The scope for action of European Parliament negotiators in the legislative process: lessons of the past and for the future Anne Rasmussen and Michael Shackleton* University of Copenhagen / European Parliament Paper prepared for the Ninth Biennial International Conference of the European Union Studies Association, Austin, Texas, March 31-April 2, 2005 Abstract The European Constitution dramatically expands the areas subject to the co-decision legislative procedure. The paper considers the potential consequences of this expansion for the ability of the Parliament as a whole to exercise control over individual members who negotiate with Council to look for agreement on legislative outputs. It looks back at the recent record of the Parliament, comparing the room for manoeuvre of its negotiators in the third and final conciliation phase of codecision with the earlier stages of the procedure. It takes as its starting point the claim in the literature that going to conciliation poses a risk for parent legislative bodies. They may lose their legislative power by delegating responsibility for solving disputes to a small number of influential negotiators. The paper argues that in fact conciliation in the EU has not proved a risky strategy for the Parliament: However, the lack of formal rules at first and second reading in co-decision increases the risk of loss of control. Hence the importance of observing the mechanisms developed by the Parliament to deal with this risk as the potential for more agreements being reached at an earlier stage of the procedure increases with the approach of the ratification of the Constitution. 1. Introduction The European Constitution provides for a very major expansion of co-decision from 40 to 89 legal bases. It will become the "ordinary legislative procedure", with important but limited exceptions (Parliament 2004c). The impact of this change has been discussed so far chiefly in terms of the increase in the role of the European Parliament in the adoption of EU legislation and the potential for increased democratic accountability. However, it is also a change that will have important implications for organisations in the EU and the way in which they interact with each other. The extent of such interactions can be expected to increase in the same way as it has done in the ten years since co-decision was introduced in the Maastricht Treaty. Thus 166 acts were adopted from 1994 to 1999 but this figure rose to 403 acts from 1999 to 2004, after the Amsterdam Treaty more than doubled the areas covered by the procedure from 15 to 32. More importantly perhaps, increased interaction is likely to lead to new patterns of negotiation between the two co-legislators, Parliament and Council, in the same way as it has done since 1994. Before 1999 co-decision acts could only be concluded at second reading or after conciliation but following Amsterdam, it became possible to conclude at first reading. Over the following five years, the range of possibilities was fully exploited, with both Parliament and Council devoting considerable energy to concluding procedures in advance of conciliation. Thus whereas 40% of co-decisions required conciliation to be concluded in the Maastricht period, this figure dropped to 20% (86 in total) over the following five years, with 80% of legislative acts adopted at first and second reading. As Figure I shows, by the end of the 1999-2004 parliamentary term almost as many pieces of legislation were being agreed at first reading as at second, reflecting the strength of a strong culture of compromise between the two organisations. Figure 1 - Evolution of the stage of conclusion 1999-2004 The pattern of the last five years would suggest that with an increased volume of legislation, covering a wider range of domains, including agriculture, fisheries and justice and home affairs, there is likely to be a continuing strong desire to conclude earlier in the procedure and thereby to avoid the heavy obligations imposed by going to the final phase of conciliation. Moreover, this desire matches the wish of individual Council Presidencies to be able to maximise the number of issues that they can claim to have resolved during their six months in charge and the tendency of Parliament rapporteurs to want to retain their central role in discussions with Council. According to the literature on bicameral bargaining, seeking to avoid conciliation in this way should be beneficial to the Parliament as a whole. Going to conciliation poses a risk for the parent bodies as conciliation committees typically enjoy substantial power over the legislative outputs. The Parliament as a whole may lose legislative power by delegating responsibility to a small number of influential negotiators for reaching a compromise, to which the legislative bodies merely get to react in a yes or no vote. And yet delegation of responsibility is not restricted to the conciliation phase of co-decision. Similar arguments would seem to apply at the earlier stages of the procedure, with the vast majority of members of the Parliament, as well as those outside it, only able to follow with difficulty the progress of negotiations before they are invited to vote at first or second reading. To evaluate these competing claims, this paper examines the room for manoeuvre of the Parliament's negotiators in conciliation as compared with the previous stages of the procedure in co-decision in the 1999-2004 parliamentary term. Despite the warnings in the literature of bicameral bargaining, it is argued here that going to conciliation does not provide key negotiators of the Parliament with greater room for manoeuvre than they have at first/second reading. In fact, contrary to these warnings, it is shown that their room for manoeuvre generally diminishes in conciliation. An effective formal and informal structure specifies the boundaries within which negotiations are conducted and ensures that negotiators act on the basis of a mandate from a representative conciliation delegation. By contrast, no similar system exists for bargaining at first and second reading, where the scope for action of key negotiators is considerably less regularized and negotiation practices more varied. This points to two conclusions. First, whereas there are often efficiency advantages in concluding early on during the legislative procedure, there are also risks for the Parliament as a whole. These are reinforced by the Constitution, whose expansion of the legislative power of the Parliament will put greater work pressure on its members, their political staffs and the Parliament's Secretariat. Hence the approach of the Constitution will oblige the Parliament to seek a balance between the desire to improve the efficiency of the procedure and the potential loss of control over outcomes by the Parliament as a whole. Second, the study refines the prominent view of conciliation committees in substantial parts of the academic literature by illustrating