Election Observation Missions – efforts made to follow up recommendations but better monitoring needed

Executive summary

I

Election observation is an important tool for promoting democracy, human rights and the rule of law. Well-managed elections can contribute to the peaceful transfer of political power. The European Union (EU) observes elections all over the world to these ends.

II

EU Election Observation Missions (EU EOMs) issue a public preliminary statement shortly after Election Day, followed two months later by a comprehensive final report containing recommendations on how to improve the framework for future elections. The recommendations are presented to the host country authorities, which are not formally committed to implementing them. Nevertheless, it is essential to follow up these recommendations in order to maximise the impact of election observation. If the host country does not address electoral shortcomings, there is a risk that weaknesses will remain and even increase ahead of future elections.

III

The main purpose of the audit was to assess whether the European External Action Service (EEAS) and the European Commission had provided support for the implementation of the recommendations by the host countries. To this end, the audit covered EU EOM follow-up in four countries: Ghana, Jordan, Nigeria and Sri Lanka, with a focus on elections observed after 2010.

IV

We examined whether the recommendations had been drafted and presented in such a way as to facilitate their implementation. In this context, we also assessed whether stakeholders in the host countries had been consulted on the recommendations so as to foster local ownership, and thereby increasing the likelihood of implementation. Subsequently, we checked whether the EEAS and the European Commission had used the tools at their disposal to support the implementation of EU EOM recommendations. The most important tools were identified as Election Follow-up Missions (EFMs), political dialogue and electoral assistance. Finally, we assessed the efforts made to monitor the results on the ground.

V

We found that the EEAS and the Commission had made reasonable efforts to support the implementation of the EU EOM recommendations and had used the tools at their disposal to this end. Presentation of EU EOM recommendations has improved in recent years, but more consultation is needed on the ground. The EEAS and the Commission engaged in political dialogue and provided electoral assistance to support implementation of the recommendations, but EFMs are not deployed as often as they could be. Lastly, there is neither a central overview of the recommendations nor a systematic assessment of their implementation status.

VI

On the basis of the audit, we recommend, among other things, that the EEAS:

• ensure, through a systematic quality control on the format before publication of the final report, that the EU EOMs follow the drafting guidelines and template;
• systematically ensure that the Core Team consults stakeholders in the host country on the recommendations before the report is finalised;
• make sure that the stakeholder roundtable is scheduled at least four working days after the report is released, giving participants enough time to familiarise themselves with the report and its recommendations ahead of the meeting;
• when possible, deploy EFMs to countries that have hosted an EU EOM, at a suitable time between elections, provided that the political or security situation in the host country so allows;
• set up a centralised depository for EU EOM recommendations; and
• track progress on implementing EU EOM recommendations on a regular basis.

Introduction

About EU Election Observation Missions

01

Election observation is an important tool for promoting democracy, human rights and the rule of law. Well-managed elections can contribute to the peaceful transfer of political power. The European Union (EU) observes elections all over the world to these ends. Since the EU election observation methodology was established in 2000, the EU has deployed 138 Election Observation Missions (EU EOMs)¹ to 66 countries in Africa, the Middle East, Asia and Latin America.

02

The United Nations’ Declaration of Principles describes election observation as: “the systematic, comprehensive and accurate gathering of information [….]; the impartial and professional analysis of such information; and the drawing of conclusions about the character of electoral processes based on the highest standards for accuracy of information and impartiality of analysis”².

03

The Handbook for EU Election Observation (“the Handbook”) describes the rationale for EU EOMs as follows: “Elections provide means for the people’s will to be freely expressed when choosing their government and provide groups with an opportunity to express their political voice in competition with others without resorting to violence, and contribute to the peaceful transfer of political power. In this regard, election observation by the EU can complement and enhance other EU crisis management and peace-building initiatives in partner countries”³.

04

Without interfering in the organisation of the election itself, EU election observers collect and analyse facts concerning the election process and provide an independent assessment.

05

Following an invitation from a host country, EU EOMs are only deployed when the EU considers that an observation mission would be useful and could help strengthen the democratic process. Before accepting each invitation, the EU carries out an Exploratory Mission (ExM), usually four to six months prior to Election Day, to ascertain whether a fully-fledged EU EOM would be useful, advisable and feasible. Typically lasting two weeks, ExMs also assess the degree to which any previous EU EOM recommendations have been implemented in the case of repeat EU EOMs.

06

On occasion the EU might decide to follow an election process closely even if all the conditions for deploying an EU EOM (for example: security conditions) are not met. In other cases, even if all the conditions are met, the EEAS might not see a need for a full EU EOM.⁴ There are other types of missions that can be deployed. In case of security concerns, an Election Assessment Team (EAT) will be considered. The EU can also deploy a smaller mission, such as an Election Expert Mission (EEM). EEMs are conducted by teams of two to four experts performing the same assessment of the election process as an EU EOM but they do not conduct Election Day observation across polling stations. EEM results are not made public, although some EEMs share their recommendations with the host country authorities. EATs also do not conduct systematic Election Day observation across polling stations but are similar to EU EOMs in that they consist of a full Core Team and possibly a Chief Observer based in the capital city.

07

The European Union can also deploy Election Follow-up Missions (EFMs) at some point between elections as part of the overall follow-up activities. The main task of the EFM is to assess the implementation of the recommendations made during a previous EU EOM. EFMs are small missions carried out by a Head of Mission (normally a Chief Observer), an electoral/political expert, a legal expert and, where appropriate, a press officer. They are also accompanied by representatives from the EEAS Election Division for Democracy and Electoral Observation and, when possible, the European Commission. The latter should focus on the assistance offered by the EU and its Member States and look for potential actions to be funded under future programming. These include short-term measures before forthcoming elections, following the recommendations of the EU EOM. EFMs can also offer additional recommendations, building on the progress made and refocusing attention on the most urgent issues ahead of new elections.

The role of the EU institutions and independence of the EU EOM

08

The High Representative for Foreign Affairs and Security Policy/Vice President of the Commission (HR/VP) has political responsibility for deciding the priority countries for annual EU election observation (after consulting the Member States and the European Parliament), makes the final decision on deployment and oversees the overall functioning of EU EOMs.

09

EU EOMs are usually led by a Member of the European Parliament (MEP), who is appointed by the HR/VP and acts as Chief Observer for the EU EOM. The EEAS provides support for political and electoral issues while the Service for Foreign Policy Instruments (FPI) is in charge of operational, security and financial aspects.

