State Regulation of Diaconal Actors in the Realm of Health and Social Services: Comparing Denmark, Norway and Sweden

Abstract

In Denmark, Norway, and Sweden, diaconal actors have been pioneers in social and health work. By applying complexity theory on managing unpredictability, the article reviews recent pioneering efforts: caring for undocumented migrants—whose legal permissions have expired. Pioneering efforts apply to services for substance abusers. With Norwegian examples, unpredictability is demonstrated for institutional and drop-in services. The second part of the article analyses different strategies for facilitating diaconal actors: the Danish so-called “in-house” system, and Norwegian and Swedish public procurement legislation that allows reservation for non-profit actors. Within public procurement, defining certain services as Services of General Economic Interest (SGEI) is an option, underutilized by Norway and Sweden. Finally, a draft directive on European cross-border associations (ECBA), including religious associations, is explained. By reviewing policy documents, legislation, court rulings, and adopted and draft EU directives and regulations, the article’s overall conclusions are: the services to undocumented migrants have not affected the overall relationships between diaconal actors and political authorities, and starting such services is not a radical shift for diaconal actors. Moreover, Denmark provides the greatest predictability for non-profit actors.

1. Introduction

In secular Scandinavia (World Values Survey 2023), the share of overall service provision by faith-based actors in the realm of health and social welfare is relatively modest, with the highest in Denmark and Norway. Norway has Parliament-approved targets on the share of health and care services provided by non-profit actors: 10 percent overall; 40 percent target for child welfare; and the largest non-profit actors are diaconal (Haugen 2024, p. 106). In Sweden, for-profit actors provide on average 20 percent of health, education and social services, and higher for child welfare and for addiction treatment (NOU 2024:17 2024, pp. 119, 278). See

Table 1. Share of services by non-profit actors (NOU 2024:17 2024, p. 119 [child welfare], p. 278 [addiction treatment], p. 308 [nursing homes]). *

	Denmark	Norway	Sweden
child welfare	46	22	3
addiction treatment	58	52	9
nursing homes	11	5	3

* For figures on schools and kindergartens, see (NOU 2024:17 2024, pp. 182, 149).

for share of services provided by non-profit actors. Von Schnurbein and Hengevoss (2020, p. 31) provides shares of all non-profit organizations involved in social services (highest in Norway); similar figures are provided by the European Commission (2022, Denmark’s figures include health) and Statistics Norway (2020, see “Satellite account for voluntary and non-profit organisations”).

Within these three services, the pioneers were diaconal actors—churches and diaconal organizations—with “diakonia” being Greek for “service”. Catholic nuns in Norway established in the years 1882–1949 23 nursing homes (Hadland 2007). During these years, the Norwegian Constitution prohibited Catholic monk orders until 1897, and Jesuits until 1956.

In Sweden, the 1862 Municipal Act stripped the Church of Sweden (CoS) of its poor relief and social care mandate. The 1929, an attempt by the Bishops’ Conference in CoS to initiate a reform program was voted down in the Parliament, implying a shift towards “defensive preservation” (Thidevall 2020; CoS 2011). According to Wollmann (2018, p. 416), since 1930, “non-public non-profit (third-sector) organisations were practically excluded from rendering personal social services.”

Scandinavians overall—between 73 and 80 percent—agree with the statement “Churches and other religious organizations play an important role in helping the poor and needy” (Pew Research Center 2018, p. 19; see also Angell 2014). The high scores among Swedes are interesting. Rambøll—quoted in Vikdahl et al. (2024, p. 194)—estimates CoS’s diaconal work to constitute SEK 1.5 billion annually. Swedes have the lowest church attendance (Inglehart 2021). Denmark, on the other hand, having the highest church attendance in Scandinavia, has the lowest share acknowledging churches’ diaconal work.

Health policy experts in Sweden and Norway include churches when analyzing health systems (Janlöv et al. 2023, p. 5; Sperre Saunes et al. 2020, pp. 4, 19). The information on Norway is not correct, however, when asserting: “…Diaconal hospital foundation in Oslo … is owned by the Norwegian Church.” What is correct is that the two diaconal hospitals—both established in the 19th century—are limited liability companies, one owned 100% by one foundation, and one owned 50% each by two other foundations. Health policy experts in Denmark refer only to “non-profit” (Birk et al. 2024, pp. 53, 96, 98).

Recent pioneering work benefits “undocumented migrants”; which is the preferred term in this article, not “illegal migrants” (Buer Haddeland 2020), “irregular migrants” (UN Committee on Economic, Social and Cultural Rights 2019, paras 62 and 63; 2020, paras 38, 39 and 29) or “irregular situation” (UN Committee on Economic, Social and Cultural Rights 2024, para 16; Hermansson et al. 2022). A Swedish proposal on public employees’ duty to report undocumented migrants was criticized (UN Committee on Economic, Social and Cultural Rights 2024, para 17; for the proposal, see SOU 2024:80 2024; see also SOU 2024:68 2024). Undocumented migrants’ legal permissions to stay have expired and the relevant political authorities know about them.

Vulnerabilities of and service provisions to these persons are studied in Denmark (Castaner et al. 2022; Funge et al. 2022), Norway (Rabe and Buer Haddeland 2023; Eick et al. 2022), and Sweden (O’Sullivan 2024; Mona et al. 2021). A meta study by Winters et al. (2018) not covering Norway or Sweden finds that Denmark provides minimum rights for undocumented migrants; acute health care is provided (Danish Ministry of Health and Development 2014).

The research questions are: Comparing legislation in Denmark, Norway and Sweden, how does the state facilitate the diaconal actors, is legislation impeding the efforts of these actors, and how are the diaconal efforts acknowledged?

