Study on the Conduct of the 2014 Elections to the European Parliament

Transcription

1 EUROPEAN COMMISSION Directorate General for Justice Multiple Framework Contract on Evaluation & Evaluation Related Services Final Report Study on the Conduct of the 2014 Elections to the European Parliament 10 March 2015

2 Contents SECTION PAGE 1. Introduction Purpose of the Study 1.2 Summary of Research Activities 1.3 Structure of the Report 2. Background to the Research European elections and involvement of citizens 2.2 Trends in Voter Turnout in the 2014 European Parliament Elections 2.3 Comparison with National, Regional and Local Elections 2.4 Effects of the Economic Crisis on European Elections 2.5 Commission s Report on the 2009 European Parliament Elections 2.6 Commission s Recommendation on the Conduct of the Elections 3. Enhancing the Democratic Conduct of the European Elections Information on the affiliation between European and national parties provided by the Member States 3.2 Information on the affiliation of political parties with European parties provided by the political parties 3.3 Candidates for President of the European Commission 3.4 Role of Civil Society in promoting the European elections 3.5 Common Voting Day and Other Issues 3.6 Conclusions Enhancing Democratic Legitimacy of the Elections 4. Role of the Media in the 2014 European Elections Restrictions on freedom of expression during an election period 4.2 Media coverage of the 2014 European elections 4.3 Media coverage of EU issues in the European election campaign 4.4 Role of the Social Media in the Election Campaign 4.5 Conclusions Role of the Media in the 2014 European Elections 5. Enhancing the Efficient Conduct of the European Elections Single Point of Contact 5.2 Timely Transmission of Data 5.3 Additional Data allowing for more Efficient Voter Identification 5.4 Technical Means for Safe and Efficient Transmission of Data 5.5 Measures taken to Inform Citizens of the Union of their Right to Vote and Right to Stand as a Candidate 5.6 Non Nationals Registering to Vote Compared with Conclusions - Enhancing the Efficient Conduct of the Elections

3 Contents 6. Overall Conclusions and Recommendations Overall conclusions 6.2 Enhancing the Democratic Conduct of the European Elections 6.3 Media coverage of the 2014 European Elections 6.4 Enhancing the Efficient Conduct of the European Elections APPENDICES PAGE A. List of Interviews 99 B. Copy of Interview Checklists 101 C. Copy of Survey Questionnaire 104 D. Supporting Information 111

4 Executive Summary Below we provide a summary of the key conclusions and recommendations set out in the final report prepared by the Centre for Strategy & Evaluation Services (CSES) for the assignment for DG Justice: Request for services JUST/2013/JCIV/FW/0077/A4 - Study on the Conduct of the 2014 Elections to the European Parliament. The research was carried out in the second half of Purpose of the Study Following each election in the European Parliament, the European Commission issues a report 1 examining how EU citizens' rights were enforced and how the relevant EU rules were implemented by the Member States. Furthermore, it assesses EU citizens' participation and the overall conduct of the elections. The present study contributes to the collection of quantitative and qualitative data for the report to be issued by the Commission on the 2014 European Parliament elections. Ahead of the 2014 European Parliament elections, the Commission issued a Recommendation on enhancing the democratic and efficient conduct of the elections to the European Parliament (2013/142/EU) 2. This was addressed to the Member States and to European and national political parties and adopted on 12 March The Recommendation sought to help increase transparency and to promote pan-european campaigns around candidates and their programmes, as well to enhance the efficient conduct of the elections. Therefore, the present study also aims to evaluate the implementation of the Recommendation 2013/142/EU in the 2014 European Parliament elections by the Member States and by national and European political parties and to analyse, inter alia, the campaign activities of national and European parties in various media, as well as the national legislative framework in place for carrying out such actions in view of implementing Recommendation 2013/142/EU. In addition, the present study also evaluates the rules of access to media during the electoral campaign period, including the rules on the allocation of broadcasting time to political parties and the time for political advertisements and on restrictions that may be imposed on freedom of expression during an election period. Based on the above findings, the study proposes possible solutions and recommendations to further strengthen the European dimension of the European Parliament elections through new actions to be implemented by the next elections in Overall Conclusions There is generally positive feedback from the research on the implementation of the Commission s Recommendations and the effects on the conduct of the 2014 European elections. This is more clear-cut in relation to the Recommendations aimed at enhancing the efficient conduct of the elections than it is with the recommendations on enhancing the democratic conduct of the 2014 European elections, where a more mixed picture emerges. Nevertheless, improvements were noted both with regard to the democratic and the efficient conduct of these elections. 1 See, most recently, the report on the 2009 EP elections: COM(2010) 605 final. 2 Recommendation on enhancing the democratic and efficient conduct of the elections to the European Parliament (2013/142/EU) OJ L 79, , p. 29. i

5 Executive Summary Against the backdrop of the continuing economic difficulties affecting most EU Member States and the growing strength of parties that are hostile to EU membership, the fact that the long-term decline in voter turnout was more or less stabilised in the 2014 European elections can be viewed as a positive outcome. At the same time, there were a number of interesting new features to the campaign such as the personalisation of the elections through the nomination of "lead candidates" by European parties and the increased proximity and possibility of interaction between candidates and voters on issues of European relevance, in part due to the campaigning efforts of the lead candidates who participated in EU-wide political events and debates with considerable media coverage and to the growing use of social media. However, it is clear that the EU institutions, Member States and political parties still face a major challenge in overcoming voter apathy and ensuring that citizens are motivated to vote in future European elections. Turnout in the Member States that joined the EU after 2004 lagged behind that in the older Member States and almost everywhere, the discrepancy between recent national elections and the 2014 European elections remained large. Also, the number of EU non-nationals registering to vote in the Member State of residence continued to be disproportionately low compared with the number of EU citizens residing in another Member State than their own. Finally, as regards the efficient conduct of the elections, in general national authorities considered that the measures taken for the efficient conduct of the European elections reduced the administrative burden and had a positive impact in terms of transfer and exchange of data. Almost all Member States said that the Commission s recommendations on the administrative part substantially improved the functioning of the mechanism of data exchange and minimized abuses, making it more efficient, which in turn simplified the bureaucratic procedures for registering voters and candidates in EU states of residence regardless of EU country of origin. Moreover almost all Member States thought that the exchange system was better in comparison to Enhancing the Democratic Conduct of the 2014 European Elections Most EU Member States took action to implement the Commission s first recommendation to encourage and facilitate that political parties provide information to the electorate on the affiliation between national parties and European political parties. Just over half the Member States (16) took non-legislative measures (e.g. informed the national parties about the recommendations or issued written guidelines) to this effect. In some cases this was combined with legislative action, to allow national parties to mention the affiliation to a European party on the ballot paper. Six Member States did not take any measures at all in response to the Recommendation, on the ground that that it is the responsibility of the political parties rather than the national authorities to raise awareness of their European affiliations. Recommendation 1 Publicizing affiliations with European political parties: National authorities should do more to encourage and enable national political parties to make known to the electorate their affiliations with European political parties and citizens about the European elections and the reasons to vote. Most Member States informed their national parties about the Recommendations via various different channels. However, a few Member States argued that it is the task of political parties to inform the electorate about the European election process. ii

6 Executive Summary With regard to the Commission s second Recommendation on the role of political parties in making their affiliation with European parties known, the actions taken varied across parties and Member States. In 5 Member States, the main political parties took comprehensive steps such as prominently displaying their affiliations in their manifestos and in political broadcasts to make the electorate aware of their memberships of a European party. However, most national parties made known their affiliation with European parties in a limited manner, e.g. mainly in manifestos. The reasons for only taking limited action were mainly practical rather than ideological. For example, in the case of some national political parties, the decision to join a European party was only taken after the 2014 European elections. In other cases, campaign materials and other information had already been prepared when the Commission made its recommendation known. Finally, a small number of parties did not mention their affiliations at all. This was mainly the case with parties that are hostile to EU membership or because some MEPs are elected without belonging to a European party. Recommendation 2 - Deadlines for deciding on affiliations with European political parties: Consideration should be given to introducing a deadline for national parties to decide on their affiliation to a European party so as to reduce the uncertainty for voters. However, there needs to be some flexibility. Some national parties argued they could not highlight their affiliation to a European party in the 2014 election campaign because they did not know which European party they would join after the results were known. This was particularly the case for new and smaller parties. Recommendation 3: Use of ballot papers to indicate affiliations with European political parties: Steps should be taken to encourage national parties to mention their link to a European party on the ballot paper (if it is allowed according to electoral law in the respective Member State). According to the research, most national parties took some actions to mention their affiliation to a European party in their campaign material. However, almost no party mentioned the affiliation on the ballot paper which is an obvious way of raising awareness. In relation to the Commission s third Recommendation, 5 of the 13 European political parties put forward candidates for President of the European Commission and also made a substantial effort to promote their candidate. A wide variety of campaigning methods were used, e.g. a campaign trail, social media, events, debates and other traditional media campaigns such as posters. The candidates of the two largest European parties, the European Socialist Party and the European Peoples Party, campaigned respectively in 20 and 18 of the EU28 Member States, and the other candidates were also very active (in total there were 246 campaign visits across the EU Member States). 3 There were also 10 televised debates that were broadcasted involving most if not all of the candidates: five debates were designed as headto-head debates between the representatives of the two main European political parties. Four others were open to all nominated lead candidates while one French-language debate was held between the candidates of the European Green Party, the Alliance of Liberals and 3 The total number of 246 visits takes all cities into account that the candidates visited in each EU Member State. iii

7 Executive Summary the Democrats for Europe Party. At the national level the picture was rather fragmented. Some parties decided to fully advertise their Spitzenkandidat by mentioning them in all the different campaigning channels while other took no action at all. In general, it can be concluded that based on the evidence of the 2014 European elections, the system of Spitzenkandidaten is a promising tool for encouraging a higher turnout. Recommendation 4 Promoting candidate for the presidency of the European Commission: More should be done to encourage national parties to promote their European party s candidate for the presidency of the European Commission. Strong campaigning by the Spitzenkandidaten mostly took place when the candidate was from a national party in their country of origin or when they spoke the language of another Member State. However, national parties across the EU need to be made aware of the benefits of campaigning irrespectively of the nationality for the Spitzenkandidat. Aiming to enhance the democratic conduct of the European elections requires informing citizens and ensuring accessibility, namely to people with disabilities. It has been argued particularly by NGOs that not enough was done to grant access to information and to polling stations to people with disabilities. While for instance many parties tried to make their campaign material more accessible, the debates between Spitzenkandidaten (broadcast by the EBU) were not made accessible for people with disabilities (e.g. no provision was made for deaf people). Recommendation 5 Voters with disabilities: Steps should be taken at the national level to enhance accessibility for voters with disabilities. For example, it needs to be ensured that campaigning material is accessible for people with disabilities. In addition, in some countries access to polling stations for disabled persons needs to be improved. Overall, the research suggests that the Commission s Recommendations aimed at Enhancing the Democratic Conduct of the 2014 European Elections had a positive effect. Despite the economic downturn, there is evidence that the 2014 elections were more European in their tone than the previous elections in The role of the European parties in putting forward a candidate for the European Commission presidency was a particularly significant contribution to this. On the other hand, for various practical and in some cases ideological reasons, many political parties did not do enough to promote awareness of their affiliation to a European party. The research suggests that actions taken by European and national parties and the national authorities to enhance the democratic conduct of the elections, and their European nature, were not enough. In particular, the measures taken were only visible to citizens who were likely to vote in any case and not to others who were less inclined to do so. 4. Media Coverage of the 2014 European Elections In the 2014 European elections, the debate on the future of Europe was more intense than in earlier elections. In terms of the coverage of EU topics during the election period, a few observations can be made. First, the role of social media increased in 2014 in comparison to Second, the introduction of the Spitzenkandidaten provided a focus point for media coverage. Third, the financial and economic crisis was a key issue in most Member States and this was far more so in 2014 than in In some Member States, the financial crisis resulted in increasing the media s coverage of the elections. iv

8 Executive Summary Recommendation 6 Role of media in highlighting EU topics: The media needs to do more to highlight EU topics during the European elections campaign rather than focusing on national and local issues. Where possible, the media should be encouraged to have special TV programmes or newspaper sections on the European elections during the campaign period. In 2014, the social media became a much more important tool than it was in the 2009 elections. First, political parties and candidates used it as a campaigning tool. Second, traditional media and journalists used the social media to increase the reach of the information. Besides the increased importance of social media it is not possible to directly correlate social media activity with a higher turnout in the elections or to correlate social media popularity with the election results. However, the social media was important, in terms of offering politicians and voters enhanced possibilities to communicate directly with each other. Voters, particularly amongst the younger generation who tend to make greater use of social media, could more actively engage in the electoral debate. An issue that emerged was related to geo-blocking. Most TV stations make their programmes available on web stream. However, some of the TV programmes providing political coverage of the 2014 European elections were not available online if accessed from another country. This is problematic for voters who are not in their country of origin in the election period but who want to obtain coverage of the election from their home media. Recommendation 7 Role of the social media: A strong emphasis should be placed on using the social media in European elections. The research suggested that social media is increasingly being used in election campaigning. It is especially important in engaging the younger generation. Recommendation 8 TV coverage and the problem of geo-blocking: Consideration should be given to asking TV stations to lift geo-blocking for the duration of the European election period in order to make information available to voters irrespective of their location. Overall, feedback from the research suggests that the media had a positive impact on the European elections in 2014 by raising awareness of EU issues and encouraging citizens to vote. However, there was criticism that the media gave too much attention to national issues in some countries. An interesting feature of the 2014 European elections was the increased use by the European parties of the social media as a campaign tool. However, because the social media is largely unregulated, it was possible to bypass some rules on electioneering (e.g. on the deadline for campaigns to stop) that apply to more traditional media forms. As regards the legal framework in place in the different Member States determining access to the media, national rules vary across the EU Member States (e.g. the amount of time allocated to political parties is linked in some countries to their representation in the national parliaments while in other countries each party is given the same amount of time. Overall, however, there do not appear to have been any major issues concerning the legal framework of the principle of the freedom of expression or access to the media during the 2014 European elections. v

9 Executive Summary 5. Enhancing the Efficient Conduct of the 2014 European Elections In total, 27 out of 28 EU Member States welcomed and implemented the single contact authority recommendation. The United Kingdom was the only Member State that did not introduce a single point of contact for transmitting information regarding EU nationals who had registered to vote. Almost all Member States agreed that setting a single point of contact reduced the administrative burden on national authorities compared with previous European elections and made the exchange of data more efficient and effective. There was, however, some less positive feedback. In particular, it was argued that transferring the tasks of handling voter information from a number of different entities to a single point of contact threatened to overload the designated body. There were some other transitional complications for the designated authorities. Overall, however, the recommendation was welcomed and had beneficial effects. The Commission s 2013 Recommendation included a number of proposals designed to ensure that the exchange of data on voters was done in a way that was timely. In general, these recommendations were successfully implemented but there were complications. Although most Member States sent and received the data on time, the major concern was with the quality of the received data that was often insufficient to identify voters, rather than the timing of the sent/received data. There was a need to receive additional (noncompulsory data) to identify voters more efficiently and help prevent double voting but five EU Member States argued that their laws or practices did not allow them to release additional information. Further complications included the use of incompatible software, differences in the spelling of names and misspellings. Over time, these complications are likely to be resolved and generally the Commission s recommendations had beneficial effects. Recommendation 9 Additional non-compulsory data on voters: There is a need for most Member States to provide additional non-compulsory data to help identify voters more efficiently and to help prevent double voting. The Commission should work with Member States to extend the definition of compulsory information on voters to include key identifiers (e.g. date of birth). Recommendation 10 Harmonisation of rules on compulsory data: There is a need for some existing rules on compulsory data to be harmonised. For example, Member States should use 'surname' in the same way (e.g. the use of diacritics in Croatia and the Czech Republic). Recommendation 11 Harmonising the last day for registration on the electoral roll: It would make the sending and receiving of data more efficient if there is a harmonization of the last day for registration on the electoral roll for all Member States (or at least the last day for registration should be at least a month before the European elections to allow a sufficient margin for sending and processing the data). With regard to the Commission s recommendations on the technical means for safe and efficient transmission of data, almost all Member States improved the security of the data exchange taking place in the run up to the 2014 European elections, by implementing a common electronic format and using the suggested encryption tool. Most Member States used the recommended character set and there were no major problems with the XML files. Likewise, all Member States complied with the encryption method. However, there were some problems. vi

10 Executive Summary Most EU Member States undertook a campaign in the run-up to the 2014 European elections to raise awareness amongst non-nationals residing in their country of their right to vote and stand as a candidate in the European elections. In almost all EU Member States (22 of the 24 providing information), the number of non-nationals residing in the respective countries was significantly higher at the time of the 2014 European elections compared with The number of non-nationals registering to vote in the European elections was also higher in 2014 compared with However, the increase in the numbers registering to vote (an average of +5.2% compared with 2009) was far lower than the increase in the number of non-nationals residing in Member States (an average of +64.6% over the same period). As a result, despite the increase in the number of non-nationals registering to vote, the proportion that this represented of the total number of non-nationals residing in the various Member States actually declined from 12.2% in 2009 to 8.5% in Moreover, the proportion actually voting to the extent that can be judged from the small number of countries providing details - was even lower. Recommendation 12 Non-national voters: Member States should take additional steps to inform non-nationals of their rights and to encourage them to vote. The proportion of non-nationals registering to vote in their country of residence dropped in the 2014 European elections compared with 2009 (8.5% in 2014 compared with 12.2% in 2009) and the proportion actually voting to the extent that can be judged from the small number of countries providing details - was even lower. Additional steps need to be taken to reverse this trend and this could include measures to help raise awareness amongst non-nationals or their voting rights. Overall, national authorities considered that the measures taken to enhance the efficient conduct of the European elections had reduced the administrative burden and had a positive impact in terms of the transfer and exchange of data. Almost all Member States said that the Commission s recommendation contributed to substantially improving the functioning of the mechanism for preventing double voting, making the registering of nonnational voters more efficient, reducing the administrative burden and simplifying the procedures for registering voters and candidates in EU Member States of residence regardless of EU country of origin. However, some national authorities considered that a significant degree of electoral fraud involving double voting was a remote risk and that the administrative burden of the information exchange system was disproportionate to the danger it was intended to prevent. The following table provides a balance sheet summary of the positive (and less positive) effects associated with the Commission s Recommendations: vii

11 Executive Summary Overall positive and less positive findings Positive Less Positive Democratic conduct of the elections Some Member States issued guidelines to inform parties about the importance of mentioning the link between the national and EU parties. Nine Member States also allowed national parties to mention their affiliation to a EU party on the ballot paper. In this way, national political parties had the possibility to raise awareness of their link to European parties. Some national parties used their Spitzenkandidaten extensively in campaigns in order to benefit from a more personalised election. The campaigns helped to create a focus on European issues for the elections. There was a common voting day across almost all EU Member States. There were 7 exceptions but all countries voted within three days of each other. The majority of Member State authorities did not allow mentioning the affiliation between parties on the ballot paper. This was because of practical and legal reasons Almost no national party made use of the right to mention their affiliation to a European party on the ballot paper where this right existed. Eurosceptic parties increased their share of the vote and they placed little emphasis on publicizing their European affiliations through campaign material. There was not enough time for national parties to include an EU focus/a EP party affiliation/ the Spitzenkandidat in their campaign material because the Recommendations were communicated rather late. Some parties only stressed national topics because shortly after the election, national elections took place. Thus, the campaign was mainly focused on national issues. The media generally had a positive impact on the European elections in 2014 by raising awareness of EU issues and encouraging citizens to vote. Compared with 2009, the social media played a much bigger role in 2014, thereby enabling more voters to be actively engaged in the campaign thus beyond just voting. Social media will most likely play a stronger role in future elections since it is the preferred tool of a younger generation. Setting up a single point of contact helped improve the efficient exchange of data on non-nationals. Compared with 2009, the administrative burden was significantly reduced in The transmission of data was much more efficient compared with Almost all Member States confirmed that Role of the media In some Member States the media focused more on national aspects of the European elections than on EU-related topics. Although more online activity took place in the 2014 European elections, there is no direct correlation between turnout/decision making on who to vote for and social media activity. Efficient conduct of the elections In some cases, it was still difficult to identify voters from the lists sent from one Member State to another because not all the required information was provided to identify individuals. The lack of a harmonised list for registration makes it difficult to send and process the data. viii

12 Executive Summary the security of the data exchange was much better in There were no major problems with the XML and encryption methods. The transmission of additional data remained a significant problem (there seems to be a need amongst most Member States to receive additional non-compulsory data, to help them identify voters more efficiently and to help prevent double voting). There were some difficulties to comply with the XML standards and encryption suggestions because every Member State has a different system. ix

13 Introduction 1 This document contains the draft final report prepared by the Centre for Strategy & Evaluation Services (CSES) for the assignment for DG Justice: Request for services JUST/2013/JCIV/FW/0077/A4 - Study on the Conduct of the 2014 Elections to the European Parliament. 1.1 Purpose of the Study Following each election in the European Parliament, the European Commission issues a report 4 examining how EU citizens' rights were enforced and how the relevant EU rules were implemented by the Member States and assessing EU citizens' participation and the overall conduct of the elections. The present study contributes to the collection of quantitative and qualitative data for the report to be issued by the Commission on the 2014 European Parliament elections. Ahead of the 2014 European Parliament elections, the Commission issued a Recommendation on enhancing the democratic and efficient conduct of the elections to the European Parliament (2013/142/EU) 5. This was addressed to the Member States and to European and national political parties and adopted on 12 March The Recommendation sought to help increase transparency and to promote pan-european campaigns around candidates and their programmes, as well to enhance the efficient conduct of the elections. Therefore, the present study also aims to evaluate the implementation of the Recommendation 2013/142/EU in the 2014 European Parliament elections by the Member States and by national and European political parties and to analyse, inter alia, the campaign activities of national and European parties in various media, as well as the national legislative framework in place for carrying out such actions in view of implementing Recommendation 2013/142/EU. The present study was also required to evaluate the rules governing access to the media during the electoral campaign period, including the rules on the allocation of broadcasting time to political parties and the time for political advertisements and on restrictions that could be imposed on freedom of expression during an election period. Based on the above findings, the study was required to propose possible solutions and recommendations to further strengthen the European dimension of the European Parliament elections through new actions to be implemented for the next elections in Summary of Research Activities The assignment was carried out in three phases: Phase 1: Preparatory tasks kick-off meeting and follow-up interviews with DG Justice, desk research and finalisation of the research tools leading to an inception report which was submitted on 29 August and discussed at a meeting with DG Justice on 2 September 2014 Phase 2: Assessment of the conduct of the 2014 European Elections a survey and an interview programme covering the European political parties, MEPs and 4 See, most recently, the report on the 2009 EP elections: COM(2010) 605 final. 5 Recommendation on enhancing the democratic and efficient conduct of the elections to the European Parliament (2013/142/EU) OJ L 79, , p