how institutional constraints can significantly curtail the power of conciliation committees in practice. In the EU, we seem to have an extreme case where informal procedures ensure the chamber as a whole high control of legislative outputs, even though its formal involvement after having delegated the responsibility to the committee is reduced to a yes or no vote on the final text. This actually leaves the chamber better off than at earlier phases, where amendments can be tabled as compromises and are voted on separately, but in practice there is rather little informal control of the work of the key negotiators. The analysis proceeds in two stages. First, we consider the distinction between formal and informal institutions as constraints on the room for manoeuvre of key negotiators in the light of the broader literature on bicameral bargaining, notably in the United States and Germany. Second, a comparison is made between the institutional constraints faced by key negotiators in Parliament at first/second reading and in conciliation. Our findings are based on a review of written sources, such as the Treaty, the joint declaration between Parliament, Council, and Commission about co-decision, internal conciliation guides from the EU bodies, annual reports of the Vice-Presidents responsible for conciliation, relevant secondary sources as well as interview data with people involved in the co-ordination of co-decision and/or involved in negotiations of specific cases. The article concludes with some remarks on the implications of the results for the future structure of negotiations between Parliament and Council as well as for work inside the Parliament. 2. Formal and informal constraints in comparative perspective The room for manoeuvre for key negotiators to exert legislative influence varies as a result of differences in the formal and informal constraints under which they act and of differences in the ability of other actors within the Parliament to influence the results. Hence the importance of identifying the nature and extent of the constraints that apply in the Parliament and the opportunities available to influence outcomes. Formal and informal constraints are often referred to as "institutions" in the literature, i.e. "a set of rules that structure social interactions" (Knight 1992: 2). These are to be differentiated from the EU Institutions themselves, which are not institutions in our terminology, but rather collective actors, here named "organisations", that are themselves subject to institutional constraints (Knight 1992: 3; North 1990: 5). We refer to both formal and informal institutions but recognise that the difference between the two is not completely clear-cut. One way of making the distinction clearer is to regard formal institutions as rules reinforced by a third party, whereas informal rules are enforced by the actors themselves. In our case, this means that the formal institutions are the Treaty provisions, whereas other types of constraints on the actors such as the Joint Declaration on the practical arrangements for co-decision, the annual Conciliations guides of the Parliament, and internal rules of procedure of the bodies are of a more informal character. Looking at the formal constraints reflected in the Treaty provision of the procedure, the Parliament has the hardest time controlling a compromise when it comes out of the conciliation committee. At this stage the chamber cannot adopt any more amendments to the text or vote on it in individual parts, but has either to adopt or reject the entire text. Hence, unless the chamber finds the compromise so bad that it prefers no legislation to it, it is forced to adopt the text as it is . On the other hand, the chamber is better equipped with formal tools to control the key negotiator at first and second readings which are not subject to such a "closed/up and down" vote, but where new amendments can be tabled and voted on separately. This is exactly the story with which we are presented in the most comprehensive, comparative study of conciliation committees in recent years by Tsebelis and Money. They argue that, even though there are differences across political systems as to how constrained conciliation committees are, these committees "have extraordinary powers, stemming from the fact that generally their proposal cannot be amended" (1997: 117). Because of their capacity to make the final proposals under closed rule they are important, and "delegating the power of agenda setting to the conference committee presents the parent chambers with a serious danger" (1997: 176). This allows the committee members to pick exactly the compromise which they like the most among all those preferred by the parent bodies to no legislation. If we look more specifically at the literature of two well-known conciliation systems such as the US and Germany, we also have clear evidence of the power that negotiators in these conciliation committees enjoy. Even though many rules have been established over the years to constrain the behaviour of US conference delegations, there has been a constant criticism during most of the history of Congress of the disproportionate influence, which members of these bodies enjoy on the policy outcomes (McConachie 1898; Luce 1922; Rogers 1922; McCown 1927; Shepsle and Weingast 1987; Tiefer 1989; Rasmussen 2005). Recently, Oleszek, a senior specialist in American Government, has argued, about the scope rules that limit conference behaviour, that "today violations of scope have become almost routine in the House, special rules regularly waive all points of order against the conference report and its consideration in the House" and that "new lawmaking by Senate conferees is now the order of the rule for the conference report" (2001: 261). He presents examples of such behavior and quotes Senator Alan K. Simpson saying about conference committees that, "They're mystical. They're alchemy. It's absolutely dazzling what you can do" (2001: 262) . Also the German conciliation committees have a reputation for being very strong and have been named the third chamber by Mastias and Grangé (1987), the same title given by Vogler to the US conference committees in his 1971 book. A judicial dispute about their scope for behaviour ended with a German High Court Ruling in 1986 favourable to their power. It argued that bills that were on the table but not approved in the Parliament could be included in the conciliation reports (Vogel 1989). Other conciliation committees such as the Japanese and the French are not as strong as the German and the US ones, but all share the important formal tool that they produce compromises which come to the chamber under a closed, yes or no, vote. This gives key negotiators some advantages over the ordinary chamber members compared to what they have at prior legislative stages. However, it is not only these formal constraints that matter. We also devote considerable attention to analysing other