resulting General Agreement on Basic Telecommunication Services (GBT) will, over time, open the provision of basic telecommunication services to international competition.  It also resulted in the adoption of a Reference Paper that sets out a common framework for the regulation of competition in this area.  Many governments in their submissions to the WTO on the preparation of the next round of multilateral trade negotiations have asked for a consideration of the wider application of the pro-competitive principles contained in the Reference Paper on Telecommunications.[5]

The negotiations on financial services were a more modest success.  While they resulted in few reductions in existing barriers to trade in financial services and in very few commitments of any kind with respect to cross-border sales of financial services, they yielded useful commitments by many governments that they would not tighten existing barriers imposed on foreign financial institutions

On the face of it, the experiences in these sectoral negotiations fly in the face of conventional wisdom in the trade policy community that sectoral negotiations are unlikely to achieve much liberalization.  While it is premature to draw any conclusions about the comparative efficacy of horizontal negotiations that cut across sectors as against sectoral negotiations, there are several observations that can be made about the organization of these negotiations.

The negotiations  on  telecommunications

• were supported by both business users and new telecommunications providers             in the development of national negotiating positions; 

• benefited from substantial empirical evidence that deregulation and competition             yielded economic benefits to both producers and consumers in the countries             that had adopted such policies. 

•  were reinforced and given a solid legal structure by the so-called Reference             Paper, which sets out a framework for competition in this sector.

In financial services, little progress was achieved until the major proponents of an agreement launched an international effort to persuade governments in developing countries of the domestic financial and economic benefits that would flow from an agreement. At the same time, one can only speculate if the development of a side agreement on prudential supervision would have established a basis for more progress.

It needs to be further noted, however, that in both the negotiations, the desire of many countries to persuade the United States to remove its reservation on MFN treatment played a major role. The United States had made it clear that it would not remove its reservation on MFN treatment in these two sectors, i.e., its right to negotiate preferential agreements with other countries, until enough countries had made substantial commitments in their national schedules in these sectors.

Beyond the concrete results of the negotiations carried out to date, it is worth observing that the negotiation of the GATS Agreement itself had a major impact in many countries in triggering a national debate on the somewhat encrusted national regulatory systems in services.  This debate in many cases led to the substantial liberalization of both internal and external trade through autonomous domestic regulatory reforms.  While national experiences with respect to the autonomous adoption of domestic reforms vary considerably, as witnessed by the different experiences of various East Asian economies, there is nevertheless an important potential lesson for future liberalization efforts.  An essential requirement for successful liberalization may well be the development of national consensus among consumers, providers, regulators, economic officials and trade officials that regulatory reform and trade liberalization in a particular sector will spur economic growth and are therefore in the country’s own interest, whatever happens in the broader global context.[6]

Turning from the impact of the GATS on the liberalization of trade in services to the perceptions of potential stakeholders, anecdotal evidence suggests that the Agreement and the associated appendixes and national schedules are not very user friendly and are somewhat opaque to the private sector.  Comments by business managers and trade lawyers who have ventured to examine the national schedules of commitments have indicated that they are more confused than illuminated by the schedules and the provisions.  Public perceptions have not been helped by what has been a somewhat confused and often ill-informed debate among the experts.  The confusion is to some extent due to the complexity of disentangling trade protection, overly restrictive approaches to national regulation, and the pursuit of legitimate public policy goals.  Nevertheless, it can also be traced to a lack of clarity in the conceptual design of the agreement and the attendant schedules, a design flaw that could be addressed in the coming negotiations.

Assessing the GATS Negotiating Procedures

            Negotiations aimed at the liberalization of barriers to trade to services under the GATS umbrella have followed two approaches two date:

(a)   a bilateral exchange of requests and offers, followed by the bilateral negotiation of market access and national treatment commitments on an item-by-item basis, across all services sectors;

(b)   plurilateral or multilateral negotiations focused on individual sectors such as telecommunications, which covered not only market access and national treatment issues but also related regulatory issues.

            As noted above, the relative success rate of these alternative approaches to the progressive liberalization of barriers to trade in services is difficult to evaluate.  The initial round of across-the-board negotiations was more focused on covering as many sectors as possible with an initial set of commitments that reflected the current level of market openness rather than in achieving a substantial liberalization of barriers. The two sectoral negotiations that were successfully concluded benefited from special factors such as the pressure of MFN exemptions and, in the case of telecommunications, the economic pressure generated by changes in technology. 

The results require careful analysis because they go against the historical experience and the firm belief of the trade policy community that crosscutting negotiations are more successful in achieving liberalization than negotiations focused on individual sectors.  Crosscutting negotiations are usually considered more successful because interests in other sectors who stand to gain from liberalization offset the opposition of vested interests in any one sector.  The underlying assumption is that in any given sector some countries are competitive and other countries are uncompetitive, and current providers in uncompetitive protected markets stand to lose as a result of negotiations. They will therefore pressure the governments involved not to reduce the protective barriers.  Broad negotiations allow sectoral trade-offs between those who lose and those who gain in every country.  The resistance of providers in uncompetitive services is compounded in many services sectors by the potential resistance of sectoral regulators who have a vested interest in preserving their existing regulations and protecting their bureaucratic turf.

