http://eiop.or.at/eiop/texte/1997-001a.htm
:
10.4.1997| Old 'foundations' and new 'rules' - For an enlarged European Union(*) |
| Philippe C. Schmitter and José I. Torreblanca |
| European Integration online Papers (EIoP) Vol.
1 (1997) N° 1;
http://eiop.or.at/eiop/texte/1997-001a.htm |
| Date of Publication in :
10.4.1997 |
| | Abstract | Back to homepage | PDF | PS | |
1
This essay presents a novel arrangement for the distribution of votes and the rules of decision-making in an enlarged European Union (EU) (1). It starts from the observation that if one merely transposes the present arrangement for the EU of 15 member states to prospective EUs with from 17 to 35 members, the result would be a very substantial distortion of one of the EU's most important "foundational principles," namely, its method for weighing the voting power of members according to their size of population. Under existing rules for qualified majority voting, it takes 62 out of 87 votes in the Council of Ministers -- needless to say, voting as a single body -- or 71% of the total to pass a measure and 26 votes out of 87 (30%) to veto one. This means that the five largest countries (Germany, France, the United Kingdom, Italy and Spain) can veto decisions, but cannot produce a qualified majority without help from either the five medium-size countries (Netherlands, Greece, Belgium, Hungary and Sweden) or the five smallest countries (Austria, Denmark, Finland, Ireland and Luxembourg).(2) Hence, the small and medium countries together cannot possibly form a qualified majority without the five largest ones, but they are able collectively to veto a decision. In other words, for the actual system to work effectively, cooperation is needed between all three size categories and, hence, the results are likely to be consensual.(3)
However, extending this system to a total of, say, twenty-seven members would violate the existing size principle in two important ways. First, the ten prospective newcomers from Central and Eastern Europe (CEECs) would alone have a veto possibility, based on an aggregate population of 105 millions or just 22% of the EU's total population. Second, it would become possible to form a qualified majority only by adding the largest countries and the medium-size ones -- not, as before, by adding the largest and either the medium or smallest ones. Any subsequent change in the direction of an absolute majority system would only further aggravate the problem by giving the medium and smallest countries, with only 26% of the population, a potential winning coalition over the seven largest countries with 74% of the EU's total population.
2
In contrast, a system that assigned votes in the Council of Ministers and seats in the European Parliament according to an uniform principle based in the square root of national populations has three advantages:
We call this arrangement: proportional proportionality.
A second reform would go further toward ensuring a consensual decision-making system and, nevertheless, allow the expanding EU to be make commitments binding on all against the resistance of individual states -- even against the will of one of its largest members. An apposite device for doing this would involve dividing the votes in the Council of Ministers -- putatively, the EU's upper chamber -- into three Colegii according to the relative size of member states. All "constitutional" decisions involving important substantive reforms, rule changes or the admission of new members would have to receive the support of all three colegii -- instead of the present unanimity rule. Thus, no measure of major significance that was binding on all members could pass the Council unless at least a simple majority of the weighted votes of small, the medium and the large countries, voting separately according to proportional proportionality, approved it. For day-to-day matters, the present qualified majority or simple majority rules could be translated into the system of Colegii, but in a less demanding fashion. Normal directives or EU laws would be approved if they obtained a simple-weighted majority in at least two of the Colegii. In practice, this would mean that the (weighted) approval of Colegio I + Colegio II or Colegio I + Colegio III would suffice; whereas, the opposition of any two Colegii (including Colegio II + Colegio III) would amount to a veto. In the event of routine decisions currently governed by simple majority criteria, the same formula would apply, but only simple majorities would be required within each Colegio and for the Council as a whole. We call this a system of concurrent majorities(4).
3
As we observed above, EU-15 works with a qualified majority voting procedure that requires 62 votes to be cast to reach a decision and 26 votes to veto a proposal. If we divide these fifteen member states -- de facto and not (yet) de jure -- into three Colegii of five members each based on their relative size, we get the distribution illustrated in Table I (in absolute terms) and in Table II (in relative terms).(5) Colegio I of the five largest states has 48 votes or 55%; Colegio II of the middle-size states has 24 votes or 28%; and Colegi III of the five smallest has 15 votes or 17% of the total 87 votes that can be cast. Any decision which receives a qualified majority in a Colegio automatically receives all of the weighted votes of that Colegio. Under such an arrangement, with the qualified majority established at 62 votes (71%), a positive decision can be reached either by adding the votes of Colegii I + II or Colegii I + III, but not by adding Colegii II + III. However, any two Colegii can, if they vote in the same direction, veto a decision.
It is our assumption that the existing system is not merely an accidental product, but reflects an underlying principle that serves to legitimate decision-making among national states of very considerable diversity in their respective populations, geographical area and market power. We believe that this basic "size principle" should be conserved in the future as the EU incorporates new members.
