Failure of Trade Liberalization: A Study of the GATS Negotiation

Transcription

1 Journal of International Business and Law Volume 10 Issue 2 Article Failure of Trade Liberalization: A Study of the GATS Negotiation Christopher F. Thornberg Ph D Frances L. Edwards J.D. Follow this and additional works at: Recommended Citation Thornberg, Christopher F. Ph D and Edwards, Frances L. J.D. (2011) "Failure of Trade Liberalization: A Study of the GATS Negotiation," Journal of International Business and Law: Vol. 10: Iss. 2, Article 6. Available at: This Legal Article is brought to you for free and open access by Scholarly Commons at Hofstra Law. It has been accepted for inclusion in Journal of International Business and Law by an authorized administrator of Scholarly Commons at Hofstra Law. For more information, please contact lawcls@hofstra.edu.

2 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION: A STUDY OF THE GATS NEGOTIATION Christopher F. Thornberg, Ph D and Frances L. Edwards, J.D.t INTRODUCTION The General Agreement in Trade in Services (GATS) was touted as an exemplary international agreement among nations by which trade may be liberalized in the service sectors. The GATS negotiations differed in many ways from its predecessor, The General Agreement on Tariffs & Trade (hereinafter "GATT"). The most significant difference rested in the large degree of autonomy that nations had in determining the scope of liberalization to which they would commit. The outcome of the GATS negotiations resulted in a far less productive liberalization in service trade among the member nations. As, a result, it was not as successful as hoped. Instead, the first round of negotiations was rather disappointing due to the emergence of unwillingness to commit (to a large degree) to openness,. The whole process seemed to highlight the fact that countries tend to hesitate to venture into immediate openness for a number of reasons. This article discusses the GATS negotiation process and outcomes. In addition, it analyzes the cross-sectional results that emerged from the initial round of negotiation in 1995 in order to reveal the primary driving forces that determined the degree of liberalization that resulted. Five potential forces are considered: 1) the net gains from trade, 2) special interest influence, 3) international free-riding, 4) internal bureaucratic resistance, and 5) regionalism. The results of statistical analysis strongly suggest that special interests within nations have had the greatest impact on the degree of liberalization. Nations that have a comparative advantage in the production of services, and are therefore more likely to be exporting services, committed to the greatest degree of liberalization in service trade. This result stems from the implicit separation of negotiations over the liberalization of trade in goods (under the GATT) from the negotiations over trade in services (under the GATS). The impact of this separation would be to increase the influence of the special interest groups by reducing the number of competing interests. At the time, the implication was quite clear- that the future rounds of GATS negotiations would be much more successful if they were directly attached to the GATT. Section I discusses the GATS negotiation and the processes by which nations make commitments under it. Section II reviews the structure of trade protection, with a preliminary look at potential forces that are indirectly considered when analyzing the decisions by nations in their commitments under the GATS. Section III discusses five theories tested as potential explanations for the cross-country patterns of liberalization found in the data collected. Section IV discusses the quantification of the results of the GATS negotiations. The data was collected from the Schedule of Specific Agreements filed by the 137 nations associated with t Many thanks to Professor William Dougan of the Department of Economics, Clemson University and participants at a seminar given at the World Service Congress, 1999 for helpful comments and suggestions. 325 Published by Scholarly Commons at Hofstra Law,

3 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESS & LAW the agreement and then quantified into a measure of liberalization that is comparable across the countries for use in a statistical analysis. (See Appendix 1). Section V presents the results of a regression analysis that tests these various hypotheses. In conclusion, the authors discuss the potential reasons for the poor results of the GATS negotiations and discuss and recommend potential future strategies for negotiating international agreements. I. THE GATS NEGOTIATIONS The reduced costs of international travel and communication, combined with increasing world wealth and the corresponding increase in the demand for services, has lead to a steady increase in the international trade of services. These services include transport, communications, finance, business and tourism. The total value of this trade in 1995 accounted for approximately 19% of total trade, which consists of merchandise and service trade combined (See Table 2.1). This percentage is up from 15% in Total world income flows, i.e. the payments from domestically owned overseas factors of production, which are arguably another form of service trade, totaled nearly the same as service trade and had been growing at a similar rate. Table 2.1: Services as Percent of Total International Trade, North Western World America Europe Asia % 13.3% 20.7% 13.3% % 19.0% 20.0% 12.7% % 22.4% 20.3% 14.2% % 22.3% 20.3% 15.3% % 23.5% 21.4% 14.6% Data source- the World Trade Organization. The World Trade Organization (hereinafter "WTO") acknowledged this trend when it set up the first round of the GATS talks in the late 1980's. 1 This agreement was adopted as an extension of the GATT with a general mandate for the progressive liberalization of international trade in services by member nations. 2 The GATS agreement set out the general obligations of the members, the most important being Most Favored Nation treatment (hereinafter "MFN"), transparency, and freedom of payments and transfers. MFN treatment under GATS was equivalent to the treatment set up under GATT in that it obligated a member to treat a supplier of services from any VTO member nation with "no less favorable" treatment than would be given to suppliers of services of any other country. 3 Transparency created a requirement that each member clearly and fully reveal any laws 1 The General Agreement on Trade in Services, Uruguay Round Final Act, Annex 113, 33 I.L.M 1130 t April 15, 1994) [hereinafter "GATS"] (started in the 1986 as Punta de Este ministerial Declaration). 2 Id. at art. XIX(1). This process is discussed in Articles II-XV of the document. It is interesting that while the agreement indicates that existing barriers should be reduced over time in order to encourage increased access, it does not provide a direct mechanism through which this should occur. ' id. at art. II(l), (2). Due to the unilateral nature of the GATS commitments and the fear of international freeriding, nations were actually given the ability to exempt themselves from MFN under certain circumstances in order to restrict the access of specific other members to domestic markets. Any of these 'Article II exemptions'