14 Introduction 1 European Parliament officials, national authorities, the media and other stakeholders to investigate the key issues set out in the terms of reference. An interim report was submitted on 20 October and subsequently discussed at a meeting with DG Justice on 5 November Phase 3: Analysis and final report detailed evaluation of the Phase 2 research findings, preparation of a draft final report, and finalisation of the report. Following discussions with DG JUST at the kick-off meeting (6 August 2014), it was agreed that certain adjustments needed to be made to the research plan The following diagram summarises the updated work plan and the key milestones: Figure 1.1: Overview of work plan Phase 1 Preparatory Tasks At the outset of the assignment, we undertook a number of preparatory tasks: A kick-off meeting with the Steering Group and follow-up interviews; Desk research to review relevant background documentation and to identify and start collecting key sources for the Phase 2 assessment; Compiling a list of key stakeholder contacts and sources of information for the Phase 2 research; Finalisation of the research tools and work plan; Preparation of an inception report and review meeting with the Steering Group. These tasks were largely completed and led to an inception report being submitted to DG Justice on 29 August As noted above, a review meeting with DG Justice took place on 2 September

15 Introduction 1 Phase 2: Research in EU Member States Phase 2 of the assignment involved: Information provided by Member States - analysis of the responses to the questionnaire sent by the Commission to national authorities. On-line survey to obtain qualitative feedback on how effectively the Commission s recommendations were implemented, and their impact on the conduct of the 2014 European elections, from MEPs, European and national political groups, movements of citizens and other key stakeholders. Interview programme - this was mainly being conducted by telephone given the relatively short timescales. The interviews were used to investigate key issues in more depth. Desk research and analysis of electoral legislation and rules, and issues relating to media coverage in relation to the various key issues set out in the Commission s terms of reference. Below we summarise tables from the CSES tender on the scope of the Phase 2 fieldwork and summarise the progress to date: Table 1.1: Research coverage Target groups Number of responses Survey Interview Total European Commission, Parliament officials National authorities Members of the European Parliament European and national political groups/parties Media, movements of citizens Academics and other experts Others (not classified) Total Responses to the online survey were below target (60 compared with a target of 250), largely because the anticipated support from the European political parties in encouraging their national parties to complete the questionnaire was not forthcoming (the target of achieving 100 responses from national political parties compares with 17 actual responses). Also, we had aimed for a response from 100 MEPs (this compares with 4 actual responses). In both cases, we were unable to persuade the European political groups to ask MEPs and national parties to participate in the survey and although we were able to develop our own lists of contacts, these were not comprehensive or numerous enough to generate a high level of survey responses. A total of 73 interviews were completed compared with a target of 81. Again, there were greater difficulties than anticipated in obtaining the agreement of political parties to participate in the interview programme. Nevertheless, the feedback obtained via the survey, interview programme and other sources (e.g. desk research) was sufficient for the research findings to be robust. 3

16 Introduction 1 Online survey The Phase 2 survey focused on the same range of key issues that were covered by the interview programme. The purpose of the survey was to allow those who could not participate in the interview programme with CSES to nevertheless make an input to the research. The following chart provides a breakdown of the survey responses: There was a good distribution of the survey responses across the EU28 Member States with completed questionnaires from 24 of the EU28 Member States. The following tables provide a breakdown of the survey responses by country: Table 1.2: Breakdown of survey responses by country (9 December 2014) Member State No % Member State No % Austria Latvia Belgium Lithuania Bulgaria Luxembourg Cyprus Malta Croatia Netherlands Czech Rep Poland Denmark Portugal Estonia Romania Finland Slovakia France Slovenia Germany Spain Greece Sweden Hungary UK Ireland Other Italy Total

17 Introduction 1 The survey questions covered all the Commission s Recommendations but focused on asking for opinions on how effectively the various Recommendations had been implemented and their effect on the 2014 European elections rather than on collecting purely factual information. Phase 2 interview programme The Phase 2 interview programme was designed to investigate key issues in more depth. It covered the European Political groups, European Parliament officials, media representatives, national authorities and national political parties, and others (e.g. academic experts). We started by selecting the two parties from each Member State that have the most MEPs in the European Parliament. Two minor parties were then randomly selected from each Member State, giving a total of four national parties in each country. This provided 71% coverage of the parties represented in the EP (according to the number of MEPs). A total of 67 interviews were carried out with a variety of different organisations: European political party officials (5); national parties (15); national authorities (18); academics and other experts (9); and NGOs and other types of organisations (8). The following tables provide a breakdown by EU Member State: Table 1.3: Breakdown of interviews by country Member State No % Member State No % Austria Latvia Belgium Lithuania Bulgaria Luxembourg Cyprus Malta Croatia Netherlands Czech Rep Poland Denmark Portugal Estonia Romania Finland Slovakia France Slovenia Germany Spain Greece Sweden Hungary UK Ireland Other Italy Total Other research DG Justice asked Member States to provide information on the implementation of the various Recommendations. All Member States provided completed questionnaires. The quality of the information provided by Member States was rather varied. Most national authorities provided detailed answers to Question 1.1 and 1.2 (Entry in the electoral roll and turnout). Sufficient information was also provided to Question 2 (Right to stand as a candidate and application of Directive 2013/1/EU). However, there were gaps in the 5

18 Introduction 1 information provided to other questions. From a different perspective, the level of detail provided in the responses varied significantly between different Member States for example, detailed answers were given by some countries (e.g. Austria and the Netherlands) while this was less so in the case of some others (e.g. Hungary, Lithuania, and Denmark). 1.3 Structure of the Report The final report is structured as follows: 2: Background to the Research examines the European elections and involvement of citizens, trends in voter turnout in European Parliament elections, the Commission s Report on the 2009 European Parliament Elections, and the Commission s 2013 Recommendation on the Conduct of the European Elections. 3: Enhancing the Democratic Conduct of the European Elections analyses the research feedback on the first set of Commission Recommendations, relating to information on the affiliation between European and national parties provided by the Member States, information on the affiliation of political parties with European parties provided by the political parties, and information on candidates for President of the European Commission indicated by political parties. 4: Role of the Media in the 2014 European Elections examines the research feedback on the extent of restrictions on freedom of expression during an election period, media coverage of the 2014 European elections, coverage of EU issues in the European election campaign, the role of the social media and other issues from the terms of reference. 5: Enhancing the Efficient Conduct of the European Elections analyses the research feedback on the second set of Commission Recommendations relating to Single Point of Contact, the timely transmission of data, additional data allowing for more efficient identification, the technical means for safe and efficient transmission of data, measures taken to inform citizens of the Union of their right to vote and right to stand as a candidate, and the proportion of non-nationals registering to vote. 6: Overall Conclusions & Recommendations overall conclusions of the research and based on the research, recommendations to help improve the conduct of future European elections. There are also a number of appendices containing a list of interviews (Appendix A), copies of the interview checklist (Appendix B), and the survey questionnaire (Appendix C). Supporting data for various sections of the report is contained in Appendix D. 6

19 Background to the Research 2 Before presenting the research findings, we first summarise the background to the research European elections and involvement of citizens, trends in voter turnout, and the Commission s 2013 Recommendations on the Conduct of the 2014 European elections. 2.1 European elections and involvement of citizens European political parties and elections are the cornerstone of the democratic process in the EU and vital link between citizens and decision-makers. 6 Over the years, turnout for European elections has declined, from an overall 62% of the eligible electorate who cast their votes in 1979 when the European Parliament was directly elected for the first time, to 49% twenty years later and 42% in 2014 (See 2.2). The Commission s Recommendations were, amongst other things, designed to address this problem and to encourage citizens to participate in the European Parliament elections (see 2.3). In its report on the 2009 European elections, the Commission argued that properly informed EU citizens are much more likely to get involved in political life. In addition to highlighting a number of best practices, it was suggested that designating 2013 as the European Year of Citizens would also help raise awareness of European Parliament elections and that this, combined with other measures, should lead to a better mobilization of the electorate in Following the 2009 European elections, the Commission also conducted a public survey to examine factors influencing the turnout. It found that the three top measures that could boost citizens' motivation to vote were: more information provided on the impact of the EU on daily life (84%); more information provided on the programmes and objectives of candidates and parties in the European Parliament (83%); and more information provided on the European Parliament elections themselves (80%). These and other factors were very much at the core of the Commission s subsequent Recommendations 8. 6 European elections are regulated by the 1976 Act on the election of the members of the European Parliament by direct universal suffrage, as amended by Euratom Council Decision 2002/772/EC, which lays down some common principles for all Member States, which lays down some common principles for all Member States, such as the obligation to use proportional representation and to ensure that polling day falls within the same period starting on a Thursday and ending on the following Sunday. 6 Detailed arrangements for allowing EU citizens to participate in European elections in the Member State where they choose to live are laid down in Directive 1993/109/EC. 7 Report on the election of Members of the European Parliament and on the participation of European Union citizens in elections for the European Parliament in the Member State of residence (Directive 93/109/EC), European Commission, 27 October Report on the election of Members of the European Parliament (1976 Act as amended by Decision 2002/772/EC, Euratom) and on the participation of European Union citizens in elections for the European Parliament in the Member State of residence (Directive 93/109/EC) 7

20 Background to the Research Lisbon Treaty and European Elections The 2014 elections to the European Parliament were the first since the entry into force of the Lisbon Treaty, which strengthened the EU s democratic foundations and reinforced the role and the powers of the Parliament. The Lisbon Treaty on the Functioning of the European Union (in particular Article 292) included a number of provisions that were designed to strengthen the democratic legitimacy of the EU institutions. Thus, Article 10(2) of the Treaty states that: citizens are directly represented at Union level in the European Parliament. This helps to ensure that democratic control and accountability occur at the level at which decisions are taken. It goes on to enhances the role of citizens of the Union as political actors, establishing a solid link between citizens, the exercise of their political rights and the democratic life of the Union. Similarly, Article 10(4) of the Treaty on European Union (and Article 12(2) of the Charter of Fundamental Rights of the European Union) confer a key role on European political parties highlighting their contribution to forming European political awareness and to expressing the will of the citizens of the Union. The Treaty of Lisbon (Article 17(7)) also stipulated that the European Parliament elections results should be taken into account in appointing a new President of the European Commission. European political parties and national parties were expected to publicize their candidates during the 2014 election campaign Right to vote and stand as a candidate in European elections for EU citizens residing in a Member State where they are not nationals An important feature of European elections is the right set out in Directive 93/109/EC 9 to vote and stand as a candidate in elections to the European Parliament for EU citizens residing in a Member State of which they are not nationals. A number of the Commission s Recommendations focused on strengthening this aspect of the European elections. The 1993 Directive states that an EU citizen has to ask to be enrolled on the electoral lists of the Member State of residence and that to prevent double voting and double candidacy, the Member States are required to exchange data to identify candidates and voters registered in both their Member State of residence and of origin. In 2006, the Commission published proposals to amend the Directive with the aim of streamlining processes and simplifying administrative procedures. This was followed in 2010 by the Commission s report on the implementation of Directive 93/109/EC which identified various ways in which the measures could be made more effective. 10 Following consultations with Member States, this led in 2012 to Council Directive 2013/1/EU which amended some of the detailed arrangements set out in 9 Council Directive 93/109/EC of 6 December 1993 laying down detailed arrangements for the exercise of the right to vote and stand as a candidate in elections to the European Parliament for citizens of the Union residing in a Member State of which they are not nationals 10 Report on the election of Members of the European Parliament (1976 Act as amended by Decision 2002/772/EC, Euratom) and on the participation of European Union citizens in elections for the European Parliament in the Member State of residence (Directive 93/109/EC) 8

21 Background to the Research 2 Directive 93/109/EC and the 2013 Recommendations which, amongst other things, include the suggestion that every EU Member State should have a single contact point the exchange of data on voters, for the implementation of Article 13 of Directive 93/109/EC Electronic Voting Another issue that could affect the turnout and which might be of relevance to future European elections - is online voting. According to our research, Estonia was the only Member State that used online voting in municipal, national and European elections (socalled Internet or I-voting). While it is difficult to establish a positive correlation between the availability of online voting and turnout, some information exists. In the European Parliament elections in 2009 the number of I-voters in Estonia had almost doubled in comparison to the previous election. More than 58,000 voters used this method (this corresponded to about 15% of all participating voters. In 2014, more than 103,000 voters used I-voting in the European Parliament elections, i.e. 31% of all voters. 11 Also conceptually, it is clear that I-voting reduces transaction costs and enhances efficiency in the voting process. Consequently the whole voting experience is more convenient for voters. 12 Besides these positive effects of I-voting, there are also some negative effects. It has been argued that I-voting reduces civic engagement and triggers privacy and security concerns. 13 In addition, making voting cheaper and faster may not necessarily lead to a higher participation in elections. In other words, for those who do not intend to vote, whatever the reason, the availability of Internet voting is unlikely to make any difference. Nevertheless, for those who consider voting time-consuming, Internet voting may reduce some of the transaction costs associated with voting. 2.2 Trends in Voter Turnout in European Parliament Elections An important consideration that influenced the Commission s Recommendations is the decline over the years in the proportion of EU citizens participating in the European elections. Turnout has steadily dropped from 62% in 1979 when the first direct elections to the European Parliament were held with the 2014 figures marking a new low point in voter participation. The 2014 turnout figure of 43.1%, based on exit polls, was subsequently revised to 42.6%, putting it slightly lower than the 43.0% turnout in Further information see: ibid. 9

22 Background to the Research 2 Table 2.1 (a): Voter turnout at European Elections Member State Austria n/a Belgium Bulgaria n/a n/a n/a n/a n/a n/a Croatia n/a n/a n/a n/a n/a n/a n/a 25.2 Cyprus n/a n/a n/a n/a n/a Czech Rep n/a n/a n/a n/a n/a Denmark Estonia n/a n/a n/a n/a n/a Finland n/a n/a n/a n/a France Germany Greece n/a Hungary n/a n/a n/a n/a n/a Ireland Italy Latvia n/a n/a n/a n/a n/a Lithuania n/a n/a n/a n/a n/a Luxembourg Malta n/a n/a n/a n/a n/a Netherlands Poland n/a n/a n/a n/a n/a Portugal n/a n/a Romania n/a n/a n/a n/a n/a n/a Slovakia n/a n/a n/a n/a n/a Slovenia n/a n/a n/a n/a n/a Spain n/a n/a Sweden n/a n/a n/a n/a UK Overall Table 2.1 (b): Member State groupings Group 1 - Over 50% turnout (8 MS) Group 2 - Between 33% to 50% (12 MS) Group 3 - Below 33% turnout (8 MS) BE, LU, MT, GR, IT, DK, IRL, SW DE, LT, AU, CYP, ES, FR, FIN, NL, EE, BG, UK, PT, RO, LV, HU, CR, SL, PL, CZ, SL As the above table shows, the decline in voter turnout in European Parliament elections during the period has been quite consistent across the EU Member States although seven countries (Finland, France, Germany, Lithuania, Netherlands, Romania, and Sweden) bucked the trend in the last two elections, i.e and 2014, with voter turnout actually increasing. In Lithuania, where the European elections coincided with the second 10

23 Background to the Research 2 round of presidential elections (as in 2004) the turnout increased by no less than 26 percentage points. When the two elections were held separately, as in 2009, the turnout in Lithuania was very low. Some Member States have a longer track record of voting in European elections than others of course, depending on when they became EU members. The following chart provides a summary: Figure 2.1: Turnout in 2014 compared with 2009 European Parliament elections (percentage point change) Looking just at the results of the 2014 European elections, turnout in the EU Member States that joined after 2004 was generally lower than in the other ones (33.4% compared with 52.0% on average or 42% compared with 33% if Member States with compulsory voting systems are excluded from the calculation). This pattern was not completely clearcut. However, whereas only two of the older Member States (Sweden and the UK) had a turnout below the EU28 average, all but three of the EU Member States joining after 2004 (Cyprus, Lithuania and Malta) were below the average. The CSES team s panel of experts put forward a number of possible explanations for this pattern: European elections are generally considered as second order, i.e. elections in which voters can vent their frustrations and vote for fringe parties. In the Member States joining after 2004, protest is regularly expressed in national elections and the European elections are, therefore, of little importance as a way of protesting. Turnout in all elections in the EU Member States that joined after 2004 tends to be lower than in the other EU Member States. A possible reason might be that 11

24 Background to the Research 2 many people make use of their right not to vote due to the election policies in the past where citizens were obliged to vote. EU Member States that joined after 2004 are predominantly small countries. The public are generally positive about the EU and its institutions, but they do not believe that their countries can make a difference to EU decision-making. The European elections are therefore of little interest to many voters. Moreover, it has been argued that there is a general distrust (manifested in disinterest) for the EU in quite a few newer Member States and therefore a malaise about voting generally which has then affected the European elections. 14 The period in the run-up to the 2014 European elections was influenced by the Ukrainian crisis. It was not the EU, but the security issues associated with the Ukraine crisis that received most public and media attention in many of the newer Member States. A further observation is that, not surprisingly, the highest turnout in the 2014 European elections was achieved by countries with compulsory voting rules (Belgium, Luxembourg) but not all the 4 EU Member States with such a system were in the group with the highest levels of voter turnout: although Greece was close to the top of the rankings, Cyprus was only just above the average. It is important to note that in Cyprus a citizen can be fined if they do not vote and in Greece it is even possible to be imprisoned. However, neither of the two countries appears to systematically enforce these rules in practice. 2.3 Comparison with National, Regional and Local Elections Comparing the turnout at European elections with the turnout in recent national and regional elections provides a useful benchmark, in addition to the trend in European elections over time Comparison with national elections Table 2.2 below provides an analysis of the difference between the turnout of voters in the 2014 European elections and the most recent national elections for each of the EU28 Member States. Table 2.2: Comparison between turnout in 2014 European elections and most recent national elections in the EU Member States MS National European Diff MS National European Diff Austria Latvia Belgium Lithuania Bulgaria Luxembourg Croatia Malta Cyprus Netherlands Czech Rep Poland See in this sense in regard to Czech Republic this analysis: 12

25 Background to the Research 2 Denmark Portugal Estonia Romania Finland Slovakia France Slovenia Germany Spain Greece Sweden Hungary UK Ireland Italy Averages Table 2.2 (b): Member State groupings Group 1 - disparity < than 10pp Group 2 - between 10pp to 25pp Group 3 - disparity > 25pp BE, GR, LU, LT, RO FR, BG, IRL, IT, MT, DE, PT PL, ES, FIN, ES, SL, CR, LV, AU, UK, DK, HU, SW, CYP, NL, CZ, SL Figure 2.2: Turnout in 2014 European Parliament elections compared with most recent national elections (percentage point change) 13

26 Background to the Research 2 As can be seen, the pattern with regard to disparities between the 2014 European elections and most recent national elections is not clear-cut: Overall, for the EU28 Member States, turnout in the most recent national elections (on average, 66.9%) was 23.6 percentage points higher than for the 2014 European elections (43.4%). Three of the 5 Member State with the smallest disparities (BE, GR, LU) were countries with compulsory voting, but the others with a disparity of less than 10 percentage points were not. Likewise, Cyprus (which has compulsory voting) was amongst the Member States with the highest disparity (35 percentage points). In terms of the older/newer Member State perspective, disparities between the most recent national elections and the 2014 European elections tended to be lower in the older EU Member States than the newer ones, but the difference was not large (21 percentage points compared with 26 percentage points). Overall, apart from the compulsory voting factor, it is difficult to identify any obvious patterns in the analysis of disparities between the turnout in national elections and the 2014 European elections. Voter turnout is influenced by multiple factors. 15 Factors that could help explain the patterns include: the length of time that had elapsed between the most recent national elections and the 2014 European elections; the extent of euroscepticism amongst the political parties participating in the elections and, linked to this, the severity of the economic crisis and how closely it was associated in voters minds with EU membership Comparison with Local and Regional Elections Another comparison that is helpful to make is between the turnout for the 2014 European elections and local/regional elections. Turnout in local/regional elections is often lower than in national elections and as such the disparity with European elections is likely to be smaller than with national elections. There have been efforts to increase the turnout of both local/regional and European elections in most EU Member States. While efforts to encourage turnout in local/regional elections usually come from national political parties, 16 efforts to encourage participation in the European elections are more driven by the national authorities and the EU institutions, supported by various campaigning groups (e.g. the European Movement) and NGOs. The comparison between turnout in European and local/regional elections is shown in the following analysis. The analysis focuses on Member States where regional elections are held separately from national and municipal elections (the other Member States do not organize regional elections). The analysis is limited to local and regional elections held in the past two years. As can be seen, in the 13 EU Member States concerned, the average For example in the UK: the recent voter engagement report of the House of Commons Select Committee on Political and Constitutional Reform assessed how to engage voters both in regional elections and EP elections. 14

27 Background to the Research 2 turnout in local or regional elections was 57.0% compared with 66.9% in national elections. As such, the disparity between recent local and regional elections and the 2014 European elections is still considerable but smaller than with recent national elections (14.0 percentage points compared with 23.9 percentage points). The following table provides an analysis: Table 2.3: Comparison between turnout in 2014 European elections and most recent local and regional elections in the EU Member States 17 Member States EP elections Turnout % Local / Regional Turnout % Year Sources Austria (Länder) 45.4% 71.2% Belgium (Région) 89.6% 88.9% Czech Republic (Kraje) 18.2% 36.9% Denmark (Amtskommuner) 56.3% 71.9% Germany (Länder) 48.1% 57.6% Spain (Comunidades autonómas) 43.8% 67.0% France (Région) 42.4% 48.8% Italy (2 autonomous provinces and 20 regions) 57.2% 61.2% Netherlands (provinciale staten) 37.3% 52.4% Poland (województwo) 23.8% 47.3% Slovakia (Kraje) 13.1% 20.1% Sweden (Landstingsfullmäktige) 51.1% 82.4% United Kingdom (county councils, unitary councils) 35.6% 35.3% Averages 43.2% 57.0% n/a n/a Source: CSES analysis Holding European elections on the same day as other elections Holding the European Parliament election on the same day as the national, regional and/or local elections, as happens in several EU Member States is also thought to give a boost to European Parliament turnout. As can be seen from the following table, seven EU Member States held national, local or regional elections on the same day as the 2014 European elections. The analysis shown in the following table does indeed confirm previous research suggesting that holding the European Parliament elections on the same day as the national, regional and/or local elections, does give a boost to European elections turnout. In the case of the seven EU Member States where elections were held on the same day, the turnout for the European elections averaged 55.6% compared with an average 17 See Appendix E for the sources of the data 15

28 Background to the Research 2 turnout of 29.4% in the other countries. Only the UK was an exception to this more general pattern. Table 2.4 Other elections held on the same day as the 2014 European elections Member State Other Elections EP elections turnout Belgium Federal and parliamentary elections 89.6% Germany Local and regional elections 48.1% Greece Local elections 60.0% Ireland Local elections 52.4% Italy Local and regional elections 57.2% Lithuania Second round of presidential elections 47.4% UK Local elections 35.6% Feedback from the survey work undertaken by CSES for this study confirms the results of the comparisons between elections in EU Member States and the 2014 European elections. The question asked did not relate specifically to the turnout of voters but this was highlighted as a key factor to be considered: 16