types of less formalised constraints, within which the negotiators act throughout the legislative process. Important to us is the extent to which formal and informal institutions interact in affecting the room for manoeuvre of the negotiators. In some cases, informal institutions, such as the 1999 Joint Declaration between Parliament, Council and Commission will strengthen the effect of the formal Treaty provisions, whereas in other cases they will modify it. Thus, it may be that even though the key negotiators of the Parliament enjoy greater formal power in the conciliation phase than at earlier legislative stages, the picture may be different when we take other types of less formalised constraints into account. This may partly explain why we are not faced with criticism in the EU literature like that in the US and German literature that conciliation committees overstep their mandates and give a few MEPs disproportionate influence on the legislative results. In many ways, the existing literature of co-decision seems to reach the exact opposite conclusion. Thus, recent pieces have referred to the democratic consequences of concluding at first reading for the control of Parliament as a whole. Farrell and Heritier, who have made the most detailed studies of the effects of first track agreements so far, argue that, "new relationships are being created between power-brokers in the European Parliament and figures within the Council, which may lead, in extreme cases to the short-circuiting of democratic processes of deliberation in committee and extensive plenary discussions" (2003a: 30). In general, they claim that "Parliament fears that it may be paying too high a price for influence, by being drawn into the kinds of secretive bargaining that better characterize inter-state negotiations than democratic deliberation" (2003a: 26) . Also Raunio and Shackleton state, "Certainly, there was and remains a recognition, for example, that 'fast-track agreements' at first reading are a legitimate element of the legislative procedure….However, the consensus is not as wide as it might appear" (2003: 176). Thus, whereas a large share of especially the US and German conciliation committee literature focuses on the negative effects of the formal conciliation procedures, the EU literature goes deeper into the informal consequences of the decision-making procedures. This may partly explain that the conciliation phase in the EU literature is portrayed in a more positive light than elsewhere. In any case, it strengthens our argument that there is a need to examine both the formal and informal aspects of co-decision to understand the net effect of these on the room for manoeuvre of key negotiators at the different stages of the procedure. Highlighting the democratic consequences of concluding early in co-decision is not new, but using somewhat similar empirical information to discuss the advantages of using different methods of reconciling differences between two legislative bodies is. 3. Institutional constraints at different stages of co-decision In order to examine how different types of institutions constrain the behaviour of key negotiators, no matter whether they are formal or informal, we distinguish between five types of institutional constraints illustrated in Table one. First, there may be institutions specifying who should be appointed as key negotiator/negotiators. Are any special institutional figures, people with a specific opinion, political colour etc. typically appointed? Second, there may be institutions specifying the scope for action of key negotiators in substantive terms. On what basis do they negotiate and what can they compromise about? Third, there may be institutions specifying the practical limits, within which negotiators act, such as time limits on their negotiations, demands to report back orally/in writing to the plenary or certain bodies within the Parliament, demands to negotiate on the basis of a mandate etc. Fourth, there may be rules specifying whether key negotiators need the support of a body within Parliament before it can move on to the plenary and, if so, what type of majority is required in such a body. Fifth, there may be institutions specifying how difficult it is for key negotiators to get their compromises adopted in plenary subsequently. What kind of majority is needed at this level? How easy it is for the compromise to get on the agenda? Is there a time limit for approval of the compromise? Seen as whole, these are the kind of institutions, which affect the room for manoeuvre of the key negotiators. Below we look at the extent to which such institutions are in place in the Parliament in the three co-decision readings. Table 1. The five types of institutions First reading Second reading Conciliation committee and third reading Type 1 Institutions: key negotiators Body in charge within Parliament Standing committee Standing committee Conciliation delegation Actors with negotiating power within Parliament Rapporteur (committee chair) Rapporteur (committee chair) Vice-president,committee chair, rapporteur Type 2 Institutions: substantive elements Scope for negotiations Not defined, no Council and Parliament position, open amendment rule. According to Parliament's Rules of Procedure (2004d), Rule 150, amendments can be tabled by the committee responsible, a political group or at least 37 MEPs. Council's common position, Parliament's possible second reading amendments Rules for who can table amendments identical to first reading but amendments restricted to those, which seek: - wholly or in part, to restore the position adopted by Parliament in first reading; or - to reach a compromise between the Council and Parliament; or - to amend a part of the common position, which was not included in – or differs from – the proposal submitted in first reading and which does not amount to a substantial change - to take account of a new fact or legal situation since first reading Parliament's second reading amendments to the Common Position Type 3 Institutions: practical limits Time limits for negotiation No time limits In total: Time limit of 3 months (4 if extended) for negotiation and voting in committee and plenary after the the Council has forwarded its common position Maximum of 8 months from second reading. The Committee has to be convened within 6 to 8 weeks of receipt of Council second reading Thereafter, the committee has 6 to 8 weeks to reach agreement Information requirements No formal requirements No formal requirements Oral and written information requirements Mandate system No No Yes Type 4 Institutions: compromise among negotiators Majority requirement Simple majority in committee Simple majority in committee Absolute majority in Parliament's delegation Type 5 Institutions: Subsequent adoption Amendment rules Open rule: All amendments accepted Restricted rule: Only certain types of amendments accepted (see above) Closed rule: No amendments accepted, the joint