ESTUDIOS SOBRE LA ECONOMÍA ESPAÑOLA

Joan Costa-Fonta Montserrat Font-Vilalta

EEE 201

March 2005

http://www.fedea.es/hojas/publicado.html

ISSN 1696-6384

Las opiniones contenidas en los Documentos de la Serie EEE, reflejan exclusivamente las de los autores y no necesariamente las de FEDEA.

The opinions in the EEE Series are the responsibility of the authors an therefore, do not necessarily coincide with those of the FEDEA.

Joan Costa-Font^a,b and Montserrat Font-Vilalta^b

^aLondon School of Economics, London, UK ^bDepartament de Teoria Econòmica & CASEPS, University of Barcelona, Barcelona, Spain

Contact Address: Joan Costa-Font (PhD), Departament de Teoria Econòmica, Universitat de Barcelona. Diagonal 690, 08034 Barcelona. E-mail: joancosta@ub.edu.

Acknowledgements: we are grateful to the comments of an anonymous referee form FEDEA, Documentos de Economia Española.

Abstract. Social and demographic changes are transforming the way societies cope with old age dependency, obliging them to extend the market, or public coverage, for long-term care insurance. This paper examines the financing alternatives for insuring long-term care in Spain and scrutinises the existing empirical evidence to evaluate the financing tools available for extending long-term care insurance. We argue that private long-term care insurance will play either a complementary or a supplementary role alongside a compulsory mainstream insurance scheme. Evidence from surveys on Spanish social attitudes suggests that the general population prefer publicly funded schemes, although these preferences are subject to significant regional heterogeneity.

Key words: old age dependency, ageing, social change and long-term care insurance.

1. Introduction

In modern, developed societies, old-age dependency requires to the need of long-term care (LTC) services. The term “long-term care” covers the attention given to provide for the physical, mental and social needs of the dependent elderly. In the past, LTC needs were traditionally met through informal caregiver networks, normally made up of relatives with an inter-generational responsibility to devote part of their lifetime to providing this care for their elders. However, social changes have shifted traditional care-giving structures in such a way that societies progressively rely on community and residential services and less on informal care giving. Both residential and community care come at an economic and social price which families may not be able to bear on their own without a loss of welfare. Insurance schemes are the obvious answer. Interestingly enough, in many European countries the financing of long-term care, unlike that of health care services, is not fully funded by mainstream public insurance schemes (though there are clear exceptions such as Germany, Comas et al, 2003). Moreover, even when public LTC services do legally exist, they seldom cover the entire cost; that is, public assistance is frequently subject to co-payments that vary according to need and financial means (Costa and Patxot, 2005).

There are several reasons for this situation. One is the fact that the need for LTC is a risk that individuals are able to foresee and something against which they are able to self-insure, although t the European public is relatively unaware of its importance. The extent of individual responsibility for finding the means to cover long-term care needs is increasingly dominating the debate on the financial reforms of the Welfare State. Indeed, it is frequently argued that those able to save for their future care needs should do so. If this is the case, the main financial policy priority of the public sector is to alert individuals to the situation and to make sure that sufficient information is readily available. The need for LTC represents a relatively new ‘social financial risk’ and there are strong arguments in favour of some sort of compulsory financial arrangement in order to cover its cost. Inadequate LTC affects individuals’ welfare at a stage in life when they are unprepared to deal with the financial consequences of their earlier lack of foresight. In this scenario, the role of the market in providing coverage for LTC might be questioned unless adequate information is provided from both the supply and demand side. Therefore, the transition from a ‘family-based’ model of LTC provision

based on informal care to a model that relies on formal services is likely to cause problems, particularly in the case of households that do not qualify for public LTC subsidies but are unable to pay for care. A variety of insurance schemes may be set up to deal with these risks, ranging from the introduction of a system based on ‘private welfare markets’ to the extension of the welfare state coverage that still maintains a place for private insurance.

