Understanding the Formation and Actions of Eurosceptic Groups in the European Parliament: Pragmatism, Principles and Publicity

This article assesses why Eurosceptic national parties form groups in the European Parliament and examines in what ways two of these groups – the European Conservatives and Reformists and Europe of Freedom and Democracy – operate in the European Parliament. It draws on interviews with politicians and group officials, roll-call votes and expert judgement data. We look at the group formation process with a focus on the British Conservatives and UK Independence Party and find that the European Conservatives and Reformists group was created with a mixture of policy-seeking and party-management aims. The UK Independence Party's interest in the Europe of Freedom and Democracy group is largely on the basis of the group's provision of distinct practical advantages, such as resources for political campaigns. We provide evidence that hard Eurosceptic and regionalist niche parties in the European Parliament struggle to agree with each other in roll-call votes on a range of subjects. Finally, we show that the hard and soft Eurosceptic parties studied here go about policy-seeking in different ways in the European Parliament in line with their differing principles on the integration process.


EUROSCEPTICS HAVE LONG BEEN PRESENT IN THE EUROPEAN
Parliament (EP) on both the left and right. But in recent years they have become more clearly organized within the European Parliament's group structure, particularly in the form of the soft Eurosceptic European Conservatives and Reformists (ECR) group formed after the 2009 European Parliamentary elections. Soft Eurosceptic parties do not oppose the European Union (EU) in principle but are against further European integration (Taggart and Szczerbiak 2008: 248). Hard Eurosceptics, favouring withdrawal from the EU, are scattered across groups and the non-attached members in a less coherent fashion but their profile has been raised particularly by the leader of the UK Independence Party (UKIP) and co-president of the Europe of Freedom and Democracy (EFD) group, Nigel Farage. Hard Eurosceptics face tricky decisions about how far to engage in an institution that they oppose but which can also bring them rewards in terms of information on EU policy developments, publicity and financial benefits useful for promoting their cause. While we know much about what happens inside the European Parliament, there has been limited research specifically on how Eurosceptics go about operating within it (for welcome exceptions to this see Benedetto 2008;Brack 2012Brack , 2013Startin 2010). This is an interesting question in the light of the conflicting pressures Eurosceptics face and given the range of Euroscepticism represented in the European Parliament -from those favouring reform to those wanting to abolish the institution itself.
Scholars have assessed reasons for forming party groups in the European Parliament (for example , Bressanelli 2012;Hix et al. 2007;McElroy and Benoit 2010) but not all of the suggested motives apply to national parties that oppose either many aspects of European integration or their country's membership of the EU. For these national parties, group formation may be motivated by a desire for publicity and information about EU policies, or a concern with national party management. This article contributes to our understanding of group formation and group action in the European Parliament by assessing, with a focus on the UK Independence Party and the British Conservatives, why Eurosceptic national parties form groups and by examining in what ways and how effectively two of these groups -the European Conservatives and Reformists and Europe of Freedom and Democracy -operate in the European Parliament.
More specifically, we assess non-policy related reasons for forming a party group in the European Parliament. We go on to ask how Eurosceptic groups use the European Parliament and to what degree they act as a coherent force. We use expert judgement data on party positions (Bakker et al. 2012), roll-call votes from the seventh term of the European Parliament and data from the VoteWatch Europe website (www.votewatch.eu). Our discussion of group formation and part of our analysis of activities in the European Parliament draw on 40 non-attributable interviews conducted during the period 2009-12 by the authors with European Conservatives and Reformists (five) and note that there are likely to be some non-policy factors at play in some national parties' decisions (McElroy and Benoit 2010: 396-7), something they do not explore. In a further analysis based on more recent expert surveys of European Parliamentary party group positions, they show that the theory holds up well for groups in the 2009-14 term (McElroy andBenoit 2011). Nevertheless, we hypothesize that this approach will much better explain the European Conservatives and Reformists than the Europe of Freedom and Democracy group on the basis that, as discussed later, the former was set up to bring together centre-right parties that agree on the need to reform the EU in a way that protects the sovereignty of member states. By contrast, of the two national parties that make up the bulk of the Europe of Freedom and Democracy group, one is hard Eurosceptic (the UK Independence Party) and the other soft Eurosceptic (the Italian Northern League; Lega Nord -LN). Given this difference, the policy congruence approach is unlikely to apply to this group, at least not on the assumption that the European integration dimension is the most salient to parties in the group, something we investigate below. In addition to this, achieving policy congruence might be made more difficult for the European Conservatives and Reformists because the larger groups had gained some national parties that the European Conservatives and Reformists might have favoured. The Finnish Centre Party MEP Hannu Takkula, for instance, joined the European Conservatives and Reformists at its beginning but left it two days later under pressure from his party back home to remain within the Alliance of Liberals and Democrats for Europe (ALDE) (EurActiv 2009). For the Europe of Freedom and Democracy group the problems are worse, with its founders having to choose partly from among those rejected by the European Conservatives and Reformists while avoiding parties with whom, for domestic reasons, they do not want to associate, such as the British National Party. These problems should be viewed in the light of the European Parliament's Rules of Procedure (Rule 30), which require party groups to consist of at least 25 MEPs drawn from one-quarter of the member states, which amounts to seven in the 2009-14 term.
