Content uploaded by Johanna Jacobsson
Author content
All content in this area was uploaded by Johanna Jacobsson on Apr 02, 2018
Content may be subject to copyright.
GATS Mode 4 and labour mobility: the significance of
employment market access
Johanna Jacobsson1
Abstract
This chapter examines the concept and scope of the so-called Mode 4 which is employed in
the WTO’s General Agreement on Trade in Services (GATS) to regulate the movement of
natural persons providing services. Whereas previous research has mainly focused on the
temporary nature of such movement and on the different categories of persons covered, the
present chapter attempts to formulate a more conceptual understanding of Mode 4. Such an
understanding is inspired by the GATS Annex on Movement of Natural Persons Supplying
Services (the MNP Annex) according to which the GATS does not apply to measures
affecting natural persons seeking access to the employment market of a WTO Member.
Taking this demarcation as a starting point, the chapter analyses whether employment market
access (or rather, non-access) can be used to conceptualize Mode 4. The focus is thus on the
interaction of international trade rules with domestic employment and immigration laws. In
addition, the results of a brief empirical study on the biggest service importers’ practice are
presented. The results show that most of these WTO Members consider Mode 4 movement to
take place outside the national labour markets.
2
CONTENTS!
1. Introduction .......................................................................................................1!
2. What is Mode 4? ...............................................................................................5!
2.1. Historic and economic background ............................................................5!
2.2. Definition under the GATS ........................................................................8!
2.2.1. The GATS provisions .........................................................................9!
2.2.2. Mode 4 as a trade instrument ............................................................15!
2.2.3. The issue of MFN ..............................................................................18!
3. Employment market access as a defining criterion .........................................21!
3.1. The categories of persons covered by Mode 4 .........................................21!
3.2. Regulation of service contracts in the national sphere .............................24!
4. Mode 4 in WTO Members’ commitments ......................................................27!
4.1. Method of the study .................................................................................27!
4.2. Results of the study ..................................................................................30!
5. Conclusion .......................................................................................................34!
BIBLIOGRAPHY ...............................................................................................37!
1
1. INTRODUCTION
The present chapter analyses the scope and coverage of the so-called Mode 4 in the
international trade in services and especially within the framework of the World Trade
Organization (WTO). Instead of presenting Mode 4 as another form of labour migration, the
focus is on the conceptualization of Mode 4 as an instrument of multilateral trade
liberalization (Betts and Nicolaïdis, 2009).2 The analysis is built upon the issue of
employment market access that is argued to distinguish Mode 4 from traditional labour
migration. It is proposed that for employment market access to occur, two criteria need to be
simultaneously met. First, the host state’s labour laws must apply, and secondly, the foreign
worker must occupy, in the host state, a post that could be taken up by a local worker. The
normative conclusions are supported by an empirical study of the biggest service importers’
Mode 4 commitments.
Within the framework of the General Agreement on Trade in Services (GATS),
Mode 4 is one of the four modes under which services are traded between WTO Members.3
The GATS became part of the multilateral trading system as a result of the Uruguay Round
negotiations that culminated in the establishment of the WTO in 1995.4 Under Mode 4, a
service is supplied by a service supplier of a WTO Member in the territory of another WTO
Member through the presence of natural persons. The GATS applies to measures affecting
natural persons who are service suppliers themselves and to natural persons who are
employed by a service supplier. Both independent service suppliers and employees of service
suppliers are thus covered. In the second case, the employer is either a natural person (an
independent service supplier) or a service company. In order to benefit from Mode 4, the
natural persons and their employer must all originate in a WTO Member.5
2
The GATS does not set any time limits for the supply of a service under Mode 4
and the WTO Members remain free to regulate the issue as they choose. In most cases the
Members’ Mode 4 commitments are limited to a specific period of time. It has therefore
become customary to speak of Mode 4 as covering the temporary movement of service
suppliers (Mattoo and Carzaniga, 2003, p. 3; Stephenson and Hufbauer 2010, p. 31).
This chapter argues that the temporary nature of the movement of service
suppliers is only one element of Mode 4 and does not suffice for its conceptualization. Instead
of concentrating on the duration covered, the focus should be on the principal issue that is
necessarily pertinent whenever the movement of people is at stake. The natural question is to
ask if the GATS regulates labour migration, i.e. the inward movement (immigration) for
employment purposes. The answer characterizes the nature of the GATS as an international
agreement and significantly affects the types of migration and transactions it can cover. This
chapter takes the position that the question should be answered in the negative. The reasoning
behind this is principally based, first, on the text of the GATS and, secondly, on its nature as a
trade agreement.
The starting point is the GATS Annex on Movement of Natural Persons
Supplying Services (the MNP Annex) which provides some clarification on the scope of
Mode 4. According to the Annex, ‘the Agreement shall not apply to measures affecting
natural persons seeking access to the employment market of a Member, nor shall it apply to
measures regarding citizenship, residence or employment on a permanent basis’. This blunt
statement seems to carve out employment market access from the agreement’s scope.
However, nowhere in the GATS is it explained what employment market access means.
While making the scope of Mode 4 extremely vague, the ambiguity is understandable
especially for the following reasons. Firstly, the Annex strikes a delicate balance between the
interests of countries at different stages of development and with different economic profiles.6
3
Without being too restrictive, the carve-out protects the most developed countries against the
demands for the liberalization of labour immigration.7 Secondly, employment market access
is impossible to define in exact terms in an international trade agreement because it is a
concept that is dealt with under national employment law and thus differs from state to state.
Thirdly, because international service transactions can be carried out through different types
of contractual arrangements, the issue necessarily enters into the field of contract law and
private international (labour) law. No internationally harmonized codes exist in either
discipline.
Notwithstanding these inherent difficulties in defining Mode 4, this chapter
builds upon the issue of employment market access that the GATS itself places at the centre
of its definition.8 In the absence of any generally agreed definition for employment market
access, and the dependence on national employment and conflict of law rules, the chapter
proposes a construction with the help of which the issue could be analysed.
Employment market access is here understood to cover at least two issues.
Firstly, it means that the migrant worker enters the regulatory framework of the host state.
From the point of view of employment regulation, this entails that the labour laws of the host
state apply in their entirety. Secondly, employment market access implies that the foreign
worker is occupying, in the host state’s labour market, a post that might equally well be
occupied by a domestic worker. It is argued that for full employment market access to occur,
both criteria need to be fulfilled.
The first criterion is a legal one. Although it is regulated differently across WTO
Members, it is usually contractual in nature: the extent to which a service supplier enters the
regulatory framework of the host state depends on the nature of the service supplier’s
relationship with the host state entity that is acting either as an employer or as a client. A
work contract entails employment in the host state, whereas a service contract provides for a
4
private transaction with a wider freedom of contract. In a service contract there is no
employment relationship between the foreign service supplier and the host state client.
Instead, a work contract may exist between the service supplier and its employees. Such an
employment relationship is primarily subject to the laws of the sending state and the labour
laws of the host state apply to the workers only to a limited extent. How widely they do apply,
greatly affects the conditions under which the service supply can take place.
The second criterion compares a Mode 4 service supplier to a person employed
in the host state (a participant in the host state’s employment market). Any exhaustive
assessment of the effects of a foreign workforce on local workers requires an economic
analysis.9 However, another way to approach the issue is more analytical. For that purpose it
suffices to ask which Mode 4 categories involve persons performing activities that could
equally well be performed by local workers.
Both criteria are examined below. In addition, a short study on the biggest
service importers’ GATS commitments is presented. The results of this study show whether
the issue of employment market access is important to the WTO Members and how it is to be
understood based on their practice, in the light of the commitments they have taken under the
GATS.
When addressing Mode 4 type movement, the article refers to service mobility.
This term encompasses both natural persons acting as service suppliers and natural persons
employed by service suppliers (which, in turn, are either companies or natural persons
themselves). In contrast, the movement of natural persons who gain access to the host state’s
labour market and are thus employed there is referred to as labour mobility. This term
basically means the same as labour migration or labour immigration. The term mobility,
however, encompasses a wider scope of activities than migration. Whereas migration is
usually understood as a movement of an often permanent nature, the term mobility lays
5
emphasis on today’s reality of a mobile workforce and on the blurriness of the limits between
permanent and temporary types of work.
The chapter starts by briefly explaining the historical and economic background
of the GATS and analyses how Mode 4 is constructed in it. It then concentrates on the central
question of employment market access. The final, empirical section starts by explaining the
method used in the study on the chosen WTO Members’ Mode 4 commitments, after which it
turns to the analysis of the results. Both the empirical results and the arguments presented in
the chapter are drawn together in the conclusion.
2. WHAT IS MODE 4?
2.1 Historical and economic background
The scope and significance of Mode 4 is best understood in the light of the negotiations that
led to the conclusion of the GATS in 1994.10 Before the advent of the GATS, trade in services
was already regulated in certain bilateral and regional schemes, the focus of which was,
however, quite narrow (Marchetti and Mavroidis, 2011, p. 690). In the 1980s it became clear
that technological development was beginning to open foreign markets to services more
widely. The United States in particular saw the opportunities that a multilateral trade
agreement on services could create.11
The GATS was negotiated at the end of the 1980s and the beginning of the
1990s, at a time when the importance of the services sector was becoming increasingly
evident. By that time services already accounted for over 60 per cent of the gross domestic
product (GDP) of Organisation for Economic Co-operation and Development (OECD)
countries.12 Even though there was a general understanding of the significant economic gains
6
that could be attained through liberalization of services, the challenge was to draft an
agreement that would meet the expectations of countries at different stages of development.
Several developing countries at first refused to negotiate on services for fear of having to
open their vulnerable services markets to foreign competition. In the course of the
negotiations they also became concerned that their own comparative advantage, abundant
labour, would not be adequately addressed in a services agenda that was mainly focused on
the liberalization of the financial sector and investment. For reasons of parity, certain leading
developing countries thus pushed for the inclusion of labour movement in the GATS (Dey,
2007, p. 84; Bast 2008, p. 576). The results of the negotiations only partially met the demands
of these countries. Whereas, in principle, the GATS allows for the movement of all categories
of natural persons, in practice, the Members have liberalized the movement of very specific
types of professionals only.