The accumulated wisdom of trade negotiators has to be tempered, however, by recognition that liberalization in many services sectors cannot be achieved without substantial domestic regulatory reforms, and that such reforms cannot be achieved on a piecemeal basis and over the determined opposition of the regulators involved.  As the experience with the less successful efforts in areas such as maritime and air transport show, without the cooperation of regulatory officials and the opportunity to make a case why regulatory reform in a particular sector is to everyone’s advantage, little progress can be made.  The more recent experience of negotiations in the accounting sector reinforces this point. While negotiations were successfully concluded, they fell short of desired results because of the determined oppositions of some regulators.[7]

Another possible lesson from past trade negotiations is that negotiations organized around negotiating targets such as tariff-cutting formulas are likely to be more successful than negotiations based on strictly item-by-item negotiations without some overall target or goal.  Targets provide benchmarks against which results based on highly detailed bilateral negotiations can be measured.  They also help improve the efficiency of negotiation, improve public comprehension, and enhance the consistency, clarity and user friendliness of the final schedules.[8]

Both of the successful sectoral negotiations benefited from overall targets that provided a benchmark against which progress in the negotiations could be measured. 

The remainder of the paper will address some of these weaknesses in the GATS and recommend practical steps for strengthening the agreement.

The Conceptual Structure of the GATS Agreement

The key difficulty in designing the GATS Agreement was that barriers to trade in services are generally embedded in domestic regulations.[9]  Unlike barriers to trade in goods, they do not take the form of transparent barriers imposed at the border against foreign services.  Cross-border flows of services are largely invisible.  Where governments have tried to protect their local service suppliers from foreign competition, they have embedded the protective measures in domestic regulations focused on the local consumption of services, or the local provision of services.  For example, regulations may stipulate that only car insurance provided by a local firm satisfies the compulsory insurance requirement for car registration, or that only locally owned and established firms may sell car insurance to consumers.

 Another challenge in designing a regime for trade in services was that domestic regulations frequently limit trade even if they do not explicitly discriminate against foreign providers. This was seen as a crucial point because the regulatory involvement of governments in the provision of services has been much more intensive in services historically than in manufacturing.  Regulations often limit : (a) the number of firms; (b) the number of employees; (c) the number of distribution outlets; (d) the services that can be sold; (e) prices; (f) marketing practices; and (g) distribution channels. These types of regulations have the effect of protecting existing firms from competition by new entrants, whether domestic or foreign.  The stated rationale for such intervention often is that consumers need to be protected from the shady practices that may result from excessive competition, or that consumers need to be assured of a stable supply of a reliable product at reasonable prices. In other cases the rationale is that the market can physically or economically support only a limited number of providers, or that all economic activities related to the provision of services that contain a monopoly component have to be regulated to protect the consumer interest.

A third challenge in designing a comprehensive regime for trade in services was that trade, investment, labor mobility and foreign consumption are much more interwoven in services than in goods.  While the major argument for the negotiation of trade rules in services was that services were becoming tradable as a result of innovations in information technology, it did not seem to make much sense to limit a trade regime to cross-border trade. After all, the bulk of services sold by foreign service providers to consumers are still being sold through locally established firms.  Interestingly enough, when confronted by a choice between cross-border sales and local establishment of foreign providers, many regulators decided they would prefer the latter.

The negotiators of the GATS thus faced a three-fold challenge in designing a regime for trade in services.  Such a regime had to remove discrimination in domestic regulations against services produced by foreign providers, it had to deal with non-discriminatory regulations that had the effect of restricting trade, and it had to deal with four different ways in which services could be traded.  While the negotiators realized that the structure of the GATT rules would have to be changed to meet these challenges, they were nevertheless a product of the GATT system.  They innovated where they had to, but otherwise they adapted legal concepts and principles from the GATT system.

            Some of the major design flaws with the GATS regime are the result of grafting GATT concepts and terminology to provisions that had been fundamentally altered.  In order to understand the GATS framework it is therefore necessary to review the underlying structure of the GATT.

The Basic Structure of the GATT Rules for Trade in Goods

The GATT rules for trade in goods are woven around a few key notions. These key notions are that : barriers should be imposed at the border; they should take the form of tariffs; the tariffs should be subject to negotiation; the tariffs resulting from such negotiations should be bound in national schedules; tariffs should be applied equally to goods imported from all member countries; and domestic regulations should not discriminate against foreign goods once they have crossed the border and paid the tariff.  The GATT rules permit various exceptions from these basic rules, and they spell out how these rules are to be applied with respect to various types of government measures, but the core rules are those that implement the central notions enumerated above.[10]

The GATT provided that the negotiation of tariffs (and quotas where permitted) should be carried out on the basis of bilateral negotiations through the exchange of requests and offers.  As a result of the experience gained from successive rounds of negotiations, it was found that that more liberalization could be achieved if such bilateral negotiations were based on multilaterally agreed targets or formulas for the negotiations.

The closest analog in goods to the central challenge in services of curbing the trade restrictive impact of domestic regulations was the need to curb the trade-distorting impact of domestic standards. The GATT rules on standards give governments the right to pursue domestic social objectives through the establishment of compulsory standards for the goods sold in domestic markets, but they require that the resulting standards achieve the desired social objective in the least trade-distorting manner possible 

The Basic Structure of the GATS Rules for Trade in Services

            The GATS addresses the trade restrictive impact of regulations at three levels – through general provisions on domestic regulation, through sectoral annexes on the regulation of particular sectors, and through the negotiation of national commitments on regulations that affect trade.  The key notions underlying the GATS rules are that : regulations which restrict trade should be subject to negotiation;[11]  the results of such negotiations should be bound in national schedules;[12] the resulting commitments should apply equally to services and service providers from all other members countries;[13] and regulations should not restrict foreign services or service providers,[14] or discriminate against them,[15] in a manner that is inconsistent with the bound commitments.  Moreover, the GATS rules define trade not only as the cross-border delivery of services, but also as the foreign consumption of services, the delivery of services by foreign-owned but locally established enterprises, and the temporary entry of natural persons for the explicit purpose of providing a service locally.[16]  It should be noted that the binding of market access in services is less precise because barriers in services take the form of discriminatory regulatory requirements rather than tariffs, as is the case in goods.