text adopted in conciliation is voted on as a whole Majority plenary requirement Simple majority Absolute majority (ie. half members of Parliament plus one) Simple majority Time limits for adoption No time limits In total: Time limit of 3 months (4 if extended) for negotiation and voting in committee and plenary after the Council has forwarded its common position Time limit of 6 weeks (8 if extended) after the date of approval of joint text by conciliation committee 3.1 Type 1 Institutions: Key negotiators Institutions specifying who the key negotiators are at the different legislative stages can indirectly have an important impact on which compromises get agreed. As stated by Tsebelis and Money about the conciliation phase, "If the committee members faithfully reflect the preferences of the parent chambers, even unconstrained conference committees will produce legislation similar to outcomes achieved in their absence" (1997: 112). The same holds for negotiators earlier in the legislative process, who can be more or less representative of the opinion of the Parliament as a whole. Of key importance here is the rapporteur, i.e. a member from the responsible committee, who takes responsibility for negotiations both inside the Parliament and between the Parliament and the Council . He/she is in charge of drafting the Parliament's report at the different stages of the procedure. Usually this MEP has a particular knowledge of the issue area in question and information about the state of play in the current negotiations, which gives him/her a unique possibility to exert legislative influence. No rules exist to ensure that the rapporteurs appointed are representative of the opinion of the Parliament as a whole, their standing committee and/or the conciliation delegation. Their selection is the result of a negotiating process inside the political groups and between the political group co-ordinators within the responsible standing committee. The Parliament's Rules of Procedure officially gives the rapporteur and the relevant committee chair the responsibility to monitor the "progress of the proposal" in the Council at first reading (2004d: Rule 54) and the possibility "to discuss amendments that have been tabled in committee with the President of the Council or his representative" before the Parliament's second reading vote (2004d: Rule 59(5)). Nevertheless, the 2000 report from the Conciliation Vice-Presidents noted that, "it has not always been clear who was actually and formally authorised to conduct the negotiations" at early stages of the procedure (Parliament 2000a: 10). Different studies have made clear that the Council is particularly interested in informal contacts and negotiations with rapporteurs and shadow rapporteurs from big political groups to make sure that deals are backed by a large range of opinion in the Parliament and are therefore more likely to be acceptable (Farrell and Heritier 2003b: 592; Raunio and Shackleton 2003: 178). At the conciliation phase the Rules of Procedure give the most important formal role to the three Vice- Presidents responsible for conciliation. One of the three Vice-Presidents chairs each of the delegations that are established to negotiate with the Council. His or her role in the discussions is likely to be somewhat different from the other members of the delegation. The rapporteur and committee chairman originate from the responsible standing committee and may have views on the issue area in question, which deviate from the chamber as a whole (for example, a member of the environment committee may be more pro the environment than the average member of the Parliament). In contrast, the Vice-Presidents can be expected to be more inclined to defend the overall opinion of the Parliament and to be less likely to share issue- specific loyalties. In 1999, they got included in the negotiating teams of the informal meetings (the so- called trialogues), which have become the centre of the discussions with the Council, partly in order to be able to fully monitor negotiations (Parliament 2000a: 13; Neuhold 2001: 15). The conciliation delegation is appointed according to fixed quotas allocated to the political groups of the Parliament and is likely to reflect the opinion of the standing committees where in practice most of its members originate from. Thus, both the standing committees and the conciliation delegations may have loyalties to a certain subject area that diverge somewhat from the opinion of the Parliament as a whole. However, just as in the trialogue negotiating teams, here too the responsible Vice-President can play an important role in defending the prerogatives of the Parliament position as a whole. Thus the Vice- President responsible for the Port Services Directive voted in the conciliation delegation in favour of the proposed agreement reached in conciliation with the Council, contrary to the position of the other members of his political group, because he took an institutional point of view, considering that the final decision should be reserved for the plenary. Hence, even though there are clear overlaps between the informal key negotiators and the membership of the parliamentary body in charge in conciliation and earlier in the procedure, both the conciliation negotiators and the delegation can generally be expected to be more likely to reflect the general interests of the Parliament. This tendency is reinforced by a difference in the level of sector specialisation at staff level in the Secretariat at the different legislative stages. Thus, whereas the staff of the different standing committees is in charge during first and second reading, the conciliations secretariat takes over in the third reading. This staff does not have a special knowledge of the detailed issues under discussion, but is a general service with an overall interest in ensuring that the conciliation system functions in such a way as to secure the best possible outcome for Parliament in the negotiating process. 