Living up to expectations? EU politicization and party Europeanization in Flanders and the Netherlands

Abstract

At the turn of the century, both academics and practitioners anticipated the Europeanization of national politics and political parties. One major expectation was that parties would adapt their organisation and behaviour to the existence of the EU and the functioning of its institutions. However, the early 2000s poured cold water on those expectations: the slacking politicization of EU affairs, it was concluded, created few incentives for parties to adapt, and so there was no meaningful Europeanization to speak of. EU politicization became the necessary pre-condition for party Europeanization. Today, however, that pre-condition seems (partly) fulfilled, as scholars are observing increasing EU politicization. Hence, we ask whether parties live up to expectations and, facing a politicised context, are today showing signs of party organizational Europeanization (POE). Based on a comparative case study of Dutch and Flemish parties, who function in differently politicized environments, we find that Europeanization remains limited. We conclude that it is likely not EU politicization holding parties back, but instead point towards the broader institutional misfit between national and European politics. We call on future research to further elaborate on this misfit, which might be the Achilles heel for EU democracy.

Introduction

Starting in the 1990s, an increasing number of Europeanisation scholars focused on different aspects of domestic change in response to the existence and development of the EU. In Ladrech’s (1994) seminal definition, Europeanisation entails “an incremental process re-orienting the direction and shape of politics to the degree that [EU] political and economic dynamics become part of the organisational logic of national politics and policymaking” (p. 69). While research focusing on the adaptation of domestic policies, parliaments, or executives proliferated (Ladrech 2012) there has been comparatively little attention for the Europeanisation of domestic political parties, and party organisation in particular. This party organisational Europeanisation (POE) entails the adaptation of the (in)formal power structures within and resource allocation (staff, money, external alliances) of parties, with the specific aim of dealing with ongoing European integration (Poguntke et al. 2007). This manifests in terms of stronger interconnections between the national and European level through, for example, better coordination between the party and its MEPs.

This organisational focus is important, as party organisations are still the essential link between citizens and government policies in modern parliamentary democracies. Given the EU’s major impact on the policy space, repertoire and instruments available for national governments and parties (Mair 2007; Ladrech 2012) one would expect parties to “adapt to the EU and organize themselves for participation in its institutions and rule-making” (Hix and Lord 1997, p. 5). From a normative perspective as well, POE is arguably desirable in the sense that parties should extend the democratic chain of delegation and accountability to the European level by establishing appropriate cross-level organisational structures. Ladrech (2007) argued that the absence of such organisational linkages contributed to an underdeveloped ideological lens to interpret EU integration, an absence of parties playing a connecting and educating role for public opinion about the EU, and a weak legitimacy of MEPs or transnational party federations.

Nonetheless, studies conducted in the early 2000s found little to no evidence of POE, rather observing an indirect strengthening of party elites and leadership through their government participation in EU decision-making (Raunio 2002; Poguntke et al. 2007; Carter and Poguntke 2010). At the time, this absence of POE was largely explained by a lack of politicisation of the EU as a domestic issue (Poguntke et al. 2007). Indeed, the second-order and (deliberately) de-politicized nature of European affairs meant that mainstream parties’ vote-, policy- or office-seeking goals were hardly affected (Ladrech 2012). Moreover, party elites also considered European affairs a liability capable of causing internal division if incorporated into existing party conflict structures (Mair 2000, 2007; Franklin and Van der Eijk 2004). As such, Ladrech (2002) pointedly argued that "there is little if anything in the way of resources that the EU possesses that can be translated into a positive gain for a political party" (p. 395). With nothing to gain and something to lose, POE was not deemed desirable or even necessary.

However, in the last 15 years there has been much talk about the politicization of European integration, although with much differentiation across time and countries (De Wilde 2011; Hutter et al. 2016). The aim of this contribution is, therefore, to assess whether these two aspects—politicization and POE—are interacting according to expectations. In other words, can we observe POE in today’s context of increased EU politicisation? To answer this question, we focus on mainstream political parties in two countries/regions—the Netherlands and Flanders (the Dutch-speaking region of Belgium)—which operate in differently politicized contexts. Methodologically, we build on a combination of quantitative indicators and qualitative interviews with party elites to assess degrees of politicization and POE.