Section 2 presents methods, working hypotheses and the broader context, whereas Section 3 introduces complexity theory. Section 4 analyzes tensions within six policy realms: Section 4.1: legal frameworks for assisting undocumented migrants; Section 4.2: diaconal actors’—and others’—care for undocumented migrants in the three states; Section 4.3: public procurement provisions, illustrated by an awkward (and cancelled) Norwegian tender process; Section 4.4: Norway’s sudden introduction of competition-based schemes for social and health services, illustrated by a drop-in service for substance abusers; Section 4.5: Norway’s and Sweden’s reserved tender processes to non-profit actors; and Section 4.6: A proposed EU directive for non-profit cross-border associations. Section 5 discusses the findings, whereas Section 6 concludes. As seen, few examples are Danish, reflecting the fact that Danish laws and policies are less contested.

2. Methods, Working Hypotheses and the Broader Context

Legal provisions and some recent court rulings in Denmark, Norway and Sweden, as well as relevant policy documents, constitute the main data material. Proposed EU directives are also included. The article is not based on interviews, but some specific information has been obtained by contacting relevant persons in Denmark and Sweden.

As regards undocumented migrants, national legal provisions will be reviewed, in addition to binding and non-binding sources from the EU, Council of Europe and United Nations’ human rights bodies. Then, outlining diaconal and Red Cross service providers—focusing on their services and what they are permitted and not permitted to do—I demonstrate one Danish example of public sanctioning.

As regards public procurement, the relevant provisions of EU’s public procurement directive (EU 2014) are reviewed, showing how implementations differ in the Scandinavian states. Reviewing unpredictable processes in Norway, the article shows—inspired by a public commission report (NOU 2024:17 2024)—that there are alternatives to contracting out service provisions for a given time period. Here, I map Sweden’s experiences regarding a register on non-profit actors.

Finally, as regards the proposed non-profit cross-border associations (ECBA), relevant EU sources are outlined. As this is an ongoing process, it is not analyzed in detail; see European Parliament (2026) for an overview and updates.

The three Scandinavian states are suitable for a most-similar-cases design, whose overall purpose is to identify and explain differences. Finland was excluded due to language barriers. Scandinavia has similar welfare state traditions. However, many scholars and policy makers—also within Scandinavia—are not familiar with the different policy solutions found in Denmark, on the one hand, and Norway and Sweden, on the other hand—and the differences between these two. Previous studies using user choice and active citizenship as variables compare service provisions by public, non-profit and for-profit actors in these states—for schools, kindergartens and nursing homes (Sivesind and Saglie 2017). Enjolras et al. (2018) and Enjolras and Sivesind (2009) have a broader European outlook. As assistance to undocumented migrants is the main focus of the article, I will review the most relevant studies analyzing their situation. As undocumented migrants are “hidden”, there is hardly any data collected by public institutions to gain insight about their situation (Gauffin et al. 2021).

There are no public procurement or public grants systems for assisting undocumented migrants. Hence, the examples in Section 4.3 and Section 4.4 describe the treatment of substance abusers. The article does not seek to review comparative scholarly work on the situation of substance abusers.

There are three working hypotheses for the article: (i) public procurement processes result in unpredictability; (ii) diaconal actors are positive to public monitoring; and (iii) the forms of service provision studied in this article—for undocumented migrants and substance abusers—have an underutilized potential for users’ self-organization and empowerment.

Four themes that are part of the broader context of non-profit actors’ service provision, will not be analyzed in the article. First, pension compensation schemes, which have been subject to much controversy in Norway over the last decade, as public pension requirements are costly for non-public actors. The Norwegian Ministry of Health and Care Services (2025, p. 64) has rejected to establish a compensations scheme for previous contracts on county or municipal levels. The Norwegian Parliament (2025, proposal no. 25) has requested a compensation scheme for non-public kindergartens—but a subsequent regulation states that to qualify, a pension funding gap of 5% over three years must be documented. Only one outcome has brought calm: a compensation scheme for contracts with state authorities adopted in 2019.

Second, recent policies benefiting for-profit actors, specifically care guarantees for and user choice in health treatment, with Denmark being the most generous by its “extended” model of a 30 days care guarantee as specified in Section 87 of the 2024 Health Act. Denmark also allows patients to choose treatment providers in regions other than one’s region of residence. Norway is making more use of for-profit health providers (NTB 2024; Norwegian Minister for Health and Care Services et al. 2024; Norwegian Directorate of Health 2023, p. 51); one result is enormous overpayments, resulting in fraud charges against Spesialistlege AS (Norwegian National Authority for Investigation and Prosecution of Economic and Environmental Crime 2025). A waiting guarantee system for Sweden will be proposed by a public commission in 2026; Nergårdh (2025) has published an initial study.

Section 5 will introduce a Swedish court case filed by a for-profit actor after the county of Stockholm reserved its user choice options in geriatrics for non-profit actors (“LOV”; Act (2008:962); for non-profit actors’ experiences, see Jutterström and Segnestam Larsson 2020; and Segnestam Larsson and Åberg 2025). On user choice, Norwegians can still choose between public and private (for-profit or non-profit) actors with an agreement with the regional health authorities (“fritt behandlingssted”), but other options were removed in 2022 (Norwegian Government 2022a, 2022b; for experiences, see Kaarbøe 2021, p. 22; Norwegian Government 2022a, p. 21; NOU 2024:17 2024, p. 283).

Third, the article discusses neither bureaucratic requirements for care providers, nor the direct links between care providers and organized crime actors in Sweden (Mogert and Strömqvist 2025; SOU 2023:34 2023, pp. 120–21, 302; NOU 2024:17 2024, pp. 100, 323, 354, 369; Brå 2016). A proposal in the Swedish Parliament (Kallifatides et al. 2025) to replace LOV with an Act on aims-based right of establishment (“målstyrd etableringsrätt”) was together with three similar proposals not approved by the Swedish Parliament’s Standing Committee on Health and Welfare (2026, pp. 4, 37).

Fourth, the article does not analyze schooling for children without legal permission. For these, Denmark and Norway only specify education up to 10th grade and do not apply the term “rights”; Norwegian Ministry of Justice and Public Security (2010) discusses these terms; see also Swedish National Agency for Education (2019). The Swedish Education Act (2010:800) specifies a right to primary education in Ch 7 Section 2, third part; a similar provision for high-school education—found in Ch 29 Section 3—is to be repealed from 1 July 2026.