In Europe, Spain and Italy are expected to be particularly affected by a rise in the elderly population in the coming decades. According to the most recent estimations of Spanish Institute of Statistics (INE) in 1999¹, 967,713 people older than 65 (and 559,434 older than 80) , suffer from severe and total dependency and thus in need of LTC. As expected, the prevalence of old age disability increases with the age individual’s age and gender. Indeed, while the prevalence of old age disability in men individuals aged 70-74 is 20%, it increases to 30% for old age women. On the other hand, the prevalence of old age disability amounts 60% for old age men and 65% for old age women, indicating that the gender gap declines with age. The estimates of the Libro Blanco de la Dependencia en España (2005), estimated that among total dependents, 54% suffer from moderate dependency, 31% severe dependency and only 15% are highly dependent. These estimates are based on the Encuesta de Discapacidades, Minusvalias y Estado de Salud (1999), which indicates that about 90% of dependent individuals live in the community (90%) and only about 10% are institutionalised.

The financial coverage of LTC is a key issue. Spain is a country where the public coverage for LTC has traditionally been scarce and has hardly ever had a primary role in public debates. Public support for old age dependents and caregivers is provided by local authorities and is subject to means tests which have assessed income, needs, and, more recently, housing assets. Furthermore, due to the decentralisation of the Welfare State, the extent of public assistance for LTC varies across regions (Casado and Lopez, 2001). Finally, the vast majority of the Spanish population are unaware of the public coverage for LTC and the extent to which public assistance is subject to copayments (Costa and Rovira, 2004).

Little has been published outside the US on the issue of LTC insurance. This paper aims to examine the role of insurance for LTC in Spain within the context of a Welfare State which may be seen as ‘omnipotent’ by the population, but which is subject to significant financial constraints. We discuss the main financial policy options available in Europe to deal with the coverage of LTC risks, providing empirical evidence based on expenditure forecasts and individual attitudes. Finally, we examine the case base of Spain, as an example of the debate taking place on the design of LTC insurance mechanisms in some European countries. The following section deals with the system of provision of LTC in Spain and provides evidence on comparative expenditure. The third section evaluates the different systems proposed for LTC financing and the article concludes with a discussion of the attitudes of the Spanish population towards LTC insurance and provision, which exhibit significant regional heterogeneity and suggest that a variety of insurance mechanisms may coexist effectively.

2. Long-term care financial risks in Spain: the starting point

The design of a LTC insurance system in Spain is a financial and social policy priority, for three reasons. The first is the relatively late but marked population ageing, which puts ‘transition generations’ at a particular risk. This generation is an age cohort that stands between two models of care giving, namely the ‘family’ model and the ‘community’ model. Second, the rapid social changes in family structures mean that family members may no longer be willing to provide care for their seniors. Finally, the capacity of welfare states to extend public coverage is limited without increasing the tax burden on the middle classes. Yet, although these features should in theory foster the development of the market for LTC and LTC insurance, the lack of individual awareness of public coverage and support for LTC limits the development of financing instruments to cover LTC risks (Costa and Rovira, 2004).

2.1 Demographic change

As Table 1 shows, the ageing phenomenon in Spain is particularly striking: it will accelerate between the periods 2030 up until 2050. Furthermore, the age cohort that is increasing the fastest is the elderly. Though morbidity patterns are unavoidably

¹ http://www.discapnet.es/documentos/estadisticas/eddes01-2.htm

uncertain, forecasts still highlight that a significant part of the Spanish population is likely to need long-term care. Furthermore, both the expected and current income (and wealth) of old age dependents are likely to be larger than that of their predecessors, suggesting that the demand for long-term care services in the near future may well be considerably higher. Furthermore, more recent data is found in the Libro Blanco de la Dependencia (2004), which indicates that from 2005 to 2020 the number of old age dependents will increase by 39%, which if examined only those with high dependency leads to a rise close to 43%.

Table 1. Central Eurostat forecasts for Spain

Source: Eurostat (2000).

