Brussels Bound

Abstract

Parties in list systems must select candidates to best accomplish their electoral, organizational, and policy goals. In particular, parties must balance nominees’ policy-making potential against other aspects of candidate quality, such as electoral viability. We exploit the unique variation in candidates and parties in European elections to study this trade-off. We develop a statistical ranking model to examine how parties facing varying strategic contexts construct electoral lists and apply it to a novel data set chronicling the political backgrounds of candidates in the 2009 European parliamentary elections. Parties that place high salience on the target legislature, are well positioned to influence policy once in office, and have less access to competing policy-making venues place particular emphasis on institution-specific policy-making experience relative to other types of candidate experience. This systematic variation in parties’ candidate nomination strategies may fundamentally alter legislative output and partisan policy influence.

Keywords

EU politics and policy political parties candidate selection

Parties channel political careers through nominations to stand for election (Best & Cotta, 2000; Gallagher, 1980; Norris, 1997; Rahat & Hazan, 2001). In turn, the identities, experiences, and qualifications of a party’s candidates condition not only its electoral success but also its ability to affect policy and governance once in office (Besley, 2005; Gallagher & Marsh, 1988; Ranney, 1981; Shomer, 2009). A party with visible, loyal, experienced, and knowledgeable candidates is likely to be successful at the polls and in government. However, not all available candidates have these desirable qualities. Parties must therefore manage their candidates as scarce resources during elections.

Parties most clearly have the opportunity to manage candidate type in list-based proportional representation systems (Gallagher, 1988). In closed-list systems, parties determine the order of candidates on the party list, and voters cast their ballot for the party, rather than a particular candidate. When lists are open, voters can use preference votes to alter candidate ranks, but the list provides a default ordering and sends the electorate signals about the relative value that the party’s electorate places on candidates (Carey & Shugart, 1995). Thus, list construction presents parties with an opportunity to behave strategically, balancing vote, office, and policy-seeking incentives (Strom, 1990). While pursuing these goals, party leaders may employ a variety of nomination strategies designed to maximize electoral competitiveness, reward party stalwarts, groom young talent, or ensure legislative policy-making ability. Moreover, leaders must often balance selection strategies across a variety of venues ranging from local council races to national elections, and sometimes, supranational competitions. Parties therefore need to ration their supply of experienced candidates when considering any one list. What determines which types of candidates parties nominate? How do party circumstances, both in the legislature at hand, and in other electoral venues, affect these decisions?

While party selectors must differentiate between potential candidates on a variety of dimensions, we focus on the fundamental choice of rewarding visible and well-connected party members and cultivating expertise within a particular legislative venue. The first strategy caters to vote and office-seeking goals whereas the second is likely to maximize policy influence. Where party competition focuses on the national parliament, long-standing national legislators will simultaneously stand near the pinnacle of the party hierarchy, be visible to voters, and wield the knowledge and institutional connections to get things done in office. But when parties must balance effectiveness in multiple policy-making venues, this connection can break down. Candidates with policy-making experience in a particular venue may not be the ones who possess the most visibility with the electorate, political connections within the party, or long-term potential for higher office. Similarly, non-linear career paths can undermine the linkage between partisan prestige and venue-specific policy-making experience in important—even national—parliaments (Samuels, 2003). Thus, parties often face a stark choice between cultivating institutional experience and rewarding intra-party stature, generalized political skill and clout, electoral visibility, or future potential. How parties manage this tension may determine whether a legislature effectively and efficiently meets evolving policy demands. In turn, parties that favor candidates with institutional knowledge and experience may exert more policy influence.

Despite the importance of list construction for party success and, indeed, the quality of democratic governance, political scientists have conducted few comparative studies of candidate nomination. We ask what factors lead parties to prioritize the maintenance of institutionally experienced delegations in legislatures and specifically investigate this question in European elections. European Parliament (EP) elections present a critical venue within which to explore the relative value that parties place on developing institutional experience within their parliamentary delegations. In European elections, voters in each of the 28 European Union (EU) member states elect national representatives to the EP. As of 2002, all member states must employ some form of proportional representation for these elections.¹ Parties in member states, therefore, present ranked lists of aspiring candidates to stand for election. Because EP elections generally take a back seat to national contests (see, for example, Follesdal & Hix, 2006; Moravcsik, 2002), and because jobs as members of the EP (MEPs) are almost universally regarded as less consequential than national parliamentary posts, European incumbents will rarely rival former national legislators in visibility or partisan clout. At the same time, because European institutions differ dramatically from the governing structures in member states and because they often deal with highly technical legislation (Ringe, 2010), parliamentary experience at the national level will often fail to seamlessly translate into efficacy within the EP. Thus, the relative value that parties place on incumbents and national politicians in EP elections can shed light on how important policy-making ability and institutional experience is to them when they select candidates.

To assess how parties select candidates, we construct a new data set of biographies of candidates in the 2009 European election. We identify the link between how parties rank-order candidates and their political contexts using a purpose-built statistical approach, explicitly modeling how parties construct lists based on the candidates at their disposal and the political environment that they face. We find that parties that emphasize the policies that a legislature controls, who have the ability to affect important legislation within the target parliament, and have fewer opportunities to place candidates in competing venues, place relatively higher priority on candidates’ policy-making potential.

Candidate Experience, Party Strategy, and List Placement

We conceptualize candidate nomination as a process in which parties determine the list rankings of a limited number of available candidates. Indeed, parties must portion the available candidates between a number of different elections at the local, regional, national, and European levels (Hix, Hobolt, & Høyland, 2012; Meserve, Pemstein, & Bernhard, 2009; Scarrow, 1997; Stolz, 2001). Broadly, parties seek three goals: votes, offices, and policy (Strom, 1990). Parties, however, do not have unlimited supplies of experienced, visible candidates (Norris, 1997), and cannot simultaneously maximize these three goals in all of the legislative venues in which they compete. Parties, therefore, must decide how to prioritize different types of candidates, depending on their electoral and legislative prospects and emphases, both in the target parliament and in competing venues.

At election time, vote and office-seeking incentives reinforce one another. Parties focused on these goals should seek to elect candidates who will maximize votes and thus control over seats in parliament. In the European context, this means choosing politicians who have held national parliamentary posts. These are often well-known public figures who are positioned to attract electoral support (Hobolt & Høyland, 2011). Conversely, policy-seeking parties should favor candidates with legislative influence, maximizing the likelihood that, after the election, the party can use its representatives to influence policy and manage public affairs effectively. Incumbency is a key predictor of policy-making ability. Candidates with previous experience in the institution understand how to navigate and manipulate the policy process and have had the opportunity to develop venue-specific policy expertise. Personal networks also play a key role in legislative effectiveness (Fowler, 2006). Incumbents have formed relationships, become embedded in the committee structure of the parliament, and have accrued seniority in the institution. Finally, parties may also make room on their lists for less accomplished candidates, including regional or local politicians, or even outright novices. This may give young talent opportunities to demonstrate their political acumen, or reward long-serving, but only moderately successful, party members. The latter strategy serves the office-seeking motivations of party loyalists whereas the former develops the party’s potential to meet broad vote, office, and policy goals in the long term.

