CHOICE AND LONG-TERM CARE IN OECD COUNTRIES: CARE OUTCOMES, EMPLOYMENT AND FISCAL SUSTAINABILITY¹ Open Access

As the number of older people in need of long-term care increases, efforts to support older people remaining in their home are intensified in most OECD countries. In this context, there is a movement towards allowing more individual choice for older people receiving publicly funded long-term care at home. Having more flexibility in terms of how to receive care can increase the older person's self-determination and that of his/her informal caregivers. Having a choice among alternative care providers can empower older people as consumers and may help strengthen the role of households in the care-management process. The main aims of this article are (i) to categorise and analyse different types of arrangements allowing home care users more choice and map the prevalence of such arrangements in OECD countries, (ii) to give an overview of some country-specific outcomes in terms of care quality for care recipients, and (iii) to discuss some implications for employment and fiscal sustainability.

As the number of older people in need of long-term care increases in OECD countries, home care continues to be the predominant – and preferred – care setting for the vast majority of households. To achieve the goal of ageing in place, a complex mix of services of acute health care, rehabilitation, long-term care, and social services is often needed, and – above a certain level of dependency – the support of a main informal carer usually becomes indispensable. Over the last 10 years, care systems in many countries have made considerable progress in moving away from the restricted choice of either receiving largely unsupported care at home from a relative or receiving formal services in a nursing home.

Against this background, the main aims of this article are (i) to categorise and analyse different types of arrangements allowing home care users more choice and map the prevalence of such arrangements in OECD countries, (ii) to give an overview of some country-specific outcomes in terms of flexibility, care quality, satisfaction and conditions for caregivers, and (iii) to discuss some of the implications for employment and fiscal sustainability.

A complete analysis of choice and flexibility in long-term care would include the provision of care both in long-term care institutions and at home and arrangements facilitating choice among alternative agencies providing formal care as well as consumer-directed employment of a personal care assistant and support of informal care. In this article, only care provided to people living at home is covered and the emphasis is put on consumer-directed employment of a personal care assistant and the support of informal community-based care.

This article covers 12 countries, namely Australia, Austria, Canada, Germany, Ireland, Japan, Luxembourg, the Netherlands, Norway, Sweden, United Kingdom and the United States. This is a subset of the countries covered in the wider OECD long-term care report (OECD 2005) and includes those countries that have experience with arrangements allowing users more choice and flexibility with regard to the way care is provided, and for which sufficient information was available.

The different ways in which countries seek to increase flexibility and choice in long-term care depend, among other things, on the weight they put on formal care provision versus informal care. This, in turn, partly reflects cultural factors, and arrangements for care can be seen to represent different family models (Millar and Warman 1995; Rostgaard 2004). In general, countries with above-average public funding for long-term care tend to rely more on formal care than others, perhaps because this is found in countries with a social-democratic welfare model, where emphasis is on public provision, but this is not the full story. As illustrated by the stylised picture of long-term care systems in Figure 1, panel A, countries can broadly be subdivided into five groups:

• •

  In Korea, Spain and other Southern European countries long-term care is, to a large extent, provided informally within families, so that although nursing homes and other residential care institutions are predominantly funded publicly, overall long-term care provision is based on private resources to a greater extent than in other OECD countries.²

• •

  Austria, Luxembourg, Germany, Ireland, the United Kingdom and Australia have considerable levels of public funding for long-term care compared with countries such as Korea and Spain. A significant share of this is in the form of payments in support of informal home care either by granting these allowances directly to care recipients, allowing them to choose whether to share them with a carer informally or to purchase other forms of care (Austria, Germany and Luxembourg), or by granting long-term care allowances to relatives providing care to older people (Australia, Ireland, the United Kingdom).

• •

  In Japan, informal care within families plays a large role as well, but it is not supported financially by public programmes. Mandatory long-term care insurance now provides for substantial formal services as care alternatives.

• •

  In the United States and Canada, informal care plays some role, but is not supported much financially by public programmes. However, the increasing orientation towards consumer direction in care has brought arrangements where in some cases relatives can be employed.

• •

  The Netherlands, Norway and Sweden have considerable levels of public funding, but channel most of this through provision of formal services for both home care and institutional care.