10

The Handbook⁵ states that “EU election observers are obliged to be strictly impartial and not to show bias towards any side in an electoral process”. It also says that “an EU EOM is politically independent from any EU-funded technical assistance projects that may be taking place in the country being observed”⁶. The independence of election observers is also enshrined in the UN Declaration of Principles⁷

11

EU EOMs are atypical in that they are independent vis-à-vis the EU. Observation missions from other groups, such as the OSCE Office for Democratic Institutions and Human Rights (ODIHR), the Commonwealth, the Carter Center and the African Union, comment on the electoral process on behalf of their main organisation. EU EOMs, however, do not speak on behalf of the EU. The Handbook states that “An EU EOM is politically independent in its findings and conclusions […]. Although there will be close cooperation with the EU institutions, an EU EOM operates under a separate and distinct mandate from that of the EEAS, the European Commission and EU Delegation present in the country⁸.”

Legal and administrative framework for EU EOMs

12

EU election observation is guided by the principles of democracy, the rule of law, human rights, gender equality and fundamental freedoms as defined in the Treaties⁹. It is also guided by the Cotonou Agreement¹⁰.

13

EU EOMs are governed in greater detail by the Regulation establishing a financing instrument for democracy and human rights worldwide (EIDHR Regulation)¹¹, which sets the objective for the EU EOMs, stating that "They are best placed to provide both an informed assessment of electoral processes and recommendations for their further improvement in the context of Union cooperation and political dialogue with third countries.”

14

Following the adoption of the EIDHR Regulation in 2014, the Commission Decision on implementing arrangements for EU EOMs¹² and its annex then provided further details, mainly regarding the roles of EU EOM staff and their remuneration.

15

While many aspects relating to EU EOMs have changed in the meantime, such as the institutional set-up following the Lisbon Treaty, the Commission Communication from the year 2000¹³ remains the main guidance document for EU EOM objectives and methodology. This communication sets out the overall policy framework for EU EOMs.

16

The most thorough methodological framework for EU EOMs is the Handbook. Following its adoption, the Handbook was first published in 2002 and has been revised on two occasions, in 2008 and 2016.

17

Since its adoption, the Commission Communication has been supplemented by other formal documents setting objectives for election observation. The most important additions were the 2009 Council conclusions on democracy support in the EU’s external relations, and the 2012 EU Strategic Framework and Action Plan on Human Rights and Democracy (updated for the period 2015-19). These documents placed greater emphasis on follow-up to EU EOMs and EU Member States’ cooperation in this regard.

EU EOM financing

18

EU EOMs are implemented by the FPI under direct management. The cost of an EU EOM varies significantly depending on the specific circumstances of each mission. However, based on the latest figures available, the average cost is 3.5 million euro. The budgetary allocation for election observation in 2015 to 2017 amounted to around 44 million euro on average per year. This allows for approximately 10-12 EU EOMs, 8 EEMs and 4 EFMs in a year. EEMs and ExMs are performed under a framework contract and cost a maximum of 300 000 euro each. The cost of an EFM does not usually exceed 100 000 euro¹⁴.

Relationship with the OSCE

19

All EU Member States are also participating States in the Organization for Security and Cooperation in Europe (OSCE). The ODIHR carries out election observation in OSCE participating States. The EU and the ODIHR use comparable methodology. For these reasons the EU does not usually observe elections in the OSCE region. Dating back to the early 2000s, this gentleman’s agreement between the EU and the OSCE is not based on a written agreement.

EU EOM structure and the election observation cycle

20

An election is not a one-day event, but a process encompassing various stages. The electoral period may be deemed to start with the launch of the election campaign, the timing of which may differ from one country to another. The electoral period ends with the closure of the complaints procedure following the elections. EU EOMs cover every phase of the electoral cycle: the pre-election phase, Election Day itself, and the post-election phase. In addition to election observation, the European Commission also provides electoral assistance, i.e. general technical or material support for the electoral process.

21

Once an invitation has been accepted, the host country signs a Memorandum of Understanding with the European Union, setting out basic details regarding the mission. This covers the security and logistical aspects of the EU EOM but not the follow-up mission or recommendations for future electoral reform.

22

EU observers can be long- or short-term observers. Long-term observers are based in regions across the host country and are present from the campaign launch to the announcement of the final results and the closure of the complaints procedure¹⁵. Short-term observers cover a wide geographical area for a few days before and after Election Day. In addition, there is a Core Team, usually comprising eight to ten experts, based in the capital¹⁶.

23

The Core Team experts specialise in different aspects of the electoral environment and may, for instance, include a legal analyst, a media analyst, and a press officer (see Figure 1). The Core Team, under the management of the Deputy Chief Observer, contributes to all mission reports, including the EU EOM’s final report.

24

A service provider contracted by the Commission provides logistical, security and administrative support for EU EOMs¹⁷.

25

An EU EOM issues a public preliminary statement shortly after Election Day (usually within two days) at a press conference led by the Chief Observer. A comprehensive final report is issued within two months of completion of the election process, containing recommendations to the host country authorities on how to improve the framework for future elections¹⁸ (see Figure 2).

Election observation recommendations and follow-up

26

According to the UN Declaration of Principles, “International election observation should, when possible, offer recommendations for improving the integrity and effectiveness of electoral and related processes, while not interfering in and thus hindering such processes”¹⁹.

27

In their final reports, EU EOMs cover the following fourteen areas of assessment²⁰: (i) political context; (ii) legal framework (including electoral system); (iii) election administration; (iv) voter registration; (v) party and candidate registration; (vi) election campaign; (vii) electoral violence; (viii) media; (ix) online election-related content; (x) complaints and appeals; (xi) human rights (including participation of women, minorities and people with disabilities); (xii) role of civil society; (xiii) election day; (xiv) results and the post-election environment. A 2012 European Parliament study estimated that around 90 % of EU EOM recommendations suggest changes in the legal framework of the host country or electoral reform²¹.

28

The countries subject to an EU EOM have not undertaken a formal commitment to implement electoral recommendations, beyond their existing obligations as signatories to international conventions such as the Universal Declaration of Human Rights and the International Covenant on Civil and Political Rights. As such, EU EOMs operate in a different environment to, for example, the OSCE, whose participating States have signed a document formally obliging them to implement recommendations²².

29

Implementation of EU EOM recommendations can be supported in many different ways. The main tools at the disposal of the EEAS and the Commission are:

• political dialogue with the host country;
• EFMs;
• EU electoral assistance.

