The market of personal assistance in Scandinavia: hybridization and provider efforts to achieve legitimacy and customers

The hybridization of the welfare services has resulted in the emergence of a market of service providers for personal assistance in the Scandinavian countries. We describe and analyse two different aspects of the market development. First, we consider the development of the providers ’ market and the different regimes of regulations in the three countries. Second, we examine how the providers handle the multiplicity of claims and expectations in order to attract potential customers in competition with each other. We analyse the different types of organizational identities the providers develop by studying their expressed core values. The findings are presented in the form of three types of values: the value of empowerment, the value of flexibility and the value of professionalism. The differences in organizational values may suggest that the providers direct their services towards different segments of users.


Introduction
The development of the welfare states in the western world is characterized by a changing relationship between state and society. In theory, these changes are termed 'governance' (Pierre and Peters 2000;Stoker 1998). From a 'superior' state, the perspective has changed to emphasize the state's dependency on other actors in the society. The welfare state is no longer the sole contributor, and responsibility is being transferred to the private and voluntary sectors and also to the citizens as co-producers of welfare services (Stoker 1998). Kiviniemi (2008, 364) refers to the rise of a 'contractual state', which has 'given ground for a multiplication of contracting out public services both to private enterprises and to voluntary associations'. This development has contributed to complex service and delivery systems (Osborne 2010), with hybrid solutions across the sectors of society. Elements from throughout the various sectors become interchangeable, and each sector's typical characteristics is challenged and reduced (Brandsen, van de Donk, and Putters 2005).
The hybridization of the welfare services in the Scandinavian countries has resulted in the emergence of a market of service providers for personal assistance (PA). This development has in turn instigated discussions about how much should be left to the market (consumers and producers) to decide, and what should be regulated and controlled by the government. This highlights the tension between the need for security, represented by quality control, against freedom of choice for the users. The tension in question has been managed in different ways in different countries.
It appears that welfare state hybridization is present in all three countries, albeit in different ways and to different extents. In none of the countries is the public sector the sole contributor of social services, as private and voluntary organizations as well as users have been given considerable responsibilities in the field. This article aims to describe and analyse two different aspects of market development in the PA field. The first concerns the development of the providers' market in the three countries: how is the market regulated by the state, and what are the implications of these regulations for user choice? That is to say, how is the tension between quality and freedom of choice managed? The second aspect pertains to the strategies utilized by the new types of providers in the market (i.e. private and cooperative) in order to gain and keep legitimacy. How do they handle the multiplicity of claims and expectations directed to themspecifically, how do they present their services in order to attract potential customers?
In the next section, we present and discuss hybridization and hybrid organizations as a theoretical basis for the analyses. After this, we present our empirical material and explain how it is used. Further on, we describe and analyse the development of the providers' market in the PA field and the different regimes of regulation that have developed in the three countries. In the fifth section, we take a closer look at how both the non-profit and for-profit providers present their objectives and means to the public. At the end, we present a summary of our findings and some concluding remarks, including a couple of recommendations for further empirical research.

Hybridization and hybrid organizations
Hybridity or hybridization is defined in different ways. We will use the concept in accordance with Evers (2008, 279): 'The word "hybrid" means that elements which usually do not go together or belong to different spheres get intertwined within one organisational being'. According to Brandsen, van de Donk, and Putters (2005, 750), 'hybridity refers to heterogeneous arrangements, characterized by mixtures of pure and incongruous origins, (ideal)types, cultures, coordination mechanisms, rationalities or action logics'. The most obvious hybridization is when elements from one sector of society 'invade' another, for instance, when market solutions are being implemented in the public sector. Another example is when public funding expands to the voluntary sector. In our understanding of hybridity, we also include organizations that combine functions that normally 'do not go together'. However, in accordance with Minkoff (2002), the elements do not have to come from different spheres or sectors of society. Important examples of such hybrids are organizations which combine policy advocacy with service provision. They 'combine features derived from distinct organizational forms' (Minkoff 2002, 381). Evers (2005, 742) has emphasized three dimensions within hybridization: resources (e.g. social capital, voluntariness and funding), goals (e.g. for-profit or non-profit) and steering mechanisms (e.g. competition, hierarchy and user influence). 'The processes of hybridization with regard to resources, goals, and steering mechanisms can finally lead to search a new and different corporate identity that reflects the multiple roles and purposes of an organization' (Evers 2005, 742).