Countries may increase flexibility by introducing choice either via support for informal care and/or via choice among providers of formal care. As illustrated by Figure 1, panel B, the one does not necessitate the other, and often reflects the path dependency of the particular social policy regime of the country. For instance, in Sweden, where the scope of choice for people receiving formal care is typically limited and where payments for informal care play a limited role, the introduction of choice by some local authorities currently implies moving horizontally (to the right) in the chart towards choice among a limited set of approved providers. For a country such as Sweden, this move towards more market provision represents some divergence from the traditional Nordic model of basically a publicly provided care system. Germany, with a much more recent public involvement in the organisation of provision and financing of care for the elderly, did not have the same public tradition to consider. Here, the development of the care system placed Germany vertically (upwards) in this chart with the introduction of the Cash Allowance for Care (Pflegegeld), and also horizontally (to the right) as the long-term care insurance reform in 1995 implied more choice among formal care providers. An important point is that improving flexibility of care via choice does not necessarily imply privatisation of funding. Whether countries maintain (or expand) public funding or coverage of long-term care costs can be a separate issue from whether they channel it through formal long-term care services.

With respect to choice among agencies providing home care or among institutions, there are very fundamental differences between countries. On the one side, the Nordic countries have build up provision of formal home care services almost exclusively via public agencies, and therefore introducing choice among alternative public and private providers comes in as an ‘additional’ element which has been a matter of policy debates over the last decade or so. On the other side, the long-term care systems in most other countries are ‘born’ with more choice among providers when older people get publicly funded formal care services. The United States has considerable levels of public funding for long-term care but it is targeted at people with low income, and the supply of care services and the market for nursing homes and other providers of home care and assisted living are therefore in part shaped by the demand from older people paying privately. Older people receiving care supported by Medicaid therefore often have a choice among different agencies. In countries such as Germany, where public funding for long-term care (as for health care) is based on social insurance funds, the purchaser–provider split (which is talked about as a policy innovation in Nordic countries) is given from the outset. This construction more easily facilitates choice among providers of formal care services as there is not one incumbent agency. That the systems are very different with respect to the extent to which they involve public or private providers of formal care services can be exemplified by the situation for institutional care (see Figure 2).

A number of fairly different arrangements are used in OECD countries to allow more choice for people receiving long-term care at home. Although they differ in many respects they can usefully be subdivided into three main groups, as shown in Table 1, panels A, B and C.³

As an alternative to provision of formal home care via a single designated agency, different arrangements can strengthen the position of older people as active consumers, making their individual demands clearer. Older people needing care can be given a personal budget to purchase care from alternative competing agencies, or they and their families can be allowed to employ a personal care assistant directly and thereby be able to hire/fire, schedule, and supervise – in other words direct – care provision.

The Personal Budget scheme in the Netherlands is the largest of the schemes in this category. In 2003, 0.8 per cent of the population aged 65 or over received home care via a personal budget – compared with 7.4 per cent receiving some form of formal care at home (see columns to the right in Table 1, panel A).

In all of the programmes listed in Table 1, panel A, care assistants have a formal employment contract, even if they are relatives of the person receiving care. Therefore, care assistants are typically paid for a specified number of hours. They can provide care to several people at the same time and their wage does not depend on what income they have from other sources, as could an income-support payment. The level of care needs covered by these programmes varies typically from 3 to 10 hours per week for the Norwegian Care Wage and up to a maximum of 66 hours per week for the Californian In-home Supportive Services programme. In practice, however, relatives and friends employed as care assistants often provide care during more hours than they are paid for.

Personal budgets may also allow the person to combine care with purchase of physical aids such as a special bed or chair and can generally support very flexible solutions. This is one important aspect where the Cash & Counseling programme differs from other consumer-directed home care programmes in the United States.⁴ In the Netherlands, a limited amount can be made directly available to the person who does not need to account for how it is spent, and from this amount some informal help may also be compensated.⁵

The table includes information on the level of payments. Under this category of schemes the level of payments or public spending is roughly similar to what it would be for agency-based formal home care, as the care assistants employed by consumers in many schemes get an hourly wage similar to that of care assistants employed by agencies. For the US Cash & Counseling programme, the table shows the average monthly payment levels per person ranging from $400 in Arkansas to $1400 in New Jersey. To compare across countries with different currencies, this table shows the monthly payment levels also as a share of average private consumption per person. The $400 corresponds to 19.6 per cent of what an average person in the United States spends for private consumption per month (see Table 1, panel A).⁶

Some countries give older people needing care the option of receiving cash to finance (part) of their expenditure on long-term care (Table 1, panel B). In Germany, people receiving support from long-term care insurance can choose between a budget for care services in-kind and the Cash Allowance for Care, and in Luxembourg those entitled to home care support under long-term care insurance may take part of this support as a cash benefit rather than services or everything as cash if found eligible for seven hours weekly or less. In Austria, all public support for long-term care to people living at home is given as cash. In practice, a substantial part of these cash payments is used to compensate informal caregivers or simply enters the household budget when care is provided by co-habiting relatives.

While there are no explicit restrictions on how the German Cash Allowance for Care is spent, the older person and their relatives are nevertheless obliged to acquire sufficient care. The health condition and wellbeing of recipients is reviewed every three or six months by an agency. If the older person is found to be receiving insufficient care in light of her/his needs then the authorities must find some provision of care services in-kind, in which case the cash allowance is withdrawn. For the UK Attendance Allowance, the requirements are looser as there are no restrictions on how the money can be spent, and it can be received even if the person does not obtain care from anyone. Provided the older person is at home (i.e., not in hospital or an institution) and has had the need for care for at least six months, she/he is eligible for the Attendance Allowance which is paid independently of the person's income and assets as for the other schemes listed in panels A and B.

Compared to schemes listed in panel A, most of the schemes listed in panel B are quite extensive. The Austrian and German schemes pay Cash Allowance for Care to 14.8 per cent and 5.7 per cent, respectively, of the population aged 65 or over. The payment levels vary considerably depending on need, with averages of 35 per cent and 27 per cent of private consumption per capita. When these payments are passed on to informal caregivers, they are in practice rarely taxed as income for the caregiver, as indeed the relation remains informal (often being within the family). The UK Attendance Allowance provides support to as many as the Austrian scheme but with a much lower maximum benefit paid.

A number of OECD countries have various forms of payments to informal caregivers in order partly to compensate for the loss of income while providing care, thereby enabling the caregiver to reduce other work activities (Table 1, panel C). Some of these allowances or cash benefits pay around half of the average private consumption per person while the Swedish temporary Care Leave pays more than that. Other allowances, however, give only a limited supplement to the income of a household, such as the Japanese Allowance for Families Caring for Elderlypaying an amount equal to 5 per cent of average private consumption per person.⁷

The essential difference between consumer or client employment of a care assistant (reviewed in Table 1, panel A) is that income support is not meant to fully compensate caregivers for the value of their work. Rather, they are meant to sustain a minimum level of income for people who are unable to have a normal full-time job as a result of providing care for someone who is near to them such as a relative or close friend. Therefore, some schemes are only available for low-income carers, namely the Australian Carer Payment, the Irish Carer's Allowance, the Japanese Allowance for Families Caring for an Elderly and the UK Carer's Allowance. To be eligible for support, the income and asset criteria may also take into account the income and assets of the carer's spouse or partner and thereby exclude carers from middle- or high-income families. Also, payments from these schemes are often combined with other forms of public income support.⁸ Other schemes are built into labour market institutions and provide an option for temporary leave from work. As such they are available to people at all income levels. The Canadian Compassionate Care Benefit and the Swedish Care Leave replace 55 per cent and 80 per cent of the caregiver's previous or normal employment income up to a maximum, while the Irish Carer's Allowance pays the same amount to all recipients.

Finally, some schemes are meant to reward or recognise the efforts of all informal caregivers, as well as those caring for people with less severe needs (e.g., the Australian Carer Allowance). Eligibility is therefore conditional only on the provision of care, not on the income or assets of caregivers. Giving the limited amount of the Carer Allowance to a wide group of people living with, and caring for, an older person at home – equal in number to 4.0 per cent of the population aged 65 or more – creates an extra incentive for the family not to seek institutionalisation that would entail higher public expenditure.

In Japan, local authorities may decide whether they wish to use a central government grant to support informal care, including the Allowance for Families Caring for an Elderly Person. This allowance has been introduced very recently and is expected to play a minor role. The major policy direction is to expand in-kind benefits through Long-term Care Insurance (see chap. 5 in OECD 2005).

In the following, Section 4.1 discusses possible outcomes in terms of flexibility, care quality, satisfaction among people receiving long-term care and conditions for caregivers, whereas Section 4.2 addresses some of the implications for employment and fiscal sustainability. For a detailed assessment, including more quantitative evidence, see Lundsgaard (2005).

When people needing care have several options, their actual choices can give an indication of what works well for them as an individual and what best meets their needs, given the care policy context. The explicit choice between services in-kind and a cash payment or a combination of the two which is open to people receiving support from the German long-term care insurance is an interesting case, as older people have a right to choose and do not have to go through additional administrative procedures if desiring one option rather than the other. The outcome has been that initially only 8 per cent of those receiving care at home choose to have services in-kind only; the remainder choose cash only or a combination of cash and services (see Figure 3), despite the fact that the value of the cash awarded is lower than the value of the service benefit. And it is remarkable that this pattern is similar for people with very extensive care needs and people with fewer care needs (see Table 2). Since the introduction of the German long-term care insurance in 1991, the proportion choosing cash only has been declining gradually, but it is still above two-thirds.

In a number of other OECD countries, including Australia, the Netherlands, the United Kingdom and the United States, popularity among users has led to the growth of personal budgets, direct payments and consumer-directed home care schemes. For example in Australia, the Carer Payment available to people who, because of their caring responsibilities, are unable to participate in the workforce, has grown rapidly over recent years from 11,740 recipients caring for people aged 65 or more by June 1998 to 18,097 by June 2002. In Sweden, on the other hand, the option of being paid an Attendance Allowance for caring for a relative was taken up by only 4980 people in 2001 compared to 20,000 people about a decade ago, presumably because of the relatively higher coverage of publicly subsidised and high-quality home care.

These trends indicate that some older people find that the cash option is preferable to the service option. Whether this is a result of poor service quality, high user charges, lack of availability or stigma, we do not know. It could indicate a preference for choice between different sources of care, be it informal care provided by relatives or market care, and that the empowerment as consumers following from being able to choose among alternative public and non-public providers is appreciated because it increases flexibility and the control that older people have over their daily lives. As mentioned above, flexibility and self-determination are important since long-term care involves the most intimate aspects of a person's life. This is confirmed by a study of the outcome of the Personal Budget scheme in the Netherlands, which found that while care quality is roughly the same as when people needing care are referred administratively to a designated agency, people receiving care via a personal budget feel less dependent because they have more control over when care is provided and notably by whom (see Figure 4). Similar outcomes have been found in the Cash & Counseling pilots under Medicaid in the United States. Foster et al. (2003) conclude that relative to agency-directed care, consumer-directed care greatly improved satisfaction and reduced most unmet needs. To some extent, the flexibility aimed for with personal budgets can be achieved via the dialogue between the person needing care and the authorities assessing eligibility. In Norway, for example, the principle of allowing users to influence the care they receive is central to long-term care policy. People needing care can express their wishes but the final decision is made by representatives of the local government's agency.

It could be asked whether leaving it in the hands of older people to find a care assistant to employ themselves would not imply a risk for vulnerable older people, particularly those with cognitive impairments, if they receive services from an untrained or possibly neglectful care assistant. How great such risks are may depend on many factors, including cultures and institutions in civil society that differ across countries. However, it is remarkable that the recent quality study of the Austrian Cash Allowance for Care – a system that goes further than in other OECD countries in leaving it to older people and their families to find appropriate care – did not find any cases of very poor hygiene or open neglect when surveying the condition of 700 older people with care needs greater than 120 hours per month. There may be other quality problems, but there is no indication of older people being neglected when relying on consumer-directed rather than agency-based care. Likewise, the experience from the Cash & Counseling pilots under Medicaid in the United States shows that consumer direction of care did not adversely affect participants’ health and safety (Foster et al. 2003). Apparently, the informal support and surveillance from relatives and others in the community is sufficient to avoid this. But it remains essential for public authorities to monitor the conditions of vulnerable older people, notably those without any remaining relatives.

Caring for a relative can be rewarding but also emotionally very demanding. In strategies to promote home- and community-based care it may therefore be desirable to have sufficient flexibility to support the type of arrangement that best suits the needs and wishes of different people – especially considering the enormous contribution from informal care resources to the overall pool of care. There are various ways in which support can reduce the strain of informal care giving. Financial support can reduce the need to maintain full-time paid work, thereby freeing up time for care giving. Likewise, some provision of formal care and home help can lift part of the burden off the informal caregiver. Notably, so-called respite care, which takes over for short periods of time, can allow the informal caregiver a break. Education of informal caregivers has been found to be another important field of support, as relatives becoming caregivers are not necessarily skilled in basic tasks such as how best to lift an immobile person in and out of bed. Informal caregivers can also benefit from better understanding the medical condition of the dependant. Support groups, often established by voluntary organisations, can contribute to this education and also provide informal caregivers with a forum for talking about the emotional strains arising from heavy care work. Finally, as pointed out by Lundh (1999), allowing informal caregivers to better enjoy the gains and positive aspects of caring for a relative can reduce emotional strains. The appropriate form of support depends on the type of informal care and who provides it. In the future, the role of children and grandchildren as informal caregivers may decrease and the role of spouses and other senior caregivers may increase, the latter group often having health problems themselves. This changing composition of the group of informal caregivers calls for new forms of care and possibly more support for informal caregivers.

Finally, as the motive for allowing more choice is to increase flexibility for users, it should not be expected that one particular scheme suits all older people needing long-term care. In particular, for older people with dementia the option of employing a care assistant themselves is of little relevance unless they have relatives or friends to act on their behalf by directing the work of the care assistant. And typically it is found that younger and adult disabled individuals are more interested in managing their care themselves. But there are many other dimensions aside from health conditions and age that matter; for example in Norway, the Care Wage is more frequently used in sparsely populated areas than in urban areas.

The long-term trends of ageing and changes in social relationships are likely to cause what some have called an ‘informal care crunch’ in many countries over the coming decades. The need for long-term care is likely to rise while the availability of informal care is likely to decline, as fewer older people live with their children and more of these children work and therefore cannot be expected to provide care to the same extent as daughters have done previously. How far this transition has gone varies significantly across OECD countries.⁹ It raises two strategic questions for the development of long-term care policies: who should be expected to provide care in the future? And should the people doing so be paid for it? Finding cost-effective answers to these questions is essential in the context of ageing and high demand and cost pressures elsewhere in health and social systems. Indeed, the economic rationale for paying informal caregivers depends much on their labour market attachment. However, the perspectives point in multiple directions:

• •

  For people who would otherwise be employed, payments for informal care, such as within a leave scheme, represent insurance against the loss of employment income they incur while providing care, and as insurance that they can return to the labour market, especially if there is a legal right to leave attached. Such payments allow families to choose informal care, and to the extent that such care replaces more expensive care that would have been provided formally and been publicly funded, the effect on public finances may be positive in the short run. Much, however, depends on the long-term labour market impacts. Particularly for those people with a loose attachment to the labour market, a prolonged period of leave can lead to subsequent unemployment as the person's skills or human capital may gradually deteriorate. Women in their 50s taking leave to care for a parent or parent-in-law may frequently be at risk in this regard.¹⁰ Very high payments to informal caregivers are therefore a mixed blessing by producing an unemployment or low-income trap that reduces the incentive for low-skilled people to maintain contact with the labour market. This problem is aggravated if payments for informal care can be received on top of unemployment benefits; cash payments to the people needing care make it essential to ensure that people receiving unemployment benefits are actively searching for regular employment in order to avoid a situation in which the combined income from unemployment benefits and some informal care giving exceeds what the person could obtain in a regular job. In leave schemes for informal caregivers, some countries confine the duration of leave to a limited period of time (such as during terminal illness) in order to avoid a situation in which caregivers loose touch with the formal labour market. Part-time leave may help carers maintain contact with the labour market; active assistance to help caregivers find work and a caregiver-friendly work culture are also important. Finally, promoting a market where quality care services can be purchased can help create new jobs and avoid forcing highly skilled people to leave their job to provide informal care, thereby reducing the return to individuals and society of their (often publicly funded) education.

• •

  For people who are more or less permanently outside employment and have other work income, such as retired people having a pension, care giving does not entail income loss and therefore there is no insurance argument for supporting informal care giving at home financially. However, to limit the need for costly formal long-term care services, some countries seek to mobilize and recognize informal caregivers for their work with payments, disregarding their labour market attachment and other income. Other countries find it essential not to engage in paying pensions to care for their relatives in order not to crowd out other spending that is more essential to maintaining a welfare society. The issue of recognising and supporting informal care giving has an important gender aspect, as most caregivers are women (Jenson and Jacobzone 2000).

There is a tendency for countries with extensive provision of formal home care but only limited financial support for informal care (such as the Scandinavian countries) to have higher employment rates for women aged 50–59 than, for example, the United Kingdom, Germany, Austria and Luxembourg which are countries in this study characterised by limited or average provision of formal home care but extensive financial support for informal care via cash allowances (see Figure 5). This finding is supported by studies using micro data which show that British informal caregivers bear costs, particularly in terms of lower wages, compared with others (Carmichael and Charles 1998,2003; Heitmueller and Inglis 2004). That is not to say, however, that more provision of formal home care services will automatically bring higher labour market participation rates. It is remarkable that among the Scandinavian countries, Sweden's employment rate for women aged 50–59 is higher than that of Norway and Denmark in both part-time and full-time jobs, despite providing publicly funded formal home care to only half as many older people as does Norway and Denmark.

