SERVICES SECTOR REFORM AND DEVELOPMENT STRATEGIES: … · Services (GATS) negotiations and for business development in the services sector of developing economies can be supported

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT

POLICY ISSUES IN INTERNATIONAL TRADE AND COMMODITIES

STUDY SERIES NO. 8

SERVICES SECTOR REFORM AND DEVELOPMENT STRATEGIES:

ISSUES AND RESEARCH PRIORITIES

by

Christopher Findlay

Asia Pacific School of Economics and ManagementAustralian National University

Canberra, Australia

UNITED NATIONS

New York and Geneva, 2001

NOTE

The views expressed in this study are those of the author and do not necessarily reflect the viewsof the United Nations.

The designations employed and the presentation of the material do not imply the expression of anyopinion whatsoever on the part of the United Nations Secretariat concerning the legal status of anycountry, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers orboundaries.

Material in this publication may be freely quoted or reprinted, but acknowledgement is requested,together with a reference to the document number. A copy of the publication containing the quotationor reprint should be sent to the UNCTAD secretariat:

ChiefTrade Analysis Branch

Division on International Trade in Goods and Services, and CommoditiesUnited Nations Conference on Trade and Development

Palais des NationsCH – 1211 Geneva

UNCTAD/ITCD/TAB/9

UNITED NATIONS PUBLICATION

Sales No. E.01-II-D-7

ISBN 92-1-112517-0

ISSN 1607-8291

Copyright 8 United Nations 2001All rights reserved

iii

ABSTRACT

The implementation of strategies for domestic policy reform, for General Agreement on Trade inServices (GATS) negotiations and for business development in the services sector of developing economiescan be supported by the empirical work which is examined or proposed in this paper. The paper makesthe case for a more coherent approach to the design of reform policy in the services sector. It stresses theimportance of considering the interaction between services activities, between modes of supply and insome cases between goods and services. It also identifies limits to the progress towards the goal ofcoherent policy making in the services sector. These include lack of information and political economyconstraints.

Moreover, the paper contains a review of recent work on the measurement of impediments to tradeand investment, including research on banking, telecommunications, engineering, architecture, accounting,legal services, distribution and maritime services. There is evidence in this work of significant impedimentsto entry into services activities by all modes of supply and of significant discrimination against foreignsuppliers. The implied price effects of these impediments in banking and telecommunications are high insome economies.

Furthermore, the paper also presents one set of modelling results of the effects of liberalization. Theseresults highlight the value of a horizontal approach to policy making and the value of a focus on marketaccess compared to national treatment issues. Suggestions are made about the next steps in a workprogramme on the nature of services sector policy and its impact. This includes wider participation bydeveloping economies in empirical projects, more testing of methodologies being employed in those projectsand the development of a framework for the consideration of issues of domestic regulation and the designof competition policy. All this work will have important spinoffs for the effectiveness of participatingeconomies in GATS negotiations.

iv

ACKNOWLEDGEMENTS

This paper was written while I was a Visiting Scholar at UNCTAD in the Division on Trade in Goodsand Services, and Commodities. Thanks go to Bijit Bora, Mina Mashayekhi and two anonymous refereesfor helpful comments on an earlier version of the paper. Helpful comments were also received duringpresentations at the World Trade Organisation and the Australian National University.

v

CONTENTS

ABSTRACT .................................................................................................................................. iii

I. INTRODUCTION ................................................................................................................... 1

II. ISSUES IN SERVICES SECTOR REFORM .................................................................... 3

A. Interactions between services activities ............................................................................... 3

B. Interaction between modes of delivery ............................................................................... 3

C. Movement of natural persons ............................................................................................. 4

D. Goods and services interactions ......................................................................................... 5

E. Qualification ...................................................................................................................... 6

F. Interaction with competition policy ..................................................................................... 6

G. Overview .......................................................................................................................... 7

III. PROBLEM – HOW TO GET THERE................................................................................ 9

A. Policy transparency ........................................................................................................... 9

B. National treatment versus market access ......................................................................... 15

C. Political economy ............................................................................................................ 16

IV. CONCLUSION ................................................................................................................... 19

A. Sample sizes.................................................................................................................... 19

B. Impact study coverage ..................................................................................................... 19

C. Sensitivities...................................................................................................................... 19

D. Templates........................................................................................................................ 20

E. Mode 4 ........................................................................................................................... 20

F. Domestic regulation ......................................................................................................... 20

REFERENCES ............................................................................................................................ 21

1

Some years ago, a developing economypolicy-maker or business-person looking at exportopportunities in services might have concentratedon transport and tourism. They might also haveexpected to export labour services. However,recent developments in technology and in theorganization of international business havebroadened the scope of developing economyinterests. Moreover, this year, negotiations onservices commitments in the GATS haverecommenced. This paper outlines some issues fordecision-makers to consider and a series of relatedresearch projects to initiate, as they respond toshifts in the market place and as they plan theirapproach to the services negotiations.

The opportunities for developing economiesin services markets were highlighted recently bythe Global Services Network (GSN). This groupidentified a number of areas in which developingeconomies have “gained a foothold in knowledge-based and labor-intensive services”. GSNconcluded that entry into these markets wasfacilitated by the opportunities created by newinformation technologies and subsequently by e-commerce. GSN cited the experience of someeconomies in establishing data processing sectors,or call centres, as well as software supply groups.In these cited examples, services were providedin the cross border mode of supply, that is, withoutthe movement of either producer or consumer.GSN went on to list other areas which are “ripefor cross border trade.”1 These areas included:

§ Software programming;

§ Data capture and repair;

§ Management of electronic medical records;

§ Translation services;

§ Technical online supports services;

§ Database management;

§ Research & development;

§ Inventory management;

§ Website design and management;

§ Medical transcription;

§ Legal transcription;

§ Backoffice services for airlines, brokeragefirms, and credit card processing.

UNCTAD, for some time, has worked on thescope for developing economies to expandservices exports through all modes of supply. TheUNCTAD secretariat (1998) paper examined howglobalization, liberalization and new informationtechnologies are providing new opportunities fortrade in services. Case studies included computerservices, back office services, environmentalservices, health care services and tourism services.Moreover, UNCTAD has undertaken a series ofexpert group meetings in relation to these services.2

The paper also examined the conditions requiredfor developing economies to take advantage ofnew opportunities in the services sector. Thisincluded reference to developing economies’participation in the negotiations associated with theGATS. It also included mention of a variety ofmeasures that could be taken at the domestic level.

Despite the emerging opportunities, somecommentators remain concerned that developingeconomies have yet to establish themselves amongthe leading services exporters. Mashayekhi (2000)for instance stresses that since the adoption of theGATS, the developing countries’ share of worldservices exports has only increased by 6 per centand that this share is mainly due to thecompetitiveness of Asian developing economies.

I. INTRODUCTION

2

Furthermore, Mashayekhi stresses that developedeconomies still account for three quarters of worldservices exports and that most of the top 20exporters are developed economies.

While their aggregate shares may be low, thereis evidence of the competitiveness of developingeconomies in many global services markets.Karsenty (1999) shows that many developingcountries, including some least-developedeconomies already have a relatively high presencein services transactions by various modes. Heshows this group has a relatively high specializationnot just through the movement of natural personsbut also in terms of the delivery of services acrossborders and the export of services in the mode ofconsumption abroad.3

Many issues are involved in expanding theability of developing economies to take advantageof these emerging opportunities in the servicessector. Market access is one issue. But also theopenness of domestic services markets can helpbuild the capacity to participate in global markets.Liberalization increases the efficiency of theoperations of suppliers in intermediate services toall export sectors. Liberalization also brings incapital and technology. New businessopportunities are not only influenced by policiesaffecting market access at home and offshore, butalso by other national policy initiatives.

