Voting patterns and alliance formation in the European Parliament

Abstract

Members of the European Parliament (MEPs) have voluntarily formed transnational political groups and invariably follow the voting instructions of these groups. This is intriguing as there are few obvious incentives for doing so. Unlike national parties, for example, the political groups in the European Parliament are not punished by the electorate if they are divided on key issues, as citizens know very little about what goes on inside the European Parliament. This paper pieces together an explanation of why the European political groups exist and why they have become so powerful by looking at the determinants of group cohesion and by undertaking a spatial analysis of voting in the European Parliament. MEPs who share preferences on a range of issues on the European Union policy agenda have an incentive to establish a division-of-labour contract and to share the costs of collecting information. Once internal party policy specialization and agenda setting has been established, MEPs have incentives to follow the voting instructions of their group owing to the advantages of cohesion in a context of repeated voting.

1. Introduction

The European Parliament is a fascinating laboratory for the study of social behaviour, in general, and the behaviour of elected officials, in particular. The European Parliament is a large and heterogeneous chamber, and has grown more heterogeneous over time. At the time of the first European Parliament elections in June 1979, there were 410 Members of the European Parliament (MEPs) from 10 European Union (EU) member states and 51 different national political parties. Approximately 30 years later, the European Parliament has 785 MEPs from 27 member states and over 170 national political parties.

Rather than sit as delegates of their member state or national party, the MEPs sit in ‘political groups’ according to their ideological preferences (table 1). Research suggests that these transnational party alliances are more than convenient umbrella organizations. The political groups have become powerful actors, able to marshal their troops in support of the policy positions of the group leaders (Raunio 1997; Kreppel 2002; Hix et al. 2005). For example, Noury & Roland (2002) estimate that if one only knows which political group an MEP belongs to, one could correctly predict her voting behaviour 90 per cent of the time, while if one knows only which member state an MEP belongs to, one could correctly predict her voting behaviour only 10 per cent of the time. In other words, a British Labour MEP is far more likely to vote with a French or German Social Democrat than with a British Conservative or British Liberal Democrat.

Table 1.

Political groups in the European Parliament in January 2008. (The ideology column describes the general policy positions of the national members' parties in each European political group.)

political group	ideology	number of MEPs	percentage of MEPs
European People's Party–European Democrats	conservatives	289	36.8
Party of European Socialists	socialists	215	27.4
Alliance of Liberals and Democrats for Europe	liberals	101	12.9
Union for Europe of the Nations	nationals	44	5.6
Greens–European Free Alliance	greens	42	5.4
European United Left–Nordic Green Left	radical left	41	5.2
Independence and Democracy	anti-Europeans	24	3.1
non-attached members	independents	29	3.7
total		785	100.0

This is intriguing for several reasons. First, reforms of the EU treaty since the mid-1980s have transformed the European Parliament from a purely consultative chamber into a powerful legislature, with the power inter alia to amend and block most EU laws. How the MEPs vote now affects the lives of approximately 500 million EU citizens. Not surprisingly, MEPs are now lobbied by governments, national politicians, interest groups and professional lobbyists. Yet, the European political groups have not collapsed in the face of these higher political stakes.

Second, the main political groups have become more internally heterogeneous as a result of EU enlargement and their expansion to bring in parties with more varied policy preferences. This is particularly the case with the largest group in the current parliament, the European People's Party (EPP), which is a broad coalition of conservatives and other parties on the centre-right.

Third, unlike most party ‘factions’ in democratic parliaments, the political groups in the European Parliament have few powers to force their members to follow voting instructions. National parties rather than the European political groups control the process of selecting candidates in the European Parliament elections. And, there are no personal electoral incentives for MEPs to vote with their European political groups, as the European Parliament elections are generally fought by national parties on the performance of national governments and not on the performance of the groups in the European Parliament (van der Eijk & Franklin 1996; Hix & Marsh 2007).

So, why do MEPs vote with their European political groups? To answer this question, the next section looks at the standard explanations in political science of why parties form and act cohesively in parliaments. Section 3 then investigates the determinants of political group cohesion in the European Parliament, while §4 employs a spatial analysis to look at individual MEP voting behaviour. Section 5 then relates the results of these analyses to the theoretical ideas. Finally, §6 contains a brief conclusion.

2. Theories of parties in parliaments

Why do political parties form and act collectively in parliaments? The answer might, at first, seem relatively trivial for anyone with a basic understanding of democratic politics. A politician has a much better chance of being elected if he or she is a member of a political party, as the party (usually) provides campaign resources (people and money), and a party label efficiently communicates a large amount of information about the retrospective and prospective policy preferences of the politicians. Then, once elected, the politician has an incentive to maintain the coherence of the party ‘brand’, for example by following the voting instructions of the party leadership. If the party is divided in the parliament, it will send mixed signals to the electorate about what policies the party will pursue in the future (cf. Schlesinger 1984; Cox 1987; Cox & McCubbins 1993; Aldrich 1995). Elected representatives have an incentive to follow party discipline owing to the potential reputation loss and negative electoral consequences, which might ensue otherwise.

