Integrative and Distributive Bargaining Situations in the European Union: What Difference Does It Make? Open Access

Governance in the European Union (EU) consists essentially of negotiations among member states and supranational institutions in the Council of Ministers (hereafter referred to as the council), which serves as the union's main decision‐making political institution and the key forum where member states can articulate and defend their interests. (This article will use the term EU when referring to the period after the Treaty on the European Union [TEU], which was signed in 1993, and will use the term European Community [EC] when referring to the period before the TEU.) Governance by bargaining in the council is characterized by compromise solutions that reconcile the differing negotiating positions of the national governments, the European Commission (hereafter referred to as “the commission”), and the European Parliament (EP) representatives.

The decision‐making process in the EU involves these three institutions. The commission has the sole formal power to propose legislation (agenda setting power) in all the issues covered by the single market, the so‐called first pillar of the European Economic Community.¹ After the commission begins the legislative process by presenting a legislative draft proposal, the council deliberates on it. From the beginning of the European integration process, the decision‐making formula has been that the commission proposed and the council disposed. Depending on the legislative procedure used, the EP's influence on legislation varies. It is only under the co‐decision procedure that the EP shares legal responsibility for legislation jointly with the council.² Nevertheless, the council remains the forum for intergovernmental (among national‐government representatives) and interinstitutional bargaining (between member states and supranational institutions). This article is solely concerned with negotiations in the EU that fall under the consultation procedure. Thus, the role of the EP will not be taken into consideration as the EP does not have decision powers in the fisheries policy and can only deliver an opinion on legislative proposals made by the commission.

Because decision making within the EU is accomplished through bargaining, it can be characterized as a negotiating polity. Furthermore, the system itself was created via negotiation among the six original member states during the Treaty of Rome negotiations in 1957. Since the EU's inception, a series of negotiations have modified the original structure and created a number of “package deals” governing institutional arrangements and a number of different policies. Accordingly, some scholars have characterized the EU as a “negotiation regime” (Scharpf 1997: 211), as a “multilateral inter‐bureaucratic negotiation marathon” (Kohler‐Koch 1996: 367), and as a “negotiated democracy” (Lodge and Pfetsch 1998: 289).

To eliminate possible confusion, I wish to clarify my use of some terms before I proceed. In this article, the terms “negotiation” and “bargaining” are used synonymously. They are defined as “a decision‐making method used by actors to reconcile their respective policy positions when they disagree about preferred outcomes.” Bargaining in the EU refers to intergovernmental and interinstitutional negotiations within the council framework, whether this takes place within working groups, the Committee of Permanent Representatives (COREPER), or at the ministerial level. This usage is consistent with other formulations. Jeffrey Rubin and Bert Brown (1975: 2), for instance, defined negotiations as a “process whereby two or more parties attempt to settle what each shall give and take, perform and receive, in a transaction between them.”David Lax and James Sebenius (1986: 11) argued that bargaining processes consist of “an attempt by two or more parties to find a form of joint action that seems better to each than the alternatives.” Andrew Moravcsik (1993: 497), in turn, referred to it as “the process of collective choice through which conflicting interests are reconciled.”

In a seminal article on the role played by the unanimity decision rule in the political system of the EU and its inherent consequences (the joint‐decision trap), Fritz Scharpf (1985a, b) drew the attention of scholars to the implications of formal rules. Decisions made by national governments under the unanimity decision rule, Scharpf wrote, will systematically lead to suboptimal policy outcomes creating “the joint‐decision trap” (Scharpf 1985b: 48). The requirement of the unanimity decision rule can be expected to produce stalemate and deadlock because joint decisions under the unanimity decision rule express the position of the most reluctant actor and preserve the institutional status quo (Scharpf 1985b).

Since then, much has been written about negotiations in the EU, with three streams of analyses dominating the discussion. State‐centric approaches either focus on the behavior of rational utility maximizers and assume that member states’ preferences are exogenously given (Moravcsik 1993, 1997, 1998; Tsebelis and Yataganas 2002), or on the domestic constraints of member states and employ a two‐level games approach (Hosli 2000; König and Bräuninger 2001).