Moving on from policy congruence theory, Bressanelli (2012) theorized that national parties may be attracted to party groups for pragmatic reasons. These include the resources allocated to groups and the opportunity for gaining influential committee positions and 235 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT rapporteurships in the European Parliament. These incentives are particularly significant in the case of the larger groups because of the European Parliament's distribution of positions and resources in proportion to group size. Party groups themselves are also motivated to attract large national parties in order to increase their size and gain the financial, staffing and office benefits that follow from growth. Bressanelli's results show that the bigger the national party, the more likely it is to join a larger group rather than a smaller one. Interestingly, Bressanelli (2012: 746) found rather more mixed results for the European Conservatives and Reformists and Europe of Freedom and Democracy groups, indicating that other factors are at work in the composition of these groups.
What might these factors be? Part of the answer lies in theories of party behaviour. Strøm (1990) sets out three types of this: office-, policy-and vote-seeking. Applied to national parties in the European Parliament, Bressanelli's (2012) argument about pragmatic motivations for joining larger groups fits with office-seeking behaviour. The policy congruence approach fits with policy-seeking priorities. However, vote-seeking behaviour is only likely to affect national parties' choice of European Parliament party group if this choice is salient among their voters (Maurer et al. 2008: 249). If the issue is salient among a party's parliamentarians and activists, then group choice may be made on the basis of party management (Lynch and Whitaker 2008). In either case, this might lead to a choice that does not maximize possible policy influence.
Once Eurosceptic groups have been formed, how can we expect them to behave in the European Parliament? Some insight into the way hard Eurosceptic groups and national parties within them might act comes from work on niche parties. While there is debate as to their definition, we start from Wagner's definition (2012: 847) of parties that 'compete primarily on a small number of non-economic issues'. This means they include regionalist, radical right, green and anti-EU parties. Jensen and Spoon's (2010) research on this suggests that niche parties in the European Parliament will act differently to their mainstream counterparts. These parties' focus on national or regional autonomy and identity means that such concerns may be particularly important for them when deciding how to vote in the European Parliament. Such parties will be less likely to compromise on policy positions than their mainstream counterparts on the basis that they risk losing their core support if they moderate policy 236 GOVERNMENT AND OPPOSITION (Adams et al. 2006), and because the organization of niche parties tends to favour influence for activists (Dumont and Bäck 2006;Ezrow et al. 2011). Jensen and Spoon (2010: 188) also find that views on European integration differ between sub-types of niche party. These various points lead us to hypothesize that there will be divisions within the Europe of Freedom and Democracy group as its two largest national parties include a regionalist party (the Italian Northern League) and an anti-EU one (the UK Independence Party), as well as the Movement for France (Mouvement Pour la France -MPF), also classified as an anti-EU niche party by Jensen and Spoon (2010).
Mü ller and Strøm's (1999: 17) approach suggests that when party organization favours activist influence -as it does in many niche parties -we are more likely to see policy-seeking rather than officeor vote-seeking behaviour. How this affects a party's approach to the European Parliament may depend on its policy goals. For hard Eurosceptic parties, which seek their country's withdrawal from the EU, we hypothesize that policy-seeking behaviour is less likely to be focused on legislative work in the European Parliament as this is not a venue in which decisions about leaving or remaining in the EU are made. For these parties, in contrast to expectations about mainstream party groups in the literature on group formation, we expect group membership to be used for publicity-seeking behaviour, primarily speaking in (high-profile) plenary sessions and channelling the financial resources that flow from being in a group into political campaigns. On this point, Startin (2010) shows how, when establishing the short-lived Independence, Tradition and Sovereignty group in 2007, radical right MEPs were motivated by the significant financial and practical advantages they felt would follow from this. Alternatively, for parties that favour reform of the EU -in our case, the European Conservatives and Reformists group -we expect policy-seeking behaviour to include attempts to influence the EU's legislative process (Benedetto 2008), such as trying to build coalitions with other party groups with the aim of affecting the outcome of votes. As the two largest parties in the Europe of Freedom and Democracy group differ in terms of their Euroscepticism, we hypothesize that there will be differences in their behaviour, with the Italian Northern League involving itself much more in the details of the European Parliament's legislative activities.

EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT
We test these expectations by looking at several aspects of Eurosceptic parties' approaches to operating in the European Parliament: forming groups, voting and engagement with legislative processes. In the next section we look briefly at the first of these, mainly in relation to the British Conservatives in the European Conservatives and Reformists group and the UK Independence Party in the Europe of Freedom and Democracy group. These constitute the largest national party in each of the two groups and allow us to compare the motives for group formation in soft and hard Eurosceptic parties. Concentrating on these two parties in examining group formation also allows us to look in detail at the motivations and consequences of putting together these two groups for the largest member of each. Our expectations concerning the UK Independence Party's motivation for group formation could equally apply to its role in the Europe of Freedom and Democracy's predecessor, Independence/Democracy (I/D), although we lack the space and data to examine the formation and actions of that group here. While the UK Independence Party was also part of Independence/Democracy's forerunner in the fifth term, the Europe of Democracies and Diversities (EDD) group, it only had three MEPs at the time, meaning that it was in a rather weaker position with regard to group formation.

FORMING THE EUROPEAN CONSERVATIVES AND REFORMISTS AND EUROPE OF FREEDOM AND DEMOCRACY GROUPS
What factors explain the formation of the European Conservatives and Reformists and Europe of Freedom and Democracy groups, and are these factors specific to groups that are Eurosceptic in outlook? This section answers these questions, drawing heavily on interviews conducted by the authors.

The British Conservatives and the European Conservatives and Reformists
Group: Principles over Influence?