2.2 Social change

Old age dependency is profoundly affected by the changes in attitudes to the idea of the family. Patterns of fertility have been modified drastically and the family can no longer be relied upon to provide a "space of social protection". In parallel, the more active role of women in formal labour markets means that they will be less likely to give up paid occupation to provide informal care for their relatives. These changes will increase demand for formal long-term care services. Some of these effects are already visible in the European Union. In southern European countries such as Spain, families provide care for roughly two-thirds (69%) of the non-institutionalised population, whilst community care remains markedly underdeveloped, though its level varies from country to country. This spectrum is in stark contrast to that of the United Kingdom where only

32% of the population receives informal aid alone, and community care is firmly established. Therefore, there is a clear north-south divide in the patterns of social provision in Europe. This is explained by the interaction of other factors such as the presence of multigenerational family structures. For instance, whereas in Spain 20% of adults live alone, in Germany the figure is 41% (Pickard, 2003). However, given that 60% of women between 25 and 44 are active in the labour market, the number of informal caregivers available in the near future is likely to decline significantly.

2.3 Public expenditure on long-term care

The financing of long-term care in most European countries is determined by debates on the limits of private responsibility and the budgetary restrictions of the Welfare State. However, it is a fact that whilst health care coverage is either universal or quasi-universal in most European countries, coverage for long-term care is highly limited, and the responsibility of families. The main reason is that unlike general health care LTC focuses more on ‘care’ than on ‘cure’, and may therefore be envisaged as a family rather than a public responsibility. In this case, the public sector in some European countries is the last resort for families unable to fund or, in certain circumstances, even to provide long-term care. Indeed, public intervention does not aim to entirely replace individuals in activities for which they normally take financial responsibility. Moreover, since public assistance to old age dependents at later stages of life might encourage them to preserve household wealth with a view to inheritance or bequests, public subsidisation is often questioned/. As a result, individual cost sharing in the funding of the costs of long-term care is established in most European countries, although no consensus has been reached on the question of determining the ‘optimal cost sharing for each individual.

Current forecasts for LTC expenditure show a level of uncertainty. It is estimated that total spending on assistance and care for elderly people varies between one and three per cent of GDP. The proportion of public funding in the United States is relatively low, at just over 50 per cent. It is higher in Scandinavian countries, Australia and the United Kingdom. Spain, Belgium and Japan are the countries with the lowest levels of public LTC spending (Table 2).

Although life expenditure may increase old-age dependency, this issue is still under debate. Indeed, old age dependency is not exclusively a matter of individual health, but is strongly determined by social change affecting informal care patterns. Currently, LTC expenditure in Spain accounts for no more than 0.65% of GDP. In a recent study (Comas et al, 2003) forecasted that the population receiving informal care will increase by 126% and the relative expenditure of old age dependency would amount to 1.3% of Spanish GDP in 2050 (close to the current figure in the US: 1.2% of GDP, Brown and Finkelstein, 2004). The underlying hypothesis assumes that informal care in Spain will fall to the level of the United Kingdom. For its part, the UK is expected to experience a rise in long-term care expenditure of about 65%, which would mean that 45% of the population would receive formal care, thus increasing public longterm care expenditure to 1.94% of GDP. Therefore, patterns of informal care have a significant impact on expenditure and the type of services chosen. Assuming that current patterns of old age dependency and informal care remain constant, that all demand will be covered by supply, that real unit costs stay stable and that the productivity adjusted with the forecasts of inflation for the European Union, Figure 1 shows patterns of long-term care expenditures for four European Union countries (Spain, Italy, UK and Germany). Interestingly, the patterns are similar.

Table 2. Total and Public expenditure on long-term care as % of GDP.

Source: Jacobzone(2000).

Figure 1. Forecasts of long-term care expenditure 2000-2050 (% GDP)

Source: Comas et al (2003)

3. Insuring long-term care

3.1 Financial schemes for insuring long-term care

The results of the previous section indicate that although the probability of serious dependency (that is, the inability to perform two or more ADLs, or activities of daily living), is not high, the potential economic cost of formal long-term care for an individual can be catastrophic (McCall et al, 1986). Some scholars argue (Brown and Finkestein, 2004) that the random and costly nature of long-term care makes it insurable, since risk-averse individuals will find the idea of insurance attractive. Though some informal caregivers inside the household may still be available, the need for complementary care and funding is urgent.

