Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The literature often sees universalism and local autonomy as the key tenets of Nordic care regimes (Burau et al. 2007); the former refers to substantive aspects of long term care policies , while the latter refers to procedural aspects. Against this background, the case of Denmark is interesting in two respects. Firstly, among the Nordic countries, long term care policies remain the most universal in terms of coverage, which is reflected in the level of public expenditure. Secondly, Denmark combines institutional change from below (nonlegislative change) with institutional change from above (legislative change).

Although Denmark , like most of the OECD countries, has been exposed to New Public Management (NPM) reforms, understood as a drive for a retreat of the state, cost containment and consumerism (Dahl 2005; Glendinning 2008), long term care policies have not been characterized by retrenchment. Yet the absence of retrenchment does not necessarily mean an absence of change. Therefore, in this chapter, we will first investigate whether there has been a change in long term care for older people in Denmark in the period 1994–2007, and if so, we will seek to identify the characteristics of this change. We will do this by investigating how any changes relate to existing institutions. Indeed, as described below, our analysis indicates that change has occurred through restructuring (Pierson 2001) . Specifically, long term care policies since the 1990s have included elements of both control/standardization and flexibility /choice , which has led to substantial changes in terms of the organization of long term care (Dahl 2005). In procedural terms, reforms represent a form of reproduction by adaptation , whereby new elements are attached to existing institutions which gradually change as a result (Streeck and Thelen 2005; Thelen 2000). Taking a broad concept of reform from neoinstitutional theory (Streeck and Thelen 2005) as our point of departure, we analyze two reforms that encompass elements of control and flexibility , and which in procedural terms point to legislative and nonlegislative forms of incremental transformation .Footnote 1

2 The Structures of Opportunities and Constraints for Reforms

In many ways, the long term care system in Demark is a classic example of the Nordic welfare state, combining universalism with local autonomy . While services are uniform and are both designed for and used by all (Anttonen 2002), services are largely controlled locally (Kröger 1997). Services are publicly funded and coverage is highly universalistic, the provision of services is public, and the organization of both funding and provision is largely the responsibility of local authorities, which also enjoy considerable powers in relation to the regulation of services.

Long term care includes both institutional and home care; although for more than five decades there has been a focus on the latter (see Sect. 4.3). Home care has a clear social care orientation and encompasses personal care as well as practical help. Access to services is based on the principle of citizenship and, by law, citizens have the right to get help if they experience difficulties with activities of daily living; in comparison, the provision of services is needs based. Home care is free at the point of use and is funded by local authorities; they are also responsible for providing services and for conducting needs assessments. Taken together, this puts local authorities in a highly influential position in terms of shaping long term care. Here, the provisions of the Social Services Act are key factors, as they offer the basis for a unified yet decentralized long term care system (Doyle and Timonen 2007). While the Act requires local authorities to provide the necessary services for its citizens, it is up to the individual authorities to determine the substance, level and organization of services. More specifically, the Act distinguishes between what local authorities have to do and what they can do (Nielsen and Andersen 2006). In practice, the long term care system in Denmark is not only highly universalistic but also generous (measured in terms of the percentage of older people receiving services) when compared with other Nordic countries (Szebehely 2003).

The decentralized nature of the long term care system means that the policy process in relation to reform occurs within the context of central–local relations . This makes the central government and the local authorities two of the key players in the policy field. Through the Association of Local Authorities (Kommunernes Landsforening, KL), local authorities have significant resources and are in an influential position by virtue of their responsibility for funding and providing long term care (Blom-Hansen 2002). Their interest in relation to long term care is twofold, namely to retain/expand existing funding and to safeguard the autonomy local authorities enjoy concerning a wide range of governing issues.Footnote 2 In contrast, the resources of central government are connected to setting the overall policy framework and to allocating additional funds to long term care. For the period we are investigating, the overall interest of central government was to develop its steering capacity vis-à-vis long term care, both in relation to containing costs and determining the substance of services. As in other Nordic countries, the balance between the central and the local level has been subject to change. While the 1980s were characterized by decentralization , detailed regulation replaced the earlier legislative framework from the 1990s onwards (Hansen and Vedung 2005). The tighter regulation of details resulted in tensions between the two levels, whereas the Association of Local Authorities has tried to actively resist the decreasing autonomy at the local level (Dahl 2011). Nevertheless, Denmark’s policy style is broadly consensus-oriented and there are multiple formal and informal channels which allow a wide range of policy actors to influence the policy process. Of these, the yearly budget negotiations between the central government and the local authorities are particularly prominent.