The enlargement from EU-12 to EU-15 already produced several important crises. Great Britain bitterly resented any decrease in its potential veto power and envisaged a dangerous future in which "spendthrift" Southerners could outvote "abstemious" Northerners. Inversely, the Spanish government wanted to prevent the disappearance of the potential "Mediterranean Veto," according to which Spain (8 votes), plus France or Italy (10 votes each), plus Portugal or Greece (5 votes each), could reach the 23 votes necessary to veto any Council decision.(6) The so-called Ionannina compromise reflected the necessity to assure some countries that the decrease in their relative voting strength weight due to an eventual increase in membership would be taken into account, even if informally. In this particular compromise, it was decided that whenever any group of states representing 23 to 25 votes would oppose a decision to be taken by qualified majority, their negative opinion would be taken into account, even if these countries lacked to capacity (or will) to cast a formal veto, and the Council would try to find a satisfactory solution which could receive at least 65 votes (instead of the 62 necessary).(7)
4
Whether this particular solution will endure, the pattern of conflict behind it illustrates very well the problems that are bound to emerge with greater intensity in the future. Enlargement to the East (and, to a lesser extent, to the South with Malta and Cyprus) has to involve states with two characteristics that are most adverse to the present EU voting system: (1) either they are small in population and, hence, would have the net effect of increasing the already very considerable over-representation of small countries; or (2) they come from a geographically and culturally "compact" part of Europe and, hence, threaten to contribute further to the already considerable fear of regional bloc formation.(8) The emergence of Northern, Eastern and Southern or Mediterranean bloc voting could well unleash a broader uncertainty about where the "center of gravity" or "core area" of Europe really was -- and disrupt the current (implicit) consensus on Franco-German duopoly. The present system has, so far, proven to be tolerable, despite its gross disproportionality at the superior and inferior extremes of country size (see Table II: at present, 79% of the population has 55% of the votes), but it could disrupted if overshadowed or displaced by regional solidarities.
Actors seem aware of the potential serious destabilizing effect of maintaining and extending the present system to future configurations of the EU. It was one of the major preoccupations of decision-makers within the so-called "Reflection Group" that prepared the current Intergovernmental Conference (IGC) on political reform. However, progress in this matter since the IGC first met in April of 1996 has been meager -- if non-existent -- due to the reluctance of member states to consider any alteration of their relative power under a new voting and decision-making system. A variety of formulae have been proposed without success, mainly because they have not been able to offer a system which is both effective and legitimate.
An examination of several alternative systems, all derived from Public Choice Theory, has recently been carried out by Torsten Peters.(9) The first one (the Shapley-Shubik Index) considers which country would be pivotal to the determination of winning coalitions, according to a given order of voting. The second (based on the Banzhaf Index) examines potential coalitions, instead of permutations, and attempts to find which countries can turn winning coalitions into losing coalitions by changing the direction of their vote. According to that system, larger countries can be pivotal more often than smaller ones. Finally, the third uses Holler's Index in an effort to correct what its author considers an overestimation of larger country capabilities inherent in the preceding two indexes and proposes a solution which results in a more equal distribution of weighted votes.
5
Peters then confronts the three indices with four different voting reform scenarios, while holding constant the number of states at fifteen. The results show clearly that any reform that reduces the number of votes necessary for a majority to be formed, will only marginally reduce the power of small states. Furthermore, if the requisite majority were reduced to a simple majority and, simultaneously, the number of countries required to form a majority is also decreased (to 9 or 8), the power of small states would even increase compared to the actual system.
In other words, the "common sense" solution to future enlargement, i.e. to reduce the number of votes and countries necessary to form winning majorities, would have precisely the opposite effect, if -- as will definitely be the case -- new members are predominantly small in size. In an EU with 27 members, decisions by simple majority would definitely be easier than by qualified majority, but they could be taken by member states representing only 26% of the EU's then total population (i.e. by 122.200.000 against 355.600.000 inhabitants). If, in reaction to this prospect, the 27 future members decide to maintain the present qualified majority system, the picture is not much more promising, since the Eastern European latecomers (Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, Slovenia) with 41 votes out of 132 and a population of 22% would have a potential clear veto on all new measures.(10)
The combination of a decision-making arrangement based on three Colegii and the weighting of votes according to the square root of population appears (to us) to be an optimal solution that would keep the EU system working according to its present principles and allow it to adjust to any future configuration.(11) Let us see why.
We begin with the premise that any reform should respect grosso modo the actual system of relative weights between the larger, medium and smaller countries. We are looking for a device whereby the largest countries would still need the support of either the small- or the medium-size countries to obtain a binding majority. This is a system that guarantees the right to preeminence of the large countries while preventing them from "tyrannizing" the others. Also, since their concurrence is necessary, it makes it worthwhile for medium and small states to remain within the EU -- rather than to "free-ride" on its periphery. But the peculiarity of the system is that it also gives a fair, i.e. a proportionately proportionate, share of voting power to the medium and small states. Together, they cannot impose their will on the large countries, but together they can veto any proposal.