Starting from the distinctions made above, the PA market in our respective countries is best described as a quasi-market since the public sector makes the purchase and has the financial responsibility for the services as well as the responsibility for quality control. As the public authority is the guarantor of acceptable services, there is at the same time an increasing pressure (from the politicians, the press and the general public) on the public sector to make certain that public funds are used as intended.
The PA market consists of both private commercial actors (for-profit) and actors that are defined as belonging to the voluntary sector (non-profit). In reality, there are multiple demands on both categories of providers, and there may be tensions between the different demands. The providers have to simultaneously satisfy several parties with different criteria for appropriate behaviour, i.e. goals and steering mechanisms (cf. Pollitt 2003). In other words, in order to gain legitimacy both non-profit and for-profit organizations have to manage the hybridity of claims and expectations currently directed to providers on the PA market.
The private commercial organizations have a responsibility, to their owners and shareholders, to make a profit. At the same time, they have to fulfil the demands from public authorities regarding quality of services. They are providers of public services and have to integrate public sector steering mechanisms and goals into their activities. Consequently, establishing short-term, profitable relationships cannot be the only steering mechanism (Evers 2005).
Furthermore, there are demands, from the disability movements and the individual users of the services, related to influence and autonomy. The areas in which further user influence is demanded include selecting the provider, choosing the assistants and deciding the tasks the assistants have to carry out. Providing public services implies that the standards for what is considered 'good performance' are set by public authorities as well as the users of the services. In this case, then, the challenge of quasi-marketing and performance measurement means that private organizations may be subject to a relatively strong regulation and different types of moral obligations (Evers 2005). The private provider has to satisfy demands from the public authorities as the purchaser, from the users to obtain influence and from the owners/shareholders for profit.
The voluntary sector organizations have also been challenged by quasi-marketing and performance measurement, and Pestoff and Brandsen (2010, 224) emphasize that: 'Longterm relationships based on trust have been replaced by short-term, contract-based relationships, changing the nature of the government-third-sector partnership'. Voluntary sector organizations have incorporated characteristics from public organizations (formalizing) and from market organizations (maximizing income without maximizing profit). Meyer (2010, 534) uses the term 'Social Movement Service Organizations' to describe multipurpose hybrid organizations that deliberately incorporate a mix of organizational features from voluntary organizations, social movements and non-profit service organizations. A main systemic feature is that service provision and advocacy are so integrated that neither of them can be carried out without the other. These kinds of organizations often try to create social change by provision of social services (Meyer 2010, 535). They create a 'critical consciousness' or 'cognitive liberation' (empowerment) among the clients, and thereby a collective identity.
The dual function creates challenges regarding balancing business (provision of services) and being an agent of social change (Meyer 2010, 539). This dualism can cause tension in allocation of resources, and the two different activities demand different competences (Mosley 2010, 512). But this dual focus may also be a competitive advantage in a service market with increasing competition (Mosley 2010, 506).
With this theoretical basis, we can begin to explore and characterize the PA context in more detail. Showing an increasing trend towards more market-based models, Sweden, Norway and Denmark have all engaged in hybridization in the way described above.

Method and material
We have used both primary and secondary data in this article. The section on governmental regulations is mainly based not only on earlier studies carried out by the authors but also draws on other relevant reports and documents. The empirical basis for the construction of the three categories of organizational identities the providers develop is derived from their self-presentations on the Internet. The categories have been primarily constructed from examinations of how providers in the PA market in Sweden present themselves on their websites. The study was conducted in 2011. The three categories are, by necessity, purifications rather than empirically clear-cut divisions. This means that the heterogeneity of the provider market is more comprehensive than is indicated in this text. The analysis has been guided by the following questions: what is the main objective for PA presented by the providers? By what means are the objective supposed to be fulfilled? What categories of PA users are implicitly or explicitly addressed by the providers?
