This chapter presents the costs of long-term care and discusses the generosity of public support in covering those costs. In the absence of public support, the costs of long-term care to be borne by an individual would be extremely high. For that reason, public systems cover a share of the costs but there is great heterogeneity in the generosity across OECD countries. Public long-term care systems tend to cover a higher share of the cost for more vulnerable individuals, that is, people with high needs and low income. However, gaps in public support remain, leaving individuals with high out-of-pocket costs. While many countries apply income and wealth-testing in their eligibility assessments, income thresholds currently still leave the most vulnerable population with insufficient public support.
Is Care Affordable for Older People?
Abstract
Populations are increasingly concerned about access to long-term care (LTC) and the financial costs it entails. LTC needs can persist for many years, with lifetime costs potentially reaching catastrophic sums. According to the OECD Risks that Matter survey (OECD, 2023[1]), 56% of respondents worry about securing good-quality LTC for older relatives. Looking ahead to the next decade, 67% and 65% of respondents worry about accessing good-quality LTC for themselves and for older family members, respectively.
In this context, public schemes play an important role in mitigating the financial burden of LTC to prevent old-age poverty and unmet needs. The private sector provides only limited options for pooling the risk of high LTC costs. In most countries, there are few private insurance options available, because of a variety of reasons. Market failures may be important, such as adverse selection. People may not plan sufficiently due to hyperbolic discounting (i.e. valuing immediate smaller rewards much more than larger long-term gains) or a myopic view of risk. Evidence shows that few people protect themselves from the potentially high costs of LTC. Only a small percentage of older households purchase insurance against these costs or accumulate savings (Webb, 2001[2]; Jensen and Theisen Cramer, 2006[3]; Cremer, Pestieau and Ponthiere, 2012[4]; Ameriks et al., 2020[5]).
This chapter provides cross-country comparable evidence on the actual cost of LTC that individuals face as they develop LTC needs. The first section examines the full LTC cost that older people would incur in the absence of any public social protection, to fully grasp the financial burden on older people with LTC needs seeking professional care. Comparing these costs to an individual’s income highlights the potential financial challenges of developing LTC needs. In the second section, LTC costs are compared to the public support available for each level of needs. This gives crucial insight into the generosity of public support across countries. Next, the remaining share of the cost is calculated to show the part of the cost that individuals have to pay for, i.e. the out-of-pocket cost. Finally, the chapter describes how the variation in the generosity is related to the set-up of public social protection systems sharing the costs of care with older people and their families.
For the majority of older people, the costs of long-term care (LTC) services would be unaffordable in the absence of public support. In virtually all the 32 OECD and EU countries and subnational areas covered in this report, median-income‑earning older individuals with severe needs would have to spend more than their income on home care or institutional care, were it not for any public support. For instance, in Denmark, Finland, the Netherlands and Sweden, the total costs of home care or institutional care for severe needs are at least triple the median disposable income of older individuals. This is precisely why most countries adopt some form of public support to ensure the financial accessibility of LTC services.
Countries differ widely in how much benefits and services cover the overall costs of long-term care. For a person with moderate needs, ten countries cover more than 90% of the costs, while in 11 countries public systems cover between 50% and 90% of the costs and in 10 countries less than 50% of the costs are covered by public systems. In three countries, none of the costs of LTC are covered by public support for individuals with moderate needs.
Public support tends to be higher for older people with more severe needs. For a person with severe needs, public social protection systems cover more than 50% of the home care costs in 22 countries and over 90% in 11 of them. For someone with low needs, 19 countries cover at least 50% of the home care costs and 6 countries more than 90%, while in 7 countries public support is below 10% or even zero. Public support for someone with severe needs tends to be higher at home than in institutional care but there is a lot of variation across countries.
Informal caregivers are often not financially compensated by public support. In half of the countries, there is no public support if care is provided by an informal caregiver, such as a child or a spouse. Even if there is public support, this represents a lower compensation than if formal care is provided. On average, only 14% of the costs of care are covered in case of informal care compared with just over 60% for formal care.