The effective participation of developingeconomies in the current services negotiations willbe a contributing factor to maintaining reform inthe services sector. A number of papers, includingthose by Hoekman and Messerlin (1999),Mashayekhi (2000), Mattoo (2000), Hoekman(1999) and Sauve (2000), have discussed thepriorities in the developing economies’ agenda inthese negotiations.

Apart from participation in multilateralnegotiations and national policy initiatives, thereare other elements in a portfolio of strategies for

services development. These include the use ofregional strategies and regional cooperativearrangements. These are topical and there isconsiderable work in progress in the applicationof regional arrangements. Important ground-breaking work, for example, is in progress inAPEC. However, detailed examination of specificregional approaches to reform is a comprehensivetopic and is beyond the scope of this paper.4

The purpose of this paper is to review someof the empirical work which can provide relevantinputs to the implementation of strategies for usein the GATS negotiations and for domestic policyreform. Section II provides a framework for theconsideration of that information. It reviews a seriesof issues related to reform strategies in the servicessector: it stresses the importance of consideringthe interaction between services sectors, betweenmodes of supply and in some cases between goodsand services. Section III identifies some issues inmaking progress toward the goal of coherent policymaking in the services sector. These issues includelack of information and political economyconstraints. The final section examines, in moredetail, a series of practical matters for furtherattention by decision makers in Government andin business.

ENDNOTES

1 This material was distributed by e-mail to mem-bers of GSN on 15 August, 2000. Subscriptions to this e-mail service are available atwww.globalservicesnetwork.com.

2A much wider range of issues is involved in thedevelopment of e-commerce. For a detailed discussion,see UNCTAD (2000).

3 Data on services FDI was not used in this study.

4 For a review of different regional approaches toreform within APEC, see Stephenson (2000a).Stephenson (2000b) contains further discussion of re-gional approaches to reform in the western hemisphere.

3

A. Interactions between services activities

It was noted above that UNCTAD hasnominated areas such as computer services, backoffice services, environmental services, health careservices and tourism services for attention bydeveloping economies. One of the key issues inthe reform of the services sector is dealing withthe interactions that exist between elements of thatsector. A successful outcome for one serviceactivity will usually depend on a package ofinitiatives or a cross-sectoral approach to thedesign of policy. The UNCTAD secretariat (1998)has also examined these sorts of linkages. It heldthat success in the nominated areas requires accessto “world class telecommunications infrastructure”,alongside financial support, the ability to movepeople temporarily and access to skilled staff.Failure to develop telecommunicationsinfrastructure will prevent developing economiesfrom gaining access to international marketsthrough electronic commerce.1

There are other examples of these sorts oflinkages. Reform of the air transport system, forexample, irrespective at policies applying to airportinfrastructure may not change market outcomes.2

UNCTAD (1998) also cites a similar issue withintourism sectors. Even when tourism operations indeveloping economies are competitively designedthey may nevertheless lose market share due tothe “pricing practices of monopoly transportationproviders bringing tourists to their markets.” Afurther example, cited in the same paper, arises inhealth insurance, where reform of the internationalhealth insurance market can help promote healthtourism.3

Linkages between services activities mean thatoutcomes of services negotiations might be moreeffective if appropriate bundles of services areconsidered in association with each other.4 Industry

better grouping of activities in services negotiations.For example the appropriate scope of energyservices has been debated, with United States firmspromoting a wide definition. There has also beendeveloped a concept of an express freight activitythat includes not just the transport component butalso complementary on-ground services. 5 Debateabout the appropriate scope to be considered inthe definitions of tourism, health and environmenthas also emerged during the UNCTAD expertgroup meetings on those activities.

The presence of interactive effects have ledto proposals for treatment in the policy makingprocess of clusters of activities. Groups of services(and possibly goods as well) should be consideredtogether in order to design a comprehensive reformprogramme. Clearly, some judgement on how topackage activities together and how to measurethe significance of various forms of impedimentswill be required.

B. Interaction between modes of delivery

Not only are there important interactions toconsider between services activities, but also linksbetween the modes of supply. Neutrality in termsof mode is important in both static and dynamicterms. Liberalization of one delivery mode withoutattention to others will divert activity through themore open mode.6

The UNCTAD secretariat (1998) has arguedthat the main emphasis under the GATS has so farbeen on improving market access conditions forthe establishment of commercial presence. It issuggested the reason for this is that commercialpresence could contribute to building domesticcapacity and expanding services exports. Thesecretariat maintains that this mode also allows for

II. ISSUES IN SERVICES SECTOR REFORM

groups make this connection. They argue for a

4

“more effective macroeconomic management andsupervisory and regulatory control.” Moreover, itobserves that most:

“Developing countries have not made use ofthe possibility of imposing conditions ofaccess (e.g. performance requirements) onforeign services and service suppliers,provided for under Article XIX.2. In grantingfurther access to their markets, developingcountries need to make full use of thesecapacity building conditions to ensureexpansion of services exports.”

Biases induced by different treatments of themodes of supply will be especially important forsome sectors where there are more options forsupply. One example is education whereconsumption abroad, cross border supply,movements of natural persons and establishmentare all important, some becoming more so. Tightrestrictions on one mode will shift consumers intoothers; for instance, restrictive rules on foreignestablishment will push students offshore, withimplications for other policy issues, such as netmovements in foreign exchange and the extent of“brain drain.”

It is also clear that technological change iscreating more scope for substitution betweenmodes. As noted earlier, these changes are thebasis for some of the new opportunities fordeveloping economies. Restrictions on othermodes of delivery will drive more people into usingdistance education, the quality of which isimproving.

The UNCTAD secretariat (1998) hasprovided a useful illustration of the importance ofthe complementarities in the linkages betweenmodes of supply. Just as in trade in goods, thereare complementarities between investment andtrade flows. In this particular example, activitiesin one mode lead to further activities in othermodes:

§ “…contract work in other economies(mode 4) is often used to build domesticcapacity by encouraging professionals to

take short-term assignments in economieswhere skills are more highly specializedand in-service training is available.

§ Returning professionals (in mode 4) withenhanced skills can then launch cross-border exports (mode 1) or meet foreignvisitors’ service expectations (mode 2)more successfully.

§ Business travel (mode 4) is also used tocomplement cross-border supply (mode1) in order to meet customer needs forpersonal contact. Similarly, mode 3(commercial presence) is often used tobuild capacity for mode 2 (where theconsumer moves) or mode 1 (cross bordersupply) by leveraging soft technologytransfer.

§ In addition, by attracting targeted foreigndirect investment (mode 3) in areas suchas software customization or healthfacilities management, developing countriescan build a reputation for quality that canbe translated into cross-border supply.”(UNCTAD, 1998).

In summary, it is valuable to pay attention tothe linkages between modes. A simple principleis to seek neutrality between all modes of supply.

C. Movement of natural persons

While neutrality between modes is important,developing economies retain their traditional stronginterest in mode 4. In Contrast, Adlung (1999b)notes that full bindings in the GATS on mode 4 arerare. He observes that the most frequent entriesin this mode are “unbound” to which somenarrowly defined exceptions are added. He saysthe focus of these exceptions is on seniorexecutives and other experts who transfer withina company, but that only limited commitments aremade to untrained or self employed people. Henotes, therefore, that most members have turnedthe negative approach generally used for schedulingmodal commitments into a positive or bottom-up

5

approach. The WTO secretariat (1998) providesmore details of the commitments in this mode.

