http://eiop.or.at/eiop/texte/2004-018a.htm
: 12.10.2004| This paper's comments page | Send your comment! to this paper |
| Treaty-Making in the European Union: Bargaining, Issue Linkages, and Efficiency (*) |
| Andreas Dür and Gemma Mateo |
| European Integration online Papers (EIoP) Vol. 8
(2004) N° 18; http://eiop.or.at/eiop/texte/2004-018a.htm |
| Date of Publication in
: 12.10.2004 |
| | Abstract |
Back to homepage |
PDF | | This paper's comments page | Send your comment! to this paper | |
1
In the course of the last twenty years, the European Union (EU) has engaged in six intergovernmental conferences (IGCs) with the aim of revising its founding treaties. A comparison of the results of these IGCs indicates that member state governments’ success in achieving agreement on thorny issues by way of substantial compromises differs across these cases. Whereas in some IGCs governments manage to include all issues on the negotiating table in an overall package deal, in other cases decisions on important issues are postponed until a later negotiation. The continuum reaches from the Treaty of Nice (2003), which had to be renegotiated immediately, to the highly efficient cases of the Single European Act (1987) and the recent Constitutional Treaty (awaiting ratification). How can one explain this variation in the efficiency of intergovernmental bargaining in the EU?
We propose that for an IGC to result in a far-reaching agreement, a set of conditions has to be given that help overcome the problem that in a bargaining situation all actors maximise individual gains rather than overall efficiency. In particular, the existence of a broad negotiating agenda and a trusted actor’s mediation during the negotiations are essential pre-conditions for efficient bargaining results based on issue linkages. In IGCs, EU member states established two procedures to facilitate the achievement of these conditions: a phase in which a preparatory group sets out possible options for a package deal precedes the actual negotiations, and the Council presidency acts as a mediator in the negotiations. Factors such as lack of time for preparation, a biased presidency, or external shocks that reduce the capacity of the presidency to guide the negotiations, however, can cause these procedures to fall short of achieving their objective, and thus to efficiency losses in IGCs.
2
In an analysis of all IGCs from the Single European Act (negotiated in 1985) to the recent Constitutional Treaty (2003-04), we find support for the resulting hypothesis that variation in the effectiveness of the preparation and of the mediation by the presidencies can explain at least a part of the observable variation in the efficiency of the IGCs.
Our argument contributes a novel perspective to an existing debate concerning negotiation efficiency between two important lines of research. On the one hand, supranationalist bargaining theory claims that intergovernmental bargaining often is inefficient because states lack expertise in technical and multi-party negotiations (Young 1999; Christiansen 2002). Governments thus need the intervention of supranational actors to achieve efficient bargaining results. On the other hand, Andrew Moravcsik (1998; 1999), who is the main advocate of intergovernmental bargaining theory, maintains that bargaining in IGCs tends to be efficient because state preferences are transparent. He argues that while expertise is abundant in IGCs, governments often fail to achieve cross-sector issue linkages because they are reluctant to impose concentrated costs on domestic constituencies even if these costs are balanced by gains for other groups.
In developing our argument, we build on both strands of research: we accept intergovernmental bargaining theory’s emphasis on the importance of state actors, but add the scepticism of supranational bargaining theory concerning governments’ capacity to achieve efficient agreements. At the same time, by stressing the importance of the preparatory phase and the mediation provided by the presidencies, we avoid some inconsistencies of these two existing theories. Supranationalist bargaining theory fails to take into account that the possibility of agent shirking should make governments cautious to accept the intervention of supranational actors in negotiations with high stakes. In addition, the theory largely leaves unclear why supranational actors should have more expertise than national officials. Intergovernmental bargaining theory, in contrast, relies on the assumption that the resistance points (the points beyond which parties reject an agreement) of all parties are well known, an assumption that is implausible because under such circumstances bargaining should hardly be observable. As a result, it ignores the possibility that due to distributional bargaining governments may fail to achieve overall efficiency even if expertise is abundant.
Besides providing some new insights to this important debate between supranational and intergovernmental bargaining theory, the paper also contributes to a growing literature on the role of the presidency of the Council of Ministers in EU decision-making (Colm 1985; Svensson 2000; Elgström 2003; Tallberg 2003). We build on this literature when stressing the importance of the presidency, but go beyond it when specifying under which conditions a presidency can increase the efficiency of intergovernmental bargaining, and when conducting a test of our argument across a large number of cases.
In the next section, we identify the puzzle underlying this paper by showing that the efficiency of IGCs varies across different cases. We then present our argument for why the preparation of the negotiations and the mediation provided by the presidencies are essential for bargaining efficiency. In the subsequent part, we introduce substantial empirical evidence that supports our contention that the success or failure of IGCs can at least in part be explained with reference to the effectiveness of the preparation and the mediation by the presidencies. We conclude with some implications of our argument for both the scholarly understanding of intergovernmental bargaining and practitioners’ attempts to reform the institutions of the EU.
3
Since the mid-1980s, EU member states have concluded six IGCs, namely the one preceding the Single European Act (1985), the one on Economic and Monetary Union (EMU, 1991), the one on Political Union (1991), and the ones preceding the Treaties of Amsterdam (1996-97) and Nice (2000), and the Constitutional Treaty (2003-04). In the following, we propose that not all of these IGCs were equally efficient; that is, not always did the governments of the EU member states manage to reach the Pareto frontier and exploit all potential gains from cooperation.
As a first step towards this argument, it is necessary to define and operationalise the term efficiency. In our usage, efficiency is an overall measure that indicates whether a group manages to maximise its joint utility function in an agreement, i.e. whether it reaches the potential welfare frontier. Efficient agreements clear the negotiating table by linking issues together on which the win sets of the participants do not intersect, while inefficient ones simply lead to the omission of controversial issues. The failure of a compromise on a specific issue because of one or two reluctant parties probably means that the participants could not reap all gains from cooperation. In such a situation, efficiency would require a deal that links several such thorny issues in a package, and thus increases joint utility. This is not to say that a package deal should encompass all issues. The inclusion of an issue in a negotiated agreement that provides gains for only a few participants, but imposes costs on many, actually reduces the efficiency of a final agreement. The reason for this is that the losses for the opposing member states are likely to outweigh the gains that the few supporters could reap from an agreement on this issue.