29 Background to the Research Effects of the Economic Crisis on the 2014 European Elections Another factor that could have influenced the turnout of voters in the 2014 European elections is the economic and financial crisis that began in To assess whether there is a correlation between turnout at the 2014 European elections and the economic crisis, we used unemployment rates as an indicator of the severity of the economic downturn affecting different EU Member States. There are of course other indicators but unemployment rates are arguably uppermost in voters minds and unlike alternatives (e.g. incomes), have the advantage of being available without a significant time-lag. The chart below shows the unemployment rates in different countries in June 2014: Figure 2.3: Unemployment rates in EU28 Member States (June 2014) Source: Eurostat An analysis of the 2014 European elections turnout and the unemployment data suggests that: Taking the 12 Member States with above the average EU28 unemployment rates, the turnout in the 2014 European elections averaged 40.4% (or 38.1% if the two countries in this group with compulsory voting, Greece and Cyprus are excluded). Conversely, with the 16 Member States with unemployment rates that were below the EU28 average, turnout in the European elections averaged 45.6% (or 39.6% if the countries with compulsory voting, Belgium and Luxembourg, are excluded). Overall, it can be concluded that there is only a weak correlation between unemployment rates and turnout in the 2014 European elections with a 5.2 percentage point difference in terms of turnout in the 2014 European elections between the two groups with relatively high/low unemployment rates (or 1.5 percentage point difference if countries with compulsory voting are excluded). The full set of data is provided in the table below: 17

30 Background to the Research 2 Table 2.5: European Elections Turnout and Unemployment Rates (June 2014) Member State European Elections Turnout Unemployment rates (June 2014) Member State European Elections Turnout Unemployment rates (June 2014) Austria Latvia Belgium Lithuania Bulgaria Luxembourg Croatia Malta Cyprus Netherlands Czech Rep Poland Denmark Portugal Estonia Romania Finland Slovakia France Slovenia Germany Spain Greece Sweden Hungary UK Ireland Italy Averages Taking a broader measure of the economic crisis, namely the six EU Member States that received bailouts suggests a closer correlation between the crisis and the 2014 European elections: whereas the EU Member States that were relatively unscathed by the crisis had an average turnout of 33.0% in the elections, the six countries receiving bailouts (Cyprus, Greece, Ireland, Italy, Portugal and Spain) had an average turnout that was considerably higher than this (48.5%). The inclusion of two countries with compulsory voting systems again affects this analysis although not to the extent of invalidating the argument. Experts who argue that the financial crisis led to an increase in voter turnout in the 2014 European elections, suggest that this is because the electorate realised that the EU has a huge impact on everyday life and this made people more inclined to vote. While turnout in countries such as Slovakia, Poland and the Czech Republic showed record lows, one expert (Stratulat 18 ) speculated that it was probably greater public awareness about the relevance of EU for national politics in the context of the crisis in countries such as Germany, Greece and Sweden, that helped keep turnout stable, or at least prevented the overall participation rate from declining again. But, at the same time, on this broader measure, it seems that Member States especially hard-hit by the economic crisis (CY, IR, PT, Spain, IT, and SK and SI among the newer Member States) all experienced a lower turnout in the 2014 European elections compared to Greece was an exception (here, turnout was higher than in 2009). This country has had strong anti-establishment and EU-critical political actors (Golden Dawn and

31 Background to the Research 2 SYRIZA) which successfully exploited the negative anti-eu feelings and this possibly raised the turnout. One expert in Slovenia suggested that although voters were strongly encouraged to vote, the 6-year economic crisis in Slovenia, together with the forthcoming national elections and local elections, had resulted in voters being not particularly enthusiastic about the European elections. The expert added that although EU issues were promoted in the European elections in Slovenia, quality of life issues (high unemployment, decreasing in purchase power, frequent changes in the national government, etc.) were more important to voters. There were also factors that could have influenced the turnout in the 2014 European elections that were specific to particular countries. For instance, if an election is held at the weekend rather than on a work day (as is the case in the UK), then the turnout of voters also tends to be higher. The turnout in European elections tends to be higher not only when the national elections are held on the same day (e.g. Belgium 2014, Luxembourg 2009, 2004), but also when any other election is held on the same day, including presidential elections (e.g. Lithuania 2004 and 2014), local/regional (Greece and Ireland 2014), and also referendums (Denmark 2009). Also notable is an experiment with e-voting, as in Estonia, which has been shown to boost turnouts as well Commission s Report on the 2009 European Parliament Elections Following the 2009 European elections, the Commission produced an assessment of aspects with a relevance to Directive 93/109/EC and more generally. 20 An important aim of the current assignment is to assess the extent to which the shortcomings identified by the Commission in relation to the 2009 European elections were rectified in the 2014 European elections. Noting the steady decline in voter turnout in the European elections, the Commission s 2010 report highlighted the results of a Flash Eurobarometer survey after the 2009 European elections identifying measures that could boost citizens' motivation to vote. These were: more information provided on the impact of the European Union on their daily life (84%), more information provided on the programmes and objectives of candidates and parties in the European Parliament (83%) and more information provided on the European Parliament elections themselves (80%). Specifically in relation to non-nationals, the report noted that even though the number of non-nationals of voting age residing in many EU Member States increased, the rate of enrolment did not rise proportionately in 2009 when compared to In relation to the systems in place to prevent double voting, the Commission s report concluded that despite the preparatory work for the 2009 elections, there had been no great improvement. Due to non-harmonised electoral timetables and differences in the data used in the Member States for identifying voters, information on EU citizens sent by 19 See above, section Report on the election of Members of the European Parliament (1976 Act as amended by Decision 2002/772/EC, Euratom) and on the participation of European Union citizens in elections for the European Parliament in the Member State of residence (Directive 93/109/EC) 19

32 Background to the Research 2 the Member State of residence to the Member State of origin very often could not be used. In particular, large numbers of non-national EU citizens were not identified by the Member States within the information exchange mechanism. 21 This was attributed to the fact that Member States were not required under the Directive to provide date of birth as part of the identification of voters and yet this was needed in most Member States to do so. The late transmission of data was identified as a problem by the authorities in Lithuania, Hungary, the Czech Republic, Denmark, Estonia, Greece, France, and Ireland. Further obstacles identified by the Commission with regard to sharing data on nonnational voters in the run-up to the 2009 European elections included: the dispatching of the information by a large number of decentralised bodies instead of one single contact point; problems accessing data sent electronically and protected by different security tools; and handling large numbers of notifications sent in paper format. Overall, it was concluded that a more efficient mechanism for exchanging data would require further common rules and possible amendments to Directive 93/109/EC. As noted earlier, it is important in this assignment to assess the extent to which the shortcomings identified by the Commission in relation to the 2009 European elections with regard to on-nationals were rectified in the 2014 European elections. 2.6 Commission s 2013 Recommendation on the Conduct of the European Elections On 12 March 2013 the Commission adopted Recommendation 2013/142/EU on enhancing the democratic and efficient conduct of the elections to the European Parliament, which was addressed to the Member States and to the European and national political parties. The European Parliament also adopted resolutions to this effect. 22 In an accompanying Communication, the Commission outlined initiatives to facilitate citizens participation in the 2014 European elections and to safeguard the respect of the democratic principles of the elections. As the Commission pointed out, it was committed to fully exploiting existing Lisbon provisions to further enhance transparency and the European dimension of the European elections, thereby reinforcing the democratic legitimacy of the EU decision-making process and bringing the system closer to Union citizen. The Recommendation on enhancing the democratic and efficient conduct of the European elections (C (2013) 1303 final) set out a number of proposed measures under two headings democratic conduct of the elections and, secondly, efficient conduct of the elections. 21 Examples quote in the report were: the Czech authorities identified only 2,500 out of the 3,800 Czech nationals notified by the other Member States, because insufficient data were available; in Ireland only 208 nationals were identified out of the 4,795 notified; Portugal identified only 38,619 nationals out of the 83,556 who were notified. 22 European Parliament resolution of 22 November 2012 on the elections to the European Parliament in 2014 (2012/2829(RSP)); European Parliament resolution of 4 July 2013 on improving the practical arrangements for the holding of the European elections in 2014 (2013/2102(INI). 20

33 Background to the Research 2 Table 2.5: Overview - Commission s Recommendations Facilitating information for voters on the affiliation between national parties and European parties 2. Informing voters about the affiliation between national parties and European political parties 3. Support for a candidate for President of the European Commission Enhancing the democratic conduct of the elections Member States should facilitate the provision of information to the electorate on the affiliation between national parties and European political parties before and during the elections to the European Parliament, inter alia by allowing and encouraging the indication of such an affiliation on the ballots used in those elections. National political parties participating in the elections to the European Parliament should make publicly known ahead of those elections their affiliation with European political parties. National political parties should prominently display their affiliation with European political parties in all campaign materials, communications and political broadcasts. European and national political parties should make known the candidate for the function of the President of the European Commission they support and the candidate's programme. National political parties should ensure that their political broadcasts in view of the European Parliament elections are also used to inform citizens about the candidate for the function of President of the European Commission they support and about the candidate s programme. 4. Common voting day Member States should agree on a common day for elections to the European Parliament with polling stations closing at the same time. Enhancing the efficient conduct of the elections 5. Single contact authority The Member States should set up a single contact authority in charge of the exchange of data on voters, for the implementation of Article 13 of Directive 93/109/EC. 6. Transmission of data Member States should take account, as far as possible, of each other's electoral arrangements, so that the Member State of residence sends the data on voters in time for the respective home Member State to take the necessary measures. 7. Additional data allowing for more efficient identification 8. Technical means for safe and efficient transmission of data The Member States of residence are encouraged to transmit, in addition to the personal data foreseen in Article 9 of Directive 93/109/EC, all relevant personal data, which may be necessary for the identification of voters. For exchanging the data as provided in Article 13 of Directive 93/109/EC the Member States should use a uniform and secure electronic means, as set out in the Annex. The Member States should transmit the data in one single package per home Member State, with one further transmission at a later stage for updates, where necessary. The first set of recommendations aimed to further strengthen the transparency of the European elections and the democratic legitimacy of the decision-making process at the EU level. The Commission recommended that ahead of the elections voters should be told about the affiliations between national and European political parties. It was suggested that 23 Communication from the Commission to the European Parliament, The Council, The European Economic and Social Committee and the Committee of the Regions Preparing for the 2014 European elections: further enhancing their democratic and efficient conduct /* COM/2013/0126 final 21

34 Background to the Research 2 national and European political parties also set out their preferred candidate for President of the European Commission and give details of their programme. National political parties were urged to use their political broadcasts to inform citizens about their party's candidate and the candidate's programme. The Recommendation also called for a common European voting day which would, it was argued, better reflect common participation by citizens across the EU Member States. The second set of recommendations was addressed to Member States and aimed to increase the efficiency of the elections by streamlining procedures for the implementation of the rights of EU citizens to vote and stand as candidates in their Member State of residence, laid down in Directive 93/109/EC, namely the data exchange mechanism for preventing multiple voting (fixed in Article 13 of this Directive). These recommendations were aimed at further improving the efficiency of the elections and, at the same time, reducing the administrative burden on Member States. Preliminary report on the implementation of the Recommendation On 27 March 2014, two months before the European elections, the Commission produced a preliminary report on the implementation of the Recommendation, Towards more democratic European Parliament elections - Report on the implementation of the Commission's recommendations of 12 March 2013 on enhancing the democratic and efficient conduct of the elections to the European Parliament (COM(2014) 196 final ). To help prepare the report, the Commission had asked the Member States to submit information, on the state of implementation of the recommendations. A total of 25 of the EU28 Member States provided at least some if not all of the requested information. However, there were considerable gaps even in these cases. For example, whilst most Member States provided information in relation to Recommendation 1 ( Member States should encourage and facilitate the provision of information on affiliation between EU parties and national parties ) and most also did so for Recommendation 2 ( Political parties should provide information on their affiliation with EU parties ), coverage of the other two main Recommendations ( Political parties - EU and national - should make known their candidates for President of the European Commission and Member States should agree on a common voting day ) was far less complete. Similarly, the information provided by the Member States related to Further strengthening the transparency and the democratic legitimacy with none being provided in relation to Further enhancing the efficient conduct of the elections 24. The Commission also gathered information from national experts on elections, as well as from European political parties. 25 Overall, it was argued that the Commission s Recommendation on enhancing the democratic and efficient conduct of European elections had been generally well received. 24 Table 2.6. in Appendix F summarises the extent to the requested data on issues relating to Further strengthening the transparency and the democratic legitimacy was provided by the different EU Member States in March A total of 18 Member States submitted information: Belgium, the Czech Republic, Denmark, Estonia, Greece, Spain, Finland, Croatia, Hungary, Luxembourg, Latvia, Malta, the Netherlands, Poland, Portugal, Slovenia, Slovakia, and the United Kingdom. From the European political groups, responses were obtained from: the European People's Party, the Party of European Socialists, Alliance of Liberals and Democrats for Europe, Alliance of European Conservatives and Reformists, the Movement for a Europe of Liberties and Democracy, and European Democratic Party. 22

35 Enhancing the Democratic Conduct of the European Elections 3 In this section we provide an assessment of the extent to which the Commission s Recommendations were implemented with regard to enhancing the democratic conduct of the 2014 European elections. Overview The structure of the assessment in this section follows the Commission s Recommendations themselves and is summarised below: Enhancing the democratic conduct of the elections 3.1 Encouraging and facilitating information for voters on the affiliation between national parties and European political parties 3.2 Informing voters about the affiliation between national parties and European political parties 3.3 Support by national and European Parties for a candidate for President of the European Commission 3.4 Common voting day In each subsection, we start by summarising the purpose of the Commission s Recommendations and then assess the extent to which they were implemented and the effects on the 2014 European elections. The assessment starts in each case with an analysis of the CSES survey data and then goes on to evaluate other research findings from the desk research, interview programme and other sources of information. Where possible, we also compare the findings in relation to the 2014 European elections with the previous elections. The Commission s Recommendations included a number of measures to further strengthen the transparency and democratic legitimacy of the 2014 European elections. 3.1 Encouragement and facilitation by Member States of the provision of information in their electoral system on the affiliation between European and national parties. The first of the Commission s Recommendations was directed to Member State authorities and designed to encourage and facilitate the provision of information to the electorate on the affiliation between national parties and European political parties before and during the elections to the European Parliament. This also includes allowing and encouraging the indication of such an affiliation on the ballots used in the elections. The type of actions Member States took mainly depended on the country s electoral laws stipulating how political parties could present themselves on the ballot paper. As noted earlier, for the Commission s preliminary report, information from Member States on the implementation of this Recommendation could be gained from all Member States except for Bulgaria (see table below). The position with regard to the implementation of the Recommendations varied across the Member States with national authorities taking different actions. As Table 2.1 below shows, there were three main actions non-legislative and legislative actions, and a situation where the Commission s Recommendation was not implemented. 23

36 Enhancing the Democratic Conduct of the European Elections 3 Table 3.1: How did Member States promote the publication of the link between national and the European Parties? National Authorities actions 1. Through non-legislative measures: - national authorities simply informed political parties of recommendations 2. Through legislative measures: the national legal framework allowed an indication of European political party affiliations on the ballot paper 3. Non-implementation: - national authorities did not take any action to inform or encourage political parties to inform the electorate on the affiliation to a EP party) Member States BE, DE, EE, EL, ES, FIN, IT, LU, LT, MT, NL, PT, RO, UK, SE, HU AT, BE, CY, EL, FR, IRL, NL, SI, UK CZ, DK, LV, SK, PL, HR Note: Some Member States supported the Commission s Recommendation both informally and formally. Therefore, they are mentioned in Category 1 and 2 and marked in bold letters. Non-legislative measures Over half the EU Member States adopted non-legislative methods to support and facilitate the provision of information to the electorate on the affiliation between national parties and European political parties. Some of these countries decided to take non-legislative measures because laws permitting affiliations with European parties already existed while others took such measures as an alternative. A total of 16 Member States indicated that they had told national political parties about the need to increase transparency and improve information for voters The way of making national parties aware took different forms: In most cases, this meant that the authorities in these Member States communicated the Commission s Recommendations to the political parties, making them aware of their existence, but there were also a few Member States were more intense efforts were made. For instance, in Greece a meeting in the Greek national Parliament was organised by the national authorities to inform parties of the Recommendation. In Italy and Finland, guidelines were issued for the national parties to make them aware of the Recommendation and encourage them to mention their affiliation to the EU party. Examples Non legislative implementation The Ministry of Justice in Finland being responsible for elections produced guidelines and disseminated information to the political parties in Finland. This included information on the Commission s Recommendation. Specifically, Finland requested national parties to make the affiliation between national and European parties more visible. The Italian Ministry of Interior strongly welcomed the Commission Recommendations. The Central Electoral Services Directorate released Communication No 2 on 'Instructions for presenting and admitting candidates' for the European Parliament elections in Italy, stating that the full name, acronym or symbol of the European political party could appear inside the registered logo of each national political party or movement. In this way, the national parties received clear guidelines on how to inform the electorate about their affiliation to the respective EU party. 24

37 Enhancing the Democratic Conduct of the European Elections 3 Legislative measures In around a third of the EU Member States the Commission s Recommendations were implemented through legislative measures. A total of 9 EU Member States introduced laws - or already had such a law in place - to allow the political parties to indicate an affiliation with a European political party on the ballot paper by mentioning the name or the logo enacted a legal act (Austria, Belgium, France, Netherlands, Spain, Greece, Cyprus, Slovenia and Ireland). However, amongst the political parties in these countries, only some parties in France, one coalition party in Greece and three parties in the Netherlands made use of this right. In Belgium, Austria, Cyprus and Spain, no party mentioned the affiliation with a European political party on the ballot paper. In Ireland the affiliations between national parties and the European parties or groupings were specified on the ballot papers. There are rules as to what can be said on ballot papers but there was no issue in this respect Four other Member States (HR, EL, DE, RO) that implemented the Recommendation using non-legislative methods also considered amending their legislation after the Commission issued its Recommendations. While Greece did implement laws allowing the mentioning of an affiliation on the ballot paper before the European elections 26, Croatia and Romania did not implement such a law. Feedback from NGOs and other organisations suggested that in Germany the authorities tried to initiate a change of the electoral law to allow the mentioning of the party affiliation between the national and the European parties on the ballot paper but the idea was rejected by the Bundestag. Consequently, only non-legislative measures were taken Examples legislative measures. The Greek authorities modified the legal framework (law 4255/2014) allowing national parties to indicate their affiliation to European parties on the ballots. In The Netherlands the parties were informed of the Commission s Recommendations. Furthermore, in line with existing law the Dutch authorities allowed party affiliations to European parties to be mentioned on the ballot paper. Of the 19 parties that contested the European election, 3 displayed the logos of the European parties that they are affiliated to on their ballot papers. The Spanish authorities informed their national parties about the Commission recommendations to make the affiliation between national and European parties more visible. National law allows names and/or logos of European parties to be displayed on the voting ballots. In this way, Spain took both formal and informal measures to ensure that parties could mention their affiliation to a European political party. The Member States that allowed parties to publicize their European affiliations either through legislative and/or non-legislative measures argued that this did increase the 26 Law 4255/2014 adopted in April. 25

38 Enhancing the Democratic Conduct of the European Elections 3 democratic conduct of the elections and made the electorate more aware of the relationship between national parties and the European Parliament. Several Member States went beyond what was suggested under the Commission s Recommendation. For instance, the Cypriot authorities took measures to directly inform the electorate about the nature of EU elections and the link between national and European parties. Non Implementation A relatively small number of EU Member States did not take any action to implement the Commission s Recommendation. These Member States neither enacted any law nor took action to inform their national parties about the Recommendation. This was the case in 6 Member States. The Member States that did not take any measures at all to encourage and facilitate the provision of information by national parties on their affiliation to European parties mainly justified this decision by mentioning that it was the responsibility of the parties and not of the authorities to do this. Most of the national election experts from countries that do not allow for the publication of such an affiliation indicated that the main reason for this was the concern that this would lead to too much interference with the mandate of political parties. In particular, in the case of the Czech Republic it was argued that it would be against their electoral law to influence how political parties conduct their election campaigns. Also in Croatia, election legislation does not allow an indication of the affiliation of national parties to the respective European political parties on the ballot papers. More specifically, in accordance with Article 41 of the Act on Election of Members of the European Parliament from the Republic of Croatia, the content of the ballot paper in elections to the European Parliament in Croatia should only include instructions on how to vote, the numbers and names of the lists taken in order from the overall list of candidates, and on each list the numbers, forenames and surnames of the candidates and the serial number. In general, European political parties are not mentioned in the Croatian elections legislation. Feedback from the CSES survey suggests that not enough was done by the EU28 national authorities to promote awareness of the links between national parties and European political parties. 26

39 Enhancing the Democratic Conduct of the European Elections 3 27

40 Enhancing the Democratic Conduct of the European Elections Information on the affiliation of political parties with European parties provided by the political parties The Commission s Recommendations urged the national political parties themselves to promote their affiliation with EU political parties as a way of informing voters about the affiliation between national parties and European political parties. Overall, in relation to this Commission Recommendation, it is possible to distinguish between three situations: Comprehensive actions - national political parties made their affiliation to a European party fully visible in all their election material, political broadcasts, etc. Limited actions - parties made their affiliation partially visible, i.e. in some of their election material, political broadcasts, etc. However, in most cases this was not done in billboard campaigns. No actions - national political parties took no actions to publicize their European affiliations. The research to examine the situation in different EU Member States involved examining websites and an interview programme for a sample of national political parties. 27 Comprehensive actions Feedback from the research suggests that only a small number of national political parties took comprehensive actions to promote awareness of their affiliations at the European level. This category consists of parties that made the electorate fully aware of their affiliation to the national party. This means that party affiliation was mentioned not only in debates, speeches and party events but also on all written material such as party manifestos, posters, brochures and social media. This category also includes national political parties that mentioned their affiliation to a European party on the ballot paper (as far as this was legally possible). For instance, feedback from the interviews with political parties in Latvia (the Homeland Union-Christian Democrats and the Liberal Movement) indicated that they, and most of the other main national political parties, did a lot to promote the affiliations to European political parties, for instance via website banners and materials, and in political broadcasts and debates about the European Parliament elections on TV and radio. In Greece, one of the ruling coalition parties highlighted such affiliations and displayed the name and emblem of the European party to which it is affiliated, and this was also the case with two of the other parties. 27 European political parties have a combined membership of just over 500 national parties. For the purposes of this research, the focus was on the two political parties from each Member State with the most seats in the European Parliament. In most countries, the two parties that won most seats in the EP are also the majority parties in the respective Member States. However, in some Member States this is not the case (i.e. France where a majority party at the national level did not win seats in the European Parliament). To help ensure representative coverage at a national level, we therefore also included a number of smaller parties in the research. 28