Pakistan, in a paper to the WTO on this issue(WT/GC/W/131), noted the modest extensionsto commitments since the outcome of negotiationsin the Uruguay Round and argued the case forfurther coverage of this mode. Pakistan stressedthe difference between GATS provisions andmigration (a distinction which is also consistent withthe Annex to the GATS on the Movement ofNatural Persons Supplying Services under theAgreement). In order to facilitate further progress,Pakistan recommended:

a) There should be a horizontal approach toeconomic needs tests applied to this mode;

b) A review of the design and implementationof visa regimes would be useful;

c) Developing countries should participate inthe design of disciplines on qualificationsand licensing.

Mashayekhi’s (2000) assessment of progresson liberalization of this mode, is that a feature ofthe GATS is the “lack of commercially meaningfulcommitments (except on intra-corporatetransferees) on the movement of natural persons”.Moreover, she argues for greater transparency withrespect to measures affecting this mode of supply.An issue of particular concern is not just thepresence of the economic needs tests but themanner of their implementation. Her commentsand those from the Pakistan paper support the casefor further work on the analysis of policy regimes,and their operation, applying to this mode.

D. Goods and services interactions

Deardorff (1999) indicates how tradeliberalization in what he calls “trade services” (e.g.transport) can create benefits, by facilitating tradein goods or other services, that are larger than mightotherwise be expected. Moreover, he points outhow lower cost services can facilitate thefragmentation of production and participation of

economies in production processes according tomuch finer definitions of activities. This observationapplies to goods production and also to servicesproduction itself. An example is the application ofinternet technology that will enable developingeconomies to participate in services sectoractivities.7

The interaction between goods and servicessector reforms is also important. For example,the scope to pursue competitiveness in dataprocessing or analysis will be inhibited if the costof the hardware required is increased byprotection. Removal of impediments to trade ingoods may have little effect on economic welfareif the distribution system for the product isprotected by legislation restricting market access.

There are many examples of the importanceof an efficient services sector for effectiveagricultural reform, especially in developingeconomies (Anderson, 1999). A major tradereform programme in India in the early 1990sremoved biases against agriculture. However,Gulati (1998) argues that the agricultural supplyresponse was weak due to policies that remainedin place and which led to restrictions in the supplyof credit, transport and communications services.Similar conditions are expected in an agriculturalreform programme in China. The programmeremoved restrictions on farmer choice ofagricultural production but failed to create a policyenvironment in which the international marketsystem, for fruit and vegetables for example, wasinvigorated (Findlay and Watson, 1999).

The APEC process provides an illustrationof these linkages between goods and servicesissues and their impact on policy development.APEC members are focussing on the notion of anAPEC food system. This system is designed toremove barriers to the growth of two-way tradein agricultural products. It considers explicitly theservices whose supply at world prices is requiredto maximize and bring forward the advent ofbenefits to local communities in importingeconomies from the liberalization of productmarkets. It proposes, in particular, to integrate

6

the construction of infrastructure with theliberalization process in importing economies.8

E. Qualification

The theme of the remarks so far is thatwidespread reform and a movement towarduniformity in the application of policy, is a valuableprinciple. There is a qualification to this argument,since some services sectors might contribute in aspecial way to growth and development. In otherwords, there is a qualification due to the presenceof externalities.9 Differential levels of support mightbe justified on the grounds that some sectorsproduce costs or benefits that are not taken intoaccount by decision-makers operating in marketsfor those activities. This support might beimplemented in a variety of forms. The obviousform is to subsidize the activity which is generatingthe external effect.

Even where specific sectoral support isjustified, trade policy is rarely the best instrumentto use. Anderson (1999) stresses this observation,among others. He summarizes “five lessons ofrelevance” for dealing with non-trade concerns,namely:

§ Where there are several policy objectives,an equal number of policy instruments istypically required to deal with them;

§ The most efficient instrument for dealingwith a particular objective, such as solvinga market failure problem, is that whichaddresses the concern most directly;

§ Trade reform will be welfare improvingwhen optimal domestic interventions todeal with non-trade concerns are in place;

§ Trade measures are rarely the bestinstrument for dealing with non-tradeconcerns;

§ Whenever Governments intervene there isa risk of Government failure (due to manyfactors, including insufficient information ordeliberate action in response to lobbying).

These general principles are relevant to theassessment of all policy choices, proposed onmarket failure grounds. However, even if theseprinciples are met, options remain for an instrumentof intervention. Some judgement will still berequired about whether the policy chosen is areasonable response to an identified problem, orwhether it is more interventionist that necessaryand as a result, protectionist.

For example, licensing and other accreditationschemes are often imposed to deal with problemsassociated with consumers’ lack of informationregarding services suppliers. A number of questionsarise. Are these actually solving the problems orare they unreasonable impediments to marketaccess? Should foreign suppliers go through amore stringent accreditation process than domesticsuppliers?

F. Interaction with competition policy

UNCTAD has stressed its concern over risksassociated with the lack of competition in someinternational services markets. For example, theUNCTAD secretariat (1998) argued that:

“Many markets for services are dominatedby relatively few firms and the trend in merg-ers, acquisitions and strategic alliances hasexacerbated this situation. There is a need todevelop national competition rules as well asmultilateral mechanisms to strengthen coop-eration amongst competition authorities todeal with transfer pricing, exclusive dealingarrangements, alliances, mergers and acqui-sitions and export cartels. To provide forcompetition safeguards, the example of thereference paper on basic telecommunicationsservices could be followed in other servicessectors.”

UNCTAD’ s expert group meeting on airtransport was also concerned with competitionpolicy issues (UNCTAD, 1999). One of itsconclusions held that the WTO reference paperon telecommunications be examined as a modelof how competition policy issues might be dealt

7

with in air transport. Issues in air transport includethe use of subsidies, denial of access to essentialservices, rules on the allocation of airport spaceand so on.

Fink, Mattoo and Neagu (2000) concludethat while unilateral policy action in the maritimesector can remove trade restrictions and while theGATS can be used to bind those commitments,“there is also a need to deal with the possibleprivate anti-competitive practices of internationalmaritime cartels.” They observe that although largercountries may be able to effectively deal with theseissues through their own competition laws, smallercountries without enforcement capacity will be ata disadvantage. These conclusions are based onan analysis of bilateral data on trade in textileproducts on shipping routes to the United States.They find that removal of restrictive cargo policieswould reduce transport prices by 11 per cent. Butremoving the price-fixing agreements betweencarriers would reduce prices by 38 per cent.

Francois and Wooton (1999a) report similarresults with respect to shipping services. Theybuild a model in which they simulate both tariffcuts on goods shipped (agricultural exports fromdeveloping economies, for instance) and changesin the extent of competition in the shipping sector.They examine the effects of a reduction in tariffsand find that when tariffs are cut, shippers withmarket power can replace the trade-tax wedge tosome extent with a higher mark-up of their own.They also examine some numerical examples. Atone extreme, they find that “approximately one-half of the gains from full and unconditional marketaccess can be lost by producers to the shippingindustry when the latter is concentrated and fullyexploits its market power.”

Although liberalization, in some cases, mightremove regulatory constraints on the operation offirms with market power, it can also removeconstraints on entry. An important feature ofservice reform is therefore that changes in rules,for example, on the terms of market access willaffect the structure of the market in which the

services are provided. A greater degree ofopenness (across all modes of delivery) is expectedto lead to more competitive markets.