Even if stringent, this definition creates the problem that a direct measurement of efficiency is impracticable. Such measurement would require information about the utility functions of all governments across all issues that possibly could be included in an agreement, an objective that can hardly be reached. As an alternative to direct measurement, we employ two proxies to estimate efficiency across our six cases. First, we deem an agreement less than efficient if it excludes issues on which a majority of participants expressed an expectation of gains from cooperation during the negotiations. Second, since efficient agreements satisfy a large majority of governments, they should remain stable in the medium to long term. Unstable agreements, on the contrary, are an indication of compromises at the lowest common denominator, which leave many governments unsatisfied, and thus before long require the renegotiation of some questions.
Based on a systematic analysis of these two proxies, we argue that three of the six most recent IGCs were efficient: the IGCs of 1985, on EMU, and of 2003-04. Concerning the first two cases, the resulting agreements resolved all major issues, and the compromises achieved have been stable ever since. The IGC of 1985 allowed for the implementation of the Single Market Programme, and the agreement on EMU already set out the stages for the adoption of a common currency more than ten years later, without requiring further treaty revisions. The IGC of 1985 also managed to tackle institutional issues, such as the extension of qualified majority voting to all decisions linked to the Single Market Program, and the inclusion in the framework of the European Communities of the provisions on European Political Cooperation.(1) The negotiations on EMU succeeded in resolving the sensitive issue of setting criteria for economic convergence including obligations to reduce government deficits and to maintain stable interest rates. The Constitutional Treaty, which, after a failed attempt to conclude the negotiations in the Brussels summit in December 2003, was approved in June 2004, also does not list leftovers. In addition, the negotiators’ immediate reactions at least indicate that in the next years it will not have to be renegotiated. Following our criteria for efficient bargaining, we classify these three negotiations as highly efficient.
4
In comparison with these three cases, we consider the IGCs on Political Union and of 1996-97 to be only medium efficient. Although each of the two negotiations achieved major compromises, they did not manage to settle some of the more important questions. The Treaty of Maastricht (1992), which resulted from the IGC on Political Union, already made explicit that the governments of the EU member countries would have to renegotiate the provisions on a common foreign and security policy in the next IGC to be convened in 1996. Further leftovers expressly mentioned in the treaty are the co-decision procedure, energy, tourism, civil protection, and the simplification of the legislative process. Similarly, the Treaty of Amsterdam, which resulted from the IGC of 1996-97, left questions open for revision in a future IGC. The proposal of the Dutch presidency actually was to “freeze” the negotiations on some issues until enlargement of the EU would be imminent (El País, 16 June 1997: 5). In a protocol to the treaty, which has been interpreted as showing that the results of the negotiations were not satisfactory (Yataganas 2001: 7), the participants agreed that there should be new negotiations on the issues left unresolved no later than one year before the next enlargement.
The IGC of 2000, which resulted in the Treaty of Nice, is arguably the most inefficient of the six cases considered. As a leftover, it postponed the decision concerning the size of the European Commission after enlargement to a vote by unanimity within the Council of Ministers after the EU would reach the number of 27 member states. In addition, in a declaration annexed to the Treaty of Nice, the governments agreed to leave four topics for a future IGC: the delimitation of powers between the EU and the member states, the legal status of the Charter of Fundamental Rights, the simplification of the treaties, and the role of the national parliaments in EU decision making. This list clearly indicates that a series of issues remained as leftovers of the negotiations. Moreover, although the negotiators achieved compromises on such issues as the reweighing of votes, these compromises were not stable, and have been challenged ever since. The treaty itself states that the governments should continue with the institutional reforms achieved in Nice, with the aim of making an enlarged EU more efficient and democratic. Only six months after the conclusion of the negotiations, Pierre Vimont (2001: 159), President of the Group of Representatives during the French Presidency, confessed: “Nice cannot be rescued and should be abandoned”. Similarly, the Financial Times (4 December 2003: 10) quoted the German Foreign Minister Joschka Fischer as saying that Nice was a “bad” and “insufficient” treaty. A large number of the decisions taken to conclude the IGC of 2000 thus had to be renegotiated in the IGC of 2003-04.
In sum, the evidence just presented suggests that not all IGCs were equally efficient in assembling an agreement that leaves no potential gains on the table. The question that derives from this finding is: what accounts for the observable variation across cases? In the following section, we develop an argument that provides a novel explanation for this puzzle.
In intergovernmental bargaining that covers a large number of issues and that includes many actors, it seems little realistic that the win sets of the participants intersect on all issues. Under such circumstances, two different strategies can lead to consensus decisions: either the participants can exclude those issues from a final agreement where at least one player opposes a negotiated agreement (principle of the lowest common denominator); or they can try to assemble a package deal that includes all issues and thus maximises their joint gains. A substantial literature has pointed out that negotiators can use this second strategy to increase bargaining efficiency (Sebenius 1983; Davis 2004), but the circumstances under which governments are able to make issue linkages remain disputed.
5
Some authors maintain that in the absence of supranational entrepreneurs high transaction costs of negotiations can inhibit issue linkages (Young 1999; Christiansen 2002), whereas others stress that issue linkages are little probable because governments are reluctant to impose concentrated costs on domestic constituencies (Moravcsik 1998; 1999). Still others emphasise that problems of incomplete contracting and the threat of reneging on agreements make the achievement of issue linkages difficult (Martin 1994).
We add to this literature by proposing that the participants’ need to concentrate on the distribution of the gains from cooperation rather than on the maximisation of joint gains often inhibits negotiation efficiency.(2) In other words, the need for “value claiming” (Lax and Sebenius 1986) makes actors unable and unwilling to propose deals that can help overcome the lack of overlapping win sets with regard to specific issues. To maximise individual gains in distributional bargaining, each party has an incentive to conceal its “resistance point”, defined as the point at which it is willing to accept an agreement, to achieve better outcomes itself. In the language of game theory, players in a bargaining situation tend to hide their true valuations of all options. Under such conditions, actors use signals sent by others to estimate their resistance points. By proposing issue linkages, participants to a negotiation signal weakness and consequently have to fear reduced gains in distributional bargaining (Morrow 1992). This makes negotiators reluctant to propose issue linkages that could help to overcome bargaining deadlock; all suggestions for how to advance in the negotiations will be highly favourable to their own interests. Under these conditions, an agreement at the lowest common denominator is highly probable.
We suggest that the problem of distributional bargaining going to the detriment of overall efficiency can best be resolved if two necessary pre-conditions for issue linkages are given. First, and quite obviously, there have to be issues that can be linked to each other. This condition is achieved most probably if a relatively large set of issues is negotiated in parallel, i.e. if the negotiating agenda is as broad as possible. Negotiators, in addition, have to possess the expertise to combine the different issues in a package deal. Second, a trusted mediator with knowledge of the resistance points of the negotiation parties, and with the capacity to propose package deals that take into account these preferences has to guide the negotiations. Only such a mediator is able to suggest compromises that increase overall negotiating efficiency without sending a signal of weakness to the other negotiating parties.
