VoteWatch Europe was launched in 2009, using sophisticated statistical methods developed by political scientists (now based at the London School of Economics and Political Science and New York University), to serve as an information source in the context of the European Parliament (EP) elections taking place that year. Since then, it has become the go-to source of information for citizens, journalists, NGOs, interest groups, politicians, civil servants, and anyone else interested in the policy positions, voting records and coalition formation trends of Members of the European Parliament (MEPs).
The website was redesigned in July 2012 to make the available information even easier to access and understand. At the same time a completely new section was added to provide information on the voting records of Member States in the EU Council of Ministers. You can now monitor the votes of MEPs and ministers on the same item of legislation in both EU legislative chambers. This website covers all voting activity in the European Parliament and the Council of Ministers since July 2009 (the start of the seventh 5-year term of the European Parliament). In addition it provides statistics for the European Parliament going back to July 2004, the start of the sixth term.
The sections below describe how the data behind this website is collected and coded as well as how it is analysed and displayed. We welcome any feedback, corrections and comments you may have. Please email us at firstname.lastname@example.org.
Most of the data covering the European Parliament is gathered automatically from the Parliament's own website. In addition the Parliament's secretariat provides VoteWatch with machine-readable files.
This website does not seek to give an overall assessment of MEPs' individual or collective performance. Many of the activities undertaken by MEPs, such as constituency work and participation in external events, are not centrally monitored, quantified and/or assessed. Furthermore, quantitative scores do not in themselves provide a sufficient basis for judging the work of an MEP. This website only provides information about those activities which are formally recorded by the European Parliament on its website. If you are interested in obtaining a more complete picture of the work of MEPs, please visit their individual websites.
The European Parliament meets in plenary session in Strasbourg twelve times every year. It also holds 'mini-plenary' sessions in Brussels (normally four times per year). During each of these plenary part-sessions it votes literally hundreds of times: on amendments, on paragraphs, on parts of paragraphs, on reports and resolutions as a whole. Most of these votes are taken by a 'show of hands', meaning MEPs put up their hand to signal their agreement or disagreement. If the result is unclear, an electronic check may be carried out. This means that MEPs are asked to press a voting button (in favour, against or abstain), and the result is established electronically. This gives an accurate overall result, but the individual votes of MEPs are not registered. In a small proportion of votes, MEPs vote by 'roll call' or 'recorded vote'. This is similar to an electronic check, but in this instance the Parliament's services also record which MEPs voted which way. This information is published later during the day, and VoteWatch processes this data and publishes it on its website. Our data covers all roll-call votes in the EP plenary, on both legislative and non-legislative issues, both final votes and separate votes (on specific paragraphs or amendments).
MEPs and others sometimes complain that as our website is based only on roll-call votes, it does not give a balanced picture of the political views of MEPs. Our response is that there is an easy solution to this, namely that the Parliament should vote by roll call more often - ideally always. That way voters can see exactly how MEPs voted on every single issue that comes before them.
'Corrections' of votes
Because they have to vote on hundreds of items every week, MEPs sometimes forget to press the voting button, or press the wrong voting button. Sometimes the voting machines malfunction. When that happens they can enter a statement into the official record, the minutes, to the effect that they had intended to vote but didn't or couldn't, or that they had intended to vote differently. However, this formal statement does not have the effect of altering their individual vote, and nor does it have an impact on the overall result. Once the President has announced the result of the vote it is final. On this website we show the votes exactly as they were cast. If an MEP has entered a statement into the minutes to the effect that they had intended to vote or vote differently, we note that with an 'info-bubble' next to the relevant vote. But like on the Parliament's own website, we do not change the individual vote or the overall result. A vote, once cast, cannot be 'corrected'.
We have defined the political line of either a European political group or a national party delegation as the position adopted by the plurality of MEPs within that Group or Delegation. Example: if 40 MEPs from the ALDE group voted "FOR", 25 MEPs voted "AGAINST" and other 27 MEPs voted 'ABSTENTION', the political line of ALDE to this vote was "FOR". In cases where an equal number of MEPs vote "FOR" and "AGAINST" in a European political group, the conclusion is that "there is no political line" and the statistics for the MEPs in the European political groups and the national party delegations are shown accordingly.