An official overview of the providers of PA has been of great help to us in the collection of the Swedish data. This overview is produced by the Independent Living Institute and covers a large percentage of the providers (www.assistanskoll.se). About 250 providers are listed with their presentations at the website. The quotations from the providers' self-presentations on their respective websites are from the Swedish material and have been translated into English by the authors.
The total number of providers in Norway and Denmark is lower, leading to a weaker empirical base, but we have been able to find examples of the different types of organizations in these countries. In the case of Norway, we have searched the Internet and made other inquiries in the field and detected about 12 private companies. By examining advertisements in membership magazines from disability non-governmental organizations (NGOs) in Denmark we have identified 13 providers.

Governmental regulations and national contexts
There are different categories of providers in the three countries. First, we will give a brief overview of the providers in the three countries. Second, we will describe and discuss the different regimes for regulating the market and some possible implications for user choice.

The providers
Sweden is the country with the highest number of documented users. It has a long tradition of user choice (since the PA reform in 1994, cf. Askheim, Bengtsson, and Richter Bjelke 2014) and the most developed market, with an increasing number of both private and cooperative providers.
The non-municipal providers of PA consist of a very heterogeneous group of organizations. The private companies range in size from small firms focused on only one user to large companies which are parts of multinational enterprises. The number of limited companies organizing services for only one user is increasing. These 'one-user companies' do not compete for the users in the market (they do compete for the assistants with other providers), and they may, therefore, be considered equivalent to users who are acting as the employer themselves. Therefore, we do not include these companies in what follows.
Traditionally, there have been a few dominant user cooperatives in Sweden: Stockholm Cooperative on Independent Living (STIL), Göteborg Cooperative on Independent Living (GIL) and the user cooperative, the JAG Association (JAG). In addition to acting as employer organizations for PAs, STIL and GIL function as special interest organizations representing their members (to the national and local authorities) and as active actors marketing PA as an arrangement. The cooperatives have a relatively modest share of the market.
There are not yet many private providers in Norway. Traditionally, there has been only one user cooperative: Cooperative on Personal Assistance (ULOBA). ULOBA was founded in 1991 on the principles of the Independent Living Movement. The organization is owned by disabled people and is open to everybody who wants PA, and the members operate as the managers for their PAs. In addition to being an employer organization for PAs, ULOBA functions as a special interest organization for the members, representing them in concerns with the national and local authorities, and also as an active actor marketing PA as an arrangement (Andersen et al. 2006). ULOBA has worked towards the development of the PA arrangement through demonstrations, lobbying and other activities on all political levels. Over the last several years, a few private companies have started up, but only on a small scale.
In Denmark the PA employers' responsibilities were solely a matter for the user until 2009. However, after a reform of the PA Act, user cooperatives and private companies became viable alternatives, allowing for the development of a new market (cf. Askheim, Bengtsson, and Richter Bjelke in this volume). The municipalities cannot be employers of user-led assistants in Denmark, but they can administer the wage payments for users who do not wish to do so themselves. Formally, in these cases, the users are still the employing party. There are several private enterprises in the market, but so far only one user co-operative: Landsorganisationen Borgerstyrt Personlig Assistance (LOBPA).
The cooperative LOBPA began functioning in 2009 and defines itself in this way: 'LOBPA is a democratic, citizen-led, non-profit organization, established and run by citizens that are all personal assistance users' (www.lobpa.dk). LOBPA is the only manifestation of the Independent Living philosophy in Denmark. It is focused on being user-led and non-profit and gets a lot of inspiration from ULOBA in Norway. Like ULOBA, LOBPA also fulfils the double functions of being both an employer organization and a special interest organization for people with PA, as PA users that do not have LOBPA as employer organization also can be members (www.lobpa.dk).