The other two main actors in the policy field are the Trade Union of Occupation and Work (Fag og Arbejde, FOA), which mainly organizes care workers, and the major interest organization of older people, called ‘DaneAge’ (Ældresagen). Occupation and Work is a specialized trade union for care workers, that is, home helpers and auxiliary nurses, and it has a total of approximately 200,000 members.Footnote 3 With its history as a trade union for public employees, the trade union also includes other professions. Indeed, there is no other trade union for care workers, only one for nurses. The trade union has traditionally had a middle-of-the road social–democratic orientation, but over the last five years has gradually become more radicalized, also defending welfare rights more broadly. The trade union has successfully put issues of work conditions—including the organization of work and the gender wage gap—on the political agenda. The membership of DaneAge is even bigger (about 600,000; DaneAge 2011), which seems to defy the dictum that the interests of older people as such are difficult to organize. Dane Age accepts members below the retirement age and thereby also lobbies on behalf of those sympathetic to the needs of the elderly, such as relatives. However, even without these additional members the organization’s membership among the older population is significant. Together with extensive fundraising activities, this also makes for significant financial resources and this gives the organization the possibility for a strong media presence as well as for conducting its own research. Not surprisingly, DaneAge is especially visible as a watchdog for older people, who encounter unfair, bureaucratic treatment or who experience a fall in their home help entitlements; indeed, the organization has several times threatened to summon a local authority. DaneAge also lobbies against retrenchment. In short, the interests of DaneAge are to promote, safeguard and expand the welfare and user rights of older people.

What are the current pressures for reform? These are best understood in the context of the historical development of the long term care system. This goes back to the period after the Second World War and the introduction of publicly funded but means tested ‘home makers’ providing temporary support to families in times of crisis (Dahl 2000; Petersen 2008). This offered a springboard for a substantial expansion of long term care over the next three decades (Hansen and Vedung 2005; Nielsen and Andersen 2006). Legislation introduced in 1958 marked a shift to homecare as both the permanent, time unlimited service and the legislation in 1964 explicitly required local authorities to offer assistance to older people, so that they could live at home as long as possible. This extended the responsibilities of local authorities and so personal care became a distinct focus of home care. The Social Services Act in 1974 brought together different provisions under one legal framework and further underlined the importance of personal care, by highlighting the need for training for care workers. Services were typically centered on the home and the principle ‘at home as long as possible’ became an entrenched principle of long term care policies in Denmark early on (Lewinter 2004). In contrast, policy developments since the early 1980s have been characterized by restructuring (Hansen and Vedung 2005) .

While, in a cross country comparative perspective, the system of long term care in Denmark emerges as the relatively cheapest and best (Sarasa and Mestres 2007), from the domestic point of view, the total amount of public funds spent on long term care is seen as the key reference point and the reason that the State is interested in influencing the provision of long term care (Nielsen and Andersen 2006). Further, with the neo-liberal turn, high public expenditure is also seen as a problem in Denmark , and concerns for containing expenditure, value for money, and responding to individual needs have become predominant (Dahl 2004, 2009). The same applies to healthcare, although shifting costs from health to social care is not an explicit issue. In addition, several factors are all seen to put more demand on long term care services (Hansen and Vedung 2005): besides rising numbers of older people and processes of urbanization, the higher labor market participation of women is especially prominent compared to other countries. For example, the percentage of women in paid labor (the employment rate) rose from 43.5 % in 1960 to 74.4 % in 2009 (Danmarks Statistik 2010a; Hansen and Vedung 2005, p. 54).

