Renegotiating state-third sector relations through collaborative partnerships: The case of reception services for asylum-seeking children in Gothenburg, Sweden

ABSTRACT

Unprecedented numbers of unaccompanied minors entering Sweden in 2015, alongside other challenging social issues, have contributed to renegotiating collaborative forms of local governance in the country. This has resulted in an emergence of collaborative partnerships between civil society and public actors (Idéburna offentliga partnerskap or IOP), as real alternative to the market-based contracts and traditional state grants. This article addresses IOP from a collaborative governance perspective and questions if and how it may cope with a major and well known governance challenge, that of balancing the different roles and principles of public and civil society realms, so as to sustain an effective and legitimate cross-sectorial partnership. It asks, may and how an IOP enable civil society organisations a more active role in co-designing and co-implementing local welfare services with public sector actors without undermining their distinctiveness? The arguments are based on a case study of, at the time, the largest IOP in Sweden initiated for early reception of unaccompanied minors in Gothenburg city. The partnership is assessed against a normative theoretical framework of balanced state-third sector relations. The results suggest that IOPs, under certain conditions, may serve as ‘spaces of hope’ for more balanced power relations in local governance models.

KEYWORDS:

• State-third sector partnerships
• collaborative governance
• power balancing
• governance innovations
• relational governance

Introduction

For several decades, the nature of the relations between civil society organizations (CSOs) and the state in addressing demands for adequate and effective welfare policies has been a focus of debate (Acheson, 2001; Lewis, 1999; Philips & Rathgleb Smith, 2012; Proulx et al., 2007). It has been argued that especially in Western European contexts collaborative logic, often grounded in corporatist traditions, has been diverted by the global marketization trend. The influential collaboration schemes, based on ‘contracting out’ and ‘competitive commissioning’ logic, have distorted the CSO distinctiveness, autonomy and advocacy role in human service provision. Subsequently, this has undermined CSO innovation and democratization potential in addressing complex social issues, thus depriving services of important public and private values (Bode & Brandsen, 2014; Hogg & Baines, 2011; Philips & Rathgleb Smith, 2012; Salamon & Toepler, 2015; Trädgård, 2012).

In response to decreasing effectiveness and legitimacy of the prevalent competition and hierarchic control-based collaboration models, collaborative state-third sector partnerships, where power is shared on more equal terms, have been suggested in countries such as United Kingdom, Scotland, Sweden, Canada and the US (Finansdepartementet, 2019; Lewis, 1999; Proulx et al., 2007; Salamon, 1999). Arguably, collaboration models based on co-joined, dialogue-based decision-making and respect for CSO specificity and independence offered better possibilities to tap into CSO innovation potential and other resources including, deep knowledge especially of disadvantaged social groups’ needs, capacity to mobilize volunteer resources and prompt action and build trustful relationships (Gavelin et al., 2010; Salamon & Toepler, 2015). Power sharing, by including CSOs in co-governance of the partnership and co-management of its outputs without undermining important aspects of third-sector autonomy, has been argued to empower the third sector and enhance its commitment to partnerships with the state (ibid; Philips & Rathgleb Smith, 2012). Local partnerships that aim for collaborative, relational governance may be seen both as an evolution of corporatist traditions in parts of continental Europe, such as Germany or France and as novel developments in other contexts, such as Hungary (ibid). Instead of occupying or substituting state functions (Klein & Lee, 2019) CSOs may become service co-producers with the state. Collaboration that goes beyond ‘zero-sum’ game relations where one sector is empowered at the expense of the other (ibid) and instead aims to enhance the capacities of both require distinct from hierarchy and market output steering. The research that focuses on collaborative policy governance through cross-sectoral networks suggests that such, often more horizontal governance, requires using a greater variety of tools, including deliberation, co-joint decision formats and state steering based on trust rather than pure control to tap into the third sector potential (Ansell, 2019; Ansell & Gash, 2008).

Despite being seen as a promising collaboration archetype within the New Public Governance paradigm, local partnerships, based on collaborative governance principles, have been notoriously difficult to establish and sustain, notwithstanding the corporatist traditions and soft regulations via recent collaboration compacts (Philips & Rathgleb Smith, 2012; Provan et al., 2007; Provan & Kenis, 2008). This is partly explained by a misfit in prevalent regulatory frameworks and the predominance of policy tools that support market logic (Acheson, 2001; Philips & Rathgleb Smith, 2012) including competition and principle-agent relations. These conditions impede attempts to equilibrate the powers of two sectors and exploit the innovation potential in collaborative models by accommodating CSO interests and resources (Ansell & Gash, 2008; Brandsen, 2010; Dickinson & Glasby, 2001; Salamon & Toepler, 2015).

While collaborative governance research should be credited for explicating important relational aspects, such as collective decision-making, ownership of the policy process and its institutionalization (Klijn, 2008; Vangen et al., 2015), important gaps remain in understanding the micro-dynamics of the power balancing game and its importance for CSOs’ commitment to a partnership with the state and partnership synergy (Ansell & Gash, 2008; Freise, 2010; Klijn, 2008; Lasker et al., 2001; Salamon & Toepler, 2015). The empirical evidence of policy tools and strategies that support the equilibration of powers in partnerships aiming for collaborative governance to achieve joined goals or co-produce public services remains scarce (Klein & Lee, 2019; Salamon & Toepler, 2015).

