The EU as an Effective Trade Power Strategic Choice of Judicial Candidates in the Context of the World Trade Organization

Working Paper No 2013/28| June 2013

The EU as an Effective Trade
Power? Strategic Choice of
Judicial Candidates in the
Context of the World Trade
Organization
Manfred Elsig
Abstract
Drawing on the literature on effectiveness, this paper studies how the European Union
chooses judges to serve on the World Trade Organizations’ key judicial institution: the
Appellate Body. Conceptually, the paper differentiates between effectiveness in
representation and effectiveness in impact. The paper shows how delegation to the
European Commission has increased the strategic agenda-setting power for championing
its preferred candidates. The paper further compares European and US practice in
nominating candidates. Overall, the paper finds that effectiveness in representation has
increased. In terms of effectiveness in impact, the paper shows how the international
environment conditions EU influence. The paper also exhibits the difficulties to study
effectiveness of EU external relations due to the peculiar decision-making processes
dominant in judicial bodies.
Keywords: International Organization, Legalization, Court, World Trade Organization,
European Union, Foreign Policy, Effectiveness.
NCCR TRADE WORKING PAPERS are preliminary documents posted on the NCCR Trade
Regulation website (<www.nccr‐trade.org>) and widely circulated to stimulate discussion
and critical comment. These papers have not been formally edited. Citations should refer
to an “NCCR Trade Working Paper”, with appropriate reference made to the author(s)


The EU as an Effective Trade Power? Strategic Choice
of Judicial Candidates in the Context of the World
Trade Organization

(Forthcoming: International Relations)

Actor, power and effectiveness
The European Union in trade politics: Actorness from day one
In the area of external trade the European Community (EC 1) has been
an influential international actor from its very beginning. As Member
States decided to create a Common Market and a functioning Customs
Union, the step of delegating sufficient power to the Community
institutions in order enable them to speak with ‘one voice’ in trade
negotiations was a logical act. Not only was there a need to coordinate
foreign economic policies among Member States, but in the area of
trade policy it was necessary to pursue a single policy on the treatment
of imports and exports of goods (in particular administrating a single
tariff system and implementing uniform product standards).
The EC quickly became accepted as a heavyweight in
multilateral trade talks conducted in the context of the General
Agreement on Tariffs and Trade (GATT). The EC was recognized as a
trade power and its presence in the 1960s, defined by the influence it
exerted on other actors, was felt instantly. While Member States
continue to be official members of the GATT and its successor
organization, the World Trade Organization (WTO), to this day, other
contracting parties quickly accepted the European Commission as the
representative voice and chief negotiator of the Community in the
respective trade rounds (eg, the Dillon Round (1960–61) and the
Kennedy Round (1964–67)).2 Therefore, the mainstream trade literature
did not pay particular attention to concepts related to actorness that

1

I refer to EC when explicit reference is made to the pre-Maastricht period,

otherwise I use the term European Union (EU).

2

Michael Johnson, European Community Trade Policy and the Article 113 Committee
(London: The Royal Institute of International Affairs, 1998); Manfred Elsig, The EU’s
Common Commercial Policy: Institutions, Interests and Ideas (Aldershot: Ashgate
Publisher, 2002).
2


developed in the foreign policy literature. 3 Today, the EU has achieved
(facilitated by various rounds of enlargement) the status of the most
powerful trading entity as measured in terms of overall trade flows in
goods and services and the market access it can offer to foreign
products.
Notwithstanding the strong presence of the EC (and later the EU)
as a trade actor, there was surprisingly little research into the
functioning and the effects of trade policy until the second half of the
1990s.4 In addition, many authors started to place EU trade policy in a
comparative perspective in order to avoid falling into some romantic
N=1 trap.5 To sum up, a growing number of students of European trade
policy accepted the similarities to statehood and questioned a
potential sui generis character of the polity.
What type of trade power?
In recent years, a growing number of scholars have studied extensively
the EU’s position in various regulatory venues (multilateral, regional,
bilateral) as well as its use of unilateral trade measures (eg in the area
of anti-dumping or related to preferential schemes such as duty-free
access for certain products from least-developed countries). As the

study on EU trade policy had a late start, the younger generation of
scholars was strongly socialized within an international relations
literature that took domestic politics seriously. Research focused on
societal preferences and the effect of the institutional setting on
preference aggregation. Various
middle-range theories from
comparative politics and international relations were successfully
applied in the field (eg, principal–agent theory, historical
institutionalism, or pluralism). Interestingly and ironically, discussions
3

Arne Niemann and Charlotte Bretherton, ‘EU External Policy at the Crossroads: The
Challenge of Actorness and Effectiveness’, this issue.

4

Stephen Woolcock and Michael Hodges, ‘EU Policy in the Uruguay Round’, in Helen
and William Wallace (eds.) Policy-Making in the European Union (Oxford: Oxford
University Press, 1996), pp. 301–24; Sophie Meunier and Kalypso Nicolaïdis, ‘Who
Speaks for Europe? The Delegation of Trade Authority in the European Union’,
Journal of Common Market Studies, 37(3), 1999, pp. 477–501; Alasdair Young, ‘The
Adaptation of European Foreign Economic Policy: From Rome to Seattle’, Journal of
Common Market Studies, 38(1), 2000, pp. 93–116.