Given these conditions, especially the institutional settings outlined above, what are the opportunities and constraints for reform? Denmark has a universal, institutionalized system of long term care, which makes any form of material retrenchment—such as reducing existing entitlements—difficult. The interests of older people and care workers are also well organized. This is compounded by a ‘care ideal’ that sees long term care first and foremost as a public rather than a private responsibility, and an ideal that privileges care in the private home over institutional care.Footnote 4 For example, family members are neither in practical nor in financial terms expected to support an elderly relative (Doyle and Timonen 2007). Concurrently, long term care has over the years become increasingly professionalized, with formal training of care workers at both basic and advanced levels. The training of home care workers has been extended from a couple of weeks to a whole year based in vocational schools (Dahl 2005). Home care workers have become social and health workers, and after graduating, they can qualify to become social and health assistants by attending 2 years of additional vocational training. Home care workers also perform some nursing functions and can take on leadership positions in care for older people or work in hospitals.

Similarly, political decision making based on consensus puts constraints on making radical reform, as the wishes of local authorities, the trade unions of care workers (FOA) and interest groups for older people (particularly DaneAge) have to be taken on board. More specifically, in relation to the parliamentary process, the consensus orientation is partly counterbalanced by the fact that, since 2001, Denmark has had a (centre-right) minority (coalition) government with a permanent support party that has not shied away from using its majority.Footnote 5 Further, the existence of a well-institutionalized and highly integrated policy arena (for a more detailed discussion, see the following section) also means that there is considerable administrative capacity at both national and local government levels. It is indicative, for example, that in 1998 the government took the initiative to get in touch with its citizens in a more direct way through the ‘preventive home visit’ (forebyggende hjemmebesøg). This allows for a more explicit codification of universalism , which otherwise tends to be defined in loose terms.

3 The Process of Reform

As outlined above, long term care policies go back to the late 1950s and have been subject to systematic expansion over a period of three decades. This makes for a policy arena that is well institutionalized, not least because the policy arena is firmly embedded in the structures of central–local relations , a feature which is one of the cornerstones of the political system in Demark.

The policy arena is also characterized by a clear focus on homecare. This is typical of Nordic care regimes and can be seen in the policy practice: ‘at home as long as possible’. In Denmark , this has its early beginning in the late 1950s when ‘older people at home’ emerged as a distinct user group, while the concrete policy practice goes back to the late 1960s and was motivated by both financial and policy considerations. In the context of the expansion of long term care, homecare was seen as less expensive than institutional care. This coincided with a widespread societal critique of care in institutions, from which home based care emerged as the ‘better’ option (Dahl 2000). Care in institutions was increasingly demonized and portrayed as an antithesis to freedom. In this dichotomous thinking, freedom in the private home was the preferred policy option. In concrete policy terms, this led to the expansion of home-based long term care together with the closing down of nursing homes from early on; legislation introduced in 1987 even prohibited the building of new nursing homes and thereby consolidated the principle of ‘at home as long as possible’ (Dahl 2000; Rostgaard 2007). Not surprisingly, the number of nursing homes has fallen dramatically, whereas the number of places in assisted housing has risen significantly (Doyle and Timonen 2007).

More recently, there have been boundary issues in relation to health care, reflecting the fact that both policy arenas have shifted towards a reform paradigm of restructuring and thereby a greater concern for public expenditure and value for money (for health care, see Strandberg-Larsen et al. 2007). Boundary conflicts are particularly centered around discharge from hospitals. Through a range of measures, hospitals were put under considerable pressure to reduce patients’ length of stay, which has had repercussions for the demand on long term care provided by the local authority (Strandberg-Larsen et al. 2007). The local authorities need to have suitable long term care services ready on discharge, as they otherwise will be penalized financially; they have to cover any additional hospital charges that patients incur because of a lack of care facilities at the local level (Dahl 2008).

As mentioned above, the substance of policy change has focused on restructuring , which Pierson (2001) defines as reforms that aim to make the welfare state more compatible with contemporary goals and demands . This can occur either through rationalization, where welfare programs are brought in line with new ideas of how to achieve existing goals, or updating, where welfare programs are adapted to changing demands and needs. The long term care reforms in Denmark have included both types of restructuring: the introduction of market mechanisms is an example of rationalization, where long term care is brought in line with neoliberal ideas about the superiority of market mechanisms for delivering welfare services; the tailoring of services to individual needs is more of a hybrid, as it is about both creating consumerism and responding to the complex needs of a more individualized society. Restructuring has occurred through a process of institutional change based on reproduction by adaptation . We look now at two reforms that are typical of this type of institutional change, but which show that institutional change can occur both through legislative and nonlegislative means.