Audit scope and approach

30

The subject of this audit was EU EOMs, with a particular focus on the follow-up of the recommendations issued by these missions. The main audit question was “Do the EEAS and the Commission make reasonable efforts to support the implementation of EU Election Observation Missions’ recommendations?”

31

Not only is effective follow-up vital for promoting electoral reform in host countries, but it is also the key to maximising the impact of the investment already made in observation missions. As explained in paragraph 18, the average allocation for EU EOMs is 44 million euro per year. This represents one-quarter of the total funds available under the EIDHR. The Handbook highlights the importance of follow-up by stating that “follow-up to EU EOM recommendations is a key priority for EU institutions”²³. Moreover, according to the OSCE²⁴ “Election observation will be beneficial only to the degree that [these] recommendations are given serious consideration and effectively addressed. If they are not, there is a real risk that electoral shortcomings and vulnerabilities will persist and, in all likelihood, increase.”

32

The Court has not carried out a performance audit in the area of electoral support since 1996²⁵. Moreover, the publication of this special report is scheduled so as to feed into discussions ahead of the mid-term review of the EIDHR Regulation to be conducted by the Commission by 31 March 2018.

33

The audit, carried out between January and May 2017, covered four host countries as cases for illustrating the effectiveness of EU EOM follow-up: Ghana, Jordan, Nigeria and Sri Lanka, with a focus on elections observed after 2010. On-the-spot checks took place in Ghana and Jordan. We used two main criteria to select countries for the audit. First, the countries should have hosted at least two previous EU EOMs with the most recent in 2015 or 2016. Second, we aimed for a geographical spread.

34

Two EU EOMs had taken place in Jordan and Nigeria in recent years. In Ghana, the previous elections in 2012 had only been covered by an EEM, but electoral reform recommendations were issued nonetheless. EEM recommendations are usually not made public. In the case of Ghana, however, the recommendations had been shared with the Electoral Commission, the EU institutions and EU Member States and could serve as a basis for EU electoral assistance. Sri Lanka was selected specifically because an EFM had been scheduled for spring 2017 and we planned to accompany the EFM. The EEAS decided to postpone the EFM in view of the situation in the country, but Sri Lanka was nonetheless covered by the audit via a desk review.

35

Considering that almost half of EU EOMs take place in Africa²⁶, we chose to cover two African countries, as well as one Asian country and one country in the European Neighbourhood. For a map of EU EOMs see Annex I.

36

First, we assessed the formulation of the recommendations, including the degree to which stakeholders in the host country had been consulted to ensure local ownership. We also assessed whether ExMs had taken previous recommendations into account in the preparation of the next EU EOM.

37

We then assessed the use of the main tools available to the EEAS and the Commission for supporting implementation of EU EOM recommendations. The tools for follow-up assessed in the audit were limited to those described in paragraph 29. We did consider that there are other tools for encouraging implementation of the recommendations, such as the special incentive under the Generalised Scheme of Preferences, GSP+²⁷, actions in international fora, the European Parliament’s political dialogue and budget support. These tools, though, are not included in the scope of this audit.

38

Lastly, we assessed EEAS and Commission monitoring of the implementation of the recommendations, the sharing of best practices and lessons learned.

39

Ultimate responsibility for implementing the recommendations lies with the host country governments. Although ensuring implementation of EU EOM recommendations is one aim of the EU EOM final reports, it is important to note that the purpose of the audit was not to assess the state of electoral reform in the host countries. Furthermore, the causal link between EU efforts to support implementation and the actual results is not strong enough for us to draw conclusions in this regard.

Observations

PART I – Recommendations have improved in recent years but more consultation is needed on the ground

40

For this section, we examined the preparation of the EU EOM recommendations from two perspectives: their presentation within the final reports and the Core Team’s consultations with host country stakeholders. A standard presentation would facilitate comparability over time. Consulting on draft recommendations is key both for ensuring accuracy and for promoting local ownership to ultimately increase the likelihood of their implementation by the host countries.

Recommendations are more comparable and their presentation more standardised but guidelines are not always followed

41

For this section, we considered the final reports from the following elections:

• Ghana: 2016;
• Jordan: 2013 and 2016;
• Nigeria: 2011 and 2015;
• Sri Lanka: 2015.

An assessment of the recommendations is presented in Annex II.

42

The EEAS, through the EU-funded Election Observation and Democracy Support (EODS) project, provides the independent EU EOM Core Team with methodological support for drafting the EU EOM final report. The EEAS is also in charge of carrying out quality control on the final reports by assessing the form of the draft report and the clarity of the proposed recommendations²⁸. The EEAS and EU Delegations comment on whether recommendations are constructive in the country-specific context and may advise on prioritisation of recommendations.

43

The EEAS (through the EODS project) introduced drafting guidelines in 2012, which included a recommendation template. The guidelines were revised in July 2016. The reports we assessed, published after 2010, had all based the presentation of recommendations on the template.

44

Overall, we found that EU EOM reports written after publication of the guidelines contained fewer and better presented recommendations.

45

Giving too many recommendations could be counterproductive for many reasons. For example, host country authorities may become overwhelmed and the work of the EU Delegations to support implementation could become more complicated. As a result, there is a risk that the most important recommendations might not be paid sufficient attention in follow-up discussions. The guidelines suggest a maximum of 30 recommendations. The post-guideline reports contained between 20 and 33 recommendations. This is an improvement on previous years, with the 2011 Nigeria report, for example, containing as many as 50 recommendations.

46

The Handbook provides a comprehensive overview of EU EOMs covering all the main aspects addressed by observers on the ground. The Handbook states that “in order to ensure accessibility for all stakeholders, the final report should avoid highly technical language or jargon”²⁹. While it is sometimes indispensable to use technical language in order to avoid misunderstandings, we found an excessive use of long words, long sentences, the passive tense and highly technical language and acronyms. Nevertheless, most stakeholders interviewed in the host countries said that they understood the recommendations, indicating that they are comprehensible for the expert reader, but not necessarily for the general public. Indeed, electoral reform can cover many different areas, involving stakeholders in various fields. Thus, recommendations that are universally understandable could be beneficial for engaging a wide range of actors in efforts to improve the electoral system.

47

The stakeholders interviewed considered that overall the recommendations were constructive and sufficiently detailed without being too prescriptive, which is also requested in the drafting guidelines.

