Top-down or bottom-up? The selection of shadow rapporteurs in the European Parliament

Abstract

Shadow rapporteurs play an important role in developing the European Parliament’s collective policy positions and in defending them in inter-institutional negotiations. This study sheds light on the ‘how’ and ‘why’ of shadow rapporteur selection. Qualitative insights from practitioner interviews and a quantitative analysis of shadow rapporteur data from the 7th European Parliament (2009–2014) indicate that the appointment process is primarily one of bottom-up self-selection by group members based on their policy interests. The party group leadership, in the form of group coordinators, plays an important coordinating role when there is competition for a shadow rapporteurship. However, the role of group coordinators is more akin to a third-party arbiter of competing demands than a mechanism of top-down control by the leadership, as suggested by principal-agent theory.

Keywords

Delegation European Parliament principal-agent theory self-selection shadow rapporteurs

Introduction

This study examines the selection of shadow rapporteurs in the European Parliament (EP), who serve as their party groups’ primary negotiators of particular legislative proposals, making them highly consequential political actors with disproportionate influence over the content of legislation. Despite their important role, however, little is known about the allocation of shadow rapporteurships. We demonstrate, using both quantitative and qualitative data, that shadow rapporteurs are not selected in a top-down process by the party group leadership; instead, shadow rapporteurships are allocated primarily through a bottom-up process in which members of the EP (MEPs) self-select to become shadow rapporteurs.

Each legislative proposal drafted by the European Commission is assigned a rapporteur in the EP’s lead committee, who drafts its official report of proposed amendments, shepherds the legislation through the law-making process, and serves as Parliament’s main negotiator in inter-institutional bargaining. Rapporteurs are selected in an ‘auction’ in which the leaders of each party group in the responsible committee (who are called coordinators) bid on reports using ‘points’ they receive according to the proportion of seats they hold, meaning that only one EP party group can secure a rapporteurship. To avoid having a single partisan actor dominate the legislative process, the other party groups may select one of their own to monitor the work of the rapporteur and negotiate on their behalf. Together with the rapporteur, these shadow rapporteurs ‘practically constitute informal sub-committees’ (Corbett et al., 2011: 159) that take the lead in negotiating particular pieces of legislation in the responsible EP committee. As such, they have the capacity to influence decision-making inside their party groups, the inter-party negotiations on the content of the collective position of the EP, and the EP’s inter-institutional bargaining process with the Council of Ministers.

Existing research recognises and highlights the influence of shadow rapporteurs in EU law-making. Jensen and Winzen (2012) show that rapporteur and shadow rapporteurs form the core of a highly hierarchical cooperation network in the EP. Ripoll Servant and Panning (2019) demonstrate how meetings of the rapporteur and the shadow rapporteurs are instrumental in developing and negotiating their EP committees’ positions. Judge and Earnshaw (2011) reveal how a coalition of shadow rapporteurs side-lined a recalcitrant rapporteur in inter-institutional negotiations with the Council. Ringe (2010) shows that rapporteurs and shadow rapporteurs not only decisively influence the substance of legislation; they also shape their party colleagues’ voting behaviour by framing its content and consequences. Hence, it is no surprise that shadow rapporteurs are recognised by the Council Presidency as important contacts to pursue and utilise in the EU legislative process (Roederer-Rynning and Greenwood, 2015: 1156) and that observers of EP politics recognise them as prominent actors with significant sway over policy outcomes (e.g. Hurka et al., 2015).

Despite the influence of shadow rapporteurs in the EP’s policy-making process, we know little about how they are selected. While the role of rapporteurs in EU law-making (e.g. Benedetto, 2005; Costello and Thomson, 2011; Mamadouh and Raunio, 2003) and their selection (e.g. Daniel, 2015; Hermansen, 2018; Høyland 2006; Kaeding, 2004; Kaeding 2005; Yordanova, 2011; Yoshinaka et al., 2010) have been recurring subjects in previous research, a lack of readily quantifiable data has, until recently, precluded similar analysis of how shadow rapporteurships are allocated. A careful investigation is thus warranted.

We examine whether the process of choosing shadow rapporteurs is a top-down assignment process or a bottom-up process of self-selection. Regarding the former, we apply a principal-agent framework to explicate the delegation logic of distributing shadow rapporteurships. The hypotheses we derive reflect the concern of party leaders about the faithful representation of their policy views as well as their interest in the effective oversight and extraction of concessions from the lead rapporteur. With respect to a bottom-up selection process, we theorise the motivations of parliamentarians to promote their policy interests and re-election chances within and outside the EP. MEPs would thus self-select to cover legislative dossiers based on policy salience, interests, and expertise.

The quantitative element of our mixed methods empirical approach analyses data on rapporteurs and shadow rapporteurs during the EP’s 2009–2014 term. Information on shadow rapporteurs was not published on a large scale by the EP prior to 2009 and has not been used extensively in research to date. Exceptions are Hurka et al. (2015), who find that MEPs from the countries that joined the EU in and after 2004 are under-represented as shadow rapporteurs; and Häge and Ringe (2019), whose study of social networks composed of rapporteurs and shadow rapporteurs finds that MEPs from small party groups are particularly central and have greater potential for brokerage. The quantitative analysis uses a novel dyadic research design that allows us to investigate the role of strategic and motivational factors in the shadow rapporteur appointment process that have not been the subject of previous research. The qualitative element relies on information from interviews with a carefully selected group of EP respondents from all but one party group and a large number of standing committees.

The findings from the analyses of both sets of data complement and confirm each other: the assignment of shadow rapporteurships is a bottom-up process based on self-selection, not a top-down delegation process with EP party groups acting as principals and individual MEPs as their agents.

Shadow rapporteur selection: Top-down or bottom-up?

Unlike the assignment of rapporteurs, the selection of shadow rapporteurs has not received systematic attention in previous research, and the factors influencing the allocation of rapporteurships are likely not the same as those influencing shadow rapporteur selection. First, the stakes in the selection of shadow rapporteurs are likely lower, both in terms of the opportunity costs for party group coordinators as they bid for lead rapporteurships and in the rewards for individual MEPs who covet the positions. Second, the process in which rapporteurships are auctioned off ascribes coordinators a strategic role in competitive inter-party bargaining, with different incentives and payoffs than the intra-party process of allocating shadow rapporteurships. Finally, shadow rapporteurs are chosen only after a report has been assigned to another party group, and the identity of the lead rapporteur is usually – but not necessarily, according to our respondents – known at that time. This temporal sequence has the potential to shape strategic considerations in the selection of shadow rapporteurs.

