http://eiop.or.at/eiop/texte/2003-010a.htm
: 24.10.2003| This paper's comments page | Send your comment! to this paper |
| The Role of the European Commission in Co-decision – A strategic facilitator operating in a situation of structural disadvantage(*) |
| Anne Rasmussen |
| European Integration online Papers (EIoP) Vol. 7
(2003) N° 10; http://eiop.or.at/eiop/texte/2003-010a.htm |
| Date of Publication in
: 24.10.2003 |
| | Abstract |
Back to homepage |
PDF | | This paper's comments page | Send your comment! to this paper | |
1
The co-decision procedure introduced by the Treaty of Maastricht in 1993 marked a fundamental change in the legislative balance between the European institutions. Most noticeably, it increased the involvement of the European Parliament (Parliament) in EU legislation, but also had the consequence that the Commission was put in a situation of structural disadvantage compared to previous legislative procedures. Hence, the Treaty, which sets out the structure of the day-to-day decision-making, gives the Commission less formal power to influence the position of the final decision-makers in this procedure, i.e. the Parliament and the Council of Ministers (Council). This is very similar to the conclusion drawn by several formal, rational choice institutionalists. As an example, Garrett finds that, "(t)he result of this institutional innovation (the conciliation committee) is that the Commission's preferences need not be taken into account because it is structurally unable to affect the decisions of the Parliament and the Council" and "(f)inally, under co-decision the Commission is effectively taken out of the game before the real bargaining over policy begins" (1995: 303 and 305). Moreover Crombez states that the Amsterdam version of the co-decision procedure, "renders the Commission irrelevant" (2001: 101).
This may partly explain why, with the exception of one unpublished study
(Rasmussen, 2001), the existing literature has dealt with the role of the
Commission relatively superficially and/or has tended to focus on the
relationship between the Parliament and the Council(1). In spite of this, the aim of this paper is to look closer at
the role of the Commission and examine the ways in which the Commission
can continue to affect legislation in co-decision despite being put in a
situation of structural disadvantage by the Treaty. We agree with the formal,
rational choice institutionalists that the Commission’s room for manoeuvre
is clearly reduced in co-decision, but we argue here that its relative
loss of power over time with the introduction of the procedure should not blur
the picture that in absolute terms it is still an important actor in the
EU policy process.
Hence, even though the models of these formal theorists
have provided valuable insights by showing how changes in the institutional
conditions – i.e. the formal day-to-day decision-making procedures –
have had important consequences for the role played by the different EU bodies
in politics over time, they are merely models. Thus, they are
simplifications of the EU policy process building on certain assumptions and
excluding other elements of the decision-making process, which can be more or
less justified (Rasmussen, 2001).
2
As recommended by the formal theorists themselves the insights developed by these models should therefore not stand alone, but should rather be seen as a starting point – or a tool to make hypotheses – for further empirical testing (Steunenberg, 2000: 369-70)(2). Moreover, even though we agree with the formal theorists that the Treaty rules should be the starting point for understanding how actors behave, we are more inspired by what we here call softer, rational institutionalists who also pay attention to how these rules are applied, interpreted and moderated through the actual behaviour of actors with different capabilities in the decision-making process (Knight, 1992; North, 1990). Instead of making an institutional analysis similar to the models of the formal theorists of whether the Commission is in a more or less favourable position in the Treaty procedure, we therefore make an empirical analysis of the extent to which it actually manages to leave its mark on the final legislative results by using both formal and informal means.(2a)
The paper is structured as follows: Firstly, we present our approach and introduce the co-decision procedure briefly. Secondly, we set up a more specific framework for measuring the legislative influence of the Commission. Thirdly, we perform the empirical analysis by looking at two parameters, i.e. A) The chances of success in co-decision without the Commission and B) The Commission’s impact on the tabling and adoption of amendments and compromise texts throughout the procedure. Finally, we conclude and discuss the ability of different theoretical frameworks to understand the role of the Commission in co-decision.
The empirical analysis of the Commission’s legislative influence covers the period from Maastricht to one year after the entering into force of the Amsterdam Treaty. The Commission is regarded as having legislative influence when it successfully manages to persuade the Council and the Parliament to adopt elements in the final legislative texts that are different from what they would have been without its participation. This does not necessarily imply that the Commission has to persuade the Council and the Parliament to change their preferences but that it must persuade them to select another final compromise within the range of their preferences than they would have done without its involvement. The analysis demonstrates that the Commission often has influence on the adopted amendments and compromises throughout the policy process through its ability to act as a strategic facilitator.
We use different methodologies and data sources to generate these findings and increase the validity of our conclusions. Generally, we both approach our data from a qualitative and quantitative angle, since we combine statistical analysis with detailed text studies and evidence from 28 interviews carried out among representatives of the formal EU bodies in Brussels in June 2000(3).
More specifically, we firstly use micro level evidence from three procedures of the Community action programme in the field of education: Socrates discussed in co-decision and completed in 1995, 1998, and 2000(4). Hence, we stress the need to make a detailed examination of the interactions of the EU bodies in different phases of specific cases to fully understand the dynamics of co-decision. The Socrates procedures are interesting because they reflect a case, which has been dealt with three times in the history of co-decision enabling us to make more controlled comparisons of the role of the Commission over time. Moreover, the Socrates cases cannot be regarded as a particularly easy for the Commission to influence compared to more technical policy areas where its high level of expertise in relation to the Parliament and the Council is typically a greater strategic advantage.