5

Dirk De Bièvre and Andreas Dür, ‘Constituency Interests and Delegation in
European and American Trade Policy’, Comparative Political Studies, 38(10), 2005,
pp. 1271–96; see also Arlo Poletti and Dirk De Bièvre, ‘The Political Science of
European Trade Policy: A Literature Review with a Research Outlook’, Comparative

European Politics, 2013, forthcoming.
3


on civilian or normative power, while prominent in other foreign policy
fields, were largely absent among students of trade policy, even
though the raison d’être for these concepts relied heavily on the
attractiveness of Europe’s economic success. With the exception of a
few scholars using critical political-economy approaches (in the
tradition of neo-Marxist and dependency theories), the field has hardly
focused on (normative) concepts related to the nature of EU trade
policy.
One of the few notable attempts to capture EU trade policy in
terms of a power concept was the contribution by Meunier and
Nicoläidis.6 While these authors call the EU a ‘conflicted trade power’,
their contribution adds to the perception that the EU is a normal, statelike trading power whose policies are shaped by domestic interests and
that, given a multitude of foreign policy objectives, a lack of coherence
(‘conflicted outcomes’) is a normal phenomenon. They argue that
‘the EU is conflicted within, as different member state
governments, influenced by a host of domestic actors, hold very
different views on how to wield such power through trade. The
EU is also conflicted between its own guiding principles, which
often appear to contradict one another – such as championing
multilateralism while blanketing the planet with bilateral trade
agreements, or promoting the cause of economic development
while protecting European agriculture.’7
The internal complexity of the EU exacerbates the lack of a grand
design or overall objective and makes trade policy appear to be simply
a negotiated result of private groups with vested interests, Member
States’ (changing) preferences and (more or less) activism by the

European Commission.8 Meunier and Nicolaïdis view the observed
trade policy outcome more as a result of efforts to replicate internal
market liberalization than crude forms of domination. Put differently,
the power pattern we witness is less a realist-type of ‘trade as foreign
policy tool’, than a result of the path-dependency linked to the decision
to create an internal market that should be ‘externalized’. In the end,
their concept casts doubt about the possibility of clearly labelling the
6

7

8

Sophie Meunier and Kalypso Nicolaïdis, ‘The European Union as a Conflicted Trade
Power’, Journal of European Public Policy, 13(6), 2006, pp. 906–25.
Meunier and Nicolaïdis, ‘Conflicted Trade Power’, p. 907.
This supports the point made in the introduction that ‘studies have exposed the
lack of (genuinely) normative intentions/commitment’ (Niemann and Bretherton,
‘Challenge of Actorness and Effectiveness’).
4


EU in terms of an exclusive ‘type’ or ‘category’ of power and therefore
they add to those voices that stress the limitations of a ‘what sort of
power’ debate in respect to trade policy.
Effectiveness in trade policy
While measuring the effects of power in trade policy, as in other fields,
is a difficult task, there is consensus in the literature that the EU has
not been ineffective. In this sense, I disagree with Smith and concur
with Ginsberg when focusing on trade policy. 9 There is enough

evidence that the EU has shown various forms of effective
representation. Effectiveness, however, is linked to the question of
institutional set-up. The development of the institutional machinery of
EU trade policy and the constitutional changes needed to switch from
the intergovernmental approach (mixed competence) to the
community approach (exclusive competence) were significantly
affected by debates about effectiveness. The recurring demands by the
European Commission in the run-up to intergovernmental conferences
to be delegated additional competence in trade policy were always
framed within a discourse centring on the lack of ‘effective’
representation.10 The drivers for increasing the EU’s competence were
on the one hand largely external, as the nature of trade politics shifted
from tearing down high tariffs to regulating trade-related concerns
ranging from intellectual property rights, public procurement, and
product standards to newer areas including investment liberalization,
competition policy, taxation or environmental protection. On the other
hand, a growing number of EU Members made consensual decisionmaking and shared competence untenable as a means to aggregate
preferences internally.
This article mainly focuses on effectiveness and addresses three
questions outlined in the introduction.11 First, how can ‘effectiveness’
be evaluated? Second, how does the external environment condition
EU effectiveness? And third, how does the EU’s effectiveness compare
9

Hazel Smith, European Union Foreign Policy: What Is It and What It Does (London:
Pluto, 2002); Roy H. Ginsberg, The European Union in International Relations:
Baptism by Fire (Lanham, MD: Rowman and Littlefield, 2001).

10

Arne Niemann, ‘Conceptualising Common Commercial Policy Treaty Revision:
Explaining the Stagnancy and Dynamics from the Amsterdam IGC to the Treaty of
Lisbon’, European Integration online Papers, 15, Article 6, 2011; Elsig, The EU’s
Common Commercial Policy.

11

Niemann and Bretherton, ‘Challenge of Actorness and Effectiveness’.
5


with that of other important powers? In addressing these questions, I
distinguish two dimensions of ‘effectiveness’: ‘effectiveness in
representation’ and ‘effectiveness in impact’. Effectiveness in
representation suggests that those speaking on behalf of the EU are
able to aggregate the different demands into a unified position. The
emphasis here is on avoiding a multitude of views being signalled
externally and abstaining from acting in an uncoordinated fashion.12 In
other words, the focus is on the nexus between the intra-EU
institutional setting and the ability of the European Commission to act
as the ‘single voice’. The idea of ‘effectiveness in impact’ can be
equated with ‘goal attainment’.13 In the EU context, the objective could
be defined as the pursuit of European societal interests at large (as
opposed to short-term gains one-sidedly pushed by special interest
groups or veto players among the Member States). Overall EU impact
(attaining its stated policy goals), however, is conditioned by two
factors: first, the degree of EU coordinated action, and second, the
external context. In other words, reaching a defined objective
(effectiveness in impact) is dependent upon effective representation
and is conditioned by external factors, such as the behaviour and