3. Complexity Theory and Non-Profit Actors

Seeking to explain how best to adapt to unpredictability, the early work of Simon (1962) developed complexity theory applied to organization theory and strategy. Contracting non-profit actors for health and social services—and disagreements on criteria for being non-profit—are subject to heated debates and relatively frequent shifts in the overall national policy frameworks in Norway and Sweden. Hence, a theory to capture such unpredictability is relevant. It will be shown that Denmark has adopted a different approach, in compliance with EU requirements.

Complexity theory—in its “European tradition”—aims at “explanation of the forces behind complexity” (Thietart and Forgues 2011, p. 58; acknowledging Mc Kelvey). Complexity theory is applied to various policy fields, or systems that operate on different levels, whose logic can be captured by the term “emergent self-organization” (Plowman et al. 2007a, p. 342). Rather than identifying one distinct complexity theory, its core is best explained by identifying its principles, that in addition to self-organization are specified as simple rules on engagement, feedback and evaluation for evolution, connectivity and interdependence, and emergence (Zimmermann 2011, pp. 622–23; acknowledging Best and colleagues). Hence, acknowledging change resulting from various factors, complexity theory can act as a guide when studying overall policy challenges related to services by diaconal actors (Lowell 2016; Allen et al. 2011; Plowman et al. 2007b; Levy 2000).

4. Six Policy Realms

4.1. Service Provision to Undocumented Migrants, Primarily Health

Denmark’s Health Act of 2024 regulates health services to persons without permanent residence, as in Sections 80 (acute treatment) and 158b (vaccines). Moreover, the Danish Immigration Act of 2024, Section 42a, states that necessary health services can be provided for free if they are residing at a known address and they—or their parents or spouse—are unable to pay. The term “undocumented migrants” or similar terms are not applied.

There are two Norwegian regulations. First, the regulation on the right to health and care services for undocumented migrants (residents without legal permission), as revised in 2024—based on the 1999 Patient and User’s Rights Act—permitting in Section 5 “very necessary assistance …”, including birth assistance and abortion. For birth assistance, payment is specified in Section 5.3 of the Specialist Health Care Act. However, hospitals are not vigilant in following up in non-payment for those giving birth.

Second, the regulation on social services for persons without permanent residence as revised in 2014—based on the 2009 Act on Social Services in the Norwegian Labour and Welfare Administration. Its Section 4 specifies short-term emergency assistance to residents without legal permission to stay.

Sweden’s Act on health care for residents without legal permission was revised in 2016 and 2019; its most inclusive provisions now apply only to children.

All three states have penalties of up to two or three years’ imprisonment for assisting a person without legal permission, with three different approaches. Denmark’s Aliens Act prohibits any housing and transportation (Section 59 stk 8–6), and if the provider achieves personal gain, this is an aggravating factor. Norway’s Immigration Act specifies gain only in the case of organized illegal activities facilitating entry (Section 108, part 5; see also part 4(a) [housing] and part 3(a) [labor]; the Norwegian Supreme Court (2021) finds a violation only for the latter). Sweden’s Aliens Act—Ch 20 Section 7—is the only one that specifies that to be considered punishable, “the act has been committed for financial gain…”; a wording that is similar to Article 1.1(b) of the 2002 Facilitation Directive (EU 2002).

All have humanitarian exceptions: Section 83 of the Danish Penal Act; Section 108 part 6 of Norwegian Immigration Act—“humanitarian assistance”; same wording as in EU 2002, Article 1.2—and in Sweden, the term “mitigating circumstances” is found in the same provision as the one quoted above.

The European Commission (2023a, see also Council of the European Union 2024) proposes a new facilitation directive; with a purpose “not to criminalise … humanitarian assistance …” (European Commission 2023a). Article 3 on criminal offences specifies the following acts to be criminalized in the context of assisting non-nationals “to enter, or transit across, or stay … where the person … requests, receives or accepts, directly or indirectly, a financial or material benefit” (Council of the European Union 2024, p. 27; author’s italics; the term “instigates” was deleted). Ministers also specified (Council of the European Union 2024, p. 8): “…Member States are free to criminalise … when no financial or another material benefit has been provided.” The proposal has been criticized by Three UN Human Rights Special Rapporteurs (2025).

Regarding undocumented migrants, the Council of Europe’s Committee of Ministers (2012) adopted a Reply to a Recommendation by the Council of Europe’s Parliamentary Assembly (2011 [38-0-1; Malta abstaining]), requesting examination of “access to education …” and “minimum health-care requirements …” (Council of Europe’s Parliamentary Assembly 2011, para 10(1) and 10(2)). The Reply emphasized “the right to subsidiary social protection, including adequate medical assistance” (Council of Europe’s Committee of Ministers 2012, para 6; para 3 refers to the “guiding principles of the United Nations Convention on the Rights of the Child [UN CRC] …”).

The European Social Committee found (7-6) that a French Act specifying “state medical assistance …” had “numerous difficulties in the implementation …” (European Committee of Social Rights 2004, para 34). This represented a violation of the revised European Social Charter Article 17 (children’s protection), specifying that Article 17 is “directly inspired by the [UN CRC]” (European Committee of Social Rights 2004, para 36).

Moreover, jurisprudence from the European Convention on Human Rights (ECHR) finds that the term “other status” in Article 14 ECHR and Article 1(1) of Protocol No. 12 on non-discrimination encompasses “immigration status” (Council of Europe/European Court of Human Rights 2024, p. 48; cases are from 2009 to 2012—with one exception [no violation]). Violation of Article 14 was found in three cases. The last decade has seen stricter policies in Scandinavian countries, notably Sweden (Hansson et al. 2023), and the rest of Europe.

The Return Directive (EU 2008) specifies minimum rights in Article 14: “emergency health care” for illness and “access to … education” for minors; on education, see also Article 17(3).