Much existing research assumes that ideology and partisanship drive political appointments in EP politics, including previous work on the allocation of rapporteurs that explicitly or implicitly views that process through the lens of principal-agent theory (e.g. Chiou et al., 2019; Hausemer, 2006; Kaeding, 2004; Kaeding 2005; Obholzer et al., 2019; Yordanova, 2011; Yoshinaka et al., 2010). This approach is appealing because it is parsimonious, intuitive, and in line with the dominant views of EP politics as revolving around strong legislative party organisations (see especially Hix et al., 2007; Kreppel, 2002; Yordanova, 2013). In a top-down process, leaders of the party group nominate members for these types of positions that, based on pre-existing characteristics, promise to best represent the leadership’s interests.

However, the uncritical application of principal-agent models to these types of hierarchical relationships in EU policy-making has also been challenged, because the assumptions of these models are a poor reflection of actual decision-making processes in the EU institutions (Häge, 2011b; Ringe, 2005). Reasons to be sceptical of a strong principal-agent logic in EP politics include the institution’s consensual decision-making culture, the high frequency of decisions passed by oversized majorities, and the fact that legislative work is not obviously linked to MEPs’ electoral fortunes. Most importantly, however, the notion that the principal has pre-existing preferences on specific policy proposals and selects agents on this basis is often empirically implausible. This assumption disregards the possibility that the principal’s preferences are, in fact, endogenously shaped by the agent in the policy-making process. If the principal’s policy preferences are only developed through the work of the agent, however, they cannot play a role in the appointment process.

Top-down assignment

From a principal-agent perspective, shadow rapporteurs fulfil a dual function. First, they represent their party group in collective decision-making processes of the committee. They formulate policy positions and negotiate on their group’s behalf with the rapporteur and their counterparts from other groups to arrive at a collective committee decision. Second, they monitor the actions of the rapporteur. The tasks delegated by committees to rapporteurs, and the prerogatives and role expectations that come with them, put rapporteurs into influential positions in the EU’s policy-making process. Rapporteurs may abuse these positions to promote their own personal or party political interests, but oversight by shadow rapporteurs limits their discretion. However, both of these functions of shadow rapporteurs can themselves be seen as being based on an explicit or implicit act of delegation. In the shadow rapporteur context, the principal is the party group as a collective actor – likely represented by the coordinator as the party group’s leader in the relevant committee – and the agent is the member nominated as shadow rapporteur.

With regard to shadow rapporteurs’ policy formulation function, the delegation logic underlying the development of legislative committees applies analogously (Gilligan and Krehbiel, 1987). Just like the establishment of legislative committees, appointing shadow rapporteurs to deal with different policy proposals is a form of legislative specialisation, except that it occurs within the party group rather than the legislature as a whole. According to this logic, principals lack policy expertise and are unsure about how policy proposals translate into practical outcomes on the ground. In order to make better-informed decisions, they delegate the policy formulation task to agents. To incentivise agents to accumulate the necessary policy expertise, principals grant them a disproportionate amount of policy-making influence. In the case of EP policy-making, shadow rapporteurs are largely put in charge of developing specific policy proposals for their party group, as well as promoting and defending them in formal and informal decision-making arenas.

However, given the informational asymmetry between principals and agents, agents might abuse these prerogatives to pursue policy goals that differ from those of their principals. In other words, shadow rapporteurs might exploit their decision-making prerogatives to promote their own policy views rather than policies that are in the best interest of their party group. Given that the party group does not know which policy solution is in its best interest, it will find it hard to identify and sanction such behaviour. Thus, the most effective way for principals to prevent agency drift is to select agents that have similar policy preferences. In the context of shadow rapporteurships, the party group leaders – in practise, that is the party group coordinators in the EP’s standing committees – are expected to appoint group members as shadow rapporteurs that hold policy positions that are representative of the group as a whole. The goal and the role of the group coordinator is to minimise agency loss by ensuring that the policy position of the person selected as shadow rapporteur does not deviate too far from the position of the party group’s median member:

H1: The closer a party group member’s policy position is to the median position of the party group, the more likely that party group member will be appointed as shadow rapporteur.

It may be the case, however, that the party group leadership is not only concerned with the characteristics of its own group members when assigning shadow rapporteurs. Applying a higher level of strategic foresight, the party group leadership might also take the interactions with the rapporteur and his or her policy positions into account. In this scenario, it is not the policy distance of the group member’s position from the group median’s position that matters, but the distance of the group member’s position from the position of the rapporteur. Much of this argument is based on Epstein and O’Halloran’s (1995) theory of strategic oversight, which models the role of interest groups for Congress’ ability to evaluate the actions of executive agencies. We can apply this model analogously to the EU law-making process. In the EU context, the rapporteur drafts a report on the Commission’s proposal or negotiates an inter-institutional text with the Council Presidency, the shadow rapporteur recommends support of or opposition to the rapporteur’s text to his or her party group, and, on the basis of this recommendation, the party group accepts or rejects the rapporteur’s text. While the rapporteur and shadow rapporteur have private information about the practical implications of the proposed policy provisions, other members of the shadow rapporteur’s party group – including its leadership – are more uncertain. For that reason, the party group leadership must be concerned about collusion between the rapporteur and shadow rapporteur and be wary of the shadow rapporteur’s recommendation.

However, the likelihood that the shadow rapporteur colludes with the rapporteur decreases with increasing distance between their policy positions. Shadow rapporteurs whose policy positions diverge from that of the rapporteur have an incentive to scrutinise the work of the rapporteur more carefully and to evaluate the content of the report more critically. All else equal, therefore, greater distance between the policy positions of shadow rapporteur and rapporteur makes it more likely that the shadow rapporteur will fulfil one of its key responsibilities associated with the position: to monitor and serve as a check on the rapporteur. Shadow rapporteurs with views that diverge from those of the rapporteur are also more likely to push harder for concessions from the rapporteur when participating in rapporteur-shadow rapporteur meetings. Such concessions, in turn, move the final policy outcome closer to the position of the party group’s median member. Taken together, the incentives to properly monitor the rapporteur’s actions and effectively defend the party group’s policy position make shadow rapporteurs with a policy position more distant from the policy position of the rapporteur more trustworthy representatives and interlocutors.