A second stream of analysis looks at institutional rules as the key variable for understanding bargaining outcomes (Christiansen and Jorgensen 1999; Garrett and Tsebelis 1996; Pierson 1996; Pollack 1997). Within this stream, three distinct new‐institutionalist approaches can be distinguished: sociological, historical, and rationalist. Although all three define institutions as rules and norms, they differ in the way in which they conceive preferences and explain how institutions are created and evolve (Aspinwall and Schneider 2000).³

Finally, other authors focus on two different bargaining dimensions. Whereas some authors (Elgström and Jönsson 2000; Scharpf 1985b) use the terms “bargaining” and “problem‐solving,” others (Richardson 1996; Sebenius 1984; Walton and McKersie 1965; Young 1989) use the terms “integrative” and “distributive” bargaining to delineate the two dimensions. In their classical study, Richard Walton and Robert McKersie (1965: vii) define integrative bargaining as a bargaining method designed to increase the joint gains available to all negotiating parties. In contrast, distributive bargaining situations are concerned with more competitive behaviors, intended to affect the division of limited resources.⁴

Despite the distinction made between integrative and distributive bargaining situations in the European Studies literature, few studies have focused specifically on how these two bargaining situations differ in the context of EU negotiating.⁵ This article attempts to close this gap by providing empiric evidence through the analysis of two key bargaining situations in fisheries policies that led to the transfer of decision making and legislation from the national to the European level (a process also known as Europeanization) of this policy field.

Decisions in the EU are subject to unanimity or qualified majority voting (QMV) in the council. The unanimity voting procedure prevents one member state being outvoted because all EU member states must agree before a policy can be adopted. Under this voting rule, each member state has an effective veto over policy decisions. Taking the unanimity voting rule as a constant parameter, this article focuses on two negotiations, one integrative and one distributive, in order to investigate first, how the unanimity rule affects patterns of negotiation and second, how these two types of bargaining situations may differ from each other. First, I focus on the negotiations surrounding the settlement of the structural policy and common organization of the market for fish products, which began in 1966 and ended in early 1970. This first case was a coordination situation without distributional conflict, in which all member states had a clear preference that “the pie” be enlarged.

The second part of the article examines the protracted negotiations over the settlement of the resource conservation and management policy, which exemplifies a distributive bargaining situation.⁶ The equal access provision and the sharing of fish resources among member states were the most controversial issues that negotiators dealt with, because each bargaining party wished to maximize its own share. Consequently, the negotiations lasted almost seven years, from December 1976 to January 1983.

In the third part of the article, I compare both bargaining situations by focusing on their similarities and differences. The analysis of these two key negotiations should reveal a great deal about decision‐making methods in the EU because most negotiations represent either an integrative or a distributive bargaining situation.

In analyzing both cases, my focus is on member states’ negotiating positions, on the role played by the commission, and on the institutional setting. Although the preferences of national‐government representatives are crucial variables in the EU negotiation process, “taking preferences seriously” (Moravcsik 1997) is not the whole story of bargaining in the EU. It is indisputable that the preferences of member states are an important variable. The “path to European integration” (Pierson 1996) has, however, placed such preferences within a complex institutional framework in which new actors (supranational institutions) and the institutional setting that includes formal (decision rules) and informal (norms, values) rules also must be taken into account.⁷

Furthermore, because very few studies (Friis 1997; Hooghe 2001; Hug 2003) have attempted to conceptualize the preference formation of the commission, I will assess its preferences when analyzing the two case studies. I assume that the commission is a rational utility maximizer and that it is more integration prone than member states. The commission regards itself as the representative of the EC because it has a legal responsibility to implement the treaties and to advance community policies. Thus, the commission is a collective body committed to the maximum total benefit of the EU. Furthermore, the commission's general interest in acting as a motor for more integration increases its own influence within it (Da Conceição‐Heldt 2004).

Finally, the variable institutional setting must also be taken into account when analyzing negotiations in the EU because the structure in which a game is played affects the scope of the bargaining outcome. The focus here shall be on the formal institutional rule unanimity voting and not on other informal rules.

The Treaty of Rome (1957) contained no reference to fisheries policy other than a remark in the agricultural policy defining fisheries as an “agricultural product.” At the time, the six original member states of the EC (Belgium, the Netherlands, Luxembourg, France, Italy, and Germany) were primarily concerned with agricultural issues.

This situation, however, changed in the mid‐1960s when the common market provisions began to indirectly affect fisheries through the common customs tariff, an external tariff for products coming from outside the EC. Because the EC's common customs tariff was more liberal than the tariffs that had up until that time protected the French and Italian fishing industries, both countries requested that the commission prepare a draft proposal for regulations complementing trade liberalization for fish products with a structural policy financed by the EC to help modernize their fishing fleets and shore‐based infrastructure. But the other four member states, particularly Germany and the Netherlands, had well‐organized fishing industries and expected to profit from trade liberalization. Germany, the Netherlands, Belgium, and Luxembourg thus opposed a further interventionist policy designed similarly to the Common Agricultural Policy that would have automatically increased their own contributions to the EC's budget, while aiding their weaker competitors in the modernization process of their fishing fleets (Da Conceição‐Heldt 2004; Leigh 1983).