The European Conservatives and Reformists group was born in July 2009 after a long process of debate in the British Conservative Party about its European Parliamentary group membership. While we do not have space to describe this in detail here, the party's decision to leave the European People's Party-European Democrats (EPP-ED) group 238 GOVERNMENT AND OPPOSITION related to its shift to a soft Eurosceptic position during the 1990s (Webb 2000: 185). This led some Conservative parliamentarians and activists to argue that they had moved away from the European People's Party-European Democrats' more pro-integration stance. The process was problematic because Conservative MEPs were divided over the issue (Lynch and Whitaker 2008) and eventual group members the Czech Civic Democrats (ODS) and the Polish Law and Justice Party (PiS) were not prepared to move in 2006 when the Conservative Party was making its initial attempts to establish a new group.
How does the formation of this group fit with the party goals set out earlier? The evidence is mixed as to whether the primary motivation for the Conservatives and Czech Civic Democrats to leave the European People's Party-European Democrats and join the European Conservatives and Reformists was office-seeking behaviour. The Conservatives lost the opportunity to lead the most legislatively active committees, such as Environment -for which they held the chair in the fifth European Parliamentary term (1999)(2000)(2001)(2002)(2003)(2004) when part of the European People's Party-European Democrats. The Czech Civic Democrats had also chaired this committee in the sixth term , when they were part of the European People's Party-European Democrats group. However, following the formation of the European Conservatives and Reformists, the Conservatives gained the chair of the Internal Market committee, a subject of particular significance for the party. A further positive point made by many Conservative interviewees was that, compared with the European People's Party-European Democrats, the European Conservatives and Reformists offered a greater chance of securing positions at the group level, specifically as group coordinators on committees and shadow rapporteurs, the numbers of which do not vary according to group size. In addition, given their electoral performance in 2009, the British Conservative Party would have found itself as only the fifth largest national party contingent in the European People's Party-European Democrats, reducing its chances of winning influential posts.
Second, there is some evidence that the formation of the group represents policy-seeking behaviour. Joining the European Conservatives and Reformists instead of a larger group would seem inconsistent with an attempt to improve a party's ability to influence the European Parliament's legislative outputs (Maurer et al. 2008).

EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT
However, if policy influence requires groups to be cohesive, national parties will also need to consider the policy preferences of others. The founding principles of the European Conservatives and Reformists, expressed in the Prague Declaration, suggest that agreement on an approach to European integration provides a coherent basis for the group. As one Conservative MEP put it, 'it has given the European Parliament something of an official opposition . . . this is the first time that a group of governing or potentially governing parties have got together to offer an alternative view of the EU.' 1 This explanation is consistent with McElroy and Benoit's (2010) policy congruence theory which we examine in more detail later.
Third, for the Conservative Party, there is little evidence that forming the European Conservatives and Reformists represents voteseeking behaviour because the EU, and by implication the party's choice of group in the European Parliament, was not a salient issue among Conservative voters at the general election before the party left the European People's Party-European Democrats. According to the British Election Study's post-campaign internet survey, 2 only 4.4 per cent of Conservative voters saw Britain's relations with the EU or the euro as the 'most important issue' in the 2005 general election campaign, and this figure was only 0.7 per cent in the equivalent survey in 2010. 3 However, the question of the party's group membership in the European Parliament was an issue in the 2005 leadership election, with David Cameron seeking votes by promising to take his party out of the European People's Party-European Democrats (Lynch and Whitaker 2008).
Fourth, in addition to policy aims, the decision to form a new group for the Conservatives seems to have been driven partly by the need to manage divisions within the party over European integration. Taking the Conservatives out of the European People's Party-European Democrats group was something Cameron could do without being in government, assuaging Conservative Eurosceptics and allowing the party, in Cameron's own words, to be 'good neighbours rather than unhappy tenants' (The Independent 2009) of the European People's Party-European Democrats. Nevertheless, creating a group generated new party management problems because not all MEPs were happy with the decision, as is clear from a Conservative interviewee's comment that 'we [now] have independence but no real influence'. 4 In addition, the hostile reaction in some of the UK media to some of the Conservatives' partners in the 240 GOVERNMENT AND OPPOSITION European Conservatives and Reformists (Bale et al. 2010) demonstrated the problems of meeting the criteria for group formation in the European Parliament while managing likely domestic media responses.
The UK Independence Party's Role in Forming the Europe of Freedom and Democracy Group: Publicity and Pragmatism The Europe of Freedom and Democracy group has its origins in the Independence/Democracy group of the sixth European Parliament term and the Europe of Democracies and Diversities group during the fifth, both of which were primarily Eurosceptic groups but which included parties not only from the right but also from the left of the spectrum, principally the Danish June Movement. The largest national party within Europe of Freedom and Democracy is the UK Independence Party, a hard Eurosceptic party advocating the UK's withdrawal from the EU. If the UK Independence Party sees the European Parliament as an institution of which the UK should not be a part, what is its rationale for being part of and providing co-leadership of a group? In interviews with Europe of Freedom and Democracy group staff and UK Independence Party MEPs, a number of motivations were expressed. Among the most prominent was speaking time in plenary sessions, particularly for the group's co-president, Nigel Farage. As one UK Independence Party MEP put it, 'If you think of all [Nigel Farage's] memorable performances in the European Parliament . . . it's all been done with speaking time in the plenary.' 5 On the basis of the European Parliament's Rules of Procedure (Rule 149), groups are at some advantage regarding speaking time compared to non-attached members, and this advantage increases with group size. A UK Independence Party interviewee pointed to the publicity the party gains beyond the UK from Farage's European Parliament performances. 6 In this respect, the existence of the group helps the UK Independence Party to promote the Eurosceptic cause and to develop an opposition within the European Parliament.