In many legislative elections, it is difficult—if not impossible—to distinctly identify the value that parties place on institutional experience because incumbents will also represent the cream of the party’s crop. That is, when examining national legislatures, it is impossible to know if parties renominate incumbents because of their institutional experience or because of their status as party insiders or well-known politicians. Thus, party nomination strategy typically yields little insight into how parties weigh vote and office-seeking incentives against policy-making potential. Our focus on European elections allows us to overcome this issue. European elections are second order (Marsh, 1998; Reif & Schmitt, 1980; Schmitt, 2005). Thus, even in European elections, it is nationally experienced politicians, not European incumbents, who are likely to be most highly regarded within the party and best positioned for electoral success. However, legislative experience is not necessarily transferable, and a background in the national legislature may not prepare a politician to be effective in the EP. The EP is an institution quite unlike most national legislatures in Europe. There is no government or opposition and rank-and-file legislators in the EP have the potential to play a more pivotal role in the legislative process than do members of many national parliaments in Europe. Committees matter, rapporteur assignments provide MEPs with substantial influence over policy outcomes, many of the relevant policy domains are highly technical, and the EP heavily amends much of the legislation that it considers. Indeed, the keys to legislative influence in the EP reside in access to rapporteurships, committee leadership positions, and party group coordinator roles (Bowler & Farrell, 1995; H. Farrell & Heritier, 2004; Ringe, 2010; Yoshinaka, McElroy, & Bowler, 2010). Tellingly, McElroy (2006) has found a seniority norm in EP committee assignments. Nationally experienced politicians, while potentially strong legislators, will often find it hard to craft European policy out of the gate, simply because their lack of seniority and under-developed networks are likely to impede their policy influence. In addition, many non-incumbents, even old party hands, will have little experience maneuvering in a supranational policy space and may have few established connections to lobbyists in Brussels or power brokers in the European Commission and Council. Therefore, parties that are primarily interested in influencing the policy process should value incumbency more than those that view the EP as a retirement venue, a well-paying job for proven party stalwarts, or a good place for inexperienced politicians to get their feet wet. This incentive structure allows us to observe variation in how parties weigh vote and office-seeking concerns against policy-making effectiveness.

Incumbency and European Policy-Making Experience

Our theoretical argument is predicated on the existence of a significant policy-making advantage for national parties that use MEP incumbents rather than non-incumbents. We argue that European policy-making knowledge and connections are largely non-transferable. While candidates with experience in other parliaments are likely to legislatively outperform truly new members, incumbents should sport a distinct advantage. To assess this claim, we compare the legislative activities of MEPs who were elected as incumbents and those MEPs who lacked experience in the EP (non-incumbents). We further divide the non-incumbent category into MEPs who had national-level legislative experience and those that did not. The evidence shows that incumbents perform more critical, and technically demanding, legislative activities than non-incumbents, even non-incumbents with previous national experience.

We split MEP activities into two categories: technical and non-technical. If our assumption has merit, incumbent MEPs should be more active in technically demanding activities. We tallied the number of codecision reports, non-codecision reports, and opinions reported by MEPs on behalf of an EP committee. These activities represent the most substantively important and technically demanding work an MEP can perform.² To measure participation in less technical areas, we code MEPs’ speeches in plenary, motions for resolution (motions to a vote, generally for non-legislative statements about human rights, democracy, or rule of law), declarations (non-binding statements by MEPs), and questions for the European Commission.

The descriptive statistics in Table 1 highlight the difference between incumbent and non-incumbent activity. For any activity requiring formal institutional action, irrespective of the technical/non-technical split, incumbents tend to be more active. For the most uncommon and technical activities—codecision reports and other rapporteurships—incumbents are far more likely to participate. Incumbents even tend to produce non-binding declarations or motions for resolution more frequently. Only for the routine, low difficulty activities—speeches in plenary or questions to the commission—do non-incumbents engage more often.³ This pattern is consistent with our assumption and highlights the importance of incumbents in the EP for any party that wishes to shape and pass legislation.

Table 1.

Average Activity by MEP Type.

Activity	Incumbent	Non-incumbent	National experience only
More technical activities
Codecision reports	0.70	0.41	0.70
Other reports	2.32	1.70	1.67
Opinions	2.05	1.93	1.77
Less technical activities
Motions	30.40	24.98	16.09
Declarations	1.24	1.00	0.84
Speech	129.41	139.88	105.33
Questions	67.17	79.16	97.70
n	375	467	43

MEP = member of the European Parliament.

The differences remain even after considering outliers. Figure 1 shows box plots of the data in Table 1. For less difficult activities, medians and quartiles are similar across incumbents and novice MEPs, though incumbents are slightly more active in declarations and motions than non-incumbents. For technically demanding activities, incumbent MEPs participate more than non-incumbents. Incumbents handle more reports, both codecision and other. The box plots show that this is not simply a story about outliers: notably, while the average production of codecision reports for non-incumbents is 60% of incumbent activity, this non-incumbent mean is actually inflated by a handful of outliers. Despite being a technical and relatively rare form of participation, opinion production appears to occur at similar rates between incumbents and non-incumbents. This could perhaps reflect the fact that the final opinion represents an entire committee and may be created with the assistance of more experienced committee members. Moreover, many opinions reflect non-binding reports that are less demanding, and valuable, than reports considered under legislative procedures.

Figure 1.

Box plots of frequency of activity.

As a further test, we performed negative binomial regressions on the most important MEP activities: codecision and other reports. Even including fixed effects for European group membership, to account for differences in agenda control and access to reports, and country specific effects, to account for proportionality norms, incumbency remains a positive and statistically significant predictor of MEP (codecision) report production.⁴

Finally, since we argue that policy-making abilities developed in national parliaments are unlikely to overcome incumbents’ policy advantages within the EP, we compared parliamentary activity by incumbents and the 43 non-incumbent MEPs in our 12-country subsample who had previously sat in national parliament.⁵ If national parliament prepares individuals to act effectively as MEPs, we should observe similar patterns of activity across these two types. The final column in Table 1 demonstrates that national-level domestic experience does not appear to prepare individuals to hit the ground running as policy makers in the EP. In virtually every measure, the amount of activity for non-incumbent national parliamentarians is well below that of incumbents. The only important exception is codecision reports, where the sample means are similar across the two groups. This lack of difference, however, is driven almost entirely by an outlier among the nationally experienced non-incumbents. Italy’s Paolo de Castro, a previous adviser of former European Commission head Romano Prodi, and a non-incumbent with substantial European policy-making expertise, produced a whopping 10 codecision reports.⁶ Removing de Castro from the sample reduces the average report production among non-incumbents with national experience to 0.48, very close to the non-incumbent mean.

Together, these data show that incumbents are more likely, on average, to perform technically demanding work in the EP. Furthermore, the data suggest that national service cannot substitute for European experience when it comes to preparing politicians to be productive in the EP. Our assumption, therefore, seems tenable: prioritizing incumbency in list orderings should be an indicator of a party’s emphasis on achieving policy goals.