interest constellation of other states and the institutional setting at the
international level (eg, decision-making procedures).
In order to address the first two questions, the empirical part of
this article focuses on the effectiveness of the EU in nominating
candidates to serve in one of the most legalized international court-like
institutions: the World Trade Organization’s Appellate Body (hereafter
AB). It addresses the performance of the EU in selecting its own
candidates. The case study that follows covers the period from 1995 to
2009. During that period the external environment remained largely
stable (the EU continued to be a key trade power), 14 while the EU’s
internal-set up underwent significant modifications. This allows us to
12

Such a conceptualization fuses two of the elements proposed to capture actorness:
‘output cohesion’ and ‘autonomy’, see Joseph Jupille and James A. Caporaso,
‘States, Agency and Rules: the European Union in Global Environmental Politics’, in
Carolyn Rhodes (ed.) The European Union in the World Community (Boulder, CO:
Lynne Rienner, 1998), pp. 213–29 ; see also Niemann and Bretherton, ‘Challenge of
Actorness and Effectiveness’.

13

Oran R. Young, International Governance: Protecting the Environment in a
Stateless Society (Ithaca, NY: Cornell University Press, 1994).

14

India and Brazil have been active members in the GATT/WTO system since the late
1980s. The Russian Federation joined the organization in 2012. Only, the impact of
China that acceded to the WTO in 2001 has been increasing in recent years.

6


draw conclusions about the nexus between internal institutions and
external outcomes. The case study shows that delegation to EU
institutions (the European Commission) increased effectiveness in
representation; the record for effectiveness in impact is, however,
more ambiguous. This is largely a result of the conditioning factors and
in particular is a reflection of the nature of courts. Finally, we discuss
effectiveness with respect to the US experience in nominating
candidates to address the third question.

Empirical discussion
The empirical part first outlines some of the key features of the WTO’s
dispute settlement system with special focus on the appeals
institution. It then describes the key processes of nominating and
selecting judicial candidates and outlines the strategic options for
powerful trade actors. This is followed by a discussion on how effective
the EU has been over time (and in comparison with the US) relying on
the above conceptualization.
Appointment of judicial candidates at the World Trade
Organization
During the Uruguay Round talks, negotiators agreed on a drastic
reform of the system for settling trade conflicts. 15 In the transformation
from a system of diplomatic bargaining to a highly legalized institution
one of the things that stands out is the creation of an appeals
institution: the Appellate Body (AB).16 This new court-like institution
makes final recommendations that de facto can not be overturned. The
AB consists of seven members who are elected for a term of four
years, renewable once. Three AB Members (ABMs) serve on individual

cases brought to the AB (called division), following a secret rotation
principle which makes it impossible for WTO ‘contracting parties’ to
15

Thomas Bernauer, Manfred Elsig and Joost Pauwelyn, ‘The Dispute Settlement
Mechanism: Analysis and Problems’, in Amrita Narlikar, Martin Daunton, and Robert
M. Stern (eds.) Oxford Handbook on the World Trade Organization (Oxford: Oxford
University Press, 2012), pp. 485–506.

16

Other reform steps include the ‘right to a panel’ (eg, agreement to abandon the
option to block the launching of a case), the automatic adoption of reports (eg, the
losing party can no longer block a report), an institutionalized ‘decentralized’
system of sanctioning, where the winning party is granted the authority to apply
sanctions against the losing party if the latter does not comply with the reports’
recommendations.
7


anticipate the composition of the AB for a specific case.17
The role of the AB in interpreting WTO law has increased over the
years fuelled by the lack of progress in trade negotiations (the
legislative arm of the WTO). Many contentious cases have been
brought to the AB and by the end of 2011 the AB had issued 105
reports as last instance. By default the AB has engaged in precedentsetting and has closed some contractual gaps which parties (in
particular losing parties) object to as judicial rule-making. As members
of international courts, such as the WTO, are difficult to control once
they are selected, WTO Members pay particular attention to the
selection of new ABMs.18

In 1994, a preparatory Committee was mandated to provide
guidance as to the implementation of the negotiated provisions. 19 It
suggested a two-stage system of nomination and selection of
candidates. While WTO Members nominate candidates (1 st stage), a
group of six officials (comprising the Director-General (DG) and the
Chairpersons of five important WTO bodies (General Council, Dispute
Settlement, Goods, Services, and Trade-related Intellectual Property
Rights)) select candidates who are subject to being accepted by
consensus by the Membership (2 nd stage). Against the background of
approval by consensus, this Committee ‘consults’ closely with
Members throughout the process. Over time, selection of candidates to
serve on the AB has become more politicized and the limited wiggle
room for the Group of Six (G6) has further decreased. The first
selection process concerned all seven ABMs at the same time, which
allowed the G6 some strategic options (as all seven ABMs had to be
accepted or rejected as a group). Over time, the selection processes
have de facto transformed into single seat competitions. As a result of
politicization, candidates today spend much more time in bilateral
meetings in Geneva (and increasingly travel to Washington, DC and
Brussels) where they undergo a careful screening process as Members
attempt to find out the exact preferences of the candidates.
The WTO AB is a ‘selective court’ as opposed to a
‘representational court’ where each country can nominate judges who
17

Given that the EU and US are often involved in disputes (as complainant or
defendant), it is not unusual to find the US or the EU ABM is sitting on a division
that hears such cases.

