Introduction
The relationships between states and non-governmental organisations (NGOs) have become increasingly central to the governance and promotion of human rights norms around the world, shaping human rights policies and practices. At the international level, NGOs work alongside states within the United Nations (UN) human rights system and mechanisms (Sweeney & Saito, Reference Sweeney and Saito2009; Holmlund, Reference Holmlund, Ramcharan, Brett, Clark and Parker2022; Ramcharan et al., Reference Ramcharan2022). NGOs provide critical input and monitoring to help UN treaty bodies and periodic review processes evaluate state compliance with human rights obligations (McGaughey, Reference McGaughey2017). Many states also actively consult with NGOs and incorporate their recommendations when drafting national reports to human rights monitoring bodies, thus benefitting from the expertise and on-the-ground information provided by NGOs. In Europe, they collaborate on a range of issues with European institutions such as the European Commission and the European Union Agency for Fundamental Rights (Thiel, Reference Thiel2014; Ruzza, Reference Ruzza2014). At the domestic level, states collaborate with NGOs as part of their efforts to implement human rights treaty obligations (Beckstrand, Reference Beckstrand2015). More broadly, NGOs can play a key role in pushing states to adopt more ambitious human rights-related policies (Böhmelt et al., Reference Böhmelt, Bernauer and Koubi2015) and they often are key implementing partners for such policies, at the national or local levels (Oomen & Durmus, Reference Oomen and Durmus2019). NGOs also cooperate with institutions like human rights commissions and ombudsman offices to strengthen accountability and redressal mechanisms.
While these collaborations demonstrate the complementary governance roles states and NGOs can play in mutually advancing human rights, in fact, interactions between states and NGOs in the human rights domain vary remarkably around the globe and can take on various forms – from cooperative partnerships advancing shared agendas, to cooptation or more adversarial roles of advocacy and monitoring by NGOs in pushing states to uphold their obligations (e.g., see Keck & Sikkink, Reference Keck and Sikkink1998; Risse-Kappen et al., Reference Risse-Kappen, Ropp and Sikkink1999; Fagan and Sircar, Reference Fagan and Sircar2011; Stroup, Reference Stroup2022; Crepaz & Hanegraaff, Reference Crepaz and Hanegraaff2022).Footnote 1 Keck & Sikkink’s (Reference Keck and Sikkink1998) influential work also highlights how transnational advocacy networks of NGOs tend to circumvent unresponsive states, mobilising international pressure to promote human rights changes domestically.
As a result of such developments, growing scholarly attention has been paid to state-NGO relations within contemporary human rights governance. Some scholars have tried to identify the specific patterns of interaction between states and NGOs, while others have rather focused on understanding why states and NGOs develop interactions in the first place (Stroup, Reference Stroup and Davies2019). A specific scholarly debate has therefore emerged on the drivers of state-NGO relations in the human rights policy field and beyond, i.e., on the factors that drive the emergence of such relations and shape them. The existing scholarship on state-NGO relations (reviewed in depth in section ‘The drivers of state-NGO relations’) has specifically identified four main sets of factors that decisively shape such relations: (a) the characteristics of the policy field in which such relations develop; (b) the characteristics of the state; (c) the characteristics of NGOs and (d) the flow of resources, particularly financial resources, between states and NGOs.
Despite the relevant insights produced, this literature has four main limitations. First, the existing scholarship mostly adopts qualitative approaches, and largely focuses on case studies, within individual countries. This allows for an in-depth understanding of the drivers of state-NGO relations in specific contexts and settings, but limits the generalisability of findings and claims. Conversely, cross-country and cross-sectoral comparative empirical studies of state-NGO relations are still scarce (among the exceptions, on rather specific dimensions of state-NGO relations, see: Chaudhry, Reference Chaudhry2022, on state crackdowns on NGOs; Murdie & Hicks, Reference Murdie and Hicks2013, on NGOs’ impact on government services; and Murdie & Bhasin, Reference Murdie and Bhasin2011, on NGOs’ effects on antigovernment protests). Second, much of this literature tends to conceptualise states as unitary actors, rather than accounting for differences among different agencies and bodies which have different mandates, interests and institutional perspectives. Third, such literature largely ignores the multilevel nature of contemporary human rights governance, i.e., does not systematically account for how interactions of NGOs and the state, and NGOs and supranational organisations like the European Union (EU), might influence one another. Fourth, such literature has so far largely ignored potential drivers of state-NGO relations related to the ideological leanings of state governments, despite the growing politicisation of human rights across the globe, i.e., the growing salience and polarisation of human rights issues in political debates, particularly in Western countries (Cliquennois et al., Reference Cliquennois, Chaptel and Champetier2024).
This article complements and challenges this existing scholarship, contributing to fill the above-mentioned gaps. To do so, the article develops an innovative quantitative study of the drivers of state-NGO relations, elaborating quantitative relational data on the frequency and nature (collaborative/conflictual) of interactions between human rights NGOs and national and European governmental actors, across 26 EU Member States, exploiting an original dataset built in cooperation with the European Union Agency for Fundamental Rights (FRA) in the framework of its long-lasting annual ‘Civic space consultation’ survey, conducted among its network of NGO partners in European countries.Footnote 2 This unique dataset, which provides a total of 2,288 cases of interactions between NGOs and state actors, allows us to systematically test hypotheses about the drivers of state-NGO relations developed by qualitative case study analyses. In doing so, the article moves beyond conceptualisations of the state as a unitary actor, distinguishing between NGOs’ interactions with different types of state actors. In addition, it simultaneously analyses relations between NGOs and state actors and between the same NGOs and (different types of) EU institutions, specifically aiming to understand how NGO-state relations and NGO-EU relations interact, thus accounting for the multilevelness of European human rights governance. Finally, it specifically develops and tests hypotheses about potential drivers of state-NGO (and EU-NGO) relations related to the political orientation of national governments, which is arguably highly relevant in the context of the growing politicisation of human rights in Europe.