The effect of reform on the nature ofcompetition was modelled in air transport marketsby Johnson et al (2000). In a partial equilibriummodel, the authors captured the effects of entrynot only on the costs of supply but also on themark-ups which were observed in those markets.Costs of supply were affected in this model bycomparative advantage factors and also bydynamic efficiency effects (a consequence ofcompetition was to force competitors closer to theircost frontiers). They were also able to capturethe opportunities for greater flexibility and thereforethe efficiency gains that more open markets created(by removing the constraints on networkingassociated with bilateralism).

G. Overview

In summary, some of the principles whichemerge from this review of the issues are:

§ It pays to aim to pursue a top-downapproach to reform, that is, to work harderto reduce the levels of assistance providedto highly assisted activities;

§ It pays to seek neutrality between modesof supply;

§ It pays to consider bundles of activitieswhich are closely related to each other.Moreover, the appropriate bundling maywell include some goods producingactivities.

Because of their nature, services also attracta considerable regulatory load. This being thecase, attention to regulatory reform, including thebuilding of institutional capacity, whichcomplements liberalization is important. There arealso important interactions with competition policy.Openness can also contribute to competition policygoals, however more efficient instruments for

8

dealing with competition policy issues may haveto be developed.

As the UNCTAD secretariat (1998) stated:“these features increase the importance of coherentpolicy and regulatory frameworks for both thedifferent services sectors and the interlinkagesbetween the goods and services sectors.” Whilethe case for greater coherence in policy design isundeniable, significant constraints are encounteredbefore getting to that outcome. The next sectionexamines a number of such issues.

ENDNOTES

1 See Thompson (1999) for a thorough discussionof classification issues in the GATS related to e-com-merce.

2 Issues in the reform of air transport systems havebeen examined by UNCTAD (1999).

3 Health services has also been the topic of an ex-perts group meeting – see Zarilli and Kinnon (1998).

4 This point is the same as the argument for theconsideration of effective rates of protection, or assist-

ance. Hoekman and Messerlin (1999) also stress the valueof an “effective rate of assistance” perspective on thedesign of services reform. This approach would takeaccount of impediments to trade and investment apply-ing to all inputs to particular activities.

5 The World Services Congress included a sessionon “How the Energy Services Industry Can Use TradeNegotiations to Secure and Expand Markets” in whichthe abstract of the session read “Energy services are forthe first time being considered subject to multilateraltrade negotiations both in the US and EU. But issues ofdefinition need to be resolved: which services and prod-ucts should be defined as energy services? What arethe most important goals for energy services companiesand energy producing countries?” The Congress alsoconsidered the definition of express freight services.

6 See for example, UNCTAD’s work on digitizableproducts.

7 The Chairman’s summary of the UNCTAD expertgroup meeting on international trade in health servicesprovides more illustrations.

8 An update on the APEC Food System is avail-able at the site of the APEC secretariat atwww.apecsec.org.sg (in Task Forces under APEC Forain APEC Activities).

9 The same issue is highlighted in UNCTAD (1998).

9

A. Policy transparency

One problem associated with formulating amore coherent approach to policy-making is thelack of transparency of current policy. Although anumber of methods have been used to assess theextent of protection afforded various servicessectors, more work remains to be done.

One method for assessing the extent ofprotection in services is to look at gross operatingmargins in different sectors. Hoekman (1999) citesthe results from a study of margins in sectors (usingthe Worldscope database). As Hoekmanobserves, a number of factors will affect marginsand not just high barriers due to Governmentpolicy. However, a correlation between barriersand margins across countries can be expected.Hoekman uses manufacturing margins as areference point, but as he also points out, foreignentrants in many cases have to establish a newbase to supply services and so margins in servicescan be expected to be higher to some extent. Theresults show that operating margins in the servicessector are generally 10-15 points higher than inmanufacturing. Furthermore, after a review of thesectoral data he concluded that “in manydeveloping economies (services) margins aresignificantly higher than in industrialized nations.”

Another methodology, used by Francois andHoekman and also quoted in Hoekman (1999),was to apply a gravity model to trade in services.They used data for trade between the United Statesand its partner countries. By comparing observedtrade with benchmarks, they estimated tariffequivalents in trade by sector. They foundconsiderable variation between economies inimplied tariff equivalents in business and financialservices and much higher rates in constructionservices.

Another much simpler approach, used thecommitments documented in the GATS to developinventories of impediments. Recent examples ofthis approach include the work by the OECD in anumber of services areas including wholesale trade,retail trade, tourism and environmental goods(OECD 2000a, b, c and d).

Further work has extended the earlierresearch of Hoekman (1995) which examined thefrequency of occurrence of various types ofimpediments, based on information in the GATS.For example, PECC (1995) reported theapplication of this methodology to APECeconomies and Marko (1998) used this approachto evaluate the telecommunications agreementunder the GATS.

Work of this type has been extended in jointresearch at the Australian National University andthe Australian Productivity Commission (seeFindlay and Warren, 2000).1 The methodologyused in these impediments to trade in servicesstudies is summarized in Figure 1.

Policy information is collected from a varietyof sources, including from the GATS and otherindustry information sources. Policy applying to aservices activity is then broken down into a numberof dimensions (for example, restrictions onestablishment, rules on on-going operations,restrictions on movement of natural persons, etc).Each dimension is given a weight for the calculationof an index. The weight represents theresearchers’ view of the economic significance ofthe rule from the perspective of the supplier of theservices. The degree of restrictiveness of eachaspect of policy for each economy included in thesample is then given a score of zero for open orone for most restrictive. An index value iscalculated using the weights for each element of

III. PROBLEM – HOW TO GET THERE

10

policy. In some cases it is possible to go furtherand relate the calculated index values to marketoutcomes, such as prices of services in eacheconomy or quantities traded. This statistical workallows for the influence of other factors likely toaffect those variables. If the policy variable issignificant in these questions, its coefficient can beapplied to work out the likely effects ofliberalization on prices in each economy. If thepolicy variable is not significant, then alternativeweights or a new scoring system can be developedand the model tested again.

Figure 2 shows the estimates of the effectson prices of restrictions on trade and investmentby groups of economies on banking (in 1997) andtelecommunications (in 1998) services derivedfrom this approach.2 Economies are groupedaccording to categories used in the World Bank’sWorld Development Report. The sample sizesare generally largest for the high income economiesand then fall to small numbers of the low incomeeconomies. Therefore, the reliability of results varybetween income groups.

The results indicate that the price effects ofthe impediments are significant. They range from5 per cent to 55 per cent in banking and from 0per cent to 138 per cent in telecommunications.Out of a sample of 38 economies, 11 in bankingand 4 in telecommunications had price effects over30 per cent.

A striking feature of the chart is that theaverage price effect within each income group risesas the category of the economies shifts from highto low income. Therefore, according to the data,the impact of impediments is much higher in lowerincome economies.

The majority of work on price effects, has sofar concentrated on banking andtelecommunications. However, index values havebeen estimated for a number of other serviceactivities. Figure 3 shows the values of the policyindex measuring the extent of impediments applyingto foreign suppliers in banking, telecommunicationsand maritime services. The index values forbanking and telecommunications were used to

Figure 1Research methodology in studies of impediments to services trade and investment

Source: Derived from work reported in Findlay and Warren (2000)

Policy

Weights/ScoresWeights/Scores

Index

Statistical work

Marketoutcomes

Sign. Not Sign.Wedges

Otherfactors

11

Figure 2Price effects in banking and telecommunications services

0

20

40

60

80

100

120

140

160

High Upper Middle Lower Middle Low

Income group

Ave

rag

e p

rice

eff

ect

du

e to

re

stri

ctio

ns o

n fo

reig

n su

pplie

rs (

%)

Banking Telecom

Source: Productivity Commission (1999a).