18

Manfred Elsig and Mark Pollack, ‘Agents, Trustees, and International Courts:
Nomination and Appointment of Judicial Candidates in the WTO Appellate Body’,
European Journal of International Relations (forthcoming 2013).

19

WT/DSB/1 (10 February 1995).
8


in turn are accepted by others without much discussion (see for
instance European Court of Human Rights).20 This implies that the
process is competitive and nominating states have to be very strategic
in choosing who they propose at the first stage. The WTO treaties
remain rather general regarding the profile and composition of the AB.
The drafters of the Dispute Settlement Understanding (DSU) agreed
that ABMs ‘shall be broadly representative of membership’ (Art. 17.3
DSU). This was partially clarified by the preparatory Committee in
1994, which suggested that ‘factors such as different geographical
areas, levels of development, and legal systems shall be duly taken
into account’.21 In addition, the expertise was circumscribed to allow
for the resolution of ‘issues of law covered in the panel reports and
legal interpretations developed by the panel.’ 22 This meant that not
only candidates with extensive legal training were qualified to act as
Court members.

The strategic options for powerful states
In terms of representation, the WTO treaties did not suggest that some
Members would have the privilege of having a permanent seat at the

AB table. Yet, from the beginning the US and the EU demanded a seat
and this was not contested by other WTO Members.23 A quasipermanent seat enlarges the agenda-setting power of the parties
concerned. Knowing that a candidate is not in competition with other
countries’ candidates, de facto gives the nominating party the
strategic advantage of proposing a candidate who closely matches its
preferences. Put differently, the nominating states (US and EU) in a
‘two-stage game’ are less constrained in agenda-setting (they have
20

Erik Voeten, ‘The Politics of International Judicial Appointments: Evidence from the
European Court of Human Rights’, International Organization, 61(4), 2007, pp. 669–
701.

21

See Elsig and Pollack, ‘Agents, Trustees and International Courts’; WT/DSB/1 (10
February 1995).

22

23

WT/DSB/1 (10 February 1995).
This acceptance is significant from a presence perspective, see David Allen and
Michael Smith, ‘Western Europe’s Presence in the Contemporary International
Arena’, Review of International Studies, 16(3), 1990, pp. 19–37. The EU and the US
even demanded two seats each in the run-up to the first selection. This demand,
however, was not seen as legitimate by other parties and was not accepted. Yet, it
bolstered the case for a permanent seat (Elsig and Pollack, ‘Agents, Trustees and
International Courts’).

9


less need to incorporate the preferences of the WTO membership).24
Besides the de facto permanent seat, the views of the powerful
states (in particular the US and the EU) are also important when it
comes to appointing the other ABMs. Similar to the US, the EU holds
power of veto over the selections of other candidates, not least
because it has no need to fear that if it behaves uncooperatively in the
selection process in one round (eg blocks a certain candidate) that
countries will veto (all) EU candidates in the next round. While all WTO
Members can potentially veto individual candidates by signalling to the
Chair of the G6 that they will not support that candidate, it is clear that
more powerful states face fewer constraints in actually using this
blocking power.
Against the background of strategic agenda-setting and gatekeeping roles (through the power to block candidates), I argue that the
EU has increased its overall effectiveness. However, it has not yet
developed the same strategic behaviour on the level of agenda-setting
as the US exhibits. The same applies in respect to its influence on the
selection of other candidates. Increased effectiveness in representation
and, partially, in impact is a result of more delegation to the European
Commission in managing this process, as shown below. In the following
section, we focus on the EU and in particular on the nomination of its
own candidates.
The EU: The relationship between delegation and effectiveness
The first selection
In the 17 years of the existence of the AB, the EU has so far been
called upon to fill three vacancies (1995, 2001, and 2009). Below, I
discuss these three individual selections by focusing on developments
in the politics of WTO dispute settlement (external environment), EUinternal changes at the level of processes (internal environment) and

resulting EU effectiveness (representation and impact). The very first
selection process was chaired by Australian Ambassador Don Kenyon.
The G6 (supported by the majority of the Membership) laid emphasis
on selecting highly qualified individuals who were able to build up the
reputation of this novel institution in the context of a member-driven
organization.25 Members of the WTO themselves were not concerned
24

Elsig and Pollack, ‘Agents, Trustees and International Courts’.

25

WT/DSB/M9 (1 and 29 November 1995): the original members of the AB were from
the EU, the US, New Zealand, Egypt, the Philippines, Uruguay, and Japan.
10


about the personal preferences of the candidates; it was largely a
diplomatic contest over which contracting party would get a seat. 26 The
European Commission favoured a highly qualified individual who would
contribute to establishing a legitimate and autonomous dispute
settlement body for the new organization which was under public
scrutiny.27 However, the European Commission played a marginal role
in the nomination of EU candidates. Agenda-setting was decentralized
as the right to make a nomination rested with the EU Member States.
Overall, EU Member States put forward a total of 13 candidates.
Belgium even nominated multiple candidates. Most candidates had a
legal background. It was largely a question of prestige for countries to
get their preferred candidate accepted in Geneva. As the G6 could only
propose one of the EU candidates, it nominated German candidate