EU politicization and party organizational Europeanization

The limits of POE in a non-politicized context

The increasing transfer of competences to EU institutions in the 1990s brought many to assume that this would lead to a Europeanisation of national politics and parties (e.g. Hix and Lord 1997). Ladrech (2002) speculated about possible areas of adaptation, ranging from programmatic changes in policy positions, to changing patterns of party competition. However, scholars were quick to discover that the adaptive capacity of parties and party systems was rather poor. As Mair (2000) noted at the turn of the century: “of the many areas of domestic politics that may have experienced an impact from Europe, party systems have perhaps proved to be most impervious to change” (p. 28).

This observation extended to party organisation. It was expected that parties would “adapt to the EU and organize themselves for participation in its institutions and rule-making” (Hix and Lord 1997, p. 5). But here too research revealed a profound lack of change. A large-scale comparative study by Poguntke et al. (2007) showed little evidence for far-reaching POE, as it remained unclear whether their main observation—increase in power for the party leadership—“is anything more than marginally connected to European integration” (Aylott et al. 2013, p. 15). This was in line with earlier research that showed that European integration increases the power and autonomy of national leadership, but also exposed a clear absence of leadership scrutiny of MEPs (Raunio 2002). In sum, they exposed parties acting “as if the EU were only a foreign policy matter, detached from core policy debates and other domestic activities” (Poguntke et al. 2007, p. 206).

At the time, the main argument for explaining this non-Europeanisation was the absence of an “overriding incentive for party elites to change” (Poguntke et al. 2007, p. 226). The permissive consensus on European integration (and hence a public neglect of the issue) combined with intra-party division over Europe within mainstream parties led mainstream party leaders to ignore or even deliberately downplay the EU as an issue, sustaining its second-order nature and low salience among the general public, and impeding parties’ organisational adaptation (Mair 2000; Franklin and Van Der Eijk 2004; Hooghe and Marks 2009). Put differently, Europe was not an issue on which parties could compete in a way relevant for electoral outcomes, while holding the risk of exposing internal division. Accordingly, the issue of Europe was taken out of the national arena and depoliticised (Mair 2000, p. 47), leaving very few repercussions for parties not monitoring European affairs and thus no real incentives for POE.

Based on these observations, Ladrech (2012) argued that reinserting the EU in the domestic arena, thereby breaking the permissive consensus and making Europe an issue of electoral consequence, would be conducive to change and may trigger POE. In others words: the moment the EU becomes domestically salient or politicized, a situation of misfit between public opinion and politics is created, serving as an enabling condition for Europeanization. Poguntke et al. (2007) similarly hypothesized that the hurdles for Europeanisation may erode in the face of the EU slee** giant waking up (p. 210). At the basis of these arguments lies the idea that parties need an external shock to be moved to action (Harmel and Janda 1994; Gauja 2017), which EU politicization is expected to be for POE. In other words, EU politicization was effectively theorized as a necessary or even sufficient condition for POE: a rise in EU politicization would shock parties into organizational Europeanization.

The differentiated politicization of European integration

Triggered by the failure of the Constitutional Treaty in 2005, the normative scholarly debate on the desirability of politicization of the EU turned into an empirical quest to measure the extent and types of conflict expansion surrounding the EU. The concept was defined (in operational terms) by De Wilde (2011) as “an increase in polarization of opinions, interests or values and the extent to which they are publicly advanced towards the process of policy formulation within the EU” (p. 566) and later argued to be constituted by three sub-processes: salience, actor expansion and polarization. This further deconstruction of the concept led to a variety of interpretations and ways of operationalizing politicization (cfr. Gheyle 2019).

Roughly speaking, a 2 × 2 matrix can be constructed, identifying four ways to study the politicization of European integration (see Table 1). A first distinction can be made between politicization as a visible, discursive phenomenon, and a more comprehensive outlook that also tracks the consequences of and potentialities for debate to emerge. In the former approach, issues have to be actively debated or contested in public, in front of an audience (e.g. De Wilde 2011; Statham and Trenz 2013; Hurrelmann et al. 2015) which is measured in parliamentary debates, protest events, or (most often studied) mass-mediated debates. In the other approach, polarization of public opinion, diverging party positions (or cleavage structures), or the rise of Euroskeptic parties, are taken as indicators that conflict on EU topics either has occurred (and had durable impact) or is susceptible to be picked up (Zürn 2016).

Table 1 Analytical framework for measuring politicization

A second distinction is one between a party-based or society-based view on politicization (Statham and Trenz 2013). In the former, politicization is present when political parties pick things up. Either in visible communication, i.e. in parliamentary debates, press releases, the media or during election campaigns. Or more structurally, looking at (diverging) party positions or the latent (instead of the short-term) salience they attach to the EU (as can be derived from expert opinions or manifestos). In a society-based view, politicization is not exclusively tied to political parties, but implies a more general expansion of the scope of conflict in society that is carried by a variety of actors. Here, mass-mediated public spheres are often the point of departure, where a variety of actors (besides parties) can make their views heard on EU topics. In practice, scholars either focus on one of the cells, or combine different ones in their interpretation of ‘politicization indexes’ (e.g. Rauh 2016).

Empirically, there is widespread evidence of EU politicisation in terms of visible conflict. Especially since 2008, scholars have studied particular highly politicized episodes (as measured in media debates, parliaments, or protest) such as the conflict over the Constitutional Treaty, the euro crisis, the Schengen crisis, free trade agreements and liberalisation policies, or the COVID-19 health crisis—also in Flanders and the Netherlands (Statham and Trenz 2013; Gheyle 2019; Hoeglinger 2016; Hutter et al. 2016; Oleart and Gheyle 2022). Given the ebb and flow of media and public attention, the phenomenon “can best be characterized as a process of punctuated politicisation, in which a significant but limited number of singular events produce high levels of political conflict for shorter periods of time” (Hutter et al. 2016 p. 283). Moreover, there is widespread recognition that EU politicisation differs across countries, with topics becoming salient and controversial in some Member States, yet hardly pop** up in others, depending on variation in political opportunity structures and resource mobilization (De Wilde et al. 2016; Zürn 2016).

On a more fundamental level, a cumulation of politicized episodes can have more lasting effects in terms of the overall salience parties attach to the EU, their positioning, or the awareness and opinions of the general public (row 2 in Table 1). In terms of public attitudes, for example, Rauh (2016) identifies increasingly polarized views over time on EU integration among citizens. Between political parties, there is growing party polarization in terms of positions on EU integration (driven most explicitly by the rise of challenger Euroskeptic parties) with more recent studies showing the increasing prominence hereof, with political conflict over European integration, immigration or trade hel** to restructure the cultural cleavage into a ‘transnational divide’ or a ‘demarcation-integration’ cleavage (Grande and Kriesi 2015; Hooghe and Marks 2018).¹ This notwithstanding, the overall salience (mainstream) parties attach to the topic remains rather low (e.g. Green-Pedersen 2019).

Overall, EU politicisation studies show that conflict over European integration (whether seen in its visible manifestations or its lasting consequences) comes in “differentiated forms, degrees, and manifestations […] depending on the time, setting and location in which it unfolds” (de Wilde et al. 2016, p. 15). It seems fair to conclude that European integration and related topics are certainly ripe for politicisation (given strengthening public and party cleavages), but that obstacles can remain in place for them to become more permanent features of visible party competition or public debate.

In sum, the Europeanization literature argued that the absence of POE observed in the early 2000s may to a large extent be attributed to a lack of politicization of European affairs, which led scholars to argue that it is a necessary condition for parties to adapt. Yet, we also showed that the potential for EU politicization has grown significantly over the past years. Hence our question: is the relationship between these two phenomena—politicization and POE—living up to expectations?

Case selection and research design

Case selection

Our study focuses on Flanders and the Netherlands for two main reasons. On the one hand, Flanders and the Netherlands constitute interesting cases from a comparative perspective. While comparable on a high number of elements (e.g. language, culture, party system, historical relation to European integration…), they have different experiences in terms of EU-related political events—and hence in EU politicization. Whereas in Belgium the EU has remained largely under the political radar, Dutch politics have been confronted with two EU referenda and increasingly vocal Eurosceptic politicians, with some even explicitly calling for a ‘Nexit’.² In the next section, we strengthen this argument of differentiated politicization by focusing on different indicators that show the variation between the two regions/countries.

On the other hand, as we argued earlier, there has been little empirical research on POE since Poguntke et al.’s (2007) major study. Studying POE requires a deep knowledge of the parties in question, and is difficult to gather at large (for a larger discussion of design issues in Europeanisation studies, please see Exadaktylos and Radaelli 2012). Therefore, we decided to focus on cases to which we have ‘easy access’ in terms of language and approachability, and of whose political systems we have a good understanding in order to contextualize findings in a sensible manner.

In terms of political parties, we focus on traditional/mainstream parties (basically: social-democrat, Christian-democratic and liberal parties). They have long been the drivers of European integration, with their elites involved at the highest European political level, but also of its domestic depoliticization. This means that, even if their programmatic emphasis on the EU was meagre (Mair 2007), we might at least expect that over time they have developed some kind of organisational connection with the European level and are aware of what is happening in the EU. As such, if we are to expect parallel evolutions of politicization and POE, this relationship is foremost expected for them. Moreover, one could also argue from a normative point of view that analysing mainstream parties is especially important, as their input and engagement with EU topics is necessary to ‘normalize’ debates and go beyond destructive and simple pro or anti EU debates (Braun and Grande 2021).