3.1 A ‘Common Language’ for Home Care

Common Language is an example of a bottom-up, nonlegislative policy change which originated from the Association of Local Authorities (KL) in 1998 (Hansen and Vedung 2005; Højlund 2004). Common Language can be seen as an attempt to define in a more objective way the level of ability of the older person and to link the levels to categories of service provision. The policy change consisted of a new initiative in which older care users were allocated to one of four newly developed categories of care needs, ranging from independence (level 1) to total dependency upon others (level 4; Petersen and Schmidt 2003). Each of these categories systematically linked individual needs with specific types and with a specific level of service. The initiative addressed concerns about securing the welfare rights of citizens, but at the same time rendered other less formalized aspects of care services invisible (Højlund 2001, 2004). The initiative also tightened the control over the delivery of care services by professionals and as such fit with the central government’s aim of introducing quality standards and preparing for the introduction of a purchaser–provider split.

The Association of Local Authorities, as an umbrella organization, also provides consultancy to its members in case of administrative and organizational problems. It is in this servicing capacity that KL in the early 1990s discovered that the field of care for older people lacked a coherent system of resource management and service provision (Højlund and Højlund 2000, p. 24). The municipalities were frustrated and within that context, KL suggested that something needed to be done. Based on a belief about the potential of technology for solving organizational problems, local authorities believed that governance problems (i.e. poor management and organization) could be resolved through the development of new information technologies. Driven by the concerns of the municipalities, KL began to investigate the more specific nature of these steering problems. In 1994, it published a report arguing that the long term care sector suffered from an extremely weak system of governance. At that time, no statistics on the costs of long term care were available, and the relationship between needs and help provided remained incomprehensible. KL decided that the introduction of new information technology could solve this problem, and initiated a developmental project. However, in 1995, KL realized that the software firms commissioned to design the new technology could not deliver the requested product, as there was no common understanding of the work delivered by home care workers available (Højlund and Højlund 2000, p. 25). There was, in short, no common language to describe what work was being done in the home with and for the older care recipient.

Consequently, the strategy was revised and the development of a common understanding of care services (Common Language ) began. The new language was supposed to provide a catalogue for the needs of older people and the corresponding services of home care workers. Common Language was launched in 1996 and represented a particular form of codification of most of the hitherto tacit knowledge and also offered a standardization of the services provided . Common Language was developed in several stages through a process based on ongoing dialogue with selected local authorities, which took part in pilot programs (Hansen and Vedung 2005). Version II was developed to improve the earlier version by adding additional dimensions, including the experienced needs and motivation of users as well as the training activities, and also expanded the applicability of Common Language to enable communication with hospital staff (Kommunernes Landsorganisation 2002a).Footnote 6

Common Language initially emerged as a purely bureaucratic and supposedly neutral tool, but soon became politicized, whereby different professional groups together with users demanded influence over its development (Kommunernes Landsorganisation 2002b). KL had to revise the decision making process and grant these groups some influence. In 2001, Common Language also became a topic in the local and national media. In the beginning, the initiative was portrayed positively as a tool that enabled better quality and comparability, but soon critical views were being voiced relating to the governance of details , the tyranny of time and the standardization of help provided (Dahlgaard 2001; Ib 2001; Pedersen 2001). In a rather atypical move, KL issued a warning that Common Language should not be misused as a control instrument . Instead, the Association argued that it should be used solely to improve the quality of care and to support the work of home helpers (Thye-Petersen 2001) .

The controversy surrounding Common Language led to it becoming one of the several key issues in the national election campaign of 2001. The leader of the opposition Anders Fogh Rasmussen criticized the incumbent Prime Minister Poul Nyrup Rasmussen on national television and blamed him for the way in which home help was governed by a “tyranny of time”. The issue was further politicized both by the media and DaneAge. This politicization was subsequently used in the electoral campaign to create an image of the government as the one favoring bureaucracy and the tyranny of time, which was contrasted with Fogh Rasmussen’s concern for freedom.

The Association of Local Authorities considered leaving the project to the National Ministry of Social Affairs, but subsequently decided against this move (Kommunernes Landsorganisation 2002c). In the process of developing Common Language, the Association became self-reflective about the system’s uses and abuses (Kommunernes Landsorganisation 2002c; Thye-Pedersen 2001) and even explicitly acknowledged some of the inherent problems, such as the dominance of nursing language and the stereotypical nature of the initiative, particularly in Version II (Kommunernes Landsorganisation 2002b).