4 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION or regulations that would affect another entering member's producers in services.' Finally, the fundamental concern of members that there would be no restrictions regarding international transfer and payments related to a member's commitments was assured under the GATS articles. 5 It appears that the goal of the initial GATS round was primarily to negotiate service trade and to create a structure that would facilitate future rounds of negotiations. As such, nations were given a large amount of leeway as to their initial commitment in the way of service trade liberalization. Each member country was required to supply the WTO with a form known as a Schedule of Specific Commitments (hereinafter "SSC"). Under these schedules, members indicated the specific service sectors in which they were willing to make commitments to the ideals of Market Access and National Treatment. 6 Market Access dealt with the treatment of the service as it entered the economy. 7 Member countries that agreed to complete liberalization in Market Access for a specific service sector agreed to not use any of the service trade barriers listed in Table 2.2. In contrast, National Treatment dealt with the manner in which the service provider was treated once it entered that market. Under the GATT, National Treatment was automatic something that was not true for GATS and distinguished the two. Liberalization implied that treatment should be no less favorable to the entering foreign service provider, once entered into the economy, than would be given to domestic or national services. According to the SSC, the members were also supposed to clearly announce any limitations to these ideals as well. 8 In practice, however, which sectors were actually listed on the schedule was left completely up to the nation in question after a series of rounds of bilateral negotiations between the various participating nations. 9 These were to be described on a sector by sector basis, were to indicate the inconsistency of the measure and its duration, must outline the conditions that created the exemption, and indicate those countries to which the exemption applies. The Council for Trade in Services has the ability to review exemptions in order to liberalize future treatment. In practice the use of Article II exemptions was actively and aggressively discouraged by the WTO, and in fact was not widely used by member nations. As such this data was not directly addressed in this analysis. See Annex on Article II Exemptions for more information. Negotiations over the financial and telecommunication sectors proved to be very contentious during the initial GATS round. As a result, a number of nations, most importantly the USA, were threatening to use the Schedule II exemptions to punish nations that were refusing to liberalize these sectors; see also, Bernard Hoekman, THE URUGUAY ROUND THE DEVELOPING CouNTRms (Will Martin and Alan Winters eds., Cambridge University Press 1996) [hereinafter "Hoekman"] Rather than allowing these problems to threaten the liberalizations already achieved in the other areas, the scheduling of commitments for these two categories was delayed until later rounds of negotiations could be completed. 4 GATS, supra note 1, at art.li(1), (2). 5 Id. at art. X1, XII (Exceptions under extreme situations of financial difficulties are addressed in Art. XII(: 1). 6 Id. at art. XVI, XVII. 7 Id. at art. XVI(1)-(2). 8 Id. at art XX. The scheduled commitments must specify terms, limitations, conditions and qualifications on both Market Access and National Treatment for the sectors listed in the schedule. In addition the scheduled commitments must specify time frames and dates of enforcement. 9 Id. at art. XVIII. The entire mechanism for scheduling commitments and limitations was quite complicated, with little assurance of any uniform outcome applicable to all members. In order to create the final schedule of commitments, a given member would submit "negotiation requests", an offer procedure that requested preliminary schedules from other member countries and might propose changes. The member countries that would receive requests for change would reflect on such requests and then offer, in the form a letter, a new draft schedule to be submitted by the requesting member. Bargaining might then ensue between two given member countries. At the same time each of those members may have been bargaining on requests made by other member countries. At the end of this elongated bilateral bargaining process there was a submission of all 327 Published by Scholarly Commons at Hofstra Law,

5 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESS & LAW schedules have been made freely available to the public and were used to create the measures of service trade openness presented in the following section of this article. Table 2.2: Market Access under GATS: Disallowed Barriers* (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; (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. *GATS Article XVI; 2(a)-2(f) Entries for each individual sector were further broken into four separate modes of supply. Mode 1 refers to the supply of services where the two parties remain in their respective territories during the transaction. Mode 2 refers to trade that occurs in the provider's territory to another nation's service consumer. Mode 3 creates a commercial presence in the service-receiving nation. Mode 4 involves the presence of natural persons within the receiving nation. 10 Nations have very little control over the first two modes of supply short of restricting international communication and travel.. In light of this, it is not surprising to find that the bulk of the GATS entries were in the latter two modes of supply, where national governments do have a large amount of control. As such, this analysis focuses primarily on these two portions of the schedules. The final set of specific commitments that member nations filed with the WTO were divided into two main sections for ease of use. The first section listed horizontal commitments. -These were the general conditions that applied equally to all the sectors that were listed in the document. The second section listed out any additional conditions that applied only to specific sectors only. The WTO required countries to classify these sector-specific commitments under twelve broad headings. Within each of these groupings there were a number of smaller sectoral classifications that corresponded roughly with 3-digit and 4-digit CPC schedules to the Negotiating Committee of the WTO for additional bilateral consultations. Due to the complexity of the system and the differences in bargaining positions of any number of member countries enmeshed in these scheduled negotiations, the entire bargaining and submission process was not only extremely complicated, but also took years for final outcomes to occur. 1o Id. at art. I(2)(a)-(d)

6 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION service industry codes.ii Both the horizontal and the sector specific entries must be taken into account when considering a country's commitment to trade liberalization within a particular sector.. On average, each nation only placed entries in slightly over a quarter of the total number of possible sectors. However, there is a wide range of variation in the number of commitments filed. Some nations had only one entry out of a total of 156 potential service sectors, while at least one nation had 128 entries in total. Table 2.3: Sector Classifications and Entry Statistics Collected from the GATS Schedules of Specific Commitments Total Sample Size- 137 Nations Total # Average # Min # Max # Broad Category Sectors Entries Entries Entries Business Telecommunication Finance Transport Other* Total * Other includes 3. Construction and Engineering, 4. Distribution, 5. Education, 6. Environment, 8. Health, 9. Tourism and Travel, 10. Recreation, Cultural, and Sporting, 12. Other For the purpose of this analysis the data has been aggregated into five major groups. Four groups correspond to the four largest WTO groupings, while the fifth is a combination of the other eight. These groups, with the corresponding WTO divisions, are displayed in Table 2.3. The first group is Business Services with 48 service sectors including such industries as computer consulting and architectural design services. The second group is Transport, with 35 sectors that include freight and passenger transport and other auxiliary services for a variety of transport media. A third group, "Other" is the combination of the smaller WTO categories including Construction, Health, Education, Entertainment, etc. It contains 32 sectors in total. The final two groups are Telecommunications and Financial Services with 26 and 15 service industries, respectively.12 HI. INTRODUCTORY ANALYSIS OF THE GATS Most legal analysis have viewed the GATS in regard to its purposes, its definition, and its outcome in terms of anecdotal cases where there have been challenges to the attempted negotiation by given nations. Lacking from this analysis appears to be a statistical view of the actual outcome and an analysis as to whether the purpose of the GATS negotiations had, in fact, been accomplished. A more proper evaluation would be to evaluate what nations have 11 There was a wide range of divergence as to how nations classified the various service sectors. To allow for comparability, judgment calls had to be made as to how to best aggregate the data. In general we used the broadest classifications used within each group possible. Additionally when a country limited liberalization to a small set of sub-sectors within one aggregate, this was taken into account when calculating the degree of openness. The total number of sectors was calculated to be Negotiations over these last two sectors proved to be very contentious during the initial GATS round. Thus these sectors differ from the others in that the length of negotiations was much longer. 329 Published by Scholarly Commons at Hofstra Law,