48

While the guidelines provide a template for recommendation layout, it is not always followed. For instance, it is not always made explicit or sufficiently clear which body the recommendation is addressing. As an example, 22 recommendations in the Jordan 2016 report are addressed to the “legislature” in general, leaving some ambiguity as to which institution should submit a legislative proposal. An analysis of the 2016 recommendations by an EU-funded electoral assistance project implemented by the UNDP shows that most of these recommendations should in fact have been addressed to the government.

49

The Ghana 2016 EU EOM report directed eight recommendations to more than one addressee³⁰: to the “Electoral Commission/Parliament”, the “Court Committee/Parliament” or “the Minister of Communications, the Parliament”. Although the addressees are jointly responsible for legislative change, it remains unclear which institution has the primary responsibility for implementing the recommendations. The EU EOM could facilitate the implementation of recommendations by determining, in consultation with the host country authorities, which institution has the primary responsibility.

50

In general, recommendations were, as set out in the Handbook, based on concrete examples given in the main body of the report to illustrate the issues triggering the recommendation³¹. However, no cross-references were used between the recommendations and the main body of the report, even though these are required by the template. This made it difficult to find the corresponding sections in the main body of the report.

51

Despite improvements in recent years, another element that was sometimes lacking was a reference to obligations based on international conventions. While the guidelines only request for this to be included where such obligations apply to the recommendations, we encountered recommendations which lacked such a reference despite being clearly embedded in international law³².

52

In the Jordan 2016 report, none of the recommendations indicate which national law is affected by the recommendation, even though the guidelines suggest that EU EOMs assess this aspect³³.

53

All reports contain a number of priority recommendations. Although prioritisation is not a straightforward exercise, it is worth noting that the reasoning behind the selection of priority recommendations was not documented. Moreover, the drafting guidelines do not require any such documentation. The Jordan 2016 report is the only one which described the rationale for prioritisation to a certain extent, and in which recommendations were split into three categories: priority, short-term, and long-term recommendations.

54

The proportion of priority recommendations in the latest EU EOM reports for the countries covered by the audit varied from 23 % to 45 % of the total number of recommendations.

55

Only two of the six³⁴ EU EOM final reports, namely Jordan 2016 and Nigeria 2011, split the recommendations into long-term and short-term recommendations. The other reports made no reference to a time horizon for implementation of the recommendations. Although not required by the guidelines, such a time horizon could facilitate prioritisation of actions by host countries and thus speed up implementation of recommendations.

EU EOMs met stakeholders extensively before Election Day but did not consult on the recommendations

56

Regarding the consultation of stakeholders, we focused on the latest EU EOMs in the countries covered by the audit in order to ensure that the information we obtained was as correct and complete as possible.

57

We found evidence that the EEAS headquarters in Brussels and, in most cases, also the EU Delegations had been consulted in all four countries on the recommendations for the latest EU EOMs throughout the process of drawing up the final EU EOM report. This was documented in emails and minor comments in documents.

Consultation before Election Day

58

Meetings were held to discuss electoral issues with a wide variety of stakeholders in the host countries prior to Election Day. These meetings are mentioned in interim and internal reports. However, none of the reports provide details on the outcome of these stakeholder consultations, despite the fact that the drafting guidelines request that diverging opinions be documented. Several Core Team experts explained at interview that the information had been documented in notes, but that these notes had remained in their possession and not been stored centrally.

Consultation while drafting the EU EOM final report

59

The guidelines explicitly state that “Discussion with stakeholders of EU EOM recommendations gives [an] opportunity for correction of any misunderstandings by the mission, and facilitates identification of priorities. It is important to ask a variety of interlocutors about each recommendation”³⁵. The EEAS has further explained that while writing the report, the EU EOM should consult local stakeholders on the recommendations at the same time as discussing them with the EEAS headquarters.

60

Although many issues were discussed with local stakeholders and the Core Team experts may already have been considering recommendations from the outset, we were unable to find evidence that they had been consulted on the recommendations in any of the countries covered by the audit. This was confirmed during our meetings with local stakeholders, who all said that they had indeed not been consulted on the recommendations. Several Core Team experts said at interview that they believed that consulting stakeholders on recommendations may be seen to contradict the principle of independence. We however do not consider that consultation on the text of the recommendations would contradict independence. After all, consulting stakeholders is a practice which is in line not only with the EEAS guidelines but also the UN Declaration of Principles, which states that “International election observation missions may also conduct private meetings with those concerned with organizing genuine democratic elections in a country to discuss the mission’s findings, conclusions and recommendations”³⁶.

61

We also found no evidence that discussions on the most technical matters had taken place in any of the four countries. As a result, experts in the field challenged some recommendations following the publication of the final reports. If the Core Team had consulted stakeholders on the technical aspects of the recommendations, technical aspects could have been clarified from the outset. Consultation during the preparation of recommendations is important as there is a risk that differences in perception that arise after publication of the final report could undermine EU EOM recommendations and their implementation.

62

The EEAS explained that, before leaving the host country, the Chief Observer pays courtesy visits to key stakeholders during which issues related to recommendations are discussed. However, we found no evidence documenting these meetings thus we could not confirm their content.

Consultation after publication of the EU EOM final report

63

Following the latest EU EOMs, stakeholder roundtables were held in all four countries, with a range of local stakeholders invited to discuss the final report and its recommendations after publication, during the Chief Observer’s return visit. They were mostly inclusive, although one political party in Jordan and one civil society organisation in Ghana claimed that they had not been invited to the meeting.

64

On the return visit, the Chief Observer typically remains in the host country for a few days. The final report is shared with stakeholders after it has been formally presented to the country’s authorities and following a press conference. As the Chief Observer only stays in the country for a short time, in all four countries, the stakeholder roundtable was scheduled too soon (immediately or 1-2 days) after the press conference to allow participants to familiarise themselves with the content of the final report and its recommendations and, consequently, the scope for useful discussion was limited. Where a Chief Observer is unable to stay longer in the host country, he/she could use the option of delegating responsibility for chairing the roundtable to the Deputy Chief Observer.

65

A return visit report is drawn up following the visit, sometimes outlining a number of lessons learned. The reports for all the countries covered by the audit stated that stakeholders had cited the issue of a lack of time between the release of the final report and the stakeholder roundtables. Although this issue had already been raised in the return visit report for Sri Lanka in 2015, the problem remained for the return visits to Jordan in 2016 and Ghana in 2017, where several stakeholders highlighted this issue in meetings with the audit team.

