Working at home: Individual level drivers of MEPs’ focus on constituency work

Abstract

This article investigates how Members of the European Parliament (MEPs) allocate resources between their parliamentary and constituency work, focusing on their staffing decisions. Specifically, it examines the distribution of MEPs’ staff between accredited assistants, based in the European Parliament (EP), and local staff working in their constituency. The study seeks to explain why some MEPs allocate a larger share of their staff to constituency work compared to others by testing several hypotheses related to the individual characteristics of MEPs while controlling for variations across Member States. Using a newly developed database with data on MEP assistants at three points during the 9th EP, the analysis reveals that Member State differences do not solely explain the variation in staff distribution. While there is no evidence that staff allocation is driven by strategic considerations to secure re-election, it depends significantly on their ideology and past political experience.

1. Introduction

This article investigates how Members of the European Parliament (MEPs) allocate their resources between work in Parliament and in their constituency. It does so by examining the effect of personal characteristics on how they distribute their staff between the two. The work of MEPs takes place not only inside the parliament but also in the constituency that elected them (Norris 1997; Brouard et al., 2013; André, Gallagher and Sandri 2014). There is indeed evidence that MEPs can devote considerable attention and resources to constituency work (Sozzi 2016; Poyet 2018). This means that MEPs must make decisions about how to allocate their limited time and resources between work in the European Parliament (EP) and in their constituency. Therefore, there is a tension between the two kinds of activity (Norris 1997; Vivyan and Wagner 2016), and MEPs likely face a trade-off between allocating resources to each. This raises the question of how to explain the balance they struck between the two places. In other words, why do some MEPs give priority to work on their constituency, as compared to work within the EP, while for others it is the opposite?

This article addresses this question and aims to explain MEPs’ choice. More specifically, following the literature on national parliaments, we expect it to be driven by MEPs individual characteristics. This literature suggests two reasons for that. On the one hand, some authors suggest that members of parliament’s (MPs) commitment to constituency work is motivated by normative reasons, linked for example to their ideology, background, and past socialization, which cause this work to be more or less valued (Giger, Lanz and de Vries 2020; Dockendorff and Lodato 2023). On the other hand, another strand of this literature assumes that this choice is primarily the result of strategic calculations on the part of MPs with the aim of favoring their re-election (Tomkova 2014). This approach assumes that MPs become more involved in constituency work when they think it promotes their re-election odds (Norris 1997; Papp 2020). In the EP’s case, this means that MEPs determine their level of investment at either level on a cost–benefit basis, which depends in part on their personal characteristics. Those who see less personal benefit in engaging in constituency work do so less, and vice versa. This article investigates whether these factors also apply to the EP, and thus whether such strategic and normative reasons linked to their individual characteristics explain why MEPs give priority to one of the two types of work.

While in many respects the EP now functions as a “normal” parliament and can therefore be compared to national parliaments (Ripoll Servent 2018), there are nevertheless some specific features regarding constituency work that can have an impact on MEPs’ arbitration choices. First, most constituencies are national. They are therefore both more populous and more heterogeneous, which can lead to more diverse demands from their members (Tomkova 2014). Second, EP work is less politicized and, as European elections remain second order, the electoral link is weaker (Lord 2018). Consequently, constituency work may be less useful for re-election (Tomkova 2014). Third, the Eurosceptic minority within the EP means that some of its members object to its policies having any power, which may affect their work (Ripoll Servent 2018). While those points warrant caution in applying theories from the parliamentary literature, it also makes the EP interesting for testing hypotheses on legislative behavior based on literature studying national parliaments (Poyet 2018), especially since the varying importance of constituency work among MEPs is not yet fully explained.