A second reason for being part of a group given by interviewees was the financial benefits and staffing allocations provided for research and administrative purposes. The financial resources are seen as important, partly, as one interviewee put it, because they would otherwise 'go to pro-federalist organizations', 7 but also because they allow the Europe of Freedom and Democracy group 241 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT to fund political campaigns such as those against the Lisbon Treaty in the second Irish referendum on this 8 and in Ireland's referendum on the Fiscal Compact. The provision of research staff means that MEPs have experts available to them who can advise on how the European Parliament works and on the details of policy proposals. For the UK Independence Party, this provides fuel for campaigns at the national level. Some interviewees referred to other office benefits, principally the allocation of places on the European Parliament's committees, where much of its legislative work is carried out. For instance, some UK Independence Party MEPs pointed to the political capital they can gain from being able to question senior figures during committee hearings. This feeds into their aims of achieving publicity and acquiring information on EU legislation.
Despite these benefits, several interviewees gave a sense that their group was formed reluctantly. A UK Independence Party MEP explained that because of the procedural privileges given to groups in the European Parliament, 'we're forced into [forming a group].
The key thing is, it's a group not a party . . . it is practically driven, not principled.' 9 Indeed, events following the formation of the Europe of Freedom and Democracy group revealed differences within the UK Independence Party delegation over the group and its composition. For instance, two UK Independence Party MEPs (Trevor Colman and Mike Nattrass) left the Europe of Freedom and Democracy group due to disagreements with the group but were permitted to remain in the UK Independence Party. 10 To a degree, this reflected disagreements within the party about how far it should involve itself in the European Parliament's structures.
This discussion suggests significant differences in the motivations of the UK Independence Party and the British Conservative Party in forming groups in the European Parliament, with the UK Independence Party reluctantly forming a group in order to gain specific practical advantages, while policy-seeking behaviour and party management explain the Conservatives' decision. We can understand further the basis of these two groups beyond a purely British focus by looking next at intragroup ideological diversity and asking how far the European Conservatives and Reformists is ideologically cohesive, as we would expect if its formation represents policy-seeking behaviour, and comparing this with the range of views within the Europe of Freedom and Democracy group. This section looks first at the ideological diversity of European Parliament groups and then examines their cohesion in roll-call votes. On the basis of the preceding discussion, we expect the policy congruence approach (McElroy and Benoit 2010) to apply to the European Conservatives and Reformists group, such that it is cohesive in policy dimensions that are of salience to its national parties. For the Europe of Freedom and Democracy group we expect lower ideological cohesion on the basis that the group includes different types of niche parties and was formed for more practical reasons.
We examine this using national parties' positions on the left-right and European integration scales from the 2010 Chapel Hill expert survey (Bakker et al. 2012). The 2010 wave of this survey is the closest to the time at which the European Conservatives and Reformists and Europe of Freedom and Democracy groups were formed and includes all member states except Cyprus, Luxembourg and Malta. The vast majority of parties represented within the seven party groups formed after the 2009 European Parliament elections are included. In all but two cases, parties missing from the data were made up of only one or two MEPs. 11 We adopt the same approach to calculating ideological diversity, albeit with different data, as Hix et al. (2005: 224), in which parties are weighted by size. We first calculated the weighted mean position of party group M i for the left-right dimension as follows: where P j is the left-right position of national party j from the Chapel Hill 2010 survey, S ji is the proportion of party group i made up by national party j. The same calculation was made, using the EU position of each national party. 12 We then calculated the ideological diversity of each group (Hix et al. 2005: 224) for left-right and EU dimensions as follows:  Table 1, where higher scores indicate more diversity. Interestingly, the Europe of Freedom and Democracy group compares well with the European People's Party (EPP), the Alliance of Liberals and Democrats for Europe and Greens/European Free Alliance (G/EFA) groups on the left-right scale of ideological diversity. This is consistent with McElroy and Benoit's results (2011: 162), which show a narrow range of views on this dimension using different data. This may be partly a result of the comparatively small number of national parties within the Europe of Freedom and Democracy group but also suggests a degree of policy congruence on this dimension, particularly in contrast with the group's predecessor, Independence/Democracy, which included some MEPs from the left. Nevertheless, Table 1 also shows that the Europe of Freedom and Democracy is the most ideologically diverse group on the EU integration scale. This does not fit with the policy congruence model, given the salience of this issue to the group. These findings are consistent with Europe of Freedom and Democracy interviewees' comments, of which the following was typical: 'We only have loose

244
GOVERNMENT AND OPPOSITION things in common and we don't share a manifesto.' 13 While the Italian Northern League is a Eurosceptic party, its position contrasts with that of the UK Independence Party. It shifted from a pro-integration view during the 1990s to a far more critical stance after 1998, particularly with regard to economic and monetary union, the common agricultural policy and EU state aid policies (Chari et al. 2004). Nonetheless, it voted in favour of the Nice and Lisbon treaties (Bartlett et al. 2012), and its scores in the Chapel Hill data differ substantially from the UK Independence Party's on several areas of EU activity, including the EU cohesion policy, where it is placed at 4.3, compared with 1.2 for the UK Independence Party on a seven-point scale with higher values representing greater support for this policy. The diversity within the European Conservatives and Reformists group is considerably more limited on both dimensions, and the group compares favourably with others on both scales. This is due to the similarity of positions among the three largest contingents in the group, the British Conservative Party, the Polish Law and Justice Party and the Czech Civic Democrats. Overall, the results suggest that policy congruence was a major factor that brought together the parties that make up the European Conservatives and Reformists. In contrast, policy similarities on European integration were not the primary motivation for forming the Europe of Freedom and Democracy group although it is comparatively cohesive in left-right terms.