However, no matter how carefully selected case studies may seem, it is always questionable to which extent findings from these can be generalized, and it is usually recommendable to combine results of these with other types of data. Secondly, we have therefore also examined general, official documents of co-decision and interviewed practitioners who have been involved in the overall co-ordination of co-decision from the three EU bodies and/or who have been involved in other specific co-decision files. Thirdly, we draw on macro level evidence from a data set for some of the first testing of the formal models produced by Tsebelis and his staff of the co-decision proposals concluded by 1998, to which the Parliament has made amendments in either the first and/or second reading(5). Because our explicit focus is on the Commission, we restrict ourselves to the 34 of the 79 files in the data set here, which have gone into conciliation(6). Hence, these files should be most difficult for the Commission to influence, since it no longer possesses any formal powers at this last stage of the procedure.
3
The initial version of co-decision set out in the Treaty of Maastricht was a highly complicated procedure compared to previous legislative procedures. It could include as much as four readings in the Parliament and two rounds of conciliation between the Parliament and the Council and it was originally applied for 15 Treaty articles. The procedure was simplified with the Amsterdam Treaty and extended to cover as many as 38 areas of legislation spread over 31 Treaty articles meaning that co-decision today is the “normal legislative procedure” for the Parliament covering more than half of Community legislation (Corbett et al. 2000: 191; Commission 1999a: 2-3).
The formal procedure begins when the Commission submits a proposal to the Parliament and the Council, who begin considering the proposal in their committees and working groups respectively. Upon recommendation of the responsible committee, the plenary of the Parliament adopts its first reading opinion typically in the form of a series of amendments to the Commission proposal. If the Council agrees with the amendments of the Parliament, it has become possible for it with the Amsterdam Treaty to adopt the proposal and the procedure stops (the so-called fast track procedure). However, usually the procedure moves on just as it always did in the Maastricht version of the procedure meaning that the Commission produces a modified proposal typically incorporating some of the amendments presented by Parliament. Thereafter, the Council adopts its so-called common position, which is the legislative basis for the rest of the procedure. After the Commission has commented, another reading starts in the Parliament and usually ends with the adoption of a number of second reading amendments to the common position. The Council can then decide either to adopt the amendments of the Parliament or to engage in conciliation with the Parliament to find a final compromise where the Commission acts as a mediator.
At both first and second reading, the Council can include amendments of the Parliament by qualified majority, if these are supported by the Commission, but needs unanimity if not. Hence the Commission exerts something, which in Rasmussen (2001) we called a formal gate-keeping power in the first and second reading(7). It has, however, become more difficult for the Commission to use its gate-keeping power at the second reading in co-decision than it was in the co-operation procedure, where it produced a modified proposal at second reading instead of “just” delivering its comments to the parliamentary amendments. This means that whereas the Council needed unanimity to remove or modify amendments in the co-operation procedure, it now actively needs to find a qualified majority to support amendments supported by the Commission in co-decision. Moreover, if the procedure goes into conciliation, the Commission’s formal role is much weaker and it has no gate-keeping power since the Parliament and the Council can agree on a compromise to be sent to the final vote in the Council and the plenary of the Parliament irrespective of what the Commission thinks. Finally, the Commission’s traditional right to withdraw legislation at any point in the legislative procedure is not valid in this conciliation phase (Commission, 1999b: 14).
In order to find out exactly which role the Commission plays in practice, we lay down a framework for assessing the legislative influence of the Commission. The measuring of legislative influence has been heavily debated in the history of social science (Hall, 1992: 205-07). Two of the most commonly used ways of measuring legislative influence include the use of either reputational or behavioural indicators. The former refer to interviewing a range of participants in the given policy area about the influence of certain actors, whereas the latter refer to using quantitative data of the actor’s behaviour to measure their influence on a proposal, for example by calculating the percentage of successful amendments (Hall, 1992: 207-08).
Reputational indicators make it possible to collect information about the influence of actors, which is often unrecorded, but typically do not capture the possible difference that may exist between an individual’s reputation of influence and that person’s actual influence. Moreover, they often produce general accounts of an actor’s influence and fail to distinguish between variation among an actor’s influence on individual files. On the other hand, the behavioural indicators do not enable a separation of the influence of an actor’s preferences from that person’s agreement with the preferences of others. In other words, it is questionable whether an individual can be said to exert influence when what he or she really wants is equal to the wishes of the other actors. Moreover, behavioural indicators cannot capture the effect of “implicit influence”, which means that an actor adjusts his/her behaviour to conform to the preferences of the other actors. This makes it impossible to judge whether the apparently successful behaviour of an actor is the result of that actor’s true preferences or merely an anticipation of what was expected of that person by other actors (Hall, 1992: 207-08).
4
Being aware of the difficulty of measuring legislative influence, we here apply a mix of both behavioural and reputational indicators, as the weakness of some will be largely compensated for by the inclusion of others. Firstly, we use an essentially “reputational” indicator to evaluate the Commission’s legislative influence by asking the respondents what the chances of success were and what the final legislative text would have looked like without the participation of the Commission in the policy process. At the outset, this indicator might seem speculative, since the respondents are asked to make counterfactual propositions by judging something, which essentially has no objective answer. However, the question is important since it encourages the respondents to think back on the role played by the Commission throughout the procedure and judge its importance. In other words, we indirectly ask the respondents to estimate whether there is a relationship between the behaviour of the Commission throughout the procedure and the final legislative outcome.
