Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 To what extent do International Trade Agreements on Services are contributing to government’s efforts to foster progressive liberalization? the case of Brazil (Executive Summary) By Roberto Echandi1 I. Introduction A fundamental function of international trade agreements concerns the national political economy arena. International Trade Agreements can act as shields for governments against pressures to adopt protectionist and populist policies -often advocated by powerful interest groups-- tempting politicians to increase their poll ratings in the short term. History shows how such policies almost always in the long run may backfire and have serious economic and social long-term effects for the country’s development. In the field of international trade in services, the need for sound and long-term policies becomes even more important given the exponential growth of trade in services’ weight in GDP and international trade, but also given their key role as inputs for tangible trade as well. Yet domestic regulations affecting trade in services remain significantly less transparent, more restrictive, and more procedurally burdensome, than those affecting trade in goods. Thus, there are two key features which makes a treaty on trade in services useful as an instrument to foster domestic reform. First, there is the extent to which it contributes to foster transparency of existing discriminatory and non-discriminatory barriers to trade in services which are often hidden and embedded in laws and regulations. Second, whether the agreement includes mechanisms such as standstill and ratchet commitments ensuring a gradual and progressive liberalization -or roll back- of existing trade in services restrictions. Within this context, a benchmark of the international agreements on trade in services negotiated by Brazil was undertaken. Such analysis focused on two concrete questions. First, to identify the gap between the regulatory status quo and the level of commitments that Brazil has undertaken in its different agreements on trade in services: GATS, MERCOSUR Protocol on Trade in Services, the Brazil-Chile FTA and the MERCOSUR-EU FTA. The higher the gap, the less useful the particular international agreement may be for the national government in its domestic efforts fostering economic reform. Second, the benchmark also compared the level of commitments undertaken by Brazil in each of these agreements with those assumed by a dozen countries through their respective Free Trade Agreements (FTAs). In particular, the benchmark covered the Members of the Pacific Alliance (Chile, Colombia, Mexico & Peru); the other BRICs (India, Russia, China and South Africa), the other major trade partners of Brazil in the Americas (Canada and the United States), and two sample countries in the Asia Pacific Region (Australia and South Korea). The purpose of such exercise was to find out whether and to what extent Brazil’s practices in international services negotiations conform or deviate from international trends. 1 Lead Trade Specialist, ETIRI, World Bank. All errors responsibility of the author. 1 Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 II. Benchmarking: Overview of the Methodology The quantitative analysis was based in what is known as the “Hoekman Index” (Hoekman, 1996, Eschenbach and Hoekman, 2006) which was originally developed to assess WTO members’ commitments in GATS.2 The study analyzed Brazil’s specific commitments in four specific trade agreements: the General Agreement of Trade in Services (GATS), the Protocol of Montevideo on Trade in Services in the Southern Common Market (MERCOSUR Services Protocol), the trade services chapter of the FTA with Chile, and the Brazilian commitments offered in the context of the services chapter of the FTA between MERCOSUR and the European Union, which was recently negotiated and it is not yet in force. For each of these treaties, each individual service subsector (based on GATS services sector classification - MTN.GNS/W/120 of 10 July 1991) and individual mode of supply were analyzed. An precise analysis of the gap between the regulatory status quo affecting trade in services in Brazil and the international commitments undertaken by the country in its international trade agreements would have required an official and detailed list of all existing measures non-conforming with the obligations of market access, national treatment, local presence, performance requirements and MFN affecting all the modes of supply of trade in services. Such information was not available to undertake the analysis. Thus, in order to overcome this challenge, a “virtual” baseline was constructed. Such “virtual” baseline aims to fix a conservative point of reference to visualize the minimum level of commitments with respect to national treatment and market access that Brazil could undertake in negotiating its trade in services 2 The “Hoekman Index” is based on an estimation for each legal obligation, mode of supply and service sector category. With 155 nonoverlapping services sectors in the GATS classification list and four modes of supply: (i) cross-border, (ii) consumption abroad, (iii) commercial presence and (iv) temporary movement of natural persons, there is a maximum of 620 commitments to be possibly assumed by a country with respect to each legal obligation. The benchmark under this study focused only on two fundamental obligations that affect the entry of foreign services suppliers into domestic markets, that is, national treatment and market access. As commitments apply to these two obligations separately, there were 1240 data points analyzed per each treaty. This index weighs each commitment undertaken by a country: if the commitment of a country establishes that there is “no” restriction for a particular sector in a specific mode of supply, such commitment is weighted as 1. If it is “unbound” or totally excluded from the coverage of the treaty is zero. If a country undertakes a commitment to bind a particular sector and mode of supply while leaving existing restrictions in place, the commitment has a weight of 0.5. Assigning such weight to binding commitments leaving restrictions in place reflects the perception that undertaking a binding commitment has value, regardless of how restrictive the measures maintained may be. An advantage of this approach is that it allows to ascribe a “value” to each treaty based on the bre adth and depth of the commitments included therein. (Hoekman 1996) However, these indicators present some limitations. First, they do not measure the country's real trade openness in services, but rather describe the country's commitments in the different service sectors. They may even underestimate the level of liberalization of a country, since there may be some sectors and modes of provision for which a country does not make any kind of commitment, but in fact there are no regulations that limit market access or national treatment. Second, the calculation assigns equal importance to all service sectors, when in fact some sectors are more relevant in the country's trade and production. Making such weighting would also require better statistics on trade in services. Regardless of these drawbacks, these indicators make it possible to assess the degree of commitment to the liberalization of trade in services between countries, or between different agreements signed by the same country. 