A Platform for Collaboration: The Views of Civil Society Organisations on a Local Compact

The first condition relates to power and resource imbalances between actors, negatively affecting collaboration. Still, even if collaborative governance aims to be inclusive, previous research shows that CSOs are underrepresented in platforms and collaborations not exclusively designed to include them (Carboni et al., 2017). Overcoming imbalances will require a ‘positive strategy of empowerment and representation of weaker or disadvantaged stakeholders’ (Ansell & Gash, 2007, pp. 551–552). Otherwise, minority CSOs might feel forgotten or excluded (Craig et al., 2005).

An essential arena for power in collaborative governance concerns the content, that is, what issues are addressed and pursued (Purdy, 2012). For example, Norfeldt (2012) argues that GOs adjust the contents of compacts to fit the present political agenda. Therefore, GOs can limit funding to activities they do not politically support. Consequently, the content depends on the political leadership currently in power (for other contexts, see Alcock, 2016).

Resource imbalance has been discussed in relation to time, as participation often is time-consuming (Ansell & Gash, 2007; Craig et al., 2005; Johansson, 2011; Klockmo & Bostedt, 2015). In Sweden, smaller CSOs without employed staff generally find it challenging to participate in compact meetings during the daytime due to work conflicts (Johansson, 2011; see also Craig et al., 2005 for an international context). Even if representatives can participate during the daytime, their travel costs are not covered (Klockmo & Bostedt, 2015). As most small CSOs do not have the budget to cover these costs, it indirectly forces representatives from smaller CSOs to use private money or refrain from participating. Consequently, collaborative governance risks becoming a matter for a small group of large CSOs or umbrella organisations (Johansson, 2011), reproducing already strong voices (Fyfe, 2005; Laforest & Orsini, 2005; Nordfeldt, 2012). If imbalances in power and resources are high, important actors are limited from participating in a meaningful way (Ansell & Gash, 2007), and the collaborative platform risks becoming irrelevant.

The incentives for participation are also affected by the potential history of conflicts between involved actors, which is closely related to trust among stakeholders (Ansell & Gash, 2007). The feeling of us versus them might limit the incentives for initiating collaboration. If conflicts are high, stakeholders are highly interdependent, and a deadlock needs to be solved; nevertheless, researchers in resource management have shown that previous conflicts are not a barrier to collaborative governance (Futrell, 2003; Weber, 2003). In other words, low levels of trust and previous conflicts might be overcome if the positive aspects of solving the problem are high. Still, the positive aspects might be difficult to grasp when the goals and potential gains from participating are unclear, like in many compact collaborations (Brock, 2020; Morison, 2000; Zimmeck, 2010). Instead, if a CSO has politically and ideologically solid opinions, they might be less likely to participate in collaboration platforms with a GO (Jönsson & Scaramuzzino, 2016). Consequently, the political environment might create conflict and a barrier to participation (Craig et al., 2005) as compacts rest on political–ideological values. On the other hand, some CSOs will always stand on the side (Morison, 2000).

Even if participation in platforms such as compacts is voluntary (Ansell & Miura, 2020; Ansell & Gash, 2007, 2018; Emerson et al., 2011; Morison, 2000; Reuter et al., 2012), the incentives for participating depend on many factors. The expectations CSOs have on potential effects or gains (Ansell & Miura, 2020) and whether the collaboration is workable and worthwhile (Gash, 2016) need to be determined. If the goals are overly broad or open-ended, which is typical of collaborative platforms (Ansell & Gash, 2018), it might be difficult for CSOs to know what they have to gain (Ansell & Gash, 2007) and how (Morison, 2000; Zimmeck, 2010). If CSOs have an alternative venue that they think is better suited for their goals, incentives might be reduced (Ansell & Gash, 2007).

The incentive might also depend on how CSOs understand their potential to affect the platform’s content. Some CSOs interpret collaborative governance through compacts as a governing tool rather than collaborative governance (Klockmo & Bostedt, 2015), making it harder to involve CSOs as they resist the idea of being governed. When negotiating for a national compact in Sweden, some CSOs argued that the compact’s framework was already fixed by the government, making it difficult to affect the final agreement. Others argued that pre-framing was vital for the negotiation and discussion (Johansson, 2011).

To conclude, collaborative platforms and collaborative governance can solve complex societal situations (Gash, 2016), but many obstacles must be addressed to succeed. Still, few studies offer evidence that collaborative governance works well to avoid top-down approaches to governance (Gash, 2016).

In total, six interviews with representatives from CSOs supporting refugees in Malmö are included (Table 1). Some CSOs started to support refugees during the increased influx of refugees in 2015–2016 and began actively supporting refugees. Others have supported refugees for a more extended period, often with the primary aim of helping vulnerable groups, including refugees (Karlsson & Vamstad, 2020).

The included CSOs can be viewed primarily as two distinct groups of organisations. Some CSOs work on human and social rights (HSR), meaning that they regularly assist with different requirements, such as food and shelter, language training, legal advice and job coaching, and are involved in refugee rights issues. Another group of CSOs work to educate, help and enable (EHE) other CSOs that support refugees. The work of EHE CSOs includes support for starting new programs for or by refugees or providing other CSOs with methods to welcome refugees in their already-established activities. Two CSOs (CSO 1 and 6) exclusively work with refugee/immigrant communities. The others organise support and activities for multiple groups in society.

Of the six CSOs, three were identified from a book (Sjöberg, 2018) describing the experiences of various CSOs from the acute situation in 2015–2016; two were traced through a network organisation for CSOs supporting immigrants and refugees in Malmö; and one was identified using snowball sampling (Bryman, 2016).

The CSOs were approached by e-mail and informed about the research aim. After showing interest in participating, they received information about ethical considerations and informed consent. Ethical aspects were raised again at the interviews, partly to guarantee that the respondents had read the information and to offer the opportunity to raise questions or concerns. All respondents provided verbal and written consent to participate. Following ethical standards, the identities of the respondents and the CSOs remain confidential in this article to protect their integrity and efforts.

The interviews with top managers or operative managers were conducted at two different times. The first round, in 2019, was conducted in Malmö. The second round of interviews in 2020 was conducted via video calls owing to the Covid-19 pandemic. The interviews were semi-structured with five broad and general thematic approaches (related to the primary research project): the organisation’s structure and activities, relations and interactions with other CSOs and LGOs (including MA), flows of information and knowledge, roles and changes since the increased influx of refugees in 2015. The interviews lasted 53–95 minutes and were recorded and transcribed. All interviews were conducted before the renegotiation process for MA. Consequently, there was no discussion of the new collaboration plan.

Before the new collaboration plan for MA was accepted at the beginning of 2021, the proposed plan was sent to CSOs and the local state administrators for a referral round. As the interviews were conducted before the renegotiation in 2021, the submitted referral documents were included to complement the interviews and to capture the views of the CSOs on the new plan. The referral documents are included in the Swedish accessibility principle (a law giving individuals, including media and researchers, the right to access information about governmental work). All submitted referral documents were downloaded from Malmö’s online citizen portal (Table 2). In total, five large and well-established CSOs in Malmö submitted documents, including two individual organisations and three umbrella organisations. All umbrella organisations support and/or include CSOs that care for refugees.

The CSOs that submitted referral documents are active partners in MA and suggested what they think needs to be rewritten, removed or renegotiated. One CSO wrote indirectly that it does not stand behind the new plan in its present shape.

Thematic analysis (Braun & Clarke, 2006) was used and, initially, only included interviews. After reading the transcripts and interview notes, all texts related to MA were selected. Two broad themes, views on MA and starting conditions, were identified based on the theoretical point of departure. Second, the interviews were recoded to identify wrongly coded or uncoded parts. Third, the two themes were separately coded into identified sub-themes. The views of MA were organised according to various opinions of MA. The starting conditions were organised based on theoretical starting conditions. However, history does not strictly predict past conflicts; rather, the views of CSOs need to be understood as a way for them to anticipate the potential future(s) with MA. Also, there was no sub-theme related to incentives to participate, as suggested by Ansell and Gash (2007). Instead, the incentives in all sub-themes are discussed throughout this article.

Finally, the referral documents were added to the empirical data and analysed based on the identified themes. In these documents, the CSOs exhibited a more positive attitude towards MA, possibly because they are active partners and are not anonymous in the referral documents. Still, potential problems noted in the interviews were identified in the referral documents. Thus, they supported the analysis, confirmed the interview findings and provided this study with a more comprehensive range of CSO voices.

In every referral document, MA is described as ‘positive’ (R1) and an ‘arena for collaboration […] that should be further developed’ (R2). Supporting refugees is a subject surrounded by different political views and opinions concerning who should be able to apply for asylum, what support refugees should receive from the official reception system and what the role of CSOs should be. Therefore, close relations between active CSOs and the LGO are crucial (cf., Gustafsson & Johansson, 2018).

Some respondents argued that the LGO in Malmö has a history of not collaborating with civil society and that MA might be a way to ‘overcome their fear of collaboration with civil society’ (I3). Even if this specific CSO representative was highly critical of the LGO and MA, the collaborative platform might contribute to a better relationship even if the CSO did not participate in MA.

