Launched at the joint initiative of the European Parliament, the Council of the European Union (hereinafter: Council) and the European Commission (hereinafter: Commission), the discourse on the future of Europe provides an opportunity to rethink the Community’s most fundamental decision-making mechanisms for more efficient and equitable functioning. The series of events, which include a number of events and debates, have high priority objectives: strengthening social fairness and equality and the democratic foundations of the European Union. Hungary, which has held the presidency of the Visegrad Group since July 2021, has a special responsibility to come up with new ideas and points of agreement that will lead to closer cooperation of the countries in the Group and more effective representation of the region. This document proposes a reform of the way voting rules are defined in the Council, in order to reduce the regional disparities that have developed over the last fifteen years and will deepen in the coming decades if the status quo is maintained.
A qualified majority is required for a Council decision (with a few exceptions), the definition of which has changed several times in recent decades and which, due to the weight of Council decisions, generates significant debates from time to time. Before the Treaty of Lisbon, countries had different numbers of votes (or country weights) that were agreed in advance, and based on these, the critical requirement for decisions was typically to reach a certain number of votes.
The Treaty of Lisbon broke with country weights and introduced the so-called “double majority” logic in which the Council decides when the number of supporting countries and the size of their population exceed pre-defined thresholds. The new voting rules of the Treaty of Lisbon have significantly increased the influence of the large Member States in Western Europe and (although to a significantly lower overall extent) of the smallest ones and reduced the decision-making power of medium-sized, typically Central European countries. The change affected the Visegrad Group particularly dramatically (Poland, Hungary and the Czech Republic are the biggest losers in the new regulation), with their combined influence falling from 17% to 13%. In addition, Brexit and unfavorable demographic trends are further widening the gap; the countries of the region are expected to see a further reduction in influence (by several percent) in the coming decades, provided the rules of the Treaty of Lisbon are maintained. More moderate influence reduces the region’s negotiating power, which can indirectly have significant negative welfare effects on its inhabitants, for example through distortions in the allocation of EU funds between Member States.
In order to correct the differences in power, this document proposes a reform of the Council’s voting rules according to the so-called Jagiellonian Compromise, previously put forward by Polish researchers. The recommendation would reintroduce country weights and determine them on the basis of the square root of the population of the Member States. The implementation of the proposal would increase the decision-making capacity of the Community (with an increase in the number of potentially decisive coalition variations) by reducing current inequalities in levels of influence at the level of European citizens, Member States and Member State interest groups.
The European Union’s current decision-making mechanisms are set out in the Treaty of Lisbon, signed on 13 December 2007. The signing and subsequent ratification of the Treaty by the Member States were preceded by a long journey, which began with the European constitutional process announced in 2001. This was because the European Union was facing the largest enlargement in its history, including the admission of ten countries, mostly former socialist countries in Central and Eastern Europe. The enlargement made it necessary to rethink institutional competencies and decision-making mechanisms and to make them more efficient, fair and flexible. The constitutional process ended in failure due to the resistance of some Member States (France and the Netherlands decided to reject it in a referendum), but most of the issues raised during it were incorporated into the Treaty of Lisbon. The decade and a half since the signing of the Treaty of Lisbon is still the most difficult period in the European Union’s existence. The global financial crisis, the migration crisis, and the coronavirus crisis have presented the Community with challenges for which it was not and could not be prepared.
The crises, and especially the responses to them, have brought to the fore unprecedented fault lines based partly on different interests and partly on different values.
These fault lines divide the Member States into at least three well-defined parts: the developed Western European core countries and the Nordic countries, the southern periphery countries with growing economic problems, and the eastern periphery Member States, which show marked differences in values. Of the latter, the V4 countries, with emphasis on Hungary and Poland, are increasingly engaging in open political and ideological debates and the conflicts that come with them to defend their interests and values. The latter are of the view that Europe’s strength and centuries-old influence at the global level have been based precisely on the diversity of the nations that live here, and they resist federalist efforts to eliminate and uniformise diversity and to centralise decision-making.