The Determinants of Party Policy-Making Preferences

European elections provide a unique opportunity to observe the relative emphasis that parties place on policy-making experience and other candidate characteristics—such as public exposure, loyalty, personal support, party influence, and campaigning skill—when constructing lists. The divorce between candidate policy-making potential and other important candidate qualities, created by the unique challenges of legislating in the EP, makes it possible to shed light on factors that affect the importance of policy making to parties. Here, we lay out a general theory that describe how parties that compete in multiple venues will prioritize different types of candidate experience in a given legislature and link our theoretical arguments to observable aspects of European parties. Throughout this section, we identify direct analogs between EP elections and domestic settings.

Institutional Emphasis

Parties that care about policy will naturally desire to maximize their influence over the process that produces it. Thus, they will want to nominate candidates with the ability to affect policy once in the legislature, prioritizing those candidates with the experience to shepherd proposals through the process. Consequently, we argue that the level of emphasis a party places on the EU, independent of its attitude toward the institution, affects its likelihood of sending effective policy makers to Europe. Parties that care about European-level policy want to place their most capable policy makers in the EP to pursue their interests. Based on the evidence introduced earlier, parties can expect MEPs with seniority as incumbents to do a large amount of substantive, technical EP policy making. As a result, parties that are highly invested in the EU will tend to rank EP incumbents higher than both high-quality candidates with national experience and low-quality, inexperienced candidates.

Hypothesis 1: Parties that emphasize Europe in their policy programs will prioritize incumbents over other types of candidates.

Critically, this hypothesis challenges the common view that parties do not take the EP seriously because they nominate incumbent MEPs who are relatively unknown and invisible to the electorate. Instead, sticking with European careerists is a prime indicator that a party is invested in European institutions. Contrary to what one would anticipate in most other electoral venues, the parties that care about policy outcomes do not necessarily rank their most well-connected party insiders, with high profile national experience, highest on lists.

But note that this EP-specific argument has clear analogs in national party politics. In multi-tiered systems, parties place varying emphasis on different institutions in which they contest elections. For example, the literature on party system regionalization finds that systems with many small regional parties may not prioritize effective public policy production at the national level, instead focusing on regional governance (Caramani, 2004; Crisp, Olivella, & Potter, 2013). We argue that this variance in institutional emphasis will affect who the parties nominate to represent them in particular legislatures; parties pursue policy more vigorously in institutions that they find most salient and nominate accordingly.

Potential for Productivity

A party’s preference for maximizing the policy-making capabilities of its delegation is also modulated by its expectation that those legislators will have the ability to use those skills once in parliament. In the context of domestic politics, parties that anticipate being in government—or holding other influential positions within parliament—face incentives to prioritize candidates who can contribute to policy making once they take power, simply because they can leverage those skills. Parties with little hope of influencing policy in a given institutions are unlikely to adopt nomination strategies that focus on cultivating policy-making experience.

There is no formal governing coalition in the EP. Instead, prior to the 2009 election, the institution was dominated by the three largest party groups: the European People’s Party (EPP), Progressive Alliance of Socialists and Democrats (S&D), and the Alliance of Liberals and Democrats for Europe (ALDE). These groups held the great majority of seats in the EP, shared party affiliation with many government ministers on the Council, and were well represented in the Commission. Their members held the highest committee positions, generated many of the most sought-after reports, and were generally well situated to influence policy. Parties from other European groups, by contrast, often find themselves in the role of the opposition and must focus more on position-taking than policy making. As a consequence, we expect that national parties belonging to one of the three largest groups rank candidates with maximal policy-making advantages—incumbent MEPs—higher on their lists.

Hypothesis 2: Parties belonging to the EPP, S&D, and ALDE will prioritize incumbents over other types of candidates.

In practice, the implication of this strategy for the composition of the EP is clear. Parties in the dominant groups will rank careerist EP incumbents in high places, relying on them to manage European legislation. Meanwhile, parties in outsider groups have less reason to rank incumbent candidates who can make effective policies. With little hope of pursuing policy goals, they should focus on vote and office-seeking, grooming talent, and rewarding party stalwarts.

Inter-Institutional Coordination

Tiered or multi-chambered legislative systems, like federal or bicameral systems, with multiple policy-making venues, may also create incentives for parties to coordinate across tiers. When a party has influence in one tier, and must also develop clout in other tiers to pass and implement its policies, it has incentives to place experienced policy makers in those other tiers to coordinate policies effectively. The EU is a multi-chambered legislative institution, requiring the cooperation of the Council and the Parliament to pass policies into law. It also oversees laws that affect individuals across Europe and may generate consequences for parties competing on the national stage and thus resembles a federation.

The Council is composed of national ministers and therefore gives national-level parties in government substantial policy influence at the European level. Council members, however, need to coordinate with the EP, which possesses substantial influence over most binding European legislation. Moreover, parties in government must rely on the EU to produce policies for which they may be rewarded or blamed in national elections. Parties in government therefore desire an EP delegation with competent policy makers to facilitate effective coordination and to promote policy that will play well to a national audience. We argue that a party’s roles, both as a member of the Council and of national-level government, alter the relative value that it places on policy experience in the EP and, therefore, its propensity to rank incumbents higher on candidate lists.

Hypothesis 3: Parties belonging to a domestic government will prioritize incumbents over other types of candidates.

Parties should favor policy-oriented candidates when they must rely on the target institution to produce policy that affects their effectiveness in other domains. This dynamic may represent pure policy-seeking by parties or a balancing of goals. In the bicameral context, parties should be more inclined to pursue policy in a given legislature if they have the necessary resources to influence the process in the other house. Under federalism, policy making in one venue may also facilitate vote or office-seeking in another. Voters may, for example, reward or punish national parties for policies that were largely built at the local or regional levels. Thus, politicians in government may prioritize policy-making acumen regionally to maintain hold over national offices. The EP context closely mirrors the sorts of coordination problems that parties that compete in multiple national venues face at home. It provides a good test for the proposition that a party’s status in one policy-making institution can affect its nomination strategies in another. Of course, we see both bicameral and federal pressures in the EU; thus, we cannot tease out the relative influence of these two mechanisms here.

Venue Options and Career Management

Parties must balance immediate policy goals and a longer-term need to manage the careers of their politicians. In particular, parties that have more opportunities to place candidates in different venues may be more likely to balance different goals through their nomination strategies. In this context, large national parties may appear to place less emphasis on the policy-making ability of their nominees to the EP than smaller parties. Although nationally successful parties may value policy making in the EP similarly to smaller parties, they also must manage a greater number of candidates and are likely to face more competing pressures than their counterparts. Thus, they may use European elections to pursue a variety of strategies, such as grooming up-and-comers, storing candidates during electoral downturns, or providing a pleasant retirement venue for party stalwarts (Daniel, 2013; Meserve et al., 2009). Smaller parties have less pressure to use their party lists for these alternative goals. For example, why would nationally unsuccessful parties groom young candidates who have nowhere to go? Similarly, nationally small parties have few politicians who need temporary housing during periods of national misfortune. These competing motivations could alter the relative emphasis that parties place on policy making.