The article produces three main findings. First, the political orientation of national governments influences the nature – but not the frequency – of their interactions with human rights NGOs. NGOs in countries governed by more conservative parties report significantly less collaborative relations with national authorities, yet they do not report fewer interactions overall. Second, NGOs working on highly politicised human rights issues, such as migration, LGBTQ+ rights, and environmental protection, have both less frequent and less collaborative interactions with national governments compared to other NGOs. This indicates that politicisation may operate as a dual constraint: it reduces opportunities for engagement and undermines the perceived nature of exchanges when they do occur. Third, NGOs based in countries with conservative governments tend to engage more frequently and more collaboratively with EU institutions, indicating that domestic political constraints may encourage NGOs to shift their advocacy and cooperation to the supranational level.
The article is organised as follows. The section that follows reviews the existing literature on the drivers of state-NGO relations. Then, we introduce the existing literature on human rights politics and governance and we develop hypotheses based on such literature. Our section ‘Methodology’ illustrates our data and methods, while our section ‘Results’ describes the findings of our multilevel regression analysis. The conclusion discusses the relevance of such findings and suggests ways forward for future research.
The drivers of state-NGO relations
This section reviews the existing literature on state-NGO relations and specifically focuses on the four sets of factors that, this scholarship suggests, shape the emergence of state-NGO relations (see Stroup, Reference Stroup and Davies2019 for a more comprehensive review). These four sets of factors – issue area, NGO characteristics, state characteristics, and resource flows – do not operate in isolation but rather intersect and interact in complex ways to shape interactions between states and NGOs (Stroup, Reference Stroup and Davies2019).
A first relevant factor identified by scholars is the issue area in which the interactions between state actors and NGOs take place. The patterns of interactions between state actors and NGOs can vary considerably depending on the specific domain, be it civil rights, environmental protection, development aid, or any other realm of global governance. Scholars such as Tarrow (Reference Tarrow1994) and Betsill & Corell (Reference Betsill and Corell2008) have observed that state-NGO relations on emerging issues tend to be less developed and less collaborative if not highly confrontational, with NGOs mostly challenging and condemning state failures or inaction. However, as the issue matures and gains traction, a shift towards negotiation and collaboration often occurs, with both parties seeking to find common ground. In the field of civil rights, for instance, NGO-State relations have been historically more underdeveloped and tense, with NGOs directly criticising and pressuring states over human rights violations (Keck & Sikkink, Reference Keck and Sikkink1998, Risse-Kappen et al., Reference Risse-Kappen, Ropp and Sikkink1999), mostly invoking international laws and often targeting diplomats and bureaucrats (Betsill & Corell Reference Betsill and Corell2008; Simmons, Reference Simmons2009). Conversely, in sectors like development aid, NGOs have often enjoyed closer cooperative relationships with states, serving as partners in service delivery and the implementation of foreign aid programmes (Edwards & Hulme, Reference Edwards and Hulme1996).
The second set of factors explored by the existing literature is related to the characteristics of the NGOs themselves. A major factor that, scholars argue, can influence the type of interactions they develop with state actors is the principled commitments upheld by these organisations, particularly their commitment to independence from state influence. As Barnett (Reference Barnett2011) and Redfield (Reference Redfield2013) noted, NGOs that prioritise maintaining their autonomy and neutrality (and therefore also adopt more confrontational or critical approaches towards states) tend to have less frequent interactions with state actors. Conversely, those willing to compromise their independence may be more inclined towards cooperation (or even cooptation) by state actors. The authority and credibility of NGOs, often related to their size (Kotzian & Steffek, Reference Kotzian and Steffek2013), also play a role in shaping their interactions with states. Bigger, prominent and well-established NGOs, often referred to as ‘leading’ or ‘gatekeeper’ organisations (Bob, Reference Bob2005, Carpenter, Reference Carpenter2014), tend to enjoy greater access and collaboration with state actors. Their expertise and moral authority lend them a degree of influence that can facilitate cooperative relationships. In contrast, lesser-known or obscure NGOs may resort to more confrontational tactics as a means of asserting their legitimacy and drawing attention to their causes (Stroup & Wong, Reference Stroup and Wong2017). Another crucial factor influencing state-NGO relations is the size and composition of the NGO population within a particular issue area. As Bush and Hadden (Reference Bush and Hadden2017) argue, the sheer number of NGOs operating in a given domain can affect their perceived importance and influence over state actors. In areas with a substantial NGO presence, such as international relief and development, states may find it increasingly difficult to ignore or side-line these organisations.
A third set of factors that shape state-NGO relations is related to the characteristics of the states themselves. The level of economic development of a country, for instance, can profoundly influence the roles and opportunities available to NGOs operating in it. In industrialised democracies, NGOs may focus more on advocacy, resource mobilisation, and influencing the policies of their home governments. In developing nations, NGOs often play a more prominent role in service delivery, capacity building, and advocating for political and social change (Lewis & Kanji, Reference Lewis and Kanji2009), and they might be more dependent on states for funding of their own activities. The regime type of a state, whether democratic or authoritarian, also can decisively impact on the nature of its interactions with NGOs (Böhmelt et al., Reference Böhmelt, Bernauer and Koubi2015). Democratic politics, founded on principles of citizen participation and self-rule, are generally more amenable to the involvement of civil society organisations (Smith & Wiest, Reference Smith and Wiest2005). However, authoritarian regimes can also engage in cooperative relationships with NGOs (Heiss, Reference Heiss and Davies2019), often driven by a desire to access external resources or project an image of pseudo-democracy (Hyde, Reference Hyde2011). Bakke et al., (Reference Bakke, Mitchell and Smidt2020) also analysed how the political regime type shapes NGO strategies, showing that they pursue more confrontational tactics against closed autocracies but collaborative approaches with semi-democratic regimes. Beyond regime type, the domestic regulatory structures and legal frameworks governing NGO activity within a state can also shape the patterns of interaction. Corporatist systems, for instance, tend to be more restrictive towards NGOs, while pluralist systems may offer greater freedoms and opportunities for engagement (Stroup & Murdie, Reference Stroup and Murdie2012).