3.2 Type 2 Institutions: Substantive elements In addition to institutions specifying who the key negotiators are, we are also interested in institutions specifying what these negotiators can agree on in practice. A comparison shows that there are very few or no restrictions on the scope for negotiation at first reading but a range of effective constraints at second reading and in conciliation. In conciliation, in particular, there is a formal institutional constraint arising from the provisions of the Amsterdam Treaty. It specified that the work of the conciliation committee should be limited to addressing the common position on the basis of the Parliament's second reading amendments, meaning that other areas than those amended by the Parliament should not be debated nor should new elements be inserted (Article 251: 4; Raunio and Shackleton 2003: 178; Parliament, 1999: 4, 2003b: 5). This change was not made because the legislative bodies wanted to control their delegations, but because the Council wanted to prevent the Parliament from being able to expand the area of negotiation in conciliation (Garman and Hilditch 1998: 282; Jacobs 1997: 21). However, in practice it restricts the scope for action of key negotiators considerably. In the words of one respondent, "In conciliation everybody knows what the starting point is: the amendments of the Parliament and the common position of the Council, and it's written down … we know where everybody is coming from and we know where they end up, but at first reading, the ground is totally unclear and indeed the rapporteur is able to determine what the ground should be". It is not merely because no scope rules exist that the ground is unclear early on in the legislative process at first reading, but also because no formal position of the Parliament or the Council has been adopted, which can be the subject of negotiation. This means that all sorts of amendments can be tabled to the Commission's proposal by the key negotiators. As in the case of the Takeovers Directive, rejected after conciliation in 2001 but adopted in a revised form at first reading in 2004, the result can be something that the Commission is extremely reluctant to accept as it diverges so far from its own proposal. At second reading, the Institutions have established a position, which restricts the free hand of the negotiators considerably, but still not as much as in conciliation. First, it is still possible to table/re-table certain types of amendments, and second there is no formal system where the negotiators collect a mandate from the relevant standing committee compared to what they receive from the conciliation delegation. This difference is reflected in the reluctance of rapporteurs to envisage any change in the system at first and second reading. According to one respondent, They do not wish there to be greater institutionalization at the moment, because they know this is what gives them ultimate power … it seems to me that it is a highly dubious instrument". 3.3 Type 3 Institutions: Practical limits Not merely, scope limits, but also practical limits on the room of manoeuvre of negotiators merit attention. First, time limits on the negotiations are markedly different. At first reading there is no time limit, the main limitation being the level of enthusiasm of a particular Presidency to conclude a dossier during its six months and the readiness of Parliament to match that enthusiasm. At second reading, the time limit is 3 to 4 months, though in practice it is usually restricted to the much shorter period between the adoption of amendments in committee and their discussion in plenary. Conciliation offers a much longer period for negotiation than is available at second reading but one subject, at every phase, to Treaty defined limits (unlike in first reading). Thus the procedure normally begins during the period of three to four months reserved for the second reading of the Council and continues in the six to eight weeks after transmission of that second reading laid down for convening of the conciliation committee. Once convened the committee has another six to eight weeks to reach a compromise, thereby effectively giving negotiators a maximum of about eight months to come to a compromise. It is extremely difficult to assess in general terms whether the length of time available for negotiation as the procedure moves along has an impact on the extent to which negotiators can influence policy results. What probably matters more are institutions specifying whether key negotiators have to inform Parliament and whether they have to collect an explicit mandate before entering into agreement with the Council. Both types of institution exist in conciliation. Thus, if conciliation is needed, the Parliament's delegation holds its constituent meeting as soon as it is known that the Council will not be able to accept its second reading amendments. The purpose of the meeting is to establish the position of the Parliament and to give a mandate to the key negotiators (vice-president, rapporteur, committee chair) to start negotiations with the Council (Parliament 1999: 9, 2000b: 10, 2002b: 11, 2003b: 11). The delegation is informed about the results of the informal negotiations on a continual basis, discusses whether to approve the results and can give explicit mandates for the subsequent negotiations. Moreover, after each conciliation and trialogue meeting the delegation receives written information about the results (Parliament 2002b: 12-14, 2003b: 12-14) , and prior to each delegation meeting, members receive a note about the aim of the meeting and the status of the negotiations plus other relevant documentation, in particular a revised version of the joint four column working document, where the positions of the different EU organisations and possible compromises are presented (Parliament 1999: 14, 2003b: 16; 2001: 18-19). As stated in the 2003 Conciliation's handbook of the Parliament, "The aim of the delegation meeting is to adopt a strategy vis-à-vis the Council's position at every stage of the procedure as well as to discuss any compromise texts" (Parliament 2003b: 13). The conciliations unit of the Parliament facilitates the overall co-ordination and makes sure that procedural guidelines are respected. The latest conciliation report notes that, "the system of the negotiating team working on the basis of a mandate by the delegation has worked well" and that "much effort has been dedicated to ensuring that the delegation can maintain confidence in the negotiators" (Parliament 2004a: 26) . On the contrary it is less clear how negotiations are handled in practice at first and second reading where the link between the informal negotiators and the standing committees is much less formalised than between the informal negotiators and the conciliation delegation of the Parliament. As stated by a respondent, "There are several reasons why many MEPs do not like to go to conciliation because for them conciliation means a loss of power …. in conciliation we have rules, we have formal delegation, regular meetings, everyone is concerned, whereas in first and second reading there is an absence of rules". The Joint Declaration on practical arrangements for the co-decision procedure: allows the EU organisations to establish "appropriate contacts (…) with a view to (..) bringing the legislative procedure to a conclusion as quickly as possible" (Parliament, Council and Commission 1999), but does not go into detail about how this is to be done in practice. Thus, there have been very few fixed practices to ensure that key negotiators keep parliamentary committees informed and act on the basis of mandates at first and second reading (Parliament 2004b: 8; 2004a: 26). This is reflected in the following comments in the 2002 conciliation report from the Parliament, The basic problems is that the positions expressed to the Council during such "unlawful" contacts do