Empirically, we address this issue by focusing on the decision by MEPs to allocate staff resources to constituency work relatively to parliamentary work. Indeed, MEPs generally employ two types of staff (Michon 2014): accredited assistants, working in the EP, and local assistants, working in their constituency. However, both types of assistants are funded from the same budget, which means that MEPs must choose how to allocate their budget between the two. As a result, the proportion of staff MEPs allocate to their constituency compared to the EP is used as an indicator to measure the priority they attach to constituency work compared to parliamentary work. This is especially true since local staff constitute one of the main resources at the disposal of MEPs for constituency work (Clarke 1978; Tomkova 2014; Rocca, Krebs and McArthur 2023). Empirically, the analysis is based on an original dataset, which includes the composition of MEPs’ team in May 2023. On this basis, we then test our hypotheses using regression analysis to assess the various explanations for the distribution of resources between constituency and parliamentary work.

The remainder of this article is structured as follows. The next section introduces constituency and parliamentary work in the EP. Section 3 then discusses local and accredited MEPs’ assistants and how the choice between the two reflects MEPs priority between the two types of work. Section 4 develops our hypotheses, based on the existing literature on national parliaments. Section 5 presents the data used for the analysis. This is followed by the analysis in Section 6, and the final section concludes.

2. The tension between constituency work and parliamentary work in the EP

MEPs, like MPs in general, engage in two types of work. On the one hand, MEPs work in the EP. This work is crucial for MEPs as it includes several tasks that are important both for the functioning of the European Union (EU), and therefore its citizens, and for the career prospects of MEPs (Ripoll Servent 2018). Foremost among these is the negotiation and development of EU policies and budgets. Second, they also scrutinize the Commission and other EU agencies. Third, MEPs can initiate, amend, and vote resolutions to express EP views on various topics. Finally, they can also take part in plenary debates, put oral or written questions to various actors, and manage the internal workings of the Parliament.

On the other hand, MEPs’ work also involves tasks that take place within the constituency that elected them. Broadly speaking, MEPs conduct two kinds of work within their constituency (Tomkova 2014; Gabriel et al., 2018): casework—that is, addressing constituents’ complaints, and communicating (in various forms) with their constituencies. Both are part of what is commonly known in the parliamentary literature as constituency work (Heitshusen, Young and Wood 2005; Dockendorff and Lodato 2023), that is the “activities MPs pursue to cultivate relations with their constituencies between elections” (Tomkova 2014: 19). Constituency work constitutes an important part of MPs work and is widely recognized as a key dimension of parliamentary mandates (Dropp and Peskowitz 2012; Poyet 2018), including by the constituents (Campbell and Lovenduski 2015; Brack and Pilet 2016).

The literature on national parliaments shows that not all MPs value the two types of work equally (Vivyan and Wagner 2016; Papp 2020). Some give more emphasis than others to their constituency and conversely. Moreover, however important it may be, constituency work requires MPs to expend resources, notably in working time and staff (Heitshusen, Young and Wood 2005). However, their resources are limited and therefore those invested in working in their constituencies cannot be invested in working in the parliament (Clarke 1978; Norris 1997). Consequently, constituency work has a potential opportunity cost, in that it can reduce the possibility of playing a relevant role in parliaments, where MPs can influence public policies.

Existing research on the EP suggests that MEPs value these two works differently, with some giving more attention to their constituency than others (Sozzi 2016; Chiru 2022; Hermansen and Pegan 2023). Yet, despite the relevance of constituency work for MEPs, the determinant of this choice is largely unknown. To date, little research has examined the activities of MEPs in their constituency as compared to their work in the EP, and even less has attempted to systematically explain differences (Brack and Costa 2018; Poyet 2018). Existing research on the MEPs’ relations to their constituency has largely focused on parliamentary activities, through analyses of the written and plenary questions they ask (e.g. Brack and Costa 2019; Chiru 2022). Hence, little attention has been paid to the work carried out within their constituencies, leaving the factors influencing the balance between these two types of work largely unexplored.