Understanding the ideological diversity within groups can only tell us so much about how effective groups are at achieving their aims in the European Parliament. The next step is to review how far these two groups act as a cohesive force by examining their behaviour in roll-call votes. As Hix et al. (2007: 101-2) have shown, ideological fractionalization did not affect group cohesion in rollcall votes for the period 1979-2004, which suggests that cohesion is not based entirely on preferences. They theorize that larger groups will be more cohesive because they are more likely to influence policy outcomes and therefore have more at stake in European Parliament votes (Hix et al. 2007: 101). Our earlier discussion of niche parties suggested that we should expect a lower propensity for them to compromise with others to achieve group positions. We can assess this in the case of national parties in the Europe of Freedom and Democracy group. We use data from the VoteWatch Europe website (www.votewatch.eu). Then we analyse coalition patterns 245 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT among national parties within the two groups based on all roll-call votes from July 2009 to December 2010.
VoteWatch Europe data on roll-call votes from the beginning of the seventh term (July 2009) up to March 2013 show that Europe of Freedom and Democracy is the least cohesive party group, in line with our expectations. The Index of Agreement (Hix et al. 2007) by which VoteWatch measures cohesion can range from 0, when a group is evenly split among the options of 'yes', 'no' and 'abstain', to 100, when the entire party group votes the same way. The Europe of Freedom and Democracy group stands out with a score of 49.7 compared with scores in the 90 s for all other groups except the Alliance of Liberals and Democrats for Europe (88.9), the European United Left/Nordic Green Left (79.2) and the European Conservatives and Reformists at 86.4. 14 The high score for the Greens/European Free Alliance group (94.6), which includes two types of niche party (green and regionalist), is out of line with the suggestion that niche parties in general will struggle to agree with each other in a transnational group setting. But our expectation of low cohesion in the Europe of Freedom and Democracy, on the basis that its two largest components are a hard Eurosceptic and a regionalist party, is borne out. This is not surprising in the light of comments made by Europe of Freedom and Democracy party group staff that there is no formal whip for the whole group 15 and that 'there is an acknowledgement . . . that every party is free to espouse its own views as long as these are not racist or anti-democratic.' 16 Furthermore, the greens in the European Parliament have a much longer history as a group and as an extraparliamentary organization through the European Federation of Green Parties (Hanley 2008: 169). The European Conservatives and Reformists group's cohesion score is much closer than Europe of Freedom and Democracy's is to the largest groups, in line with our expectations.
Next we turn to what underpins the cohesion, or lack of it, in these two groups.   (Table 2). This makes sense in terms of the softer Euroscepticism of these parties compared with the UK Independence Party. 17 Table 3 shows equivalent figures for the European Conservatives and Reformists group, which are much higher than those for the Europe of Freedom and Democracy group, as would be expected on the basis of the group's index of agreement scores. Of the group's three main parties, the British Conservative Party and the Czech Civic Democrats vote with each other in about nine out of ten votes, slightly more than either of them does with the Polish Law and Justice Party.
These figures cover votes on a range of topics and therefore might mask differences within the groups over particular policy areas. Given the presence of niche parties in the Europe of Freedom and Democracy group, we expect differences on issues that might have national or regional repercussions. The differing types of Euroscepticism between the UK Independence Party and the Italian Northern League also lead us to expect disagreement between the two. We consider roll-call votes from the same data set under three types of policy, allowing us to test for economic left-right, national or regional and integration-based differences: the regulation of the single market in the form of environment and public health policy (156 votes), distributive policies measured via votes on the budget (167 votes) and agricultural policy (67 votes), and integration issues, assessed through votes on constitutional and inter-institutional matters (40 votes).
In the votes on environmental policy, majorities of national parties voted the same way in more than nine out of ten votes for most combinations in the European Conservatives and Reformists group. In the Europe of Freedom and Democracy group, majorities in the two largest parties, the UK Independence Party and the Italian Northern League, voted the same way in only around onefifth of votes (20.8 per cent). In the 167 roll-call votes on budgetary matters, the UK Independence Party and the Northern League managed a slightly higher degree of similarity than in all votes, with a score of 41.7 per cent. In the European Conservatives and 248 GOVERNMENT AND OPPOSITION So, while Europe of Freedom and Democracy is commonly understood as a Eurosceptic group, on issues concerning EU integration and other areas of policy there are considerable differences in voting behaviour within it. This is consistent with our expectations based on the presence of hard Eurosceptic and regionalist niche parties within the group (the UK Independence Party, the Italian Northern League, the Movement for France) and the differing types of Euroscepticism espoused by its two largest national parties. The European Conservatives and Reformists group, by contrast, has a cohesive core of national parties but experiences slightly higher levels of disagreement on distributive policies in the sample of roll-call votes used here. Given this contrast and our expectations about how hard and soft Eurosceptic parties might make use of the European Parliament in different ways, the next section looks at how far and in what ways these groups, and some of the parties within them, take part in the European Parliament's business.