The out-of-pocket costs in some countries remain large despite public support. For individuals with moderate needs, in 11 countries out-of-pocket costs would be at least 50% of median income. In 16 countries, for individuals with severe needs, out-of-pocket costs for home care exceed 50% of median income while they are above median income in seven countries.
In most countries, generosity for older people with a low income would be greater than for those with higher incomes but progressivity across the income distribution can be improved. In 17 countries and subnational regions, public share of the cost is higher for those on low incomes than for those in the top 20th percentile of the income distribution. In a number of countries, public coverage decreases sharply as income goes above a certain threshold. In others, public support decreases only for incomes which are already high above the median income while several public support decreases progressively along the income distribution. At the same time, there are some gaps in public support for individuals with low income. In seven countries, an older person with moderate needs and low income would have to devote more than half of their income to pay for home care, leaving less than half of their already lower income to cover basic living expenses.
While only 40% of countries perform wealth-testing, this has a high impact on out-pocket costs for such individuals. In 11 countries that implement wealth-testing, individuals with median income and median wealth will have out-of-pocket costs that exceed their income. This means that such people would need to cash their wealth to pay for long-term care. Wealth-testing could be better designed to achieve a more efficient use of public LTC systems without causing the risk of financial distress.
This section gathers information on LTC costs from both national and sub-national authorities through a detailed survey. The costs covered include all forms of care, from direct services in Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs) to addressing social needs. The costs refer to care costs that individuals would face without any governmental social protection measures if they had to pay for it themselves. These expenses, referred to as the costs of LTC, are examined in relation to individuals’ income to assess potential financial challenges as they develop LTC needs. To allow for cross-country comparisons, it relies on a methodology of typical cases of need (Cravo Oliveira Hashiguchi and Llena-Nozal, 2020[6]) which are described briefly in Chapter 1, Box 1.1.
3.1.1. Total costs of formal long-term care are high, especially for care provided at home
Without public support, the costs of LTC for older people would be unaffordable in most countries. The costs of LTC can pose a substantial financial challenge for older people, often consuming a large portion of their disposable income. For individuals with severe needs, equivalent to 41.25 hours of care weekly, these LTC costs can be almost seven times the median disposable income of people aged 65 years and more, depending on the country (see Figure 3.1). The highest LTC costs are observed in Sweden, Italy (South Tyrol), and Czechia. Conversely, in countries like Greece, the Slovak Republic and California (United States), the cost of care is more affordable for someone with the national median income. However, even in these countries, the cost of LTC is almost equal to or higher than the median income.
Compared to the national median disposable incomes for older people, without public support
Note: Severe needs correspond to 41.25 hours of care per week. Detailed descriptions of care recipients’ needs are available in Annex A. An older person with severe needs receiving LTC at home is assumed to live with a spouse who can provide 24‑hour supervision, help with taking medicines, and manage the finances, but cannot provide any other ADL/IADL care. Data on the cost of institutional care for severe needs in Israel is missing.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
In the majority of OECD countries, the cost of home care is higher than the cost of staying in a nursing home for an older person with severe needs. Even when including the costs of board and lodging, Figure 3.1 illustrates that, in a number of countries, institutional care may be an overall less costly way to meet more severe LTC needs in old age. For example, in Italy (South Tyrol), Czechia and Slovenia, the total cost of providing 41.25 hours of home care per week is more than double that of institutional care. Formal home care can thus be very expensive when needs are severe and involve many hours of care each week. Professional carers must travel between care recipients’ homes, which in some countries can take significant amounts of time during which they are not providing care. This limits the number of older people they can care for at any given time.
To protect older people against these potentially high costs, public social protection systems cover a certain share of the total costs. The generosity of these systems varies both between and within countries, depending on a person’s needs, income, and wealth. This section explores the extent to which public long-term care systems are generous to older individuals with different levels of long-term care needs.