3. The allocation of staff allowances by MEPs as an indicator of their orientation between parliamentary work and constituency work

To perform their tasks, MEPs receive allowances, that is public funds that enable them to carry out the duties arising from their mandate. In particular, these allowances enable them to hire personal assistants (or staff). Personal assistants are recruited directly by MEPs; under contracts linked to their parliamentary mandate (Michon 2014). As with other legislators, the organization of MEPs’ staff reflects the division of labor between parliamentary work and constituency work (Tomkova 2014; Papp 2020). MEPs can indeed hire two types of personal assistants: those who work in Parliament (accredited assistants) and those who work in the constituencies (local staff). Each type of assistant performs distinct tasks specific, to each type of MEP work. This leads MEPs to hire assistants tailored to these specific roles (Michon 2014).¹ The respective numbers of each type of staff can therefore be seen as an indicator of the priority MEPs give to constituency work as compared to legislative work (Hermansen and Pegan 2023).

Accredited assistants work in the EP. They generally deal with the positioning of their MEP in the European political arena and in the conduct of the EP’s tasks, in particular policy-making (Busby and Belkacem 2013; Pegan 2016). In contrast, local staff carry out constituency work in their political territory (Michon 2014; Hermansen and Pegan 2023). They are a way for legislators to voice citizens’ grievances, cultivate their trust, and connect them with representative institutions (Tomkova, 2014; Kaslovsky, 2021). They can also act on behalf of MEPs and be the bearers of symbolic representation, for example when they attend events or hold office hours (Hermansen and Pegan 2023). Lastly, they help MEPs to communicate with their constituents on EU activities (Poyet 2018).

These two types of assistants differ not only in the activities they perform but also in their working conditions and labor costs (Michon 2014; Hermansen and Pegan 2023). Accredited assistants have an established contract with the EP and therefore relatively uniform and established salary rules. Constituency staff are employed under a private law contract drawn up directly with the MEPs and subject to their country’s legislation. Their remuneration—and therefore the number an MEP can hire—therefore varies between member states. While MEPs have thus some leeway in adjusting the number of both types of staff, it should be noted that this leeway is much greater for local staff than for accredited assistants. While the total number of accredited assistants is capped at four (and MEPs must justify the fourth), this is not the case for local staff, for whom there is no numerical limit, although local expenses are capped. Overall, the EP’s rules establish a malleable—but sticky—staffing system that leaves MEPs with considerable leeway to organize their team according to their own preferences² (Hermansen and Pegan 2023). The parliamentary teams of MEPs thus form veritable political enterprises around their elected representatives (Michon 2014, see also Salisbury and Shepsle 1981).

In the EP, local and parliamentary staff are financed over the same budget. This means that MEPs must make a choice as to where to allocate their assistant between their constituency and the EP. Elected representatives generally shape their teams according to their political agenda, particularly between the different political arenas (Mayhew 1974; Fenno 1977). As a result, the distribution of teams between assistants in the EP and in the constituencies differs considerably from one MEP to another (Michon 2014; Hermansen and Pegan 2023). While national difference in electoral rules and labor cost explain a great deal of this variation, the composition of the teams also reflects the priorities of the elected representatives (Fenno 1977; Tomkova 2014). MEPs build their teams “in their own image, according to what they have been, what they are and what they aspire to be” (Michon 2014). The allocation of MEPs resources between the two kinds of staff illustrates therefore MEPs’ broader trade-offs between constituency and parliamentary work (Tomkova 2014). This makes the balance between local and accredited assistant among MEPs an opportunity to empirically assess their behavior in that regard.³

Yet, little research has looked at the choice between local and accredited assistants. One exception is the work of Tomkova (2014). Using survey data of MEPs, she found that MEPs declare employing varying numbers of both types of assistants, but on average the same number in Brussels as in their constituency. Another notable work is the study by Hermansen and Pegan (2023), on the effect of electoral agendas on the number of MEPs’ local staff. They found that the number of local staffers increases before elections, confirming that MEPs carry out constituency work in part to promote their re-election. However, they focused on local staff, which represents only half of the overall picture at stake here. Given that MEPs do not have to spend all their budget, the number of local and accredited assistants is not necessarily correlated (as the data confirms). Therefore, examining the distribution of both kinds of assistants is necessary to examine and explain MEPs’ choices between parliamentary and constituency work.