41 Enhancing the Democratic Conduct of the European Elections 3 In Spain, the PP (People s Party; Partido Popular) mentioned its affiliation with the EPP in its manifesto. In fact, it highlighted that the EPP is the largest European party in Europe and noted the importance of this representation for the future as more and more decisions are made by the European Parliament. It stated that to "be part of European parties that will shape these parliamentary majority is the greater and better guarantee to effectively defend the interests of Spain in Europe". Furthermore, the affiliation to the EPP was mentioned in public events during the campaign, as well as broadcasts and TV political debates about the European elections. Moreover, the candidate for the Presidency of the European Commission from the EPP participated in public events during the campaign with PP and these activities were widely covered by the media. Examples - Comprehensive actions In the case of the majority party DISY (The Democratic Alliance; Δημοκρατικός Συναγερμός) in Cyprus, the party s manifesto made direct references to membership of the European Peoples Party. The webpage dedicated to the European elections also included a translated document of the manifesto of the European Peoples Party and the action plan of the EPP for the period The banner of the EPP is also rather prominently displayed in the party s website and it was also included in all campaign material. In Portugal, Bloco (Left-wing Bloc; Bloco de Esquerda) and CDU (The Democratic Unitarian Coalition; Coligação Democrática Unitária) made visible their link to the GUE/NGL in their manifesto, websites and also in other material (e.g. a newspaper that was especially produce by Bloco for the campaign). Limited actions A larger proportion of national political parties took more limited actions to publicize their affiliations with European political parties. This second category includes parties that promoted their affiliation to a European political party in a rather limited way, i.e. usually in party manifestos rather than more widely and more publicly (e.g. on billboards, campaigning events and/or in their election material). This meant that awareness of the European affiliations tended to be restricted to better informed citizens rather than the general public, a view which is confirmed by experts we consulted from NGOs and other organisations. For instance, in Germany both of the majority parties, the SPD (Social Democratic Party; Sozialdemokratische Partei Deutschlands) and CDU (Christian Democratic Union; Christlich Demokratische Union Deutschlands), as well as smaller parties revealed the affiliation to European parties in some way. While billboard campaigns largely omitted to mention the affiliation, it was made clear in the respective party manifestos, and in some cases in TV debates and campaigning events. A similar situation existed in other Member States. For instance, in the Czech Republic all three parties that have been analysed (CSSD (Czech Social Democratic Party; Česká strana sociálně demokratická), ANO 2011 and TOP09) only undertook limited efforts to promote their Spitzenkandidat. 29

42 Enhancing the Democratic Conduct of the European Elections 3 In Hungary, feedback from NGOs and other organisations suggests that MSZP (Hungarian Socialist Party; Magyar Szocialista Párt) did make some references to their European party but only on the side in public debates or at events and not in the general campaigning material. Fidesz (Hungarian Civic Alliance; Fidesz Magyar Polgári Szövetség) mentioned their affiliation also at events and in public debates but no clear connection was evident in printed material or posters. In Denmark, Venstre, the Social Democrats and the Danish People s Party undertook some limited actions to make the link to their EU affiliated party visible, mainly through highlighting the affiliation in their party manifesto. A similar situation existed in Malta where both the main parties mentioned their affiliations with European parties, but only in passing. No actions A relatively small proportion of the national political parties made no mention at all of their European affiliations. Several of them were Eurosceptic parties. Examples include Jobbik in Hungary (which is generally anti-eu) and UKIP in the UK. However, there were also other reasons why national political parties did not publicize their affiliation to a European political party, either not at all or only in a rather limited way. Firstly, in some cases there was not enough time to take action. The Commission s Recommendations were only communicated to the national authorities and parties in December 2013, i.e. only a few months before the campaigning for the EU elections began. Consequently, in some cases the election campaign was already designed and could not be changed at that relatively late stage. This was mentioned, for instance, by SPÖ in Austria. Secondly, the affiliation to a European party might have not yet existed at the time when the campaign for the 2014 European elections started. This could have been especially the case for some smaller parties and/or for parties that that were not previously represented in the European Parliament. Likewise, a small number of national parties did not belong to any European party even though they had/have seats in the European Parliament (their MEPs are referred to as Non-attached members. In this case, absence of publicizing a European affiliation is obvious). 28 Last but not least, another explanation is related to domestic political considerations, i.e. some national parties did not want to mention a European party affiliation because at the Member State level they were in opposition to a party that was a member of the same European party at the EU level (examples that were provided to us included parties belonging to the Alliance of Liberals and Democrats for Europe (ALDE)). Besides assessing whether the parties in the Member States publicised their affiliation to the European parties, another pertinent question is the extent to which the national parties focused on European topics during the election campaigns. In many EU Member States (particularly good examples include Austria, Germany and Finland) the political 28 For a list mentioning all NA members, see: 30

43 Enhancing the Democratic Conduct of the European Elections 3 parties focused very much on European issues in their election campaigns (e.g. issues relating to youth unemployment, the financial crisis, migration). However, in other Member States (good examples being the Czech Republic and Hungary) the election campaigns were marked by a focus of national topics. In the case of Hungary a reason for this, according to feedback from our research, is that since the national elections took place shortly before the European elections, the two campaigns blended into each other. Other parties, e.g. Jobbik in Hungary deliberately ignored European topics due to their hostility or distant position to the EU. Overall, feedback from the research suggests that there was more information on the affiliations between national and European parties in 2014 than in the previous 2009 European elections. However, there was a considerable diversity in the way the Commission s Recommendation was implemented. Most national parties mentioned their affiliation to a European party, at least in a limited way. But a significant (albeit difficult to quantify) number of national political parties did not, for practical and/or ideological reasons. This latter explanation is especially because of the financial crisis affecting Europe and the increased strength of political parties that are hostile to the EU. This conclusion is supported by the CSES survey responses with around two-thirds of those completing the questionnaire agreeing that the political parties did enough to promote an awareness of their affiliation with European political parties. 31

44 Enhancing the Democratic Conduct of the European Elections Candidates for President of the European Commission With the entry into force of the Lisbon Treaty, a silent revolution took place through a change of procedures of the European elections. Josef Janning (2014) argues that the fact that after Lisbon European parties could nominate a Spitzenkandidat for the Commission Presidency made the EU less intergovernmental and strengthened the democratic nature of the elections. 29 It is argued that this constitutional change was positively perceived by a majority of Europeans. 30 Based on this constitutional change, one of the Commission s Recommendations was that the European and national political parties should publicize their candidates for the role of President of the European Commission. It was envisaged that this would help make it clearer to citizens how their vote could influence EU decision-making and give the President a democratic mandate, albeit an indirect one European Parties and the Spitzenkandidaten All the European parties that put forward a candidate launched EU-wide election campaigns to raise awareness of their Spitzenkandidaten and his or her manifesto. Table 3.2: Campaign visits by Spitzenkandidaten to EU Member States 32 MS PES EPP ALDE EL EGP MS PES EPP ALDE EL EGP Austria Latvia Belgium Lithuania Bulgaria Luxembourg Cyprus Malta Croatia Netherlands Czech Rep Poland Denmark Portugal Estonia Romania Finland Slovakia France Slovenia Germany Spain 29 arliament_c281# Six of the 13 European political parties put forward candidates for President of the European Commission: Jean-Claude Juncker (EPP), Martin Schulz (PES), Guy Verhofstadt (ALDE), José Bové and Ska Keller (Greens), and Alexis Tsipras (European Left). The Movement for a Europe of Liberties and Democracy (MELD) and the Alliance of European Conservatives and Reformists (AECR) indicated they would not nominate a candidate. 32 Source: Transparancy International ( (Google Doc)). A further table with more details on the visits can be found in Appendix D. 32

45 Enhancing the Democratic Conduct of the European Elections 3 Greece Sweden Hungary UK Ireland Italy Total The Party of European Socialists (PES) for the first time had a campaign where a large number of volunteers directly spoke to the electorate in each EU Member State through door-to-door visits. This initiative, called Knock the Vote 2014, was supplemented by a large-scale social media campaign and the organisation of events in collaboration with their national member parties. As part of the wider election campaign, the PES organised a far-reaching campaign with Martin Schulz who visited 20 countries and many towns and cities in them. The European Peoples Party (EPP) also stressed the importance of conducting a transnational campaign to make the European party more visible and focused on European instead of national topics. Another part of the EPP campaign was The Juncker for President campaign trail which started at the beginning of April 2014 and ended on the 25 May The trail passed through the following cities in 18 Member States: Porto and Lisbon (Portugal), Athens (Greece), Hamburg, Saarlouis 33 Berlin, Braunschweig, Düsseldorf, Rotenburg and der Fulda, and Munich 34 (Germany), Nivelles, Antwerp 35 and Brussels (Belgium), Esch (Luxembourg), Strasbourg 36 Paris and Bordeaux (France), Helsinki 37 (Finland), Riga 38 (Latvia), Poznan (Poland), Sofia (Bulgaria), Maastricht (the Netherlands), Valletta (Malta), Nicosia (Cyprus), Bratislava (Slovakia), Rome and Florence (Italy), Vienna (Austria), Madrid (Spain). 39 The Alliance of Liberals and Democrats for Europe (ALDE) campaign I vote liberal also promoted the Spitzenkandidat through speeches, events and social media. Nevertheless, in contrast to the PES, less emphasis was put on the Spitzenkandidat in billboard campaigns. The ALDE campaign focused more on political objectives. The Campaign trail of ALDE s Spitzenkandidat, Guy Verhofstadt, covered 15 countries (see Table 3.3). The European Left Party (EL) put most of its efforts on online campaigning. It believed that grassroots methods created the most spontaneous participation. Nevertheless, the EL also organised a transnational campaign trail. The Spitzenkandidat, Alexis Tspiras, organised events throughout Greece and in 10 other Member States (see Table 3.3). The European Green Party (EGP) campaign was called 2014 European Elections: this time it s different! In 2004, the Greens were the forerunners of pan-european ibid. 38 Ibid

46 Enhancing the Democratic Conduct of the European Elections 3 campaigning while in 2014 it was the only party to organise an open Europe-wide Online Primary to select the two Spitzenkandidaten. This allowed all supporters of the Greens to select their preferred candidates for EU presidency. Besides this innovative tool the Greens also conducted a pan-european campaign trail by conducting events in 16 Member States (see Table 3.3) Televised debates between the Spitzenkandidaten The fact that Spitzenkandidaten were involved in TV debates in various countries was an important new aspect of media coverage of the 2014 European election. The main TV debate took place in the European Parliament s plenary chamber on 15 May 2014 and was hosted by the European Broadcasting Union (EBU). It was shown on 49 TV channels and in 24 languages. 41 In addition to the EU-level debate and the individual appearances of the candidates, there were also a number of national televised debates between the Spitzenkandidaten. In total, there were 10 events in five different countries at which at least two of the candidates debated election issues. Table 3.3: Debates between the Spitzenkandidaten Date Host Participants Location Language 9 April France 24 and RFI Juncker and Schulz Brussels French 9 April France 24 Juncker and Schulz Brussels English 28 April European Youth Forum, City and University of Maastricht, Euronews 29 April Euranet Plus Juncker, Schulz, Verhofstadt, and Keller Juncker, Schulz, Verhofstadt, and Keller Maastricht Brussels English English 8 May ZDF and ORF Juncker and Schulz Berlin German 9 May EUI Juncker, Schulz, Verhofstadt and Bové Florence English 13 May LCI and RFI Juncker and Schulz Paris French 15 May EBU Juncker, Schulz, Verhofstadt, Keller and Tsipras Brussels English 19 May France 2 Verhofstadt and Bové Paris French 20 May ARD Juncker and Schulz Hanover German Source: Wikipedia As can be seen from the table, five debates were designed as head-to-head debates between the representatives of the two leading European political parties, Jean-Claude Juncker of the European People's Party and Martin Schulz of the Party of European 40 See: 41 BBC News, European elections: Rivals for EU Commission president;, 19 May 2014 at 34

47 Enhancing the Democratic Conduct of the European Elections 3 Socialists. Four others were open to all nominated lead candidates, while one Frenchlanguage debate was held between José Bové of the European Green Party and Guy Verhofstadt of the Alliance of Liberals and Democrats for Europe Party. It should be noted that these events were also broadcast on other national television channels, beyond those hosting the events (e.g. the debate on 9 April that was held in English was also broadcast on the UK s BBC) as well as online. While the traditional media in the form of broadcast TV still plays the most important role in informing potential voters, many people now watch these programmes online. A study conducted by the German TV programme ZDF came to the conclusion that offering TV programmes online ensures that there is greater coverage of target audience (including all age and social groups). 42 This is because most people have access to the TV but especially younger people prefer to use the Internet. 43 The study further revealed that broadcasting TV programmes online not only guarantees access to a wide range of citizens but that this option is made greater use of than TV. In Germany, for instance, 52% of Internet users aim to gain information about the latest news. These statistics are similar to the European average where 60% of Internet users claim to obtain information on current events online. 44 Of those persons that obtain information online, 65% users retrieve information on EU political matters from official news websites. 45 Based on findings related to the impact of TV debates between candidates in the US, it can be argued that TV debates involving the Spitzenkandidaten was important in terms of informing citizens and influencing the turnout. 46 Less positively, some TV broadcasters made use of geo-blocking, i.e. denying users access to their website s content in third countries. 47 There are various reasons why TV broadcasters deny access of certain programmes to Internet users in other countries. The reason is mainly related to commercial/contractual practices and distribution strategies. For example, the BBC blocks the use of some web-services because outside the UK because users based abroad do not pay licence fees. Also, the German ZDF blocks some live streams and web broadcasts so they are not always available abroad. However, the ZDF has indicated that this practice is currently under revision. 48 The practice of geo-blocking is likely to have a negative impact on citizens living in a Member State which is not their country of origin and who want to be informed about 42 Konzept der Telemedienangebote des ZDF Retrieved from: p. 6 and p Research has shown notably that TV debates between candidates has a huge impact in the US. See for instance: 47 Geo-blocking is defined as [t]he practice of preventing users from viewing Web sites and downloading applications and media based on location. Used by countries to block foreign material as well as by movie and TV studios to restrict viewing to specific regions, geo-blocking is accomplished by excluding targeted Internet addresses

48 Enhancing the Democratic Conduct of the European Elections 3 EU topics via their national media. Research suggests that in general EU citizens trust the media of their home country as an objective source of information on EU topics. 49 Also in relation to the TV debates of the Spitzenkandidaten it is most likely that the debate in the respective home countries would have been preferred because of language issues and the topics addressed in the discussion. Consequently, geo-blocking could have prevented expatriates from obtaining information on EU-related topics National Parties and the Spitzenkandidaten The national parties were also involved in promoting the Spitzenkandidaten. The following situations arose: Very active - some national parties were very active in widely promoting their Spitzenkandidat, i.e. both in the billboard campaign as well as in events, debates and in the campaign material. Quite active - other national parties publicized their Spitzenkandidat to a more limited extent, i.e. only in some debates and events. This led to a situation where only highly-informed voters knew about the party s Spitzenkandidat. Not active at all - a few national parties decided to not inform the electorate at all about their Spitzenkandidat. Examples - National Parties promotion of Spitzenkandidaten In Germany, the Social Democratic Party (SPD) was very active in promoting the PES Spitzenkandidat. The candidate was advertised by the SPD via different tools and on every occasion. He was not only promoted in the traditional billboard campaign but also during events and in the social media. A good example of high visibility of the Spitzenkandidat although the candidate was neither from the same party nor country was in Italy. The other Europe with Tsipras was the name of the Italian list supporting Alexis Tsipras. His position was communicated to the electorate by making his name highly visible in the symbol used on the ballot papers. Equally, the list supporting Alexis Tsipras publicized his programme through a website and other forms of political broadcasting (e.g. videos and blogs) but mostly thorough public speeches. The Liberal Movement in Lithuania was also very active in promoting Guy Verhofstadt as the candidate for ALDE and used his name in campaign literature (e.g. website and printer materials) in specific contexts, for instance, emphasising his position on Ukraine compared with other candidates p

49 Enhancing the Democratic Conduct of the European Elections 3 The following table summarises the position for the same sample of national political parties as used earlier. 50 Table 3.4: Extent to which national parties promoted the Spitzenkandidaten Key: Very active: ; Quite active: ; Not active at all: x. Member State Rating Member State Rating Austria Italy Belgium Latvia Bulgaria Lithuania Croatia x Luxembourg Cyprus Malta x Czech Rep x Netherlands x Denmark x Poland x Estonia Portugal Finland Romania x France Slovakia Germany Slovenia Greece Spain Hungary x Sweden Ireland x UK x The reasons for either publicising or omitting to promote the Spitzenkandidat are similar to the reasons for promoting the party s affiliation to a European party and are summarized below. However, whilst in relation to the latter, there is a distinction between ideological and practical reasons, the reasons for following or not the Commission s Recommendation on promoting the Spitzenkandidaten are mostly practical and strategic. Apart from the obvious reasons for supporting a Spitzenkandidat, namely to extend the principle of democratic accountability to the appointment of the President of the European Commission, via the European Parliament, some national parties decided to make the electorate fully aware of their Spitzenkandidat since the candidate came from their Member State. For example, the German SPD advertised in all campaign materials their Spitzenkandidat, Martin Schulz. This was at least in part because he was not only the Spitzenkandidat for the PES but also a member of the German SPD itself. Consequently, it was easier for the SPD to make the link to the EU level. Similarly, the Austrian SPÖ (also a member of the PES) also advertised Martin Schulz heavily in its campaign materials - although Martin Schulz is not an Austrian national he is Germanspeaking which was believed to help voters to identify with him more easily. Likewise, SYRIZA in Greece also promoted their Spitzenkandidat, partly at least because he was a member of the party. 50 The same methodology in selecting the sample of national parties as earlier is applied (See 23.1). Also in this table, not all Member States are covered as relevant information was missing. 37

50 Enhancing the Democratic Conduct of the European Elections 3 Conversely, there were a number of reasons why political parties did not support a Spitzenkandidat. Firstly, one reason why some national parties decided to not publicize their Spitzenkandidat was because they did not belong to any European party even though they had seats in the European Parliament (i.e. were Non-attached members). Also, some European parties did not put forward any candidates. This is the case for parties belonging to the EP parties MELD and AECR. Thirdly, in some cases, the Spitzenkandidat had not been selected at the time when the campaign material was prepared. This is the case with for instance the CDU in Germany who elected their Spitzenkandidat only shortly before the election campaigns started. There were also instances of where a national party did not support the same candidate as their European party. For instance, the German CDU supported the EPP s Spitzenkandidat, Jean-Claude Junker, only towards the end of the campaign, 51 having supported a different candidate earlier on. 52 Certain national parties did not want to publicise a Spitzenkandidat because they preferred to focus their campaign on national issues. For instance, in the Czech Republic the CSSD ((Czech Social Democratic Party; Česká strana sociálně demokratická), ANO 2011 and TOP09 did not promote a Spitzenkandidat at all until after the European elections. Last but not least, the European parties that were hostile to the EU did not even select a Spitzenkandidat for reasons that are obvious. The CSES survey suggested that most respondents did not think that, overall, political parties had done enough to promote the candidates for the Presidency of the European Commission. This view was slightly more pronounced in relation to the national political parties than the European ones. It was felt that there had been only a very limited effect on voters. 51 Wittrock, P. (2014) Unterstützung für Juncker: Merkel beugt sich dem Druck, vorerst

51 Enhancing the Democratic Conduct of the European Elections 3 39

52 Enhancing the Democratic Conduct of the European Elections Role of Civil Society in promoting the European elections Civil society was actively involved in efforts to mobilise voters in the European elections in their home countries and abroad at the EU and national levels. Below are examples of EU level campaigns followed by examples of campaigns at the national level. This section also analyses the possible effects of these campaigns Civil society initiatives carried out across the EU The EU-wide European Students' Forum AEGEE promoted the Y vote 2014 project, which aimed to encourage young people to get involved and to turn them into important actors of the upcoming European Parliament Elections. 53 The project involved organizing bus tours in Spain and the UK, and youth conventions, to encourage debates on topics important to young people, training events and local activities throughout Europe. Moreover, AEGEE developed a special online voting guide to inform young people, especially first time young voters. The European Anti-Poverty Network (EAPN) - in order to raise the trust in the European Union, and because of their belief that democratically elected representatives are part of the solution to poverty, combatting xenophobia, inequality, etc., the EAPN called on people to make their vote count in the European elections. They included in their campaign manifesto a call on candidates to show their commitment and sign up a pledge that said: If elected, I commit to make the fight against poverty, social exclusion and inequality a priority item on the Parliament and EU agenda by defending the development of an explicit European anti-poverty strategy and an annual hearing of people experiencing poverty to monitor its implementation and progress. 54 The European Disability Forum (EDF) is an independent NGO that represents the interests of 80 million Europeans with disabilities. The role of EDF is important in monitoring all EU initiatives and in proposing new legislation to advance disabled people s rights. In regard to the EU elections, EDF pointed out three main problems for people with disabilities. First, most Member States deprive people with intellectual disabilities of the right to vote. Second, there is a lack of accessibility to campaign materials for persons with disabilities. For example, only one of the televised debates between the Spitzenkandidaten was accessible for people with disabilities. There was also a lack of accessibility of polling stations for disabled persons in some countries. EDF pointed out that these three issues taken together possibly led to a very low participation of disabled persons in the 2014 European elections. EDF also stressed that in conjunction with the CJEU judgment in Spain v. UK (2010) and the COM Disability Strategy , these obstacles need to be addressed. They argued that media, political parties and national authorities need to cooperate to make campaign material, information and polling stations more accessible for people with disabilities