This seems straightforward enough. However, why do politicians decide to form party organizations inside legislatures independently of the process of democratic elections? The first modern parties emerged in parliaments in the early nineteenth century, long before universal suffrage and mass electoral politics (LaPalombara & Weiner 1966). As the Federalists and Republicans in the early US Congresses, pre-democratic ‘parliamentary parties’ in Europe included the Whigs and Tories in the British House of Commons, the National Liberals and Progressives in the German Reichstag and the Conservative and Left (Venstre) parties in the Scandinavian parliaments (Duverger 1954 [1951]). Similarly, in the European Parliament, the political groups formed and began to act cohesively before the first elections in 1979, and have continued to develop despite almost no reward or punishment for their activities at the ballot box. In other words, an explanation of parliamentary parties must be based on incentives that are largely internal to parliamentary politics rather than mainly external, relating to the electoral process.

The first such explanation is that legislative parties reduce the volatility of policy and thus increase the predictability of parliamentary politics. In any session of a parliament, there are a high number of items on the agenda. Also, each one of these items could have multiple sub-issues within it with various policy implications. If an issue is multidimensional, there may be no stable policy equilibrium if anyone is free to propose amendments (McKelvey 1976; Riker 1980). In other words, in the absence of some form of collective organization, legislative politics is potentially highly chaotic and uncertain. Legislative parties create voting cohesion among their members, which strongly reduces the dimensionality of voting and also increases stability and predictability.

By establishing an organization (a party) with other politicians with similar policy preferences across a range of issues, a politician has a much greater chance of achieving her policy goals in the legislative process, and at a lower cost. It is very costly for a politician to carefully consider her position on each issue and then cast her vote sincerely. The politician would have to collect a huge amount of information to be able to work out her position on every item. Also, the politician is unlikely to know before a vote whether she will be on the winning side, the losing side or pivotal. If she could persuade some colleagues to vote the same way as she does, her chances of being on the winning side are likely to increase. The group of politicians could agree to delegate to the person among them who has the most expertise on a particular issue to instruct the others how to vote. The group could then be confident that they will all vote the right way and that their chances of being on the winning side would be much higher.

However, two members of the group might claim to be experts on the same issue yet disagree on how the group should vote and so issue competing instructions. Which way should the others vote now? It immediately becomes apparent that since the behaviour is repeated, a formal organization is required, with a leadership structure and rules governing how these sorts of issues will be resolved.

A solution is a division-of-labour contract, where back-benchers provide labour and capital (such as policy expertise), and leaders distribute roles to back-benchers (such as positions in committees), communicate party positions (issue voting instructions) and enforce the terms of the party organization contract (such as expelling members who do not following the instructions) (cf. Kiewiet & McCubbins 1991; Cox & McCubbins 1993). Even without external electoral incentives, then, party organizations inside parliaments reduce transaction costs for politicians and allow politicians to overcome collective action problems. In this sense, party organizations in parliaments are similar to firms in the labour market (Coase 1937).

However, when observing a high level of party line voting, it is often difficult to identify the precise effect of the party division-of-labour organization independently of the effect of legislators’ preferences. As Krehbiel (1993) explains:

In casting apparently partisan votes, do individual legislators vote with fellow party members in spite of their disagreement about the policy in question, or do they vote with fellow party members because of their agreement about the policy question? In the former case,…partisan behavior may well result in a collective choice that differs from that which would occur in the absence of partisan behavior. In the latter case, however,…the apparent explanatory power of the variable, party, may be attributed solely to its being a good measure of preferences. (1993: 238, italics in original).

Specifically, if all the members of a party have the same preferences when faced with a choice between a status quo policy and an alternative proposal, then if the members have the same information about the policy implications of the two choices, all the members will decide to vote the same way, whether or not their party has instructed them to do so and threatened to enforce this instruction.

Nevertheless, even with heterogeneous policy preferences among its members, a party organization may shape policy outcomes by controlling what issues get to the floor of the chamber (Cox et al. 2000; Amorim Neto et al. 2003; Cox & McCubbins 2005). If a party leader has a monopoly on agenda setting, she will only make proposals that will lead to policy outcomes that are closer to her ideal policy than the current policy status quo. One side effect of this power is that a party leader does not need to work hard to enforce party cohesion, since, if her party is divided on an issue, it is unlikely that the issue would be put to the chamber. Some bills may be passed with the support of parties that do not hold agenda-setting power. On these bills, leaders will not need their party to act cohesively if there is a sufficient majority in the chamber in favour of the proposal. In general, though if control of the agenda is concentrated in the hands of one party or a coalition of parties, the members of the party (or parties) holding this power are likely to vote the same way on most bills.