The financial schemes for long-term care can be classified in three types: personal, such as savings or self-insurance, non-personal, such as non-contributing schemes and taxation, and finally interpersonal transfer, among them social or voluntary insurance. The different systems for funding long-term care in the European Union and the US differ in the role of the public sector in funding long-term care as well as in the role of insurance markets (Kane et al, 1998; Lutzky and Alecxih, 1999). Nevertheless, in all of them some form of public insurance complements or substitute’s private insurance contracts. Private long-term care insurance has been the most widely developed instrument in the US, whereas European countries have some sort of public system to cover long-term care expenditure in the form of contributing schemes (e.g.,

Germany) and non-contribution schemes based on general taxation (for instance, Scandinavian countries and the United Kingdom).

3.2 From self-insurance to insurance markets

The financing of long-term care may be the responsibility of individuals and families if they are able to save or invest in assets to pay for future and foreseeable long-term care expenditures. However, because a share of the population may refuse to think of the future, face liquidity constraints or wish to leave some assets for inheritance (or might free-ride in the presence of some sort of public scheme) this option is limited. Furthermore, from the perspective of insurance theory, self-insurance will be a less efficient option than a potential insurance market for long-term care because saving in order to pay long-term care does not avoid the uncertainty surrounding future funding needs. Hence insurance markets appear to be relatively more efficient mechanisms for covering risks of this kind. However, evidence from the US suggests that the market development of long-term care insurance is slow. It was introduced in the US in 1974, but today it barely covers 6% of the population and accounts for no more than 1% of total expenditure on long-term care (Rivlin, A.M and Weiner, 1998). Only about 10% of elderly have private LTC insurance (Brown and Finkelstein, 2004). Some authors find evidence in the US market of a significant reluctance among the middle-aged to purchase LTC insurance (Merril, 1992) although in a seminal article, Pauly (1990) argues that Medicaid crowds out LTC insurance by creating incentives to exhaust one’s individual savings so as to be eligible for public funding. Interestingly, a recent study by Brown and Finkelstein (2004) suggests that although supply factors may limit the development of the market for long-term care they do not explain the small market size. Private LTC insurance has been developed as well in countries where the public sector plays a more active role, such as the UK, though it is still in its infancy there.

Current research indicates that private long-term care insurance may be affected by many factors; by moral hazard, in that it may discourage family members from providing care; by adverse selection, if the probability of dependency varies with insurance; by uncertainty about the duration of long-term care needs, which complicates insurance pricing; by the possibility of risk-selection by insurers ‘cream skimming’ potential insurees who present the lowest risks of future LTC; and by barriers for

individuals to choose from a variety of hypothetically suitable contracts to cover a relatively unfamiliar risk such as old age disability (Norton, 2000). Finally, potential consumers may short-sightedly underestimate their potential long-term care needs; they may misperceive the extent of public sector coverage and consider the LTC insurance contracts ‘too expensive’. Evidence from “willingness to pay” studies in Spain suggest that only about 20% of the population is prepared to fund LTC insurance (Costa and Rovira, 2004). Finally, Finkelstein and McGarry (2003), find evidence of ‘preferencebased’ selection in insuring for LTC, according to which risk perceptions correlate positively with the purchase of LTC insurance.

Yet this evidence does not imply that the market for long-term care insurance is not suitable or efficient in itself: it suggests rather that the development of this market, at least in the initial stages, will require public information campaigns, the design of insurance contracts that cater for the preferences of specific populations, and the inclusion of incentives that may contain informational asymmetries (e.g., cost-sharing and deductibles) but do not put off potential consumers. Contract designs should involve the public sector and find ways to persuade individuals not to exhaust their resources so as to become eligible for public assistance (e.g., Medicare in the US). Alternatively, a system of tax incentives could be included to speed up the development of an insurance market. It could be argued that these policies introduce fiscal incentives as an indirect way of raising public funds for the system, but at the same time the development of a market for LTC insurance is likely to reduce the pressure of future long-term care expenditure on the public sector.

3.3 Alternative to insurance markets: social insurance and taxation

In most European countries, welfare is traditionally provided by the public sector. However, within the mechanisms used by the public sector, a distinction can be made between contributory systems such as social insurance and non-contributory funding systems such as taxation. The former are based on the assumption that people’s entitlement to long-term care derives from their earlier contributions to the social funding pool, while the latter sees the Welfare State as a system in which long-term care is part of the public assistance network to which all are entitled, regardless of their contributions. However, this latter system is often modulated by the existence of cost sharing mechanisms (e.g., means-testing and co-payments), which often depend on an

individual’s income; it is often under-funded and, in the case of Spain, varies significantly from region to region in spite of the fact that funding is uniform (Costa and Patxot, 2005).

A clear-cut example of a contributory mechanism is the German long-term care system, first implemented in 1994. The German model is based on extending the social insurance scheme to cover LTC in the form of compulsory insurance for employers and employees, although high income employed are able to opt out and enter parallel private LTC insurance schemes. The system overcomes some of the limitations of long-term care insurance markets though it is still subject to the inefficiencies with the design of insurance contracts and is dependent on the business cycle. On the other hand, taxfunded systems such as those of Spain, Scandinavia and the UK rely on uniform funding schemes although they tend to be run by local authorities and are largely means tested. However, the nature of means testing may enhance some perverse incentives to exhaust one’s household wealth so as to become eligible for public funding. The role of cost-sharing has a significant impact on the use of long-term care; for instance, in Denmark access to residential care is free and 20% take this option, but in the United Kingdom, where abundant co-payments are in place, the figure is only 10%. On the other hand, means tests can give rise to the development of a complementary private insurance although, as noted above, this is still in the initial stages.

3.4 What does the public prefer?

The above section suggests that no system is without its drawbacks. Indeed, the system in place in a particular country may be the product of the preferences of the general public, which ideally are channelled through the political system, though of course there is also a risk of misinformation. To determine which system is best suited to Spain, we need to look at the preferences of the public. This issue is dealt with in the survey known as the Health Care Barometer (Barometro Sanitario, 2000), which asked interviewees that long-term care funding option they prefer. The responses indicate that 43% of the public are willing to contribute in line with their economic possibilities, 27% think that the contribution should be a percentage of their pension, 18% believe that they should not have to contribute at all, and 6% would be prepared to pay a fixed amount per month. In all cases the contribution vary according to the ability to pay. However, there is significant regional heterogeneity. Table 3 shows that the preferred

option is a relatively redistributive system, in which funding depends on economic capacity, although some regions such as Aragon and Extremadura prefer a system linked to the pensions system. In some autonomous communities such as Canary Islands, Madrid, Asturias, Catalonia, the Basque Country and La Rioja, a sizeable share of the population is unwilling to contribute to the funding of long-term care.

Table 3. Public preferences for public long-term care financing

Source: CIS, Barómetro Sanitario 2000.

4. Discussion

The funding of long-term care is a priority issue in an ageing Europe and in western societies. However, as we have suggested in this paper, there is no ideal system and several (private and public) insurance mechanisms may coexist. The share of public funding is limited by the public’s preferences for redistribution at later stages of life although in systems such as the Spanish one, where the public sector is the preferred alternative for covering long-term care financial risks, the key issue under debate is the optimal rate of cost sharing. The US experience suggests that the development of the private LTC market largely depends on the incentives that the system provides to the middle classes to find complementary or supplementary funding in addition to that

offered by the public sector. Since the population seems slow to purchase long-term care insurance at early ages (most begin at ages close to 60) more information should be made available on the extent of public coverage and the potential benefits of LTC insurance. Currently, LTC insurance is infrequent and costly.

The existence of a public system, and the fact that the coverage that it should provide is not clearly defined, are the strongest limitations on the design of private insurance schemes. The question that remains is how far the State should intervene in insuring LTC, and above all whether the financing of old age dependency is the responsibility of the individual or the State. Within this debate, and assuming that some form of public (or mixed) insurance will be developed in many European countries in coming years, the key question is how the design of a LTC insurance policy will interact with other public financing tools such as the taxation of inheritance and the provision of pensions. The Spanish Ombudsman Report in 2000 (Informe del Defesnor del Pueblo) pointed out a social insurance system as the benchmark model to follow. On the other hand, the recent report commissioned by the Ministry of Labour and Social Affairs ( Libro Blanco de la Dependencia, 2005) points towards the design of a universal and public system ( so called fourth pillar or public social service system) although the access to long-term care will still be monitored by the set up of costsharing schemes. However, the role of the effects of the future insurance design on the inter-generational equity as well as the potential role of the private sector is still subject to significant uncertainty. A recent study (Brown and Ficklestein, 2004) suggest on the basis of a calibration of an actuarial model for long-term care insurance that in the US content, Medicare is responsible for a limited demand for private long term care insurance whilst Medicare in itself is far from comprehensive coverage. Overall, evidence from both Spain and the US suggests that the development of public coverage for long-term care exert a non neutral effect over the development of the private insurance market and inter-generational equity which needs to be tackled in future research.