We therefore expect that parties that control a large proportion of seats in their national legislature will exhibit relatively less bias toward incumbents because they are using the EP lists for numerous purposes simultaneously. Smaller parties, meanwhile, will be more likely to maintain a policy-making focus because they are not juggling other ranking strategies.

Hypothesis 4: Parties with a smaller share of national legislative seats will prioritize incumbents.

In general, parties with limited electoral options in other venues will face fewer trade-offs in pursuing their goals through nomination strategies. When you have limited success in other institutions, you will maximize votes, obtain the most offices, and influence policy most effectively by investing in candidates with previous success in the current institution.

Attitudes Toward Europe

Finally, it seems sensible to account for the possibility of ideological differences in strategies of candidate rankings. Individual parties take policy positions with respect to the EU. Some parties take strongly euroskeptic positions, opposing European expansion whereas other parties push for deeper integration and stronger powers for supranational institutions. Parties outline their policy positions vis-à-vis the EU during elections, on the campaign trail, and in manifestos. In general, we have weak expectations about differences in candidate ranking priorities or party emphasis on policy making stemming from euroskeptic or pro-integration views. Euroskeptics may have just as much interest in nominating capable MEPs as pro-European parties, if only to more effectively undermine European institutions. Although we include this variable mostly as a control, we have weak expectations that pro-Europe parties will emphasize policy-making ability in the EP more than euroskeptics.

Hypothesis 5: Pro-EU parties will prioritize incumbents when nominating EP candidates

Note that there is a fundamental distinction between approving of an institution and emphasizing it in your policy program; anti-system parties may, in general, find existing institutions highly salient venues for pursuing policy goals. Including attitudes in our analysis allows us to parse out how these related—but different—aspects of a party’s approach toward Europe correspond with their nomination strategies. This variable helps us distinguish the correlated, but differential, effects of emphasis and attitude.

The EP Biographies Project

Evaluating list placement strategies requires extensive data on the characteristics of candidates. Most existing studies of the EP focus on elected MEPs rather than candidates (Daniel, 2013; Hix et al., 2012; Patzelt, 1999) or examine a single country (Gherghina & Chiru, 2009; Hrbek & Schweitzer, 1989; Kauppi, 1996; Linek & Outly, 2006; Westlake, 1994).⁷ Attempts to survey candidates have had low response rates (D. M. Farrell, Hix, Johnson, & Scully, 2006; Norris & Franklin, 1997). Even the largest candidate survey project covering European elections, the European Parliament Candidate Election Study (EECS), has response rates ranging between 0% and 40% of each party’s candidates (Giebler, Haus, & Wessels, 2010), and few parties are represented by more than one or two respondents. Thus, although these surveys generate reasonable pictures of candidate characteristics on average, they provide limited traction for questions about how candidate selection strategies vary across parties.

Therefore, we collected original data describing EP candidate characteristics and experiences. In the months before the 2009 EP election, we gathered native language candidate information from Internet sources for virtually all national parties predicted by Hix, Marsh, and Vivyan (2009) to receive a single seat in the 2009 EP election.⁸ We drew the bulk of the biographical information from official party websites, but also used other sources, including blogs, personal websites, media reports, and governments’ candidate rolls. The biographical materials ranged from short descriptions to full CVs.

Fluent language speakers then coded information about salient political positions, career histories, and demographic variables for each candidate. Coding biographies is extremely time and resource intensive, so we focused those efforts on a sample of EU countries including both the largest, most powerful countries, and representatives of the major regions of Europe save Scandinavia. The sample includes 3,092 candidates from 71 national parties in 12 countries: Bulgaria, Czech Republic, France, Germany, Greece, Hungary, Ireland, Italy, the Netherlands, Romania, Spain, and the United Kingdom (see Table 2).⁹ Although we include descriptive statistics about Ireland here for illustrative purposes, it is not included in any of our statistical models because it uses single transferable vote (STV) for EP elections.¹⁰

Table 2.

Number of Candidates per Country.

Country	Number of candidates coded
Bulgaria	51 (2%)
Czech Republic	188 (6%)
France	991 (32%)
Germany	199 (6%)
Greece	132 (4%)
Hungary	78 (3%)
Ireland (excluded from analysis)	30 (1%)
Italy	428 (14%)
Netherlands	190 (6%)
Romania	153 (5%)
Spain	300 (10%)
The United Kingdom	352 (11%)
Total	3,092

We distinguish between viable and non-viable candidates. For a candidate to be classified as viable, she must be placed no more than four positions below the lowest successfully elected candidate on national lists.¹¹ This cutoff serves as proxy for whether or not a candidate’s list position was “in the ballpark” of electability. In our subsequent analysis, we focus on explaining how parties select and order their most viable candidates from all of the candidates included in the list. Thus, although our models consider every candidate listed, they focus on ordering decisions at the top of the list. This approach captures the idea that differences in rank-order at the top of the list are substantially more important than ordering decisions lower down the list. Thus, we break down candidate characteristics by viability in the descriptive statistics that we present here.

We recorded all elected political positions mentioned in the candidate biographies. We differentiated between local, regional, and national positions. Although 43% of candidates had held some type of elected position (Table 3), 69% of viable candidates had been elected to office. Individuals near the top of their lists frequently had multiple elected positions in their biographies. Yet, there were significant cross-national differences. In Greece, Spain, and the Netherlands, only about one third of candidates reported elected experience while in Ireland, the United Kingdom, and France, experience in an elected office was far more common.

Table 3.

Candidates With Elected Experience.

	All candidates		Highly viable candidates
Country	Elected experience	No experience	Elected experience	No experience
Bulgaria	14 (27%)	37 (73%)	10 (42%)	14 (58%)
Czech Republic	62 (33%)	126 (67%)	32 (64%)	18 (36%)
France	422 (43%)	569 (57%)	86 (70%)	37 (30%)
Germany	128 (64%)	71 (36%)	105 (81%)	25 (19%)
Greece	34 (26%)	98 (74%)	22 (48%)	24 (52%)
Hungary	47 (60%)	31 (40%)	21 (66%)	11 (34%)
Ireland	25 (83%)	5 (17%)	25 (83%)	5 (17%)
Italy	217 (51%)	211 (49%)	89 (87%)	13 (13%)
Netherlands	62 (33%)	128 (67%)	27 (44%)	34 (56%)
Romania	59 (39%)	94 (61%)	33 (72%)	13 (28%)
Spain	76 (25%)	224 (75%)	44 (59%)	30 (41%)
The United Kingdom	176 (50%)	176 (50%)	89 (70%)	38 (30%)
Total	1,322 (43%)	1,770 (57%)	583 (70%)	262 (30%)

Other Data

We supplemented our biographical data with information from a variety of sources, all measured and varying at the level of the national party. First, we measured national party size at home (% Seats), and party governing status (In Government) from the European Journal of Political Research. We coded EP group identity (Big 3 Group) from the time of the election in 2009. We also made extensive use of the PIREDEU group’s 2009 European Election Studies (EES; 2010), drawing our measures of support for, and emphasis on, European institutions from PIREDEU’s EP election manifesto study. These manifestos are unique EP election manifestos drafted by national parties, as European groups do not campaign on single manifestos.¹² We operationalized EU attitudes (Pro/Anti EU) in terms of the proportion of sentence fragments in the party’s manifesto that the PIREDEU coders classified as pro-integration minus the proportion of sentence fragments that they coded as integration-sceptic.¹³ We measure a party’s emphasis on Europe (EU Emphasis) in terms of the total proportion of sentence fragments in their manifestos classified as pertaining to EU institutions. Table 4 summarizes the independent variables in our analysis. The supplementary appendix (at www.sneps.net/cps1supplemental.htm) contains summary, and covariance, statistics for these variables, and a discussion of collinearity.