In IGCs, EU member states established certain procedures that facilitate the achievement of these pre-conditions. First, a preparatory phase in which the member governments sketch out their negotiation positions, and in which lower level negotiators manage to outline options for how to create different possible package deals, generally precedes an IGC. For this purpose, governments establish different types of preparatory groups, be they composed of member states’ representatives or of a multitude of actors such as in the case of the Convention. These groups elaborate reports that, if they are to serve the purpose of facilitating future package deals, have to list not only the demands of the participants, but also have to suggest options for how to link different issues. An effective preparation phase thus not only facilitates the achievement of the first pre-condition for efficient bargaining set out above, but even has the potential of reducing the distributional part of bargaining to a choice among several well-specified options.
Second, the presidency of the Council of Ministers has the task of serving as mediator in the IGCs (Svensson 2000; Tallberg 2003; Mateo 2004).(3) For this objective, it has at its disposal a substantial number of tools. In particular, together with the General Secretariat of the Council, it elaborates documents throughout the negotiations, and it presents the draft treaty according to which the detailed negotiations proceed.(4)
6
Furthermore, it controls which of the many proposals made by the member states are admitted to the negotiations, sets the agenda of meetings, and determines the calendar of the IGCs. Most importantly, however, to make sure that other governments are willing to disclose their preferences to the presidency, it is bound by an informal rule of neutrality.(5) The reason for this norm is that all other governments have to be confident that the presidency does not use its privileged access to information to achieve negotiation outcomes close to its own preferences. The presidency then can explore what the member states are willing to accept, and engineer the final package based upon this information. Assuming the role of an agenda setter, and due to its privileged access to information, the presidency thus can help governments discover possible solutions to impasses in the negotiations.
For several reasons, however, we maintain that the preparatory phase and the mediation provided by the presidency are not always equally effective in facilitating issue linkages in IGCs. First, due to lack of time or because a preparatory group tries to take away the distributional part of the bargaining by already suggesting a complete package deal, the preparation can fail to assemble enough issues for the negotiations and to suggest possible ways of linking them. Lack of time is plausible if some exogenous event makes the achievement of agreement on some issues particularly urgent, while classical principal-agent problems can make a preparatory group overstep its mandate. Second, since the state that holds the presidency is also a participant in an IGC, it is credible as honest broker only if it takes a neutral stance on key issues of the negotiation. Actors involved in bargaining only accept to reveal their resistance points in informal contacts and confessionals with a neutral presidency. A government with important outlier preferences, however, is likely to ignore the neutrality norm and defend its interests instead even if it occupies the presidency.(6) Consequently, a country with no significant outlying preferences, in practice most often a small country, can better live up to the role of mediator than a country with important stakes in the negotiation. Besides lack of neutrality, exogenous events such as elections or external crises, or simply low administrative capacity, can reduce the capability of a presidency to fulfil its tasks.
This discussion shows that two conditions have to be present for governments to reach efficient bargaining results in IGCs. We propose that only if the preparatory group’s work, and the presidencies’ mediation are effective, member states can achieve a large package of issues strategically linked to each other, rather than an agreement at the lowest common denominator. Formulated as a hypothesis, the degree of effectiveness of the preparation and of the mediation by the presidencies in IGCs explains whether conflicts of interest lead to the exclusion of certain issues from a final agreement or to the inclusion of new issues to achieve a larger package deal. This is not to say that there is no limit to the capacity of the preparatory group and the presidencies to facilitate package deals: even highly effective preparation and mediation by the presidencies cannot facilitate compromises if the negotiations have to tackle zero-sum issues, or if the number of issues that can be included in an agreement is not large enough to make all participants gain from a package deal. We posit, however, that for the case of the IGCs these constraints on the argument are of minor relevance since generally a large number of issues of a positive-sum character exist that allow for issue linkages and thus for agreements beyond the lowest common denominator.
To examine the explanatory power of our argument, we show the importance of the preparatory phase and of the mediation by the presidencies for bargaining efficiency in six IGCs. We provide a systematic analysis of all IGCs that have been concluded since the mid-1980s, and find substantial support for our hypothesis. Table 1 presents a correlation between the effectiveness of the preparation and of the mediation by the presidencies on the one hand, and the efficiency of the negotiations on the other.
7
This correlation demonstrates that far-reaching results are most likely in the cases in which both factors are effective in facilitating future issue linkages. Ineffective preparation and the lack of a mediator in the negotiations, on the contrary, lead to the exclusion of issues from the final agreement.
The following narrative fulfils two major goals. First, we provide substantial evidence concerning a set of indicators that we used to assign values to our independent variables (see table 2). The main indicator for effective preparation is whether the respective preparatory group managed to produce a document that lists the positions of the participants, existing conflicts among governments, and, importantly, possible ways to resolve these conflicts. We employed several indicators to evaluate the effectiveness of a presidency’s mediation: whether a presidency tried to exclude from the negotiations issues and proposals that it disliked; whether it built on the work undertaken by the preparatory groups or by preceding presidencies; and other negotiators’ evaluations of whether or not the presidency acted in a biased way. Second, besides legitimising the correlation presented in table 1, we muster evidence to support our contention of a causal link between effective preparation and effective mediation by the presidencies on the one hand, and higher efficiency in the negotiations on the other. In continuation, we present these two types of evidence for our six case studies.
We propose that the IGC of 1985 was highly efficient in achieving far-reaching compromises, largely because both the preparation and the mediation provided by the presidencies were effective in facilitating the negotiations. The European Council at Fointainebleau in June 1984 decided to establish a committee, known as Dooge Committee for its president, an Irish Senator called James Dooge, with the task of preparing the negotiations on institutional questions. The committee’s members, both former and current politicians, and senior officials in the national administrations, secured efficient preparation due to their expertise. Despite huge difficulties caused by diverging preferences among the members, most observers agree that the preparatory work carried out by the Dooge Committee in eleven meetings was very successful (Keatinge and Murphy 1989: 224). The report prepared by the committee included such far-reaching proposals as the extension of majority voting in the Council of Ministers, the expansion of the powers of the European Parliament, and an extension of cooperation on foreign and security policy (Financial Times, 30 November 1984: 1). With the opinions of opposing member states often relegated to footnotes, the text both allowed for later negotiations across a broad spectrum of issues (De Ruyt 1989: 55) and made explicit the conflicts that had to be resolved. Exactly because the document presented by the committee did not try to resolve all outstanding issues, but rather simply showed how different package deals could look like, it provided the basis for efficient bargaining in the later negotiations.