4. Hypotheses

Broadly speaking, the existing literature on national parliaments identifies at least three alternative reasons for the differences in the distribution of parliamentary and constituency work by elected officials (Farrell and Scully 2010). First, electoral systems create different incentives for MPs to invest in their constituency (Cain et al., 1987; Bowler and Farrell 1993). Second, it can be due to cultural differences across parliaments in the expectations and demands placed on elected representatives (Katz 1999). Third, scholars also explain this variation by the individual characteristics of the MPs (Farrell and Scully 2010; Brack and Pilet 2016; Brack and Costa 2019), for example in the personal constraints they face in being re-elected, or in their sense of duty to different types of work. Regarding the PE, Michon (2014) suggests that these latter factors are crucial for the composition of the MEPs’ teams.

In this article, we focus on the third reason, the two others largely depending on MEPs’ member states. More specifically, we distinguish four individual characteristics of MEPs that may explain a greater focus on one of the two types of work. The first two are linked to normative reasons—MEPs basing their choices on certain values rather than separable consequences⁴ (Giger, Lanz and de Vries 2020; Dockendorff and Lodato 2023). The last two are linked to strategic incentives on the part of MEPs, depending on the functional constraints they face to be re-elected (Giger, Lanz and de Vries 2020; Papp 2020).

First, we expect the extent of MEPs Euroscepticism, that is their opposition or rejection of current and/or future European integration (Taggart 1998), to affect how they allocate resources between parliamentary and constituency work. In other words, we expect the degree of MEP’s opposition to the EU to influence whether they hire relatively more local staff or accredited assistants. The literature indeed argues that the choice between parliamentary and constituency work can also be the result of ideological motivations (Brouard et al., 2013). These motivations can be based either on symbolic or instrumental reasons (Crosson et al. 2020). It can be a way for MEPs to signal their commitment to less (or more) EU capacity or a way to reduce (or increase) the EU’s capacity for action by devoting less (or more) resources to its operation.

More generally, Eurosceptic MEPs tend to be less interested in participating in the work of the EP. Instead, they see the EP as a source of income and publicity, helping them to promote their ideas in their home country (Ripoll Servent 2018). They focus on the local and national levels, which they regard as the most important for democracy (Brack and Costa 2019). Hence our first hypothesis (h1): the more Eurosceptic an MEP is, the higher the proportion of local staff.

The second individual characteristic which likely affects the proportion of MPs’ local staff is their past local experience (Brack and Pilet 2016). Such experience in local politics influences how MEPs perceive their job and politics in general (Chiru 2022). MEPs with local responsibilities are more inclined to develop an interest in local issues (Norris 1997; Navarro 2015), as well as local ties with and loyalty to that level (Binderkrantz et al., 2019). This, in turn, drives them to invest more resources in constituency work (Borghetto et al., 2020; Sanches et al., 2024). In addition, they are more likely to consider returning to this level during their career (Russo 2011), and therefore have a greater need to be known, respond to, and therefore be aware of voters’ preferences. That is why in our second hypothesis (h2) we expect MPs with local experience to have a higher proportion of local staff.

The last two factors are linked to strategic considerations on the part of MPs to further their prospects of re-election. A great deal of the literature on parliaments assumes that MPs primarily seek re-election (Heitshusen, Young and Wood 2005; Papp 2020). From this perspective, they act strategically to achieve this goal. This means that MEPs would allocate resources to work in their constituency only to the extent that they see it as benefiting their re-election odds⁵ (Norris 1997; Dropp and Peskowitz 2012; Sozzi 2016). In addition to the electoral system, MEPs face other constraints which may explain why some believe that working in their constituency is more electorally profitable than working in the EP—while others do not, and therefore have more incentive to hire relatively more local staff (Poyet 2018). We identify two such reasons.