3.2.1. Public support tends to be greater for older people with more severe needs
Public support varies greatly across countries. All OECD and EU countries covered in this report provide some public support for individuals with LTC needs. For a person with severe needs, public social protection systems cover more than 50% of the costs in two‑thirds of countries and over 90% in one‑third of them and one‑third cover less than 50% of the costs. Similar proportions hold for those with moderate needs. The highest generosity for all levels of care need is granted by systems in the Nordic countries (Denmark, Finland, Iceland, Sweden), as well as Luxembourg, Canada (Ontario) and the Netherlands. Generosity is more limited for a person with any level of needs and median income in Estonia, Czechia and Croatia.
Most support is granted to individuals with more severe needs, for whom care is usually more expensive and more likely to be unaffordable (see Figure 3.2). In the United States (California and Illinois), Croatia and Greece, public support is limited for low and moderate needs, but more generous for people with severe needs. In ten countries, the relative generosity is greater for individuals with low needs than for those with more severe needs. Limits to the number of hours of care that can be covered through public LTC benefits can contribute to this pattern, especially in France, Korea and Slovenia. In Latvia and Lithuania, social protection is more generous with individuals with median income but only when they have at least moderate needs.
On average, the generosity of public benefits is greater for care provided at home, compared with institutional care (Figure 3.2, Panel B). On average, around 66% of home care costs are covered, while the average covered cost of institutional care is over 10 percentage points lower. By contrast, the share of costs covered by public LTC systems is higher for institutional care in Estonia, Portugal, Czechia, the United States (California and Illinois), Korea, Croatia, Italy and Germany. The public support for institutional care is the highest in Sweden, Ireland, the Netherlands, Germany and Denmark. At the other end of the scale, there is no public support for institutional care for a person with median income in Israel, Poland and Greece. The support is also very limited (below 20%) for older people in the Slovak Republic, Slovenia and Spain. This large difference is partly due to the design of the LTC systems, some of which aim at covering all costs, while others are designed to provide partial support with the expectation that older people organise most of their support themselves.
For individuals with median income and zero wealth
Note: Low, moderate, and severe needs correspond to 6.5, 22.5 and 41.25 hours of care per week, respectively. An older person with severe needs receiving LTC at home is assumed to live with a spouse who can provide 24‑hour supervision, help with taking medicines, and manage the finances, but cannot provide any other ADL/IADL care. Detailed descriptions of care recipients’ needs are available in Annex A. Data on the cost of institutional care for severe needs in Israel is missing.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
3.2.2. Financial public support for the provision of informal care is limited
Informal care is widespread in many countries.1 Providing care involves a substantial investment of resources for family and friends and has manifold effects on their health, labour market outcomes and social life. Adequate public support for these informal carers is not always forthcoming in OECD countries (Rocard and Llena-Nozal, 2022[7]). Indeed, they often receive little or no training and psychological support, likely resulting in a lower quality of care for recipients. While most OECD countries provide some financial support to informal carers through cash benefits, these rarely constitute a remuneration for the carers’ investment and neither comes close to the full costs of formal care.
Most countries see the support for informal carers as a recognition that providing care involves costs for carers and as a (partial) compensation for the opportunity costs of providing care – that is, for lower incomes caused by reduced working hours. Low support for informal carers is partially explained by the risk that high compensations could trap carers in a role that is comparatively low-paid, informal (e.g. with no social security contributions) and for which they have no training. In addition, countries are reluctant to make informal care too attractive because care could have been provided even without compensation, also because many older people prefer to be cared for in their own homes by their relatives and friends. It is difficult to strike a good balance between adequately compensating informal carers and avoiding incentivising informal employment.
Public support for the provision of informal care is limited, especially compared with formal care and tends to be low (Figure 3.3). In nearly half of the countries, an adult child providing 22.5 hours of care to an older parent would receive no public support, even though the adult child would have to reduce working hours or even resign from work to provide care. Even some countries with generous support for formal care like Iceland and Sweden do not provide support for informal carers.