53 Enhancing the Democratic Conduct of the European Elections Civil society initiatives carried out at national level In the Czech Republic, the Association for International Affairs played a significant role in highlighting European issues. It organized an event with five candidates from the strongest political parties and asked them to comment on specific EU related topics. In Estonia, the e-governance Academy completed a project A Guidance to ecitizenship. The main goal of the project was to increase the awareness and motivation and develop skills of young people for e-activeness. However, to reach the main target group and project s goals, the campaign focused on teachers of civic education. In cooperation with Narva College of the University of Tartu, the teachers were trained and equipped with updated, interactive teaching material (containing videos of case-studies, etc) about e- state and e-participation mainly targeted for the use in classroom. In the long term this might be a good preparation for e-voting if it is introduced for European elections. 55 Another example on the role of civil society is COFACE, which collaborated with Eurofound and AGF to organise a European elections debate in Germany "2014: Citizen action for better family policies." The event focused on how to tackle the current political distrust and disengagement of European citizens as well as to motivate voters to get engaged, learn more about the European Union and vote at the EP Elections. 56 In Greece, the civil society organization Interalia organized a project called The Election is coming... to vote or not to vote? A survey was conducted in a period close to the European elections and the findings attempted to highlight: (a) how people see the EU, (b) how they feel about being part of it; (c) how important it is to participate in the EU processes and (d) in which ways this could be done. 57 In Italy, the UEF Italy (Movimento Federalista Europeo - MFE) played an active role during the 2014 European Elections. For example, it organised a campaign event in Rome called European Citizen Convention for European Federation on 5 April 2014 when hundreds of citizens and representatives of civil associations attended a public debate with the aim of increasing awareness on key issues in the campaign. The Italian President Giorgio One of the questions was Are you going to vote for the EP election in May 2014? 58% of the respondents answered positively, 35% were undecided and 4% said that they would not vote. Moreover 55% of the respondents said that their primary purpose to vote in the EP elections is their democratic duty to do so, while 23% claimed that they would be exercising their voting right because they are interested in the result; 11% of the respondents said that they would vote because they support a particular political party. On the question to whether the new role of the European Parliament (2014) in the decision-making process will change, 49% of the respondents said that it will remain the same, 33% said that it will increase a little and 12% said that it would increase a lot. An interesting question was raised in the survey which said How many Members of the European Parliament coming from your country can you name? from the 470 respondents only 13 people were able to name 6-10 members of the EP coming from their countries, 19 people could name 1-2 members of the EP coming from their country, while 33 people were able to name 3-5 members. 41

54 Enhancing the Democratic Conduct of the European Elections 3 Napolitano was also involved, sending a strong political message of support toward the movement. After the elections, the MFE maintained active through other campaign actions such as We Are the European People 58 and Meet your MEPS. In Lithuania there was a good example of a civil society organization White Gloves to strengthen transparency in the democratic conduct of the European elections. 59 There was evidence in the past of election irregularities during the national election campaign in 2012 in some areas (vote buying). Recognising that civil society can help to strengthen transparency, 700 election observers participated in the 2012 national parliamentary elections and subsequently a larger number in the 2014 European elections. In the recent elections, White Gloves succeeded in having votes suspected of irregularities excluded from the count. The organization has also launched a website and a hotline to report on irregularities. Whereas previously most irregularities were identified through observers, the majority of irregularities identified in the EP elections were from people who had heard about the initiative through a campaign to highlight the problem and felt a civic duty to report irregularities. The campaign encouraged people to observe what was happening during the elections and to report any suspicious activities in the interest of promoting greater transparency. The initiative targeted young people who would like to see positive changes. This resulted in stronger civic engagement, with young people incentivized to galvanize their friends and relatives to vote. In Romania, the civil society organization NACAB campaigns to encourage citizens to make use of their European rights within the framework of the European Year of the Citizens There are two ongoing projects that are dealing with this subject matter. First, EPIC-European Participation Initiatives for Citizens running from September 2013 until February 2015 focuses on informing citizens on their right to vote in EU parliamentary elections. It also focuses on mobilising grassroot organisations which shall inform citizens. 60 Second, JoiEU- Joint Citizen Action for Stronger, Citizen-Friendly Europe is a project running from August 2013 until January The aim of this project is among others to increase opportunities for active citizen participation in the political life of the EU to enhance interest in the EP elections. In order to obtain this objective, NACAB collaborates with civil society centers in various Member States to organise twelve transnational debates on common EU issues. These debates shall reach a broad spectrum of society. Therefore, the debates will be available as webinars and on social media. 61 In the UK, New Europeans organised various campaigns in the 2014 European elections to promote EU topics and to encourage people to vote. One example is the Talk Europe Project that aims to generate an interest in and an informed discussion about the Moreover their aim is to encourage civic participation and interest in different elections (including the European one). They worked to promote engagement with young people, which was also linked to civic participation. White Gloves is a civil-society driven initiative started in The main aim of the organization is to raise awareness about the importance of transparency during elections

55 Enhancing the Democratic Conduct of the European Elections 3 opportunities and impact of European citizenship. In addition, New Europeans organised the campaign Vote Denied which aimed to raise awareness of the problems related to voting in the European elections encountered by non-national EU citizens residing in the UK. 62 Civil society organisations relied heavily on the social media as a tool to conduct their campaigns and raise awareness most importantly because it is a cost-efficient tool and it is perceived to be accessible by a wide range of citizens. However, besides being widely accessible, research suggests that accessibility does not mean that it is widely accessed. Social media has a tendency to be an echo chamber, i.e. certain topics on social media only reach the part of society that is already involved in this topic. 63 For instance, politically active individuals tend to gain a lot of information from social media while politically inactive people do only marginally retrieve information on politics from social media. Thus, if civil society organisations use social media as a tool for their EP elections campaigns, the effect is mostly to inform politically active citizens and help them to take decisions that are based on facts. It is less likely that civil society campaigns can mobilise non-voters on a large scale. 3.5 Common Voting Day and Other Issues The fourth of the European Commission s Recommendations designed to enhance the democratic conduct of the elections was that Member States should agree on a common day for elections to the European Parliament with polling stations closing at the same time. As the following table shows, in all but seven EU Member States, the European elections took place on 25 May Table 3.5: Date of 2014 European elections in Member States Member State EU Elections Member State EU Elections Austria 25 May Italy 25 May Belgium 25 May Latvia 24 May Bulgaria 25 May Lithuania 25 May Cyprus 25 May Luxembourg 25 May Croatia 25 May Malta 24 May Czech Rep. 23/24 May Netherlands 22 May Denmark 25 May Poland 25 May Estonia 25 May Portugal 25 May Finland 25 May Romania 25 May France 25 May Slovakia 24 May Germany 25 May Slovenia 25 May Greece 25 May Spain 25 May Hungary 25 May Sweden 25 May Ireland 23 May UK 22 May More specific information on social media, see

56 Enhancing the Democratic Conduct of the European Elections 3 Where there were differences, these largely arose because of national traditions. For example, in Germany, elections of any type take place on a Sunday whereas in the UK, there is a tradition of elections almost always taking place on a Thursday. 3.6 Conclusions Enhancing the Democratic Legitimacy of the European Elections In regard to the first of the Recommendations examined in this section, Member States chose between three different options. First, some Member States took non-legislative measures and informed parties about the existence of Commission Recommendations via different channels. Second, other Member States took legislative measures and implemented electoral laws that allow national parties to mention the affiliation to a European party on the ballot paper. A few Member States did not take any measures at all in response to the Recommendation and argued that it is the responsibility of the political parties to raise awareness of their European affiliations. With regard to the Commission s second Recommendation, which was directed towards national and European parties, the actions taken varied across parties and Member States. Some national parties made the electorate fully aware of their affiliation to a European party. Most parties, however, decided to mention the affiliation only partially. As has been shown, the reasons for that were mainly practical rather than ideological. A small number of parties did not mention their affiliation at all. This was mainly the case with parties that are hostile to EU membership. In relation to the third Recommendation, while some European parties did not put forward a candidate at all, other EU parties put a substantial effort into promoting their candidate. This involved a wide variety of campaigning methods, i.e. a campaign trail, social media, events, debates and other traditional media campaigns such as posters, etc. At the national level the picture is more fragmented. Some parties decided to fully advertise their Spitzenkandidat by mentioning them in all the different campaigning channels. Other parties decided to only partially advertise their Spitzenkandidat which resulted in a lower profile for the Spitzenkandidaten amongst the less informed voters. Some parties did not advertise the Spitzenkandidat at all. At the aggregate level, it would be hard to conclude that Spitzenkandidaten had a profound effect on the European elections turnout, as some expected (and many hoped). The decreasing trend of the turnout was not reversed but it was brought nearly to halt. The available evidence suggests that the Spitzenkandidaten made some difference (i.e. higher turnout) in the country of their birth and, secondly, that in the countries they visited and in which they strongly campaigned. For instance, both in Austria and Germany the awareness of the Spitzenkandidaten was high since the main candidates Junker and Schulz are German-speaking and in the latter case a German national. This seems to have raised more interest in the European Parliament elections in German speaking countries. In general, it can be concluded that based on the evidence of the 2014 European elections, the system of Spitzenkandidaten is a promising tool for encouraging a higher turnout. An indication of the success of the new procedure is the fact that in countries most closely associated with the Spitzenkandidaten, the turnout of voters increased compared with Where the Spitzenkandidaten were particularly active, the 44

57 Enhancing the Democratic Conduct of the European Elections 3 campaign was also more European. To the extent that there were shortcomings with regard to the effects in 2014, this can be possibly attributed to the profile of candidates 64 and the fact that it was not obvious that they would actually become the next Commission President. In 2009 everything was left to the national parties whereas in 2014 the European political parties played a bigger role. This focused on promoting the Spitzenkandidaten. The wider European parties were not especially active or effective in campaigning and the effort to promote the Spitzenkandidaten therefore mainly involved the European Parliament s Political Groups that form the core of the wider parties. The Commission s Recommendations aimed at Enhancing the Democratic Conduct of the 2014 European Elections had an overall positive effect. Despite the economic downturn, there is evidence that the 2014 elections were more European in their tone than the previous elections in The role of the European parties in putting forward a candidate for the European Commission presidency was a significant contribution to this. On the other hand, for various practical and in some cases ideological many political parties did not do more to promote awareness of their affiliation to a European party. Overall, the research suggests that actions taken by European and national parties and the national authorities to enhance the democratic conduct of the elections, and their European nature were not enough. In particular, the measures taken were only visible to citizens who were likely to vote in any case and not to others who were less inclined to do so. The long-term objective of enhancing the democratic conduct of the elections is to increase the turnout of the European election. In this context it has to be noted that the European dimension in the 2014 election was strengthened through the existence of the Spitzenkandidaten. The fact that turnout did not increase might suggest that probably not enough was done by the political parties to promote awareness of EU topics. However, this might also suggest that it is not merely the absence of a European dimension to the elections that caused the decline in the turnout. Other factors need to be considered as well. For instance, a report produced by the UK s Political and Constitutional Reform Select Committee 65 concludes that citizens only vote if they think that their vote matters The two top candidates had little appeal and as explained later in the media section, they did not have too many controversial differences. Therefore, they might have appeared less interesting An example given in the report showing that people vote if an issue matters to them is the Scottish Referendum. It concerned Scottish people greatly in various aspects of their lives (i.e. politically, economic, culturally) and resulted in a turnout of 85%. The relatively low turnout in EU elections suggests that people do not feel that their vote matters i.e. they do not feel that their vote has a direct and immediate impact on their lives. On the contrary, people feel that the EU is regarded as something far away that is not related to their lives. This perceived distance between the civil society and the EU has been termed proximity deficit. 45

58 Enhancing the Democratic Conduct of the European Elections 3 Thus any initiative to increase voter engagement needs to address this fundamental issue - does the vote matter? In this respect, if the present Commission's actions are perceived by European citizens as having had a significant impact on issues of relevance to them, the importance of the Spitzenkandidaten as a way of engaging voters is likely to play a much greater role in the 2019 election. Political campaigns in 2019 will be able to focus on evaluating the present Commission's action and there is more likely to be a feeling that the vote in 2014 had a real (positive or negative) impact on political developments at the EU and national levels. As such, it could be that the real effects of the Spitzenkandidaten (in conjunction with other measures to raise awareness on EU issues) will be more apparent in the coming elections. 46

59 Role of the Media in the 2014 European Elections 4 In this section we examine issues in the Commission s terms of reference relating to the media coverage of the 2014 European elections. 4.1 Restrictions on freedom of expression during an election period Article 11 of the European Charter of Fundamental Rights 67 guarantees freedom of expression and information. It states that: Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers. The freedom and pluralism of the media shall be respected. From the research we have conducted it is clear that freedom of expression was largely guaranteed in all Member States. The responses to the CSES survey support this conclusion: 67 The European Charter of Fundamental Rights 67 brings together in a single document the fundamental rights protected in the EU. The Charter contains rights and freedoms under six titles: Dignity, Freedoms, Equality, Solidarity, Citizens' Rights, and Justice. Proclaimed in 2000, the Charter has become legally binding on the EU with the entry into force of the Treaty of Lisbon, in December

60 Role of the Media in the 2014 European Elections Political Parties Access to Media during the 2014 European elections In general, there were no restrictions for political parties to gain access to the traditional media during the 2014 European elections. In most Member States, the political parties had the right to free advertising space on public radio and television stations, although the system of allocating time varied from one Member State to the other. Legal frameworks were in place in all cases to ensure fair access to the media. These conclusions are supported by the responses to the CSES survey: In most EU Member States, there are some restrictions on media coverage of political parties and election campaigns. These restrictions either determine the amount of time to allocate to political parties according to their representation in the parliaments, or give each of the parties the same amount of time. A summary of the legal frameworks in EU Member States relating to media coverage of the European elections is provided below (a more detailed assessment is provided in the appendices to this report). In Austria, the activities of broadcast and print media are mainly regulated by the Federal Law on the Press and Other News Media (Media Act), the Federal Act on the Austrian Broadcasting Corporation (ORF Act) and the Federal Law Stipulating Provisions for Private Televisions (Private Television Act). None of the above laws, however, provides any specific provisions regarding media coverage of elections or media access for political parties and candidates

61 Role of the Media in the 2014 European Elections 4 In Belgium there is apparently no regulation to the access of political parties to the media and broadcasting, so there was no limit to the amount of time that individual candidates for the European elections, as well as for the Federal/regional elections, could be given with regard to media coverage. All mainstream political parties have broadly equal access to the media (however, as mentioned above, equal media airtime is not guaranteed by law). Minor parties do not have equal access to media, as the main TV stations reserve the right to exclude such political parties from broadcasts. 69 Printed media also offers broad and mostly balanced coverage of political parties. In Bulgaria, II and III of the Bulgarian Electoral Code regulate access to media and election material. V regulates the election campaign coverage, and prohibits commercial advertising during the election campaigns. It regulates the order of presentation of candidates, which is determined by the Central Election Commission, and regulates the opening and closing of campaign messages, news briefs, other spots, debates and the allotment of time by other electronic media. 70 All media outlets have to provide time/space for paid political advertisements under equal conditions to all parties. The Bulgarian national TV and radio are also obliged to organize and broadcast campaign debates, free of charge, for at least four hours on national channels and for at least one hour on regional television and radio stations. 71 In Croatia, The Election of Representatives to the Croatian Parliament (LERCP) and the 2003 Rules of Procedure for Electronic Media with National Concession in the Republic of Croatia during the Election Campaign (amended in 2007) regulate media coverage during the official campaign period. 72 The legal framework requires that the national broadcast media, both public and private, provide coverage of all political parties on an equal basis. Financing of the electoral campaign is regulated by the provisions of the Political Activity and Election Campaign Financing Act (Official Gazette Narodne novine, No. 48/13). 73 The national electronic media, both public and private, are legally required to provide equal coverage of all candidates and parties. 74 In Cyprus access to the media for different parties is regulated by the Law of 1998 on Radio and Television Stations and its Regulations of 2000 on Radio and Television Stations. However, equal treatment is not explicitly stipulated by the law and in general, access to electronic and online media is not regulated August-27-29/Responses_NVs_2012/ _Bulgaria_English_4.pdf

62 Role of the Media in the 2014 European Elections 4 In the Czech Republic, the Broadcasting Act, Electronic Communications Act, Czech Television Act, and Czech Radio Act regulate the activities of media during an election campaign. Political parties and movements (and coalitions) are eligible for indirect public funding, in the form of free broadcasting time and space to put up posters during the election campaign period ( 16, Law No. 247/1995 on Elections to the Parliament). Article 59 of the Act of Law No. 62/2003 Coll. of February 18th, 2003, on Elections to the European Parliament and on Change of Selected Acts of Law stipulates that all parties and movements, which have registered candidates lists for elections to the Czech parliament, are entitled to free broadcasting time. An amount of 14 hours on the Czech Radio and 14 hours on Czech Television is distributed equally between the running parties, movements and coalitions by drawing lots. 76 In Denmark, all Danish broadcasters are subject to the Radio and Television Broadcasting Act while the print and broadcast media in Denmark are regulated by the Media Liability Act of This law specifies that the content and conduct of mass media shall be in conformity with sound press ethics. 77 All political parties that plan to take part in elections have the right to equal programming time on the radio and on television. In Estonia the National Broadcasting Act regulates the public service broadcaster, while the recently enacted Media Services Act regulates the public and private broadcasting services. 78 There is no licensing of print media. The Election Act establishes the legal basis for the Internet voting and is supplemented by an NEC decree and operational guidelines that provide more detailed instructions for some stages of the process. In Finland, the media law covers state-owned media and stipulates that state media need to adhere to a number of criteria with regard to political coverage. The Act on the Exercise of Freedom of Expression in Mass Media regulates publishing and sets conditions for the right to reply and correction, and the Act on Radio and Television Activities regulates TV programme quotas. 79 The Finnish Communications Regulatory Authority (FICORA) is the media regulatory body that supervises broadcasters compliance with the media legislation, in particular Act on Television and Radio Operations, and issues broadcasting licenses. In France - The Conseil supérieur de l Audiovisuel (French broadcasting authority, CSA) has established rules on broadcast media coverage of general election campaigns. Equal and fair coverage of candidates and their parties, in terms of voice time and total airtime is required. In Germany political campaigning is regulated by federal legislation. Article 5 of the Political Parties Act (Parteiengesetz, PPA) requires that where a public authority provides facilities or other public services for use by one party, equal treatment must be accorded to all parties and that "media time is allocated both equally and proportionally (share of (1) of the Media Liability Act 78 The Media Services Act, passed in 2010, replaced the 1994 Broadcasting Act. There were no major changes regarding the regulation of the election campaign

63 Role of the Media in the 2014 European Elections 4 seats). According to the Constitutional Court, equal opportunities imply that all parties running for a specific election receive some media time. In Greece access to the media is regulated by the General Election Law that also applies to European elections and allocates broadcasting time to political parties that participate in the elections. Print media operates largely on the basis of self-regulation but radio and television are under the direct control of the state. 80 This control is carried out primarily by the National Council for Radio and Television (NCRTV). According to the 1998 law during the election campaign (3 months period), airtime must be allotted to parties according to their respective share of parliamentary seats. In Hungary media conduct during the election campaign is regulated by the Election Procedures Act, the Freedom of the Press, the Fundamental Rules of Media Content Act and the Act on Media Services and Mass Media (this was revised on 24 May 2012). 81 Act XXXVI of 2013 on Electoral procedures provides for fair and balanced political coverage for all public media is imposed by the media regulations with regard to the number, appearing order, timeframe and time of broadcast of political advertisements offered to nominating organisations that put forward candidates and the independent candidates. In case of joint candidates the nominating organisations are entitled jointly to order political advertisement. Private media is not entitled to broadcast paid political advertisements but can provide free airtime to all electoral contestants on an equal basis, although none chose to do so in the 2014 European elections. 82 In Ireland, the Electoral Act 1997 (amended in 2001), regulates access to media during elections. Broadcast time is allocated to political parties on the basis of the first preference results from the previous general elections". 83 The new Broadcasting Act came into force in 2009 and established the new Broadcasting Authority of Ireland. The Broadcasting Act 2009 provides for a significant expansion of the role and responsibility of the old Broadcasting Commission of Ireland, which had no responsibility for Public Service Broadcasting, namely RTE. The Italian electoral laws on political campaign (especially Law No. 28 of 22/02/2000)84 aim at regulating the allocation of broadcasting time to each national political party, which varies according to the votes received during the previous election. This legal framework is called par condicio.85 In Latvia Chapter II of the Pre-election Campaign Law from December 2012 regulates Broadcasting Time in Electronic Mass Media Programmes. 6 addresses the Right to the State Ensured Free of Charge Broadcasting Time for the list of election candidates. The funding for pre-election broadcasts in the public remit is granted in accordance with the provisions and procedures laid down in 71 of the Electronic Mass Media Law. The Pre-election Campaign Law ensures that the Latvian state provides broadcasting time 80 Article 15(2) of the Constitution Italian Legal framework: legge 22 febbraio 2000, n

64 Role of the Media in the 2014 European Elections 4 free of charge for campaign broadcasts on TV programmes in both the public and commercial electronic mass media. In Lithuania the electoral regulations comprises the 1992 Constitution, which was last amended in 2004, the election law 7 as amended last in March 2012, the 2002 Law on the Central Electoral Commission as amended in March 2012, the 2004 Law on the Funding of Political Parties and Political Campaigns, and Control of Funding as amended in 2012, and the Law on Political Parties as amended in The legal framework is based on the principle of ensuring equal access time to all political parties to public broadcasting (TV, radio). In Luxembourg, elections are primarily regulated by the Constitution (adopted in 1868 and last amended in 2011), a unified electoral law (adopted in 2003 and last amended in 2011) and a number of other regulations. The electoral law regulates different aspects of the organization of parliamentary, local and European Parliament elections. However, the law does not provide any regulation of the election campaign and its coverage in the media. The coverage of the electoral campaign by the media is not regulated by the law and is agreed among the media and political parties in a process co-ordinated by the Ministry of State s Information and Press Service. In Malta the Broadcasting Authority oversees political broadcasts during electoral campaigns and grants access to all political parties and independent candidates contesting elections. Article 13(4) of the Broadcasting Act (chapter 350 of the Laws of Malta) stipulates that it is a duty of the broadcasting authority to ensure that political broadcasts (including political spots) w fairly apportion facilities and time between the different political parties represented in Parliament. In Netherlands, Article 39g of the Media Law requires that political parties with one or more seats in either chamber of the States General be allotted time on the national broadcasting stations during the parliamentary term, provided that they participate in nationwide elections. The Commission for the Media ensures that political parties are given equal media access free from government influence or interference (Article 11.3). In Poland Chapter 10 of the Act of 23 January 2004 on Elections to the European parliament regulates election campaign and access to the media during the EP elections. The Act specifies that candidates to the EP are proposed by party election committees or voter groups. Each committee can propose one list of candidates in each constituency. Article 78 states that the Election committees have the right to free airing of their election programs on radio and television in the form of political broadcasts and political programs. The total air time for broadcasting election material is 1) on nationwide channels 15 hours broadcasted by the Polish Television Joint-stock Company from 17:00 to 23:00 and 20 hours by the Polish Radio Joint-stock Company; 2) on every regional channel - 10 hours broadcasted by the Polish Television Joint-stock Company and 20 hours by regional broadcasting companies, set out in Article 26, paragraph 2, subparagraph 2 of the Act of 29 December 1992 on Radio and Television. Airtime for broadcasting election materials is divided into equal parts among the election committees. 52

