Sage Journals: Discover world-class research

Abstract

Gender equality in political representation is a cornerstone of contemporary democracies. Although much scholarly attention has been devoted to understanding gender inequality in government, less is known about gender representation within interest groups. We posit that gender representation in interest groups is influenced by the gender diversity of government officials. To test this hypothesis, we utilize a unique dataset that includes the gender composition of the entire population of government officials and interest group representatives in the European Union (EU). We examine it along with data from more than 4000 consultation meetings between lobbyists and EU officials. Using an instrumental variable design, we find that interest groups are more likely to engage female lobbyists when interacting with more gender-diverse policy-making bodies. Conceptually, this paper advances discussions on the link between political representation and gender mainstreaming. Empirically, it offers one of the first comprehensive analyses on the links between lobbyists access to government and gender.

Keywords

Gender representation EU lobbying shift-share instrument

Introduction

Despite a notable body of work that assesses the interaction between policy-makers’ characteristics and interest group access (Baumgartner et al., 2009; Getz, 1997; Hillman et al., 1999; Mahoney, 2004), the gender dimension remains largely out of focus, despite some notable exceptions (Celis et al., 2008; De Bruycker, 2024; Dunlop, 2024; Elomäki and Kantola, 2020; Hanegraaff et al., 2022; Junk et al., 2021; Lombardo and Verloo, 2009). As such, we know little about why/how gender impacts interest groups’ political strategies (Hillman et al., 1999; Oliver and Holzinger, 2008). Moreover, research leaves unaddressed theoretical discussions on interest group representation, gender mainstreaming, and political inequality (Krook and Mackay, 2011; Mansbridge, 1999; Waylen, 2017; Weldon, 2002).

In this paper, we ask whether gender impacts interest groups’ access to policy-makers. We argue that interest groups employ lobbyists with certain individual characteristics, including gender, to improve their chances to access policy-makers in key positions. To test our hypothesis, we focus on the European Union (EU). This political system consists of a complex institutional setting, with a very diverse interest group population. In addition, there is a rich literature that explores the conditions under which interest groups are able to gain access to EU institutions. This makes the EU an ideal case to test our argument.

We rely on a unique dataset that we construct containing information on: (i) the staff of all EU Commission Directorate Generals (DGs), akin to government departments; (ii) the lobbyists these policy-makers interact with (drawing from the EU’s Joint Transparency Register (JTR)). This dataset is complemented with information on the meetings that take place between these policy-makers and lobbyists. We use information from the censuses of various countries and employ machine learning techniques to extract policy-makers’ and lobbyists’ genders. The resulting dataset contains the gender composition of government officials and lobbyists in the EU, along with information on roughly 4000 meetings between them.

First, we match the interest groups registered in the JTR with the relevant DGs based on their policy areas of interest. Lobby groups must specify their policy areas of interest in the JTR. We test whether the DG’s gender composition affects the interest group representatives’ gender. We find a statistically significant and positive effect using a Bartik shift-share instrument, where we interact the gender composition of a DG in 2010 with the increase in the number of female staff in other DGs over time.

Second, we focus on the meetings between lobbyists and policy-makers, testing whether access to these meetings depends on the lobbyist’s gender. We find that employing a female lobbyist makes it likelier the interest group gains access to meetings. Interestingly, this holds true regardless of the gender composition of the policy-makers involved, suggesting an interesting mismatch between the demand and supply of gender representation.

The findings provide evidence indicating that gender is used strategically by interest groups to improve their access to decision-makers. These findings speak to the body of work in political science and management studies that examine the interaction between lobbyists’ individual characteristics, representation, and gender mainstreaming policies.

Gender mainstreaming as a lobbying strategy

Access to policy-makers is crucial for interest groups (Coen et al., 2021a; Levy and Egan, 2003; Mahoney, 2004; Reimann, 2006; Vogel, 1996), and hence the latter tailor their lobbying strategies to maximize access. Because access is dependent on trusted relationships between key individuals, interest groups adapt their strategies by adjusting staff while considering policy-makers’ characteristics (Walker and Rea, 2014). Interest groups take into consideration qualities, such as education, experience, and position (Baumgartner and Leech, 2001; Christiansen, 1997; Egeberg et al., 2013), and this is true also in the EU’s context. Moreover, companies tend to prefer individuals with work experience in the private sector who can easily relate to decision-makers on technical issues in similar policy areas (Coen et al., 2021b; Coen and Vannoni, 2016, 2020)

Whereas a substantial body of work examines interest group strategies vis-a-vis access to policy-makers, including research focussing on the individual level characteristics, there is less work untangling the impact of gender at the individual level. The main hypothesis put forward here is that the policy-maker’s gender will affect the lobbyist’s gender. Put simply, interest groups are likely to employ more female lobbyists if they deal with a more gender diverse set of policy-makers. Recent work has found evidence of this in the US: Strickland and Stauffer (2021) find that interest groups hire female lobbyists when working in more diverse legislatures.

Formally:

Hypothesis 1

(H1): Interest groups are more likely to use female lobbyists when dealing with female policy-makers

We argue that interest groups use gender as a heuristic to choose which individuals get more access (Strickland and Stauffer, 2021). There are three main reasons why female lobbyists are likelier to be given access by female policy-makers. First, gender diversity in politics may produce an environment more conducive to the emergence of women lobbyists and women’s issue groups (Celis et al., 2008; Wiener, 2021). This may lead to a positive representative spiral where there is overlap in the types of issues women frequently lobby for and legislate on. As such, interest groups may see women policy-makers as natural allies to women lobbyists and vice versa due to issue homophily (Ertug et al., 2022).

Second, policy-makers are likelier to give access to lobbyists with whom they share previous professional linkages (e.g. a lobbyist who interned at a policy-maker’s office) or issue based connections (e.g. being part of a common advocacy coalition) (Krook and Mackay, 2011; Strickland and Stauffer, 2021). Female policy-makers are more likely to share these links with female lobbyists. Hence, we might see that female policy-makers are more likely to work with female lobbyists, regardless of the policy area/issue or the government department (Kantola and Agustín, 2019; Minto and Mergaert, 2018).

Third, gender homophily may reduce information and search transaction costs that DG staff use to identify credible lobbyists to provide reliable information and expertise (Chalmers, 2013). Individuals tend to form groups with other individuals with whom they share common characteristics, gender being a key variable to be considered. This may act as a driver that enhances interactions between individuals who share this characteristic.

Data and measurement

We focus on the EU for theoretical and practical reasons. Theoretically, the EU represents a complex multi-level political system with high variation in structure and functions across its political institutions. In turn, this has an effect on the interest group population and its lobbying strategies. In the Appendix, we provide detailed descriptive statistics and a discussion on gender representation in the EU. In line with work on policy gendering, we find that more technical issues are dealt by male policy-makers and lobbyists, whereas ’caregiving issues’ see more female policy-makers and lobbyists.

To test our hypotheses, we need information on the gender distribution of staff in the different DGs in the European Commission and on the meetings between those staff members and lobbyists. Significantly, we require information on the specific characteristics of individual policy-makers and lobbyists including their gender and position. Furthermore, we require information on interest group type, budget, and overall staff.

The EC is the executive of the EU, drafting and proposing legislation. It works as a cabinet government, with 27 ministers (called Commissioners). Almost 33,000 civil servants work for the EC and are divided into Directorates General (DGs), akin to departments or ministries. There are two broad categories of civil servants: Directors (AD) and Assistants (AST). Directors develop policies, liaise with stakeholders, and take part in international negotiations, whereas Assistants have more of an executive role. Director General and Deputy Director General are the highest-ranking Directors and are roughly 100 in total.

For the policy-makers, we collect data on the gender composition of each Commission DG from 2010 to 2018 from its official website. We collect information on meetings that took place between EC staff and interest groups from November 2014 to February 2021. A total of 30,881 meetings were recorded (around 4000 took place in 2018 for which we have information on the lobbyists from the JTR). We collect information on: the decision-maker involved; their policy portfolio; the subject of the meeting; the organization that participates in the meeting.¹

We then match all this information with the data from the EU Joint Transparency Register (JTR) across different years. The JTR is the EU’s mandatory lobbying register: any organization that aims to lobby or meet any EU official must be registered on the JTR. The JTR contains detailed information on each interest group: the organization’s budget; the number of full-time employed lobbyists; the details of the government’s affairs representative; the legal representative; the lobbyists with accreditation to enter the European Parliament (EP); the headquarters’ address; the type of organization; the policy fields they are interested in.

In order to allocate a gender (male or female) to the lobbyists (i.e. the EU affairs representative and the lobbyists with accreditation to enter the EP from the JTR) and policy-makers, we employ the genderize R package.¹ This software predicts the gender of a person on the basis of their first name (and assigns a probability to that prediction), drawing on a database of first names from different censuses around the world. More information on this is provided in the Appendix.

From this procedure, we derive three measures of the gender of the interest group representative. The first measure is the gender of the EU representative, who is the main lobbyist employed by the organization and typically responsible for EU affairs as a whole. The second measure is the gender of the EP-accredited lobbyists, referring to those individuals accredited to enter the premises of the European Parliament. Since most organizations list more than one accredited individual, we calculate an average for this group. The third measure is an overall average derived from these two. For our analysis, we use data on individuals from the JTR for the year 2018.

Analysis

We run two analyses. The first analysis takes the ’interest group-policy area-DG’ as unit. We start from the interest groups registered in the JTR and we link them to the relevant DGs through their fields of interest (as registered in the JTR). It should be noted that most of the interest groups list more than one field of interest and that a DG can work on different policy areas. Moreover, a policy area can be the competence of two or more DGs. The following is the baseline model:

G e n d e r I n t e r e s t_{ipd} = α G e n d e r L e g_{d} + β X^{'} + δ_{i} + ϕ_{i} + γ_{p} + ε_{ipd}

(1)

where GenderInterest_ipd is the gender of the representative of interest group i in policy area p and DG d. GenderLeg_d is the average gender of the staff working for the DG d. We use fixed effects for: type of interest group (δ_i), its country of origin (ϕ_i), and the policy area (γ_p). We also include a vector of control variables related to the interest group (X′), specifically: the budget of the organization and the number of lobbyists.

The inclusion of fixed effects is very important, as we can exclude some sources of endogeneity and omitted variable bias (but not all). By including the policy area fixed effects, we control for the fact that female lobbyists and policy-makers might be attracted to work (i.e. they self-select) in the same areas, for instance. Yet, other sources of endogeneity and omitted variable bias might be present.

Therefore, we take a Bartik shift-share approach. We instrument the gender composition of a DG by interacting the shares of assistant and director female staff in that DG in 2010 with (the increase in) the number of female policy-makers in those categories in the other DGs from 2010 to 2018. The contemporaneous gender composition of lobbyists within an interest group in a specific policy area does not affect the gender composition of the DG working in that area a decade earlier, nor does it affect changes in the gender composition of other DGs. Yet, it is linked to changes in the gender composition of the DG of interest. In other words, we isolate the supply of female policy-makers from the demand.

This is in line with many current applications of shift-share instruments. For instance, Basso and Peri (2015) instrument the population of immigrants in a region with the population of immigrants by nationality at t0 across economic regions and the increase in these populations by the aggregate growth factor of immigrants from that nationality in the US between t0 and t. Acemoglu and Linn (2004) instrument the market size for a drug with the age profiles of users for each drug category and then compute the implied market size from aggregate demographics and income changes given these time invariant age profiles.

More formally, these are the two main terms (see equation (2)). Female_jp0/Tot_jp0 is the share of female staff in DG j in position p (we have data on administrators (AD) and assistants (AST)) at the beginning of the sample (at time 0, namely 2010). These are the so-called shares. Δ Female_j−1pt is the average increase in female staff in all the other DGs j − 1 in position p from time 0 to time t. These are the so-called shifters.

G e n d e r L e g I n s t r u m e n t_{j} = \sum_{p = 1}^{P} \sum_{j = 1}^{J} \frac{F e m a l e_{jp0}}{T o t_{jp0}} (Δ F e m a l e_{j - 1 p t})

(2)

We then instrument GenderLeg_d in 1 with GenderLegInstrument_jt in 2.

We perform some balance tests in the Appendix. We regress the shifts and the shares of the instrument on the contemporaneous value of the gender of the EU representative. We find a statistically significant relationship, but this relationship vanishes when using policy area fixed effect. This is in line with our research design. We can then claim that our instrument is exogenous to the outcome within policy areas.

The second analysis relies on the data on the meetings between lobbyists and EC policy-makers. We test whether the gender of the lobbyist is related to access to those meetings. More formally:

M e e t i n g_{idp} = α G e n d e r I G_{i} + γ G e n d e r L e g_{d} + β X^{'} + η_{i} + λ_{i} + δ_{d} + ϕ_{p} + ε_{idp}

(3)

where Meeting is whether interest group i attends a meeting with policy-makers from DG d in the policy area p. The main explanatory variable is GenderIG_i, which is the gender of the representative of the interest group i. GenderLeg_d is the average gender of the staff working for the DG d. We use fixed effects for: the DG (δ_d), the policy area (ϕ_p), the type of interest group (η_i) and its country of origin (λ_i). Finally, we include a vector of control variables related to the interest group (X′): the budget of the organization and the number of lobbyists.

Results

Results for the analysis using the Bartik shift-share instrument are shown in Table 1. We find a statistically significant and positive relationship between the DG’s gender (instrumented with the Bartik shift-share) and the lobbyists’ gender. We can reject the null hypothesis that the instrument is a weak instrument and we find a very strong first stage.

Table 1.

IV reg: gender DG (instrumented with shift-share) and gender EU representative.

	(1)	(2)	(3)
Gender DG	0.163^*** (0.039)	0.151^* (0.038)	0.395^** (0.079)
IG type	No	Yes	Yes
IG country	No	Yes	Yes
Policy area	No	No	Yes
IG covs	No	Yes	Yes
N	74,486	66,778	66,778

We find empirical support for our Hypothesis 1. This finding suggests that the relationship between the gender of policy-makers and that of lobbyists goes beyond the demand and supply factors seen above. For instance, this finding suggests that female lobbyists are more likely to get access to female policy-makers not (only) because they deal with the same issues, such as education and healthcare. In the Appendix, we provide the results also for the other interest groups’ representatives. Tables A3 and A4 respectively show the results when we use the gender of the EP representatives and the averages of the gender of the EU representative (used in the main specification) and that of the EP representatives. The effects indicate that a DG with only women is about twice as likely to be lobbied by a female lobbyist compared to a DG with only men.

The second part of the analysis investigates whether these strategies work in securing access to the decision-making process. Table 2 shows that gender is positively related to the likelihood of getting access to a meeting with an EC decision-maker. Results hold also when controlling for the gender of the DG with which the interest group interacts, with stable estimates across specifications. In other words, those interest groups that employ female lobbyists are more likely to get access to the EU decision-making process. Having a female EU representative increases the likelihood of a meeting by about 35 per cent compared to when the representative is male.

Table 2.

EC meetings: meeting and gender EU rep.