The transition from the preparatory phase to the actual negotiations, however, was not smooth. The three member states most opposed to the proposals included in the final document of the Dooge Committee, namely Denmark, Greece, and the United Kingdom, resisted the start of an IGC. Only due to the ability of the Italian presidency to convene an IGC despite these objections, the opposition from the recalcitrant states to the start of negotiations could be overcome. In the ensuing negotiations, the Luxembourg Presidency played an important role as facilitator (De Ruyt 1989: 70; interview, Brussels, 1 April 2003).
8
At the beginning of September 1985, and thus three months before the final summit, already, the Presidency was able to present a unified document with precise proposals. This document clearly tried to offer a compromise between diverging positions, and as such was criticised by some governments as being too conservative, while others maintained that it was too far-reaching.(7) As the Financial Times (2 December 1985: 19) wrote, however, “If an agreement can be negotiated, it will probably look something like the Luxembourg proposals.” The existence of such a compromise text facilitated the ensuing negotiations because it permitted the governments to concentrate on resolving the distributional questions. Given this success, Sir Geoffrey Howe (1994: 454), British Secretary of State in these negotiations, referred to the Luxembourg presidency as a “tiny but immensely efficient […] machine”. To quote once more the Financial Times (5 December 1985: 14), “The fact that an agreement was finally reached in the short space between two European Summits, is in part a tribute to the presidential skill of the Luxemburgers.” That until the end governments struggled over the final document does not contradict our analysis, but rather shows that a compromise, which linked the liberalization of the internal market, demanded by the United Kingdom, to institutional reform, asked for by France and Germany, could be found despite diverging national preferences in an overall package deal. This case, therefore, provides some initial evidence that demonstrates that effective preparation and mediation by the presidency can augment the efficiency of intergovernmental negotiations.
Similarly, effective preparation and mediation by the presidencies were important for negotiation efficiency in the IGC on EMU (1991). Concerning preparation, in June 1988 the European Council in Hanover decided to appoint an expert committee composed of the twelve governors of the central banks and of three independent experts, and formally headed by Jacques Delors, President of the Commission. The Delors Committee, as it was soon called, had the task of studying and preparing the first steps towards EMU. On 17 April 1989, after nearly one-year of discussions, and despite many still unresolved issues, the committee presented the Delors Plan, which was a good starting point for the negotiations on monetary union. Hans-Dietrich Genscher, the German foreign minister, was not alone in describing the Delors report as “constructive and far-reaching” (Financial Times, 20 April 1989: 2). A detailed history of the British involvement in the IGC concludes, “[T]he [Delors] report increased the pressure on the British government to concede to the majority will [within the EU] on the question of EMU” (Forster 1999: 50). The similarity between the report presented by the Delors Committee and the resulting treaty on EMU (Italianer 1993) provides some further proof for the importance of effective preparation for the following negotiations. Even after in June 1989 the European Council of Madrid endorsed the Delors Plan, and convoked an IGC on EMU to start in December 1990, however, the preparatory phase continued. Two more committees, one headed by the then French minister for European affairs, Elisabeth Guigou, and the other composed of the heads of the central banks of the then twelve member states, discussed such issues as the statute of the future European Central Bank. In accordance with our argument, Giulio Andreotti, prime minister of Italy, clearly expressed the importance of the results achieved in the preparatory phase for the following negotiations on EMU (Doutriaux 1992: 45). Although some sensitive issues remained to be settled during the IGC, the excellent preparation paved the way for future issue linkages.
A couple of months after the start of the actual negotiations, in May 1991, the Luxembourg presidency presented a first draft for a future treaty on EMU that was clearly based on the prior preparatory work. Despite important advances introduced by this draft treaty, the following Dutch presidency still had to resolve several disputed issues.
9
At the beginning of September, the Dutch presidency submitted a new draft for a treaty that was very similar to the one presented by the Luxembourg presidency (Baun 1996: 65). A substantial number of mainly monetarily less stable countries raised objections to the draft because it suggested that any six countries that met the conditions for EMU could move to establish the common currency. In the end, the two sides found a compromise that linked acceptance of monetary union to financial compensations in form of cohesion funds. The Dutch presidency’s revised draft treaty of 28 October, some months before the final summit that was scheduled to approve of the results of the IGC in December 1991, could thus already resolve the majority of the outstanding questions (Corbett 1993: 41).
Given the long and detailed preparation, even the British delegation, the most sceptical with regard to the objectives of EMU, saw the European Council in Maastricht mainly as a “rubber-stamping exercise” (Forster 1999: 69). Not astonishingly, then, the final negotiations went smoothly, and resulted in such far-reaching decisions as the creation of a single currency by 1999. This ease in finalising the treaty stands in stark contrast to the previous conflicts mainly between France and Italy on the one hand, and Germany and the Netherlands on the other on such issues as convergence criteria and a timetable for starting the introduction of a common currency (Dyson and Featherstone 1999: 6). The fact that member states could find an agreement despite initially very diverging views confirms our point that effective preparation and mediation by the presidencies rather than simply converging preferences explain the success of the IGC on EMU.
In contrast to the IGC on EMU just presented, the outcomes of the parallel IGC on Political Union were less significant, and member states failed to achieve compromises on several outstanding questions.(8) In this section, we demonstrate how lack of preparation and ineffective mediation by the Dutch presidency can explain these efficiency losses. The preparation was less effective than in other IGCs because the decision to negotiate on Political Union was only taken in April 1990, nearly two years after the start of the preparations for the IGC on EMU, although both IGCs were supposed to end at the same time. Since the timeframe available for the preparation thus was very limited, instead of a specifically appointed preparatory committee the permanent representatives of the member states in Brussels had to undertake most of the preparatory work, only marginally supported by the General Secretariat. The foreign ministers then adopted a “reflection document” on which basis the negotiations were to be started, but which had no major impact on the following negotiations. Most observers therefore agree that at least some of the IGC’s weaknesses were due to the lack of an “orderly process” in preparing the treaty (Forster 1999: 6; see also Dyson and Featherstone 1992: 32; Laursen, Vanhoonacker and Wester, 1992).