6
The obvious solution is to find a device or devices that could permanently insure the reproduction of such decision-making principles, no matter how many members the EU decides to admit in the future. An examination of the effect of weighing states' voting power according to the square root of their populations demonstrates the viability of such system in a Europe of 17, if and when the two small states of Malta and Cyprus are admitted; a Europe of 21, by adding the Visegrad Four; and a Europe of 27 that would include the ten Eastern European and Baltic states, but not the CIS, the Ukraine, Moldova, Croatia, Serbia-Montenegro, Macedonia, Albania or Moldova. It even works with a mega-Europe of 35 countries and 572 million inhabitants!
This can be accomplished by combining a qualified majoritarian system at the EU level with a qualified majority-bloc voting system at the level of the Colegii. The qualified majority rule would still hold globally because it would continue to be necessary for any directive to obtain 71 percent of the weighted votes for a proposal to become binding. Larger countries would still have to ally with either medium or small countries to obtain that 71%. Also in common with the present system, the medium and smaller countries could veto a proposal agreed to by the larger countries, if they voted together.
The basic difference is that voting in the Council would also take place within each of the three Colegii under a modified first-past-the-post system. Any coalition of countries whose weighted votes approved a directive/law within a Colegio would carry all the votes of that Colegio to the next stage.(12) As we shall see below in greater detail, this produces two net advantages. First, it would break up the possibility of forming dominant or veto blocs based on geographic or cultural regions, since the Northerners, Southerners and Easterners would be distributed among the three Colegii according to their relative size. By itself, this should suppress the fears derived from the typical instability of triadic systems which the prospect of three regional groups would surely unleash. Second, within each Colegio, the range of possible coalitions should be much more varied, permitting countries to align according to their particular preferences with regard to the policy issue at stake, rather than according to relatively fixed strategic calculations of relative power and status. For example, in EU-27, Colegio I would have seven countries. Germany plus the U.K, France and Italy would have an enabling majority, but Spain, Poland and Romania could also be in the winning coalition if they could convince any two of the four above to join them. In Colegio II with nine middle-size countries, any five countries could form the necessary majority, since they are so close to each other according to the size principle. Finally, in Colegio III with its eleven small members, the possible winning coalitions would range from any four of the largest to the five smallest plus at least two of the larger ones. The combinations are, thus, quite diverse -- one is tempted to say, pluralistic -- but relations between the Colegii should ensure a strong bias toward consensual solutions rather than the winner-take-all mentality of simple or even qualified majority systems in a single representative body.
7
To become a law or "directive" of the EU, a proposal would have to obtain a concurrent majority (CM) of 71% of the Council votes as a whole and a simple majority in at least two of the Colegii. For matters of constitutional importance, the criteria could be strengthened to include the requirement that a simple majority of the weighted votes in all three Colegii would be necessary.
The entire system rests on proportional proportionality (PP) derived from the square root of each country's population. We are assuming that this universalistic standard, while it deviates from the usual "one person, one vote" criteria applied in virtually all liberal democracies, would prove acceptable to the EU as a whole. In effect, what PP would ensure is that representation would be equal for similar proportions of the citizenry of different size political units. For example, ten percent of the population of Luxembourg (40,000 persons) would have approximately the same voting weight/number of seats as ten percent of the population of Germany (816,000 persons). Not only can this criterion be defended normatively -- provided all agree that both Luxembourg and Germany deserve to persist in their present territorial/ demographic configuration -- but it is a more objective and easy to measure basis for differentiation than level of development, geographic location, religious conviction or any other cultural trait. Europeans have long lived in national societies of different size and this is predictably associated with varying intensities of feeling with regard to such things as language use, cultural preferences, institutional peculiarities, and so forth. If the larger countries refuse to recognize this proportionately fair criterion for over- and under-representation and insist on strictly numerical equality, any peaceful and voluntary effort to democratize the Euro-polity is bound to fail.
If the proposed PP system where applied to the present EU-15 and the number of votes in the Council of Ministers remained unchanged at 87, the previous system maintains its basic features -- which is precisely the goal we have been seeking. Germany's votes would be increased to twelve (from 10) and Greece, Belgium, Portugal and Luxembourg would lose a single vote. A quick look at Table V shows how little variation in votes the new system would produce.(13) Thanks to the Colegii arrangement, it would still be necessary for one of the two smaller units to enter into a coalition with Colegio I for any measure to pass. Moreover, as we have discussed above, coalitions within each Colegio should be very fluid over time and across issues, given that any three of their five members can form a simple majority and obtain all the votes of that Colegio.(14) Now, let us see how the system would work when faced with three possible and successive enlargements.