	Meeting
	(1)	(2)	(3)	(4)	(5)	(6)
Gender EU rep	0.303*** (0.030)	0.302*** (0.030)	0.343*** (0.030)	0.270*** (0.035)	0.260*** (0.035)	0.258*** (0.035)
Gender DG		X	X	X		X
IG covs			X	X	X	X
IG country				X	X	X
IG type				X	X	X
DG					X
Policy area						X
N	78,260	78,260	70,375	70,375	70,375	70,375

This is a rather interesting finding, suggesting that interest groups hire more female lobbyists when dealing with more gender-diverse policy-makers. In addition, female lobbyists are more likely to get access to the EU decision-making process regardless of the gender of the policy-makers in charge. This finding elicits interesting considerations on the link between the demand and supply of representation. While interest groups supply more gender diversity when faced with gender diversity, policy-makers demand it regardless of the context. This is an important finding that calls for further research on policy-makers’ expectations in terms of representation and their effects on the decision-making process.

Conclusion

This paper is the first to take a gender perspective and assess on an individual micro-level lobbyists’ access to policy-makers in the European Union. We argue that interest groups are likely to use the lobbyists’ gender to improve their likelihood of gaining access to policy-makers.

By cross-referencing data from the JTR with a dataset on meetings held by specific policy-makers with interest group representatives in the European Commission we create a unique dataset to test our argument. In so doing, we provide a nuanced perspective that assesses the specific lobbyist and policy-maker interacting in a specific meeting. Moreover, we determine in detail lobbyists’ individual characteristic and their organizations as well as for the policy-makers and their institutions. Using innovative machine learning techniques, we extract the gender of the lobbyists and policy-makers.

To test our hypothesis, we take an instrumental variable approach. We find that female lobbyists are likelier to gain access to meetings when the policy-maker is female. The results lend support to the expectation that interest groups adapt their strategies to improve their lobbying success (access to policy-makers), with gender being a notable characteristic taken into consideration.

Supplemental Material

Supplemental Material - Gender representation and lobbying in the European Union

Supplemental Material for Gender representation and lobbying in the European Union by David Coen, Alexander Katsaitis and Matia Vannoni in Journal of Research & Politics.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Matia Vannoni would like to acknowledge funding from Leverhulme Research Fellowship scheme RF-2023-035\7.

ORCID iDs

Alexander Katsaitis

Matia Vannoni

Supplemental Material

Supplemental material for this article is available online.

The replication files are available at:

Carnegie Corporation of New York Grant

This publication was made possible (in part) by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.

Note

References

Acemoglu

Linn

(2004) Market size in innovation: theory and evidence from the pharmaceutical industry. Quarterly Journal of Economics 119(3): 1049–1090.

Basso

Peri

(2015) The Association between Immigration and Labor Market Outcomes in the United States. Bonn, Germany: IZA. https://ftp.iza.org/dp9436.pdf

Baumgartner

Leech

(2001) Interest niches and policy bandwagons: patterns of interest group involvement in national politics. The Journal of Politics 63: 1191–1213.

Baumgartner

Jeffrey

Hojnacki

, et al. (2009) Lobbying and Policy Change: Who Wins, Who Loses, and Why. Chicago, IL: University of Chicago Press.

Celis

Childs

Kantola

, et al. (2008) Rethinking women’s substantive representation. Representation 44(2): 99–110.

Chalmers

(2013) With a lot of help from their friends: explaining the social logic of informational lobbying in the European Union. European Union Politics 14(4): 475–496.

Christiansen

(1997) Tensions of European governance: politicized bureaucracy and multiple accountability in the European commission. Journal of European Public Policy 4(1): 73–90.

Coen

Vannoni

(2016) Sliding doors in Brussels: a career path analysis of EU affairs managers. European Journal of Political Research 55(4): 811–826.

Coen

Vannoni

(2020) The strategic management of government affairs in Brussels. Business & Society 59(4): 612–641.

10.

Coen

Lehmann

Katsaitis

(2021a) Deliberative layering: explaining diverse interest mobilization across the European Parliament’s policy cycle. Journal of Public Affairs 21: e2139.

11.

Coen

Katsaitis

Vannoni

(2021b) Business Lobbying in the European Union. Oxford, UK: Oxford University Press.

12.

De Bruycker

(2024) Access for all? Socio-demographic biases and interest group access to the European Parliament. Journal of European Public Policy 1–25. Available at: https://doi.org/10.1080/13501763.2024.2409822

13.