Secondly, we look more specifically at the different ways in which the Commission exerts legislative influence. Since the Commission is not one of the final decision-makers, we argue that it exerts influence when it succeeds in persuading the Council and the Parliament to take its viewpoints into account and adopt amendments and compromise texts different from what they would otherwise have done. Hence, we question the respondents about the extent to which the Commission has affected the drafting of compromise texts and parliamentary amendments, which were successful in the end. Moreover, we analyse quantitatively the extent to which the Commission’s opinion on an amendment has an effect on its adoption in the final text.
Having set out our framework for measuring the legislative influence of the Commission, we now turn to a presentation of the results of our quantitative and qualitative examination.
Closely linked to our definition of influence, we start by addressing the hypothetical question what the chances of success and the legislative content of the proposals would be without the presence of the Commission in the policy processes after it has presented its original proposal. In fact, most of the respondents from Council, Parliament, and Commission do not exclude the possibility that the files, in which they have been involved, could have been concluded without the Commission.
However, the respondents generally agree that often the compromises would have been harder to reach. Hence, an official of the Commission states that, “Sometimes the positions of the Council and the Parliament are so far apart that only the Commission can find common ground”. This is supported in the following comment from a member of Parliament (MEP) who notes that conciliation without the Commission, would be more a war of attrition”, and by a Council representatives who notes that, “I also think that it would be more difficult to reach a compromise, because the Commission, when it acts in the right way, is relied upon by other parties”. This was for example the case in Socrates II where the Commission’s involvement in the negotiations impacted on the willingness of the Council and the Parliament to come to a final agreement. A Council official explains how “on a request from the Council and the Parliament, the Commission presented a document about the budgetary needs…with this document it was considerably easier to agree on a budgetary increase on the Council side, and for the EP it became easier to relax its demands. Therefore, the Commission played an extremely important role”. Other respondents note that the deadlines in the co-decision procedure could probably not have been kept without the involvement of the Commission.
Moreover, a number of the respondents think that the compromises in co-decision would have been different, had the Commission not been involved. In this way, it is recognised that an important side effect of the Commission’s task to reconcile the positions of the other parties is that it is given the opportunity to influence the substance of the final text. As an example, a Commission official comments on the procedures for selecting projects in Socrates II and notes that:
“I think that there, the compromise found between the two institutions would have been different, had the Commission not been there..they would have taken the Council’s position more into consideration, which would have made the selection procedures more complex, longer, and less user-friendly”.
5
It is, however, important to point out that the importance of the Commission to the successful completion of the files and its ability to influence the contents of the files seem to have declined over the years. Hence, the development of informal practices and interactions between the key players throughout the procedure has not always been beneficial to the Commission, who used to have a privileged position since the Parliament and the Council had very few direct contacts and were somewhat sceptical to engage in direct dialogue between each other in the beginning. However, over the years the relationship between the Council and the Parliament has loosened up, and increased informal contacts between them have developed. As noted by an EP official, I think clearly one of the impacts of co-decision has been to throw the Parliament and the Council together on a much wider scale that never happened in the past”. This is underlined by a Commission official who was involved in the negotiations of all Socrates files, “At that first stage (Socrates I) it went really through us and I would say uniquely through us… Afterwards (Socrates II) when they found each other and realised that they could do things directly between them, they did not see the Commission as so strong”. Therefore, there are occasions where the Parliament and Council have got together to negotiate files informally without the presence of the Commission(8). Generally speaking it is therefore likely that there will be variation in the influence of the Commission between individual cases, and the Commission’s advantage of acting as “the messenger” and link between the Parliament and the Council has come under increased pressure over the years.
However, in the period studied, it was the norm that the Commission participated in the informal negotiations about the legislative files between Council and Parliament, which has provided it the possibility of influencing the files. Moreover, the Commission has the advantage of being the single body which participates in both the internal deliberations of the Council and the Parliament, which gives it a unique information advantage in the legislative process and a possibility to promote its viewpoints to the two other bodies separately.
After having examined what the situation might have looked like without the participation of the Commission, this section considers the influence of the Commission in more detail by looking at some of the means by which the Commission can influence the final text. Thus, the Commission’s impact on the tabling and adoption of amendments and compromise texts is examined.
As mentioned, the Treaty gives the Parliament a possibility to table amendments to the Commission’s proposal in its first reading and to the Council’s common position in its second reading. Despite the closer relationship between the Council and Parliament over the years, the respondents confirmed that often the Commission actively assists the Parliament in tabling these amendments. In the first reading, this can be an important way for the Commission to ensure that the substance of its initial proposal is not changed fundamentally. Moreover, Commission officials can sometimes have an interest in encouraging the Parliament to present amendments, which would be politically sensitive for the Commission itself to include in its initial proposal. One MEP notes: “..the Commission expects that the Parliament will help it to get more (than the original proposal). I don’t think that it is completely satisfied with its original proposal, which is a bit on the low side. That’s why the Commission expects the Parliament to intervene..”. This is confirmed by a Commission who notes, “Often we can use the parliamentarians to pass certain messages, because they have a right to say what they like..there is a very good co-operation”. Furthermore, several respondents explained how, in the second reading, the Parliament can also be a valuable partner for the Commission by tabling amendments, which reintroduce elements from the Commission’s original proposal, not included in the Council’s common position. As a parliamentary official noted, “Of course, the Commission unofficially proposes or draws the attention of the rapporteur (the responsible MEP for the file) to things, which are not in anymore”.