3. Determinants of political group cohesion in the European Parliament

How far do these theories explain the organization and development of the political groups in the European Parliament? To answer this question, a good starting point is to look at how MEPs vote, and how far their voting behaviour is shaped by the policy positions of the European political groups, as opposed to the interests of the member state governments or national political parties.

There are three types of votes in the European Parliament: a ‘show of hands’ vote; ‘electronic votes’, where MEPs press the yes, no or abstain buttons on their desks and the vote outcome is announced but how each MEP voted is not recorded; and ‘roll-call votes’, where the vote choice of each MEP is reported in the minutes. Under the European Parliament's rules, only certain votes are required to be taken by roll-call. However, a ‘political group’ or a fifth of all MEPs can request any vote to be taken by roll-call. In practice, roughly a third of votes are by roll-call.

To measure party cohesion in roll-call votes, we use the Rice (1928) index defined as follows:

$$A{I}_i=\frac{|Y_i-N_i|}{Y_i+N_i},$$

where Y_i denotes the number of yes votes expressed by group i on a given vote and N_i is the number of no votes. The index equals 1 if all the members of a group vote together and equals 0 if the members are equally divided between these voting options.

If there is not much variation in the overall majority size, party cohesion in one period can easily be compared with that in another period. However, if there are large variations in the overall majority size, measuring the absolute level of cohesion is deceptive, as parties will be measured as more cohesive if votes are ‘lopsided’ than if votes are evenly split. The cohesion of the European Parliament as a whole has varied considerably since 1979: rising between 1979 and 1987 and then declining until 2004. So, to compare group cohesion in different periods, Hix et al. (2005, 2007) calculated a ‘relative cohesion’ index by dividing the basic cohesion index of a group in a particular period by the cohesion score for the European Parliament as a whole in the same period. The resulting score was then divided by 2, so that the relative cohesion scores were predominantly between 0 and 1.

First, looking at the average relative cohesion scores of the European political groups (figure 1), transnational party cohesion in the European Parliament is relatively high and has risen since the late 1980s. By way of comparison, the Democrats and Republicans in the US Congress have voting cohesion scores around 0.80 while most parties in national parliaments in Europe score above 0.90. Second, as the figure also shows, the European political groups vote in a more cohesive way than do national groups of MEPs, and the gap between voting as European political groups and voting as national groups of MEPs has widened since the mid-1980s.

Figure 1.

Party and national voting cohesion in the European Parliament in the first five directly elected European Parliaments. The figure shows the average cohesion of the political groups and national groups of MEPs in all the roll-call votes in each European Parliament between 1979 and 2004. The figure was calculated from the data used in Hix et al. (2007).

To understand why party cohesion has risen and what might explain variances between the groups, two types of analysis can be undertaken (cf. Hix et al. 2007). First, a time-series analysis, looking at changes in political group cohesion between different periods. This allows us to look at the effects of the changing powers of the European Parliament, the enlargement of the EU, the changing sizes of the political groups and the internal ideological and national make-up of the groups. Second, a cross-sectional analysis, looking at variances in political group cohesion in all votes in a given period. This allows us to look at the effects of agenda setting, pivotality, political group size and the EU's legislative procedures.

To illustrate the cross-sectional method, we estimate a statistical model of the determinants of the relative cohesion scores of the six main groups on each roll-call vote in the fifth European Parliament. Three sets of factors are likely to have influenced party cohesion: (i) roll-call vote characteristics, (ii) political group characteristics, and (iii) the power of the European Parliament.

Regarding roll-call vote characteristics, we included the following variables: absolute majority; final reading; legislative vote; participation; and whole bill. We expect cohesion to be higher when an absolute majority of MEPs is required to pass a bill, when participation is higher (indicating that the issue is more important for the MEPs), when a vote is on the final reading of a bill and when a vote is on a complex high-dimension issue (for example on a whole bill rather than on an individual amendment). By contrast, we expect cohesion to be lower on a legislative vote (as opposed to a European Parliament own resolution), when the proposal has been put forward by the Commission rather than initiated internally by the MEPs.

Four dummy variables capture political group characteristics: whether the group was an agenda setter (if the group proposed the bill or the amendment); whether the group requested the roll-call vote; whether the group was pivotal in the vote (if the result of a vote would have been different if the group had voted on the other side); and finally whether the group was on the winning side. Here, we expect all these factors to increase group cohesion.

Regarding the power of the European Parliament, two variables are included: (i) a dummy variable for votes under co-decision procedure, where political stakes are greater as the European Parliament has more powers and (ii) a dummy variable for consultation procedure. Here we expect cohesion to be higher under the co-decision procedure (when the European Parliament has equal legislative power with the EU governments) and lower under the consultation procedure (when the European Parliament is weaker than the Commission and the EU governments).

Finally, we included eight dummy variables corresponding to policy areas including economic, environmental, social, external, agricultural, institutional, internal and budgetary votes.

