Lobbying cycles in Brussels: Evidence from the rotating presidency of the Council of the European Union

Abstract

The Council presidency holds direct responsibility for the Council’s functioning and moves between EU member states via a six-month rotation scheme. We argue that this rotating Council presidency causes a lobbying cycle among interest groups at the European level, whereby national interest groups from the country holding the presidency temporarily become active at the European level. Using a unique dataset including almost 16,500 registrations of interest groups in the European Transparency Register over the 2008–2017 period, we confirm that holding the Council presidency increases the number of interest groups from that member state in the Transparency Register. We also find that national interest groups generally have a higher likelihood to exit the register following the end of their country’s presidency.

Keywords

Council presidency European Union European Transparency Register lobbying

Introduction

Within the European legislative framework, the presidency of the Council of the European Union (henceforth ‘Council presidency’) holds a particularly important position. It not only involves the crucial managerial aspect of ensuring the smooth functioning of the Council but also entails considerable agenda-setting powers and a key role in coordination and mediation between member states (Häge, 2017; Hayes-Renshaw and Wallace, 2006; Tallberg, 2003; van Gruisen et al., 2017; Westlake and Galloway, 2004). As a fundamental institutional characteristic, the Council presidency is held by European Union (EU) member states on a six-month rotation system. In this article, we maintain that this rotating presidency has important consequences for the composition of the interest group population in Brussels over time. Interest groups are organisations that aim to influence policy outcomes ‘through frequent informal interactions with politicians and bureaucrats’ (Beyers et al., 2008: 1105–1106), and previous research has highlighted their remarkable ability to adapt when opportunity structures change (Mazey and Richardson, 1993; Richardson and Mazey, 2015). Viewing the rotating presidency as a prominent and predictable change in this opportunity structure at the European level, we argue that national interest groups are likely to become more active in Brussels during (or in the immediate run-up to) the Council presidency of their home country and have a higher probability to leave Brussels again shortly after their country’s presidency comes to an end.

From a theoretical perspective, such a ‘lobbying cycle’ among national interest groups might arise for a number of reasons. First, interest groups as rational actors focus their attention, energy and resources on where and when it matters most for policy formulation and implementation (Coen and Richardson, 2009; Mazey and Richardson, 1993; Richardson and Mazey, 2015). From this perspective, temporary shifts in power across EU member states deriving from the rotating Council presidency reflect a window of opportunity for national interest groups desiring to gain access at the European level. The reason is that the ‘ear of a sympathetic presidency can be a useful tool for an interest group wishing to promote a particular issue’ or ‘to prevent the reaching of an agreement’ (Hayes-Renshaw, 2009: 83). National interest groups thus have an incentive to take advantage of the rotating presidency’s considerable – albeit temporary – increase in their home country’s power within the EU. Second, from a demand-side perspective, the considerable workload attached to holding the Council presidency (Batory and Puetter, 2013; Hayes-Renshaw, 2009) may stimulate government actors’ reliance on national interest organisations to increase their capacities (Lewis, 2016) or tap into a natural and accessible source of relevant (technical) information and expertise (Jensen and Nedergaard, 2017). Government actors may also attract national interest groups to the European level during their presidency by acting as entrepreneurs for causes of particular concern to their national constituencies (Mahoney and Baumgartner, 2014), or via expectations of increased goodwill at the national level in return for interest groups’ support during the presidency (Jensen and Nedergaard, 2017). Finally, national interest groups are likely to have well-established long-term relationships with government officials in their home countries, which may ‘travel’ to – and become formalized at – the EU level during the Council presidency. Crucially, as we discuss in more detail below, the relative importance of these various forces may differ depending on the type of interest group under consideration.¹

We empirically test for the presence of a lobbying cycle in the population ecology of interest groups at the European level using a unique new dataset on the timing of entry and exit decisions of almost 16,500 interest groups in the European Transparency Register over the 2008–2017 period. Our main findings show that holding the Council presidency causes an increase in the number of interest groups from the presidency’s country in the Register. This upsurge is particularly pronounced among professional consultancies and non-business groups (such as non-governmental organisations (NGOs), think tanks and research institutions). Trade unions and professional associations are significantly more likely to register in the semester prior to their home country’s presidency. Furthermore, following the end of a country’s Council presidency, interest groups from that country generally have a lower likelihood of renewing their registration in the Transparency Register. Both findings are consistent with the presence of a substantively meaningful lobbying cycle in national interest groups’ presence at the European level linked to the rotating Council presidency.

Our analysis contributes to two main strands within the lobbying literature. First, a growing body of studies investigates the population ecology of interest groups – including their entry and exit decisions – at both the (sub)national (Binderkrantz et al., 2015; Gray and Lowery, 2000) and the international level (Berkhout et al., 2018; Berkhout et al., 2015; Berkhout and Lowery, 2010; Hanegraaff et al., 2017; Wonka et al., 2010). Such studies generally explain variations in interest group density over time in terms of socio-economic as well as policy demand factors (Bernhagen et al., 2015; Kluger Dionigi, 2017). We show that (temporary) shifts in national interest organisations’ opportunity structure due to the rotating Council presidency likewise impact on their entry and exit decisions at the European level. Second, within the European context, considerable attention has been awarded to national governments as a lobbying channel – particularly for lobbyists with predominantly national interests (Hayes-Renshaw, 2009). This literature finds that the consultation of interest groups is a common administrative practice at the national level, which also takes place ‘during the process of domestic preference formulation in EU-level policy making’ (Braun and van Der Berg, 2013: 763). Mazey and Richardson (2006: 264) posit that the value of this national channel varies according to ‘the policy issue, the type of interest group, the period in the policy process and the institutional structure of the government itself’. Our findings indicate that a country’s control over the Council presidency should be viewed as an additional element on this list. That is, the rotation principle of the Council presidency constitutes an institutional driver for national interest groups and this built-in feature of the European governance system critically affects interest groups’ decision-making.