7 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESS & LAW pledged under the GATS in order to determine if the purpose of the GATS was realized. In taking this view, it is useful to view the entire GATS negotiation in economic terms, analyzing the reality of the GATS outcomes. It is standard mantra in the economics profession that when two countries engage in international trade, it is beneficial to both of the trading nations. Thus, it seems contradictory that nations typically exert such efforts to restrict the flow of trade. Stigler sets up a model to explain this behavior by considering the market for political influence. In his model, politicians set policy in order to maximize political support. Small groups of highly interested individuals often wield influence disproportionate with their size due to the free-riding problems that afflict larger groups. Hence, nations often choose policies that favor these small interest groups despite the inefficiencies necessarily involved.13 The structure of trade protection is a natural setting in which to explore this political influence model. This article continues this tradition by testing a number of potential explanations for the results of the first round of the GATS negotiations. GATS, while technically an extension of the GATT negotiations, differed from that agreement in a number of ways. One of the most important differences was that individual nations had a large degree of autonomy in the choice of which sectors to commit to liberalization in trade, as well as the extent to which they liberalized. A large variation in the extent and patterns of liberalization on the part of the member countries resulted.. This article seeks to explain the variance in these crosscountry patterns of commitment. Much of the work in the area of political influence focuses on cross-sectional differences within a country.1 4 It is useful to consider the structure of protection across nations. Implicitly this type of empirical analysis not only seeks to determine the driving forces behind governmental activity, but also tests the more fundamental idea that the rules that govern the behavior of governments are consistent across nations. There exists a small amount of literature discussing cross-country patterns of trade protection.' 5 These studies are made difficult by the problems with trying to collect a comprehensive set of trade barrier data that is consistent across nations. This article is one of the first studies that extensively examines barriers to service trade rather than goods trade. Therefore, this study would not have been possible without the natural experimental setting created by the GATS negotiations and the near-simultaneous filing of the SSC by each of the 135 members and 2 provisional members of the WTO. 13 George J. Stigler, The Theory of Economic Regulation, 2 BELL J. OF EcON. & MGT. SCIENCE, 3-21 (Spring 1971). 14 See ROBERT E. BALDWIN, THE POLITICAL ECONONOMY OF U.S. IMPORT POLICY (MIT Press, 1986); Richard E. Caves, 9 Economic Models of Political Choice: Canada's Tariff Structure, CANADIAN J. OF EcON.., (1976); William R. Dougan, Tariffs and the Economic Theory of Regulation, 6 RESEARCH IN L. & EcoN ; Real P. Lavergne, The Political Economy of U.S. Tariffs: An Empirical Analysis, THE J. OF EcON. HISTORY (Cambridge Univ. Press, 1984). 15 See John A.C. Conybeare, Tariff Protection in Developed and Developing Countries: A Cross-sectional and Longitudinal Analysis, 37 INT'L ORG., (1983); see also Bela Balassa, Tariff Protection in Industrial Countries: An Evaluation, 6 J. OF POL. EcON (1965)

8 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION IIL. FORCES CONSIDERED IMPACTING THE GATS NEGOTIATIONS BY NATIONS Five potential forces are indirectly considered when analyzing the decisions by nations in their commitments under the GATS: 1) the net gains from trade, 2) special interest influence, 3) international free-riding, 4) internal bureaucratic resistance, and 5) regionalism. 16 The gains-from-trade theory states that countries that have the most to gain from international trade in services-those with a comparative disadvantage in the production of service-will liberalize the most. The theory of special interest influence states that nations that have politically powerful minority interest groups, and that would be harmed by expanded trade in services, will be the least likely to liberalize. International free-riding addresses the concerns voiced at the GATS negotiations that smaller countries would have the incentive to abstain from direct negotiations with other nations and not liberalize while still enjoying access to markets opened through the negotiations among the major players. Internal bureaucratic resistance stems from the relatively heavy regulation of service sectors that is observed within most nations. This theory states that there will be constituencies within governments that will be opposed to liberalization of international trade in services due to the almost certain resultant loss of regulatory power that will be suffered. The prediction here is that nations with a greater degree of regulation will liberalize the least. Finally, regionalism addresses the concerns voiced by many economists that the growth of regional trade associations is having a negative impact on the drive for global trade liberalization. If so, nations within regional trade agreements will liberalize less.' 7 The result of the analysis paints a rather grim picture of the outcomes of the first round of the GATS negotiations. Rather than significantly opening world trade in services, the results show that the majority of nations failed to commit to the ideals of Market Access and National Treatment in but a few of their service sectors. Furthermore, the large number of 'unbound' entries confounded the secondary goal of GATS negotiations-transparency. The results of the regression analysis performed on the pattern of liberalization add to this grim picture. These patterns strongly imply that special interests within nations had by far the greatest influence on the degree of liberalization seen. The countries that were most willing to open their borders to service trade were those with relatively large endowments of human capital, i.e. those nations with a comparative advantage in the production of skill-intensive service goods and those nations who already have a large surplus in the net export of services. The 16 This paper has taken the additional step of empirically examining these measures within a multiple regression format to attempt to identify the primary forces that determined the patterns of results across nations. 17 BERNARD HOEKMAN & K.P. SAUVANT, THE POLrTICAL ECONOMY OF THE WORLD TRADING SYSTEM (P.A. Messerlin eds., 1995); see also Hoekman, supra note 3. In a number of ways this research expands on the work first begun by Bernard Hoekman. These two Hoekman papers provide the first preliminary analysis of the results of the GATS negotiations, as well as provide many helpful insights into the negotiation process, the structure of the final agreements, and the primary motivations of the major parties involved. This paper expands on Hoekman's work by introducing a more sophisticated algorithm to create comparable measures of service trade openness across nations, a task complicated by the fact that the difficulty in quantifying service flows has lead to the proliferation of the use of non-tariff barriers to trade. The resultant index takes into account more fully the extent of liberalization to which member nations committed to within specific sectors, rather than considering only if some commitment took place or not. To do so, a number of different types of limitations had to be controlled for, including basic negative restrictions on trade, to use of positive commitments in unbound sectors, and limitations on the number of sub-sectors to which a specific entry applies. 331 Published by Scholarly Commons at Hofstra Law,