GOVERNMENT AND OPPOSITION (DIS)ENGAGING WITH THE EUROPEAN PARLIAMENT'S LEGISLATIVE AND NON-LEGISLATIVE BUSINESS? EUROPEAN CONSERVATIVES AND REFORMISTS AND EUROPE OF FREEDOM AND DEMOCRACY APPROACHES
Earlier we set out expectations that hard Eurosceptics will be largely reactive and/or engaged only for the purpose of gathering information to feed into campaigns. Soft Eurosceptics can be expected to take part in the policy-making process in an attempt to move policy closer to their ideal point. While other research has considered the actions of individual Eurosceptic MEPs (for example, Brack 2012Brack , 2013Navarro 2012), here we focus on groups' and national parties' activities in the first half of the seventh term. We have space only to sketch some examples of these groups' approaches to the European Parliament's activities, but they serve to indicate the variation between the two groups.

The European Conservatives and Reformists Group: Attempting Legislative Influence
On the basis of the Prague Declaration -the European Conservatives and Reformists group's founding statement -we would expect this group to be engaged with policy-making in the EU in an attempt to make changes in line with its preferences for free enterprise, minimal regulation and the sovereignty of member states. Here, drawing on our interviews, we focus on the British Conservative Party in the group, although the approaches described are not exclusive to it. We should expect policy-seeking Conservative MEPs to attempt to gain influence in areas of concern to the party and its voters. We expect this to be particularly important for areas where the Conservatives differ from many in the European People's Party, such as on European integration and the regulation of financial services and protection of the City of London. Given that the allocation of positions in the European Parliament is based on proportional representation of party groups, the European Conservatives and Reformists' place as the fifth largest group with around 7 per cent of the seats means that high-profile legislative reports are unlikely to come its way. Conservatives used to working in the much larger European People's Party-European Democrats have therefore had to be smarter about their approach to the European Parliament's procedures. Interviewees pointed to two methods for attempting 251 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT influence: first, making strategic use of own initiative reports, and second, attempts at building alliances with others where there is common ground.
Many of our interviewees referred to the use of own initiative reports (OIRs), particularly those defined in the European Parliament's Rules of Procedure (Annex XVIII) as strategic, drawn up on the basis of items included in the Commission's annual legislative programme and therefore expected to be the subject of legislation in the near future, or implementation reports concerning the implementation and enforcement of EU laws. Data from the European Parliament's Legislative Observatory for the first three and a half years of the seventh term show that the European Conservatives and Reformists received a roughly proportional share of own initiative reports. However, European Conservatives and Reformists group staff suggested that the group is strategic about the reports it tries to win, aiming to gain them in areas where significant legislation is imminent. As one Conservative MEP put it, 'the own initiative route is a very good tool to set the legislative agenda down a path you would like it to go . . . it is easier to have a Commission document that starts in a good place so you can tweak it.' 18 Reports in the European Parliament are distributed on the basis of a bidding process in which groups have points that they can spend and which are allocated to groups in proportion to their size. Non-legislative reports are easier to obtain for a smaller group as they cost fewer points than the legislative reports that follow. Interviewees explained that the Commission will sound out rapporteurs and shadow rapporteurs on own initiative reports covering topics about which they plan to legislate, giving an indication of how these reports are used by the Commission to measure opinion in the European Parliament.
Two particular cases help to demonstrate this approach. First, an own initiative report on trading in financial instruments (2010/ 2075(INI)) 19 was drawn up by British Conservative MEP Kay Swinburne ahead of the Commission's plans to revise its Markets in Financial Instruments Directive (MiFID). Interviews with European Conservatives and Reformists group staff suggested that although compromises and changes were made, the own initiative report set the framework for the ensuing draft legislation. 20 In a media interview in February 2011, Swinburne described having seen how own initiative reports could be used to prepare the way for legislation that follows.

GOVERNMENT AND OPPOSITION
She explained that the legislation revising the Markets in Financial Instruments Directive would be highly significant for the City of London and the UK economy and this was the reason for making sure the group won this own initiative report, which she hoped would 'shape the initial direction in which [the legislation] was going' (O'Hara 2011). The Commission's first draft of the new legislation proposed, among other things, tougher rules on highfrequency trading (HFT), representing a development of calls for an assessment of the pros and cons of high-frequency trading in Swinburne's own initiative report (Spicer 2010). At the time of writing it is at first reading stage.
Second, an own initiative report on the implementation of the Professional Qualifications Directive (2011/2024(INI)) was drawn up by British Conservative MEP Emma McClarkin ahead of Commission plans for legislation on this subject. Interviewees argued that the presence of an early warning mechanism on individuals' malpractice and a requirement for language testing in the Commission's subsequent legislative proposal -both of which were retained by the European Parliament at first reading in committee 21 (Stamatoukou 2013) -are evidence that key points favoured by the European Conservatives and Reformists group rapporteur were taken on board.
In the second of these two cases, however, interviewees were careful to point out that there was broad support in the European Parliament for much of the European Conservatives and Reformists' position, making it easier to ensure this was carried forward. In addition, due to a lack of support from the two largest party groups, the European Conservatives and Reformists did not get its way in the own initiative report on the need for a detailed impact assessment of the proposed 'professional card' -designed to ease professionals' movement around the EU -prior to its introduction.
