Andrew Duff -  On Governing Europe

Over the years, step by step, the European Parliament has won a share of real constitutional power. At times, as in 1984 with the Spinelli Draft Treaty and in 2002-03 in the Convention on the Future of Europe, Parliament has had a decisive influence on the constitutive development of the European Union. At other times, MEPs have found it just as difficult as the European Council to make constitutional sense of a Union which comprises both federal and confederal elements. EU governance is weak because its institutions are unable to manage the dichotomy between supranational and intergovernmental.

Today, circumstances have thrown the European Parliament a golden opportunity to take a major step in the federal direction – but it looks as though MEPs are going to retreat again.

As long ago as 1951, the Treaty of Paris gave the nascent European Parliament the job of drawing up proposals for election by direct universal suffrage in accordance with a uniform procedure. After much grunting and groaning, the first elections were held in 1979. But the electoral procedure was not uniform. A degree of uniformity was introduced in 1999 when, thanks to Tony Blair, all fifteen states held elections according to one or other system of proportional representation. But the elections still remain more national than European: the candidates are selected by national political parties, campaign on national platforms on mainly national issues, and are elected according to different (albeit proportional) national procedures. Although confederations of national political parties have been created at EU level, these organisations are not real political parties: they have next to no influence on the outcome of the European elections; they are actually prohibited from campaigning directly inside member states; and their relationship with the party political groups in the Parliament remains casual and tenuous. Undeceived by these Potemkin type European parties, the media continues to focus on national politics and does a disservice to any citizen elector wishing to vote on European issues.

Seat apportionment

The Treaty of Lisbon, bless it, changed the designation of MEPs from being representative of “the peoples of the States brought together in the Community” into being “representatives of the Union’s citizens”. Lisbon maintained Parliament’s right to initiate a uniform electoral procedure (Article 223(1) TFEU). But it also vested the Parliament with a new power to propose the reapportionment of seats in the House to reflect demographic change as well as to cater for the accession or secession of member states (Article 14(2) TEU). The divvying up of parliamentary seats between states had always been subject to unseemly bartering, usually in the early hours of the morning at the close of some fractious intergovernmental conference called to amend the treaties. Several in the European Council at Lisbon knew they were gifting the Parliament a poisoned chalice – especially as the treaty added the rejoinder that the “representation of citizens shall be degressively proportional” with a minimum of six seats per state and a maximum of 96.

Needless to say, the apportionment of seats in the Parliament has never yet met the criterion of the federalist principle of degressive proportionality. The present composition of the Parliament elected in 2014 is in serious breach of the treaty. Degressive proportionality means that MEPs from larger states should represent more people than MEPs from smaller states, and, conversely, that more populous states should have more MEPs than less populous ones. As things stand today, however, French, British and Spanish MEPs represent more folk than German MEPs; Dutch MEPs represent more than Romanian; Swedish and Austrians more than Hungarian; Danes more than Bulgarian; and Irish more than Slovak.

After a bruising row about the number of seats to be afforded Croatia when it joined up in 2013, the European Council took a decision, in accordance with Parliament’s requests and with its consent, to try to reach agreement on “establishing a system which in future will make it possible, before each fresh election to the European Parliament, to allocate the seats between Member States in an objective, fair, durable and transparent way”. In order to have the formula operative in time for the 2019 elections, Parliament was asked to come up with proposals before the end of 2016 – a target which it signally failed to meet.[1]

The arithmetical formula needs to be as straightforward and intelligible as possible.

Parliament commissioned leading mathematicians in the field who came up with what we know as the Cambridge Compromise (CamCom) to meet these criteria. A base of five seats would be given to each state, and additional seats would be accorded in a ratio proportionate to total resident population, rounded up. Because the immediate application of CamCom would have ended the current over-representation of the middling size states, it was not possible for a majority to be found in the House for this proposal, or indeed for any other more complicated and less transparent version.

Transnational lists

Parallel to the debate about seat apportionment support has been growing for the federalist proposal that a certain number of MEPs be elected for a pan-European constituency from transnational party lists. The purpose is to breathe life into the European dimension of electoral politics and thereby strengthen the democratic legitimacy of the Parliament. The reform would pitch the EU-level parties into competition with each other for ideas, votes and seats. It would electrify the European Parliamentary election campaigns and stimulate increased turnout. It would satisfy the treaty objective of a uniform electoral procedure and give practical expression to the concept of an MEP as representative of all EU citizens. It would give each citizen a second vote in the election, additional to his or her traditional vote for candidates standing in national or regional constituencies: a tangible manifestation of EU citizenship. The installation of transnational lists would give real meaning and direction to the role of Spitzenkandidaten, controversially introduced by partial experiment in 2014, whereby the EU parties select champions for the campaign.