Once strategic interactions with the rapporteur are taken into account, a shadow rapporteur whose policy position diverges from the rapporteur’s is even more beneficial to the party group than one whose position is tightly aligned with the position of the party group median. A party group with more strategic foresight should, therefore, select a shadow rapporteur with policy positions that diverge as much as possible from those of the rapporteur.

H2: The more distant a party group member’s policy position is to the rapporteur’s position, the more likely that party group member will be appointed as shadow rapporteur.

Bottom-up self-selection

It is possible, however, that the assignment of shadow rapporteurs is much less of a top-down process than the principal-agent perspective suggests. If the process is more bottom-up and MEPs self-select into shadow rapporteurships, they will choose to cover proposals based on some combination of policy interests, policy salience, and policy expertise.

An indicator that policy salience drives self-selection would be that MEPs select to become shadow rapporteurs for proposals of particular salience in national politics. Becoming engaged in the formulation of policies that have stark implications for nationally important economic sectors or receive much attention in national public and political debate may enhance MEPs’ public visibility, generate opportunities for credit-claiming, and galvanise the support of important interest groups, which in turn should improve MEPs’ reelection chances. Independent of their party affiliations, Nordic MEPs may thus be particularly inclined to volunteer for fisheries legislation, Mediterranean MEPs for dossiers on tourism, and German MEPs for proposals affecting the car industry. Unfortunately, we cannot measure the national salience of individual policy proposals directly. However, if the national salience of policy is an important factor in both becoming rapporteur and shadow rapporteur, then the coincidence of interests at the dyadic level should result in a high probability of rapporteurs being matched by shadow rapporteurs from the same country.

H3: Party group members with the same national background as the rapporteur are more likely to become a shadow rapporteur than party group members with a different national background.

MEPs’ policy interests are also difficult to measure directly, but we can observe the types of policy proposals MEPs have been working on as rapporteurs or shadow rapporteurs and treat the content of these proposals as manifestations of MEPs’ policy specialisms. If policy interests have a substantial effect on MEPs’ choices in taking on rapporteurships and shadow rapporteurships, then we would expect rapporteurs to be matched with shadow rapporteurs that have a report and opinion portfolio that is similar in terms of policy content.

H4: The more similar a party group member’s policy interests are to the rapporteur’s policy interests, the more likely that party group member will become a shadow rapporteur.

Qualitative analysis

We conducted ten in-depth, semi-structured interviews in the EP and exchanged several emails with one MEP who was not available to meet in person. Our sample was carefully and purposefully selected and, therefore, includes respondents from seven (of eight) political groups and ten member states who have been or are involved with ten (and thus half) of the EP’s standing committees. Among our respondents were MEPs, MEP assistants, party group advisors, and members of the EP secretariat. The interviews provide analytical leverage by offering insight into the actual practice of assigning shadow rapporteurs and by allowing for methodological triangulation. Moreover, the quantitative and qualitative data complement one another, such that we can use each to make better sense of the other.

We received highly consistent responses across interviews, in particular with regard to the selection process for shadow rapporteurs, which our respondents described as primarily bottom-up.¹ Independent of each other and consistently across party groups and committees, respondents described a process whereby MEPs in the responsible committee indicate which reports they would like to cover and succeed much of the time. One political group advisor, for example, describes that ‘it is a bottom up approach, not top down, but it is a matter of self-organisation where people turn up to do their job’ (Respondent #3). To start, MEPs indicate at the beginning of a new legislative term which policy areas and topics they would generally prefer to focus on (#7, 8, 11). Then they ‘come forward to express their interest’ in particular reports (#2), either in response to lists of upcoming reports that are shared in advance (#3; also #9, 11, 10) or when they are ‘asked who is interested to become the shadow’ in meetings of committee members from the same party group (#10; also #8). On this basis, a decision is made in a process that respondents agree is generally collaborative and ‘collegial’, as one MEP put it (#4). A number of different criteria are applied such as

‘other files [MEPs] are working on to ensure a balance in workload, interest in the particular field, expertise in the relevant field, how much they participate in the working group – votes, attendance, etc. We try to strike a balance, also geographically and across member states’ (#11).

Other respondents similarly listed some or all of these criteria: relative workload and the distribution of other reports (#6, 8, 10); previous level of engagement in legislative activities (#2, 3); and the geographic and national makeup of the negotiating team as a whole (#4, 6). Concerning the latter, it is notable that respondents explicitly indicated that ‘you don’t want a whole group of negotiators who are all from the same country’ (#6, also #11), which limits the extent to which members may choose reports based on national priorities and thus casts doubt on H3. In line with H4, however, substantive expertise is a major consideration (#3, 6, 8, 10), and there are ‘definitely some “go to” people for specific policy areas; for example, somebody is the ‘‘carbon capture guy’’’ (#7; also #1, 3, 9). These criteria are not formal rules, however, but conventions that reflect best practices (#2, 3, 7, 8, 9, 10).

Another consideration is that everybody who wants to serve as shadow has to be given the opportunity at times, to ‘keep people happy’ (#3, 11; also #8) and make sure nobody is ‘feeling like they are excluded’ (#9, also #6). In other words, it would be seen as a major problem if the party group ‘ignore[d] [some] members or favor [others]; everybody has to be able to work’ (#11). Ensuring this is one of the responsibilities of party group coordinators. They are crucial in coordinating the bottom-up process of shadow rapporteurship allocation (#2, 3, 8, 9, 10, 11) – in which they are supported by party group advisors (#3, 11) – especially when more than one MEP requests a particular shadow rapporteurship and a choice has to be made. An MEP from the Greens explained that ‘if there are two or three members who show the same interest as well, there is a discussion and normally the coordinator […] comes up with a proposal, which he checks with the colleagues’ (#10). A major consideration is, again, who is covering other reports and ‘what’s in the pipeline’ (#3). And while it is not the case that coordinators simply declare ‘you get that, you get that’ (#8), they can try to steer particular reports to specific MEPs. Especially when the report concerns ‘a big legislative file, we make sure we get the right person, how efficient a member might be’ (#11). Similarly, ‘if only some [committee members] are active and engaged, allocation is based on the nature of dossier, but also if the coordinator thinks the person will actually work’ (#3). Others concurred, explaining that they consider ‘who I would like to have this done by’ (#8) or if somebody is ‘not so good at negotiating’ (#10).