Table 4.

Determinants of Candidate Ranking on EP Lists.

Hypothesis	Operationalization	Variable name
Institutional emphasis	Proportion of manifesto sentence fragments mentioning EU	EU emphasis
Potential for productivity	Big 3 EU group dummy	Big 3 group
Inter-institutional coordination	National governing status dummy	In government
Venue options and career management	Proportion of seats held in national lower house	% seats
Pro/anti EU position	Manifesto pro-integration sentence fragment proportion minus anti-integration proportion	Pro/anti EU

Modeling List Construction

Our goal is to model the decisions that parties make when ranking candidates on their EP lists. Parties may rely on more or less centralized selectorates and a variety of decision-making processes when developing lists (Rahat & Hazan, 2001). Thus, while our modeling framework treats parties largely as unitary actors, we emphasize that this approach represents a deliberate over-simplification of the true process. This model pays substantial dividends by focusing our attention on how broad party circumstances translate into nomination behavior. In some parties, this process may represent explicit strategy by a small leadership, in others, the result of a more complex choice process.¹⁴

We model which candidates parties selected for viable—as defined in the previous section—list spots, and how parties rank candidates within these viable positions. There is a set I = {1, 2,…, n} of available candidates across all parties, with each potential nominee indexed by i ∈ I. For simplicity, we assume that parties make selections in list order—that they choose the candidate heading the party list first, and so on. Furthermore, we assume that a function, $f (Θ_{t}^{p}, Ψ_{t}^{p}, x_{p}, i) = \Pr (i_{p t} = i)$ , probabilistically determines party p’s choice of the candidate at list position t, where $Θ_{t}^{p} \subset I$ is the set of candidates on party list p after choice t − 1, $Ψ_{t}^{p} \subset I$ is the set of party p’s potential candidates at choice t, x_p is a vector of covariates describing party p, and i_pt ∈ I is the candidate that party p selects for list position t.¹⁵

Each element of $Ψ_{0}^{p}$ , the party’s pool of available candidates, is associated with a K-vector, γ_i, representing candidate i’s membership in each of K ideal types, or groups. In this article, we group potential viable candidates in terms of their previous elected experience at the local/regional and national levels, and their incumbency status in the 2009 European election.¹⁶ Specifically, available candidates could hold membership in either the no-experience category or in some combination of the local/regional, national, and incumbent categories. So, we represented every candidate with no record of previous elected experience by the vector γ_i = (1, 0, 0, 0), while an incumbent with previous elected experience at both the local/regional and national levels obtained the coding γ_i = (0, 1, 1, 1), and so on. Thus, multiple group membership is possible, except for candidates with no prior elected experience. This is an important modeling decision with respect to our theory, as it allows for the possibility that incumbents are chosen based on previous national, or other, electoral experience.

In general, party p’s choice of nominee for list place t might depend both on the characteristics of the remaining available candidates, $Ψ_{t}^{p}$ , and those of the members already on the list at point t, $Θ_{t}^{p}$ . For example, parties might wish to balance the experience composition of their lists. Nonetheless, in this work, we make the simplifying assumption that parties’ selections are independent of the choices that they have already made and that they consider only their remaining available candidates when making list selections (i.e., $f (Θ_{t}^{p}, Ψ_{t}^{p}, x_{p}, i) = f (Ψ_{t}^{p}, x_{p}, i)$ ). Building on standard statistical models of choice, we assume that

f (Ψ_{t}^{p}, x_{p}, i | β) = {\begin{array}{l} 0 & if i \notin Ψ_{t}^{p} \\ \frac{\sum_{k = 1}^{K} (γ_{i k} \cdot e^{x_{p} β_{k}})}{\sum_{j \in Ψ_{t}^{p}} [\sum_{k = 1}^{K} (γ_{j k} \cdot e^{x_{p} β_{k}})]} & otherwise . \end{array}

Equation 1 implies that parties make nomination choices in terms of how much affinity they feel toward candidates of each ideal type and that affinity compounds additively for individuals that belong to multiple groups. Parties’ characteristics determine their preferences, and, in particular, each β_k is a vector of m coefficients that captures the extent to which parties value candidates representing group k, as a function of party characteristics x_p. We represent a party’s overall bias toward a potential nominee in terms of the sum of the party’s affinity toward each of the K types, weighted by the potential candidate’s degree of membership—described by γ_i—in each group. The probability that party p selects candidate i for list position p is simply this bias divided by the party’s overall affinity toward the candidate pool that remains at choice t.

Note that this model is a generalization of multinomial logit (see, for example, Long, 1997). Indeed, if, at every time t, every $Ψ_{t}^{p}$ contains K candidates, each of which is a full member of just one of the K candidate groups, and no two members of $Ψ_{t}^{p}$ belong to the same group, Equation 1 simplifies to the functional form assumed by multinomial logit. Therefore, one can interpret the coefficient matrix β in the model that we present here similarly to coefficients in a multinomial logit; specifically, they capture the relative affinity that parties sporting a particular set of characteristics have for full representatives of each of the K candidate groups, given the counterfactual situation in which party p has the opportunity to select a single candidate from a full set of ideal types.¹⁷ This characteristic of the estimator is crucial to understanding how we use the model to examine European nomination behavior. In particular, we do not directly address questions of candidate supply in this article. Rather, we ask, “Who is at the top of the list, and why, given the menu of candidates available to the party?” Thus, for example, the fact that small parties are less likely to have incumbents does not reduce our estimate of the value that such parties place on experienced candidates. The lack of pertinent observations makes our estimates less certain, but the model can capture the fact that small parties value the incumbents that they do have very highly. Put another way, the quantity of interest here is not who, on aggregate, different types of parties place into European office. Rather, we ask who they would prioritize were they given the chance to choose their ideal type. Ideally, we would predict parties’ list placements in terms of the universe of potential candidates available to them. Unfortunately, this set is generally impossible to observe and even harder to collect biographical data about. Therefore, we analyze how parties select viable candidates—or how they fill the places at the top of their lists that have some probability of providing nominees with seats in the legislature—from the set of candidates that appear on the entire list. Our results, therefore, speak to how parties prioritize among candidates that are willing to appear on the list.