not necessarily represent the points of view of the majority of the members of the committee concerned" (Parliament 2002a: 18). Also Farrell and Heritier's comment that even though there are "emerging patterns of interaction" in the Parliament over the years (Farrell and Heritier 2003a: 24), "Negotiations on early agreement dossiers are almost entirely informal – it is often extremely difficult for others within the Parliament, let alone outsiders, to have any idea of what exactly is going on" (2003a: 8). Some of these disadvantages have led to criticism within the Parliament and the annual reports from the Conciliation Vice-Presidents have included recommendations to regulate things further. (Parliament 2004b: 7; 2004a: 27-34; 2001: 22-24; 2003a: 12). The 2001 report states, for example: "Other informal contacts with the Council …should, however, be reported in committee, and should be the subject of reports, preceded by a clear mandate given by the committee" (Parliament 2001: 28). As we shall see below, some first steps have since been taken to ensure that key negotiators at first and second reading provide information to parliamentary committees and ensure their support when reaching deals with the Council. . However, it seems likely that the key negotiators of the Parliament will still enjoy a considerably greater room for manoeuvre in the two first readings than in conciliation. 3.4 Type 4 Institutions: Compromise among negotiators Institutions regulating how difficult it is to get compromises adopted within the relevant parliamentary body in charge can also affect the room for manoeuvre of the key negotiators. Thus, the easier it is for the key negotiators, the greater the chance that their deals move on to adoption in the plenary. Also this type of institution gives the key negotiators greater room for manoeuvre early on in the procedure. Thus, whereas a simple majority is sufficient to move the file out of the standing committee (Parliament 2004d: Rule 59(4)), i.e. a majority of those present, an absolute majority is needed within Parliament's conciliation delegation, i.e. a majority of all conciliation delegates no matter whether they are present or not. This requirement is now stated explicitly in the Parliament's Rules of Procedure (2004d: Rule 64(7)) but its establishment was agreed only after the rejection by the plenary of the Takeovers Directive, following concern about the level of attendance at delegation meetings. No such concerns have been expressed up to now about equivalent attendance in committee. 3.5 Type 5 Institutions: Subsequent adoption When a compromise is adopted in the relevant parliamentary body in charge, the next stage is to get it adopted in plenary. This means that institutions regulating how this takes place become important in determining the chances that negotiators get their compromises adopted. The smaller the possibility for the chamber to reject the legislative compromise, the greater the room for manoeuvre the key negotiators have. Different rules determine the possibility of adopting a compromise, such as time limits on the amount of time available for adoption, the possibility of tabling further amendments to the compromise and the majority needed in the plenary for adoption. As far as time limits are concerned, these are again most liberal at the beginning of the process. At first reading, there is no time limit, at second reading there is a 3 (4 if extended) month time limit for both negotiations and adoption in committee and plenary, and at third reading there are 6 (8 if extended) weeks after the date of approval of the joint text in the conciliation committee to get it adopted. As already indicated, the time available to verify an agreement is normally shortest at second reading, with the vote often taking place only a matter of days after the conclusion of negotiations. Necessarily this reduces the scope for challenges to an agreement, with other members obliged to accept on trust that they should vote in favour. By contrast, the time available in conciliation after an agreement is normally a number of weeks, enabling opposition to any agreement to mobilise (as happened with the Takeovers and Port Services Directives). The possibility of not of being able to table amendments also has an effect on the room for manoeuvre of negotiators. Here the formal rules give negotiators more room in conciliation than at first and second reading. Thus, whereas conciliation compromises are subject to a yes or no vote, texts coming out of parliamentary standing committees can be amended at first and second reading. As already discussed, having the possibility to present non-amendable compromises to legislative bodies is where conciliation committees derive their importance according to the literature. However, there is a difference between having a formal and a practical possibility to table amendments. Thus, even though there is a formal possibility to table amendments in plenary at both first and second reading, it is not necessarily a feasible strategy to do so in practice. Thus, one consequence of doing so for the Parliament is likely to be that the Council will not be able to accept all of Parliament's amendments thus pushing the procedure into the next reading. Therefore, there will often be a considerable pressure on members of the Parliament not to use their power to table amendments but to agree to the compromise which has been agreed with the Council at first or second reading. As explained by Raunio and Shackleton about a concrete case where negotiations were settled without going to conciliation, "Only at the end of the discussion was the proposal returned to committee with a package that was effectively presented on a take it or leave it basis. Any successful amendment would have undermined the agreement with the Council" (2003: 179). Thus, even though the Treaty with respect to the possibility of tabling amendments clearly gives key negotiators greater formal room for manoeuvre in conciliation than at earlier stages, this may not really be so in practice where amendments can be ruled out for practical reasons. As far as the majority required for final adoption is concerned the key negotiators are most constrained at second reading, where an absolute majority is required whereas merely a simple majority is required at both first and third reading. Interviews reveal that key negotiators are very aware of this and plan their strategy accordingly, "They know if they can get certain amendments accepted in committee in heavy opposition to the plenary at first reading they can still go through on a simple majority so they know that at first reading a simple majority makes deals more easily acceptable than at second reading where you need an absolute majority" The fact that compromises are more easily adopted at first reading can increase the pressure on the plenary not to use its formal power and table amendment to the first reading compromise, which may not have sufficient support behind it to win an absolute majority. 