The commission, however, seized this opportunity and prepared a regulation draft entitled “Report on the Situation in the Fisheries Sector of EEC Member States and the Basic Principles for a Common Policy” in 1966. In this draft proposal, the commission combined the demands for a structural policy and a common market organization and also added the basic principles for a common policy to be managed at the supranational level by the commission itself. The most original feature of the commission's proposal was the suggestion that fishing grounds should be equally accessible to all. This was the point of departure for a four‐year bargaining process.

This first key fisheries negotiation constitutes an integrative bargaining situation without distributional conflict. All six original member states expected that a coordination of their policies on the aforementioned issues would generate joint gains. They had, however, divergent preferences on how to coordinate. These actors (national governments and the commission's representatives) had certain preferences, meaning they preferred one alternate to another and could be expected to try to maximize their utility, when they tried to reach bargaining outcomes that were as close as possible to their preferred bargaining position. Every actor considers the consequences of all the possible outcomes from which a choice can be made, and ranks these outcomes in order of preferences, from the most to the least preferred. This preference ordering is transitive, that is, if a is preferred to b and b to c, then a must also be preferred to c.

Actors’ preferences become apparent through their policy or negotiating positions on different issues. I have used council minutes and secondary sources (especially Da Conceição‐Heldt 2004; Farnell and Elles 1984; Leigh 1983) to assess the negotiating positions of the actors involved. Foreign policy decisions traditionally fall within the domain of the executive, and decision making in the EU does not constitute an exception to this rule. Thus, states (understood as national governments) are treated as rational and unitary actors, that is, they try to maximize their policy preferences and speak with one voice in EU negotiations.

Table One lists the policy positions of national‐government representatives of the six member states and of the commission at the beginning of the bargaining process. The negotiations on the structural policy centered around the three different proposed methods of financing the policy. They were:

• 100 percent EC financing (represented in the table as x),

• 50 percent EC financing (y), or

• 0 percent EC financing (z), which was the status quo.

Germany, the Netherlands, Belgium, and Luxembourg preferred to have a policy financed entirely at the national level.⁸ Each had the following transitive preference on the structural policy: z was preferred to y and y to x (z>y>x). France and Italy, in contrast, wished to have a structural policy financed entirely by the EC (x), but 50 percent financing from the EC (y) was better than keeping the status quo (z) (x>y>z). The commission, in turn, preferred to have greater integration in fisheries policy. Accordingly, it preferred x>y>z because 100 percent financing by the member states of the structural policy at the European level would have put the responsibility for administration and monitoring at the supranational level, in the commission's hands (Da Conceição‐Heldt 2004: 57f.).

In the second policy issue under discussion, the common organization of the market, three options were also on the table: more market liberalization (m), less market liberalization (l), or maintenance of the status quo (s). Whereas Germany, the Netherlands, Belgium, and Luxembourg preferred m to l and l to s (m>l>s), France and Italy preferred s to l and l to m (s>l>m). Those member states against a structural policy financed entirely by the EC wished to have a common market organization with more market liberalization, and those member states in favor of a structural policy financed by the EC vetoed the common market organization. Whereas the main aim of the common market organization was to have a free movement of fish products within the EC, the structural policy would have helped member states adapt to this change by providing financial assistance to modernize their fishing fleets.

The commission had no clearly delineated preference regarding the common organization of the market. The central issue for the commission was whether the decision‐making competence in the fisheries policy should be transferred to the European level. The question of whether more or less market liberalization should take place was rather secondary for the commission. Most importantly, the commission used its formal agenda setting powers to establish the basic principles for a common policy guaranteeing equal access to fishing grounds, which went far beyond the demands of both groups for a structural policy and for a common market organization.⁹ In this way, the commission, a bureaucracy with a long‐time horizon, used this short window of opportunity to achieve what Oran Young would call “entrepreneurial leadership,” which he defines as the ability to “frame issues in ways that foster integrative bargaining and to put together deals that would otherwise elude participants endeavoring to form international regimes through institutional bargaining” (1991: 292).

