Missed opportunities? State licencing on the Swedish residential care market

ABSTRACT

In many countries, residential care is subject to outsourcing and consequently, states are increasingly engaged in controlling activities. A central instrument is licencing; a procedure determining membership to markets. By analysing all formal applications submitted to the Swedish national Inspectorate between the years 2013–2016, the aim of the article is to describe the characteristics and influx of applicants as well as analysing how the inspectorate, conceptualised as a market agency, contributes to shape the residential care market. The findings show that applicants often are not forced to revise their care in conjunction with the licencing procedure, and slightly more than half of the applicants are granted a licence. Large companies are in general more successful than small companies, as are those presenting specialised target groups. The primary focus in licencing appears to be target group descriptions, management and ensuring the absence of methods with repressive elements. Licencing does not, or only to a limited extent, address issues such as schooling and health support, staffing levels, children’s contacts with the birth family, the scientific base of methods and measurement of client outcomes. The discussion focuses on how licencing contributes to the market formation as well as care aspects omitted in the controls.

ABSTRAKT

I många länder utspelar sig institutionsvård för barn på en marknad och i spåren av det ägnar sig offentliga myndigheter alltmer åt kontroll och styrning. Ett centralt instrument är tillståndsprövning, vilket är den samhällsinstitution som beslutar om vilka företag som ges tillträde till marknaden. Syftet med föreliggande studie är att – baserat på analyser av samtliga tillståndsansökningar inskickade till den svenska tillsynsmyndigheten mellan 2013–2016 –beskriva kännetecken för dem som ansöker och beviljas tillstånd samt analysera hur tillsynsmyndigheten genom prövningen bidrar till att forma barnavårdsmarknaden. Studien visar att drygt hälften av samtliga ansökningar beviljas tillstånd. Stora företag är generellt sett mer framgångsrika än små företag, liksom är fallet med dem som presenterar mer avgränsade målgrupper. Prövningens primära fokus riktas mot målgruppsbeskrivningar, ledningens kompetens samt att inga arbetsmetoder förekommer som riskerar kränka barns fri- och rättigheter. Prövningen riktar inte alls, eller mycket begränsat, in sig på vårdaspekter som skol- och hälsostöd, bemanning och personalens kompetens, vetenskaplig säkring av metoder samt uppföljning av klienter. Diskussionen behandlar de sätt på vilka tillståndsprövningen bidrar till att barnavårdsmarknaden formas och de vårdaspekter som försummas i granskningen.

KEYWORDS:

• Residential care for children
• licencing
• privatisation
• quality

NYCKELORD:

• Institutionsvård för barn
• Tillståndsprövning
• Privatisering
• Kvalitet

Introduction

In recent decades, outsourcing has affected the provision and steering of social services in many countries (Gingrich, 2011). With regard to residential care for children – a far-reaching intervention targeting vulnerable children exposed to maltreatment and/or with behavioural problems – the role of private providers has increased in many nations (see, e.g. Nordstoga & Stokken, 2011; Stanley et al., 2013; Toikko, 2017). This is also the case in Sweden, where the residential care field has gone from being almost exclusively dominated by the public sector in the 1980s, to becoming a market with a significant proportion of private providers (Meagher, Lundström, Sallnäs, & Wiklund, 2016).

As a result of marketisation, the public sector has increased its engagement in various monitoring activities in order to control and manage markets (Levy, 2006). In Sweden, calls for control have also become more urgent due to e.g. high profit-levels and uncertainty regarding the outcomes of care (Pålsson, 2018). The control activities targeting residential care include for example municipal procurements and follow-ups, but also state regulation and inspections (Pålsson, 2018). One crucial activity is licencing, which is the first public control meeting new private care providers wishing to enter the market to sell their residential care services to local authorities. In Sweden, as well as in many other countries, having a licence is mandatory for all residential care companies, and the application process involves presenting the planned services for an independent national inspectorate (in Sweden The Health and Social Care Inspectorate, IVO). The Swedish licencing process aims at ensuring ‘good quality and safety’ (https://www.ivo.se/om-ivo/) among the care companies that establish on the market.