2 Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 agreements, provided that assuming those commitments do not entail roll-back of any existing non- conforming measures. The baseline presupposes that there is no legal impediment for any government wishing to leverage a services trade treaty for purposes of fostering domestic reform to bind all 155 non-overlapping services sectors in all modes of supply, thereby mirroring the existing normative reality in the country. Thus, in principle, such virtual baseline would have no service sector or mode of supply excluded a priori, leaving the baseline with a binary coding: either a service sector would be bound with restrictions or bound with no restrictions. To enable comparison between the baseline and the examined treaties, the baseline also relies in the “Hoekman Index” referred to above, by which an open sector would be graded with a “1”, and a bound sector with restrictions would be graded with a “0.5”. This approach also enables the analysis to ascribe a value to the level of openness reflected by the baseline, subsequently allowing to compare it with the respective “values” ascribed to each analyzed treaty. For purposes of facilitating the reading of the results, most of the conclusions and graphics identifying the “water” or gaps between the baseline and the commitments undertaken were presented on the basis of percentages calculated with respect to the numerical value of the baseline. Thus, the logic would be that if Brazil negotiated a services treaty fully reflecting the baseline, such agreement would reflect a 100 percent of all the concessions Brazil could in fact undertake in such negotiation without entailing any rollback of existing restrictions . In such situation, the gap between the baseline and the negotiated treaty would equal to zero percent. Conversely, the higher the percentage of the gap, the lower the level of commitment that Brazil opted to undertake in a specific treaty. For example, a gap of 70 percent between the baseline and a particular treaty analyzed should be interpreted as Brazil being only willing to grant its negotiating partner(s) only 30 percent of all the concessions it could have otherwise be able to give. Based on the spreadsheet applying the methodology described above, the baseline used as point of reference for Brazil for purposes of this analysis represents a value of 794.5 points, representing the 100 percent of the level of national treatment and market access commitments Brazil could undertake without being forced to change any existing restriction. Such figure results from 891 out of 1240 data points showing restrictions (each with a value of 0.5), 456 affecting market access and 435 national treatment, added to 349 data points (each with a value of 1) showing no measures inconsistent with those two legal principles. A pattern worth noting is that in the case of Brazil, most of the restrictions affecting both market access and national treatment are horizontal and in practice tend to affect all services sectors both in modes 3 and 4, but also some in modes 1 and 2. That explains why in modes 1 and 2, Brazil could actually bind open trade for many sectors. Further, judging by the level of commitments in the context of the negotiation with MERCORSUR, Brazil’s international agreement with a wider and deeper scope of sector- specific commitments, five services sectors concentrate the majority of the “unbounds”. Without any particular order, they are: (i) research and development services related to genetic activities or aiming at exploiting certain minerals, (ii) health and social services, (iii) certain cultural services, (iv) space transport, (vi) transportation through pipelines and (vi) financial services -both insurance and banking. 3 Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 III. Summary of Key Findings The analysis found a significant gap between the level of concessions on market access and national treatment assumed by Brazil in its international agreements and the level that the country could actually take, even under the assumption of not undertaking any commitment to reform any existing law or regulation. The “water” between applied policies and international commitments is the lowest in the case of the MERCOSUR Services Protocol, equivalent to 38 percent, 62 percent in the case of the Brazil-Chile FTA, 87 percent in the case of the MERCOSUR-EU FTA and as high as 95 percent in the case of GATS. Further, because Brazil has not undertaken the same level of commitments assumed under the MERCOSUR Services Protocol in its other FTAs or GATS, the benchmarking finds that, keeping such high gaps between applied and bound policies in services trade for the rest of countries negatively affects Brazil’s ranking position in the sample of 12 countries included in the comparative exercise. Indeed, of all countries examined, Brazil comes up in the lowest position except for India. 4 Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 Figure 2: Benchmark of International Services Agreements Negotiated by Brazil and Peer Countries Source: Author on the basis of official texts Further, the analysis also shows that Brazil’s treaties tend to not include mechanisms to ensure full transparency of non-liberalized barriers to trade in services nor to lock-in unilateral openings which may occur in the future -ratchets. However, the study also finds that Brazil may make its services treaties significantly more useful to complement domestic trade reforms with minor technical steps that should not entail neither reform to the regulatory services status quo nor any political cost. With those technical steps, Brazil could quickly catch up with the level of commitments shown in the benchmarking assumed by countries of the Pacific Alliance and South Korea. In other words, Brazil could quickly position itself among the international leaders of countries undertaken services commitments in its trade agreements without having to undertake a single domestic reform to any national law or domestic regulation 5 Reforming services trade policies for development (P172000) Reform oriented operational support: Example Dec.2020 Figure 3. VI. Conclusion Brazil’s significant evolution in the architecture of its trade in services chapters could be better leveraged to make these treaties more useful to complement the government’s current efforts to liberalize trade in services, attract investment and foster policy coherence. Without undertaking any political cost associated with eliminating any additional restriction to trade in services, trade in services agreements negotiated by Brazil could be adjusted to significantly improve their capacity to: (i) lock-in the existing level of opening for trade in services in Brazil, (ii) significantly increase the level of confidence of services providers by improving regulatory transparency and predictability of remaining barriers to trade and (iii) lock-in future domestic reforms reducing the level of protectionism for trade in services in the country. 6