3.6 Constraints and key negotiators in the three readings As the comparison of constraints has shown key negotiators are generally more constrained in conciliation than at both first and second readings, thereby enabling the Parliament as a whole to exercise more control over legislative outputs. The definition of who the key negotiators in conciliation are implicates in the process actors, in particular the Vice-Presidents responsible for conciliation, whose concerns extend beyond the particular interests of the committee or the rapporteur, who will have dominated the procedure up to that point. The scope of negotiations is, as a result of Treaty revision, strictly circumscribed to prevent the introduction of elements that were not subject to amendment at second reading, a limitation that does not apply at first or second reading. The time limits laid down for the approval of the results of conciliation enable the whole Parliament to become familiar with them in a way that is not possible where agreements are reached very shortly before a plenary vote. The conciliation mechanism imposes heavy information and mandate requirements which make it extremely difficult for negotiators to escape a broader control, whereas these requirements are much lighter and less formalised at committee level. The majority requirements in a conciliation delegation are more severe than in a committee where a simple rather than an absolute majority is enough to adopt a compromise. The fact that amendments cannot be tabled to the results of conciliation, which can be accepted by a simple majority, only gives negotiators in conciliation the formal advantage identified in the literature. In reality, practical concerns often severely constrain the possibility to table amendments at first and second readings. As shown in Figure one, the conciliation stage is the stage where the key negotiators are most subject to informal and formal constraints and where other actors within the Parliament have the most powers to influence the results giving the negotiators a smaller room for manoeuvre than at previous legislative stages. In this way, the EU is an interesting case, where the lack of informal procedures at first and second reading actually modifies the impact of formal rules and makes control of key negotiators by the chamber very difficult. Thus it does not help the Parliament much at earlier stages of the legislative process that it has the possibility to table amendments and to vote on them separately, when deals are often made with the Council that are very difficult to change in practice. It also shows that conciliation is not per se a risky strategy for a legislature. There is nothing inherently problematic in delegating responsibility to a subset of legislators in a conciliation delegation. One cannot argue that different kinds of reconciling differences between the legislative bodies have fixed consequences for the control of the chambers in general. It is all about how it is handled in practice. Thus, it is not enough to draw a conclusion about a chamber's control of legislative outcomes to know whether conciliation committees were used to reconcile differences with the other chamber. Instead, it is necessary to examine the formal and informal constraints in each political system for the different kinds of reconciliation throughout the legislative process and see what the net effect is for the control of the legislature as a whole. 4. Implications for the future What are the potential implications of these conclusions for the future of the relations between Parliament and Council and for the way that the Parliament organises its work, when both are confronted with the prospect of a major expansion of co-decision under the terms of the Constitution? In particular, how will the balance be struck between the greater efficiency implicit in the early conclusion of legislative procedures and the pressure for the Parliament as a whole to retain control over the activity of key negotiators in circumstances where the number, range and heterogeneity of negotiations can be expected to grow markedly? One conclusion can already be drawn and that is that the new Article III-396 of the Constitution itself does not does not alter the shape of the formal constraints under which the EU will operate when using the "ordinary legislative procedure". By contrast, there has been some movement towards altering the informal structure of constraints applicable to negotiations at first and second readings. Within the Parliament the desire to avoid the risk of the organisation as a whole losing control over the legislative process is reflected in a set of guidelines for best practice relating to agreements at first and second readings that were adopted by the Conference of Presidents, the leading decision-making body of the Parliament, in November 2004 (Parliament 2004b: 9). These guidelines address a number of the issues already discussed in this paper, in particular the question of a mandate for negotiators, the provision of information to the committee as a whole and the treatment of any agreement reached with the Council. As far as the mandate is concerned, the guidelines specify that "concrete negotiations should not usually take place until the committee has adopted its first or second reading amendments. This position can then provide the mandate on the basis of which the committee's representatives can negotiate with Council and Commission" (emphasis added) (Parliament 2004b: 9). In other words, there is a presumption that committees will not enter into formal discussions with the other body until they have established their own position, thereby establishing more clearly the room for manoeuvre of the negotiators. The provision of information to all the political groups in the committee is also given a high priority, with the rapporteur expected to "report back regularly on the state of negotiations, if necessary to the whole committee. Any significant changes in the negotiating position should have broad political support" (Parliament 2004b: 9). Again the commitment to greater transparency puts a premium on ensuring that the influence of negotiators is mediated through the political structures of the committee. There is an attempt to avoid the situation where the Parliament is effectively obliged to say yes, even though it has the formal right to amend. Thus "any compromise amendments required as a result of the agreement reached should be the subject of written information to all committee members. If they cannot be approved by the committee for submission to plenary, they should be co-signed by the rapporteur and shadow rapporteurs or coordinators on behalf of their political groups to demonstrate that the amendments enjoy broad support" (Parliament 2004b: 9). Again the provisions are not binding but they establish a yardstick against which the