Another positive aspect raised was that MA might be one of the few venues in which participation is not forced into a specific shape, but instead, collaboration may take different forms:

‘We [society] try to fit everything into a system, and that is not sustainable over time […] When we take all the commitment and try to squeeze it into a pipe, well, it dies. Malmöandan may be an example of when we do not do that’ (I1).

Thus, from the perspective of CSOs, MA can increase collaboration and include various types of collaboration. In theory, participating in collaborative platforms should be beneficial for CSOs because it would benefit relations with the LGO and other CSOs (Ansell & Gash, 2007, 2018; Emerson et al., 2011; Gash, 2016), resulting in equal power distribution and horizontal relations (Ansell & Miura, 2020). Participating in compacts allows CSOs to affect change from within (Johansson, 2011) and be a part of the ‘official’ arena. Changing from within might be essential for CSOs, since ‘the political climate is becoming harsher, with some CSOs subjected to threats directed at their organisation and activities’ (I1, I6). Participation could be a way for CSOs to raise their voices, affect the political agenda and collaborate to solve complex problems. Furthermore, participation could show the LGO that the CSO is a legitimate partner for collaboration and agree to the game’s rules. As a result, some CSOs participate despite raising critical concerns regarding MA.

The central problem and scepticism raised are that it is difficult for CSOs to understand what they have to gain from participating (cf., Ansell & Gash, 2007, 2018). In this empirical case, the vagueness was related to the formulations in the plan that are the foundation of the collaborative platform and how actual outcomes are (not) presented.

Even if the CSOs generally were less sceptical in their referral documents compared to interviewed CSOs, they were highly doubtful concerning the content and results.

Compacts as governing tools are often criticised and described as a ‘handsome face with not much behind it’ (Zimmeck, 2010, p. 131). When potential gains and results are clear, CSOs are more likely to find collaborative platforms valuable and spend time, energy and resources to participate (Ansell & Gash, 2018). Still, according to the empirical data, only CSOs that are active partners raised concerns about the results of MA. Non-participating CSOs did not consider the possible effects produced by MA. Instead, they decided not to participate based on other values. 7

The vague formulations do not only concern what the collaborative platforms achieve but they also make it difficult for participants to see how they can influence the content, that is, how power is distributed. Put differently, they make CSOs question for whom and by whom the collaborative platform is constructed.

Collaborative governance must be driven by demand (Gash, 2016). The respondents mainly argued that there is a lack of communication regarding possibilities to influence. There is a risk that CSOs interpret the platform as a top-down project (cf., Klockmo & Bostedt, 2015) without horizontal collaborative dimensions, which are essential for collaborative governance (Ansell & Gash, 2018; Ansell & Miura, 2020). According to Ansell and Gash’s (2007) model for collaborative governance, the vague content should affect the incentives for participation (see also Klockmo & Bostedt, 2015, for an empirical example in Sweden). Still, for CSOs participating, the unclear power distribution was insufficient to terminate their participation. For CSOs that are non-members of MA, their decision to participate was instead based on financial and personal resource asymmetry.

Representatives from the LGO often participate in collaborative governance as part of their job. For CSOs, conditions are often different (Craig et al., 2005; Johansson, 2011; Klockmo & Bostedt, 2015; Purdy, 2012). Many CSOs struggle to cover basic costs; therefore, participating in collaborative platforms is challenging, especially for smaller CSOs struggling financially. If a CSO’s future is unclear, entering new platforms is not prioritised.

Even if participation in MA can be beneficial in the long run due to increased collaboration and project funding, most CSOs have to participate in their spare time, which CSO representatives criticise.

Understanding collaborative platforms’ power and resource imbalances is essential (Ansell & Gash, 2007; Purdy, 2012). When CSOs lack the necessary means to participate in a meaningful way and perceive an unwillingness from the LGO to work actively to include them, CSOs might risk becoming underrepresented (Carboni et al., 2017). In this empirical case, limited time and resources made it challenging for CSOs to participate. Therefore, unequal access to participation might result in already strong voices reproduction (Morison, 2000). Consequently, there is a risk that ‘less heard voices’ will be lost in setting the agenda and discussing how to solve complex societal challenges.