66

There are only limited guidelines in place establishing the structure of the return visit. The EU Delegations found the guidelines unclear regarding responsibility for organising the return visit; some EU Delegations made all the practical arrangements, while others used the EU EOM service provider, which was still in operation at the time of the return visit. In Ghana, the stakeholder roundtable was organised by the Special Election Working Group co-chaired by the EU and the UNDP. The timing of the return visit also varied between the countries depending on the availability of the MEP acting as Chief Observer. The return visits to Ghana, Jordan and Sri Lanka took place approximately two months after Election Day, while the visit to Nigeria took place after six months.

PART II – Political dialogue and electoral assistance were used as tools to support implementation of the recommendations, but EFMs were not deployed as often as they could have been

67

For this section, we assessed the main tools used by the EEAS and the Commission to support the implementation of EU EOM recommendations: political dialogue with host country authorities, EFMs and electoral assistance. To maximise the impact of election observation, it is essential to address EU EOM recommendations in political dialogue with the host countries, and to ensure complementarities of electoral assistance with the EU EOM recommendations.

68

There are many reasons why a recommendation might not be implemented. Host countries may, for example, be unwilling or unable to follow up electoral recommendations owing to:

1. a lack of political will or support in parliament;
2. a lack of professional expertise, financial resources or time to implement reforms;
3. an outbreak of war, instability or civil unrest;
4. concerns among national counterparts that recommendations are unconvincing, insufficiently targeted or inappropriate to the national context³⁷.

69

While the implementation of EU EOM recommendations ultimately depends on the host countries’ political will and capacity for implementation, the EEAS and the Commission have a number of tools at their disposal for supporting the host countries. These include political dialogue with national authorities carried out primarily by the EU Delegations and EFMs, involving discussions in which both the EEAS and the Commission participate.

70

In addition, implementation can be supported by different types of EU funding, such as through electoral assistance programmes, managed by the Directorate-General for International Cooperation and Development (DG DEVCO) and the Directorate-General for Neighbourhood and Enlargement Negotiations (DG NEAR). We assessed the link between the EU EOM recommendations and programming of EU-funded electoral assistance. In this context, we also examined the EU Delegation’s efforts to coordinate electoral assistance with the EU Member States.

Political dialogue took place, but certain host country bodies would like further discussion

71

The EU EOM operates under a separate and distinct mandate from that of the EU institutions. It is only following presentation of the final report in the host country that it becomes “up to the EU Delegation and Member States to plan and carry out activities to support follow-up to the EU EOM recommendations”³⁸.

72

EU Delegation support for the implementation of EU EOM recommendations only begins once an EU EOM is closed and the final report has been delivered to the national authorities. From that point on, the EU’s political dialogue with the host country plays an important role in supporting the implementation of the recommendations.

73

The EIDHR Regulation states that EU EOMs are “best placed to provide both an informed assessment of electoral processes and recommendations for their further improvement in the context of Union cooperation and political dialogue with third countries”³⁹.

74

Political dialogue is crucial for promoting national ownership and follow-up to EU EOM recommendations, as host countries are not formally committed in this respect.

75

A recent evaluation of EU EOMs suggested “making a reference to the issuing and importance of the recommendations” in the Memorandum of Understanding preceding each EU EOM⁴⁰. We believe that such a reference to the recommendations in the Memorandum of Understanding could provide the EU Delegations with a useful starting point for conducting discussions with the host country authorities after elections.

76

Political dialogue with the host country is provided for in the Cotonou Agreement⁴¹ for Ghana and Nigeria, the Association Agreement with Jordan⁴² and the Cooperation Agreement with Sri Lanka⁴³. According to Article 8 of the Cotonou Agreement, the objective of this dialogue is “to exchange information, to foster mutual understanding, and to facilitate the establishment of agreed priorities”.

77

The EU EOM recommendations, or issues addressed by them, were brought up at the annual high-level political dialogue meetings in all four host countries covered by the audit. They were also discussed in human rights working groups.

78

In Ghana, for instance, the Election Expert Mission recommended in 2012 that legislation on the right to information should be developed to improve the transparency and accountability of public affairs, including the electoral sector. Progress made in this regard was welcomed during the Article 8 political dialogue held in June 2016.

79

In Jordan such issues are discussed at the Subcommittee on Human Rights, composed of representatives from the Jordanian authorities, the EEAS and EU ambassadors to the country. At its latest meeting in October 2016, the EU Delegation and the EEAS geographical desk raised several key issues from the preliminary statement which subsequently became EU EOM recommendations in the final report. These included restrictions on young candidates’ right to stand, scope for enhancing equality of vote, and restrictions on domestic observers’ access to the tabulation of results⁴⁴.

80

Similarly, at the EU-Nigeria Human Rights Dialogue in November 2016, the EU Delegation and the Nigerian Ministry of Foreign Affairs exchanged ideas on how to sustain electoral reform, for example by reinforcing the channels of civic engagement, an issue which was addressed by an EU EOM recommendation⁴⁵.

81

In Sri Lanka the Working Group on Governance, Rule of Law and Human Rights met for the first time in January 2016. At this meeting, the Sri Lankan authorities and the EU discussed electoral reform and follow-up to the EU EOM recommendations as one of 14 items on the agenda. The working group agreed to take two steps: to introduce electoral reform under the new constitution and to set a minimum quota of 25 % for the parliamentary representation of women.

82

The EU Delegations also organised other working-level meetings with key counterparts, such as ministries and the Electoral Commission, in each of the countries covered by the audit to discuss issues related to EU EOM recommendations. However, they did not document the outcome of these working-level meetings.

83

While a large part of the political dialogue concerned other priorities, we found that host country bodies were keen to discuss the EU EOM recommendations further. In both Ghana and Jordan, this wish was expressed by civil society organisations and the parliamentary legal committees. These committees proposed organising workshops to discuss the possible implementation of EU EOM recommendations with the EU Delegation, as they were not familiar with the final report prior to our audit visit. This shows that there is scope for increasing activities related to political dialogue to support the implementation of the recommendations⁴⁶.

EFMs were not deployed as often as they could have been

84

The Handbook states that EFMs may be deployed to take stock of the way the recommendations in the final report have been dealt with, to contribute to ongoing political dialogue with the country on democracy and human rights, and to identify areas for future reform and intervention. Follow-up missions may, when possible and appropriate, be headed by the former Chief Observer of the latest EU EOM deployed in the given host country. Approximately four EFMs are carried out annually, meaning that not all countries hosting an EU EOM will automatically be revisited for an EFM. The EEAS, in close cooperation with the EU Delegation, decides on the deployment and timing of EFMs.