Figure 3Index of impediments to foreign suppliers of services

Source: Productivity Commission (1999a).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

High Upper Middle Lower Middle Low

Income groups

Fo

reig

n s

up

plie

r in

dex

val

ue

Banking Telecom Maritime

12

calculate the price effects in Figure 2. Price effectestimates in this format are not available formaritime services necessitating index values to bepresented instead.

Figure 3 shows the rising trend in the indexvalues of restrictiveness for banking andtelecommunications as the income category shiftsto lower levels. The index values for maritimeservices follow the same trend. The columnsindicate higher degrees of restrictiveness in lowerincome economies.

Policy indices have also been estimated forprofessional and distribution services. These areshown in Figure 4. The levels of restriction appearto be similar for all income groups in the case ofengineering, architecture and legal services. Thereis, however, some evidence of rising degrees ofrestrictiveness in impediments applying to foreignsuppliers of accounting services and distributionservices.

The structure of policy impediments is similarin all the listed professions, making it possible to

use the indices to formulate comparisons acrossactivities. The data reported in Figure 4 indicatethat the degree of restrictions applying to foreignsuppliers is relatively low for engineering andarchitecture but higher for accounting and legalservices.

The methodology applied allows impedimentsto be divided into two groups, those applying toestablishment and those applying to on-goingoperations. The contribution of these types to theaverage scores over all economies for all activitiescovered in this work is shown in Figure 5.

The data indicate that typically the relativeimportance of these two types of impediments issimilar for banking, telecommunications anddistribution services. But for all other services,the restrictions on establishment tend to be the moreimportant contributors to the overall index values,while for maritime services a larger contributioncomes from restrictions on on-going operations.However this variation is mainly a reflection of theweights used in the calculation of the indices, andthese weights reflect the judgement of the

Figure 4Foreign supplier index values for listed professions and distribution by income group

Source: Nguyen-Hong (2000) and Kalirajan (2000).

0

0.2

0.4

0.6

0.8

1

Engineering Architecture Accounting Legal Distribution

Fo

reig

n in

dex

val

ue

High Upper Middle Lower Middle Low

13

Figure 5Contributions of restrictions on establishment compared to operations to average index values

Source: Author’s calculation based on Figures 2-4

Figure 6The share of the discriminatory component of the total foreign index value

Source: Author’s calculation based on Figures 2-4

0

0.1

0.2

0.3

0.4

0.5

Bankin

g

Telec

om

Maritim

e

Engine

ering

Archite

cture

Accou

nting

Lega

l

Distrib

ution

Fo

reig

n s

up

plie

r in

dex

val

ue

Establishment Operations

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Telec

om

Accou

nting

Maritim

e

Engine

ering

Distrib

ution

Lega

l

Archite

cture

Bankin

g

Sh

are

of

dis

crim

inat

ion

in t

ota

l fo

reig

n in

dex

va

lue

High Upper Middle Lower Middle Low

14

reseachers about the importance of various typesof impediments.

Trade negotiations on impediments to tradeand investment in services made a distinctionbetween impediments to market access whichapply to all suppliers, domestic and foreign andthose which discriminate against foreign suppliers.The index methodology also captures thisdistinction. The foreign index scores includeimpediments which apply to all potential entrantsand also includes any special provisions applyingto foreigners. Figure 6 summarizes the contributionof the discriminatory component to the total foreignindex value.

Figure 6 shows that in most cases thediscriminatory component accounts for the bulkof the overall value of the index representing thedegree of restrictiveness applying to foreigners. Fora number of sectors (maritime, engineering,distribution, legal and architecture), this elementaccounts for about two thirds of the total value.Its contribution is especially high for banking.However, the discriminatory element is elativelylow for telecommunications and accounting. Thesignificance of the discriminatory element inprofessional services is similar to the weight placedon those sorts of measures in the index calculations.The significance of the discriminatory element fordistribution, banking and maritime service is muchhigher than its weight in the index. In those sectorstherefore, the impediments observed are relativelyhighly concentrated among the discrimantoryinstruments.

Similar methodology has been applied in otherresearch on the effect of policy on the performanceof various activities within the services sector. Thework of Fink, Mattoo and Neagu on maritimeservices has already been referred to and theimpact of restrictive trade policies on prices wasreported to be significant and of the order of 11per cent. In another transport sector study, Goneneand Nicoletti (2000), the performance of the airtransport industry in OECD economies isexamined. They found that stricter regulatoryregimes are associated with higher values of

estimates of the degree of inefficiency in theindustry in particular economies. A number ofeconomies have highly restrictive regimes whichare associated with expected inefficiency levels ofthe order of 40 per cent. Higher inefficiency isexpected to be correlated with higher prices. Atthe route level, they find that the policy environmentalso contributes to significant effects on faresthrough its contribution to actual and potentialcompetition.

The following key observations are resultsfrom research which is relatively now. There is asubstantial research agenda which can help confirmthese results and the final section of the paperoutlines some options for its development.

§ There are significant impediments to tradein services, particularly intelecommunications, banking and transportas well as some of the professional services.High levels of impediments in these fieldsare likely to have significant effects on othergoods and services exporters. A servicessector reform programme which ignores thehighly protected sectors could even reducethe efficiency of the allocation of resources;

§ In policy applying to the services sector,there is usually significant discriminationagainst foreign suppliers, although intelecommunications, general market accessconditions are relatively more important;

§ The relative importance of barriers toestablishment and operation vary betweensectors, with establishment barriers beingmore important in the professions;

§ In general, impediments to internationalservices transactions tend to fall as incomerises, except in some professional servicesactivities. Thus, developing economies,even if they wish to focus on various crossborder activities, also face some importantdomestic policy challenges to reduce theburdens that are now carried by thoseexportable sectors.

15

B. National treatment versus marketaccess

Impediments to trade in services may bedivided into market access restrictions andderogations from national treatment.3 Part III ofthe GATS explicitly introduces the concepts ofmarket access and national treatment into theinternational services trade architecture.Surprisingly, as Warren and Findlay (2000) pointout, the GATS does not specifically define marketaccess. Article XVI (1) of the GATS, simplyobliges members to grant market access toscheduled industry sub-sectors, while article XVI(2): (a)-(f) contains a list of measures consideredto be limitations on market access. Article XVII(1) defines national treatment as treatment no lessfavourable than that accorded to like domesticservices and services providers subject to thelimitations and conditions set out in the country’sschedule of commitments.

The implication of Part III is that marketaccess and national treatment are broader in scopethan were the corresponding market access andnational treatment provisions in the GATT. To beginwith, the GATS provision on national treatmentdoes not draw a distinction between frontier andinternal constraints but embraces all policies thatmight discriminate between domestic and foreignsuppliers. In contrast national treatment in theGATT extended to matters of internal taxation andregulation only. In effect, the GATS article onnational treatment encompasses both nationaltreatment and market access as normally defined.

Box 1Examples of impediments (in the accounting services sector)

Market access National treatment

Barriers toestablishment

Rules about the form ofcorporate organizationfor accounting firms

Restrictions on who can bean owner of an accountingfirm

Barriers to on-going operation

Requirements that parti-cular services be providedby staff with specificqualifications

Restrictions on the servicesthat foreign accounting firmscan and cannot provide

More importantly, the GATS article on marketaccess extends beyond traditional concerns ofaccess for foreign services suppliers to encompassall policies which restrict access to a market. Thisis a major extension of multilateral trade disciplinesinto the realm of domestic policy, in particular,competition policy.