A history of conflicts might limit CSOs’ incentives to actively participate in collaborative platforms (Ansell & Gash, 2007). Discussions concerning previous conflicts were not evident in this specific material or discussed as limiting incentives. Instead, how CSOs relate to previous collaboration experiences with the LGO and their understanding of MA were important. The problematisation of participating relates to two intertwined dimensions: (i) political values to distinguish themselves from a specific political agenda and (ii) ideological values and ideas of how society should be organised. 8

MA is a collaborative platform that aims to support negotiation and discussion for future collaboration. Thus, in theory, collaboration platforms and collaborative governance should facilitate equality of power and horizontal relations (Ansell & Gash, 2007, 2018; Purdy, 2012). However, several CSOs interpreted MA as a way for the LGO to control and govern their actions. Other CSOs reflect on how they might be governed by participating in MA:

‘If they [LGO] promote a policy you disagree with, and you must adjust your activities, you might not be able to do what you want to do…. [And] you must do what someone else thinks is needed’ (I3).

Even if MA does not govern CSOs to act in a specific way, this was the fear of CSOs. When asking if collaboration, including funding, would restrict them as an organisation from supporting refugees, the answer seemed obvious:

‘Yes, most definitely […] [F]or us, it is much easier to go outside the box and follow the needs of the target group, and we would not be able to do that […] [W]e want to be able to do what we want to, and our work frequently entails activism’ (I6).

Because MA aims to create LGO–CSO collaboration, legitimisation and exhibiting a willingness to participate by the rules are not considered necessary; therefore, participation is unnecessary (cf., Morison, 2000). As the CSOs in this study aimed to help and support refugees, how they discussed MA addressed the worries of being governed, the potential effects MA will have and the values and ideological dimensions behind the platform.

For example, several respondents were critical of how the municipality supported refugees during the refugee situation in 2015/2016 and argued that the LGO was not doing enough, that is, they opposed their politics. In this context, being governed by political actors with the ‘wrong’ political agenda is unwelcome, and it creates a feeling of ‘us’ versus ‘them’ (cf., Nordfeldt, 2012). Others pointed out that since Sweden Democrats, a right-wing party in Malmö that seeks to ‘pause’ immigration and establish ‘return centres’ for all residents not born in Sweden, signed the document, there is clear evidence that the document only consists of beautiful words with little meaning (cf., Morison, 2000; Zimmeck, 2010). For CSOs working exclusively with refugees, this is of great importance as political leaders are a part of the steering group. To summarise, the ideological and political values that MA is based on are essential to understand the relations of CSOs to MA. The CSOs feared that active participation could limit their ability to work as they wished.

Another essential aspect of the viewpoint of CSOs to MA is related to how CSOs think society should be organised and what relationship CSOs in Malmö should have with the LGO. Some CSOs believed that MA had replaced previous ways to initiate partnership discussions. When discussing different types of collaboration between Malmö and CSOs and whether MA had replaced many of the previously established relations, one CSO representative said:

‘Yes, at least, it seems like it. When I asked for a meeting regarding a CSPP [with the LGO], I got the answer “MA [already] exists”. So, what you can say is that, at least in Malmö, it is like that’ (I2).

By comparing the past ways of initiating collaborations with their understanding of the present with MA, the CSOs were critical of how MA, against its aims, seemed to limit the possibility of CSOs forming a CSPP. The respondents’ statements could be understood from two perspectives. One understanding is that MA has replaced CSPPs, that is, limited the possibilities of CSOs to collaborate with the LGO. Another interpretation is that MA, as a collaborative platform serves as a gatekeeper for the LGO and that all types of new collaborations should pass through MA. Consequently, if CSOs do not participate in MA, they have no opportunity to form a CSPP. Either way, the incentives for CSOs to participate might have increased.

Nevertheless, the incentives are not based on voluntary participation (cf., Ansell & Gash, 2007, 2018). Rather, participating is indirectly forced as it might be the only way to form a CSPP. The combined narratives of the CSOs indicate that they understand the collaborative platform as a top-down project, attempting to govern the actions of CSOs and organise society by LGO political ideology.

In the interviews, members and non-members discussed how they anticipate the future, what challenges there are, and how they think complex societal challenges should be solved. Generally, they recognised that social problems are increasing. CSOs meet more individuals in acute need of support struggling with homelessness, drug use and mental instability. In the opinion of CSOs, this has to do with GOs at national, regional and local levels withdrawing from the market and LGOs trying to shift their responsibility to CSOs. When the market (here CSOs) takes greater responsibility for vulnerable groups, collaborative platforms like MA fit well as a governing tool. Therefore, the traditional role of CSOs in Sweden as the voice for vulnerable groups rather than a service provider is challenged (Wijkström, 2000, 2007; von Essen, 2019). The question raised is what role voluntary actions should have in a society and if it is reasonable for an LGO to hand over greater responsibility to CSOs. As one respondent argues, ‘there is a risk that MA bites its tail if voluntary action becomes a kind of must’ (I2).