In short, Common Language was not initiated by the state; the Ministry of Social Affairs merely provided some initial seed money for supporting the development of the project. Instead, KL developed Common Language as an administrative tool for its members; the local authorities were free to choose, or to reject, the tool. As such, the initiative was formally voluntary. However, KL was indirectly setting a norm for good governance when it advised its members to link Common Language to new information technologies (Kommunernes Landsorganisation 1999, pp. 3–4). Today, the initiative remains part of the software package offered to local authorities, and is therefore difficult to avoid (Nielsen and Andersen 2006, p. 41).

3.2 The Introduction of ‘Free Choice’ in Home Care

In 2003 Denmark witnessed the introduction of Free Choice (frit valg), which can be seen as an example of top down, legislative change (Højlund 2004, 2006; Rostgaard 2006). The legislation allowed citizens greater choice of service provider and extended the choice to the range of services to be provided. The legislation required local authorities to act as purchasers and to contract services not only from public but also from private providers; local authorities also have to define quality standards.

Where did Free Choice come from and why was it introduced? Free choice was not a new instrument in Danish welfare policies. There had been a long tradition for free choice of schools in Denmark, which goes back to the 1824 People’s Schools Act (Folkeskoleloven). This legislation required children to be educated, but did not specify where; thus, education could also be delivered outside of publicly funded schools. Importantly, for the present context, free choice was further extended under the Social Democratic government in the late 1990s, notably in relation to hospitals and home care itself. This was embedded in the notion of putting users at the center stage (Petersen 2008, p. 171), but the basic acceptance of the principle as such also made it difficult for the Social Democrats to turn their back on free choice under the subsequent center-right government (for more information on this point, see Green-Pedersen 2002). The free choice of hospital was introduced in 1993 and allowed patients to choose among public hospitals, not only limited to their own county (Vrangbæk and Østergren 2006). The approach to free choice was pragmatic and cautious, as reflected in a number of significant safeguards and restrictions; hospital choice could therefore best be characterized as ‘extended’ rather than ‘free’. The same applies to so-called ‘flexible home care’ (fleksible hjemmehjælp), introduced in 2000, which allows the older people to have more say about the type of services they receive, although the concrete procedures are heavily prescribed: choice is confined to practical help and does not include choice of provider; the adjustments have to remain within the original time frame and ‘necessary services’ cannot be replaced; the local authority also retains the overall and final responsibility for the delivery of services (Højlund 2004).

Yet, there were two things which were new: first, to connect the principle of free choice to the marketization of welfare services—more specifically, to use free choice as a mechanism to stipulate competition; and second, to substantially extend free choice from a possibility that existed in individual local authorities to a general right across local authorities. On the one hand, this reflects the fact, that the introduction of Free Choice was part of the modernization program of the new conservative centre-right government , which came into power in 2001 . The program emphasized the centrality of freedom and promised to put citizens at the centre in public services, by combining existing responsibilities of society with new responsibilities of the individual (Petersen 2008). The notion of ‘free choice ’ is based on two conditions (Højlund 2004): transparency, in that users need to be able to know precisely what they choose among; and competition, in that free choice only makes sense if there are multiple providers to choose between . The program can be seen as a clear ‘ideological marker’ of the policies which were to come later. On the other hand, the government was also driven by budgetary considerations. As mentioned above, the long term care system is both highly universal and generous and competition was seen as a welcome measure of cost containment (Hansen and Vedung 2005). In this respect, the emergence of a strong budgetary discourse treating care as a commodity was also indicative of the new conservative government’s ideological slant (Dahl 2005).

With the top-down nature of the policy initiative, the policy process leading to the introduction of Free Choice was mainly concentrated within the parliamentary arena (Politiken 2001) . The decision making process stretched from late February 2002, when the government presented the first draft legislation to parliament, to May 2002, when the parliament passed the revised legislation, which came into force in January 2003. During this period, the draft legislation was debated in Parliament twice, when a wide range of institutional interests submitted evidence, and the Parliament’s Select Committee on Social Affairs met four times.

The parliamentary decision making process progressed smoothly, for a number of reasons. The basic principle of free choice remained largely uncontroversial in the political debate , which instead focused more on issues relating to curtailing the autonomy of local authorities as well as to technical issues concerned with implementation. Furthermore, the strong majority held by the new center-right government, the main actor driving the reform was a significant factor contributing to the successful passage of the initiative through the Parliament. Indeed, the phenomenon of a significant majority was rather unusual in modern Danish political history. The coalition included the Liberal Party (Venstre) and the smaller Conservative Party (Konservative) and, thanks to the support of the extreme right wing Danish People’s Party (Dansk Folkeparti), the government had the majority in parliament; old age care policies were also a central policy for the support party.

This may also highlight the fact that the decision making process was in fact not very consensus oriented. The draft legislation was formally negotiated with the Association of Local Authorities as the subcentral level of government and beyond that, only the Social Democratic Party (Socialdemokraterne) succeeded in getting some of its concerns addressed; in contrast, the criticisms from other stakeholders were ignored. However, the Social Democrats also supported the need for greater choice. Both local authorities, as the funders and providers of care services, and the interest groups of older people supported the initiative. Nevertheless, all parties also expressed concerns in relation to the reform, but had little weight in the ensuing decision making process.

For the Social Democratic Party, a major concern was that the legislation required all local authorities to implement free choice and thereby to forcefully intervene in local autonomy (Petersen 2008, p. 200) . The more specific fear was that this would undermine the coordination of services, resulting in unnecessary and costly bureaucracy (Folketingsdebat 2002). Similarly, in relation to the extension of choice of the range of service, the Social Democrats criticized the fact that increased flexibility did not necessarily translate into more time for care (Petersen 2008, p. 199) and that it was precisely the time that was lacking in home care (Folketingsdebat 2002). Significantly, the party succeeded in addressing only some of these concerns in later drafts of the legislation (Jyllands-Posten 2002): the number of private providers per local authority was restricted to between 2 and 5; the implementation of the law was postponed to 1 January 2003; and the legislation was set to come up for review after operating only for a short period of time (in the autumn in 2004).

The concerns of the Association of Local Authorities mirrored those of the Social Democrats. The Association criticized not only the fact that introduction of free choice was to be imposed from the top-down but also about a range of practical implications (Høringssvar 2002; Socialudvalget 2002). This included the costs associated with defining quality standards and price levels as well as with contracting private providers (Ritzau 2002); there were additional concerns about how best to organize the entry of new providers into the sector and also about competition between providers.

For their part, the interest organizations of older people, from large umbrella organization DaneAge, to smaller organizations related to specific diseases such as the Alzheimersforening, saw free choice as part and parcel of strengthening the rights of older people (Kristeligt Dagblad 2001). They specifically demanded greater transparency and a stronger professional orientation of the needs assessment by local authorities; that local authorities would retain ultimate responsibility for ensuring that services are delivered; and the introduction of minimum quality standards. Significantly, however, the evidence submitted by the organizations to the parliamentary subcommittee did not seem to be taken into consideration in the subsequent decision making process.

At the same time, only a few actors rejected the reform outright. This included the Social Liberal Party (Socialistisk Folkeparti) and the left wing Unity Party (Enhedslisten). Indeed, the Unity Party argued that it was paradoxical to use public funds to open private businesses (Petersen 2008, p. 199), while the Social Liberal Party accused the government of introducing privatization through the back door (Folketingsdebat 2002) . However, both were small opposition parties without much influence or power. In addition, the trade union of care workers was outspoken in its opposition to the principle of free choice (Høringssvar 2002) . The organization expressed a view that the provision of home care was best kept in public hands to safeguard the coordination of services and also to maintain a high standard of care. From this perspective, the central problem in homecare was the lack of resources, which put care workers under undue pressure. Compared to the small opposition parties, the trade union was potentially more influential, but the absence of a consensus orientation within the decision making process gave it little leverage.