Because the most important issue for member states was to try to achieve a compromise deal on the other two issues — and because altering the original proposal of the commission would have required unanimous agreement — the proposal was not changed, and the discussion on how to deal with the equal access to fishery resources was postponed until the next bargaining round. This confirms Paul Pierson's (1996) thesis that national‐government representatives have short time horizons and are primarily concerned with the next elections, so they pay little attention to the long‐term future and find themselves “locked in” to the integration process because future policy changes are linked to previous ones. As time unfolds, the probability of continuing along the same path increases, while the probability of significantly deviating from the established paths decreases (Da Conçeicão‐Heldt 2004: 51).

Given these bargaining positions, how might a compromise agreement have been found? And what role did the formal unanimity voting procedure, which protects member states from being outvoted in the council, play in the final bargaining outcome? It is generally assumed that under unanimity voting, every actor is a veto player, defined by George Tsebelis (2002) as “an actor whose consent is necessary to change the status quo.” So bargaining under this rule can be extremely time‐consuming and lead to stalemate. Surprisingly, in this particular negotiation, member states did not make use of their veto power. Instead, two coalitions were built, so that issues could be linked. Before discussing the actual bargaining process, let me briefly introduce some theoretical considerations about the ways in which coalition building and issue linkage may occur.

Coalition building is a useful instrument actors employ to maximize their total payoff. Players aggregate their preferences and form coalitions when they want to either reach or prevent an agreement. Coalition building makes bargaining easier by reducing the number of parties and preferences involved. Following James Morrow (1986), I assume that a coalition only arises when the status quo across one or several issues can be challenged. Coalition building in the EU revolves around long‐term interest groupings that can be geographic (northerners versus southerners), interest‐oriented (farmers, industrialists, etc.), based on budgetary issues (net contributors versus net receivers), or on size (larger versus smaller nations) (Hosli 1996; Spence 1995).

Member states built two coalitions in this integrative bargaining situation: France and Italy built what I call here Coalition A; and Germany, the Netherlands, Belgium, and Luxembourg formed Coalition B. The primary objective of Coalition A was to have a structural policy fully financed by the EC, in order to obtain communitarian aid for the modernization of fishing fleets and to redeploy redundant labor. France and Italy were ready to accept a common organization of the market for fish products if withdrawal prices (minimum prices below which fish cannot be sold) were introduced. The primary aim of Coalition B, on the other hand, was to avoid excessive financial communitarian intervention and to foster a minimal coordination of the structural policy. This was clearly reinforced by the fears of its members that they would have to pay the largest share of the new system with a higher contribution to the EC budget (Leigh 1983: 26).

In other words, Coalition B had no interest in adopting a structural policy primarily advantageous to Coalition A. Germany, Belgium, and the Netherlands, with modern and well‐organized fishing industries, needed little financial assistance and were reluctant to finance a structural policy at the European level. This would seem to confirm the taxpayer model developed by Thomas König and Thomas Bräuninger (2001) in which they argue that the fewer gains a member state receives from a certain policy, the less willing it will be to fund it.

The second negotiating tool used to overcome the unanimity or joint‐decision trap, which would have maintained the institutional status quo, was the linkage of issues. In an issue‐linkage situation, an actor is willing to give up something of value with respect to one issue if he will receive concessions with respect to another more important issue in exchange. Decisions are simultaneously taken on different issues so that they can be accepted by all players. Without the existence of issue linkages, member states that benefit in one policy area but receive negative payoffs in another would only support the first issue (Milner 1997). As Young (1989: 365) observed, “linking together disparate issues sometimes opens up possibilities for mutually acceptable arrangements by creating opportunities for the international equivalent of logrolling and the formulation of package deals.”

Numerous studies on issue linkage in international relations indicate that issue linkage might be successful in a bargaining situation when the parties involved value the issues differently (Tollison and Willett 1979; Weber and Wiesmeth 1991). James Sebenius (1983), for instance, argued that adding issues to a negotiation can yield joint gains and change the zone of possible agreement among the parties involved. Two forms of issue linkage are possible: internal linkage, which refers to linking two different issues within one policy sector, and external issue linkage, which involves the linkage of different policy sectors. Because of the highly polarized preferences in this particular integrative bargaining situation, internal issue linkage was used as a negotiating tool for overcoming the “unanimity trap.” Detailed analyses of other policies in the EU have revealed that issue linkage has often been a very useful bargaining tool.¹⁰

The council and its working groups discussed the proposals for a common fisheries policy (CFP) for almost four years. Although the proposals were revised many times, the divergent negotiating positions of the two coalitions prevented an agreement from emerging. By early 1970, the beginning of negotiations for membership (accession negotiations) with new EC candidates Denmark, Ireland, Great Britain, and Norway, that each had a strong interest in fishing issues and control of rich fishing grounds, pushed reluctant member states, especially Germany, the Netherlands, Belgium, and Luxembourg, into accepting a compromise.