8
Let us first speculate with what would happen should Malta and Cyprus enter the EU-15 as planned.(15) Under the present system, both countries, one with 400,000 and the other with 700,000 inhabitants, would be entitled to two votes in the Council. This would quite substantially alter the voting strength of the small member states since, by adding only slightly more than one million inhabitants to the Union (0.3%), it would increase their collective votes by 27%, i.e. from 15 to 19. Faced with that prospect, the fifteen could opt to do nothing. The sub-set of large countries could still not be overridden by a potential coalition of medium and smaller countries, and they would still need the votes of one or the other "size blocs." Needless to say, the (implicit) Southern or Mediterranean Veto would be strengthened to the delight of Spain and Italy and the dismay of Britain and the Scandinavians.
If PP were applied to EU-17, winning majorities could be formed without including as many of the larger countries. The twelve medium and small-size members would have 42% -- if they voted together -- and the five large ones would have the remaining 58% votes in the Council.(16) Of course, according the rules of the Council of Ministers, it would still be highly unlikely that its members would pass anything that could negatively affect the group as a whole and, on constitutional issues, this group would still have a veto. The strongest argument in favor of introducing the reform already for EU-17 is that it would serve to familiarize members with their new voting weights and allow them to begin experimenting with the formation of coalitions within each of the Colegii. Moreover, this would set a precedent under relatively easy conditions for negotiation since the immediate impact would be so negligible. A very important and contentious issue that is bound to plague much more complex enlargements in the future would have been resolved in a way that is universal, predictable and defensible in terms of democratic theory. Each potential member state would know before it entered into negotiations what its weighted vote would be and where it would fit within the decision-making procedures as a whole. The "shadow of the future" will have been clarified considerably, if not dissipated altogether -- and the conflicts surrounding eventual inclusion and exclusion should (hopefully) turn to other, more substantive, matters.
If the IGC on institutional reform were to adopt PP and CM, they might also choose a further simplification of EU rules. For example, they might set the number of weighted votes in the Council of Ministers at a fixed sum -- say, one hundred -- and make it even easier for everyone to calculate the possible impact of any enlargement.(17) This way, all eventual changes in membership would result in a (predictable) redistribution of weighted votes among the pre-existing members, as well as the assignment of new "quotas" to those just entering. Under the Colegii system, countries would be assigned to one of the three sub-groups according to the size of their respective populations.(18) If it was felt to be desirable to keep the Colegii more-or-less equal in terms of the number of members, specific countries could even be promoted from medium to large or small to medium status, although in so elevating themselves they would lose some of their proportional clout. Perhaps, one could allow them the choice of remaining a relatively large fish in a small pond or becoming a small fish in a larger pond! In any case, all of the fish -- no matter how small -- would get at least one vote.(19)
9
For EU-17, the results of PP and CM would be as follows, according to Table VI:
The overall principle is protected, however. If would still take a majority in Colegio I plus either Colegio II or III to win, and the combination of Colegio II and III can still veto.
Now, the institutions of the EU are being put to a stiffer test, that of incorporating the so-called "Visegrad Four", i.e. the Czech Republic, Hungary, Poland and Slovakia.(20)
The basic rule of qualified majority would still hold, since Colegio I (with Poland as its sixth "Great European Power") would have 54% of the total votes, Colegio II (now joined by the Czech Republic and Hungary) would have 29%, and Colegio III (with incoming Slovakia and descending Austria) 17%. Thus, the small and medium countries could veto but not decide; whereas, the largest countries would still need one of the two small Colegii to reach the necessary majority. Also, coalitions within each Colegio would be potentially quite varied -- at least without reference to any specific substantive issue.(21) In Colegio I, the potential threat of a majority bloc composed exclusively of Mediterranean countries (France + Italy + Spain) is dissipated. Either the smallest four or the biggest three among the Great Powers can form a winning coalition. In Colegio II, the picture has become even more varied: out of seven countries, any four could produce a majority and the only stable alliance one might imagine would involve its four older members (Greece, Netherlands, Portugal and Belgium). This is not very likely given their behavior in the EU to date. In Colegio III, the biggest four countries could dominate, but so also could the smallest three if they could convince any three of the five remaining ones. Thus, the system of relative weights within Colegii plus the "first-past-the post" bloc voting provision assigned to each Colegio should guarantee respect for the one of the foundations of the present EU system, despite a major change in the number of players. There are still the internal checks and balances between different countries grouped by size, but no greater a propensity for stalemate or blackmail by hardcore recalcitrants.
