Introduction

Civil society organisations (CSOs) are used and funded by national states to provide public services. However, their role as watchdogs has been mentioned in numerous cases (Cuttitta, Reference Cuttitta2018; Fehsenfeld & Levinsen, Reference Fehsenfeld and Levinsen2019; Hasenfeld & Gidron, Reference Hasenfeld and Gidron2005). Even though this ambiguity is present in many fields where CSOs operate, it is a more salient and urgent issue in highly politicised topics, such as migration and the integration of foreigners. This dilemma is not only an analytical one, but it is also part of CSOs’ existence and decision-making. The so-called migrant crisis in 2014/2015 represents one of the wicked policy problems that could hardly be tackled without collaboration between the public sector and private and third-sector organisations (Brandsen & Johnston, Reference Brandsen and Johnston2018).

We are interested in how CSOs balance these contradictory aspects of their work. Since we are concerned with professional and formal organisations, we discuss non-governmental organisations (NGOs) instead of the more general term civil society organisations. Specifically, we focus on the role of umbrella organisations as meta-organisations—organisations whose members are also organisations (see Ahrne & Brunsson, Reference Ahrne and Brunsson2008). We propose a hypothesis that in highly politicised and fragmented fields, such as migration policy, meta-organisations create platforms that can criticise the government, effectively negotiate with it and protect their individual members. We propose that hybrid organisations, such as meta-organisations, play an important role in these fields. In the article, we follow key research questions: (1) What functions do meta-organisations fulfil within the migration policy field? and (2) How do external constraints shape these functions?

In this paper, we focus on the policy work of Czech migration NGOs and their umbrella organisation, the Consortium of Migrants Assisting Organisations. Our analysis draws on semi-structured interviews and desk research. More specifically, insights into the attitudes of NGO representatives have been gained through semi-structured interviews with staff members holding different positions in a relevant number of organisations, identified during our analysis of publicly available sources, such as lists of successful grant applicants, lists of members of governmental advisory bodies, the media, roundtables and conference programmes and policy documents. We have structured this paper as follows. First, we theoretically discuss the dynamics of three dominant functions of the ecology of games approach (Berardo & Lubell, Reference Berardo and Lubell2019): resource distribution, cooperation and social learning and the role of meta-organisations in governance systems. Second, we describe invisible policy work in a hostile environment. In the analytical part, we used the ecology of games approach to describe relations between NGOs and the state and transaction costs theory (Williamson, Reference Williamson1985; North, Reference North1990; Jaffee, Reference Jaffee, Jaffee and Morton1995) to examine the role of the umbrella organisation Consortium of Migrants Assisting Organisations. We conclude that the mere presence and initiatives of umbrella organisations representing the entire field of NGOs can enforce the development of the critical capacities of the third sector and allow the avoidance of external control connected with financial resources. We also introduce new types of transaction costs that are significant for third-sector meta-organisations.

CSOs and Public Services: Between Collaborative and Critical Policy Work

Interactions between governments and CSOs have materialised in diverse forms, ranging from hierarchical positions to fully embraced partnerships (Brandsen & Johnston, Reference Brandsen and Johnston2018). The article employs the ecology of games approach (Berardo & Lubell, Reference Berardo and Lubell2019), which can integrate multiple perspectives. It describes the policy field as the ‘ecology of games’ that consist of interlinked games played simultaneously by actors eager to achieve their individual goals. This theory recognises that policy decisions are made by networks of actors interacting in ‘multiple forums at different geographic scales ranging from local to global, addressing a myriad of interconnected issues’ (Berardo & Lubell, Reference Berardo and Lubell2019, p. 7). The structure of the ecology of games refers to three functions of a complex governance system: (1) distribution of resources, (2) cooperation and (3) learning. Resource distribution refers to how benefits and costs are distributed in the governance system. Cooperation involves shaping the behaviour of multiple actors in ways that increase the likelihood that they will collaborate to achieve desirable outcomes. Learning takes place when actors process new information about the problems they care about and how other actors think about those problems.