The actual negotiations on Political Union were started in parallel with those on EMU. Despite the ineffective preparation, the Luxembourg presidency was very effective at the beginning of the IGC. An indication for this is that in April 1991, after only three months of negotiations, it presented a first proposal for a draft treaty with a view of achieving Political Union. Since the document presented by the Luxembourg presidency was a compromise between the opposing camps, all member states agreed to continue the negotiations on basis of this draft treaty (Le Monde, 26 June 1991: 21). An efficient agreement, however, was made impossible once the following Dutch presidency decided to ignore this compromise, and instead to propose a different draft treaty that would have brought foreign policy and internal affairs into the Community framework. This new text, which was closer to the outlier preference of the Netherlands, was no good reflection of the preferences of all participants.
10
On 30 September, a day that has become known as the “Black Monday” of Dutch diplomacy, ten of the twelve member states of the EU rejected the text proposed by the Dutch presidency, compelling the latter to abandon its proposal and to return to the Luxembourg draft as the basis for the negotiations. Besides its outlier preference, the Dutch presidency of the IGC on Political Union was also crippled by difficulties within the coalition government (Smith 2002: 123).(9) Consequently, three months into the Dutch presidency, and already close to the Maastricht summit at which the IGC was to be concluded, little progress on the important issues had been made. Although in the end the member states succeeded in concluding the negotiations on Political Union in December 1991, the case demonstrates that with ineffective preparation and partly ineffective mediation during the negotiations, intergovernmental bargaining is unlikely to be fully efficient.
According to our classification, the IGC of 1996-97 led to an intermediate result. As we will show in the following, mainly the lack of neutrality of the presidencies in the final negotiations, rather than ineffective preparation as in the case of the IGC on Political Union, caused this outcome. The pre-history of the IGC starts in June 1994, when the European Council of Corfu decided to establish a group of member states’ representatives to determine the agenda for a new IGC. After fourteen meetings, and six months of deliberations, the Reflection (or Westendorp) Group presented its report to the European Council of Madrid in December 1995. The report fulfilled the objective of effective preparation since it not only provided the negotiators with ample expertise and information about the positions of all governments, but also outlined possibilities of how to resolve existing conflicts of preferences. According to one negotiator’s evaluation, some 95 percent of the key ideas of the reflection group were approved in the Amsterdam summit (Elorza 1998: 44). As Marcelino Oreja (1998: 86; interview, Madrid, 24 September 2002) maintained, moreover, the decision of the Reflection Group to define the agenda broadly had a positive effect on negotiating efficiency because it facilitated the finding of compromises in this IGC.
Ineffective presidencies, however, crippled the following negotiations.(10) The national elections on 21 April 1996 that resulted in a change of government explain a large part of the ineffectiveness of the Italian presidency (Financial Times, 8 January 1996: 2). The German Chancellor, Helmut Kohl, also expressed this view when he criticized the presidency’s lack of organization and coordination in structuring the negotiations, which he blamed upon the national elections (Corrado 2002: 231). In addition, Michel Barnier (Agence Europe, 9 May 1996: 5), French Representative, declared that during the Italian presidency the IGC remained in an “observation phase”. The following Irish presidency, although it fared better than the Italian one, tried to avoid controversial questions rather than tackling them directly, and thus made the achievement of later compromises more difficult. Opposing a change in the size of the Commission, its objective was to create a “mini-package” of issues on which consensus could be reached (McDonagh 1998: 73). The draft treaty presented by the Irish presidency consequently completely excluded most aspects of institutional reform, a fact that was heavily criticised mainly by France and Germany (El País, 7 December 1996: 4). Moreover, the Irish presidency lacked neutrality on the Schengen agreement and on the issue of social provisions, where it lined up with the position of the United Kingdom (Dehousse 1999: 8).
The Dutch presidency, which had the task of concluding the negotiations, failed for still different reasons. First, its ability to mediate in the negotiations was reduced due to problems within the ruling coalition (Van Keulen and Rood 2003: 85). Second, the Dutch presidency lacked neutrality on the key issue of the reweighing of votes in the Council of Ministers. Its main objective was to achieve a higher number of votes than its smaller neighbour, Belgium.
11
When the Dutch presidency saw itself unable of achieving this aim, it preferred to have the negotiations postponed rather than to agree upon a compromise that did not take into account this Dutch preference. Consequently, even though several member countries rejected the proposals it presented in May 1997, it did not provide further suggestions for how to achieve an agreement prior to the Amsterdam summit (Elorza 1998: 48). In accordance with our hypothesis, despite effective preparation, mostly ineffective mediation by the presidencies meant that the treaty agreed upon in Amsterdam excluded several issues, which had to be negotiated in a further IGC.
In this section, we propose that the IGC of 2000 failed to achieve an efficient agreement mainly due to lack of preparation and bad handling of the negotiations by the French presidency. The preparation of the negotiations suffered from the fact that, facing an accelerated enlargement process, governments decided to convene a new IGC earlier than originally planned. The short time span available, together with initial disagreement on how to prepare the IGC, made the governments decide to have the Finnish presidency prepare a report on its own. In parallel, the Commission assembled a small group of sages – known as the Dehaene group – to elaborate a report, which was presented in October 1999, but which had no influence on the ensuing negotiations (Smith 2002: 177; interview, Brussels, 3 April 2003). The Finnish presidency, instead of offering options for how to find a compromise on difficult questions, as had done the reflection groups in other IGCs, simply listed the issues that remained disputed among the member states (Gray and Stubb 2001: 9). These issues were the reweighing of votes in the Council of Ministers, the composition and size of the Commission, and the extension of qualified majority voting in the Council of Ministers. Since the Finns decided to be very conservative in drafting their report, Javier Elorza (2000: 46) concluded that the report could “only serve as one of the support documents of the IGC”, insinuating that it was insufficient to facilitate issue linkages in the ensuing negotiations.
In December 1999, the European Council in Helsinki hardly discussed the report prepared by the Finnish presidency, obviously because several governments deemed it inadequate to structure the following negotiations (interview, Brussels, 29 April 2003). Later, when the limited outcome of the IGC was already known, Agence Europe (13 December 2000) wrote that given the ineffective preparation, nothing more than a “mediocre” reform of the institutions at the Nice summit could have been expected. Because of this failure of preparation, it remained undecided whether the IGC should discuss such issues as the common security and defence policy, the Charter of Fundamental Rights, and enhanced cooperation. Since thus most discussions remained limited to the Amsterdam leftovers, in the later negotiations few issue linkages could be found. As Francisco Seixas da Costa (cited in Conselho Económico e Social 2000: 85), Portuguese representative in the IGC, affirmed, a less limited agenda could have made the finding of a package deal and thus the achievement of consensus easier. This confirms our argument that far-reaching agreements can only come about if an effective preparation of the negotiations manages to put issues on the negotiation table that can serve as side payments in a package deal.