References

1. Brown, J and Finkelstein, A 2004. Supply or Demand? Why is the Market for Long-Term Care Insurance so Small?. NBER, Working Paper Series, 10782 (September).
2. Casado, D and Lopez Casasnovas, G, 2001. Vejez, dependencia y cuidados de larga duración. Colección Estudios Sociales, 6, Fundació La Caixa, Barcelona.
3. Chen, Y .2002. Finding long-term care, Paper presented at Retirement Implications of Demographic and Family Change Symposium, San Francisco, June
4. CIS Encuesta sobre la soledad de las personas mayores, CIS (Centro de Estudios Sociológicos) 1998. www.cis.es.
5. Comas, A; Costa-Font, J; Gori, C ; di Maio, A; Patxot, C; Pickard, L; Pozzi, A; Rothgang, H and Wittemberg, R . 2003. European Study of Long-term care Expenditure. Report to the European Commission, Employment and Social Affairs DG & PSSRU Discussion Paper 1840.
6. Costa-Font, J and Rovira 2004, J Willingness to Pay for Long-Term Care Coverage: the Role of Private Information and Self-Insurance Faculty of Economics and Business, Working Paper, Universitat de Barcelona e04/125.
7. Costa-Font, J and Patxot,J 2005. The design of the long-term care system in Spain: policy and financial constraints. Social Policy & Society, 4(1): 11-20 .
8. Feder, J 1999. Improving financing for long-term care: the political challenge. Journal of Ageing & Social policy, 10(3): 1-6
9. Finkelstein, A. and McGarry, K. 2003 ‘Private Information and its Effect on Market Equilibrium: New Evidence from Long-Term Care Insurance,’ National Bureau of Economic Research, WP 9957.
10. Jacobzone S. 1999.Ageing and care for frail elderly persons: an overview of international perspectives. OECD: Labour market and Social. Occasional papers nº 38, Paris: OCDE.
11. Jacobzone, S (2000). Health and Aging: international perspectives on long-term care. ISUMA, Volume 1 N° 2 (Autumn).
12. Kane RL, Kane R, Ladd RC, Veazie WN, 1998 . Variation in state spending for long-term care: factors associated with more balanced systems. Journal of Health Policy, Politics and Law, 23: 363-389.
13. Lutzky, S and Alecxih, L, 1999. Enabling informed consumer choice in long-term care insurance market. Journal of Ageing and Social Policy, 10:27-44.
14. Manton, K, 1982. Changing concepts of morbidity and mortality in the elderly population. The Milbank Quarterly, 60: 183-244.
15. Ministerio de Trabajo y Asuntos Sociales (2005). Libro Blanco de la Dependencia en España .
16. McCall, N , Rice, T and Sangl, J, 1986. Consumer Knowledge of Medicare and Supplemental Health Insurance Benefits. Health Service Research, 20: 633-57.
17. Merril, J. 1992 ‘A Test for our Society: How Long and for whom we Finance Long-Term Care,’ Inquiry 29: 176–87.
18. Norton, E. C. 2000 ‘Long-Term Care’ in A.J. Cuyler and J.P. Newhouse (eds), Handbook of Health Economics, Volume 1B. New York: Elsevier Science B.V., pp. 956–94.
19. Pauly, M.V 1990 ‘The Rational Non-Purchase of Long-term care Insurance,’ Journal of Political Economy, vol. 98, pp. 153–68.
20. Pickard, L 2003. ‘Trends in and projection for informal Care’. European Study of Long-Term Care, European Commission, Brussels, 2003.
21. Rivlin, A.M and Weiner, J.M 1988. ‘Caring for the disabled elderly : who will pay?’. Washington D.C: The Brookings institution.