Degressive proportionality: EU enlargement will increase European electoral inequality – but the problem can be solved

Deutsch

By Manuel Müller

Disputes over national seat quotas in the European Parliament could become a problem for EU enlargement. How can this be avoided?

The allocation of national seat quotas in the European Parliament has long been a thorny issue. First, there is the question of “degressive proportionality”: According to Art. 14 (2) TEU, the Parliament is “composed of representatives of the Union’s citizens”, whose representation “shall be degressively proportional, with a minimum threshold of six members per Member State”. In practice, this means that larger member states have more seats than smaller ones, but smaller ones have more seats per population – an obvious problem for the transnational election equality. At the two extremes, a vote cast in Malta (6 seats per 0.4 million population) can have about 13 times the impact on the overall outcome of the election as a vote cast in Germany (96 seats per 83.2 million population). The inequality of voting power between constituencies in the European Parliament is therefore among the highest of any democratic parliament in the world.

Haggling over seat quotas every five years

Second, there are an infinite number of ways to organise a degressive-proportional allocation of seats – depending on how flat or steep the curve from the smallest to the largest member state is drawn. However, member states have never been able to agree on a fixed mathematical formula to operationalise the requirements of the treaty text.

Instead, political negotiations take place every five years to decide which country should be allocated how many seats in the European elections, in the light of changing population figures. This regularly leads to months of haggling. And since no country wants to lose any seats, the usual way out is to increase the total number of seats. In 2024, for example, the Parliament will grow from 705 to 720 MEPs.

With EU enlargement, countries will have to give up seats

This latest increase was possible because the UK’s withdrawal from the EU freed up seats that could be redistributed to other countries. However, this approach is not sustainable in the long term: Firstly, the Parliament is already approaching the maximum size of 751 seats provided for in the EU treaty – especially as MEPs also would like to use 28 seats for pan-European lists.

And secondly, the EU is set to welcome several new members in the coming years. Instead of 27, there could soon be 30, 35 or 40 EU countries. Even without pan-European lists, this will only be possible if the existing member states give up some of their seats in the European Parliament.

In search of a clear mathematical formula

In order to make this politically feasible, it makes sense to develop a clear mathematical formula that permanently regulates the allocation of national seat quotas in the European elections in a comprehensible and as fair as possible way. In fact, several expert proposals for such a formula have been on the table for years. The best known among them is the so-called Cambridge Compromise of 2011. In somewhat simplified terms, it suggested that each member state would get a fixed basic quota of five seats, and the remaining seats would be distributed proportionally according to population (rounding up, so that the smallest countries would get six seats).

However, implementing this proposal would result in a few large countries gaining and many medium-sized countries losing seats compared to the status quo. The Cambridge Compromise therefore proved politically unviable. As a result, various approaches were developed to modify it or to find alternative formulas – for example, using the square roots instead of the real populations of the member states. In a new legislative procedure, the European Parliament is currently working on a formula for degressive proportionality that is also politically acceptable.

Why use degressive proportionality at all?

But how can degressive proportionality be justified at all if it is obviously a violation of democratic electoral equality? A common argument is that degressive proportionality is a compromise between the “equality of citizens” and the “equality of states”, both of which are constitutive of the EU.

But this argument is not very convincing. According to the treaty, MEPs are not supposed to represent the member states (that’s the job of the Council of the EU), but the Union’s citizens. And also in practice, voting in the Parliament does not take place along national lines, but primarily along the lines of the transnational political groups.

Communicating to national party and media publics

However, degressive proportionality can also be functionally justified in terms of European party democracy. MEPs have not only a representative function, but also a communication function: They serve as a transmission belt between European and national parties as well as between the European parliamentary public and the national media public.

In this context, MEPs from smaller states face particular challenges. For example, national media in larger states have a wider reach, making it easier to reach many citizens at once. Moreover, MEPs from small countries also have a harder time to become visible within their own national parties: While in Germany there are 96 MEPs for 630 national MPs (a ratio of around 1:6.5), in Malta the ratio is 6 to 65 (around 1:11) and in Estonia 8 to 101 (around 1:12.5).

It is therefore a priori easier for MEPs in large member states to have an impact and to make European perspectives visible within their national party networks and public spheres. Degressive proportionality can thus be understood as a compromise between the democratic principle of electoral equality on the one hand and the particular challenges posed by the nationally fragmented party and media public spheres in the EU on the other hand.

Inequality will increase further with enlargement

But this compromise is always precarious and will be reaching the limits of its viability in view of the upcoming enlargements. Among the candidates for EU membership, there are a few large countries (Ukraine, and possibly Turkey in the future), one medium-sized country (Serbia) and many small ones (Georgia, Bosnia, Albania, Moldova, North Macedonia, Montenegro, as well as Kosovo at some point).

At the same time, the total number of seats should not exceed 751; after all, the European Parliament is already now one of the largest parliaments in the world. Logically, this can only be achieved in one of two ways: Either the minimum number of seats for the smallest countries is reduced – which is undesirable for the reasons given above. Or the large countries will have to give up more seats – but this will lead to more degressivity and exacerbate the lack of transnational electoral equality.

This in turn will jeopardise the public acceptance of the European Parliament, especially in the larger member states. Public voices in Germany already regularly complain about the lack of electoral equality and sometimes use this to deny the democratic legitimacy of the European Parliament. A further distortion could easily overstep the mark and become a threat to the acceptance of supranational parliamentarism itself.