On the one hand, the number of previous mandates held by MEPs is likely to affect the need to invest in their constituency to be re-elected. This is a classic explanation for the differences in involvement in constituency work (André, Gallagher and Sandri 2014). Incumbents are likely to have less need to work in their constituency since their position is more established there (Papp 2020; Chiru 2022). They have already had opportunities to build a base of sympathy in their constituency and a reputation for listening at the local level (Heitshusen, Young and Wood 2005). On the other hand, new MEPs are probably less established and known to voters. They must therefore spend more resources within their constituency to develop their positioning and personal reputation therein (Tomkova 2014; Brack and Pilet 2016) and widen their basis of supporters (Russo 2021). Hence the third hypothesis (h3): “the more terms an MEP has already spent in the EP, the lower the proportion of local staff.”

On the other hand, the proportion of local staff may also depend on whether an MEP holds a leadership position. MEPs who occupy a leadership position in the EP, such as (vice-) president or committee chair, or in its political group, such as group chair, are expected to devote fewer resources to constituency work than rank-and-file MEPs. These leading positions are indeed likely to bring greater visibility, recognition as political experts, and influence within their party(Costa and Poyet 2016; Poyet, 2018; Russo, 2021). As a result, MEPs who occupy them must worry less about being selected as candidates and being re-elected and instead seek to defend their position in parliament or acquire more prominent ones (Russo 2021). Additionally, fulfilling and retaining these positions requires devoting more resources to parliamentary work, which might tilt the balance more in favor of hiring accredited assistants (Heitshusen, Young and Wood 2005; Gabriel et al., 2018). Hence, MEPs with leadership positions have fewer career-related incentives to serve their constituency and are more likely to pay less attention to it.

On the contrary, MEPs who do not hold such positions must compensate for the absence of the benefits they provide as they seek re-election. Strategically, one way of doing this is through constituency service, as having a strong local base and being recognized locally increases their chances of securing re-selection by their party and of being re-elected (Russo 2011, 2021). As a result, they have more incentive to prioritize constituency work by hiring local staff. Hence our fourth hypothesis (h4): MEPs with leadership positions within the EP hire a smaller proportion of local staff.

Finally, we control for three additional variables. First, we add a fixed effect for MEPs’ member states. It is indeed one of the main explanations for variations between staffing choices, for many reasons, that we seek to control for. For example, one of the most obvious is the difference in labor costs, which makes it possible to hire more—or less—local staff. More generally, differences in electoral rules and calendars as well as the distance between constituencies and parliaments are an important factor in explaining MPs’ constituency orientation. Second, we control for the size and extremism of their national party. These two factors are linked, respectively, to the voting power and coalition capacity of parties and can thereby affect parties’ ability to achieve favorable outcomes in the policy making process and, consequently, the need for their members to conduct constituency work (Ripoll Servent 2018; Brack and Costa 2019).

5. Data and method

Our dependent variable—the balance between local staff and accredited assistants within each MEP’s team—is measured by the proportion of local staff in a team. Specifically, this is calculated as the number of local staff divided by the total number of local staff and accredited assistants.⁶ The identities and functions of MEPs’ assistants are publicly accessible on the EP website, and we retrieved this information using web scraping in November 2023.⁷ Since 2009, MEPs have been able to share “grouped” assistants and staffers. When this is the case, we divide the value of the assistant by the number of MEPs employing that assistant. We obtained information for 700 of the 705 MEPs, as 5 of them had no information available on their page on the EP website.

To check the robustness of the analysis, we also carried it out at two other points in time during the 9th EP, to ensure that the relationships we found were consistent and not explained by omitted contextual factors. We collected data for 1 November 2020 and 1 May 2022. The first November was chosen to correspond to our original data collection (the first November 2023), while not being too early in the mandate so that MEPs’ teams had already been constituted and their operation in place. May 2022 falls just in the middle of the other two dates. We identified the MEPs in office on these two dates using the EP website.

Since the EP website does not provide information on changes in the teams of MEPs, we collected the information using “the wayback machine” website.⁸ This website allows the user to go back consult older versions of websites. In other words, it archives the different versions of a website. In our case, we selected the version of each MEP’s webpage which was the closest to the dates we chosen and retrieved the information on their assistants. It is important to note that we were unable to find the information for all MEPs. However, for both dates, we found information for around 90% of the MEPs and there is no substantial difference in the independent variables between the missing MEPs and the rest of the samples.

Regarding the independent variables, to measure MEPs Euroscepticism (h1) we rely on the 2019 Chapel Hill Expert Survey (CHES) dataset (Jolly et al., 2021). This dataset codifies the positions of national parties on several dimensions, including vis-à-vis European integration. Although the use of party-level data carries the risk of ignoring possible heterogeneity within parties (Carroll and Kubo 2017; Willumsen 2022), national parties attitudes toward European integration are generally shared by their MEPs (Hermansen and Pegan 2023) and existing research confirms that MPs’ preferences tend to align with those of their parties (Kukec, Kocijan and Verzichelli 2018; Hix, Whitaker and Zapryanova 2024). Therefore, in the absence of exogenous individual data for this term on MEPs’ ideology, such as surveys, it is standard in EP studies to use party-level data (e.g. Häge and Ringe 2020; Basedow and Hoerner 2023; Hermansen and Pegan 2023).⁹

We measure local experience (h2) using the dataset from Salvati (2022), which includes whether MEPs hold a political position at the local level. We manually completed the dataset for MEPs that entered the EP after its publication. Regarding our third independent variable, MEPs’ experience in the EP, we use the number of previous terms served by each MEP, which is available on the EP website. Finally, for our fourth hypothesis, we take into account two types of positions that are recognized in the literature on the EP as particularly relevant (Ripoll Servent 2018, see also Wilson, Ringe and Thomme 2016): (1) positions that are important for the functioning of the EP, i.e. chair or vice-chair of a committee, or a member of the EP Bureau, which is composed of the EP president, vice-presidents, and quaestors; or (2) leadership positions in political groups, that is, MEPs who are no rank-and-file members of their group.

Regarding the control variables, in addition to the member states of the MEPs, we use the number of seats held by its national party as a proxy for its weight. To measure MEPs’ extremism, we also rely on the CHES dataset (Jolly et al., 2021). We calculated the distance between MEPs’ national party position and the weighted average of the position of all parties in the EP for the two main dimensions of conflict within the EP: pro-anti EU and left–right.

6. Results

Fig. 1 displays the distribution of the number of local staff and accredited assistants. It confirms that MEPs make different choices when it comes to allocating their staff. The means are relatively similar (2 for local staff and 2.6 for accredited assistants), and both the number of local and accredited assistants varies, although the variation is smaller for the latter (standard deviation of, respectively, 2.1 and 1). This difference in variation, very probably due to the cap on the number of accredited assistants, confirms that MEPs have more flexibility in hiring local staff, as the number can go up to 12. Moreover, it is noteworthy that a significant number of MEPs (around 25%) do not employ local staff, while almost all MEPs have at least one accredited assistant.

More interestingly for us, Fig. 2 shows that the proportion of local staff also largely varies. This confirms that the allocation of resources between the EP and their constituency differs from one MEP to another. On average, the proportion of local staff is 35%, increasing to 47% when excluding cases without local staff. This aligns with Tomkova’s (2014) observation that, on average, the number of local staff and accredited assistants is similar in MEPs’ teams. However, this is only true on average, as the proportions largely vary across MEPs, as indicated by the distribution (standard deviation of 0.26–0.19 without MEPs with no local staff). It is also noteworthy that the correlation between the number of both kinds of staffs is low, both at the aggregate levels (correlation of −0.07, −0.2 without MEPs with no local staff) nor within member states (average correlation of −0.1, −0.17 without MEPs with no local staff, with exceptions obviously). This result confirms the relevance of examining the balance between both types of staff.