We model the selection and ordering of n_p top list positions from N_p total list spots for each party, p.¹⁸ In so doing, we assume that the universe of potential nominees to top list positions, $Ψ_{0}^{p}$ , is captured by each party’s full list.¹⁹ Combined with Equation 1, this strategy leads to the observed data likelihood

\prod_{p \in P} \prod_{t = 1}^{n_{p}} \frac{\sum_{k = 1}^{K} (γ_{c (p, t) k} \cdot e^{x_{p} β_{k}})}{\sum_{j \in Ψ_{t}^{p}} [\sum_{k = 1}^{K} (γ_{j k} \cdot e^{x_{p} β_{k}})]},

where c(p, t) is a function mapping party p’s nominee at list position t into I. Note that this likelihood makes two key modeling assumptions explicit. First, as we mention above, we assume that parties make their viable list placement decisions in order and that each choice is independent of previous list placements. Second, each party’s nominations are strategically independent of other parties’ decisions. That is, we model list construction as decision-theoretic, rather than game-theoretic. Both of these assumptions are restrictive. Nonetheless, we think that they represent a reasonable foundation for systematic investigation of the determinants of party list construction.

We estimated the model using a Bayesian approach and adopted diffuse normal priors on the coefficients, β. Specifically, after making the identifying restriction that the first row of the parameter matrix β₁ = 0, we assumed that each β₂, β₃,. . . , β_K ~ N_m(0, 25 · I_m), a priori. The supplemental appendix provides a detailed description of the sampling algorithm that we used to fit the model. We used a basic Metropolis–Hastings algorithm to fit the model and implemented the sampler using the Scythe Statistical Library (Pemstein, Quinn & Martin 2011) and the R Language for statistical computing (R Development Core Team, 2010).

Results and Discussion

We applied this model to the European Biographies data, including the independent variables mentioned in Table 4 in the embedded regression equation.²⁰ We also included country dummies in our specification to control for unobserved cross-national variation in appetites for different types of candidates. These controls are important for a variety of reasons but serve primarily to account for differences in European electoral procedures across states. In particular, fixed effects control for any possible differences between open- and closed-list systems in nomination patterns.²¹ Moreover, the inclusion of country fixed effects means that our results are driven by inter-party differences within countries rather than cross-country differences in party characteristics, as the fixed effects fully account for the latter form of variation.

Figure 2 displays the model coefficient estimates graphically, suppressing the country dummy coefficient estimates for readability.²² The model generates a set of coefficient estimates for each of the K experience-based groups of candidates. Figure 2, which can be read like a regression plot for a multinomial logit model, uses pure incumbents as the reference category because our foremost interest is in examining when parties prioritize policy making in their list creation strategies by preferring incumbents to other candidate types. We mark coefficients for nationally experienced candidates, candidates who have attained either regional or local offices, and inexperienced nominees, with N, R, and 0, respectively. These coefficients represent the marginal relative tendency for parties to rank members of each of these candidate groups above an incumbent, given an idealized choice between representatives of each ideal type, in terms of party characteristics. We plot three such coefficients for each independent variable that we included in the analysis. Vertical lines represent 95% highest posterior density (HPD) regions. In general, the average effects displayed in Figure 2 correspond to our expectations and are consistent with the idea that parties that have good reasons to value policy making in the EP prioritize incumbents more than other parties.

Figure 2.

Regression coefficients (base category: incumbent).

First, Figure 2 shows that, on average, parties become more biased toward incumbents as the emphasis that they place on the EU in their programmatic documents grows. Indeed, all three coefficients for this variable are negative, and their 95% HPD regions do not cover zero, indicating a consistent increasing preference for incumbents over all three alternative candidate types. Moreover, this effect is substantively overwhelming. Figure 3 lays plain the fact that parties that place substantial emphasis on the EU in their platforms display a dramatic preference for incumbents. This figure plots predicted decision probabilities for an average—with respect to size and ideology—out-of-government German party that is a member of a big 3 group, facing a choice between four ideal types. That is, it must choose between a non-incumbent, a candidate with no higher than local or regional experience, a candidate with only national experience, or an incumbent with no non-European background.²³ The x-axis of Figure 3 spans the within-sample variation in the salience that parties place on Europe in their manifestos and vertical lines represent 95% HPD intervals around the predicted probabilities. Parties that are especially uninterested in the venue are just as likely to choose national, and in extreme cases, local or regional candidates, as they are incumbents. Yet, as EU emphasis increases, incumbents quickly become the dominant option. Indeed, the probability of choosing an incumbent rises from around one half for EU-indifferent parties to nearly one among parties that place extremely high value on the venue. For parties that really take EU policy seriously, incumbents are, by far, the first choice. These results provide resounding support for Hypothesis 1. Generally, given the fact that incumbents have distinct advantages in policy production within the EP, this behavior is consistent with the idea that parties that especially care about a particular institution will use nomination strategies to develop policy influence within a given legislature.

Figure 3.

Party emphasis on electoral venue.

Turning to Hypothesis 2, we find that parties that belonged to the three dominant groups in the EP prior to the 2009 election did tend to select incumbents at higher rates than other parties. All three coefficients in Figure 2 fall below zero, and the 95% HPD regions for the novices and candidates with regional experience exclude zero. Furthermore, the model estimates that the probability that the coefficient for nationally experienced candidates is negative is greater than 0.9.²⁴ Thus, it is probable that, looking across candidate types, parties within the big 3 groups place a higher emphasis on incumbency than do parties hailing from smaller groups. In terms of predicted probabilities, Figure 4 shows that, at standard levels of statistical confidence, the predicted probability that such a party will choose an incumbent over a national candidate is statistically distinguishable for parties in dominant groups but not for their less influentially placed counterparts. In fact, the estimated probability of selecting an incumbent over another option increases by about 26 percentage points across party group types, although the variability around that estimate is quite large (95% HPD = [0.03, 0.52]). We can also use the model to calculate how the difference in the probability of selecting ideal candidate types differs across party group membership. For example, when faced with a choice between an incumbent and a national candidate, a representative big 3 party is roughly 61% more likely to select an incumbent, while a representative party belonging to another group is only 28% more likely to select the incumbent. These estimates are, of course, subject to error, but the model indicates that the probability that the big 3 premium on incumbency—relative to national experience—exceeds that of other group members is almost 96%. This difference in incumbency premium is even larger for regional and novice candidates. We have, therefore, strong evidence to support the contention that parties in big 3 groups value incumbency over other forms of political experience, including a history holding national legislative office. This is consistent with the general contention that parties that have the potential to influence policy within a given venue are most likely to strategically nominate candidates with the connections or aptitude necessary to effectively produce policy once in the target office. Although less emphatic than the evidence with respect to Hypothesis 1, our findings show that Hypothesis 2 is consistent with the data.

Figure 4.

European party group membership.