The commission argued persuasively that, if agreement to the CFP package was to be delayed until the beginning of accession negotiations, the candidates for membership would demand to be consulted. Furthermore, a package deal taking the interests of the candidates into account might have been far less favorable to the six present members than a deal which they might agree upon before the accession. On June 30, 1970, the day before accession negotiations started, the six reached an agreement on the CFP creating a fait accompli to be accepted by the accession candidates. But the way in which the new regulation was hastily put together created misgivings in the candidate countries and gave rise to acrimonious debate on fisheries after the accession.

The adoption of a CFP with a structural policy and a common market organization for fish products represents a typical compromise deal containing something for everyone: for France and Italy, there was a structural policy financed 50 percent by the EC, while, with the common organization of the market, Germany, the Netherlands, Belgium, and Luxembourg achieved their preferred position. Member states were able to reach a compromise because concessions on one policy issue (structural policy) were balanced by concessions on another (common organization of the market).

The international negotiations on the United Nations Convention of the Law of the Sea in 1973, which led to an extension of national fishing zones from a 12‐ to a 200‐mile exclusive economic zone (EEZ) in 1982, posed major challenges to the EC. The member states lost access to their traditional fishing waters in such far away areas as Iceland, the Barents Sea, and the banks of Newfoundland. It was this external event that led to the beginning of negotiations at the European level on the common management of fish stocks.

In November 1976, the commission presented a proposal for a council regulation establishing a community system for the conservation and management of fishery resources. Fishing grounds situated in territorial waters, contiguous, or reserved zones were to be considered community fishing grounds to which the access and the exploitation should be based on nondiscriminatory rules (Leigh 1983: 26). The objectives of the proposed regulation were to protect fishing grounds by setting conservation measures through total allowable catches (TACs), which annually specify the maximum quantity of each fish stock that can be fished, and also to establish a national quota system to distribute the volumes of the catches among the member states. Such conservation measures were deemed necessary because the loss of foreign fishing grounds was expected to put added pressure on European fishing grounds.

The most controversial issues that arose during the negotiating process were linked to the rules on the access to fishing waters, specifically to the question of whether there should be restricted access to fish stocks in the EC's waters, and to the division of catch quotas among member states. Although the council began to deal with these issues for the first time in December 1976, the member states did not approve the regulations on the conservation and management policy until January 25, 1983 — following six years of bargaining processes.

Distributive‐gains negotiations involve at least a two‐step process. In the first step, the central issue is the identification and definition of the parameters of disagreement and the set of possible bargaining outcomes that involved parties can agree upon. In the second negotiating step, players must divide the gains among themselves (Cross 1996: 156). In other words, although there is a consensus that policy coordination can bring joint gains to all involved parties, agreement on its distribution has to be reached before the gains can be achieved.

In the distributive bargaining over the conservation and management system of fishery resources, the central goal was to ensure that fish stocks would not be depleted in the present at the expense of the future. Recognizing the need to manage fish resources, however, was only the beginning. The member states still had to agree on how to divide the TACs among themselves and on how to apply the principle of equal access to fish stocks in EC waters.

There were two distinct negotiating phases: in the first phase, Great Britain and Ireland opposed the principle of equal access, and in the second negotiating phase, the focus was on how much of the total catch quotas should be given to Denmark.

At the beginning of negotiations in 1976, the policy positions of member states could be divided into two major groups. On the one hand, a majority of member states with interests in fisheries issues, including Belgium, the Netherlands, France, Germany, and Denmark, favored an unrestricted application of the equal‐access principle because they had shorter coastlines or poor fish stocks and were unable to satisfy their fish needs in their own coastal waters. On the other hand, Great Britain and Ireland asked for more fish catch quotas and a restricted application of the principle of equal access to fish stocks in order to keep certain areas for their exclusive use (Da Conceição‐Heldt 2004).

Because British and Irish coastal waters are rich in fish resources, their fishing vessels did not need to enter the waters of other member states. A change in the status quo was Pareto‐inferior for both countries because most of the catch of the EC had occurred in British and Irish waters before the settlement of the 200‐mile EEZ. Accordingly, both countries demanded an EEZ of 50 miles within the future 200‐mile EEZ of the EC. In addition, Great Britain wished to have a principle known as “fair return” applied to fisheries and asked for 60 percent of the EC total catch quotas (Wise 1988).