Finally, the UN Committee on the Rights of the Child (2005, para 56; para 18) specifies that “…unaccompanied and separated children have the same access to health care as children who are […] nationals”; see also UN Human Rights Committee (2005, para 3) and UN Committee on the Elimination of Racial Discrimination (2004, para 36). The UN Committee on Economic, Social and Cultural Rights (2009, para 36) asserts that “children … with an undocumented status, have a right to receive education and access to adequate food and affordable health care.”

In this context, it is relevant that children of undocumented migrants have significantly higher perinatal mortality rates—28 of 1000—than children of other women—5 or 6 of 1000 (Eick et al. 2024). Perinatal deaths occur between the 22nd pregnancy week and first week after birth.

4.2. Health Care and Social Services from Diaconal Actors and Others for Undocumented Migrants

All states have penalty provisions and humanitarian exceptions provisions, permitting certain activities by diaconal actors. Activities by non-religious Red Cross are also outlined.

Denmark’s inadequate provision to undocumented migrants (Lægeforeningen et al. 2010)—vaccines and urgent treatment is provided only to those having permanent residence—made the Danish Red Cross initiate a Health Clinic for Migrants in 2011, now operating in four cities (Danish Red Cross 2025), and Caritas Denmark runs a health clinic and a mobile clinic in Copenhagen (DanChurch Social 2025).

In 2017, DanChurchSocial (Kirkens Korshær) was ordered by the Danish Authority of Social Services and Housing to pay back DKK 170,000, as Kirkens Korshær could not guarantee that there were no undocumented migrants sleeping in the emergency shelter in Hellig Kors Church in Copenhagen in the period November 2016—March 2017. Hence, the prohibition against assisting undocumented migrants had been violated.

Turning to Norway, the inadequate treatment in the public health sector led in 2009 to—with the three largest political parties protesting—the establishment of the Health Center for Undocumented Migrants in Oslo, by the Church City Mission and Norwegian Red Cross. A similar Health Center was established in Bergen, Norway’s second largest city, in 2013 (Norwegian Directorate of Health 2022, 2019 [Bergen not listed]). Decisions to facilitate access are made by Bergen City Council (2023; Berge 2023 [incl. decisions by Vestland County Council]; Meisfjord 2024; on proposed cost sharing). Volunteers were medical professionals; surgeries were done at diaconal hospitals, without payment. The Health Center in Oslo is since 2017 funded by local authorities (Braathen and Dommerud 2017). Trondheim municipality, Norway’s third largest city, provides health services, housing and room in nursing homes to undocumented migrants (Adresseavisen 2023).

A 2023 regulation—enacted in accordance with the 2018 Local Government Act—regulates the operation of emergency shelters, but only in Oslo. The actors for more than a decade are the Church City Mission, Salvation Army and Red Cross. In two other cities, the Church City Mission provides shelter. The Norwegian Ministry of Labour and Social Inclusion (2025, p. 184) proposed that no financial support will be provided from 2027; this and other cut proposals for 2026 were revised in the Parliament (Norwegian Parliament’s Standing Committee on Finance 2025, p. 16; see also p. 46 [securing 2026 funding to Amathea; was explicitly Christian/diaconal 1978–2003; named Alternativ til abort]).

Swedish Red Cross operates three health clinics for undocumented migrants, two in Stockholm and one in Gothenburg (Swedish Red Cross 2025). Additionally, a CoS congregation in Western Sweden turns its offices into a health consultation office twice a week. The Church City Mission runs Crossroads, a shelter specifying that it is only for EU citizens, in Malmö and Stockholm.

A Swedish study based on 22 interviews—seven with NGO employees—in Gothenburg, Malmö and Stockholm regarding service provision for EU/EEA citizens shows a complex picture (Ekendahl et al. 2020). The two main findings are summarized as “desperate measures for desperate times” and “the limits of universalism”. Even if the number of EU/EEA citizens is higher in Sweden than in Denmark and Norway, the same overall patterns apply to numerous EU/EEA citizens who come to Scandinavia to earn a living—for themselves and for their families.

Summing up, DanChurchSocial has been fined for its alleged non-control of shelter users in 2017–2018. Overall, however, providing for undocumented migrants has not affected Scandinavian diaconal actors’ abilities to enter into agreements with public authorities.

We now widen the focus to the ways governmental actors can facilitate or impede services and activities by diaconal actors.

4.3. Exceptions Within the EU’s Public Procurement Rules—And Unpredictable Processes

The Swedish IOP (non-profit/public partnership) is a cooperation somewhere between annual grants and a procurement (SOU 2019:56 2019, pp. 277–329; SOU 2019:56 Annex 2019; see also Swedish Association of Local Authorities and Regions 2025 and NOU 2024:1 2024, p. 59). The Swedish Government (2026) has initiated a process on an IOP Act. Swedish National Agency for Public Procurement (n.d.) specifies for shelters for battered women that in situations where IOP is the preferred framework for cooperation, the two IOP parties must make analyses before entering into IOP on whether procurement and state support rules apply (Swedish National Agency for Public Procurement n.d., see “Upphandling och IOP”). In principle, Swedish authorities consider such shelters to be an economic service (Swedish National Agency for Public Procurement n.d.; see “Upphandling av tjänsten skyddat boende”).

Four specific exceptions of Directive 2014/24/EU (EU 2014) will be highlighted: (i) a general exception; (ii) an exception in accordance with Article 74–77 (light touch regime for social and other specific services); (iii) a reserved contract under Article 20; and (iv) public contracts between entities within the public sector under Article 12.

First, recital 6 specifies that “Services of General Economic Interest [SGEI] should not fall within the scope of this Directive.” This phrase is not defined and is applied only in a note (EU 2014, Annex XIV; Haugen 2018 [EU sources on SGEI]).

Second, reviewing Articles 74–77, Article 75 is on information requirements, Article 76(1) is on transparency, and Article 76(2) list 17 principles on characteristics of the service provider, the service and the user. Article 77 on reserved contracts for certain services is less attractive, as it specifies contract “duration … [no] longer than three years.”