Claus-Dieter Ehlermann, who gained the widespread support of the
Membership due to his profile and his performance in the talks with the
Selection Committee. It helped that Ehlermann was a national from an
important EU Member State.28 As he was not the only German
candidate, the chair of the G6 asked the German WTO Ambassador
whether ‘it was ok if we went for Ehlermann; and the Ambassador said
he agreed.’29 Overall the G6 proposed a mix of elderly and experienced
trade practitioners, lawyers and even a former constitutional judge to
serve on the bench.30 While the European Commission was supportive
of the European candidate chosen, after the selection it criticized the
make-up of the AB as a whole for its lack of balance. The European
Commission’s spokesperson complained that five seats had gone to
nationals from APEC countries and one seat to Egypt, which was
another close ally of the US. The European Commission announced
that it would elaborate a new proposal for rebalancing the composition,
26

‘It was political in the sense that countries wanted to be represented; (…) it was a
question of prestige, (…) and ambassadors worked hard for pushing one of their
nationals through’ (Interview, 1 July 2010).

27

The EU Commissioner for External Trade Leon Brittan and the USTR Mikey Kantor
agreed to support candidates who were working for the integrity of the system and
would not represent WTO Members (Interview, 23 September 2010).

28

Interview 4 June 2010; Interview, 7 July 2010. Ehlermann was a former DirectorGeneral of the Legal Service of the European Commission and had acted as

Director-General of the Directorate-General Competition. Around the time of
appointment he also was appointed professor at the European University Institute
in Florence.

29

Interview, 7 July 2010.

30

Elsig and Pollack, ‘Agents, Trustees and International Courts’.
11


to be discussed in the upcoming Singapore Ministerial in 1996, a
proposal that did not see the light of day.31
In summary, the European Commission was pleased by the
selection of a former EU official; however, the European Commission
stood on the sidelines during this first appointment procedure, and its
role was marginal (even though it supported Ehlermann). 32 In terms of
candidates for the AB nominated by other WTO Members, European
states as well as the European Commission lacked influence in the
proceedings as witnessed by the sharp EU criticism following the
selection. To sum up, effectiveness by representation was low.
Effectiveness by impact is more difficult to judge, not least
because the WTO has no records of individual positions taken by
judges on the various cases. This makes it notoriously difficult to
evaluate the performance of single members of the court and one has
to rely on anecdotal evidence. Judging by the prevailing interests of the
leading WTO Members (including the European Commission) in 1995 in

having an ‘independent court’ (goal attainment), Ehlermann was
probably contributing to this objective as he had retired from the
European Commission and was an autonomous thinker. 33 In particular,
the first seven ABMs were not only able to draft the Working
Procedures of the AB, they also actively ‘embraced’ the Vienna
Convention rules on treaty interpretation. Both acts were very strong
signals of building legitimacy.34 At the same time, Ehlermann and his
colleagues invested ample time and resources in establishing a
credible new body that would not shrink from ruling against the
powerful states.35
The second selection
31
32

33

WT/DSB/M9 (1 and 29 November 1995).
Ehlermann himself had good relations with the German State Secretary for
Economic Affairs and former EU trade negotiator Hugo Paeman who encouraged
him to run. He also knew WTO Director-General Renato Ruggiero who served on the
G6 (Interview, 4 June 2010).
Interview, 4 June 2010.

34

Peter Van den Bossche, ‘From Aftermath to Centerpiece: The WTO Appellate Body
and its Rise to Prominence in the World Trading System’, University of Maastricht
Working Paper, Faculty of Law, 1. 2005

35

James Bacchus, Trade and Freedom (London: Cameron May, 2004); Former ABM
Julio Lacarte, public statement at a Conference on ‘Face to Face with AB Members’
organized by the Graduate Institute of International and Development Studies,
Geneva, 27 May 2008.
12


In 2001, the terms of office for the ABMs from the EU, Uruguay and the
Philippines ended.36 By 2001, given an increasing case-load and
important decisions with distributional consequences for the Members,
the attention given to the AB (and therefore to the selection process)
increased. This was in particular reflected in the growing use of the
bilateral screening process where Members tested candidates’
positions vis-à-vis a number of important judicial matters (eg, general
approach to gap-filling, how the AB dealt with unsolicited briefs, and
the growing number of trade remedy cases where the US was strongly
opposing the developing AB jurisprudence). Candidates also started to
travel to the key capitals to meet with capital-based officials.
Candidates with strong views on some of the key issues raised by
Members had a hard time gathering sufficient support.37
The European Commission’s overall preference had not changed
since the first selection; it continued to support an autonomous body.
However, the European Commission attempted to exert more influence
on the selection of candidates. The European Commission invited EU
Member States to cooperate more in the nomination of EU candidates
and urged them to put forward fewer candidates. 38 In addition, the
European Commission managed to invite internal candidates to
Brussels for an informal exchange. However, this incremental change
was not yet a formalized process and took place ‘over a cup of coffee

not taking more than 30 minutes.’39 In Geneva, the EU delegation
offered its services, but the campaigning was left to delegations (and
the capitals) of those EU countries which put forward candidates. 40 The
European Commission continued to maintain a low profile regarding
the use of the diplomatic machinery to influence other Ambassadors in
Geneva regarding their priorities.41
EU Member States continued to be mostly concerned with
diplomatic prestige. This time, the successful EU candidate was
36

These three judges were selected initially for two years and were reappointed for
another four years. The idea was to decouple future selections and not to face the
situation again of having to select seven new ABMs at once. However, the fact that
two ABMs died during their tenure and that some did not seek reappointment (or
were no longer supported by the nominating state for a second term) meant that
over time selections were no longer grouped in large packages.