Notwithstanding the significant economic potential that freer movement of
service suppliers holds, many authors have drawn attention to the extremely modest level of
liberalization that was reached under Mode 4 in the Uruguay Round (Dey, 2007, p. 88; Bast,
2008, p. 576; Self and Zutshi, 2003, p. 27; Chanda, 2001, p. 639). Very few WTO Members
have taken any significant commitments under Mode 4 and the existing commitments mainly
relate to the movement of intra-corporate transferees (most often executives, managers and
specialists), business visitors and highly-skilled self-employed persons. Moreover, Mode 4
commitments generally do not even bind the status quo but often reflect stricter entry
conditions than the access granted in practice (Carzaniga, 2003, p. 21). Considering that the
GATS commitments were made in the mid-1990s, the difference between bound and applied
entry conditions is likely to have become wider since then. Taking into account the modest
level of liberalization achieved so far, it is not surprising that trade in Mode 4 represents less
than 5 per cent of overall services trade (Kelsey, 2010, p. 18).13
7
It can be argued that, for the sake of increased relevance, the WTO Members
should extend their Mode 4 commitments to include temporary employment with nationally-
owned host state employers.14 The supporters of this point of view emphasize the economic
gains stemming from the liberalization of labour migration and the difficulties in
distinguishing service contracts from employment contracts.15 In this chapter it is proposed
that the placing of Mode 4 outside the ambit of labour mobility follows the logic of a trade
agreement and can lead to benefits that participation in the local employment market would
not offer. There is a growing need to efficiently transfer expertise and labour force
internationally, either through the temporary relocation of specialists or for the purposes of
contractual service supply.16 Even though such transactions do not replace traditional
immigration for labour purposes, they can increase gains and bring additional income to both
sending and receiving states. A more practical consideration is that it would be unrealistic to
expect OECD countries to open their markets to labour mobility in a multilateral, legally
binding agreement. Instead of arguing that WTO Members should do this, it would be more
productive to call for deeper commitments that remain purely within the sphere of services
trade.17
A significant impediment to a more liberalized Mode 4 is the lack of incentive.
Instead of taking binding commitments on a multilateral basis in accordance with most-
favoured nation (MFN) treatment,18 WTO Members have the possibility to do so unilaterally
or bilaterally by granting access to certain nationals only. It seems that even if access to local
employment markets is ruled out, the requirement of MFN creates an issue of trust. Unilateral
market opening or specifically negotiated bilateral migration schemes have the advantage that
they can be easily reversed and are concluded with certain countries only. Such arrangements
provide the flexibility that is often needed for the regulation of movement of people.19
8
Mode 4 does, however, have certain advantages over bilateral migration
agreements. Being part of a trade deal, Mode 4 commitments can be used as a trade-off for
other goals: the commitments are part of a wider negotiation agenda and hence contribute to
the attainment of a quid pro quo (Panizzon, 2010; Trachtman, 2008). Moreover, a trade
agreement can be an easier instrument to use politically than an agreement liberalizing
immigration outright.
Multinational companies and developing countries share an interest in having a
more open Mode 4. Whereas the latter wish to increase the movement of their nationals in
ways unrelated to commercial presence abroad, multinational companies would like more
scope for the international movement of their personnel (related to Mode 3) (Mattoo and
Carzaniga, 2003). Moreover, both big and smaller companies are increasingly interested in
deploying foreign personnel for short, specific projects with tight deadlines (Chaudhuri et al.,
2004). An attractive prospect for these companies would be the possibility to bypass
frustrating visa procedures and employer’s obligations. Considerable economic gains have
been demonstrated to be achievable even by modest liberalization of Mode 4 if the relatively
abundant medium-skilled and less-skilled workers from developing countries were allowed to
provide their services in developed countries (Winters, 2008, p. 480). In the developed parts
of the world, temporary access is often preferred to more permanent types of immigration due
to the lower social and political costs involved. In developing countries, on the other hand,
temporary outward movement can lead to less brain-drain than permanent emigration (WTO,
1998, p. 6).20
2.2. Definition under the GATS
9
2.2.1 The GATS provisions
Article 1 of the GATS defines trade in services under Mode 4 as the supply of a service by a
service supplier of one Member, through presence of natural persons of a Member in the
territory of any other Member. In addition, the MNP Annex states that the Agreement does
not apply to measures affecting natural persons seeking access to the employment market of a
Member, nor to measures regarding citizenship, residence or employment on a permanent
basis.21
The GATS does not define the concepts ‘access to the employment market’,
‘residence’ and ‘employment’. Whereas the exact scope of all GATS modes of supply is
somewhat open to interpretation, in the case of Mode 4 the lack of clarity culminates in a very
fundamental issue. Based on the reading of the GATS provisions dealing with Mode 4, it is
not entirely clear to what extent measures pertaining to the field of labour mobility were
intended to be covered.
The obscurity is principally due to two separate provisions of the GATS. First, it
is sometimes claimed that Article 1:2(d) read together with the first paragraph of the MNP
Annex leaves it unclear whether employment in the service of a host state employer can be
considered part of Mode 4.22 Secondly, the second paragraph of the MNP Annex has given
some commentators reason to infer that only permanent employment in the host state is
excluded.23 They have, a contrario, concluded that temporary employment in the host state
would be covered by Mode 4.
Some backing for such a proposition can be found in certain WTO Members’
GATS schedules. The most notable example is the Mode 4 commitment of the United States
that provides on an annual basis for the temporary employment of 65 000 persons in specialty
occupations (the so-called H1-B visa). Moreover, a significant number (43) of mostly
10
developing Members have included in their schedules a versatile group of executives,
managers and specialists (EMSs) without specifications as to the nationality or location of
their employer (WTO, 2009, p. 23). Section 5 of this chapter will show that in the case of the
biggest service importers, no access to the host state employment market is allowed in the
majority of their commitments.
A combined reading of the MNP Annex and Article I:2(d) can be seen as
resulting in the coverage by Mode 4 of the following categories of persons:
1) foreign self-employed persons supplying services to host state companies of
individuals (independent professionals, IPs)
2) employees of a foreign service supplier who are temporarily transferred to the
supplier’s commercial presence in another Member (intra-company
transferees, ICTs)
3) employees of a foreign service supplier who enter the host state to supply a
service pursuant to a service contract between their employer and the host-
country client (contractual service suppliers, CSSs) and
4) business visitors (BVs) and services salespersons who seek entry for the
purpose of setting up a commercial presence or negotiating the sale of a
service (WTO, 2009, p. 6; Carzaniga, 2008, p. 477).24
As already mentioned, some Members have also included in their schedules the
group of executives, managers and specialists (EMSs) who appear on some occasions to
overlap with ICTs and on other occasions to provide for the possibility of entry to the host
state employment market. Notwithstanding these differences in the Members’ practice and
certain scholars’ desire for an extended coverage of Mode 4, the majority view among
11
commentators appears to be that host-country employers are not covered.25 Self and Zutshi
(2003, p. 34) point out that during the GATS negotiations the issue of employment of
foreigners by host state employers did not even come up for a detailed examination because
access to the employment market was understood to be outside the coverage of the agreement.
The most common conceptualization of Mode 4 is built around its temporary
character. The GATS, however, does not set any time limits and the definition of period of
stay is left to the discretion of the Members.26 A look at the Members’ schedules reveals that
there is considerable variety in the periods of time covered: they extend from a couple of
months to several years. Bilateral migration schemes, such as temporary worker programmes,
usually also limit the maximum period of employment and they are categorically
conceptualized as providing temporary employment in the host state.27 Trade agreements,
however, do not follow an entirely similar logic to that of migration agreements; it is thus
questionable whether the most accurate conceptualization of service supply under a trade
agreement should be built similarly to migration schemes.
According to the MNP Annex, employment market access is the factor that
distinguishes labour mobility (labour immigration) from service mobility (Mode 4).
Employment market access would entail employment with host state entities. Even though the
majority view is that domestic employment, and thus labour mobility, is not covered, making
the distinction between service and labour mobility is largely neglected in the scholarly
analysis of Mode 4.28 This neglect results from difficulties in distinguishing between the two
phenomena, as their boundaries vary between the different legal cultures. Furthermore, the
distinction tends to be further blurred by economic analyses that do not take the conceptual
differences between labour and service mobility into account.29
Analyses of Mode 4 usually focus on the formation of an employment
relationship between the Mode 4 entrant and the service recipient. According to Bast (2008),
12
‘the decisive element in distinguishing employment from other forms of personal service is
the degree of freedom from the instructions of a superior since a substantial degree of
independence is usually seen as a typical feature of self-employment’.30
The analysis by Bast is correct. Based on a case-by-case analysis, and in
accordance with the host state law, it helps to establish whether a specific situation is covered
by Mode 4 or not. There are, however, practical considerations which limit the usefulness of a
blunt differentiation between work and service contracts. For these reasons, the analysis
should be taken further. There are two groups that require specific attention. The first are
intra-company transferees (ICTs) who appear in most Members’ schedules.31 ICTs are a
special case in the sense that they are employed in the host state. Their employer, however, is
a foreign-owned entity whose presence in the host state is often linked to Mode 3.32 If one
were to distinguish Mode 4 entrants from labour migrants purely based on their type of
contract, ICTs would necessarily have to be categorized as labour migrants because they are
employed in the host state.
A second complex category of Mode 4 entrants are contractual service suppliers
(CSSs). So far they appear in only a few schedules, the EU being the most prominent
example. Nevertheless, the biggest potential for further liberalization of Mode 4 lies in this
category (Winters, 2008, p. 519). This would be the case especially if WTO Members opened
their markets to less-skilled contractual workers in sectors such as construction, health-care or
any other sector where there is significant labour input. In these sectors, however, there is a
strong prospect for the contractual relationship between the Mode 4 entrant and the host state
client to become blurred. To stay in the field of Mode 4, the person performing the service
should not become a subordinate of the host state entity but should retain a strong degree of
freedom and take regular instruction only from the employer based in the sending state.
13
In fields where CSSs work closely with the host state entity, the maintenance of
independence can be problematic. In the construction field, for example, part of a project can
be outsourced to a foreign service supplier. When the employees of the foreign service
supplier work side-by-side with local workers and under the supervision of the same master
builder, they become easily identifiable with the locally employed workers. Even if formally
in a contractual relationship, such workers may be categorized as labour migrants by local
authorities. This can be detrimental to their status as Mode 4 entrants. A similar problematic
situation can arise in the case of independent professionals (IPs). IPs, however, are usually
highly-skilled professionals, such as architects and lawyers. Since they usually perform their
services independently, the formation of an employment relationship is avoided. In the case of
less independent IPs, similar problems to those faced by CSSs may arise.