9 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOuRNAL OF INTERNATIONAL BUSINESS & LAW same result holds true for trade openness in the capital-intensive telecommunication industry and the labor-intensive business service industries. In other words, the nations that were willing to reduce import barriers were nations who are in the best position to export large quantities of service goods in the event of liberalization and those who already are exporting large quantities of services. Conversely, those nations that stood to gain the most from liberalization in the imports of services-capital scarce nations-hardly opened their borders to service trade at all. These results stem from the implicit separation that occurred between the negotiations over liberalization in trade in goods (GATT) from the negotiations over trade in services (GATS). Omnibus trade agreements are successful in part because they create competing special interest groups, and thereby limit the influence of any one group. By separating GATT from GATS, this process was effectively circumvented. The implication is clear: future rounds of GATS must be directly attached to the GATT to be successful. IV. QUANTIFICATION OF THE SCHEDULED COMMITMENTS Service trade differs from regular trade in goods in many ways. Services, in general, are very difficult to quantify.1 8 Apart from its often-nebulous nature, trade in services is often associated with the movement of some portion of the factors of production into the importing nation, thus making the "imported" portion of the service difficult to measure.' 9 Therefore, the quantity of imports can no longer be controlled at the border. This has resulted in a situation in which the majority of barriers to service trade are regulatory in nature, including licensing and operation limitations. This section details how, for the purposes of this article, these regulatory limitations were converted into numerical measures of service trade openness that are comparable across nations. The first step was to classify the limitations into four categories based on the degree of severity. Each of these categories was then assigned a numerical value. There is necessarily some degree of subjectivity in these assessments. In order to reduce this potential error, the data was further divided into ten measures for each country, one for each of five aggregated sectors within each of the two modes of supply that were analyzed (mode 3 and 4). Following the framework mandated by the GATS, the information entered into the schedule for each sector was broken down into the four separate modes of supply. 20 This analysis will concentrate primarily on two modes of supply: commercial presence and the presence of natural persons. A nation essentially had four options for each entry. If the country entered "None", then this country agreed fully to the ideals of either Market Access 18 The measurement of tourist service imports in the United States, for example, comes from surveys handed out to a small number of departing US citizens in which they are asked to estimate how much money they will spend while abroad. 19 Rural France, Up In Arms, THE ECONOMIST (Sept. 9, 1999), One of the ironies of the trashing of the McDonalds Restaurant in France is that 80% of the ingredients that go into the food produced by these establishments in France are purchased from French suppliers. The remainder is purchased primarily from Europe. The question then is what makes this an American venture? The answer is brand capital, imported directly from the US. 20 The data was further broken into separate Market Access (MA) and National Treatment (NT) categories. This turned out, however, to be a false division. If a nation places severe limitations on National Treatment, this is equivalent to a severe restriction on Market Access, and vice-versa. Therefore for this analysis the two entries are combined into one

10 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION ("MA") or National Treatment ("NT") for this sector under that mode of supply. The second possible entry was some specific limitation or condition. 2 1 In this case the country agreed to the basic principles of NT or MA, but limited this treatment in some manner that must be fully explained by the country in the text of its Schedule. In practice, these limitations ranged from very minor conditions such as a simple licensing restriction to truly onerous limitations such as only allowing minority foreign ownership when establishing a commercial presence, or requiring nationality conditions for a majority of managers. In either case, the nation implicitly agreed that no further limitations would be put into place in the future and also to negotiate in good faith during future rounds of negotiations for the removal or reduction of any barriers that are in place. At the opposite end of the spectrum was an entry of "Unbound". This stated that the country did not agree to bind itself to either National Treatment or Market Access. Note that this does not necessarily imply that there did, at the time, exist substantial barriers to entry in that industry even though there might have been. What it did say was that the country reserved the right to erect such barriers in the future if it so chose. It is the authors' interpretation of these GATS articles that any commitments made in the horizontal portion of the SSC documents still applied to sectors that have been included in the schedules, even if they were listed as unbound. 22 The final option, of course, was not to include the sector in the schedule at all. This was effectively the same as remaining unbound, except that any positive commitments made in the horizontal section of the schedule did not apply to this sector. For the purposes of this article, these entries needed to be codified in some comparable manner. To avoid inserting too much subjectivity into this process, this data was converted into a simple cardinal scale that ranged between 1 and 4 on an entry-by-entry basis. The basic definitions of these codes are laid out in Table 3.1. A value of 4 represents an entry of "None". A value of 3 was entered if only a minor limitation was placed on an entry, while a 2 was used to represent a major barrier to entry. In general, the difference between a 2 and a 3 depends on the authors' estimates of the impact that the restriction would have on the long run viability of the venture. A 3 would be entered if the limitation were judged to be a minor fixed cost of entry such as a licensing restriction or a small local labor requirement. A 2 was used when the limitation directly impacted the long run control and/or profitability of the firm. An example of this would include a minority ownership restriction or strict limitation on the number of branches that could be opened. In the case of presence of natural persons this could include strict limitations on the length of a work visa for a foreign national or a requirement of nationality. 21 Note that these limitations specifically apply only to foreign service providers. Foreign suppliers are also required to comply with any national regulations that apply to domestic service providers. 22 It is unclear about individual nations' interpretation of the use of 'unbound' in specific sectors and the impact of horizontal commitments. Some nations used both "unbound" and "unbound except as listed in the horizontal commitments" in their documents. Whether these two entries were meant to be freely interchangeable is debatable; however, the GATS statutes clearly define horizontal commitments as applying equally to all sectors specifically listed in the document. Therefore we adhered to this interpretation. In practice most of the horizontal commitments in regards to Commercial Presence are negative in nature, so this has little impact on the result. However for Mode 4, presence of natural persons, many of the horizontal commitments are positive in nature. Under this case and entry of unbound may not be as bad as it sounds since under certain circumstances the nation does commit to either NT or MA. 333 Published by Scholarly Commons at Hofstra Law,