An MEP is considered 'loyal' to his/her European political group or national party delegation if his/her voting option is identical to the political line of the political group or party delegation, respectively. Example: if P.Moscovici votes "FOR" and the political line of S&D to that vote is also "FOR", then he is loyal to his political group. On the contrary, if he had voted "AGAINST" or "ABSTENTION", he is considered as 'rebel'. The use of the terms 'loyal' and 'rebel' does not imply a value judgment as to whether MEPs were 'right' or 'wrong' to vote they way they did; the terminology is merely designed to help you understand how MEPs behave.
Members who attended less than 100 votes (for instance because they left the Parliament shortly after being elected) are not included in the loyal/rebel scores.
'Cohesion' refers to how united a European political group is in voting situations. This feature can be defined in various ways, but we have here calculated cohesion of the European political groups in two steps: at each vote, we establish an 'Agreement Index' for each European political group according to the Hix-Noury-Roland formula: Ai=(max(Y,N,A)-(0.5((Y+N+A)-max(Y,N,A))))/(Y+N+A), where Y = number of votes "FOR", N = number of votes "AGAINST", and A = number of "ABSTENTIONS". Then, the cohesion rate of a European political group is determined by the arithmetical average of the scores of the Agreement Index. We provide results regarding both the overall cohesion rates and by various policy areas. In the latter case the cohesion rate is being calculated only in relation to the votes cast in that specific policy area.
Unlike the political dynamics at national level, the EU level does not have politics of an explicit 'government-opposition' nature. Coalitions are instead formed on differently from one policy area to another and at times even from one proposal to a next. VoteWatch has identified and publicly displays the composition of the majority for each roll-call vote cast. Based on these votes we calculate a series of trend statistics, as follows:
Vote-matching between European political groups (one-on-one)
This measures shows the absolute vote convergence between two European political groups, which means that it shows the number of votes and the % out of the total number of roll-call votes in which the political lines of these two European political groups coincided. Example: 'on civil liberties, EPP matched ALDE in 60% of the votes'.
Votes on the winning side
Each time a vote is cast, we record the way in which each European political group (its plurality) voted on the one hand and the result of the vote (adopted or rejected) on the other hand. We then compare to see which European political groups (and individual MEPs) were on the winning side, and each of the European political groups (and individual MEPs) on the winning side are considered to be part of the "winning coalition". Example of result: 'A winning coalition was formed by EPP+ALDE+ECR'.
'Loyal' or 'rebellious' behaviour of national party delegations
For each vote, we calculate if the political line of each national party delegation (its plurality) within an European political group matches the political line of the group as a whole. If it does, then that national party delegation is considered 'loyal', otherwise it is considered 'rebel'.
In addition to voting data, VoteWatch also collects statistics regarding the attendance and participation record of MEPs. We publish two percentages for each MEP.
The attendance rate shows on how many plenary meeting days an MEP was present in the Parliament. In order to determine whether an MEP was present on a specific day, we first check the attendance register available online. If they have signed the register they are automatically registered in our database as 'present'. If an MEP is found not to have signed the register (which sometimes happens, for instance because they forgot), we then also check whether she/he took part in at least one roll-call vote that day. If this is the case, then she/he is also automatically registered as 'present'. For each MEP, we calculate the percentage of days she/he was present from the total number of plenary days which took place. For those MEPs who left the European Parliament before the end of the mandate, or for new MEPs who joined the Parliament during the course of the term, we include only their attendance during the relevant period. These are all marked with a special icon. MEPs who have been Members for less than 3 months are not included in any of our attendance calculations.