85

The only country in the sample that had hosted an EFM was Nigeria. During the audit, a possible EFM to Sri Lanka was under discussion, but the EEAS decided to postpone it since the country was undergoing significant political reform. This decision was supported by the Sri Lankan EU Delegation.

86

In Nigeria, where an EFM had taken place from May to June 2014, the EU Delegation confirmed that it was widely considered to have been useful and timely. The EFM had carried out high-visibility meetings, identified critical areas that needed to be addressed immediately, and made recommendations for consideration by the Nigerian authorities, the National Assembly, the media and civil society⁴⁷.

87

The EFM drew up a chart on the implementation status of previous EU EOM recommendations and organised a roundtable with stakeholders to discuss and endorse the recommendations. The EFM findings were also used by the subsequent EEM deployed to Nigeria from 19 November to 16 December 2014.

88

In the countries covered by the audit that had not hosted an EFM in the period between elections, the vast majority of national stakeholders interviewed considered that an EFM would have provided a useful political impulse through high-visibility meetings. As explained in paragraph 18, considering the money already invested in EU EOMs, EFMs do not represent a substantial cost, but could greatly benefit follow-up if used more frequently.

Electoral assistance supported the implementation of EU EOM recommendations

89

The EU is one of the leading actors in the field of electoral assistance and provides technical and material support for electoral processes in many partner countries. Contrary to the EU EOM, which focuses on the elections themselves, electoral assistance may be provided throughout the electoral cycle and is usually broader in scope. Electoral assistance may include professional support in establishing a legal framework, the provision of voting material and equipment, or help in registering political parties and voters. It may also include support to civil society organisations in areas such as voter education or the training of local observers, as well as support to the media⁴⁸. In most cases, electoral assistance is financed from the geographical funds – the European Development Fund, the Development Cooperation Instrument and the European Neighbourhood Instrument.

90

The Handbook notes a significant increase in the funding and coverage of EU electoral assistance in recent years⁴⁹. Nevertheless, the Commission could not provide us with a financial overview of electoral assistance for all recipient countries⁵⁰. The reason for this is that support focusing on other areas also includes components aimed at promoting electoral reform. Moreover, not all EUDs in the countries covered by the audit kept a documented overview of EU and Member States’ electoral assistance programmes.

91

Although an EU EOM is politically independent from any EU-funded technical assistance projects that may be taking place in the country being observed⁵¹, the Commission has identified complementarity as a key factor. It has concluded that election observation and electoral assistance “are different activities but essentially they should be considered and programmed in a complementary manner”⁵².

92

More recently, the importance of complementarity was reiterated in the Commission’s guide on electoral assistance⁵³ and in the 2014 EIDHR Regulation⁵⁴.

93

Since the majority of EU EOM recommendations relate to legal reform in the host countries, their implementation depends far more on the political will of the host country government than on the provision of EU funding. Nevertheless, we found several areas where it was possible to link funding and recommendations, and where the link had also often been made in programming and formulating EU electoral assistance (see Box 1).

94

In addition, the EU programming exercise follows a timetable over the year, which does not necessarily coincide with the publication of EU EOM final reports following an election. However, recommendations from previous EU EOMs may be used in subsequent programming exercises.

95

Importantly, we found no cases where EU electoral assistance was in direct contrast with the recommendations. This is due to the fact that the information collected during the EU EOM was already known to some extent by the programme managers as a result of their work with the host country. For example, we came across a project in Sri Lanka which had been signed three months prior to the finalisation of EU EOM recommendations. The project was fully in line with the recommendations.

96

As regards the EU Delegation’s coordination with EU Member States, our analysis of previous and ongoing electoral assistance in the countries covered by the audit shows that there has been no direct overlap between EU assistance and that financed by EU Member States. This is attributable to the fact that the EU Delegations have organised meetings with EU Member States present in the country to discuss electoral assistance.

97

As an example of coordinated assistance, in 2016 the Commission adopted a comprehensive programme called Enhanced Support to Democratic Governance in Jordan with an EU contribution of 15 million euro. To increase synergies, EU Member States present in Amman were asked if they would be interested in implementing and co-financing the programme. The Spanish Cooperation Agency AECID decided to co-finance the programme with an additional 2 million euro and to implement the component on support to civil society.

PART III – There is no central overview of the recommendations and their implementation status

98

In this part of the audit, we analysed the efforts made by the EEAS to monitor the EU EOM recommendations at a centralised level and to track their implementation in the host countries. Awareness at central level of the progress made helps better target future follow-up activities. In this context, we also assessed the identification of best practices and the lessons learned processes in place.

There is no centralised depository for EU EOM recommendations

99

EU EOMs remain a particularly challenging field when it comes to preserving institutional memory due to staff mobility. For instance, it is relatively rare for a staff member to remain in the same EU Delegation for two consecutive elections. The EEAS Division for Democracy and Electoral Observation provides an important focal point within the EU institutions for election observation⁵⁶.

100

Moreover, as the MEP is temporarily nominated as Chief Observer and Core Team experts are recruited for one EU EOM at a time, there is a strong need to document and centrally store the work carried out for future reference. The EU EOM activities are reported in interim, internal and return visit reports. However, these reports do not contain sufficient detail on stakeholder consultations.

101

Back in 2012, a study commissioned by the European Parliament suggested that there should be a database in place to ensure a central tracking of the implementation status of the recommendations⁵⁷. However, no centralised depository for EU EOM recommendations yet exists.

102

The EODS project has a website with an extensive archive. This can mitigate the impact of staff mobility since it functions as an alternative to in-house institutional memory. The project experts train Core Teams and other observers in order to build up EU election observation capacity and methodology.

103

As there is no centralised depository for EU EOM recommendations, Core Team experts preparing new recommendations on similar topics do not have easy access to recommendations by country and/or theme. The same is true for colleagues in other EU institutions with a particular interest in the country.

104

As one example, the Organization of American States (OAS) has set up a database of this kind for storing recommendations. Database users can search by country, year and the following eight thematic areas: electoral justice, media, political financing, electoral organization, gender, under-represented groups, electoral technology, and political and legal systems.