The results indicate that pivotality and group cohesion are positively correlated regardless of the ideological position of a political group (see table A1 in appendix A). However, the groups are less cohesive when participation is higher. This suggests that the groups are less cohesive on high-stakes issues as it is reasonable to assume that participation is higher on important votes. We also find that cohesion is higher on whole bills rather than on individual amendments, which suggests that on complex high-dimension issues MEPs are more likely to follow the instructions of their leaders. For some variables, our analysis leads to mixed results: being on the winning side increase cohesion only for socialists, conservatives and nationalists. Other variables, such as being an agenda setter or requesting a roll-call are not statistically significant, again independently from MEPs’ ideological preferences.

Table A1.

Determinants of political group cohesion. (The dependent variable is the relative Rice index for a political group in a vote. Robust t-statistics in parentheses. ^*p≤0.05, ^**p≤0.01. Policy area variables (including economic, social, agricultural, institutional, internal and budgetary variables) are included in the estimations but their coefficients are not reported.)

	PES	EPP-ED	ELDR	UEN	G/EFA	EUL/NGL
majority size	−0.289	−0.367	−0.220	−0.375	−0.392	−0.372
	(15.63)**	(19.49)**	(14.15)**	(17.00)**	(24.28)**	(19.97)**
absolute majority	−0.014	0.030	−0.034	0.013	−0.006	0.018
	(1.69)	(3.11)**	(3.84)**	(1.14)	(0.79)	(2.03)*
final reading	0.012	0.031	0.017	0.007	0.014	0.016
	(1.10)	(2.64)**	(1.66)	(0.47)	(1.49)	(1.23)
whole bill	0.020	0.030	0.000	0.019	0.022	0.023
	(2.94)**	(4.30)**	(0.02)	(2.08)*	(3.36)**	(3.15)**
legislative	−0.080	−0.073	−0.007	0.042	0.030	−0.047
	(2.95)**	(1.90)	(0.22)	(0.91)	(0.65)	(0.86)
agenda setter	−0.005	−0.006	−0.009	−0.014	−0.002	0.015
	(0.99)	(0.99)	(1.28)	(0.40)	(0.20)	(1.35)
party calling RCV	0.017	0.012	0.003	0.017	0.007	0.012
	(2.12)*	(2.05)*	(0.44)	(1.24)	(1.41)	(1.46)
party is winner	−0.036	−0.027	0.005	0.028	−0.002	0.006
	(5.54)**	(4.01)**	(0.93)	(4.12)**	(0.39)	(1.15)
party is pivotal	0.098	0.043	0.181	0.176	0.232	0.227
	(17.11)**	(7.79)**	(26.16)**	(9.61)**	(38.71)**	(26.76)**
participation	−0.060	−0.067	−0.044	−0.142	−0.068	−0.080
	(4.07)**	(4.10)**	(3.13)**	(6.43)**	(4.84)**	(4.68)**
co−decision	0.056	0.037	−0.004	−0.025	−0.023	0.060
	(2.11)*	(0.97)	(0.14)	(0.55)	(0.49)	(1.10)
consultation	0.044	−0.001	−0.016	−0.030	−0.076	0.003
	(1.70)	(0.03)	(0.55)	(0.67)	(1.64)	(0.06)
constant	0.880	0.927	0.772	0.887	0.962	0.913
	(44.70)**	(45.74)**	(42.36)**	(33.90)**	(54.30)**	(42.88)**
observations	5163	5163	5160	5082	5159	5159
R-squared	0.20	0.14	0.27	0.10	0.35	0.25

By contrast, a time-series analysis of political group cohesion suggests that larger political groups are on average more cohesive than smaller political groups, and that the political groups have voted more cohesively since the major increase in the powers of the European Parliament in the Maastricht Treaty (cf. Hix et al. 2005). Meanwhile, ideologically centrist groups are neither more nor less cohesive than more extreme groups, and variations in internal ideological or national heterogeneity between the political groups do not have a significant effect on the voting cohesion of the groups. Similarly, as the results for the second model show, at the political group level, as a group grows bigger its voting cohesion increases, although not significantly so. Nevertheless, while increased ideological heterogeneity does not decrease the voting cohesion of a group, increased national heterogeneity does.

4. Spatial analysis of voting in the European Parliament

With the new availability of parliamentary voting data on the internet, the growth of computer power and the development of new scaling technologies, there has been an explosion of research in the last decade on the geometric scaling of parliamentary votes (e.g. Londregan 2000; Voeten 2000; Schonhardt-Bailey 2003; Clinton et al. 2004; Morgenstern 2004; Rosenthal & Voeten 2004; Poole 2005; Spirling & McLean 2007). One of the most popular scaling methods, which was developed by Keith Poole and Howard Rosenthal for studying voting in the US Congress, is known as NOMINATE (from NOMINAl Three-Step Estimation) (Poole & Rosenthal 1985, 1997, pp. 233–51). W-NOMINATE (the non-dynamic version of the method) estimates the ‘ideal point’ of each individual parliamentarian as follows.