10
Now let us see what happens to the properties of the EU system when PP and CM are applied to what many regard as its most likely final configuration -- at some indefinite moment in the Twenty-First Century. Besides the Visegrad Four, two Southeastern European countries (Bulgaria and Romania) and the three Baltic states (plus Slovenia in the very near future) have all signed association agreements with the EU. No doubt, their inferior economic conditions (not to mention the problematic consolidation of democracy in several of these countries) might defer their entry for some time; nevertheless, the EU's formal stance is that they will all become members.
And this is where and when the decision-making problems will surely become especially acute, unless the existing rules are changed. If not, as we have discussed above, the new "Eastern Bloc" of ten countries (out of 27) would be so over-represented that collectively they could veto all initiatives within the Council (while only having 22% of the EU-27's total population). This is presumably the menacing "shadow of the future" that all existing members have an interest in avoiding.
And the proposed combination of PP and CM would do just that -- without resorting to any explicitly discriminatory treatment. Especially if it could be implemented early enough, its effects should be perceived as generic and fair. It would tend to impede the formation of any geographically or culturally based "sub-regional" veto, just as it would continue to encourage the larger countries to take into account the preferences of their medium and small "partners."
In EU-27 with PP and CM, Colegio I joined by Romania would have a weight of 54%; Colegio II, now with Bulgaria in its ranks, would have 29% and Colegio III with four new members (the Baltic Republics + Slovenia) would have 17%. With the qualified majority remaining at 71% and the prospective veto at 30%, all of the checks and balances discussed above would continue to function. Within each Colegio, a considerable variety of winning and vetoing combinations have become possible -- and, moreover, they do not cluster by either level of development or geo-cultural region. In Colegio I, a coalition of Germany plus the next three "bigees" or one formed by the smallest three plus any two of the four largest countries would be able to pass directives and win all of its 53 votes. In Colegio II, any five of the nine countries composing it could produce a simple majority and obtain its 15 votes. In Colegio III, the necessary majority would be 9 out of 17 votes. Its five Eastern and Baltic members would have to attract either Denmark, Finland, Ireland, Cyprus, Luxembourg or Malta to their ranks in order to produce a majority -- and even then they would still have to join with a majority in either Colegio I or Colegio II to enact their preferred measures. Most likely, the pattern of voting in Colegio III with its more numerous (and one must add very heterogenous in terms of level of development) membership should be quite varied with possible majorities coming from any four of its two-vote members plus one of its five one-vote governments, while all of the five smallest countries in that Colegio would need at least two of the larger ones to reach an effective majority.
11
We are now pushing the envelope of possible "Europes." And no one questions that the issue of such a "mega-enlargement" lies far in the future. EU-35 would include all countries that, for one reason or another and with one degree of seriousness or another, have expressed or might express in the future a desire to enter the European Union. These prospective members include: Turkey, the Federal Republic of Yugoslavia (Serbia and Montenegro), Switzerland, Croatia, Norway, Macedonia, Albania and Iceland.(22)
Table IX shows that the application of PP and CM to a putative Euro-polity of this enormous scale -- 572 million inhabitants according to present figures! -- would not produce radically different potential political outcomes, even if one could question whether any mechanism of democratic representation and accountability could function effectively for such a demographically numerous and culturally varied population. To demonstrate that EU-35 is possible is not the same thing as arguing that it would be desirable.
The increase in number of countries belonging to Colegio III by five does not alter substantially the situation with respect to EU-27 as described above. Colegio I would get just one new member (Turkey), but only diminish its proportion of the total Council votes from 54% to 47%. Again, it would only be able to reach a qualified majority with the help of winning majorities among the medium or small states in either Colegio II or Colegio III. The former would have 26% of the votes (as opposed to 28% in EU-27) with 17% of the expanded European population. The latter would now have a total of 28 of the 104 votes in the Council, up from 17 of 98 in EU's previous configuration. Moreover, its 27% would correspond to 60 million Europeans, while Colegio II with almost 95 million would have one less vote. Despite this anomaly, by making voting concurrent among the three Colegii, its effect is nullified. As before, the PP & CM system works to prevent the larger states from being placed in a minority and to ensure that the medium and smaller states will not be marginalized or ignored. Those eight "Great European Powers" with 73% of the total population would have to be content with 47% percent of the voting power, while the rest of weighted votes would be almost equally distributed between the nine medium and the eighteen small states.
* * *
Until proven wrong, we are convinced that the combination of proportional proportionality and concurrent majority offers the best solution for allocating weighted votes and for making binding decisions in the Council of Ministers of a continuously enlarging European Union. Its basic principles are consistent with those already in use and defensible in terms of democratic theory; its calculations are simple, predictable and open-ended; its procedures should promote flexibility and consensual decision-making; its results are likely to be sufficiently fair and other-regarding to encourage those who are outside the EU to join it and those who are inside the EU to remain within it.