The lessons of Brexit are worth drawing for everyone. The leaving of the British is proof that the process of European integration could become reversible and fragile if minority opinions are systematically disregarded and majority positions are excessively imposed. It was in such a crucial and conflict-ridden period that the Conference on the Future of Europe, a series of dialogues, with the clear aim of strengthening the EU and enhancing its internal cohesion by listening to and integrating different views. Such a conference can only be effective if it allows everyone to express their views, whether they envisage the future of Europe in the form of a federation or a cooperation of nation states, and it is also important that there be an honest and democratic debate on the various views. Part of this is to reconsider and review the voting weights of the Member States within the Council, because in the long run it does not serve the strengthening and interests of the Community, if Member States or certain groups of them feel that changes in proportions put them at a disadvantage.
Proactive participation in the Conference on the Future of the European Union and the search for long-term strategies for the Visegrad Group to make a constructive contribution to solving the problems faced by the European Union are priorities in the programme of the Hungarian presidency of the Visegrad Group. The programme also emphasises that joined-up thinking can only work if it is combined with respecting the principles enshrined in the Treaties, equality between Member States, the constitutional framework of the nation states and the balance between the institutions. This document reviews the Council’s decision-making rules, analyses their effects on the influence of each Member States, and proposes the introduction of new rules that will lead to more balanced and fairer decision-making.
Development of decision-making rules in the Council of the European Union
The Council of the European Union, alongside the European Parliament, is the EU’s main decision-making body, with Member States’ ministers deciding on legislative proposals prepared by the European Commission, coordinating countries’ policies, shaping the Community’s foreign and domestic policies and adopting the EU budget. The Council has no permanent members; the countries are represented by the ministers responsible for the policy topic of a given meeting (hence the forum was formerly called the Council of Ministers). The participants in the forums typically decide by qualified majority voting, the rules of which were originally laid down in the Rome Treaty, which has been amended twice so far, under the Treaty of Nice and the Treaty of Lisbon. Given the weight of Council decisions, the definition of “qualified majority” has from time to time caused significant conflicts between Member States. During the evolution of voting rules, the following four main expectations were formulated for the system:
1. Population proportion
The population of the Member States is different, so the ministers attending Council meetings represent a different number of EU citizens. It is a general expectation that a proposal that is rejected by members representing a large proportion of the Union’s population is not accepted.
2. Member State sovereignty
The European Union is based on cooperation between sovereign countries, so it is a natural expectation on the part of the Member States to be able to assert their national positions in Council decisions. It is therefore appropriate to establish a system in which all Member States can influence decisions.
3. Decision-making power
In order to ensure the EU’s adaptability and rapid response time, it is worth introducing voting rules in which decisions can be taken reasonably and effectively.
It is worth using a formalised voting mechanism that can automatically deal with changes in the members of the European Union and demographic trends. Otherwise, the rules will have to be renegotiated
with each entry and exit, resulting in further conflicts.
The main difficulty in designing voting rules is the contradiction between the first three expectations.
Population proportionality and the sovereignty of the Member States necessarily conflict with each other, since if the former applies, the influence of the Member States with large populations will increase to the detriment of the smaller ones, and vice versa. And decision-making means broadening the interpretation of “majority”, as the more potential coalitions of Member States that cross the majority line there are, the more issues can be decided. The increasing number of decisions due to the lowering of the majority limit could, however, conflict with the interests of more and more people of more and more countries; the third expectation, therefore, limits the first two. And while the condition related to automated country weights does not cause any controversy, it makes it more complicated to design the right mechanism.
The rules for the operation of the Council were originally laid down in the Rome Treaty, which assigned the country weights shown in Table 1. below to each Member State. This meant that each Member State could cast one vote, but the value of the vote was more or less according to country weight. The document defined qualified majority in 12 units (i.e. 71% of the weights). For issues not discussed by the Council on a proposal from the Commission, four supporting Member States were needed for decision-making in addition to the 12 units.