The left part of Fig. 3 displays the distribution of the proportion of local staff by Member states. This enables us to examine whether the observed variation is merely due to differences between member states, notably in terms of labor cost and electoral rules. The figure confirms that there are differences between member states, and thus they likely explain a great deal of the variation. However, member states do not explain all this variation, as the proportion of local staff also largely varies within member states. This supports a potential effect of individual factors and thus the rationale of our hypotheses.

The right part of Fig. 2 presents the distribution according to political groups, showing that members of different groups tend to make different choices in the way they allocate staff. Members of the RENEW and Green/EFA groups tend to have a higher proportion of accredited assistants, while the nonaligned, members of the ECR, I&D, and GUE/NGL groups exhibit a lower proportion of EP assistants. This difference already provides a first indication of an ideological motive behind staff allocations, as these groups are, on average, more eurosceptic.

In the remainder of this section, we test our hypotheses and attempt to explain the differences in the proportion of local staff between MEPs. To do this, we use zero-inflated beta regressions (Cook et al., 2008; Ospina and Ferrari 2011). Beta regression models are specifically designed to model continuous proportional variables, that are limited between ]0, 1[ (Ferrari and Cribari-Neto 2004). The zero-inflated part is added to account for the sizeable presence of 0 in the data. Indeed, beta regressions cannot model data that contain exactly 0 or 1, and zero-inflated models enable to model distributions with a large share of 0, such as this one. Zero-inflated beta regressions model a mixture of data-generating processes¹⁰ first a logistic regression model that predicts if an outcome is 0 or not and second a beta regression model that predicts if an outcome is between 0 and 1 for the cases that are higher than 0.

We could not obtain reliable confidence intervals and P-values using a frequentist framework with such a model. Therefore, we decided to fit the model in the Bayesian framework and derive credible intervals on the estimates. Conclusions are then drawn on these 95% credible intervals as we would in the frequentist framework, considering as significant all effects for which the interval does not include 0. Fig. 3 presents the estimate and the credibility interval for our independent variables.¹¹

Fig. 4 shows the effect of the independent variable on the probability of having at least one local staffer on the right-hand side, and on the proportion of local staff to accredited assistants when they have local staff on the left-hand side. Regarding the former, none of our independent variables has a consistently significant effect on the probability of having at least one local staff as compared to none. This means that none of our hypotheses are confirmed when it comes to the MEPs’ decision to allocate at least some staff resources to their constituency. In other words, we found no evidence that the personal characteristics we identified influence the probability of having local staff or not.

We now turn to the balance between the two types of staff. First, our two first hypotheses are significantly supported. On the one hand, more eurosceptic MEPs tend to hire a higher proportion of local staff. This positive effect is true for all periods but becomes more pronounced over time. While it is positive but not significant in November 2020, it is significant and increasingly important in the two following periods. This suggests that MEPs who oppose European integration tend to invest fewer resources in its functioning, instead focusing on other levels they support more, and conversely. It also indicates that this effect becomes more pronounced over the course of the parliamentary term, the difference between eurosceptic MEPs and those who support European integration increasing as their mandates progress.

On the other hand, MEPs with experience of local politics have on average, a significantly higher proportion of local staff across all three periods. This suggests that their links with this level make them more inclined to devote resources to work therein. Those results confirm that the choice MEPs make in allocating resources between constituency work and parliamentary work is partly driven by ideological considerations and political experience. Consequently, this means that the focus MEPs place on either work in their constituency or in the EP is influenced by their normative reasons on their own.

As far as strategic incentives are concerned, our hypotheses do not receive the same support. Our third hypothesis is not confirmed: the number of mandates held by MEPs has a significant effect, it goes in the opposite direction to the hypothesis put forward: the greater the number of mandates held by an MEP, the greater the proportion of local staff. This is true for the three periods. This is also the case when we binarize the variable into either the first term or not. One explanation could be that new MEPs, having less knowledge of the EP’s workings and politics, are compelled to hire a higher proportion of accredited assistants to manage their parliamentary tasks. In contrast, more experienced MEPs who are less reliant on such support can allocate more resources to work in their constituency.

