Rory Creedon 

Should we interpret high consensus in the European Council as reflecting cultural norms, thus making power index approaches irrelevant?

There is a wealth of qualitative evidence documenting that despite formal procedures for decision making by qualified majority in The Council of Ministers (“the Council”), there is a cultural norm that dictates that decisions should strive to be arrived at by consensus. Indeed the web portal for the European Union (“EU”) states that “The European Council decides by consensus”[1], which confirms the existence of the norm although it cannot speak as to its cultural or other roots. However, despite this fact there are those who argue that when analyzing power within the Council it is still productive to take a power index approach which abstracts away from how power is actually exercised to make statements about the power each member “possesses by virtue of the [decision] rule itself”[2]. Characteristic of the debate is the differing stance taken by Moberg as against Holsi & Machover. The former argues that the power index approach is inferior to a cultural norm based analysis of Council decision-making because the underlying assumption of power indices is that every potential voting outcome has the same probability of occurrence, whereas in reality states have a fairly stable set of interests. This means that “the vast majority of the millions of theoretically conceivable coalitions are highly unlikely”[3]. The latter maintain that in the long-run, national interests can change and also the issues which the council will decide upon are varied and “highly unpredictable”.[4] As such, power indices provide us with the best estimates of how power will be exercised in a wide time horizon. This essay argues that due to the institutional design of the Council it is impossible to verify categorically which approach is most successful. Specifically the system of pre-negotiation in Le Comité des Représentants Permanents (COREPER) only allows us to observe successful legislative motions, that is those where a consensus can be achieved. Without data on the proposals rejected in the COREPER stage that never made it to the main Council negotiations, and the reasons for their being so rejected, it cannot be observed empirically what the implicit decision rule is in committee stage and hence whether there is a true cultural norm based drive for consensus which would question the relevance of a power indices approach. That being said, upon inspecting the data that are available on Council decisions, certain inferences can be made regarding the rejection of proposals in the COREPER stage which indicate that a cultural norm based analysis is more successful. These inferences are based on the assumption that we would see more contestation in Council level voting data if COREPER negotiations were based to some large extent on the explicit decision rules as opposed to a cultural preference for consensus. The strength of this assumption, however compelling, is challenged only weakly by the ability of states to bargain across issues.

Data on decisions taken by the Council indicate that there is a strong tendency to consensus. Of those decisions formally subject to qualified majority voting in the 1998-2004 period, 75-80 per cent were not explicitly contested.[5] In the post enlargement EU this figure has actually risen.[6] In some sense it is futile to argue that this observed consensus is not driven at least in part by cultural norms that operate within the Council; countless interviews with those involved in the day-to-day workings of the Council confirm that there exists a culture of compromise, and a “consensus reflex”[7]. Whilst the origins of this culture are fascinating and frequently debated, to try to determine them would be outside the scope of this essay. Therefore, this essay starts from the non-controversial assertion that such a culture does exist, a fact accepted even by scholars who advocate analysis based upon power indices (Renshaw et al, Hosli & Machover). This notwithstanding, much time and energy has been devoted in the EU Intergovernmental Conferences (“IGCs”) to debating, upon successive enlargements, how voting weights should be distributed and at what level the threshold for passing legislation by qualified majority voting (“QMV”) should be set. This would appear to affirm that voting procedures continue to be important in spite of the existence of the informal norm of consensus which would seem to make them a somewhat less contentious agenda item. Heisenberg argues that the consensus norm renders voting weights “largely irrelevant”[8], and that they are argued over as no more than an insurance policy in the event that stable repeated interactions break down in the Council. Whilst this may indeed be case, the existence of the consensus culture, the disagreements as to procedure, the testimony of officials and the voting weights as insurance argument are not logically inconsistent with each other. Thus it remains to be determined whether a cultural norm based approach is sufficiently strong in explaining consensus in the Council thereby making analysis based upon power indices redundant.

The problem for scholars thus far has been that this is impossible to so determine as the voting data are highly biased. This is due to the structure of decision making in the Council. Prior to a decision being taken by the Council, legislation is discussed in various committees and specialised working groups . The most important of these committees is COREPER[9] which is comprised of permanent representatives from each member state. In practice nearly everything is agreed in the COREPER stage. “Real negotiations” take place in these pre-Council stages which can last for “months and sometimes years”.[10] Although COREPER never formally votes it does take decisions and has “de facto legislative competencies”[11]. This institutional design means that observed voting data at the Council level do not have much descriptive power as COREPER is unlikely to send a proposal to the Council if it is likely to fail subsequent to Council negotiations. This is evidenced by the fact that a majority of points passed to the Council for deliberation are “A points” on which no further discussion is needed.[12] What is observed then in Council level voting data does not include data on failed proposals, as proposals anticipated to fail would not be passed from committee to Council, and as such it is highly selective sample.

