Impediments to Trade in Services

Measurements and Policy Implications

Christopher Findlay, Tony Warren

• Publisher: Taylor & Francis
• Year: 2013
• Print ISBN: 9780415240901
• eBook ISBN: 9781134550166
• Edition: 1
• Topic: Pedagogía
• Subtopic: Educación general

First Published in 2004. The economic impact of barriers to world trade and investment in services has been thought impossible to measure. As a consequence, significant global policy initiatives such as the General Agreement on Trade in Services have been concluded in an information void. This book challenges the view that impediments to services trade cannot be quantified, detailing how these barriers can be measured and their significance estimated. The book contains studies measuring impediments to trade and investment in a variety of sectors, including telecommunications, finance, shipping, education and air transport. The authors explain how the measures were calculated and show how the results could be used in sophisticated economic models. The final part of the book looks at current issues in services negotiations in the World Trade Organisation and the Asia Pacific Economic Cooperation forum. The measurements and methods detailed in this work have clear relevance to policymaking on services liberalisation and could be used by both international and regional organisations in services negotiations. This work will consequently prove to be an extremely valuable addition to the literature of the field.

1 Introduction

Christopher Findlay and Tony Warren

Studies of the measurement and impact of impediments to trade and investment in services are rare. Yet, for successful reform to take place, it is crucial that policies are transparent. Greater information, in terms of the detail of policy and the analysis of its effects, helps mobilise key countervailing interests against protectionist forces in domestic economies, aids political leaderships in constructing coalitions for reform and adds to the policymaker’s confidence as strategies are designed and implemented. The urgency of work in this area has been increased by the focus on services in the in-built agenda of the World Trade Organisation (WTO) and by the increasing awareness of the need to drive reform in service industries so as to ensure rising standards of living around the globe. The research challenge addressed in this volume is to characterise, assess and then measure the economic impact of policies affecting services trade.

The barriers impeding trade in services are opaque, given the nature of the transactions involved. Some information is available – often in a qualitative form and from a diverse range of sources. This book shows how this sort of information can be combined into robust assessments of policy that prove to be powerful explanators of market outcomes. This opens up new opportunities to use these sorts of measures in the design of reform programs and in the international negotiations associated with their implementation.

This chapter reviews the nature of trade in services and the impediments involved, and spells out the case for greater transparency. A method is outlined for achieving that transparency and the chapter illustrates its application to a number of sectors. Related modelling issues are discussed and the chapter summarises the implications for the negotiating process.

WHAT ARE BARRIERS TO TRADE IN SERVICES?

Services transactions

A service is ‘an economic activity that adds value either directly to another economic unit or to a good belonging to another economic unit’.¹ Consequently, services have as a defining feature the requirement for direct interaction between producers and consumers before the service can be rendered (Hirsch 1989).

This need for interaction influences how international transactions in services are conducted. If a producer in one economy can offer a service, then a consumer, resident in another country, must somehow interact with the producer to acquire it. Article I of the General Agreement on Trade in Services (GATS), substantially following Bhagwati (1984) and Sampson and Snape (1985), applied a four-part typology of how this meeting can occur internationally (also known as modes of supply):

• through cross-border communications in which neither the producer nor the consumer moves physically, interacting instead through a postal or a telecommunications network;

• through the movement of a consumer to a supplier’s country of residence;

• through the movement of a commercial organisation to the consumer’s country of residence; or

• through the movement of an individual service supplier to the consumer’s country of residence.

Unfortunately, data are not available on the volume of trade in services according to each of these possible modes of supply. We need to examine standard trade and investment data to develop a clearer picture of international services transactions – and even these data are incomplete.² Furthermore, data on services transacted by the movement of labour are almost nonexistent at an industry level. Chapter 2 in this book summarises the services trade and investment data that is available and highlights a number of features of the data, including the relatively rapid growth of services transactions. It is observed that for a sector once considered to be untradable, there has been a dramatic increase over the past decade in the proportion of total trade accounted for by services. This is true for both industrialised and developing economies. In some Pacific economies, the ratio of services trade to total trade is now over 30 per cent (Table 2A1).

WHY MEASURE BARRIERS TO TRADE IN SERVICES?

As international services transactions encompass foreign direct investment and the movement of labour, as well as traditional cross-border transactions, any policy that impedes service producers and consumers interacting through any of these channels (or modes of supply) is considered an impediment to trade.