When the actual negotiations started in February 2000, the Portuguese presidency’s bias on some issues, which was most visible in its defence of the right for a commissioner for all member countries even in an enlarged EU, aggravated the lack of preparation. As the issue of the size and composition of the Commission was tightly linked to the reweighing of votes in the Council of Ministers, the Portuguese presidency did not tackle the latter issue either. Instead, its strategy was to have these issues negotiated under the next presidency, and rather to advance the discussions concerning the extension of majority voting in the Council of Ministers. This biasness was heavily criticised by some observers, not least because it set precedence for the following French presidency, which had its own strong views on these issues (Agence Europe, 26 April 2000).
12
More seriously still, the French presidency defended strong interests on practically all of the issues under negotiation (Smith 2002: 201). In the question of the reweighing of votes, France was reluctant to accept fewer votes in the Council of Ministers than Germany, despite the difference in population sizes. Its strategy was to have this issue decided in the last moment, hoping that by this time Germany would be more willing to yield (Tallberg 2003: 21). One way of doing so was by simply ignoring the report prepared by the previous Portuguese presidency (Gray and Stubb 2001: 11). Some member states, when witnessing the French presidency’s attempt to omit this question before Nice, accused the presidency of violating the neutrality rule (Financial Times, 14 December 2000: 23). France also defended specific interests concerning the size and the composition of the Commission, and the extension of majority voting in the Council of Ministers.
In the final summit in Nice, the French presidency’s lack of neutrality made it propose a solution to the questions concerning the size of the Commission and the voting power in the Council of Ministers that was rejected by nearly all other countries. The French presidency not only lacked neutrality but also capacity to mediate due to difficulties between the socialist Prime Minister, Lionel Jospin, and the conservative President, Jacques Chirac (Costa et al. 2003: 128). Ten days before Nice, Jospin threatened to end the cohabitation between the two. The French presidency, consequently, was heavily criticised for the “deplorable political balance of the summit in Nice” (Le Monde, 29 December 2000). Given these occurrences it is no wonder that an observer talked of the “largest defeat of French diplomacy” when commenting the Treaty of Nice (Agence Europe, 21 December 2000). Tony Blair, with regard to the Nice summit, concluded, “[W]e cannot do business like this in the future” (quoted in Church 2001: 75). Even the French President Chirac conceded that the summit showed the need for changes in the procedures followed in future negotiations (Agence Europe, 11/12 December 2000). It was for these reasons that most governments accepted the Treaty of Nice only as a provisional agreement. Already before finishing the IGC, a new debate emerged about the future of the EU, which finally resulted in the establishment of the European Convention and the start of a new IGC in 2003. This case, consequently, confirms our argument that a combination of ineffective preparation and presidencies leads to inefficient outcomes in intergovernmental bargaining. Expressed in form of a counter-factual, it at least seems plausible that more effective preparation and mediation by the presidencies would have facilitated the finding of a more stable package deal.
The IGC of 2003-04 started with a decision by the European Council of Laeken in December 2001 to establish the European Convention with the objective of preparing a reform of the Treaty of Nice. The Convention, which encompassed representatives of the governments of the 15 member states, of the national parliaments, and of the European institutions, immediately set itself the objective of presenting a draft for a constitutional treaty.(11) After deliberating for 18 months, the president of the Convention, Giscard D’Estaing, presented such a draft treaty in June 2003. Despite the long time span available for preparation, since the leadership of the Convention aimed at concluding the negotiations, rather than at effectively preparing the distributional part of the bargaining, the Convention did not live up to the criteria of effectiveness set out in this paper. Rather than outlining several options of how to achieve an agreement, the Convention only put forward one possible compromise. For the key institutional issues, the decision to postpone their discussion to the end of the preparatory phase (El País, 13 December 2003: 4) aggravated this weakness, since this move eliminated the possibility for a debate among the members of the Convention that could have helped uncover the positions of the various governments. Whereas the resulting draft treaty had the potential of facilitating the conclusion of the negotiations on many issues, it thus also created problems since it made later distributional bargaining more difficult.
13
The following Italian presidency, although it did not have outlier preferences, failed to take into serious consideration the Spanish and Polish threats to veto a change in the decision-making rules in the Council of Ministers as set down in the Treaty of Nice. The Italian presidency thus fell short of adequately preparing the summit that was supposed to conclude the negotiations (Le Monde, 27 March 2004). Since the Italian presidency also mismanaged the European Council of Brussels in December 2003, the French President Jacques Chirac went so far as to threaten with his early return to Paris (El País, 14 December 2003: 4). Especially the Italian presidency’s claim to have a compromise formula available in the case that the other proposals for the reweighing of votes in the Council of Ministers were not accepted, coupled with its incapacity to outline this alternative formula in the critical moment of the negotiations, irritated most governments. Since the preparation was only medium effective, and the first presidency equally fell short of being a valuable mediator in the negotiations, it is no wonder that the first attempt at concluding the Constitutional Treaty failed.
The failures of the Italian presidency meant that the following Irish presidency had to start its semester with a round of consultations to resolve the blockage in the IGC. Only after this round, the presidency saw itself capable of presenting a document on the state of the negotiations with possible solutions for how to achieve a package deal. In this process, the presidency’s neutrality on main questions, which it maintained despite having some outlier preferences, helped it present a series of draft texts that only left few issues open for the final negotiations. Most contemporary observers thus highlighted the effective work of the Irish presidency (Financial Times, 14 June 2004: 6). Shortly before the summit in June 2004, Anne Anderson, Ireland’s ambassador to the EU, even was able to say that there would be no repetition of the Nice summit, referred to as the “five-shirt summit”, because it involved so many meetings (Financial Times, 17 June 2004: 9). The persistence of some conflicts involving among others France, Germany, and the United Kingdom serves as an indicator that agreement was due to issue linkages, rather than a convergence of preferences. In the end, the effective work of the presidency enabled the negotiators to approve a constitutional treaty that is likely to remain stable for some time and to be considered efficient in the long run.
In this paper, we have set out a novel explanation for the observable variation in the efficiency of the EU’s six most recent IGCs based on the idea that in bargaining situations, negotiators fail to propose issue linkages whenever they fear that such a step signals weakness. The reason for this is that giving the impression of weakness reduces the share of the cake an actor is likely to gain in distributional negotiations. Based on this reasoning, we have proposed two pre-conditions for efficient intergovernmental bargaining: a broad negotiating agenda that secures the existence of a large number of issues that can be linked to each other and a credible and trusted actor’s mediation during the negotiations.