Regimes for regulating the market
As an introduction to the descriptions of the three national contexts, we will outline two principal methods used by the state to regulate the provider market in the welfare sector. One concerns the prerequisites for entering the market, and the other the prerequisites for operating in the market: (1) Authorization of providers. This refers to the need for an actor, who wishes to be a provider of a welfare service, to be approved by an agency with the authority to do so. Normally, this will be a public authority, and the authorization may be carried out at the national (state) or local (municipal) level.
(2) Control and supervision of the providers and the way the services are carried out.
This involves controlling the quality of services to ensure they are satisfactory, as well as supervising the use of resources to confirm they are used in accordance with agreements (to avoid misuse of public funding). Additionally, it concerns oversight of the service workers' working conditions to ascertain that they conform to regulations. This control may also be carried out at the national or local level.
The way these methods are implemented will indicate what kind of regime of regulation is being observed. With no authorization or control of the providers, it is a free market with almost free user choice of provider. With an authorization and control of the providers at the national level, it is a centralized regime of regulation, and when the authorization and control are conducted at the local level, it is a decentralized regime. As mentioned in the introduction, there is a tension between freedom of choice for the users and regulation of the market through authorization and control of the providers. A customer perspective will focus on obtaining the largest possible freedom for the users to choose a provider and, therefore, may argue against any regulation of the market that will restrict the users' choice. At the same time, an unregulated market may put the users in a risky position because it allows the presence of providers that are not serious. Providers that deliver low-quality service are dishonest, and/or misuse of public funding are not acceptable from the public point of view, and they will even undermine the serious actors in the market. In such a perspective, certain types of regulations of the market may be preferable.
The three countries differ concerning regimes of regulation, and there have also been changes over time in the countries.
In the beginning, the provider market was virtually unregulated in Sweden. But in the 2000s, important questions about how the market was functioning have been raised at the governmental level. The state authorities have focused on three types of problems: the working conditions of the personal assistants, the competence of the assistants and the aptitude of the private providers. The latter also includes some of the private companies' financial dispositions and social responsibilities (Hugemark 2006). This started a shift from a market discourse to a regulation discourse, where market goals and marketsteering mechanisms are questioned. In 2011, Sweden entered into a nationwide governmental authorization and supervisory arrangement with providers of PA (National Board of Health and Welfare 2011). Sweden has thus implemented a centralized regime.
We do not yet have any clear understanding of the implications, for the market and the users, of the new regulatory regime. At the end of 2012, about one-third of the providers had their applications refused (Assistanskoll 2012). As yet, no official report has been presented on the kinds of companies that have been refused. However, statements from the responsible authority indicate that the most common reasons for rejection are lack of skills and incomplete applications (Assistanskoll 2011). Some substantial related concerns are the questions of what kind of control and supervision system will ultimately be implemented, and how it will be possible to effectively control the providers at the national level.
Norway differs from the other countries in this group by the fact that it does not legislate user choice at the national level. The municipalities are the most important actors among the public authorities in this field. They have to either approve the choice of provider in each separate case or establish an arrangement for the authorization of a group of providers from which the users can choose. Several municipalities do establish authorization of providers and user choice themselves, so there is a tendency towards user choice of PA even in Norway (Guldvik and Andersen 2013). The municipalities also have the responsibility of overseeing the existing providers. Subsequently, there is a decentralized regime in Norway.
The locally based regulation regime results in big differences between municipalities in both what kind of providers the local authorities accept and how the control of the providers is implemented in practice (Andersen et al. 2006;Guldvik and Andersen 2013). Therefore, the development of the market of PA providers and its geographical diffusion nationwide is, to a certain degree, decided by the municipalities. They can open the market by authorizing private and non-profit providers for the users to choose between, or they can restrict the market to just the municipal provider, taking the choice away from the users.