The final factor explored by the literature is the flow of resources, particularly financial resources, between states and NGOs (see Fagan & Sircar, Reference Fagan and Sircar2011; Crepaz & Hanegraaff, Reference Crepaz and Hanegraaff2022). In recent decades, there has been a marked increase in government funding channelled towards NGOs, often as a means of outsourcing or supplementing state functions in areas like development aid or service delivery (Ali & Gull, Reference Ali and Gull2016; Bloodgood & Tremblay-Boire, Reference Bloodgood and Tremblay-Boire2017). This influx of resources – which can provide NGOs with much-needed financial support but also carries the risk of cooptation as organisations may align their priorities and programming with the preferences of their donors – tends to favour the emergence of collaborative relations between states and NGOs (Gent et al., Reference Gent2015).
Theoretical framework
The ideological leaning of national governments, in fact, is not typically listed among the key drivers of State-NGO relations (e.g., see Stroup, Reference Stroup and Davies2019). More broadly, none of the scholarly works reviewed in the previous section, to the best of our knowledge, have thus far focused on the link between human rights politics and governance or, in other words, on the role of domestic political divides in potentially shaping human rights governance. We do, instead, argue in this section of the paper that there are reasons to believe that this factor can be potentially crucial in accounting for the emergence of different patterns of interactions between state actors and NGOs in the field of human rights.
Following Oberleitner (Reference Oberleitner2007), human rights governance refers to the structures and procedures aimed at safeguarding human rights developed by public and non-state actors like NGOs, advocacy networks, social movements and civil society groups.Footnote 3 Other definitions of governance (which could be well applied to the case of human rights) prioritise an understanding of governance as a process where multiple public and nonpublic actors (e.g., NGOs in the field of human rights) work together to produce and implement policy (Ansell & Torfing, Reference Ansell and Torfing2016; Stoker, Reference Stoker2018). Moving from this latter perspective, one must acknowledge that governance as a process implies fundamentally political choices on the collective goals to be pursued (Peters et al., Reference Peters2022), for instance on which human rights measures should be adopted and how human rights-related policies should be implemented. Such a political dimension of human rights governance has become highly relevant in the past decade following the growing political contestation and politicisation of human rights (Freedman, Reference Freedman2024), mostly related to the growth of radical right parties (Chaney, Reference Chaney2014; Tuori & Karjalainen, Reference Tuori and Karjalainen2025). It follows that party politics has a critical role in shaping who will work with whom, and how, in the making of policies on human rights-related issues (on how party politics dynamics can shape governance relations see: Caponio et al., Reference Caponio, Schiller and Talleraas2025; Pettrachin & Solano, Reference Pettrachin and Solano2025).
More specifically, we derive from existing works on party politics and human rights (discussed in the next subsections) that two main factors can potentially influence patterns of interaction between state actors and NGOs in the field of human rights – and specifically the frequency and nature of such interactions (about which we do have information in our dataset). These are introduced in the subsections that follow.
The political affiliation or ideology of national governments
Existing studies on human rights electoral politics have highlighted how political parties’ stances on human rights are influenced by their position on the Left-Right ideological spectrum, mostly suggesting a positive correlation between left-wing ideology and support for human rights (Moghaddam and Vuksanovic, Reference Moghaddam and Vuksanovic1990; Dunne & Wheeler, Reference Dunne and Wheeler1999; Steiner et al., Reference Steiner, Alston and Goodman2007; Chiru & Stoian, Reference Chiru and Stoian2019; Østergaard-Nielsen et al., Reference Østergaard-Nielsen, Ciornei and Lafleur2019).
Left-wing parties are traditionally more inclined to build a politics of universal human rights and to prioritise upholding welfare rights and expanding civil rights (Dean, Reference Dean2007). Left-party strength is traditionally seen as an indicator of a progressive, reformist public policy agenda associated with promoting equality and human rights (cf. Esping-Andersen, Reference Esping-Andersen1990; Phillips, Reference Phillips1994). Chaney’s (Reference Chaney2014) analysis of UK political manifestoes, for instance, unveils a pronounced Left-Right political cleavage in issue salience and policy framing, with parties on the Left attaching greater priority to human rights promotion and employing different framing practices.
As to centre-right parties, the existing literature has provided mixed findings about their positions and stances on human rights. Scholarly works published until the mid-2010s have shown that at times these parties did advocate for the application of human rights, especially in foreign policy (Quinn, Reference Quinn2008; Kerr et al., Reference Kerr, Byrne and Foster2011 ; Duranti, Reference Duranti2017), that they were often internally split on these issues and that, overall, characterising their positions as anti-human rights was misleading (Chaney, Reference Chaney2014). Other studies have instead revealed centre-right or conservative parties’ concerns regarding human rights which often stem from potential loss of sovereignty, clashes with personal liberty notions, fears of oppressive political correctness, upholding the rule of law and safeguarding property rights (Green, Reference Green1996; Cole, Reference Cole2005; Quinn, Reference Quinn2008; Nash, Reference Nash2016). Such motivations are at the basis of the more pronounced opposition to human rights policies and measures of radical right parties (RRPs). These parties have increased their electoral consensus across most European countries in the past decade (Tuori & Karjalainen, Reference Tuori and Karjalainen2025), and, when in government, they have sometimes introduced repressive laws targeting civil society organisations (Chaudhry, Reference Chaudhry2022). At the same time, recent scholarly works suggest that mainstream centre-right parties have often responded to RRPs’ electoral growth by coopting some of their exclusionary position on several human rights domains, such as migrants’ access to the welfare state (Schumacher & van Kersbergen, Reference Schumacher and van Kersbergen2016; Lefkofridi & Michel, Reference Lefkofridi, Michel, Banting and Kymlicka2017; Herman et al., Reference Herman, Hoerner and Lacey2021).
While accounting for the potential role of party politics in shaping human rights governance, one must consider, however, that, while most of the existing literature tends to focus on a unidimensional Left-Right ideological spectrum, European party politics is in fact characterised by two major cleavages or dimensions that shape party positioning. The first is the cultural or value-based cleavage, often referred to as the ‘liberalism-authoritarianism (or conservatorism)’ divide or, more recently, the ‘integration-demarcation’ divide (Kriesi et al., Reference Kriesi2012). This dimension captures parties’ stances on cultural issues like civil rights, immigration, multiculturalism, national identity, and the role of traditional values in society. Parties on the one hand of the spectrum (more conservative or authoritarian parties) advocate for greater cultural protectionism, stricter immigration controls, and preserving traditional national identities. On the other side are more libertarian or progressive parties favouring cultural openness, cosmopolitanism, and promoting individual civil rights and diversity. The second major cleavage is the economic left-right divide over the role of state intervention in the economy versus free market capitalism. This divides economically left-wing parties supporting expansion of social rights and robust social welfare policies, economic redistribution, and labour market regulation against economically right-wing parties advocating for economic liberalisation, deregulation, and limiting state involvement in the economy. Overall, parties can combine different positions along the two dimensions, for instance they can be culturally liberal but economically right-wing (as the case of parties affiliated to Renew Europe in the European Parliament), or culturally conservative but economically left-wing (as several parties in Eastern Europe).