            This conceptual framework appears clear enough.  The major problems of the GATS are not the result of deficiencies in this basic framework. They are rather the result of the manner in which this underlying framework was implemented through the drafting of the legal provisions in the GATS and the negotiation of the associated national schedules.  After correctly analyzing the unique requirements of an effective GATS regime for services, the negotiators frequently fell back to GATT terminology and legal drafting, even where that did not provide the best fit. The ambitious nature of the undertaking and the continuing rapid pace of technological change in many services have also led to inevitable shortcomings that will need to be addressed in future negotiations.

A more detailed assessment of the key shortcomings of the current architecture of the GATS, and some recommendations for possible approaches to overcoming the weaknesses are treated below.

The Hierarchy of Commitments in the GATS

            The GATS provides for the negotiation of commitments at three levels – general provisions which are part of the GATS text, annexes which set out rules for particular sectors (e.g., telecommunications) or policy instruments (e.g., visas for temporary service providers), and national schedules of commitments on market access and national treatment.  This architecture reflects the practical realities associated with the negotiation of increased disciplines. National schedules reflect the reality that the liberalization of trade barriers cannot be achieved over night and that not every country can move at the same pace. Sectoral agreements reflect the fact that liberalization of trade in some sectors cannot be divorced from the establishment of compatible regulatory regimes, and in some cases the modification of the current international regulatory regime. Sectoral agreements also may provide the best channel for achieving significant liberalization where the conditions are right, but the ability of any one country to liberalize its own regulations is facilitated either in economic or political terms by simultaneous reforms in other countries. The drafting of general GATS provisions for all services is, of course, the most efficient approach for advancing the liberalization of trade in services where such principles can be applied to all services.

            The three-tier structure is analytically sound, though it does not answer the question whether more progress can be achieved with respect to the  particular objectives by concentrating negotiating resources at one level or another.  The negotiation of the provisional agreement on accounting, for example, has raised the question whether more progress could be made by focusing on another professional service, by developing a cross-cutting agreement for all professional services, or by improving the provisions of Article VI dealing with domestic regulation overall.  No a priori answers can be provided on this question, and the real answer probably is that it depends on a political judgment of what can be achieved in different contexts, given current political and economic realities.

            The three-tiered structure does raise questions about the relationship among the commitments at the three levels.  For example, to what extent can reservations in national schedules override general provisions of the GATS, or of sectoral annexes?  Some provisions of the GATS or of Sectoral Annexes obviously refer to the right of members to enter reservations in the national schedules, but what about the other provisions?  If member countries agree to adopt new definitions in the GATS agreement or in an Annex, would that alter the definitions employed in the national schedules where the country did not specify that it was using a national terminology? It may well be desirable to spell out the hierarchy of commitments that would apply under different circumstances.

Market Access and National Treatment

            The GATS recognizes that trade in services can be hampered by either discriminatory regulatory requirements imposed only on foreign services or by restrictive regulations that are imposed on both domestic and foreign services. It confuses the issue, however, by transposing GATT concepts and terminology in ways that neither correspond clearly to their application in the GATT Agreement, nor clarify the intent of the commitments in the services context.

All quantitative limits on services or service providers are dealt with under the rubric of market access, regardless of whether such limits are being imposed on foreign services on a discriminatory basis or on both domestic and foreign services on a non-discriminatory basis.  National treatment is defined in the traditional GATT way as the non-discriminatory application of domestic regulations to foreign services or service providers.  This means that quantitative limits placed on foreign services or service providers fall under both the market access and national treatment provisions.[17] 

The drafters of the agreement realized that this overlap could create a potential source of confusion in the scheduling of commitments, by raising a question whether national commitments regarding the discriminatory application of quantitative regulatory controls should be entered as market access or as national treatment commitments in the schedules.  To avoid duplication and confusion, Article XX:2 of the GATS on the scheduling of commitments indicates that “measures inconsistent with both Articles XVI [on market access] and XVII [on national treatment] shall be inscribed in the column relating to Article XVI.” While this fix removes a potential confusion with respect to the scheduling of discriminatory quantitative restrictions, it introduces a discrepancy between the text of the provisions on market access and national treatment on the one hand, and the content of the market access and national treatment columns in the national schedules on the other hand. This difference between the Articles and the schedules has introduced a conceptual confusion on the relationship between market access and national treatment commitments. Some countries have wrongly assumed that national treatment commitments become operative only after they have made a commitment on market access.

The intermingling of commitments on discriminatory and nondiscriminatory barriers has the further effect of intermingling two laudatory, but separate goals – trade liberalization and domestic regulatory reform. Removing discriminatory regulation, whether in quantitative or qualitative form, is all about trade liberalization. Removing nondiscriminatory restraints on services is frequently an exercise in domestic regulatory reform. The use of nondiscriminatory quantitative restraints more often than not reflects a country’s approach to the regulation of activity in a services sector, and telling a country to eliminate such restraints is tantamount to saying it must reform its approach to the regulation of that sector.

The intermingling of commitments on discriminatory and nondiscriminatory quantitative restraints raises the hurdle for countries that are willing to tackle the liberalization of trade barriers or the reform of domestic regulation in a services sector, but not both simultaneously.  It is tantamount to saying to countries that they have to liberalize trade and reform their domestic regulations on services simultaneously.  While nothing in the agreement requires them to do both simultaneously, the organization of the schedule does not easily allow countries to highlight progress on either trade liberalization or domestic reform.