It should, however, be noted that even though the Commission is superior in terms of expertise and specific knowledge of the proposal, the MEPs are not merely the Commission’s tools. Hence, of course the Parliament also drafts a lot of amendments against the wishes of the Commission, which is shown by the fact that the Commission accepted or largely accepted 43 and 53 per cent of the Parliament’s first reading and second reading amendments respectively in the 34 analysed co-decision files. Thus, the MEPs do not automatically transfer the Commission’s wishes into amendments, but can often be very receptive towards its wishes. In addition, the MEPs will typically be more open towards the Commission when the two bodies share common concerns, which makes it difficult to evaluate the degree of Commission influence on the basis of this measure alone.
6
This general picture of the Commission’s impact on the tabling of reports and amendments is confirmed when we look specifically at the Socrates files. Here, respondents have explained how, in all the three files, the Commission officials were assisting the MEPs in drafting their reports and amendments. In Socrates II for example, a Commission official notes how, “the relationship between xx (the parliamentary rapporteur) and the DG (Directorate General of the Commission) was just optimal. It could not have been better, which meant the Parliament did very much rely on the expertise of the Commission in drafting the amendments”.
In all the three Socrates files, the Parliament introduced second reading amendments to the budget, which were in line with or even more favourable than the Commission’s original proposal. Hence, the Parliament deleted the Council’s introduction of a figure into the legislative act in Socrates I arguing in line with the Commission that the budget should be decided through the annual budgetary procedure, whereas in both the mid-term review and in Socrates II it proposed to increase the budget of the common position. It is likely that the Commission has encouraged the Parliament to table these amendments. As noted by a Council official, “If the Parliament offers a lot of money, the Commission says that it is too much, but it is clear that the Commission appreciates this very much”. It should, however, be noted that the opinions of the Parliament and the Commission about the budget do not differ fundamentally, wherefore this does not necessarily signify that the Commission has exerted considerable legislative influence.
Nevertheless, we have also found evidence that other issues of less concern to the Parliament were introduced by it in amendments. A Commission official explains how in Socrates II, “The Parliament certainly asked us to state the key issues that we would see as worthy to reintroduce into the debate. If the Parliament had not asked us we would have told them, so it does not really matter who picked up the phone first”. As an example, Parliament reintroduced the more federalist “European education area” from the proposal as the goal of the programme instead of the Council’s proposal for a “European area of co-operation in the field of education and training”. Several officials supported the view that this amendment was a clear priority by the Commission and not so much of the Parliament. One of the Council officials even noted that the “European area of co-operation” could be regarded as a “hobbyhorse” of the Commission. Moreover, the Parliament fought against the change of the selection procedures and proposed to delete the Commission’s obligation to consult the member states in cases of disagreement between it and the national agencies on a project. Again, it is likely that the Commission had an impact on the tabling of these amendments, which is reflected in the following comment by a Commission official, “Whether the Parliament would have been quite so keen on that one without us having suggested it, I am not sure, but certainly as soon as we had explained it, they were very keen”.
On the whole, the Commission therefore does seem to have a certain impact on the behaviour of the Parliament when it tables amendments. However, this is not enough to claim that the Commission has exerted legislative influence as defined here. Hence, the Commission’s effect on the behaviour of the Parliament must also result in changes in the substance of the actual proposals. In other words, are the amendments inspired by the Commission adopted in the end?
At a general level, it is of course very difficult to test whether the amendments inspired by the Commission are adopted. A calculation on the basis of Tsebelis’ data shows that 25 and 42 per cent of the Parliament’s first and second reading amendments respectively were accepted or largely accepted by the Council in the 34 files(9). Nevertheless, we can only guess, how many of these were influenced by the Commission. From examining the key amendments of the Socrates files, it looks as if the answer is mixed.
If one looks at the issue of the budget, the Commission succeeded in getting a higher figure than the Council declared itself willing to accept from the very beginning in all the files, even if it did not get exactly what it proposed initially. It is clear that this was to a large extent a result of the fact that the Commission and the Parliament had convergent preferences in this respect. At the same time, it was stated in section 5.1 that the presence of the Commission had a large impact on the Council and Parliament’s willingness to agree on a compromise on this issue. Instead, the amendments in Socrates I aiming at removing the budgetary figure and at changing the comitology procedure into an advisory committee were not successful from the point of view of the Commission. Thus, Socrates I did not get a purely advisory committee but a combined management and advisory committee as proposed by the Council and an explicit budget was set in the final legislative act(10). The other Commission inspired amendments in Socrates II also had a mixed history. For instance, the amendment on the overall goal of the programme was given up by the Parliament at the conciliation stage, which declared itself willing to change the wording from a “European areas of co-operation” to a “European dimension in education and training”. On the other hand, the amendment regarding the selection procedures was successful in the sense that the extra consultation of the member states was made conditional upon a specific request from these and that it would have a maximal duration of two weeks.
7
Generally, the analysis shows that the Commission does indeed try to exert legislative influence by affecting the tabling of the parliamentary amendments. It has, however, also demonstrated that such a strategy does not necessarily lead to a successful adoption of the Commission inspired amendments in the end. Nevertheless, on the whole the Parliament has become an important ally to the Commission in a number of legislative areas with the introduction of co-decision. Thus, the Commission has often been able to use the Parliament body to push forward its concerns to a much greater extent than in other legislative procedures where the Parliament’s opinion is merely advisory or where the Parliament does not have a final veto. This is recognised in the following statement by a Commission official about Socrates I, “it was difficult to get the budget, but it was a great thing that Parliament had a role to play for the first time”.