Analytical framework and hypotheses

The Council presidency alternates in a predetermined six-month rotation between EU member states. This rotation can have notable implications for national interest groups’ activities and representation at the European level, since the Council presidency represents the most direct link between interest groups and the European institutions (i.e. via the national government holding the presidency; Hix and Høyland, 2011; Jensen and Nedergaard, 2017). Three main arguments can be brought forward to support this general proposition and develop empirical predictions about the nature of the presidency’s effect on national interest groups’ EU-level representation.

First, the Council presidency involves considerable agenda-setting powers and a central role in coordination and mediation between member states (Häge, 2017; Tallberg, 2003; van Gruisen et al., 2017). This ‘power of the chair’ holds the promise that a member state will have more influence on negotiation outcomes during its time as Council President (König and Proksch, 2006; Thomson, 2008; Warntjen, 2008). As such, it may be viewed as a valuable and predictable ‘window of opportunity’ for national interest organisations by creating a pathway towards influencing decision-making at the European level during a particular moment in time. This provides a strong incentive to enter the European lobbying arena during the period one’s country holds the Council presidency. Clearly, however, this incentive disappears again when a country’s presidency ends. National interest groups can thus be expected to take the ‘European route’ to a larger extent when their country’s presidency improves their chances to influence European decision-making – and leave this route again when the window of opportunity for potential privileged EU access via ‘their’ Council presidency closes.

Second, holding the Council presidency comes with a considerable workload for national governments (Batory and Puetter, 2013; Hayes-Renshaw, 2009; Jensen and Nedergaard, 2017). From a demand-side perspective (Gray et al., 2005), this may increase government actors’ demand for the expertise, knowledge and assistance of national interest groups during the country’s presidency (‘government information needs’). Reliance on national interest organisations not only allows government actors’ temporarily to increase their capacities (Lewis, 2016), but these organisations also offer a natural and accessible source of relevant (technical) information and expertise during European-level negotiations (Jensen and Nedergaard, 2017). This is particularly important when national governments act as entrepreneurs for causes of particular concern to their national constituencies (Mahoney and Baumgartner, 2014) or wish to push the European agenda on issues most beneficial (or least harmful) to their own interests (Alexandrova and Timmermans, 2013; Ferreira-Pereira, 2008). National interest groups’ willingness to provide support during the presidency may thereby increase when it is expected to increase goodwill at the national level (Jensen and Nedergaard, 2017) or when additional resources are made available to them during their country’s six months in office (a key source of funding for many groups; see Beyers and Kerremans, 2007). Clearly, national interest groups’ support, information and/or expertise will no longer be required at the European level after the Council presidency ends. Hence, this line of argument reinforces the expectation expressed above that entry (exit) of national interest groups in the European lobbying arena increases during (after) the period in which their country holds the Council presidency.

Finally, from the perspective of national interest groups, the domestic policy domain may be of more immediate interest. National interest groups can not only rely on well-established, long-term relationships with government officials in their home countries but also will have more limited networks of EU officials and face more competition from other groups at this level. Nonetheless, previous research on the Europeanisation of domestic interest groups has shown that ‘Euro-level networks of domestic interests are substantially related to their structural location within the domestic realm’ (Beyers, 2002: 585; see also Beyers and Kerremans, 2012; Kanol, 2016). As such, pre-existing connections with national government officials may well ‘travel’ to – and become formalized at – the EU level during a country’s Council presidency. Naturally, this pathway to the EU closes again with the end of the presidency.

H1: Holding the Council presidency increases the number of national interest groups from that member state present at the European level.

H2: The end of a country’s Council presidency decreases the number of national interest groups from that member state present at the European level.

To differentiate between the three theoretical arguments brought forward – i.e. the ‘power of the chair’, ‘government information needs’ and ‘Europeanisation of domestic interest groups’ – it is important to observe that their empirical relevance may differ depending on the type of interest group. For instance, a member state’s control over the Council presidency will have very little impact on national interest organisations maintaining a permanent presence in Brussels. This includes, for instance, large private-sector companies as well as important trade and business associations employing their own lobbyists. Such organisations are usually active across a broad range of policy fields and arguably cannot afford to exit the European level due to the breadth of their policy engagement or the need to communicate European policy developments to their members. Given this long-term perspective on EU-level lobbying, such organisations are less likely to base their entry and exit decisions in the European lobbying scene on the rotation of the Council presidency. In sharp contrast, interest groups that work on a narrower range of policy fields or are more managerial in their outlook (Walker et al., 2011) may be particularly sensitive to the power member states derive from holding the Council presidency. These groups – most prominently including professional consultancies – are often committed to a more ‘business style of interest intermediation […] with a more combative, conflictual approach’ (Lahusen, 2002: 697). This gives them a keen interest in capitalizing on the temporary increase in their country’s power within the EU, since that is the time best suited to benefit their clients via improved access to the European institutions.

In a similar vein, governments’ information demands play a more prominent role for expertise-based or issue-focused national interest groups, such as NGOs and research and academic institutions. These groups generally focus on a narrow or specialized set of policy fields, and this expertise may be particularly valuable when preparing dossiers and developing arguments since it ‘would tend to come with a promise [of] objectivity and transparency’ (Trondal et al., 2015: 27). The direct empirical implication is that the effect of holding the Council presidency – in terms of changes in the number of national interest groups entering the EU-level lobbying scene – will be heterogeneous across different types of national interest groups.

H3: National interest groups with narrow and/or shifting policy field involvement show stronger responses to their country holding the Council presidency.

Note that our line of argument here implies that, although patterns of entry and exit from the EU-level lobbying scene could be similar across subsets of national interest groups, the mechanism driving these patterns might well differ. This is consistent with Bouwen’s (2004) argument about different types of ‘access goods’ possessed by distinct interest groups. We return to this in more detail during the discussion of our empirical results.

Empirical analysis

Data: European Transparency Register

We assess our hypotheses using a unique new dataset extracted from the European Transparency Register. This register was established in 2011 when the European Parliament joined the Commission’s Register of Interest Representatives (itself launched in 2008) and aims to represent one place documenting all interest groups seeking access to the European institutions (European Commission, 2011). Each entry in the register is provided with a unique identification number and contains detailed information about, among other things, the day of registration, the interest group’s name, legal status and location (including the exact address of its head office), as well as the type of interest group.² While the register is well known to have a number of weaknesses – including its voluntary and self-reported nature – these are of lesser importance to our analysis. The main reason is that we only make use of an organisation’s presence in the register at a certain point in time, its self-reported head-office location and its self-reported type. Although interest groups’ self-classification within the Register might raise particular concerns, Bloodgood and Tremblay-Boire (2017) argue that organisations have little strategic incentives to misrepresent their type. Likewise, we also cannot think of any reasons why they would mis-report their head-office location. Two additional elements make that the register can constitute a credible source for our specific purpose (Berkhout et al., 2018; Bloodgood and Tremblay-Boire, 2017; Greenwood and Dreger, 2013). First, several European institutions require a European Transparency Register number from interest groups that want to gain accreditation and access. This makes registration a de facto requirement for such groups. Second, registration is free and occurs via a simple online procedure, such that the barriers to registration are low.³

The European Transparency Register currently contains approximately 11,500 entries. The oldest dates back to June 2008, when the Commission’s Register of Interest Representatives was launched. Importantly, interest groups have to renew their registration on a yearly basis and are removed if they fail to do so. As a result, approximately 1300 de-registrations occur every year. This not only keeps the register up to date but also implies that the online version of the register at any given day no longer contains interest groups that did not renew their registration. Using historical versions of the Transparency Register at specific points in time, we are able to recover all interest groups that left the register since May 2013. Specifically, we were provided access by the Joint Transparency Register Secretariat to backup files containing the complete register in May 2013, July 2014 and July 2015, and downloaded the complete register ourselves in August 2016 and July 2017.⁴ By comparing entries across the different versions of the register, we can observe which interest groups did not renew their registration. We then retain them in our dataset along with their exact time of registration, and a time of exit approximated by the time period between two historical versions of the register. Our final dataset includes 16,462 interest groups entering the European Transparency Register (or its Commission-run predecessor) at some point since June 2008, of which 5183 exited the Register between May 2013 and July 2017. Our complete dataset contains 14,859 interest groups with headquarters in an EU28 country, of which almost 81% is located outside of Belgium. The latter observation strongly suggests that most entries concern national interest groups (see also Dür and Mateo, 2016).

Empirical models

Our analysis of interest group exits from the Transparency Register (hypothesis 2) is modelled at the organisation level of observation. It is based on discrete-time hazard models of the following form (with subscript i for interest groups and t for time)

E x i t_{i} {, t}_{=} a + β_{1} P r e s i d e n c y_{x x y y i} {, t}_{+} γ C o n t r o l s_{i} {, t}_{+} ε_{i}_{, t}

where Exit equals 1 if a group left the Register between two consecutive observations of the Transparency Register (say, July 2014 and June 2015), 0 otherwise. Each interest group contributes one observation for each year it is present in the register and is coded as having exited when it fails to reappear in future iterations of the Transparency Register.⁵ The key independent variable in this model is ‘presidency_xxyy’. This is an indicator variable equal to 1 when the interest group’s country of origin held the presidency in the period indicated by the subscript (i.e. the second half of year xx or the first half of year yy), 0 otherwise. This concerns Cyprus and Ireland for presidency₁₂₁₃, Lituania and Greece for presidency₁₃₁₄, Italy and Latvia for presidency₁₄₁₅ and Luxembourg and the Netherlands for presidency₁₅₁₆. National interest groups from these countries should be more likely to exit with the end of their countries’ Council Presidencies. This implies that β₁ > 1 (when expressed as odds ratios) would provide evidence in favour of hypothesis 2.

The set of control variables in this model, first of all, includes the natural logarithm of a variable counting the time periods included in the dataset (which is constrained to the period where we can observe exits). This controls for the fact that exit probabilities might change with the passage of time. Then, we also control for interest groups’ country of origin. This captures the possibility that interest groups in different sets of member states have different exit probabilities. Finally, we include separate dummies for different types of interest groups, which control for potential heterogeneity in distinct interest groups’ exit probabilities. Summary statistics are included in the Online appendix.

To assess whether holding the Council presidency causes an increase in the number of interest groups from that country in the European Transparency Register (hypotheses 1 and 3), we cannot rely on an empirical approach at the organisation level. The reason is that we do not know which interest groups exist in each country at each point in time, and thus cannot know the set of organisations ‘at risk’ of entering the register. In the absence of information about organisations not entering the register, it is unfeasible to estimate an organisation-level probability of entry in function of a country’s presidency. Instead, we follow Van de Wardt et al. (2017) in evaluating entry by means of a count variable at the country level and estimate the following negative binomial count model

\begin{array}{l} E n t r y_{c, t} = α_{c} + β_{1} B e f o r e P r e s_{c, t} + β_{2} P r e s i d e n c y_{c, t}_{- 1} \\ + β_{3} P r e s i d e n c y_{c, t}_{+ 1} + β_{4} A f t e r P r e s_{c, t} + δ_{t} + υ_{c, t} \end{array}

where subscript c stands for country and t for time (counted in semesters due to the six-month terms of Council Presidencies). Our dependent variable Entry_c,t is the number of interest groups from country c present in the Transparency Register in semester t.⁶ The key independent variables relate to the Council presidency of the country where an interest group’s headquarters is located. Specifically, ‘presidency_c,t_–1’ and ‘presidency_c,t₊₁’ are equal to 1 in the semester immediately preceding or succeeding the Council presidency of the interest group’s country of origin, while ‘BeforePres’ and ‘AfterPres’ are equal to 1 in all other semesters preceding or succeeding this Council presidency. The semester of the country’s presidency acts as the excluded reference category.⁷ We also include a full set of semester dummies (δ_t) that capture all time-specific factors affecting national interest groups similarly across countries, and country fixed effects (α_c) to control for unobserved sources of heterogeneity in registration decisions across countries. This is critical to obtain valid inferences and implies that our inferences are drawn from changes over time in the number of interest groups from each member state. Support for hypotheses 1 and 3 would be reflected in negative point estimates for coefficients β₁ to β₄. Summary statistics for all variables are included in the Online appendix.

Results

Entry into the European Transparency Register

The key findings for the entry analysis are summarized in Table 1. Column 1 includes all entries in the Transparency Register independent of the type of interest organisation. Subsequent columns focus on particular types of interest groups. Specifically, column 2 only includes hired lobbyists (defined as professional consultancies, law firms and self-employed consultants), column 3 looks at private businesses as well as trade and business associations, column 4 covers unions and professional organisations, while column 5 includes non-business organisations (i.e. think tanks, research and academic institutions as well as religious and local public organisations).⁸ From the perspective of hypothesis 3, interest groups in columns 2 and 5 are least likely to have a permanent representation in Brussels since they work with a narrower and shifting set of policy fields or adapt their strategy depending on the demands of changing clients. Throughout the analysis, we restrict the sample to interest groups with their head office located in one of the EU28 countries, since non-EU countries naturally cannot hold the Council presidency. Yet, as interest groups from Belgium or with a Brussels office may not necessarily be domestic organisations, these are always excluded from the sample.⁹ Since our unit of observation is a country-semester, the maximum number of observations equals 513 (i.e. 19 semesters for 27 EU countries, excluding Belgium). Still, some columns have fewer observations due to missing data for certain countries and/or semesters.

Table 1.

Results of entry analysis.

	ALL	Hired lobbyists	Business interest groups	Unions and professional associations	Non-business interest groups
	(1)	(2)	(3)	(4)	(5)
BeforePres	−0.198** (−2.38)	−0.447*** (−3.30)	−0.107 (−0.91)	−0.122 (−0.50)	−0.195* (−1.94)
presidency_t _–1	−0.048 (–0.41)	–0.246 (–1.20)	0.027 (0.16)	0.573** (1.92)	–0.147 (–1.01)
presidency_t (ref. cat.)	–	–	–	–	–
presidency_t ₊₁	–0.149 (–1.33)	–0.253 (–1.37)	–0.042 (–0.26)	0.138 (0.46)	–0.227* (–1.65)
AfterPres	–0.175** (–2.15)	–0.252* (–1.94)	–0.029 (–0.25)	–0.075 (–0.33)	–0.263*** (–2.67)
Semester FE	YES	YES	YES	YES	YES
Country FE	YES	YES	YES	YES	YES
N N countries Wald test	513 27 1517.15***	513 27 537.45***	513 27 694.54***	486 27 359.72***	513 27 918.06***

Note: The analysis is based on panel negative binomial count models with the dependent variable equal to the number of interest groups from country c entering the Transparency Register at time t. The key independent variables relate to the timing of the Council presidency of the country where an interest group’s main office is located: ‘presidency_t_–1’ and ‘presidency_t₊₁’ are equal to 1 in the semester preceding or succeeding the Council presidency of the interest group’s country of origin, while ‘BeforePres’ and ‘AfterPres’ are equal to 1 in all other semesters preceding or succeeding the Council presidency of the interest group’s country of origin (the semester of the country’s presidency is the reference category). Each column focuses on particular subsets of interest groups: ‘Hired lobbyists’ consists of consultancies and law firms, ‘Business interest groups’ includes private firms and business/trade associations, ‘Unions & professional associations’ are self-explanatory and ‘Non-business interest groups’ cover NGOs, think tanks, academic institutions, and religious and local public organisations. *** is significant at 1%, ** at 5%, and * at 10%.

Looking first at the aggregate results in column 1, we find that the point estimates for coefficients β₁ to β₄ are all negative. Moreover, they are statistically significant at 95% confidence when interest groups’ country of origin is multiple semesters away from holding the Council presidency (relative to when this country controls the presidency). This implies that the number of interest groups from a country in the European Transparency Register tends to be lower before and after that country holds the Council presidency. As the coefficient estimates reflect the difference in the logs of the expected number of national interest groups when the indicator variable for holding the Council presidency goes from 0 to 1, they allow us to calculate the change in the incidence rate keeping all else constant. In terms of such effect size, the number of national interest groups in the Transparency Register on average increases with approximately 16% to 18% during the country’s presidency (compared to when interest groups’ country of origin is multiple semesters away from holding the Council presidency).¹⁰ Overall, therefore, these results provide substantial evidence in favour of hypothesis 1.

Nevertheless, the remaining columns in Table 1 indicate that this aggregate result obscures substantial heterogeneity in the effect of holding the Council presidency across different types of interest groups – in line with hypothesis 3. The strongest results are observed for hired lobbyists (i.e. consultancies and law firms), which appear in the Transparency Register in significantly lower numbers when their country of origin is multiple semesters away from holding the Council presidency (relative to when it controls the presidency). In terms of effect size, taking control over the Council presidency is linked to an upsurge of 22% to 36% in the number of interest groups from that country in the Transparency Register working with a project-based managerial approach. This is consistent with the theoretical argument that such organisations ‘shoot where the ducks are’ (Coen and Richardson, 2009; Richardson and Mazey, 2015) and are keen to capitalize on the temporary increase in their country’s power within the EU to the benefit of their clients.