65 Role of the Media in the 2014 European Elections 4 In Portugal political parties have the right of access to public and private radio and television stations for election campaigning. The rules are laid down by the law on elections to the Assembly of the Republic (Law no. 14/79 of 16 May 1979). In Romania campaign coverage is regulated by the Election and Broadcast Laws and the National Audiovisual Council (CNA). They ensure that candidates and parties enjoy equal access to the media and other means of communication. Art. 38 (1) from Law no 35/2008 on the election of the Chamber of deputies and the Senate, political parties, political alliances, electoral alliances, the national minority s citizens participating in elections for the Chamber of Deputies and Senate and independent candidates guarantees free access to public radio and television services in proportion to the number of candidates proposed. 86 Moreover, electoral law stipulates that public broadcasters provide free coverage to parties and that private broadcasters charge the same rates to all contestants. Furthermore, the broadcast Law requires that the airtime offered to parties by private broadcasters must be congruent with the time allowed by public broadcasters. In Slovakia media legislation follows from Law No. 308/ on Broadcasting and Retransmission (which was amended). Law No.331/2003 Coll. on Elections to the European parliament regulates the European parliament elections campaign. 88 The law stipulates that Slovak Radio 89 and Slovak Television 90 shall allocate each contesting party or coalition not more than thirty minutes. The maximum time for time political broadcasting is 5 hours. All costs of party political broadcast and time schedules should be identical and fair for all political parties. Other relevant laws are Law No. 610/2003, on Electronic Communications; Law No. 211/ on free access to information and on amendments and supplements to certain acts (Freedom of Information Act), which regulates campaigns in the media before various elections (presidential, parliamentary, regional self-government and local elections) In Slovenia the legal framework for the media coverage of the election campaign includes provisions of the Elections and Referendum Campaign Act (ERCA), as well as rules for public broadcasters contained in the Radio and Television Corporation of Slovenia Act (RTV Act). The European elections are regulated by the Electoral campaign Act. Law no.40/2004 regulates and ensures equal rights to candidates and political parties during the election campaign. ERCA also regulates paid and free coverage in public media and the publication of opinion polls. 92 The public broadcaster also has to offer parties free airtime during the campaign period based on their representation in the National Assembly and the European Parliament Act No. 308 of 14th September 2000 on Broadcasting and Retransmission and on Amendments of Act No. 195/2000 Coll. on Telecommunications 88 Amended by Law No. 515/2003, 324/2004, 464/2005, 445/2008 and 599/2008) 89 Act No. 619/2003 Coll. on Slovak Radio 90 Act No. 16/2004 Coll. on Slovak Television 91 Slovakia, Law No. 211/2000 Coll. on Free Access to Information (Zákon č. 211/2000 Z.z. o slobodnom prístupe k informáciám),

66 Role of the Media in the 2014 European Elections 4 In Spain the Central Election Commission (CEC) is the primary media regulatory body during the electoral campaign period. The CEC allocates free airtime to parties in the public media, based on the proposal of a Radio and Television Committee. Law No. 5/1985 of 19 June (General Electoral Act)93 regulates access to public television and public radio networks during electoral campaigns. It also determines the ballots forms for elections to European Parliament. It provides that during elections to the European Parliament, coverage of political parties should be based on the total number of votes obtained by each party (Article 62). For example, 30 minutes of media coverage is allowed for parties, federations and coalitions that obtained between 5 and 20% of the votes in the last elections, or 40 minutes if the share was over 20%. In Sweden media legislation is based on the tradition of press freedom. Rights to freedom of press are enshrined in the Freedom of Expression Act, with further details being set out in the Radio and Television Act. The coverage of election campaigns by media outlets is largely unregulated. The Swedish Radio and TV Authority is the main agency that legally regulates broadcast media. In the United Kingdom relevant legislation includes sections 319(2)(c) and 320 of the Communications Act 2003, the Broadcasting Act of 1996, (establishes two bodies to license and regulate private broadcasting in the UK, the Radio Authority (RA) and the Independent Television Commission (ITC), the Office of Communications (OFCOM). Broadcasters should have regard to relevant sections of the Representation of the People Act 1983 (as amended) ("RPA"). This is mentioned in particular in sections 66A, 92 and 93 (which is amended by section 144 of the Political Parties, Elections and Referendums Act 2000).94 Paid TV advertising is prohibited for political parties who can only advertise in newspapers. However, major parties are granted a certain amount of free time for TV advertising, a concession that is not available to minor parties. Coverage on television has to be fair and balanced, and is policed by Ofcom. Broadcasters are required to be balanced in their coverage of parties, especially at election time. No such restrictions exist for the print industry. The table below provides an overall summary of the legal frameworks for key aspects of the media coverage of the 2014 European Parliament elections

67 Role of the Media in the 2014 European Elections 4 Table 4.1: Summary of legal frameworks media coverage of the 2014 European Parliament elections Key (freedom of expression): Fully guaranteed: Partially guaranteed: Not guaranteed: Country Austria Rules of access to media during the election campaign Federal laws on Press and Media Act, Federal Act on the Austrian Broadcasting Corporation, Private Television Act Rules on the allocation of broadcasting time during election campaigns Federal laws on Press and Media Act, Federal Act on the Austrian Broadcasting Corporation, Private Television Act Belgium No Particular laws No Particular laws Bulgaria II and III of the V of Electoral Code Bulgarian Electoral Code Law of 1998 on Radio and Television Stations and its Law of 1998 on Radio and Television Stations and its Cyprus Regulations of 2000 on Radio and Television Stations Regulations of 2000 on Radio and Television Stations Croatia Czech Rep. Denmark Election of Representatives to the Croatian Parliament/2003 Rules of Procedure for Electronic Media with National Concession in the Republic of Croatia during the Election Campaign Election Law/ Broadcasting Act, Electronic Communications Act, Czech Television Act, and Czech Radio Act. Radio and Television Broadcasting Act Electronic Media Act (OG 153/09; 84/11 Article 59 of Law No. 62/2003 Act of Law 247/1995 Media Liability Act of 1998 Is freedom of expression guaranteed? Estonia Media Services Act Media Services Act Finland Act on the Exercise of Freedom of Expression in Mass Media/ Act on Radio and Television Activities Finnish Communications Regulatory Authority 55

68 Role of the Media in the 2014 European Elections 4 France Germany The Conseil supérieur de l Audiovisuel (French broadcasting authority, CSA) Political Parties Act (Parteiengesetz, PPA) The Conseil supérieur de l Audiovisuel (French broadcasting authority, CSA) Political Parties Act (Parteiengesetz, PPA) Greece General election law General election law Elections Procedures Act/ Article 147 of Act XXXVI of Hungary Freedom of the Press/Fundamental Rules of Media Content Act/ Act on Media Services and Mass Media Elections Procedures Act Hungarian Media Law on electoral Procedure Ireland Electoral Act 1997 Electoral Act 1997 Italy Electoral Law Law No. 28 of 22/02/2000 Latvia Pre-election Campaign Chapter II of the Preelection Campaign Law Election Law Election law, Law on the Funding of Political Parties/Political Lithuania Campaigns/Control of Funding and the Law on the Provision of Information to the Public. Luxembourg Electoral Law No specific laws Malta CHAPTER 354 Article 13(4) of GENERAL ELECTIONS ACT Broadcasting Authority Media Act 2008/ Media regulation 2008/ individual media outlets decide themselves how Netherlands broadcasting time decrees (zendtijdbesluiten much attention they pay to political parties and candidates Poland Portugal Romania Election Code Act of 23 January 2004 on Elections to the European Parliament Law no. 14/79 of 16 May 1979 Election and Broadcast Laws and the National Audiovisual Council Article of Act of 23 January 2004 Articles 252, 253 and 254 of Chapter 10 of the Polish election code Law no. 14/79 of 16 May 1979 Law no 35/

69 Role of the Media in the 2014 European Elections 4 (CNA). Law No. 308/20095 Law No. 619/2003 Law No. 16/2004 Law No.331/2003 Slovakia 24 of the Act 333/2004 Law No. 220/2007 No. 610/ /2007 The Elections and Referendum Campaign Law no. 40/2004 (amended by ZVPEP-B 109/2009) Slovenia Act Radio and Television Corporation of Slovenia Act Spain General Electoral Act Law No. 5/1985 Sweden Swedish Radio and TV Broadcasting authorites Authority Communications Act 2003 sections 319(2)(c) and 320 of the Communications Act UK Broadcasting Act 1996 Office of Communications (OFCOM) 2003 Turning to other more general issues, although traditional media coverage of the European elections and political parties is regulated in most EU Member States, this is not the case with social media. In Latvia, for example, there are restrictions on how political parties expenditure on media coverage (e.g. only half the political campaign budgets can be spent on TV programmes). All expenditure must be published online by the Electoral Commission. However, social media is not regulated. Even on Election Day, messages were being published on social media sites run by political parties and this was not monitored and controlled. Although there were rules in Latvia to say that campaigning should stop a day or two before the elections, SMS messages were sent out the day before the elections, which was illegal but difficult to prevent and monitor. There were also issues concerning the role of the public and privately-owned media in the European elections. In Austria, for example, there are no specific rules on access to the media. While it is not allowed to publish political adverts on state-run TV programmes, all parties can have campaigning spots in programmes produced by the private TV companies with no restriction on the length of broadcasts. However, political parties have to pay for such coverage and according to the feedback from our interviews, some smaller parties felt discriminated against and not given appropriate coverage because they do not have the required financial resources. Against this, it was argued that 95 Act No. 308 of 14th September 2000 on Broadcasting and Retransmission and on Amendments of Act No. 195/2000 Coll. on Telecommunications 57

70 Role of the Media in the 2014 European Elections 4 TV coverage plays a less important role because of the growing importance of the social media. 4.3 Media coverage of EU issues in the European election campaign The 2014 European election differed significantly from the 2009 election mainly due to four factors. These four factors might not only have influenced the behaviour of voters but also how the media covered the election: The role of social media increased in comparison to 2009; 96 The introduction of the Spitzenkandidaten provided a focus point for media coverage; The financial and economic crisis was a key issues in most Member States and this was far more an issue in 2014 than 2009 given its growing intensity; The debate on the future of Europe by 2014 this had also become more intense and was a feature of media coverage in many Member States. 97 Overall, feedback from the research suggests that the media had a positive impact on the European elections in 2014 by raising awareness of EU issues and encouraging citizens to vote. However, there were criticisms that the media gave to much attention to national issues in some countries and this is reflected in the responses to the CSES survey. 96 See section in this report on social media Mirte-van-den-Berge.pdf 58

71 Role of the Media in the 2014 European Elections 4 In the past years a comprehensive body of knowledge has accumulated on how the media cover the EU European elections. 98 In considering the extent to which the role of media changed in comparison to 2009, it is helpful to consider various features, namely the type of media used, the visibility of European elections news and the tone of the coverage. In regard to the type of media used, statistics show that traditional media such as TV, newspaper (including online news) and radio play a substantial role in comparison to Moreover, the 2014 European elections saw a significant increase in the use of the social media (explained in 4.4. below). Turning to the visibility of the European elections, in 2009, on average about 16% of the television news coverage in the three weeks leading up to the elections dealt with EU issues or the European elections specifically. 100 At the time of writing, there was no similar analysis for the 2014 but as noted earlier the visibility of the European elections was particularly strong in the countries of origin of the two main Spitzenkandidaten : Jean-Claude Junker (Luxembourg) and Martin Schulz (Germany) while in other countries media coverage was less intense. In terms of the European dimension, as a result of the economic and financial crisis, campaigns at national level were much more 'Europeanized', especially in eurozone countries. The increased prominence of eurosceptic parties broadened the election debate beyond the realm of the European election campaign and into the national political arena in many Member States. 101 This increased visibility of the European elections is reflected in statistics. In 2009, only 53% of EU citizen felt sufficiently informed about EU politics while in 2014, 57% of EU citizens felt informed about EU politics See: De Vreese, C. et al. (2010) 2009 European Election Campaign Study, Scenario I: Data and Documentation, Amsterdam: University of Amsterdam. De Vreese, C. H. & Tobiasen, M. (2007). Conflict and Identity. Explaining turnout and anti-integrationist voting in the Danish 2004 elections for the European Parliament. Scandinavian Political Studies, 30(1), Hollander, B. A. (2007). Media Use and Political Involvement. In R. W. Preiss, B. M. Gayle, N. Burrell, M. Allen & J. Bryant (Eds.), Mass Media Effects Research: Advances Through Meta-Analysis (pp ). Mahwah, N.J.: Lawrence Erlbaum Associates. Schuck, A. et al. (2011) Media visibility and framing of the European Parliamentary Elections 2009:A media content analysis in 27 countries. In: Maier, M., Strömbäck, J. & Kaid, L.L. (Eds.). Campaigning in Europe: Strategies, Effects, and Media Coverage in Parliamentary Elections. Ashgate Publishing Limited. 99 See section in this report on social media. 100 Schuck, A. et al. (2011) Media visibility and framing of the European Parliamentary Elections 2009:A media content analysis in 27 countries. In: Maier, M., Strömbäck, J. & Kaid, L.L. (Eds.). Campaigning in Europe: Strategies, Effects, and Media Coverage in Parliamentary Elections. Ashgate Publishing Limited (See Figure 4, for details, range 7-48%) Mirte-van-den-Berge.pdf c_annex_en.pdf (p.110) 59

72 Role of the Media in the 2014 European Elections Media coverage of the European elections in Member States In Austria, all types of media showed an interest in EU topics during the elections and covered the elections thoroughly. As was pointed out by one interviewee, the political parties and media focused in 2014 on the direction the EU should take, while in 2009 the topic was mainly yes or no to the EU. The interest of the media in the elections was driven by the fact that two of the leading Spitzenkandidaten were German-speaking. The TV debate between Junker and Schulz was broadcasted in ORF which collaborated with the ZDF in Germany. While the market share in Germany was only 5.8%, the market share in Austria was more than 10%. This is an interesting result because (although German-speaking) none of the candidates was Austrian. Nevertheless, the interest in Austria was higher in the debate than in Germany. The same trend was not, however, reflected in the election result. While in Austria 45.4% people voted, in Germany 48.1% voted in the European elections. In Germany interviews with both print/online media as well as TV broadcasters illustrate that media played an overall positive role during the 2014 elections in raising awareness of EU topics. One interviewee stressed that the European election was taken very seriously by print and online media as well as TV. Tabloid media discussed the elections in a more unsophisticated manner or only marginally. It has been stated that media was even more active in discussing the European elections than the political parties themselves. In 2014, there was an increase in media coverage in comparison to the 2009 elections because of the existence of the Spitzenkandidaten and the fact that they were both German-speaking. Although media was generally more interested in the European elections than in 2009, the TV debates have not been very popular. The ZDF debate between the two Spitzenkandidaten on the 8 May 2014 took place at prime time and had an audience share of 1.7m and a market share of 5.8% whereas a chancellor debate usually results in 2-3 times higher audience share and around 15% of market share. As one interviewee pointed out, the reason for the relatively poor results is that most people still doubted the relevance of the European Parliament while national elections do not raise such questions. Also, heated debates between the candidates did not take place leading to less interest by the public. Media was relatively inactive in the Netherlands. There were some programmes on television with debates about the EU issues but otherwise little coverage of the election campaign. In Latvia a considerable amount of attention was given in the media to highlighting European issues. However, the focus tended to be on issues of Latvian national interest rather than on strengthening awareness of the manifestos of the European political parties. In Portugal, between 28 April and 2 May, the public radio broadcasted a series of debates on the 2014 European elections 103 in which several guests discussed various themes and raised issues about the present and future of the European Union. The first debate had the participation of the different candidates of the main political parties. It

73 Role of the Media in the 2014 European Elections 4 was possible to listen to these debates live and also at a later stage on the website of the radio. In addition, the Portuguese 24-hour news channel, RTP Informação (available on basic cable and satellite), operated by the Portugal s public service broadcasting organisation Rádio e Televisão de Portugal, S.A. (RTP) broadcast the debate between the five candidates for the Presidency of the European Commission that was viewed by 6,800 people. 104 An interesting approach was adopted in Romania where a video entitled Ambassador of a European Romania was disseminated. The video, broadcast both on Romanian massmedia and the PEA s online platform, was produced to help Romanian citizens understand how the European Parliament functions and its importance as an institution within the EU, as well as the right and responsibility of every Romanian to elect their representatives. Alongside this video, the ambassadors from a number of other EU Member States agreed to film a video testimonial in which they shared some of their thoughts with voters. The campaign was active across online media, with most embassies disseminating the video via the social media channels that they make use of. In Spain, campaign events of political parties were widely covered by the media, both public and private. There were also a number of televised debates before and during the campaign (both in public and private television). Most of these debates took place with the participation of representatives of minority parties, but there were also debates with the participation of head of lists of the majority parties and a single face to face debate between the two leading candidates (candidates PP and PSOE) in the Spanish public television (TVE) which was seen by nearly 2 million people with 10.2% audience share. 105 There was also an online debate organised by the digital diary Público. Previously, the channel Canal 24 horas (from the TVE) broadcasted the debate among the five candidates for the Presidency of the European Commission that was seen by 89,000 people (representing only 0.6% of the audience). 106 The Swedish campaign was similar to the Austrian one: it was mainly about the direction the EU should take. In the UK, the top candidates debate hosted by the EBU (European Broadcasting Union) was broadcast in the UK on the BBC s Parliament channel. This is a terrestrial channel available in the UK on freeview. There are no viewing figures available for the programme (this is true of most of the programmes broadcast on the Parliament channel). The Parliament channel would only get recordable viewing figures for something of interest like the live House of Commons vote on whether to go to war in Iraq. It is worth pointing out, however, that the BBC did encourage viewers by having the broadcast trailed on the Today Programme (Radio 4), which has a huge daily audience, where it was mentioned on a couple of occasions by the presenter Justin Webb. The broadcast of the debate was also trailed in radio bulletins

74 Role of the Media in the 2014 European Elections 4 While the UK media made an effort to get people interested in the TV debate, the public was not particularly mobilized. The relatively low level of interest in the programme may have been due to the fact that there was no British Spitzenkandidat and the UK national parties tended to downplay the European perspective in the election. Two of the speakers in the debate did not wish to speak in English and translation does create difficulties in terms of audience attention. There was no problem of geo-blocking. The BBC s Democracy Live site can be accessed anywhere. Recently all the output of the Parliament channel has been made available on iplayer, the BBC s catch-up facility where programmes can be viewed for 30 days after they are broadcasted. However, in May 2014 this was not the case and only a small selection of the BBC Parliament channel s output was on iplayer. This also affected the European elections debate broadcasted by the BBC Content of Media Coverage in European Elections In terms of tone of the European elections coverage, in 2009 the coverage was largely neutral. 107 As already stated in the previous section, the financial and economic crisis had a significant impact in the 2014 elections. While it increased the visibility of EU issues in the media it also led to a stronger and more negative polarization of the coverage than in According to a Eurobarometer survey, European citizens felt marginally more encouraged to vote by the media in 2009 (67% felt encouraged) than in 2014 (65% felt encouraged). 108 The economic crisis played a crucial role in the 2014 European elections campaign in several countries. In all the bail-out Member States, the debate was whether the EU (Troika) should limit national sovereignty; in the debtor Member States (e.g.. Germany, NL), the debate focussed on the burden of financial transfers. In other countries (UK, France, NL, Sweden), the issues of immigration played an important role. The European election campaigns lasted around four weeks in most Member States which seems to be relatively short to raise real interest. In several countries, Eurosceptic parties started their campaign ahead of the other parties. Their campaigns were often also more intensive than other parties' campaigns (e.g. in the Czech Republic, France and the UK). 109 The fact that the campaigns were longer and more intense resulted in more extensive coverage in the media in these countries. In Germany two representatives of different media organisations mentioned that media showed much more interest in the European elections than in Nevertheless, it was mentioned that there is a discrepancy between the perception of the EU of media correspondents in Brussels and the German home media. While the Brussels offices of some German media organisations produce balanced articles on events at the EU level, they are sometimes difficult to sell to the home media AFCO_NT%282013% _EN.pdf c_annex_en.pdf (p.118) Mirte-van-den-Berge.pdf 62

75 Role of the Media in the 2014 European Elections 4 In general, it can be observed that while the importance of social media grew, TV and printed/online press were still the most important media for covering the European elections. Overall, the 2014 European elections were more visible than in However, this coverage was often of a negative nature. There was also a variation between countries, media types, and specific outlets. 4.4 Role of the Social Media in the Election Campaign Social media when used as a political campaigning tool mainly involves Facebook and Twitter. 110 Social media is used by the media, by political candidates/ political parties, by political activists and the general public. Although social media provides vast amounts of unstructured text about political issues, it is possible to differentiate between three broad categories of campaigning: Coverage of campaign events - for example, the debates held by the Spitzenkandidaten and MEPs standing for election; Reporting on EU topics this included issues, for example, to do with the role of the EU in the economic crisis; Reporting about political parties and candidates more general coverage of candidates and parties, e.g. their election manifestos. The European political parties made increased use of social media in the 2014 European Parliament elections in comparison to The social media had already been used quite extensively in 2009 but this increased still further in An example is the European Parliament Facebook page which attracted 45,000 visits just before the 2009 elections compared with 1.5m just before the 2014 elections. During the period from 31 March to the 2014 elections, there were 1m tweets using on #EP2014. Some European parties were more advanced than others in using the social media but there is no obvious pattern. Use of the social media by European parties in the 2014 elections The PES Knock the vote 2014 campaign (referred to earlier) was supplemented by a large-scale social media engagement. It dominated Twitter discussions with an undisputable lead over the other European parties. For instance, Martin Schulz scored high with 418 mentions. 111 On Facebook, Martin Schulz received 89% of likes. 112 The EPP was weaker on Twitter: it received 79 mentions while its Spitzenkandidat, Jean-Claude Junker received 121 mentions. On Facebook, on 110 See: Barberá, P. (2014). Estimating Quantities of Interest Using Social Media Data. Social Media and Political Participation Lab New York University. Available at: Refer to this report for all data on Facebook activity 63

76 Role of the Media in the 2014 European Elections 4 the other hand Jean-Claude Junker received 84% of Likes. In the cases of the ALDE campaign, I vote liberal, the social media campaigns dealt primarily with political objectives that ALDE promoted in the 2014 elections. Furthermore, ALDE s Spitzenkandidat received 134 mentions on Twitter and 82% of Likes on Facebook. The European Left Party put almost exclusively efforts on online campaigning which was reflected in the highest amount of Likes on Facebook with 88% for its Spitzenkandidat. The European Green Party was the only party to organise an open Europe-wide online primary to select the two leading candidates for EU presidency. It also conducted a pan-european social media campaign. Nevertheless, its Spitzenkandidat only received 79% Likes on Facebook. A Pew Research Centre analysis of the conversation on Twitter about the 2014 European elections suggests mixed feelings toward the EU and a general lack of interest in the Spitzenkandidaten. 113 By analysing more than 1.2 million tweets in English, French and German (collected between 1- and 14 May 2014), a relatively fragmented view about the EU emerged. For example, in English, 31% of the posts on Twitter about the EU were positive compared with 39% that were negative and 30% that were neutral. 114 Further analysis of the Englishlanguage discussion about each of the five candidates for the European Commission presidency shows the degree to which the parties draw more attention than individual candidates. None of the candidates received more than 12,886 English-language tweets in the period that was examined. The tone of Twitter sentiment toward each of the candidates was mostly neutral with between 65% and 79% of the tweets not being either positive or negative. It is far from clear whether social media activity influenced the turnout of voters in the European elections or even influenced the decision about who to vote for. Firstly, people who used social media to inform themselves are usually people who are already very interested in EU politics and therefore more likely to vote anyway. 115 The highest volumes of conversations on social media were recorded in France, Italy, Spain and the UK. The most used languages recorded were English (29%), Spanish (21.7%), French (19%) and Italian (10.3%). 116 There was significant online campaigning during the 2014 European elections in France, Finland and Spain (in all of these countries, the turnout Compare to: French: 33 % positive, 39% negative and 28% neutral and German: 39% positive, 5% negative. 115 Academics and journalists often describe social media as echo chamber. This means that contrary to the common perception that social media is able to reach different levels of society, it is likely to be used in the same way as traditional media. In other words, when social media is used to convey political information it usually reaches only those who are already politically active. Therefore, the metaphor of an echo has been used