The current article is a part of a larger research project exploring how the Swedish residential care market is formed (Lundström, Sallnäs, & Shanks, 2018; Shanks, Lundström, Meagher, Sallnäs, & Wiklund, forthcoming). While the other studies investigate the development of the residential care market and the motivations and strategies of private providers, our interest in this study concerns the role of the state in the market formation and specifically the interplay between private companies applying for licence and state activities directed towards them. One of the authors has previously published a study (Pålsson, 2017) investigating the process of licencing of residential care. This article extends the previous study as it uses new and more comprehensive data to explore the features of care providers seeking access to the market as well as explicitly analysing the role of the state in the market organisation. In the article, we employ the concept market agency (Andersson, Erlandsson, & Sundström, 2017) to analyse how the IVO through the licencing procedure affects the market and the residential service. We solely focus on treatment-oriented residential care units (RCUs), i.e. units with the explicit aim to provide assessment and/or treatment in order to rehabilitate children and youth. By analysing all formal applications concerning treatment-oriented RCUs submitted to the IVO between the years 2013–2016, the aim of the article is to describe the characteristics and influx of applicants as well as analysing how the IVO as a market agency contributes to shape the residential care market. The following research questions are posed:

1. How can the applicants be characterised in terms of ownership, and how do they describe their target groups and professional care content?

2. Why are applicants rejected and what characterises RCUs that gain licence?

Internationally there has been an increased focus on licencing and public standard-setting as one important way to improve residential group care (see, e.g. Boel-Studt & Tobia, 2016; Whittaker et al., 2016). However, apart from a few important exceptions (e.g. Pålsson, 2017), empirical studies exploring licencing of residential care are scarce. This article contributes with valuable knowledge concerning the range and impact of the state’s influence on residential care markets through its licencing procedure.

Controlling the residential care market

In Sweden, as in many other countries, residential care is an institutionalised service in the out-of-home care system for children (Sallnäs, 2009). The service is predominantly used by teenagers (Thoburn, 2007) with various needs and problems (e.g. substance abuse, behaviour problems, experiences of abuse), but there is also a share of units enrolling toddlers together with their parents. The market for residential care has been described as turbulent and heterogeneous. It consists of a wide range of facilities with varying ownership structures, sizes, target groups and staff qualifications (Lundström et al., 2018; Meagher et al., 2016). As in many other countries, the general educational level among staff has been depicted as low and there is a plethora of methods and programmes used (IVO, 2013; Sallnäs, 2009). Also, in Sweden as well as elsewhere, relatively little is known about the outcomes of the care and few methods have been evaluated and proved effective (James, 2011; SBU, 2016). Nevertheless, there is cumulative knowledge showing that e.g. educational achievements (Forsman, Brännström, Vinnerljung, & Hjern, 2016), health support (Vinnerljung & Hjern, 2018), care stability and leaving care transition (Kääriälä & Hiilamo, 2017), as well as working with the family of origin, forming a positive social climate among adolescents and providing trauma-informed care (Whittaker, del Valle, Holmes, & Gilligan, 2015) are all vital aspects in order to make residential care effective.

In recent years, the public control of Swedish residential care has been reformed and a reinforced audit policy put in place (Hämberg, 2017; Pålsson, 2018). The increase in external control has parallels in many other western countries (see, e.g. Hood, Nilsson, & Habibi, 2019; Nordstoga & Stokken, 2011; Power, 1999). This development can be understood against the backdrop of privatisation creating a need for coordination, but also as a response to public inquiries exposing evidence of serious maltreatment in, and a deficit of control of, out-of-home care in a historical perspective (Sköld, 2013). In Sweden, the IVO as central inspectorate constitutes the main instrument for this control. The agency issues licences and conducts annual inspections of each residential unit, and is formally entitled to issue sanctions to those residential units that fail to comply with the regulatory standards (by, for example, depriving a company of its licence). The activities aim at improving safety and care quality, but current knowledge indicates that licencing and inspections have difficulties affecting core care aspects, such as care-giver-care-taker-relations, children’s preferences and quality aspects identified in research (Pålsson, 2017, 2018).