Research design

The main goal of this paper is to evaluate whether differences in terms of politicization result in similar differences in terms of POE. To tackle this question, we use a mixed method study design, building on a combination of quantitative indicators derived from the Chapel Hill Expert Survey (CHES) and Eurobarometer for studying EU politicization, and 27 qualitative interviews with party elites in the period 2017–2019 for the study of POE.

Regarding EU politicization, our focus is firmly on manifestations indicating the lasting effects of politicization and its potential for conflict to erupt (second row of Table 1). The reason has to do with the nature of the concept we link to politicization, namely changes in party organization caused by (the politicization of) European integration. Crucially, we do not expect that party organization shifts with the ebbs and flows of visible public debate, but that it rather correlates to more structural and durable changes in party systems and society. Hence, we look at how salient and polarized the EU is in the party system, and with respect to public opinion.

For the party-based indicators, we build on the one hand on the Chapel Hill Expert Survey (CHES) trend file 1999–2019 (Jolly et al. 2022). Information on the salience of the EU and party positioning is gathered by asking experts in the respective countries. Salience and positioning are, therefore, indirect measures, yet also valid ones as experts are assumed to combine general knowledge of party positions with real-time developments of what parties find important, hence producing a better overall picture than only focusing on, for example, party manifesto data (Netjes and Binnema 2007). However, to cross-check developments, we also include data from the Manifesto Project Database (MPD), which measures salience based on the occurrence of quasi-sentences in party manifestos (Burst et al. 2021). For the society-based indicators, we rely on standard Eurobarometer data, with a focus on two questions: one inquiring how often people discuss European political matters with friends (indicating salience of EU topics) and one inquiring which image the EU commonly conjures with them (ranging from very positive to very negative, to gauge positions and polarization).

Regarding POE, we take our cue from the operationalization presented in Poguntke et al. (2007), who in turn expand on previous studies (e.g. Hix and Lord 1997; Raunio 2000) and form the basis for later studies on EU party politics (e.g. Mühlböck 2017). Poguntke et al. (2007) operationalize POE as formal and informal organizational adaptation to European integration, focusing particularly on the increase in intra-party power for (1) party elites, who engage in top-level European negotiation and bargaining, and (2) ‘EU specialists’, i.e. “a heterogeneous group of actors … characterized by the fact that a considerable part of their political activity is related to the process or substance of European governance” (p. 11). Concretely, they looked at the inclusion of EU specialists in the party leadership bodies, EU specialists’ access to resources and EP candidate selection procedures, as well as the informal behaviour of these actors.

In this paper, we look at approximately the same indicators of POE, apart from EP candidate selection. The reason for this is that recent studies confirmed that EP candidate selection remains a strongly centralised and exclusive process (Euchner and Frech 2020). Instead, we focus on three indicators: (1) the extent to which party leadership bodies discuss European affairs (e.g. through the presence of MEPs), (2) the existence of specific cross-level liaison structures (as resources dedicated to managing European affairs), and (3) informal contacts across levels. One important element to consider is that we do not look at change over time, but rather conduct an analysis of the current state of POE in Flanders and the Netherlands. Nonetheless, aligning our indicators with those of previous studies allows our analysis to assess whether changes in EU politicization over the past years resulted in a new status quo regarding POE.

We use two main sources of information for our POE analysis: party statutes and semi-structured interviews. In total we interviewed 27 individual respondents. We selected four people from six mainstream parties in Flanders and the Netherlands, plus one person from each of the Europarties of which these national parties are a member (Table 2). National party respondents were selected to represent a variety of perspectives on how the party deals with European affairs, including EU specialists both in the EP and ‘at home’ (e.g., MEPs, international secretaries, MPs with an EU portfolio), as well as other party elites with a broad knowledge of the organization and functioning of the party (e.g., party directors, general secretaries). Europarty respondents were selected based on their familiarity with and expertise on relations with member parties. To safeguard the anonymity of the respondents, their names or any other information that allows them to be traced, cannot be divulged. A full list of respondent characteristics can be found in appendix.

Table 2 Overview of parties including in qualitative study

Politicization in Flanders and the Netherlands

To evaluate differences in POE between mainstream parties in Flanders and the Netherlands, we first strengthen the argument of the variation in politicization between these two countries/regions. This section is divided in three parts, the first and second parts party-based and societal politicization, while the third part focuses on parties’ internal division on the issue of Europe.

Party-based EU politicization

Figure 1 shows the salience of the EU averaged over all political parties, and displayed over time. With a maximum salience of 10, both Flanders and the Netherlands are in the middle range. However, post-2002, salience in the Netherlands has consistently been higher than in Flanders, floating around the 6-point line, while for Flemish parties, the average over time is about 4.5. This discrepancy can be attributed to two factors. First, both radical left and especially radical right parties in the Netherlands have consistently attached a higher salience to the EU. Radical right parties attach between 7 and 8.5 salience since 2006. In Flanders, the radical right VB maxed at (only) 4.8 in 2014, but otherwise scores lower. The second reason is D66, a left-liberal Dutch party established in the 60s that is often attributed salience scores above 7 because of its outspoken pro-EU stance. The high salience attributed to the EU of these parties hence suggests that the above-mentioned transnational cleavage is stronger in the Netherlands.

Fig. 1

Source Jolly et al. (2022)

Average EU salience in CHES.

On the basis of MPD, a similar picture emerges. On average, the percentage of sentences devoted to the EU post-2006 has been higher in the Netherlands (see Fig. 2). Yet if we track specific party positions (Figs. 3 and 4), we again see that in the Netherlands this is related to the extreme right PVV and (to a lesser extent) the left-liberal D66. In Belgium, it is only the VB and N-VA that show an increase in salience for the EU, but relatively speaking, to a low extent. All in all, one can arguably conclude that both countries resemble each other in terms of salience attached by mainstream parties, yet with the Netherlands differing due to the presence of extremes on the transnational cleavage (PVV and D66).

Fig. 2

Source Burst et al. (2021)

Average EU salience in MPD.

Fig. 3

Source Burst et al. (2021)

EU Salience in the Netherlands.

Fig. 4

Source Burst et al. (2021)

EU Salience in Flanders.

From CHES data we derive information on the positions that parties took on EU integration (ranging from 1 to 7, the higher the more supportive). Figure 5 shows that there is not much difference over time on average between Flanders and the Netherlands, but that Dutch parties are, on average, less supportive of EU integration than the Flemish parties. Again, this firstly has to do with radical right parties, where the position of the Dutch LPF, PVV, and FvD has been consistently lower (close to 1) than VB (around 2.5).

Fig. 5

Source Jolly et al. (2022)

Evolution in average positions.

Deconstructing this average reveals the almost complete consensus among Flemish parties on their pro-EU stance (Fig. 6). In the Netherlands, support is consistently lower, but also remains very high. Moreover, the Dutch averages hide individual differences, with again D66 broadly supportive (+ 6.5) and interestingly the liberal VVD as the least supportive before 2014 (lower than 4). After that date, it becomes more supportive again (+ 5). Similar to the salience figures, the positions of radical right (and left) parties in the Netherlands are consistently low (almost 1), while Vlaams Belang is a bit more supportive. The same intermediate conclusion appears: mainstream Dutch and Flemish parties are similar (with a slight exception for the VVD), yet the Dutch face a different situation with extremely pro and anti-EU parties (D66 and PVV/FvD) among their midst.

Fig. 6

Source Jolly et al. (2022)

Average positions of centre parties.

Society-based salience

For indicators tracking more durable changes in society, we rely on Standard Eurobarometer data. Figure 7 shows the evolution over time on the percentage of people that indicate they either ‘occasionally’ or ‘frequently’ discuss European political matters with their friends and family, which we take as a proxy for how salient the EU is in that country. The picture that emerges indicates that in the period 2010–2020 (which is the only time period this question was asked) more Dutch citizens discuss European politics than their Belgian counterparts, and this stays relatively stable over time. Deconstructing this further, we see that this mostly relates to the group of citizens that ‘frequently’ discuss European politics: 23% of Dutch citizens, compared to 10% of Belgian citizens. Society-based salience of the EU is therefore clearly different between the two areas.

Fig. 7

Source Standard Eurobarometer 95 - Spring (2021)

Percentage of people that discuss European political matters frequently/occasionally.

To this salience-based indicator, we also add information on positions and polarization by looking at the question which image the EU conjures for citizens. Taking together responses indicating a ‘very positive’ and ‘fairly positive’ stance (together seen as ‘total positive’) and subtracting from this the ‘total negative’, renders a ratio of percentages that indicates the general feeling towards the EU (kee** out neutral cases). This shows that Belgian citizens³ were much more supportive of the EU than Dutch citizens, up until the financial and eurocrisis (Fig. 8). After that, the image of the EU clearly deteriorated in both countries, roughly moving together ever since. Taken together, we conclude that Dutch citizens are more aware of the EU (politics) and have long been more critical than their Flemish counterparts.

Fig. 8

Source Standard Eurobarometer 95 - Spring (2021)

Ratio 'total positive' vs 'total negative' on general image of the EU.

Internal division

As argued above, mainstream parties have long depoliticized the EU (Braun and Grande 2021), partly because of internal divisions on the topic and ownership of the topic by more radical parties (Whitefield and Rohrschneider 2019; Brack and Startin 2015). Figure 9 shows how, especially in the Netherlands, traditional mainstream parties (and particularly VVD) show signs of significant internal division on the EU topic. Europe thus remains a difficult issue for mainstream parties to include in their own identity and positioning, especially because of radical parties on the right and left ‘own’ the Euro-critical or -sceptic positions and discourse.

Fig. 9

Source Bakker et al. (2020)

Internal dissent over EU issues.

All in all, these data not only indicate that European affairs remain cause for internal division, but also show a more politicized public environment in the Netherlands than in Flanders. On the one hand, societal salience is consistently higher in the Netherlands, while the Dutch public has also long been more critical than their Belgian counterparts. On the other hand, the indicators of party-based politicization show that Dutch mainstream parties face more competition of pro and anti-EU parties than Flemish parties. Whether this difference also leads to differences in party organizational Europeanization is discussed in the next section.

POE in Flanders and the Netherlands

The EU in leadership bodies

Key decisions are generally made by the party leadership, which can meet in various forms depending on the division of power within the party, such as a party board or parliamentary group meeting. Regardless of their form, these leadership bodies convene at least weekly and usually bring together the broad elite and various sections of the party, thus (theoretically) serving as ideal occasions for discussing and coordinating European affairs. No doubt, in all parties under study MEPs have a standing invitation to join the national parliamentary group meeting, while almost all statutes foresee in MEP membership of the board, either ex officio or made possible through elections. Both the Flemish and Dutch social-democrats prescribe a reporting duty MPs and MEPs alike directly to the party leadership, although in practice this occurs more ad hoc than systematically.

Indeed, statutes only tell half the story. In practice, leadership bodies hardly address European affairs, paying significantly more attention to domestic affairs. Accordingly, MEP absenteeism is common. While in some parties MEPs “try to have at least one of us present every time” (politician, personal communication), others delegate their assistants to be their eyes and ears, or do not show up at all:

In theory the interest for the EU is high, but you can see that in practice, they do not have time for it and the importance of the EU is very small. (board member, personal communication)

In fact, respondents indicate that these meetings predominantly address domestic politics, to the detriment of time spent on the EU. If MEPs attend the meeting, they can usually manage to briefly share what is happening in ‘Brussels’, but respondents generally confirm the marginalisation of European affairs in these leadership meetings. As one Flemish respondent said:

Europe is the end of the line; first there are Flemish and federal issues, then EU issues. … Only when things become very, very acute is there a big discussion, for example on Brexit, but that rarely happens. (board member, personal communication)

As such, one can hardly say that these leadership meetings show any sign of meaningful ‘Europeanisation’. Often caught up in domestic politics, they generally only address the EU at length when a European issue has become a politically relevant issue, which apparently does not happen very often. As a Dutch respondent clearly stated: "Europe is not exactly a subject on which you can score in terms of publicity; there is only a minor public interest … Although Brussels is close by, mentally it is very far away" (politician, personal communication)—which goes also for Flemish parties.

Cross-level liaisons

Although party elites have not fundamentally adapted their behaviour to European integration, this has not stopped them from investing additional resources in staying connected with the European level and its people in the EP. Many parties created (semi-)formal intermediaries, liaison officers such as a ‘European secretary’ aimed at staying on top of their European activities, although both the nature and institutionalisation of this role varies.

In Flemish mainstream parties (sp.a, OVLD, CD&V), being European or international secretary is a full-time position, often supported by a few assistants or an international office, while in most Dutch parties this position is combined with other responsibilities, such as board member (VVD), advisor (CDA) or MEP (PvdA). Yet, the biggest area of variation is the supposed role of these secretaries. In most Flemish parties, they formally perform the double role of managing the parties’ international relations (with sister parties and the Europarty) and staying on top of what its MEPs are doing in the EP, thereby acting as bridges between European and national politics. In Dutch parties, this latter role is absent—even though it formally exists within PvdA, in practice the distinction is not made.