These caveats indicate the importance of building coalitions in the European Parliament in order to influence outcomes. An example of coalition building cited by interviewees concerns the EU's most recent proposals on maternity leave. In March 2010, the European Parliament's Women's Rights Committee voted on the Commission's proposals to extend the length of maternity leave on full pay to 18 weeks. The committee supported raising this to 20 weeks but without the support of the European Conservatives and Reformists. This group demanded an impact assessment be carried out, to which the Commission agreed (Petitjean 2010 Many interviewees also mentioned the 'single seat' issue as one on which the European Conservatives and Reformists had been proactive in seeking a coalition of support. This concerns the aim of ending the European Parliament's trips to Strasbourg for plenary sessions 12 times each year. Among several votes on this issue, in an own initiative report on the Multiannual Financial Framework (2010/ 2211(INI)) for the post-2013 period, MEPs held a roll-call vote on the European People's Party's proposal to remove a paragraph that referred to the 'significant savings that could be made if the European Parliament were to have a single seat', with the European Conservatives and Reformists on the winning side (353 votes against removing the paragraph, 282 in favour and 38 abstentions). Furthermore, against the wishes of the European People's Party, 23 sufficiently widespread support was gained in the Conference of Presidents to permit an own initiative report on the European Parliament seat issue (2012/ 2308(INI)) with the rapporteurship being shared between British Conservative MEP Ashley Fox and German Greens/European Free Alliance member, Gerald Häfner.
These cases demonstrate approaches used by British Conservatives in their attempt at legislative influence in areas of significance for the party, such as the cost of the European Parliament and the regulation of the single market insofar as it affects employers' costs and the City of London. But European Conservatives and Reformists group staff noted that influence over legislation affecting the latter was hard enough for the Conservative Party when it was part of the European People's Party-European Democrats due to differences with others in the group. In the European Conservatives and Reformists group, the combination of what some described as a more difficult environment for British Conservatives since David 254 GOVERNMENT AND OPPOSITION Cameron's veto of a draft fiscal compact in 2011 and being the fifth largest group means that making a difference in the European Parliament is not easy. VoteWatch Europe data show that the group is on the losing side in a majority of roll calls in areas related to European integration, namely, constitutional affairs, the budget, budgetary control and the European Parliament's rules. The European Conservatives and Reformists group is also on the losing side most of the time in roll calls on issues of culture, economic and monetary affairs and gender equality. This mixture of indirect influence in some areas and the experience of being on the losing side in others was summed up by one Conservative MEP explaining that the group is 'not an unmitigated disaster but not a runaway success either'. 24

The Europe of Freedom and Democracy Group: Limited Engagement
It's like walking on a tightrope. I won't be voting for any EU legislation . . . but it's not credible to go there and just object to everything. 25 We know from previous work on MEPs' roles that hard Eurosceptics take a range of approaches to the European Parliament -from absence to pragmatic involvement (Brack 2012(Brack , 2013Navarro 2012). Such MEPs face a tension between gaining finance and publicity via the European Parliament and avoiding the impression of accepting the legitimacy of the institution. As indicated in the discussion of the European Conservatives and Reformists above, soft Eurosceptics, in contrast, are more likely to attempt to achieve influence in the legislative process. Both types of party are represented in the Europe of Freedom and Democracy group such that we expect to see differences in approach to the European Parliament's activities within the group.
The European Parliament's Legislative Observatory shows that the Europe of Freedom and Democracy group was under-represented in the distribution of reports in the first three and a half years of the seventh term, gaining around 2 per cent of these. If we look at the 17 reports taken on by Europe of Freedom and Democracy MEPs in the responsible committee in this period, differences between national parties are clear: 12 of these were written by MEPs from just two parties, the Italian Northern League (seven) and the Greek Popular Orthodox Rally (five). The UK Independence Party took on one report, in the hands of Marta Andreasen on the Budgetary Control Committee. Andreasen defected to the Conservatives and European 255 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT Conservatives and Reformists group in February 2013. Reports taken on by the Europe of Freedom and Democracy group are therefore distributed disproportionately within the group, with little role for the UK Independence Party in this area of activity. Europe of Freedom and Democracy officials indicated that the group takes a pragmatic approach to this: 'We try to influence legislation and to diminish its harmful effects or to block it.' 26 Referring to the UK Independence Party, another explained with regard to amendments, 'If something is in line with our outlook, [the MEP] should vote yes, such as with amendments that delete text and make policy more free market. It is rare in committee that [we] produce an entirely negative voting list.' 27 Officials and MEPs indicated that the UK Independence Party also proposes amendments in committee where it feels this will further its aims. Interviewees suggested that the group does not normally engage in coalition building but indicated that it may respond by supporting actions taken by others, including Eurosceptics on the left and libertarian parties such as the Swedish Pirate Party, when they are in agreement.