Let s denote the number of policy dimensions (k=1,2, …, s), p denote the number of parliamentarians (i=1,2, …, p) and q denote the number of roll-call votes (j=1,2, …, q). Let parliamentarian i's ideal point be x_i, which is a vector of length s. The ideal point for a given parliamentarian is fixed on any given dimension. Call z_jy the policy outcome of dimension s, where y refers to the policy outcomes associated with a yes vote. W-NOMINATE then assumes that parliamentarian i, who votes sincerely, has a utility function over outcome y on vote j of

$$U_{ijy}=u_{ijy}+{ϵ}_{ijy}=\beta \text{exp}\left[\text{-}{d}_{\text{ijy}}^{\text{2}}\right]+{ϵ}_{ijy},$$

where u_ijy is the deterministic portion of the utility function; ε_ijy is the stochastic (idiosyncratic or error) portion; and the d_ijy term is the Euclidean distance between x_i and z_jy. The coefficient β is a constant, which acts as a signal-to-noise ratio: as β increases, the deterministic element of the function increases relative to the stochastic element and perfect spatial voting results, and as β decreases, voting becomes more random. The usefulness of outcome n on vote j is defined simply by substituting n for y where z_jn is defined accordingly. The stochastic term ε_ijy is assumed to have an extreme value distribution. The constructed likelihood function is then maximized (using a standard algorithm such as Gauss–Newton method) to obtain the parameters of the model: the dimensions of the political space and a ‘score’ for each parliamentarian on each recovered dimension.

Figure 2 shows the results of applying W-NOMINATE to the roll-call votes in the first, second and fifth elected European Parliaments. As in most other democratic parliaments, voting in the European Parliament is predominantly one-dimensional, and increasingly so. The W-NOMINATE scores on dimension 1 correctly predict approximately 85 per cent of votes in the first elected European Parliament and approximately 90 per cent in the fifth elected European Parliament, while the scores on dimension 2 only predict an additional 6 per cent in the first Parliament and 2 per cent in the fifth.

Figure 2.

Revealed ideal point estimates of MEPs in the three European Parliaments (a) the first elected European Parliament, 1979–1984, (b) the third elected European Parliament, 1989–1994, (c) the fifth elected European Parliament, 1999–2004. These figures are the result of applying the W-NOMINATE geometric scaling metric to all the roll-call votes in each of these parliaments. Each dot is the estimated location of an MEP on the two main dimensions of voting in a particular European Parliament. The method is an inductive scaling technique, and as such does not provide an a priori substantive meaning of the dimensions. The colours in the figures indicate the political groups: red dots, the social democrats; blue dots, the conservatives and Christian democrats; yellow dots, the liberals; dark red dots, the radical left; light blue dots, the British conservatives and their allies (who then joined the EPP); navy blue dots, the national conservatives; green dots, the greens and regionalists; purple dots, the extreme right; pink dots, the anti-Europeans; grey dots, the non-attached MEPs.

But, what is the meaning of the first and second dimensions estimated by W-NOMINATE? One weakness of such inductive scaling methods is that they cannot provide a substantive interpretation of the content of the dimensions. The relative location of the political groups in the maps suggests that the first dimension in the European Parliament is the left–right, as the parties seem to be ordered from left to right exactly as one would expect. Meanwhile, the second dimension might be related to pro-/anti-European positions, with the pro-EU groups at the top of the figures and the more anti-EU groups at the bottom.

Nevertheless, the substantive meaning of the dimensions can be investigated more systematically by looking at the exogenous determinants the W-NOMINATE scores. Here, the dependent variable is the average score of each national party's group of MEPs on a dimension in a European Parliament. The unit of analysis is a national party's group of MEPs, as opposed to individual MEPs, because reliable exogenous measures of the policy positions of actors in each European Parliament only exist for national parties, and because national parties have a powerful influence on the behaviour of their MEPs via the control of candidate selection in European elections.

There are two main types of independent variables: (i) policy positions of national parties and (ii) the institutional interests of national parties (whether their party is in national government in a particular period or whether their party has a European Commissioner). The former influence the policy preferences of the MEPs and the latter influence what types of issues MEPs are voting on, as national governments (in the EU Council) and the European Commission are the external agenda setters in the EU legislative process. The national party policy positions are taken from political scientists’ estimations of parties’ positions on two dimensions: left–right, and pro/anti-European integration, scaled between 0 (most left/anti-European) and 1 (most right/pro-European) (Marks & Steenbergen 2004). For the institutional variables, two dummy variables are included: the first coded 1 if a national party was in government for the majority of a particular European Parliament, 0 otherwise; and the second coded 1 if a national party had an EU Commissioner during a particular European Parliament, 0 otherwise. Finally, to examine whether these factors shape MEP behaviour within the political groups as well as between them, separate models with dummy variables for each political group are estimated.