behaviour of negotiators, above all, their ability to keep the confidence of their colleagues, both inside and outside the committee, can be judged. These issues of practice show that in the particular case of the European Parliament in the EU the concerns of loss of control of the whole house over members delegated to negotiate on its behalf is becoming ever more a question to be addressed before negotiations reach the conciliation phase. Hence the importance of developing an understanding of the mechanisms that allow such delegation and the conditions under which it can work successfully. The full impact of the new guidelines has yet to be seen. However, it is probably fair to say already that they are likely to reduce some of the outlined differences in the scope for action of parliamentary negotiators at the different stages of co-decision.. In many ways, it can be expected that negotiations at first and second reading will become more similar to the already established pattern in the conciliation phase. Bibliography ? Council, Parliament and Commission (1999) Joint Declaration on practical arrangements for the new codecision procedure (Article 251 of the EC Treaty), OJ C 148, 28.05.99, p. 1. ? Farrell, Henry and Adrienne Heritier (2003a) "The Invisible Transformation of Codecision: Problems of Democratic Legitimacy", report 2003: 7, Swedish Institute for European Policy Studies. ? Farrell, Henry and Adrienne Heritier (2003b) "Formal and Informal Institutions Under Codecision: Continuous Constitution-Building in Europe", Governance, 16(4): 577-600. ? Farrell, Henry and Adrienne Heritier (2004) "Interorganizational Negotiation and Intraorganizational Power in Shared Decision Making. Early Agreements Under Codecision and Their Impact on the European Parliament and the Council", Comparative Political Studies, 37(10): 1184-1212. ? Garman, Julie and Louise Hilditch (1998) Behind the scenes: an examination of the importance of informal processes at work in conciliation, Journal of European Public Policy, 5(2): 271-84. ? Jacobs, Francis B. (1997) "Legislative Co-decision: A real step forward", Paper presented at the ECSA conference. ? Knight, Jack (1992) Institutions and Social Conflict, Cambridge: Cambridge University Press. ? Luce, Robert (1922) Legislative Procedure. Parliamentary Practices and the Course of Business in the framing of Statutes, Boston: Houghton Mifflin Company. ? Mastias, Jean and Jean Grangé (1987) Les secondes chambres du Parlement en Europe Occidentale, Paris : Economica. ? McConachie, Lauros Grant (1898) Congressional Committees : a study of the origins and developments of our national and local legislative methods, Boston : Thomas Y. Crowelll and Company. ? McCown, Ada (1927) The Congressional Conference Committee, New York : Columbia University Press. ? Neuhold, Christine (2001) "The 'Legislative Backbone' keeping the Institution upright? The Role of European Parliament Committees in the EU Policy-Making Process", European Integration Online Papers, 5 (10), http://eiop.or.at/eiop/texte/2001-010a.htm ? North, Douglass C. (1990) Institutions, Institutional Change and Economic Performance, Cambridge : Cambridge University Press. ? Oleszek, Walter J. (2001) Congressional Procedures and the Policy Process, Washington DC : Congressional Quarterly. ? Opfermann, Wilhelm (1995) "Das europäische Vermittlungsausschuss" in Rolf Grawert, Bernhard Schlink, Rainer Wahl, and Joachim Wieland Offene Staatlichkeit. Festschrift für Ernst-Wolfgang Böckenförde zum 65. Geburtstag, Berlin: Duncker und Humbolt. ? Parliament (1999) Conciliations Handbook, 1st edition. ? Parliament (2000a) Delegations to the Conciliation Committee: Activity Report 1 May 1999 to 31 July 2000. ? Parliament (2000b) Conciliations Handbook, 2nd edition. ? Parliament (2001) Delegations to the Conciliation Committee: Activity Report 1 August 2000 to 31 July 2001. ? Parliament (2002a) Delegations to the Conciliation Committee: Activity Report 1 August 2001 to 31 July 2002. ? Parliament (2002b) Conciliations Handbook, 4th edition. ? Parliament (2003a) Delegations to the Conciliation Committee: Activity Report 1 August 2002 to 31 July 2003. ? Parliament (2003b) Conciliations Handbook, 5th edition. ? Parliament (2004a) Delegations to the Conciliation Committee: Activity Report 1 May 1999 to 30 April 2004. ? Parliament (2004b) Conciliations and Codecision. A Guide to how Parliament co-legislates, November. ? Parliament (2004c) Codecision News. Newsletter from the European Parliament Conciliations and Co- decision Secretariat, October/November. ? Parliament (2004d) Rules of Procedure. 16th edition. ? Rasmussen, Anne (2005) "Conference committees in the United States and the European Union. Multiple principals and (dis)loyal delegates", paper to be presented at the Ninth Biennial International Conference of the European Union Studies Association, Austin, Texas, March 31-April 2, 2005. ? Raunio, Tapio and Shackleton, Michael (2003) "Co-decision since Amsterdam: a laboratory for institutional innovation and change", Journal of European Public Policy, 10(2):171-188. ? Rogers, Lindsay (1922) "Conference Committee Legislation", The North American Review¸215(3): 300- 307. ? Shepsle and Weingast (1987) "Why are Congressional Committees powerful?", American Political Science Review, 81(3): 929-45. ? Tiefer, Charles (1989) Congressional Practice and Procedure. A Reference, Research and Legislative Guide, New York: Greenwood Press. ? Tsebelis, George and Jeanette Money (1997) Bicameralism, Cambridge: Cambridge University Press. ? Vogel, Friedrich (1989) "Der Vermittlungsausschuss" in Vierzig Janre Bundesrat, Tagungsband zum wissenschaftlichen Symposium in the Evangelischen Akademie Tutzing vom 11 bis zum 14 April 1989, Baden-Baden: Nomos Verlagsgesellshaft. ? Vogler, David J. (1971) The Third House: Conference committees in the United States Congress, Evanston: Northwestern University Press. * The views expressed in this paper are strictly personal and do not necessarily reflect the position of the European Parliament. Most notably in the areas of taxation and own resources, where Council will also decide by unanimity. Parliament only rejected the outcome of two conciliation negotiations (out of 86) between 1999 and 2004 – the Takeovers Directive in 2001 and the Port Services Directive in 2003. The quote is from Marc Lacey (1996) "Senate Panel Opts to Split Bill on Immigration", Los Angeles Times, March 15, 1996. For similar statements, see also Farrell and Heritier 2004. See Parliament's Rules of Procedure, Rule 42 and 48. The 1999 and 2000 conciliation guides do not include exactly this wording. Instead, they state that "The results of the trialogues are discussed and possibly approved at the meetings of the respective delegations.." (1999: 10, 2000b: 11). See also Parliament, 2000a, p. 13-14. The importance of having such a majority was clarified following a note from the Parliament's Legal Service (Parliament, 2001: 15). / 23