This study questions whether currently emerging particular forms of partnerships in human services between voluntary and public sectors in Sweden (idéburna offentliga partnerskap, hereafter IOPs) may offer cases of relational and more power equilibrated collaborative governance, and if so, how, including tools, strategies and their impact on CSO commitment. The IOP model distances itself from collaboration based on competitive contracts and the principle-agent model, also seen as mistrust-based steering (Montin, 2016), by favouring the more trust-based steering and relational principles of the compacts. It aims at co-joined decision-making, implementation and governance in a covenant format (Ansell, 2019) to address complex social issues. In legal and financial terms, IOP falls under a state grant scheme but involves shared resource contributions of both sectors. Instead of contracts, such partnerships build on mutual agreement on longer-term commitment and learning with the possibility to adjust its outputs (Forum, 2016).

Drawing on the longitudinal analysis of a strategically selected real-life IOP operating 2015-2018 for the reception and integration of unaccompanied asylum-seeking children the study contributes to research on state–third sector partnerships and collaborative governance by offering insights into tools and strategies to support equilibration of powers to sustain especially CSO commitment to partnership with the state. The study develops a synthesized analytical framework for assessing and informing a reconfiguration of state-third sector partnerships towards more equilibrated powers in collaborative governance.

In what follows, the article first presents a conceptual framework for more power-balanced collaborative governance. A section on methodology follows, and the article moves on to illustrate some novel relational aspects of IOP in the Swedish context of the historically established state-third sector collaboration. It then explores and assesses the selected partnership model from the perspective of collaborative governance. The concluding section sums up the findings and discusses the implications for further research.

Theory

State-third sector collaboration is prompted by sectors’ complementarity in addressing such issues as public sector accountability failures, knowledge gaps or institutional capacity. Partnerships aiming at more relational, collaborative governance may especially appeal to less powerful CSOs that are eager to impact local policies (Ansell & Gash, 2008). Yet, collaborative governance attempts frequently fail due to strong steering by the local state or adherence to market principles (Bode & Brandsen, 2014; Salamon & Toepler, 2015).

Balancing powers, interests and roles has been found essential for partnership ‘internal success’ (Peters, 2012), in terms of its functionality, legitimacy (not least among CSOs) and sustainability (Salamon & Toepler, 2015). A more balanced partnership, in turn, may increase its ‘external success’ (Peters, 2012), including legitimacy, innovativeness, resource mobilization and implementation effectiveness (Bode & Brandsen, 2014; Hogg & Baines, 2011; Salamon & Toepler, 2015; Trädgård, 2012). The large literature on collaborative governance and cross-sectorial partnerships (see for example Ansell & Gash, 2008; Dickinson & Glasby, 2001) suggests further exploring and theorizing tools and mechanisms support the balancing of sector’s powers, interests and roles as a major research concern, in the face of divergent sectorial interests and governance logics (Mörth & Sahlin-Andersson, 2006; Philips & Rathgleb Smith, 2012; Salamon & Toepler, 2015; Vangen et al., 2015).

Power relations may be seen as multidimensional and there are still too few studies developing tools to assess power balancing in state-third sector collaborations (Salamon & Toepler, 2015), with Bar-Nir and Gal (2011) and Proulx et al. (2007) as few exceptions. While providing useful indicators for assessing individual actor power in cooperation or classifying varieties of state-third sector interface, such studies devote limited attention to specific power balancing mechanisms at a micro level to achieve more internal success in collaborative arrangements.

Salamon and Toepler (2015) highlight three key normative criteria with guiding principles to initiate and sustain collaborative arrangements to balance state and third-sector powers, disparate interests and roles.

The first criterion requires engaging CSOs in both the implementation of government programmes and their design, effecting a collaborative style of governance and management (ibid., p. 2174). Salamon and Toepler do not provide much further guidance, but collaborative governance elsewhere is defined as ‘a governing arrangement where one or more public agencies directly engage non-state stakeholders in a collective decision-making process that is formal, consensus-oriented, and deliberative and that aims to make or implement public policy or manage public programs or assets.’ (Ansell & Gash, 2008, p. 544). Agreements set in national compacts (Regeringskansliet, 2009; 2010) are illustrative macro-strategies (Salamon & Toepler, 2015). Similarly, decision-making based on commonly agreed ground rules and collaborative leadership (Ansell & Gash, 2008) has been emphasized among local strategies.

Drawing from governance network literature (Sørensen & Törfing, 2009), this study distinguishes between two types of domains for collaborative governance: institutional design and process management. A Joint institutional design is here taken to imply a more indirect, hands-off partner influence on partnership institutional procedures, such as its composition, scope, ground rules and contents (cf. Ansell & Gash, 2008; Klijn & Edelenbos, 2012; Sørensen & Törfing, 2009). Collaborative process management and leadership refer to allowing partner impact a variety of hands-on governance strategies that facilitate daily partner collaboration, such as activation of resources, creation of organizational arrangements, joint knowledge production, mediation in conflicts or trust creation (Klijn & Edelenbos, 2012; Sørensen & Törfing, 2009).

The second criterion requires flexibility and adjustment of partnership tools and steering mode, such as performance requirements and degree of discretion, to CSO needs (Salamon & Toepler, 2015). Public actors are to balance between exercising a legitimate degree of democratic control via regulative and financial tools and trust-based steering strategies. Adjusting financial tools may imply a shared exercise of discretion over the use of public resources. Salamon and Toepler (2015) do not specify the mutual strategies that might be appropriate for adjusting the steering mode and building trust. Other literature suggests it may involve steering techniques contrary to authoritative or mistrust-based steering specific to principle-agent relations in regulatory policies (Montin, 2016). The Swedish National Compacts (Regeringskansliet, 2009; 2010) and Social Forum (Forum, 2010) suggest adherence to the following relational norms: (i) mutual openness and access to information; (ii) dialogue-based decision-making; (iii) long-sightedness in relationships avoiding detailed steering; (vi) respect for sectorial differences allowing civil society distinctiveness and an independent critical voice, (iv) accountability.