More troubling however, is that if negotiations progress by implicit voting in the COREPER committee stage as qualitative studies indicate,[13] then Council level voting data do not contain data that can allow a substantiation of the claim that decisions are arrived to pursuant to a culture of consensus. [14] Indeed, the structure of pre-Council negotiations can be interpreted as either a mechanism for reaching consensus, or a series of unofficial votes[15]. However, even if that latter interpretation is correct, there is no way of knowing what the decision rule is for the series of pre-Council implicit votes. If the implicit decision rule is the same as the explicit decision rule (a qualified majority in the case of decisions to be taken by QMV), then proposals in the COREPER stage will have be abandoned or killed if it appears that  there would be a blocking minority in the Council stage. Were this to be the case, then in spite of a high level of consensus at the Council level, a power index model would still have a lot of descriptive power as the a priori power derived from the decision rule is still of key importance at committee stage if not Council stage. If on the other hand there is “issue specific” power which leads to different “preference intensities”[16] and these preferences are respected, then even a single state or minority coalition not sufficient for a blocking minority in the Council, may have their point of view considered in a “compromise by accommodation of divergent interests”.[17] An implicit decision rule that accommodates interests even when such interests are pursued by a number of states insufficient to form a blocking minority in the Council, would seem to suggest a culture of compromise, thus making power index analysis somewhat redundant. Without specific data on how implicit decisions are taken in the pre-Council negotiations, and particularly in the COREPER stage it is impossible to conclude definitively whether analysis of Council decisions is adequately explained by the cultural norms thus rendering power index analyses irrelevant. This is the basic conclusion of this essay.

However, in spite of the data problem outlined above, it seems that it should in fact be possible to infer conclusions about the decision rule in the COREPER stage even though there is no extant data pertaining specifically those negotiations. As Hayes-Renshaw et al point out only 25 per cent of Council decisions decided upon by QMV were subject to a contested vote.[18] This figure seems to imply that the implicit decision rule in the committee stages is founded upon a desire to achieve consensus rather than based upon the explicit qualified majority rules that pertain at Council level. If the implicit decision in rule COREPER were qualified majority, then we would expect negotiations to proceed as follows: a proposal is discussed, negotiated and tabled which may or may not command a supporting majority. If it does command a majority then the proposal is passed to the Council with no further amendments. If it does not, then additional negotiations are held up until the point where it commands a majority, and no blocking minority coalition is feasible. At this point the proposal is passed to the Council. If on the other hand the implicit rule is consensus then following the tabling of the proposal, negotiations would proceed until all views that are capable of being accommodated without contradiction, would be included as amendments, and only at the stage where no further agreement is possible would the proposal be passed to the Council for approval.

Whilst it is not possible to conclude categorically from the voting data available which mechanism is in fact at work, it seems that the number of contested votes is simply too small to support the argument that the implicit COREPER decision rule is QMV. Absent a culture of consensus, there would be no incentive for those who form a majoritarian coalition of states supporting a proposal to allow for negotiations to proceed beyond the point where a majority is garnered. Yet of the votes decided at Council level by QMV, only a small fraction are contested. Moreover, 47 per cent of those contested decisions are only contested by a single state and a further 19 per cent by only two states[19]. The assumption here is that if decision making in the committee stages were based upon an implicit QMV, the number of contested votes in the Council would be significantly higher, and in particular the number of instances where a coalition of member member states simultaneously recording their contestation would be a lot higher. It is argued here that assumption is strong even if it is the case that member states that are in the losing coalition do not wish to record the fact officially in the vote, for sake of preserving domestic political capital. Thus it seems that even from the voting data that are available we can lean toward a conclusion that when looking at decision making in the Council an analysis based on a cultural norm of consensus is the most appropriate, and this tends to make power index analyses unwarranted.

The issue not capable of firm resolve at this point however as it is further clouded by the ability of states to bargain across interests. It is well known that inter-temporal vote swapping is a method used in pre-Council negotiations in order to achieve agreement.[20] This involves member states who might well be a part of a minority coalition opposing a bill, agreeing to the proposal in exchange for some consideration of their interests in another area being decided upon by the Council. Interpreting such behaviour through the lens of Robert Putnam’s two-level game theory[21], it seems clear that for issues that are not significantly salient to the domestic constituencies, member states negotiating in COREPER can effectively “lose out” in one area under negotiation if they are compensated for this loss by a “win” in a different area. In the language of Putnam, the Level II constituency win-set is contoured such that a level I bargain involving vote-swapping and issue-trading is possible. This is problematic for the analysis presented above because a lack of contestation in the voting data, rather than supporting the assertion that consensus is the implicit decision rule in COREPER, could instead be reflecting the ability of states to trade votes. If states are willing to support proposals they would otherwise vote against, in return for concessions in another area, and this system of compensation occurs in the COREPER stage then it is to be expected that levels of contestation at Council level should be low. This is because, states that form a potential blocking minority, can be persuaded to agree to a proposal through vote trading.