Theoretically, this study is influenced by perspectives proposing that markets – contrary to the traditional view of them as emerging spontaneously – can be understood as substantially organised and formed by their institutional context (Ahrne, Aspers, & Brunsson, 2015; Brunsson & Jutterström, 2018). In this respect, the state plays an important role in formulating market policies. These policies (i.e. approaches and strategies towards markets) are increasingly effectuated by what has been conceptualised as market agencies (Andersson et al., 2017), i.e. public authorities that to various degrees intervene on the market and impact the services and products that are sold. Market agencies differ in focus and purpose; some are established to enhance the effectiveness of the markets, whereas others are more involved in protecting other types of values, such as people’s health and well-being. In organising markets, the state (or other organisers) can use one or several organisational elements. Relevant organisational elements pertinent to licencing includes deciding membership (by approving or rejecting entrance to market) and rule-setting (working out standards that providers must follow) (c.f. Ahrne et al., 2015; Brunsson & Jutterström, 2018). Through deciding on who can become a member of the market the supply-side is affected, and by setting rules the agencies attempt to impact the product or service (in this case residential care). In addition to these organisational elements, market agencies often provide information to sellers and buyers in order to affect their behaviour (Andersson et al., 2017). For example, the IVO on their website provides information regarding current rules, expectations, etc.¹ In the article, we use this theoretical view of markets as organised to analytically reason about the range and impact of licencing on the residential care market.

The licencing process

To gain access to the residential care market, all companies must apply for a licence for each individual unit that they intend to open. The licencing procedure is based on completed application forms and filed appendices. The process does not involve any appointments in person between the IVO and the applicant. In addition to brief information about the owner and the premises, the applicant must have a manager enrolled and describe the target group, care content, process of admission, how the children’s educational needs will be monitored, predicted staffing, etc. The companies’ descriptions are in one sense structured and pre-defined (aligned with the regulations on the area), yet there is room for reasonably lengthy accounts of the services that the company aims to provide. The descriptions are prospective, entirely based on ideas regarding services that do not yet exist (see Pålsson, 2017).

Licencing is based on controlling applicants’ compliance with regulations. These are specified primarily in the Social Services Act (SSA 2001:453), the Social Services Ordinance (SSO 2001:937) and regulations promulgated by the National Board of Health and Welfare (NBHW). The SSA declares that a licence can be granted if the care assures ‘good quality and safety’ (SSA 2001:453). In the SSO and the rules and advice by NBHW, there are specifications on enrolment procedures, premises, collaborations between with social services and home environment of the child, monitoring of health and schooling of children and that methods should be based on best available research and respect personal integrity (SSO 2001: 937; NBHW 2016:55). There are also rules on manager competence (degree in social work or an equivalent education, and work experience from relevant areas) and staffing (NBHW 2016:55; SFS 2007:171), quality management systems (NBHW 2011:9), management of irregularities (NBHW 2011:5) and documentation (NBHW 2014:5). However, in many aspects, the regulations are rather general, meaning that it becomes the IVO’s task to concretise what compliance means.

Material and methods

The study is based on analyses of formal licence applications and decisions regarding treatment-oriented RCUs. In order to collect data the authors, during spring 2018, contacted the national unit at the IVO that administers licencing. The application documents are open to the public and all applications processed by the authority during the years 2013–2016 were requested. 2013 was selected as the starting year since the responsibility for licencing was relocated to the IVO during that year. 2017 was not part of the material, since, at the time of data collection, all cases had not yet been processed.

An employee of the IVO transferred the information to file and submitted it to the authors. The material consisted of 335 new applications. Among these, 53 had either been cancelled by the applicant before decision or had been written off by the IVO due to applicants’ failure to complete the application or submitting necessary appendices. Hence, the material analysed amounted to 282 licence applications, coming from 256 unique RCUs (during the study period, we identified 30 units that reapplied due to previous rejection/s). For each application we obtained the following documents: (a) the licence application and associated appendices wherein the applicant describes the premises, manager, target group, care content, etc., (b) the formal licence decision comprising that the IVO either approves or rejects the application and (c) in case of approval, the licence of the company (i.e. a description of what the company obtained licence for).

Both authors studied the applications and transferred relevant information to a statistics programme (SPSS). The data analysed in this article was information on (a) ownership and size of company,² (b) age and gender of children, (c) target group description (data were inductively categorised by the authors into ‘psychosocial and behaviour problems’, ‘substance abuse/criminality’, ‘toddlers and parents’ and ‘protected residence’), (d) care content (methods, therapeutic approaches, etc.), (d) education of proposed manager and (e) rejection/approval of application and, if rejection, the authorities’ motives for this.

The material was analysed quantitatively and qualitatively focusing both on characteristics of the RCUs applying for a licence (in terms of ownership, target group, care content and manager qualifications), and on the nature of the standards that the state set up at licencing. Quantitative bivariate analysis and multivariate logistic regressions were performed to assess possible correlations between characteristics of applications and approvals/rejections.

The rejection decisions were analysed qualitatively and illustrative excerpts of the IVOs’ reasoning were selected in order to shed light on how licencing contributes to shaping the market. Inferences were drawn abductively (Alvesson & Sköldberg, 2009) by applying the aforementioned theoretical concepts on the extracted data. We also paid analytical attention to care aspects not addressed in the licencing procedure. As the data consists of documents, there are some caveats that need to be borne in mind (cf. Bowen, 2009). For example, there may be information missing, and the relation between how applicants describe their service at the time of licencing and the actual residential practice to come is of course not straightforward. Furthermore, data consisted of rather lengthy descriptions (e.g. regarding target groups, methods, etc.) which had to be reduced into categories in order to be analysed quantitatively. It is, of course, difficult to rule out that another way of categorising data would have yielded somewhat other results. Nevertheless, the extensive material enabled a comprehensive picture of important traits of the RCUs that have entered the residential care market in recent years. Moreover, it gives knowledge regarding the actual role, operations and impact of the state in managing and steering the market through licencing.

Results

In this section, the findings are presented. The presentation begins with a description of how RCUs aspiring to enter the market present themselves and their service. This is followed by an analysis of the formal reasoning of the IVO when refusing RCUs access to the market, and an exploration of the characteristics of applicants finally gaining a licence.

RCUs aspiring to enter the market – ownership and care characteristics

During the years of 2013–2016, the IVO made decisions regarding 282 licence applications submitted by 256 RCUs. A general impression when studying the content of these applications is that they in many ways resemble each other. Apart from institutional pressures to conform with normative values on the organisational field (Brunsson & Jutterström, 2018), this is likely to be due to that in applying for a licence, all applicants are faced with a pre-structured applications form (concerning set questions related to e.g. target group, care content, admission and discharge, quality systems, how the children’s educational needs will be monitored, staffing, etc.). In addition, on their website the IVO provides information regarding current expectations on residential care, by e.g. explicating the applicable legislation, as well as instructions on how to complete the application form. Hence, the given frames confirm the applicants’ descriptions of their offerings and the information provided by the IVO can be viewed as a way to shape the service in advance (cf. Andersson et al., 2017).

The majority of applications came from small companies (which in this case means companies owning fewer than three RCUs) (Table 1). Around one in six applications (n = 47, not in table) came from one of the five multinational companies that during the last decade have significantly increased their market shares in Sweden (Lundström et al., 2018). There were also a small number of applications from not-for-profit organisations. In Sweden, the small share of applications from these types of organisations is not surprising as they, in contrast to in many other countries, are marginal providers of residential care and have been so for several decades (Lundström et al., 2018).

Table 1. Characteristics of RCUs applying for licence 2013–2016.