If we look in more detail at other activities in the European Parliament, based on data adapted from VoteWatch Europe, 28 differences within the group are clear. For instance, the mean attendance at plenary sessions in the seventh term up to March 2013 was 76 per cent for UK Independence Party MEPs and 92 per cent for those in the Northern League. We see similar variations on other measures. In the same time period, UK Independence Party MEPs asked an average of 101 questions each, while the figure for the Northern League was 359. A similar gap is evident in the numbers of reports amended, which for the average UK Independence Party MEP is four compared with 76 for those in the Northern League. In addition, the mean numbers of motions for resolution signed by MEPs are, respectively, 0.7 and 33.0. While we would expect some differences in these measures between national parties in any group, the variations among European Conservatives and Reformists parties are much smaller, with mean attendance levels hardly varying from one national contingent to another and an overall mean attendance of 87 per cent. Indeed, UK Conservative MEPs sometimes criticize their UK Independence Party counterparts for non-attendance at votes. Intragroup differences in the Europe of Freedom and Democracy group suggest that its national parties differ in their views about how to operate in the European Parliament and what the 256 GOVERNMENT AND OPPOSITION group is for. The Northern League and other non-UK Independence Party MEPs in the group appear to be engaging to some degree with the options open to MEPs to attempt to influence the European Parliament's outputs. UK Independence Party MEPs involve themselves much less in these activities, a difference acknowledged by Europe of Freedom and Democracy group staff in interviews. 29 These findings are consistent with our expectations on the basis of the different types of Euroscepticism associated with these parties and the likely effects of this on their policy-seeking behaviour. CONCLUSION Using interviews, roll-call voting data, expert judgements, information on MEPs' activities and some brief legislative case studies, this article has demonstrated the different approaches that Eurosceptic parties and groups take to operating within the European Parliament. These differences with regard to group formation, cohesion and engagement with European Parliament activities make sense in terms of variations in Euroscepticism and the presence of niche parties in the Europe of Freedom and Democracy group. Returning to our theoretical framework based on Strøm (1990) and others, we have seen that, especially for the British Conservative Party, the European Conservatives and Reformists group's formation represents a sometimes uneasy mixture of policy-seeking and party-management aims, rather than office-and vote-seeking. For the UK Independence Party, the Europe of Freedom and Democracy group was formed largely on the basis of distinctive practical advantages, such as speaking time in plenary for publicity more than policy and resources for political campaigns rather than mainly for attempts to influence legislation. These differ somewhat from the pragmatic motivations that Bressanelli (2012) refers to among mainstream groups, as they are about seeking publicity more than policy influence in the European Parliament and they concern resources for extra-European Parliament campaigning.
The different motivations for group formation are reflected in differences between the groups' cohesion. The European Conservatives and Reformists group is largely cohesive in roll-call votes and has a level of ideological diversity that compares favourably with others. By contrast, views within the Europe of Freedom and Democracy group are wide-ranging on EU integration, although its diversity is comparatively limited in left-right terms. As we expected 257 EUROSCEPTIC GROUPS IN THE EUROPEAN PARLIAMENT on the basis of two types of niche party being represented in the group, there is no group whip and voting cohesion in roll-call votes is low on distributional policies as well as in the examples of market regulation and EU integration issues we considered here. This effect of niche parties seems to be specific to the presence of hard Eurosceptics and regionalists in the same group as we observed high cohesion in roll-call votes in the Greens/European Free Alliance.
While it would be wrong to suggest that Europe of Freedom and Democracy is entirely disengaged from the European Parliament's activities, it is hard to speak of this group as an individual actor as its two largest parties take rather different approaches. This is consistent with our suggestion that policy-seeking behaviour in the European Parliament will vary by type of Euroscepticism. The soft Eurosceptic European Conservatives and Reformists group engages much more with the European Parliament's work and -as we saw through examples of British Conservative MEPs' activitiesattempts to influence policy outcomes by coalition building and trying to set the agenda for forthcoming proposals in some areas. But when it comes to policy influence, numbers matter in the European Parliament. The comparatively small size of the European Conservatives and Reformists combined with its stance on European integration, which differs from that of the three largest party groups, mean that there are many areas in which it finds itself on the losing side.
The existence of the European Conservatives and Reformists means that party competition in the European Parliament now involves a coherent soft Eurosceptic group, made up partly of governing parties, which attempts to influence policy but which is limited by being only the fifth largest group. The basis of the group as primarily Eurosceptic and its relative cohesion on this make it different from the Union for a European of the Nations. The presence of a hard Eurosceptic co-leader of the Europe of Freedom and Democracy group also means that there is now a voice in the European Parliament's Conference of Presidents and at the European Parliament's high-profile plenary sessions arguing against the entire integration project. With hard and soft Eurosceptic views present via the leadership of European Parliament groups, we may be seeing better representation of these stances via Europe's parliament. Whether voters are aware of the very different approaches 258 GOVERNMENT AND OPPOSITION taken to the European Parliament by hard and soft Eurosceptic MEPs is less clear and requires further research.
Looking ahead to the situation after the 2014 European elections, the European Conservatives and Reformists group seems likely to remain in place but its chances of growing substantially would appear to be slim. Cameron's veto in 2011 and the group leadership style of Martin Callanan, who has tended to diverge from the Conservative Party leadership on some issues -for instance, he called for Greece to leave the euro -may make it more difficult for the European Conservatives and Reformists to gain parties currently in the European People's Party and Alliance of Liberals and Democrats for Europe. The group will therefore have to continue attempting to gain influence via some of the approaches set out here. While some of its new MEPs either arrived with or have begun to develop policy expertise, several of the Conservatives' most experienced members have chosen not to stand in 2014, including Sir Robert Atkins, Giles Chichester, Malcolm Harbour, Struan Stevenson and Robert Sturdy. Their replacements, if elected, will need to learn quickly how to operate in the European Parliament if the group is to continue trying to influence policy through specialism in particular areas. Both the European Conservatives and Reformists and Europe of Freedom and Democracy are vulnerable groups due to the number of single representatives of member states in them. This means their future composition -and survival in the case of the smaller Europe of Freedom and Democracy -is dependent upon how these very small parties perform in 2014. While it seems likely that the UK Independence Party and the Italian Northern League will be present in reasonable numbers, it is difficult to predict with certainty the other members of any future group they develop. Should there be higher numbers of Eurosceptic MEPs after the 2014 ballot, the decisions that Eurosceptics have to make about group choice and how to approach the European Parliament, highlighted in this article, may become even more significant. 27 Interview with Europe of Freedom and Democracy group staff, 10 November 2011. 28 These data are available at: www.votewatch.eu/en/activity-statistics.html. 29 Interview with Europe of Freedom and Democracy group staff, 28 September 2010.