The results are summarized in figure 3 and reported more fully in table A2 in appendix A. As suggested by the spatial maps (figure 2), MEP scores on the first dimension are strongly explained by exogenous left–right policy positions of national parties. In fact, 1 s.d. change along the left–right policy dimension corresponds with a 78 per cent standard deviation change on the first W-NOMINATE dimension. As model 2 shows, left–right policy positions also explain variations in MEP voting behaviour within the European political groups. In other words, a national party that has a policy position to the left (right) of the average member of a European political group will be revealed to vote slightly to the left (right) of the average member of their group. By contrast, institutional interests of national parties do not correlate with scores on the main dimension of voting. And, once one controls for European party positions, European integration preferences are not relevant explanatory factors on the main dimension of voting in the European Parliament.

Figure 3.

Determinants of national party voting in the European Parliament. The figure plots the coefficients from four OLS regression models of the average W-NOMINATE score of a national party's MEPs in a particular European Parliament ((a) dimension 1 scores, (b) dimension 2 scores). Dummy variables for each European Parliament are included in all models. There are 352 observations (national parties) in each model. The figure plots the coefficients for each variable and the 95% confidence interval. The full results are reported in table A2 in appendix A. (a) Grey circles, model 1 (adj R²=0.696); black circles, model 2 (model 1+ EP group fixed effects; adj. R²=0.890); (b) grey circles, model 3 (adj R²=0.290); black circles, model 4 (model 3+ EP group fixed effects; adj. R²=0.632).

Table A2.

Determinants of national party voting in the European Parliament. (The dependent variable is the average position of a national party's group of MEPs on dimension 1 or 2 in a particular European Parliament (5-year period). The positions of the MEPs on the dimensions are produced by applying the W-NOMINATE method of ideal point estimation to all the roll-call votes in each European Parliament. The parameters are estimated by linear OLS regression. Standard errors in parentheses. ^*p≤0.05, ^**p≤0.01. Dummy variables for each European Parliament are included but not reported.)

	dimension 1 scores		dimension 2 scores		variable summary statistics
	(1)	(2)	(3)	(4)	mean	s.d.	min	max
dependent variable: average W-NOMINATE score of a national party's group of MEPs in a European Parliament				dim.1 scores:	0.018	0.490	−0.903	0.863
				dim.2 scores:	−0.055	0.449	−0.953	0.927
left–right position of national party	1.686** (0.053)	0.826** (0.062)	−0.206* (0.089)	−0.257* (0.104)	0.503	0.229	0	1
EU integration position of national party	0.193** (0.056)	−0.033 (0.049)	0.601** (0.079)	0.174* (0.081)	0.716	0.277	0.000	1.000
national party is in government	0.052 (0.033)	0.019 (0.020)	0.129** (0.047)	0.036 (0.034)	0.381	0.486	0.000	1.000
national party has a commissioner	0.064 (0.038)	−0.006 (0.025)	0.253** (0.053)	0.017 (0.042)	0.216	0.407	0	1
constant	−0.893** (0.062)	−0.317** (0.062)	−0.575** (0.095)	−0.457** (0.103)
political group fixed effects	No	Yes**	No	Yes**
member state fixed effects	No	No	No	No
observations	352	352	352	352
adjusted R-squared	0.696	0.890	0.290	0.632

Meanwhile, EU integration preferences matter more than left–right preferences on the second dimension of politics in the European Parliament. A 1 s.d. change along the European integration policy dimension corresponds with a 37 per cent standard deviation change on the second W-NOMINATE dimension, whereas a 1 s.d. change on the left–right dimension corresponds with only a 10 per cent standard deviation change on the second W-NOMINATE dimension. Also, MEPs from national parties in government and who have Commissioners are located towards the top of the second dimension, while MEPs from parties in opposition are located towards the bottom. However, these institutional interests are not significant once party dummies are introduced, which reveals that these institutional interests do not produce voting conflicts within the political groups. Again, as with the first dimension, a large proportion of the variance is explained by the location of the political groups.

5. The effect of repeated interactions and lopsided votes

So, MEPs increasingly vote along transnational political lines rather than national lines, and these European political groups are strong determinants of MEP behaviour on the two main dimensions of politics in the European Parliament. What is intriguing is that the European political groups have few powers to enforce party line voting. Unlike most national parties, the European political groups do not have the power to prevent MEPs from standing at the next election if they do not follow voting instructions. The groups can influence the allocation of office rights inside the European Parliament, such as committee positions and rapporteurships (MEPs who prepare a report on the bill and propose amendments). However, even this power is shared with national parties, as committee positions and rapporteurships are allocated to national parties, who then decided which of their MEPs will hold what offices. In other words, the emergence and cohesion of the European political groups is largely a result of the voluntary and strategic actions of the MEPs and their national parties.