However, whilst this may challenge the above interpretation of low levels of contestation indicating a decision rule founded on consensus, the fact remains that nearly half of all contested votes are only contested by one member state. If the implicit decision rule in COREPER were QMV then states that support a proposal that currently cannot garner a majority would only be willing to compensate members of a blocking minority until such a point where the minority was no longer capable of voting down the proposal in the Council. Given that a blocking minority consists of 50 per cent of states[22], absent a cultural norm for consensus it would be rational for supporting states only to grant concessions to the minority until 50 per cent minus one state would agree to the proposal (assuming the voting weight/population conditions are also satisfied). Yet what is observed is that in the majority of contested decisions, only one state is willing to go on the record as a dissenter. This strongly suggests that the implicit decision rule in the committee stages is consensus and not QMV, and this consensus is reached both by trying to accommodate divergent views, and vote trading across issues.

This essay has argued that it is not possible to categorically state whether decision making in the Council is adequately analysed using a cultural norm based approach to understanding consensus thereby making a power index approach irrelevant. This is because the institutional design of decision making in the Council means that much is decided in pre-Council committees, particularly COREPER. As it is not possible to observe how decision making occurs in those bodies and specifically what the implicit decision rule is, it cannot be stated with certainty that decisions are made predominantly pursuant to an informal rule of consensus, or rather based upon the explicit rule of QMV. If it is the former then power index analysis is not useful, whereas if it is the latter, then power indices can still be of use to scholars analysing decision making. That being said, low levels of contestation in votes in the Council seem to imply that consensus informs decision making in committee stage rather than the explicit voting rules. This assertion rests on the assumption that contestation would be higher at council level if COREPER decision making were based on implicit QMV. This conclusion is challenged somewhat by the ability of states to vote trade across policy issue. However, the number of states that contest a vote appears too low to plausibly support the thesis that it is QMV rather than consensus that is driving decision making in COREPER. This is because absent a cultural norm that dictates consensus, it would be irrational for supporting states to grant more concessions than strictly necessary in order to gain a majority. Therefore although it cannot be proved empirically, consensus in the Council is best understood from a cultural norm perspective, and this leaves analysis based on power indices somewhat irrelevant.

[2] Felsenthal & Machover A Priori Voting Power: What is it all about? Political Studies Review Vol. 2 (2004) p.3

[3] Moberg The Nice Treaty and Voting Rules in the Council Journal of Common Market Studies Vol. 40, No. 2 (2002) p. 261

[4] Holsi & Machover The Nice Treaty and Voting Rules in the Council

[5] Hayes-Renshaw, Van Aken & Wallace When and Why the EU Council of Ministers Votes Explicitly, Journal of Common Market Studies Vol. 44, No. 1, (2006) p. 163

[6] Hagemann & De Cleck-Sachsse Decision-Making in the Enlarged Council of Ministers: Evaluating the Facts Centre for European Policy Studies, Policy Brief No. 119 (2007) p. 3

[7] Lewis The Methods of Community in EU Decision-Making and Administrative Rivalry in the Council’s Infrastructure, Journal of European Public Policy, 7:2 (June 2000), pp. 270-271

[8] Heisenberg The Institution of ‘Consensus’ in the European Union: Formal versus informal decision-making in the Council, European Journal of Political Research 44, (2005)  p.71

[9] Mattila Contested Decisions: Empirical analysis of voting in the European Union Council of Ministers European Journal of Political Research 34 (2004) p. 30

[10] Moberg Op cit. p. 277

[11] Lewis Op cit. p 264

[12] Matilla Op cit. p. 30

[13] Moberg Op cit. p. 278

[14] Hayes-Renshaw et al Op cit. p. 164

[15] Holsi & Machover Op cit. p 512

[16]Tallberg Bargaining Power in the European Council Journal of Common Market Studies Vol. 46 No.3, pp. 692-3

[17] Lewis Op cit. p.271

[18] Hayes-Renshaw et al Op cit. p165

[19] Ibid. p. 169

[20] Heisenberg Op cit. p.69

[21] Putnam Diplomacy and the Logic of Two-Level Games International Organization, Vol. 42 No. 3 (1988) pp. 427-460

[22] Kirsch The New Qualified Majority of the Council of the EU


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: