![Sample our Area Studies journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5922/TOC-Subject-Sample-2021-Area-Studies.jpg"})
ABSTRACT
How much do Italian interest groups undertake their advocacy/lobbying activities at the EU level? How often have groups gained access to different EU level institutions? This paper presents an original conceptualisation for the concept of ‘interest group Europeanisation’, which takes into account both the percentage of EU lobbying and access to EU institutions, and assesses the role of national centrality (i.e. access to national institutional venues and self-perceived influence in national policy-making) in determining whether there is more or less interest group Europeanisation. Original data from a national survey conducted on around 500 Italian interest groups are provided. Groups that are at the core of the national interest system are less likely to undertake a large part of their lobbying activities at the EU level, but more likely to gain EU access.
Disclosure statement
No potential conflict of interest was reported by the author.
Supplementary Material
Supplemental data for this article can be accessed here.
Notes
1. More precisely, those countries are: Belgium, Germany, Italy, Lithuania, Montenegro, Netherlands, Poland, Portugal, Slovenia, Spain, Sweden and the UK.
2. In the literature, the distinction between ‘direct’ and ‘indirect’ lobbying is generally based on the fact that the former implies face-to-face interactions between groups and policymakers (either political or bureaucratic actors), whereas the latter implies grassroots mobilisation and/or media campaigns (Binderkrantz Citation2005). However, in this context ‘direct lobbying’ should be intended as any lobbying activity which is undertaken by national interest groups without involving Euro-associations.
3. However, Beyers and De Bruycker (Citation2018) and Greenwood (Citation2017) – when studying EU lobbying – focus on EU-level associations rather than national interest groups.
4. One of the foremost aims of groups is gaining access to EU institutions to influence policymaking (Bouwen Citation2004).
5. Organisational changes due to Europeanisation cannot be analysed in this paper: organisational changes generally need time to be implemented, yet this paper – being based on survey data carried out in 2016–2017 – does not present any diachronic analysis.
6. However, the choice to lobby at the EU level may be (also) constrained by issues’ characteristics and issue jurisdiction. Firstly, the more a particular issue is politically salient, as well as the more it implies high policy change, the more groups face incentives to develop a multilevel lobbying strategy (Beyers & Kerremans Citation2012). Secondly, interest groups lobbying on issues whose jurisdiction is (totally or partially) at the EU level are more likely to be ‘Europeanised’ than interest groups lobbying on issues whose jurisdiction is exclusively at the national level (Constantelos Citation2004).
7. Resource-rich associations can be expected to have the necessary resources to monitor and try to influence legislation at all relevant levels of governance.
8. At best, Roux and Verzichelli (Citation2010) investigated Italian economic elites’ judgement of the EU and European issues.
9. See, among others: Baumgartner et al. (Citation2009); Klüver (Citation2013); Dür, Bernhagen and Marshall (Citation2015).
10. What is extremely relevant here is that the amount of EU lobbying is not conceived in absolute terms, but in relative terms. It could be that groups at the core of the national interest system lobby the EU more than groups that are marginal in the national arena. However, it is the balance between national lobbying and EU lobbying that is expected to systematically vary between the former and the latter. More precisely, groups at the core of the national system are expected to focus their lobbying activities at the domestic level at the expense of EU lobbying, with the aim of taking advantage of their dominant position. On the contrary, groups which are nationally marginal are expected to focus their lobbying activities at the EU level at the expense of domestic lobbying, with the aim of compensating for their relative irrelevance in the national arena.
11. Yet, the direction of causality might also be reversed: it could be that European centrality impacts on national access as well as on influence on national policymaking. In other words, groups that get access to European institutional venues might think of themselves as more influential than their allies/rivals both at the supranational and national levels.
12. The decision to use the INTERARENA classification depends on the fact that – similarly to this article – that project uses an organisational definition of interest groups. The coding process was developed as follows: first, each interest group was coded separately by two different researchers (the author and a colleague who is unanimously regarded as an expert on interest groups in Italy); second, contradictory cases – i.e. interest groups included in different categories by the two coders (7 per cent of the whole sample) – were solved jointly.
13. This includes only the ‘business group’ category of the INTERARENA coding scheme.
14. This includes the categories ‘identity groups’, ‘religious groups’ and ‘public interest groups’ of the INTERARENA coding scheme.
15. This includes all the other categories – ‘unions’, ‘institutional groups’, ‘leisure groups’ and ‘occupational groups’ – of the INTERARENA coding scheme.
16. The first invitations were sent in mid-January 2017, while the online survey remained open until the end of March 2017. Over the course of this time period, four reminders were sent to non-respondents every two weeks in order to solicit their participation in the survey. This gave rise to five ‘waves’ of data collection, among which the first included the majority of respondents. Only legal representatives (namely, presidents or national secretaries) were allowed to answer the questionnaire.
17. Business groups: 38.1 per cent (135/354). Citizens’ groups: 43.3 per cent (165/381). Other groups: 32.8 per cent (178/542).
18. More precisely, this additive index depends on the answers to the following question(s): ‘During the last 12 months, how often has your group gained access to: a) ministers; b) parliamentarians; c) civil servants in national ministries?’. Also in this case, with regard to each institutional actor, respondents were allowed to indicate one of the options between (i) ‘we did not do this’ (= 1) and (v) ‘at least once a week’ (= 5). Thus, the independent variable ‘National access’ ranges from 3 (respondents indicating ‘we did not do this’ for ministers, parliamentarians and civil servants in national ministries) to 15 (respondents indicating ‘at least once a week’ for all institutional venues).
19. Respondents were allowed to select one of the following options: i) under 10,000; ii) between 10,000 and 50,000; iii) between 50,000 and 100,000; iv) between 100,000 and 500,000; v) between 500,000 and 1 million; vi) between 1 million and 5 million; vii) between 5 million and 10 million; viii) over 10 million.
20. Age is considered a proxy for the concept of ‘reputation’ on the basis of organisational ecology theory. Scholars following this approach argue that newly established groups face a higher threshold in being recognised as legitimate players by policymakers and other organisations (Hannan & Freeman Citation1993).
21. It could be questioned whether staff size is a good proxy for the concept of ‘expertise’: for example, a group could have a large staff size but only a few policy experts or a small staff comprised almost entirely of experts. However, this generally represents an exception rather than the rule: the vast majority of scholars, when operationalising expertise, share this operational choice (Baumgartner et al. Citation2009; Klüver Citation2013). The assumption in this case is that the higher the number of employees, the higher the possibility that those same employees are endowed with all the information and knowledge that are necessary in the policymaking.
22. Yet Benelux groups clearly profit from their geographical proximity to the EU capital Brussels.
23. The issue of missing values is not of secondary importance: the groups answering these questions may be systematically different from the groups in the rest of the sample. If this is the case, responding groups would not be representative of the sample as a whole. However, as the online Appendix (available at https://doi.org/10.1080/13608746.2020.1728958) demonstrates, there are no significant differences between groups responding to questions on EU lobbying and EU access, on the one hand, and the whole population of groups analysed in this study, on the other. Statistical results presented in Model 1 and in Model 2 are therefore representative of the whole population of interest groups.
24. With respect to Model 2, even with a high number of ordinal categories, an ordinal regression is the mathematically correct model specification. However, in this case the interpretation of the results would be messy given the dependent variable’s large number of ordered values; this is why both models are OLS regressions. Yet, to confirm that OLS results are robust, a robustness test has been conducted by running the same model specification as an ordinal regression. Empirical results do not differ significantly: both independent variables have the expected sign; ‘national access’ is highly statistically significant.
Additional information
Notes on contributors
Andrea Pritoni
Andrea Pritoni is an Associate Professor of Political Science at the Department of Cultures, Politics and Society of the University of Turin (Italy). He also collaborates with the Istituto Carlo Cattaneo and is the Coordinator of the Editorial Committee of ‘Politica in Italia’ (Italian Politics). His main research interests focus on Italian politics, comparative interest group politics and public policy analysis.