77 Role of the Media in the 2014 European Elections 4 increased). Nevertheless, similar activity did not increase the turnout in Italy. Also, despite more limited social media activity in the European elections in Lithuania and Greece, turnout increased there more than anywhere else. This suggests that election turnout and social media activity cannot be positively correlated. 117 Secondly, it has been shown above that the PES was most active of the European parties on Twitter and that Martin Schulz also received the highest amount of likes on Facebook. In addition, the European Left was also very active on Twitter and Facebook. Nevertheless, this activity is not reflected in the election results. Consequently, it is not possible to identify a direct correlation between Twitter activity and election results. Nevertheless, various studies show that, in national and regional elections, social media mattered. One study conducted in the US (involving 61 million Facebook users) revealed that people were more likely to vote if they saw a message showing their friends had voted (using the I voted button). Facebook used a similar tool in these elections and the I voted message was seen by nearly 90 million people. 118 Besides these more measurable effects, the potential impact of social media in contributing to a higher turnout can also be derived from the nature of social media. Social media allow political actors, particularly smaller parties or less well-known candidates, to bypass mass-media filters. 119 Furthermore, social media is especially well-suited to being targeted at the young who are the largest group of social media users and often least likely to vote. 120 Specifically in relation to the 2014 European elections, in a study of around three million tweets about the elections were tracked. More than a million tweets about the European elections (not just using the official #EP2014 hashtag) were sent during the week of the elections, with around half of those sent on Sunday 25 May. The European Parliament has also published some top-line statistics, such as 11 million views of its elections video. The most important effect of these posts is that they create a snowball effect in a sense that re-posting and the borderless nature of social media manages to reach many more people than traditional media instruments. In addition, journalists and others can make use of social media material and distribute it on other platforms. All in all, the evidence on how widespread social media reported on the European elections and social media s possible snowball effect suggest an increasingly important contribution to raising awareness about the elections. In 2010 a Study was conducted on the impact of the social media on the 2010 Dutch elections. 121 It was found that social media had a marginal but significant impact on the number of preferential votes received by candidates. They found that for every 1,000 Twitter followers there was an extra 190 votes, whereas for every 1,000 followers on Hyves there was an extra 1,343 votes. It was argued that when candidates use social media more [actively], the effect of the number of followers on the number of preferential votes increases (and becomes ibid. 119 For example the Pirate Party in Germany, see: Niels Spierings and Kristof Jacobs 65

78 Role of the Media in the 2014 European Elections 4 significant). Every tweet during the campaign would result in another 11 extra votes per 1,000 followers. 122 In conclusion, social media activity was extensively used by all European Parties in the 2014 election but some parties made use of this tool more than others. There is no doubt that there has been an increase in the use of social media for campaigning in Very few MEPs do not have a twitter or Facebook account; there was also significant online campaigning from Spitzenkandidaten. It is not possible to directly correlate an increased social media activity with a higher turnout in the elections or to correlate social media popularity with the election results. 123 The available studies also show that some 70% of citizens still obtain most information on the EU-related matters from the TV and from newspapers. Nevertheless, a number of studies acknowledge the importance of social media in elections generally even if is the impacts are not quantifiable. One example is the snowball effect of social media. Furthermore (and more importantly) social media is primarily important for young people. Indeed, since the young people are the voters of tomorrow it can be assumed that the impact of social media will be more visible in the next European elections. 4.5 Conclusions Role of the Media in the 2014 European Elections Overall, with a few exceptions, in almost all EU Member States there do not appear to have been any major issues concerning the principle of the freedom of expression or access to the media during the 2014 European elections. However, rules determining access to the media varied across the EU Member States (some determining the amount of time to allocate to political parties according to their representation in the parliaments while in other countries each of the party was given the same amount of time) and in some countries there was unequal coverage of political campaigns in the privately-owned media. In terms of coverage of EU topics during the election period, a few observations can be made. First, the role of social media increased in 2014 in comparison to Second, the introduction of the Spitzenkandidaten provided a focus point for media coverage. Nevertheless, journalists we interviewed considered that the Spitzenkandidaten were too similar in their views on EU issues which hindered an interesting debate. Furthermore, some argued that the candidates also lacked charisma. 124 Third, the financial and economic crisis was a key issue in most Member States and this was far more so in 2014 than In some Member States, the financial crisis had an effect in terms of raising the media s attention. However, in other Member States the financial crisis predominately helped Eurosceptic parties to gain more attention. Overall, in 2014 the debate on the future of Europe was more intense than in earlier elections. In 2014, the social media became a much more important tool than it was in the 2009 elections. First, political parties and candidates used it as a campaigning tool. Second, It can be assumed that some smaller parties might have benefitted from social media campaigning, but mainstream parties probably did not

79 Role of the Media in the 2014 European Elections 4 traditional media and journalists used the social media to increase the reach of the information. Besides the increased importance of social media it is not possible to directly correlate social media activity with a higher turnout in the elections or to correlate social media popularity with the election results. However, the social media was important, particularly for the younger generation. Overall, feedback from the research suggests that the media had a positive impact on the European elections in 2014 by raising awareness of EU issues and encouraging citizens to vote. However, there was criticism that the media gave too much attention to national issues in some countries. An interesting feature of the 2014 European elections was the increased use by the European parties of the social media as a campaign tool. 67

80 Enhancing the Efficient Conduct of the European Elections 5 In this section we provide an assessment of the extent to which the Commission s Recommendations were implemented with regard to enhancing the efficient conduct of the 2014 European elections. Overview The structure of the assessment in this section follows the Commission s Recommendations themselves and is summarised below: Enhancing the efficient conduct of the elections 3.5 Single contact authority 3.6 Efficient transmission of data 3.7 Additional data allowing for more efficient non-national voter identification 3.8 Technical means for safe and efficient transmission of data The European Commission put forward a number of proposals to help ensure that the 2014 European elections were conducted as efficiently as possible, particularly in relation to Directive 93/109/EC granting EU citizens the right to vote or stand as a candidate in European elections either in their country of residence or the country of nationality. Ahead of the 2014 elections all Member States carried out tests, coordinated by the Commission, to prepare for the implementation of these recommendations and to allow for the putting in place of the necessary tools, including IT. As in 3, in each subsection of 4, we start by summarising the purpose of the Commission s Recommendations and then assess the extent to which they were implemented and the effects on the 2014 European elections. The assessment starts in each case with an analysis of the CSES survey data and then goes on to evaluate other research findings from the desk research, interview programme and other sources of information. Where possible, we also compare the findings in relation to the 2014 European elections with the previous elections. 5.1 Single Point of Contact The Commission s recommendation suggested that all Member States should set up a single contact authority in charge of the exchange of data on voters, for the implementation of Article 13 of Directive 93/109/EC. According to the Commission s preliminary report, all Member States agreed to have a single contact authority for notifications through the information exchange mechanism except the UK which indicated it would send notifications about EU voters via local electoral authorities. However, the UK established a single contact point for receiving of information from other Member States concerning UK nationals. The table below, based on an analysis of the responses to the DG JUST questionnaire that was sent to Member States, summarises the situation: 68

81 Enhancing the Efficient Conduct of the European Elections 5 Table 5.1: Single Point of Contact Member State Single Point of Contact Member State Single Point of Contact Austria Italy Belgium Latvia Bulgaria Lithuania Croatia Luxembourg Cyprus Malta Czech Rep Netherlands Denmark Poland Estonia Portugal Finland Romania France Slovakia Germany Slovenia Greece Spain Hungary Sweden Ireland UK 125 Source: analysis of DG JUST questionnaire In total 27 out of 28 EU Member States implemented the single contact authority recommendation. The United Kingdom was the only Member State that did not introduce a single point of contact for transmitting information regarding EU nationals who had registered to vote. In the UK this responsibility rested with each local Electoral Registration Officer. However, a single contact point (the Cabinet Office Elections & Parliament Division) was set up for receiving information from other Member States concerning UK nationals registered elsewhere in the EU. In most Member States the nominated single contact points were the Electoral Commissions and/or the Ministry of Interior. Almost all Member States agreed that setting a single point of contact reduced the administrative burden on national authorities compared with previous European elections and made the exchange of data more efficient and effective. There was, however, also some less positive feedback. Firstly, it was argued that transferring the tasks of handling voter information from a number of different entities to a single point of contact had threatened to overload the designated body. Thus, according to the authorities in the Czech Republic, although data was more coherent, the process did create a greater burden on the Ministry of the Interior as the single contact point, moreover without temporarily reinforced staff capacities (which the Ministry was able to plan in advance) it would not have been possible to handle. The heavy workload of the single contact point included not only receiving and processing 125 As explained below, in the UK there was no single contact point through which information was sent to other Member States. The UK only had a single point of contact ( SPOC - the Cabinet Office fulfilled this role) for receiving data about UK citizens from other EU Member States. 69

82 Enhancing the Efficient Conduct of the European Elections 5 information from various Member States, but also transmitting information regarding the nationals of other Member States, which in the case of Czech Republic does not have a central electoral roll, constituted a considerable two-way flow of information between the central point and the municipalities. It was also necessary to provide technical guidelines for the municipalities in advance to explain the procedure for the 2014 European elections and differences to the procedures for the 2009 elections when the municipalities sent information on EU nationals directly to the authorities in the other Member States. Moreover the Czech Republic, as well as Austria and Belgium questioned the benefits of setting of a single point of contact. It was argued in one case that there had been no quantifiable increase in the efficiency of the information exchange. One expert said that each Member State still had different administrative processes ( administrative covert ). Nevertheless, it seems that the consequences of not establishing a single point of contact by the UK in respect of outward data transmission did cause difficulties for other Member States. Thus, the authorities in France, for example, explained that they had received information, not all in XML format, from different UK authorities at different times, which had created significant problems. 126 All other countries had sent information this time from one central point. In conclusion, the research confirms that the setting up of single points of contact for the exchange of data on voters was beneficial although there were some transitional complications for the designated authorities. 5.2 Timely transmission of data The Commission s 2013 Recommendation included a number of proposals designed to ensure that the exchange of data on voters was done in a way that was timely. It was clearly important that voter data would be exchanged earlier enough in the run-up to the 2014 European elections to give receiving authorities enough time to process the information and to ensure EU citizens wishing to make use of the provisions of Directive 93/109/EC, and entitled to do so, could vote accordingly. All the EU Member States considered that they sent out the notifications in a timely way. The following table provides a summary, based on the DG JUST questionnaire responses, of the extent to which Member States received data on voters in a timely way from other countries: 126 Not having implemented a single contact point and with different local authorities in charge of data exchange, it was also difficult for the UK to observe other Commission recommendations such as those concerning the timely transmission of data and using a standardized format for the transmission of data and to allow for more efficient identification. 70

83 Enhancing the Efficient Conduct of the European Elections 5 Table 5.2: Member States which declared having received the notifications from the other Member States in time for taking the necessary steps (i.e. removing their nationals from the electoral rolls who registered to vote in their MS of residence) Member State Member State Austria Italy Belgium Latvia (See below) Bulgaria Lithuania Croatia Luxembourg Cyprus (see below) Malta Czech Rep Netherlands Denmark Poland Estonia Portugal Finland Romania France (with exceptions Slovakia See below) Germany Slovenia Greece Spain Hungary Sweden Ireland UK (with exceptions See below) As can be seen, judging by the questionnaire responses, most EU Member States complied with the Commission s Recommendation. However, there were a number of shortcomings. Examples of the late transmission of data include: Data from the United Kingdom was received in Cyprus right up to the day before the 2014 European elections which left no time to process it or take measures to prevent double voting; According to the authorities in France, some countries (Romania, Hungary, Malta, Poland, Ireland, Germany and Denmark) sent their files too late to be considered; In Latvia some data were received from other countries after 19 May 2014 when it was no longer possible to make corrections to the Latvian electoral roll and it was therefore pointless to try to identify the persons concerned; In the UK, files received from HR, MT and RO were not processed because they were sent after the deadline. Moreover second batches received from DE, GR and SI were also not analyzed because they were sent after the deadline. Overall, it seems that data was generally transmitted and received in a timely way. In addition to the feedback from the completed DG JUST questionnaires and our interviews, this conclusion is supported by the survey responses with just over two-thirds (68%) of the respondents stating that Member States had done enough to comply with the Commission s Recommendations. Looking ahead, it was, however, argued that it would be much better if there is a harmonization of the last day for registration on the electoral roll for all Member States, or at least the last day for registration should be at least a month 71

84 Enhancing the Efficient Conduct of the European Elections 5 before the European elections to allow a sufficient margin for sending and processing the data. 127 Although most Member States sent and received the data on time, the major concern was with the quality of the received data that was often insufficient to identify voters, rather than the timing of the sent/received data. Moreover, the non-harmonization of sent data created difficulties in identifying EU voters. The need was highlighted to receive additional (non-compulsory data) to identify voters more efficiently and help prevent double voting. The following table provides an analysis of the data available from DG JUST s questionnaire on the extent to which different Member States were able to identify their nationals from the information that was sent to them by other Member States. Table 5.3: Statistics on the efficiency of the double voting mechanism Key: A - No. of your nationals residing in other MS notified to your authorities as being registered in the electoral rolls of their Member State of residence (2014); B - No. of your nationals notified by their MS of residence that your authorities were able to identify and therefore, deleted from your Electoral rolls (2014); C - No. of your nationals residing in other MS notified to your authorities as being registered in the Electoral rolls of their Member State of residence (2009); D - No. of your nationals notified by their MS of residence that your authorities were able to identify and therefore, deleted from your Electoral rolls (2009) MSs A B % C D % Austria 11, % Croatia 6, % Czech Rep 4,125 + about 1,106 ** 25% 3,800 2, % 300 (UK) Cyprus 1, % Estonia 3,346 32,22 (124 not 96% identified) Finland 18,245 17, % France 93,132 73,572 79% Germany % Greece 17,164 15,019 88% Ireland 8, % 4, % Italy 129, ,260 68% Latvia 6,651 2,522(identified), 20% 4,129(uniden.), 1,334(deleted) Lithuania 9, % 127 In their response to the DG Just questionnaire the Austrian authorities suggested that a final deadline for exchanging information is necessary to improve the efficiency. At the moment, the Member States are unsure about when they can expect to receive the last information that is relevant to updating their own electoral roll. It would have been more efficient if there were a harmonization of the last day for registration on the electoral roll for all Member States, or at least the last day for registration should have been a month before election day, to allow a sufficient margin for sending and processing the data 72

85 Enhancing the Efficient Conduct of the European Elections 5 Malta n/a n/a n/a Netherland 50,725 2, 286 * 1 4.5% Romania 137,765 49, % Slovakia 5,288 2,283 43% Poland 62,780 44,813(Identified. 71% ) 171,750 (Uniden.) Portugal 103,997 42,036 40% 83,556 38, % Spain 50,111 34,300 68% Slovenia ,102 ( Ident) % (Unidentified) Sweden 12,042 4,305 identified 36% Source: Data for 2014 is taken from the responses to DG Justice questionnaire; data for 2009 is taken from the Commission s 2009 report (data was only provided for Czech Republic, Ireland and Portugal). As a complete set of data is available for only 3 countries (Czech Republic, Ireland and Portugal) it is not possible to make an overall assessment of whether there was a significant improvement in the identification of voters in 2014 compared with While in Portugal there was an increase in the number of identified voters from 38,619 to 42,036 (+8.8%), in the Czech Republic, the number of nationals notified by their Member State of residence that authorities were able to identify and therefore, delete from the electoral rolls went down from 65.8% (2009) to 25% (2014). 128 In Ireland, the situation was the same the numbers went down from 4.3% to 1.7%. In Latvia, data received from other Member States allowed the identification of only 38% of voters. The UK authorities argued that, in considering whether the information exchange is achieving its key objective of preventing double voting across Member States, further research is needed. The evidence currently available does not suggest that double voting by EU citizens is an issue. Moreover, from the UK s experience, the Information Exchange system might not be effective in tackling attempts at double voting. Consequently the UK authorities suggested that the Information Exchange scheme should be abolished altogether and replaced with a solution that relies on the use of a declaration by the voter (as to which country they will vote in) supported by sanctions to prevent double voting. More generally, difficulties with the identification of voters arose from the need of additional data (e.g. date of birth). The efficiency of preventing double voting still relies Czech Republic said that they were able to identify 20-25% of voters notified to the Czech Republic by other Member States under the exchange of information 129 As indicated in an earlier study on the Franchise and Electoral Participation of Third Country Citizens Residing in the European Union and of European Union Citizens residing in Third Countries AFCO_ET(2013)474441_EN.pdf) 73

86 Enhancing the Efficient Conduct of the European Elections 5 on suitable information exchanges between national electoral authorities. One solution would be the integration of national electoral registers for European Parliament elections into a single register, but the obstacles to doing this are considerable. 130 Additional noncompulsory data is deemed by Member States as important to identify voters (examples are provided in 5.3) In some Member States, there was an increase in the number of questions regarding dual EU nationality in the 2014 European elections. But in most EU Member States there are no real sanctions available for those with dual nationality who vote in both Member States. There was no exchange of information on this group. 5.3 Additional non-compulsory data allowing for more efficient identification Article 9(2) of Directive 93/109/EC requires any EU citizen wishing to register to vote in the Member State in which he is resident, but of which he is not a national, to provide as well as his or her family name and first name, nationality and address in the state of residence, and the locality or constituency in his home Member State where he or she was last registered to vote 131 The Member States of residence were encouraged to transmit, in addition to the personal data referred to above, all relevant personal data (i.e. not limited to the first names and family names), which could be necessary for the identification of voters by the authorities of their home Member State. As noted earlier ( 2), the most important personal data in this respect is the individual s date of birth which most Member States require to identify a person but which was omitted from the scope of the Directive. In some other countries (e.g. Spain and Portugal), passport numbers are also used for electoral registration purposes. The table below identifies the Member States that sent additional non-compulsory data for the identification of voters: Table 5.4: Additional data allowing for more efficient identification Key: = Member States that did send additional data = Member States that did not send additional data Member State Member State Austria Italy Belgium Latvia * See below Bulgaria Lithuania Croatia Luxembourg Cyprus Malta Czech Rep Netherlands AFCO_ET(2013)474441_EN.pdf) 131 He must also produce the same documents as are required from a national of the Member State where he wishes to register. 74

87 Enhancing the Efficient Conduct of the European Elections 5 Denmark Poland Estonia Portugal Finland Romania France Slovakia Germany Slovenia Greece Spain Hungary Sweden Ireland * (see below) UK Amongst the Commission s Recommendations designed to improve the efficiency of the European elections, this remained an area of difficulty to many Member States and cannot be said to have been implemented as effectively as other Recommendations. Different Member States still register citizens using different types of personal information. Some record the identity card number of the individual, others the passport number. The lack of a common approach creates difficulties because if voter identification details are sent without showing the full name, date and place of birth of the individual, the receiving Member State may not be able to make a match with a person wishing to register as a voter for the European elections. Various complications arose. Firstly, four Member States were only able to provide compulsory data on individuals under Directive 93/109/EC without sending any additional data. This was due to either the national laws of the country that only allow sending compulsory data (e.g. Hungary, Ireland) or for other reasons. For example in the UK there is no national law that directly prohibits the transmission of additional data but there seems to be reluctance to transmit non-compulsory data. In the UK the transmission of data on EU residents is currently carried out by the Electoral Registration Officers at a local level. This was what happened for the 2014 European elections. What is transmitted is information from the electoral register and this only includes compulsory data. The policy of the UK authorities (the Elections and Parliament Division of the Cabinet Office) has been that only data on the electoral register can be released. A new voter registration system was introduced from June 2014 in the UK. Under this new system, known as the Individual Electoral Registration System (IER), which was brought in by the Electoral Registration and Administration Act 2013, electors register individually (rather than by households as previously). They are required to provide their date of birth (DOB) and their National Insurance (NI) number to verify their identity, in other words to match applications for registration to other information held on the individual concerned. However, this verification information is not entered on the electoral register For future European elections, if the UK's method of transmitting information for the EU Information Exchange System (IES) remains as it is now, with information being sent out by the local EROs, the DOB and NI information which will have been used for verification will not be sent. Whether there is any legal reason why this non-compulsory verification information could not be sent if the IES system changed in the UK is not clear from the information provided by the authorities. However, there are some indications of government sensitivity regarding this data as the 75

88 Enhancing the Efficient Conduct of the European Elections 5 Also in Lithuania and Romania, there seems to be no law that prohibits the transmission of non-compulsory data to other Member States. In the case of Ireland, only one item of non-compulsory information was collected, namely the Date Of Birth (DOB). This was transmitted by the Irish under the IES. Other types of non-compulsory information, such as national ID card numbers, are not covered by Irish electoral law and it would require a change in this law to collect these data. There were also gaps in the compulsory data that was provided which meant that it was not possible to proceed with the identification of voters and their application for registration therefore had to be cancelled. Examples of problems with the registration on non-nationals arising from a lack of additional data The authorities in Croatia complained that certain countries failed to submit sufficient data to allow them to identify their nationals and failed to submit the agreed and established basic dataset required for the exchange. Problems arose because the data did not include the personal ID number (OIB or JMPG), which is the main identifier for Croatian citizens in the electoral rolls. The data they received corresponded to the countries' own electoral systems and the information available in those systems rather than to the agreed model. The Czech Republic complained that some Member States did not provide the data they were obliged to provide pursuant to Directive 93/109/EC,, notably data on the last place of residence,. In France, voter registration captures the first name, family name, date and place of birth and the commune of registration, whereas in the UK what is recorded is the first name, family name, and residential address. Thus data relating to French residents in the UK provided by UK electoral registration officers did not include the date and place of birth or the French commune where the voter was enrolled. Consequently no French voter of the 13,699 notified by the UK to the French authorities was identified by them 133 ; as a result they could not check whether they were registered on their electoral rolls and the information sent was impossible to use. In Germany, only 17% of the data sent from other Member States had the sufficient info to allow the authorities to identify voters. Ireland s authorities required information on the constituency where the Minister concerned stated in the House of Commons in September 2010 "it should be made absolutely clear that no new national databases will be created [by the IER system] and that no additional information will be placed on the electoral register as a result of the changes to the system". This suggests that there may be reluctance to transmit this type of non-compulsory data even if it could be produced. 133 Denmark, Malta, and Slovakia all experienced similar difficulties with data from the UK. 76