Hypothesis 3 maintains that party governing status positively predicts policy-making emphasis and a preference for incumbents. Governing parties engage in European policy making within the Council and require competent partners in the EP if they wish to be most effective. They also bear the brunt of national distaste for European policy outcomes. Here, we find weak support for our argument. Again, all three coefficients in Figure 2 are negative, as expected, yet all three 95% HPD regions cover zero. Using our Bayesian approach, we find that the probabilities that the coefficients for novice, regionally experienced, and nationally experienced candidates are less than zero are 0.58, 0.95, and 0.79, respectively. Thus, although the balance of evidence is consistent with Hypothesis 3, it fails to meet conventional null-hypothesis-test-based standards for statistical significance. It is possible that multicollinearity between our independent variables is undermining our ability to draw strong statistical conclusions here; we address this issue in the supplementary appendix.

However, as Hypothesis 4 predicts, parties with minimal national representation prefer incumbents to other types of candidates, even nationally experienced politicians. Remember that the model predicts choices based on the options at hand. Thus, although small national parties are likely to have limited supplies of both incumbents and candidates with other experience, when they do face such an option, they prioritize incumbency. This is consistent with the idea that parties that are not successful in multiple venues will have less reason to adopt strategies that involve prioritizing outsiders over those with venue-specific backgrounds. Figure 5 shows that nationally larger parties choose not only nationally experienced politicians but also those with regional and local experience at rates that are statistically indistinguishable from incumbents. For very large parties, even the 95% HPD intervals for incumbents and novices overlap. This makes sense because parties with strong outside options have incentives to pursue cross-venue candidate management strategies at the expense of cultivating policy experience. They can benefit at the national level from grooming young talent in the EP, and nationally experienced members can ride out electoral dry spells with work at the European level or use European seats as a hedge against poor national electoral fortunes. Furthermore, obtaining a reputation for placing party members that are nearing retirement in EP jobs—in other words, for rewarding aging party stalwarts for their years of service—might enhance an organization’s ability to recruit candidates at all levels and meets individuals’ office-seeking demands. In addition, because they will also tend to have more seats in the EP than their small national counterparts, large parties are better able to sacrifice an incumbent or two to pursue competing goals.²⁵

Figure 5.

Success in a competing electoral venue.

Finally, we find no clear relationship between attitudes toward Europe and preference for incumbents, contra Hypothesis 5. This seems like an unintuitive result. Why should euroskeptics engage in long-term careers and perhaps even develop policy-making advantages, within the EP? However, this result exposes the key distinction between attitudes and emphasis. What matters for nomination decisions is not so much whether or not you like an institution, but if you care about the policies that the institution oversees.

Conclusion

Experienced candidates are a scarce commodity. Thus, effective parties will rank candidates on lists to best accomplish their electoral, organizational, and policy goals. The unique character of the EP allows us to identify party motivations based on the career backgrounds of the candidates ranked toward the tops of party lists. Clearly, this informs our understanding of European politics, but it also addresses comparative questions about how, and under what circumstances, parties use nominations to pursue votes, offices, and policy. We find that the relative emphasis that a party places on the EU, the party’s expected potential to be productive in the EP, and the party’s status at the national level have implications for what sorts of candidates they prefer to provide with valuable electoral nominations. There is also weak evidence that parties that need to coordinate policy production across governing institutions prioritize institutional experience when selecting candidates.

More generally, our results imply that partisan variation in attitudes toward particular offices, the ability to affect policy in an institution, and the opportunity to place candidates across electoral venues, should all help to explain the types of politicians that parties choose to represent them in a given legislature. Thus, we propose—and provide evidence in support of—a broad framework for understanding how parties that compete in a variety of electoral contests may strategically use nomination to achieve over-arching goals. Moreover, parties that seek policy in one venue may prioritize other goals elsewhere. Thus, our results should encourage scholars to further develop a research agenda that acknowledges that goal-seeking behavior by parties is often context-dependent, and although parties sometimes appear one dimensional when evaluated in light of a single institution, they may pursue a more complex strategy across venues. Furthermore, our results have broad implications for understanding patterns of legislator behavior. Because the power to nominate provides parties with overwhelming control over who actually gets into office in many systems, party-based mechanisms may be more comparatively influential in shaping how legislators behave in office than incentives that derive from personal vote cultivation. To enjoy career success, legislators must be aware of party goals and invest their time and resources to meet those goals. Indeed, party goals may provide incentives for legislators to invest in policy-making expertise in some situations but to emphasize electoral visibility, constituency service, or fund-raising in others (Adler & Wilkerson, 2012).

Candidate nomination strategies should, in turn, affect legislative policy-making effectiveness. For instance, our finding that parties that expect to govern nominate candidates with policy-making advantages suggests that surprise electoral victories are likely to reduce legislative competence as the unexpected winner likely nominated candidates with qualities other than legislative expertise. This is an intuitive argument, but one previously based on little empirical evidence. Similarly, our emphasis on party nomination strategies may also help explain why some parties enjoy more policy influence than others within a legislature, even if they hold similar numbers of seats. Parties that prioritize candidates with policy-making expertise possess advantages in understanding complex issues, manipulating parliamentary procedures, and developing personal networks. Parties that nominate candidates on the basis of other goals—like electoral visibility or fund-raising capability—may be left at a disadvantage when it comes to passing legislation. Thus, how parties choose to rank candidates has downstream implications for legislative output and policy influence. Understanding those nomination choices, therefore, is critical if we wish to account for patterns of legislative policy making within and across countries, and over time.

Finally, we introduce data that provide researchers with a rich vein of information about candidates in European elections and statistical tools that are applicable to the study of candidate nomination in many contexts. Given that the EP biographies data set contains a number of characteristics beyond legislative experience of candidates, it allows researchers to investigate many other theoretical questions about the way parties prioritize candidate characteristics or skills. For example, when do parties prioritize female candidates? Do parties nominate individuals with particular professional backgrounds to accomplish their goals in parliament? Do candidates’ educations matter for nomination? The data set that we introduce will allow researchers to investigate a multitude of questions about what different types of parties value in their candidates. And, as candidate-level electoral data become more readily available, scholars can leverage the methods that we introduce here to study nomination strategies in other elections.

Footnotes

Acknowledgements

We thank Cliff Carrubba, Matt Gabel, Simon Hix, David Leblang, and numerous seminar participants at the annual meeting of the Midwestern Political Science Association, London School of Economics, University of Illinois, University of Oregon, Texas Tech University, and Vanderbilt University for helpful comments. We also thank the European Union Center at University of Illinois for supporting this project.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded in part by the European Union Center of Excellence at the University of Illinois through a grant from the European Union.

Supplemental Material

The supplementary appendix is available at www.sneps.net/cps1supplemental.htm and

Notes

Author Biographies

Daniel Pemstein is an assistant professor of political science at North Dakota State University. His primary research interests are legislatures, political economy, and European Union politics.

Stephen A. Meserve is an assistant professor of political science at Texas Tech University. His research interests lie in comparative political economy, comparative legislatures, and democracy measurement.

William T. Bernhard is a professor of political science and Associate Provost for Faculty Development at the University of Illinois. His research focuses on international relations, political economy, and legislative studies.

References

Adler

E. S.

Wilkerson

J. D.

(2012). Congress and the politics of problem solving. Cambridge, UK: Cambridge University Press.