            It is easy to see how a GATT mindset about traditional notions of market access and national treatment could lead to the current structure and terminology in the GATS.  However, the net effect has been to create a great deal of confusion not only by businesses, but also by trade officials and other trade experts who are not thoroughly familiar with the details of the GATS agreement.  They naturally equate market access in services with the tariff bindings in the GATT, and national treatment in services with its counterpart in the GATT. They fail to grasp that market access in services potentially is a more far-reaching commitment, since it covers both discriminatory and non-discriminatory restraints on the production and sale of services; and that the deviations from national treatment in services are not as damning as they would be in a GATT context, since national treatment in many services equates to the complete absence of barriers to trade, and that it grants nondiscriminatory treatment not only to services but also to service providers.

            To cure the problems with respect to market access and national treatment, the rule in Article XX should be reversed, and measures that impose quantitative limits on a discriminatory basis should be listed under the national treatment, rather than under the market access column of the schedule. This would allow countries to highlight progress on trade liberalization through the scheduling of commitments under the national treatment column, and to highlight progress on domestic regulatory reform though the scheduling of commitments under the market access column. Ideally, the market access column would be relabeled as nondiscriminatory quantitative restrictions on services, and the sequencing of the two columns would be reversed, with the national treatment column preceding the column on nondiscriminatory quantitative restraints. In an ideal world, the two relevant provisions of the GATS would also be retitled and reversed. Making all these changes may prove too difficult, but at a minimum the scheduling convention in Article XX needs to be reversed.

Top Down Vs. Bottom Up or Negative List Vs. Positive List Approaches

            One of the hotly debated issues in the course of the negotiations was whether the national schedules should list only deviations from an ideal state of national treatment and market access, or whether they should list only positive commitments to provide national treatment and market access for a particular service and mode of supply.[18] The first approach was referred to as the top down or negative list approach, while the second was referred to as the bottom up or positive list approach.  Those who favored a maximum degree of liberalization generally argued for the top down, negative list approach, while those who were more reluctant to see much liberalization argued for the bottom up, positive list approach. 

A top down approach is more liberalizing because new services or services that no one really cares about are automatically provided national treatment and market access under such an approach. Moreover, a top down approach provides more information for potential exporters, since importing countries must indicate in their schedule any deviation from national treatment or the use of any quantitative restraints with respect to any tradable service. Under a bottom up or positive list approach new services, services no one cares about, or services on which the importing country does not wish to make a commitment are not covered in a country’s national schedule. Foreign exporters therefore have no information on the treatment they might expect on such products.

            Aside from differences over the desired degree of liberalization in services, the negotiators ran into a practical problem.  Some countries were simply not equipped to identify all national and local laws that might conflict with the national treatment and market access provisions.  In some cases they did not have the personnel to do it.  In other cases their regulatory systems had not evolved to a state where they could identify potential inconsistencies, and they did not want to constrain their future ability to develop their own regulatory approach, or to put themselves in the uncomfortable position of having to renegotiate their commitments. 

In the final result, the negotiators chose to adopt a hybrid approach. In sectors where countries are prepared to make commitments, they must list their reservations to market access and national treatment.  However, they assume no commitments with respect to sectors not inscribed in their schedule.  One might call this a sector-by-sector, top down or negative list approach.

            In sectors inscribed in their schedules, countries can list limitations on market access and national treatment by spelling out in some detail the products on which they are not making a market access or national treatment commitment, and/or the precise manner in which market access or national treatment commitments are circumscribed with respect to such products.  Alternatively, a country may list the specific provisions in its laws for which it is taking a reservation with respect to either market access or national treatment.  In some cases a country may state that the commitment applies only to certain products or subsectors within a sector, and that it is making no commitments in other products or sub sectors.

            What an examination of the schedules reveals is that the issue is not primarily whether a top down or bottom up approach should be taken, but rather the degree of precision with which a country describes its limitations on market access and national treatment.  There is a huge difference whether a country declares a whole subsector as unbound, takes a reservation for a broad law without listing the specific provisions that violate national treatment or impose quantitative restraints, or spells out in some detail the specific ways in which a particular law violates national treatment or imposes quantitative restraints.

            Precision in the drafting of the reservations on national treatment and market access is desirable both because it provides greater assurance against arbitrary protectionist measures and because it gives potential exporters more information on the provisions they must satisfy.  In other words, precision is desirable for the achievement of transparency, maximum liberalization, and greater certainty. The scheduling guidelines issued by the secretariat as guidance for the development of schedules specifically indicates that countries should not merely list a law or measure that contains inconsistent provisions, but should spell out the specific provisions that are inconsistent.[19] Many of the existing schedules do not meet this standard, and one of the objectives of the coming negotiations should be to improve the quality of the schedules in this regard,

            It is not worth revisiting the debate over top down vs. bottom approaches.  Not only would the outcome likely be the same, but it would also miss the more important issue regarding the care and precision with which a country spells out the limitations it is putting on national treatment and market access, and the precise provisions of its laws which give rise to these limitations.  This cannot be accomplished through broad conceptual arguments.  It should be addressed through a detailed and labor-intensive peer review of national schedules, and the negotiation of a commitment by countries to bind a substantial portion of their schedules.  Ironically, the schedules of new entrants have received a great deal more scrutiny than the schedules of existing members, and therefore come closer to providing the kind of information that is desirable.

The Use of Negotiating Targets, and the Negotiation of Rules

            As we noted above, one of the challenges for negotiators is to find an efficient approach to the negotiation of commitments on market access and national treatment are either negotiated. Each of the approaches that have been used – bilateral negotiations based on an exchange of requests and offers a sectoral negotiation based on sectoral targets – has its advantages and disadvantages. The strength of the bilateral negotiating approach is that it provides a broad basis for establishing a negotiating outcome that meets the interests of Members, but its weakness is that it is very labor intensive and it is easy to lose sight of the forest as the negotiators argue over each tree, The strength of the sectoral approach is that it sets a common target against which progress can be measured and it allows countries to address common regulatory issues that might block progress in liberalizing restrictive regulatory measures. Its weakness is that it limits the possibility of satisfying the varied economic interests of members.  The telecommunications sector may well be somewhat unique in that most countries found it in their economic interest to shift to more open competition domestically as well as internationally.