37

Elsig and Pollack, ‘Agents, Trustees and International Courts’.

38

Interview, 30 August 2010.

39

Interview, 15 July 2010.

40

Interview, 2 June 2010.

41

Interview, 30 August 2010.
13


nominated by Italy (Giorgio Sacerdoti). He was not known in Brussels
and Geneva and he had no former working relationship with DG Trade
unlike other candidates from Belgium and Germany. 42 The successful
candidate was a professor of international economic law with
consultancy experience with the Organisation for Economic Cooperation and Development (OECD). He was also an experienced
arbitrator in investment cases. The other two prominent EU nominees
were not only closer to the European Commission; they were also
experienced trade law experts and former GATT and WTO panellists.
Moreover as academics, they had written widely on the WTO. Both
these candidates were rejected by the US. 43 In the end, many Members
felt that independence from the European Commission was important
as the EU was often involved in legal disputes.44 In addition, Sacerdoti’s
experience as an arbitrator was a sign interpreted by some WTO
Members to mean that the candidate would be less likely to engage in
law-making and be more attentive to parties’ interests in preserving
their negotiated rights and obligations. Finally, the successful
candidate had not published in the field of trade law; therefore he had
not taken a firm position on a trade issue that could potentially offend
a powerful WTO Member.
In a nutshell, the second selection shows the first timid attempts
by the European Commission to be more active in the process for

selection of the EU candidate. Yet, WTO Members did not select the
most desirable candidate from a DG Trade perspective, a Belgian trade
lawyer who had a close working relationship with DG Trade. 45 This time,
the WTO Members did go against the European Commission’s
preferences. As to the European Commission in its role in the selection
of other WTO Members’ candidates, DG Trade became more informally
involved. Yet, even when candidates came to see the EU
representative, the European Commission did not signal its preferences
to the G6 in any meaningful or strong way. 46 Effectiveness by
representation slightly increased as the EU managed to put forward
fewer candidates; however, the selected candidate was clearly not
explicitly endorsed by the European Commission.
42

Interview partners recalled candidates from Belgium, Germany and Italy.

43

Interview, 6 July 2010; Interview, 15 July 2010.

44

Interview, 30 August 2010.

45

Interview, 30 August 2010.

46

Interview, 30 August 2010.
14


Effectiveness by impact is again hard to measure. For once, the
AB gained in prominence over time and WTO Members were quite
vigilant, as witnessed in the infamous amicus curiae brief episode. 47
WTO Members started to limit the autonomy of the AB by voicing
increasing concerns over substantive issues. This observed change of
external ‘environment’ partially explains why the second generation of
ABMs adopted a more cautious approach to the evolving case law
which led to less criticism from the WTO Members. They proved more
pragmatic in their interpretative approach and pushed less towards a
judicial style by building precedents. 48 Overall, the decisions of the AB
continued to be oriented towards finding consensual positions within
the AB with a view to offering long-term guidance on treaty obligations.
The third selection
The political environment for the third selection was characterized by
controversy over how the AB dealt with a number of trade remedy
cases. In particular, the US lamented that an established practice of
calculating dumping margins (the so-called zeroing method) had been
ruled by the AB to be WTO-incompatible. In addition, questions such as
whether to allow dissenting opinions or restriction of the interpretation
space came up during reform discussions that started in parallel to the
Doha trade negotiations in 2001. This was reflected by the increased
interest of WTO Members in screening candidates. As one of the
candidates recalled
‘Members quiz candidates on the role of the dispute settlement
body and WTO jurisprudence, about the role of AB vis-à-vis
negotiated agreement, they have some strong concerns about

filling gaps and the AB making law, they want to learn whether
candidates have experience in dispute settlement and they ask
about specific issues, including agricultural policy or zeroing (…).
One issue that was also important for some Members, was the
issue of dissenting views, whether this should become more
possible in the future.’49
47

WTO Members did not like the position of the AB that unsolicited briefs by nongovernmental organizations were to be accepted by panels, see for instance Petros
Mavroidis, ‘Amicus Curiae Briefs before the WTO: Much Ado about Nothing’, NYU
Law School Jean Monnet Working Paper, 2, 2001.

48

One ABM suggested that the AB was more holistic in its approach towards
interpretation, Interview, 6 June 2010.

49

Interview, 25 June 2010; there are only a few examples where a single ABM would
15


The European Commission’s overall preference for a candidate, wellversed in trade law and socialized within a broader European view, did
not change. Yet, its attempts to manage the process more actively had
become successful in the meantime. In the run-up to the EU
nominations, the European Commission had this time persuaded
Member States of the need to coordinate the internal nomination
process better. In particular, the European Commission argued that the
EU had to increase its ‘effectiveness’.50 Therefore, nominations had to

be discussed in the main Committee for coordination between the
European Commission and the Member States (the Trade Policy
Committee (TPC)). First, nominees from Member States were invited to
attend a formal interview in Brussels with a Committee composed of
officials from DG Trade, the Legal Service, and a representative of the
Presidency.51 The European Commission then ranked the candidates
and made a nomination proposal to the TPC explaining the criteria for
the selection. Through this change of procedure the European
Commission obtained control over who was to be nominated.
For the successor to Giorgio Sacerdoti, the European Commission
nominated two candidates who were approved by the TPC, one from
Belgium and one from the Netherlands. The nomination process,
however, took a bit longer, as the Spanish representative in the TPC
demanded that a third candidate be sent to Geneva. 52 After a couple
of weeks, Spain gave up its opposition. For the Geneva process, the
European Commission adopted a low profile as usual. One candidate
remembered that ‘DG Trade said that they would not take sides – that
would have backlashed and Member States would have been
offended.’53 In addition, the two candidates met and agreed that their
countries would not organize active lobbying in Geneva. 54 Both
candidates went to the numerous bilateral meetings with other WTO
Ambassadors, accompanied by a representative of the European
offer a dissenting or concurring opinion. This has been a deliberate policy of the
first generation of ABMs. However, the pressure on candidates to dissent is
increasing.
50