At the time of the Rome Treaty, due to the small number of Member States, there was little problem of quorum, but the rules were heavily criticised for all three further expectations. The last two columns in the table show that the country weights reflected the population of each Member State to a limited extent: for example, although Germany had eight times the population of Belgium, it received only twice as many units. Despite the fact that Luxembourg did exceptionally well in terms of population, the majority requirement harmed the country’s sovereignty the most, as there was no coalition that would have needed Luxembourg to adopt the Commission’s proposals (i.e. to acquire 12 units); this country became, in the case of certain votes, weightless. Finally, in the absence of a formalised vote-determining logic, country weights had to be renegotiated with each new entrant, so the expectation of automatism was not met either. As the Community grew, the maximum number of votes increased, but the qualified majority ratio remained almost unchanged: It ranged from 70% to 72%.
The first major reform of the voting system took place in the Treaty of Nice, during which two different approaches emerged. One argued in favor of maintaining and expanding the existing system based on country weights, which meant renegotiating voting weights and changing majority thresholds. The other proposal argued for the abandonment of country weights and the application of a new so-called “double majority” logic, which tied decisions to the joint support of a predetermined number of Member States and the population they represent. Eventually, the Treaty included a combination of the two proposals; so, a Council decision required that the following three criteria are simultaneously met:
- Those in favour had to obtain 74% of the units based on the redefined country weights.
- Most Member States were required to support the proposal.
- The population represented by supporters had to reach 62% of the population of the European Union.
The country weights set out in the Treaty of Nice (and accepted by all Member States) and the new 62% rule resulted in a more balanced situation in terms of population proportionality, in which sovereignty improved, in addition to the majority criterion. The price of this, however, was a drastic drop in decision-making power. The most common measure in literature to measure the decision- making effectiveness of a Community is the action power index proposed by Coleman (1971), which shows the ratio of the number of potentially decision-making coalitions1 to the number of all possible coalitions. The value of the index exceeded 13% when the Rome Treaty was made, then its value gradually fell below 8% with new entrants, and then fell to 2% as a result of the Treaty of Nice. In addition, the new structure still did not offer an automatic solution to weight changes due to new entrants (or potential exits). The low level of decision-making efficiency and the lack of automatism prompted the European Community to further reform the Council’s voting rules, the framework of which was set out in the Treaty of Lisbon, which is still in force. The new approach broke with previous country weights and built its voting rules entirely on the “double majority” logic. Under the Treaty of Lisbon, therefore, a decision must meet all the following conditions2:
- At least 55% of Member States must support a proposal.
- Those supporting it must represent at least 65% of the EU population.
- A maximum of three Member States may oppose the proposal.
The new rules significantly increased the Council’s decision-making efficiency (or its action power index) to 13%, resolved the previous problem of automatism and also satisfied the expectation of proportionality.
However, the price of this was a serious violation of Member State sovereignty: the abolition of country weights caused a shift in the European Union’s balance of power, overturning the previous system, balanced in terms of Member State influence, and allowing some countries to gain significant influence to the detriment of others.
Development of national and regional influences
Due to conflicting expectations of the Council’s voting system, it is difficult to determine which decision-making rules can be considered fair. Perhaps the best starting point may be the country weights laid down in the Treaty of Nice, as this was the last time that the Member States had jointly agreed on a level of distribution of influence that was acceptable to all of them. As country weights were, however, removed from the Treaty of Lisbon, a new indicator applicable to both sets of rules should be used to compare the two periods. This paper uses the relative Banzhaf index (1965), which is widely used in the literature to measure the capacity of individual Member States to influence decisions, and shows in percentage terms how often each player plays a critical role in potential coalitions for majority decision-making3.
Of the decision rules in the Treaty of Nice, the strongest criterion was the 72% of country weights; the population requirement that additionally emerged did, therefore, only minimally increase the influence of the large Member States. All Member States could, therefore, have a say in the Council’s decisions to approximately the same extent as the Parties had agreed in advance. Thus, if the agreement on country weights can be considered fair, the system deviated only minimally from the ideal. Figure 1 demonstrates the relative country weights in the Treaty of Nice (i.e. per Member State out of the total number of votes allocated) as well as the Member State Banzhaf indices calculated on the basis of multi-component voting rules.