Lastly, our final hypothesis—that the proportion of local staff is linked to the positions held by MEPs—is not supported. Holding a leadership position within the EP or a political group does not significantly affect the proportion of local staff. Therefore, MEPs’ preference for local staff over accredited assistants (or vice versa) does not seem to be driven by strategic incentives related to their (lack of) positions in the EP and the benefits those positions bring.

7. Conclusion

The aim of this article was to explore and explain how MEPs balance their work between the EP and their constituencies by analyzing how they allocate staff to these two places. Specifically, we examined how MEPs divide their staff between accredited assistants, who work in the EP, and local staff, who operate in their constituencies. To achieve this, we developed a novel database containing information on all assistants for all MEPs in 2023. We then tested several hypotheses related to individual factors to explain why some MEPs prioritize constituency work more than others, while controlling for differences due to MEPs’ member states.

As a result, we found that MEPs differ in how they allocate their staff between constituency and parliamentary work, and that this variation is not solely explained by differences between member states. Looking at individual determinants, we found no empirical evidence that strategic or normative incentives increase the likelihood that MEPs choose to employ at least one local staff member. Regarding the proportion of local staff to accredited assistants, we also found no evidence for our strategic hypotheses. First, there is no significant relationship between holding a leadership position and the proportion of local staff. Second, less experienced MEPs do not tend to employ a greater proportion of local assistants but, in fact, they tend to employ a significantly higher proportion of accredited assistants. In contrast, we find that MEPs’ staff allocations depend on their ideology and past experiences. First, the degree of Euroscepticism of MEPs is significantly and positively related to the proportion of local staff. Second, MEPs with local political experience tend to have a significantly higher (lower) proportion of local staff (accredited assistants).

These results therefore suggest that MEPs’ choice between parliamentary and constituency work is not only strategic but also based on other considerations (André, Gallagher and Sandri 2014; Brack and Pilet 2016). Some MEPs believe that investing resources at the local level is more worthwhile, while others prioritize the parliamentary level, not merely because they expect greater personal benefits. If one considers that both types of work constitute important parts of MEPs’ function, these results raise normative questions. Indeed, if investing less time in parliamentary work reduces MEPs ability to play a significant role in the EP, this may pose a problem for parliamentary decision-making. It would mean that MEPs with local ties or a certain ideology—and therefore their voters—would be less well represented in the EU decision-making (Hermansen and Pegan 2023). Similarly, if some categories of MEPs tend to invest significantly less in constituency work, this could create inequality in the way citizens can rely on this work in the EU. This is especially true since constituency work is part of the MEPs mandate and is generally perceived as important by EU citizens.

Overall, those results also mean that further research is needed on this issue. First, given that none of our variables were significantly related to the choice not to assign staff to their constituency, further work should be devoted to explaining this. Second, a limitation of this work is that we measure MEPs’ Euroscepticism at the party level. At the aggregate level, more Eurosceptic MEPs tend to hire a higher proportion of local staff, but this result may thus also partly reflect shared norms, strategies, or rules within Eurosceptic or Europhile parties rather than just MEPs’ personal ideologies (Raymond and Overby 2016). Future research could therefore disentangle to what degree decisions reflect each of these dimensions. Third, one could further study the reasons why politicians with local ties or eurosceptic MEPs make these choices, and what their rationales are. Similarly, to better assess the normative consequences of these choices, it is necessary to examine their consequences. In other words, how does MEPs’ allocation of assistants between the EP and their constituencies affect their work? Finally, one could also investigate the activities that local staff undertake in their constituencies. While we know that this type of staff is important to MEPs, we know very little about how they organize it. This would provide a better understanding of MEPs’ choices between constituency work and parliamentary work, and why they differ.

Acknowledgements

The authors would like to thank Tom Delreux, Marine Bardou, Ferdinand Teuber, Nicolas Arens, and two anonymous reviewers for their helpful comments and suggestions on earlier versions of this article.

Conflict of interest statement. None declared.

References

Footnotes