Besley

(2005). Political selection. Journal of Economic Perspectives, 19(3), 43-60.

Best

Cotta

(2000). Parliamentary representatives in Europe 1848-2000. Oxford, UK: Oxford University Press.

Blomgren

(1999, March 26-31). Candidate selection to the European Parliament: A comparative study of Ireland, the Netherlands and Sweden. Paper presented at the ECPR Joint Workshop, Mannheim.

Bowler

Farrell

D. M.

(1995). The organizing of the European Parliament: Committees, specialization and coordination. British Journal of Political Science, 25, 219-243.

Caramani

(2004). The nationalization of politics: The formation of national electorates and party systems in Western Europe. Cambridge, UK: Cambridge University Press.

Carey

J. M.

Shugart

M. S.

(1995). Incentives to cultivate a personal vote: A rank ordering of electoral formulas. Electoral Studies, 14, 417-440.

Crisp

B. F.

Olivella

Potter

J. D.

(2013). Party-system nationalization and the scope of public policy. Comparative Political Studies, 46, 431-456.

Daniel

W. T.

(2013). When the agent knows better than the principal: The effect of education and seniority on European Parliament Rapporteur Assignment. Journal of Common Market Studies, 51, 832-848.

10.

European Election Studies. (2010). European Parliament election study 2009, candidate study, advance release (Technical report). Available from www.piredeu.eu

11.

Farrell

D. M.

Hix

Johnson

Scully

(2006). EPRG 2000 and 2006 MEP surveys dataset (Technical report). Retrieved from http://www.lse.ac.uk/government/research/resgroups/EPRG/MEPsurveyData.aspx

12.

Farrell

Heritier

(2004). Interorganizational negotiation and intraorganizational power in shared decision making: Early agreements under codecision and their impact on the European Parliament and Council. Comparative Political Studies, 37, 1184-1212.

13.

Follesdal

Hix

(2006). Why there is a democratic deficit in the EU: A response to Majone and Moravcsik. Journal of Common Market Studies, 44, 533-562.

14.

Fowler

J. H.

(2006). Connecting Congress: A study of cosponsorship networks. Political Analysis, 14, 456-487.

15.

Gallagher

(1980). Candidate selection in Ireland: The impact of localism and the electoral system. British Journal of Political Science, 10, 489-503.

16.

Gallagher

(1988). Introduction. In Gallagher

Marsh

(Eds.), Candidate selection in comparative perspective: The secret garden of politics (pp. 1-19). London, England: Sage.

17.

Gallagher

Marsh

(1988). Candidate selection in comparative perspective: The secret garden of politics. London, England: Sage.

18.

Gherghina

Chiru

(2009). Where is the difference? The profile of the Romanian candidates for the 2009 European elections. Romanian Journal of European Affairs, 9(4), 79-99.

19.

Giebler

Haus

Wessels

(2010). 2009 European election candidate study—Codebook, advance release (Technical report). Berlin, Germany: WZB. Available from www.piredeu.edu

20.

Hix

Hobolt

S. B.

Høyland

(2012, August 30-September 2). Career paths and legislative activities of members of the European Parliament. Paper presented at the Annual Conference of the American Political Science Association, New Orleans, LA.

21.

Hix

Marsh

Vivyan

(2009). Predicting the June 2009 European Parliament elections (Technical report). Available from http://www.predict.edu

22.

Hobolt

S. B.

Høyland

(2011). Selection in European Parliament elections. British Journal of Political Science, 41, 477-498.

23.

Hrbek

Schweitzer

(1989). Die Deutschen Europa-Parlamentarier: Ergebnisse einer Befragung der deutcshen Mitglieder des Europaeischen Parlaments [The German MEPs: Results from a Survey of German Members of the European Parliament]. Aus Politik und Zeitgeschichte, B3, 3-18.

24.

Kauppi

(1996). European Institutions and French political careers. Scandinavian Political Studies, 19, 1-24.

25.

Linek

Outly

(2006). Selection of candidates to the European Parliament in the main Czech political parties. Contemporary European Studies, 1, 5-31.

26.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage.

27.

Marsh

(1998). Testing the second-order election model after four European elections. British Journal of Political Science, 28, 591-607.

28.

McElroy

(2006). Committee representation in the European Parliament. European Union Politics, 7, 5-29.

29.

Meserve

S. A.

Pemstein

Bernhard

W. T.

(2009). Political ambition and legislative behavior in the European Parliament. Journal of Politics, 71, 1015-1032.

30.

Moravcsik

(2002). In defence of the “democratic deficit”: Reassessing the legitimacy of the European Union. Journal of Common Market Studies, 40, 603-634.

31.

Norris

(1997). Passages to power. Cambridge, UK: Cambridge University Press.

32.

Norris

Franklin

(1997). Social representation. European Journal of Political Research1, 32, 185-210.

33.

Patzelt

W. J.

(1999). Recruitment and retention in Western European Parliaments. Legislative Studies Quarterly, 24, 239-279.

34.

Pemstein

Quinn

K. M.

Martin

A. D.

(2011). The Scythe Statistical Library: An open source C++ library for statistical computation. Journal of Statistical Software, 42(12), 1-26.

35.

Rahat

Hazan

R. Y.

(2001). Candidate selection methods: An analytical framework. Party Politics, 7, 297-322.

36.

Ranney

(1981). Candidate selection. In Butler

Penniman

H. R.

Ranney

(Eds.), Democracy at the polls: A comparative study of competitive national elections (pp. 75-106). Washington, DC: American Enterprise Institute.

37.

R Development Core Team. (2010). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available from http://www.r-project.org

38.

Reif

Schmitt

(1980). Nine second-order national elections: A conceptual framework for the analysis of European election results. European Journal of Political Research, 8, 3-44.

39.

Ringe

(2010). Who decides, and how? Preferences, uncertainty, and policy choice in the European Parliament. Oxford, UK: Oxford University Press.

40.

Samuels

(2003). Ambition, federalism, and legislative politics in Brazil. Cambridge, UK: Cambridge University Press.

41.

Scarrow

(1997). Political career paths and the European Parliament. Legislative Studies Quarterly, 22, 253-263.

42.

Schmitt

(2005). The European Parliament elections of June 2004: Still second-order? West European Politics, 28, 650-679.

43.

Shomer

(2009). Candidate selection procedures, seniority, and vote-seeking behavior. Comparative Political Studies, 42, 945-970.

44.

Stolz

(2001, April 6-11). Parliamentary careers in Europe between the regional, national, and supranational level. Presented at the ECPR Joint Sessions of Workshops, Grenoble.

45.

Strom

(1990). A behavioral theory of competitive politics. American Journal of Political Science, 34, 565-598.

46.

Westlake

(1994). Britain’s emerging euro-elite? The British in the directly elected European Parliament. Aldershot, UK: Dartmouth.

47.

Yoshinaka

McElroy

Bowler

(2010). The appointment of rapporteurs in the European Parliament. Legislative Studies, 35, 457-486.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.21 MB