In other words, even though all respondents emphasised the bottom-up dynamic in the allocation of shadow rapporteurships, most also acknowledged that coordinators can and do influence who is put in charge of a particular report, especially when multiple MEPs explicitly request it. None of our respondents maintained, however, that shadow rapporteurs are selected in light of their policy positions vis-à-vis the party group median or the rapporteur. Thus, H1 and H2, about the effects of member’s policy position, found no support. Neither did H3 about the role of national salience on the selection of shadow rapporteurs. In contrast, policy interests and expertise are most often mentioned as reasons for volunteering or being selected as shadow rapporteur, which supports H4. Overall, the interview responses describe the allocation of shadow rapporteurships primarily as a process of bottom-up self-selection, and secondarily as a top-down coordination process that is principally aimed at ensuring that shadow rapporteurs are competent and engaged.

Quantitative data and methods

The dataset analysed in this study is based on information about characteristics of co-decision reports and MEPs in the EP’s 7th term (2009–2014). Through computer scripts, we downloaded, extracted, and merged information from two separate sources on the EP website: the ‘Legislative Observatory’ and the ‘History of Parliamentary Service’ pages of the directory providing MEPs’ biographical information. Our dataset takes a dyadic form, where the two dyad members defining a row in the data matrix are the rapporteur and a potential shadow rapporteur from another party group. In principle, all full and substitute members of the relevant committee at the time of the adoption of a file are potential shadow rapporteurs. However, we exclude all non-aligned committee members, who are not a member of a party group, and committee members from party groups that did not appoint a shadow rapporteur for a particular file. The theoretical expectations derived from principal-agent theory are not applicable and the related explanatory variables are not defined for MEPs that are not members of a party group.

The elimination of dyads including committee members from party groups that did not nominate a shadow rapporteur ensures that variation in the party group’s choice of whether to nominate a shadow rapporteur is not conflated with variation in the party group’s choice of who to nominate. Focusing on this sub-sample is conceptually more appropriate, as our theoretical arguments relate to the selection of shadow rapporteurs from a pool of potential shadow rapporteurs, not to the decision to appoint or not appoint a shadow rapporteur in the first place. Based on these selection criteria, the sample contains 53,522 ‘rapporteur-potential shadow rapporteur’ dyads, which make up 2031 choice sets related to 491 co-decision reports. Choice sets consist of all members of a particular party group, within a particular committee, dealing with a particular report. In other words, they constitute the pool of party group members of a committee from which a shadow rapporteur can be selected for a certain report. Our quantitative analysis employs a conditional logit model, which uses choice sets as a stratification variable. Table 1 indicates how the size of the choice set varies within and across party groups. While choice sets can be as small as two MEPs for the smaller party groups, such small sets are the exception rather than the rule. Even the smallest party groups were able to select on average from three to four MEPs when nominating a shadow rapporteur for a particular report.

Table 1.

Choice set size by party group.

Party group	Minimum	Mean	Maximum
EPP	15	33.7	68
S&D	10	23.6	44
ECR	2	7.1	15
GREENS	3	7.0	13
GUE-NGL	2	4.3	10
ALDE	5	10.0	20
EFD	2	3.6	6

Note: The table shows the minimum, mean, and maximum number of party group members across choice sets. The choice set consists of members of a particular party group, within a particular committee, dealing with a particular report. The total number of choice sets is 2031.

ALDE: Alliance of Liberals and Democrats for Europe; ECR: European Conservatives and Reformists; EFD: Europe of Freedom and Democracy; EPP: European People’s Party; GUE-NGL: Confederal Group of the European United Left/Nordic Green Left; S&D: Progressive Alliance of Socialists and Democrats.

The dependent variable in the statistical analysis is a simple binary variable, indicating whether (1) or not (0) a committee member was nominated by his or her party group to shadow a particular rapporteur. The two explanatory variables testing different versions of the principal-agent argument relate to ideological differences between the rapporteur, the shadow rapporteur, and the party group median. In the absence of exogenous ideology scores for individual MEPs, we use data from the 2010 Chapel Hill Expert Survey (CHES) about the ideological positions of their national parties (Bakker et al., 2015). In contrast to ideology measures based on EP roll call votes, CHES party positions are unlikely to be affected by MEPs’ activities as rapporteurs or shadow rapporteurs. Experts of national party systems estimate parties’ positions based mainly on their communication and behaviour in domestic politics. Given the small size of many national party delegations, these data still provide considerable variation of positions within European party groups. Indeed, for the 7th term of the EP, general left-right positions based on national party positions show a higher degree of within-party group variation than W-Nominate scores for individual MEPs. In addition, at the party group level, the positions from the CHES data have better face validity than the positions based on roll call votes, and they allow for the consideration of a wider array of conflict dimensions. We compute three versions of the position variables, based on different dimensions of party competition often identified in existing studies of EU politics (e.g. Hix et al., 2006; Marks et al., 2006): the general left-right dimension, the pro-/anti-European integration dimension, and the GAL-TAN (Green-Alternative-Libertarian vs. Traditional-Authoritarian-Nationalist) dimension.²

If the party groups’ strategic considerations focus only on the characteristics of its members, we would expect party group leaders to select shadow rapporteurs that are ideologically close to the party group median. We operationalise this variable as the absolute value of the distance between the ideological position of the national party of the potential shadow rapporteur and the ideological position of the national party of the party group median. If party groups also take the interactions with the rapporteur into account, party group leaders are better off selecting shadow rapporteurs with positions that are far away from the position of the rapporteur. This variable is operationalised as the absolute value of the distance between the ideological position of the national party of the rapporteur and the ideological position of the national party of the potential shadow rapporteur.

The variable measuring joint national background takes a value of 1 if the rapporteur and the potential shadow rapporteur come from the same member state and 0 otherwise. The variable measuring the similarity of policy interests is based on the subject codes of policy proposals. The EP’s legislative observatory assigns each decision-making process one or more hierarchically structured policy subject codes. If an MEP was a rapporteur or shadow rapporteur for a report or opinion related to a particular policy proposal, the MEP is linked to this proposal’s subject codes. The number of times an MEP was linked to each of 372 subject codes constitutes the MEP’s subject code profile. The similarity of two MEPs’ policy interests is then measured as the correlation between their subject code profiles, which is in turn based on the extent two MEPs acted as rapporteurs and/or shadow rapporteurs for policy proposals of similar content.