It can be seen that the difference between the columns is relatively small and that the influence of large population countries only slightly exceeds their relative weight. In the Treaty of Lisbon, uniformly accepted country weights were replaced by a pure “double majority” logic, which significantly rearranged powers. The system favored the most and least populous countries due to population and Member State majority requirements, and as groups of countries reaching 65% of the population mostly reach the 55% threshold, the real winners of the new system, in terms of influence, were the large Member States.
Medium-sized countries have lost the most in the new regulation, whose decision-making power has previously been offset by country weights, sacrificed to increasing action power.
Figure 2 above shows the percentage change in the Banzhaf index in each Member State as a result of the Treaty of Lisbon.
With the Treaty of Lisbon, drafted under the German presidency, Germany’s influence has increased the most, from 7.8% to 11.3% by Banzhaf, corresponding to a relative increase of 45%. In addition, the increase in the significance of the Member State threshold had a positive effect on Malta, Slovenia and Latvia (countries with small populations), whose influence has increased by 67%, 42.4% and 41.6%. In contrast, the biggest loser in the new regulation was Poland, with a Banzhaf index reducing by 28%, from 7.4% (similar to that of the largest Member States) to 5.4%, hardly half of their influence after the reform. The influence of Hungary and the Czech Republic decreased by 26% and 25%, respectively4. It is striking that the Treaty of Lisbon has broken balanced regional relations. While in the past, the decision-making ability in the Council was almost evenly distributed between the northern5, western6, eastern7 and southern8 regions which often represented shared interests and values, the new regulation made a significant part of the influence of the eastern bloc (about 4%) go to the western bloc9. In addition, as shown in Figure 3, the gap between the two regions is expected to widen further as a result of demographic trends.
The trend is particularly detrimental to the Visegrad Group, whose influence has fallen from 17% to 13% and is expected to decline by a further 2% percent in the coming decades if existing regulations are maintained. It is important to note that the latest available literature measuring the influence of Member States using the Banzhaf index, and thus the basis for this study, does not yet include the UK’s exit and Croatia’s admission, so the changes in influence caused by these two events are not included in Figures 3, 4 and 5. Research using other methodological approaches (see, for example, Kóczy, 2021), however, addresses the former effects and results show that the influence of the Northern region, weakened by the British exit, was largely transferred to the Western region, while the Croatian entry did not cause a significant change in the influence of each region.
As a result of growing inequalities between regions, Eastern Europe (and the Visegrad Group in particular) is increasingly unable to assert its own interests in the Council’s policy, budgetary and other decisions, which, in addition to distorting decisions, may adversely affect the well-being of the Eastern bloc (for example through reducing budget payments).
It is therefore worth rethinking the voting rules in the Treaty of Lisbon so that the region can again represent its interests with due weight.
Proposal to reform the Council’s voting rules
The lesson of the power changes brought about by the Treaty of Lisbon is that country weights should be reintroduced to a fair voting system that allows the sovereignty of the Member States and the balanced influence of the regions of the European Community. The main argument for a “double majority”, i.e. that this is the only way to ensure population proportionality, decision-making and automatism, is wrong, because there is a formalised method for determining the votes of the countries that ensures that all expectations are met more effectively. The reform proposal came from two Polish researchers at the Jagiellonian University, so it became known as the Jagiellonian Compromise in the literature. The recommendation of Wojciech Słomczyński and Karol Życzkowski (2013) is based on an early mathematical recognition, the Penrose (1946) square root rule, the essence of which is that in systems where democratically elected representatives from countries of different populations vote10 equality between citizens is best ensured if the number of votes of representatives is equal to the square root of the population of their respective countries. The reason for this, in short, is that a voter of a smaller country has a greater influence in elections in his/her country than a voter of a larger country, and the difference between the two is quadratic due to the growth dynamics of coalition combinations.
Differences in citizens’ decision-making power could, therefore, be compensated for by giving representatives the same weight in the higher level of voting as the square root of their country’s population. (The calculated country weights based on the rule are shown in Table 2).