11 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESS & LAW Table 3.1: Coding Definitions by Case Code Negative Limitations Positive Exceptions Sub-Sector Limitation 0 Unbound Unbound No Entry.25 Major Limitations Unbound w/ Minor Minority of Sub-Sectors Exception.75 Minor Limitations Unbound w/ Major Majority of Sub-sectors Exception 1 None All Sub-sectors Since sectors that are "Unbound" do not list any barriers that are currently in place, we were faced with the problem of how to code these entries relative to sectors that did enter limitations. Here we chose to code "Unbound" as a 1; as if it were a more substantial limitation than an entry condition that was codified as a 2. This was done even though the entry "Unbound" does not necessarily imply that there were, in fact, any substantial barriers currently attached to that particular sector, only that the nation reserved the right to erect these barriers in the future. There are two reasons we chose to code the Unbound data with a 1. This first reason stemmed directly from the mandate of the WTO. When the original GATT agreement was drawn, there were three major goals. The first was the reduction of trade barriers currently in place, the second was to prevent barriers from being raised in the future and the third was the goal of transparency. While a substantial limitation on service trade contradicts the first goal, the nation was in line with the third goal of transparency by recording a limitation in the schedule and, by implicitly agreeing to not increase barriers in the future, the nation was in line with the second. Remaining Unbound clearly violated both the second and third goals, and may have violated the first. 23 The second reason we chose to code Unbound as a I was that the existence of sunk entry and exit costs implied that the simple threat of future barriers was enough to prevent entry into the market, hence remaining unbound was by itself a major barrier to entry. 24 Mode % 19.4% 31.2% 3.4% Unfortunately, the bound-with-negative-restriction was not the only type of limited entry used in the schedules. A second type of entry that occurred was an unbound-with-positive-commitment. In other words, in some cases a sector was listed as being unbound except for under certain circumstances where the country agreed to be bound. This type of limitation was most commonly used under supply Mode 4 relating to the presence of natural persons. For example, the nation may have remained unbound; however, it would make an exception and agreed to be bound in regards to the entry of high-level managers and for specialists 23 Hoekman, supra note 3. It is worth noting that Hoekman states that the nations that chose to remain unbound also typically had severe barriers in place. 24 NAFTA is a good reference point in analyzing the GATS and its impact on barriers to entry. The tariffs on goods coming from Mexico into the US were negligible prior to the NAFTA negotiation, and the immediate impact of NAFTA on these barriers was minimal. However despite this the NAFTA agreement immediately saw a large increase in the flow of investment into Mexico. This can be explained by the fact that NAFTA was not an agreement to simply lower current barriers, but rather more importantly an agreement to not raise barriers in the future

12 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION necessary for the operation of the firm. For a "minor positive exception" a 2 was entered, while for "major positive exceptions" a 3 was entered. The same criteria for rating minor and significant exception entries were used here as they were for the negative limitations. Exceptions that allowed the firm to operate more or less normally were rated as significant. Otherwise, they were rated as minor. A final type of restriction used needs to be addressed. Within each of the sectors listed by the GATS are a number of sub-sectors at the 4 or 5 digit industry level. Some nations limited their entries to apply only to a portion of the sub-sectors within that particular sector. For example, the entry in the Legal Services sector may be limited to work in International Law only, indicating that other Legal Services sub-sectors were effectively Unbound. If the entry was limited to a minority of sub-sectors then a 2 was entered. If a majority, but not all, of the sub-sectors were included then a 3 was entered. If this occurred along with a negative limitation or positive commitment then the minimum of the two entries was used as the final entry. When multiple restrictions applied, the minimum score was used. Table 3.2: Distribution of Entries Aggregated for 137 nations Entry Value Sector Mode Total Mode % 6.7% 10.0% 6.8% Mode % 9.3% 15.6% 1.3% Other Mode % 5.3% 10.8% 7.8% Mode % 9.3% 16.4% 1.3% Transport Mode % 2.4% 5.6% 3.6% Mode % 4.2% 7.9% 0.6% Communication Mode % 8.9% 7.3% 6.6% Mode % 7.9% 13.2% 1.9% Finance Mode % 23.5% 6.9% Business Mode % 6.1% 9.7% 8.2% Mode % 10.2% 16.7% 0.8% In total there were 137 members who filed schedules with the WTO, with approximately 156 sectors to which the members could commit, and 2 entries in each category-one each for the two modes of supply that were included in this study. Most of these entries were simply left blank by the member countries. Table 3.2 presents some summary statistics for the codified data. On average, countries chose to remain Unbound in about 75% of all sectors for both modes of supply. However, there was a high variance to this average, with the range going from 98% for some countries to as low as 30% for others. The patterns of commitments varied across groupings as well, with Finance having the largest degree of overall commitment and Transport having the least. These results show the outcome of the GATS negotiations to be substantially worse than what was calculated by Hoekman. 25 This is mainly due to the inclusion of the limitations on openness embodied in the data in the authors' determination. 25 Hoekman, supra note Published by Scholarly Commons at Hofstra Law,

13 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINEss & LAW V. THEORIES OF SERVICE TRADE LIBERALIZATION AS RELATED TO THE GATS COMMITMENTS The GATS Schedules of Specific Commitments provide an excellent setting in which to identify the determinants of trade policy because the results are (roughly) comparable across a large number of nations, and the results were all filed with the WTO within a fairly short period of time. Also importantly, each nation had a wide degree of latitude in choosing the degree and extent of liberalization. This variance in the patterns of results across nations can reveal much about the underlying motivations for the degree of openness in these types of trade agreements among nations. This section outlines a number of competing explanations for these patterns. One of the primary sources of the gains from international trade stem from specialization in production. Trade allows nations to concentrate their scarce resources in the production of the goods and services in which they possess a comparative advantage. They then trade internationally for those goods and services that they have a comparative disadvantage in producing. According to the Hecksher-Ohlin theory of trade, countries will have a comparative advantage in the production of goods that use the inputs, such as labor or capital, which are relatively abundant compared to the rest of the world. 26 Gains from trade can also arise from economies of scale by which a firm expands its operations. By specializing in the production of a set of products a nation can lower average costs below what could be achieved without trade in autarky. 27 Some stylized facts regarding the factor intensity of services are helpful prior to discussing potential explanatory theories behind the pattern of results seen in the GATS data. To create a rough estimate of the factor intensity of the tradable service sectors, data on education and capital investments were collected for a variety of US service and manufacturing industries. Figure 4.1 shows the average years of education per worker on a sector-bysector basis calculated from data taken from the Current Population Surveys crossed with the log of per worker capital investments taken from the Bureau of the Census's Annual Capital Expenditures Many other explanations for the patterns of trade also exist and work to explain such issues as intra-industry trade. Unfortunately the data on international trade in services is not detailed enough to allow these theories to also be examined despite their importance and potential relevance. Eli Heckscher, The Effect of Foreign Trade on the Distribution of Income, in READINGS IN THE THEORY OF INTERNATIONAL TRADE (1949) [hereinafter "Heckscher"]. 27 An autarky in economic terms is a closed economy that often views itself as self-sufficient. As a result, this type of economy often does not participate in international trade. 28 See Figure 4.1, U.S. CENSUS BUREAU, Current Population Surveys and Annual Capital Expenditures, available at Investment calculated as aggregate of 3 years investment, Number of workers and average education calculated from combined sample. Therefore average capital per worker must be viewed as a relative value only. Investment was calculated as an aggregate of three years investment, Number of workers and average education was calculated from a combined sample. Therefore, the average capital per worker must be viewed as a relative value only

