Matthew Gabel& Clifford Carrubba* The European Parliament and Transnational Political Representation: Party Groups and Political Conflict O ne of the central issues for the emerging Europolity is whether policy-making at the European level can or has been transformed so that a new political space is available for the articulation of organized social interests that might transcend national borders. In particular, we might ask whether the European Parliament(EP) is a suitable arena for contestation among transnational political inte rests, where European national political parties re-align themselves in European parties reflecting transnational political conflict and provide voters with an electoral connection to EU policy-making. This issue is the general focus of this paper. Specifically, we re-examine the conclusion drawn in the literature that party groups—coalitions of national party delega tions in the European Parliament—successfully organize legislative behavior in the EP such that nationality is dominated by shared transnational political interests defined mainly along traditional left-right political lines. Based on analysis of a novel set of data regarding EP legislative behavior, we will argue that the empirical basis for these conclusions is dubious. 1 In addition, we discuss how party group management of legislative votes affects the transparency of EU policymaking and therefore the quality of accountability and representation in the EP. The major conclusion on this front is that party groups hide the vast majority of legislative votes from the eyes of voters, therefore obfuscating legislative behavior. Thus, while the EP is often identified as a source of democratic accountability for EU policy-making because its members are directly elected, our findings suggest that in practice party groups significantly obstruct this channel of popular control over policy-making. The paper is divided into three parts. The first section surveys a variety of recent studies of voting beha vior in the EP designed to examine positive questions about the level of party group cohesion and the cha r* University of Kentucky, Department of Political Science; Emory University, Department of Political Science. 1 The dataset is described more generally and used to address more ge neral issues of roll-call vote analysis in Carrubba, et. al.(2003). acter of party group competition. They examine whether a supranational party system exists and how it organizes political conflict. These data have also been used to address more general questions of normative democratic theory related to European integration. Adopting a“responsible party” model of democracy for the EU, this research has concluded that party groups, at least in terms of organizing the legislature, function in a manner consistent with normative expe ctations about party cohesion and competition. The second section revisits the empirical basis for these positive and normative conclusions. Previous analyses of voting behavior in the EP rely on roll call votes. But roll call votes represent only a small fraction of EP votes and, importantly, the selection of votes for roll call may be endogenous to the level and type of conflict among MEPs. Thus, before we draw conclusions from roll call votes about the character of the party system in the EP, we need to evaluate whether roll call votes are a representative sample of voting behavior. Using a novel dataset of information about votes not decided by roll-call, we conduct such an analysis. 2 The third section of the paper discusses how the sampling properties of roll call votes affect our conclusions about party group cohesion and competition. We also discuss how the prevalence and pattern of roll call vote requests affects the transparency of policy-making in the EP. The concluding section discusses what voting behavior and the use of roll calls indicate about whether the EP provide a forum for the articulation and contestation of transnational political interests and an electoral connection between voters and policymakers in the EP. 2 The EP uses one of four methods for casting votes: 1) voice; 2) hand; 3) electronic; or 4) roll call. Voice and hand votes only record whether the motion passed, electronic votes record the final tally of the vote, and roll call votes record exactly how each legislator voted. The default method of voting is by voice or hand. Electronic votes occur when a voice or hand vote is too close to call, and RCVs occur if a party group or thirty-two members issue a formal request for a roll call. A small number of votes—e.g., the Commission investiture—require a RCV.