The question is whether the strength of the sectoral approach in setting a target could be applied on a more horizontal, cross-sectoral basis.  In tariff negotiations under the GATT umbrella, tariff-cutting formulas, which were applied across the board, served this purpose. Since barriers in services do not take the form of tariffs, and instead take the form of a wide variety of regulatory measures, the use of targets is likely to involve the establishment of a more varied set of objective criteria against which progress in the negotiations can be measured. Some of these criteria could take the form of quantitative targets, while others may take the form of qualitative targets for the treatment of certain types of common regulatory issues.

One type of quantitative target could be based on the number of sectors in which countries have made binding commitments, and the degree to which such commitments cover both market access and national treatment and the four modes for the delivery of services. Negotiators might agree that developed countries cover say 95% of all sectors and sub sectors, and that developing countries cover say 40% to 60% of their sectors and sub sectors, based on their level of development. Similar percentages might be developed  for the four modes.[20]

Another approach to the establishment of quantitative targets might be based on the types of quantitative restrictions included in Article XVI as the kind of measures that countries may not impose without listing such restrictions as exceptions in their national schedules in sectors where they have made any commitments. The types of restraints covered includes :

    (a)    limitations on the number of service suppliers whether in the form of numerical quotas, monopolies, exclusive service suppliers or the requirements of an economic needs test;

    (b)    limitations on the total value of service transactions or assets in the form of numerical quotas or the requirement of an economic needs test;

    (c)    limitations on the total number of service operations or on the total quantity of service output expressed in terms of designated numerical units in the form of quotas or the requirement of an economic needs test;

    (d)    limitations on the total number of natural persons that may be employed in a particular service sector or that a service supplier may employ and who are necessary for, and directly related to, the supply of a specific service in the form of numerical quotas or the requirement of an economic needs test;

    (e)    measures which restrict or require specific types of legal entity or joint venture through which a service supplier may supply a service; and

     (f)    limitations on the participation of foreign capital in terms of maximum percentage limit on foreign share holding or the total value of individual or aggregate foreign investment. 

Some targets or formulas may take the form of a suggested approach to a common regulatory issue or problem. Such qualitative targets would serve as negotiating guidelines, against which national measures could be compared. Negotiating guidelines do not become a binding part of the agreement.[21] If enough countries found a qualitative negotiating guideline useful, however, it could be given the status of a rule by being incorporated in a sectoral annex to the GATS agreement.  The Reference Paper, which forms a part of the General Agreement on Basic Telecommunications (GBT), is an example of such a rule. The proposed  rule could be open to any Member for signature, and apply only to Members that signed, or it could be made a universal obligation, which would bind all Members who did not specifically enter a reservation in their national schedule.       

Improving the Rules on Domestic Regulation

Given the importance of the regulatory dimension in the liberalization process, Article VI on Domestic Regulation plays a key role in the liberalization of trade in services. Article VI is the essential third leg of three-legged stool, along with Article XVI on Market Access and Article XVII on National Treatment.  While Article XVI disciplines the use of quantitative restrictions, and Article XVII disciplines discriminatory treatment of foreign services and service providers, Article VI disciplines more hidden forms of protection buried in domestic regulations and their administration.[22] Article VI:1 requires that “in sectors where specific commitments are undertaken, each Member shall ensure that all measures of general application are administered in a reasonable, objective and impartial manner.

Article VI calls for the negotiation of more detailed disciplines that would help ensure that all measures are reasonable, objective and impartial. Article VI even spells out the objectives of such disciplines, by indicating that such disciplines shall aim to ensure that regulatory requirements are, inter alia :

a)     based on objective and transparent criteria, such as competence and the ability to supply a service;

b)     not more burdensome than necessary to ensure the quality of the service;

c)      in the case of licensing procedures, not in themselves a restriction on the supply of the service.

Pending the negotiation of the disciplines called for in VI:4, VI:5 applies the objectives listed in a) to c) above to any new measures that may nullify or impair commitments made in the schedules.[23]

            The negotiation of the disciplines called for in Article VI above should clearly receive a high priority in the new round of negotiations in services, and the negotiators would do well to start by converting the objectives listed under a) to c) above into disciplines that would apply to all domestic regulatory measures with an impact on trade in services. Beyond these disciplines, they might consider the addition of a number of related principles.[24]  Some key principles that might be added to Article VI are :

·        transparency of regulatory objectives. The social objective served by a particular law or regulation should be transparent, i.e., that it should be clearly stated at the time the regulation is adopted.   A clear statement of the desired social objective helps to remove possible confusion over the purpose of the regulation.  It also makes it a great deal easier to judge whether a regulation is the least burdensome necessary to accomplish the desired social objective. This principle is contained in the WTO agreement on accountancy services, which has been adopted on an ad ref basis.

·        appropriate use of market mechanisms. Governments should use market mechanisms to promote desired social objectives, whenever that is feasible.  The use of across the board economic incentives and disincentives is an economically efficient method of accomplishing desired social goals because it allows market forces to determine the most economically efficient manner of accomplishing the desired social goal.  A corollary of this principle is that scarce resources should be auctioned whenever possible rather than allocated to incumbent firms on the basis of historic shares. An auction process is more likely to enable the most efficient firms to gain access to such resources, and is more likely to ensure that the opportunity cost of using the scarce resource is properly reflected in the cost of supplying the service and the prices charged to consumers.  Where regulations give existing producers or sellers preferential treatment in the allocation of scarce resources, they not only create domestic economic inefficiencies by discriminating against potentially more economically efficient new suppliers; they also distort international trade and competition. Moreover, there is a significant risk that the economic inefficiencies and trade distortions are magnified by political/interest group pressures and corruption.