14 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION Figure 4.1. Education and Capital by Sector, 1996 Data calculated for United States * Manufacturing and Mining Tradable Services O 15 0 O O&O*O %14 O 13-0 O Ln (Investment / Worker) It is not surprising to find that the tradable service sectors 29 tend to be skill intensive relative to manufacturing sectors with only a few exceptions. 30 On average, service sectors tend to be roughly equal in capital intensity compared to manufacturing sectors. However, the overall distribution of capital intensity is much larger for services, and includes some of the most capital intensive sectors (for example, telecommunications and utilities) and some of the least (higher education, medical and legal services). The business and other sectors are very labor intensive relative to manufacturer intensive, while communications are more capital intensive. This result, however, may still understate the true role of human and physical capital in the export of services. Service flows usually entail the movement of some portion of the factors of production to the receiving market. Transport costs for factors of production and the risks inherent in international factor flows are sufficiently high to ensure that the factors that will be moved will be those that are most specialized. As such, a large portion of service trade will be concentrated in the specialized subsets of service production. For example, it is likely that the US might export management services for cleaning crews, even if they employ the local cleaning crews to provide the actual service to foreign buyers. The actual flows of services will tend to occur within these fairly specialized sub-fields in an industry-a form of high-end outsourcing or procurement of products or services from an outside manufacturer or supplier. 3 1 In viewing trade liberalization, it is helpful to look at five concepts to explain why countries are open to liberalizing. The concepts this paper will analyze concerning liberaliza- 29 See Figure 4.1, Services excluding wholesale, retail and personal services. 30 The exceptions include certain transport sectors (railroad, trucking, and water), hotels and water supply services. 31 These procurements are made in order to cut costs. 337 Published by Scholarly Commons at Hofstra Law,

15 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESS & LAW tion are 1) Net Gains from Trade; 2) Political Economic Political Economic Forces; 3) Strategic Free Riding; 4) Regional Trade Regional Trade Agreements and Trade Trade Diversion; and 5) Internal Bureaucratic Internal Bureaucratic Resistance. Net Gains from Trade That there are some economic gains to be had from international trade is shown by the simple fact that agents willingly engage in it. However, trade often has a mixed impactthat is, while some groups benefit, others are harmed. The normative-positive pressures for liberalizing trade stem from the view held by economists that the gains from trade largely outweigh the costs. Governments that are concerned with national growth and development will seek to take advantage of these potential net gains by liberalizing trade in services. Of course, if these pressures were the only influence on government policy, all countries would be largely free-trade. However, the very existence of trade talks such as the GATS implies that countries do not have perfect free-trade economies and that there are other forces within economies that tend to be anti-trade. On the other hand, it can be thought that the countries that will enjoy the greatest net gains from trade will have the greatest incentives to counteract these other influences and pursue a more liberal service trade policy. It follows then that countries that are relatively scarce in the factors that are used intensively in the production of a good would benefit the most from reducing import barriers for those goods. For services, this would include nations that are scarce in human capital and for some service sectors, physical capital. 32 There are other sources of gains from trade beyond those relating to productive endowments (such as land, labor and capital) used to produce goods and services in a given country. Trade can allow countries to take advantage of external scale economies through increased specialization. 33 Trade also increases market competition, leading to more efficient markets and greater rates of innovation. 34 These latter two sources are especially important for small countries and for those that have small service sectors. Trade also represents one means by which technology can flow from rich to poor countries allowing for faster growth. In this sense, liberalizing trade is of greater benefit for technologically poor countries. 35 At one level the gains-from-trade theory predicts that small countries, poor countries, and countries that are comparatively scarce in human and physical capital will receive the greatest gains from trade, and should be more likely to liberalize trade in services. Alternatively, nations with a comparative advantage in the production of services may have no reason to erect barriers to service imports, and thus we might expect no relationship between measures of comparative advantage and service trade openness. 32 It might be argued that a more direct measure of comparative advantage, such as a vector of prices, would be helpful here. However, these direct measures are nearly impossible to collect for such a broad selection of nations. Therefore we must rely on indirect measures of comparative advantage such as factor endowments, including productive endowments (i.e. land, labor and capital) used in the production of goods and services. 33 See Paul Krugman, Scale Economies, Product Differentiation and the Pattern of Trade, 64 Am. EcoN. REv. 950 (1980). 3 See Elhanan Helpman, Increasing Returns, Imperfect Markets, and Trade Theory, in HANDBOOK OF INrERNATIONAL EcONOMIES (Ronald W. Jones and Peter B. Kenen, eds. 1984). 35 Nobuo Minabe, Capital and Technology Movements and Economic Welfare, 64 Am. EcON. REv (Dec. 1974)

16 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION Political Economic Forces Another important force that may impact the degree of openness is the political economy of trade such authors as Caves, 36 Lavergne, 37 and Baldwin 3 8 and reviewed in detail by Magee 39 and Rodrik. 40 While there are net gains to be had from expanding international trade, these gains also involve substantial transfers between the various constituencies associated with it. Not surprisingly, individual groups are more concerned with these transfers than the net gains or losses to the nation as a whole. As a result a whole new market forms-a market for political influence over the regulatory bodies that determine trade policy. Groups compete to have the regulatory bodies enact policies that will lead to positive transfers of wealth to them, or at least to prevent negative transfers away from them. Under the GATS we can expect that this type of rent-seeking 4 1 behavior will be pursued by groups that will be negatively impacted by increased service trade. These rent-seekers will lobby to erect or maintain existing barriers to trade, while those groups that would benefit from increased service trade-exporters and consumers-will encourage the removal of barriers. The relative degree of influence between these interest groups will determine the extent of trade liberalization. One of the more orthodox beliefs in the field of political economy is that government policy is often inefficient because small, concentrated groups tend to have a degree of political influence on the actions of government out of proportion with their size relative to larger groups within the economy. This is due to the public-good nature of public policy. 42 When a particular trade policy is put into place, the impact-whether negative or positive-applies equally to all individuals within some interest group. This inability to exclude political nonparticipants from the gains that might be collected from lobbying efforts creates the incentive for each individual within an interest group to behave strategically, relying on others to pay the costs necessary to influence policy while fully enjoying the benefits from the gains. During this process, free-riding occurs. In general, the larger and more diverse the group, the greater the free-riding problem; both because of the greater costs of organizing and the fact that the average gain from successful lobbying tends to be relatively small, as the total gains must be spread across a larger group. Interest groups that would be anti-trade include the scarce factors in the economy, as these are the groups that would find themselves competing directly against imported service products and be faced with falling wages as a result. 43 As services are human capital intensive, this would include skilled workers in human-capital scarce economies. Capital owners in 36 Richard E. Caves, Economic Models of Political Choice: Canada's Tariff Structure, 9 CANADIAN J. OF EcoN (1976). 3 REAL P. LAVERGNE, THE POLITICAL ECONOMY OF U.S. TARIFFS: AN IMPHUCAL ANALYSIS (1983). 38 ROBERT E. BALDWIN, THE POLITICAL ECONOMY OF U.S. IMPORT POLICY (1985). 39 Stephen Magee, The Political Economy of Trade Policy, in SURVEYS IN INTERNATIONAL TRADE (David Greenaway and L. Alan Winters ed. 1994). 4 Dani Rodrik, Political Economy of Trade Policy, in HANDBOOK OF INTERNATIONAL ECoNOMIES 3 (1995). 41 Rent-seeking occurs when some parts of the community wish to gain certain favors at the expense of others that are a part of the same community. 42 See Mancur Olson, The Logic of Collective Action, in CONTEMPORARY SOCIOLOGICAL THEORY 111 (Blackwell Publishing 1965). In Economics, a public good is one that if censured by one individual does not exclude its availability for individual consumption by others. Furthermore, individuals cannot be excluded from its use. Although this concept might not work perfectly in a real world situation, yet it is useful for economic analysis. 43 Heckscher, supra note Published by Scholarly Commons at Hofstra Law,