Another way of measuring whether the Commission exerts legislative influence is to look at its impact on the overall adoption of the parliamentary amendments, which is done next. We have chosen to exclude first reading amendments to entries of the text, which have been changed again at second reading, since we merely look at the relationship between the Commission’s opinion on an amendment and its inclusion in the final text(11). Altogether, our analysis is therefore based on 1,155 first and second reading amendments from the 34 examined files. Table 2 presents a cross tabulation of the relationship between the Commission’s opinion on an amendment and its degree of adoption in the final text.
It shows that an amendment is adopted or largely adopted in the final text in 12 per cent of the cases, in which the Commissions rejects an amendment, whereas it is rejected in 72 per cent of the cases. At the other extreme, the Commission’s recommendation to accept an amendment has resulted in the amendment being either adopted or largely adopted in as much as 61 per cent of the cases. Hence, the data reveal a high degree of convergence between the opinion of the Commission on an amendment and an amendment’s chance of getting adopted in the final text. However, they also show that it is easier for the Commission to convince the Council to reject than to accept an amendment, even though it is influential in both cases.
As mentioned above, Tsebelis et al. (2001) and Kreppel (2002) have also made analyses of the data set themselves. These analyses cannot be compared accurately with the results here, since they also include files, which have not gone into conciliation. Moreover, both studies use regression analysis even though essentially differences in an amendment’s degree of adoption cannot be quantified precisely on an integral scale. Bearing this in mind, we have instead calculated the simpler ordinal correlation co-efficient tau b, which is 0.46. In other words, our analysis shows that the Commission’s opinion can be used in approximately half of the times in the 34 of the examined files to predict an amendment’s final degree of adoption. It is clear that such crude data do no take the importance of the amendments into account and that correlation does not necessarily mean causation. However, seen in combination with our qualitative examinations of the means by which the Commission participates in the policy process, we believe there is evidence to conclude that there is a fairly substantial relationship between the Commission’s involvement in the amending phase and the content of the final legislative results.
As discussed initially, the relationship between the Commission’s opinion on an amendment and its degree of final adoption can probably partially be explained by the fact that the Commission does not make its decisions on the amendments in a vacuum but tries to anticipate the amendments’ chances of success later on. That was confirmed by several Commission officials and is reflected in the following statement by one, “(t)o a certain extent, the Commission also in accepting amendments or refusing them firstly has a view to what its original position was, but secondly also has to think in terms of what the final structure is going to be”.
Another way to partially explain why the Council followed the Commission’s advice on the parliamentary amendments could be that the preferences of the two were convergent. However, we believe that the relationship between the Commission’s opinion on the amendments and their degree of adoption in the final text is so substantial that it cannot exclusively be seen as an example of anticipated reactions or coherent initial preferences. The respondents also confirmed that the Commission does indeed have an impact on which amendments are incorporated into the final text. Hence, a Commission official noted how, “(...c)ouncil does not even examine the amendments, which the Commission does not take up in its modified proposal. They even do not look at them”. Even though this may not always be the case, it is not very exceptional either. A Council official states, “(u)sually, we examine the amendments supported by the Commission, for example if there are 100 amendments and the 60 are supported by the Commission, it is useless to examine the others, but it happens sometimes that we do”(12).
8
The general relationship between the Commission’s opinion on an amendment and its chances of adoption is also confirmed if one looks at the Socrates files. Firstly, table 3 of Socrates I shows that in the vast majority of cases, in which the Commission rejects an amendment, the Council follows the same strategy. In addition, table 3 demonstrates that the Council adopted as much as 60 and 83 per cent of the first and second reading amendments respectively, which the Commission accepted.
Secondly, table 4 shows that none of the amendments, which the Commission did not support, were accepted in the first and the second reading in the mid-term budgetary revision of Socrates. Moreover, the Council accepted all the substantial amendments, which the Commission supported in the first reading.
Finally, table 5 of Socrates II also shows a clear relationship between the Commission’s opinion of an amendment and its final adoption. Hence, none of the amendments, which were rejected by the Commission, were included in the final text. Moreover, as much as 74 per cent and 77 per cent of the first and second reading amendments respectively, which the Commission supported, were incorporated into the final text.
Thus, there seems to be a fairly substantial relationship between the Commission’s opinion on an amendment and its degree of adoption in the final text at a general level as well as in the three specific Socrates files. Even though part of this correlation may again be a result of anticipated reactions and/or coherent initial preferences, the quantitative analysis supported by the qualitative interviews has left evidence that the Commission’s opinion on an amendment does in fact seem to influence the Council’s willingness to incorporate it in the final text. Hence, a number of final legislative texts might have looked different if Parliament and Council had reconciled them on their own without the involvement of the Commission.
Another way for the Commission to continue to exert influence on the substance of the final text is through the drafting of compromise texts. Even though the increased contacts between the Council and the Parliament over the years have put the Commission in a less beneficial position, a parliamentary official stated, “I don’t recall any conciliation where the Commission compromises were not a central part of the discussion”. The co-decision guide from the conciliation secretariat of the Parliament also states that, “the Commission is often invited to produce compromise texts to be discussed in the delegations or at the next trialogue meeting” (Parliament, 2000: 11)(13). This may explain why the Commission does not automatically try to avoid the conciliation phase. As stated by a Commission official, “...sometimes it is in the Commission’s interest to try and get closer to its proposal in a conciliation process. In fact, there are times when we say, well, it must go into conciliation because then we will get something out of it”.