In IGCs, member governments established procedures with the intention of attaining these pre-conditions. On the one hand, IGCs are normally preceded by a phase in which preparatory groups set out the negotiating agenda and propose how different issues can be linked to each other. On the other hand, the presidency of the Council of Ministers is to assume the role of honest broker in IGCs with the necessary powers to mediate and propose package deals. Nevertheless, for several reasons, sometimes these negotiations are less than efficient: the time span for the preparation may be too short, the presidency may not be credible as honest broker, or the presidency can be inhibited in its role by exogenous shocks that reduce its capacity to guide the negotiations.
14
Our explanation both complements and contrasts with existing theories of intergovernmental bargaining. Moravcsik’s intergovernmental bargaining theory, in particular, negates the possibility of efficiency losses in intergovernmental bargaining, while we demonstrate that not all IGCs are equally efficient. Moreover, whereas supranational bargaining theory emphasises negotiators’ lack of expertise, in our view individual actors’ attempts to maximise their personal gains is the main reason for efficiency losses.
One of the paper’s most important conclusions thus is that the constellation of preferences, while important, is not the only determinant of negotiation outcomes. The presence or absence of two pre-conditions determines whether an agreement is found at the lowest common denominator, or whether governments can overcome differences in preferences by way of issue linkages. Neither, as a comparison of the IGC of 2000 and the IGC of 2003-04 demonstrates, which both tackled so-called “political issues”, is the substance under negotiation the main explanatory variable for variation in the efficiency of negotiations. We submit, instead, that our argument, although it does not cover all influential factors, provides a sound explanation of the bargaining processes in the six most recent IGCs. Idiosyncratic events such as elections in member states, problems within coalition governments, foreign policy crises, and others always have some influence on how an IGC develops. Rather than embracing an eclectic approach that includes all of these factors in a more or less organized manner, in this paper our objective has been to delineate the systematic element behind often complex negotiations.
A practical implication of our conclusions is that recent calls for reducing the importance of the presidency in the IGCs (Financial Times, 10 November 2003: 2) appear to be counterproductive. Equally, the changes in the role of the presidency included in the recent Constitutional Treaty can be problematic if further revisions of the treaties are contemplated, since having separate presidencies for the European Council and the Council of Ministers may increase confusion rather than effectiveness. Instead, an alternative institutional change could endow smaller member states with more responsibility in the IGCs for two reasons. First, due to their higher credibility as honest brokers they often make for better presidencies. Second, smaller member states invest more effort in bringing negotiations to a positive end since the reputational gains from being considered a “successful” presidency are more important for them than for the larger member states. Even if this alternative is unlikely to be realised, what is clear is that institutional changes matter for the IGCs and that their consequences have to be considered in detail.
Our findings also have implications that extend beyond the narrow case of the six IGCs treated here. With the growth of global interdependence, negotiations among states have become more relevant to avoid negative externalities from national policies in such areas as the environment, migration, and the economy. Given the growing influence on the lives of all people of these negotiations, in which also in the future majority decisions will remain the exception rather than the rule, increasing the efficiency of consensus decisions becomes ever more important. Finding ways to enhance efficiency in negotiations consequently appears to be an important task for future research in the social sciences; we hope that the present paper makes one step into this direction.
Some 50 interviews with experts, participants in the negotiations, and politicians contributed to the information contained in this paper.
Baun, M. (1996) An Imperfect Union: The Maastricht Treaty and the New Politics of European Integration, Boulder: Westview Press.
Beach, D. (2004) ‘The Unseen Hand in Treaty Reform Negotiations: The Role and Influence of the Council Secretariat’. Journal of European Public Policy 11 (3): 408-39.
Budden, P. (2002) ‘Observations on the Single European Act and “Relaunch of Europe’: A Less ‘Intergovernmental’ Reading of the 1985 Intergovernmental Conference’. Journal of European Public Policy 9 (1): 76-97.
Christiansen, T. (2002) ‘The Role of Supranational Actors in EU Treaty Reform’. Journal of European Public Policy 9 (1): 33-53.
Church, C. (2001) ‘Intergovernmental Conferences and Treaty Reform: The Nice Experience’, in M. Bond and K. Feus (eds.), The Treaty of Nice Explained, London: The Federal Trust: 75-96.
Colm, O. (ed.) (1985) The Presidency of the European Council of Ministers and Implications for National Governments, London: Croom Helm.
Conselho Económico e Social (2000) ‘Mesa redonda sobre a Conferência Intergovernamental’, Lisboa: Conselho Económico e Social.
Corrado, L. (2002) ‘Italy: From “Hard Core” to Flexible Integration’, in F. Laursen (ed.) The Amsterdam Treaty: National Preference Formation, Interstate Bargaining and Outcomes, Odense: Odense University Press: 225-55.
Corbett, R. (1993) The Treaty of Maastricht From Conception to Ratification: A Comprehensive Reference Guide, London: Longman.
Costa, O., Couvidat, A. and Daloz, J-P. (2003) ‘The French Presidency of 2000: An Arrogant Leader?’, in O. Elgström (ed.) European Union Council Presidencies: A Comparative Perspective, London: Routledge: 120-37.
Davis, C. (2004) ‘International Institutions and Issue Linkage: Building Support for Agricultural Trade Liberalization’. American Political Science Review 98 (1): 153-69.
De Ruyt, J. (1989) L’Acte Unique Européen, Bruxelles: Université de Bruxelles.
Dehousse, F. (1999) Amsterdam: The Making of a Treaty, London: Kogan Page.
Doutriaux, Y. (1992) Le Traité sur l’Union Européenne, Paris: Armand Colin.
Dyson, K. and Featherstone, K. (1999) The Road to Maastricht: Negotiating Economic and Monetary Union, Oxford: Oxford University Press.
Elgström, O. (ed.) (2003) European Union Council Presidencies: A Comparative Perspective, London: Routledge.
Elorza, J. (1998) ‘El Tratado de Ámsterdam: Una evaluación española’, in A. Matutes et al. (eds.) España y la negociación del Tratado de Ámsterdam, Madrid: Política Exterior: 35-58.
Elorza, J. (2000) ‘The Position of Spain with Regard to the IGC’, in E. Best, M. Gray and A. Stubb (eds.) Rethinking the European Union: IGC 2000 and Beyond, Maastricht: EIPA: 43-52.
Forster, A. (1999) Britain and the Maastricht Negotiations, Basingstoke: Macmillan.