Denmark does not have any system for the authorization of private providers. The municipalities have the responsibility of controlling the providers, with regard to both quality of service and use of resources. This can be done partly by reports every year and partly by access and control on demand. As a consequence of the localization of this responsibility, the degree and type of control varies between municipalities (Deloitte 2012). Serious actors in the market welcome an efficient system of control that could uncover fraudulent activities and exclude incompetent providers from the market (www. lobpa.dk).
The development of the market in Denmark will be dependent on whether it becomes more attractive to the users to choose a provider to handle the employer responsibilities instead of handling it themselves. A report, based on a user survey performed in 2011, shows that 36% of the users have chosen a provider to handle the employer responsibility. There is a tendency for younger and new PA users to choose a provider to take the responsibility, so the share may be higher in the future (Deloitte 2012). Thus, the market of PA providers will probably increase.

Management of multiple demands
Despite the differences summarized above, providers of PA in all three countries have to manage multiple demands from the public, private and voluntary sectors in a way that legitimizes them. With reference to Evers (2005, 742), we can ask the following question: how is 'the process of hybridization with regard to resources, goals, and steering mechanisms' reflected in PA providers' corporate identities? That is, to what extent are their multiple roles and purposes reflected in their organizational identities and values towards PA as presented in public?
In this section, we start out with an examination of how non-profit PA providers present their systems of organization and goals in public. We then continue with a corresponding examination of the for-profit providers. As mentioned earlier, we consider the way the providers present themselves to be a manifestation of their strategies towards attracting potential customers and showing how they fulfil governmental regulations. Through these presentations, they hope to reach levels of legitimacy and resources that will allow them to survive in the PA market.

Non-profit organizations
The typical non-profit organizations in our material are represented by the user cooperatives, i.e. organizations in which user influence is the basic steering mechanism and the members represent an important resource. On the basis of the presentations of these organizations, we call the core value in the provision of PA 'the value of empowerment'. These organizations strongly emphasize disabled peoples' right to make their own decisions and highlight user control as their steering principle. Here is one example of how this is expressed in connection with PA: 'The cooperative is user controlled, meaning that it is the users of personal assistance that are members and determine the direction of the cooperative' (Rikskooperativet Blå/Sverige). The cooperative stresses that the users, in a concrete sense, are managers for their assistants. The manager function includes control of the employment and the work of the PAs, as well as control of how the financial resources are distributed and administrated. But the user cooperatives also put a great deal of importance on respect for the collective agreements in working life, which give some formal borders for the work managers' ability to control the assistants' work and working conditions. Thus, non-profit organizations explicitly address two different external, societal claims that stand in a rather complicated relationship to each other. One is related to evident political demands in the field of disability, i. e. demands that played an important part in the background of the PA scheme. Another is related to agreements in working life, a crucial element in 'the Nordic welfare model'. However, in the presentation of 'the value of empowerment', the claims for certain working conditions, etc., for PAs appear as rather unproblematic and nothing that hinders fulfilment of genuine user control.
'The value of empowerment' is further enhanced by arguments that the organizations in question offer something qualitatively special compared to the PA services provided by the municipality. Empowerment is presented as a value that is most safely guaranteed by a user cooperative: 'Do you think that your present provider, for example the municipality, decides too much about your assistance?' (Kooperativet Frihet/Sverige). 'The value of empowerment' is based on the belief that the disabled person herself/ himself has the undisputable right to decide what problems are to be addressed and how they should be solved: 'We are the real experts of our needs and must take the responsibility for our development' (Kooperativet STIL/Sverige).
Some of the non-profit organizations are cooperatives with members that are dependent on other people in order to express themselves and to function as work managers. In their presentations, these organizations, for example, describe that they 'adjust qualified personal assistance to persons with multiple impairments and to persons with intellectual disabilities' (Kooperativet JAG/Sverige). In those cases, we find an additional element involved in order to achieve 'the value of empowerment', namely, representatives for the user of PA. A relative, guardian or another trusted person may in these cases assist the user or take full responsibility for the work management. The importance of the disabled persons' own influence and empowerment is, however, as highlighted in these organizations as in other user cooperatives.