To compute the correlation between policy subject code profiles of pairs of MEPs, we use the concordance correlation coefficient (Lin, 1989). The concordance correlation coefficient can be interpreted as a chance-corrected agreement index for valued data (Häge, 2011a). It expresses the agreement in policy subject codes covered by MEPs beyond the agreement expected based on the marginal distributions of their policy subject code profiles. The coefficient adjusts the agreement score for the differential propensity of MEPs to be linked to a policy subject, and for the overall low propensity of all MEPs to be linked to any policy subject. The former feature makes sure that differences between the dyad members in the total number of policy subjects they are linked to are not treated as dissimilarity in terms of policy content; the latter feature ensures that the preponderance of the joint absence of links to policy subjects is not unduly treated as a form of similarity. In other words, the chance-correction ensures that the score of the correlation coefficient reflects similarity in the content of policy subject profiles rather than similarity in the frequency of MEPs’ engagement in policy work.

To alleviate somewhat concerns about the endogeneity of this variable to shadow rapporteur appointments, it should be noted that the data basis for the policy interest similarity variable is much broader. Whereas the dependent variable focuses on shadow rapporteurships for codecision reports only, the policy interest similarity variable is based on rapporteurships and shadow rapporteurships for any type of report or opinion, including those for non-legislative proposals, own-initiative reports, consultation and consent procedure proposals.³ As the original policy subject profile variable is highly positively skewed, we transform the profile scores to their natural logarithm before computing the concordance correlation coefficients.

We also include a range of monadic control variables used in previous research. The first is seniority, operationalised as the number of terms an MEP has served in the EP. Less senior members might be more likely to become shadow rapporteurs, seeing it as a first step on the ladder of influential office positions in the EP. Next, the size of the national party delegation within the European party group might affect their members’ chances of becoming a shadow rapporteur, with larger national parties expected to secure a disproportionally larger share of shadow rapporteurships. This variable is operationalised as the share of European party group members that are members of the national party delegation. MEPs in committee leadership positions are also supposed to be in an advantageous position to obtain desirable rapporteurships for themselves. The committee leadership variable indicates whether or not an MEP was a committee chair or vice-chair. In contrast, party group leadership positions are supposed to make involvement in committee work less likely. The party group leadership variable indicates whether or not an MEP was a member of the bureau of a political group. Furthermore, substitute committee members might generally be less likely to get involved in committee work than full members. A dummy variable indicating whether or not an MEP was a substitute committee member captures this difference. All position and membership variables are measured at the point of adoption of the report. In general, MEPs that engage in EP work more generally might also be more likely to become shadow rapporteurs. Thus, we include a variable for absenteeism, which measures the percentage of roll call votes missed by an MEP. As the distribution of the variable is concentrated near its minimum value and has a strong positive skew, we transform it by taking the natural logarithm.

Finally, to focus the analysis of shadow rapporteur selection on its intra-party group variation within committees, we employ a conditional logit model to analyse the data. The stratification in conditional logit models allows for the incorporation of fixed effects without the need to explicitly estimate them. This property of conditional logit models is particularly useful in situations like ours, where the inclusion of a large number of dummy variables to estimate coefficients for fixed effects would result in estimation problems. Our stratification variable indicates a party group within a committee, dealing with a particular report. As a result, stratifying the analysis by this variable holds all report, committee, and party group characteristics constant. The stratification also adjusts for many of the dependencies caused by the multi-level structure of the data. To further investigate whether the effects of explanatory variables are heterogeneous across party groups, we do not only report regression results for the full sample of dyads, but also for individual party groups.⁴ This form of analysis allows for possible multi-level interactions. In the interpretation of results, we prioritise the results of the individual party group analyses, as party groups are the appropriate context in which the nomination of shadow rapporteurs occurs.

Determinants of shadow rapporteur appointment

Table 2 presents the results of a conditional logistic regression with shadow rapporteur appointment for codecision reports as the dichotomous dependent variable. In general, the results provide support for the bottom-up self-selection perspective, while evidence for the top-down delegation perspective is inconsistent and not robust. The model estimates based on the full sample of codecision reports produce statistically significant effects for some top-down delegation variables, but those turn out to be fragile when the analysis concentrates on sub-samples of individual party groups.

Table 2.

Conditional logistic regression of shadow rapporteur appointment.

	Full sample	GUE-NGL	Greens	S&D	ALDE	EPP	ECR
Top-down
Distance to party group
Left-right	0.94 (0.05)	1.11 (0.43)	1.00 (0.12)	1.24 (0.30)	0.70** (0.09)	0.74 (0.13)	0.99 (0.36)
EU Support	0.79*** (0.05)	1.00 (0.33)	0.72* (0.11)	0.36*** (0.09)	1.95*** (0.37)	0.91 (0.16)	0.43* (0.15)
GAL/TAN	0.92* (0.03)	1.11 (0.15)	0.84 (0.14)	1.07 (0.12)	0.89 (0.06)	1.04 (0.12)	0.66* (0.13)
Distance to rapporteur
Left-right	1.07 (0.04)	0.99 (0.32)	1.07 (0.14)	0.77 (0.12)	1.12 (0.07)	0.98 (0.12)	1.13 (0.18)
EU support	1.01 (0.05)	0.67 (0.18)	1.18 (0.16)	1.13 (0.23)	0.99 (0.14)	0.79 (0.10)	0.93 (0.27)
GAL/TAN	0.98 (0.03)	0.81* (0.07)	1.47* (0.27)	0.91 (0.07)	1.10 (0.06)	0.98 (0.07)	1.17 (0.18)
Bottom-up
Similar national salience	1.01 (0.10)	0.81 (0.26)	1.14 (0.27)	1.21 (0.28)	0.81 (0.18)	1.22 (0.29)	0.97 (0.37)
Similar policy interests	1.07*** (0.00)	1.06*** (0.01)	1.07*** (0.01)	1.07*** (0.01)	1.06*** (0.01)	1.07*** (0.00)	1.06*** (0.01)
Control variables
Seniority	1.01 (0.03)	1.05 (0.17)	1.30** (0.12)	1.00 (0.07)	0.86 (0.07)	0.93 (0.06)	0.88 (0.07)
Committee leader	0.96 (0.09)	0.76 (0.55)	0.92 (0.24)	0.59* (0.14)	1.12 (0.24)	1.05 (0.27)	1.23 (0.25)
National delegation size	0.91** (0.03)	1.09 (0.27)	0.83 (0.08)	0.82* (0.07)	1.13 (0.12)	0.91 (0.06)	0.82 (0.11)
Substitute member	0.62*** (0.04)	0.67* (0.12)	0.62** (0.10)	0.52*** (0.08)	0.60*** (0.08)	0.83 (0.12)	0.49*** (0.09)
Party group leader	0.99 (0.07)	1.07 (0.25)	0.63* (0.14)	1.11 (0.36)	0.82 (0.11)	1.02 (0.15)	1.96*** (0.33)
Vote absenteeism	0.91* (0.04)	1.20 (0.33)	0.51*** (0.08)	1.25^* (0.13)	0.92 (0.10)	1.05 (0.11)	1.13 (0.18)
AIC	6566.99	523.34	874.09	1371.27	1321.37	1484.11	856.76
No. of appointments	1854	245	317	273	358	254	308
No. of observations	23737	1026	2144	6305	3546	8274	2152
Missings	2735	176	244	667	243	966	310