The examples of ICTs, CSSs and IPs demonstrate that it is problematic to use
the type of contract as the main criterion for distinguishing Mode 4 entrants from more
traditional labour migrants. Even if the distinction is legally correct, it leads to situations
where further liberalization of Mode 4 would be challenged due to its identification with
labour immigration. To keep Mode 4 in the field of services and within the scope of the
GATS, Members should aim to construct their commitments in a way that does not entail the
formation of an employment relationship with a host state entity.
To help in this endeavour, additional criteria for defining ‘employment market
access’ in the GATS context could be used. Special attention should be paid to the cross-
border element of Mode 4 that is present in all service trade. The cross-border element
separates service mobility from labour mobility. Whereas employment-based work
performance is regulated by host state norms, in service contracts the employment
relationship remains under the regulation of the sending state. Even if host states sometimes
extend the application of their most important labour norms to foreign workers performing
14
services on behalf of their home state employers, the home state rules on the employment of
the workers continue to apply. One can take the example of CSSs and IPs. If they retain a
sufficient degree of independence from their customers, they are usually not considered to be
in employment in the host state. This means that they continue to be subject to the rules of the
sending state, even if certain mandatory host state rules, for example on minimum pay, are
often applied.
The first aspect specific to Mode 4 entrants thus relates to the regulatory
framework in which these entrants operate. The second aspect is the competition with local
workers. Employment market access can occur only in cases where the Mode 4 entrant is
occupying a post that could potentially be filled by a local worker. With CSSs and IPs this is
often the case since most services could also be provided by domestic workers. In the case of
ICTs, however, the situation is different. When they are required to possess knowledge
specific to the foreign company, no local worker is replaced.33 The situation is different when
ICTs are simply categorized as experts or managers with no special connection to any foreign
entity belonging to the same group of companies as the receiving entity in the host state. If
they are in an employment relationship with the group’s host state entity and their job could
be performed by a local worker, under the criteria presented here, employment market access
can be said to occur.
As the examples of CSSs, IPs and ICTs show, both criteria put forward here
need to be simultaneously met for employment market access to occur in a specific case. In
the case of CSSs and IPs, attention should be paid to the applicable regulation. Where closer
links to the sending state prevail, no employment market access occurs even if the job could
be performed by a local worker. In the case of ICTs, a conceptual differentiation between
labour entrants and Mode 4 entrants relates to their capacity to provide expertise that could
not be easily supplied by local workers.
15
Before going further into the issue of employment market access, it is necessary
to explain why the issue is important and why employment market access (or rather, non-
access) should be seen as the core criterion in the definition of Mode 4. Two issues arise in
particular: trade logic and MFN. Each will be dealt with separately below.
2.2.2 Mode 4 as a trade instrument
Mode 4 suffers from a lack of clear conceptualization. This is evident both in the light of the
Members’ schedules and the literature. The absence of generally agreed definitions makes it
hard to distinguish between different categories of Mode 4 and to compare them across the
Members’ commitments. Particular terms, such as specialists and executives, and open-ended
notions, such as economic needs tests, are not used consistently even by individual Members.
The problems in classification and in determining the exact scope of Mode 4 allow for
administrative discretion and loss of relevance (WTO, 2009).
The main challenge is the location of Mode 4 in the middle of trade and
migration, due to which it is often poorly coordinated between the authorities administering
the two fields. Even though it is part of a trade instrument, the fate of the persons entering a
country under Mode 4 is in the hands of immigration authorities. Even though some countries
have special procedures for entrants under Mode 4 (for example, the GATS visas in the UK
and Australia), service suppliers may need to be fitted into immigration categories designed
for employment-based movement.
The variety of actors involved is thus a challenge (Betts and Nicolaïdis, 2009).
Another factor blurring the distinction between labour and service mobility is that developing
country Members of the WTO, as well as several commentators, are pushing for further
liberalization of labour mobility. Bringing labour mobility under the GATS would make any
16
commitments on immigration binding and subject to compulsory WTO dispute settlement.
This would greatly increase the relevance of Mode 4 as a migration instrument considering
that there are no other legally enforceable migration agreements (Trachtman, 2009).
It is, however, hard to see many countries being willing to tie their labour
migration schemes to a multilaterally fixed level enforced with MFN. The overemphasis on
Mode 4 as a labour migration scheme actually limits its use in trade negotiations: Mode 4
already risks becoming a non-trade issue in some countries. For example, in the US, as a
consequence of the opposition of the Congress, no preferential trade agreement negotiated by
the US since 2002 has contained a chapter aimed to facilitate the movement of natural persons
(Stephenson and Hufbauer 2011, p. 282).34 In the EU, on the other hand, the Union’s
exclusive competence to conclude trade agreements does not extend to agreements
liberalizing immigration. Agreements entering into the field of labour mobility would thus
need to be signed and approved separately by each Member State, which otherwise is no
longer a necessity in services trade.35 This would mean that any national parliament that was
discontented with the immigration aspect of the trade deal could veto it in its entirety
(Bungenberg, 2010, p. 132).
Limiting the scope of Mode 4 to service mobility would not eliminate
opposition but could increase the tradability of Mode 4. Moreover, such a limitation more
closely follows the logic of a conventional, cross-border trade agreement. In an international
trade transaction a sales contract is concluded between economic actors in (at least) two
different countries. In labour mobility the economic actor (the migrant) moves to the
regulatory framework of the host state and no formal tie with the country of origin necessarily
remains. The economic benefits of such a move accrue primarily to the migrant himself or
herself. Naturally, the migrant’s family and the country of origin often benefit in the form of
remittances, but such transfers are subject to individual discretion.36
17
It is worth emphasizing that the GATS is an agreement that provides for the
liberalization of trade in services between different countries. Trade in services under Mode 4
is defined as the supply of a service by a service supplier of one Member, through presence of
natural persons of a Member in the territory of any other Member. In the case of natural
persons acting as employees of a service supplier, the service being liberalized is the one
supplied by their employer, not by themselves.37 If domestic service suppliers (other than
foreign-owned entities established under Mode 3) could also appear as employers, there
would be no trade in the sense of the GATS.
For the sending state it is relevant that the service supplier and its employees
remain in the field of application of home state regulation. In this case, the employees’ work
performance, as well as the value created by the service supplier, can be subjected to home
state taxes and social security contributions.38 If the service supplier or its employees instead
enter the host state employment market, these connections to the sending state are in most
cases lost. It is hard to see what is left to be considered as trade in such circumstances.
It is important to note that Mode 4 entrants are often subject to the host state’s
minimum wage and certain other labour standards.39 Low-wage countries are thus, to a certain
extent, stripped of their comparative advantage: being subject to mandatory rules imposed by
the host state, means that they cannot necessarily benefit from their home state’s more lenient
regulation. Lower wages are, however, only one factor that encourages companies to engage
in cross-border services trade. Gains in productivity, and especially the availability of persons
with different levels of skills, are often more relevant considerations.40 Significant benefits to
both sending and receiving countries could be attained especially if the movement of less-
skilled persons was facilitated. Developing countries could benefit more were they to pursue
deeper Mode 4 commitments that would remain strictly in the field of services trade but
would be extended to cover categories of less-skilled workers. Ideally, developed countries
18
would also open their labour markets to more entrants from developing countries. In the
context of a trade agenda, to strive for employment market access can, however, put at risk
the attainment of a more liberal Mode 4 regime overall.41
2.2.3 The issue of MFN
Another consideration in drawing the limits of Mode 4 stems from the need to avoid clashes
with bilateral and regional migration schemes. There are numerous bilateral and regional
agreements, in the field of both trade and migration, which provide for different types of
movement of labour.42 If Mode 4 of the GATS was considered to cover labour mobility, the
signatories to such agreements would risk violating one of the main cornerstones of the
GATS, the most-favoured nation obligation (MFN).
The MFN principle, enshrined in Article II of the GATS, requires the following:
With respect to any measure covered by this Agreement, each Member shall
accord immediately and unconditionally to services and service suppliers of any
other Member treatment no less favourable than that it accords to like services
and service suppliers of any other country.
Thus, WTO Members have to extend the treatment they afford to a service
supplier of any country to all similar service suppliers originating in other WTO Members.
The reach of the MFN can be avoided through preferential trade agreements
(PTAs), labour markets integration agreements (LMIAs), MFN exemptions and waivers.
Article V of the GATS regulates the conditions for PTAs in the field of services. It requires
overall liberalization of services trade and does not apply to agreements limited in their scope.
19
Most labour mobility schemes would thus not qualify as PTAs under Article V GATS. Very
few of them would qualify as LMIAs either. LMIAs, regulated under Article V bis of the
GATS, exempt the citizens of the parties from requirements concerning residency and work
permits, and typically provide them with a free entry to the parties’ employment markets.
LMIAs thus require full liberalization of labour markets between the contracting parties.43
An MFN exemption, on the other hand, releases the Member from applying
MFN as far as a specific measure has been prescribed as an exemption in the Members’
schedule. All MFN exemptions had to be notified before the entry into force of the GATS and
new ones cannot be added.44 Moreover, according to the GATS Annex on MFN exemptions,
such exemptions should not, in principle, exceed a period of ten years. The availability of
MFN exemptions is, therefore, limited.45
The last option for deviating from the MFN obligation is to apply for a waiver.
A waiver is a permission granted by all WTO Members for a certain WTO Member not to
comply with its normal commitments. Waivers are difficult to obtain and they are limited in
time. A waiver has been successfully adopted with regard to the least-developed WTO
Members (the so-called LDCs). At the eighth WTO Ministerial Conference of 2011 the
Members adopted a waiver that provides for a departure from the MFN principle and allows
Members to give preferential treatment to services and service suppliers of LDCs.46
Favourable treatment of LDC service suppliers, however, depends on the willingness of the
other Members to make preferential concessions in this regard.
At the conclusion of the Uruguay Round in 1994 certain labour mobility
schemes were inscribed as MFN exemptions, most notably by European countries towards
their former colonies and with regard to traditional recruitment areas.47 Some scholars claim
that the presence of such MFN exemptions demonstrates that WTO Members understood
labour migration to be covered by Mode 4 since otherwise they would not have included
20
them. They thus conclude that bilateral labour mobility schemes are covered by the WTO
disciplines (especially by MFN) and consequently several of them appear to be clear
violations of WTO law.48
This is not a viable position. If labour mobility (employment market access) was
part of the scope of the GATS, WTO Members could no longer favour labour immigration
from certain countries only. It is difficult to believe that the purpose of the GATS negotiators
was to render all labour mobility schemes, including those between countries with especially
close cultural ties, potential violations of WTO law.