17 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINEss & LAW capital-scarce economies would lobby against the import of capital-intensive services, while labor owners in labor-scarce economies would lobby against in the import of labor-intensive services. In general this implies that nations with small endowments of human capital will be less open to service trade. The degree to which these anti-trade groups will be successful, however, depends on their political influence within the economy. If these interest groups are very small, they will usually have very little impact on the overall results of the negotiations. Hence we might expect a non-linear relationship between factor endowments and service trade openness. Those nations with a very small endowment of human capital may be more open to service trade than those with larger, and thus more politically powerful, human capital endowment. External economies of scale may also play a role in the determination of trade openness. Nations with relatively large service sectors will be able to take advantage of scale economies internally and thus will have an initial cost advantage over smaller nations. Smaller nations are also less likely to have competitive service markets, leading to rent-seeking activities by domestic monopolists. As a result, interest groups within smaller nations will have more to gain from lobbying to prevent the opening of trade. Strategic Free-Riding One interesting feature of the GATS negotiation is that while the commitments were determined unilaterally, the results automatically applied to all member nations of the WTO under the Most Favored Nation clause. One potential result of this type of structure was that some countries might choose to free-ride on the negotiations between other member nations."' 4 A nation could have sat out of the negotiation stage, chosen not to reduce trade barriers in any sector, and thereby enjoy unilateral access to other nations' service sectors. 45 If the countries had made their commitments simultaneously and without communication with other members, the entire process would have mimicked a large scale prisoner's dilemma. 46 In this case no country would have committed to any substantial degree of liberalization for fear of having competitive losses. However, the extensive negotiations that occurred prior to the filing of the schedules and the ability to amend the schedules over time as information was revealed by other nations ensured that the outcome was cooperative rather than competitive. Those countries that worked to open foreign markets for their service exports would find it in their best interest to focus their negotiation efforts on reducing barriers to the richest potential markets for their exports. The prediction that arises from this is that small and poor countries-those that do not represent fertile markets for service industriesare the nations who would be most able to free-ride on the results of the negotiations between the larger, richer countries. 4 Hoekman, supra note Id. It is clear that the Article H MFN Exemption was entered into the GATS agreement out of the concern of some member countries regarding free-riding behavior on the part of smaller nations in the negotiations. However, according to Hoekman, the use of these exemptions was actively discouraged by the WTO and in the end played little role in the outcome of the negotiations. 46 The prisoner's dilemma is one in which there are mixed motives by the participants and where the payoffs for cooperation among the parties fall between the payoffs for cooperation for one participant (who tends to lose the most in the trade) and competition by the other participant (who tends to gain the most from the trade)

18 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION Regional Trade Agreements and Trade Diversion Another possible influence on the degree of liberalization committed to is membership within a regional trade association (RTA) such as the European Union or the North American Free Trade Area. Regional trade groups may have the impact of lessening support for a more comprehensive degree of trade liberalization by lessening the potential benefits of freer trade overall and the benefits for specific pro-trade constituencies who are already enjoying access to a variety of foreign markets within the RTA. 4 7 Also, the complex internal negotiations that would necessarily have to occur between the member nations of an RTA to determine the extent of liberalization may further serve to impede the commitment process. Internal Bureaucratic Resistance A final source of potential conflict may arise from within the governments' own bureaucratic structures. Governments make efforts to regulate their economies in order to either promote a given leader's political self-interest or to ensure economic stability, or part of both. Opening domestic industries to competition from foreign nations threatens this regulatory control. Internal forces within the government will naturally be opposed to any action that threatens their control. This issue is particularly true of service sectors as service industries tend to be some of the most heavily regulated within an economy, especially in industries such as telecommunications and financial services. Overall the greater the degree of government intervention in the economy, the less likely liberalization will occur under this scenario. Statistical analyses were used to test the above ideas. Two sets of regressions were run, one using factor endowment measures of comparative advantage, and the other using more direct measures in order to reveal comparative advantage. VI. CONCLUSIONS The results of the statistical analysis highlighted in this conclusive article are an initial step in creating a comprehensive analysis of the results of the first GATS round of negotiations. Yet the work accomplished here still paints a strong picture. Based on the analysis, it appears that the GATS negotiation was a failure both in form and in function in its effort to begin the systematic opening of international trade in services. The first failure of the GATS was created by the nature of the GATS agreement itself. At its core the creators of the GATS were either unwilling or unable to create uniform schemes on how to deal with certain issues that are intrinsic in any importation of services into a foreign country, such as the repatriation of capital, differing tax treatment and governmental screening through authorization. The GATS gave a large amount of leeway to individual members regarding the scope of liberalization to which they committed. Nations unilaterally chose which sectors to which to commit, what types of exceptions to make to those commitments, and were permitted to remain 'Unbound' in whatever sectors they chose-in essence making no commitment at all. Furthermore, the GATS had no general mandate of enforcement to prevent the overuse of this leeway by member nations during the negotiation process. In fact the only punishment mecha- 47 See Jagdish Bhagwati, Regionalism and Multilateralism: An Overview, in NEw DIMINESION IN REGIONAL INTEGRATION (Jaime De Melo & Arvind Panagaryia eds., 1993); Robert E. Baldwin, A Domino Theory of Regionalism, in EXPANDING MEMBERSHIP IN THE EUROPEAN UNION (Richard E. Baldwin, et. al. eds., 1995). 341 Published by Scholarly Commons at Hofstra Law,