Combining insights introduced in the previous paragraphs and moving from the definition of ‘governance’ outlined above, we define our first hypotheses, expecting that the political affiliation of national governments – and specifically the positions on the cultural axis of European party politics of political parties in government – might be correlated with the frequency and nature of governments’ interactions with NGOs in the field of human rights:
H1a: State-NGO interactions in the field of human rights will be more frequent in countries with national governments supported by more culturally progressive political parties and less frequent in countries with national governments supported by more culturally conservative political parties.
H1b: State-NGO interactions in the field of human rights will be more collaborative in countries with national governments supported by more culturally progressive political parties and less collaborative in countries with national governments supported by more culturally conservative political parties.
The specific human rights issues at stake and the level of politicisation of such issues in contemporary political debates.
A second key factor that we expect might influence the frequency and nature of interactions between NGOs and state actors is the level of politicisation of the specific human rights issues around which NGOs operate. Indeed, while all human rights issues ‘can be political’ (Valentini, Reference Valentini2012), some are more politicised – i.e., more polarising and salient in political debates – than others (Terman and Byun, Reference Terman and Byun2022). The existing literature suggests that the different levels of politicisation of human rights issues have powerful implications for the enforcement of human rights norms (Terman & Byun, Reference Terman and Byun2022). While the existing literature does not explicitly identify links between politicisation levels and the frequency and nature of governmental actors’ relations with NGOs, a lower inclination by governments to deal with specific human rights issues is likely to have implications for the kind of relations that establish with NGOs operating in such fields (Tarrow, Reference Tarrow1994; Betsill & Corell Reference Betsill and Corell2008;– see also works on state-NGO relations in the field of civil rights, where relations tend to be more tense and underdeveloped, e.g., Risse-Kappen et al., Reference Risse-Kappen, Ropp and Sikkink1999; Simmons, Reference Simmons2009, and works on state-NGO relations in less politicised fields such as development aid, describing more collaborative relationships, e.g., Edwards & Hulme, Reference Edwards and Hulme1996).
Specifically, our review of the existing literature on human rights politics suggests that migrants’ rights have been among the most highly politicised human rights issues in Europe in the last decade (Lutz, Reference Lutz2025). As underlined by Abou Chadi and colleagues (Reference Abou-Chadi, Breyer and Gessler2021), sexual and reproductive rights and LGBTQ+ rights have been also increasingly politicised in the past decade in both Eastern and Western Europe, mostly because of politicisation strategies enacted by mainstream right and radical right parties, resulting in increased polarisation of views across the political spectrum (Alonso & Espinosa-Fajardo, Reference Alonso and Espinosa-Fajardo2021). Finally, a third key area marked by high levels of politicisation and polarisation, especially after the approval of the European Green Deal, is environmental protection (Witajewska-Baltvilka et al., Reference Witajewska-Baltvilka2024). A study of societal polarisation of different human rights-related issues recently conducted by the MIDEM project on several Western and Eastern European countries confirms that migration, climate change and rights of sexual minorities are the three most polarised issues currently in European societies (Herold et al., Reference Herold2023).
This leads us to our second set of hypotheses:
H2a: State-NGO interactions in the field of human rights will be less frequent for NGOs working on highly polarised/politicised human rights issues (migration, LGBTQ+ and sexual rights, environment).
H2b: State-NGO interactions in the field of human rights will be less collaborative for NGOs working on highly polarised/politicised human rights issues (migration, LGBTQ+ and sexual rights, environment).
NGOs and venue shopping
Besides analysing the effects of electoral politics on relations between state actors and NGOs, we are also interested in analysing whether the political affiliation of national governments might also play a role in altering and shaping the same NGOs’ relations with EU institutions. A few qualitative works on NGOs and advocacy (Buffardi et al., Reference Buffardi, Pekkanen and Smith2015; Kagan Reference Kagan2024) have indeed suggested that NGOs can engage in ‘venue shopping’, i.e., shifting their activities and particularly advocacy initiatives, when facing obstacles, to more favourable policy venues (Schattschneider, Reference Schattschneider1960; Pralle, Reference Pralle2006), where policy venues are conceptualised as ‘the institutional locations where authoritative decisions are made concerning a given issue’ (Baumgartner & Jones, Reference Baumgartner and Jones2009, 32). In our case, this means that facing an ideologically hostile government might lead NGOs to shift their activities to the EU level. This leads us to our third and final set of hypotheses, according to which the ideological leaning of national governments is expected to influence the frequency and nature of relations between NGOs and the EU level:
H3a: Interactions between NGOs and EU-level institutional actors in the field of human rights will be more frequent for NGOs operating in countries with national governments supported by more culturally conservative political parties.
H3b: Interactions between NGOs and EU-level institutional actors in the field of human rights will be more collaborative for NGOs operating in countries with national governments supported by more culturally conservative political parties.
This article conducts an innovative systematic comparative analysis which aims to test the above-mentioned hypotheses across 26 EU Member States. It does so by applying a quantitative methodological approach that is described in the next section.