Within the GATS structure, commitments aresought in both areas of market access and nationaltreatment. Are there priorities? Some indicationof the impact of the different priorities in a reformprogramme is available in the modelling work byDee and Hanslow (2000).4 As previouslyexplained, impediments can be divided into thoseaffecting establishment and those affecting ongoingoperations. The barriers affecting establishmentare modelled as a tax on capital. Barriers tooperation apply to foreign owned firms who supplythe services sector in the host economy. They alsoaffect foreign firms who supply services acrossnational borders and also affect local firmssupplying their home market. These impedimentsare modelled as a tax on output.

There is a further two-way split betweenbarriers affecting access to the market whether bylocal or by foreign firms, that is, impediments toentry into services markets by whatever modewhich are non-discriminatory and barriers whichare discriminatory. Box 1 contains examples ofthe types of impediments that might be involved,with reference to accounting services.5

The results of simulations of various patternsof reform examined by Dee and Hanslow are

16

reported in Table 1. Four partial reformprogrammes are examined, reflected by the cellsof the table where the percentage refers to the sizeof the gain relative to the gain from comprehensivereform.

The data in Table 1 underscore thatwidespread reform is more efficacious than partialreform, since the returns to join reforms (the columnor row marked “both”) are greater than the sumof the parts. If a more specific choice is necessary,then the modelling results illustrate the followingpossibilities:

1. The largest gain is available from the reformof market access rules relating toestablishment, followed by market accessrules affecting ongoing operations.

2. Relatively large gains are available fromreform of all arrangements that violatemarket access compared to those whichaffect national treatment.

3. Reform of policies affecting bothestablishment and operations for eithermarket access or national treatment (thatis, down the columns) are much greaterthan the sum of the parts.

As Dee and Hanslow note, it is surprising thatthe returns to market access reform are so high,

given the relatively greater significance, in general,of the distortions to national treatment (evident inFigure 6). The problem is that removing distortionson national treatment in the presence of distortionselsewhere can shift resources in the wrongdirection. For example, this occurs when policyapplied to domestic firms does not change butextra distortions are removed from foreign firms.The foreign sector gets bigger, supply increases,market price falls and the domestic sector couldshrink even further.6

Furthermore, this set of results suggests thereis greater benefit in applying a reform programmethat affects a category of barriers (especiallymarket access) and applying the reform to allmodes of delivery (that is, moving down thecolumns in Table 1). The relative size of thesepayoffs is clearly an interesting topic to monitor asfurther modelling results are released.

C. Political economy

A familiar argument for a commitment touniformity is that it helps constrain the forces ofpolitical economy that will otherwise bias policychoices. As Tarr (1989) explains, in the contextof tariff policy:

“A uniform tariff conveys a number ofadvantages, the most important of which is

Table 1Partial reform in the services sector:

gains as a percentage of the gain from comprehensive reform

Marketaccess

Nationaltreatment

Both

Barriers toestablishment

43% 3% 48%

Barriers to ongoingoperation

20% 8% 29%

Both 76% 13% 100%

Source: Dee and Hanslow (2000)

17

that if the tariff is uniform, the gains to industrylobbying are much smaller (and may benegative), creating a kind of free-riderproblem for the lobbying industry anddramatically reducing the incentive to lobbyfor protection. Then: (1) the level of protectionis likely to be lower; (2) there is a directsaving of resources from the reducedlobbying; (3) the reduction to the gains fromlobbying for protection provides a vastlyimproved signal to valuable entrepreneurialtalent which will thus be encouraged to createbetter and cheaper products; and (4) thereduction in resources devoted to lobbyingwill result in less corruption in Government,which may have positive spillover effects intoother dimensions of Government activity.”

Impediments to services transactions, becauseof their quota characteristics, tend to create rentsfor incumbent suppliers. Barriers of the type usuallyused to protect the services sector will createprivileged positions for insider suppliers.Incumbents retain the revenue that might haveaccrued to Governments through a tariff, if onehad been applicable. Governments could capturethe rents in protected services by auctioning therights to supply, although this option is rarelychosen. The presence of these rents will strengthenthe incentive of incumbents to argue to retainprotection.

The “insiders” in these markets include notonly local firms but also foreign firms. Liberalizationof all modes of a services sector in a developingeconomy will create efficiency gains, reduce rentsearned by incumbents and also reduce rents earnedby foreign suppliers (through cross border tradeand establishment). Conversely, partialliberalization can have offsetting effects; for examplea removal of restrictions on establishment whileretaining restrictions on cross border supply (andalso impediments to market access by local firms)might simply transfer rents to foreigners and reducenational welfare.

A topic for further empirical work is toestablish the size of the rent transfer in variouscircumstances and to consider the implications of

the results for the design of reform strategies.7

The political economy of services negotiationsis more complicated than simply an alignment ofdomestic interests against the owners of foreigncapital. The current structure of protection canlead to the formation of alliances between domesticand foreign incumbents, both opposed to furtherreform.8 This situation is similar to that which arosein relation to quota protection in textile and clothingmarkets or where foreign investors haveestablished plants behind tariff barriers.

In the situations just described, there couldbe constraints from the political process to reform.At the same time, the interactions discussed earlierbetween goods and services markets, and amongstservices markets, create opportunities to mobilizepolitical support for reform. This support couldcome from those with interests in other marketsand who gain from reform, for example in marketsin which they buy inputs. Packaging of sectors, orclustering of activities, is also important for thesepolitical economy reasons. Overall, therefore, thepolitical economy considerations reinforce thecase on efficiency grounds for a uniform approachto policy making.

ENDNOTES

1 Some key results were recently summarized forthe OECD in a note by the Commission; see the OECDpaper TD/TC/WP(00)26. The World Bank has launcheda programme of research on trade in services, includinga data base on measures affecting trade in services whichcan be used in further quantitative work on their impact(see the World Bank’s note to the OECD TD/TC/WP/RD(00)7). UNCTAD is developing a data base of policymeasures called MAST.

2 The Productivity Commission reports data in astandard format. The original sources are McGuire andSchuele (2000) and Kalirajan, McGuire, Nguyen-Hongand Schuele (2000) for banking and Warren (2000a andb) for telecommunications.

3 For en early example of this kind of distinction inservices see UNCTAD and the World Bank (1994), chap-ters 4-7.

4 The OECD (2000e) has recently provided a re-view of a number of modelling projects of different types.

18

Here, the focus is on one of the modelling exercises whichuses the impediment measures described in the previ-ous part of this section.

5 For more detail, see White (1999) and Colecchia(2000)

6 This possibility is illustrated in Dee, Hardin andHolmes (1999). Dee and Hanslow point out that the re-sults summarized in Table 1 do not apply to all econo-mies but that for 13 of the 19 regions in the model, theremoval of market access restrictions on establishment

(the top left cell in the table) is the best partial reformoption.

7 Dee and Hanslow under extreme assumptionsabout the distribution of rents find significant lossesfrom services sector reform for the prominent home econo-mies of services sector foreign direct investment.

8 The distribution of rents earned by foreign andlocal suppliers in various policy scenarios is modelledby Francois and Wooton (1999b).