19 Journal of International Business and Law, Vol. 10, Iss. 2 [2011], Art. 6 THE JOURNAL OF INTERNATIONAL BUSINESs & LAW nism available was the Schedule II Exemptions that take the form of unilateral measures by individual nations to impose additional limitations on specific members of the GATS. The use of this mechanism was, thankfully, actively discouraged by the WTO leaving at least one basic principle of the GATS in place, that of Most Favored Nation status for all participating members. 48 The result of the leeway given by the WTO as to the degree and nature of the patterns of liberalization by individual members was the creation of a series of schedules that had a wide multiplicity of outcomes in terms of the degree of liberalization and the types of exceptions that were created. The schedules had little direct comparability among nations in the types of entries or even in the definitions of the individual service sectors that the entries were made for. This was unnecessarily complicated even further through the separation of entries into the four modes of transport and between exceptions that impacted market access and national treatment. 49 In contrast, the GATTGATS has always actively pursued uniformity through, for example, its tariffication efforts: 5 0 As a result there was little ability to compare openness for a single service sector across nations. This runs counter to the WTO's efforts to uniformly open markets and create transparency. The second failure of the GATS was in regard to the actual results that emerged from the initial round of negotiation. Overall the average level of participation was very small, with each nation only entering positive commitments in 25% of the total number of service sectors. The WTO claimed that this was due in part to the relatively short time period available for negotiation available to participants in the initial round. 5 1 However the results are not significantly different for the financial and telecommunication sectors for which the participating nations did have a longer period of negotiation available. Rather than nations pursuing policies that would most benefit their economies, the data seem to have indicated that special interests inside each nation almost completely dictated the degree of liberalization to which nations committed. The nations who committed to the greatest degree of liberalization of service imports were those that possessed a comparative advantage in the production of the service goods. In other words, the nations that would most benefit from liberalizing the import of services (those nations with a comparative disadvantage in the production of services) liberalized the least, while those who would benefit the least (those already with a comparative advantage in the production of services and who would be exporting these services if there was free trade) liberalized the most. This pattern was consistent across all sectors. Nations that possessed a large stock of human capital liberalized more than those nations with a relatively small stock. Nations with a large stock of physical capital liberalized more in capital intensive telecommunications sector, but less in 48 Coalition of Service Industries Research & Education Foundation, Services Will Dominate the 21'st Century Global Economy, Address before World Services Congress I (Nov. 1-3, 1999) [hereinafter "World Services Congress"]. 49 The differences between MA and NT were sufficiently unclear that on numerous occasions similar exceptions would be found in either category. In fact, many nations simply gave up trying to sort between the two and use 'as in MA' as an entry in the NT category and vice-versa. 50 Admittedly, creating uniformity in services is difficult because of the inherent problems of measuring quantities, and the fact that the provision of services often takes the form of the movement of factors across national borders. Hence tariffication is not necessarily an achievable goal. However the original GATS could have clearly listed types of acceptable non-tariff restrictions along with detailed explanations and limitations on their implementation rather than allowing nations to unilaterally define them. s World Services Congress, supra note

20 Thornberg and Edwards: Failure of Trade Liberalization: A Study of the GATS Negotiation FAILURE OF TRADE LIBERALIZATION labor intensive business and other service sectors. Nations already possessing a significant trade surplus in services liberalized more than those with a trade deficit in services. The influence of special interests cannot be underestimated. This was most openly reflected in the demonstrations and riots that occurred in Seattle during the start of the 2000 WTO talks. However, true influence was not always wielded by those as vocally or as clearly in the public's eyes as these Seattle demonstrators. As a commentary stated regarding these talks, while "world leaders may not show up [for] the conference... up to 100 members of Congress and representatives of many of America's largest corporations will be holed up at the hotels of Seattle, trying to influence the outcome of the talks." 52 The reality of this study is clear. If nations had been left to their own resources without any sort of multilateral agreement, countries would have behaved in much the same way as that which was accomplished by the GATS negotiations. In other words, the GATS failed to accomplish anything more than what might have been accomplished without any central organizational body at all. One major reason for this was the fact that the GATS was negotiated independently of the GATT agreement. This phenomenon severely limited many of the reciprocity tools that might have been used to increase the overall degree and consistency of openness and thereby the overall success of the GATS. To induce nations that would import services to reduce barriers to trade requires the offering of foreign access for their export sectors. For example, India might be induced to open up to trade in Consulting Services by the European Union by offering reduced barriers to the markets of these developed nations for the type of labor-intensive products India has a comparative advantage in producing. Such a broad-based scheme has the impact of bringing offsetting special interests into the negotiation process, and thereby limiting the influence of any one group. This is the primary motivation behind the omnibus negotiations that have characterized the various rounds of the GATT negotiations. The question then arises why the two agreements were not more closely tied at the outset to create consistency. in their outcomes on trade. This failure to link GATT to GATS might be most likely explained by the influence of special interests in creating the GATS agreement in the first place. Interestingly, this de-linking of GATS from the GATT may be thought to have created a circumstance under which it wasn't even necessary for there to be special interests directly involved in the negotiations for these patterns of results to have occurred. Trade negotiators are often judged by public opinion based not on the optimality of what they have accomplished, but rather on the trophies with which they return in the form of new export markets being opened. By limiting the trade negotiations to service goods only, the WTO effectively removed any potential trophies to be gained by the negotiators from nations without a comparative advantage in the production of services, and as a result these nations failed to substantially liberalize. Against this background, the question that arises is what did the GATS actually do? One argument might be that it was set up to simply lure nations into a frame of mind to begin the difficult process of lowering the barriers to service trade. The idea was to encourage nations to come to the negotiating table by allowing them enough freedom to not initially bind or limit themselves. The hope would be that an overall progressive liberalization would occur in the future. Theoretically, the result of such a strategy on the part of the WTO would be that 52 Martin Crutsinger, Trade Trouble Brewing in Seattle, on Eve of Worldwide Conference, Clinton Shoots for Success but could Face Disaster, AKRON BEACON J., Nov. 28, 1999, at Al. 343 Published by Scholarly Commons at Hofstra Law,