The user cooperatives can be understood in their role as non-profit organizations, but some of them are also part of a social movement, promoting the collective interests of people with impairments. This as well is an important aspect of the self-representation of the cooperatives that are included in the Independent Living Movement. The value of individual as well as collective empowerment is presented to the public. With reference to Hirschman (1970), we can speak of organizations that put a strong emphasis on 'voice', i.e. that attach great importance to the members' opportunities to control and achieve changes by making their voices heard. This is an affiliation with movements oriented towards equal status for dominated and marginalized groups in society and movements which introduce a new social identity and promote political demands on behalf of the group (cf. Andreassen 2004, 23). In other words, organizations claim equal status and citizen rights for their members in accordance with the civil rights movement (Shakespeare 1993).

For-profit organizations
The for-profit organizations are by definition privately owned, i.e. the property right is the basic steering principle, albeit complemented by different types of labour laws introduced in order to empower the employees. Thus, while the members are important resources in the cooperative, the employees are indispensable resources in the for-profit organization. On the basis of the presentations of these organizations, we can find two different values in the provision of PA, each corresponding to organizations of rather different character when it comes to owner structure. We call them 'the value of flexibility' and 'the value of professionalism'.
The organizations that express the value of flexibility are typically rather small. The small-scale character is also something that is presented as an important prerequisite for their ability to offer PA guided by the value of flexibility. The small size is an important feature in the presentations of some of these organizations: 'Eva & Cathrine welcome you to the small assistance company with the big ambitions … Welcome to a small assistance company with a big heart and a small administration' (SveGar Assistans/Sverige). Some also make explicit references to PA providers that are described as guided by values other than flexibility. One example is criticism of the loss of flexibility and freedom of choice at municipal agencies. Some small-scale organizations present themselves using references to the employees' previous experiences working in municipal agencies: 'We have been working with disabled persons in municipal agencies for many years. We know that it is not easy to make personal assistance function well' (Pyradacka/Sverige). Such an organization may also refer to its customers' experiences: 'We know that several of our users have chosen us because they are tired of the municipal agencies', and also a lot of the big assistance companies', loss of flexibility. We prefer flexibility and have chosen to work on a small scale because we are always able to construct flexible solutions that are personally adapted to each individual user' (Prompera/Sverige).
Strong user influence is an important part of this type of organization's profile. Nevertheless, this influence has a different meaning than in the user cooperative. The steering is flexible, and the users' opportunities to influence the PA scheme are emphasized in the small-scale private organization. But it is not the discourse of citizenship that is this type of organization's basis, but rather a market-orientated discourse; they are motivated to speak of 'exit' rather than 'voice' (cf. Hirschman 1970). It is the consumer that is making choices in the market that the organization takes into account: 'We talk about the disabled as customersnot patients, care receivers or clients. They are customers because they decide. If they are not satisfied they have the opportunity to choose another service provider' (Pyradacka/Sverige). But this type of organization also points out that the customer's choice is integrated into the steering of the services: 'We know that the assistance never can be the same for our different customers, so it differs from person to person, as the rainbow's different colours' (Din Assistans/Sverige). The degree of engagement and influence is decided by the customer. The customer may choose to have a big influence, or she/he may choose not to have any influence at all.
A few of the small-scale for-profit organizations do, however, show some similarity with some of the cooperatives. The similarity in these cases is that they do not only provide assistance but also offer juridical support and help when their customers apply for PA.