As the theory of agenda control would suggest, the leaderships of the political groups could promote collective group action by preventing issues from getting to the floor on which the group is internally divided. However, unlike the US Congress, the parties in the European Parliament do not control the legislative agenda. In the EU legislative process, the European Commission has the right of initiative and the EU Council can amend legislation. This means that the political groups in the European Parliament are often forced to vote on issues proposed by others. And, even when an issue arrives in the European Parliament, agenda-control rights inside the chamber (such as committee chairs and rapporteurships) are allocated between the groups on a broadly proportional basis rather than monopolized by one particular group. This means that a group leadership cannot prevent other groups proposing amendments or requesting a roll-call vote on an issue on which the group might be split. In fact, the groups vote less cohesively on amendments proposed by other groups and in roll-call votes called by other groups (Hix et al. 2007, ch. 6). Hence, if a group or a coalition of groups could control the agenda, they would vote even more cohesively than they do.

Nevertheless, the spatial analysis of voting suggests that the European political groups are based on the preferences of MEPs and national parties about the direction of social and economic policies at the European level, as captured by the traditional ‘left–right’ dimension of politics. However, the analysis of group cohesion suggests that declining preference homogeneity inside the largest groups (as a result of expanding group membership) did not have a negative effect on voting cohesion. How can these results be reconciled? One possible interpretation is illustrated in figure 4. A significant proportion of MEPs in the two largest political groups in the fifth elected European Parliament have centrist policy preferences, such as the British Labour MEPs in the Party of European Socialists (PES) and the Benelux Christian Democrat MEPs in the EPP (figure 4a). In their voting behaviour, however, these centrist MEPs tend to vote with their European political group colleagues rather than with the MEPs in the other group, who may have closer policy preferences.

Figure 4.

Comparison of posited preferences and revealed behaviour of MEPs in the 2004–2009 European Parliament. (a) A kernel density plot of the preferences of socialists (PES) and conservatives (EPP) in the sixth elected European Parliament (posited preferences of MEPs), as measured by the self-placement of the MEPs on a 10-point left–right scale, where 0 equals the furthest left and 10 equals the furthest right. The data come from a survey of the MEPs that was conducted between March and June 2006 (see Farrell et al. 2006; and the European Parliament Research Group website: http://www.lse.ac.uk/collections/EPRG/survey.htm). (b) A kernel density plot of the preferences of the same PES and EPP MEPs in the sixth elected European Parliament (revealed behaviour of MEPs), as measured by applying W-NOMINATE to all roll-call votes between July 2004 and December 2006. The correlation between the first dimension W-NOMINATE scores and the left–right self-placement scores in this period is 0.70.

An anecdote helps to explain why this happens. In May 2005, the European Parliament voted on the working time directive. A key provision of the legislation was the removal of a British opt-out from the existing rules governing the maximum number of hours certain employees could work. The British Labour government wanted to keep the opt-out, but had lost the vote in the EU Council. The Labour leadership put pressure on the 19 British Labour MEPs to vote against the directive in the European Parliament. In pure policy-preference terms, the British Labour MEPs agreed with the position of their party leaders in London. However, the national party delegation of MEPs decided to vote with the other members of the PES group in support of the directive. They did this for two reasons. First, there was a large majority in favour of the legislation (the vote passed by 378 to 262), so a vote against the bill would have been purely symbolic. Second, voting against the PES on such an important piece of legislation would have threatened the future influence of the British Labour MEPs, as the other members of their group would be reluctant to allow Labour MEPs to be rapporteurships or committee chairmen in the future.

In other words, national parties rather than the European political groups ultimately control the MEPs. However, owing to the sheer volume of decisions that have to be made in the European Parliament, and the potential for unstable and chaotic outcomes, national parties have an incentive to join a political group that broadly shares their policy preferences on a range of issues that are likely to come up on the EU agenda, and to delegate organizational and leadership powers to the European political groups, to share information, allocate agenda-setting rights between the member parties and set the broad guidelines of policies. The result is that national parties might be forced to vote against their policy preferences on some issues, but on average will vote according to their policy preferences in the knowledge that they are more likely to achieve these preferences as their colleagues in the group will be voting the same way.

This internal group bargain is self-policing for two main reasons. First, legislative policy making involves repeated interactions. So, if a national party breaks away from the group on a key issue, they may gain in that particular vote but would risk losing out in the future. Second, in most votes, the outcome is easily predictable because the positions of the political groups are usually clear before a vote is taken. If a vote is likely to go a particular way, a national party is unlikely to be pivotal if they vote against their European political group, yet may risk future benefits by doing so. Hence, national parties voluntarily choose to vote with their European political groups approximately 90 per cent of the time.