With respect to distributing resources, there is literature on NGOs using resource dependence theory (Froelich, Reference Froelich1999; Pfeffer & Salancik, Reference Pfeffer and Salancik2003). Resource distribution affects the goals of the organisation (Atia & Herrold, Reference Atia and Herrold2018; Khieng & Dahles, Reference Khieng and Dahles2015), autonomy and programmatic flexibility (Feiock & Jang, Reference Feiock and Jang2009), structure (Mitchell, Reference Mitchell2014) and vulnerability to economic shocks (Hodge & Piccolo, Reference Hodge and Piccolo2005). Dependency on public resources contributes to the development of more bureaucratic structures and increases competition among NGOs (Mitchell, Reference Mitchell2014). However, it increases the organisation’s sensitivity to the environment and its ability to adapt to it (Hafsi & Thomas, Reference Hafsi and Thomas2005). As Mitchell (Reference Mitchell2014) concludes, NGOs employ three kinds of strategies in response to resource distribution: (1) adaptation (alignment, subcontracting, perseverance), (2) avoidance (revenue diversification, funding liberation, geostrategic arbitrage, specialisation, selectivity) and (3) shaping (donor education, compromise). Paradoxically, some attempts to avoid external control and maintain organisational autonomy might reinforce the professionalisation process and competition between the entities for obtaining limited resources (Maya-Jariego et al., Reference Maya-Jariego, Holgado-Ramos and González-Tinoco2020).

In relation to cooperation, NGOs may differ from coordination and consultations (Chorianopoulos & Tselepi, Reference Chorianopoulos and Tselepi2019; Geyer, Reference Geyer2001) through contracting out (see Gronbjerg, Reference Gronbjerg1993; Gutch, Reference Gutch1992; De Hoog & Salamon, Reference De Hoog, Salamon and Salamon2002) and commissioning (Gillam & Siriwardena, Reference Gillam and Siriwardena2013; Murray, Reference Murray2009) to co-management (Plummer et al., Reference Plummer, Crona, Armitage, Olsson, Tengö and Yudina2012; Soliku & Schraml, Reference Soliku and Schraml2020) and co-production (Brandsen & Honingh, Reference Brandsen and Honingh2016; Brandsen et al., Reference Brandsen, Verschuere and Steen2018). Co-productive modes of governance are often promoted as a tool to increase the legitimacy of public policy (Börzel & Risse, Reference Börzel and Risse2010; O’Flynn, Reference O'Flynn2009), bring more targeted and innovative solutions (Bingham et al., Reference Bingham, Nabatchi and O’Leary2005; Fung, Reference Fung2006) and limit third-sector autonomy and legitimacy (Johnston, Reference Johnston2015). Hamilton and Lubell’s (Reference Hamilton and Lubell2018) findings emphasise that transaction costs and an increase in the transaction costs among participants in an ecology of games can imperil their capacity to cooperate.

In relation to social learning, civil society actors appear to play five roles in domain learning: (1) identifying issues, (2) facilitating the voice of marginalised stakeholders, (3) amplifying the importance of issues, (4) building bridges among diverse stakeholders and (5) monitoring and assessing solutions (Brown & Timmer, Reference Brown and Timmer2006). NGOs are also increasing their influence over public policy through their ability to spread ideas and information (Dolowitz & Marsh, Reference Dolowitz and Marsh2000) and facilitate the scaling up of collaboration and social innovation (Lee & Restrepo, Reference Lee and Restrepo2015).

The system of patronage might be associated with the NGO-isation of civil society, which occurred in the 1980s and the 1990s in Western countries (Bebbington et al., Reference Bebbington, Hickey, Mitlin, Bebbington, Hickey and Mitlin2008; Howell & Pearce, Reference Howell and Pearce2001) and in the 1990s and the 2000s in Central and Eastern Europe (Grünberg, Reference Grünberg2000; Jacobsson & Saxonberg, Reference Jacobsson and Saxonberg2016). The NGO-isation of civil society has been reinforced by the availability of European Union (EU) funding (Karlberg & Jacobsson, Reference Karlberg and Jacobsson2015). Europeanisation through funding entails channelling towards formalisation and professionalisation (Císař & Vráblíková, Reference Císař and Vráblíková2010; Hašková and Křížková, Reference Hašková and Křížková2008), towards more structured and less militant forms of activism (Císař, Reference Císař2010; Císař & Vráblíková, Reference Císař and Vráblíková2013; Císař & Navrátil, Reference Císař and Navrátil2015; Koubek, Reference Koubek2020) and contributes to the shift in agendas to fit funding opportunities (Sudbery, Reference Sudbery2010).

In such fragmented fields, the existence of meta-organisations seems crucial. They might be important actors in balancing dependency and autonomy, decreasing transaction costs for cooperation and actors facilitating social learning.

Meta-Organisation as Civil Society Actors

As Laurent et al. (Reference Laurent, Garaudel, Schmidt and Eynaud2020) pointed out, meta-organisations contribute to the coordination of the various interests of the stakeholders involved in each field and act as a legitimate and collective voice to address public authorities. In relation to the third sector, Karlberg and Jacobsson (Reference Karlberg and Jacobsson2015) discuss civil society umbrella organisations as specific meta-organisations. Ahrne and Brunsson (Reference Ahrne and Brunsson2008, pp. 62–77) point out three general purposes for which a meta-organisation may be formed: interaction among members, collective action among members or creation of a collective identity. Significant economies of scale are among the most important motives for the formation of meta-organisations to lower the costs that individual organisations must pay (Ahrne & Brunsson, Reference Ahrne and Brunsson2008, p. 69). The creation of meta-organisations might also mean introducing a new actor who may fulfil new functions that have been absent in the field. The functions of meta-organisations may be classified as inward functions, such as cooperation, coordination and regulation among members, and outward functions, relating to the management of relationships with external stakeholders (Laurent et al., Reference Laurent, Garaudel, Schmidt and Eynaud2020). These two functions are reflected in two dimensions of the legitimacy of meta-organisations—internal and external. Legitimacy corresponds to the last purpose of meta-organisations mentioned by Ahrne and Brunsson (Reference Ahrne and Brunsson2008).

Meta-organisations can also be viewed, in line with the ecology of games approach, through the lens of transactional cost theory. In the context of civil society, transaction costs can be defined as costs from managing the field itself. They can be divided into six broad categories (Williamson, Reference Williamson1985; North, Reference North1990; Jaffee, Reference Jaffee, Jaffee and Morton1995): (1) search costs (a lack of information), (2) screening costs (uncertainty about actors), (3) bargaining costs (conflicting objectives and interests), (4) transfer costs (transfer of ideas), (5) monitoring costs (watchdog activities) and (6) enforcement costs (influence on other actors). In this paper, we propose that meta-organisations can bear such transactional costs and are crucial in a system where individual organisations’ costs for such activities are too high.

NGOs’ and Migrants’ Integration into the Czech Republic

The Czech migration policy is highly centralised, with the Ministry of the Interior being a crucial actor. Czech migration policy is inconsistent in terms of time switching between liberal and restrictive tendencies (Baršova & Barša, Reference Baršova and Barša2005; Bauerova, Reference Bauerova2018), and it is marked by the coexistence of national and European levels. On the one hand, efforts have been made to introduce more restrictive conditions of entry and stay, while on the other hand, Czech migration policy must follow EU regulations that strengthen civil rights and support permanent settlement (Kušnirákova, Reference Kušnirákova2014). Similar to policies of other CEE countries, Czech migration policy can be described as ‘mainstreaming by accident’ (see Jóźwiak et al., Reference Jóźwiak, Sánchez-Domínguez, Sorando, Scholten and van Breugel2018) in which mainstreaming measures are caused by economic reasons rather than political ones.

The number of foreigners has been growing since 1990, with a short break during the financial crisis. The biggest groups by nationality are consistently Ukrainians, Slovaks, Vietnamese and nationals of the Russian Federation, who altogether constitute around 75% of non-EU migrants in 2018. There are important streams of labour migration from other EU countries: Romania, Bulgaria and Poland. In 2019, the total number of foreign nationals living in the Czech Republic was 593,000, approximately 5.4% of the population (Czech Statistical Office, 2020).

An active civil society and the support of NGOs and local communities are all conducive to successful integration in the EU. In cooperation with local authorities, they play a key role in the social integration domain, delivering social integration measures in local communities (Fóti et al., Reference Fóti, Bešić, Vasileva, van Nierop, Puts, Rose, Ulcica, Vajai and Konle-Seidl2019). The Czech government recognises NGOs as crucial players contributing to MRAS (Migrants, Refugees and Asylum Seekers; Czech Government, 2015). Currently, 25 organisations deal with the integration agenda in the Czech Republic. The NGOs involved in the integration agenda have a relatively homogeneous offering of services. Most NGOs are involved in providing direct services and working with individual clients. Only a limited number of NGOs are active political actors whose aim is to shape the current integration landscape through advocacy and participation in policymaking.

As mentioned, most NGOs are principally involved in service provision, offering direct services to foreigners (the most widespread of which are individual social and legal counselling), followed by language courses and the organisation of social events for both migrant communities and the public. The services provided by NGOs depend on funding from a variety of public resources, provided by ministries, regions and municipalities. Several NGOs benefit from EU funding, such as the Asylum, Migration and Integration Fund (AMIF) and the European Social Fund (ESF). Individual NGOs are successful in securing resources from other international funding schemes, such as European Economic Area (EAA) Grants or Norway Grants. Funding from private donors represents a small share of the finances available to NGOs.