·        Minimizing the Scope of Regulations. Governments should seek to minimize the regulatory burden by limiting the scope of any regulation to what is necessary to accomplish the desired social objective.  This principle could complement and reinforce objective b) cited above, namely that government regulatory measures are “not more burdensome than necessary to ensure the quality of the service”.  This corollary principle would require governments to limit the scope of regulations to what was necessary for the achievement of the regulatory objective.  Minimizing the scope of regulations to the minimum necessary for the achievement of the desired social objective helps to minimize the economic cost of such regulations. The application of this principle is particularly relevant to the regulation of infrastructure services such as water, gas, electricity, telecommunications and rail transportation.  The tendency in the past has been for governments to regulate all aspects of economic activity in these sectors because the network for distributing these services often constituted a natural monopoly.  In more recent years many governments have recognized that they can more efficiently accomplish their objective of protecting consumers by separating the construction and operation of the distribution network from provision of services over that network.  By separating the production of these infrastructure services from their distribution through the monopoly network, the government can regulate access to and use of the network monopoly, while leaving the supply of the services involved open to market competition. Efforts to minimize the scope of regulations to the minimum necessary to achieve the desired social objective help to minimize the economic cost of such regulations and the potential distortion of international trade and competition.

·        use of international regulatory standards.  Governments should use international standards where such standards would satisfy the desired social objectives.  This principle is already included in Article VII:5, which deals with the negotiation of agreements recognizing the authorization, licensing or certification of services suppliers by other governments. It requires that “wherever appropriate, recognition should be based on multilaterally agreed criteria.”  This principle is also a core principle embedded in the GATT Agreement on Technical Barriers to Trade, otherwise referred to as the Standards Code.

Domestic Regulation and Electronic Commerce

            As previously described, trade in services is defined as encompassing four modes of trade – foreign consumption, cross-border trade, the sale of services through a foreign owned local establishment, and the provision of a service by the resident of another country who has gained temporary entry. Each of these modes is treated as a separate commitment in the national schedules. Countries thus have to specify the market access and national treatment commitments that apply with respect to each mode of delivering services in a particular sector. (Where a country is prepared to commit itself with respect to a particular mode across all sectors or where a reservation applies across all sectors, the schedule may contain a horizontal entry that applies to all sectors.)

By addressing itself to the four modes, the services framework covers a much wider ground than the GATT.

The advent of electronic commerce has created the need for a more precise definition of the four modes.  Before the Internet and the burgeoning of electronic commerce, the definition of the four modes was fairly straightforward.  Foreign consumption of a service occurred when a traveler to another country purchased and consumed a service abroad. Cross-border trade occurred when the purchaser of a service was in one country and the seller in another country, and the service was transmitted from one country to the other country through a phone call or the mail.  The sale of a service through local establishment took place when a foreign-owned enterprise established a facility or a legal entity to produce or sell services in the local market.  Finally, the sale of services through the temporary entry of a natural person took place when a provider physically traveled to the importing country to produce the service.

            With the advent of the Internet and electronic commerce these distinctions have become blurred, and this has raised questions such as which set of market access and national treatment commitments apply under different circumstances? It also raises the question as to whose regulatory jurisdiction applies under those circumstances. Take the issue of foreign consumption.  If a consumer buys a service from a provider in another country over the Internet, under what circumstances should that transaction be treated as foreign consumption, or as cross-border trade, or as a purchase from a locally established enterprise? 

Some have argued that direct marketing to consumers in a particular country should be considered as local establishment, subjecting the production of the service to the local regulatory jurisdiction of the country of the consumer.  Alternatively, if the purchaser acquires a service from a foreign producer who does not engage in targeted marketing, the sale might be considered as cross-border trade, subjecting the sale of the service to the regulatory jurisdiction of the consuming country.  But what if the consumer in question intends to consume the service abroad? Should the buyer’s current physical location at the time of consumption?

            To make things even more complicated, how does the physical location of the Internet Service Provider used by the consumer affect all of the above distinctions. Let us assume that a consumer in country A acquires access to the Internet from an Internet Service Provider in country B, using a local address in country B. For all practical purposes the purchaser will appear to the service provider as a local resident. To what extent can the government of country A hold the seller responsible for meeting its regulatory requirements?

            One solution to the questions being raised is to follow strictly the treatment of analog transactions involving more traditional methods of acquiring services or goods abroad.  The problem is that it is not always clear where a service was produced and where the service provider is located.  This is particularly the case with respect to the sale of information or other form of intellectual property by multinational firms established in many different countries. 

            An alternative approach would be to let the market sort out regulatory issues by allowing sellers and buyers to choose the regulatory jurisdiction that would apply to a particular transaction. Such an approach might be complemented by the development of private codes, which might receive some kind of recognition by individual governments, and would be enforced by all governments that subscribed to a particular code.

The Application of the MFN Principle

            Another hotly debated issue in the negotiation of the GATS was the application of the MFN principle.  For many countries the MFN principle is the cornerstone of the multilateral system. In economic terms, it avoids a distortion of the relative prices applicable to goods and services imported from or exported to different foreign countries.  In political economic terms it puts all members of a multilateral system on a more equal footing. Small countries in particular feel that the MFN principle gives them more equal standing with large countries, and protects them from excessive bilateral pressure.  The MFN principle also helps prevent the adverse political fallout of competitive bilateral negotiations, which historically contributed to a breakdown of global trade and exacerbated political tensions among countries between the two world wars.