There is no systematic assessment of progress in implementing recommendations

105

We also assessed whether the EEAS regularly assessed and documented the implementation status of EU EOM recommendations.

106

As far back as 2014, the Commission’s Internal Audit Service reported that there should be a robust central follow-up of recommendations. It also suggested that the EU Delegations should report to the EEAS and the FPI on the response host countries give to the EU EOM recommendations⁵⁸.

107

The ExM report is currently the only place where an assessment of the implementation status is systematically required. For all ExM reports, the EEAS provides a template with a structured grid for assessing the implementation status of all previous EU EOM recommendations for that country, one by one. However, as explained in paragraph 5, it should be noted that the main purpose of ExM reports is not to assess previous recommendations. ExMs serve as a basis for deciding whether or not to deploy an EU EOM.

108

In three of the countries covered by the audit (Ghana 2016, Jordan 2016 and Nigeria 2015), the grid in the ExM report had not been filled in. Instead, the reports contained a paragraph with a general description of the implementation status. For Sri Lanka, the grid had been filled in but no comprehensive analysis had been carried out. As a result, reporting in all four countries was not comprehensive, providing a summary rather than a detailed assessment of implementation status. We therefore conclude that the ExMs missed the opportunity to assess the implementation status of previous recommendations. While the implementation status was not assessed in the ExM reports, we came across examples of such assessments in other documents. The 2014 EFM in Nigeria, for example, carried out an analysis of this kind.

109

In Jordan, the EU-funded electoral assistance project implemented by the UNDP can serve as one relevant example of how such an assessment might be done. Four reports were prepared between March 2016 and January 2017 assessing the implementation status of each of the 2013 EU EOM recommendations. Using a traffic-light system to assess implementation status (see Figure 3), this “tracking project” provides the EU Delegation in Amman and EEAS in Brussels with a timely and detailed overview of progress. According to the UNDP, the project uses a flexible methodology and so could potentially be replicated in other countries. The project also analysed the 2016 recommendations with a view to facilitating their implementation ahead of the local elections in August 2017.

110

While there are initiatives assessing implementation status, there is no continuous process in place whereby such an analysis is systematically carried out after each EU EOM. The EEAS has explained that EU Delegations are generally not sufficiently staffed and sometimes lack the in-depth knowledge required to assess electoral reform in the country.

111

Implementation status is occasionally assessed in electoral assistance projects financed by the European Development Fund or the European Neighbourhood Instrument, but this is not the case in all EU EOM host countries benefiting from electoral assistance.

112

In June 2017, DG NEAR approved a project under the Instrument for Pre-Accession Assistance⁵⁹ to assess the level of implementation of recommendations (primarily from ODIHR EOMs⁶⁰) in candidate countries and potential candidate countries in the Western Balkans.

113

In order to measure the performance of EU EOMs in general, the Commission uses indicators set out in the Multiannual indicative programme⁶¹. These are output indicators, such as the number of observers trained. However, there are no clear result indicators for EU EOMs allowing the Commission to measure their performance. The uneven quality of results indicators at instrument level was also noted in a recent evaluation on EIDHR⁶².

Best-practice exercise carried out recently but lessons learned process does not systematically cover follow-up

114

In June 2017, the EEAS published its work “Beyond Election Day”. This provides the first comprehensive overview of the tools in place for following up EU EOM recommendations. The document not only explains in depth the tools at the EU’s disposal, but also provides numerous examples of best practice in different contexts and shows a variety of ways in which follow-up could be fostered. The document will act as a source of information for those tasked with following up EU EOM recommendations, such as the EU Delegations.

115

A lesson learned exercise is carried out after every EU EOM. This exercise, managed by the FPI, is systematic and covers all logistical and security aspects of the mission. In addition, after each EU EOM, the EEAS Headquarters organises a debriefing with the Core Team to discuss the content of the EU EOM final report and the recommendations. These meetings may address methodological issues, but they take place before the report is finalised and are not documented for future reference.

116

Lessons learned seminars on methodological aspects are regularly organised by the EEAS with EODS, most recently in April 2017.

117

In April 2016, the EEAS requested input from the EU Delegations that had hosted an EU EOM regarding their action to support recommendation follow-up. Standardised and carried out regularly, such an exercise has the potential to serve as a starting point for sharing lessons learned from follow-up activities across the EU Delegations. However, while these reports were useful and helped provide an overview of the state of play at a certain point in time, they were inconsistent in the level of detail they provided.

Conclusions and recommendations

118

We conclude that the EEAS and the Commission made reasonable efforts to support the implementation of the EU EOM recommendations using the tools at their disposal. Nevertheless, certain adjustments could be made to the process to enhance this support further. We are therefore making a number of recommendations to the EEAS.

EU EOM recommendations have improved in recent years but more consultation is needed on the ground

119

Final reports and the presentation of recommendations have improved in recent years, following the introduction of the EEAS drafting guidelines. This is likely to facilitate comparability over time in the future. However, there are areas where the guidelines are not followed. Moreover, the reports are not consistent in referring to a time horizon for implementing recommendations. The Core Team is also not required to document the rationale behind prioritisation (see paragraphs 41 to 55).

120

The Core Teams extensively discussed issues relating to the recommendations with stakeholders in the host countries prior to Election Day, but did not document the outcome of these meetings or share their notes with EEAS headquarters. The EU EOMs covered by the audit consulted the EEAS and, in most cases, EU Delegations, but did not consult stakeholders in the host countries on the draft recommendations (see paragraphs 56 to 62).

121

Stakeholder roundtables took place in each country following presentation of the final report and were mostly inclusive. However, they were scheduled too soon after the press conference to allow stakeholders to familiarise themselves with the content of the final report and its recommendations. Moreover, there are only limited guidelines in place for the organisation of the return visit (see paragraphs 63 to 66).

Political dialogue and electoral assistance were used as tools to support implementation of the recommendations, but EFMs were not deployed as often as they could have been

122

The EU Delegations raised issues contained in EU EOM recommendations in political discussions with host country authorities. Yet there is still scope for further increasing activities related to political dialogue since we observed a willingness to discuss the possible implementation of EU EOM recommendations in host country parliaments (see paragraphs 71 to 83).

123

Nigeria was the only country in the sample to which an EFM was deployed, and the EU Delegation deemed it timely and useful. In the other countries covered by the audit, the vast majority of national stakeholders interviewed welcomed the possibility of an EFM between elections to raise awareness about the recommendations through high-visibility meetings. EFMs do not entail substantial costs, but could greatly benefit follow-up if used more frequently (see paragraphs 84 to 88).