While the for-profit organizations that present flexibility as their leading value are typically small scale and often locally based, the characteristics of our second group of for-profit organizations are quite the opposite. The organizations whose presentations are based on 'the value of professionalism' are more often part of larger companies (sometimes even a multinational enterprise) and have a profile of professional care. These organizations do not present themselves as radically different alternatives to the municipal agencies, but rather as supplements: 'It is our firm belief that we offer a strong alternative to the traditional ways of working in the municipal agencies. The competition has led to a reasonable division of work between public and private care agencies' (Förenade Care/Sverige). Another way of putting this is that this type of organization asserts that the private owner engagement is important in order to stimulate the development of the care and service sector overall, rather than arguing that it is an alternative with qualitatively different content than the municipal services. The organizations that express the value of professionalism for PA often use their organizational size as a positive indicator in itself: 'We are a part of ISS Sverige AB which is a nationwide service company with activities in 400 places in the country' (ISS Care Service/Sverige).
Unlike the organizations that stress the values of empowerment or flexibility, the organizations that highlight the value of professionalism offer many other services in addition to PA. They also have wider target groups, including especially the elderly. In these organizations' presentation, it is sometimes even difficult to recognize PA as something qualitatively different from the other provided services. Often, care for the elderly and the disabled is presented as a common field of services, rather than describing PA as something special and unique. Like the other providers of PA described above, the large-scale providers also profess a commitment to freedom of choice and respect for individual integrity and influence: 'We design your assistance on the basis of your personal needs and preferences. You have strong influence on how and when you want to receive your assistance. We assist you in your contact with public authorities, the social security agency and the municipality' (Attendo/Sverige).
However, in the name of professionalism, the users' dignity and safety are also emphasized: 'Our goal is always to give our care receivers a dignified and safe life' (ISS Care Partner/Sverige).
A prominent element of how these organizations present themselves is the emphasis that is put on the degree of competence of the personnel, which is much wider than what is generally found in the assistance field. One way this is done is to highlight the quality of services by referring to the staff's education and earlier work experiences from treatment and care. Another is to point out the personnel's ethical attitudes, characterized by humanity and respect. To further emphasize the importance of service quality, some of the organizations refer to certifications in accordance with international quality standards: 'As a result of the high quality we offer in the care services, we achieved the ISA 9002 certification in 1999, as one of the first care service companies in the Swedish market' (Förenade Care/Sverige).
Some of the organizations grouped under the value of professionalism differ from the others in a particular way. They accentuate the possibility that PA may lead to progress and development for persons with impairments. These organizations place added focus on professional knowledge of rehabilitation and activity. In this way, they especially target persons with impairments who have a likelihood of profiting from activity and training: 'We have developed a nationwide program for personal assistance with focus on activity … Through a well-functioning assistance our participants may continue to develop technique, strength and motivation' (Frösunda/Sverige). The intention is to present themselves as offering PA in a way that helps the users to develop themselves and overcome difficulties to as great an extent as possible.
The organizations which give expression to the value of professionalism uniformly underline that they are large and well established and possess well-developed routines to achieve their goals. One advantage they present is having a group of personnel with a wide competence base, including especially personnel with knowledge of disability and medical and rehabilitation issues.

Conclusions
We have seen that the PA markets in the three Scandinavian countries are developing with many different categories of providers. We will characterize this as a quasi-market with a hybrid character, albeit with different forms and contents. In Sweden and Norway the municipalities are still important providers of PA, while they are not permitted to be providers in Denmark. The private companies vary in size, profile and other characteristics. The cooperatives also vary in size and scope, from small and locally based to large and nationwide. There are many and varied user cooperatives in Sweden, while Norway and Denmark have just one nationwide user cooperative each.
The three countries also have different types of regimes for regulating the provider markets. Sweden has progressed from an unregulated market to a regime of state regulation with a centralized arrangement for approval and control of providers. In Norway, there is a decentralized regime in which the municipalities have the responsibility to impose an approval arrangement on, and also to control, the providers. The municipalities also decide if they want to implement a system of user choice or not. Denmark is in a position between the two, without a system for approval of the providers, but with a municipal responsibility to control them.
Both the market sector and the volunteer sector organizations can be categorized in different ways. We have divided the providers into three main categories, based on how they market themselves in a hybrid context and how they try to cope with the multiple demands they are confronted with as they struggle for legitimacy and resources in relation to potential customers. The first category includes the typical non-profit organizations in this field, which are represented by the user cooperatives, expressing the value of empowerment. The for-profit organizations make up the other two categories, expressing the values of 'flexibility' and 'professionalism', respectively.