In a similar vein, we find that significantly fewer non-business organisations from a given country are represented in the European Transparency Register in the semesters before and after that country’s Council presidency – relative to the period when their country of origin controls the presidency. In fact, the negative point estimates throughout column 5 again suggest that such organisations record their highest presence in the register during their home country’s Council presidency. In terms of effect size, the number of these national interest groups on average increases with approximately 20% during their country’s presidency. This is in keeping with the idea that public agencies and national administrations act as ‘brokers’ in interactions with such interest groups (Braun, 2012, 2017). These ‘brokers’ make it easier to contact selected interest groups with the required expertise and/or political resources and attract them towards the European arena during a country’s presidency (Jensen and Nedergaard, 2017; Mahoney and Baumgartner, 2014). With the end of the presidency, these interest groups’ ‘access goods’ become less valuable since their presence at the European level is no longer required (Bouwen, 2002, 2004).

Column 3 of Table 1 indicates that business interest groups do not respond to their country’s Council presidency – at least not in terms of their decision to enter the European Transparency Register. This may in part reflect the wide diversity of groups under this category, which covers large companies, sectoral groups and business associations. Yet, the mere presence within many such organisations of employees specifically hired and paid for their lobbying activities draws attention to their long-term perspective on lobbying activities and their activity in a broad set of policy fields – both at the EU level and elsewhere. As such, these organisations are most likely to maintain a permanent presence in Brussels (see also below), which helps explain why they do not respond to changes deriving from their country’s Council presidency.

Finally, column 4 in Table 1 suggests that unions and professional associations increase their European presence in the semester immediately prior to their country’s presidency. This does not entirely match our initial expectations. While it might reflect some degree of anticipatory behaviour in preparation of the upcoming presidency (Hayes-Renshaw, 2009: 82), Jensen and Nedergaard (2017: 45) also note that the last few months before the presidency are considered ‘the busiest period’ by those involved in the preparations. Hence, government actors’ demand for support, input and network building with designated national interest organisations might increase particularly ‘just before the execution phase of the presidency’ (Jensen and Nedergaard, 2017: 49). Our data unfortunately do not allow us to analyse this in more detail and we consider such potential anticipation effects an important avenue for further research.

Before we turn to interest groups’ exit decisions, it is worth highlighting that our analysis thus far looks only at a country holding the Council presidency. Nonetheless, since January 2007, member states holding the Council presidency work in groups of three over an 18-month period (called ‘Trio’ or ‘Tri-presidency’). While each country prepares its own six-month programme, the Trio-system aims to increase cooperation and longer term policy consistency across Council Presidencies (Batory and Puetter, 2013; Jensen and Nedergaard, 2014). Clearly, the Trio-system may prolong the period within which a member state plays a more prominent role in the EU legislative system (Van Gruisen et al., 2017). Hence, one might wonder to what extent the effects in Table 1 extend to a country’s membership of a Tri-presidency. The Online appendix indicates that membership of a Tri-presidency generally provides statistically insignificant results. Hence, from the perspective of national interest groups, the Council presidency is more important than being part of a Tri-presidency. This null result is consistent with Jensen and Nedergaard’s (2017: 67) finding that the ‘Trio function played a limited role for the execution of the [Danish and Polish] presidency’, and that most Presidencies thus appear mainly to follow their own six-month programme.

As a robustness check, we also replicated the analysis using information on the number of interest groups from country c holding an access pass to the European Parliament (EP) at time t. As only 15.9% of all interest groups in our sample (claim to) hold such an access pass for at least one employee, this drastically reduces the size of the available sample. The results in the Online appendix confirm that significantly fewer consultancies and law firms from a given country hold an EP access pass when their country of origin is multiple semesters away from holding the Council presidency. Hence, consistent with the findings in Table 1, demand for access to the EP increases when interest groups’ country of origin controls the Council presidency.

Exit from the European Transparency Register

In Table 2, we turn to interest groups exiting the Transparency Register and use discrete-time hazard models to assess hypothesis 2. The four columns differ mainly in terms of the sample under analysis. Columns 1 and 2 include all interest groups with their head office located in one of the EU28 countries, and differ in terms of the inclusion of country fixed effects in column 2. Columns 3 and 4 exclude interest groups from Belgium or with a Brussels office, respectively. Throughout the analysis, we provide separate estimates for the four sets of Council Presidencies where historical observations of the Transparency Register allow us to infer exits (see above). Note that we report coefficient estimates in terms of the odds ratio. This implies that point estimates below (above) 1 indicate a reduced (increased) probability of exit relative to the probability of no exit. Standard errors are clustered at the level of the interest group’s home country throughout the analysis.

Table 2.

Results of exit analysis.

	All interest groups	All interest groups	Exclude interest groups from Belgium	Exclude interest groups with Brussels office
	(1)	(2)	(3)	(4)
presidency₁₂₁₃	1.561*** (3.53)	1.234* (1.66)	1.222 (1.55)	1.217 (1.45)
presidency ₁₃₁₄	2.931*** (6.31)	2.458*** (13.79)	2.442*** (13.55)	2.526*** (9.79)
presidency ₁₄₁₅	1.831*** (6.95)	1.402*** (19.70)	1.407*** (19.16)	1.393*** (15.70)
presidency ₁₅₁₆	0.823** (–2.32)	0.911** (–2.38)	0.920* (–1.88)	0.871** (–2.09)
LnTime	1.271 (1.46)	1.281* (1.67)	1.225 (1.18)	1.131 (0.84)
Business interest groups	0.554*** (–8.01)	0.571*** (–7.26)	0.598*** (–6.46)	0.587*** (–6.79)
Unions and professionalassociations	0.397*** (–7.36)	0.384*** (–7.57)	0.410*** (–7.21)	0.380*** (–7.81)
Non-business interest groups	0.776*** (–4.63)	0.773*** (–4.28)	0.788*** (–3.37)	0.751*** (–5.32)
Country FE	NO	YES	YES	YES
N Nr. of exits Log likelihood	33,789 4579 –13,228.09	33,789 4579 –13,045.21	26,086 3812 –10,585.70	30,016 4438 –12,241.43