		Proportion (%)
Organisation (n = 282)	Large company (n = 94)	33
	Small company (n = 180)	64
	Not-for-profit organisations (n = 8)	3
Manager’s education (n = 245)	Social work/psychologist (n = 166)	68
	Other degree in social sciences (n = 59)	24
	Nursing degree (n = 11)	5
	Other (n = 9)	4
Target group, sex (n = 281)	Girls (n = 45)	16
	Boys (n = 103)	37
	Mixed (n = 133)	47
Target group, age (n = 280)	Children, 12 years and younger (n = 58)	21
	Children, 13 years and older (n = 222)	79
Target group, services (n = 280)	Various psychosocial and behaviour problems (n = 228)	81
	Specialised towards substance abuse/criminality/ protected residence (n = 33)	6
	Toddlers and parents (n = 19)	7
Methods/approaches	Cognitive/behavioral oriented	75
	MI (Motivational Interviewing)	52
	Less standardised approaches	56
	Milieu therapeutic	55
	Family-based	26
	Pychodynamically oriented	17
	Substance abuse/criminality oriented	13
	Attachment theory-based	8
	Trauma-informed	4
	Other	45

Almost all companies (except a small number of facilities for toddlers and their parents) declare that they target youth from 13 years and over, which is aligned with the predominance of teenagers in Swedish residential care (Thoburn, 2007). Further, it is common among organisations to target both girls and boys; around half of the units declare that they accept both sexes. The descriptions of the target groups’ difficulties are often broad, indicating that many companies apply for approval that covers youth with a diverse mix of needs. However, most facilities targeting substance abuse/criminality, protected residence and families with young children appeared to plan on offering specialised services (Table 1).

Based on the descriptions of care content among RCUs aspiring to enter the market, the traditional eclectic nature of Swedish residential care is notable (cf. Sallnäs, 2009). Approximately 100 different treatment methods/approaches were identified in the 282 applications, and it is common among applicants to intend to utilise several methods and approaches concurrently. Also, it is apparent that the evidence-based movement has impacted the field, as methods commonly referred to as evidence-based are widely mentioned. Unsurprisingly due to their rather hegemonic position in the psychological field, the most common methods and techniques (mentioned in three-quarters of applications) are those that can be subsumed under the cognitive/behavioural tradition (e.g. aggression replacement training/ART, dialectic behavioural therapy/DBT). Likewise, motivational interviewing (MI) is mentioned in half of the applications. However, both milieu therapy and other less standardised approaches (e.g. salutogenic and systemic perspectives), which historically have had quite a strong position in the Swedish residential care, maintain a clear influence on companies aspiring to enter the market (mentioned in around half of the applications respectively). Trauma-informed methods, which internationally have been highlighted as important for the residential care population (Whittaker et al., 2015), are however virtually absent in the care descriptions (4%). Important to stress is that how the mentioned methods actually will be used in the day-to-day care is unclear and prior studies show that staff often lack education in specific methods (IVO, 2013; Lundström et al., 2018).

The IVO as market agency – how does licencing impact the service?

Explicit reasons for rejection

During the study period, the IVO rejected around half of the applications (47%). An analysis of the explicit rejection arguments provides an understanding of the IVO’s assessments, as well as an understanding of what aspects of the care presentations that the IVO considers to be inappropriate.

Although the IVO often presents more than one reason for rejection, the most common rejection argument is related to target group descriptions (used in 54% of the rejections). By not allowing an RCU to enter the market if the planned target groups are viewed as too diverse (with regard to e.g. needs or ages), the IVO contributes to forming a market with increasingly narrowed target groups. For example, the IVO does not approve of mixing children with substance abuse problems with children that have other difficulties:

…  the IVO concludes that the facility intends to accept girls with substance abuse problems. Yet, the owner has not restricted the target group to girls with substance abuse problems. The target group’s composition entails a risk that children with less burdensome problems, who for example are placed for having neglectful parents, will be mixed with children who have destructive behaviour, possibly also with drug abuse problems.

Nor does the IVO allow a target group that falls outside the jurisdiction of the social services, e.g. children with merely neuropsychiatric problems, which in Sweden should be treated within the health services.