The Jagiellonian Compromise fulfills the expectation of automatism because, based on the square root rule, demographic changes as well as new entrants and exits can be easily tracked. And whether the proposal meets the remaining three expectations better than the rules set out in the Treaty of Lisbon depends on two factors: the choice of the qualified majority threshold and the interpretation of population proportionality. For the threshold, Słomczyński and Życzkowski propose 61%, where decision-making efficiency (also known as the action power index) exceeds 17%, which is more than the 13% resulted by the Treaty of Lisbon. At this threshold, Member State sovereignty also applies better. Figure 4 shows how the new rules would change the influence of each country (Banzhaf indices).
The Member States’ Banzhaf indices under the proposal are typically between those calculated under the Treaty of Nice and the Treaty of Lisbon. The Jagiellonian Compromise corrects the population-related gap, reversing the lower influence of medium-sized countries. Country weights based on the square-root rule also improve regional balance: as shown in Figure 5, the Jagiellonian Compromise rebalances Europe’s major regions, and the Visegrad Group’s influence increases from 13% to 16%.
Thus, with a well-defined majority threshold, the Jagiellonian Compromise would improve the Council’s decision-making capacity, the sovereignty of the Member States and ensure a more equal representation at European citizen level. Although the new regulation would reduce the influence of large Member States, the criterion of population proportionality would apply through the application of the square root rule. The Jagiellonian Compromise approach would implement the Council’s „democracy of democracies” spirit and thus be more in line with the values of the European Union in political terms.
1 For a simple majority, the action power index is 50%.
2 Exceptions to the voting rules are procedural and administrative matters, where a simple majority (i.e. 14 Member States) is sufficient, and sensitive issues (such as foreign policy), the adoption of which requires the support of all Member States.
3 In addition to the Banzhaf index, there is another similarly popular indicator, the Shapley-Shubik (1954) index, which, although differing in mathematical detail, leads to very similar results to those presented.
4 It is important for the Visegrad Group that Slovakia’s decision-making capacity also decreased slightly, by 0.01%.
5 Austria, Denmark, United Kingdom, Estonia, Finland, Ireland, Latvia, Lithuania, Sweden
6 Belgium, France, Netherlands, Luxembourg, Germany
7 Bulgaria, Czech Republic, Poland, Hungary, Romania, Slovakia, Slovenia
8 Cyprus, Greece, Malta, Italy, Portugal, Spain
9 groups were selected based on Kóczy (2012), where Croatia was not yet included in the calculation.
10 Penrose originally studied the functioning of the United Nations General Assembly.
Banzhaf, J. F. (1965). Weighted voting doesn’t work: A mathematical analysis. Rutgers Law Review, 19, 317–343.
Coleman, J. S. (1971). Control of collectives and the power of a collectivity to act. In B. Lieberman (Ed.), Social Choice (pp. 192–225). Gordon and Breach. Eurostat. 2020. Population and social conditions. https://ec.europa.eu/eurostat/data/database? node_code=lan_use_ovw Last viewed on 15.08.2021.
Kóczy, L. Á. (2011). Lisbon outlooks. Közgazdasági Szemle [Hungarian Economic Review], 58 (10), 1045–1058. http://www.kszemle.hu/tartalom/cikk.php?id=1281
Kóczy, L. Á. (2012). Beyond Lisbon: Demographic trends and voting power in the European Union Council of Ministers. Mathematical Social Sciences, 63 (2), 152–158.
Kóczy, L. Á. (2021). Brexit and Power in the Council of the European Union. Games. 12 (2), 51. https://doi.org/10.3390/g12020051
Penrose, L. S. (1946). The elementary statistics of majority voting. Journal of the Royal Statistical Society, 109 (1), 53–57. http://www.jstor.org/stable/2981392
Shapley, L. S., & Shubik, M. (1954). A method for evaluating the distribution of power in a committee system. American Political Science Review, 48 (3), 787–792. http://www.jstor.org/stable/1951053
Słomczyński, W., & Życzkowski, K. (2013). Jagiellonian compromise - An alternative voting system for the council of the European union. Institutional Design and Voting Power in the European Union, September, 43–58.
Please find the full report here: