Research problem

It is safe to say that any kind of user influence over the delivery of welfare services requires at least some basic capabilities on the side of the receiver of the service. Different individuals with different personal characteristics, health status and resources have different capabilities but their ability to be authoritative in the relationship with their service provider is also decided by their level of dependence on the service and the staff that provide it. This is an especially significant issue in elderly care because older people with care needs often have a direct and daily need for the service they are receiving, which makes it more difficult for them to voice complaints or to change service provider. A related and often observed factor in the elders' influence over their services is that they are generally grateful that care is provided to them at all, and therefore they are reluctant to ‘complain’ about the service quality or even to voice their needs and preferences (Möller Reference Möller1996, Reference Möller1999). Some research also suggests that older people in Sweden are especially grateful that the professional and publicly funded care for older people relieves them of the need to ‘burden’ their family with their needs (Berggren and Trägårdh Reference Berggren and Trägårdh2006). Older people with care needs are, in other words, a group for which it is difficult to facilitate user influence, and it is therefore problematic to hand over the choice of service provider to them. The power to decide on a quasi-market is a consumer power, or the power to make a rational, benefit-maximising choice of service provider and to decline services that do not match the consumer's needs and preferences, in favour of another service provider. This specific type of consumer power is what Hirschman (Reference Hirschman1970) refers to as practising ‘exit’, to leave an undesired service. Hirschman differentiates ‘exit’ from ‘voice’, which means to keep such a service while trying to influence it in a desired direction through communication. Both ‘exit’ and ‘voice’ are possible consumer strategies in Swedish home care but the last 20 years of marketisation have set a norm for ‘exit’ as the dominate strategy. How are older people with care needs using ‘exit’ and ‘voice’ and how do they view their role as consumers? Is it at all possible to make consumers of older people with care needs? These are the questions to be addressed in this paper.

Previous research

The issue of how older people experience freedom-of-choice systems in Sweden has been the subject of a number of public evaluations and reports. The Swedish National Board on Health and Welfare (Socialstyrelsen) has performed three major studies in the area in recent years. One of them combines official statistics with interviews with older people to determine how they choose their home care providers and how they experience the choice situation. The general conclusion from the report is that the older people with care needs show little resemblance to critical consumers. They lack useful knowledge and information and they rely instead on various informal sources of information, such as neighbours and friends (Socialstyrelsen 2012a ). The choice is difficult for older people to understand because they do not see the value in choosing between anonymous service providers; they are more interested in which physical persons will be performing the home care. The choice is made even more difficult because it is carried out under less than perfect circumstances, as in cases of sudden illness, as is, of course, often the case when the care need arises (Socialstyrelsen 2012a : 29–32). The report further suggests that older people with care needs have difficulty switching from one home care provider to another due to loyalty and dependence, which by and large removes any positive effect on quality through competition (Socialstyrelsen 2012a : 7). A second study is a meta-study that arrives at much the same conclusions. This study compares the hypothetical, active consumer role with the actual, much more passive role that the older people with care needs take in the choice situation due to failing abilities and lack of interest (Socialstyrelsen 2011: 12–13). Another result, hardly unexpected, was that people with high income and good education are both less troubled by the choice process and more satisfied with it (Socialstyrelsen 2011). A third report shows, using focus groups, that time constraints and other difficult circumstances during the choice situation affect the decision negatively and the respondents find it difficult to switch home care provider due to loyalty and dependence (Socialstyrelsen 2012b : 7–14).

The Swedish Association of Local Authorities and Regions (Sveriges Kommuner och Landsting) shows in its study that only 4 per cent of older people switch from one home care provider to another, a result that is explained once more by loyalty and dependence but also lack of energy and knowledge (Sveriges Kommuner och Landsting 2010). The Swedish Competition Authority (Konkurrensverket) approached the issue of freedom of choice in the home care sector a little differently in their evaluation from 2013. Their focus is, naturally, on competition between different providers on the quasi-market for home care but their results also correspond clearly with those from previously mentioned reports. Older people with care needs are found to be passive consumers that lack understanding of why they are choosing, as well as the difference, if any, between the alternatives, and they also seem to have difficulty switching providers due to emotional bonds to the staff (Konkurrensverket 2013). Hjalmarsson and Wånell (Reference Hjalmarsson and Wånell2013) write in their report to the Stockholm County Administrative Board's Centre for the Elderly (Stockholms läns äldrecentrum) that older people think of the provider as the individual members of the staff, that they choose a provider under less than ideal circumstances and that they have difficulty telling the difference between the different home care providers. A final set of results is provided by the Swedish National Pensioners' Association (Pensionärernas riksorganisation) which, in a survey of their members, asked a series of questions about the choice between different home care providers. As many as 71 per cent replied that the content of the care, in other words what type of services they would receive and how, was the most important aspect of the choice, while only 20 per cent thought that the choice between different providers was the most important aspect (Pensionärernas Riksorganisation 2012). Many in the latter category might also have been referring to the distinction between the general categories of public and private providers, not all the individual service providers. The pensioners' association also concludes that there is too much uncertainty surrounding the choice situation and that their members generally seem to be unknowledgeable about the choice and their part in it (Pensionärernas Riksorganisation 2012).

The results from these reports and evaluations from the government agencies and interest organisations are also reflected in some of the, rather few, research publications on the topic. Meinow, Parker and Thorslund (Reference Meinow, Parker and Thorslund2011) show that the freedom of choice in Swedish elderly care is largely illusionary as about 90 per cent of the older people have reduced capability as consumers at the time of the choice situation. They also reflect on the fact that the older people with the greatest care needs have the greatest interest in influencing their home care and in picking the right one, but also the worst circumstances when doing so (Meinow, Parker and Thorslund Reference Meinow, Parker and Thorslund2011). Winblad and Blomqvist (Reference Winblad, Blomqvist, Rönnberg, Strandberg, Wihlborg and Winblad2013) similarly argue that the information about the choice alternatives, especially under these circumstances, is difficult to interpret and is thus disregarded as irrelevant. The lack of understanding of the freedom-of-choice system leads older people and their close ones to adopt various coping strategies. These strategies typically involve leaving out large parts of the available information and taking ‘short cuts’ to simplify the choice, such as asking friends and neighbours which providers they are using (Winblad and Blomqvist Reference Winblad, Blomqvist, Rönnberg, Strandberg, Wihlborg and Winblad2013: 66–67). One of the main reasons for this behaviour could be, as suggested by Fotaki (Reference Fotaki2009), that the quasi-market expects all consumers to assume the role as consumer but older people and other receivers of health and social care simply do not think of themselves in this way. Fotaki and Boyd (Reference Fotaki and Boyd2005) show that difficulties like these among older people in both Sweden and the United Kingdom (UK) are to some extent written off by policy makers as an acceptable price to pay to achieve the normatively more important objective of individualising the care services.

Stolt, Blomqvist and Winblad (Reference Stolt, Blomqvist and Winblad2010) can show that the choice between different care providers is increasingly important in Sweden because of the growing difference in quality between providers that has followed the large-scale privatisation of the home care sector in the last 20 years. They can also show that the freedom-of-choice system has led the care providers to give priority to their more visible service functions, such as more specialised or personalised services, at the expense of more structural quality aspects of the home care such as number of staff, apparently as a means to be competitive on the quasi-market (Stolt, Blomqvist and Winblad Reference Stolt, Blomqvist and Winblad2010). These adverse effects of the quasi-markets should be considered alongside some of the effects of competitive tendering, which has been the other marketisation strategy in Swedish elderly care. Competitive tendering has in large part been replaced by the freedom-of-choice systems, and the needs and preferences of older people has become a factor in the competition. Erlandsson et al. (Reference Erlandsson, Storm, Stranz, Szebehely, Trydegård, Meagher and Szebehely2013: 29) can show that competitive tendering is ‘characterised by price competition rather than quality competition’ and that the price usually wins over quality even in tendering processes where both criteria are included. A similar conclusion is reached by Szebehely (Reference Szebehely and Hartman2011), who shows that competitive tendering did lead to some reductions in cost early on in the marketisation process but only temporarily and at the expense of lower quality. She also concludes, however, that the motives behind the marketisation at the local level actually are ‘political rather than economic’, or in other words, that the marketisation reforms are primarily ideologically motivated (Szebehely Reference Szebehely and Hartman2011: 236). Andersson and Kvist (Reference Andersson and Kvist2014) and Edebalk and Svensson (Reference Edebalk and Svensson2005) agree that there are neo-liberal, ideological reasons behind the establishment of freedom-of-choice systems. The latter can, however, also show that young people with disabilities, unlike older people, have been part of a political movement for greater influence and choice in home services (Edebalk and Svensson Reference Edebalk and Svensson2005). The older people appear to be less interested or less able to get behind freedom of choice as a political issue.

The overall impression given by the previous research on freedom-of-choice systems in Swedish elderly care is that older people with care needs are seriously limited in their role as consumers. The evidence for this already seems convincing and the contribution of this paper is therefore not only to further test this conclusion but to clarify it by combining the quantitative approach with a qualitative one.

Methods and data

The results presented in this paper come from a large data-set collected for a three-year research project conducted in three major Swedish cities with large and well-established freedom-of-choice systems in home care for older people. The empirical parts used for this paper come primarily from a telephone interview study conducted in 2012 in which 324 older people with home care services in these three cities were asked about their choice of care provider. The interviews were structured according to a question sheet and took anywhere from 15 minutes to an hour, depending on how capable the respondents were of answering our questions and on how many additional comments they wanted to add in the comment fields connected to many of the questions. We found this technique for data collection useful because it offered a balance between structure and flexibility. Everyone received the same questions but the helpful telephone interviewers could guide the respondents through the question sheet, using the same set of instructions, and the respondents had plenty of room to freely add descriptions of their personal experience. The random selection was drawn from lists of persons above the age of 75 with home care, lists that were requested and received from the three local authorities under a written agreement that the information was to be used only for research purposes. All calls that were made, where we actually made contact with the respondents, were included in the selection, while the calls where we had no reply after three attempts were not included. Since the only information we know about the people we were unable to contact is their names, we cannot conduct an analysis of why they did not respond. We reached a total of 364 older people and conducted 324 interviews which gave a response rate of 89 per cent.

The respondents had in common that they were capable of participating in a sometimes rather long and possibly tiresome interview, which tells us that they were at least in reasonably good health. The respondent's own, subjective, opinion of their health status was more or less evenly distributed between the three categories ‘good’, ‘neither good nor bad’ and ‘bad’. This refers to the health status of the respondents at the time of the interview; we did not ask about their health status at the time of the choice. A majority of the respondents, 63.3 per cent, were women and this reflects the gender distribution among home care users fairly well, since more women than men have home care in Sweden.Footnote ¹ A majority of 85.8 per cent of the respondents were born in Sweden, which is reasonably representative for this age group. However, most of the others had been living in Sweden for a very long time. For instance, several were World War Two refugees and almost all of them came from other parts of Europe, especially the Nordic countries and especially Finland. The most common country of birth in the very small category of people born outside Europe was the United States of America, while the representation from Third World countries was minimal. The respondents were relatively well educated for their age group, they had a significant interest in local political issues, and slightly more of them said that they leaned to the right than to the left ideologically. This group of respondents seemed, all in all, to be a relatively ‘strong’ group among those receiving home care in that they were well integrated in Swedish society, educated, interested and capable. Their ability as consumers was, therefore, likely to be better than average. The bias in this relatively small, random sample could be seen as a limitation but this ‘strong’ group can also add valuable insight. The previous research presented above has already established that older people with care needs do not act as good consumers, but is this still true if one focuses exclusively on a capable group like this one? This sample could, in other words, be seen as a group where older people were more likely to act as consumers, a group for which the freedom of choice was likely to work relatively well.

The interview questions concerned the respondents' experience of choosing a home care provider, how they got their information, what made them decide and how satisfied they were. The question sheet was constructed based on results from two focus groups in one of the three cities and it was tested on a small panel of older people before being put to use. The interviews were coded into statistical material. They were not recorded but comments were written down in full text. All of the interaction between interviewer and interviewee was, in other words, not documented, since much of it concerned explanations of what turned out to be a rather abstract topic for the respondents. In order to still, to some extent, add this aspect of the interviews to the interpretation of them, we also performed a lengthy group interview with all five interviewers. The purpose of the following empirical segments is to present parts of this study in a mainly descriptive way. The presentation is meant to give an overall impression of the consumer behaviour of the respondents, which is why not all the results for every alternative reply are presented in the text. The descriptive statistics will be mixed with comments from the interviews in a qualitative effort to demonstrate better how older people with care needs experience the choice of care provider.

The theoretical frame of the analysis

Freedom-of-choice systems in the form of quasi-markets are often criticised for allegedly treating the service market, particularly the welfare service market, as if it worked according to the same logic as a regular market for goods, even though welfare services and goods are two distinctly different commodities. Considering this objection, it is both a challenge and an opportunity to frame the analysis according to Hirschman's theory on exit, voice and loyalty. The challenge is, of course, that this theory has been used for both goods and services. The opportunity lies in the fact that Hirschman goes beyond the narrower view of the market as being defined strictly by rational, benefit-maximising ‘economic men’. Instead, he considers the possibility of other consumer strategies, which opens up the possibility of a more nuanced analysis of service markets. Using his own terminology one could say that he took the conventional understanding of the market as based on the consumer strategy ‘exit’ and added ‘voice’ and loyalty’ to the list of options, which in many ways was a criticism of the neo-liberal market ideas made popular by Milton Friedman, Friedrich von Hayek and others at the time. Exit means to leave a good or a service that one does not want to consume and thereby reduce demand and thus the price for that particular good or service. Price is, however, not a factor on a quasi-market like the one for the Swedish home care since price levels are standardised and the costs are heavily subsidised by the local authority, but exit is important for quality control in the market model. Hirschman compares the relation between exit and quality to that between supply and demand in classical economic theory; one varies as a direct effect of variation in the other, due to competition in quality (Hirschman Reference Hirschman1970: 22–23). Exit applied to our issue would be when the older person leaves one service provider for another on account of the quality being too low, according to his or her personal preferences. Voice is described by Hirschman as ‘any attempt at all to change, rather than escape from, an objectionable state of affairs’ (1970: 30). Voice is, in other words, an attempt to influence and alter the provided good or service that in most cases is either used when the service receiver finds it the most efficient strategy, or conversely when exit is somehow difficult. Voice applied to our example would be when an older person chooses to bring up complaints or other forms of feedback with the provider in order to change the manner in which the home care is provided. This could be done both in direct contact with the care-giver or through contact with management. Loyalty is a strategy by which the user maintains use of a good or a service in spite of it being of low quality, which is, of course, a strategy with which Hirschman contradicts the neo-liberal expectations of the consumer as rational and benefit-maximising. The relation between loyalty, exit and voice is that loyalty reduces the use of exit but also increase the use of voice, Hirschman (Reference Hirschman1970: 78) explains that voice becomes a more attractive strategy when exit is blocked by loyalty. Loyalty applied to our example would be when an older person has the means and at least some reason to change provider but for whatever reason does not. It might be hard to imagine such a case of loyalty but it is possible that a service receiver might, for instance, favour the public, municipal home care over one of the many private alternatives even though the quality is less than perfect. The choice between the three strategies is dependent on the abilities or perceived abilities of the individual service receiver to utilise them, and there are usually a wide range of factors that affect what one is able to do. Several of these factors are related to age, for example the ability to gather information, to express concern in a care relation and to evaluate quality of services. The question of how older people with care needs act as consumers of home care is therefore answered by a clarification of their ability and willingness to use voice or exit.

Further understanding of this question can be gained by viewing this particular consumer role as one defined by two opposing logics; the logic of choice and the logic of care. Mol (Reference Mol2008) argues that the core problem with offering care services according to a choice or market logic is that markets require the product to be clearly defined. It must, as she says ‘have a beginning and an end’ (Mol Reference Mol2008: 23). The older person must view the offered service in itself and decide if it is an objectively good one or one that he or she desires. The care logic gives the older person an opportunity to interact in a process with a health professional in which the service can be designed and redesigned according to actual care needs, a process with substantial elements of what Hirschman refers to as ‘voice’ (Mol Reference Mol2008). Care is an open-ended process while choice is a one-off transaction in which the customer is ‘always right’ and free to regulate demand but where the service is an off-the-shelf product that might or might not work as desired, once tried. Tronto (Reference Tronto1993) makes a similar differentiation between choice based on preferences and interests and care based on needs and dependence. ‘Humans are not fully autonomous but must be understood in a condition of interdependence’, she argues, while describing care services, much like Mol (Reference Mol2008), as an ongoing process (Tronto Reference Tronto1993). Markets assume consumers to be independent, rational and benefit-maximising, which is, in other words, contrary to the care logic. This means that while Hirschman's theory gives a broadened understanding of the workings of the quasi-market, the care logic theories offer opposition and an alternative to the idea of care service quasi-markets as a whole.

Familiarity with the choice among older people

Our study of older people receiving home care shows that, along with difficulties understanding the specifics about the choice situation, the respondents also have difficulty grasping the general idea of choosing a provider of home care; 12.7 per cent of the respondents did not know that they could choose, although in most cases they had. Some of these might have had the same provider since before the freedom-of-choice system was introduced, but a fair number of them must have gone through the choice process unknowingly. This is also supported by the fact that 23.8 per cent of them claimed that the municipal authority had made the choice for them. Some of them can, as mentioned, refer to having had the same provider for a long time, going back to when the civil servant at the municipality did indeed pick a provider for them (usually the municipality's own home care) but this cannot explain why almost one in four of the respondents held this view. Another 7.7 per cent stated that people close to them had chosen for them and less than half of them, 46.3 per cent, claimed that they had chosen all on their own. Considering these numbers, it is important to remember that these respondents represent an ‘elite’ among the older people; the older citizens that we would expect to have the fewest problems with choosing.

The respondents were asked a series of questions designed to determine how knowledgeable they were about the choice situation. One of these questions concerned the number of alternatives that the respondents perceived there to be. The most frequent answer by far was ‘Don't know’, which is what 41.4 per cent stated in the telephone interviews. Among the ones that did make an estimate, 23.5 per cent stated ‘10 or more’, which in all three cities was the correct answer. The three cities in which the study was performed offer between 15 and over 60 different home care providers, even though it is difficult to actually state an exact number since providers come and go and some of the registered home care providers do not seem to be operational. Almost as many of the respondents, 20.4 per cent, thought that the number of alternatives was ‘2 to 9’ and 14.2 per cent believed there was only one alternative. To know the exact number of available alternatives is, of course, not that easy and also not necessary to make a choice, but the lack of knowledge in this respect might mirror a general lack of knowledge and ability, as well as interest, in the choice of provider. The results certainly do not portray the respondents as active consumers, and neither does their stated ability to distinguish between different services and their varying qualities. The respondents were asked to estimate the difference between the available service providers, using a five-level ordinal scale ranging from ‘very small’ to ‘very large’. As many as 44.4 per cent of the respondents stated that they did not know how big a difference there was between the different alternatives, and among the ones that did have an opinion, 21.3 per cent stated that the difference was ‘very small’. Only 5.2 per cent recognised ‘some difference’ and even fewer, 3.7 per cent, of the respondents thought that the difference was ‘very large’. In conclusion, the respondent's familiarity with the choice situation was far from good and it can be asserted that if the choice of home care provider is a source of influence then it is one that is little known by the older people themselves.

How the choice was conducted

The choice of home care provider seemed to be made with a high degree of chance. This is not least apparent from the type of sources of information that the older people used. The respondents were given a wide range of possible sources of information to consider in the telephone interviews and they could pick just one or any number of them as sources they used. The two most commonly mentioned sources of information were printed brochures (30.2 per cent) and spoken information from civil servants (25.3 per cent). Only 15.4 per cent stated ‘relatives, friends and acquaintances’ and 17.9 per cent claimed that they had not used any source of information at all. All the respondents had potential access to information since all three cities provide similar catalogues of providers and information online, and there is certainly no lack of information from the home care providers themselves. The fact that the respondents still claimed that they had little or no information therefore seems to indicate that they were too passive to access it, not that there is an actual lack of information. The respondents could add sources of information not covered by our questionnaire, and 16.7 per cent of the respondents chose to do so. The list of additional sources of information was rather long and varied. The single most common reply was ‘at the hospital’, which is interesting since it tells us something about the conditions under which the choice is sometimes made. The care need sometimes evolves over a long enough period of time to allow older people to make a decision after careful consideration, but as seen from the example with the respondent in hospital the need can arise urgently and unexpectedly. These older people must sometimes struggle both with poor health conditions and a narrow time frame when choosing a home care provider.

The older people were also asked about their reasons for choosing the way they did. The most common response among the alternatives in the questionnaire was that the geographic location of the home care provider decided the choice, but only 18.5 per cent stated this. Only 13.9 per cent stated the second most common response, that the chosen home care provider had a good reputation. Instead 39 per cent of the respondents decided to reply ‘other reason’, and these other reasons were very mixed. Many of them were simply variations on randomness; the respondents stated that they ‘just picked one’, ‘made a guess’, ‘just put my finger in the middle of the list’, ‘picked the top one on the list’, and so on. Other strategies were slightly less random without being very informed. The respondents described spontaneous recommendations from friends, neighbours and relatives as well as advertisements and chance interactions with the home care providers of people in their proximity. The sources of information and the reasons for choosing provide further evidence that the choice situation does not in any way resemble that of an active and informed consumer.

Influence over home care services

The shortcomings in the familiarity with the choice situation and the manner in which the choice was conducted were matched by equally grave shortcomings in the respondent's understanding of why they were choosing a home care provider. The data suggest that older people want to be able to influence their services but they do not regard the choice of provider as a means to do so. Several comments from the interviews revealed a desire to alter the home care but these were limited exclusively to two finely defined themes. The respondents wanted to influence: (a) the content and to a lesser degree the amount of care, and (b) the specific member of staff that carried out the care. The latter wish was simply a matter of having people one likes around, people that can be trusted and who are not replaced all the time; a wish for continuity is a recurring theme. There was, of course, a link between the staff and the choice of provider but this link was indirect and not clear to the older people. Some comments were appreciative of the opportunity to choose the home care provider but mainly because that allows you to exit a service that is not living up to your expectations. Twelve per cent of the respondents said that had changed their home care provider after their first choice but most of these did so because the provider had closed down their business or simply because the respondents themselves had moved to a different area. Less than half of these 12 per cent switched home care provider as a reaction to a poor service. One reason why so few older people changed their home care was no doubt that they were in general very satisfied with the care that they were receiving; only 3.4 per cent claimed that they were dissatisfied or very dissatisfied with their home care. One must assume that this high level of satisfaction was due to genuinely high quality levels in Swedish home care, but we also know from previous research that older people express deep gratitude that they are taken care of at all. Another finding from previous research was that older people also refrained from changing home care service because they were locked into a social relation with the home care staff on whose services they were dependent.

The gratitude is perhaps also reflected in the fact that so few claimed to be dissatisfied but that as many as 37 per cent still claimed to have had complaints or opinions about the home care. Another 10.2 per cent stated that they had an opinion without communicating it to the provider or anyone else. In other words, things were not perfect but the respondents were satisfied anyway. The fact that as many as 37 per cent of the respondents used voice as a strategy is difficult to interpret. Their opinions might very well concern both major and minor issues, most of which can be directed straight to the staff in their homes. That 50.9 per cent had no opinions whatsoever, not even minor ones, and that 10.2 per cent had opinions without bringing them up suggests that voice is not a very significant source of influence even if it is much more accessible than exit. This conclusion is supported by the fact that only 60 per cent of the 37 per cent who communicated opinions thought that their concerns were considered by the provider. Combining these numbers shows that just one in five of the respondents had opinions that were communicated and dealt with.

Analysis

The results from our study are well in line with those from previous studies in the area in that they highlight some of the many problems that older people have when participating in a freedom-of-choice system. The exit function is almost entirely non-functioning in the quasi-market for home care services simply because older people with care needs are not acting like the consumers they are supposed to be. They are unfamiliar with the choice, they do not know how to choose and they do not understand why they should choose. The reason for this lack of understanding is a shortage of ability and interest, but also the fact that the choice is often conducted under less than perfect circumstances, during periods of illness, and with time restraints. Exit seems to be a poor source of influence for the older people since they very rarely switch from one home care provider to another. This may be due to poor circumstances and lack of abilities, but gratitude and personal bonds to the staff are also effective in blocking the possibility of exit. Hirschman states that blocked exit should lead to an activation of voice or, in other words, if older people are unable to exit their home care service they should instead voice their concerns in order to alter the care they are receiving. There are signs of this occurring but not to the extent that one would expect. The reasons for the lack of voice are the same as for the lack of exit; older people have limited capabilities and they are generally satisfied and grateful. The link between exit and voice in Hirschman's description is associated with the concept of loyalty; when loyalty stops consumers from exiting they instead resort to voice. Loyalty is, however, a poor description of the older people's behaviour in the freedom-of-choice system. They have the formal possibility to exit their service but they do not have the actual possibility, which means they are locked in rather than loyal. The fact that they are also not fully using their voice suggests that they are not so much loyal as indifferent. The free choice of home care provider is abstract and foreign to the older people with care needs; they do not act as consumers in a free market because they simply do not see themselves as such and they also do not realise that they are expected to be consumers. The home care they are receiving is a key component of their everyday life but free choice in itself is not important enough for them. A reason for this is that the general quality of Swedish home care services is rather high, and older people are justifiably satisfied with their services, but this fact still undermines the rationale of the freedom-of-choice system.