As the third criterion for balanced partnerships, Salamon and Toepler (2015) identify the ability of public actors to make use of the CSO’s capabilities in ways that allow preserving their distinctive character, especially independence to act as a critical ‘voice’ towards government on behalf of important segments of society.

In sum, a relevant management style and a careful selection of collaborative tools to alleviate CSOs’ access to and influence in welfare service delivery without undermining their specificity are deemed necessary.

This study adopts a normative framework of analysis (see Table 1) based on the three relational cues or criteria and associated principles to guide in particular public partners’ towards more equilibrated collaborative partnerships. The framework is a synthesis of ideas by Salamon and Toepler (2015), Klijn and Edelenbos (2012), the National Compact (Regeringskansliet, 2009) and Forum (2010) and is here used to explore and assess IOP practices to further refine collaborative strategies to equilibrate powers.

Table 1. A framework of analysis for balanced collaborative governance in CSO and public sector partnerships. ³

Criteria for balanced partnerships ‘WHAT’	Guiding principles ‘HOW’
Collaborative style of governance	Joint institutional design Shared leadership and management Adjusting financial tools Counterbalancing influence
Adjusted steering mode and tools	Trust-based steering Monitoring through dialogues and avoiding detailed regulations Respect, openness and transparency Long-term perspective Focus on service quality, diversity, coherence Facilitative leadership
CSO’s abilities to sustain their voice	Allowing independent CSO’s voice in partnership governance and management Sustaining CSO’s critical voice towards public policies Preserving a level of independence in partnership service delivery

Methodology

The innovative IOP partnership for the reception of unaccompanied asylum-seeking children in Gothenburg city was selected strategically for the study, as an inspirational case (Alvesson & Sköldberg, 2008) because of its collaborative and relational governance ambitions, a large number of CSOs (9), its relative longevity (3 years) and operation under politically and financially fluctuating circumstances – from favourable to adversarial. It was an IOP with the largest public and private financing in Sweden at the time (2015–2018) that operated in a heavily regulated reception policy area. The research has been conducted as a longitudinal, real-time study during 2016-2018 focusing on the strategies and occasional struggles of power balance at different partnership stages.

Departing from an abductive approach (Alvesson & Sköldberg, 2008), data collection was guided by variables deemed essential to a well-functioning and integrated partnership in the collaborative governance literature (Peters, 2012). The principles and collaborative strategies employed in the IOP were further explored and assessed against a synthesized conceptual framework for power balancing in partnerships (Table 1) seeking a deeper understanding of its mechanisms and appeal especially from the CSO perspective. The adopted approach also allowed for testing the relevance of the suggested criteria for power-balanced partnerships– collaborative style of governance, adjusted tools and steering mode, and sustained CSO independence – and identifying complementary strategies.

The results of the study largely build on data from 40 qualitative semi-structured interviews, additional conversations with the participants from the joined IOP Steering Group, Collaboration Group, Working Group for housing directors, four focus group interviews with CSO housing staff and four focus group interviews with targeted children. Interviewees were chosen on the basis of their role in IOP structures and services. Also, systematic participant observations conducted at partnership meetings in Steering and Collaboration groups between August 2016 and June 2017 were included. Secondary data sources were governmental bills, partnership documents, such as agreements, meeting minutes, partner organization activity reports, and documented advocacy activities.

The interviews were conducted at several partnership phases and included questions about IOP initiation, local authority roles and relationships with the CSOs, perceived IOP challenges, coping strategies and partner and targeted group assessment of partnership success. Success here is understood not only as a positive partnership impact on service implementation and targeted youngsters, but primarily in terms of ‘internal success’ or sustaining collaborative relations and equilibrating powers. Participant observations focused on what questions, problems and solutions participants chose to highlight, as well as on their relationships with the partners. The collected texts were analyzed by multiple readings and manual coding of all relevant utterances of arguments and strategies under the three conceptual categories: a collaborative style of governance, steering mode and tools and sustained CSO independence.

The possibility to follow the partnership over a period of time facilitated access and understanding of its internal dynamics. To validate the findings, the analysis of partnership relations was discussed at two seminars with the partnership participants in September 2016 and May 2017. Such data triangulation served to increase data reliability and relevance.

The study’s reliance on a single case prevents a broader generalization of the power balancing criteria and mechanisms of this partnership format, which is clearly a limitation. Nevertheless, the study offers a synthesized analytical framework of potential relevance for exploring relational partnership models in Sweden and elsewhere and opens avenues for its further development.

Background

In Sweden, the roots of a close collaboration between the state and CSOs may be found in CSO involvement in the public welfare building since the early twentieth century. Yet, the Swedish and the Nordic welfare model was characterized by an increasingly limited and mostly complementary role of CSOs in welfare service delivery (Johansson, 2001; Trädgård, 2012; Wijkström, 2012), prevailed by the state. Corporatism has opened for CSO’s influence on public policy formulation, such as labour market issues (Larsson & Bäck, 2008), but their primary role has been expressive (Arvidson et al., 2018) by voicing citizen or member concerns and policy critique.