The question is: what is the cause and what is the effect? The lower employment rates of women aged 50–59 in countries such as Austria, Germany and Luxembourg probably to some extent reflect a cultural factor, reflecting a greater preference for living in one-earner couples. In this context, cash allowances may allow older individuals to acquire care from these people informally in a way that is much less costly to society than if formal services were to be provided. The Netherlands clearly shows that the availability of formal home care is not the only factor determining the employment rates of females in their 50s.

A particular concern is the adverse incentives for labour supply that can arise from programmes where people needing care are given cash, part of which they pass on to informal caregivers. Indeed, where the person needing care and the informal caregiver are living in the same household, a cash allowance for care naturally enters the common household budget. Also, if the informal caregiver is a close relative such as a daughter or son of the person needing care, the people involved may not think of passing on the money as something that should be reported to any authorities – and indeed tax and social security legislation may not take such payments into consideration. The consequence can be that the combination of taxation, unemployment benefits, social assistance and cash allowances for care redirected informally to the informal caregiver reduces the incentives for some unemployed people to actively search for employment while providing informal care for 15–20 hours a week, for example (Lundsgaard 2005).¹¹

The type of choice arrangements for home care, the level of support they provide and the number of users were shown to vary enormously across and within OECD countries – reflecting also that long-term care provision differs more significantly across countries than acute health care systems. Choice arrangements (e.g., in the Netherlands) are preferred by a number of elderly people, where some report that they feel more satisfied and less dependent when having a say about how, when and notably by whom care is provided. Indications of shortfalls in care quality or outright neglect of frail older people were shown in one study to be remarkably rare, but informal caregivers – most being female – are known to carry sometimes large burdens in terms of physical pain and emotional strain from feelings of excessive responsibility, overload and isolation when working alone.

The implications for employment and fiscal sustainability are fairly complex. Giving older people a budget or cash to pay informal caregivers can help tap into a wider pool of human resources where there are shortages of professional care workers. On the other hand, a functioning market for formal home care services (or public supply of such services) is essential to allow relatives of older people in need of care to maintain their attachment to the normal labour market. And payments for informal care can risk creating ‘incentive traps’ that attract informal caregivers away from the normal labour market, if the interaction between informal care payments, taxes, unemployment benefits and other transfer incomes is not controlled well. After having been away from a normal job for a period of time, it can prove difficult to return.

Future ageing of populations together with demands for active ageing will increase the policy relevance of these issues. The large cohorts moving into old age in the coming decades may bring the fiscal sustainability of current health and long-term care systems into question. In this context, publicly funded formal home care, and, in particular, payments for informal care, will have to be carefully targeted. With more years in retirement, of which many are spent without major disabling conditions, the growing number of healthy and active senior citizens represents a potentially very valuable resource as informal caregivers. Finding the best way of nurturing this potential, and thereby shifting the task of providing informal care away from working-age children and towards able seniors (spouses, neighbours and others in the local community) may well prove to be key to achieving fiscal sustainability.

The preparation of this article has benefited from valuable comments and discussions with Manfred Huber and other members of the OECD Secretariat, including Martine Durand, Mark Pearson, Peter Scherer, Junichi Izumi, Patrick Hennessy, Weonjong Kim and Andres Fuentes, but first and foremost it would not have been possible to produce without the input from the representatives of the OECD member countries in the long-term care expert group and the Ad Hoc Group of the OECD Health Project. The article has also benefited from discussions with participants at the ‘Consumerism of Care for the Elderly’ symposium organised by the Danish National Institute of Social Research (SFI) in Copenhagen, August 2004, where an early draft was presented. The project has been funded partly by the Office of the Assistant Secretary for Planning and Evaluation (ASPE), Department of Health and Human Services, in the United States.

Jens Lundsgaard is an economist at the OECD. His work concerns public finance, health and long-term care as well as economic policies more broadly. His current responsibility is to prepare the OECD Economic Surveys for the United Kingdom and Finland. As part of the OECD Health Project he has analysed the role of choice and support for informal care in OECD countries. It is the results of this analysis that are presented in this article. Before joining the OECD he worked for the Ministry of Finance in Denmark from 1998 to 2001.