On average, only 14% of the costs of care are covered in case of informal care compared with just over 60% for formal care. The most generous support for informal carers is estimated for Malta (more than 90% of formal care cost), followed by Belgium, the Netherlands, Germany and Denmark, with their informal carer support covering more than 40% of formal care cost. In the meantime, Greece and the United States (California and Illinois) provide limited support not only for formal care but also for informal care.
As share of formal home care costs for older people with moderate needs, median income and zero wealth
Note: Moderate needs correspond to 22.5 hours of care per week. Median income refers to the country disposable median income of older people. The care is provided by adult child who earns country median income of the entire population. Detailed descriptions of care recipients’ need, and the informal carer’s characteristics are available in Annex A.
Source: OECD analyses based on the OECD Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database and the OECD Wealth Distribution Database.
While public social protection systems tend to provide greater support for those with higher needs, even small out-of-pocket payments can represent a large proportion of the incomes of older people with limited financial resources. This section looks more in detail at these out-of-pocket costs – the shares of the total LTC costs that are left for older people to pay even after considering public support relative to the older people’s income.
3.3.1. Even with public support, home care for severe needs could be too costly
Out-of-pocket costs remain limited for individuals with low needs in most countries and for moderate needs in two‑thirds of the countries. For individuals with low needs, in all countries, out-of-pocket costs are below the median income (Figure 3.4, Panel A). The median income is taken as a benchmark because, if costs are higher, this would mean that the remaining disposable income left might not be enough to cover basic living costs such as rent or utilities, food, or clothing. For individuals with moderate needs, in 11 countries, out-of-pocket costs would be at least 50% of median income (i.e. relative poverty line). Given that old persons’ median income is generally lower than that of working-age individuals, it is unlikely that they can afford these basic living costs after spending a significant part of their income on LTC.
Out-of-pocket expenses are high in a majority of countries for individuals with severe needs at home. In 16 countries and subnational regions, out-of-pocket costs exceed 50% of median income while they are above median income itself in seven countries and subnational regions. On the other hand, in Denmark, Canada, Finland, Iceland, Luxembourg, Malta and Belgium, out-of-pocket costs for those with severe needs at home are below 10% of the median income.
As a share of median income, after receipt of public support
Note: Low, median and high incomes mean the bottom 20th, 50th and 80th percentile of net disposable income among individuals aged 65 years and over, respectively. Low, moderate, and severe needs correspond to 6.5, 22.5 and 41.25 hours of care per week, respectively. An older person with severe needs receiving LTC at home is assumed to live with a spouse who can provide 24‑hour supervision, help with taking medicines, and manage the finances, but cannot provide any other ADL/IADL care. Detailed descriptions of care recipients’ needs are available in Annex A. In Panel A, an estimated out-of-pocket cost for home care for severe needs is 468% for Czechia, which is very high compared to other estimates and thus suppressed for better readability. In Panel B, estimated out-of-pocket costs for institutional care for low, medium, and high-income individuals were 1 107%, 579% and 338%, respectively and thus they are suppressed for better readability.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
3.3.2. The financial burden of institutional care is unevenly distributed
In most countries, out-of-pocket costs for institutional care are below the national median income. Only in six countries (Estonia, Portugal, Greece, the Slovak Republic, Poland and Israel), the out-of-pocket costs of institutional care for people with severe needs are estimated to be greater than their median income – see Figure 3.4, Panel B. This out-of-pocket cost already includes the costs for board and lodging, so it is safe to assume that older people in institutions do not have many other additional costs. At the same time, some countries do consider that a minimum amount of income equivalent to 10‑20% of median income might be necessary to face other expenses such as clothing and other. In 17 OECD countries, individuals would not be able to afford these other purchases, as their out-of-pocket costs for institutional care would be above 80% of median income.
While institutional care tends to be affordable for those in the top income quintile across nearly all OECD countries (with the exception of Israel), individuals with low income face challenges to afford institutional care. In just over half of the OECD and EU countries, an older person with low income may need to spend around 80% of their disposable income on institutional care. This raises significant concerns and underscores the need for policy measures to mitigate poverty among the older individuals (see detailed discussion on poverty reduction in Chapter 4).
Many countries make people’s out-of-pocket expenditure dependent on their income. Requiring older people with higher incomes to contribute more to the cost of their care is a good way to increase efficiency and effectiveness of LTC systems. Notably, in France, Germany, Latvia, the Netherlands, Finland, Iceland, Austria, Hungary, the United Kingdom (England), Croatia and the United States (California and Illinois), individuals with higher incomes are expected to spend a greater proportion of their income on care compared to those with lower incomes.
In ensuring that public support contributes to mitigating the financial burden of LTC services (see Section 3.2), OECD countries employ a wide range of methods to fund their public LTC systems: e.g. taxation (general or earmarked tax revenues reallocated to LTC), health insurance, long-term care insurance, and various combinations thereof (OECD, 2020[8]; Lee, Chon and Kim, 2023[9]) – see Chapter 5). At the same time, some mechanism of cost-sharing with care recipients is implemented by almost all countries covered in this report, where LTC beneficiaries contribute to a portion of the costs for LTC services they are expected to receive. Cost-sharing with care recipients typically involves a process to assess their financial capacity in determining the level of support from the public system and out-of-pocket costs from the care recipients. This procedure is widely known as means-testing, and it is implemented in many jurisdictions to impose a higher level of cost-sharing from individuals with a higher level of income and/or wealth.
3.4.1. Means-testing is widely used aiming to efficiently allocate support
Income‑testing is the more common form of means-testing. Table 3.1 summarises public LTC systems in OECD countries or subnational areas covered in this report, including details regarding whether any income‑testing and wealth-testing are conducted. Czechia, Denmark, Germany, Poland and Portugal are the exceptions where neither type of means-testing is conducted in determining the degree of public support. There is a general tendency with countries (and regions) only implementing income‑testing, such as Austria (Vienna), Canada (Ontario), Estonia (Tallin), Finland, France, Iceland (Reykjavik), Israel, Latvia, Malta, the Slovak Republic and Sweden. Meanwhile, wealth-testing appears to be more extensively used to assess the financial standing of institutional care residents. This is the case with Estonia, (Tallin), Greece, Hungary, Ireland, Italy (South Tyrol), Japan, Lithuania, Spain, the United Kingdom (England) and the United States (California, Illinois).
Note: Benefits and schemes that are not applicable to the typical cases of LTC needs used in this report are not included in this table. Benefits that are marked as income‑ or wealth-tested, if income‑ and/or wealth-tested are taken into account for at least one of the typical cases of LTC needs in this report. * In Germany, means-testing can be used for other services, but it is not applied to the provision of long-term care services.
1. 1. In France, tax reductions depend on income.
2. In Iceland, there is no income‑testing for institutional care but a specific ceiling is placed on out-of-pocket costs.
3. In Japan, wealth-testing is applied to determine the eligibility of food and accommodation fee reductions at nursing homes.
4. Although the benefit is not wealth-tested when it is given in Luxembourg, all sums paid through this benefit can be recovered from the care recipient’s estate upon death.
Source: OECD analyses based on the OECD Long-Term Care Social Protection questionnaire.
3.4.2. Public support is generally lower for older people with higher incomes but it is less so for people with greater wealth
Of the 32 OECD and EU countries and subnational areas covered in this report, at least 28 use income‑testing so that care recipients with higher incomes have less public support and those with lower incomes greater public support. The conventional design of income‑testing directs people with higher income (defined by particular thresholds) to pay more towards the total costs of care and thereby contribute more of their higher income to easing the fiscal burden of LTC support systems.
In 18 countries, the generosity of public systems measured by the share of the LTC costs covered by public support (for home care) is higher for those on low incomes than for those in the top 20th percentile of the income distribution (Figure 3.5, Panel A). Differences in the level of public support for low-income and high-income individuals are significant in Estonia (100 percentage points) and the United States-Illinois – (72 percentage points), England, the United Kingdom (69 percentage points) and Spain (65 percentage points). Additionally, the generosity of public systems in these countries moderately differs between low-income and median-income individuals, albeit with great variance including those observed for Finland (1 percentage point) and Estonia (100 percentage points).
Share of total home care costs that would be covered by public social protection, for respondents with zero wealth, by income, and for respondents with median income, by wealth, for older people with moderate needs
Note: Low, median, and high incomes refer to the upper boundary of 20th, 50th and 80th percentile of the income distribution among older people, respectively. Median wealth refers to the upper boundary of 50th percentile of the wealth distribution among older people. Moderate needs correspond to 22.5 hours of care per week. Detailed descriptions of care recipients’ needs are available in Annex A.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
By contrast, in the other 14 countries, our estimation indicates that the public coverage for home care is the same for all income groups. This can be partially explained by the absence of income‑testing (e.g, Czechia, Denmark, Germany, Poland, Portugal) or by the fact that income‑testing only applies to institutional care residents (e.g. Canada (Ontario), Ireland, Luxembourg, Malta, the Slovak Republic). Iceland, the Netherlands and Sweden do have income‑testing for home care but are still relatively indiscriminate with respect to LTC beneficiaries’ income.
When it comes to wealth-testing, more than half of the OECD countries and subnational areas incorporate it into their LTC systems but changes in the generosity are rather limited. In fact, gaps in the estimated public support between people with no wealth and people with median wealth noticeably change only in four countries (Figure 3.5, Panel B): namely, the United States (Illinois), Italy, Spain and the United Kingdom (England). Generally speaking, these differences are small compared to those from income‑testing – ranging from 7 percentage points (Illinois) to 54 percentage points (England). The underlying cause of these outcomes may be related to the setting of thresholds for wealth-testing – see more discussions in 3.4.4.
Conversely, the existing setting of income‑testing thresholds greatly affects the extent to which public support can be provided (Section 3.4.4 explores the specific details of such thresholds). To visualise the effect of income‑testing thresholds on public support and to best recapitulate three different approaches to threshold setting, Figure 3.6 illustrates curves of public support coverage relative to income for home care in Croatia, Israel and Lithuania. Note that the subsequent analysis generally excludes countries that do not conduct income‑testing at all or do so only for institutional care.2
First of all, in countries such as Croatia, Estonia, Greece, Latvia, Slovenia and the United States (California and Illinois), individuals with low income receive the maximum level of public support, but for people with median income, the level of public support is significantly reduced. In Croatia, for instance, an older individual with median income sees 8% of the total LTC costs of home care covered by the state (Figure 3.5, Panel A). Those who earn less than one‑third of the median income would be fully covered, although this threshold effectively excludes people around the poverty line (see Figure 3.6).
For an older person with moderate needs and zero wealth, receiving care at home, as a share of the maximum social protection for three example countries (Croatia, Lithuania and Israel)
Note: Moderate needs correspond to 22.5 hours of care per week. Detailed descriptions of care recipients’ needs are available in Annex A. Median income refers to the country disposable median income of older people.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
In the second approach, public support for home care is at the maximum level or nearly so for people with median income and remains generous for high-income individuals with a reimbursement rate of approximately 50% or above. The public LTC systems of Israel, Finland, Iceland, Italy (South Tyrol), Japan, Korea, the Netherlands and Sweden share this feature and the generosity of their public systems are high across all income levels. Almost the same can be also said about the coverage of institutional care costs, except for Israel since there is no public support for institutional care.
Lastly, in countries like Lithuania, France, Spain and the United Kingdom (England), the share of public support declines in proportion to income and people receive limited public support once they acquire high income. In Lithuania, for instance, the coverage of public support progressively declines from 72% for those whose income is 50% of the national median income, to 46% for median income earners and to less than 30% for those with double the median income (Figure 3.6).