Representation of member states and representation of parties

At first glance, this legitimacy dilemma looks like an insoluble problem, and indeed it is unlikely to be solved by a mathematical formula for degressive proportionality alone. But a solution is not completely impossible. Taking a closer look, the apparently opposing needs of the small and large member states are really about two different aspects: respectively, about the representation of the member states and the representation of the European parties (or political groups) in the European Parliament.

The smaller countries need representation of the member states (in the form of national seat quotas) to be degressively proportional in order to have a certain guaranteed minimum number of MEPs. In this respect, therefore, the electoral system is necessarily unequal. However, considering that in practice the Parliament works is not structured in national delegations but in transnational parliamentary groups, this unequal representation of member states is not decisive for the Parliament’s democratic legitimacy. What matters more is the representation of the European parties: Does the seat share held by the political groups in the European Parliament correspond to the vote share that their corresponding party families received in the European elections?

Decoupling state and party representation

In the current electoral system, state and party representation are structurally linked: Political groups whose member parties are elected mainly in small member states (and in member states with low voter turnout) are over-represented in the European Parliament, while political groups whose member parties are elected mainly in large member states (and in member states with high voter turnout) are under-represented.

As most political groups have member parties from both large and small countries, this effect is partially offset. Still, the remaining distortions can be significant: In the 2019 European election, the group with the “most expensive” seats (the Left) needed around 1.35 times as many votes per seat as the group with the “cheapest” (the EPP). In 2014, the distortions even led to the EPP becoming the largest group (and Jean-Claude Juncker becoming Commission president) when in fact the Socialists had won the most votes across Europe.

These problems of electoral equality could be solved by decoupling the representation of states and parties – in other words, by developing an electoral system that retains the degressive proportionality of national seat quotas but eliminates distortions between the vote and seat shares of the European political groups.

Model 1: The “tandem system”

Two models for such an electoral system are currently under discussion. The first is the so-called “tandem system”, developed by the mathematician Friedrich Pukelsheim and the former MEP Jo Leinen (SPD/PES), and based on the “biproportional apportionment” method used, for example, in the Swiss canton of Zurich.

In the tandem system, the number of seats to which each European party (or group) is entitled would first be determined using a standard proportional method on the basis of the votes cast for its member parties across Europe. An individual divisor would then be determined for each member state and for each European party. Put simply, the larger the member state, the higher the state divisor; and the more likely the party is to be elected in smaller countries, the higher the party divisor. Finally, to determine the number of the seats of its national party, its number of votes is divided by both its country divisor and its party divisor.

Side effect: Strong distortions at the national level

With the right divisors, it is thus possible to ensure that each European party (or group) receives the number of seats to which it is entitled, while national seat quotas are also respected. The tandem system thus skilfully combines the degressive proportional representation of the member states with the direct proportional (i.e., democratically equal) representation of the European parties.

However, the price for this is that in some cases there can be strong distortions between the vote and seat shares of parties at the national level: To put it somewhat hyperbolically, the tandem system would mean that the voters of the larger member states determine which parties from the smaller member states win seats in the European Parliament. This contradicts a central purpose of degressive proportionality, namely that the national party systems of the smaller member states should be adequately represented in the Parliament.

Model 2: Proportional compensation through EU-wide lists

The second model, which I myself have written about on several occasions, leaves national seat quotas untouched and relies instead on a European proportional compensation using EU-wide (“transnational”) lists. The basis for this is the introduction of an additional EU-wide seat quota, as has long been discussed and was formally proposed by the European Parliament in 2022.

However, in the proportional compensation model, these European seats would not be allocated separately from the national quotas. Instead, each EU-wide list would be allocated so many seats that the overall seat share of each political group (i.e. its seats from the national quotas plus its seats from the European quota) corresponds exactly to the share of votes that this group’s transnational list has received across Europe. A similar system of proportional compensation already exists, for example, in the elections to the Austrian national parliament.

Substantial number of European seats needed

Thus, the proportional compensation model also combines degressive proportionality in the representation of member states with EU-wide electoral equality with regard to the European parties. Moreover, as it leaves the allocation of national seat quotas unchanged, it has fewer side effects on the representation of national party systems than the tandem system.

However, full proportional representation through EU-wide lists will only work if the European seat quota is a substantial part of the total number of seats. Instead of the 28 seats currently proposed by the European Parliament, the number of European seats should be more in the region of 75 to 100 – which would of course further reduce the number of seats available for the national quotas. Whether the Parliament and the member states are prepared to introduce such a large EU-wide quota is ultimately a question of political will.

Reconciling degressive proportionality and electoral equality

Meanwhile, one thing is clear: With the forthcoming enlargement of the EU, the problem of degressive proportionality will become even more explosive than it is today. A further increase in voting inequality could jeopardise the democratic legitimacy and acceptance of the Parliament. At the same time, however, the arguments in favour of a significant minimum number of seats for small countries remain valid – simply reducing it, as has sometimes been proposed, would be the wrong answer to the problem.

In developing a permanent formula for the allocation of national seat quotas, MEPs and national governments should therefore not get too comfortable, but must find ways to reconcile degressive proportionality and electoral equality. In doing so, the tandem system and the model of EU-wide proportional compensation offer possibilities of a sustainable solution.

Next 14 February, the European Parliament’s Committee on Constitutional Affairs (AFCO) is organising an expert workshop on the planned permanent system for the allocation of seats in the European Parliament, which I will also be attending. Like all public committee meetings, the event will be web-streamed.

Picture: © European Union 2019 – Source: EP [CC BY 4.0], via Flickr.