            On the other hand, it was argued that differences in regulatory philosophy could lead to incompatible market structures.  Also, differences in the quality of the regulations that are applied to service suppliers in different countries can affect the degree to which imported services meet the domestic regulatory objectives of the importing country in such areas as consumer protection and prudential supervision.  Moreover, it was pointed out that international trade in certain services, such as international air passenger service, was currently governed by bilateral agreements, and that the application of the MFN principle in these sectors would require a new international regime.

            The outcome of the debate was another compromise. The MFN principle was incorporated as a generic principle of the GATS. However, countries were permitted to enter reservations on the application of the MFN provision in particular sectors.  A number of countries availed themselves of this opportunity in heavily regulated sectors. It should be noted, however, that as a result of the subsequent negotiation of sectoral agreements on telecommunication and financial services, most countries withdrew their MFN reservations in these sectors.

Another part of the compromise was the introduction of two provisions that allow the negotiation of bilateral or plurilateral agreements on the mutual recognition of regulatory standards. Countries are allowed to negotiate such agreements without automatically extending the benefits to third countries, provided third countries are given the opportunity to negotiate accession to such agreements on equivalent terms. The first such provision is included in Article VII of the GATS, which deals with the recognition of experience or education obtained, requirements met, or licenses or certifications granted in connection with the authorization, licensing or certification of services suppliers. The second such provision is included in the Annex on Financial Services, which allows countries to recognize prudential measures of any other country for the purpose of determining whether a financial institution from that country meets requisite prudential standards

The GATS, like the GATT, provides for an MFN exception for free trade agreements, which in the GATS are called economic integration agreements.  It is worth noting that the conditions set out in Article V of the GATS for economic integration agreements are poorly drafted.  This is another case where one suspects that the drafters borrowed heavily from the equivalent GATT provision without thinking through how the provision would be applied in practice in the case of services. It is difficult to see how the conditions set out in the Article can be effectively monitored with the kind of data that are normally available on trade in services.  Moreover, there is a considerable lack of clarity with respect to the kind of barriers that an economic integration agreement should be expected to eliminate, and the degree of integration that should be expected in areas such as labor mobility and the recognition of professional qualifications 

As in the case of the previously cited debate over the top down vs. bottom up approach, there is little point in holding another conceptual debate over the application of the MFN principle.  What should be done with respect to national MFN exceptions is to establish the conditions under which countries are prepared to eliminate their MFN reservations in individual sectors, and to organize negotiations designed to satisfy those conditions.

With respect to mutual recognition agreements, the negotiations should focus on the development of model instruments for accomplishing the regulatory objectives of member governments. Few bilateral mutual recognition agreements have been negotiated under the terms of either GATS Article VII or the Annex on Financial Services to date, and there is no precedent for the accession of third parties to date. The negotiation of mutual recognition agreements in services between the United Stats and the European Union under the umbrella of the Transatlantic Dialogue represents one such effort. The Basle Accord among the BIS countries could be seen as a prototype agreement on prudential supervision. It is not clear how the accession of third parties under the terms of either Article VII or the Annex on Financial Services should be approached.  It would be desirable to encourage a great deal of experimentation in this area.  Countries should explore a range of alternative approaches, including the development of voluntary codes and regulatory models.

Finally, the drafting of Article V should be improved. Negotiators should clarify the conditions which economic integration agreements should be expected to meet.  Negotiators should also examine the kind of data that have been submitted by countries participating in such agreements as well as the kind of data that might be readily available, and, on the basis of such an analysis, develop more meaningful guidelines for the examination of such agreements.

A Plan for the Progressive Future Integration of the GATT and the GATS

            Services are different from goods and this has led to the development of the GATS as a separate instrument for liberalizing trade in services and disciplining government regulations affecting trade in services. The thesis advanced earlier in this paper was that the negotiators did not go far enough in recognizing the differences between goods and services, and this led to some confusion in the drafting of some of the GATS provisions.

            In light of that thesis, it may come as a surprise that the paper should end with a discussion of the progressive integration of the GATT and the GATS in future negotiations.  There are several reasons for believing that it makes sense to start thinking about the development of a common framework for disciplines that may apply in the future to both trade in goods and trade in services. This is not to suggest that the development of common disciplines should take up significant negotiating resources in the next round of negotiations. However, the fact that the current negotiations on transparency in government procurement apply to both goods and services is an indication that it is not too early to think about the accommodation of crosscutting agreements within the WTO structure.

There are a number of reasons for believing that in the future the current distinction between goods and services will be less important. As we saw in the earlier discussion of the conceptual foundation of the GATT framework, much of the GATT is built around the negotiation of tariffs and the maintenance of the benefits that accrue from tariff bindings. As tariffs on goods are eliminated, many of the provisions of the GATT, which uniquely apply to trade in goods, will no longer matter. Moreover, much of the substance of trade negotiations in goods will shift to regulatory issues and issues of industrial structure policy, issues that form the core substance of the GATS.

            While foreign investment and labor mobility were particularly important issues for trade in services; they are also increasingly relevant for trade in manufactured goods.  It is somewhat of an anomaly that a firm that produces services can secure WTO commitments on the right to invest and establish, or on the entry of foreign professional personnel, but it cannot do so if it produces goods. What about firms that produce both goods and services on an integrated basis? Can they really separate management staff that supervises or advises manufacturing units from those who supervise or advise units producing services?

            Moreover, many goods come bundled with a package of services.  For many purposes such services are treated as goods when they are bundled together with the goods, and they are treated as services when they become separated.  Where the unbundled goods and services are subject to different regulatory requirements, the difference in treatment can distort the most efficient way of packaging the components.