19

As previous work in UNCTAD has stressed,developing economies are offered newopportunities for gains from trade and investmentin services. There are a number of issues involvedin taking up those opportunities, including the termsof access to international markets for suppliers andinvestors from developing economies. However,market access at home also matters. Thediscussion in this paper has focussed on thecontribution of the openness of the domesticservices sector in developing economies to thecompetitiveness of the outward orientedproducers. For the purpose of achieving greatergains from a move towards a higher degree ofopenness, some principles of services sectorreform were stressed, including the value of a “top-down” approach, the importance of neutralitybetween modes and the value of bundling activities.Interactions between services reform and policydevelopment in other areas were also highlighted,especially the importance of policy in relation tocompetition and regulation. Impediments todomestic reform were noted, included the lack ofpolicy transparency and political economyconstraints. Recent empirical work on the patternsof impediments was reviewed. Some points onselecting policy priorities between initiatives relatedto market access and national treatment were alsomade, based on modelling results.

A theme of the paper is that the key to thedevelopment of coherent policy is to make moretransparent the effects of current policy and of thelikely impacts of alternatives. Some work onassessing services sector policy and assessing itsimpact has been completed and part of that bodyof work has been reviewed above. There aresome directions in which it would be useful toextend that work

A. Sample sizes

The first of these steps is to increase the sizeof the samples of economies in the empirical worklinking policy indices to market outcomes. Thedata sets are dominated by developed economies,since they tend to have more transparent policyregimes. Greater participation by developingeconomies will provide their policy-makers moreinformation about the nature of policy in theireconomy and its impact on other services sectors.It will help policy-makers establish priorities forattention, negotiation and commitment. Forexample, a tourism promotion strategy might haveas its first step national initiatives to reform the airtransport and telecommunications sectors. Thiswork will also facilitate developing economies’participation in the negotiations in WTO processes,or in regional arrangements, on market access.

B. Impact study coverage

Another target is to study the impact ofimpediments in a larger number of servicesactivities. So far, as noted already, this work ismore extensive in relation to banking andtelecommunications. Further work on transportsectors to yield country specific results would beuseful. Other sectors can be added to the list.These could include not only the professions anddistribution but also the areas nominated byUNCTAD for special attention. Education servicescould also be added to the list. Further work onimpacts will also help the design of clusters ofservices to be considered concurrently. 1

C. Sensitivities

Changes in the specifications of the policyindices could have significant effects on results of

IV. CONCLUSION

20

impact studies. Nevertheless, the sensitivity of theresults to these changes has not been testedextensively. Changes in weights or scoring systemscould be tested against a wider range of opinion,including business people with a knowledge of themarkets. The weights should also be comparedbetween the perspectives of developed anddeveloping economy business sectors. Thestatistical work could be extended to test forchanges in results, under different assumptionsabout the importance of the various elements ofpolicy.

D. Templates

A further possibility is to focus this activity topromote transparency by developing a negativelist reporting exercise. Hoekman and Martin(1999) observe that negotiators are not likely tochange the way that commitments are scheduledin the GATS to a negative list and that indeed suchan approach could be counter-productive. Insteadthey suggest that a less binding reporting processmight be possible. The process of constructingthe policy indices used in the work reported herewill facilitate the implementation of this reportingapproach. It will help the reporting of impediments,for example, in the form of model templates. Oneof the benefits of moving toward an index approachto describe a set of policies is that it forces theadoption of the “three c’s” - consistency,comparability and completeness into the collectionand recording of data. The weights and the scoringsystems applied to the calculation of the indicescan always be debated, although if these are notagreed, a key contribution of the overall indexapproach is its contribution to transparency throughthe “three c’s.”

E. Mode 4

Policy indices, constructed so far, recordimpediments at aggregate levels. Furtherdisaggregation is possible. A key interest ofdeveloping economies is to break downimpediments by mode of supply. Particularattention could then be paid to mode 4. Even usingthe existing data, this disaggregation could be

provided and the extent of impediments comparedacross policy areas.

F. Domestic regulation

A further impediment to reform is the set ofgenuine concerns about the sequencing and designof services sector reform. Some of these concernsrelate to the design of domestic regulation andwhether the appropriate regulatory structure is inplace. Aspects of the application of domesticregulation to the service sector were discussedpreviously.

The challenge is to design the right sequenceof change in reform policies, without losing themomentum of reform. A prescription that all theright policy institutions must be in place beforeliberalization occurs will only provide excuses to“go slow”.

Case studies of successful reform, in wholeeconomies or particular sectors, are one methodof transferring knowledge about the sequencingand institution building challenges in services sectorreform.2 This is particularly important in relationto the design of policy to deal with market failureproblems, such as lack of information amongconsumers about service supplier characteristics,or market power problems. There are importantspin-offs from this work for the GATS negotiationson issues related to domestic regulation.

ENDNOTES

1 Hoekman and Messerlin (1999) outline how thisapproach might be implemented.

2 The Pacific Economic Cooperation Council inconjunction with the APEC Studies Centres preparedtwo services sector studies, one on telecommunicationsin the Philippines and the other on the financial sector inPeru. These studies included details of the experiencesof particular firms that could be used to illustrate theimpact of reform and the business opportunities created.These studies can be downloaded from the site of theAustralian APEC Studies Centre. Chadha (1999) andTohamy (1999) report other interesting case studies ateconomy level.

21

REFERENCES

Adlung, Rudolf (1999a), “Liberalizing trade in services: from Marrakech to Seattle”, Intereconomics,September/October, pp. 211-222.

Adlung, Rudolf (1999b), “Adjusting to services trade liberalization: developed and developing countryperspectives”, paper presented to the World Services Congress, Atlanta, November 1-3, 1999(downloadable from www.worldservicescongress.com).

Anderson, Kym (1999), “Agriculture, developing countries and the WTO Millennium Round”, CIESDiscussion Paper No. 99.28, University of Adelaide (downloadable fromwww.adelaide.edu.au).

Chadha, Rajesh (1999), “GATS and developing countries: a case study of India”, paper presented tothe World Services Congress, Atlanta, November 1-3, 1999 (downloadable fromwww.worldservicescongress.com).

Colecchia, Alessandra (2000), “Measuring Barriers To Market Access For Services: A Pilot Study OnAccountancy Services”, in Findlay and Warren.

Deardorff, Alan V. (1999), “International provision of trade services, trade and fragmentation”, paperprepared for a World Bank Project on WTO 2000, October.

Dee, Philippa and Kevin Hanslow (2000), “Multilateral Liberalization of Services Trade”, ProductivityCommission Staff Research Paper, Ausinfo, Canberra http://www.pc.gov.au/research/staffres/multilatlib/index.html.

Dee, Philippa, Alexis Hardin and Leanne Holmes (1999), “Issues in the application of CGE models toservices trade liberalization”, in Christopher Findlay and Tony Warren (eds) (2000).

Findlay, Christopher and Andrew Watson (eds) (1999), Food security and economic reform: thechallenges facing China’s grain marketing system, Macmillan, London.

Findlay, Christopher and Tony Warren (eds) (2000), Impediments to trade in services: measurementand policy implications, Routledge, forthcoming.

Fink, Carsten, Aaditya Mattoo and Ileana Cristina Neagu (2000), “Trade in international maritimeservices: how much does policy matter?”, Development Research Group, The World Bank.

Francois, J. (1999), “Estimates of barriers to trade in services”, mimeo (quoted by Hoekman andMartin, 1999).

Francois, Joe and I. Wooton (1999a), “Trade in international transport services: the role ofcompetition”, mimeo (downloadable from www.intereconomics.com/francois/text9902.pdf).

Francois, J. and I. Wooton (1999b), “Market structure trade liberalization and the GATS”, paperpresented to the World Services Congress, Atlanta, November 1-3, 1999 (downloadable fromwww.worldservicescongress.com).

22

Gonene, Rauf and Guiseppe Nicoletti (2000), “Regulation, market structure and performance in airpassenger transportation”, Economics Department Working Papers No. 254, ECO/WKP(2000)27, OECD, Paris.