It is more likely that the MFN exemptions were inscribed for other reasons,
some of which were probably purely political. An easily understandable reason is that the
scope of Mode 4 was unclear from the beginning and several Members felt that in the
presence of doubt, it was better to protect themselves against accusations of MFN violations
(even though this was only for a period of ten years). Another completely rational reason may
be the degree of overlap between service and labour mobility: many labour mobility schemes
facilitate the movement of service suppliers as well. By exempting the entire arrangement, the
Member can preserve the preferential arrangement for service suppliers originating in a
specific country without violating its MFN obligation towards other Members.49
It is worth noting that the GATS does not prevent Members from regulating the
entry of service suppliers from different countries.50 In the case of two similar suppliers, the
main distinguishing feature being their origin, the MFN obligation, however, prevents
favouring one of them with respect to the supply of a service. Were the MFN discipline to be
extended in the field of labour mobility, unexpected consequences would follow as the same
treatment would need to be extended to all similar labour migrants originating in WTO
Members and entering service sector activities covered by the GATS.
21
It is also important to note that the MFN obligation is binding with regard to all
services covered by the GATS, irrespective of a Member’s commitment to liberalize a
specific sector. Schmitz takes the example of a Ugandan lawyer who has legally entered the
EU and cannot consequently be discriminated against in favour of a lawyer from any other
Member with respect to the provision of legal services, even if no commitment in legal
services exists.51 Similarly, immigration and visa policies are covered by the MFN obligation
even if no specific commitment has been taken.52
3. EMPLOYMENT MARKET ACCESS AS A DEFINING CRITERION
3.1 The categories of persons covered by Mode 4
It has been proposed that employment market access under the GATS necessitates two
separate criteria: full application of the host state’s employment regulation and occupation in
the host state of a post that could be taken up by a local worker. Both of these criteria need to
be fulfilled before a foreign worker can be said to have entered the local employment market.
The first criterion is especially complex since it is at the crossroads of both
(national) public and private law and public and private international law. It points to the fact
that the type of contract the foreign service supplier is engaged under cannot be disregarded
since, depending on the national legislation, it can bring about very distinct regulatory
outcomes. The principal difference is that while host state employment contracts are in the
field of application of local labour laws, the employment aspects of service contracts are to a
certain extent beyond their reach. A necessary condition for a service contract is that its
connections to the host state are limited. That goes hand in hand with the assertion that no
22
employment market access occurs. The two regulatory outcomes are analysed in more detail
in the following sub-section which concentrates on the question of conflict of laws.
The consideration of the second criterion of employment market access,
replacement of a local worker, is here more limited. For the purposes of the conceptual
analysis of Mode 4, it is relevant to ask which Mode 4 categories involve persons performing
activities that could potentially be performed by local workers.
Business visitors (BVs) are most clearly performing tasks that cannot easily be
done by local workers. BVs represent themselves or their employers and go on short visits to
prepare the establishment of a subsidiary or to negotiate a trade deal. They are the group of
service suppliers that are most clearly outside the application of host state labour laws. ICTs,
by contrast, are employed by local entities and are therefore usually covered by the host state
labour laws. They do not, however, enter the local employment market because their place of
work cannot be occupied by a local worker. As Bast (2008) points out, ICTs cannot, on the
basis of Members’ GATS commitments, present themselves to other potential employers in
the host state. At the point of entry, they are already employed by the foreign entity
established in the host state.53
CSSs can most easily be seen as replacing local workers. Their movement is not
related to Mode 3, as in the case of ICTs, and their service supply is usually much closer to a
work performance than is the case for BVs. They are engaged in the type of projects that can
potentially create jobs for local workers. Maybe for these reasons few Members have so far
facilitated the movement of CSSs and, where they have done so, it is limited it to a few
sectors. In 2009, the number of such Members was 31 (including all EU Member States).54
Since CSSs’ employers compete directly with domestic suppliers, the liberalization of the
movement of CSSs carries similar types of risks to the opening of domestic labour markets.
Although conceptually a separate regime, the risk that CSSs pose to domestic workers can be
23
considered even higher than in the case of traditional labour immigration since foreign
suppliers may, in certain cases, benefit from differences between their home state’s and the
host state’s operating environments (OECD et al., 2004, p. 85).
Certain Members have scheduled a category of EMS (executives, managers and
specialists) whose movement is not limited by an obligation of prior employment in the
sending state. When employed by a host state entity and with no special link to a foreign
establishment, in the case of such employees, as in the case of the US H1-B visa holders,
employment market access can be seen to occur.
Some Members use quotas or economic needs tests (ENTs) to make sure that
service suppliers are not replacing local workers. In the case of Mode 4, ENTs are usually
applied in the form of labour market tests which allow entry only when it is not expected to
have a negative impact on the domestic labour market. ENTs therefore generally establish as
the sole criterion the non-availability of suitably qualified persons in the local employment
market (WTO, 2009, pp. 14, 22).
In the case of ICTs, a few Members have inscribed quotas or ENTs, concerning
mostly specialists. Specialists can potentially replace local employees where the required
expertise is very general in nature. The requirement of ENTs is thus understandable from the
point of view of avoiding employment market access. In the case of BVs, only two Members
impose ENTs. No quotas are applied. For CSSs and IPs the use of quotas and ENTs is rare,
even though by definition CSSs and IPs can win contracts that could create jobs for local
workers. The requirement of an ENT would, however, make the relevant commitment largely
obsolete since it would not create any opportunity for foreign service suppliers to compete
with local providers. Access would be granted only if no domestic service suppliers were
available.55
24
The examples of the various categories covered by Mode 4 demonstrate that
both criteria of employment market access used here have to be met simultaneously. It is
worth stressing that the absence of employment market access does not mean that the
presence of foreign service suppliers, especially CSS and IPs, in the host state does not have
any implications on local jobs. Instead, the purpose is to show how the issue of employment
market access can be used to conceptually differentiate Mode 4 from labour mobility.
3.2 Regulation of service contracts in the national sphere
Typically the regulatory outcome differs depending on whether the service is supplied
through a work contract or a service contract.56 From a legal point of view, the basic
distinction is that a work contract is regulated by labour laws, many of which are mandatory,
whereas a service contract is in the field of private law. Another important distinction can be
made based on the choice of law. In cross-border transactions the parties usually select the
law applicable to the contract. In employment contracts the choice of law is typically more
restricted. Where there is no cross-border element involved, all compulsory employment
norms of the state of employment usually apply. For example, in cases where a work contract
is concluded between a host state employer and a foreign worker employed in the host state,
the work contract is regulated by host state labour laws notwithstanding the foreign origin of
the worker. In service contracts the contractual relationship is between the client and the
service supplier. The employment relationship, on the other hand, is between the service
supplier and its employees. The work contract is thus subject to the rules of the sending state.
In many legal systems it is accepted that an employee temporarily sent abroad
by his or her employer remains subject to the law of the home state. Several regional and sub-
regional social security conventions establish a similar principle. There are, however,
25
variations regarding what is considered temporary. Some social security conventions, for
example, limit the application of the state-of-origin rules to six months or one year with a
possibility of extension. There is, however, great variety in how countries treat contracts with
foreign elements. Each state’s choice of law rules finally determines which country’s law is
applied.57
Even in situations where foreign service suppliers are performing under a
service contract, the host state has an interest in ensuring that the most crucial norms of the
country are adhered to. Such norms can be referred to as mandatory or overriding rules58 and
they apply for public policy reasons even in situations where a foreign law would otherwise
apply (Hepple, 2005, p. 155).59 In the field of employment such generally applicable rules can
relate to crucial issues such as minimum wage, working hours and safety at work. Even when
no employment relationship exists in the host state, the employees of foreign service suppliers
may be subject to such overriding rules of the host state. The basic feature of Mode 4,
however, is that the law primarily applicable to the employment relationship is that of the
home state.
For a trade scholar, employment law and private international law can be
unfamiliar areas. Labour laws, as well as choice of law rules, are not issues dealt with by the
WTO. It is, however, important to keep in mind that trade in services is crucially different
from trade in goods. Barriers to services are not tariffs but instead take the form of
regulations. To understand how to liberalize services, one needs to understand regulations.
Therefore, it is crucial to determine whose regulations apply. Certainly, it would be unrealistic
to expect WTO Members to come to a joint agreement on the issue. It suffices to look at the
EU where the Member States did not manage to agree on any clear-cut rule for the choice of
law in the long-awaited Services Directive adopted in 2006.60
26
Instead of aiming to adopt any WTO-wide choice of law rules, WTO Members
could look at Mode 4 from the perspective of their own national laws and formulate their
commitments in a way that would be in line with the GATS as a trade instrument. Naturally,
nothing would prevent Members from making commitments that would provide for
employment market access. Such commitments would, however, be so-called WTO+ issues
that are outside the scope of the GATS.61
Countries interested in further liberalization of Mode 4 should do a careful
analysis of the types of migratory movements that are most beneficial to them.62 Remittances
sent home by their nationals who are immigrants in richer countries are attractive but they do
not necessarily bring similar, long-term and widely distributed benefits to those to be gained
from a more developed services economy.63
At present, the most successful bilateral migration schemes are considered to be
those that are holistic in their approach. In addition to providing adequate protection and
benefits to the workers, they enforce temporariness and facilitate formal channels for
transferring remittances and for directing them towards productive investments in the sending
state (Chanda, 2009). Their aim is to make sure that the state of origin and its society as a
whole benefits from its nationals working abroad, not just the migrant and the receiving state.
Similar perspectives should be adopted with respect to Mode 4. It is in the interests of labour-
abundant countries to develop such institutional and practical mechanisms that would help in
persuading other WTO Members to let in those countries’ service suppliers.64 Nothing
prevents countries from simultaneously aiming at labour mobility agreements. Keeping Mode
4 in the field of trade would, however, bring different sources of income that would contribute
to the development of a services economy.
Trachtman (2008) notes that it is highly unlikely that a one-size-fits-all approach
to migration liberalization would suit all states. Neither would it be consistent with
27
maximizing global welfare. According to him, the WTO negotiation procedure with request–
offer-type negotiations seems attractive since it is country-specific and allows the exchange of
liberalization commitments in migration for commitments in other areas, such as investment
or goods (Trachtman 2008, 33). It is, however, questionable how practical it is to have a very
diverse and inconsistent group of Mode 4 commitments from different countries. Mode 4
already suffers from lack of common understanding as to its scope, which from the point of
view of the service suppliers undermines its legal predictability. The main Mode 4 categories
need to be clearly defined, and for the sake of clarity and consistency, they should correspond
to the scope of the GATS.65 If the categories could be made clearer, there would still be
plenty of room for negotiations on questions such as length of stay, professional experience,
economic needs tests and covered sectors. The national treatment commitments would ideally
take a stand on the question of to what extent host state labour laws apply to workers sent to
perform a contract. The situation is different in each country and binding the national rules
would greatly improve transparency and predictability.