Table Two shows the transitive preferences of the member states and of the commission regarding the application of the equal‐access principle. Germany, the Netherlands, Belgium, Luxembourg, France, Italy, and Denmark preferred an unrestricted (u) application of the equal‐access principle over a restricted application of the equal‐access principle (r) and the status quo (s) of no catch limits in European fishing grounds. In contrast, Great Britain and Ireland's preferences were restricted, unrestricted, and status quo (r>u>s). The commission was indifferent between the restricted (r) or unrestricted (u) application of the equal‐access principle because both were better than keeping the status quo (s). The most important issue for the commission was conservation of fish resources, so the commission's goal was a centrally monitored and enforced quota system, in which the commission would propose total catch quotas for the entire EC every year.

In an effort to move the negotiations forward, the commission proposed a package of proposals for an informal meeting of the Fisheries Council in Berlin in January 1978 that British representatives did not attend. The commission, with the acquiescence of all present member states, offered Ireland substantial quota increases, a fishing area reserved for Irish fishermen (the Irish box), and the prospect of EC financial aid within the framework of the structural policy for fisheries. The Irish representative accepted this compromise package and abandoned its demand for an exclusive 50‐mile zone (Leigh 1983).

Obtaining Irish consent had been relatively easy and now the commission and the other member states still had to find a way to buy off Britain's consent. After the legislative elections in May 1979, the new British government replaced the expression “exclusive 50‐mile zone” with “dominant preference” for the 50‐mile zone and asked that fishing areas around the Shetland and Orkney boxes be reserved for British fishermen. Great Britain obtained this concession and accepted a final compromise giving it 36.1 percent of the total fish allocation, the highest catch quota granted to any member state (Leigh 1983: 91ff.).

Because both countries achieved side deals in exchange for the lifting of their vetoes, it seemed that all obstacles for settling a conservation and management policy had been cleared. The Danish government, however, now perceived the settlement of the Shetland and Orkney boxes as clearly disadvantageous to its fishing industry, and Denmark rejected the whole compromise. Instead, it asked for additional fishing licenses in Britain's restricted‐access area and threatened that if it were not guaranteed satisfactory access to mature herring in British waters, unrestricted fishing of the juvenile herring would be permitted in Danish waters (Holden 1996: 237), which would obviously be an unsound practice from a conservation standpoint. In June 1982, the commission gave Denmark 23.5 percent of the TACs in its new proposals for the distribution among member states. The Danish government, in turn, asked for 30 percent. When the council met on October 25 and 26, 1982, the commission suggested instead that the fishing licenses in the Orkney and Shetland boxes should be reduced by three hundred square kilometers, a change benefiting Danish vessels. But the Danish government representative refused to accept the modified proposal (Leigh 1983).

Irritated with this Danish intransigence, the other member states and the commission used threats and deadlines to try to bring the negotiations to a successful end. Under the leadership of the commission, the other member states gave Denmark until November 5, 1982 to accept the “last” compromise proposal put on the negotiating table by the commission, threatening that the rest of the EC would otherwise implement the commission's previous proposals, which were less favorable to Denmark.

Threats are often employed as a tool to obtain concessions in negotiations. They send a signal that the sender (in this case the EC) will take a certain action harmful to the other negotiating party (Denmark) if the latter does not accept a certain compromise proposal (see O’Neill 1991). But for the other party to accept the offer, the threats must be credible and the other party must believe that it will be fulfilled and sanctions will be applied (Schelling 1960).

In order to make the threat credible, the other member states agreed to adopt the conservation and management policy in January 1983, leaving Denmark outside the agreement. But even though the threat proved credible, the stalemate remained. Denmark held the presidency of the council and, supported by France, refused to proceed to a vote by referring to the Luxembourg compromise. The permanence of EU negotiations leads to the prevalence of such strong consensus norms as the Luxembourg compromise, which states that member states are not required to vote on a commission proposal if a vital national interest, which does not have to be specified, is affected.¹¹ It is likely that an agreement might have been reached earlier if the council had not been functioning under such provisions throughout these negotiations.

Although the EC's offer had been made on a take‐it‐or‐leave‐it basis, there were nevertheless further attempts to break the deadlock. The commission offered more concessions to Denmark, specifically more catch quotas and 24.1 percent of the TACs (Leigh 1983). Denmark responded to the EC's “final offer” with requests for additional tons of fish with “staying power,” that is, continuity for more than one year.