Gulati, A. (1998), “Indian agriculture in an open economy: will it prosper?”, in I.J. Ahluwalia and I.Little (eds), India’s economic reforms and development: essays for Manmohan Sing,London, Oxford University Press.

Hoekman, Bernard (1995), “Tentative First Steps: An Assessment of the Uruguay Round Agreementon Services”, paper presented for the World Bank Conference, The Uruguay Round and theDeveloping Economies, 26-27 January.

Hoekman, Bernard (1999), “The next round of services negotiations: identifying priorities and options”,paper presented at the Federal Reserve Bank of St. Louis conference on Multilateral TradeNegotiations: Issues for the Millennium Round, October 21-22, 1999 (revised 4 November1999).

Hoekman, Bernard and Will Martin (1999), “Some market access issues for developing countries in theMillennium Round; Results from Recent World Bank Research”, presented to the AnnualMeetings of the Latin American Economic Association, Santiago, October, 1999.

Hoekman, Bernard and Patrick Messerlin, “Liberalizing trade in services: reciprocal negotiations andregulatory reform”, paper presented to the World Services Congress, Atlanta, November 1-3,1999 (downloadable from www.worldservicescongress.com).

Johnson, Martin, Tendai Gregan, Geraldine Gentle, and Paul Belin (2000), “Modelling the benefits ofincreasing competition in international air services”, in Findlay and Warren.

Kalirajan, K. (2000), “Restrictions on trade in distribution services”, Productivity Commission StaffResearch Paper, Ausinfo, Canberra, August.

Kalirajan, K., Greg McGuire, Duc Nguyen-Hong and Michael Schuele (2000), The price impact ofrestrictions on banking services, in Findlay and Warren.

Marko, M. (1998), “An evaluation of the Basic Telecommunications Services Agreement”, CIESDiscussion Paper 98/09, Centre for International Economic Studies, University of Adelaide.

Mashayekhi, Mina (2000), “GATS 2000: progressive liberalization”, (DN; source to be identified).

Mattoo, Aaditya (1999), “Developing countries in the New Round of GATS Negotiations: towards aPro-Active Role”, The World Economy, vol. 23, no. 4, April 2000, pp. 471-491.

McGuire, Greg, Michael Schuele and Tina Smith (2000), “Restrictiveness of International Trade inMaritime Services’, in Findlay and Warren.

McGuire, Greg and Michael Schuele (2000), “Restrictiveness of International Trade in Banking Serv-ices”, in Findlay and Warren.

Nguyen-Hong D. (2000), “Restrictions on trade in professional services”, Productivity CommissionStaff Research Paper, Ausinfo, Canberra, August.

OECD (2000a), “Wholesale trade services”, TD/TC/WP(99)18/FINAL.

OECD (2000b), “Assessing barriers to trade in services: retail trade services”, TD/TC/WP(99)41/FINAL.

23

OECD (2000c), “Assessing barriers to trade in services: tourism services”, TD/TC/WP(2000)10/FINAL.

OECD (2000d), “Environmental goods and services: an assessment of the environmental, economicand development benefits of further global trade liberalization”, COM/TD/ENV(2000)86/FINAL.

OECD (2000e), “Quantification of costs to national welfare from barriers to services trade: a literaturereview”, TD/TC/WP(2000)24.

Pacific Economic Cooperation Council (1995), Survey of impediments to trade and investment inthe APEC region, APEC secretariat, Singapore.

Productivity Commission (1999), Trade and Assistance Review 1998-99, Annual Report Series 1998-99, Ausinfo, Canberra.

Sauve, Pierre (2000), “Developing countries and the GATS 2000 Round”, Journal of World TradeLaw, April.

Stephenson, Sherry (2000a), “A comparison of existing services trade arrangements within APEC”, inFindlay and Warren.

Stephenson, Sherry (eds) (2000b), Services trade in the western hemisphere: liberalization, inte-gration and reform, Brookings Institution Press, Washington, 2000.

Tarr, David (1998), “The design of optimal tariff policy for Russia”, World Bank, August(downloadable from www1.worldbank.org/wbiep/trade/documents/TARR-uniform-tariff.pdf).

Thompson, Rachel (1999), “Services liberalization to support electronic commerce”, paper presentedto the World Services Congress, Atlanta, November 1-3, 1999 (downloadable fromwww.worldservicescongress.com).

Tohamy, Sahar (1999), “Case study of Egypt’s service liberalization, service barriers and implementa-tion of the GATS agreement”, paper presented to the World Services Congress, Atlanta, No-vember 1-3, 1999 (downloadable from www.worldservicescongress.com).

UNCTAD (1999), Report of the expert meeting on air transport services: clarifying issues to define theelements of the positive agenda of developing countries as regards both the GATS and sectorspecific negotiations of interest to them, TD/B/COM.1/25.

UNCTAD (2000), Building confidence: electronic commerce and development, UNCTAD, Ge-neva.

UNCTAD and the World Bank (1994), Liberalizing International Transactions in Services: A Hand-book New York, United Nations.

UNCTAD secretariat (1998), “Scope for expanding exports of developing countries in specific servicessectors through all GATS modes of supply, taking into account their interrelationship, the role ofinformation technology and of new business practices”, TD/B/COM.1/21, July.

UNCTAD secretariat (1999), “Air transport services: the positive agenda for developing countries”,TD/B/COM.1/EM.9/2, April.

Warren, Tony (2000a), The Identification of Impediments to Trade and Investment in Telecommunica-tions Services, in Findlay and Warren.

24

Warren, Tony (2000b), The impact on output of impediments to trade and investment in telecommuni-cations services, in Findlay and Warren.

Warren, Tony and Christopher Findlay (2000), “Measuring impediments to trade in services”, in PierreSauve and Robert M. Stern (eds), GATS 2000: new directions in services trade liberaliza-tion, Centre for Business and Government, Harvard University and Brookings Institution Press,Washington DC.

White, Lawrence, J. (1999), “Reducing the barriers to international trade in accounting services; why itmatters and the road ahead”, paper presented to the World Services Congress, Atlanta, No-vember 1-3, 1999 (downloadable from www.worldservicescongress.com).

WTO secretariat (1998), “Presence of Natural Persons (Mode 4)”, Background note, S/C/W/75,December.

Zarilli, Simonetta and Colette Kinnon (1998), International trade in health services: a developmentperspective, UNCTAD, Geneva, 1998.

UNCTAD Study Series on

POLICY ISSUES IN INTERNATIONAL TRADE AND COMMODITIES

No. 1 Erich Supper, Is there effectively a level playing field for developingcountry exports?, 2001.

No. 2 Arvind Pangariya, E-commerce, WTO and developing countries, 2000.

No. 3 Joseph Francois, Assessing the results of general equilibrium studies ofmultilateral trade negotiations , 2000.

No. 4 John Whalley, What can the developing countries infer from theUruguay Round models for future negotiations?, 2000.

No. 5 Susan Teltscher, Tariffs, taxes and electronic commerce: Revenueimplications for developing countries, 2000.

No. 6 Bijit Bora, Peter J. Lloyd, Mari Pangestu, Industrial policy and the WTO,2000.

No. 7 Emilio J. Medina-Smith, Is the export-led growth hypothesis valid fordeveloping countries? A case study of Costa Rica, 2001.

No. 8 Christopher Findlay, Services sector reform and development strategies:Issues and research priorities, 2001.

No. 9 Inge Nora Neufeld, Anti-dumping and countervailing procedures – Useor abuse? Implications for developing countries, 2001.