The second most common cause of rejection has to do with care content (i.e. methods and other work routines, used in 41% of the rejections). Firstly, applications are systematically rejected if the IVO finds that the care is too restrictive or intrusive (e.g. containing obligatory drug tests, searching through rooms/clothes, or interventions of a punitive nature). Secondly, the IVO sometimes has objections regarding the suitability of treatment methods. RCUs that fail to convince the IVO that the care in an approximate way is apt for the placed children will not be allowed to start their businesses. A typical argumentation is that applicants describe the care content in too general terms, which e.g. leads to the IVO finding it difficult to assess whether the enrolled children will receive the treatment they need in order to combat their difficulties. Such a line of reasoning is exemplified in the following extract:

The unit will provide an active spare time where the youth will receive support in forming contacts with the local citizens associations. The IVO does not find that the mentioned interventions are sufficient in order to help the youth to overcome social problems or give the youth tools to cope with their problems outside their home environment.

Despite roughly assessing the suitability of methods for the target group, the licencing does not involve any profound assessment of treatment methods used in the care and their relation to client outcomes. Nor is the IVO as a market agency endeavouring to shape the residential service by prompting the applicants to explicate how they intend to measure and produce good client outcomes. Having such ambition would be in line with the evidence-based policy officially embraced by Swedish central authorities (cf. Bergmark, Bergmark, & Lundström, 2012).

Around one fifth of all applications were rejected as a result of failing to present a manager with sufficient qualifications. Additionally, previous misconduct (economic and/or related to the quality of care in previous establishments, etc.) of owners/representatives appears to be regarded as unacceptable (18% of applications were rejected for this reason). Hence, by being reasonably strict in its rule-setting with regards to managers’ qualification and owner reliability, the IVO intends to secure that there is some sort of basic professionalism with regard to the management of the RCUs.

It is important to note that a rejection does not necessarily mean that a unit will not eventually enter the market. One quarter (24%) of applicants who initially were rejected reapplied during the study period.³ Of these units, a substantial share (70%) was granted a licence at the second or third time of application. Systematically, the IVO provides the applicant with (at least some of) the precise reasons for rejection, giving the applicant clear insights into the demands of the IVO (which in theoretical terms can be regarded as a form of information through which the IVO intends to shape the service). In general, the rejections concerned parts of application that can be revised solely by reformulating the written description (e.g. presentations of intended target groups and care contents rather than the construction of premises), making a revision very manageable for those wishing to reapply.

Care aspects that are ignored in the licencing process

Over the studied years, there are several aspects of care offerings that, despite being covered by the application form, never or rarely cause rejection and hence are not directly impacted by the licencing procedure. With regard to these aspects, the IVO does not promulgate any clear rules regarding what is an acceptable or non-acceptable way of describing the service (cf. Pålsson, 2017). For example, imprecise descriptions of how an RCU intends to deal with the children’s contacts with the family of origin, or the monitoring of the children’s educational and health needs are never used as rejection arguments. Hence, care aspects that the international literature often describe as crucial – which, in addition to contacts with the family of origin and educational/health needs of the children, include care stability and leaving care services (Whittaker et al., 2015) – are more or less omitted in the licencing procedure. To give an example that concerns schooling, which has been shown to be of great importance and where comprehensive support may be needed (Forsman et al., 2016), merely mentioning that the children are going to be assigned to a local school appears to be enough information to satisfy the IVO. No further information regarding how the children will be supported with regards to homework etc. appears to be necessary. Below is a typical (and according to the IVO sufficient) description of how the applicant intends to cater to the children’s educational needs:

If required, the student can be offered a customized study program in a small class through a municipal school or through a private school. As our facility is situated on the border between two counties, several options for schools are available.

Also, licencing only exceptionally control and evaluates staffing levels or educational levels among staff. This should be viewed in light of residential care being a field where the professional level of staff has been depicted as low in relation to the target groups being children with complex needs that are notably difficult to rehabilitate. Further, staffing levels are an area in which private providers have incentives to cut costs, at the same time as it often is depicted as a crucial quality dimension for human services (see, for example, Meagher & Szebehely, 2013).

Characteristics of RCUs gaining a licence

Exploring characteristics of the companies that actually gain a licence gives an indication of the output of the IVOs assessments, and gives a picture of what kind of RCUs that have been able to enter the market during the last years. Table 2 gives an overview of characteristics (based on information from the application form) that differ between successful and unsuccessful applications.

Table 2. Statistically significant differences in characteristics of applicants gaining/not gaining licence.