NGOs are highly professionalised, subsidised by the state and only rarely based on volunteerism. Only during the so-called migrant crisis were several rather informal initiatives started (e.g. Pomáháme lidem na útěku [Help for People on the Run], Iniciativa Hlavák [The Main Train Station Initiative]). However, they were established largely as attempts to pass through the crisis period immediately provide migrants with urgent humanitarian aid and help meet their basic needs.

The integration services that target migrants are not evenly distributed. Most service-oriented NGOs operate in the two largest cities of the Czech Republic—Prague and Brno (Bauerová, 2018); however, more than half of migrants live outside these two areas, in cities and towns close to the borderline (Czech Statistical Office, 2020). The offered services, thus, do not reflect the needs of all MRAS, who have increasingly spread across semi-peripheral areas of the Czech Republic (see Čermák & Janska, Reference Čermák and Janska2011). Moreover, migrants have access to a limited spectrum of social services. While social counselling services are available in all major cities, subsequent services that are important for successful labour market integration, such as social housing or street work, are provided almost exclusively in major cities or only in the Czech language (Fišarova, Reference Fišarová2018; Klinecky, Reference Klinecký2012). Migrants in peripheral areas are, thus, more vulnerable to social exclusion and more dependent on their personal connections and networks.

NGOs’ Policy Work and The Ecology of Games of Migration Policies

Following the ecology of games approach, we examine three dominant functions of the system: (1) resource distribution, (2) cooperation and (3) social learning. We seek evidence of these functions at both the state and local levels and disparate fora. These functions, though, are heavily influenced by their context. The hostile context in which NGOs operate creates an environment for invisible and less publicised policy work.

Czech NGOs were affected by the so-called 2014–2015 migrant crisis. Although the immigration trends in the Czech Republic have remained almost unchanged during and even after the crisis, Czech NGOs have had to face the indirect effect of the crisis and, more specifically, the hostile approach towards NGOs. In the media, refugees were depicted as a security threat and an administrative burden partly imposed by the European Union (Jelínková, Reference Jelínková2019). This discourse was replicated by the president’s or political parties’ leaders (see Čada & Frantová, Reference Čada, Frantová, Law, Easat-Daas, Merali and Sayyid2019; Tkaczyk, Reference Tkaczyk2017). In comparison with other EU countries, Czechs scored the highest in hostility towards migrants (see Pospíšilová & Krulichová, Reference Pospíšilová and Krulichová2018). According to the results of the Eurobarometer poll (2018), immigration remains the most important issue facing the EU for 38% of European citizens, with the rate found in the Czech Republic (58%).

Several actors have attempted to damage the public image of NGOs. The negative impact of the public debate can be well illustrated by a remark made by a director of an NGO that was significantly engaged in service provision: ‘It is uncomfortable. Our clients are often people in difficult situations, so it is a hard job to deal with attacks from the public at the same time. We are also worried about what will happen if some political subjects that are calling for the destruction of the civic sector are successful in the election’. All our interviewed NGO representatives admitted to having experienced verbal and symbolic abuses. Prague-based organisations encounter hate speech on their Facebook sites. The negative attitudes towards NGOs are not expressed online only. For example, a local service provider’s office door was glued with anti-refugee stickers in broken Arabic. Similarly, members of another organisation found a hangman figure on their entrance door.

Several NGOs initially aimed to actively contrast the stigmatisation of their work by contributing to the public debate and to attempt to change the hostile public perception of migration. However, they have withdrawn one by one. Having no resources and capacities to contest the negative stereotypical image of NGOs, sometimes legitimised, and even propagated by the highest political representatives, they have decided to focus almost exclusively on less visible initiatives.

As the director of a local service provider stated, ‘Publicity never brought us any good’. Her colleague from a citywide organisation says that this is a task for other organisations specialising in public relations but that ‘their work should speak for itself, and those who matter will get to know about it’. This also explains NGOs’ relatively low involvement in activism and their primary focus on service provision. As argued by a social worker in an organisation that bears the term ‘refugee’ in its name: ‘Our office moved 2 years ago, and we decided not to put the full name of our organisation anywhere around the doors’.

In a hostile environment, publicity is a danger for an organisation. Hostile environments create a crucial context for ecologies in which NGOs play their games, representing an important variable structuring the rules of field and strategies of NGOs. Costs and risks which relate to publicity are perceived as so high that they can ruin the organisation. These costs endanger the survival of the organisation and the stability of personnel. The environment is also characterised by high insecurity. For these reasons, the organisation turns away from activism and public work towards invisible work, which can be associated with transactional activism—lobbying, participating in expert bodies or influencing policy as social service providers.