Even though the majority of the respondents confirmed that the Commission is often the author of the compromise text, they also stressed that very often they are not presented as such. A Council official explains in detail how this often works, “(b)efore the trialogues or the COREPER meetings, we often have informal meetings between the presidency, the Commission and the Secretariat General of the Council, where we work on compromise texts. Sometimes, these texts are entirely written by the Commission. However, they are presented as proposals from the presidency”. The reasons for doing so are political. Hence, it is typically easier to reach agreement on presidency than Commission compromises in the Council. Generally speaking, a parliamentary official notes how, “(t)he Commission is actually doing quite a lot of stuff, but it is not necessarily being given credit for it. It has to live with that. It is being seen as useful”. The Socrates files confirm that the Commission is very active in the drafting of the compromise texts. Hence, an official states how in Socrates II, “the Commission always co-operated with the Austrian presidency in order to modify the text. The Commission was always invited to drafting meetings when new texts were presented...basically it was the Commission redrafting its own text and the presidency accepting the redrafting of the Commission’s texts”.
9
On the whole, the Commission does therefore exert influence on the final text through the drafting of compromise texts in a number of cases even though there are examples of cases where the actual compromises have been settled directly between the parliamentary rapporteur of the proposal and the presidency of the Council. Moreover, compromise texts written by the Commission do not necessarily get fully incorporated into the final text, but should rather be seen as a starting point for the discussion between the parties. Finally, it should be remembered the even though Commission unquestionably writes the compromise texts with its initial proposal in mind, it also has to consider what it will be able to get through in the end. This may for example explain why the Commission also drafts presidency compromise texts in cases where the two do not have very convergent preferences, for example in Socrates II.
The empirical analysis performed here has demonstrated that the Commission should not be ignored in co-decision despite being put in a situation of structural disadvantage by the Treaty. Formal, rational choice institutionalists, who argue that it is “irrelevant” and that its “preferences need not be taken into account” as stated in the introduction, are therefore not right. Instead, it has been fruitful to contrast their theoretical conclusions with empirical evidence both at the aggregate and individual level.
That being said, it is also clear that the Commission in no way controls the legislative process of co-decision. Thus, the analysis also showed that co-decision has in fact put the Commission in a situation of structural disadvantage compared to previous decision-making arrangements since the Treaty limits its room of manoeuvre considerably. This means that the Commission cannot be certain to persuade the Parliament and the Council to bring forward its viewpoints throughout the procedure or to adopt any type of proposal, which it puts forward. This is in accordance with the recent analysis on the data set by Tsebelis’ larger research group as well as Kreppel herself, which both show that the opinion of the Commission on amendment seems to matter less in the co-decision procedure than under the previous co-operation procedure (Tsebelis et al., 2001: 595-96; Kreppel, 2002: 805)(14). Moreover, it has been pointed out in our analysis that the Commission’s prominent position as the messenger and link between the Council and the Parliament has come under increasing pressure over the years as the two have started to co-operate closer.
However, the respondents generally believed that often the compromises would have been harder to reach and would have looked different without the Commission. More specifically, it was indicated that the Commission has exerted legislative influence by strategically influencing the tabling and adoption of amendments and compromise texts throughout the legislative process in such a way that the same results could not have been expected without its participation. In other words, it is underlined that in absolute terms the Commission is still an important actor in the daily integration of the EU policy process, despite its relative loss of power with the introduction of co-decision. Moreover, we have gone a step further into examining the dynamics of the daily politics of co-decision by moving beyond theoretical accounts of the actors’ scope of action to more specific consideration of the ways in which one of the actors involved, i.e. the Commission, can influence the final legislative result.
The results of this work stresses the need not just to base further research of co-decision on a theoretical understanding of the formal Treaty provisions, but to examine in more detail how these provisions are used, interpreted and moderated in practice. As noted, we believe therefore that so-called softer, rational institutionalist analyses can be a valuable complement to the formal studies discussed initially. Thus, even though both of them share the same assumptions about actor behaviour etc., their different analytical frameworks allow them to focus on different elements of the decision-making context.
Hence, soft, rational institutionalists typically pay attention to a wider range of institutions than the formal theorists by including the effects of informal institutions such as norms and standard operating procedures in their understanding of actor behaviour (Knight, 1992; North, 1990). In the case of co-decision, the actors are not merely influenced by the formal Treaty provisions in regard to co-decision, but also by less formalized institutions such as interinstitutional agreements between the decision-makers, informal practices etc. As an example, our analysis showed that it was typically the norm that the Commission participates in the formal and informal negotiations between the Council and the Parliament in the period studied, which has given it a possibility to affect these meetings. Moreover, it demonstrated how it is standard practice that the Commission as the only body participates in the internal proceedings of the Council and the Parliament, which often gives it an information advantage in the decision-making process. Thus, the informal arrangements may to a certain extent modify the effects of the formal Treaty provisions.
10
Moreover, contrary to the formal models of co-decision softer, rational institutionalists do not merely examine the actors as a product of their formal institutional powers, such as their voting power, but also take their possession of strategic resources into account (Knight, 1992: ch. 5). Thus, the institutions will not determine the behaviour of an actor but rather give it a certain room for manoeuvre within which it can use its resources strategically (Scharpf, 1997: 42; Knight, 1992: 58-59). Here, it was shown how often the Parliament has relied on the expertise of the Commission in tabling amendments and so have both the Council and the Parliament in the drafting of compromise texts later in the procedure. This gives the Commission a possibility to persuade the two to adopt a result different from what they would otherwise have done.
In conclusion, formalistic frameworks are less appropriate for making conclusions about the Commission in co-decision because the formal weakening of the Commission by the Treaty means that it has increasingly had to rely on informal sources of influence to strategically persuade the two final decision-makers to bring on board its concerns. Such informal ways of exerting influence are not captured by the existing formal models, which explains why their authors have reached different conclusions about the role of the Commission than the ones presented here.