Interview, 2 June 2010.

51

Interview, 2 June 2010; Interview, 6 July 2010.

52

E-mail exchange, 30 April 2012. Spain also had a candidate who was not ranked in
the top 3. The intervention was an attempt to push for its preferred candidate
knowing that the WTO Ambassadors in Geneva might opt for this candidate.
Interview, 23 April 2012; e-mail exchange 30 April 2012.

53

Interview, 18 June 2010

54

Interview, 18 June 2010
16


Commission.55 The candidates also flew to Washington, DC. Both had
quite similar profiles. The European Commission favoured having WTO
insiders on the bench. Unlike Sacerdoti, both nominees were well
known trade lawyers and had gained insights into the WTO by working
for the AB Secretariat and the Legal Division of the WTO respectively.
One of them was even involved in the negotiations on the creation of
the DSU during the Uruguay Round. It seemed clear that the European
Commission wanted a strong personality with extensive expertise in
WTO law who could shape the court’s deliberations. The key difference
was that the candidate from the Netherlands was perceived to be

much closer to the European Commission as he had worked for many
years for the Commission.
The Members opted for Peter van den Bossche from Belgium.
Similar to the 2001 EU selection, a key factor was that the successful
candidate had not worked for the European Commission and was not
perceived as a ‘Commission candidate’.56 In addition, the successful
candidate received a lot of support from developing countries as he
had provided extensive technical assistance and training in the past.
The US was not happy with either of the candidates and China had
some concerns with the successful candidate, though both abstained
from blocking.
To sum up, in the most recent nomination and selection process,
the European Commission took over the control and management of
the internal nomination. This allowed the European Commission to field
a candidate who it believed represented the European Commission’s
views on the role of the AB. Given the information about the
candidates, the problem of adverse selection was also minimized. 57 In
the third selection, the European Commission did not fully attempt to
use its strategic agenda-setting by nominating only one candidate.
This could be explained by ‘established practice’ in the case of the US
and the EU to allow some choice to the Membership and to offer an
alternative should Members not support one of the candidates.
This third case shows that effectiveness in representation was
increased. Both candidates were acceptable to the European
Commission, highly competent in WTO law, and it was not possible to
divide the European position. In regard to effectiveness in impact, it is
55

Interview, 18 June 2010

56

Interview, 18 June 2010; Interview, 2 June 2010; Interview, 5 July 2010.

57

Adverse selection exists when principals (Members) choose the wrong agent
(candidate) who is able to hide his true intentions before delegation takes place.
17


too early to judge how the European ABM will influence the debates
within the group of seven ABMs. To sum up, overall effectiveness (in
particular in representation) seems to have been improved as a result
of allowing the European Commission to coordinate the process and
act as a gate-keeper in the internal nomination game.
Comparing effectiveness: The case of the US
As outlined above, agenda-setting power depends whether a state has
de facto a permanent seat (US, EU, and potentially China in the
future)58 and whether it can push through its preferred candidate even
against the will of others. Evidence suggests that the US has been able
to follow its preferred strategy since the creation of the AB, 59 whereas
the EU (and the European Commission) has only recently been able to
achieve a similar strategic position. At the beginning the choice was
determined by actors outside Brussels’ influence.
If we focus on how preferences developed in the context of the
US nomination procedures, we witness important changes over time. In
the first selection procedure, the US (and in particular Mikey Kantor,
United States Trade Representative (USTR)) shared the widespread
concern for selecting a first group of ABMs that were characterized by

strong personalities reflecting both expertise in law and diplomacy.
USTR had a preference for one of its own candidates, a former Member
of Congress who helped to push the WTO treaties through US
ratification (James Bacchus).60 USTR was also concerned that the AB
would have to deal with the political realities of trade policy and
therefore needed to have an excellent first group of ABMs. The USTR
Mikey Kantor sat down with EU Commissioner for External Trade Leon
Brittan and WTO Director-General Renato Ruggerio to find candidates.
They agreed that ‘the WTO dispute settlement system was the most
important part and that needed to be protected.’61 Yet, USTR also
anticipated that the US would be more on the winning than on the
losing side.62 The US deliberately put forward two candidates neither of
58

Japan has lost this privilege (strictly speaking it never had a permanent seat);
Brazil and India rotate with other countries in their respective regions.

59

Elsig and Pollack, ‘Agents, Trustees and International Courts’.

60

The other US candidate was a well-known academic in the field of international
economic law, but not a WTO insider.

61

Interview, 23 September 2010.