Note: The analysis is based on discrete-time hazard models with the dependent variable equal to 1, if an interest group left the Transparency Register at a given point in time, 0 otherwise. Coefficient estimates are provided in terms of odds ratios. ‘presidency_xxyy’ equals 1 when the interest group’s country of origin held the presidency in the period indicated by the subscript (i.e. the second half of year xx or the first half of year yy). This concerns Cyprus and Ireland for presidency₁₂₁₃, Lituania and Greece for presidency₁₃₁₄, Italy and Latvia for presidency₁₄₁₅, and Luxembourg and the Netherlands for presidency₁₅₁₆. LnTime is the natural logarithm of a variable counting the time periods included in the dataset. ‘Hired lobbyists’ consists of consultancies and law firms, ‘Business interest groups’ includes private firms and business/trade associations, ‘Unions & professional associations’ are self-explanatory and ‘Non-business interest groups’ cover NGOs, think tanks, academic institutions, and religious and local public organisations. The dataset is constrained to the period for which we can observe exits (i.e. July 2013 onwards). Standard errors are clustered at the level of the interest group’s home country. *** is significant at 1%, ** at 5% and * at 10%.

The results in Table 2 indicate that national interest groups from countries recently handing over the Council presidency are generally more likely to exit (or less likely to renew their registration in) the European Transparency Register – supporting hypothesis 2. The observed effect sizes are substantively meaningful in the first three rows of Table 2. The fourth row, however, indicates that interest groups from Luxembourg and the Netherlands appear roughly 8 to 18% less likely to exit when their country hands over the Council presidency. Assessing the exit probabilities across the available interest group types in more detail, we find that this unexpected result is driven by business lobbyists, NGOs and think tanks. One possible explanation may be that these interest group types from both countries are so well represented and well established at the EU-level that (the end of) a Council presidency hardly matters to them. From a population ecology perspective, this might reflect that a saturation point in the development of these countries’ organisational population has been reached (Hanegraaff et al., 2017; Hannan and Freeman, 1989). Given the limited time period for which we can infer exits in our dataset, we unfortunately cannot evaluate this tentative explanation in greater detail, which remains an important avenue for future research. In sharp contrast, however, the group of Dutch and Luxembourg consultancies and law firms is – in line with the remaining results in Table 2 – significantly more likely to exit following the end of their country of origin’s Council presidency. This striking divergence with other Dutch/Luxembourg types of interest groups is consistent with such organisations’ higher sensitivity to power changes within the EU system also documented in Table 1.

Before concluding, we should also point out that throughout Table 2, business lobbyists consistently show the lowest odds of leaving the Transparency Register (while the reference group of consultancies and law firms shows the highest odds of exit). This finding is consistent with our results in Table 1, where business lobbyists were found to be unresponsive to their home country holding the Council presidency. Taking both sets of results together, we thus obtain strong evidence that business lobbyists maintain a permanent presence in Brussels and remain in the European Transparency Register for the long term.

Concluding discussion

In this article, we have shown that the composition of the EU-level interest group population over time is characterized by a ‘lobbying cycle’ linked to the rotating Council presidency. This contributes to – and links together – the previously distinct literatures on (a) interest groups and lobbying activities in the EU; and (b) the Council presidency within European legislative politics. Our findings first of all shed new light on the unintended implications of the six-month rotation scheme imposed on the Council presidency. While this rotation explicitly intends to ‘bring Europe to the people’ via its member states, it clearly also brings professional consultancies to Europe – which from a normative/societal point of view is very different. This is important since the Council represents one of the EU’s main policy-making institutions, and its presidency has considerable powers to influence legislative decisions (Häge, 2017; Hix and Høyland, 2011; van Gruisen et al., 2017; Westlake and Galloway, 2004). A positive interpretation of the uncovered lobbying cycle might be that it provides a clear time for otherwise relatively peripheral domestic actors to make a meaningful contribution to the European policy process. A more negative interpretation, however, might view it as symptomatic of these actors’ conventional exclusion from the EU-level policy process when their country does not hold the presidency. Even so, a broader normative assessment would have to take into account the nature and intentions of the interest groups taking the ‘European route’ during their country’s presidency, which are likely to differ for hired lobbyists (i.e. consultancies and law firms) and non-business interests (such as NGOs). While our findings cannot make definite statements on this important point, more in-depth qualitative research is required to gauge the exact influence – if any – of the observed temporal shifts in the EU-level interest group population for European decision-making.

Our analysis also contributes to the empirical literature on the population ecology of interest representation at the European level. Such studies have generally explained interest groups’ entry and exit decisions based on economic and/or policy demand factors (Berkhout et al., 2018; Berkhout et al., 2015; Berkhout and Lowery, 2010; Bernhagen et al., 2015; Dionigi, 2017; Wonka et al., 2010). We show that member states’ control over the Council presidency also features prominently in the decision of national interest groups to be present – even if temporarily – at the EU-level. This complements recent work by Toshkov et al. (2013: 48) showing that ‘interest groups neither lead nor lag bursts in legislative activity in the EU’. Indeed, our results highlight that it is not necessarily European-level legislative activity per se that drives interest group entry and exit, but that member states’ control over the Council presidency plays an important role. Moreover, at least part of this effect appears driven by demand-side factors, whereby government actors attract national interest groups to the European level during their presidency (Mahoney and Baumgartner, 2014; Jensen and Nedergaard, 2017).

Naturally, our analysis has limitations, which point towards a number of avenues for further research. First, registration into the Transparency Register need not imply that interest groups also actively lobby EU-level policy-makers. An interest group could register without ever showing up in Brussels. Still, this does not affect the interpretation of our findings. The group’s registration decision would still respond to changes in the opportunity structure generated by the rotating Council presidency. It effectively wants the option to become active at the EU level, and this option already has some value. Nonetheless, future research should assess whether our findings also hold for more direct measures of interest group activity – including, for instance, scheduled meetings with Commissioners or Members of the European Parliament.