12
The critical (and democratically concerned) reader will surely have noticed that we have given scant attention to the European Parliament (EP) in this discussion of possible reforms in the rules of decision-making and voting. The observant reader, however, will have noticed that in each of the tables we have displayed the effect of applying PP to the distribution of seats in that body. In these tables, he or she can readily see how various future enlargements of the EU might be translated into changes in the distribution of seats for Members of the European Parliament (MEPs) if the method of weighing according to the square root of the country's total population were applied and if the eventual reformers were sensible enough to limit the total size of the EP to 700 MEPs. We see no reason why, if the principle of proportionate proportionality is acceptable for distributing votes in the Council of Ministers, it should not be used for same purpose in the Parliament. A quick perusal of the results it produces from EU-15 all the way to EU-35 would show that -- as intended -- it overrepresents the smallest countries in particular, but it does not radically alter the present system. Perhaps its major difficulty would have to be faced initially since such "biggees" as Germany (88 from 99), the United Kingdom (74 from 87), France (74 from 87) and Italy (73 from 87) would have to face a considerable reduction in their number of seats. It should be noted, however, that these large delegations could still be expected to wield a considerable amount of informal influence via party channels and that, in any case, the EP is not designed to represent "national" populations as such, but individual citizens.
What would not be appropriate would be to transfer the system of concurrent majorities to the European Parliament. The main purpose of the EP is to insert within the political process of the EU a different "partisan" mode of representation, one that can potentially moderate and counteract the effects of representation according to national states. In the EP, it is the individual citizen that is being (indirectly) brought into the process through the internal divisions of each country's party system and (eventually) through the formation of partisan alliances along transnational lines of cleavage and solidarity. Representatives in the EP are not expected to vote along national lines and it would make no sense to force them to do so through the concurrent majority system.
We can only think of two instances in which its division into three Colegii might make some sense:
Otherwise, the procedures of the European Parliament should be as close as possible to those of a "normal" national parliament with its self-elected leadership, its party caucuses, its standing committees, its majoritarian decision-making, and so forth. Admittedly, this will demand some major reforms in the future, especially with regard to the nature and uniformity of its electoral system, but to discuss them here would take us far beyond the intent of this short article.(23)
(*)This is a draft manuscript of
work-in-progress. However, you are encouraged to comment, circulate and
cite it as much as you please, since we are very anxious to use it as a
vehicule for stimulating debate on the subject of institutional reform in
the EU. This essay will appear shortly in the working paper series of the
Centro de Estudios Avanzados en Ciencias Sociales of the Fundación
Juan March (Madrid) where it was originally written and presented, and in
the working paper series of the Robert Schumann Centre of the Istituto
Universitario Europo (Florence) where it was subsequently discussed and
revised. Eventually, it will appear as an appendix to a lengthier work by
the senior author, tentatively entitled: How to Democratize the European
Union and Why Bother, which should be finished by early 1998.
Please send any comments, criticisms or corrrections by mail to
Professor Philippe C. Schmitter, Dipartimento di Scienze Politiche e
Soziale, Istituto Universitario Europeo, Badia Fiesolana, I-50016 San
Domenico di Fiesole, Italia; by fax to ++39/55/46.85 201 or by email to:
schmitte@datacomm.iue.it.
(1) Actually, as Schmitter was reminded when he presented these ideas at a seminar of the Cellule de Prospective in Bruxelles, the arrangement it proposes is not really so novel. One participant came up afterward and remarked: "You know what you have done? You have re-invented the Holy Roman Empire!" Subject to further research on the matter, the accusation might not be as damaging as it sounds. After all, the HRE lasted for quite awhile, was a genuinely European creation, and probably made a net positive contribution to the integration of the European State System. Admittedly, it was not so good at preventing wars (and may even have encouraged a few), was consistently dominated by its larger members, and it had a strong reputation for corruption -- but no arrangement is perfect!
(2) The Treaty (Art. 148) establishes the qualified majority at 62 votes whenever a Commission proposal is involved and 62 votes representing at least ten member states in the rest of the cases. The clear intention is to avoid "government by size," i.e. that the largest eight member states acting together would be able to make rules binding all fifteen.
(3) It should be noted that this "hidden"
rule operates whether or not the actual voting patterns vary according to
the size of country. They are, so to speak, embedded in the context of
voting in the EU's Council of Ministers and presumed to affect the
behavior of member states by anticipation and habit, i.e. they "structure"
actor preferences in advance. In any case, since Council voting is "opaque,"
there is no systematic and reliable way of verifying present behavior and,
hence, of predicting how the change of rules we propose will affect future
behavior -- which is not to say that those who have been directly involved
will not be able to estimate their effect.
For a comprehensive assessment of how the Council operates, see
Fiona Hayes-Renshaw and Helen Wallace, "Executive power in the
European Union: the functions and limits of the Council of Ministers,"
Journal of European Public Policy, Vol. 2, No. 4 (December 1995), pp.