The rest of the chapter is dedicated to the results of an empirical study on the
biggest WTO service importers’ Mode 4 commitments. The results show that most of the
Members reviewed seem to differentiate Mode 4 from labour mobility and formulate their
commitments in a way that is aimed to prevent the establishment of an employment
relationship in the host state.
4. MODE 4 IN WTO MEMBERS’ COMMITMENTS
4.1 Method of the study
We now turn to an analysis of the issue in the light of WTO Members’ commitments.
28
The countries chosen for review include the ten WTO Members with the highest
share in imports of commercial services in 2011.66 Ranked according to volume of imports
from the biggest to the smallest these ten countries are the EU-27 (extra-EU imports), the
United States, China, Japan, India, Singapore, Canada, Republic of Korea, Russian Federation
and Brazil. Imports were considered most relevant for the study since each Member’s
schedule of commitments defines the level of liberalization applied to imports only. It is,
however, noteworthy that an almost identical group of countries appears in the list of the top
ten services exporters.67
Most Members’ schedules were formulated in 1994 at the conclusion of the
GATS. China and Russia joined the WTO during the Doha Round and their schedules date
from 2001 and 2011, respectively. Offers made during the Doha Round were not reviewed in
this study. It is likely that since 1994, the conditions for service supply under Mode 4 have
changed in many Members to a large extent. However, rather than enquiring into currently
applicable entry conditions, the purpose of this small study is to examine the Members’
approach to the question of employment market access. Despite the study’s limitations, the
Members’ perception of the scope and purpose of Mode 4 at the time of its negotiation can be
illustrated in the light of the original GATS commitments.68
The review is limited to the chosen Members’ horizontal Mode 4 commitments.
Most Members include the categories of persons covered under Mode 4 in the horizontal
section of their schedule of commitments. The exact depth of liberalization for a specific
service sector is, however, revealed only by reviewing the specific commitment for the sector
concerned. Under Mode 4, in those cases where a commitment exists, it is common practice
to refer to the schedule’s horizontal section, which includes the common rules for the types of
movement that the Member accepts under Mode 4.69 These rules contain the categories of
persons covered and the conditions for their access to the host state. The categories of persons
29
usually apply to all sectors although some Members limit some categories of persons to
certain sectors or to certain professional groups.70
National treatment is usually provided for the categories of persons covered in
the horizontal commitments. This is usually expressed in the horizontal commitments’
national treatment column in the following blanket reference to immigration law or similar:
‘Unbound except for the measures concerning the entry and temporary stay of natural persons
who fall into the categories referred to in the market access column’. Such blanket references
are problematic since they allow for broad discretion which diminishes legal security.
Moreover, giving foreign service suppliers the same treatment as one’s own nationals does
not mean that foreign qualifications, education and experience were accepted. Especially in
regulated professions the Members generally require national qualifications or limit
considerably the scope of activity allowed. This is usually expressed in the horizontal
commitments by noting that the natural person must possess the necessary academic
qualifications and professional experience as specified for the sector or activity concerned.
The possibility to practise one’s profession in the Member in question is thus revealed only by
reviewing the applicable requirements in the relevant service sector.
Even though qualifications, education and professional experience are relevant
to whether foreign workers can practice their professions in other countries, for the purposes
of this chapter the review is limited to issues directly relevant for employment market access.
Such issues include the type of contract that the service supplier is engaged under, the foreign
or national ownership of the entity engaging the workers, the existence of a link between the
service provider and an employer in the sending country (e.g. the requirement of prior
employment), the requirement of an economic needs test (to check for the availability of
similarly qualified workers in the host state), the period of stay, the source of remuneration
and the status of the service supplier in the company hierarchy (for ICTs).
30
4.2 Results of the study
The categories that appear most often in the Members’ schedules that were reviewed are
ICTs, IPs and BVs. ICTs are included in 60 per cent of all Members’ schedules and they are
clearly the category of persons most frequently covered (Carzaniga, 2008, p. 481). The
movement of ICTs is almost always connected to commercial presence in the host state (see
Persin 2010, pp. 786–787). BVs are the second-biggest group, whereas CSSs and IPs appear
only in a small number of Members’ schedules. Some Mode 4 commitments provide for the
possibility of an employment relationship in the host state, the most notable one being the US
quota for highly-skilled persons in specialty occupations (H-1B visa). Such bindings,
however, seem to represent a small minority.
The results of the present study show that the biggest service importers follow
the MNP Annex’s demarcation and cover in their horizontal commitments only categories of
persons who do not access the national employment market. There are, however, two
important exceptions. Firstly, the H-1B visa category of the US allows employment market
access for up to 65,000 persons annually in specialty occupations.71 Secondly, Brazil allows
foreign specialized technicians and highly qualified professionals to work under temporary
contracts with all legal entities established in Brazil, whether of national or foreign capital.72
The commitment of the US is peculiar in the sense that it clearly provides for
access to the country’s employment market. It concerns persons engaged in specialty
occupations that require theoretical and practical application of a body of highly specialized
knowledge and the possession of a bachelor's or higher degree in the specialty as a minimum
for entry into the occupation in the US.73 The persons and their US employers must comply
with a number of conditions regarding wages and conditions of work. In addition, the
31
employer must not have laid off or displaced workers during a specific period before or after
engaging the foreign worker. The employer must also take timely and significant steps to
recruit and retain sufficient US workers in the relevant specialty occupation. The length of
stay is limited to three years.
Brazil’s commitment is more open-ended. It states that foreign specialized
technicians and highly qualified professionals may work under a temporary contract with
legal entities, whether of national or foreign capital, established in Brazil. The company has to
justify the need to contract such professionals and technicians in relation to similar
professionals and technicians available in Brazil. In addition, the contract must be approved
by the Ministry of Labour. With regard to certain service activities (e.g. communications, land
transportation, commercial stores, hotels and restaurants), the engaging entities must obey the
proportionality requirement of at least two Brazilians for every three employees. There are no
specifications as to wage parity or other working conditions. In the light of the terminology
used (work under a temporary contract and the proportionality of two Brazilians for three
employees) it seems that the commitment is meant to cover employment relationships
between local companies (also of national capital) and foreign workers.74
None of the other schedules reviewed include such clear examples of
employment market access. The most commonly occurring category of persons in them are
ICTs. All the schedules reviewed include ICTs and all of them cover companies established in
the host state. In most cases the receiving entity must be in the form of a subsidiary,
dependent company or branch of a juridical person of another Member performing the intra-
corporate transfer.
In the schedules reviewed, the category of ICTs covers only persons who are not
easily replaceable by local workers. They need to be highly qualified and possess experience
and knowledge relevant to the company in question (evident in light of the obligation to have
32
a prior employment in the sending company). Even if such persons are in the field of
application of local labour laws, they can hardly be seen as occupying in the employment
market a place that could equally well be occupied by a local worker and thus be displacing a
domestic worker. Therefore, there is generally no requirement to apply an economic needs
test before this type of transfer. There is, however, some haziness concerning the employment
market access of ICTs in the schedule of the EU. There, certain EU Member States (Finland,
Latvia and Poland) state that either an economic needs test is applied or that the post needs to
be open for applications before the transfer can take place.75
Another group of service suppliers that can in certain cases take away jobs from
local workers and thus become comparable to employees are CSSs.76 This seems to be the
view of some EU Member States at least. The EU’s schedule states that an economic needs
test for CSSs will not be required except where otherwise indicated for a specific sub-sector.
Such tests can be found in certain sectors. For example, the United Kingdom requires an
economic needs test for persons admitted under Mode 4 in engineering services.
The rather strict limits imposed on the use of CSSs (for example, maximum
three months in any 12-month-period for most EU Member States) reflect the sensitivities
relating to the use of foreign workers through service contracts. CSSs are engaged in work
that can usually be similarly performed by domestic workers. Even though several countries
require that local labour standards must be adhered to, the control of such standards is much
harder where no employment relationship to a domestic company exists.
It seems that the intra-EU development inspired the formulation of the EU’s
GATS offer of 2005.77 There the EU has in two instances (CSSs and IPs) defined that the
offer applies to natural persons engaged in the supply of a service as employees of a juridical
person as long as the juridical person is not an agency engaged in the placement and supply
services of personnel (service sector CPC 872). The exception appears to reflect the EU’s
33
internal case law according to which the making available of labour (supply of personnel) is
not within the scope of free movement of services but comes in the field of free movement of
workers.
The EU case law in question relates to the use of non-EU national workers in the
intra-EU provision of services. In its case law, the Court of Justice of the EU (CJEU) has built
its legal analysis of cross-border service supply around the question of employment market
access.78 In a relatively recent case the CJEU ruled that hired workers (temp-agency workers)
are a group of service suppliers that specifically seek access to the host state’s employment
market and thus belong to the category of workers.79 In cases where such workers are not EU
nationals and are sent from one EU Member State to work in another, work permits may still
be required (a requirement otherwise prohibited in intra-EU provision of services). Moreover,
the host-Member State’s labour laws apply in their entirety.
Interestingly, Canada has also paid attention to avoiding the formation of a
relationship that resembles employment, between the client and the foreign service supplier’s
workers, by excluding temporary agency work. Canada’s commitment on professionals
engaged by a services contract obtained by a juridical person of another Member with no
commercial presence in Canada excludes agencies engaged in the placement and supply
services of personnel (service sector CPC 872). In addition, Canada clarifies that
professionals may not engage in secondary employment while in Canada.
Even though the GATS covers the supply services of personnel, i.e. manpower
services, there are especially strong tensions that relate to the liberalization of such services.
In the EU, the CJEU has determined that the rules concerning the free movement of workers,
rather than services, should in some respects be applied to hired workers. If one applies the
criteria for employment market access established in this chapter, one can conclude that hired
workers do in certain cases enter the host state employment market. Countries may provide
34
for the full application of local labour laws.80 In addition, hired workers are usually engaged
to perform jobs that could be done by local workers.
Certain clear exceptions (most notably the US) and certain borderline cases
(executives, managers and specialists) notwithstanding, the Members reviewed have
formulated their Mode 4 bindings so as to avoid the formation of employment relationships
between service suppliers and host state employers. At the same time it is notable that there
are considerable differences in the periods of time for which access is allowed, ranging
between 90 days and several years. The results give reason to conclude that the biggest
importers of services consider that movement under Mode 4 is something that takes place
primarily outside the host state’s employment market.