Impediments to services trade come mostly in the form of non-tariff barriers (NTBs), reflecting the difficulties inherent in imposing tariffs directly upon either the service consumer or the service supplier as they interact across borders. NTBs are notoriously difficult to identify and measure. There have been very few systematic attempts to collect information on barriers to entry beyond the periodic trade reviews conducted by national trade negotiators.³ No equivalent of the United Nations Conference on Trade and Development (UNCTAD) database on NTBs affecting tradable goods yet exists for the services sector.⁴ As a consequence, very few studies have identified the barriers that exist or assessed the impact of these barriers on economic outcomes. This is concerning for several reasons.

At a policy development level, the lack of information on the extent and impact of impediments to trade in services undermines the liberalisation process. Evidence is available to suggest that service industries remain protected for the standard political economy reasons – protection is primarily afforded to uncompetitive industries that have significant political muscle (Warren 1996, 1997, 1998). For those involved in multilateral and regional negotiations, the evidence tends to confirm what many already suspect; negotiations on services encounter the same barriers to progress that are so familiar in negotiations on merchandise and agriculture. Powerful domestic interests limit the extent to which commitments to liberalisation will be made. If anything, the barriers to progress in services are even greater because of the relatively widespread involvement of the private sector in service provision and the public sector in service regulation.

Overcoming the forces of protection is never a trivial task and there is no simple solution.⁵ However, it is generally considered useful in the domestic political process to make the costs of protection as transparent as possible. Not only does this help build coalitions of interest for liberalisation, it allows policymakers to have greater confidence in the implications of any decisions they may make.⁶ The lack of information on impediments to trade and investment in services, and the consequent impact on the economy of these impediments, reduces the set of tools that policy reformers can use when pushing for liberalisation.

At a negotiating level, there is evidence that the desire for reciprocity has played a major part in determining the pattern of specific commitments made under the auspices of the GATS, as it has in other areas of the multilateral trade negotiations (Hoekman 1995; Warren 1996). It is generally accepted that in such a negotiating framework, multi-product negotiations (whereby concessions in one industry are traded for concessions in another) lead to more liberal outcomes by extending the set of industries over which concessions can be traded.⁷ As Nau (1987) has argued:

The across-the-board approach has clearly enjoyed the most success. It establishes politically salient overall goals early in negotiation while permitting great flexibility in subsequent negotiations to deal with individual products, sectors, barriers, or framework and institutional issues. By contrast, the product and sector approaches are, taken alone, unlikely to generate enough political interest and momentum to move negotiations forward at an early stage.

When dealing with NTBs, multi-product negotiations become technically much more difficult, as issues with the comparability of concessions arise (Olechowski 1987; Hoekman and Kostecki 1995). During the Uruguay Round, the problem of NTBs affecting agriculture was confronted directly with the development of the aggregate measure of support (AMS) and attempts to have this measure encompass all non-tariff and tariff barriers affecting agriculture (Croome 1995). Negotiators on services trade had no equivalent to the AMS, or even usable industry-level measures of impediments. During the Uruguay Round this may not have been too significant a problem, as negotiations seemed to have focused on developing the necessary framework and the specific commitments made by members appear overwhelmingly to have been examples of binding the status quo. However, as the new round of services negotiations approaches, the lack of information on impediments to services trade will undermine the potential for negotiated liberalisation.

HOW TO MEASURE BARRIERS TO TRADE IN SERVICES?

The measurement of impediments to trade in services involves a three-step process:

• First, available qualitative evidence that compares the way nations discriminate against potential entrants in various service industries is collected. This evidence is then transformed into a frequency-type index, with every attempt made to weight discriminatory policies by their economic significance.

• Second, the impact of the policies, as measured by the frequency indexes, is assessed against cross-national differences in domestic prices or domestic quantities, with the effect of other factors explaining these differences explicitly taken into account.

• Third, the measured impact of the frequency indexes (the coefficient) on prices or quantities is incorporated into a partial or general equilibrium model to assess the economy-wide impacts of the policies at issue. Where possible, partial equilibrium modelling is also undertaken to allow for the uncertain effects of liberalisation to be more clearly understood.

Chapters 4, 5, 10, 12, 14 and 15 provide examples of the first step of this method for measuring price impacts, generating frequency indexes of barriers to trade in services for various industries. Chapter 3 outlines a general approach to the second step of this method, giving particular attention to the identification of a benchmark against which price impacts can be assessed. Chapters 7, 11, and 13 use these frequency indexes for the telecommunications, financial services and maritime industries, respectively, to generate price-impact measures. Chapter 6 uses the frequency data on impediments to trade in telecommunications services to generate quantity-impact estimates. Chapter 8 presents an example of a partial equilibrium model and Chapter 16 discusses the challenges involved in the use of impact measures in general equilibrium models. The final three chapters of the volume examine policy issues.