559-582. They estimate that contentious issues actually leading to formal
voting by ministers -- whether by unanimity or qualified majority -- only
amount to 10 to 15% of the total. The rest is settled consensually by
senior officials at one level or another (p. 562). They also conclude that
they could find "no systematic cleavage between smaller and larger
members" (p. 577) -- which does not, we believe, argue against the
utility of the Colegii system we propose. Our purpose is to preclude and
preempt conflict among different sized countries, not to mirror or
institutionalize this line of cleavage.
For the conclusion that all previous enlargements have enhanced the
relative voting power of smaller EU members, see Madelaine O. Hosli, "Admission
of European Free Trade Association states to the European Community:
effects on voting power in the European Community Council of Ministers,"
International Organization, Vol. 47, No. 4 (Autumn 1993), p. 631.
(4) Unfortunately, the origins of this concept go back to an effort to structure the decision-making system of the United States in such a way as to ensure the perpetuation of slavery. Not exactly an auspicious start! John C. Calhoun, A Disquisition on Government (New York: Peter Smith, 1943), originally published in 1853.
(5) Population figures for all the Tables have been extracted from 1996 Keesing's World Record of Events, which uses United Nations data.
(6) Prior to the Nordic enlargement, qualified majority stood at 54 votes and veto at 23 with a total number of votes of 76. Spain argued that Norway, Finland, Sweden and Denmark, with only 23 million inhabitants, would have 13 votes at the Council; whereas, Spain would count on only 8 votes having a population of 39 million. With the support of the United Kingdom, it fought unsuccessfully for new veto rights be awarded according to population criteria -- more specifically, whenever coalitions of two big and one small country representing 100 million inhabitants opposed a particular decision. Not coincidentally, Spain plus France or Italy plus Greece or Portugal made up that figure (see El País 2 March 1994, p.4).
(7) Council Decision of 19 March 1994, amended on 1 January 1995. See Official Journal of the European Communities, No. C 105 of 13 April 1994 and No. C 1 of 1 January 1995.
(8) As we observed above (FN 2), the issue is not whether the Eastern (or, for that matter, the Mediterranean or the Northern) countries "really" form a voting bloc, but whether they might be anticipated to do so on specific issues or at some specific moment in the future.
Moreover, the present system offers some perverse incentives. For example, the former Czechoslovak Federal Republic would have received five votes in the Council; whereas, the Czech and the Slovak Republics now that they have "velvetly divorced" might receive eight votes between them in EU-27. By submerging them within the weighted voting of the Colegii system these differential advantages would become less significant -- but not be eliminated altogether.
(9) Torsten Peters. "Voting Power after the Enlargement and Options for Decision Making in the European Union," paper presented at the ECPR Joint Session of Workshops, Oslo, April 1995. For further discussion of these competing indices of "voting power," see Geoffrey Garrett and George Tsebelis, "An Institutional Critique of Intergovernmentalism," International Organization, Vol. 50, No. 2 (Spring 1996), pp. 269-299 and Madelaine Holsti, "Coalitions and Power: Effects of QMV on the Council of the EU," Journal of Common Market Studies, Vol. 34, No. 2 (June 1996), pp. 255-274. As dazzling and "scientific" as these may seem, they are only as good as their initial premises -- which, to us, do not seem realistic. Much more promising is the approach taken by Kirman and Widgrén which attempts to put some substantive content into the simulations: "European economic decision-making policy: progress or paralysis?" Economic Policy, No. 21 (October 1995), pp. 423-460. For a knowledgeable critique of rational choice simplifications applied to Council voting, see the article cited in FN 3 by Fiona Hayes-Renshaw and Helen Wallace.
(10) Table III shows the authors' estimated weights. For another estimation, which excludes Malta and Cyprus, see Richard E. Baldwin. Towards an Integrated Europe (London: CEPR, 1992), pp. 186-187. Baldwin gives the ten CEECs 47 votes in an EU-25 of 134 votes. The differences seem to be due to the population figures used. For example, if Bulgaria were to reach a population of 9 million, it would obtain one more vote in the Council. Also, Baldwin is systematically more generous than we are in assigning votes to the three Baltic Republics.
(11) One might refer to this as "the Lampedusa Ploy" in honor of the author of Il Gattopardo who advised his fellow conservative Sicilians that "if things are going to remain the same around here, they are going to have to change." The trick for the EU, faced with enlargement, will be how to change its decision-making rules so that the present principle of equilibrium by size remains unchanged.
(12) It should be noted that there would be no necessity for the Colegii to meet formally and separately in order to tally the vote, although it would not be surprising if a certain amount of informal caucusing took place among large, medium and small countries. It also goes without saying that nothing in this scheme would prevent larger and medium-size countries from seeking to influence the voting of their smaller brethren in Colegii II & III. What the device does do, however, is to guarantee the relative bargaining power of the smaller countries in these informal log-rolls and trade-offs.