Second, registration in the Transparency Register is required for all activities ‘carried out with the objective of directly or indirectly influencing the formulation or implementation of policy and the decision-making processes of the EU institutions’ (Official Journal of the European Union, 2014). As mentioned, this explicitly excludes activities targeting the member states and their Permanent Representations. Many interest groups appear to make use of this legal loophole. A recent study by the Alliance for Lobbying Transparency and Ethics Regulation (ALTER-EU, 2016: 8) suggests that 7 to 21% of meetings with interest groups at the ‘Dutch, Polish, and Romanian permanent representations […] were held with lobbyists who were not registered in the EU lobby register’. This suggests that our results may only show the tip of the iceberg. The high visibility of the register data arguably generates a strong test of our theoretical propositions. Additional effects may well play through less visible channels available to interest groups beyond the Transparency Register.

Supplemental Material

EUP-17-1427.R1 replication - Supplemental material for Lobbying cycles in Brussels: Evidence from the rotating presidency of the Council of the European Union

Supplemental material, EUP-17-1427.R1 replication for Lobbying cycles in Brussels: Evidence from the rotating presidency of the Council of the European Union by Michelle Hollman and Zuzana Murdoch in European Union Politics

Supplemental Material

eup-17-1427-File002 - Supplemental material for Lobbying cycles in Brussels: Evidence from the rotating presidency of the Council of the European Union

Supplemental material, eup-17-1427-File002 for Lobbying cycles in Brussels: Evidence from the rotating presidency of the Council of the European Union by Michelle Hollman and Zuzana Murdoch in European Union Politics

Footnotes

Acknowledgements

We are grateful to the editor (Gerald Schneider), three anonymous referees, Jan Beyers, Joost Berkhout, Sara Connolly, Benny Geys, Hussein Kassim, Kirsten Lucas, Paul Stephenson, Anchrit Wille, Arndt Wonka and participants of the 24th International Conference of Europeanists (Glasgow, 2017) for insightful comments. The usual caveat applies. All research materials related to this paper (data and do-files) can be obtained from the journal’s website.

Funding

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

Supplemental material

Supplementary material for this article is available online.

Notes

References

Alexandrova

Timmermans

(2013) National interest versus the common good: The presidency in European council agenda setting. European Journal of Political Research 52(3): 316–338.

ALTER-EU (2016) National Representations in Brussels: Open for Corporate Lobbyists. Brussels: Alliance for Lobbying Transparency and Ethics Regulation.

Batory

Puetter

(2013) Consistency and diversity? The EU’s rotating trio Council Presidency after the Lisbon Treaty. Journal of European Public Policy 20(1): 95–112.

Berkhout

Beyers

Braun

et al . (2018) Making inference across mobilization and influence research: Comparing top-down and bottom-up mapping of interest systems. Political Studies 66(1): 43–62.

Berkhout

Carroll

Braun

et al . (2015) Interest organizations across economic sectors: Explaining interest group density in the European Union. Journal of European Public Policy 22(4): 462–480.

Berkhout

Lowery

(2010) The changing demography of the EU interest system since 1990. European Union Politics 11(3): 447–461.

Bernhagen

Dür

Marshall

(2015) Information or context: What accounts for positional proximity between the European commission and lobbyists? Journal of European Public Policy 22(4): 570–587.

Beyers

(2002) Gaining and seeking access: The European adaptation of domestic interest associations. European Journal of Political Research 41(5): 585–612.

Beyers

Eising

Maloney

(2008) Researching interest group politics in Europe and elsewhere: Much we study, little we know? West European Politics 31(6): 1103–1128.

10.

Beyers

Kerremans

(2007) Critical resource dependencies and the Europeanization of domestic interest groups. Journal of European Public Policy 14(3): 460–481.

11.

Beyers

Kerremans

(2012) Domestic embeddedness and the dynamics of multilevel venue shopping in four EU member states. Governance 25(2): 263–290.

12.

Binderkrantz

Christiansen

Pedersen

(2015) Interest group access to the bureaucracy, parliament, and the media. Governance 28(1): 95–112.

13.

Bloodgood

Tremblay-Boire

(2017) Does government funding depoliticize non-governmental organizations? Examining evidence from Europe. European Political Science Review 9(3): 401–424.

14.

Bouwen

(2002) Corporate lobbying in the European Union: The logic of access. Journal of European Public Policy 9(3): 365–390.

15.

Bouwen

(2004) Exchanging access goods for access: A comparative study of business lobbying in the European Union institutions. European Journal of Political Research 43: 337–369.

16.

Braun

(2012) The captive or the broker? Explaining public agency-interest group interactions. Governance 25(2): 291–314.

17.

Braun

(2017) The demand-side hypothesis of lobbying and interest group influence. In: Keynote presented at ACCESS EUROPE Symposium ‘Lobbying Influence: New research questions, new lobbying practices? 20 October 2017. Amsterdam: UvA University Library.

18.

Braun

van den Berg

(2013) At a snail's pace. On the limited Europeanisation of domestic consultation practices. Public Administration 91(3): 763–781.

19.

Coen

Richardson

(2009) Lobbying in the European Union: Institutions, Actors and Issues. Oxford: Oxford University Press.

20.

Dionigi

(2017) Lobbying in the European Parliament. The Battle for Influence. London: Palgrave Macmillan.

21.

Dür

Mateo

(2016) Insiders versus Outsiders: Interest Group Politics in Multilevel Europe. Oxford: Oxford University Press.

22.

European Commission (2011) Press Release: Commission and European Parliament Launch Joint Transparency Register to Shed Light on all Those Seeking to Influence European Policy. Brussels: European Commission. Available at: http://europa.eu/rapid/press-release_IP-11-773_en.htm?locale=en (accessed 11 July 2017).