Denmark's next moves confirm both Thomas Schelling's (1960) arguments about the power of weakness (an executive can point to a recalcitrant legislature to extract greater concessions in international negotiations) and Robert Putnam's (1988) model of “the two‐level game,” which explains how domestic politics may influence an actor's bargaining position in international negotiations. Denmark played the internal‐weakness card credibly and was consequently able to bargain with the EC from a position of strength by arguing that domestic pressures (from interest groups and also from the Danish Parliament) prevented the country from agreeing to any disadvantageous compromise solutions.

At the Fisheries Council meeting held in December 1982, the Danish government representative still refused to agree on the TACs. Given the possibility that another year might end without a compromise solution, the commission called upon the member states to implement its proposals by national measures. Moreover, the council even discussed approving the conservation and management policy package by qualified majority.¹² Although the other member states could not change the rules without the consent of the Danish government, they were able to adopt interim conservation policy measures for January 1983 and put pressure on the Danish government in this way. This, together with the side deals in the form of more catch quotas, was crucial in persuading Denmark to accept the last compromise proposal.

After almost seven years of acrimonious negotiations, on January 25, 1983, the EC finally approved the regulations establishing the conservation and management policy of the CFP. All member states were exhausted by the long negotiations, and the resulting compromise was compared to a “house of cards” (Holden 1996: 56). In order to avoid future stalemate situations, the member states decided to abandon unanimity voting in fisheries issues.¹³

I draw two major conclusions from the study of this distributive bargaining situation. First, when a member state blocks an agreement, the commission may act as a broker searching for compromise. In the last bargaining phase with Denmark, the commission played a central role in bringing negotiations forward by suggesting that member states agree to the proposed legislative framework, with the prospect that this might be modified later according to demands.¹⁴ Second, this bargaining situation shows how veto players can resist making concessions by assuming a hard negotiating line. This confirms James Fearon's (1998) hypothesis that, in international negotiations, the longer the “shadow of the future,” the more likely the use of a tough bargaining strategy. Before transferring a competence to the European level, which will bind member states for a long time, they will seek to ensure that the settlement is not Pareto‐inferior for them. If a certain member state, however, considers itself disadvantaged by a particular regulation or directive proposal, blockade can only be overcome through compensation in the form of side deals.

Examining the impact of adding players to or subtracting them from a bargaining situation allows one to analyze the dynamics of reaching agreement under unanimous decision making. James Sebenius (1983) hypothesized that the more parties (and issues) involved in a negotiation, the higher the costs and the longer the time it takes to reach an agreement.

In the integrative bargaining situation with seven actors (six member states and the commission) there were essentially two diverging policy positions: Germany, the Netherlands, Belgium, and Luxembourg wished to have a liberalization of markets for fish products, while France and Italy wanted to have a structural policy to modernize their fishing fleets totally financed by the EC. The commission was indifferent to market liberalization, but was highly interested in a policy administered at the European level. These divergent interests led the member states to build two coalitions in order to maximize their total payoff: the costs of linking issues (common market organization with structural policy) were not too high and these two policy dimensions could be linked together in a package deal facilitating agreement. In this way, overcoming the unanimity decision rule was not a problem.

In the distributive bargaining situation, however, some member states (Great Britain, Ireland, and Denmark) made use of their veto power under the unanimity decision rule to temporarily block a policy proposal that they perceived as disadvantageous to them. For that reason, the settlement on a conservation and management policy was delayed for more than half a decade.

Although there were six repeat players (Belgium, France, Germany, Italy, Luxembourg, and the Netherlands), the three new member states (Denmark, Ireland, and Great Britain) dominated the protracted negotiations. These three countries have rich fishing grounds and made use of their veto power so that an agreement could only be reached by buying their consent with side deals. Member states had no incentive to build coalitions, because only one policy issue was bargained upon and, more importantly, each member state was primarily concerned with its relative gains, that is, how much it would get from the total fish allocation compared with what the others would get. In other words, every member state was more concerned with maximizing its own benefits and with minimizing its own losses, a classic relative gains situation.

Integrative dynamics in the EU are different from distributive ones. During the negotiations on the integrative bargaining situation from 1966 to 1970, the focus was on maximizing the benefits for all member states from a common policy. In contrast, the distributive bargaining situation from December 1976 to January 1983 dealt with a fixed‐sum process of dividing the fisheries resources among member states.

During the last bargaining stage in 1982, under the leadership of the commission, the other member states threatened Denmark with the adoption of the regulations with QMV in order to persuade it to accept the CFP settlement. This was, however, not put into practice. It was only a way of putting pressure on the Danish government.