Commission (1994) "Observation de la Commission sur la position commune du Conseil concernant la proposition de decision du Conseil arretant le programme d'action communautaire Socrates", SEC(1994)1217.
Commission (1999a) ”Communication on the implementation of the Treaty of Amsterdam. The co-decision procedure, article 251 CE”, SEC(99)1083.
Commission (1999b) ”Codecision procedure, article 251. Guide to internal rules for the co-decision procedure”, SEC(99)1684
Corbett, Richard (2000) “Academic Modelling of the Codecision Procedure: A Practitioner’s Puzzled Reaction”, European Union Politics, 1 (3): 373-81.
Corbett, Richard (2001) “A response to a reply to a reaction (I hope someone is still interested!)”, European Union Politics, 2 (3): 361-66.
Corbett, Richard, Francis Jacobs and Michael Shackleton (2000) The European Parliament, John Harper Publishing, 4th edition.
Crombez, Christophe (1996) "Legislative Procedures in the European Community", British Journal of Political Science, 26 (2): 199-228.
Crombez, Christophe (1997) "The Co-Decision Procedure in the European Union”, Legislative Studies Quarterly, 22 (1): 97-119.
Crombez, Christophe (2000) “Institutional Reform and Co-Decision in the European Union”, Constitutional Political Economy, 11 (1): 41-57.
Crombez, Christophe (2001) “The Treaty of Amsterdam and the Co-decision Procedure” in Gerald Schneider and Mark Aspinwall (eds.), The rules of integration. Institutionalist approaches to the study of Europe, Manchester University Press.
Earnshaw, David and David Judge (1996) ”From co-operation to co-decision. The European Parliament's path to legislative power” in Jeremy Richardson, European Union – power and policymaking, Routledge.
Farrell, Henry and Adrienne Héritier (2002a) “Formal and informal institutions under codecision: continuous constitution building in Europe”, European Integration online Papers (EIoP), 6 (3) http://eiop.or.at/eiop/texte/2002-003a.htm (05/09/03).
Farrell, Henry and Adrienne Héritier (2002b) “Interorganizational Cooperation and Intraorganizational Power: Early agreements under codecision and their impact on the Parliament and the Council”, Preprints from the Max-Planck project group “Recht der Gemeinschaftsgüter”, Bonn, 2002/14 http://www.mpp-rdg.mpg.de/pdf_dat/2002_14.pdf.
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.
Garrett, Geoffrey (1995) “From the Luxembourg Compromise to Codecision: Decision Making in the European Union”, Electoral Studies, 14 (3): 289-308.
Garrett, Geoffrey and George Tsebelis (1996) "An institutional critique of intergovernmentalism", International Organization, 50 (2): 269-99.
Hall, Richard L. (1992) ”Measuring Legislative Influence”, Legislative Studies Quarterly, 17 (2): 205-31.
Jacobs, Francis B. (1998) ”Legislative Co-decision: A real step forward”, Paper presented at the ECSA conference 1998.
Knight, Jack (1992) Institutions and Social Conflict, Cambridge University Press.
Kreppel, Amie (2002) “Moving Beyond Procedure. An Empirical Analysis of the European Parliament Legislative Influence”, Comparative Political Studies, 35 (7): 784-813.
Moser, Peter (1997) “The Benefits of the Conciliation Procedure for the European Parliament. Comment to George Tsebelis”, Aussenwirtschaft, 52 (special issue): 57-62.
North, Douglass C. (1990) Institutions, Institutional Change and Economic Performance, Cambridge University Press.
Parliament (2000) Conciliations Secretariat, Conciliations Handbook, 2nd edition July 2000, 420330EN.doc.
Rasmussen, Anne (2001) The Role of the Commission in Co-decision, unpublished dissertation, University of Copenhagen.
Scharpf, Fritz (1997) Games Real Actors Play – Actor-Centred Institutionalism in Policy Research, Westview Press.
Scully, Roger M. (1997a) “The European Parliament and the Co-Decision Procedure: A Reassessment”, Journal of Legislative Studies, 3 (3): 58-74.
Scully, Roger M. (1997b) “The European Parliament and Co-Decision: A Rejoinder to Tsebelis and Garrett”, Journal of Legislative Studies, 3 (3): 93-104.
Shackleton, Michael (2000) ”The Politics of Co-decision”, Journal of Common Market Studies, 38 (2): 325-42.
Steunenberg, Bernard (1994) "Decision Making Under Different Institutional Arrangements: Legislation by the European Community", Journal of Institutional and Theoretical Economics: 150 (1): 642-69.
Steunenberg, Bernard (2000) “Seeing What You Want to see: The Limits of Current Modelling on the European Union”, European Union Politics, 1 (3): 368-73.
Tsebelis, George (1997) “Maastricht and the Democratic Deficit”, Aussenwirtschaft, 52 (special issue): 29-56.
Tsebelis, George and Geoffrey Garrett (1997) “Agenda Setting, Vetoes and the European Union’s Co-Decision Procedure”, Journal of Legislative Studies, 3 (3): 74-93.
Tsebelis, George and Geoffrey Garrett (2000) “Legislative Politics in the European Union”, European Union Politics, 1 (1): 9-36.
Tsebelis, George and Geoffrey Garrett (2001a) “Understanding better the EU legislative process”, European Union Politics, 2 (3): 353-61.
Tsebelis, George and Geoffrey Garrett (2001b) “The Institutional Foundations of Intergovernmentalism and Supranationalism in the European Union”, International Organizatio, 55 (2): 357-90.