Methodology
Data and questionnaire
The quantitative analyses conducted in this article use original survey data collected in cooperation with the European Union Agency for Fundamental Rights (FRA) between January and February 2024 as part of its ‘Civic space consultation’ survey. The ‘Civic space consultation’ is an annual survey among the FRA’s partner network of NGOs in 30 European countries: the 27 EU Member States, Albania, North Macedonia and Serbia. Since 2018, FRA run this survey annually among key players in civil society on their experiences of civic space through its Fundamental Rights Platform. Participation in the survey is voluntary but key players in the field are mapped and invited to participate. Therefore, the sample should not be considered representative for each country, but it includes the most prominent NGOs working in the broad field of human rights in Europe. The consultation gathered information from 332 NGOs. Among these, 27 were EU-level umbrella organisations, 240 were based in the EU27 countries and 65 were based the three non-EU countries. Responses were received from all EU Member States except Cyprus, with an average of nine NGOs per country. The umbrella organisations and the NGOs based in non-EU countries were excluded from the analysis, as our study focuses specifically on national-level and EU-level dynamics. The final sample consisted of 240 NGOs operating across 26 EU Member States.
As part of the survey, questions were asked about the overall situation of civil society organisations operating in the human rights field, including their experiences with threats or attacks and their impact; challenges in accessing financial resources and the need for changes to funding frameworks; the legal and regulatory environment in which they operate; and their ability to participate in decision-making processes at local, national, and EU levels. As part of the broader survey, we specifically asked questions about the relationship between the NGOs and various institutional actors. Participants were first asked about the frequency of their organisation’s ‘interactions, exchanges, or discussions’ with a range of institutional actors over the previous 12 months. The frequency was measured on a temporal scale of 0–5.Footnote 4 Next, we requested respondents to describe the nature of these interactions. Respondents were invited to assess whether the exchanges were conflictual or collaborative on a scale of 1–5: (1) Very conflictual; (2) Rather conflictual; (3) Neither conflictual nor collaborative; (4) Rather collaborative; (5) Very collaborative. Both questions were asked regarding NGOs’ interactions with a wide range of actors: different types of EU institutions, different actors within national and local governments, and human rights bodies. In the analysis on State-NGO interactions (see Findings), we specifically focus on NGOs’ interactions with elected members of government and non-elected government officials (Models 1, 416 interactions were therefore considered). Analyses on interactions between NGOs and the EU focus on interactions between NGOs and three EU institutions: the European Commission, the European Parliament, and the EU Agency for Fundamental Rights (Models 2, 623 interactions were therefore considered).
This information helps assess both the frequency and the nature of the interactions between NGOs operating in the field of human rights and national and EU-level institutional and governmental actors. Since the dataset includes information on the respondents (the NGO) and their interactions with other actors, across different countries, it has a nested (multilevel) structure.
Variables
This subsection describes the variables we use in our quantitative analyses.
Dependent variables
Our two dependent variables in analyses testing the first and second hypotheses (Models 1) are:
-
• The frequency of interactions between the NGO and national governmental actors (elected members of national governments or governmental officials) (‘Frequency’), measured on a scale of 0–5 (where 0 = never and 5 = very frequent interactions).
-
• The nature of these interactions between the NGO and national governmental actors (elected members of national governments or governmental officials) (‘Nature’), measured on a scale of 1–5 (where 1 = very conflictual and 5 = very collaborative). We considered as missing values all the respondents that declared to have no interaction to the frequency question.
Our two dependent variables in analyses testing the third hypothesis (Models 2) are very similar but consider the frequency and nature of interactions between the NGO and EU-level institutional actors (European Commission, European Parliament, or the FRA).
On average (median value), interactions were not so frequent (2–3 times per year) and rather collaborative.
Main independent variables
Our main independent variables are:
-
• Political orientation of the national government. This variable describes the political affiliation of political parties controlling the national government in each country. Following discussions in section Theoretical framework above, we opted for focusing on party positions on the cultural dimension of European party politics, relying on the ‘liberty-authority’ index used in the Parl-Gov dataset (Döring & Manow, Reference Döring and Manow2024). We calculated the political orientation of the government as the weighted average of the scores in the liberty-authority index of the governing parties, using each party’s share of seats in the national Parliament relative to the total seats held by the governing coalition.
-
• Areas of activity. NGOs were asked to indicate their main area of activity across 28 predefined areas, including a wide range of domains (e.g., access to justice, health, economic and social rights, women’s rights…). They reported whether they were active or not in each of these areas. Specifically, we constructed a binary classification, distinguishing between highly politicised areas (areas related to migration, sexual and reproductive health and LGBTQ+ rights, and environmental protection – see section above) and less politicised areas (all the others).
Control variables
The following additional independent variables are included in our models to control for other theoretically grounded determinants of local integration policymaking (see section The drivers of state-NGO relations). On the one hand, we do include a number of control variables at the level of individual NGOs:
-
• Size. This variable describes the number of employees of the NGO: 0 (ref.); 1 to 10; 11 to 50; Over 50.
-
• Main source of funding. This variable identified the main source of funding that the NGO reported to rely on, including: National government (ref.); European Union; EEA/Other international donors; Private donors; Own resources/local community.
-
• Target actor. This variable distinguishes among the different types of governmental actors with whom NGOs had interactions. Within national governmental actors (Models 1, testing H1 and H2), we distinguished between members of government (ref.) and government officials. As to the EU, we distinguish between the European Commission, the European Parliament and the EU Agency for Fundamental Rights (Models 2, testing H3).
We also included several control variables at the country level, which are also assumed to potentially influence NGO-state relations:
-
• GDP per capita (in USD).
-
• The Corporativism Index (CorpEUplus), derived from Jahn, Reference Jahn2016.
-
• Electoral Democracy Index (Polyarchy) retrieved from the V-DEM dataset (Coppedge et al., Reference Coppedge2016).
Table 1 illustrates the descriptives for the independent variables. Several other variables at both the individual and country levels were introduced in our robustness checks (see section ‘analytical strategy’).
Descriptives of the independent variables
Table 1 Long description
The table presents descriptives of independent variables at both state and NGO levels. For state-level variables, it includes GDP with a mean of 54,441 and a standard deviation of 24,084, ranging from 23,424 to 132,414. Political orientation of the national government has a mean of 5.62 and a standard deviation of 1.61, ranging from 1.72 to 8.25. The corporatism index has a mean of -0.05 and a standard deviation of 0.67, ranging from -0.94 to 1.11. The electoral democracy index has a mean of 0.72 and a standard deviation of 0.18, ranging from 0.44 to 0.92. For NGO-level variables, it includes the number of employees categorized into four groups: 0 employees, 1 to 10 employees, 11 to 50 employees, and over 50 employees, with respective counts of 29, 139, 50, and 18. The main source of funding is categorized into national government, EU, EEA/other international donors, private donors, and own resources/local community, with respective counts of 45, 31, 29, 33, and 51. Areas of activity are categorized into not or moderately politicised areas and highly politicised areas, with respective counts of 37 and 171.
Analytical strategy
To test our hypotheses regarding the frequency and nature of interactions between NGOs and different actors, we estimated multilevel models including three levels: NGOs-actors interactions level; NGO level; country level variables. As such, we conducted our multilevel analyses on a sample of Na = 2,288 NGO-actor interactions nested within Nb = 208 NGOs, which in turn were nested in Nc = 27 countries. The first set of models (Models 1, Table 2) focuses on NGO’s interactions with national governmental actors, while the second set of models (Models 2, Table 3) focuses on NGOs’ interactions with EU-level governmental actors.
Models 1. Multilevel model for frequency and nature of NGO-national government interactions (n = 416)
Table 2 Long description
The table presents a multilevel model analyzing the frequency and nature of interactions between NGOs and national government actors. It includes data on the number of employees, main source of funding, main actor for interactions, areas of activity, and state-level indicators. The table has 13 rows and 3 columns, with headers for Frequency, Nature, and various categories such as Number of employees, Main source of funding, Main actor for interactions, Areas of activity, and State-level indicators. Notable trends include significant interactions for NGOs with 1-10 employees, those funded by the EU, and those interacting with government officials. The adjusted R-squared value is 0.29.
Note: ***p <.001; **p <.01; *p <.05.
Models 2. Multilevel model for frequency and nature of NGOs-European institutions interactions (n = 623)
Table 3 Long description
A table with three columns labeled Frequency, Nature, and Number of employees, Main source of funding, Main actor for interactions, Areas of activity, State-level indicators, and Frequency of interactions. The table presents data on the frequency and nature of interactions between NGOs and European institutions. It includes rows for different categories such as the number of employees, main source of funding, main actor for interactions, areas of activity, state-level indicators, and frequency of interactions. Each row provides specific values and trends for these categories, highlighting the relationships and interactions within the dataset.
Note: ***p <.001; **p <.01; *p <.05.
The number of countries is rather low (N = 27) and this can be potentially problematic in multilevel analysis, but the existing literature confirms that our model should be statistically robust. In a recent analysis of sample size in multilevel modelling published in the quantitative methods-focused journal European Sociological Review, Bryan & Jenkins (Reference Bryan and Jenkins2016, 19–20) conclude that a minimum of 25 cases are required at level 1 (the country level in our case) for multilevel linear models with no interaction terms. The number of NGOs is well above the minimum identified by Bryan & Jenkins (Reference Bryan and Jenkins2016).
Nevertheless, we conducted a set of robustness checks. First, to further reduce biases related to the number of cases at the country level, we re-estimated our models using restricted maximum likelihood (REML) estimation for the linear mixed versions of the dependent variables (Heisig et al., Reference Heisig, Schaeffer and Giesecke2017; Elff et al., Reference Elff2021) – see Table A1, online Appendix. Second, we produced ordinal multilevel regression treating frequency and nature as ordered outcomes (see Table A2, online Appendix). In the main analyses we treat our frequency and nature variables as continuous variables but in fact they are only semi-continuous. Third, we developed models including additional theoretically relevant controls (see Tables A3–4, online Appendix). In a first set of checks we added a categorical measure distinguishing between NGOs focused exclusively on social rights, exclusively on civil rights, or on both (ref: civil rights only), as well as a variable measuring the political ideology of the national governments on the economic left-right dimension of European party politics (following a similar strategy as done for the cultural dimension). In a second set of checks we added to our main models additional variables related to other factors that, the literature suggests, can potentially influence the frequency and nature of state-NGO interactions, namely: funding concerns and principled commitment (data for these variables are only available for a lower number of NGOs, which is why they were not included in the main models). Table A0 (Online Appendix) describes all the additional variables introduced.
Finally, in the main analysis, three different highly politicised issues (migration; sexual and reproductive health and LGBTQ+ rights; and environmental protection) are collapsed into a single binary variable. In additional analyses, we disaggregated the three issue areas and included them as separate predictors in the model, rather than as a single binary indicator (see Tables A5–6, online Appendix).
These specifications allow us to assess the stability of our results to changes in outcome specification, estimation approach, and model covariates.
Results
Main analyses
To test Hypotheses 1 and 2, we first focus on the frequency and nature of interactions between NGOs and national governmental actors. Table 2 reports the regression coefficients for both dependent variables: frequency and nature of interactions between NGOs and national governments.
We hypothesised that interactions would be more frequent and more collaborative in countries with governments controlled by more (culturally) progressive political parties, and less so in countries with governments controlled by more conservative parties. This expectation finds partial support in the results. Our political orientation variable is indeed negatively associated with the nature of interactions between national governmental actors and NGOs (b = −.15, p <.05), indicating that a more conservative political orientation of national governments is associated with less collaborative relations between governmental actors and NGOs. However, this variable does not significantly affect the frequency of these interactions, suggesting that, while the tone and nature of state-NGO relations may change depending on the political orientation of the national government, the frequency of interactions remains relatively stable. Thus, our H1b is supported, but our H1a is not.
We further hypothesised that NGOs working on highly politicised human rights issues –specifically migration, LGBTQ+ and sexual rights, and environmental issues – would report less frequent and less collaborative interactions with state actors. The results support this expectation. NGOs active in these polarised issue areas are significantly less likely to frequently engage with national governments (b = −.35, p <.01), and their interactions tend to be less collaborative (b = −.35, p <.05). These findings support both H2a and H2b.