The user cooperatives emphasize their fulfilment of both the demands of disabled people's movements and the users' demands for management. They must have a sufficient income to survive in the market, but they have other values related to user control and liberation which are important to focus on. Thus, we name their expressed core value as 'the value of empowerment'. The user cooperatives also stress the importance of escape from municipal governance and emphasize that the users are the managers (work leaders) that control the services and, thereby, also evaluate the quality of the services. This will probably be the cooperatives' advantage in their competition with other providers in the market. The user cooperatives thus try to gain legitimacy by appealing to users that are ideologically attracted to the social movement of disabled people and their struggle for social rights and that are concerned with controlling their own services.
The organizations expressing the value of flexibility fulfil demands from neoliberal actors for choice of providers and the possibility of 'exit'. These may be small limited companies which have to carefully manage their finances, but they do not necessarily have external shareholders claiming a share of the profits. These companies focus on nearness, flexibility and individual adjustment of the services. They may also specialize by working with specified target groups of users and utilize this as a niche strategy in the market competition. The organizations of flexibility thus try to gain legitimacy by appealing to users that feel like customers in a market and appreciate the nearness and the possibility for an individual adjustment of the services, but at the same time see themselves as capable of deciding to what degree they may influence the services. And if they are not satisfied, they may choose to exit and select another provider.
The organizations expressing the value of professionalism focus on their attention to professional quality. These are usually big companies with a for-profit orientation. They are focused on traditional care values and directed towards elderly and disabled people and emphasize safety combined with a service orientation and opportunities for the customers to influence the service provision. These organizations thus try to gain legitimacy by appealing to users that are concerned about quality, dignity and safety and that trust in professional competence rather than being concerned with either controlling their services themselves or being able to exit in case of dissatisfaction.
The organizations' ways of presenting themselves through their expressed values indicate that the providers direct their services towards different segments of users. From earlier studies of the users of PA, we know that they have different preferences and emphasize different aspects of the arrangement. On the basis of empirical data from Norway, Guldvik (2003) has constructed three ideal types of users of PA: 'ideologues', 'traditionalists' and 'rehabilitationists'.These types illustrate that the users differ concerning how they would like their assistance to be carried out (see also Askheim et al. 2012). The 'ideologues' emphasize that they are the main experts on their own lives and that they have the right to decide over all aspects of their assistance. They also place great importance on the flexibility of the assistance. They are labelled 'ideologues' because their attitude is in accordance with the ideology of Independent Living. Providers expressing the value of empowerment will appeal to users with the ideological attitude. The 'traditionalists' believe that the assistance should be developed in cooperation between the assistants and the users, and some of the providers expressing the value of flexibility may match their preferences. The 'rehabilitationists' are primarily concerned about that assistance is given in a safe and secure way, and the providers expressing the value of professionalism may appeal to this category of users. This implies that the competition between both for-profit and non-profit providers in this quasi-market creates diversity and thereby gives the users possibilities for choosing a provider that matches their preferences. However, further research is needed before we can make such a conclusion. We lack sufficient knowledge of how the users comprehend the differences between the providers and of how the providers' practices match up with their claims. This raises several important questions for further empirical studies. To what extent do the providers offer adequate information about their services to the potential users, so the users can understand the differences between them? Do the providers actually function as they say they do, and are they genuine alternatives to each other?
The development of such a diversified provider market may also be regarded differently depending on ideological viewpoint. On the one hand, the development may be seen as positive because different categories of users can choose providers with profiles that match their preferences. On the other hand, one possible objection is that the development of providers founded on traditional care values may represent a threat to the ideological basis of PA and, therefore, can be regarded as unwanted in the PA market. Another and quite opposite possible objection is that the heterogeneity in the PA field may represent a threat to other welfare values, i.e. equal treatment and safety.