The consensus rule was a common feature of both bargaining situations. Hard bargaining characterized the distributive bargaining situation, but the consensual style of decision making predominated in the integrative situation. Although member states have different preferences on policy issues, the consensus‐seeking approach of the council encourages the development of compromise solutions that they can all accept. Both bargaining processes indicate that EU negotiations only lead to agreement if each party expects an outcome that is no worse than the status quo. A member state can be persuaded to accept a policy outcome that may be difficult to defend at home if it is sweetened by side deals (substantial quota increases in the case of Denmark and financial aid within the framework of structural policy for fisheries in the case of Ireland). So, if a member state cannot block a decision indefinitely, it may demand a price for its consent.

Furthermore, both bargaining situations illustrate how the commission is able to influence negotiations in the EU. In the integrative bargaining situation, it exploited the differences between member states and introduced its own preferences into the regulation draft. When asked by France and Italy to prepare a regulation draft on the structural policy and by Germany, the Netherlands, Belgium, and Luxembourg to prepare a proposal on the common organization of the market, the commission used this window of opportunity to include elements for a general common policy administered at the supranational level by the commission itself. During the bargaining process, the commission acted merely as a mediator between member states.

In the distributive bargaining situation, in turn, the commission had a stronger influence than in the integrative negotiation. Confronted with the use of veto power to block an agreement, the commission not only kept the negotiations going by redrafting policy proposals, but it also took on the role of a broker. It used the exclusion threat by proposing to vote with qualified majority and thus bypass the Danish veto. Because the commission's compromise proposals remained blocked at the council, it persuaded the member states to implement other aspects of the resource and conservation policy on their own. In this way, the commission played a central role at the negotiating table by keeping the negotiations going and by trying to move the member states toward an agreement.

The primary aim of this article was to show how integrative and distributive bargaining situations can differ. The two key negotiations that led to the Europeanization of the fisheries policy show how different negotiating tools can be used under unanimity voting and how the commission influenced bargaining outcomes. Based on my study, I draw four conclusions.

First, the integrative bargaining situation shows that when the preferences of member states are strongly polarized, the number of actors involved is small, and two or more policy issues are negotiated, coalition building and issue linkage are the negotiating tools used to reach an agreement.

Second, the distributive bargaining situation demonstrates that when only one policy dimension is bargained, deadlocks can only be broken if the reluctant member states obtain side deals (more catch quotas and restricted fishing areas in this case) that buy their consent. The unanimity voting rule enhances the bargaining power of each actor. This feature explains why the costs of reaching agreement under unanimity can be extremely high.

Third, in this article I have highlighted the central role played by the commission as an autonomous player and as a broker during the negotiations. The commission acted as an entrepreneurial leader to help member states reap the bargain's surplus. In the integrative bargaining situation, it functioned as a mediator between diverging preferences and as an agenda setter by shaping the form in which issues were presented for bargaining and by devising innovative policy options to overcome bargaining obstacles through the use of issue linkage. Furthermore, it did not act as a simple agent waiting for tasks from the principals. Because its compromise proposals remained blocked in the council, it assumed an important role in persuading the member states to implement other aspects of a common policy, which were subsequently approved.

Finally, the bargaining dynamics of the two cases examined here differed significantly. The focus in the integrative bargaining situation was on increasing the joint gains available to all from a CFP. It describes a coordination situation without distributional conflict, in which all member states had a clear preference that TACs as well as their national quotas be increased. The distributive bargaining situation, on the other hand, dealt with more competitive bargaining behavior. There was a consensus among member states that policy coordination brought joint gains to all, but they disagreed on how to divide fish stocks gains among themselves. While the integrative bargaining situation was linked to the priority of absolute advantages, member states were concerned about their relative advantages in the distributive bargaining situation, that is, with how well they would fare when compared to other member states.

European Union member states have the ability to make flexible use of unanimity voting and to overcome the so‐called “join‐decision trap.” In contrast to Fritz Scharpf's findings, I assert in this study that joint decisions do not automatically lead to the preservation of the institutional status quo. In the integrative bargaining situation, member states were easily able to overcome the joint‐decision trap through coalition building and issue linkage. In the most controversial distributive bargaining situation, member states, because of the active role played by the commission, reached agreement with the help of side deals and threats of excluding the most reluctant member state. These findings illustrate that there are some ways in which the unanimity or joint‐decision trap can be overcome, although unanimity requirements in multiparty negotiating situations do present many obvious difficulties. Although there is a tendency in European studies to focus on member states’ preferences or on the institutional rules, in this article I have tried to show that bargaining situations in themselves are critical. Consequently, we should not underestimate the impact that the type of bargaining situation will have on outcomes.