		Rejection (%)	Approval (%)	p
Applications (n = 282)	2013–2016	47	53
Organisation (n = 282)	Large company (n = 94)	34	66	*
	Small company (n = 180)	53	47
	Not-for-profit organisations (n = 8)	38	62	–
Manager’s education (n = 245)	Social work/psychologist (n = 166)	42	58	*
	Other degree in social sciences (n = 59)	30	70
	Nursing degree (n = 11)	82	18
	Other (n = 9)	67	33
Target group, age (n = 280)	Children, 12 years and younger (n = 58)	69	31	***
	Children, 13 years and older (n = 222)	40	60
Target group, services (n = 280)	Various psychosocial and behaviour problems (n = 228)	48	52	*
	Specialised towards substance abuse/criminality/ protected residence (n = 33)	24	76
	Toddlers and parents (n = 19)	53	47
Methods/approaches (n = 282)	Cognitive/behavioral oriented	42	58	*
	MI (Motivational Interviewing)	37	63	**

* = p < 0.05, ** = p < 0.01, *** = p < 0.001, - = excluded from the analysis due to too few observations to allow chi² tests.

The rates of approved/rejected applications relate relatively well with what could be expected based on the formal rejection arguments. For example, companies that could present a manager with a qualification in social work or an adjacent field, are to a higher degree granted a licence compared with those that could not. Also, applicants describing more specialised offerings, i.e. older children and delimited target groups (specialised towards substance abuse, criminality or protected residence) have increased possibilities to be granted a licence.

The analysis of the presented methods shows that the approval rate is higher than the rejection rate for almost all categories of methods/approaches (not in table), which confirms the finding that the label of the method is of lesser importance. However, applications mentioning MI or cognitive behavioural oriented methods were more successful than those abstaining from mentioning these methods. Even though the IVO does not explicitly reject an application based on presented methods (above), the higher success rate for applications mentioning certain methods could be related to popular methods (such as MI) lending an air of professionalism to the application, which may influence the licence decision. However, contrary to what may be expected, it seems acceptable that applicants describe a myriad of methods or have methods that are incompatible (e.g. in terms of how the problems of the target group are conceived and should be treated).

An aspect that could not be discerned from the analysis of formal rejection arguments was the much higher success rate among applications from larger companies compared to smaller (68% and 47%, respectively). The higher success rate among large companies could not be explained on the basis of differences in characteristics described in the application forms (such as described treatment methods, target groups, etc.⁴), and may instead be related to circumstances that are not easily captured in quantitative analysis, such as subtle know-how regarding how applications should be written. It is not far-fetched to assume that large companies succeed better as a consequence of their size and resources, which may enable them to describe how they intend to meet requirements in a way that better satisfies the IVO (e.g. by having experience from other RCUs and staff/units working solely with applications and other quality issues).

Summary and discussion

Like in many other countries, Swedish residential care for children has been subject to a far-reaching privatisation in recent decades. This marketisation does not appear to have reduced the need for organisation of the field. Instead, it has spurred new organisational initiatives, and in many countries, audit activities have become increasingly important instruments used to organise markets and steer social services. In Sweden, membership of the market is determined through a licencing procedure, meaning that all private residential units need a licence to be able to compete on the market. In international research, there is increased awareness of public rule-setting such as licencing as one important way to secure the quality of residential group care (see, e.g. Boel-Studt & Tobia, 2016; Whittaker et al., 2016), but knowledge regarding the impact of licencing is still scarce (for an exception, see Pålsson, 2017).

The aim of this article has been to describe and analyse the characteristics and influx of applicants as well as the way licencing aims to impact the Swedish residential care market for children. By drawing on empirical material covering all applications concerning treatment-oriented RCUs processed by the national licencing unit during the years 2013–2016, the study has a well-substantiated base for drawing conclusions. A caveat is that it is difficult to state the actual impact of licencing on care provision and the bearing the assertions of applicants have on the care to come.