Several control variables also exhibit significant associations. Larger NGOs (in terms of staff size) report more frequent interactions with state actors, with coefficients rising from.46 (1–10 employees) to.95 (over 50 employees), all statistically significant at the p <.001 level. However, staff size does not significantly influence the nature of interactions. Funding source matters for the nature of state-NGO relations. Compared to NGOs funded primarily by national governments, those relying on EU funding (b = –.87, p <.01), private donors (b = −.90, p <.001), or own resources (b = −.88, p <.01) report less collaborative interactions. Interestingly, such interactions are, however, not less frequent. These patterns suggest that NGOs that are financially independent from national governments may face more adversarial or cautious engagement, or that state funding might more easily reach NGOs that are perceived to have more collaborative relations in place with the government. Finally, interactions with non-elected government officials, as opposed to elected members of government, are associated with a higher frequency of interaction (b =.51, p <.001), although the nature of interactions does not vary the types of governmental actors engaging with NGOs.
Table 3 displays the results of the multilevel models for the frequency and nature of interactions between NGOs and EU-level actors. Hypotheses H3a and H3b proposed that NGOs operating in countries with governments controlled by more culturally conservative political parties would be more likely to interact with EU-level actors, expecting an effect on both the frequency (H3a) and nature of interactions (H3b), with NGOs using the EU as a venue for support or pressure when national spaces are more hostile.
The results provide support for these expectations. NGOs in countries with governments controlled by more conservative parties are indeed significantly more likely to interact with EU-level institutions (b =.11, p <.01), and these interactions are also perceived as more collaborative (b =.15, p <.05). This suggests that NGOs facing domestic political resistance may actively turn to supranational institutions (and especially the European Commission) for engagement and advocacy, and that EU institutions are relatively receptive to NGOs operating under national constraints. Conversely, issue politicisation does not significantly shape NGO–EU relations: NGOs working on polarised topics (migration, LGBTQ+ and sexual rights, environmental issues) are neither more nor less likely to interact with EU-level institutions, and these interactions are not perceived as less collaborative.
Moving to our controls, organisational capacity once again plays a strong role. NGOs with larger staff sizes are much more likely to engage with EU actors. Those with over 50 employees report the highest frequency (b =.61, p <.001) and also more collaborative relationships (b =.55, p <.01), indicating that capacity facilitates both access and nature of engagement. The main institutional target of interactions also matters greatly as NGOs report far less frequent contacts with the European Parliament (b = −.17, p <.001), or the EU Agency for Fundamental Rights (b = −.28, p <.001) compared to the European Commission. Smaller organisations (1–10 employees) do not report more frequent contact, but they do perceive more collaborative relationships with EU-level actors (b =.69, p <.05). Interestingly, funding dynamics have a more nuanced role at the EU level. NGOs with private donor funding report slightly more frequent interactions with the EU level (b =.22, p <.05), while those relying on their own or community resources report significantly less collaborative relations (b = −.86, p <.001). Lastly, one state-level structural factor influences NGO–EU interactions: NGOs based in countries with a more functioning electoral democracy (higher score in the Electoral Democracy Index) report significantly more collaborative interactions with EU institutions (b =.40, p <.05), suggesting that democratic institutions may facilitate mutual engagement between NGOs and supranational actors.
Additional analyses
In the main analyses, three highly politicised issue domains (migration; sexual and reproductive health and LGBTQ+ rights; and environmental protection) are combined into a single binary indicator. In these supplementary analyses, we disaggregate these issue areas and include them as separate predictors in the model rather than as a single composite measure (see Tables A5–A6 in the online Appendix). This specification allows us to assess whether the observed effects are driven by a particular policy domain (e.g., migration) or whether the patterns are consistent across issue areas.
In the main specification, politicised issue domains (relative to non-politicised issues) are associated with significantly lower frequency and less collaborative forms of NGO–national government interaction. When the domains are disaggregated, migration emerges as the primary driver of the reduced frequency of interactions, whereas sexual and reproductive health and LGBTQ+ rights appear to be most strongly associated with less collaborative interactions (see Table A5, online Appendix).
By contrast, consistent with the main analysis, the effect of politicised issue domains (relative to non-politicised issues) on NGO–European institution interactions is not statistically significant when replicating the analysis with the three domains disaggregated. This null finding is confirmed for both the frequency and the nature of interactions (see Table A6, online Appendix).
Discussion and conclusion
This article examined the drivers of state–NGO relations in the field of human rights, building on and extending existing scholarship that is, so far, largely based on qualitative case studies. Specifically, it investigated the role of the political orientation of national governments in influencing such relations, again so far largely neglected by scholars. We developed and tested hypotheses related to the influence of national governments’ political orientation on the frequency and nature of NGOs’ interactions with both state actors and EU institutions (to explore the dynamics of venue shopping related to governments’ political affiliations) and assessed whether the level of politicisation of the human rights issues addressed by NGOs influences these interactions. We tested these hypotheses using original cross-national survey data on interactions between NGOs and institutional actors in 26 EU Member States, applying a multilevel statistical framework. The analysis generated three central findings.
First, the political orientation of national governments influences the nature – but not the frequency – of interactions between state actors and human rights NGOs. NGOs in countries governed by more conservative parties report significantly less collaborative relations with national governmental actors, yet they do not report fewer interactions overall. This finding suggests that governments’ political orientation influences state-NGO relations, but in a more nuanced way compared to what we had predicted in our hypotheses grounded in the existing literature on party politics and human rights (e.g., Chaney, Reference Chaney2014; Schumacher & van Kersbergen, Reference Schumacher and van Kersbergen2016; Lefkofridi & Michel, Reference Lefkofridi, Michel, Banting and Kymlicka2017; Herman et al., Reference Herman, Hoerner and Lacey2021). The absence of an effect on frequency suggests that formal or procedural avenues for interaction may remain accessible even in less ideologically hospitable environments. This could be due to legal obligations, institutionalised consultation mechanisms, or the pragmatic needs of governments to draw on NGO expertise regardless of political alignment (see e.g., Heiss, Reference Heiss and Davies2019). In fact, these findings challenge arguments about the supposed ‘end of the NGO age’ recently raised by Bush & Hadden (Reference Bush and Hadden2025). However, the tone of engagement shifts depending on governments’ political orientation, with the nature of state-NGO collaborations deteriorating in countries controlled by more conservative governments. In other words, the political context shapes not whether but how state-NGO engagement unfolds, with potential downstream effects on the effectiveness of joint initiatives and the sustainability of partnerships in human rights governance. Overall, this finding connects with earlier qualitative studies that emphasise how ideological distance can undermine trust and mutual recognition between actors interacting in policy networks (e.g., in the migration policy field: Pettrachin & Hadj Abdou, Reference Pettrachin and Hadj Abdou2024; in the climate policy field: Wagner & Ylä-Anttila, Reference Wagner and Ylä-Anttila2018), while also pointing to the resilience of interaction frequency in the face of political divergence.