Tsebelis, George, Christian B. Jensen, Anastassios Kalandrakis and Amie Kreppel (2001) “Legislative Procedures in the European Union: An Empirical Analysis”, British Journal of Political Science, 31: 573-99.
(*) The author would like to thank Ove Kaj Pedersen, Michael Shackleton, Una O'Dwyer, Ole Elgström, Derek Beach, Karina Pedersen, Morten Kelstrup and two anonymous referees for comments on earlier drafts.
(1) Roughly speaking, the literature can be divided into studies that use formal modelling such as Crombez 1996, 1997, 2000, and 2001, Steunenberg 1994, Garrett 1995, Garrett and Tsebelis 1996, 1997, 2000, 2001a and 2001b, Tsebelis 1997, Scully 1997a and 1997b, and Moser 1997 and qualitative studies such as Garman and Hilditch 1998, Shackleton, 2000, Earnshaw and Judge 1996, Jacobs 1997, and Farrell and Heritier 2002a and 2002b.
(2) Some of the first testing of the models have been carried out by the formal theorists themselves, most importantly the question of whether the Parliament has won or lost power with the introduction of co-decision (Tsebelis et al, 2001 and Kreppel, 2002).
(2a) To be fair, it should be noted that a series of the formal, rational choice institutionalists are aware that the Commission may exert informal influence beyond the formal Treaty provisions. As an example, Tsebelis and Garrett state "(t)he remaining influence of the Commission over legislation is thus likely to rely more on informal channels - asymmetries of information, persuasion and deal-brokering – than on formal roles written into the various procedures" (2000: 26). Thus, the problem is not that the formal institutionalists are ignorant towards the fact that informal means of influence exist but that they often draw very clear conclusions about the role of an actor totally ignoring these, and that they have not made any attempts to study these elements of the EU decision-making context further, for example by incorporating them into their formal models.
(3) The 28 interviewees included 4 administrators from the Secretariat General of the Commission involved in the co-ordination of co-decision and relations to the other institutions, 7 desk officers from various DGs of the Commission, who had been involved in concrete co-decision files, 3 parliamentary co-ordinators from various DGs of the Commission, 2 members of Commission cabinets in charge of relations to the European Parliament, 5 members of the Parliament, 2 administrators from the European Parliament and 3 from the Council working with the conciliation procedure, as well as 2 deputy permanent representatives from the Member States.
(4) COD/1994/0001, 1997/0103, and 1998/0195. 12 of the 28 interviewed have been involved in the negotiations of Socrates.
(5) Some files that fit these criteria have been omitted when some crucial citation for one of the stages was missing from the sources available to the coders.
(6) In fact, there were 36 files altogether, which have been in conciliation. However, one has been excluded from our analysis, as it did not include information about the Commission’s opinion on the parliamentary amendments and another because it was Socrates I, which will be dealt with separately.
(7) There are a few exceptions to this rule, since there were always a few treaty articles (currently four and after Nice three) where co-decision was applied with unanimity.
(8) Examples of cases in which the Commission has played a relatively minor role in the negotiations include Novel Foods (COD 00/426) and Postal Services (COD 95/221) adopted in January and November/December 1997 respectively.
(9) In the first reading there is a relatively high number of observations where the Council’s degree of adoption in the common position is coded ”unclear”. This means that the first reading figure should be treated cautiously.
(10) Instead, the result of the dispute was the so-called “modus vivendi” agreement, which increased the Parliament’s general right of scrutiny in comitology moderately (OJ C 43, 20.02.95, p. 37), whereas the question of whether budgetary figures should be written into the legislative proposals was solved by another agreement on “Amounts deemed necessary” stating that multiannual programmes in co-decision like Socrates should include an amount, which serves as “the principal point of reference” (OJ C 68, 20.03.95, p. 8).
(11) Hence, it would be wrong to use data showing that the Commission’s opinion on an amendment did not have an impact on whether the Council included it in its common position at the first reading stage, if the amendment supported by the Commission was reinstated at second reading by Parliament and actually included in the final legislative text after conciliation. Thus, parliamentary amendments in the first and second reading are usually not independent, primarily due to internal restrictions in the Parliament on the tabling of second reading amendments (Parliament’s Rules of procedure, 15th edition, rule 80). This means that, with very few exceptions (approximately 5 per cent in the current data set), second reading amendments to text pieces, which Parliament has also tried to amend at first reading, are typically reintroductions of previous amendments (with or without modifications).
(12) However, if the file goes into conciliation, the Council has to examine all the Parliament amendments, as these along with the common position serve as the reference point for the discussion in the conciliation committee (Article 251 [4]). Before the Treaty of Amsterdam, a similar wording can be found in paragraph four of “The 1993 Inter-Institutionel Agreement on Arrangements for the Proceedings of the Conciliation Committee under Article 189b” (OJ C 329, 06.12.93, p. 132).
(13) In fact, it is even written into the joint declaration for the practical arrangements of co-decision that the Commission may draft compromise texts (OJ C 148, 28.05.99, p.1). Trialogue meetings are smaller meetings between the key players from the Parliament, the Council and the Commission to prepare the work of the actual conciliation committee.
(14) More specifically, Tsebelis et al. show that the Commission’s behaviour predicts the overall fate of a parliamentary amendment 85 per cent of the time under the co-operation procedure, whereas this figure is reduced to 70 per cent of the time under co-decision (2001: 595-96). For the reasons stated in section 5.2 their figures on co-decision cannot be directly compared to our statistics.