Notes: Statistical significance: ***P < 0.001, **P < 0.01, *P < 0.05; the dependent variable indicates the appointment of an MEP as shadow rapporteur for a particular codecision report; cell entries present odds ratios and standard errors are reported in parentheses; all model specifications include a stratification variable that captures the relevant choice set: MEPs of a particular party group in a particular committee being potentially appointed as shadow rapporteur for a report the committee is currently dealing with; the sample consists of dyads of rapporteurs and potential shadow rapporteurs from other party groups in the same committee.

ALDE: Alliance of Liberals and Democrats for Europe; ECR: European Conservatives and Reformists; EPP: European People’s Party; GUE-NGL: Confederal Group of the European United Left/Nordic Green Left; S&D: Progressive Alliance of Socialists and Democrats; GAL/TAN: Green-Alternative-Libertarian vs. Traditional-Authoritarian-Nationalist.

Regarding the top-down perspective, the results do not support the idea that the distance in positions between the potential shadow rapporteur and the party group median along the left-right or the GAL/TAN dimension matters in the selection of shadow rapporteurs. Only the variable measuring the distance along the pro-/anti-EU dimension shows the expected effect. However, the sub-sample analyses of individual party groups indicate that this finding is not robust: for three party groups, the effect is not statistically significant, and amongst the other three groups, one analysis (ALDE) reports a large statistically significant effect in the opposite direction.⁵ This variation in estimation results across party groups is difficult to reconcile with H1. The analyses also provide no support for an effect of ideological distance between the rapporteur and the potential shadow rapporteur, regardless of which ideological dimension is considered. H2 is thus not supported either.

With respect to the bottom-up perspective, coming from the same country as the rapporteur does not affect the probability of becoming a shadow rapporteur, which contradicts H3. Being involved in the formulation of reports with similar policy content as the rapporteur, however, has a positive effect on being nominated as shadow rapporteur. In fact, similarity in policy interests is the only explanatory variable that retains a consistently strong and statistically significant effect across all party group sub-sample analyses. H4 is thus supported by the data. As Figure 1 illustrates, keeping all other variables constant, a change from the 5th (−2.4) to the 95th percentile (46.9) of the similarity of policy interest variable increases the probability of becoming shadow rapporteur from 27% to 89% (i.e. by 62% points). The figures of bivariate relationships between shadow rapporteurship appointment and the main explanatory variables in the Online appendix demonstrate that these results are not affected by the inclusion of particular control variables or dependent on the assumptions of the statistical model.

Figure 1.

Effect of similar policy interests on shadow rapporteur appointment.

The results for the monadic control variables are mostly negative. Seniority and serving as committee or party group leader has no effect on shadow rapporteur selection. Estimates based on the full model indicate statistically significant results for national party delegation size and absenteeism. However, these results are not robust when party groups are analysed separately (i.e. in the context in which the nomination of shadow rapporteurs takes place). Only the type of committee membership shows a consistent and almost always statistically significant effect in the expected direction: not surprisingly, being a substitute rather than a full committee member more than halves the odds of becoming a shadow rapporteur.

Conclusion

This study sheds light on the selection of shadow rapporteurs, which play an essential and influential role in EP policy-making. Existing research on the appointment of MEPs to important positions in the EP often takes a top-down principal-agent perspective, assuming that the party group leadership selects suitable candidates from among its membership to ensure that the group’s policy positions are adequately represented and defended in intra- and inter-institutional negotiation processes. We contrast this top-down appointment with a bottom-up self-selection process, where taking up influential positions in EP committees is mainly a result of the policy motivations of individual MEPs. While it is analytically useful to differentiate between the two ideal-typical selection processes, we do not expect that real selection processes follow either template in their entirety. The question is not one of either/or, but one of degree.

The study’s findings from interviews with practitioners indicate that the assignment of shadow rapporteurs is largely a bottom-up process in which MEPs volunteer for particular reports on the basis of policy interests, and generally receive those reports if nobody else claims them. If there is a choice to be made, shadow rapporteurships are assigned in a ‘collegial’ process that takes into account factors such as policy expertise, relative workload, overall engagement in legislative activities, and a geographic balance in the EP’s negotiating team. Party group coordinators play an important coordinating role in this process and help shape outcomes; a limited top-down dynamic was thus highlighted in our interviews. Yet, not a single respondent mentioned ideological views as a selection criterion. Rather than imposing the position of the party group leadership, the role of the party group coordinator seems to be one of facilitating coordination and collaboration amongst party group members and ensuring the efficient use of group resources.

This conclusion confirms and complements the findings from our quantitative analyses, which found no consistent evidence for an effect of strategic considerations of the party group leadership on shadow rapporteurship appointments. Neither the ideological distance between party group members and the party group median, nor the ideological distance between party group members and the rapporteur are criteria for the selection of shadow rapporteurs, as suggested by principal-agent theory. In contrast, the quantitative evidence is consistent with a bottom-up self-selection process. Having policy interests similar to those of the rapporteur has a strong and robust positive effect on the probability of becoming a shadow rapporteur. This observed pattern is a direct implication of the bottom-up argument that engagement in policy-formulation work within committees is driven by committee members’ policy interests.