In short, the centre-right minority government was the clear “winner” of the reform, as it succeeded in pushing through its own agenda. Thus, the small opposition parties and the trade unions, with their outright rejection of the legislation, can arguably be seen as the clear “losers” of the reform. The position of the Social Democratic Party, the Association of Local Authorities and the interest organizations of older people is more ambivalent. They emerged well out of the process in that they supported the basic principle of free choice , though at the same time, their concerns, especially about safeguarding local autonomy, went unheeded. The overall picture of the reform process was similar when it came to the revision of the legislation in 2004/2005, which was concerned with spelling out the procedures for free choice (Petersen 2008).

4 The Substance and the Result of the Reforms: Substantial Change through Restructuring

In 2003, the percentage of older people over 67 receiving home care in Demark was 25 %; by far the highest percentage compared to other European countries (Nielsen and Andersen 2006, p. 57). The picture was similar in relation to the extent of the help provided (Nielsen and Andersen 2006, p. 63). Significantly, the percentage of older people over 65 receiving home care has remained at a high level since the mid-1990s (Nielsen and Andersen 2006). Analyses even suggest that the absolute rise in the number of older people receiving home care since the mid-1990s mostly reflects an increase in coverage, as demographic factors played a minor role (Indenrings- og Sundhedsministeriet 2004, 1). The picture is similar when looking at more recent figures, though no direct comparison is possible because of data problems; between 2005 and 2008, the coverage of home help remained at around 22 % for those aged 65 and more (Danmarks Statistik 2010b). The most recent figure (based on our own calculations) for home help use for people 65 years or older is 20 % (Danmarks Statistik 2011a, b). The typical recipient obtains 4 h per week if he lives in his own home, and approximately 20 h of help if in institutional care or sheltered dwellings. Those in institutional care are typically older and more fragile; approximately every second person above 80 lives in a nursing home (Danmarks Statistik 2008).

The figures above strongly demonstrate that there has been no retrenchment. We therefore need to look more closely at the specific type of restructuring associated with the reforms . The reforms we analyze were aimed at increasing efficiency, quality, and participation, and we thus need to connect any change to specific tools of implementation, such as the type of regulation of service provision. As is typical of Nordic care regimes, citizens are at the centre of such regulatory activities (Højlund 2004, 2006). Yet, regulation has two potentially contradictory sides; it is concerned with both ‘securing’ and ‘extending’ the welfare rights of citizens and, as a consequence, encompasses both measures of control and measures of choice /flexibility . In the following section, we argue that these contradictory building blocks have offered a springboard for gradual, yet substantial change through adaptation , which significantly alters the organization of long term care.

4.1 Common Language

As noted earlier, Common Language was a nonlegislative change that was both initiated and implemented by the Association of Local Authorities, while the Ministry of Social Affairs only provided financial support for some of the initial development of the system. Common Language added a new element (the needs assessor) to existing institutions, namely the home care worker and the policy principle of ‘as long as possible at home’. This has gradually led to a change in the way in which care is delivered, altering the status of the home care worker and, unintentionally, changing the overall goal of care work, as well as the routines of the home care worker, as outlined below.

The effects of Common Language are manifold and complex as implementation varies locally. However, the literature clearly identifies the more general effects of the reform: increasing bureaucratization and less time for concrete care (Nielsen and Andersen 2006, p. 45), the standardization of care for users resulting in less flexibility and responsiveness to individual needs (Petersen and Schmidt 2003) and more transparency for older people and their relatives in relation to the care services provided. These impacts reflect the ambivalence of restructuring through both control and flexibility .

The policy principle of ‘at home as long as possible’ thus remains, but the basis on which it is implemented has been significantly transformed. The standardization encapsulated in Common Language reduces the autonomy of home care workers and, thus, also changes their status. More specifically, their flexibility is limited as they are not able to respond to emergent and unassessed needs. Further, the time available for hands on care is being limited as more time is being used for assessing needs and documenting the care delivered. This is an unintended effect which changes the goal of care work from providing care responsive to the immediate needs of the user to provide care based on standardized packages, and from a focus on delivering care to a focus on documenting care. From the user’s point of view, an equally substantial change has occurred in the direction of greater autonomy ; Common Language has strengthened the position of the user through the introduction of a written assessment specifying what they can expect (and demand). In this sense, Common Language reflects the consumerism inherent in NPM (Glendinning 2008) together with a Nordic institutional context, where the welfare state secures services through rights , rather than exclusively relying on administrative practices and discretion.