Information about MEPs who are absent for longer periods of time (for instance because of maternity leave or illness) is not available directly from the European Parliament. We believe that this information should be made available to allow voters to form a fuller and fairer picture of the reasons for an MEP's non-attendance. As long as this is not the case, VoteWatch has taken a policy decision to reflect 'justified' absences in its attendance statistics. If and when an MEP informs us of a 'justified' absence, substantiated with a supporting letter from the EP's College of Quaestors, we adjust the relevant statistics and add an icon to explain that the MEP's attendance record has been adjusted to reflect their absence from the EP during certain plenary days. In addition to cases of long-term absence for personal reasons such as maternity leave or illness, we also do this when an MEP has been absent on official EP business, such as participation in an international conference on behalf of the Parliament. We can only take these absences into account if and when MEPs contact us directly, and we reserve the right to make a final judgment about whether the information supplied justifies a manual correction to our data.
We welcome suggestions from MEPs and members of the public on how the presentation of statistics on this website can be improved further.
The participation rate is the percentage of roll call votes an MEP has taken part in, irrespective of the number of days she/he attended the plenary. While the attendance rate only measures whether or not an MEP was 'in the building' on any given plenary day, the participation rate shows to what extent they attended the voting sessions and the degree to which they took part in all the votes being held.
Amendments to reports
We record the number of reports to which MEPs have submitted amendments, whether they have been adopted or not.
Under each MEP's profile we show how often they have delivered a speech in the EP plenary. Speaking time is generally allocated in proportion to the size of each European political group and each national party delegation. Explanations of vote, which take place after each voting session, are also considered as speeches. Many of these are delivered in writing to save time.
We show how many times an MEP has submitted a written question to the European Commission and/or the Council of Ministers.
Our team has collected data on the voting records of all 27 EU member state governments in the Council from July 2009 to the present. It should be noted that the Council currently releases information only on final votes, at ministerial level, on legislative and budgetary issues. For this reason, the data presented on this website is limited to these decisions.
This website shows all legislation passed by the Council; how each government has voted on each piece of legislation; who voted on behalf of that government; as well as more detailed information about the dossier (Council configurations in which it was debated and voted on, number of readings, working groups involved in preparing the legislation, etc). The data is aggregated to produce statistics on coalition patterns between Member States, to compare, pair-wise, how frequently governments voted against each other, and whether or not individual governments often find themselves in a minority.
The Council voting data is compared to votes on the same piece of legislation in the European Parliament, which allows you to see, for example, to what extent the policy preferences of a government were followed by MEPs from the governing party or coalition.
We publish new Council voting data as soon as it becomes available. It is collected manually from the Council website (minutes, fiches de vote and summaries of legislative acts), PreLex and the EU Who is Who. We also use data from ParlGov for the dates and the party-political composition of EU governments. As the Council does not publish all of its information immediately following the vote, it can take some time for full information on a vote to be available on this website. As long as votes are processed manually, errors may occur. We will correct any errors as soon as we are made aware of them.
The Council meets in ten different configurations, based on policy area. Different voting rules apply depending on the policy area. In some sensitive areas unanimity still applies, but most decisions are taken by qualified majority (QMV). Under this system each Member State has a number of votes based on population size. A qualified majority is reached if:
- a simple majority of Member States supports the proposal (in some cases a two-thirds majority is required) and
- a minimum of 255 votes is cast in favour, out of a total of 345 votes.
In addition, a Member State may ask for confirmation that the votes in favour represent at least 62 % of the total population of the Union.
Despite the complex voting system, the Council does not always vote in the formal sense of raising hands. A proposal can be declared adopted by the Chairperson when she or he knows that there is a sufficient majority. In practice, a formal vote only takes place in cases where adoption of a proposal is guaranteed.
There is one important difference between the unanimity and QMV systems. When the decision rule is unanimity, abstentions do not count as votes against a proposal. This means that decisions can be taken with only some countries actually voting in favour, as long as no country actively opposes it. The opposite is true for QMV, where the high threshold for a proposal to be adopted means that abstentions are in practice equivalent to 'no' votes. Furthermore, if a proposal is accepted, members who wish to oppose, abstain or who have serious concerns about the decision can record their views officially by making formal statements. Formal statements are usually made immediately after a decision has been adopted, and are either included directly in the minutes of the meeting or posted separately on the Council website.