To the extent that policy interests go together with policy expertise, the finding of a positive effect of policy interests could also be interpreted as supporting principal-agent theory. One of the main reasons for principals to delegate tasks to agents is to take advantage of the latter’s higher level of expertise. However, the finding that divergent policy preferences have no effect on the nomination of shadow rapporteurs contradicts the core tenet and the very raison d’être of principal-agent theory, which problematises the moral hazard created by the divergent views of principals and their agents. Once preference divergence becomes irrelevant, principal-agent models reduce to simple, non-strategic optimisation models. Principals will then almost trivially choose the agent with the highest level of expertise from the pool of available candidates. The principal’s choice becomes one of selecting the most competent person for the job. Control and monitoring of agent behaviour, which is at the heart of principal-agent theory, is not an issue in these situations anymore. In this scenario, the question of why MEPs have different levels of policy expertise is of more causal relevance for the selection of shadow rapporteurs than the quasi-automatic ‘choice’ by group coordinators. This system of shadow rapporteurship allocation might very well induce MEPs to develop expertise in some policy area, but given the lack of top-down direction, the choice of which policy area to gain expertise in – and thus what type of reports will be allocated to them as a consequence – is primarily a function of individual MEPs’ motivations.

The qualitative findings that group coordinators do not select shadow rapporteurs based on their political views but based on their expertise and their general track-record of engagement in legislative activities is quite in line with this characterisation of the role of ‘principals’ in principal-agent models when preference divergence is absent. Furthermore, the qualitative findings show that the role of group coordinators is even more circumscribed than such a simple optimisation model would suggest. These findings suggest that, in game theoretical terms, a model in which group members move first and the group coordinator moves second would be a better representation of ‘real world’ selection processes. Group members first decide about whether to put themselves forward for a certain report, and then the coordinator selects one member from among the pool of volunteers. In this way, the behaviour of group members severely restricts the choice set of the group coordinator. In the limiting (but empirically common) case where only one member volunteers, the coordinator’s ‘choice’ is predetermined, and his or her role is redundant. But even if more than one group member volunteers, the coordinator only optimises over the set of candidates that actually put themselves forward, not over the entire range of party group members in the committee.

To conclude, the study results indicate that group coordinators are not a mechanism through which the party group leadership controls and monitors rank-and-file members, but rather a device to overcome coordination problems amongst group members (the hint is in their title). Party group coordinators act as mediators and ultimate arbiters when the generally bottom-up process of self-selection into shadow rapporteurships results in conflicting demands by group members and requires binding third-party adjudication. In this view, party group coordinators merely facilitate the bottom-up self-selection process, and they play this already rather limited role only in those relatively rare instances where such arbitration is required.

The quantitative analyses also show that shared nationality has no effect on shadow rapporteurship appointment, which could mean that policy motivations are more important than re-election concerns when it comes to MEPs selecting into shadow rapporteurships. Again, this finding is consistent with evidence from the qualitative analysis. Moreover, the qualitative analysis adds an important nuance to the finding’s interpretation. It suggests that there are limits to self-selection in that party groups strive for a geographic and national balance in the EP’s negotiating ‘teams.’ The very fact that geographical balance is mentioned by interviewees as a selection criterion suggests that national policy salience is an important motivating factor for MEPs to volunteer for shadow rapporteurships, but the consequences of these motivations are curbed by group coordinators’ desire to avoid national biases in the policy formulation process. While the application of this selection criterion constitutes a form of top-down control, it is not well represented by principal-agent theory either. Rather than a desire to prevent moral hazard on behalf of the selected group member, this form of coordination across party groups seems to be driven by norms of proportionality and representation.

The study’s general conclusion is that shadow rapporteurships in the EP are assigned in what is primarily a bottom-up process of self-selection based on policy interests. The appointment of shadow rapporteurs thus differs from the selection of rapporteurs, a process in which ideological closeness to the party median or party loyalty has been found to matter. Unlike in the allocation of reports, the role of party group coordinators in the selection of shadow rapporteurs is to help coordinate when more than one member requests a report, and to ensure that the subgroup of lawmakers to whom the committee delegates policymaking authority is not too overtly biased toward particular nationalities. They do not, however, assign shadow rapporteurships strategically based on group members’ policy positions (or party loyalty). Concerns over the faithful representation of the party group’s position in the policy formulation process, or the effective monitoring of and the extraction of concessions from the rapporteur, do not play a role in this selection process.

Hence, our research not only contributes to a better understanding of how MEPs are selected into a particular position of influence in EU law-making, it also adds to a growing body of work that cautions against assuming strong party control when it comes to the allocation of influential legislative positions in the EP: in addition to our findings about shadow rapporteurs, ideological closeness to the party median or party loyalty have been found not to be associated with the inclusion and placement of incumbent MEPs on national party lists (Wilson et al., 2016), committee assignments (Whitaker, 2019), or the selection into committee leadership positions (Chiru, 2019; Treib and Schlipphak, 2019; Whitaker, 2011), including as group coordinator (Daniel and Thierse, 2018). These findings do not negate the important role of legislative parties in EP politics, but they call into question the ability or willingness of EP party groups – and the national party delegations of which they are composed – to systematically use their sway over the allocation of important positions to enforce party discipline.

Supplemental Material

sj-zip-1-eup-10.1177_1465116520934865 - Supplemental material for Top-down or bottom-up? The selection of shadow rapporteurs in the European Parliament

Supplemental material, sj-zip-1-eup-10.1177_1465116520934865 for Top-down or bottom-up? The selection of shadow rapporteurs in the European Parliament by Frank M Häge and Nils Ringe in European Union Politics

Supplemental Material

sj-pdf-2-eup-10.1177_1465116520934865 - Supplemental material for Top-down or bottom-up? The selection of shadow rapporteurs in the European Parliament

Supplemental material, sj-pdf-2-eup-10.1177_1465116520934865 for Top-down or bottom-up? The selection of shadow rapporteurs in the European Parliament by Frank M Häge and Nils Ringe in European Union Politics

Footnotes

Acknowledgements

We would like to thank the editor of EUP and four anonymous reviewers for their constructive suggestions, as well as Michael Kaeding for very helpful discussions. Earlier versions of the paper were presented in 2019 at the European Studies Association Biennial Conference in Denver and the American Political Science Association Annual Meeting in Washington. We are grateful to Amie Kreppel, Stefanie Bailer, and other panel participants for their comments at these occasions. Finally, we would like to thank the MEPs, their assistants, and EP officials who took time out of their busy work schedule to participate in our interviews.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Frank M Häge

Supplemental material

Supplemental material for this article is available online.