General Secretariat of the Council (2001) Council Guide I: The Presidency Handbook, Luxembourg: Office for Official Publications of the European Communities.
Gray, M. and Stubb, A. (2001) ‘Keynote Article: The Treaty of Nice - Negotiating a Poisoned Chalice?’ Journal of Common Market Studies 39 (Annual Review): 5-23.
Howe, G. (1994) Conflict of Loyalty, London: Macmillan.
Italianer, A. (1993) ‘Mastering Maastricht: EMU Issues and How They Were Solved’, in K. Gretschmann (ed.) Economic and Monetary Union: Implications for National Policy-Makers, Dordrecht: Kluwer Academic Publishers: 51-113.
Keatinge, P. and Murphy, A. (1989) ‘The European Council’s Ad Hoc Committee on Institutional Affairs (1984-85)’, in R. Pryce (ed.) The Dynamics of the European Union, London: Routledge: 217-37.
Laursen, F., Vanhoonacker, S. and Wester, R. (1992) ‘Overview of the Negotiations’, in F. Laursen and S. Vanhoonacker (eds.) The Intergovernmental Conference on Political Union: Institutional Reforms, New Policies and International Identity of the European Community, Maastricht: EIPA: 3-24.
Lax, D. and Sebenius, J. (1986) The Manager as Negotiator: Bargaining for Cooperation and Competitive Gain, New York: The Free Press.
Martin, L. (1994) ‘Heterogeneity, Linkage and Commons Problems’. Journal of Theoretical Politics 6 (4): 473-93.
Mateo, G. (2004) ‘Un análisis institucional de las Conferencias Intergubernamentales en la Unión Europea: De Ámsterdam a Niza’, PhD Thesis: Universidad Autónoma de Barcelona.
McDonagh, B. (1998) Original Sin in a Brave New World: The Paradox of Europe. An Account of the Negotiation of the Treaty of Amsterdam, Dublin: Institute of European Affairs.
Moravcsik, A. (1998) The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, Ithaca: Cornell University Press.
Moravcsik, A. (1999) ‘A New Statecraft? Supranational Entrepreneurs and International Cooperation’. International Organization 53 (2): 276-306.
Morrow, J. (1992) ‘Signaling Difficulties with Linkage in Crisis Bargaining’. International Studies Quarterly 36 (2): 153-72.
Oreja, M. (1998) ‘El desarrollo de la CIG’, in M. Oreja (ed.) El Tratado de Ámsterdam: Análisis y comentarios, Madrid: McGrawHill: 73-116.
Smith, B. (2002) Constitution Building in the European Union, The Hague: Kluwer Law International.
Sebenius, J. (1983) ‘Negotiation Arithmetic: Adding and Subtracting Issues and Parties’. International Organization 37 (2): 281-316.
Svensson, A-C. (2002) In the Service of the European Union: The Role of the Presidency in Negotiating the Amsterdam Treaty 1995-97. Uppsala: Actas Universitatis.
Tallberg, J. (2003) ‘The Power of the Presidency: Brokerage, Efficiency, and Distribution in EU Negotiations’, Paper prepared for presentation at the EUSA’s 8th International Conference, Nashville, March 27-29, 2003.
Van Keulen, M. and Rood, J. (2003) ‘The Netherlands’ Presidency of 1997: Between Ambition and Modesty’, in O. Elgström (ed.) European Union Council Presidencies: A Comparative Perspective, London: Routledge: 71-86.
Vimont, P. (2001) ‘Misunderstanding Nice: A French Assessment of the Results’, in M. Bond and K. Feus (eds.) The Treaty of Nice Explained, London: The Federal Trust: 159-68.
Yataganas, X. (2001) ‘The Treaty of Nice: The Sharing of Power and the Institutional Balance in the European Union - A Continental Perspective’, Harvard Jean Monnet Working Paper 1. <http://www.jeanmonnetprogram.org/papers/01/010101.html>
Young, O. (1999) ‘Comment on Andrew Moravcsik, ‘A New Statecraft? Supranational Entrepreneurs and International Cooperation’ ’. International Organization 53 (4): 805-09 .
(*) We would like to thank Dirk De Bièvre, the participants of a colloquium at the Mannheim Centre for European Social Research (MZES), and three anonymous referees for their comments on earlier versions of this paper. Gemma Mateo is also grateful to the MZES, where she was a guest researcher while working on this paper, and would like to acknowledge financing from the project "Constitutional Politics of the EU", directed by Carlos Closa at the Univerity of Zaragoza (Ministry of Science and Technology, Contract Number SEC2003-0253)
(1) Nevertheless, some questions were deferred to a later IGC. In particular, a group of member states wanted to go further with regard to economic and monetary union, and foreign and environmental policy (Smith 2002: 109-110). These efficiency losses, however, were more than offset by the IGC’s far-reaching achievements.
(2) The main point here is that individual participants may gain most if an agreement is less than efficient; maximizing individual utility thus does not automatically lead to efficient bargains.
(3) In the preparatory phase, where negotiations take place within a preparatory group, the presidency of an IGC is mostly of little importance; its significance increases with the start of the actual negotiations, and thus the need for value claiming.
(4) Both Beach (2004) and Mateo (2004) stress the importance of the General Secretariat in this process.
(5) The handbook for the presidency prepared by the General Secretariat of the Council (2001: 5) confirms the existence of this rule when stating that “[t]he Presidency must, by definition, be neutral and impartial.” See also Svensson 2000: 25.
(6) The costs of reputation linked to a violation of this norm serve as a constraint on the extent to which a government in the role of the presidency can defend its own interests. Only if it really has a strong stake in an issue, it is thus willing to defy the neutrality norm.
(7) While in our view this document illustrates the essentially neutral position of the Luxembourg presidency, Budden (2002: 89) emphasises that the presidency’s pro-integrationist stance was critical for the successful conclusion of the negotiations.
(8) The IGCs on EMU and on Political Union were not completely separate; agreement in the one was linked to agreement in the other. The dynamics behind the two IGCs, however, are very different; this warrants the separation undertaken here.
(9) The ongoing Yugoslav crisis also hampered the negotiations (Corbett 1993: 31); however, there is little evidence that suggests that the crisis caused the efficiency losses in this IGC that we try to explain here.
(10) It would be too strong a claim to say that the IGC only suffered from ineffective mediation by the presidencies. The IGC also faced other difficulties: elections were pending in the United Kingdom, France held anticipated elections, and the mad-cow disease distracted governments from the negotiations.
(11) The representatives of 13 applicant countries had observer status during the meetings of the Convention.