The main results show that slightly more than half of the applicants were granted a licence during the studied years; large companies tend to be more successful compared to small companies, and applicants describing specialised target groups tend to be more successful than those describing broad target groups. Contrary to target group descriptions, presenting a wide array of methods does not appear to be compromising with regards to gaining a licence; many applicants display a highly eclectic approach to treatment methods and it is not unusual that the same unit present several methods originating from various theoretical schools. There is a conspicuous influence of methods connoting scientific evidence, but other less standardised approaches are also markedly present. In the licencing procedure, the IVO intervenes on the market primarily through its rules regarding specialised target groups, manager qualifications, care that to some extent addresses the needs of the children, care that does not involve working methods that restrain the freedom of children, and by excluding owners/representatives that previously have had substandard conduct. However, the licencing does not, or only to a limited extent, address issues such as schooling and health support, staffing levels, children’s contacts with their birth family, the use of evidence-based methods and the measurements of client outcomes.

Licencing is a central instrument in controlling and forming markets. Through this instrument, the state decides on membership and has the capacity to affect the scope of a market and the traits of its new members. By in practice benefiting larger companies, the licencing system may consolidate the position of these on the care market, which in time may cause a significant reduction of smaller family-like establishments – which in Sweden historically has had a strong position (see also Lundström et al., 2018). The application fees that have been introduced after the investigated period in order to reduce the number of inferior applications, is likely to further contribute to this development. Further, the liberal attitude to various methods and approaches can be conceived as somewhat surprising, given the contemporary trend of emphasising standardised methods (cf. Bergmark et al., 2012). By omitting to evaluate the scientific basis of methods and not demanding any type of follow up of client outcomes, the system does not contribute to secure that the methods used on the field have a scientific coupling or have shown beneficial effects. As a result, these considerations become the responsibility of the market actors. Neither does the system forcefully demand any particular educational level among staff and the licencing does not attempt at impacting crucial care aspects which could contribute to an improved professionalisation of the service and a care content more aligned with the current research base. This result corresponds to previous research showing that the Swedish control system fails to forcefully audit many care aspects that are stressed in research (Pålsson, 2018). The practical consequences of this is a licencing system that approves of residential care that can consist of almost any method, have an absence of client follow up, and low educational level among staff.

To sum up, even though the actual impact of licencing on the care to come is largely unknown (cf. Pålsson, 2017), the IVO as a market agency can be said to contribute to an increasing concentration of large companies and a narrowing of targets groups as well as establishing that new RCUs have management that to some extent have professional knowledge of the services. However, it scarcely impacts care aspects that arguably are at the heart of the residential service. Hence, although licencing is the main instrument invoked by the state to govern an outsourced service, its actual possibility of forming the residential care market is restricted. Therefore, one can pose the question of whether it, in its current design, in fact, constitutes a missed opportunity to improve the care quality for children.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

We would like to thank the Swedish Research Council for Health, Working Life and Welfare (FORTE) (Forskningsrådet om Hälsa, Arbetsliv och Välfärd) for their funding of the research.

Notes on contributors

David Pålsson

David Pålsson holds a PhD in Social Work and is a lecturer at the Department of Social Work, Stockholm University. His research interests include out-of-home care for children, privatisation of social services, control and follow up of child welfare services and children in families with social assistance.

Emelie Shanks

Emelie Shanks holds a PhD in Social Work and is employed as a researcher at the Department of Social Work, Stockholm University. Her main areas of research include staffing in the social services, social work management, privatisation of social services and residential care for children and youth.

Notes

1 Other organisational elements and methods to influence the market used by market agencies include e.g. surveillance (through inspections control compliance with standards), the ability to issue sanctions (such as injunctions, fines or revocation of licences) and producing images (mapping and reports of a market which in turn is assumed to affect buyers and sellers). IVO utilises all mentioned organisational elements and can hence be viewed as a sort of archetype of a market agency.

2 In line with former publications in the research project (Authors own 2018, forthcoming), we distinguish between ‘small companies’ and ‘large companies’ where the latter owns more than three units.

3 Units that did not occur more than once in the register 2013–2016, nor stated in their application that they had applied before, could not be identified as re-applicants, indicating that this may be an underestimation.

4 Analysed through a multivariate logistic regression, controlling for target group (age, sex, services), manager qualification, and methods/approaches.