Third, Article 20—part of “General rules”—specifies: “Member States may reserve the right to participate in public procurement procedures to sheltered workshops and economic operators whose main aim is the social and professional integration of disabled or disadvantaged persons…”, as specified in recital 36.

Fourth, Article 12 on public contracts between entities within the public sector—part of “Exclusions”—specifies three conditions on control exercised by a contracting authority for a legal person to be considered as being part of the public sector. This system is termed “in-house” (DK: “selvejende”; authorities exercising “udvidet egenregi”) and implemented by Section 12 of the Danish Public Procurement Act (LOV nr 1564 af 15/12/2015; see NOU 2024:1 2024, p. 49). Lack of independence and predictability is the outcome for the service providers.

Overall, notwithstanding the demanding requirements that the diaconal actors must fulfill in public contracts, a positive view is expressed by Arvidson and Linde (2021, p. 207) on the closer scrutiny resulting from public contracts: “control and autonomy are each other’s companions rather than antagonists.”

However, the problem with public procurement is the unpredictability, illustrated by a Norwegian procurement process worth at least 11 billion NOK (1 billion Euro) that resulted in complaints, cancellations and three court rulings (Bertheussen Isachsen 2025 [after the first judgement that ruled against the cancellation]). The outcome was a new procurement process and extension of current contracts to June 2027 (South-Eastern Norway Regional Health Authority 2026 [SEN-RHA]). SEN-RHA did not appeal the first two judgments; the third judgment ruled in favor of SEN-RHA.

In another procurement process, the Norwegian Parliament’s Standing Committee on Health and Care Services (2026, Labour party and Conservative party voting against) cancelled a procurement process for rehabilitation in SEN-RHA and one other RHA. Two reasons were given. First, the calls set prices to be weighed much more than quality (70 v. 30 percent). Second, the calls implied a shift from inpatient to outpatient rehabilitation—without the Parliament being consulted in advance (Norwegian Parliament’s Standing Committee on Health and Care Services 2026, pp. 5–6). The Norwegian Minister for Health and Care Services (2026) emphasized that the annual costs for the cancellation would exceed 426 million NOK (38 million Euro).

As unpredictable was the sudden introduction of a competition-based scheme, as will now be shown.

4.4. Norway’s Sudden Introduction of Competition-Based Scheme for Social and Health Services

The Norwegian Government announced with the state budget for 2023—launched in October 2022, without any prior notification—competition-based schemes for several services, many of which were initiated after governmental requests. The new schemes were justified by restrictions on state aid (Norwegian Ministry of Finance 2022, p. 99; Article 61(1) of the European Economic Area (EEA) Agreement; similar to Article 107(1) of the Treaty on the Functioning of the European Union (TFEU)—that prohibits “any aid … which distorts or threatens to distort competition.”).

There are two alternatives to EEA Article 61(1). First, applying the exception provision specifying what is “compatible with the functioning of this Agreement”, including “aid having a social character, granted to individual consumers, provided that such aid is granted without discrimination related to the origin of the products concerned” (Article 61(2)(a)/Article 107(2)(a)).

Second, these services—as they do not affect EEA trade; cf. Article 59(2)/Article 106(2)—could have been defined as SGEI in accordance with Article 59/106. A Norwegian Commission recommends legislation on state support to such SGEI (NOU 2024:17 2024, p. 365), clarifying when financing undertakings entrusted with the operation of SGEI is compatible with Norway’s EEA obligations. The basis is the principle of solidarity and the so-called Altmark criteria (Court of Justice of the European Union 2003: (i) clear obligations; (ii) fair compensation; (iii) reasonable profit; and (iv) costs not exceeding what is expected by a well-run undertaking).

Moreover, the Norwegian financial management rules specify as one of the exceptions to competition-based calls: “individual grants where the grant recipient is named in the allocation decision or referred to in the [state] budget proposal in such a way that the grant must be considered earmarked for this recipient” (Norwegian Ministry of Finance 2022, pp. 109–10, Section 6.4.2; see also Norwegian Agency for Public and Financial Management 2022, p. 6).

Grants—without competition—is by the Norwegian Agency for Public and Financial Management (2024) termed “an exception”, and by the Norwegian Ministry of Health and Care Services (2022, p. 28) as “appropriate in some cases …”.

There are two reviews of the system of annual grants (Norwegian Agency for Public and Financial Management 2022; National Audit Office of Norway 2022).

Other grant receivers were informed in 2023 with the state budget for 2024 about new competition-based schemes. After applying, one service by the Church City Mission for substance abusers, 24sju, was informed in March 2024 that it would not receive funds for 2024, as it fell outside the relevant competition-based scheme. A new scheme—with its application date in August—was decided in the revised state budget, including “costs incurred in 2024…” (Norwegian Parliament’s Standing Committee on Finance 2024, p. 4) The Church City Mission opened a new service titled The Street Clinic (“Gateklinikken”), funded through this new competition-based scheme (NRK 2025a; Norwegian Directorate of Health 2025). Hence, efforts that involved the Labour Party in the Norwegian Parliament in 2024 made the Norwegian Government—led by the Labour Party—initiate this new scheme.

Changes in systems for public funding—from grants to contracts—must be expected (Neumayr and Pennerstorfer 2020, 441), but the unpredictability caused by the sudden—and late—introduction of the Norwegian competition-based scheme is unwarranted.

A totally different approach—resulting in a multi-year contract without a tender process—was taken in 2026 for the national air ambulance service (Norwegian Air Ambulance Helicopter 2026). NAAC is 100% owned by a membership-based non-profit foundation and has a similar responsibility for air ambulances in Denmark.

4.5. Laws Facilitating Non-Profit Actors: Reservation Provisions

Denmark stands out with its generous system for tax deductions on donations to non-profit actors (LBK nr 806 af 08/08/2019 § 12, stk 2–7), representing 15% percent of one’s income. Moreover, the in-house system—the fourth exception listed above (EU 2014, Article 12)—might be seen as a form of facilitation for non-profit actors, but it will not be discussed further.