62

Interview, 23 September 2010.
18


whom would have any issue with ruling against the US if necessary.
Over time, the concerns of the US, related to evolving case law,
grew. In order to address this US-internal worry, USTR proposed, in the
second round of selections, one candidate who was a critic of the AB
and another candidate who was a bit closer to USTR preferences.
However, USTR anticipated that the Members would abstain from
selecting an AB critic, so it was de facto a single candidacy. The WTO
membership, as widely expected, opted for the candidate who was
more in line with the overall orientation of the system. 63 However, it
turned out that USTR was not happy with the performance of the new
US ABM (Merit Janow). She sat on a number of trade remedy cases that
went against the US. Ultimately, she did not seek re-appointment after
the first four years, so 2007 saw the start of a new selection process. In
the third round of nominations, USTR put forward two former USTR
negotiators. The choice of the Membership was again very limited.
While the Membership expressed concerns about the strong links to
USTR, a former textile negotiator, USTR chief legal council and member
of the US International Trade Commission was finally accepted (Jennifer
Hillman). As the US more openly encouraged ABMs to use the option to
dissent, the US ABM reacted to these calls and wrote in one important
trade remedy case a subtle dissenting view. After only one term, USTR
officially blocked the re-nomination of Jennifer Hillman in summer
2011. They were not entirely happy with her performance and started
a new internal nomination process where the need for dissenting views

in AB reports has become even more openly advocated.64
To summarize, the US was more effective (in representation) than
the EU in getting their preferred candidates selected. For the EU it took
longer and a delegation to the EU level was necessary. The EU process
shifted from a decentralized, Member States-dominated one to a more
streamlined Commission-orchestrated approach. The strategic gap was
observable in the first two selections. While, in the first selection,
major WTO powers had a shared understanding of the need to have
independent judges, the EU candidate was in line with EU preferences.
In the second selection, within a changing context, the European
Commission ended up with a little-known candidate, which represented
some risks. Finally, by gaining significant control over the process, in
the most recent selection DG Trade was able to present two candidates
who were largely in line with EU preferences.
63

Elsig and Pollack, ‘Agents, Trustees and International Courts’.

64

Inside US Trade, 9 September 2011.
19


Effectiveness in impact is more difficult to evaluate and compare.
Like the EU, the US can only nominate one national to sit on the bench.
While it tries to influence the choice of other ABMs, it is hard to gauge
the effects of its intensified screening over time. Its discourse on the
AB filling contractual gaps and the will of the negotiators has, however,
led to the AB exercising more care in using expansive interpretation

methods, including building precedents.65

Conclusion
The article conceptualizes effectiveness by differentiating between
effectiveness in representation and effectiveness in impact. The article
shows empirically that with increasing delegation to the European
Commission effectiveness in representation in particular has increased.
Over time, by centralizing the nomination process, the European
Commission has achieved more strategic leeway to put its preferred
candidate forward. This probably affects effectiveness in impact. Yet,
as the case study illustrates, effectiveness in representation is
dependent largely upon internal institutional factors, whereas
effectiveness in impact is conditioned by the external environment. In
this case, the external environment is a group of seven AB Members
who develop, mostly by consensus (and with great caution), case law.
A major difficulty in measuring the impact is the nature of legal
processes which are buffered widely from public attention due to the
secrecy surrounding internal deliberations. In addition, the few
dissenting views do not offer sufficient observational data points for
studying individual behaviour. This might change in the future due to
pressure from the US and other countries. However, given the high
degree of politicization of the selection process, the US nominees are
put in an uncomfortable position knowing that the demands from USTR
and the demands from the AB as a collegial team are more and more
difficult to reconcile. The US hire-and-fire attitude towards its AB
members could paralyse the work of the AB in the future more than
USTR might want; thereby contributing to ineffectiveness of the body it
had once pushed to create. Put differently, if USTR achieves its shortterm objectives by appointing a judicial agent close to its preferences,
this might complicate the inner workings of the AB, lead to more
65

Richard Steinberg, ‘Judicial Lawmaking at the WTO: Discursive, Constitutional and
Political Constraints’, American Journal of International Law, 98(2), 2004, pp. 247–
75; Elsig and Pollack, ‘Agents, Trustees and International Courts’.
20


conflicts and undermine longer-term objectives of a stable predictable
system by weakening the legitimacy of the AB.
Finally, will the Lisbon Treaty challenge the current EU internal
nomination process? It is quite possible that the European Parliament,
given its new prerogatives, reflected for instance in co-decision powers
in trade negotiations, will also demand more influence over the
nomination of EU candidates. At a time when public actors are
increasingly subject to democratic control, the European Parliament
might be tempted to use its privileged position to demand a say in the
internal nomination process. This would most likely weaken
‘effectiveness in representation’ as more voices will favour the search
for acceptable candidates rather than the most suitable candidates,
curtailing the strategic agenda-setting power gained by the European
Commission in recent years. However, given that final decisions over
nomination are still in the hands of EU Member States and that the
European Commission is now dominating the process, these EU
institutions will not easily give in to parliamentary participation. In
addition, the Parliament is currently more interested in influencing the
course of bilateral and regional trade agreements, as exemplified by
the debates on ratification of the EU–Korean free trade agreement. 66
The distributional consequences of these negotiations automatically
translate into greater process involvement by the directly elected
representatives of the EU’s citizens.

66

Manfred Elsig and Cédric Dupont, ‘European Union Meets South Korea:
Bureaucratic Interests, Exporter Discrimination and the Negotiations of Trade
Agreements’, Journal of Common Market Studies, 50(3), 2012, pp. 492–507.
21