Since the 1980s, in line with NPM ideology, the state has started acknowledging its greater dependency on market actors and CSOs (Johansson, 2001). Legal acts of public procurement and user-choice (re)diversified CSO’s roles by drawing attention to service delivery as in the pre-welfare state times. In the Swedish context, it implied a renegotiation of CSO’s role in public service provision based on competitive bids and contracts as key collaboration schemes (Wijkström, 2012). This has effected an asymmetric collaboration model in power terms that was not necessarily equipped to exploit CSO innovation, distinctiveness and advocacy potential (Henriksen et al., 2012; Kelly, 2007; Trädgård, 2012). The market-based relational and financial model is one of the reasons, besides state ideological reluctance to open up services for new actors, for the role of CSOs remaining comparatively low in welfare service delivery¹ (Finansdepartementet, 2019; Trädgård, 2012; Wijkström, 2012), although the current relations between the two sectors in Sweden do not unilaterally fall under the marketization trend (Reuter, 2012) as there are local variations due to the historical context (Arvidson et al., 2018; Johansson, 2001).

The recent national collaboration compacts (Regeringskansliet, 2009; 2010) stress a more active role for CSOs in welfare policies and under different conditions than those afforded by state grants- or market-based contracts (Riksrevisionen 2014:3) and specify key guiding norms for more relational collaboration including dialogical, deliberative decision-making. Consequently, IOPs have, in Sweden, emerged as a distinct model of such relational collaboration between value-based CSOs and the (mostly local, regional) government agencies with its first case at Västerås municipality in 2012. Social Forum, an umbrella organization for value-based CSOs, has been an eager early promoter of the IOP model (Forum, 2010; Forum, 2016).

IOPs are aimed at complex or new social challenges, such as immigrant integration, the homeless or early school dropouts. Juridically, such partnerships are reserved for the collaborative development of public services of general interest (defined by national legislations) where there is no established market but where the initiatives are in line with municipal policy goals.

In practice, the IOPs are still under development and no specific governance model or strategies have been distinguished to guide its implementation. As of June 2016, at least 51 IOPs were signed in three regions out of 21 and 13 out of 290 municipalities (Sandberg, 2016) and by 2020 they numbered several hundred, but official statistics are missing. The central government has been hesitant in meta-governing this collaboration form by clarifying its judicial status and application rules (Finansdepartementet, 2016; Forum, 2016; Kulturdepartementet, 2016). The initiation of IOPs has rather been justified politically on a case-by-case basis.

The study results are based on a longitudinal study of the governance of the largest IOP operating 2015-2018 for the reception and integration of unaccompanied asylum-seeking children (juridically individuals under age 18 who arrived without accompanying adults) in Gothenburg, Sweden, under dynamically shifting political and demographic context.

IOP in Gothenburg

In the past decade, Sweden has faced an unprecedented inflow of asylum seekers especially unaccompanied children increasing dramatically from 388 in 2004 to 35 369 in 2015, resulting in the largest number per capita in the EU (Narbutaité Aflaki & Freise 2021). This effected law changes in 2014 granting the Swedish Migration Agency the right to allocate unaccompanied children to any local municipality for implementing the newcomer reception policy, yet disregarding the lack of bilateral agreements (Justitiedepartementet, 2012). Gothenburg, the second largest city, was now under pressure to act as a major reception municipality. The immenseness of the challenge and uncertainty regarding the best ways to address the newcomer needs opened a window for innovative solutions to complement municipal reception services with prompt delivery of housing facilities by voluntary and for-profit actors. Local CSOs saw here a possibility to offer housing services complemented with effective care and integration services. For this purpose late in 2014, they initiated a row of dialogues to explore the city municipality’s alternative collaboration models with multiple CSOs. Finally, IOP was chosen as the most attractive alternative.

Unique in the Swedish reception services a joint five-year IOP agreement was signed in May 2015 between initially seven, later nine, CSOs and one municipal unit, Gothenburg City Social Resource Department, hereafter SRD (IOP Göteborg, 2015) All CSO partners entrusted with running housing facilities for the children were financed on non-competitive basis. Its initiation was facilitated by the political board, local collaboration compact (Göteborgs stad, 2014) and by established trust between SRD and several of the CSOs. The partnership covered two different types of services and steering relationships: (1) Publicly financed residential children care homes with basic service and care, and (2) complementary and innovative need-tailored integration services, such as training for staff at residential care homes, long-term children support from volunteering mentor families, crisis counselling, organized leisure time activities and on-the-job internships. The housing services were run by six partner CSOs (Bräcke Diakoni, Skyddsvärnet, Reningsborg, Goteborgs Räddningsmission, Karriärkraft, Göteborgs Stadsmissionen) under the supervision of the municipal SRD that remained accountable for the service quality to the National Health and Social Care Inspectorate. Complementary integration services and training were offered by a mixture of CSOs, including those traditionally associated with advocacy roles (The Swedish Red Cross, Save the Children, Individuell Människohjälp). The IOP services were to be tested for a bracket of (initially 100 but later ca 270) ca 1 500 unaccompanied minors² allocated by the central government to Gothenburg city.

In contrast to commissioned services and state grants the IOP agreement provided possibilities to help uphold a more comprehensive reception services system that was experimental and flexible. To achieve its social goals the partners aimed at exploring a qualitatively new, collaborative governance model and an innovative financing tool (IOP Göteborg, 2015; Anonymous 2019). A major novelty in its relational goals was granting both sectors influence and responsibility in decision-making to enhance sensitivity to user needs, and advance innovative solutions by pooling multiple resources.

A top municipal civil servant underlined the IOP novelty and political courage: ‘We are in a process of building a [partnership] model that does not exist anywhere else.’ [in Sweden] (Socialresursförvaltning, 2015). While its comprehensive approach and collaborative relations appealed to CSOs, the inherent power asymmetries stemming from differences in sector roles and resources presented challenges for joined institutional design and management.