Not incidentally, most Dutch parties speak of an ‘international’ rather than ‘European’ secretary. The CDA is an interesting case in this regard. Up until the 2014 elections it had both an international and European secretary, whose job specifically was to bring ‘Brussels’ closer to ‘The Hague’. After 2014 the position was scrapped for several reasons, disappointing electoral results and diminishing funds being one of them, but also the fact that the party leadership noticed increased internal tension over European affairs: “if you [mobilise] a lot of people [to discuss European affairs], then of course there is going to be some racket, and the leadership doesn't like racket” (staffer, personal communication). Hence, no more European secretary.

However, again the formal state of affairs does not necessarily reflect reality. Particularly, despite some formal variation, the fact is that (1) nearly all parties free up resources to deal with European affairs one way or another, but also that (2) these resources and their effect remain limited overall. As mentioned, in many parties being a European liaison is a part-time position at best, some simply add it to pre-existing positions or even rely on ‘full-time’ volunteers (sp.a). Moreover, even among parties with full-time liaison staff, respondents across cases indicate that, regardless of formal role, it is very difficult to get the party going on European affairs and even more so to get the party leadership to take liaising with the EP on a regular basis seriously. As a result, many of these European or international secretaries operate in a bubble of ‘EU specialists’ within their own party, only connecting with the broader party or leadership on the most important issues:

The party does not invest a great deal in trying to break that bubble. The party has limited resources and a limited number of people and mandataries at its disposal, so it directs those resources to those issues that are politically useful—and preferably immediately so—to the detriment of EU issues which usually aren't. (staffer, personal communication)

Informal cross-level contacts

With formal leadership bodies hardly addressing European affairs in a systematic and comprehensive way, and (semi-)formal liaison positions largely working in an EU bubble inside their party, it comes as no surprise that much of the interaction between the national and European levels occurs informally, often sporadically. Respondents in all parties declared that exchanges on European affairs usually happen in the margins of more formal gatherings, and that more direct interaction usually goes through phone calls or WhatsApp groups: "usually it is enough to just phone or text someone”; "it's really much easier to make contact through phone or e-mail" (different respondents, personal communication). Of course, this also means that interaction between levels depends heavily on the personal network and gravitas of the MEP or other ‘EU specialist’ involved: “If you're never there, you can't do that. … It requires also personal commitment” (staffer, personal communication). Consequently, socialisation, rather than coordination, is often the real aim of cross-level contacts:

The best way of getting [the national party] to pay attention is … making sure that everybody has each other’s phone numbers, that they know each other and have met each other. It makes a direct call so much easier. That's how you do it. You create a framework. They call, they text, they app. (politician, personal communication)

Importantly, these informal contacts increase in both frequency and intensity each time a ‘big moment’ happens at the EU level—for instance, a plenary session of the EP or a European Council meeting—and particularly when particular domestically politicized topics are on the agenda. In other words, national parties are in no way oblivious to what is going at the EU level, but organizational connections are only (and briefly) made in times of high salience. So far, our study finds very little traces of these peaks in informal contacts leading to more institutionalised linkages, let alone fundamental adaptation of the national party organization.

In sum, what can we say about the state of POE in Flanders and the Netherlands? Two things: there is no clear evidence for fundamental adaptation of party organization related to European integration, and organizational practices for dealing with European affairs is remarkably similar in Dutch and Flemish parties. Despite some formalistic differences stemming from different individual historical traditions and the institutional contexts in which they operate, parties in both Flanders and the Netherlands still struggle with unabated leadership focus on domestic affairs, European liaisons working in EU bubbles, and cross-level coordination based on informal and ad hoc contacts. There is no indication that Dutch and Flemish parties fundamentally differ in their organizational approach to dealing with Europe.

Moreover, although some respondents argue that minds have changed and that “the time you … could say 'them in Brussels' is over” (politician, personal communication), today this does not seem to have translated into lasting or fundamental adaptations of internal party organization or power relations. As such, our findings resonate strongly with Poguntke et al.’s (2007) earlier conclusions that “alignments of the supranational and national arenas are the exception” (p. 207) and that national parties “continue to practice party politics as if the EU were only a foreign policy matter, detached from core policy debates and other domestic activities” (p. 206).