Although the European Parliament has backed the concept of transnational lists in theory, in practice its steps towards their introduction have been faltering.

Electoral reform is always difficult, of course, because serving parliamentarians risk losing their seats. Moreover, conservative forces in the current Parliament are stronger than the progressive, and the rise of the nationalists seems to have intimidated the federalists. But in June 2014 the European Council committed itself to revisiting the method of choosing the next Commission President, as the treaty enjoins, “taking into account the elections to the European Parliament and after having held the appropriate consultations” (Article 17(7) TEU). A decent proposal for transnational lists headed by Spitzenkandidaten would settle the matter satisfactorily. Fully worked-out proposals to adjust the statute governing elections to the Parliament have been published already.[2]

Brexit

One of the more interesting aspects of Brexit is to consider what can be achieved by the EU without the British that it has not been able to achieve with the British. Electoral reform is a very good example. British ministers in the Council were without exception opposed to the federalist concept of transnational lists, as was a very large majority of British MEPs. The departure of the UK will not only remove these political obstacles but will also leave vacant 73 seats in the next Parliament. Those empty seats can be put to good use.

Seizing the initiative, Emmanuel Macron, true to his manifesto, has taken up the proposal for a transnational European constituency at the level of the European Council.

The Italian government, with its long-serving Minister of Europe Sandro Gozi, launched the matter in the first place. The Belgian and Spanish governments are also in favour. Martin Schulz for the SPD, along with the German Liberals and Greens, back the plan. In the Parliament, the federalist cause is led by the Liberal leader (and Brexit coordinator) Guy Verhofstadt. He won a vote in November 2015 to include a reference to a “joint constituency” in the Parliament’s most recent report on electoral reform (which otherwise merely tinkered annoyingly on the edges of the problem).[3] But now, just at this moment when the time appears to be ripe for radical change, MEPs are in danger of losing the plot.

The most recent draft report on the matter from the Constitutional Affairs Committee disingenuously appears not to accept that the UK is leaving the European Union before the next elections in May 2019. The co-rapporteurs, Danuta Hübner (EPP) and Pedro Silva Pereira (S&D), use the excuse of uncertainty surrounding Brexit to put off until 2024 the two critical decisions about the formula for seat apportionment and the introduction of transnational lists. Instead, they would merely use 51 of the UK’s 73 vacant seats to reduce (but not eliminate) the abuse of the principle of degressive proportionality.[4] This adjustment to the distribution of seats among member states would take place at some unspecified date after the actual 2019 election. Such prevarication amounts to a weak political fix that, if adopted, will put the Parliament in disrepute – and even at risk of being found by the European Court of Justice of having failed to act correctly. One hopes that wiser and bolder MEPs will amend the draft report appropriately. And British MEPs may choose to abstain from voting on this one.

As President Macron has identified, the fact is that the 73 ex-British seats gives the Union both the perfect chance and ample scope not only to introduce the arithmetical formula but also to install a transnational list of, say, 25 MEPs, as well as reducing the overall size of the House to save money. The European Parliament should be pushing forward on all three fronts. MEPs should maintain their earlier decision to back the pan-European constituency, and neither they nor the Council can dodge the need to apportion seats in advance of the 2019 elections. The European Council and Parliament together can act constructively to consolidate the Spitzenkandidat exercise.

In the end, these reforms need consensus in the Council as well as the consent of MEPs, but starting the process is the job of Parliament. Not to trigger urgent negotiations with the Council on this democratic package will cast serious doubt on the European Parliament’s constitutional purpose.

 

————————

This table shows how, post-Brexit, a transnational list of 25 MEPs could be introduced in 2019 along with the introduction of the arithmetical formula (Base + Power Compromise model (using 0,8 as the power). The overall size of the House is reduced from 751 to 735. Degressive proportionality is respected:

European Parliament Transnational List, post-Brexit.

*****
[1] European Council Decision No 2013/312/EU of 28 June 2013.

[2] The 1976 Act introducing direct elections has the status of primary law of the Union. For its revision both with respect to transnational lists and to seat apportionment, see The Spinelli Group, A Fundamental Law of the European Union, Bertelsmann Stiftung, 2013, pp. 273-280. For a description of CamCom, see Andrew Duff, Friedrich Pukelsheim & Kai-Friederike Oelbermann, The Electoral reform of the European Parliament: composition, procedure and legitimacy, European Parliament AFCO Paper, 2015, PE 510.002.

[3] Resolution of 11 November 2015 on the reform of the electoral law of the EU, P8_TA-PROV(2015)0395.

[4] Draft Report on the composition of the European Parliament, PE608.038, 7 September 2017. Sweden’s MEPs will still represent more people than those from smaller Hungary.

Tweet about this on TwitterShare on Facebook111Share on Google+0Share on LinkedIn11
Author :
Print

Leave a Reply