            For all the reasons above, it does not make much sense to negotiate commitments on establishment and labor mobility for service providers, but not goods providers. It is not clear at this stage how Ministers in Seattle will wish to address issues related to investment. At a minimum, Ministers should decide to pursue investment commitments on a sector-by-sector basis, much along the lines currently possible under the GATS.

Should WTO Ministers decide to launch comprehensive negotiations on investment, they would need to decide whether such negotiations should cover both goods and services within a single framework, or whether such negotiations should be divided between goods and services. While any common agreement on investment would have to be reconciled with existing commitments in the GATS under mode 3 dealing with establishment, it is difficult to imagine the negotiation of a comprehensive agreement on investment that did not apply to both goods and services.

            Moreover, as the WTO regime for trade in goods is extended to investment, many of the provisions of the GATS will become highly relevant for trade in goods.  For example, the GATT deals with standards but not with regulations that affect the production process.  Once investment in manufacturing is covered, the provisions of GATS Article VI on Domestic Regulation and the provisions of Article VII on Recognition of Certificates and Licenses could apply equally to manufacturing enterprises.  Conversely, many of the provisions of the GATT Standards Code could be applied to trade in services.        

Similar considerations apply with respect to trade and competition policy.  The GATS has pioneered the integration of competition provisions into trade disciplines in the telecommunications area.  Presumably if and when the WTO embarks on a broader consideration of competition issues, it would want to do so on an integrated basis, and ultimately integrate the existing provisions under the GATS into the new joint framework.

            The globalization of production, which results in the unbundling and distribution of different steps in the production process to different countries gives manufacturing many of the same characteristics as the production of services.  Many of the steps in the unbundled manufacturing process involves activities that may have been classified as manufacturing activities in the past, but as a result of unbundling fall into the category of business services.  Moreover, it can be argued that partial processing of manufactured goods is more effectively treated as the production of a value-added service rather than as the manufacturing of a good.  The traditional concepts surrounding trade in goods center on he proposition that goods have an identifiable country of origin.  While that is increasingly a questionable proposition, the whole existing GATT structure is built on that proposition.  It might be better to view partially manufactured goods that are shuttled from country to country as involving trade in services, rather than as a good that originated in the last port of call.

            All of the above argues for a gradual process for integrating GATT and GATS rules on a step-by-step basis.  One could imagine the development, over time, of new instruments that would establish common provisions for trade in both goods and services.  Specific provisions of the GATT and the GATS could be replaced by the new common provisions, as negotiators succeed in reconciling differences between GATT and GATS provisions that do not arise from fundamental differences between goods and services.  Ultimately provisions that are specific to either goods or services could be treated as special provisions contained in separate subsidiary chapters of the new common framework pf disciplines.

            These are long-term ideas that are unlikely to affect the near term negotiations in either goods or services. However, the advent of negotiations on common issues such as transparency in government procurement argues for the establishment of an institutional structure to accommodate agreements that apply to both goods and services.  Joint jurisdiction by both the Council for Trade in Goods and the Council for Trade in Services may be a good interim solution, but ultimately it may be more advisable to establish a Council for Common WTO Disciplines, just as the Dispute Settlement System falls under a joint body, i.e., the Dispute Settlement Body of the WTO.

Conclusion

            The negotiation of the General Agreement on Trade in Services during he Uruguay Round was a major achievement. The negotiators of the agreement made creative adjustments in the tool kit of GATT concepts and terms to fashion a set of disciplines suited for trade in services. In the end they proved too conservative in applying the GATT tool kit, by not realizing the full implication of the changes they were required to make to fit the needs of the services world. The result has been that the GATS suffers from ambiguities in key provisions, which undermine the value of negotiated commitments and the effectiveness of the agreement. This article has described some modest fixes to the agreement and the negotiating methodology that would help strengthen the agreement and the associated schedules of national commitments.          

References

Feketekuty, Geza, International Trade in Services: An Overview and Blueprint for Negotiations. Cambridge, Mass: Ballinger, 1988.

Group of Negotiations on Services, The Uruguay Round of Multilateral Trade Negotiations, “Scheduling of Initial Commitments in Trade in Services: Explanatory Note”, MTN.GNS/W/164. Geneva: GATT Secretariat, September 3, 1993.

Hoekman, Bernard and Patrick Messerlin, “Liberalizing Trade in Services: From Reciprocal Negotiations to Domestic Regulatory Reform”

Howse, Robert, “Comments on Feketekuty, Nicolaidis/Trachtman, and Zampetti Papers”.

Low, Patrick and Aaditya Mattoo, “Is there a better Way? Alternative Approaches Under the GATS,”

Mattoo, Aaditya, “National Treatment in the GATS: Corner Stone or Pandora’s Box?” Journal of World Trade Law, Volume #?, Year?

Nicolaidis, Kalypso and Joel P. Trachtman, “From Policed Regulation to Managed Recognition: Mapping the Boundary in GATS”,

Thompson, Rachel, “Formula Approaches to Improving GATS Commitments: Some Options for Negotiations”

Wilcox, Clair, A Charter for World Trade. New York: Macmillan, 1949.

World Trade Organization, The Results of the Uruguay Round of Trade Negotiations: The Legal Texts, Geneva: WTO, 1995.

World Trade Organization, WT/GC/W/189, “Preparation for the 1999 Ministerial Conference: EC Approach to Services.” Geneva: WTO, June 2, 1999.

World Trade Organization, WT/GC/W/116, “Preparation for the 1999 Ministerial Conference: Communication from Australia.” Geneva: WTO, November 28, 1998.  

top

return to Articles & News