Findings

In what follows, the article presents the study findings on the Gothenburg IOP and assesses internal success from a collaborative governance perspective focusing on strategies for balancing both sectors’ powers and interests even under dramatically shifting circumstances.

Collaborative Style of Governance

The study results indicate that collective decision-making was practised from the initial informal dialogues to the later established formal platforms with representatives from the involved partners, ranging from strategic issues in the Partnership Steering Board to tactical and operational issues in the Collaboration Group and Working Group for Housing Services as well as in an ad hoc group for renegotiating partnership operational conditions. Because of the prevalent trust, openness and collaborative spirit at the early partnership stages, CSOs were allowed significant influence on partnership aims, partner contributions and some overarching partnership institutional design arrangements. This included joined formulation of the guiding partnership values and operational principles stipulated in the partnership agreement, such as respect for CSO uniqueness, transparency, consensus-based decision making and voluntariy withdrawal.

Additionally, CSOs were allowed an important role in ‘jointly defining the partnership social goals’ (SRD) and partner contributions including the format and contents of offered integration services, their operational conditions and duration. This included making the IOP agreement an unusually long-term, five-year a time, commitment (with possibilities to renegotiate its conditions after three years). In the context of later policy reforms, this saved the partnership from premature dissolving. Such influence positively affected partnership relations and CSO’s sense of safety (Klijn & Edelenbos, 2012).

CSOs were also tempted to join the partnership for unique possibilities to, at least partially, influence the partnership financial tool, in this case, a novel model for redistribution of compensation for residential care home services towards value-adding integration services. This model became one of the major factors that raised IOP attractiveness as it enabled service contributions from multiple CSOs including those that otherwise fear dependency on state financing to hamper their advocacy role.

It is the first now that we felt that we could do something worthwhile for as many of the unaccompanied as possible together with others and that is why we joined in. (Voice CSO)

The bulk of the financing for the housing services provided by initially six, later seven, partner CSOs came from state funding via the National Migration Board. The state compensated municipalities based on a lump sum per child and day. In 2015–2016, the national regulations allowed compensation for vacant places to sustain service provider readiness to receive more municipally allocated children.

In the Gothenburg case, a ‘novel financial model’ (SRD) was designed for pooling public and CSO resources for what partners called ‘added value services’. Firstly, the CSOs managed to jointly negotiate a higher municipal payment for their residential care home services – an extra 50 SEK per child and day compared to other direct award contracted housing providers. Secondly, this ‘surplus’ was used to co-finance supplementary integration services for unaccompanied children in the partnership responsibility. The remaining part was covered by CSO budgets or with a voluntary workforce. This model was formalized in individual agreements with each partner and the SRD. As the municipal representative put it:

In the model we designed, we had the idea to make value-adding efforts together and discussed the part of the funding that should be jointly pooled. The fact that it then amounted to exactly SEK 50 [per child and day] happened to be what the CSOs had proposed. The municipality apparently thought it was reasonable.’ (SRD representative 3)

Initially, high numbers of asylum-seeking children generated resources for up to six such supplementary services provided by individual CSOs. This financial model was, however, vulnerable both to a decrease in the number of children in partnership care and shifts in government asylum policies regulating the compensation tariffs for housing services.

What is more, CSO representatives conveyed that through their participation in both strategic and operative partnership structures, they had a say on numerous aspects of partnership management in its various phases. They jointly managed the pooled resources to support any partner capable of developing supplementary integration services as long as this was in line with partnership goals and jointly agreed on partnership platforms. Most importantly, CSO partners felt they could impact the implementation conditions and, to some extent, the contents of the centrally regulated housing services.

There were no specific goals or specifications [by municipality] of services to be delivered, rather we worked out this agreement together based on the youngsters’ needs. (CSO)

The services were shaped by sharing partner experiences, strategies and methods, and by jointly learning from some openly exposed service effects.

Together we can guarantee a certain quality level in this agreement. I feel confident that we are on the same line. (CSO)

The CSOs also used a special strategy in collaborative decision-making which is organizing pre-meetings to coordinate their views and proposals before facing the municipal partner in the Partnership Board. Such a concerted voice helped to equilibrate their power towards the municipal SRD power advantage (augmented by its political authority and mandated monitoring role) and increased CSOs’ weight in shared leadership. By taking over the management of some partnership decision processes the CSOs diminished the need for municipal mediation (Klijn & Edelenbos, 2012; Sørensen & Törfing, 2009). Disputes could arise on what services to prioritize in the face of the increasing complexity of children’s asylum status and needs. Later, when the national government introduced austerity measures in asylum policies CSOs also influenced the composition of the IOP partnership re-negotiation group.

Yet, not unexpectedly, the municipality did make some attempts to demonstrate its authoritative power, particularly in the housing services. This included asking CSOs to receive children without granted asylum status because of the significant delays in asylum processing and municipal lack of housing. Although this added service demands beyond the initial partnership agreement, the CSOs agreed to it because of the value they saw in this partnership format and the collaborative ‘partnership spirit’. In other cases, such as when the SRD used its authority to invite two new CSOs in the partnership to better handle the peak of the newcomer children inflow but without consultation with the Partnership Board, it faced resistance. The partner CSOs counteracted such municipal authority attempts in the partnership institutional design by demanding consensus for any further IOP expansion and restricting the use of the pooled resources to the extant integration services.