3.4.3. Out-of-pocket costs can still be substantial despite means-testing
Despite the prevalence of income‑testing, public support occasionally fails to identify low-income individuals and disadvantages low-to-median-income individuals. For example, out-of-pocket costs for low-income earners exceed 50% in seven out of 32 OECD and EU countries and subnational areas (Figure 3.7). This result may be expected for countries with no income‑testing (Croatia, Czechia and Poland), but it suggests that, even for countries that do conduct income‑testing (United States-California, Croatia, Czechia, Lithuania and Slovenia), their public support falls short of supporting people with low income.
Moreover, ten countries see out-of-pocket costs for median-income earners surpassing 50%. Except for Lithuania and Spain, the LTC systems of these countries are typically designed to target low-income people (see Section 3.4.2). Indeed, the share of out-of-pocket costs is substantially lower for people with low income than for people with median income in Estonia, Latvia and Illinois. However, in the United States (California), Croatia and Czechia, there seems to be some room for improvement on the income‑testing and generosity of their public LTC systems.
As a share of disposable income after public support, for older people with moderate needs and zero wealth
Note: Low, median, and high incomes refer to the upper boundary of 20th, 50th and 80th percentile of the income distribution among older people, respectively. Moderate needs correspond to 22.5 hours of care per week. Detailed descriptions of care recipients’ needs are available in Annex A. For Czechia, estimated out-of-pocket costs are 354% for low-income, 276% for median-income and 208% for high-income earners, which are very high compared to other estimates and thus suppressed for better readability.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
About 40% of the countries adjust the generosity of public systems with respect to a person’s wealth for institutional care (Figure 3.8). The impact of this wealth testing tool is on average resulting in out-of-pocket costs being 54 percentage points higher for individuals with median wealth. There is great heterogeneity on differences in out-of-pocket costs, depending on wealth with differences being small for Belgium (Flanders) and Spain but being particularly large for England (112 percentage points) and the United States (approx. 110‑120 percentage points). Moreover, out-of-pocket costs for those with median wealth exceed median income in the eleven countries that perform wealth-testing for institutional care. At the same time, Croatia, Greece, Japan and Slovenia have a wealth-testing scheme but there is no change in the estimated out-of-pocket costs, which may be affected by how wealth-testing thresholds are set in these countries.
As a share of disposable income after public support, for an older person with severe needs
Note: Median wealth refers to the upper boundary of 50th percentile of the wealth distribution among older people. Severe needs correspond to 41.25 hours of care per week. Detailed descriptions of care recipients’ needs are available in Annex A. For Israel, estimated out-of-pocket costs are 579% for people with no wealth and people with median wealth due to the lack of information on support for institutional care, which is very high compared to other estimates and thus suppressed for better readability.
Source: OECD analyses based on the Long-Term Care Social Protection questionnaire, the OECD Income Distribution Database, and the OECD Wealth Distribution Database.
3.4.4. Thresholds should be better set for cost-sharing without harming the vulnerable
The setting of means-testing thresholds should ideally achieve, i) general accessibility and affordability, ii) the identification of those who have low income and/or limited wealth to reduce their out-of-pocket costs, and iii) the targeting of people who have sufficient income and wealth to pay more for LTC services so that the fiscal sustainability and efficiency of public LTC systems can be reinforced (Oliveira Hashiguchi and Llena-Nozal, 2020[10]). As discussed in the previous section, however, out-of-pocket costs for the provision of care are rather substantial for people with low-to-median income earners and people with median wealth.
Figure 3.7 indicates that the estimated relative user contributions are 50% higher for people with median income than people at the top 20th income distribution, in 12 out of 32 OECD and EU countries and regions. As for income‑tests, it may be worth considering setting a threshold for high-income individuals to target public support at low-income counterparts in a more efficient manner. Imposing such thresholds does not immediately compromise the adequacy of public systems. For instance, the co-payment rate for high-income individuals (who accounted for 9.6% of old-age population in 2015‑16) changed from 10% to 20% in Japan, which statistically significantly reduced the intensity of LTC services but only by 0.46% (Sano et al., 2022[11]).