Dunlop

(2024) LGBTIQ+ lobbying: advocacy, advice and regulation. In: Handbook on Lobbying and Public Policy. Cheltenham, UK: Edward Elgar Publishing, 471–486.

14.

Egeberg

Gornitzka

Trondal

, et al. (2013) Parliament staff: unpacking the behaviour of officials in the European Parliament. Journal of European Public Policy 20(4): 495–514.

15.

Elomäki

Kantola

(2020) European social partners as gender equality actors in EU social and economic governance. Journal of Communication and Media Studies: Journal of Common Market Studies 58(4): 999–1015.

16.

Ertug

Brennecke

Kovács

, et al. (2022) What does homophily do? A review of the consequences of homophily. The Academy of Management Annals 16(1): 38–69.

17.

Getz

(1997) Research in corporate political action: integration and assessment. Business & Society 36(1): 32–72.

18.

Hanegraaff

Berkhout

van der Ploeg

(2022) Exploring the proportionality of representation in interest group mobilization and political access: the case of the Netherlands. Acta Politica 57(2): 254–276.

19.

Hillman

Zardkoohi

Bierman

(1999) Corporate political strategies and firm performance: indications of firm-specific benefits from personal service in the US government. Strategic Management Journal 20(1): 67–81.

20.

Junk

Romeijn

Rasmussen

(2021) Is this a men’s world? On the need to study descriptive representation of women in lobbying and policy advocacy. Journal of European Public Policy 28(6): 943–957.

21.

Kantola

Rolandsen Agustín

(2019) Gendering the representative work of the European Parliament: a political analysis of women MEP’s perceptions of gender equality in party groups. Journal of Communication and Media Studies: Journal of Common Market Studies 57(4): 768–786.

22.

Krook

Mackay

(2011) Gender, Politics and Institutions: Towards a Feminist Institutionalism. Berlin, Germany: Springer, Vol. 1.

23.

Levy

Egan

(2003) A neo-gramscian approach to corporate political strategy: conflict and accommodation in the climate change negotiations. Journal of Management Studies 40(4): 803–829.

24.

Lombardo

Verloo

(2009) Institutionalizing intersectionality in the European Union? Policy developments and contestations. International Feminist Journal of Politics 11(4): 478–495.

25.

Mahoney

(2004) The power of institutions: state and interest group activity in the European Union. European Union Politics 5(4): 441–466.

26.

Mansbridge

(1999) Should blacks represent blacks and women represent women? A contingent “yes”. The Journal of Politics 61(3): 628–657.

27.

Minto

Mergaert

(2018) Gender mainstreaming and evaluation in the EU: comparative perspectives from feminist institutionalism. International Feminist Journal of Politics 20(2): 204–220.

28.

Oliver

Holzinger

(2008) The effectiveness of strategic political management: a dynamic capabilities framework. Academy of Management Review 33(2): 496–520.

29.

Reimann

(2006) A view from the top: international politics, norms and the worldwide growth of NGOs. International Studies Quarterly 50(1): 45–68.

30.

Strickland

Stauffer

(2021) Legislative diversity and the rise of women lobbyists. Political Research Quarterly 75(2): 531–546.

31.

Vogel

(1996) The study of business and politics. California Management Review 38(3): 146–165.

32.

Walker

Rea

(2014) The political mobilization of firms and industries. Annual Review of Sociology 40: 281–304.

33.

Waylen

(2017) Analysing gender in informal institutions: an introduction. In: Gender and Informal Institutions. London, UK: Rowman & Littlefield, 1–22.

34.

Weldon

(2002) Beyond bodies: institutional sources of representation for women in democratic policymaking. The Journal of Politics 64(4): 1153–1174.

35.

Wiener

(2021) Getting a high heel in the door: an experiment on state legislator responsiveness to women’s issue lobbying. Political Research Quarterly 74(3): 729–743.

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.

0.00 MB

1.30 MB