In contrast to housing services, to counterbalance the heavy CSOs’ influence in the integration services, the municipal partner was allowed a say on operative issues through its representation in the Collaboration Group. Also here, the municipal partner occasionally used its authority to influence partnership management, such as in prioritizing between municipal and CSO care, home staff, access to the popular partnership training in ‘Trauma aware care offered by a CSO’. The CSOs initially accepted municipal priorities (due to their insufficient preparedness to partake in the training) but finally took back their power by deciding to prioritize the partner CSO staff.

Collaborative governance was sustained although not without challenges, during the crucial partnership adjustment and downscaling period. In 2016, the second year of the partnership, followed by a decreased newcomer inflow, a sudden shift in national policies significantly lowered compensation tariffs (per child and day) and quality requirements for housing and care of unaccompanied children while sustaining municipal responsibilities for reception readiness (SFS, 2017:193). The partnership was confronted with budget cuts and a tough challenge to readjust its housing volumes and integration service ambitions. This put to test the shared leadership and adjustment management. The SRD started scrutinizing its financial capacity and responsibilities to adhere to the principle of dialogue and consensus in the partnership agreement. Unlike in the procured services, the municipality could not simply unilaterally cancel the IOP agreement (Anonymous, 2019). The first joint crisis management strategy was to set up a joint partnership re-negotiation group balancing municipal against CSO interests. Although the municipal partner drafted the adjustment proposals, CSOs retained significant influence on their contents in the negotiations.

We cannot sustain the [IOP] agreement under very much lower tariffs (CSO).

But we want to find a mutual solution (SRD).

Sustaining the dialogue is crucial now (CSO).

The CSO representatives did push for sticking to the original agreement in the first three years, but finally conceded to lowered compensation tariffs, sharing with the municipality some costs for vacancies, and providing fewer service-demanding housing formats. The partnership terms of operation were altered based on the joint agreement about sharing the adjustment costs.

The CSOs had rejected several initial proposals before they were offered a better deal because of their incurred long-term investment costs for facilities and staff. The agreement allowed CSOs some time to downscale, adjust and later withdraw (some of) their partnership services. The municipality too became a victim of government policy shift and yet a winner as its self-incurred cover-up costs were lower than if the original IOP agreement had been sustained during the three initial years.

In sum, in contrast to the procured services partnership with the municipality also called ‘big brother’, when acting in concert, allowed the CSOs to exert significant influence in shaping the partnership institutions and process management both during and after the covenants were signed. This has enhanced collaborative partnership attractiveness and legitimacy among the CSOs.

Adjusting Partnership Steering Mode

A balanced partnership even when operating in strictly steered service provision requires steering it in ways that are better adjusted to CSO and user needs. This implies steering that is distinct from traditional rule-based steering in public administration or result-based steering in outsourced services.

The quality of housing services to newcomer unaccompanied children was monitored by The National Health and Social Care Inspectorate while municipalities retained operational responsibility for their own services and services under direct award contracts and IOPs. The SRD made attempts to adjust the steering of its CSO partners in service delivery based on two major principles: trust and facilitative leadership (Ansell & Gash, 2008).

Firstly, the SRD applied a softer, more trust-based steering than in services provided under direct award contracts or public procurement. The Gothenburg IOP was characterized by less formalization, transparent information, trust in CSO capabilities and dialogue-based quality monitoring. During the 2015 immigration wave and its aftermath, partners encountered increasingly complex unaccompanied children’s needs and related service challenges, also fuelled by shifts in national asylum policies. Under such ambiguous circumstances, the SRD, as the ultimately accountable party, showed considerable flexibility for CSO distinctiveness by allowing CSOs some time to adjust to service requirements.

Here we are used to act based on user needs and not routines or regulations. (CSO Housing Director 1)

Sometimes I perceive that the SRD treats us as one of their own [housings]. You get an insight into how municipal units organize services but we are allowed to be more flexible, and have more possibilities to adjust the services. (CSO Housing Director 2)

I do not experience that the municipality is steering us so much. Sure, the Act of Social Service applies, but there is space to operate a little bit our own way. (CSO Housing Director 3)

The national service norms were communicated and complementary local norms defined in dialogue with CSOs, in contrast to CSOs’ experiences of services under direct award contracts or procured by the same municipality.

We have communicated these [requirements] within the partnership, but here it is kind of another way of working. (…) We do not send them [CSOs] the documents, really, even if the same conditions [as in procured/direct award] apply here too. We communicate these requirements in a different way, via the partnership agreement but also through an ongoing dialogue. (SRD, 1A).

Attempts to sustain dialogue on reception capacities and special needs also characterized the CSO interaction with the municipal unit responsible for children’s allocation to suitable housing forms. As CSO housing directors stated, ‘We do not have that much influence but nevertheless we feel their support’ and that they most of the time ‘were having a good dialogue’ (CSO Housing directors 1,2,3).

Consequently, the partnership accountability practices differed from those in contractual relations. The municipality indicated high trust in CSO commitment and the quality of services without, at least initially, implementing regular, formal inspections. The SRD realized that formal social policy requirements misfit unaccompanied children as a new user group and unrealistic to live up to, given the time needed for premise and service adjustments and competence development. Municipal representatives conveyed that regular dialogues and knowledge exchange in the partnership platforms provided them with a good, ‘perhaps even better insight into the [CSOs] services as compared to monitored procured service providers’ (SRD representative 1B).

Secondly, as the responsible authority and more experienced social service provider the municipal SRD undertook facilitative leadership by offering IOP partners support with advice or common solutions without neglecting their distinctive care methods.

Because we carry the responsibility for both municipal and partnership housing service provision we also provide CSOs some support and try to offer [housing] staff training to secure that all our [housing] personnel have the same knowledge level […]. But because we collaborate in an IOP we have a different discussion there, it is about what [trainings] we have and what CSOs can offer additionally (SRD representative 3).