23.

Ferreira-Pereira

(2008) Portugal and the 2007 EU presidency: A case of constructive bridge-building. Journal of Common Market Studies 46(s1): 61–70.

24.

Gray

Lowery

(2000) The Population Ecology of Interest Representation: Lobbying Communities in the American States. Ann Arbor, MI: University of Michigan Press.

25.

Gray

Lowery

Fellowes

et al . (2005) Legislative agendas and interest advocacy. Understanding the demand side of lobbying. American Politics Research 33(3): 404–434.

26.

Greenwood

Dreger

(2013) The Transparency Register: A European vanguard of strong lobby regulation? Interest Groups & Advocacy 2(2): 139–162.

27.

Häge

(2017) The scheduling power of the EU Council Presidency. Journal of European Public Policy 24(5): 695–713.

28.

Hanegraaff

(2015) Interest groups at transnational negotiation conferences: Goals, strategies, interactions, and influence. Global Governance: A Review of Multilateralism and International Organizations 21(4): 599–620.

29.

Hanegraaff

Braun

De Bièvre

et al . (2015) The domestic and global origins of transnational advocacy. Comparative Political Studies 48(12): 1591–1621.

30.

Hanegraaff

Vergauwen

Beyers

(2017) Should I stay or should I go? Explaining entry and exit of non-state actors at United Nations climate summits. Mimeo.

31.

Hannan

Freeman

(1989) Organizational Ecology. Cambridge, MA: Harvard University Press.

32.

Hayes-Renshaw

(2009) Least accessible but not inaccessible: Lobbying the council and the European Council. In: Coen D and Richardson J (eds) Lobbying in the European Union: Institutions, Actors and Issues. Oxford: Oxford University Press, pp.70–88.

33.

Hayes-Renshaw

Wallace

(2006) The Council of Ministers. Basingstoke: Palgrave Macmillan.

34.

Hix

Høyland

(2011) The Political System of the European Union. Hampshire: Palgrave Macmillan.

35.

Jensen

Nedergaard

(2014) Uno, duo, trio? Varieties of trio presidencies in the council of ministers. Journal of Common Market Studies 52(5): 1035–1052.

36.

Jensen

Nedergaard

(2017) EU Presidencies between Politics and Administration: The Governmentality of the Polish, Danish and Cypriot Trio Presidency. London: Routledge.

37.

Kanol

(2016) Europeanization of domestic interest groups. Interest Groups & Advocacy 5(2): 165–171.

38.

König

Proksch

S-O

(2006) Exchanging and voting in the council: Endogenizing the spatial model of legislative politics. Journal of European Public Policy 13(5): 647–669.

39.

Lahusen

(2002) Commercial consultancies in the European Union: The shape and structure of professional interest intermediation. Journal of European Public Policy 9(5): 695–714.

40.

Lewis

(2016) The European council and the council of the European Union. In: Cini M and Borragán NP-S (eds) European Union Politics. Oxford: Oxford University Press, pp.138–154.

41.

Mahoney

Baumgartner

(2014) Partners in advocacy: Lobbyists and government officials in Washington. Journal of Politics 77(1): 202–215.

42.

Mazey

Richardson

(1993) Introduction: Transference of power, decision rules and rules of the game. In: Mazey S and Richardson J (eds) Lobbying in the European Community. Oxford: Oxford University Press, pp.3–26.

43.

Mazey

Richardson

(2006) Interest groups and EU policy making: Organizational logic and venue shopping. In: Richardson J (ed.) European Union: Power and Policy-Making. London: Routledge, pp.217–237.

44.

Official Journal of the European Union (2014) Agreement Between the European Parliament and the European Commission on the Transparency Register for Organisations and Self-employed Individuals Engaged in EU Policy-making and Policy Implementation (L 277/11) . Brussels: European Commission. Available at: http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32014Q0919(01)&from=en (accessed 11 July 2017).

45.

Richardson

Mazey

(2015) Shooting where the ducks are. EU Lobbying and institutionalised promiscuity. In: Richardson J and Mazey S (eds) European Union. Power and Policy-making. London: Routledge, pp.419–444.

46.

Tallberg

(2003) The agenda-shaping powers of the EU Council Presidency. Journal of European Public Policy 10(1): 1–19.

47.

Thomson

(2008) The Council Presidency in the European Union. Responsibility with power. Gender, Work and Organization 46(3): 593–617.

48.

Toshkov

Lowery

Carroll

et al . (2013) Timing is everything? Organized interests and the timing of legislative activity. Interest Groups & Advocacy 2(1): 48–70.

49.

Trondal

Murdoch

Geys

(2015) Representative bureaucracy and the role of expertise in politics. Politics and Governance 3(1): 26–36.

50.

Van de Wardt

Berkhout

Vermeulen

(2017) Ecologies of ideologies: Explaining party entry and exit in West-European parliaments, 1945–2013. European Union Politics 18(2): 239–259.

51.

Van Gruisen

Vangerven

Crombez

(2017) Voting behavior in the Council of the European Union: The effect of the trio presidency. Political Science Research and Methods. Epub ahead of print 3 May 2017. DOI: 10.1017/psrm.2017.10

52.

Walker

McCarthy

Baumgartner (2011) Replacing members with managers? Mutualism among membership and non-membership advocacy organizations in the United States. American Journal of Sociology 116(4): 1284–1337.

53.

Warntjen

(2008) The Council Presidency: Power broker or burden? An empirical analysis. European Union Politics 9(3): 315–338.

54.

Westlake

Galloway

(2004) The Council of the European Union. London: John Harper Publishing.

55.

Wonka

Baumgartner

Mahoney

et al . (2010) Measuring the size and scope of the EU interest group population. European Union Politics 11(3): 463–476.

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.

7.09 MB

0.00 MB

0.09 MB