89 Enhancing the Efficient Conduct of the European Elections 5 individual was last registered in order to be able to trace the person. Largely because of the absence of such information in the data received, in only 147 out of 2,524 cases was any matching possible. Italy did not request any additional data, but mentioned that often even the compulsory information was not sent. The Latvian authorities did not receive all the compulsory data from the UK. They only received the first name and surname of the Latvian nationals. Similarly, if the UK authorities were not provided with information as to a UK citizen s last constituency of registration in the UK they also were unable to trace the person concerned. None of the 36,615 individuals notified by the French authorities could be traced in the UK. France fared better in matching data received to their own electoral records. However, even in France around 26% of records received did not result in the individual being deleted from the relevant electoral roll either because of late receipt of the data or on account of the information received being insufficiently compatible with the data held in the French system for a match to be made. The UK only has the compulsory data on the register and the authorities cannot send out any additional data as they do not collect it. There is now a collection of Date of Birth (DoB) and National Insurance Number (NiNo) for the new registration process but these data are not put on the register and the NiNo is not even seen let alone retained by the elections office when the application to register is made online. The UK takes the view that if it is not on the register then it cannot be released as it is not part of the registration data but, in any event, most people that are registered will not even have given a DoB when applying in the past and may not need to re-apply if their circumstances remain the same for the future. So that data is not collected or held. There were a number of other complications. This included: The use of incompatible software (for example, in a few cases the UK used software programs that the Latvian authorities would have had to buy in order to open the files they received) or in some cases the use of various file formats that were not recommended by the Commission (in Finland, for example, the notifications were not sent using the XML file format suggested in the Commission guidelines); Differences in the spelling of names (e.g. the Croatian language has diacritic marks, and when such marks are omitted from names, automatic identification is impossible) or misspellings (e.g. the Latvian authorities complained that first name and surname were written according to the language and spelling of the relevant Member State and not the language and spelling of Latvia as it was used in identity documents issued in Latvia); Changes in surnames (e.g. there were instances where a person s maiden name was given in the Latvian electoral roll, but they had changed their name following 77

90 Enhancing the Efficient Conduct of the European Elections 5 marriage in another Member State. The persons concerned had not fulfilled their legal obligation to report any change in the information in the register of population so as to allow it to be updated). 134 Several of those we spoke to also highlighted the issue of multiple nationalities and the risk that citizens falling into this category vote more than once in the EP elections. Finland signalled a further problem: when removing people from the electoral roll, it was discovered that many people who were registered to vote in other Member States had already moved back to Finland. The problem is that when Finns move back to Finland, their registration to vote in their previous country of residence remains in force, and so their previous country of residence notifies them as using the right to vote in that country, which leaves them without the right to vote in Finland. The authorities received a large number of complaints from Finnish citizens about this. (Since by law a person may only have the right to vote in another Member State if he or she lives there, this suggests that a procedure should be applied whereby a person who has actually moved away from a Member State is removed from the electoral roll directly on the basis of a notification by the authority of the other EU Member State. Citizens could then deal directly with the authorities of their home country, and would not have them removed from the electoral roll of their previous country of residence through their current country of residence.) It was also argued that not all Member States understand the information exchange in the same way, in particular whether the information exchange always involves information on all voters from other Member States entered on the electoral roll in the Member State of residence, or only information on the new voters entered on the electoral roll; in such a case there is no mechanism in place to inform the home Member State that its national has requested to be deleted from the electoral roll in the Member State of residence as there is no exchange of information on deletions of voters from voter lists. Overall, there remain problems with the transmission of additional data required to identify voters stemming from differences between national systems for voter registration, and other factors such as incompatible software, misspelling of names, etc. The scale of the problem is not, however, clear although many of the Member States responding to the DG JUST questionnaire, as well as national authorities who we interviewed, highlighted such issues. 5.4 Technical means for safe and efficient transmission of data The Commission s Recommendations stipulated that for exchanging the data as provided in Article 13 of Directive 93/109/EC, Member States should use a uniform and secure electronic means, transmit the data in one single package per home Member State, with 134 One suggestion made to us as a way of improving the exchange of information would be for all Member States to use 'surname' in the same way. In some Member States, for example, married women are registered under their husband's surname. However, in The Netherlands and in some other Member States married women are registered on the electoral roll under their maiden name. If all Member States interpreted 'surname' in the same way it would be easier to check whether someone was registered to vote in more than one Member State or not. 78

91 Enhancing the Efficient Conduct of the European Elections 5 one further transmission at a later stage for updates, where necessary. Table 5.5: Use of technical means for safe and efficient transmission of data and other issues (XML file schema, encryption and character set) Member State Member State Austria Italy Belgium Latvia Bulgaria ( See below) Lithuania Croatia Luxembourg Cyprus Malta Czech Rep Netherlands Denmark Poland Estonia Portugal Finland Romania France Slovakia Germany Slovenia Greece Spain Hungary Sweden Ireland UK Almost all Member States improved the security of the data exchange taking place in the run up to the 2014 European elections, by implementing a common electronic format and using an encryption tool recommended by the Commission. Most Member States used the recommended character set and there were no major problems with the XML files. Likewise, all Member States complied with the encryption method. However, there were some problems. Examples of problems with the transmission of data In Austria the use of the prescribed encryption method caused serious problems as the encryption software was not compatible with the security requirements of the Ministry of the Interior. The decryption procedure had to be performed on a stand-alone workstation and the unencrypted data subsequently transferred into the Ministry's computer system. In Belgium there were difficulties in the decryption of some files. New files were requested but never received. In Bulgaria, the XML file schema was not used on a regular basis, but when files were received in this format they were opened using Excel, which they believed showed in a simple and ordered manner all data of interest relating to other Member States. However the Encryption method Cryptotool was very convenient and user friendly. In Estonia there were difficulties and the authorities had to revert to manual options to deal with the data. In Germany the system worked well. However there were some issues with the encryption method (it had to be repeated twice). Moreover Germany said that 79

92 Enhancing the Efficient Conduct of the European Elections 5 Croatia and Slovakia did not send XML files but rather used the excel format. In Sweden the tool and the XML files worked well for the information exchange. However there were difficulties in the encryption tool of some file which hindered the exchange of information for the candidates. Moreover, the naming system (the Commission's proposal) was not used by a number of countries when exchanging files. In some cases, it was difficult to distinguish files for eligible voters from those for candidates. Some Member States complained that the UK did not provide the data in XML and did not provide sufficiently detailed personal information to make identification possible. The UK sent only a few files encrypted using the key, with the majority of files sent being protected by their own password and, in a few instances, not password-protected at all. More generally, feedback from the interviews with national experts suggests that XML standards and encryption suggestions are difficult to comply with because every Member State has a different system. (Sometimes the XML files were not formatted correctly, the encryption key was not transmitted or the data could not be matched as only first and surnames had been provided). 5.5 Measures taken to inform citizens of their right to vote and right to stand as a candidate Most EU Member States undertook a campaign in the run-up to the 2014 European elections to raise awareness amongst non-nationals residing in their country of their right to vote and stand as a candidate in the European elections. In some cases in particular, those EU Member States with a large non-national population - the emphasis was on general publicity rather than contacting individuals. Such publicity included posters, information brochures, media advertising, and websites. Ireland provides a good example of the types of measures taken: Example of awareness-raising campaign in Ireland The registration form used by the authorities for the compilation of the register of electors for the 2014 European elections included details on voting and standing at the 2014 elections for EU citizens living in Ireland. Potential European electors identified on the register of electors since the last European elections in 2009 were sent the relevant registration form. A multi-lingual prompt card was available int17 languages (Arabic, Czech, English, French, German, Hungarian, Irish, Italian, Latvian, Lithuanian, Mandarin, Polish, Portuguese, Romanian, Russian, Slovakian and Spanish) to facilitate the contacts of staff engaged by registration authorities in their door-to-door registration work. Two information leaflets were updated in a number of different languages - The Register of Electors and a brochure called How Ireland s MEPs are Elected. These leaflets were also made available on the Department of the Environment, Community and Local Government website and on registration authorities websites. A further information leaflet on voting and standing in the European 80

93 Enhancing the Efficient Conduct of the European Elections 5 elections was distributed to all EU Member States embassies. An advertising campaign urging people to check whether they were registered drew attention to the fact that the register would be used for the European elections in Various posters were distributed to registration authorities, elected members, youth organisations and multi-cultural groups. Advertisements were placed on local radio stations to bring the registration message to young people in particular. The use of on-line advertising (Leaderboard and MPU (mid page units)) on the main news sites. Awareness of the registration process was promoted through the Department s website and Twitter account. Newspaper advertising was undertaken in 2014 to inform EU citizens of their entitlement to get on the register of electors via the supplement. The Internet was an important tool for raising awareness of non-nationals rights. In Croatia, the Electoral Commission adopted a Notice on the arrangements for nationals of other EU Member States to exercise their right to vote and stand as candidates in the European elections and sent it to all media outlets for general publication and also publicized it via a website. In Italy, all EU citizens were able to fill in the standard application form directly online via the Ministry of Interior s website so that it could then be printed out and signed before being forwarded to the municipal authorities. A more targeted approach was adopted in some countries. For example, in the Czech Republic, the embassies of other Member States and relevant government departments receiving applications for temporary or permanent residence were contacted and informed about the rights of non-residents. Local authorities, citizens rights entities and other local bodies were also heavily involved in raising awareness amongst non-nationals living in their areas of their European elections voting rights electoral registration procedures. In Denmark, a special call centre was in operation to provide information. In some EU Member States, there was a targeting of individual non-nationals in the run-up to the 2014 European elections. Thus, in Italy the municipal authorities were informed of the need to send personal letters to all EU citizens residing in their municipal areas who were not yet entered on the relevant electoral list. They were provided with suitable supporting material including a standard letter (in four languages: English, French, German and Italian) accompanied by the standard application form already duly translated and separate for each Member State which sent the translation to the Ministry of the Interior. The municipal authorities were asked to attach the standard application of the addressee's Member State to each letter (to be sent in all four languages), if available. In the event that the standard application was not available in the language of the addressee's country, the municipal authorities were asked to attach it in Italian and English. To take another example, in Cyprus, a total of 7,465 personal letters were sent to nationals of other Member States who were entered on the special electoral roll for municipal elections and were not included on the register for European elections, together with the relevant registration form. Information on how non-nationals could participate in the European elections was also available on a website (euroelections2014.gov.cy) in English, Greek and Turkish. 81

94 Enhancing the Efficient Conduct of the European Elections 5 In Denmark, EU citizens from other Member States who are resident in Denmark are informed individually of their being able to vote and stand in the European election by posting cards to every EU citizen in Denmark. EU citizens from other EU Member States who had previously applied to be included on the electoral roll for European election in Denmark and who had since resided uninterruptedly in the country were automatically included on the electoral roll for the 2014 European election (the cards also included information on the possibility of being removed from the electoral roll if, for example, the individual wanted to vote in his or her country of origin. Overall, the procedure was considered to have worked satisfactorily and did not give rise to any particular problems In Lithuania, a similar approach was adopted with the Central Electoral Commission sending personal letters to all 3,993 nationals from other EU Member States that were entitled to take part in the 2014 European elections, informing them of the possibility and of the personal details that they would have to provide to be included in the Lithuanian electoral roll. These letters were sent together with the registration forms to the addresses which the EU nationals in question had declared as their place of residence in Lithuania. Both the letters and the forms were sent in English, German, French, Latvian, Polish and Italian. Full information was also published on the official website of the Central Electoral Commission (also in English). In Latvia, this approach was also adopted with individual letters being sent to non-nationals in 15 languages (Bulgarian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Italian, Lithuanian, Polish, Romanian, Spanish and Swedish). Application forms by which they could ask to exercise the right to vote at the 2014 European Parliament elections in Latvia were sent to them in English. Information for nationals of other Union Member States regarding the procedure for registration for the 2014 European Parliament elections was also published on the Central Election Commission s website ( 5.6 Non Nationals Registering to Vote Compared with 2009 To assess the effects of Directive 93/109/EC on the turnout in the 2014 European elections, it should ideally be possible to calculate: (a) the number of EU citizens living in a Member State other than the one where they are nationals; (b) the number of non-nationals registered to vote; and (c) the number of registered non-nationals who actually voted in the 2014 European elections. However, whilst information exists on (a) and (b) for most EU Member States, it is not available for (c). Most Member States said that they do not know, and have no way of finding out how many citizens from other EU Member States who registered to vote actually voted. Likewise, data on the nationalities of non-nationals who voted are not collected. Our analysis therefore has to rely on a comparison between (a) and (b) Registrations by non-nationals in country of residence The following table provides an analysis of the proportion of non-nationals living in each of the EU28 Member States who registered to vote in the European elections in 2009 and In the case of 2 Member States (AT and PL) no information is available on the number of non-nationals living in the countries concerned or on the number who registered to vote in the 2014 elections. (The Austrian authorities did not provide the number of nonnationals who registered to vote, while the information from Poland did not include both 82

95 Enhancing the Efficient Conduct of the European Elections 5 the number of non-nationals registered to vote and the number of non-nationals residing in Poland). Table 5.6: Proportion of non-nationals registered to vote in the 2009 and 2014 European Elections (by Member State) Member State 2014 European Elections 2009 European Elections Registered Voters No. Non- Nationals % Registered Voters No. Non- Nationals Austria n/a 34,643 n/a 30, , Belgium 68, , , , Bulgaria 68 26, n/a n/a Croatia 8 5,293 0 n/a n/a n/a Cyprus 7, , ,449 77, Czech Republic , , Denmark 15, , ,744 97, Estonia 1,191 20, , Finland 7,341 71, ,193 45, France 245,063 1,406, ,148 1,156, Germany 172,110 3,168, ,425 2,142, Greece 13, , , Hungary 1, , , Ireland 71, , , , Italy 80,356 1,287, ,904 n/a n/a Latvia 326 8, , Lithuania 278 3, , Luxembourg 24, , , , Malta 7,868 45, ,087 19, Netherlands 48, , n/a 241,495 n/a Poland n/a n/a n/a , Portugal ,930 84, Romania , , Slovakia 33 55, , Slovenia , , Spain 337,748 2,119, ,792 1,970, Sweden 49, , , , UK 160,000 1,921, ,043,629 n/a n/a Totals/average 2,730,834 10,869, ,051,833 7,580, Source: the 2014 data has been collected from the responses to the DG Just questionnaire; the 2009 data comes from the European Parliament AFCO_ET(2014)493047_EN.pdf). % 83

96 Enhancing the Efficient Conduct of the European Elections 5 Note: for some Member States (FR, IT, NL, SK) data for the number of non-nationals was not provided in the completed DG JUST questionnaire and we have therefore used Eurostat statistics which are as at 1 January by_ group_of_citizenship. The following chart highlights the percentage of non-nationals who registered to vote in the Member State of residence for the 2014 European elections. As noted above, estimates could not be made for two EU Member States (Austria and Poland) because of a lack of data from the countries concerned. Figure 5.1: Percentage of non-nationals who registered to vote in the Member State of residence for the 2014 European elections 84

97 Enhancing the Efficient Conduct of the European Elections 5 From the above analyses it can be seen that: In almost all EU Member States (22 of the 24 providing information), the number of non-nationals residing in the respective countries was significantly higher at the time of the 2014 European elections compared with The exceptions to this trend were Austria and Hungary. The number of non-nationals registering to vote in the European elections was also higher in 2014 compared with However, the increase in the numbers registering to vote (an average of +5.2% compared with 2009) was far lower than the increase in the number of non-nationals residing in Member States (an average of +64.6% over the same period). As a result, despite the increase in the number of non-nationals registering to vote, the proportion that this represented of the total number of non-nationals residing in the various Member States actually declined from 12.2% in 2009 to 8.5% in As can be seen from the above chart, there is a big difference between the total number of non-national citizens residing in a country and the total number of non-national citizens entered in the electoral roll in a country. There is also a difference, of course, between registrations and actual votes although the data on this is very limited. Cyprus, Lithuania and Malta were the only Member States which were able to provide information on the number of non-nationals who actually voted in the 2014 European elections. Registration of non-national voters and actual votes caste In Cyprus, the estimate of the number of people who actually voted was 2,711 which was 35.1% of the 7,712 people who were registered. In Lithuania 175 non-nationals actually voted, i.e. 62.9% of those who registered. In Malta, 530 non-nationals voted which was 6.7% of the 7,868 people who had registered. In other Member States, the number of non-nationals who actually voted in the European elections is not known. In the UK, for example, there is a marked register of who voted but access to it is restricted with people looking at it for electoral research purposes under supervision during the one year period it is retained. We cannot make copies of the aforementioned (nearly 400) registers available. They are not held by the UK centrally but by each local authority. They are on paper and not electronic. The positive trend in registrations by non-nationals that was evident in the 2014 elections built on a similar trend between the 2004 and 2009 European elections. In its report on the 2009 elections, the Commission had noted that in 2009, almost everywhere there were more EU citizens registered to vote in their Member State of residence, compared to the previous elections. However, the Commission s report noted that in some Member States, even though the number of non-nationals of voting age residing on their territory increased, the rate of enrolment did not rise proportionately in 2009 when compared to 2004 (for example, in Denmark, the number of non-nationals living in the country increased 85

98 Enhancing the Efficient Conduct of the European Elections 5 by 66% between whereas the number of non-residents registering to vote increased by 7%). 135 In the following table we provide a summary of the situation in the 2014 European elections. It should be noted that the analysis is limited to 17 Member States for which data is available. The analysis suggests that in a majority of EU Member States (12 of the 17 or 70%), the number of non-nationals registered to vote increased over the period. However, except in six Member States (Spain, Greece, Latvia, Malta, Portugal and Romania), this increase was not proportionate to the increase in the number of nonnationals residing in the countries concerned. Compared with , the differences may have narrowed. Table 5.7: Comparison between change in number of non-nationals residing in Member States and change in number of those amongst them registered to vote in European elections, Decrease in the number of registered voters No. non-nationals No. non-nationals registered to vote Denmark +33.0% -5.0% Lithuania +21.8% -21.5% Ireland % -2.0% Hungary -0.8% -71.3% Slovakia % -94.4% Increase in the number of registered voters Belgium +15.7% +3.3% Cyprus +44.2% +19.5% Spain +7.5% +19.1% Luxembourg +56.9% +39.9% France +21.7% +9.8% Finland +56.2% +18.5% Greece +22.8% % Latvia +4.5% +31.4% Estonia % +25.2% Portugal +22.7% % Malta % 277.0% Romania +28.4% 596.8% Source: DG JUST questionnaire and Eurostat. The Commission had also noted in its 2010 report that rates of participation in the European elections in the Member State of residence could be influenced by the fact that, 135 European Commission Report on the election of Members of the European Parliament (1976 Act as amended by Decision 2002/772/EC, Euratom) and on the participation of European Union citizens in elections for the European Parliament in the Member State of residence (Directive 93/109/EC). COM (2010) 605 final,

99 Enhancing the Efficient Conduct of the European Elections 5 because double voting is prohibited, EU citizens have to choose whether to vote in their Member State of origin or in the Member State to which they have moved (i.e. for different sets of candidates). A 2010 Eurobarometer survey revealed that people are extremely divided on the candidates and lists they prefer to vote for in European elections. About 44% said that, if living in a Member State other than the one they hold the nationality of, they prefer to vote on the list of that Member State, while a similar percentage (46%) indicated that they would still prefer to exercise their right to vote in their home Member State Non-nationals standing candidates in the 2014 European elections Whilst there is very little information available on the turnout of non-nationals in the European elections, almost all Member States said that it was easy for EU citizens living in another Member State to stand as candidates in the EP elections, and the relevant data is readily available: Table 5.8: Number of EU non-nationals standing as candidates for the 2014 European elections in countries of residence Member State No. Non-Nationals standing as candidates Member State No. Non-Nationals standing as candidates Austria 15 Latvia 0 Belgium 31 Lithuania 0 Bulgaria 0 Luxembourg 3 Croatia 0 Malta 0 Cyprus 0 Netherlands 5 Czech Rep 6 Poland 3 Denmark 0 Portugal 3 Estonia 3 Romania 2 Finland 1 Slovakia 1 France 24 Slovenia 0 Germany 30 Spain 16 Greece 7 Sweden 3 Hungary 1 UK 13 Ireland 1 Italy 2 Total 170 Source: analysis of DG JUST questionnaire responses. Overall, comparing the 2014 European elections with previous elections, there was a positive trend with regard to the number of non-nationals standing as candidates in their country of residence. As can be seen, for the EU Member States that provided data, a total of 170 non-nationals put their names forward as candidates in the 2014 elections. This compares with 81 in the previous 2009 European elections, 57 in 2004 and 62 in Conclusions - Enhancing the Efficient Conduct of the European Elections Overall, national authorities considered that the measures taken to enhance the efficient conduct of the European elections reduced the administrative burden and had a positive impact in terms of the transfer and exchange of data. Almost all Member States said that the Commission s recommendation contributed to substantially improving the functioning 87

100 Enhancing the Efficient Conduct of the European Elections 5 of the mechanism for preventing abuses such as double voting, making the registering of non-national voters more efficient, reducing the administrative burden and simplifying the procedures for registering voters and candidates in EU Member States of residence regardless of EU country of origin. However, some interviewees from national authorities considered that a significant degree of electoral fraud involving double voting was, in any event, a remote risk and at least one felt that the administrative burden of the information exchange system was disproportionate to the danger it was intended to prevent. Member States said that although the recommendations did not have a significant impact on the level of participation in the elections, they reduced the administrative burden considerably. Moreover almost all Member States thought that the exchange system was better in comparison to These overall conclusions, based on Member States responses to DG JUST s questionnaire and our interview feedback, are also supported by the survey feedback although, as can be seen from the chart below, there was a quite high proportion of don t knows. Given the relatively technical nature of many of the Recommendations, this is not surprising. As can be seen from the chart, Member States were seen as having done most to implement the Commission s Recommendations relating to the common voting day and the timely transmission of data. However, the variations across the different Recommendations are within a quite narrow band. The data needs to be treated with some caution given the relatively small sample size. Turning to the effects, it is clear from the research that the measures taken by Member States to implement the Commission s Recommendations on enhancing the efficiency of 88

101 Enhancing the Efficient Conduct of the European Elections 5 the 2014 European elections had a significant effect in reducing the administrative burden on national authorities. The effect of a more efficient exchange between Member States of data on voters on the turnout of non-nationals in the European elections was, perhaps not surprisingly, less pronounced. However, overall, the conduct of the 2014 elections was generally perceived as being better than in This is illustrated by the survey responses: 89