5. CONCLUSION
The object of this chapter has not been to rule out the possibility of regulating migration for
employment purposes through GATS commitments. Members remain free to do so since they
can schedule the types of movement they prefer. As the study conducted for the purposes of
this chapter showed, some Members have indeed done this. Where such commitments exist,
they should be considered binding. An analogy can be drawn to new Members’ accession
protocols which remain binding in their entirety even if they sometimes contain elements that
are outside the general WTO disciplines.
The chapter has aimed at forming a more conceptual understanding of Mode 4,
inspired by the specific context it is part of. The analysis of Mode 4 often suffers from
analytical confusion regarding the differences between labour and service mobility. The
purpose is not to deny the significant and desirable benefits that would follow from deeper
liberalization of labour mobility. Instead, it is considered that imprudent extension of Mode 4
35
commitments in the field of employment-based migration carries the risk of making Mode 4
less relevant and more prone to violations of WTO law. Labour migration is currently rarely
administered on an MFN basis. In WTO law, however, the treatment of a temporary labour
migrant of one specific nationality would need to be extended to all similarly positioned
nationals of WTO Members. Moreover, the availability of MFN exceptions is limited and
new exceptions are hard to add. In practice, several WTO Members already seem to operate in
a legally grey area. Restricting the scope of Mode 4 to genuine service transactions would
clarify the types of situations it covers and improve the credibility of Mode 4 as a trade
instrument which, instead of providing for a status quo, could entail real liberalization.
As the study on the Members’ commitments demonstrated, there seems to be a
certain distinction between service mobility and traditional labour mobility built around the
issue of employment market access. Drawing the line between employment and supply of
services can, however, prove difficult and often necessitates a case-by-case analysis. The
issue is resolved in accordance with the host state’s legislation, which means that similar
situations may be judged differently in different countries. A good example of difficulties
relating to the differentiation of service supply from an employment relationship is the EU
experience with the posting of third-country workers. In its case law on posted third-country
nationals, the CJEU has drawn a subtle and somewhat blurry distinction between work carried
under a service contract and a service carried out with the sole purpose of entering the host
state’s employment market.
Employment market access almost always leads to wage parity. Temporary
service contracts, in contrast, can sometimes be used to benefit from differences in wages and
other conditions of work. This is, however, not self-evident under Mode 4 as many countries
understandably require wage parity for ICTs and CSSs even if no employment market access
is seen to occur. Independent professionals and business visitors (such as sale negotiators) are
36
the groups that in most cases remain outside the application of domestic labour laws and thus
easily fall under Mode 4. One can, however, perceive that movement of natural persons in
conjunction with cross-border trade (Mode 1) and commercial presence (Mode 3) is capable
of creating a growing number of situations where added value is brought about by the
temporary relocation of personnel from one Member to another. One can take the example of
IT specialists who usually provide cross-border services, but occasionally travel to the
location of the client. In such cases the natural persons are easily left outside the application
of the host state labour laws. When it comes to CSSs, the most interesting category of Mode 4
especially for low-skilled workers, it is unlikely that developed countries would give up on
requirements of wage parity.81
Even if widened employment market access were to bring about significant
economic gains and open up attractive possibilities for workers, especially in developing
countries, from a trade perspective the liberalization of contractual service supply would be
especially beneficial if it allowed the movement of low-skilled workers into a wide spectrum
of service sectors. It is often conceived that employment market access is the most sensitive
issue relating to cross-border movement of natural persons and especially hard to justify to
national constituencies. This is understandable since employment market access is usually
reserved for immigrants and any extension of immigration rights is likely to cause
controversies. The resistance to immigration may, however, turn against itself when
companies seek to benefit from cheaper ways to perform jobs. Thus, when there is a specific
need for a foreign workforce (as in the US for IT specialists), an increase in labour
immigration may to a certain extent prevent the outsourcing of the same jobs to foreign
service suppliers. When the need for foreign workers grows in the developed countries,
opening up their employment markets may in practice be the most effective way to ensure
compliance with local wage and labour laws. For labour-rich countries, on the other hand, the
37
extension of possibilities to supply services on a contractual basis outside the regulatory
jurisdiction of the host country should appear at the top of their Mode 4 agenda. This would
allow them to benefit from their abundant labour force without losing the workers to host
countries’ employment markets.
BIBLIOGRAPHY
Bast, J. (2008) ‘Commentary on the Annex on Movement of Natural Persons Supplying
Services Under the Agreement’ in R. Wolfrum, P.-T. Stoll and C. Feinäugle
(eds) WTO – Trade in Services (Leiden/Boston: Martinus Nijhoff) 573–595.
Betts, A., and K. Nicolaïdis (2009) ‘The Trade-Migration Linkage: GATS Mode IV’, in M.
Eagleton-Pierce, E. Jones and K. Nicolaïdis (eds) Building Blocks for a Global
Trade Ethics (A Compendium of the Oxford Programme on Global Trade
Ethics) 65–74.
Bradford Jensen, J. (2011) Global Trade in Services: Fear, Facts, and Offshoring
(Washington, DC: Peterson Institute for International Economics).
Bungenberg, M. (2010) ‘Going Global? The EU Common Commercial Policy After Lisbon’
in C. Herrmann and J.P. Terhechte (eds) European Yearbook of International
Economic Law 2010 (Heidelberg: Springer), 123–151.
Carzaniga, A. (2003) ‘The GATS, Mode 4, and Pattern of Commitments’ in A. Mattoo and A.
Carzaniga (eds) Moving People to Deliver Services (Washington, DC: World
Bank and Oxford University Press) 21–26.
Carzaniga, A. (2008) ‘A warmer welcome? Access for natural persons under PTAs’ in J.
Marchetti and M. Roy (eds) Opening Markets for Trade in Services: Countries
38
and Sectors in Bilateral and WTO Negotiations (Cambridge, New York:
Cambridge University Press) 475–502.
Chanda, R. (2001) ‘Movement of Natural Persons and the GATS’, The World Economy,
24(5), 631–654.
Chanda, R. (2004) ‘Movement and Presence of Natural Persons and Developing Countries:
Issues and Proposals for the GATS Negotiations’, Trade-related agenda,
development and equity (T.R.A.D.E.) Working Papers (19).
Chanda, R. (2009) ‘Mobility of Less-Skilled Workers under Bilateral Agreements: Lessons
for the GATS’, Journal of World Trade, 43(3), 479-506.
Chaudhuri, S., A. Mattoo and R. Self (2004) ‘Moving People to Deliver Services: How can
the WTO Help?’ Journal of World Trade, 38 (3), 363–393.
Dey, D. (2007) ‘The Movement of Natural Persons (Mode 4) Under GATS: Advantage
Developing Countries!’ in D. Dey (ed.) GATS: An Introduction (Hyderabad:
Icfai University Press), 78–104.
Espa, I. (2012) ‘The Appellate Body Approach to the Applicability of Article XX GATT In
the Light of China – Raw Materials: A Missed Opportunity?’ Journal of World
Trade, 46(6), 1399–1423.
Gibson, J., and D. McKenzie (2011) ‘Eight Questions about Brain Drain’ Journal of
Economic Perspectives, 25(3), 107-128.
Grynberg, R., and V. Qalo (2007) ‘Migration and the World Trade Organization’, Journal of
World Trade, 41(4), 751–781.
Hepple, B.A. (2005) Labour laws and global trade (Oxford: Hart).
Kelsey, J. 2010. ‘How “trade in services” transforms the regulation of temporary migration
for remittances in poor countries’, in M. Kolsky Lewis and S. Frankel (eds)
39
International Economic Law and National Autonomy (Cambridge: Cambridge
University Press).
Liukkunen, U. (2004) The Role of Mandatory Rules in International Labour Law. Helsinki:
Talentum.
Liukkunen, U. (2006) Cross-Border Services and Choice of Law (Frankfurt am Main: Peter
Lang).
Magdeleine, J., and A. Maurer (2008) ‘Measuring GATS Mode 4 Trade Flows’, WTO Staff
Working Paper ERSD-2008-05.
Marchetti, J.A. and P.C. Mavroidis (2011) ‘The genesis of the GATS (General Agreement on
Trade in Services)’, European Journal of International Law 22 (3): 689–721.
Mattoo, A. and A. Carzaniga (2003) Moving People to Deliver Services (Washington, DC:
World Bank and Oxford University Press).
Mukherjee, N. 1996. ‘Exporting Labour Services and Market Access Commitments under
GATS in the World Trade Organization’, Journal of World Trade 30(5), 21–42.
OECD, IOM, and World Bank (2004) Trade and migration: building bridges for global
labour mobility, Paris: OECD.
Panizzon, M. (2010) ‘Trade and Labor Migration: GATS Mode 4 and Migration
Agreements’, Friedrich-Ebert-Stiftung Occasional Paper N° 47, January 2010.
Persin, D. (2010) ‘Market Access for Small versus Large Service Enterprises: The
Preferential and Multilateral Trade Liberalization Tracks Compared’, Journal of
World Trade, 45(4), 785–819.
Reyna, J.V. and T.P. Stewart (1993) The GATT Uruguay Round: a Negotiating History
(1986–1992): Services (Deventer: Kluwer Law and Taxation).
40
Self, R.J. and B.K. Zutshi (2003) ‘Mode 4: Negotiating Challenges and Opportunities’ in A.
Mattoo and A. Carzaniga (eds) Moving People to Deliver Services (World Bank
and Oxford University Press) 27–58.
Stephenson, S.M. and G. Hufbauer (2010) ‘Increasing Labor Mobility: Options for
Developing Countries’ in O. Cattaneo, M. Engman, S. Sáez and R.M. Stern
(eds) International Trade in Services: New Trends and Opportunities for
Developing Countries (Washington, DC: World Bank), 29–66.
Stephenson, S.M. and G. Hufbauer (2011) ‘Labor Mobility’, in J.-P. Chauffour and J.-C.
Maur (eds) Preferential Trade Agreement Policies for Development: a
Handbook (World Bank: Washington, DC), 275–306.
Trachtman, J.P. (2008) ‘The Role of International Law in Economic Migration’, Society of
International Economic Law, Online Proceedings, Working Paper (No. 24/08).
Trachtman, J.P. (2009) The International Law of Economic Migration: Toward the Fourth
Freedom (Kalamazoo, MI: W.E. Upjohn Institute for Employment Research).
Winters, A.L. (2002) ‘Liberalising Labour Mobility under the GATS’, Commonwealth
Economic Paper Series (London: Commonwealth Secretariat).