The CSOs too took their share in partnership leadership, including actions to identify additional staff and children’s needs and organizing training, for example, in preventive crisis counselling for children and crisis-management training for housing staff.

The SRD also offered its partners help with quality standards in housing services, such as staff competence, security procedures or children allowances, in accordance with municipal practices. The SRD approach was that of shared responsibility stating that ‘here we are creating services together’ (SRD representative 2). The CSOs welcomed some standardization advice, especially during the initial partnership stages when pressure for services was great and time for testing alternatives was short.

Yet, sometimes CSOs had to defend their distinctiveness and autonomy by resisting what was seen as an illegitimate municipal exercise of authority in attempts to converge the municipal and CSO operative routines in housing and care services. This includes a procedure for the opening of new group housing to accommodate the concerns of the neighbourhoods; compensation schemes for children’s free time and sports expenses and extending housing staff responsibilities towards the children without additional payment, such as accompanying them to school or hospitals. The CSOs response was ‘This is not acceptable’ or, ‘if we do this we need additional resources’ (CSO Housing Director 4).

The municipal partner, however, retained some of its authoritative power in steering such as when urging CSOs to provide reception and integration services to a much more divergent group than originally anticipated, including children without residence permits and in need of psychosocial care. Contrary, in integration services where the partnership gained from CSO entrepreneurial spirit and ‘thinking out of the box’ the CSOs succeeded in preserving their distinctiveness.

The steering of housing services delivery was under continuous adjustment, balancing between municipal and CSOs interests, roles and power. Despite its challenges, especially financial ones, such relational governance in combination with supportive municipal leadership appealed to the CSOs more than other collaboration formats.

CSOs’ Abilities to Sustain Their Independent Voice

Voicing critique may be a challenge in a partnership with public authorities that aims for dialogue and consensus. One of the advantages of an IOP partnership was the CSOs’ possibility to influence public service design and implementation by retaining their advocacy on behalf of targeted social groups. Here, we focus on the major strategies in the Gothenburg partnership that CSOs used to safeguard their independence and voice.

One strategy to safeguard the CSO’s independence and advocacy role was to avoid providing housing services in which the government could circumscribe their independence.

If you provide housing you become very much an agent. (Interview, CSO)

The advocacy CSOs stressed that they ‘like to work differently’ such as by providing the IOP partners with evidence-based knowledge about the targeted children’s needs and the impact of services or lack thereof. As the partnership progressed, this included providing supplementary need-tailored integration and care services.

We do not start any support activities for children that would not lead to increased knowledge and possibilities to make an impact. (Interview, CSO)

The partnership governance model allowed CSOs to stick to their human-centric and out-of-the-box approach in tailoring services. For example, CSOs unanimously resisted cuts in children allowances as an effect of new state policies in 2017 (SFS, 2017:193) by pointing to their freedoms stated in the partnership agreement in realising partnership goals:

We should not do exactly as the municipality, but instead, see what best fits our [children] aims. Active children feel better.’ (CSO, Collaboration Meeting Notes).

Yet, apart from service tailoring, CSOs did not take so many opportunities to exercise their critical voice within the partnership structures. There were a few major disagreements during the partnership initiation and implementation period and partners were ‘careful not to step on each other’s toes’ (CSO, Collaboration meeting notes) in a trust-building process. Later in 2016, when unfavourable to the partnership goals, changes in national asylum policy forced adaptation of the partnership agreement regarding service scope, financial conditions and CSOs’ contributions (see further in Narbutaité-Aflaki, et al., 2017) the consensus was temporarily disrupted. The CSOs fought for their view of service quality strongly critiquing early adjustments suggested by SRD before the partners settled for a jointly renegotiated agreement early in 2017.

During the partnership adaptation and termination phases, CSOs started openly critiquing the fragmented municipal participation in the partnership when the actions of the municipal department responsible for children’s placement outside the partnership started colliding with the IOP agreement. This included rapidly sending away unaccompanied youngsters aged 18 (and not 21 as was municipal practice) from partnership care to state housings outside the municipality and depriving them of partnership services before they finished their obligatory secondary education which violated the original IOP agreement. Yet, being partners with a municipal SRD unit CSOs felt obstructed in expressing formal protests against another municipal unit, and instead relied on at that stage rather unsuccessful dialogues.

At the same time, the common partnership values and activism among its service staff strengthened all CSOs as ‘opinion influencers’ in a social and political environment increasingly hostile towards asylum seekers. Frustrated by the changes in asylum policies and their effects on the unaccompanied children, some partner organizations chose to critique the government in national newspapers (Göransson, 2016), established political meeting platforms (Almedalen), via social media or direct appeals to State Migration Agency and the municipal board. Additionally, partnership relations enabled CSOs to join forces in a collective critique of government through membership in a social protest action ‘#Vistårinteut’ [we cannot stand it], for which the partnership seemed too narrow as an arena.

The municipal ability to accommodate CSO influence in various partnership phases without significantly curbing their singularity reassured CSOs of the partnership value and prominence over contractual relations based on market logic.

Conclusions and Discussion

This article responds to the call in research for the need to explore whether and how collaborative state-third sector governance with better equilibrium between the interests, roles and powers of CSOs and public agencies may be achieved also in contexts with prevailing contracting out. It provides some answers to these questions and the state’s ability to tap into CSO potential in addressing complex social needs within the remit of an IOP model.