Norway’s regulation on public procurement (FOR-2016-08-12-974) has a reservation provision: Section 30-2(a). It has three parts, part 1 reading (extract): “to achieve social goals, the common good and budgetary efficiency”, and part 2 allows non-profit actors to (extract):

…reinvest any surplus in activities that fulfill the organization’s social purpose. A non-profit organization can, to a limited extent, run commercial activities that support the organization’s social purpose.

This provision—or rather a more comprehensive proposed definition (NOU 2024:1 2024)—is proposed to be a part of a new public procurement act (NOU 2023:26 2023, p. 49).

Norwegian Supreme Court (2025) found unanimously that the Oslo municipality was right in reserving a call for tenders on the construction and operation of nursing homes with up to 800 beds to non-profit actors. This reservation was challenged by two for-profit service providers. Borgarting Appeals Court (2024, dissenting judge; 2-1), drawing upon the Court of Justice of the European Union (2023), ruled in favor of the for-profit actors, as it found that the justification for the reserved call was not adequately based on the wording of Section 30-2(a), rather being “political ideological”. The Appeals Court ruled against Oslo District Court (2023), which emphasized budgetary efficiency, as the three participating diaconal actors specified capacities that in sum resulted in only 342 hospital beds in the specifications made by the non-profit actors. Oslo District Court held that Oslo municipality could have applied Section 30-1(3)—based on Article 76(2) of Directive 2014/24/EU—specifying numerous characteristics of the service and the users.

Section 30-2(a) of the Regulation on public procurement complies with Norway’s EEA obligations (EFTA Court 2023, para 86), as it encompasses economic efficiency, universality and solidarity.

Sweden’s non-profit register (Act 2022:900; see also SOU 2019:56 2019) can identify which non-profit actors can participate in reserved tender competitions. It works in tandem with Act (2016:1145) Ch 19 Section 25a, introduced through Act 2022:902, specifying the non-profit actor’s “public purpose” and “purpose of the service to be provided.”

The relatively few entries into the register—29 by 3 June 2026 (Kammarkollegiet 2026)—is explained by the overall structure of the Swedish non-profit landscape and the costs of being in the register (initially SEK 12,000; then 1600 SEK annually); initially the annual fee was SEK 12,000 (Gustafsson 2023).

The Swedish Act (2022:900) specifies in Section 8 to whom the non-profit actor can transfer funds, namely other non-profit actors and research.

One Swedish court case (Stockholm Court of Appeals 2025; Uppsala Administrative Court 2025) found that the Uppsala municipality acted legally when reserving a tender process for non-profit actors. The four for-profit companies that complained listed seven requirements that they held were derived from the directive (EU 2014; see Stockholm Court of Appeals 2025, pp. 6–7 for the alleged requirements). They argued that, for instance, budgetary efficiency was not met if the overall costs were not lower than in a situation with service provision by the municipality itself (Stockholm Court of Appeals 2025, pp. 8–9; Uppsala Administrative Court 2025, p. 20). The Stockholm Court of Appeals (2025) held that the core was the overall legal framework, which was in place and correctly applied.

Moreover, the companies’ allegation that Section 8 of Act 2022:900 (see above) is EU-incompatible, was not supported. The Supreme Court rejected the appeal (Uppsala Municipality 2025).

Another court case was on “LOV”, introduced in Section 2 above. For LOV to apply, there must be a contract between the care provider or medical doctor and the relevant municipality or region. Ch 1 Section 2a of LOV permits the municipality or region to reserve its “vårdvalssystem” to non-profit actors—specifying one health service where this is not permitted, namely primary medical doctors (“fast läkarkontakt”). Such a reservation is exactly what the Stockholm region decided upon in 2024 when institutional geriatric care was reserved for non-profit actors, and for-profit actors were immediately informed that their contracts would be terminated. This was challenged by a for-profit actor. The same seven requirements as presented in the tender competition court case were referred to in this case; they were held to be derived from the non-discrimination principle practiced by the EU (Stockholm Administrative Court 2025, p. 3); two of the lawyers were involved in both cases. The for-profit actor lost.

In this context it is relevant that the Swedish Agency for Economic and Regional Growth (2012, pp. 43–44) recommended: (i) enhanced capital access for non-profit actors; (ii) all municipalities should apply “LOV” more than the (then) Act of public procurement (“LOU”) to strengthen the role of non-profit actors. A 2026 review by the Swedish Competition Authority of non-profit actors’ impediments in the context of “LOV” and “LOU” is ordered by the Swedish Government (Famna 2026); a total of seven orders were given to public institutions, all on non-profit actors.

In summary, Norway provides the greatest flexibility, by allowing for reservations and by allowing non-profit actors to run commercial activities, but only to a limited extent. A proposal to prevent any surplus, by revising the Act on foundations (Norwegian Ministry of Trade, Industry and Fisheries 2021, p. 37) was not endorsed by numerous actors, including the Norwegian Gambling and Foundation Authority (2021). While awaiting legislative proposals based on NOU 2024:17 2024 and NOU 2024:1 2024, the Norwegian Parliament (2026) made eight decisions to strengthen non-profit actors. Decision 584 requested the RHAs to reserve procurement processes for non-profit actors where appropriate.

Sweden stands out by permitting non-profit actors to distribute their funds to other non-profit actors or to research. All states have penalty provisions and humanitarian exceptions provisions, permitting certain activities.

4.6. Laws Facilitating Non-Profit Actors: EU Initiative on European Cross-Border Associations

The European Commission (2023b) proposes that non-profit associations can apply for registration as European cross-border associations (ECBA).

Before outlining the core of this proposal, it is relevant that none of the Scandinavian states have specific legislation on associations—but all have constitutional protection of freedom of association. Denmark has an Act on social entrepreneurs, whose Section 5 lists five cumulative requirements: (i) social purpose; (ii) for-profit; (iii) independent; (iv) involving and accountable; and (v) reasonable surplus.