Matthew Gabel& Clifford Carrubba The European Parliament and Transnational Political Conflict Europäische Politik (03/2004) 2 Party Groups, Voting Behavior, and the ReThis conclusion is based on several analyses of vote sponsible Party Model patterns among MEPs across a variety of issue areas. The basic methodology of these analyses is to evaluate Empirical research on MEP voting behavior has gener- how PG affiliation relates to these vote patterns. If ally supported two conclusions regarding party group MEPs of the same PG commonly vote together and (PG) cohesion and competitiveness. First, PG cohesion they vote differently from MEPs of other PGs on some is higher than cohesion by nationality, is objectively issues, then this indicates competitiveness. And, if the high for the major party groups, and has generally in- policy areas that account for whether and which PGs creased over time. Studies of RCVs in the EP in the differ in voting behavior are those that commonly de1980s and early 1990s showed high PG cohesion(At- fine the left-right dimension, then the character of this tina 1990; Brzinski 1995; Raunio 1997). Raunio(1997: competition is left-right. 34) showed that, in three-quarters of RCVs for the ma- A broad set of studies, based on RCVs, shows that jor PGs, 90% of MEPs voted with the PG line. Kreppel indeed PG ideological distinctions affect MEP voting and Tsebelis(1999) used correspondence analysis of behavior and PG coalition behavior. Raunio(1997) 100 RCVs from 1989-1994 to confirm that PGs are a showed that coalitions of PGs on RCVs are explained stronger influence on voting coalitions than nationality. by their proximity on the left-right dimension. Kreppel Kreppel(2001) analyzed 300 RCVs from 1980-1996, and Tsebelis(1999) and Kreppel(2000; 2001) add imalso using correspondence analysis and demonstrated portant context to this conclusion, showing that the high PG cohesion over this period. level of left-right ideological competition varies tempoRecent studies confirm these findings. Hix(2001; rally. In particular, the frequency of“grand coalitions” 2002), analyzing RCV data from July 1999-July 2000, between the two largest PGs has increased over time. showed that PG was a stronger determinant of voting But left-right ideology remains an accurate characteribehavior than nationality or individual MEP ideology. zation of policy differences and competition in the EP. Similarly, Noury(2002) and Thomassen, Noury, and Based on a larger number of RCVs, Noury(2002) and Voeten(2002) showed that PGs were the strongest de- Thomassen, Noury, and Voeten(2002) identify four terminant of vote choice from 1989-1999 and that PG dimensions to competition in the EP, but show that the cohesion has increased over time. Finally, Hix, Noury, dominant dimension is left-right. Hix(2001), analyzing and Roland(2002) have analyzed all RCVs since the RCVs from 1999, and Hix, Noury, and Roland(2002), advent of direct election to the EP. This comprehensive using the full set of RCVs from 1979-2001, show that study showed that PG cohesion has remained high inter-PG competition in the EP has been along a single since 1979 and has increased with the legislative po- left-right ideological dimension. This pattern has been wer of the EP. stable over time. Notably, these studies show that nationality is not a This characterization of the EP, if accurate, has imstrong determinant of voting behavior among MEPs. portant implications for our understanding of legislaThis is true on roll call votes generally, and across issue tive politics in the EU, the development of a Europeanareas that one might consider relevant to national as level party system, and the quality of representation in opposed to ideological interests. The clear implication the EU. It also speaks to the prospects for a wellfrom these findings is that party groups appear to functioning parliamentary democracy at the EU level. manage their membership—either through enforcing With the growth in the scope of EU competency at the some form of discipline or managing cross-national expense of national legislative authorities, a variety of logrolls—so as to dilute national conflict over policy. academics, journalists, and politicians have expressed The second general conclusion is that legislative concerns about the quality of democratic control over politics in the EP is competitive, with PGs generally dis- EU policy-making(Weale and Nentwich 1998; Blondel, tinguishing themselves along one main ideological di- Sinnott, and Svensson 1998; Schmitt and Thomassen mension that reflects traditional left-right political con- 2000). Many scholars consider the PGs as essential to flict found at the domestic level in the EU member- improving the quality of democracy in the EU. In difstates. That is not to say there are no other cleavages ferent forms, these scholars appeal to a“responsible in voting behavior. Most studies identify minor cleav- party” model when evaluating the quality of democages that are distinct from the left-right dimension. But ratic control in the EU(e.g., Thomassen, Noury, and the basic conclusion is that this left-right dimension is Voeten 2002; Schmitt and Thomassen 1999). 3 Accordthe major source of conflict among MEPs and the predominant characterization of party group competition. 3 Note that our purpose here is not to advocate or critique this normative model. We describe the model because it is the

Internationale Politikanalyse International Policy Analysis Unit ing to this model, parties serve as the crucial conne ction between voters’ interests and policy-making. To do this, they must provide different policy programs to voters, voters need to vote based on their preferences over policy programs, and parties must behave cohesively in executing their programs when involved in policy-making. These RCV results suggest that PGs meet the requirements of the responsible party model in terms of legislative behavior. MEPs vote according to their PG affiliations – resulting in high internal PG cohesion – and PGs differentiate themselves from each other according to their ideological positions on the Left-Right dimension. Consequently, Hix, Noury, and Roland (2002) conclude that the EP functions as a“normal” parliament, resembling the legislatures of the EU member-states in terms of party cohesion and the character of competition. And, as Thomassen(2002:18) argues, a single left-right dimension to policy conflict is important, as it facilitates voter-party correspondence in policy positions. To see this, note that representation is facilitated if voters’ preferences across the range of policy areas under EU authority are captured by pa rticular parties. This is difficult to attain if the policy space is multidimensional. But if voters and parties share a common ideology, which serves as a shorthand for different packages of policies, then voters can more easily connect their policy preferences with the policy programs on offer by parties. Thus, the apparent commonality between the left-right ideological dimension characterizing voters in the national arena and party groups in the EP is advantageous from the view point of the responsible party model. 4 In sum, the recent empirical studies suggest that party groups in the EP have successfully developed a transnational pa rty system in the legislature, with ide ological interests dominating national interests in policymaking. What is more, these findings indicate that, if European elections were reformed to allow PGs, rather than national parties, to organize and contest the elections, we could imagine a dramatic improvement in the connection between voter preferences and policymaking in the EU. However, as we argue in the next section, the reliability of these conclusions hinges critically on the quality of RCV data. Reconsidering MEP Voting Behavior For most legislative years, over 3/4 of votes in the EP have not been by roll call. This obviously raises the question of whether roll call votes tell us anything meaningful about party group organization of MEP voting behavior, the character of legislative conflict, or the importance of nationality as a political cleavage in the EP. Clearly, roll call votes are an attractive source of information about behavior, precisely because the behavior is observable. But what if the selection of votes for roll call is based on exactly those characteristics of voting behavior and legislative competition that we wish to study? Specifically, what if party groups, which are the most common source of roll call vote requests, have select votes for roll call based on their expectations of the level of party group cohesion and of the type of political conflict associated with the vote? If so, then the votes that are hidden from view—i.e., those not decided by roll call—may be very different from roll call votes in terms of party group cohesion and the character of political conflict. Consequently, roll call votes may not provide a reliable source of information about legislative behavior and the impact of ideology and nationality on EP politics. We are concerned that party groups do indeed select roll call votes based on strategic considerations that would generate misleading inferences about legislative behavior. Previous research on the EP provides several arguments about the selection of votes for roll call. Here we highlight two of the most common 5 ones. First, PGs use RCVs to influence legislative outcomes. As Kreppel(2001:128) states, PG leaders have the ability to reward or punish their membership through a variety of means. 6 However, PG leaders cannot exercise party discipline without some way of monitoring their members’ behavior. Thus, PG leaders have an incentive to request RCVs when they want to enforce party discipline. They will try to employ this tactic on legislation they consider important, for which the outcome of the vote is uncertain, and where they anticipate inducing party cohesion. 7 Second, PGs use RCVs to signal their or other groups’ policy positions to a third party, such as a national electorate or another EU institution(Kreppel 2002: 128). In particular, a PG may want to publicize its policy agenda, to embarrass a rival PG by revealing motivation for much of the empirical analysis we discuss below. 4 But see Gabel and Anderson(2002) and Gabel and Hix 5 See Kreppel(2002: 128-9) for an extremely thorough discussion of these possible strategic motivations. (2002), as a well as Marks and Steenbergen(2003), for a 6 For example, PG leaders control the granting of committee more thorough discussion of the level of correspondence be- seats. tween party conflict in the EP and voter preferences over EU 7 One could also imagine that PG leadership might use roll calls policy. to monitor whether members honor vote-trades. 3

Matthew Gabel& Clifford Carrubba The European Parliament and Transnational Political Conflict Europäische Politik (03/2004) 4 its low cohesion on a particular policy, or to distinguish legislative behavior from RCVs if the italicized condithemselves publicly from other PGs on particular poli- tions do not obtain. Thus, we would like some empiricies they deem significant. cal evidence regarding these conditions. If these arguments are correct, selection bias is likely Fortunately, some relevant information is available to undermine inferences about intra-party group unity, about non-RCVs. Carrubba, et. al.(2003) collected denational conflict, or inter-party group cleavages based scriptive information about non-RCVs for the 1999on RCVs. For example, one of the main findings in the 2000 legislative year. These data allow one to compare EP literature is an objectively high level of PG cohesion. RCVs with non-RCVs on several dimensions relevant to If the arguments presented above are correct, we can- the level of party cohesion and the chara cter of policy not trust such conclusions because the decision to re- conflict. Carrubba, et. al.(2003) present these data quest a roll call is endogenous to the expected level of and provide the statistical analysis supporting the ensucohesion. Furthermore, the signaling arguments imply ing discussion. Here, we simply summarize the analysis that the sample of RCVs is endogenous to the policy and focus on the implications for our understanding of agendas of the PGs requesting roll calls. That is, PG parties in the EP and the previously described concluleaders, pursuing their political agenda, choose among sions regarding the party system at the European level. votes for roll call so as to highlight or conceal policy We divide the discussion into three parts, organized conflict or consensus among PGs or MEPs. This behav- around different characteristics of EP votes. Note that ior could cause RCV samples to lead to incorrect infe r- our conclusions apply whether one focuses on simply ences about the character of policy conflict and the final legislative votes or on votes over amendments dimensionality of voting cleavages. and final legislative votes in the EP in that year. Is this a serious problem? Obviously, we cannot examine this question directly, since we cannot observe how individual MEPs voted on legislation that was not Roll Call Votes and the Type of Motion decided by roll call. And, although theory suggests we might expect a selection bias due to strategic concerns Votes in the EP vary by the type of motion, which varby party groups, one could imagine these effects being ies with legislative procedures. Some votes are on nonsmall or actually canceling each other out. For exam- binding legislation while other votes could effectively ple, some roll call votes might be requested because veto legislation. Clearly, if we are interested in how party group leaders want to induce or demonstrate MEPs vote on legislation, we want to know whether high cohesion in their own party group while other RCVs show us that type of behavior. Comparing RCVs time party groups might request a roll call to display with non-RCVs on type of motion allows us to address disunity in another party group. two important questions: Are RCVs representative of As a re sult, previous studies, while sometimes rec- legislative voted? And, as Hix, Noury, and Roland ognizing the possibility of selection bias, dismiss the is-(2002) claim, are all important votes by roll call? sue. For example, Hix, Noury, and Roland(2002) state: 8 First of all, RCVs are disproportionately called on We cannot exclude the possibility that MEP behavior is different in roll call votes than in other votes. However, it is reasonable to assume that roll call votes are used for the most important decisions. Also, roll call votes are the only votes we can study in detail, the number of roll call votes has increased over time, roll call votes are called on the full range of issues in the European Parliament, and roll call votes do not appear to be called disproportionately by any party group. We can thus be confident that the systematic analysis of roll call votes provides an accurate picture of the European Parliament. non-legislative resolutions(58.9% of all RCVs on final votes and 86.7% of RCVs on final votes and amendments). That is, RCV analyses of legislative behavior are based on many votes that are not on legislative texts that are part of the formal EU legislative procedure. Carrubba, et. al.(2003) describe this distinction further and consider different ways of defining legislative votes. But the simple story is that much of MEP legislative voting behavior is not by roll call and that much of their non-legislative voting behavior is by roll call. This is an important finding since we might imagine that party groups and MEPs approach non-legislative votes differently than legislative votes. Furthermore, the polTherefore, according to Hix, Noury, and Roland icy agenda is probably different on non-legislative reso(2002), one cannot draw reliable inferences about EP lutions than on legislative votes, which can only be initiated by the European Commission. 8 Hix(2001a; 2002a) provides a very similar defense of RCVs as a reasonable sample of MEP legislative behavior. Second, among legislative votes, RCVs are disproportionately called on texts for which the EP vote has

Internationale Politikanalyse International Policy Analysis Unit little or no consequence. The most obvious evidence of this is that a disproportionately large share of roll call votes are on measures decided under the Consultation procedure and a disproportionately small number of texts considered under the latter stages of the Codeci9 sion procedure are decided by roll call. Specifically, only one out of sixteen Assent votes and three out of 619 Codecision II votes had roll calls requested. Thus, this evidence strongly suggests that PG leaders are making sure that these most important votes are specifically not decided by roll call. Thus, if we are interested in understanding MEP legislative behavior, this RCV dataset is simply inappropriate. We would be trying to infer legislative behavior from a sample of votes that are procedurally the least legislatively important votes. Table 1. Percent and Number of Votes and RCVs by Type of Legislative Vote Roll Call Votes and Issue Areas of Legislation Type of Motion All Final Votes RCVs on Final Votes All Votes All RCVs Votes in the EP vary by issue area. A simple proxy for issue area is the identity of the responsible committee for the legislative text. Thus, we can examine whether EP votes are representative of all legislative votes by isConsultation 45.83 60.00 45.24 59.54 sue area by examining the distribution of RCVs and Assent 176 36 1088 103 4.17 1.67 0.67 0.58 non-RCVs by responsible committee. This is particularly important for whether RCVs are reliable sources of information regarding the character of policy conflict in Codecision I 16 18.75 72 1 31.67 19 16 27.90 671 1 37.57 65 the EP. If legislation for some issue areas is rarely decided by roll call, then we might expect RCVs to fail to capture policy conflict related to these issue areas. The distribution of votes across responsible commitCodecision II 12.76 1.67 49 1 25.74 1.73 619 3 tees indicates that RCVs are substantially biased toward legislation from particular committees(see Table 2, p.6). Three committees(Constitutional Affairs; CitiCodecision III 2.86 1.67 0.46 0.58 zen´s Freedoms and Rights, Justice and Home Affairs; 11 1 Codecision II, III, and Assent 23.5 5.2 11 1 26.9 3.9 Economic and Monetary Affairs) account for almost two-thirds of RCVs on legislation, but less than onethird of all legislative votes. Furthermore, some issue areas receive roll call votes very rarely. For example, 76 3 646 5 there were fifty votes on amendments or final texts Total Chi-Square 324 58 22.02(p<.001) 2405 173 88.57(p<.001) from the committee on Women’s Rights and Equal Opportunities, but none by roll call. This finding is clearly important for unde rstanding legislative behavior. Because some issue areas are to9 Resolutions are EP motions not directly associated with any piece of legislation, for example, general statements of position on some issue of the day. Consultation procedure votes are also largely symbolic. While the EP can slow down the le gislative process by delaying consideration of legislation, the EP can only issue non-binding opinions on the legislative proposals under this procedure. Under the Assent procedure, the EP can veto the motion under consideration. Under the Codecision procedure, the effect of an EP vote depends on whether it is a round I, II, or III vote. Codecision I is similar to the Contally unrepresented in the RCV sample, 1) RCV analysis necessarily will miss dimensions defined by those issues, and 2) if those issues happen to have unusually high or low levels of intra-party conflict, cohesion will be mischaracterized as well. And, clearly, this evidence is contrary to the assumption of Hix, Noury, and Roland(2002) regarding the distribution of votes across issue areas. sultation procedure in that the EP issues an opinion. However, if the EP and Council of Ministers do not reach an agreement, a second round of deliberation occurs. At that point, the EP Party Group Requests for Roll Call Votes can amend or reject the Council’s common position. If agreement is still not reached, then Codecision III begins with the bill being referred to a Conciliation Committee, which is Hix, Noury, and Roland(2002) also claimed that roll comprised of members of the Council and EP. If a comprocall votes are not disproportionately called by any one mise is reached in the form of a joint text, it is referred back to the EP and Council for a final vote. If an agreement cannot party group. While we could imagine defining“probe reached, the bill falls. portional” in a variety of ways, under any reasonable 5

Matthew Gabel& Clifford Carrubba The European Parliament and Transnational Political Conflict Europäische Politik (03/2004) 6 Table 2. Responsible committee by all votes and RCV definition the pattern of RCV requests is not propor(Legislative Final Votes and Amendments) tional. Furthermore, if you compare requesting patterns between final votes and amendments, different Committees All votes RCVs groups are using RCVs for different purposes. The PPE requested almost half of the RCVs on final votes, while the Verts/ALE, the TDI, and the ELDR requested the Agriculture and Rural Develop- 5.26% 5.99% bulk of the RCVs on amendments(see Table 3). The ment(AGRI) 212 65 PSE makes only very limited use of RCV requests, comBudgetary Control(CONT) Budgets(BUDG) 2.36% 95 10.79% 435 3.87% 42 6.82% 74 pared to its relative importance in the EP. These findings strongly reaffirm our concerns that the policy space may not be accurately characterized. Unless we assume the legislative agendas of all the party groups are identical, the characterization of the Citizens' Freedoms and Rights, 13.22% 16.04% Justice and Home Affairs(LIBE) Parliament's delegation to the Conciliation Committee 533 0.27% 11 174 0.09% 1 Table 3. RCV Requesting Groups on Final Votes and Amendments(% and raw numbers) Conference of Presidents Constitutional Affairs(AFCO) 0.20% 8 9.20% 0.00% 0 33.00% RCV Requesting Group Party Final Grou Votes p Size Amend Total ments Votes 371 358 Europe of De16 2.33% 6.81% 5.88% Culture, Youth, Education, the Media and Sport(CULT) 2.16% 87 0.00% 0 mocracies and Diversity 4 48 52 Development and Cooperation (DEVE) Economic and Monetary Affairs (ECON) Employment and Social Affairs (EMPL) Environment, Public Health and Consumer Policy(ENVI) Fisheries(PECH) Foreign Affairs, Human Rights, Common Security and Defence Industry, External Trade, Research and Energy(ITRE) Legal Affairs and the Internal Market(JURI) Regional Policy, Transport and Tourism(RETT) Women's Rights and Equal Opportunities(FEMM) Total 1.54% 62 5.93% 239 2.58% 104 19.02% 767 3.67% 148 4.74% 191 8.87% 358 4.61% 186 0.09% 1 14.84% 161 1.47% 16 0.55% 6 2.76% 30 2.76% 30 5.81% 63 3.50% 38 4.36% 1.24% 50 2.40% 0.00% 0 4033 1085 European Liberal 53 Democratic and Reform Parties Far Left 50 Party of Euro175 pean Socialists Technical Group 32 of Independent Members Union for a Euro 22 pe of Nations Greens 45 European Peo233 ples’ Party President MEPs Not Available Total 626 9.50% 17 11.77% 11.31% 83 100 5.59% 10 16.20% 29 3.35% 6 7.80% 55 9.79% 69 12.91% 91 7.35% 65 11.09% 98 10.97% 97 2.23% 4 11.73% 21 47.49% 85 1.12% 2 1.12% 2 0% 0 179 9.79.% 69 26.95% 190 9.93% 70 0% 0 2.41% 17 1.84% 13 705 8.26% 73 23.87% 211 17.53% 155 .23% 2 2.15% 19 1.47% 13 884

Internationale Politikanalyse International Policy Analysis Unit policy space almost certainly is being distorted. The EPP Committee, none of these votes were by roll call. Condisproportionately dominates the sample consisting sequently, it seems extremely unlikely that RCVs would only of final votes and small parties like the Greens and capture a cleavage among the MEPs characterized by TDI disproportionately dominate the RCVs on amend- this libertarian-traditional dimension. But, given that ments. the survey of MEPs revealed such a dimension, we might well expect that, had these 50 votes been reImplications for Existing Findings Regarding the Party System in the EP corded, an analysis of RCVs would reveal a different dimensionality of policy conflict. That is, a libertariantraditional issue domain may indeed characterize legislative policy conflict, but the selection bias in requestIn describing these characteristics of roll call votes, our ing roll-calls would hide it from view. Consequently, main priorities were to demonstrate the existence and our analysis provides an explanation for this aspect of severity of the sampling problem. These findings both incongruence in the findings reported in Thomsassen, provided systematic evidence that there is a major stra- Noury, and Voeten(2002). And, even more importegic component in the decision to request RCVs and tantly, this demonstrates how the observed RCV bias demonstrated that RCVs are likely to be misrepresent- by responsible committee could generate the general ing legislative behavior. In this section, we evaluate finding of one dimension of conflict in RCV studies. specific results in the EP literature in light of what we Third, the other major conclusion in the literature is learned from this RCV analysis. As we will demon- that intra-party cohesion is high, national divisions are strate, the results of our analysis completely undermine low, and inter-party cohesion is relatively low. There the central findings in the literature. are several reasons to doubt this conclusion. The first First, all three sets of findings are inconsistent with point is a recapitulation and extension of a point made the conditions Hix, Noury, and Roland(2002) assume previously. Simply put, most scholars are inte rested in in justifying the use of RCVs to study MEP voting be- studying RCVs because they want to understand how havior. RCVs are not used for the most important deci- the EP does or might influence legislative outcomes in sions, they are not called on the full range of issues, the EU. Answering this question requires examining and they are not called proportionately by party group. behavior on votes that actually have legislative conseThus, according to their own standards, their evidence quences. Without such consequences, MEPs could easof a“normal” EP party system is highly suspect. ily be engaging in cheap talk. And, by implication, Second, one of the two major conclusions regarding what we can learn from those votes is highly problemvoting behavior in the EP is that policy conflict in the atic. Thus, any study interested in explaining cohesion EP is predominantly over left-right issues. Our results and conflict in the EP necessarily should be examining indicate this is, at best, dubious. To see this, consider a votes MEPs care about casting. However, the analysis recent study by Thomassen, Noury, and Voeten(2002). by types of motion demonstrated that legislatively imThis study assessed the congruence between the di- portant votes are a very small proportion of the pool of mensionality of MEP attitudes regarding different poli- votes and are systematically selected out of the RCV cies, as reported in surveys, and MEP voting behavior, sample. Thus, conclusions over levels of intra-party coas indicated by RCVs. They found that MEP policy atti- hesion and inter-party conflict are being drawn from a tudes are structured by three issue domains: integra- particularly bad sample of data. tion-independence, socio-economic Left-Right, and lib- Moreover, the analysis shows that standard statistiertarian-traditional. In contrast, like previous RCV stud- cal analysis of RCVs cannot accurately evaluate coheies, they found only a single left-right dimension sion. To begin with, note that traditional measures of underlying MEP voting behavior. Why this difference in party cohesion(e.g., agreement scores) should be dimensionality? treated as sample statistics, because RCVs are only a Our analysis provides an explanation for part of this sample of legislative votes. But when we infer ge neral incongruence in findings. To see why, consider the fact characteristics of voting behavior from this sample, we that only the surveys found a libertarian-traditional di- should report some indication of the uncertainty of our mension. One of the policy issues that defined the estimates. By way of analogy, notice that public opin“libertarian-traditional” issue domain involved ion polls generally report confidence intervals around women’s rights—specifically, a woman’s fre edom to the estimate of, say, the popularity of the governing decide on abortion. In the 1999 EP, there were 50 party. votes on legislation for which the responsible commit- This is important be cause, we are unaware of any tee was the Women’s Rights and Equal Opportunity published studies that report party cohesion scores and 7

Matthew Gabel& Clifford Carrubba The European Parliament and Transnational Political Conflict Europäische Politik (03/2004) 8 report a standard error for the estimate. Lacking a hesive in voting behavior and competition is along trameasure of sample variability, we cannot assess with ditional left-right lines. In addition, we cannot conficonfidence the level of our inferences regarding party dently claim, as the existant literature has, that nationcohesion or make meaningful comparisons of cohesion ality is of little importance in structuring EP legislative across time or across PGs. behavior. Instead, what our evidence shows is that PG This problem would be easy to correct if RCVs were leaders are strategic in how they use roll calls. This may a random sample of legislative votes. All one would indicate some organizational sophistication, but it need to do is calculate a standard error for the esti- means that much of the important legislative voting mate of party cohesion and then re-interpret the find- occurs out of public view. This is bad for the study of ings accordingly. But the analysis indicates that, for EP legislative behavior and bad for normative concerns RCVs in the EP, this solution is unavailable because the regarding the responsible party model described earsample of votes is not random. This means that, even if lier. we were to calculate standard errors for the estimates Recall that RCV studies have indicated that PGs fit of party coheision, we have no reason to believe they the responsible party model in terms of legislative pa rare accurate. Indeed, they would likely be underesti- ties. This evidence has been used to support argumates of the uncertainty in the level of party cohesion. ments for institutional reforms to strengthen the EP so Thus, lacking any measure of uncertainty, we simply as to improve the link between voters and policycannot learn very much about the level of party cohe- makers in the EU. Our analysis indicates that such consion or its differences over time or across party groups clusions and arguments are suspect. We really do not from roll call votes. know very much empirically about how nationality or Finally, the analysis has important implications for transnational political interests affect legislative beha vwhether policy-making in the EP contributes to the ior in the EP. Under these conditions we cannot distintransparency and accountability of EU authority. As guish one good theoretical argument from another, discussed above, many journa lists and scholars identify and it seems reasonable that national interests could, the EP as a critical source of democratic legitimacy for at least in theory, be relevant to MEP behavior. the EU, at least in part because it allows public scrutiny The study also has an important implication regardof policy-making and it provides voters with an elec- ing the value of roll call votes as a tool to enhance detoral connection to policy-making. As in democracies mocratic accountability. Our results show that the segenerally, roll call votes are an important institution for lection of roll call votes by party leadership can unde rtransparency and accountability in the EP(Carey forth- mine this transparency. In the EP, we see that PG leadcoming-a, b). ers seldom request roll calls on the votes that are preThe results of our analysis suggest otherwise. Most sumably of greatest import to voters and instead often of EP votes are not available for scrutiny, thus denying request roll calls on the votes that have the least legisvoters the opportunity to monitor the behavior of their lative impact. Clearly, this does not enhance transparelected representatives. That is not to say that party ency. Indeed, it undermines it, by allowing party leadgroups obfuscate all legislative behavior—e.g., ers to hide legislators’ beha vior selectively from voters, speeches or committee reports. But party groups selec- which may be more insidious than holding only secret tively reveal – and hide- one of the most important votes. legislative acts, voting. Our results indicate that the se- As a final note, we support using RCVs to analyze lection of votes is far from random, thereby providing EP legislative behavior. We believe the next logical step voters with an atypical view of behavior. Obviously, the is to derive and incorporate an explicit model of RCV biased selection of votes for roll-call also limits the transparency of policy-making. 10 requests that will allow scholars to control for these biases when they analyze RCVs. Fortunately, previous studies provide valuable information about the motivaConclusion tions and context of roll call vote requests(e.g., Kre ppel 2002) that inform a model of roll call vote requests. The appropriate theoretical model of roll call vote reGiven these findings, we are far from confident that quest could then be used to estimate Heckman selecthe EP is a“normal” parliament, where parties are co- tion bias models of legislative voting behavior that rely upon RCV data. Such models would temper our infe r10 Of course, one might consider this sort of obfuscation a ne cessary evil, allowing party groups to maintain cohesion in the face of divisive pressures from varying constituent demands ences about party cohesion and the dimensionality of party conflict in light of the selection process. (Carey forthcoming-a).

Internationale Politikanalyse International Policy Analysis Unit References Attina, Fulvio. 1990.“The Voting Behaviour of the European Parliament Members and the Problem of the Europarties.” European Journal of Political Research 18: 557-79. Bardi, Luciano. 2002.“Transnational Trends: The Evolution of the European Party System.” In Bernard Steunenberg and Jacques Thomassen, eds. The European Parliament: Mov ing Toward Democracy in the EU . Oxford: Rowan and Littlefield. Blondel, J., Sinnott, R. and Svensson, P. 1998. People and Parliament in the European Union. Oxford: Oxford University Press. Bowler, Shawn and David Farrell. 1995.“The Organization of the European Parliament: Committees, Specialization, and Co-ordination.” British Journal of Political Science 25(2): 219-243. Brzinski, Joanne Bay. 1995.“Political Group Cohesion in the European Parliament, 1989-1994.” In Rhodes, Carolyn and Sonia Mazey(eds.). The State of the European Union . London: Lynne Reinner. Carey, John. Forthcoming-a.“Discipline, Accountability, and Legislative Voting in Latin America.” C omparative Politics . Carey, John. Forthcoming-b.“Transparency versus collective action: Fujimori’s legacy and the Peruvian Congress.” Comparative Political Studies . Carey, John. 2002.“Getting Their Way or Getting in the Way?: Presidents and Party Unity in Legislative Voting.” Paper presented at the annual meeting of the American Political Science Association, August 2002, Boston, MA. Carrubba, Clifford, Matthew Gabel, Lacey Murrah, Ryan Clough, Elizabeth Montgomery, and Rebecca Schambach. 2003.“Off the Record: Unrecorded Legislative Votes, Selection Bias, and Roll-Call Vote Analysis.” Working Paper. Emory University. Collie, Melissa. 1985.“Voting Behavior in Legislatures.” In Lowenberg, Gerhard, Samuel Patterson and Malcolm Jewell. Handbook of Legislative Research . Cambridge: Harvard University Press. Corbett, Richard, Francis Jacobs, and Michael Shackleton. 2000. The European Parliament, 4 th Edition . London: Catermill. Gabel, Matthew and Christopher Anderson. 2002.“The Structure of Citizen Attitudes and the European Political Space,” Comparative Political Studies 35(8): 893-913. Gabel, Matthew and Simon Hix. 2002.“Defining the EU Political Space: An Empirical Study of the European Elections Manifestos, 1979- 1999,” with Simon Hix. 2002. Comparative Political Studies 35(8): 934-964 . Harcourt, Alison. 1998.“EU Media Ownership Regulation: Conflict over the Definition of Alternatives.” Journal of Common Market Studies 36(3): 369-389. Hix, Simon. 2002a.“Parliamentary Behavior with Two Principals: Legislator Preferences, Parties and Voting in the European Parliament.” American Journal of Political Science 46(3): 688- 698. Hix, Simon, Abdul Noury, and Gerard Roland. 2002b.“A ‘Normal’ Parliament?: Party Cohesion and Competition in the European Parliament, 1979-2001.” EPRG Working Paper ,no 9 . Hix, Simon. 2001a.“Legislative Behaviour and Party Competition in the European Parliament: An Application of Nominate to the EU .” Journal of Common Market Studies . 39(4): 663- 88. Hix, Simon. 2001b.“Which Electoral System for the European Parliament? How Electoral Rules Shape MEP Voting Behaviour.” Presented at Conference on Democratic Inst itutions for a Large European Union. Florence, Italy. Hix, Simon. 1999. The Political System of the European Union . New York: St. Martin’s Press. Kreppel, Amie. 2002. The European Parliament and Supranational Party System . Cambridge: Cambridge University Press. Kreppel, Amie. 2000.“Rules, Ideology and Coalition Formation in the European Parliament: Past, Present and Future .” European Union Politics 1(3): 340- 362. Kreppel, Amie and George Tsebelis. 1999.“Coalition Formation in the European Parliament.” Comparative Political Studies : 32(8): 933-66. Marks, Gary and Marco Steenbergen, Editors. 2003. European Integration and Political Conflict . Cambridge: Cambridge University Press. Noury, Abdul. 2002.“Ideology, Nationality, and E uroParliamentarians.” European Union Politics 3(1): 33-58. Raunio, Tapio. 1999.“The Challenge of Diversity: Party Cohesion in the European Parliament.” In Shaun Bowler, David Farrell and Richard Katz, eds . Party Discipline and Parliamentary Government . Columbus: Ohio State University Press. Raunio, Tapio. 1997.“Aggregating Interests Within and Between Party Groups in the European Parliament: MEP Voting Behaviour, 1989-1994.” The European Perspective: The Transnational Party Groups in the 1989-1994 European Parliament. Aldershot: Ashgate. Raunio, Tapio. 1996. Party Group Behavior in the European Parliament: An Analysis of Transnational Political Groups in the 1989-94 Parliament . Acta Universitatis Tamperensis, Vammalan Kirjapaino, Vammala, Finland. Schmitt, Hermann and Jacques Thomassen. 1999. Political Representation and Legitimacy in the European Union . Oxford: Oxford University Press. Scully, Roger. 1997.“Policy Influence and Participation in the European Parliament.” Legislative Studies Quarterly 22(2): 233-252. Thomassen, Jacques, Abdul Noury, and Erik Voeten. 2002. “Political Competition in the European Parliament: Evidence from Roll Call and Survey Analyses.” In Marks, Gary and Marco Steenbergen, Eds. Dimensions of Contestation in the European Union . Cambridge: Cambridge University Press. Thomassen, Jacques. 2002.“Parties and Voters.” In Thoma ssen, Jacques and Bernard Steunenberg, Eds. The European Parliament . London: Rowan and Littlefield. 9