The study results evidence that despite some limitations when operating in a strictly regulated policy field, the Gothenburg partnership scored rather well on all three major balanced collaborative governance clues or criteria (Table 1) from its initiation to renegotiation and adjustment: (i) collaborative style of governance, (ii) steering arrangements adjusted to CSO’s distinctiveness, needs and partnership aims and (iii) CSO’s abilities to exercise independent voice.

The attractiveness and success of the IOP model rested on allowing CSOs a significant influence not only in co-managing partnership commitments and its daily operations but also in co-governing its institutional design, including the financial model, complementary integration services and quality readjustments in housing services forced by unfavourable national policies.

Furthermore, the adopted governance model made it possible to counterbalance the lesser CSO discretion in regulated reception services by granting them significant decision-making freedom in the complementary integration services. The partnership was rather successful in accommodating civil society’s distinctiveness and independence as advocated by the national collaboration compacts by adjusting municipal control and accountability practices. Trust-based steering (Montin, 2016), including monitoring through dialogues and facilitative leadership (Ansell & Gash, 2008), proved to be helpful municipal strategies without neglecting municipal precedence in monitoring the regulated aspects of reception policy.

A notable partnership consensus on local reception policy progress did not significantly curb the CSOs’ advocacy role. Instead, due to their role in complementary services CSOs gained more confidence and power in self-critically examining partnership services as a whole and in voicing critique externally of the national government. They also attempted internal critique towards the dissonant organization of the public partner and lack of comprehensiveness, even if partnering with one municipal unit delimited their full voice potential.

What is more, insights from the Gothenburg case suggest that sustaining power-balanced collaborative governance may require a trade-off of sectors’ powers. The voice and influence of individual CSOs may be enhanced by joint advocacy in partnership structures and by allowing them some distinctiveness in state-regulated service provision. The municipal partner power is retained instead by partaking in partnership design and governance, influencing the terms of trust-based steering of CSO service delivery and offering facilitative leadership. Importantly, achieving and sustaining balanced collaborative governance requires embedding partnerships in favourable policy contexts, such as soft-institutional norms (here the compacts), local political approval and mutually beneficial financial models. Lack of clarity on IOP legal status and ‘rules of the game’ as well as abruptly shifting political and financial support mechanisms may corrupt collaborative governance efforts.

To conclude, the Gothenburg case indicates that a more power-balanced collaborative governance model that uses a range of CSO resources may be possible within the remits of local IOP agreements. IOPs may offer a contrast to misbalanced hierarchic, control-based or ‘the winner takes it all’ (Ansell & Gash, 2008, p. 544) relations. In congruence with collaborative and relational governance ideas, IOPs may offer shared policy ownership and responsibilities as well as more equal, deliberative and consensus-oriented relations (ibid.) as commonly agreed and nurtured principles (Vangen et al., 2015).

The study findings also confirm the relevance of the three explored clues or criteria for more balanced and relational governance in partnerships (Table 1). The empirical findings add to the adopted model the principles of counterbalancing state power by such strategies as concerted CSO action and greater influence in non-regulated service areas; trust-based steering including dialogue strategies; facilitative leadership including guidance and training, as well as allowing CSOs to exercise their critical voice within the partnership governance and management, especially by joined efforts and some autonomy in drafting their service models, to guide power equilibration in collaborative governance and to enhance CSOs commitment to partnership with public agencies.

Moreover, the Gothenburg case arguably enriches the prevalent (Ansell & Torfing, 2021; Ansell & Gash, 2008) conceptualization of collaborative governance by illustrating a more complex version within the remit of nationally regulated and locally delivered public services. It showcases that innovative collaborative governance initiatives may come not only from public agencies but also from engaged CSOs and result in self-selective multi-actor partnerships with complementary services. The management of partnership challenges may be addressed by shared cross-sectoral leadership exercised through partnership structures. Local states may retain elements of control in governing partnership design and service delivery, yet they need to counterbalance its power by including elements of trust, decision delegation and support.

Arguably, IOPs, may under favourable circumstances offer more relational and collaborative ways of interacting between value-based CSOs and public agencies in service co-production, co-management and co-governance than market- or grant-based models in the Swedish collaboration context and possibly beyond. Although integrating elements from historical corporatist arrangements (Wijkström, 2012), IOP goes beyond cross-sectorial collaboration in policy formulation networks. The IOP offers a distinct relational model that includes joined consensus-based governance of the partnership and co-management of its outputs (here new service models), including trust-based steering and supportive leadership, adjusting financial and resource-sharing tools to co-create more value for the targeted groups. This implies renegotiating the state-third sector relationships again, especially in contexts, such as Sweden, where lately CSOs have mainly been cast in a subordinate service agent role by market logic (Bode & Brandsen, 2014) or operate under the financial state grants model but are rather disconnected from the state in drafting innovative service designs and their implementation.

The IOP spread in Sweden indicates a counter-trend to NPM pervasiveness in large parts of Europe where at least some IOPs may serve as ‘spaces of hope’. Because of their novel aspects and varieties in Sweden, there is a further need to explore the evolving IOP governance models using both the suggested and alternative frameworks and a wider variety of research tools.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by The European Commission Horizon 2020 project Innovative Social Investment – Strengthening Communities in Europe: [Grant Number 649189]; Swedish Innovation Agency (VINNOVA): [Grant Number 2016-03494].

Notes

1 Only 29% of civil society financing in 1990s came from the public sector (compared to 45% in Great Britain) and up to 60% from membership fees.

2 Göteborg stad årsredovisning 20f15. https://cutt.ly/Wdj9Zuw.

3 Highlights in the table indicate the study findings that complement the selected analysis framework.