Table 2. Number of associations and social enterprises, and constitutional provisions, in the three states. *

	Denmark	Norway	Sweden
associations	100,000	101,000	159,298
social enterprises	600/411	295	300
constitutional provision	78	101	1974:152, Ch 2, Sect. 1(5)

* Sources: European Commission (2022, pp. 50–51); Garvik (2025; quoting Statistics Norway); Sociale Entreprenører Danmark (n.d.); Hulgård and Chodorkof (2019, p. 35); Kobro (2019, p. 28); Eimhjellen and Loga (2016, p. 6); Gawell (2021, p. 222).

provides figures:

Most notably concerning the ECBA for this article is the European Parliament’s decision to remove the EU Commission’s exception stating that the directive does not apply to religious organizations and associations of such entities (European Parliament 2024a, Article 3.1(a); recital 17). The initial vote in the European Parliament for the revised proposal gave a result of 490-69-64 (European Parliament 2024b).

In summing up the hearing to this proposal, to which most were positive, the proposed directive represents “a recognition of the non-profit sector at large” (European Commission 2024, p. 5 [p. 2 in attached document]). This opportunity to provide services in other states will have considerable economic benefits (European Commission 2023b, pp. 10–11).

The discussions in the Council of the European Union started in 2025; one question raised by the EU Legal Service (2025, p. 5) is:

whether the relevant proposal is “aiming to create a company governed by Community law with capacity to operate throughout the Community or whether, on the contrary, it is aiming to enable national companies with a uniform statute to be set up.”

The term “uniform statute” must be read to imply that there are certain principles that must be included in the national legislation in order for companies established in accordance with national law to function as an ECBA. As specified by the European Parliament’s (2026) Legislative Train Schedule: “The proposal has not been included within the joint declaration on legislative priorities for 2026, signed in December 2025.”

5. Discussion

Starting with the three working hypotheses specified in Section 2, I find that the first—public procurement processes results in unpredictability—is strengthened. Norway’s experiences over the last years should be clear enough evidence. The second—diaconal actors being positive to public monitoring—is also strengthened. Arvidson and Linde (2021) and the responses made by diaconal actors in connection with the decision to end the free treatment and user choice (Norwegian Government 2022a, 2022b) all affirm this. The third working hypothesis—the underutilized potential for users’ self-organization and empowerment among undocumented migrants and substance abusers—is strengthened for undocumented migrants (Berge 2023). For substance abusers, studies from outside of Scandinavia affirm that self-organization and empowerment is important (Hunter et al. 2013; Bröring and Schatz 2008).

Public authorities face numerous challenges when deciding on health and care services. The provisions for undocumented migrants are regulated in legislations in all three states, all having in principle humanitarian exceptions for assisting migrants, with the Danish exception found in the Penal Code. As seen, the EU Council also does not take a position on national legislation that criminalizes assistance in the absence of “financial or another material benefit …” (Council of the European Union 2024, p. 8); the criminalization extends to assisting non-nationals to “stay”. As regards providing for undocumented migrants, much responsibility lies with diaconal and other non-profit actors in all three states.

Complexity theory applies to systems that operate on different levels. One peculiar aspect of assistance to undocumented migrants, particularly in Norway, is the distinct solutions found at the sub-national level. First, Oslo municipality co-funds the Health Center in Oslo. Second, Vestland County Council facilitates its own services—dental services and high schools—as well as transportation from outside of Bergen to treatment in Bergen (Berge 2023). Third, Trondheim municipality’s own services extend to undocumented migrants, despite the restrictions in the two Norwegian regulations.

However, unlike Plowman et al. (2007a, 2007b) who studied a radical change of becoming “ministry with the marginalized”, and Lowell’s (2016) emphasis on organizational change, the diaconal actors engaging with and for undocumented migrants and substance abusers have not shifted focus from their previous responsibilities but simply engaged with new groups of people, initiating new services as pioneers. Complexity theory contributes to explaining diverse responses to unpredictable decision-making within given policy fields, resulting in locally based responses.

The five principles listed by Zimmermann (2011, pp. 622–23)—self-organization, simple rules on engagement, feedback and evaluation for evolution, connectivity and interdependence, and emergence—all seem relevant and applicable to explain these differences and the gap between these three cities and the Norwegian regulations. The Vestland County Council’s decision includes establishing a coordination forum with relevant actors to increase overall knowledge. Bergen City Council (2023) asks for feedback from two actors; one being a diaconal actor, and one having a high-profiled former bishop involved. Moreover, the information on the health centers provided by the Norwegian Directorate of Health (2022 [Oslo and Bergen], 2019 [Oslo]) is particularly interesting in light of the restrictions found, particularly in the regulation on the right to health and care services.

Moving to the supra-national level, we have seen that the EU/EEA legal framework is the most important explanation for why numerous Norwegian grant systems were replaced with competition-based calls—without due notice less than three months before the start of the budget year.

In Sweden, the shelter for battered women is understood by the Swedish authorities to be an economic service, and not an SGEI. This article has not engaged the SGEI requirements in detail; as noted by Guy (2025, p. 351), “the SGEI mechanism places a certain degree of responsibility on the Member States in designating particular activities as SGEI—the Commission’s role is essentially to review the national decision …”.

According to Guy (2025, p. 351), Denmark stands out from other EU member states with a “consistent rejection of the SGEI mechanism and indeed the state aid rules as applicable to the Danish healthcare system.” We have seen above that Denmark stands out in numerous ways from the systems applied in Norway and Sweden, with the Danish system being more predictable. Denmark has less political debates on the role of for-profit actors and non-profit actors, and much fewer court cases, typically with cases brought by for-profit actors against the relevant public authorities; court cases are brought by public authorities against for-profit actors regarding inadequate personal data protection. Hence, complexity theory seems less able to describe or explain the Danish system in the realm of health and care services, as Denmark has simply chosen predictability over flexibility. Many non-profit actors have been included in an overall public provision that is recognized by Article 12 of the EU public procurement directive (EU 2014). An implication of this so-called in-house system is that non-profit actors lose some of their independence. NOU 2024:17 (2024, pp. 55–57, 360–62) presents French, German, Italian and Swedish alternatives.