Winters, A.L. (2008) ‘The Temporary Movement of Workers to Provide Services (GATS
Mode 4)’ In A. Mattoo, R.M. Stern and G. Zanini A Handbook of International
Trade in Services (Oxford: Oxford University Press).
Wolfrum, R., P.-T. Stoll and C. Feinäugle (2008) WTO – trade in services, Max Planck
Commentaries on World Trade Law (Leiden/Boston: Martinus Nijhoff).
Worster, W.T. (2006) ‘Conflicts Between United States Immigration Law and the General
Agreement on Trade in Services: Most-Favored Nation Obligation’, The Texas
International Law Journal, 42(1), 55–117.
41
WTO (1998) Presence of Natural Persons (Mode 4), Background Note by the Secretariat,
Council for Trade in Services, WTO Doc. S/C/W/75 of 8 December 1998.
WTO (2009) Presence of Natural Persons (Mode 4), Background Note by the Secretariat,
WTO Doc. S/C/W/301 of 15 September 2009.
WTO, and OECD. 2005. ‘Background Note on GATS Mode 4 and its Information Needs.’
United Nations Deparment of Economic and Social Affairs, Statistics Division
no. TSG 2/8.
1 Johanna Jacobsson, PhD candidate, Department of Law, European University Institute. Contact information:
johanna.jacobsson@eui.eu. The author is grateful for the invaluable comments of Marion Panizzon, Antonia
Carzaniga and Petros C. Mavroidis on an earlier draft of this chapter.
2 Betts and Nicolaïdis portray Mode 4 as a shift to a market-based logic to govern human mobility. See Betts and
Nicolaïdis 2009.
3 Article I.2(d) of the GATS. The GATS differentiates between four modes of supply: 1) from the territory of
one Member into the territory of any other Member (cross-border trade); 2) in the territory of one Member to the
service consumer of any other Member (consumption abroad); 3) by a service supplier of one Member, through
commercial presence in the territory of any other Member (commercial presence, or investment) and 4) by a
service supplier of one Member, through presence of natural persons of a Member in the territory of any other
Member. Jan Schmitz provides an illustrative explanation of the GATS modes in chapter X, p. X.
4 The relevance of cross-border movement of labour was evident early on in the GATS negotiations but the
question of the scope of Mode 4 took longer to resolve. See Reyna and Stewart 1993, 27. As this chapter will
show, the exact scope of Mode 4 remains uncertain.
5 The natural and juridical persons covered are defined in Article XXVIII of the GATS. In the case of natural
persons, the criteria are based on nationality and/or permanent residence.
6 Bast asserts that the MNP Annex represents a specific legal regime in which the contracting parties maintain a
careful balance between the sensitive issue of migration and the narrow confines of trade. The ambiguous
language of the Annex is understandable in the light of the diverging interests of different WTO Members that
the drafters of the Agreement had to accommodate. See Bast 2008, 575.
42
7 The Annex rules out access to the employment market but does not define what such access means. WTO
Members can thus largely determine the exact scope of Mode 4 through their own labour and immigration
regulation.
8 Perhaps due to these challenges in the definition of employment market access, the extensive literature on
Mode 4 has largely neglected its conceptualization despite its importance for the understanding of the GATS.
9 Several such analyses have been conducted by economists. See useful overviews in Bradford Jensen 2011 and
Trachtman 2009.
10 See a more extensive account of the GATS and inclusion of Mode 4 in the GATS framework by Jan Schmitz
in chapter X (sections 1 and 2).
11 Commentators share the view that the US financial services sector played a prominent role in the inclusion of
services in the multilateral trade negotiation agenda. See Marchetti and Mavroidis 2011, 692.
12 Since then the economic significance of services has only grown: at the moment services account for
approximately 75% of the GDP in OECD countries and over 70% of global output (The World Bank’s World
Development Indicators, available at http://databank.worldbank.org).
13 Magdeleine and Maurer (2008) note the challenges relating to the ‘adequate translation’ of the Mode 4 legal
provisions into statistical concepts. In the economic literature, compensation of employees and migrant workers’
remittances are often used as statistical indicators to estimate the value of Mode 4 trade. They are not suitable
indicators, however, since they are labour income measures. To grasp Mode 4 type movement, Magdeleine and
Maurer concentrate on two aspects which aim to measure trade flows (transactions): the value of the service
provided and the number of natural persons moving. Under Mode 4, most value is created in contractual service
supply since other types of movement are usually related to Mode 3 (ICTs) or to transactions that take place at a
later stage (business visitors and service sellers) (p. 8).
14 On the scope of Mode 4 see Self and Zutshi 2003, Chaudhuri et al. 2004, Grynberg and Qalo 2007, Mukherjee
1996 and OECD et al. 2004.
15 See especially Winters 2008, 515.
16 The EU’s Communication to the Special Session of the Council for Trade in Services, WTO Document
S/CSS/W/45 of 14 March 2001, p. 1.
17 Chaudhuri, Mattoo and Self make a similar argument: Chaudhuri, Mattoo and Self 2004, 381-382.
43
18 According to the MFN principle, WTO Members cannot discriminate between other Members but have to
grant the same treatment to everyone. Along with national treatment (no discrimination between domestic and
foreign service suppliers), MFN is a cornerstone of the GATS and WTO law in general.
19 See chapter X by Ryszard Cholewinski in this volume.
20 Recent empirical research, however, questions the gravity of the brain-drain phenomenon and offers evidence
of the benefits that skilled migration can offer both for migrants and for sending countries. Nonetheless, a
comprehensive understanding of the effects of brain-drain on sending country growth and development is still
lacking. See Gibson and McKenzie 2011.
21 For an overall explanation of GATS rules on Mode 4, see section X of chapter X by Jan Schmitz.
22 Article 1:2(d) defines Mode 4 to cover the supply of a service by a service supplier of one Member, through
presence of natural persons of a Member in the territory of any other Member. The first paragraph of the MNP
Annex states that it applies to measures affecting service suppliers of a Member, and natural persons of a
Member who are employed by a service supplier of a Member, without specification as to the origin of the
employing service supplier. This paragraph, however, needs to be read together with Article 1:2(d) which clearly
states the presence of a cross-border element. See the entire text of the MNP Annex in chapter X, p. X.
23 The second paragraph of the MNP Annex clarifies that the GATS does not apply to measures affecting natural
persons seeking access to the employment market of a Member, nor to measures regarding citizenship, residence
or employment on a permanent basis.
24 Similar categorization was also proposed by the EU, Canada and a group of developing countries led by India
during the Doha Round negotiations (see p. X of chapter X by Jan Schmitz in this book).
25 See, for instance, Carzaniga 2008, 479; Chanda 2001, 2004 and Chaudhuri, Mattoo, and Self 2004, 370. Also
the WTO Secretariat’s background notes on Mode 4 adopt this view. See WTO 1998 and WTO 2009.
26 The duration of stay should be included in the Members’ commitments. In the absence of such a
determination, it can be considered that no binding is undertaken thereof. See paragraph 34 of the Scheduling
Guidelines, document S/L/92, dated 28 March 2001.
27 See the examples of the Canadian seasonal agricultural worker programme and the Spain–Ecuador Agreement
on Migratory Flows in Chanda 2009.
28 Some commentators bring up the issue but do not go into it very deeply. See, for example, Chanda 2004, 16;
and OECD et al. 2004, 105. Bast enquires into the GATS concepts of ‘employment’ and ‘employment market’
but does not consider how the formation of an employment relationship affects different categories of Mode 4
44
entrants (Bast 2008, 581). Chaudhuri et al. analyse differences between service and work contracts and provide
an interesting analysis of existing commitments on contractual service suppliers. See Chaudhuri et al. 2004, 381–
384. In addition to Bast and Chaudhuri et al., useful information on how to distinguish service contracts from
employment contracts is included in the UN Background Note of 2005 (WTO and OECD 2005) and in the WTO
Secretariat’s Background Note of 2009 (WTO 2009).
29 In the WTO Secretariat’s latest background note on Mode 4 it is pointed out that several empirical studies
have attempted to quantify the potential economic impact of liberalizing Mode 4. It is, however, stressed that at
the outset all these studies seek to estimate the effects of freeing up the temporary movement of workers (labour
mobility), rather than the effects following from liberalization of Mode 4. See WTO 2009, 13.
30 Bast notes that a natural person offering services on a regular basis to one particular customer under the
customer’s instructions is usually seen as being engaged in an employment relationship, any other denomination
used in the contract notwithstanding. Bast 2008, 581. See an enlightening example of the types of criteria that are
used in one jurisdiction: WTO 2009, 5.
31 An overview of all Members’ commitments in 2002 revealed that almost 280 out of 400 entries in the
Members’ GATS schedules pertained to executives, managers and specialists. Of these, 168 entries explicitly
related to ICTs (42% of aggregate entries). See Carzaniga 2003, 24–25.
32 Carzaniga notes that the economic value of commitments on ICTs is dependent on access conditions for Mode
3. Carzaniga 2003, 24.
33 The company-specific specialty is usually demonstrated with a prior employment requirement. Of the ten
biggest service importers whose Mode 4 schedules were examined for the purposes of this chapter, eight
explicitly require prior employment in the sending state for a minimum of one year before the transfer. Two
(Brazil and China) do not mention any prior employment criterion but limit transfers of ICTs to senior
employees only.
34 In 2005 the Congress started to oppose the inclusion of immigration measures in trade agreements especially
because of the limited possibility to exercise control. This has blocked the inclusion of Mode 4 commitments in
US trade agreements altogether. See Sarah Anderson, U.S. Immigration Policy on the Table at the WTO,
http://www.globalpolitician.com/default.asp?21446-immigration/, date accessed 31 July 2013.
35 The ambit of the EU’s Common Commercial Policy (CCP) was extended in the Treaty of Nice to cover all
modes of services. Prior to that, the European Court of Justice (ECJ) had held that only cross-border trade (Mode
1) fell within the CCP since it was ‘not unlike trade in goods’ and involved no movement of persons. See
45
Opinion 1/94, Opinion of the Court of 15 November 1994 – Competence of the Community to conclude
international agreements concerning services and the protection of intellectual property [1994] ECR I-05267. In
Opinion 1/08 the ECJ confirmed that the Community was, as a result of the Nice Treaty, competent to conclude
international agreements relating to trade in services supplied under modes 2, 3 and 4. Opinion 1/08, Opinion of
the Court (Grand Chamber) of 30 November 2009 – Opinion pursuant to Article 300(6) EC [2009] ECR I-
11129, paragraph 119. With the Lisbon Treaty, the most sensitive sectors, such as social and human health
services, were also brought into the exclusive competence of the EU.