Meta-Organisations and Transaction Costs

The whole sector is represented by the umbrella organisation—Consortium of Migrants Assisting Organisations. The Consortium has a lean office with a team consisting of five persons (some with part-time contracts only) and uses a diverse portfolio of funding resources, such as small grants from the municipality of Prague, international donors or EU funds. The consortium currently gathers 18 NGOs that are active in migration and represents them in their advocacy initiatives. In fact, the NGO representatives stress that they do not attempt to influence the policymaking process separately and that they prioritise (thanks to common perspectives on the MRAS’ agenda) the representation of NGOs’ voices through the consortium. ‘Services give organisations good knowledge of praxis, and they [the consortium] spread it out through their research and advocacy activities’ (a director of an NGO).

Such an approach might help them to avoid conflating their service provision initiatives—which require a certain degree of alignment with the state—with more critical, advocacy-driven initiatives. ‘This might sound cowardly, but I believe that we need to enter the debate cautiously. I do not want my organisation to face [a] difficult situation if I would make some strong statement about Czechs as xenophobes’ (an NGO representative).

The Consortium organises policy roundtables and conferences. More recently, the consortium has called for the stronger involvement of actors, such as local political representatives and employers, who have only rarely joined integration debates and initiatives. The consortium members are involved in research projects and have published policy papers, participated in the evaluation project and published an overview of legal changes. The role of the consortium is also informational. In particular, the representatives of non-Prague NGOs recognise the importance of the umbrella organisation. They view the consortium as a valuable source of information about legislative changes, migration policy and activities of other organisations. As such, the umbrella organisation has an important networking function and is a source of shared inspiration. ‘Of course, there are differences in approaches of different organisations, but we can talk it through’ (an NGO representative).

Through the lens of transaction costs theory, the Consortium actively searches for information that it shares among its members, looks for policy actors who are willing to cooperate, participates in policy dialogue, transfers ideas through media, and monitors and evaluates policy programmes (see Table 1). Such activities save costs for individual members or provide them with services for which costs will otherwise be too high for individual members. In terms of civil society meta-organisations, search costs relate to research and policy reports, informing members of news in migration policies and providing information on migration law. Screening costs are tied to looking for partners from different fields and establishing a dialogue with them. Bargaining costs are associated with establishing and maintaining platforms for exchange and dialogue among member organisations. Transfer costs are linked to communication with the media and the broad public and the transfer of best practices from NGOs to policymakers. Evaluating public policy and monitoring legislative changes correspond to monitoring costs, and organising policy advocacy is associated with enforcement costs.

Table 1

Meta-organisation and transactional costs

Type of transactional costs	Activities
Search	Preparing policy reports and research compilations on varied topics connected to migration Publishing a newsletter about members’ activities and recent policy changes Providing information on migration law
Screening	Participating in and initiating meetings with relevant stakeholders among politicians, civil servants, experts, businesspeople, academics and others Providing journalists with contacts among experts in the field
Bargaining	Engaging in cooperation and exchange of good practices among members by organising regular meetings of four working groups: social work, legal counselling, advocacy and lobbying, and communication and media
Transfer	Organising debates and seminars on the current most relevant topics Holding cultural events and educational activities Holding special meetings for journalists and publishing press releases Providing background materials on migration-related topics Transferring NGOs’ experiences to policymaking
Monitoring	Evaluating public policies and monitoring new legislative changes
Enforcement	Engaging in policy advocacy, negotiations with policymakers and power relations
Performance	Representing the field symbolically and institutionally undertaking critical policy work
Development	Conducting professional training and workshops Strengthening reflexivity and innovation Evolving the entire field

Other types of transaction costs also are mitigated: (1) performance and (2) development. Performance costs represent the field symbolically and institutionally. When the performance costs for critical action are too high due to the dependency of individual organisations on funding and their precarious conditions, the meta-organisation can perform critically on their behalf. Performance costs are particularly relevant in the Czech context, where the attempts of the third sector are viewed with hostility. Finally, maintaining the meta-organisation is also a way to make the field itself visible. Development costs are related to the transformation processes and the temporal evolvement of organisations. The consortium organises professional training and workshops on topics relevant to the work of its member organisations. The participation in meta-organisations provides individual member organisations with a platform for reflection, innovation and development.