Wealth-tests can also contribute to better targeting social protection but there is a risk of incentivising distortionary saving behaviour around wealth-testing thresholds (Oliveira Hashiguchi and Llena-Nozal, 2020[10]). The current wealth-test criteria for institutional care in England, for example, requires an older individual with a capital of GBP 23 250 or more to be fully self-funded. This might induce eligible individuals to stop saving or transfer their capitals elsewhere (although the risk will be lower as the planned threshold of GBP 100 000 is significantly more generous). English local councils can charge more if institutional care residents are found to be engaged in such purposeful wealth depletion, but establishing the causal link is not straightforward (Foster, 2024[12]).
That being said, well-designed wealth-testing could incentivise a more efficient usage of LTC services without a significant risk of financial distress. The 2012 Dutch co-payment reform which includes increasing user contributions for those with a taxable amount of wealth resulted in a less intensive use of institutional care and a small decrease in public expenditure, while average lifetime out-of-pocket costs significantly increased (Tenand, Bakx and Wouterse, 2021[13]). To avoid the risk of disproportionately high care costs, countries could consider a lifetime cap on user contributions.
References
[5] Ameriks, J. et al. (2020), Long-Term-Care Utility and Late-in-Life Saving, Journal of Political Economy, https://doi.org/10.1086/706686.
[6] Cravo Oliveira Hashiguchi, T. and A. Llena-Nozal (2020), “The effectiveness of social protection for long-term care in old age”, OECD Health Working Papers, No. 117, OECD, Paris, https://doi.org/10.1787/2592f06e-en.
[4] Cremer, H., P. Pestieau and G. Ponthiere (2012), The economics of long-term care: A survey, Nordic Economic Policy Review, https://orbi.uliege.be/bitstream/2268/146126/1/core%20Nordic.pdf.
[12] Foster, D. (2024), Paying for adult social care in England, https://researchbriefings.files.parliament.uk/documents/SN01911/SN01911.pdf.
[3] Jensen, G. and A. Theisen Cramer (2006), Why Don’t People Buy Long-Term-Care Insurance?, The Journals of Gerontology: Series B, Volume 61, Issue 4,, https://academic.oup.com/psychsocgerontology/article/61/4/S185/603693.
[9] Lee, S., Y. Chon and Y. Kim (2023), “Comparative Analysis of Long-Term Care in OECD Countries: Focusing on Long-Term Care Financing Type”, Healthcare, Vol. 11/2, https://doi.org/10.3390/healthcare11020206.
[1] OECD (2023), Main Findings from the 2022 OECD Risks that Matter Survey, OECD Publishing, Paris, https://doi.org/10.1787/70aea928-en.
[8] OECD (2020), Long-term care and health care insurance in OECD and other countries, OECD Publishing, Paris, https://doi.org/10.1787/3eabc286-en.
[10] Oliveira Hashiguchi, T. and A. Llena-Nozal (2020), “The effectiveness of social protection for long-term care in old age: Is social protection reducing the risk of poverty associated with care needs?”, OECD Health Working Papers, No. 117, OECD Publishing, Paris, https://doi.org/10.1787/2592f06e-en.
[7] Rocard, E. and A. Llena-Nozal (2022), “Supporting informal carers of older people: Policies to leave no carer behind”, OECD Health Working Papers, No. 140, OECD Publishing, Paris, https://doi.org/10.1787/0f0c0d52-en.
[11] Sano, K. et al. (2022), “Effects of cost sharing on long-term care service utilization among home-dwelling older adults in Japan”, Health Policy, Vol. 126/12, pp. 1310-1316, https://doi.org/10.1016/j.healthpol.2022.10.002.
[13] Tenand, M., P. Bakx and B. Wouterse (2021), “The impact of co-payments for nursing home care on use, health and welfare”, CPB Discussion Paper, https://doi.org/10.34932/GDWF-PP65.
[2] Webb, A. (2001), The impact of the cost of long-term care on the saving of the elderly, https://users.nber.org/~confer/2001/si2001/webb.pdf.