36 For an excellent account of the welfare economics of migration see Trachtman 2009. The economic effects of
temporary v. permanent migration in home and destination countries are debated among economists. There are
no simple conclusions but the theory suggests that the biggest benefits accrue to the migrants themselves. See
Trachtman 2009, 48. Temporary migration, or, as in this case, service mobility, does not necessarily make the
home country better off. Temporary migration, however, appears better suited than permanent migration to
ensure that the arrangement benefits developing countries. See Chaudhuri, Mattoo and Self, 381.
37 In the MNP Annex persons covered by Mode 4 are defined either as natural persons who are service suppliers
of a Member or as natural persons of a Member who are employed by a service supplier of a Member, ‘in respect
of the supply of a service’. See Bast 2008, 579.
38 The host state may on certain occasions require similar contributions. However, in many cases where work is
subcontracted to a foreign company, social security contributions are paid by the foreign company according to
the provisions of its home country. See OECD et al. 2004, 85.
39 This consideration is mostly relevant for CSSs since ICTs and IPs are usually highly-skilled and highly-paid
specialists. Fifty WTO Members have prescribed the application of domestic minimum wages in their schedules,
often coupled with some other laws regarding working conditions. See OECD et al. 2004, 84. In principle,
scheduling is not necessary since Members are allowed to apply their national laws to Mode 4 entrants. In the
absence of an explicit statement, it is, however, hard to know to what extent local wages and other working
conditions are extended to temporary service suppliers.
40 The US, for example, imports most of its services from other high-wage, skill-abundant countries. Jensen
notes that even the US companies establishing affiliates overseas to export services back to the parent company
are usually located in high-income countries and thus pay relatively high wages. See Jensen 2011, 156.
41 In Bast’s opinion the legal construction of the terms of the MNP Annex should aim at preserving the fragile
compromise represented therein. Winters shares the view that bringing the WTO into the contested field of
46
domestic labour law without proper support by the Members puts at risk the attainment of other trade objectives.
See Bast 2008, 577 and Winters 2002, 29.
42 See Cholewinski, Battistella and Hazan et al. in this volume.
43 Article V bis of the GATS was tailored especially to meet the needs of the Nordic countries that had deeply
integrated labour markets and did not want to risk the compatibility of their arrangement with the GATS. Bast
also mentions India’s integrated labour market with Nepal (Bast 2008, 154).
44 An exception is made for new Members who have the option to inscribe their own MFN exemptions when
acceding to the WTO. See Wolfrum, Stoll, and Feinäugle 2008, 570.
45 Wolfrum notes that the wording ‘in principle’ and the fact that MFN exemptions remain subject to future
negotiations imply that the ten-year period is not indefinite. In fact, most Members’ exemptions were introduced
as indefinite. Ibid., 571.
46 Preferential Treatment to Services and Services Suppliers of Least-Developed Countries, Ministerial
Conference Decision of 17 December 2011, WTO Document WT/L/847. The existence of the exceptional
circumstances justifying the waiver is reviewed annually. The waiver terminates in 15 years from the date of
adoption.
47 See a comprehensive account of such exemptions in Grynberg and Qalo 2007. See also Annex of WTO 2009.
48 See especially Grynberg and Qalo 2007, 758.
49 Similar reasons are likely to be behind the inclusion of Article V bis in the GATS (labour markets integration
agreements). Since entire labour market integration agreements are exempted, the countries involved can
liberalize the movement of service suppliers as well. This may be a practical necessity since many countries’
immigration schemes do not differentiate between employment-based movement and service supply.
50 The MNP Annex specifies that the agreement does not prevent a Member from applying measures to regulate
the entry of natural persons into, or their temporary stay in, its territory. Visas, for example, can be administrated
differently among different nationalities. Border and entry measures cannot, however, be applied in such a
manner as to nullify or impair the benefits accruing under the terms of the Member’s specific commitment.
Excessive border control measures or work permit procedures could, therefore, potentially be the subject of a
GATS dispute. See Worster 2006, 68.
51 See p. X in chapter X. It should, however, be noted that the GATS provides for a comparison with like service
suppliers. Considering the significance of a service supplier’s personal characteristics (such as education and
skills), it can be extremely difficult to prove that likeness exists in any given case.
47
52 Some commentators have referred to the MNP Annex, which allows Members to regulate the entry and stay in
their territory, and stated that immigration and visa policies are covered by specific commitments only. This can
be contested since MFN is a general obligation and applies to all measures relevant for the application of the
GATS. See Worster 2006, 74.
53 Bast interprets the MNP Annex’s carve-out of employment on a permanent basis to refer mainly to ICTs who
are formally employed by a host state entity but who do not seek entry to the employment market and whose stay
is temporary by definition. Bast 2008, 586–587. Schmitz shares the view. See chapter X, p. X of this book.
54 See WTO 2009, 24. In addition to the EU, several newly acceded Members have inscribed a commitment on
CSSs, which may imply that this category is becoming more widely acknowledged.
55 The numbers of applied quotas and ENTs can be found in WTO 2009, 22–24.
56 As has been pointed out above, in practice an employment relationship can, however, be seen to form between
the customer and a service supplier.
57 Choice of law rules are determined by private international law. Each country has its own private international
law that is a branch of national law.
58 Different terms, such as lois de police, lois d’application immediate and Eingriffsnormen, are used. See
Liukkunen 2006, 76.
59 This approach was adopted in the Rome Convention on the law applicable to contractual obligations adopted
by EC Member States in 1980. See Liukkunen 2004.
60 Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on services in the
internal market. The Commission originally proposed a country-of-origin principle, which stipulates that service
providers should be subject to the laws of their own country rather than of the country where the service is
provided.
61 However, where such WTO+ commitments are made, they should, a priori, be considered binding. One can
draw a parallel to commitments made by new WTO Members in their accession protocols. The new Members’
commitments sometimes go further than the WTO rules but are still binding. Such rules relate, for example, to
export duties regarding which China at least was required to make commitments that do not exist for original
WTO Members on the basis of the GATT. The binding nature of China’s WTO+ commitments was upheld in
the WTO dispute settlement. See Espa 2012.
62 Betts and Nicolaïdis note that, India excluded, developing countries are not doing research on demand and
supply to find out what would benefit them most. India, by contrast, is on a path significantly different from
48
most other developing countries but tries to speak on behalf of the developing world as a whole. Betts and
Nicolaïdis 2009, 70.
63 Kelsey writes about the ‘fetishisation’ of turning Mode 4 into labour migration in exchange for remittances.
She points out how ‘the broader rights and interests of workers, their communities and the long-term
development vision of the country are often left out of the equation altogether’. Kelsey 2010, 286.
64 For CSSs, Chanda proposes a form of juridical affiliation for less-skilled workers in the home country. Such
affiliation could sponsor a worker and appear as the contracting party for the overseas client. The return of the
Mode 4 entrants could be encouraged through different incentives for the entrant or through obligations imposed
on the sending country. Chanda 2009.
65 The EU notes in a communication to the Special Session of the Council for Trade in Services that the
definitions of terms such as personnel, administrators, managers, and specialists are often neither clear nor
consistent between different country schedules. According to the EU, the vague terms and definitions lay the
ground for administrative discretion and can thus be subject to arbitrary and discriminatory application by
regulatory authorities. See WTO Document S/CSS/W/45 of 14 March 2001.
66 The EU is counted as one Member and intra-EU trade is excluded. See ‘Leading exporters and importers in
world trade in commercial services (excluding intra-EU(27) trade), 2011’, Appendix Table 6 of the World Trade
Report 2012 (p. 33), http://www.wto.org/english/res_e/reser_e/wtr_e.htm, date accessed 31 July 2013.
67 In the list of the top ten importers of commercial services, Russia is number 9, whereas among exporters it is
number 11. Brazil is number 10 in the list of top importers but only number 18 in the list of exporters.
68 It is, however, noteworthy that the categories of persons covered have not radically changed in the Doha
Round offers. Instead, certain Members have aimed at solidifying the four most established categories (ICTs,
BVs, IPs and CSSs). See p. X of chapter X by Jan Schmitz in this book.
69 In many schedules this is expressed by inscribing in the relevant sector ‘Unbound, except as indicated in
horizontal commitments’.
70 For example, Canada limits the entry of professionals to foreign legal consultants, urban planners and senior
computer specialists.
71 The United States of America, Schedule of Specific Commitments, WTO Document GATS/SC/90.
72 Brazil, Schedule of Specific Commitments, WTO Document GATS/SC/13.
49
73 In addition to persons engaged in specialty occupations, the US commitment covers fashion models who are of
distinguished merit and ability. There are no further requirements or specifications and it is unclear whether
fashion models are to be engaged as independent professionals or under employment contracts.
74 Brazil, Schedule of Specific Commitments, WTO Document GATS/SC/13.
75 European Communities and their Member States, Schedule of Specific Commitments, WTO Document
GATS/SC/31.
76 Other schedules reviewed do not include CSSs, but the category of ‘professionals’ in the schedules of Japan
and Canada seems to include persons who may be in an employment relationship in the sending country and thus
in a situation similar to that of CSSs. See WTO Documents GATS/SC/46 (Japan) and GATS/SC/16 (Canada).
77 WTO Document TN/S/O/EEC/Rev.1 of 29 June 2005. The Doha Round offers are not otherwise reviewed as
part of this study.
78 See especially Case C-113/89, Rush Portuguesa [1990] ECR I-1417 and Case C-43/93, Raymond Vander Elst
v Office des migrations Internationales [1994] ECR I-3803.
79 Joined Cases C-307/09 to C-309/09, Vicoplus [2011] ECR I-00453.
80 For example, the Temporary Agency Work Directive of the EU requires equal treatment of temporary agency
workers. Workers hired out by temporary work agencies must be assured the working and employment
conditions that would have applied if they had been recruited directly by the user undertaking to perform the
same work. Directive 2008/104/EC of the European Parliament and of the Council of 19 November 2008 on
Temporary Agency Work.
81 An interesting issue is whether the national treatment obligation of the GATS actually requires wage parity. If
foreign service suppliers are to be treated similarly to domestic employees, all domestic labour laws would need
to be applied. Winters mentions that the exclusion of foreign service suppliers from the host state’s social
security systems may have to be registered with the WTO as exception to national treatment (Winters 2008,
516). One can, however, also conclude that conflict-of-law rules are part of the host state regulations and should
therefore be normally applied when a specific situation points towards the application of the laws of a different
state.