Notes

References

Bakker

de Vries

Edwards

, et al. (2015) Measuring party positions in Europe: the Chapel Hill expert survey trend file, 1999–2010. Party Politics 21(1): 143–152.

Benedetto

(2005) Rapporteurs as legislative entrepreneurs: the dynamics of the codecision procedure in Europe’s Parliament. Journal of European Public Policy 12(1): 67–88.

Chiou

Hermansen

SSL

Høyland

(2019) Delegation of committee reports in the European Parliament. European Union Politics (online first).

Chiru

(2019) Loyal soldiers or seasoned leaders? The selection of committee chairs in the European Parliament. Journal of European Public Policy (online first).

Corbett

Jacobs

Shackleton

(2011) The European Parliament. London: John Harper Publishing.

Costello

Thomson

(2011) The nexus of bicameralism: Rapporteurs’ impact on decision outcomes in the European Union. European Union Politics 12(3): 337–357.

Daniel

(2015) Career Behaviour and the European Parliament. Oxford: Oxford University Press.

Daniel

Thierse

(2018) Individual determinants for the selection of group coordinators in the European Parliament. Journal of Common Market Studies 56(4): 939–954.

Epstein

O’Halloran

(1995) A theory of strategic oversight: Congress, lobbyists, and the bureaucracy. Journal of Law, Economics, and Organization 11(2): 227–255.

10.

Gilligan

Krehbiel

(1987) Collective decisionmaking and standing committees: an informational rationale for restrictive amendment procedures. Journal of Law, Economics, and Organization 3(2): 287–335.

11.

Häge

(2011a) Choice or circumstance? Adjusting measures of foreign policy similarity for chance agreement. Political Analysis 19(3): 287–305.

12.

Häge

(2011b) Politicising Council decision-making: the effect of European Parliament empowerment. West European Politics 34(1): 18–47.

13.

Häge

Ringe

(2019) Rapporteur-shadow rapporteur networks in the European Parliament: the strength of small numbers. European Journal of Political Research 58(1): 209–235.

14.

Hausemer

(2006) Participation and political competition in committee report allocation: under what conditions do MEPs represent their constituents? European Union Politics 7(4): 505–530.

15.

Hermansen

SSL

(2018) (Self-)selection and expertise among decision-makers in the European Parliament. Journal of Legislative Studies 24(1): 148–172.

16.

Hix

Noury

Roland

(2006) Dimensions of politics in the European Parliament. American Journal of Political Science 50(2): 494–520.

17.

Hix

Noury

Roland

(2007) Democratic Politics in the European Parliament. Cambridge: Cambridge University Press.

18.

Høyland

(2006) Allocation of codecision reports in the fifth European Parliament. European Union Politics 7(1): 30–50.

19.

Hurka

Kaeding

Obholzer

(2015) Learning on the job? EU enlargement and the assignment of (shadow) rapporteurships in the European Parliament. Journal of Common Market Studies 53(6): 1230–1247.

20.

Jensen

Winzen

(2012) Legislative negotiations in the European Parliament. European Union Politics 13(1): 118–149.

21.

Judge

Earnshaw

(2011) Relais actors and co-decision first reading agreements in the European Parliament: the case of the advanced therapies regulation. Journal of European Public Policy 18(1): 53–71.

22.

Kaeding

(2004) Rapporteurship allocation in the European Parliament: information or distribution? European Union Politics 5(3): 353–371.

23.

Kaeding

(2005) The world of committee reports: rapporteurship assignment in the European Parliament. Journal of Legislative Studies 11(1): 82–104.

24.

Kreppel

(2002) The EP and Supranational Party System: A Study in Institutional Development. Cambridge: Cambridge University Press.

25.

Lin

(1989) A concordance correlation coefficient to evaluate reproducibility. Biometrics 45(1): 255–268.

26.

Mamadouh

Raunio

(2003) The committee system: powers, appointments and report allocation. Journal of Common Market Studies 41(2): 333–351.

27.

Marks

Hooghe

Nelson

, et al. (2006) Party competition and European integration in the east and west: different structure, same causality. Comparative Political Studies 39(2): 155–175.

28.

Obholzer

Hurka

Kaeding

(2019) Party group coordinators and rapporteurs: discretion and agency loss along the European Parliament’s chains of delegation. European Union Politics 20(2): 239–260.

29.

Ringe

(2005) Policy preference formation in legislative politics: structures, actors, and focal points. American Journal of Political Science 49(4): 731–745.

30.

Ringe

(2010) Who Decides, and How? Preferences, Uncertainty, and Policy Choice in the European Parliament. Oxford: Oxford University Press.

31.

Ripoll Servent

Panning

(2019) Preparatory bodies as mediators of political conflict in trilogues: The European Parliament’s shadows meetings. Politics and Governance 7(3): 303–315.

32.

Roederer-Rynning

Greenwood

(2015) The culture of trilogues. Journal of European Public Policy 22(8): 1148–1165.

33.

Treib

Schlipphak

(2019) Who gets committee leadership positions in the European Parliament? Evidence from the 2014 selection process. European Union Politics 20(2): 219--238.

34.

Whitaker

(2011) The European Parliament’s Committees: National Party Influence and Legislative Empowerment. Abingdon: Routledge.

35.

Whitaker

(2019) A case of ‘you can always get what you want’? Committee assignments in the European Parliament. Parliamentary Affairs 72(1): 162–181.

36.

Wilson

Ringe

van Thomme

(2016) Policy leadership and re-election in the European Parliament. Journal of European Public Policy 23(8): 1158–1179.

37.

Yordanova

(2011) Inter-institutional rules and division of power in the European Parliament: Allocation of consultation and co-decision reports. West European Politics 34(1): 97–121.

38.

Yordanova

(2013) Organising the European Parliament. Colchester: ECPR Press.

39.

Yoshinaka

McElroy

Bowler

(2010) The appointment of rapporteurs in the European parliament. Legislative Studies Quarterly 35(4): 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.

2.34 MB

0.00 MB

0.41 MB