A final important aspect of the EU in the context of diaconal actors is the change in the draft directive on ECBA brought by the European Parliament (2024a, draft article 3.1(a) and draft recital 17), removing its non-application to “religious organisations and associations of such entities.” Its outcome is not given, but if the directive should be adopted, the associations have to operate under diverse legal and bureaucratic systems—even if the possible directive would foster certain forms of harmonization. The distinctions are numerous, and include diverse requirements on professionals’ conduct, and on transparency and other requirements when justifying decisions—and complaining about such decisions—within health and care services. In this context it is relevant that the General Data Protection Regulation (GDPR) sets a high common standard.

An appropriate understanding of the “forces behind complexity” (Thietart and Forgues 2011, p. 58) implies that sub-national differences and often rapid changes must be acknowledged. As one example, when a new City Government took power in Oslo 2023, it reversed the eight-year policy of the previous City government of phasing out for-profit actors in nursing homes. A recently established for-profit company won the tender competition for two nursing homes. When responding to the criticism that it was inexperienced, it specified its owners’ previous role in another for-profit company (NRK 2025b), to which this company protested (Sjursen Skiphamn 2025).

Both Norway and Sweden permit that tender processes for multi-year contracts on health and social services are reserved for non-profit actors—and the attempts of challenging this have been unsuccessful (Norwegian Supreme Court 2025; see also EFTA Court 2023; Stockholm Court of Appeals 2025; Uppsala Municipality 2025). Oslo has one example of a for-profit actor winning a contract for operating a nursing home in a congregation foundation’s facilities, with reactions elsewhere: congregation-owned nursing homes in Bergen have specified that this will not apply to Bergen, as they will not permit a for-profit actor to run a nursing home in their buildings (Thorsen 2024).

Plowman et al. (2007a, p. 342) introduced the term “emergent self-organisation”; the West coast of Norway can be an example of how undocumented migrants work between themselves and with diaconal actors in forms of self-organization (Berge 2023). ETUI (2021) explains self-organization in the context of workers’ rights on national levels. Norway’s Mennesker i Limbo, being explicitly mentioned in the decision by Bergen City Council (2023) has been partly successful on reimbursements of taxes paid by undocumented migrants who previously got paid for work and did not receive social services. More research is needed on the empowering impact of diaconal actors’ efforts (Lo et al. 2025, p. 8; Brandsen et al. 2025, p. 22; Clarke 2000) and on undocumented migrants’ self-organization and claims-making.

Finally, I agree with Levy (2000, p. 84) that rational actor models or behavioral approaches are not sufficient to explain the organizational responses discussed in this article. Rather, as indicated by Levy (2000, p. 83): “Complexity does show significant promise in furthering our understanding of processes by which organizations adapt to their environment.” Diaconal actors, who have been pioneers, have adapted to new needs that they have seen and documented (Seilskjær 2023). The fact that most diaconal actors have a solid financial foundation, are taken seriously by political authorities, and have abilities to recruit volunteers, cannot be underestimated. These findings regarding societal impacts for those experiencing ostracization—documenting innovative services to meet some of their needs—supplements the findings on overall impact of the third sector identified by Enjolras et al. (2018) and Enjolras and Sivesind (2009), and the findings on active citizenship by Sivesind and Saglie (2017). This article stands out from these studies by analyzing legal provisions and overall policy frameworks, and strategies applied primarily by diaconal actors.

6. Conclusions

Returning to the research topic on states’ facilitating, impeding and acknowledging diaconal actors, Denmark is particularly characterized by two different tendencies. As regards the assistance to undocumented migrants, the article has found that Denmark and Norway are stricter than Sweden, as the relevant provision specifies that assistance is punishable only if “the act has been committed for financial gain.” Norway stands out from Denmark by its humanitarian exception within the relevant provision (Norwegian Supreme Court 2021). The overall finding on assistance to undocumented migrants is that assistance to them are supplements to many other services by diaconal actors.

As regards the overall facilitation of diaconal actors in the three states, Denmark’s in-house system stands out. Overall, Denmark provides the strongest predictability for diaconal actors. The complexity theory has been found to be more relevant for the Norwegian and Swedish contexts, as compared with the Danish context. Less predictability is found in Norway, as seen by the awkward 2024–2025 procurement process and the 2022–2025 replacement of grant schemes with competition-based schemes, without prior information.

Are there examples of national authorities impeding diaconal actors? Taking Norway as an example, the Norwegian Parliament has reversed proposals or instructed the government to do so, for instance in the context of medical assistance to substance abusers (Norwegian Parliament’s Standing Committee on Finance 2024, p. 4; Norwegian Directorate of Health 2025; NRK 2025a). Hence, it is not possible to state that Norwegian authorities are actively impeding diaconal actors. Moreover, the fining of DanChurchSocial (Gotfredsen 2017; Busk Stie 2017) cannot be said to constitute an impediment. As regards Sweden, it is too early to predict the outcome of the Swedish authorities’ explicit invitation to churches to assist in facilitating (voluntary) the return of non-nationals (Janzon 2025). Most likely, there will not be any sanctions for churches’ less than enthusiastic response to the invitation.

As regards acknowledgement of diaconal actors, the European Parliament (2024a, Article 3.1(a)) stands out by including “religious organisations and associations of such entities” in the proposed ECBA directive, unlike the European Commission (2023b). The outcome of the process to establish a legal framework for ECBA is not given (European Parliament 2026); two different options can be identified (EU Legal Service 2025).

The debate in the coming development of this proposal—and its implementation when adopted—is an obvious candidate for further research. I highlight two other themes for further research: First, the shockingly high perinatal mortality rate as reported for Norway by Eick et al. (2024, 2022) implies that similar research should be conducted in other countries. Second, comparative studies on the legal framework for engaging non-profit actors—beyond Scandinavia—should be undertaken. An initial hypothesis is that other EU member states provide more predictability than Norway and Sweden—without the lack of independence that is implied in Denmark’s in-house model.