Discussion

From a comparative analysis of indicators of (comprehensive) politicization in Flanders and the Netherlands, we concluded that Dutch mainstream parties found themselves in a much more politicized environment than Flemish parties, both from a societal or party-based perspective. Yet despite these differences in EU politicization we found that mainstream parties organize themselves in very similar ways to deal with Europe, and show limited signs of POE. As such, on the basis of these cases, we invalidate the stated hypothesis about EU politicization as a necessary or even sufficient condition for POE: despite clearer indications of EU politicization in the Netherlands, the Dutch parties exhibit equally limited POE compared to their Flemish counterparts.

These findings also lead us to question the often-heard argument that there simply is ‘not enough’ politicization to push mainstream parties towards organisational change. The indicators for societal salience and polarization clearly indicate that Dutch citizens are aware of the EU and that they have different and outspoken opinions on it. This potential is picked up by anti-European parties à la PVV and FvD, and even constituted potential for left-liberal parties like D66 to build a profile as pro-European party. Today, the Dutch party system hosts another openly pro-EU party (VOLT), clearly indicating the political uptake of the EU issue. As a rather pessimistic conclusion then, one could ask how much more we are expecting from citizens and dedicated EU-minded parties for mainstream parties to meaningfully engage with EU topics?

Yet we think that the answer can also be found on a more fundamental level, namely that there are major institutional misfits between national and European politics that withholds parties from organisationally adapting to the existence of the EU, regardless of whether it is politicized or not. This means the problem is not to be found on the demand side, but rather on the supply side. The main barrier for mainstream parties to Europeanize is not so much that citizens or competing parties do not demand this from them, but rather that there is a misfit between national and European politics that holds parties back from supplying that demand. Below, we identify two elements of this ‘Achilles heel’ of EU democracy that can arguably contribute to understanding why mainstream parties hardly respond to European integration.

First, even though national parties are the dominant political actors in their respective political systems, those same parties often have few opportunities to directly influence EU decision-making, which remains a hybrid constellation of EU officials, Member State representatives, and MEPs. In such a set-up, politicization often functions to strengthen intergovernmentalist actors, and executives in particular (Oleart and Gheyle 2022), emphasizing the absence of (domestic) party-political conflicts. Of course, in specific episodes and under certain circumstances, we might see national politicians step** in to cash on the temporary salience, but it is difficult to see how their engagement can be made more sustainable in an institutional set-up where their voice is institutionally limited. Even parties’ own MEPs demand a certain degree of autonomy to act on their European mandate independently from their national parties. More scrutiny by and institutional power for national parliaments could partly solve this issue, as it would give national parties and party elites a more direct and visible channel of influence on EU decision-making (Bellamy and Kröger 2014; Winzen 2021).

Second, the political reality of the ‘EU bubble’ does not translate very well into the political reality of the ‘national bubble’. As respondents indicated, both diverge immensely not only in terms of timing, but also in terms of topics discussed and actors involved. For instance, EU decision-making often takes years, whereas national politics follow the ups and downs of the news. Moreover, the EU’s compromise-driven politics are at odds with parties’ domestic competitive behaviour. A politicized environment does little to alter this institutional situation, and there is no easy workaround. Any solution would essentially involve introducing a majority-versus-opposition dynamic at the EU level, which not only raises democracy-related questions, but also conflicts with what the EU has been created to do, namely fostering compromise (Wolkenstein 2018).

In conclusion, given these institutional misfits between national and European politics, parties have few incentives to Europeanize their organization, regardless of EU politicization. In other words, as parties’ direct influence on the EU is limited anyway, there is no need to have complex and resource-intensive connections to the EU, no matter whether it is politicized or not. Our study indicates that these institutional misfits may be a greater barrier to POE than EU politicization. This qualifies the traditional politicization-based hypothesis: rather than politicization serving as a pre-condition for POE, we suggest the importance of institutional misfits that mitigate adaptation incentives (including, but possibly not limited to EU politicization). Driving this point home, in and of itself EU politicization can thus be considered as neither a sufficient nor even a necessary condition for POE. The EU might hence remain an elephant in the room for some time to come.

Future research will have to further study this institutional misfit and, particularly, how to bridge the gap between national and European politics that is preventing parties from Europeanising. As we suggested, giving a more prominent role to national parliaments in EU decision-making can be a fruitful avenue, but one has to consider also how this chimes with recent developments in the area of Europarties and transnational lists for European elections. Moreover, we have focused exclusively on the relation between EU politicisation and POE, while there is also a broader debate on what can explain party (organisational) change. The seminal work by Gauja (2017) highlighted a range of explanatory factors, such as electoral defeat or leadership change, which we do not include in our study. To come to a comprehensive understanding of what drives specific party organisational responses to European integration, future research will need to engage also with this broader literature.