However, if a key vote is likely to be close, this second calculation changes. For example, this happened in a high-stakes vote in July 2001 on the takeover directive (table 2). This issue—the harmonization of national rules on hostile takeover bids—was certainly ideological, with the MEPs on right broadly in favour of proposed new rules and the MEPs on the left broadly opposed. As a result, the groups to the right of the EPP voted cohesively in favour, while the groups to the left of the PES voted cohesively against. However, the two largest groups were split down the middle along national lines, with the Danish, British, Swedish, Portuguese, Irish, Luxembourg and Finnish MEPs overwhelmingly supporting the legislation and the Dutch, Belgian, Austrian, Greek and German MEPs overwhelmingly opposing it.

Table 2.

Vote on the third reading of the Takeover Directives on 4 July 2001. (The political groups and member states are sorted from the most to the least in favour of the directive in terms of who they voted on the bill. Because the vote was a tie, the directive was rejected. Using the Rice index to calculate voting cohesion in the European Parliament, the average cohesion score of the political groups in the vote was 0.56 (not counting the non-attached MEPs), while the average cohesion score of the member states was 0.63.)

		yes	no	abstain	cohesion score
political group	liberals (ELDR)	45	0	0	1.00
	nationals (UEN)	16	0	0	1.00
	non-attached MEPs	16	4	6	0.60
	anti-Europeans (EDD)	9	5	3	0.29
	social democrats (PES)	80	84	0	0.02
	conservatives (EPP-ED)	98	119	4	0.10
	greens and regionalists (G/EFA)	8	31	2	0.59
	radical left (EUL/NGL)	1	30	7	0.94
member state	Denmark	13	0	1	1.00
	United Kingdom	72	6	0	0.85
	Sweden	19	0	3	1.00
	Portugal	19	1	2	0.90
	Ireland	10	2	0	0.67
	Luxembourg	5	1	0	0.67
	Finland	11	2	1	0.69
	France	45	26	10	0.27
	Italy	32	36	3	0.06
	Spain	26	31	1	0.09
	The Netherlands	9	22	0	0.42
	Belgium	5	16	0	0.52
	Austria	4	14	1	0.56
	Greece	2	21	0	0.83
	Germany	1	95	0	0.98
	total	273	273	22

The two largest groups collapsed because the outcome of the vote was highly uncertain—it ended in a tied vote, which meant that the legislative was rejected. Unlike the story of the British Labour MEPs and the working time directive, in such a close vote, every vote decision by a national party was potentially pivotal rather than purely symbolic. Also, once several national parties signalled that they would vote against the position of their group, the threat of the group withholding the future benefits declined, because the group leadership would find it difficult to withhold rapporteurships or committee chairs from a large number of national parties. In this situation, the national parties were willing to risk the future benefits to try to secure a short-term policy gain.

Interestingly, voting behaviour on close votes as opposed to lopsided votes seems to be opposite in the European Parliament and the US Congress. Snyder & Groseclose (2000), for example, looked at lopsided votes in the US House of Representatives to measure the ideological preferences of Congressmen, on the assumption that the party caucuses will work hard to enforce party discipline in tight vote, and Congressmen will follow instructions in these votes so as not to undermine the electoral appeal of the party. In lopsided votes, by contrast, Congressmen would be freer to vote according to their personal policy preferences. In the European Parliament, by contrast, where the European Parliament elections have very little to do with the performance of the European political groups, the European political groups cannot enforce cohesion in close votes. One might think that MEPs, like US Congressmen, would be also be free to vote how they like in a lopsided vote, as the group knows the outcome. However, an MEP has an interest to keep her policy preferences private in these votes, as revealing a difference of opinion with her colleagues on an important policy issue might lead her colleagues not to trust her to hold a key agenda-setting office on a future issue. Hence, the fluidity of the allocation of agenda-setting rights in the European Parliament, via rapporteurships and committee assignments, as opposed to the more stable and monopolized system of agenda-setting rights in the US Congress, provides an incentive for the MEPs to vote collectively in lopsided votes.

6. Conclusion

The most persuasive explanation of the formation and operation of the political groups in the European Parliament is that they are vehicles for the promotion of the policy preferences of MEPs and their national political parties. MEPs who share preferences on a range of issues on the EU policy agenda have an incentive to establish a division-of-labour contract, to share the costs of collecting information. Once this division of labour has been established, which involves policy specialization and the division of agenda-setting powers inside the political group, MEPs are likely to follow voting instructions from their political group leaders. They do this because they reasonably expect that these instructions are the positions they would come to if they had the time and resources to work out their position on the complex legislative issue. On some issues, an MEP may receive information (e.g. from interest groups) that their preferences in a particular issue are different from those being communicated by their political group, and the frequency of this conflicting information has probably increased as the European Parliament's power has increased. However, an MEP is often unlikely to follow these alternative instructions because the chances of being pivotal in a vote are usually small, and because voting against her group would send a negative signal about the preferences of the MEP relative to the group. If the other members of a group believe the MEP has variant preferences from them, then they will cease to trust the information the MEP is sharing on the issues on which she has specialized knowledge and will prefer to allocate internal party agenda-setting rights to the members who are closer to the average party member.