Second, NGOs working on highly politicised human rights issues, such as migration, LGBTQ+ rights, and environmental protection, have less frequent interactions with national governments, and the nature of these interactions (when they occur) is perceived to be significantly lower than in less politicised domains. This finding highlights the particularly constraining effect of issue politicisation on both the incidence and tone of engagement, suggesting that politicisation operates through a combination of access restriction and diminished the substantive quality of policy dialogue and collaboration once contact occurs. Governments may deliberately limit engagement with NGOs on contentious issues to avoid political costs, and when they do interact, the heightened ideological or symbolic stakes of these policy areas may foster more guarded, transactional, or conflict-prone exchanges from both sides (e.g., Risse-Kappen et al., Reference Risse-Kappen, Ropp and Sikkink1999). This reinforces the view that politicisation is not merely about agenda-setting constraints but also about relational strain within governance processes (Betsill & Corell Reference Betsill and Corell2008; Simmons, Reference Simmons2009; Terman & Byun, Reference Terman and Byun2022).
Third, NGOs based in countries with conservative governments tend to engage more frequently and more collaboratively with EU institutions, indicating that domestic political constraints may encourage NGOs to shift their advocacy and cooperation to the supranational level. This third finding seems to provide clear evidence of multilevel ‘venue shopping’ and that supranational venues may not only serve as alternative arenas for advocacy when national spaces are constrained but can also offer comparatively receptive and supportive environments for NGOs in politically adverse contexts. These findings resonate with venue-shopping theories in policy studies (Baumgartner & Jones, Reference Baumgartner and Jones2009; Buffardi et al., Reference Buffardi, Pekkanen and Smith2015; Kagan Reference Kagan2024), extending them to the domain of human rights governance in Europe. Importantly, the absence of an effect of issue politicisation on NGO–EU relations indicates that politicisation at the national level does not translate directly into barriers at the EU level, at least for the policy areas examined here. This highlights the distinctive role of EU institutions (at least in the time period analysed, before the last European elections held in 2024) as a counterbalance to national-level hostility. This finding also connects with the existing literature on spiral and boomerang models of transnational advocacy (see e.g., Pallas & Bloodgood, Reference Pallas and Bloodgood2022), indicating that when NGOs are thwarted in their advocacy efforts they seek to link up with foreign governments or international organisations. While this is often discussed in a non-Western context, our findings suggest that this dynamic also applies to the European context.
Taken together, these findings advance the literature on state–NGO relations in two ways. First, they demonstrate that the ideological orientation of national governments and the degree of politicisation of specific human rights issues are significant determinants of interaction patterns, complementing the structural, institutional, and resource-based factors emphasised in earlier work (Stroup, Reference Stroup and Davies2019). Second, they suggest the need to conceptualise human rights governance as a genuinely multilevel system, where national and supranational arenas interact and where NGOs can and do reallocate their engagement in response to shifting political opportunities. This moves the study of human rights governance closer to political process approaches that focus on strategic behaviour, opportunity structures, and inter-arena dynamics (e.g., Bevir & Rhodes, Reference Bevir and Rhodes2003; Kingdon Reference Kingdon2014).
At a broader level, the results suggest that the politicisation of human rights governance in Europe is not only a matter of rhetorical contestation or policy divergence (e.g., Chaney, Reference Chaney2014; Schumacher & van Kersbergen, Reference Schumacher and van Kersbergen2016) but also manifests in the micro-level patterns of engagement between NGOs and state actors. While such politicisation can constrain certain forms of interaction, especially in highly contested issue areas, it also stimulates adaptive strategies by NGOs, including the cultivation of alliances at the EU level. For scholars of state-NGO relations, this underlines the importance of viewing these interactions as embedded within wider political fields where national and supranational logics intersect. The findings also suggest that measuring the effects of politicisation requires attention not only to public debates and electoral competition but also to everyday governance practices and organisational strategies. For scholarship on the politicisation of human rights, our findings underscore the need to address (e.g., through qualitative works) its impact on both structural access to decision-making and governance relations more broadly.
This study has several limitations. The survey data, while unique in scope, are not fully representative of all NGOs in each country, as participation is limited to members of the FRA’s Fundamental Rights Platform. Despite other studies having applied a similar methodology (see e.g., Pettrachin & Solano, Reference Pettrachin and Solano2025; Martínez-Ariño et al., Reference Martínez-Ariño2019), the reliance on self-reported measures of interaction frequency and nature introduces the possibility of perceptual bias, particularly if NGOs’ political orientations influence their evaluations of collaboration. The cross-sectional nature of the data also limits our ability to make causal claims about the direction of influence between political context and interaction patterns.
Future research could address these limitations in several ways. Expanding the dataset to include a broader range of NGOs, including those outside formal FRA networks, would enhance representativeness. Longitudinal designs could capture how changes in government composition or shifts in the politicisation of issues affect state–NGO relations over time. Further disaggregation of interaction partners – both within national governments and within EU institutions – could reveal more precise patterns of selective engagement or avoidance. Finally, qualitative comparative studies could complement these quantitative findings by unpacking the mechanisms through which politicisation and ideology shape not only whether and how interactions occur, but also their substantive outcomes for human rights policy and practice.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1755773926100502
Data availability statement
The data that support the findings of this study are available on request.
Acknowledgements
We would like to express our gratitude to the European Union Agency for Fundamental Rights (FRA) for the cooperation on the survey and in particular to Waltraud Heller.
Funding statement
David De Coninck was supported by funding from the Research Foundation Flanders (FWO) with grant number 1219824N (DeMiSo).
Competing interests
No conflict of interest.
Ethical standards
The study did not require approval from an ethics committee because no personally identifiable information was collected. Participation was voluntary and NGOs participating were part of the FRA (European Union Agency for Fundamental Rights)’s civic space consultation.
Open access