Frequency measures

The measurement of NTBs using a frequency index involves:

• the collection of qualitative information on the impediments to trade; and

• the conversion of this qualitative information into a numerical index.

Collecting qualitative information has long proved an insurmountable hurdle in relation to services trade. To begin with, services trade issues were virtually ignored in policy and academic circles until the beginning of the Uruguay Round (Drake and Nicolaïdis 1992). It was not until the mid-1980s that any serious attempt was made to identify impediments to trade in services. Furthermore, the definition of what constitutes an impediment to trade in services continues to be a point of contention. For example, are prudential restrictions on offshore financial services firms or qualification requirements for foreign-trained doctors impediments to trade in services or legitimate regulatory instruments? Finally, it is a costly exercise to collect and verify the necessary information.

In relation to goods trade, significant efforts were made to bridge the information gap with the construction of such databases as the UNCTAD TRAINS NTM database. In relation to services, it was not until the end of the Uruguay Round that a significant international database on the incidence of NTBs became available. The requirement of the GATS that countries list in their individual schedules those sectors in which they were prepared to make commitments, and any specific barriers they wish to retain, produced the first systematic, if incomplete, database on impediments to trade in services.

The conversion of the qualitative information provided by the GATS schedules into a numerical index began with the pioneering work of Hoekman (1995) who developed a three-category weighting method as a means of assessing the extent of GATS commitments. Hoekman examined all GATS schedules and, for the purposes of assessment, allocated a number to each possible schedule entry (i.e. each possible commitment on market access or national treatment in each mode in each industry sub-sector). Specifically:

• Where a member has agreed to be bound without any caveats, a weight of 1 is allocated. A weight of 1 is also allocated in circumstances where a member declares that a particular mode of supply is ‘unbound due to lack of technical feasibility’, if other modes of supply are unrestricted. A common example of this situation is the cross-border supply of construction and related engineering services.

• Where a member has agreed to be bound but specific restrictions remain, a 0.5 weight is allocated. If a mode of supply is bound but specific reference is made to the horizontal commitments, a weight of 0.5 is allocated. This is commonly the case for commitments on the movement of natural persons, where immigration constraints continue to apply.

• Where a member has explicitly exempted that particular entry from the operation of the GATS by recording an entry of ‘unbound’ or by simply failing to make any commitments at all, a weight of 0 is allocated.

Hoekman used these measures to quantify the extent of commitments (the greater the number, the more commitments made). However, other researchers quickly realised the potential to use this information to construct a frequency index of impediments to trade in services. The Pacific Economic Cooperation Council (1995), for example, utilised the Hoekman analysis to highlight the number of commitments that have not been made (the greater the number the more illiberal the economy).

Welcome as these first steps were, these studies have several key limitations, which were identified immediately by Hoekman (1995), the Pacific Economic Cooperation Council (1995) and others. They are also reviewed in this volume with respect to barriers to foreign investment in services by Hardin and Holmes in Chapter 4. There are two major concerns:

• First, the coverage of the GATS schedules. The positive-list approach adopted for the GATS schedules means that countries only schedule information in those industries where they agree th...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Contents
• List of figures
• List of tables
• List of contributors
• Foreword
• Abbreviations
• 1 Introduction
• 2 The services sector in output and in international trade
• 3 Price-impact measures of impediments to services trade
• 4 Assessing barriers to services sector investment
• 5 The identification of impediments to trade and investment in telecommunications services
• 6 The impact on output of impediments to trade and investment in telecommunications services
• 7 A price-impact measure of impediments to trade in telecommunications services
• 8 Modelling the benefits of increasing competition in international air services
• 9 Regulatory reform in the maritime industry
• 10 Restrictiveness of international trade in maritime services
• 11 Price impact of restrictions on maritime transport services
• 12 Restrictiveness of international trade in banking services
• 13 The price impact of restrictions on banking services
• 14 Trade in education services and the impacts of barriers to trade
• 15 Measuring barriers to market access for services: a pilot study on accountancy services
• 16 Issues in the application of CGE models to services trade liberalisation
• 17 A comparison of existing services trade arrangements within APEC
• 18 Services issues in APEC
• 19 The WTO agenda: next steps
• References
• Index