(13) Votes and seats have been rounded up in all tables. This explains why there are only 85 votes and 624 seats in Table IV and why, in the following tables, the totals do not always add up to 100 (for votes ) or 700 (for seats). Also, in the Council of Ministers, when the very smallest countries would be mathematically entitled to less than one half a vote, their weights have been increased to one.
(14) Except for the case of Luxembourg, whose diminutive size would naturally limit its coalition possibilities.
(15) Negotiations were scheduled to start six months after the conclusion of the 1996 IGC on institutional reform. However, the IGC will surely finish well into 1997, which would carry the opening of negotiations into January 1998 at the earliest. Moreover, the recent national elections in Malta seem to have resulted in the withdrawal of its application for membership. The Cypriot bid has long been stalled over the splitting of the island into Turkish and Greek parts, and a recent recrudescence of violent conflict would seem to have prolonged any serious consideration of entry. Nevertheless and however unlikely its eventuality now seems, we have gone ahead simulated the impact of their joint entry into the EU.
(16) Due to the rounding up of figures, Austria would maintain its 4 votes. However, given that in future enlargements, Austria would be corrected to 3 votes and sent down to Colegio III, we propose: (1) to maintain it in Colegio II but decrease its weight to 3; (2) sent it down to Colegio III while maintaining its weight at 4, or (3) maintain its 4 votes and let it stay in Colegio II, but raise the veto threshold to 31%. Any of these three modifications would void the possibility that Colegio II alone would be able to veto EU legislation.
(17) They might also choose to limit the total number of seats in the European Parliament to, say, 700 in order to facilitate discussion and internal procedures. For a similar suggestion, see Mathias Dewatripont et al., Flexible Integration: Towards a More Effective and Democratic Europe (London: CEPR, 1995), p. 172. It should be pointed out that the authors of this collective report were predominantly economists and their concern was much more with effectiveness than democracy.
(18) Actually, it would be more democratic to assign votes, seats and Colegio membership according to the size of their respective citizenries. For most members, this would not change much since the criteria for eligibility for citizenship in Europe are rather standard these days, but it would have the effect of penalizing those countries that discriminate against "denizens" and have in their midst a substantial proportion of residents who are not eligible to vote or stand for office. Countries such as Luxembourg and Switzerland with their mercenary proletariats of "guest-workers" and those such as the Baltic Republics that discriminate against resident Russians, Ukrainians and Belorussians would be especially affected -- and for a good democratic reason.
(19) It might eventually be necessary to modify this rule if the mini-states of Europe became serious about joining the EU. Just imagine the effect of Andorra, Monaco, San Marino, the Vatican, the Isle of Man, the Faröes Islands and Liechtenstein on Colegio III!
(20) Though no exact dates have been given for entry, these four countries are often assumed to be in the best position for early entrance. Since December 1994 Essen's European Council, all the Central and Eastern European Countries (CEECs) enjoying association agreements with the European Union are placed on a 'pre-accession' path consisting of a 'structured dialogue' with the EU, a free trade area and legal convergence. At the Madrid Meeting of the European Council (December 1995), it was agreed that the 15 would try to make CEECs' entry negotiations coincide with those dealing with Malta and Cyprus. Thus, the start date would be equal for all and the accession date will depend on economic criteria. The general principles agreed upon at the European Council were: (1) non-discrimination; (2) early date for political entry; and (3) long-transitional periods (see: Commission of the European Communities. European Council -Madrid 15 & 16 December 1995- Presidency Conclusions, Brussels, 16 December 1995, SI 95 1000).
(21) For an imaginative effort at simulating how different structurally determined preferences might affect voting on external trade and budgetary issues in an enlarged EU, see the article by Alan Kirman and Mika Widgrén cited in FN 8.
(22) Bosnia-Herzegovina has not been included due to the difficulty in determining the population and area that might define that polity in the future. Belorussia, the Ukraine, Moldova, Russia and the other former republics and units of the USSR have been considered "beyond the pale" of all possible future Europes. Recently, the heads of six "center-right" or "Peoples" parties (Belgium, Germany, Ireland, Italy, Luxembourg and Spain) announced that Turkey would never be admitted to the EU. Financial Times (5 March 1997), p. 2.
(23) Several proposals in this sense have been advanced in a longer manuscript of which this is an appendix. There, the senior author argues that the formula of concurrent majorities can also be applied to two other contentious areas of EU decision-making: (1) the selection of Commissioners (by giving each Colegio collectively the right to nominate 1/3 of them); and (2) the composition of the Presidency of the Council of Ministers (by establishing a three person collective executive for a fixed term with each of the chief executives chosen by one of the Colegii).