4.2 Free Choice

The introduction of Free Choice also represents a gradual yet substantial change through adaptation . More specifically, universalism and localism, the institutional cornerstones of Nordic care regimes, are complemented by the new elements of free choice , standards and centralism; the reform actively sponsors the shift from “citizens” to “consumers”; the redefinition of the role of local authorities as purchasers; and the expansion of the regulative power of the central level. Taken together, this substantially changes the organization of long term care, although the change itself has occurred rather gradually. The possibility of contracting out as part of Free Choice has existed since 2003 , but started out rather slowly. However, a major organizational reform of central–local relations in 2007 meant that local authorities grew in size and as a result the market for long term care services has become more attractive for private providers (Dahl 2008) . Between 2008 and 2010, the number of private providers rose by 34 % (Danmarks Statistik 2011c) and private providers are no longer marginal in the delivery of home help. Now, every third recipient of home help chooses a private provider, although this applies exclusively to practical help such as cleaning; only 4 % of older people receiving personal care choose private providers (Danmarks Statistik 2010b).

Again, in substantive terms, restructuring occurs through both control and flexibility (Højlund 2004, 2006) . On the one hand, to stipulate competition among providers of home care, the local authorities are required to develop a purchasing function, which is separate from the provider function. This means splitting up two formerly integrated functions, which also offers a basis for breaking the local authorities’ monopoly over the provision of services. Yet, since homecare services remain publicly funded, this requires a ‘controlled market entry’. The local authorities define quality standards. In addition, local authorities have to define procedures to ensure the due process of law, in relation to choosing providers, conducting needs assessment and to stepping in when providers fail to deliver. Taken together, this involves ceding the provider monopoly of local authorities and encouraging a more mixed provision of homecare services, while strengthening the organizational controls on the part of local authorities.

The picture is equally ambivalent when it comes to giving individual older people greater choice. On the one hand, users can now freely choose among providers, yet choice is combined with control . The user can only choose between the providers approved by the local authority and in relation to the services which the local authority has allocated as part of the preceding needs assessment. Similarly, users have to some extent the option to choose precisely which services they would like to receive. However, choice is conditional: the choice of individual users has to be approved by a care worker as a professionally sound and practical help cannot be exchanged for personal care tasks if these are not included in the initial needs assessment. Further, if an individual user persistently rejects a task, the local authority may decide to conduct a new needs assessment.

5 Discussion

From the analysis above, reform of long term care policies in Denmark can be seen as a case of substantial change through restructuring rather than retrenchment . The supposed need for restructuring originates from cost containment which is related to the dominant, transnational discourse of NPM. Reforms draw on elements of both control and flexibility, and this is gradually yet substantially changing the organization of long term care. More specifically, the existing institutions of universality, localism, the home helper and the principle of ‘at home as long as possible’ are redefined with the introduction of new elements, including free choice , standards, and centralism. In procedural terms, reforms draw on both legislative and nonlegislative change .

While the system of long term care in Denmark continues to perform favorably in comparison with other countries (Sarasa and Mestres 2007), the number and extent of reforms over the last 20 years have left the system strained. The redefined institutions often exist uneasily and individual elements draw into different directions. In relation to both points, the ambivalent interplay between control/standardization and flexibility /choice is central (Hansen and Vedung 2005; Højlund 2004, 2006); this is associated with a number of tensions, which potentially impact negatively on the effectiveness of the system of long term care. The tensions between control and flexibility become visible especially in two areas: the relations between central and local levels, and the interactions between users and professionals. Free Choice encapsulates a radical form of decentralization ; it focuses on individual service providers rather than merely the local authorities. Yet, the flexibility inherent in microlevel market interactions is severely limited by public control of the emerging market . In addition, central government also regulates the activities of local authorities in an increasingly detailed way (Dahl 2011). Here Common Language adds further mechanisms for control, as it helps to standardize services. The tensions between control and flexibility are equally strong in the interactions between users and professionals. The standardization inherent in Common Language limits the possibility of care workers to exercise their professional judgment and to react to needs in a flexible way. However, the very same standards give more flexibility to the care users, as standards increase the transparency of the care system and thereby help to enable user choice. Nevertheless, user choice is severely ring fenced; it is limited in scope and depends on an initial needs assessment by professionals in the purchaser–provider model.