124

We found numerous cases where it was possible to link electoral assistance with EU EOM recommendations, and where the Commission had considered EU EOM recommendations when designing the programmes. This enhanced complementarity between election observation and electoral assistance (see paragraphs 89 to 97).

There is no central overview of the recommendations and their implementation status

125

Given the independent nature of EU EOMs and staff mobility in EU Delegations and the EEAS, there is a strong need to preserve institutional memory through other means. There is no centralised depository for storing recommendations in order to make them easily accessible to Core Team experts or EU staff (see paragraphs 99 to 104).

126

While we found good examples of work tracking the implementation status of recommendations, there was no systematic process for regularly assessing and storing status details (see paragraphs 105 to 113).

127

The EEAS recently carried out comprehensive work to list the tools in place and collect best practices for following up EU EOM recommendations, but there is no ongoing systematic lessons learned process for follow-up (see paragraphs 114 to 117).

Recommendation 1 - Drafting guidelines

The EEAS should:

• ensure, through systematic quality control on the final report’s format before it is published, that the EU EOMs follow the drafting guidelines and template;
• update the drafting guidelines to request that EU EOMs document the rationale behind the prioritisation and discuss with stakeholders on a time horizon (short, medium, or long-term) for implementing recommendations.

Time frame: End of 2018.

Recommendation 2 - Consultations with stakeholders

The EEAS should:

• systematically ensure that the Core Team consults stakeholders in the host country on the recommendations before the report is finalised;
• request that the outcome of EU EOM consultations with stakeholders be documented and stored centrally.

Time frame: Mid-2018.

Recommendation 3 - Return visits and roundtables

The EEAS should:

• clarify roles and responsibilities for organising the return visit;
• ensure that the stakeholder roundtable is scheduled at least four working days after the report is released, giving participants enough time to familiarise themselves with the report and its recommendations ahead of the meeting.

Time frame: Mid-2018.

Recommendation 4 – Refining tools

The EEAS should:

• make reference in the Memoranda of Understanding to the fact that the upcoming EU EOM will submit recommendations, and emphasise their importance;
• when possible, deploy EFMs to countries that have hosted an EU EOM, at a suitable time between elections, provided that the political or security situation in the host country so allows.

Time frame: Start 2019.

Recommendation 5 – Monitoring and results

The EEAS should:

• set up a centralised depository for EU EOM recommendations;
• track progress on implementing EU EOM recommendations on a regular basis;
• in the context of EU coordination, ask the EU Delegations to report regularly on the activities carried out in the host country to support implementation of the recommendations. This should feed into a lessons learned process on follow-up.

Time frame: Start 2019.

This Report was adopted by Chamber III, headed by Mr Karel PINXTEN, Member of the Court of Auditors, in Luxembourg at its meeting of 16 November 2017.

Acronyms and abbreviations

DG DEVCO: Directorate-General for International Cooperation and Development

DG NEAR: Directorate-General for Neighbourhood and Enlargement Negotiations

EAT: Election Assessment Team

EEAS: European External Action Service

EEM: Election Expert Mission

EFM: Election Follow-up Mission

EIDHR: European Instrument for Democracy and Human Rights

EU EOM: European Union Election Observation Mission

EODS: Election Observation and Democracy Support

ExM: Exploratory Mission

FPI: Service for Foreign Policy Instruments

HR/VP: High Representative for Foreign Affairs and Security Policy/Vice- President of the European Commission

MEP: Member of European Parliament

OAS: Organization of American States

ODIHR: OSCE Office for Democratic Institutions and Human Rights

OSCE: Organization for Security and Co-operation in Europe

UNDP: United Nations Development Programme

Audit team

The ECA’s special reports set out the results of its performance and compliance audits of specific budgetary areas or management topics. The ECA selects and designs these audit tasks to be of maximum impact by considering the risks to performance or compliance, the level of income or spending involved, forthcoming developments and political and public interest.

This performance audit was produced by Chamber III – headed by ECA Member Karel Pinxten – which is responsible for the audit of the external actions’ and security and justice spending areas.

The audit was led by ECA Member Ville Itälä, supported by Turo Hentilä, head of private office; Helka Nykänen, private office attaché; Beatrix Lesiewicz, principal manager; Kristina Maksinen, head of task. The audit team consisted of Balázs Kaszap, Jiri Lang and Erika Söveges.

Contact

EUROPEAN COURT OF AUDITORS
12, rue Alcide De Gasperi
1615 Luxembourg
LUXEMBOURG

Tel. +352 4398-1
Enquiries: eca.europa.eu/en/Pages/ContactForm.aspx
Website: eca.europa.eu
Twitter: @EUAuditors

More information on the European Union is available on the internet (http://europa.eu).

Luxembourg: Publications Office of the European Union, 2017

PDF	ISBN 978-92-872-8763-2	ISSN 1977-5679	doi:10.2865/551559	QJ-AB-17-022-EN-N
HTML	ISBN 978-92-872-8725-0	ISSN 1977-5679	doi:10.2865/09843	QJ-AB-17-022-EN-Q

© European Union, 2017.

For any use or reproduction of photos or other material that is not under the European Union copyright, permission must be sought directly from the copyright holders.

GETTING IN TOUCH WITH THE EU

In person
All over the European Union there are hundreds of Europe Direct Information Centres. You can find the address of the centre nearest you at: https://europa.eu/european-union/contact_en

On the phone or by e-mail
Europe Direct is a service that answers your questions about the European Union. You can contact this service

• by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),
• at the following standard number: +32 22999696 or
• by electronic mail via: https://europa.eu/european-union/contact_en

FINDING INFORMATION ABOUT THE EU

Online
Information about the European Union in all the official languages of the EU is available on the Europa website at: http://europa.eu

EU Publications
You can download or order free and priced EU publications from EU Bookshop at: http://bookshop.europa.eu. Multiple copies of free publications may be obtained by contacting Europe Direct or your local information centre (see http://europa.eu/contact)

EU law and related documents
For access to legal information from the EU, including all EU law since 1951 in all the official language versions, go to EUR-Lex at: http://eur-lex.europa.eu

Open data from the EU
The EU Open Data Portal (http://data.europa.eu/euodp/en/data) provides access to datasets from the EU. Data can be downloaded and reused for free, both for commercial and non-commercial purposes.